diff options
| author | Mazdak Farrokhzad <twingoow@gmail.com> | 2018-08-19 18:34:46 +0200 |
|---|---|---|
| committer | GitHub <noreply@github.com> | 2018-08-19 18:34:46 +0200 |
| commit | 08b1d83a46848dd7bd778aeae67a1e529e95d8cd (patch) | |
| tree | 9153a34f91860b175afb24f904fd50ac09e77c4e /src/libcore | |
| parent | ac64ef33756d05557153e00211cdf8fcf65d4be3 (diff) | |
| parent | b355906919927ab3c879becd14392f023af883a1 (diff) | |
| download | rust-08b1d83a46848dd7bd778aeae67a1e529e95d8cd.tar.gz rust-08b1d83a46848dd7bd778aeae67a1e529e95d8cd.zip | |
Merge branch 'master' into feature/core_convert_id
Diffstat (limited to 'src/libcore')
116 files changed, 25263 insertions, 6320 deletions
diff --git a/src/libcore/Cargo.toml b/src/libcore/Cargo.toml index 5af63aa970f..321ed892ea9 100644 --- a/src/libcore/Cargo.toml +++ b/src/libcore/Cargo.toml @@ -2,6 +2,8 @@ authors = ["The Rust Project Developers"] name = "core" version = "0.0.0" +autotests = false +autobenches = false [lib] name = "core" @@ -16,3 +18,6 @@ path = "../libcore/tests/lib.rs" [[bench]] name = "corebenches" path = "../libcore/benches/lib.rs" + +[dev-dependencies] +rand = "0.4" diff --git a/src/libcore/alloc.rs b/src/libcore/alloc.rs new file mode 100644 index 00000000000..35e4eea756d --- /dev/null +++ b/src/libcore/alloc.rs @@ -0,0 +1,1233 @@ +// Copyright 2015 The Rust Project Developers. See the COPYRIGHT +// file at the top-level directory of this distribution and at +// http://rust-lang.org/COPYRIGHT. +// +// Licensed under the Apache License, Version 2.0 <LICENSE-APACHE or +// http://www.apache.org/licenses/LICENSE-2.0> or the MIT license +// <LICENSE-MIT or http://opensource.org/licenses/MIT>, at your +// option. This file may not be copied, modified, or distributed +// except according to those terms. + +//! Memory allocation APIs + +#![stable(feature = "alloc_module", since = "1.28.0")] + +use cmp; +use fmt; +use mem; +use usize; +use ptr::{self, NonNull}; +use num::NonZeroUsize; + +/// Represents the combination of a starting address and +/// a total capacity of the returned block. +#[unstable(feature = "allocator_api", issue = "32838")] +#[derive(Debug)] +pub struct Excess(pub NonNull<u8>, pub usize); + +fn size_align<T>() -> (usize, usize) { + (mem::size_of::<T>(), mem::align_of::<T>()) +} + +/// Layout of a block of memory. +/// +/// An instance of `Layout` describes a particular layout of memory. +/// You build a `Layout` up as an input to give to an allocator. +/// +/// All layouts have an associated non-negative size and a +/// power-of-two alignment. +/// +/// (Note however that layouts are *not* required to have positive +/// size, even though many allocators require that all memory +/// requests have positive size. A caller to the `Alloc::alloc` +/// method must either ensure that conditions like this are met, or +/// use specific allocators with looser requirements.) +#[stable(feature = "alloc_layout", since = "1.28.0")] +#[derive(Copy, Clone, Debug, PartialEq, Eq)] +#[lang = "alloc_layout"] +pub struct Layout { + // size of the requested block of memory, measured in bytes. + size_: usize, + + // alignment of the requested block of memory, measured in bytes. + // we ensure that this is always a power-of-two, because API's + // like `posix_memalign` require it and it is a reasonable + // constraint to impose on Layout constructors. + // + // (However, we do not analogously require `align >= sizeof(void*)`, + // even though that is *also* a requirement of `posix_memalign`.) + align_: NonZeroUsize, +} + +impl Layout { + /// Constructs a `Layout` from a given `size` and `align`, + /// or returns `LayoutErr` if either of the following conditions + /// are not met: + /// + /// * `align` must not be zero, + /// + /// * `align` must be a power of two, + /// + /// * `size`, when rounded up to the nearest multiple of `align`, + /// must not overflow (i.e. the rounded value must be less than + /// `usize::MAX`). + #[stable(feature = "alloc_layout", since = "1.28.0")] + #[inline] + pub fn from_size_align(size: usize, align: usize) -> Result<Self, LayoutErr> { + if !align.is_power_of_two() { + return Err(LayoutErr { private: () }); + } + + // (power-of-two implies align != 0.) + + // Rounded up size is: + // size_rounded_up = (size + align - 1) & !(align - 1); + // + // We know from above that align != 0. If adding (align - 1) + // does not overflow, then rounding up will be fine. + // + // Conversely, &-masking with !(align - 1) will subtract off + // only low-order-bits. Thus if overflow occurs with the sum, + // the &-mask cannot subtract enough to undo that overflow. + // + // Above implies that checking for summation overflow is both + // necessary and sufficient. + if size > usize::MAX - (align - 1) { + return Err(LayoutErr { private: () }); + } + + unsafe { + Ok(Layout::from_size_align_unchecked(size, align)) + } + } + + /// Creates a layout, bypassing all checks. + /// + /// # Safety + /// + /// This function is unsafe as it does not verify the preconditions from + /// [`Layout::from_size_align`](#method.from_size_align). + #[stable(feature = "alloc_layout", since = "1.28.0")] + #[inline] + pub unsafe fn from_size_align_unchecked(size: usize, align: usize) -> Self { + Layout { size_: size, align_: NonZeroUsize::new_unchecked(align) } + } + + /// The minimum size in bytes for a memory block of this layout. + #[stable(feature = "alloc_layout", since = "1.28.0")] + #[inline] + pub fn size(&self) -> usize { self.size_ } + + /// The minimum byte alignment for a memory block of this layout. + #[stable(feature = "alloc_layout", since = "1.28.0")] + #[inline] + pub fn align(&self) -> usize { self.align_.get() } + + /// Constructs a `Layout` suitable for holding a value of type `T`. + #[stable(feature = "alloc_layout", since = "1.28.0")] + #[inline] + pub fn new<T>() -> Self { + let (size, align) = size_align::<T>(); + // Note that the align is guaranteed by rustc to be a power of two and + // the size+align combo is guaranteed to fit in our address space. As a + // result use the unchecked constructor here to avoid inserting code + // that panics if it isn't optimized well enough. + debug_assert!(Layout::from_size_align(size, align).is_ok()); + unsafe { + Layout::from_size_align_unchecked(size, align) + } + } + + /// Produces layout describing a record that could be used to + /// allocate backing structure for `T` (which could be a trait + /// or other unsized type like a slice). + #[stable(feature = "alloc_layout", since = "1.28.0")] + #[inline] + pub fn for_value<T: ?Sized>(t: &T) -> Self { + let (size, align) = (mem::size_of_val(t), mem::align_of_val(t)); + // See rationale in `new` for why this us using an unsafe variant below + debug_assert!(Layout::from_size_align(size, align).is_ok()); + unsafe { + Layout::from_size_align_unchecked(size, align) + } + } + + /// Creates a layout describing the record that can hold a value + /// of the same layout as `self`, but that also is aligned to + /// alignment `align` (measured in bytes). + /// + /// If `self` already meets the prescribed alignment, then returns + /// `self`. + /// + /// Note that this method does not add any padding to the overall + /// size, regardless of whether the returned layout has a different + /// alignment. In other words, if `K` has size 16, `K.align_to(32)` + /// will *still* have size 16. + /// + /// # Panics + /// + /// Panics if the combination of `self.size()` and the given `align` + /// violates the conditions listed in + /// [`Layout::from_size_align`](#method.from_size_align). + #[unstable(feature = "allocator_api", issue = "32838")] + #[inline] + pub fn align_to(&self, align: usize) -> Self { + Layout::from_size_align(self.size(), cmp::max(self.align(), align)).unwrap() + } + + /// Returns the amount of padding we must insert after `self` + /// to ensure that the following address will satisfy `align` + /// (measured in bytes). + /// + /// E.g. if `self.size()` is 9, then `self.padding_needed_for(4)` + /// returns 3, because that is the minimum number of bytes of + /// padding required to get a 4-aligned address (assuming that the + /// corresponding memory block starts at a 4-aligned address). + /// + /// The return value of this function has no meaning if `align` is + /// not a power-of-two. + /// + /// Note that the utility of the returned value requires `align` + /// to be less than or equal to the alignment of the starting + /// address for the whole allocated block of memory. One way to + /// satisfy this constraint is to ensure `align <= self.align()`. + #[unstable(feature = "allocator_api", issue = "32838")] + #[inline] + pub fn padding_needed_for(&self, align: usize) -> usize { + let len = self.size(); + + // Rounded up value is: + // len_rounded_up = (len + align - 1) & !(align - 1); + // and then we return the padding difference: `len_rounded_up - len`. + // + // We use modular arithmetic throughout: + // + // 1. align is guaranteed to be > 0, so align - 1 is always + // valid. + // + // 2. `len + align - 1` can overflow by at most `align - 1`, + // so the &-mask wth `!(align - 1)` will ensure that in the + // case of overflow, `len_rounded_up` will itself be 0. + // Thus the returned padding, when added to `len`, yields 0, + // which trivially satisfies the alignment `align`. + // + // (Of course, attempts to allocate blocks of memory whose + // size and padding overflow in the above manner should cause + // the allocator to yield an error anyway.) + + let len_rounded_up = len.wrapping_add(align).wrapping_sub(1) + & !align.wrapping_sub(1); + len_rounded_up.wrapping_sub(len) + } + + /// Creates a layout describing the record for `n` instances of + /// `self`, with a suitable amount of padding between each to + /// ensure that each instance is given its requested size and + /// alignment. On success, returns `(k, offs)` where `k` is the + /// layout of the array and `offs` is the distance between the start + /// of each element in the array. + /// + /// On arithmetic overflow, returns `LayoutErr`. + #[unstable(feature = "allocator_api", issue = "32838")] + #[inline] + pub fn repeat(&self, n: usize) -> Result<(Self, usize), LayoutErr> { + let padded_size = self.size().checked_add(self.padding_needed_for(self.align())) + .ok_or(LayoutErr { private: () })?; + let alloc_size = padded_size.checked_mul(n) + .ok_or(LayoutErr { private: () })?; + + unsafe { + // self.align is already known to be valid and alloc_size has been + // padded already. + Ok((Layout::from_size_align_unchecked(alloc_size, self.align()), padded_size)) + } + } + + /// Creates a layout describing the record for `self` followed by + /// `next`, including any necessary padding to ensure that `next` + /// will be properly aligned. Note that the result layout will + /// satisfy the alignment properties of both `self` and `next`. + /// + /// Returns `Some((k, offset))`, where `k` is layout of the concatenated + /// record and `offset` is the relative location, in bytes, of the + /// start of the `next` embedded within the concatenated record + /// (assuming that the record itself starts at offset 0). + /// + /// On arithmetic overflow, returns `LayoutErr`. + #[unstable(feature = "allocator_api", issue = "32838")] + #[inline] + pub fn extend(&self, next: Self) -> Result<(Self, usize), LayoutErr> { + let new_align = cmp::max(self.align(), next.align()); + let pad = self.padding_needed_for(next.align()); + + let offset = self.size().checked_add(pad) + .ok_or(LayoutErr { private: () })?; + let new_size = offset.checked_add(next.size()) + .ok_or(LayoutErr { private: () })?; + + let layout = Layout::from_size_align(new_size, new_align)?; + Ok((layout, offset)) + } + + /// Creates a layout describing the record for `n` instances of + /// `self`, with no padding between each instance. + /// + /// Note that, unlike `repeat`, `repeat_packed` does not guarantee + /// that the repeated instances of `self` will be properly + /// aligned, even if a given instance of `self` is properly + /// aligned. In other words, if the layout returned by + /// `repeat_packed` is used to allocate an array, it is not + /// guaranteed that all elements in the array will be properly + /// aligned. + /// + /// On arithmetic overflow, returns `LayoutErr`. + #[unstable(feature = "allocator_api", issue = "32838")] + #[inline] + pub fn repeat_packed(&self, n: usize) -> Result<Self, LayoutErr> { + let size = self.size().checked_mul(n).ok_or(LayoutErr { private: () })?; + Layout::from_size_align(size, self.align()) + } + + /// Creates a layout describing the record for `self` followed by + /// `next` with no additional padding between the two. Since no + /// padding is inserted, the alignment of `next` is irrelevant, + /// and is not incorporated *at all* into the resulting layout. + /// + /// Returns `(k, offset)`, where `k` is layout of the concatenated + /// record and `offset` is the relative location, in bytes, of the + /// start of the `next` embedded within the concatenated record + /// (assuming that the record itself starts at offset 0). + /// + /// (The `offset` is always the same as `self.size()`; we use this + /// signature out of convenience in matching the signature of + /// `extend`.) + /// + /// On arithmetic overflow, returns `LayoutErr`. + #[unstable(feature = "allocator_api", issue = "32838")] + #[inline] + pub fn extend_packed(&self, next: Self) -> Result<(Self, usize), LayoutErr> { + let new_size = self.size().checked_add(next.size()) + .ok_or(LayoutErr { private: () })?; + let layout = Layout::from_size_align(new_size, self.align())?; + Ok((layout, self.size())) + } + + /// Creates a layout describing the record for a `[T; n]`. + /// + /// On arithmetic overflow, returns `LayoutErr`. + #[unstable(feature = "allocator_api", issue = "32838")] + #[inline] + pub fn array<T>(n: usize) -> Result<Self, LayoutErr> { + Layout::new::<T>() + .repeat(n) + .map(|(k, offs)| { + debug_assert!(offs == mem::size_of::<T>()); + k + }) + } +} + +/// The parameters given to `Layout::from_size_align` +/// or some other `Layout` constructor +/// do not satisfy its documented constraints. +#[stable(feature = "alloc_layout", since = "1.28.0")] +#[derive(Clone, PartialEq, Eq, Debug)] +pub struct LayoutErr { + private: () +} + +// (we need this for downstream impl of trait Error) +#[stable(feature = "alloc_layout", since = "1.28.0")] +impl fmt::Display for LayoutErr { + fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result { + f.write_str("invalid parameters to Layout::from_size_align") + } +} + +/// The `AllocErr` error indicates an allocation failure +/// that may be due to resource exhaustion or to +/// something wrong when combining the given input arguments with this +/// allocator. +#[unstable(feature = "allocator_api", issue = "32838")] +#[derive(Clone, PartialEq, Eq, Debug)] +pub struct AllocErr; + +// (we need this for downstream impl of trait Error) +#[unstable(feature = "allocator_api", issue = "32838")] +impl fmt::Display for AllocErr { + fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result { + f.write_str("memory allocation failed") + } +} + +/// The `CannotReallocInPlace` error is used when `grow_in_place` or +/// `shrink_in_place` were unable to reuse the given memory block for +/// a requested layout. +#[unstable(feature = "allocator_api", issue = "32838")] +#[derive(Clone, PartialEq, Eq, Debug)] +pub struct CannotReallocInPlace; + +#[unstable(feature = "allocator_api", issue = "32838")] +impl CannotReallocInPlace { + pub fn description(&self) -> &str { + "cannot reallocate allocator's memory in place" + } +} + +// (we need this for downstream impl of trait Error) +#[unstable(feature = "allocator_api", issue = "32838")] +impl fmt::Display for CannotReallocInPlace { + fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result { + write!(f, "{}", self.description()) + } +} + +/// A memory allocator that can be registered as the standard library’s default +/// though the `#[global_allocator]` attributes. +/// +/// Some of the methods require that a memory block be *currently +/// allocated* via an allocator. This means that: +/// +/// * the starting address for that memory block was previously +/// returned by a previous call to an allocation method +/// such as `alloc`, and +/// +/// * the memory block has not been subsequently deallocated, where +/// blocks are deallocated either by being passed to a deallocation +/// method such as `dealloc` or by being +/// passed to a reallocation method that returns a non-null pointer. +/// +/// +/// # Example +/// +/// ```no_run +/// use std::alloc::{GlobalAlloc, Layout, alloc}; +/// use std::ptr::null_mut; +/// +/// struct MyAllocator; +/// +/// unsafe impl GlobalAlloc for MyAllocator { +/// unsafe fn alloc(&self, _layout: Layout) -> *mut u8 { null_mut() } +/// unsafe fn dealloc(&self, _ptr: *mut u8, _layout: Layout) {} +/// } +/// +/// #[global_allocator] +/// static A: MyAllocator = MyAllocator; +/// +/// fn main() { +/// unsafe { +/// assert!(alloc(Layout::new::<u32>()).is_null()) +/// } +/// } +/// ``` +/// +/// # Unsafety +/// +/// The `GlobalAlloc` trait is an `unsafe` trait for a number of reasons, and +/// implementors must ensure that they adhere to these contracts: +/// +/// * It's undefined behavior if global allocators unwind. This restriction may +/// be lifted in the future, but currently a panic from any of these +/// functions may lead to memory unsafety. +/// +/// * `Layout` queries and calculations in general must be correct. Callers of +/// this trait are allowed to rely on the contracts defined on each method, +/// and implementors must ensure such contracts remain true. +#[stable(feature = "global_alloc", since = "1.28.0")] +pub unsafe trait GlobalAlloc { + /// Allocate memory as described by the given `layout`. + /// + /// Returns a pointer to newly-allocated memory, + /// or null to indicate allocation failure. + /// + /// # Safety + /// + /// This function is unsafe because undefined behavior can result + /// if the caller does not ensure that `layout` has non-zero size. + /// + /// (Extension subtraits might provide more specific bounds on + /// behavior, e.g. guarantee a sentinel address or a null pointer + /// in response to a zero-size allocation request.) + /// + /// The allocated block of memory may or may not be initialized. + /// + /// # Errors + /// + /// Returning a null pointer indicates that either memory is exhausted + /// or `layout` does not meet allocator's size or alignment constraints. + /// + /// Implementations are encouraged to return null on memory + /// exhaustion rather than aborting, but this is not + /// a strict requirement. (Specifically: it is *legal* to + /// implement this trait atop an underlying native allocation + /// library that aborts on memory exhaustion.) + /// + /// Clients wishing to abort computation in response to an + /// allocation error are encouraged to call the [`handle_alloc_error`] function, + /// rather than directly invoking `panic!` or similar. + /// + /// [`handle_alloc_error`]: ../../alloc/alloc/fn.handle_alloc_error.html + #[stable(feature = "global_alloc", since = "1.28.0")] + unsafe fn alloc(&self, layout: Layout) -> *mut u8; + + /// Deallocate the block of memory at the given `ptr` pointer with the given `layout`. + /// + /// # Safety + /// + /// This function is unsafe because undefined behavior can result + /// if the caller does not ensure all of the following: + /// + /// * `ptr` must denote a block of memory currently allocated via + /// this allocator, + /// + /// * `layout` must be the same layout that was used + /// to allocated that block of memory, + #[stable(feature = "global_alloc", since = "1.28.0")] + unsafe fn dealloc(&self, ptr: *mut u8, layout: Layout); + + /// Behaves like `alloc`, but also ensures that the contents + /// are set to zero before being returned. + /// + /// # Safety + /// + /// This function is unsafe for the same reasons that `alloc` is. + /// However the allocated block of memory is guaranteed to be initialized. + /// + /// # Errors + /// + /// Returning a null pointer indicates that either memory is exhausted + /// or `layout` does not meet allocator's size or alignment constraints, + /// just as in `alloc`. + /// + /// Clients wishing to abort computation in response to an + /// allocation error are encouraged to call the [`handle_alloc_error`] function, + /// rather than directly invoking `panic!` or similar. + /// + /// [`handle_alloc_error`]: ../../alloc/alloc/fn.handle_alloc_error.html + #[stable(feature = "global_alloc", since = "1.28.0")] + unsafe fn alloc_zeroed(&self, layout: Layout) -> *mut u8 { + let size = layout.size(); + let ptr = self.alloc(layout); + if !ptr.is_null() { + ptr::write_bytes(ptr, 0, size); + } + ptr + } + + /// Shink or grow a block of memory to the given `new_size`. + /// The block is described by the given `ptr` pointer and `layout`. + /// + /// If this returns a non-null pointer, then ownership of the memory block + /// referenced by `ptr` has been transferred to this alloctor. + /// The memory may or may not have been deallocated, + /// and should be considered unusable (unless of course it was + /// transferred back to the caller again via the return value of + /// this method). + /// + /// If this method returns null, then ownership of the memory + /// block has not been transferred to this allocator, and the + /// contents of the memory block are unaltered. + /// + /// # Safety + /// + /// This function is unsafe because undefined behavior can result + /// if the caller does not ensure all of the following: + /// + /// * `ptr` must be currently allocated via this allocator, + /// + /// * `layout` must be the same layout that was used + /// to allocated that block of memory, + /// + /// * `new_size` must be greater than zero. + /// + /// * `new_size`, when rounded up to the nearest multiple of `layout.align()`, + /// must not overflow (i.e. the rounded value must be less than `usize::MAX`). + /// + /// (Extension subtraits might provide more specific bounds on + /// behavior, e.g. guarantee a sentinel address or a null pointer + /// in response to a zero-size allocation request.) + /// + /// # Errors + /// + /// Returns null if the new layout does not meet the size + /// and alignment constraints of the allocator, or if reallocation + /// otherwise fails. + /// + /// Implementations are encouraged to return null on memory + /// exhaustion rather than panicking or aborting, but this is not + /// a strict requirement. (Specifically: it is *legal* to + /// implement this trait atop an underlying native allocation + /// library that aborts on memory exhaustion.) + /// + /// Clients wishing to abort computation in response to a + /// reallocation error are encouraged to call the [`handle_alloc_error`] function, + /// rather than directly invoking `panic!` or similar. + /// + /// [`handle_alloc_error`]: ../../alloc/alloc/fn.handle_alloc_error.html + #[stable(feature = "global_alloc", since = "1.28.0")] + unsafe fn realloc(&self, ptr: *mut u8, layout: Layout, new_size: usize) -> *mut u8 { + let new_layout = Layout::from_size_align_unchecked(new_size, layout.align()); + let new_ptr = self.alloc(new_layout); + if !new_ptr.is_null() { + ptr::copy_nonoverlapping( + ptr, + new_ptr, + cmp::min(layout.size(), new_size), + ); + self.dealloc(ptr, layout); + } + new_ptr + } +} + +/// An implementation of `Alloc` can allocate, reallocate, and +/// deallocate arbitrary blocks of data described via `Layout`. +/// +/// Some of the methods require that a memory block be *currently +/// allocated* via an allocator. This means that: +/// +/// * the starting address for that memory block was previously +/// returned by a previous call to an allocation method (`alloc`, +/// `alloc_zeroed`, `alloc_excess`, `alloc_one`, `alloc_array`) or +/// reallocation method (`realloc`, `realloc_excess`, or +/// `realloc_array`), and +/// +/// * the memory block has not been subsequently deallocated, where +/// blocks are deallocated either by being passed to a deallocation +/// method (`dealloc`, `dealloc_one`, `dealloc_array`) or by being +/// passed to a reallocation method (see above) that returns `Ok`. +/// +/// A note regarding zero-sized types and zero-sized layouts: many +/// methods in the `Alloc` trait state that allocation requests +/// must be non-zero size, or else undefined behavior can result. +/// +/// * However, some higher-level allocation methods (`alloc_one`, +/// `alloc_array`) are well-defined on zero-sized types and can +/// optionally support them: it is left up to the implementor +/// whether to return `Err`, or to return `Ok` with some pointer. +/// +/// * If an `Alloc` implementation chooses to return `Ok` in this +/// case (i.e. the pointer denotes a zero-sized inaccessible block) +/// then that returned pointer must be considered "currently +/// allocated". On such an allocator, *all* methods that take +/// currently-allocated pointers as inputs must accept these +/// zero-sized pointers, *without* causing undefined behavior. +/// +/// * In other words, if a zero-sized pointer can flow out of an +/// allocator, then that allocator must likewise accept that pointer +/// flowing back into its deallocation and reallocation methods. +/// +/// Some of the methods require that a layout *fit* a memory block. +/// What it means for a layout to "fit" a memory block means (or +/// equivalently, for a memory block to "fit" a layout) is that the +/// following two conditions must hold: +/// +/// 1. The block's starting address must be aligned to `layout.align()`. +/// +/// 2. The block's size must fall in the range `[use_min, use_max]`, where: +/// +/// * `use_min` is `self.usable_size(layout).0`, and +/// +/// * `use_max` is the capacity that was (or would have been) +/// returned when (if) the block was allocated via a call to +/// `alloc_excess` or `realloc_excess`. +/// +/// Note that: +/// +/// * the size of the layout most recently used to allocate the block +/// is guaranteed to be in the range `[use_min, use_max]`, and +/// +/// * a lower-bound on `use_max` can be safely approximated by a call to +/// `usable_size`. +/// +/// * if a layout `k` fits a memory block (denoted by `ptr`) +/// currently allocated via an allocator `a`, then it is legal to +/// use that layout to deallocate it, i.e. `a.dealloc(ptr, k);`. +/// +/// # Unsafety +/// +/// The `Alloc` trait is an `unsafe` trait for a number of reasons, and +/// implementors must ensure that they adhere to these contracts: +/// +/// * Pointers returned from allocation functions must point to valid memory and +/// retain their validity until at least the instance of `Alloc` is dropped +/// itself. +/// +/// * `Layout` queries and calculations in general must be correct. Callers of +/// this trait are allowed to rely on the contracts defined on each method, +/// and implementors must ensure such contracts remain true. +/// +/// Note that this list may get tweaked over time as clarifications are made in +/// the future. +#[unstable(feature = "allocator_api", issue = "32838")] +pub unsafe trait Alloc { + + // (Note: some existing allocators have unspecified but well-defined + // behavior in response to a zero size allocation request ; + // e.g. in C, `malloc` of 0 will either return a null pointer or a + // unique pointer, but will not have arbitrary undefined + // behavior. + // However in jemalloc for example, + // `mallocx(0)` is documented as undefined behavior.) + + /// Returns a pointer meeting the size and alignment guarantees of + /// `layout`. + /// + /// If this method returns an `Ok(addr)`, then the `addr` returned + /// will be non-null address pointing to a block of storage + /// suitable for holding an instance of `layout`. + /// + /// The returned block of storage may or may not have its contents + /// initialized. (Extension subtraits might restrict this + /// behavior, e.g. to ensure initialization to particular sets of + /// bit patterns.) + /// + /// # Safety + /// + /// This function is unsafe because undefined behavior can result + /// if the caller does not ensure that `layout` has non-zero size. + /// + /// (Extension subtraits might provide more specific bounds on + /// behavior, e.g. guarantee a sentinel address or a null pointer + /// in response to a zero-size allocation request.) + /// + /// # Errors + /// + /// Returning `Err` indicates that either memory is exhausted or + /// `layout` does not meet allocator's size or alignment + /// constraints. + /// + /// Implementations are encouraged to return `Err` on memory + /// exhaustion rather than panicking or aborting, but this is not + /// a strict requirement. (Specifically: it is *legal* to + /// implement this trait atop an underlying native allocation + /// library that aborts on memory exhaustion.) + /// + /// Clients wishing to abort computation in response to an + /// allocation error are encouraged to call the [`handle_alloc_error`] function, + /// rather than directly invoking `panic!` or similar. + /// + /// [`handle_alloc_error`]: ../../alloc/alloc/fn.handle_alloc_error.html + unsafe fn alloc(&mut self, layout: Layout) -> Result<NonNull<u8>, AllocErr>; + + /// Deallocate the memory referenced by `ptr`. + /// + /// # Safety + /// + /// This function is unsafe because undefined behavior can result + /// if the caller does not ensure all of the following: + /// + /// * `ptr` must denote a block of memory currently allocated via + /// this allocator, + /// + /// * `layout` must *fit* that block of memory, + /// + /// * In addition to fitting the block of memory `layout`, the + /// alignment of the `layout` must match the alignment used + /// to allocate that block of memory. + unsafe fn dealloc(&mut self, ptr: NonNull<u8>, layout: Layout); + + // == ALLOCATOR-SPECIFIC QUANTITIES AND LIMITS == + // usable_size + + /// Returns bounds on the guaranteed usable size of a successful + /// allocation created with the specified `layout`. + /// + /// In particular, if one has a memory block allocated via a given + /// allocator `a` and layout `k` where `a.usable_size(k)` returns + /// `(l, u)`, then one can pass that block to `a.dealloc()` with a + /// layout in the size range [l, u]. + /// + /// (All implementors of `usable_size` must ensure that + /// `l <= k.size() <= u`) + /// + /// Both the lower- and upper-bounds (`l` and `u` respectively) + /// are provided, because an allocator based on size classes could + /// misbehave if one attempts to deallocate a block without + /// providing a correct value for its size (i.e., one within the + /// range `[l, u]`). + /// + /// Clients who wish to make use of excess capacity are encouraged + /// to use the `alloc_excess` and `realloc_excess` instead, as + /// this method is constrained to report conservative values that + /// serve as valid bounds for *all possible* allocation method + /// calls. + /// + /// However, for clients that do not wish to track the capacity + /// returned by `alloc_excess` locally, this method is likely to + /// produce useful results. + #[inline] + fn usable_size(&self, layout: &Layout) -> (usize, usize) { + (layout.size(), layout.size()) + } + + // == METHODS FOR MEMORY REUSE == + // realloc. alloc_excess, realloc_excess + + /// Returns a pointer suitable for holding data described by + /// a new layout with `layout`’s alginment and a size given + /// by `new_size`. To + /// accomplish this, this may extend or shrink the allocation + /// referenced by `ptr` to fit the new layout. + /// + /// If this returns `Ok`, then ownership of the memory block + /// referenced by `ptr` has been transferred to this + /// allocator. The memory may or may not have been freed, and + /// should be considered unusable (unless of course it was + /// transferred back to the caller again via the return value of + /// this method). + /// + /// If this method returns `Err`, then ownership of the memory + /// block has not been transferred to this allocator, and the + /// contents of the memory block are unaltered. + /// + /// # Safety + /// + /// This function is unsafe because undefined behavior can result + /// if the caller does not ensure all of the following: + /// + /// * `ptr` must be currently allocated via this allocator, + /// + /// * `layout` must *fit* the `ptr` (see above). (The `new_size` + /// argument need not fit it.) + /// + /// * `new_size` must be greater than zero. + /// + /// * `new_size`, when rounded up to the nearest multiple of `layout.align()`, + /// must not overflow (i.e. the rounded value must be less than `usize::MAX`). + /// + /// (Extension subtraits might provide more specific bounds on + /// behavior, e.g. guarantee a sentinel address or a null pointer + /// in response to a zero-size allocation request.) + /// + /// # Errors + /// + /// Returns `Err` only if the new layout + /// does not meet the allocator's size + /// and alignment constraints of the allocator, or if reallocation + /// otherwise fails. + /// + /// Implementations are encouraged to return `Err` on memory + /// exhaustion rather than panicking or aborting, but this is not + /// a strict requirement. (Specifically: it is *legal* to + /// implement this trait atop an underlying native allocation + /// library that aborts on memory exhaustion.) + /// + /// Clients wishing to abort computation in response to a + /// reallocation error are encouraged to call the [`handle_alloc_error`] function, + /// rather than directly invoking `panic!` or similar. + /// + /// [`handle_alloc_error`]: ../../alloc/alloc/fn.handle_alloc_error.html + unsafe fn realloc(&mut self, + ptr: NonNull<u8>, + layout: Layout, + new_size: usize) -> Result<NonNull<u8>, AllocErr> { + let old_size = layout.size(); + + if new_size >= old_size { + if let Ok(()) = self.grow_in_place(ptr, layout.clone(), new_size) { + return Ok(ptr); + } + } else if new_size < old_size { + if let Ok(()) = self.shrink_in_place(ptr, layout.clone(), new_size) { + return Ok(ptr); + } + } + + // otherwise, fall back on alloc + copy + dealloc. + let new_layout = Layout::from_size_align_unchecked(new_size, layout.align()); + let result = self.alloc(new_layout); + if let Ok(new_ptr) = result { + ptr::copy_nonoverlapping(ptr.as_ptr(), + new_ptr.as_ptr(), + cmp::min(old_size, new_size)); + self.dealloc(ptr, layout); + } + result + } + + /// Behaves like `alloc`, but also ensures that the contents + /// are set to zero before being returned. + /// + /// # Safety + /// + /// This function is unsafe for the same reasons that `alloc` is. + /// + /// # Errors + /// + /// Returning `Err` indicates that either memory is exhausted or + /// `layout` does not meet allocator's size or alignment + /// constraints, just as in `alloc`. + /// + /// Clients wishing to abort computation in response to an + /// allocation error are encouraged to call the [`handle_alloc_error`] function, + /// rather than directly invoking `panic!` or similar. + /// + /// [`handle_alloc_error`]: ../../alloc/alloc/fn.handle_alloc_error.html + unsafe fn alloc_zeroed(&mut self, layout: Layout) -> Result<NonNull<u8>, AllocErr> { + let size = layout.size(); + let p = self.alloc(layout); + if let Ok(p) = p { + ptr::write_bytes(p.as_ptr(), 0, size); + } + p + } + + /// Behaves like `alloc`, but also returns the whole size of + /// the returned block. For some `layout` inputs, like arrays, this + /// may include extra storage usable for additional data. + /// + /// # Safety + /// + /// This function is unsafe for the same reasons that `alloc` is. + /// + /// # Errors + /// + /// Returning `Err` indicates that either memory is exhausted or + /// `layout` does not meet allocator's size or alignment + /// constraints, just as in `alloc`. + /// + /// Clients wishing to abort computation in response to an + /// allocation error are encouraged to call the [`handle_alloc_error`] function, + /// rather than directly invoking `panic!` or similar. + /// + /// [`handle_alloc_error`]: ../../alloc/alloc/fn.handle_alloc_error.html + unsafe fn alloc_excess(&mut self, layout: Layout) -> Result<Excess, AllocErr> { + let usable_size = self.usable_size(&layout); + self.alloc(layout).map(|p| Excess(p, usable_size.1)) + } + + /// Behaves like `realloc`, but also returns the whole size of + /// the returned block. For some `layout` inputs, like arrays, this + /// may include extra storage usable for additional data. + /// + /// # Safety + /// + /// This function is unsafe for the same reasons that `realloc` is. + /// + /// # Errors + /// + /// Returning `Err` indicates that either memory is exhausted or + /// `layout` does not meet allocator's size or alignment + /// constraints, just as in `realloc`. + /// + /// Clients wishing to abort computation in response to a + /// reallocation error are encouraged to call the [`handle_alloc_error`] function, + /// rather than directly invoking `panic!` or similar. + /// + /// [`handle_alloc_error`]: ../../alloc/alloc/fn.handle_alloc_error.html + unsafe fn realloc_excess(&mut self, + ptr: NonNull<u8>, + layout: Layout, + new_size: usize) -> Result<Excess, AllocErr> { + let new_layout = Layout::from_size_align_unchecked(new_size, layout.align()); + let usable_size = self.usable_size(&new_layout); + self.realloc(ptr, layout, new_size) + .map(|p| Excess(p, usable_size.1)) + } + + /// Attempts to extend the allocation referenced by `ptr` to fit `new_size`. + /// + /// If this returns `Ok`, then the allocator has asserted that the + /// memory block referenced by `ptr` now fits `new_size`, and thus can + /// be used to carry data of a layout of that size and same alignment as + /// `layout`. (The allocator is allowed to + /// expend effort to accomplish this, such as extending the memory block to + /// include successor blocks, or virtual memory tricks.) + /// + /// Regardless of what this method returns, ownership of the + /// memory block referenced by `ptr` has not been transferred, and + /// the contents of the memory block are unaltered. + /// + /// # Safety + /// + /// This function is unsafe because undefined behavior can result + /// if the caller does not ensure all of the following: + /// + /// * `ptr` must be currently allocated via this allocator, + /// + /// * `layout` must *fit* the `ptr` (see above); note the + /// `new_size` argument need not fit it, + /// + /// * `new_size` must not be less than `layout.size()`, + /// + /// # Errors + /// + /// Returns `Err(CannotReallocInPlace)` when the allocator is + /// unable to assert that the memory block referenced by `ptr` + /// could fit `layout`. + /// + /// Note that one cannot pass `CannotReallocInPlace` to the `handle_alloc_error` + /// function; clients are expected either to be able to recover from + /// `grow_in_place` failures without aborting, or to fall back on + /// another reallocation method before resorting to an abort. + unsafe fn grow_in_place(&mut self, + ptr: NonNull<u8>, + layout: Layout, + new_size: usize) -> Result<(), CannotReallocInPlace> { + let _ = ptr; // this default implementation doesn't care about the actual address. + debug_assert!(new_size >= layout.size()); + let (_l, u) = self.usable_size(&layout); + // _l <= layout.size() [guaranteed by usable_size()] + // layout.size() <= new_layout.size() [required by this method] + if new_size <= u { + Ok(()) + } else { + Err(CannotReallocInPlace) + } + } + + /// Attempts to shrink the allocation referenced by `ptr` to fit `new_size`. + /// + /// If this returns `Ok`, then the allocator has asserted that the + /// memory block referenced by `ptr` now fits `new_size`, and + /// thus can only be used to carry data of that smaller + /// layout. (The allocator is allowed to take advantage of this, + /// carving off portions of the block for reuse elsewhere.) The + /// truncated contents of the block within the smaller layout are + /// unaltered, and ownership of block has not been transferred. + /// + /// If this returns `Err`, then the memory block is considered to + /// still represent the original (larger) `layout`. None of the + /// block has been carved off for reuse elsewhere, ownership of + /// the memory block has not been transferred, and the contents of + /// the memory block are unaltered. + /// + /// # Safety + /// + /// This function is unsafe because undefined behavior can result + /// if the caller does not ensure all of the following: + /// + /// * `ptr` must be currently allocated via this allocator, + /// + /// * `layout` must *fit* the `ptr` (see above); note the + /// `new_size` argument need not fit it, + /// + /// * `new_size` must not be greater than `layout.size()` + /// (and must be greater than zero), + /// + /// # Errors + /// + /// Returns `Err(CannotReallocInPlace)` when the allocator is + /// unable to assert that the memory block referenced by `ptr` + /// could fit `layout`. + /// + /// Note that one cannot pass `CannotReallocInPlace` to the `handle_alloc_error` + /// function; clients are expected either to be able to recover from + /// `shrink_in_place` failures without aborting, or to fall back + /// on another reallocation method before resorting to an abort. + unsafe fn shrink_in_place(&mut self, + ptr: NonNull<u8>, + layout: Layout, + new_size: usize) -> Result<(), CannotReallocInPlace> { + let _ = ptr; // this default implementation doesn't care about the actual address. + debug_assert!(new_size <= layout.size()); + let (l, _u) = self.usable_size(&layout); + // layout.size() <= _u [guaranteed by usable_size()] + // new_layout.size() <= layout.size() [required by this method] + if l <= new_size { + Ok(()) + } else { + Err(CannotReallocInPlace) + } + } + + + // == COMMON USAGE PATTERNS == + // alloc_one, dealloc_one, alloc_array, realloc_array. dealloc_array + + /// Allocates a block suitable for holding an instance of `T`. + /// + /// Captures a common usage pattern for allocators. + /// + /// The returned block is suitable for passing to the + /// `alloc`/`realloc` methods of this allocator. + /// + /// Note to implementors: If this returns `Ok(ptr)`, then `ptr` + /// must be considered "currently allocated" and must be + /// acceptable input to methods such as `realloc` or `dealloc`, + /// *even if* `T` is a zero-sized type. In other words, if your + /// `Alloc` implementation overrides this method in a manner + /// that can return a zero-sized `ptr`, then all reallocation and + /// deallocation methods need to be similarly overridden to accept + /// such values as input. + /// + /// # Errors + /// + /// Returning `Err` indicates that either memory is exhausted or + /// `T` does not meet allocator's size or alignment constraints. + /// + /// For zero-sized `T`, may return either of `Ok` or `Err`, but + /// will *not* yield undefined behavior. + /// + /// Clients wishing to abort computation in response to an + /// allocation error are encouraged to call the [`handle_alloc_error`] function, + /// rather than directly invoking `panic!` or similar. + /// + /// [`handle_alloc_error`]: ../../alloc/alloc/fn.handle_alloc_error.html + fn alloc_one<T>(&mut self) -> Result<NonNull<T>, AllocErr> + where Self: Sized + { + let k = Layout::new::<T>(); + if k.size() > 0 { + unsafe { self.alloc(k).map(|p| p.cast()) } + } else { + Err(AllocErr) + } + } + + /// Deallocates a block suitable for holding an instance of `T`. + /// + /// The given block must have been produced by this allocator, + /// and must be suitable for storing a `T` (in terms of alignment + /// as well as minimum and maximum size); otherwise yields + /// undefined behavior. + /// + /// Captures a common usage pattern for allocators. + /// + /// # Safety + /// + /// This function is unsafe because undefined behavior can result + /// if the caller does not ensure both: + /// + /// * `ptr` must denote a block of memory currently allocated via this allocator + /// + /// * the layout of `T` must *fit* that block of memory. + unsafe fn dealloc_one<T>(&mut self, ptr: NonNull<T>) + where Self: Sized + { + let k = Layout::new::<T>(); + if k.size() > 0 { + self.dealloc(ptr.cast(), k); + } + } + + /// Allocates a block suitable for holding `n` instances of `T`. + /// + /// Captures a common usage pattern for allocators. + /// + /// The returned block is suitable for passing to the + /// `alloc`/`realloc` methods of this allocator. + /// + /// Note to implementors: If this returns `Ok(ptr)`, then `ptr` + /// must be considered "currently allocated" and must be + /// acceptable input to methods such as `realloc` or `dealloc`, + /// *even if* `T` is a zero-sized type. In other words, if your + /// `Alloc` implementation overrides this method in a manner + /// that can return a zero-sized `ptr`, then all reallocation and + /// deallocation methods need to be similarly overridden to accept + /// such values as input. + /// + /// # Errors + /// + /// Returning `Err` indicates that either memory is exhausted or + /// `[T; n]` does not meet allocator's size or alignment + /// constraints. + /// + /// For zero-sized `T` or `n == 0`, may return either of `Ok` or + /// `Err`, but will *not* yield undefined behavior. + /// + /// Always returns `Err` on arithmetic overflow. + /// + /// Clients wishing to abort computation in response to an + /// allocation error are encouraged to call the [`handle_alloc_error`] function, + /// rather than directly invoking `panic!` or similar. + /// + /// [`handle_alloc_error`]: ../../alloc/alloc/fn.handle_alloc_error.html + fn alloc_array<T>(&mut self, n: usize) -> Result<NonNull<T>, AllocErr> + where Self: Sized + { + match Layout::array::<T>(n) { + Ok(ref layout) if layout.size() > 0 => { + unsafe { + self.alloc(layout.clone()).map(|p| p.cast()) + } + } + _ => Err(AllocErr), + } + } + + /// Reallocates a block previously suitable for holding `n_old` + /// instances of `T`, returning a block suitable for holding + /// `n_new` instances of `T`. + /// + /// Captures a common usage pattern for allocators. + /// + /// The returned block is suitable for passing to the + /// `alloc`/`realloc` methods of this allocator. + /// + /// # Safety + /// + /// This function is unsafe because undefined behavior can result + /// if the caller does not ensure all of the following: + /// + /// * `ptr` must be currently allocated via this allocator, + /// + /// * the layout of `[T; n_old]` must *fit* that block of memory. + /// + /// # Errors + /// + /// Returning `Err` indicates that either memory is exhausted or + /// `[T; n_new]` does not meet allocator's size or alignment + /// constraints. + /// + /// For zero-sized `T` or `n_new == 0`, may return either of `Ok` or + /// `Err`, but will *not* yield undefined behavior. + /// + /// Always returns `Err` on arithmetic overflow. + /// + /// Clients wishing to abort computation in response to a + /// reallocation error are encouraged to call the [`handle_alloc_error`] function, + /// rather than directly invoking `panic!` or similar. + /// + /// [`handle_alloc_error`]: ../../alloc/alloc/fn.handle_alloc_error.html + unsafe fn realloc_array<T>(&mut self, + ptr: NonNull<T>, + n_old: usize, + n_new: usize) -> Result<NonNull<T>, AllocErr> + where Self: Sized + { + match (Layout::array::<T>(n_old), Layout::array::<T>(n_new)) { + (Ok(ref k_old), Ok(ref k_new)) if k_old.size() > 0 && k_new.size() > 0 => { + debug_assert!(k_old.align() == k_new.align()); + self.realloc(ptr.cast(), k_old.clone(), k_new.size()).map(NonNull::cast) + } + _ => { + Err(AllocErr) + } + } + } + + /// Deallocates a block suitable for holding `n` instances of `T`. + /// + /// Captures a common usage pattern for allocators. + /// + /// # Safety + /// + /// This function is unsafe because undefined behavior can result + /// if the caller does not ensure both: + /// + /// * `ptr` must denote a block of memory currently allocated via this allocator + /// + /// * the layout of `[T; n]` must *fit* that block of memory. + /// + /// # Errors + /// + /// Returning `Err` indicates that either `[T; n]` or the given + /// memory block does not meet allocator's size or alignment + /// constraints. + /// + /// Always returns `Err` on arithmetic overflow. + unsafe fn dealloc_array<T>(&mut self, ptr: NonNull<T>, n: usize) -> Result<(), AllocErr> + where Self: Sized + { + match Layout::array::<T>(n) { + Ok(ref k) if k.size() > 0 => { + Ok(self.dealloc(ptr.cast(), k.clone())) + } + _ => { + Err(AllocErr) + } + } + } +} diff --git a/src/libcore/any.rs b/src/libcore/any.rs index 338e5c7fd95..6b26093439e 100644 --- a/src/libcore/any.rs +++ b/src/libcore/any.rs @@ -120,7 +120,7 @@ impl<T: 'static + ?Sized > Any for T { /////////////////////////////////////////////////////////////////////////////// #[stable(feature = "rust1", since = "1.0.0")] -impl fmt::Debug for Any { +impl fmt::Debug for dyn Any { fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result { f.pad("Any") } @@ -130,13 +130,20 @@ impl fmt::Debug for Any { // hence used with `unwrap`. May eventually no longer be needed if // dispatch works with upcasting. #[stable(feature = "rust1", since = "1.0.0")] -impl fmt::Debug for Any + Send { +impl fmt::Debug for dyn Any + Send { fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result { f.pad("Any") } } -impl Any { +#[stable(feature = "any_send_sync_methods", since = "1.28.0")] +impl fmt::Debug for dyn Any + Send + Sync { + fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result { + f.pad("Any") + } +} + +impl dyn Any { /// Returns `true` if the boxed type is the same as `T`. /// /// # Examples @@ -196,7 +203,7 @@ impl Any { pub fn downcast_ref<T: Any>(&self) -> Option<&T> { if self.is::<T>() { unsafe { - Some(&*(self as *const Any as *const T)) + Some(&*(self as *const dyn Any as *const T)) } } else { None @@ -233,7 +240,7 @@ impl Any { pub fn downcast_mut<T: Any>(&mut self) -> Option<&mut T> { if self.is::<T>() { unsafe { - Some(&mut *(self as *mut Any as *mut T)) + Some(&mut *(self as *mut dyn Any as *mut T)) } } else { None @@ -241,7 +248,7 @@ impl Any { } } -impl Any+Send { +impl dyn Any+Send { /// Forwards to the method defined on the type `Any`. /// /// # Examples @@ -301,7 +308,7 @@ impl Any+Send { /// ``` /// use std::any::Any; /// - /// fn modify_if_u32(s: &mut (Any+ Send)) { + /// fn modify_if_u32(s: &mut (Any + Send)) { /// if let Some(num) = s.downcast_mut::<u32>() { /// *num = 42; /// } @@ -325,6 +332,89 @@ impl Any+Send { } } +impl dyn Any+Send+Sync { + /// Forwards to the method defined on the type `Any`. + /// + /// # Examples + /// + /// ``` + /// use std::any::Any; + /// + /// fn is_string(s: &(Any + Send + Sync)) { + /// if s.is::<String>() { + /// println!("It's a string!"); + /// } else { + /// println!("Not a string..."); + /// } + /// } + /// + /// fn main() { + /// is_string(&0); + /// is_string(&"cookie monster".to_string()); + /// } + /// ``` + #[stable(feature = "any_send_sync_methods", since = "1.28.0")] + #[inline] + pub fn is<T: Any>(&self) -> bool { + Any::is::<T>(self) + } + + /// Forwards to the method defined on the type `Any`. + /// + /// # Examples + /// + /// ``` + /// use std::any::Any; + /// + /// fn print_if_string(s: &(Any + Send + Sync)) { + /// if let Some(string) = s.downcast_ref::<String>() { + /// println!("It's a string({}): '{}'", string.len(), string); + /// } else { + /// println!("Not a string..."); + /// } + /// } + /// + /// fn main() { + /// print_if_string(&0); + /// print_if_string(&"cookie monster".to_string()); + /// } + /// ``` + #[stable(feature = "any_send_sync_methods", since = "1.28.0")] + #[inline] + pub fn downcast_ref<T: Any>(&self) -> Option<&T> { + Any::downcast_ref::<T>(self) + } + + /// Forwards to the method defined on the type `Any`. + /// + /// # Examples + /// + /// ``` + /// use std::any::Any; + /// + /// fn modify_if_u32(s: &mut (Any + Send + Sync)) { + /// if let Some(num) = s.downcast_mut::<u32>() { + /// *num = 42; + /// } + /// } + /// + /// fn main() { + /// let mut x = 10u32; + /// let mut s = "starlord".to_string(); + /// + /// modify_if_u32(&mut x); + /// modify_if_u32(&mut s); + /// + /// assert_eq!(x, 42); + /// assert_eq!(&s, "starlord"); + /// } + /// ``` + #[stable(feature = "any_send_sync_methods", since = "1.28.0")] + #[inline] + pub fn downcast_mut<T: Any>(&mut self) -> Option<&mut T> { + Any::downcast_mut::<T>(self) + } +} /////////////////////////////////////////////////////////////////////////////// // TypeID and its methods @@ -341,7 +431,7 @@ impl Any+Send { /// /// While `TypeId` implements `Hash`, `PartialOrd`, and `Ord`, it is worth /// noting that the hashes and ordering will vary between Rust releases. Beware -/// of relying on them outside of your code! +/// of relying on them inside of your code! #[derive(Clone, Copy, PartialEq, Eq, PartialOrd, Ord, Debug, Hash)] #[stable(feature = "rust1", since = "1.0.0")] pub struct TypeId { @@ -367,7 +457,8 @@ impl TypeId { /// } /// ``` #[stable(feature = "rust1", since = "1.0.0")] - pub fn of<T: ?Sized + 'static>() -> TypeId { + #[rustc_const_unstable(feature="const_type_id")] + pub const fn of<T: ?Sized + 'static>() -> TypeId { TypeId { t: unsafe { intrinsics::type_id::<T>() }, } diff --git a/src/libcore/ascii.rs b/src/libcore/ascii.rs new file mode 100644 index 00000000000..6ee91e0b22f --- /dev/null +++ b/src/libcore/ascii.rs @@ -0,0 +1,147 @@ +// Copyright 2013-2014 The Rust Project Developers. See the COPYRIGHT +// file at the top-level directory of this distribution and at +// http://rust-lang.org/COPYRIGHT. +// +// Licensed under the Apache License, Version 2.0 <LICENSE-APACHE or +// http://www.apache.org/licenses/LICENSE-2.0> or the MIT license +// <LICENSE-MIT or http://opensource.org/licenses/MIT>, at your +// option. This file may not be copied, modified, or distributed +// except according to those terms. + +//! Operations on ASCII strings and characters. +//! +//! Most string operations in Rust act on UTF-8 strings. However, at times it +//! makes more sense to only consider the ASCII character set for a specific +//! operation. +//! +//! The [`escape_default`] function provides an iterator over the bytes of an +//! escaped version of the character given. +//! +//! [`escape_default`]: fn.escape_default.html + +#![stable(feature = "core_ascii", since = "1.26.0")] + +use fmt; +use ops::Range; +use iter::FusedIterator; + +/// An iterator over the escaped version of a byte. +/// +/// This `struct` is created by the [`escape_default`] function. See its +/// documentation for more. +/// +/// [`escape_default`]: fn.escape_default.html +#[stable(feature = "rust1", since = "1.0.0")] +pub struct EscapeDefault { + range: Range<usize>, + data: [u8; 4], +} + +/// Returns an iterator that produces an escaped version of a `u8`. +/// +/// The default is chosen with a bias toward producing literals that are +/// legal in a variety of languages, including C++11 and similar C-family +/// languages. The exact rules are: +/// +/// * Tab is escaped as `\t`. +/// * Carriage return is escaped as `\r`. +/// * Line feed is escaped as `\n`. +/// * Single quote is escaped as `\'`. +/// * Double quote is escaped as `\"`. +/// * Backslash is escaped as `\\`. +/// * Any character in the 'printable ASCII' range `0x20` .. `0x7e` +/// inclusive is not escaped. +/// * Any other chars are given hex escapes of the form '\xNN'. +/// * Unicode escapes are never generated by this function. +/// +/// # Examples +/// +/// ``` +/// use std::ascii; +/// +/// let escaped = ascii::escape_default(b'0').next().unwrap(); +/// assert_eq!(b'0', escaped); +/// +/// let mut escaped = ascii::escape_default(b'\t'); +/// +/// assert_eq!(b'\\', escaped.next().unwrap()); +/// assert_eq!(b't', escaped.next().unwrap()); +/// +/// let mut escaped = ascii::escape_default(b'\r'); +/// +/// assert_eq!(b'\\', escaped.next().unwrap()); +/// assert_eq!(b'r', escaped.next().unwrap()); +/// +/// let mut escaped = ascii::escape_default(b'\n'); +/// +/// assert_eq!(b'\\', escaped.next().unwrap()); +/// assert_eq!(b'n', escaped.next().unwrap()); +/// +/// let mut escaped = ascii::escape_default(b'\''); +/// +/// assert_eq!(b'\\', escaped.next().unwrap()); +/// assert_eq!(b'\'', escaped.next().unwrap()); +/// +/// let mut escaped = ascii::escape_default(b'"'); +/// +/// assert_eq!(b'\\', escaped.next().unwrap()); +/// assert_eq!(b'"', escaped.next().unwrap()); +/// +/// let mut escaped = ascii::escape_default(b'\\'); +/// +/// assert_eq!(b'\\', escaped.next().unwrap()); +/// assert_eq!(b'\\', escaped.next().unwrap()); +/// +/// let mut escaped = ascii::escape_default(b'\x9d'); +/// +/// assert_eq!(b'\\', escaped.next().unwrap()); +/// assert_eq!(b'x', escaped.next().unwrap()); +/// assert_eq!(b'9', escaped.next().unwrap()); +/// assert_eq!(b'd', escaped.next().unwrap()); +/// ``` +#[stable(feature = "rust1", since = "1.0.0")] +pub fn escape_default(c: u8) -> EscapeDefault { + let (data, len) = match c { + b'\t' => ([b'\\', b't', 0, 0], 2), + b'\r' => ([b'\\', b'r', 0, 0], 2), + b'\n' => ([b'\\', b'n', 0, 0], 2), + b'\\' => ([b'\\', b'\\', 0, 0], 2), + b'\'' => ([b'\\', b'\'', 0, 0], 2), + b'"' => ([b'\\', b'"', 0, 0], 2), + b'\x20' ..= b'\x7e' => ([c, 0, 0, 0], 1), + _ => ([b'\\', b'x', hexify(c >> 4), hexify(c & 0xf)], 4), + }; + + return EscapeDefault { range: 0..len, data }; + + fn hexify(b: u8) -> u8 { + match b { + 0 ..= 9 => b'0' + b, + _ => b'a' + b - 10, + } + } +} + +#[stable(feature = "rust1", since = "1.0.0")] +impl Iterator for EscapeDefault { + type Item = u8; + fn next(&mut self) -> Option<u8> { self.range.next().map(|i| self.data[i]) } + fn size_hint(&self) -> (usize, Option<usize>) { self.range.size_hint() } +} +#[stable(feature = "rust1", since = "1.0.0")] +impl DoubleEndedIterator for EscapeDefault { + fn next_back(&mut self) -> Option<u8> { + self.range.next_back().map(|i| self.data[i]) + } +} +#[stable(feature = "rust1", since = "1.0.0")] +impl ExactSizeIterator for EscapeDefault {} +#[stable(feature = "fused", since = "1.26.0")] +impl FusedIterator for EscapeDefault {} + +#[stable(feature = "std_debug", since = "1.16.0")] +impl fmt::Debug for EscapeDefault { + fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result { + f.pad("EscapeDefault { .. }") + } +} diff --git a/src/libcore/benches/iter.rs b/src/libcore/benches/iter.rs index b284d855c45..6c597301ac2 100644 --- a/src/libcore/benches/iter.rs +++ b/src/libcore/benches/iter.rs @@ -281,3 +281,32 @@ bench_sums! { bench_take_while_chain_ref_sum, (0i64..1000000).chain(1000000..).take_while(|&x| x < 1111111) } + +// Checks whether Skip<Zip<A,B>> is as fast as Zip<Skip<A>, Skip<B>>, from +// https://users.rust-lang.org/t/performance-difference-between-iterator-zip-and-skip-order/15743 +#[bench] +fn bench_zip_then_skip(b: &mut Bencher) { + let v: Vec<_> = (0..100_000).collect(); + let t: Vec<_> = (0..100_000).collect(); + + b.iter(|| { + let s = v.iter().zip(t.iter()).skip(10000) + .take_while(|t| *t.0 < 10100) + .map(|(a, b)| *a + *b) + .sum::<u64>(); + assert_eq!(s, 2009900); + }); +} +#[bench] +fn bench_skip_then_zip(b: &mut Bencher) { + let v: Vec<_> = (0..100_000).collect(); + let t: Vec<_> = (0..100_000).collect(); + + b.iter(|| { + let s = v.iter().skip(10000).zip(t.iter().skip(10000)) + .take_while(|t| *t.0 < 10100) + .map(|(a, b)| *a + *b) + .sum::<u64>(); + assert_eq!(s, 2009900); + }); +} diff --git a/src/libcore/benches/lib.rs b/src/libcore/benches/lib.rs index 201064e823b..ced77d77918 100644 --- a/src/libcore/benches/lib.rs +++ b/src/libcore/benches/lib.rs @@ -8,10 +8,7 @@ // option. This file may not be copied, modified, or distributed // except according to those terms. -#![deny(warnings)] - #![feature(flt2dec)] -#![feature(slice_patterns)] #![feature(test)] extern crate core; diff --git a/src/libcore/borrow.rs b/src/libcore/borrow.rs index 61558034e63..f45a32d4b94 100644 --- a/src/libcore/borrow.rs +++ b/src/libcore/borrow.rs @@ -14,24 +14,154 @@ /// A trait for borrowing data. /// -/// In general, there may be several ways to "borrow" a piece of data. The -/// typical ways of borrowing a type `T` are `&T` (a shared borrow) and `&mut T` -/// (a mutable borrow). But types like `Vec<T>` provide additional kinds of -/// borrows: the borrowed slices `&[T]` and `&mut [T]`. +/// In Rust, it is common to provide different representations of a type for +/// different use cases. For instance, storage location and management for a +/// value can be specifically chosen as appropriate for a particular use via +/// pointer types such as [`Box<T>`] or [`Rc<T>`]. Beyond these generic +/// wrappers that can be used with any type, some types provide optional +/// facets providing potentially costly functionality. An example for such a +/// type is [`String`] which adds the ability to extend a string to the basic +/// [`str`]. This requires keeping additional information unnecessary for a +/// simple, immutable string. /// -/// When writing generic code, it is often desirable to abstract over all ways -/// of borrowing data from a given type. That is the role of the `Borrow` -/// trait: if `T: Borrow<U>`, then `&U` can be borrowed from `&T`. A given -/// type can be borrowed as multiple different types. In particular, `Vec<T>: -/// Borrow<Vec<T>>` and `Vec<T>: Borrow<[T]>`. +/// These types provide access to the underlying data through references +/// to the type of that data. They are said to be ‘borrowed as’ that type. +/// For instance, a [`Box<T>`] can be borrowed as `T` while a [`String`] +/// can be borrowed as `str`. /// -/// If you are implementing `Borrow` and both `Self` and `Borrowed` implement -/// `Hash`, `Eq`, and/or `Ord`, they must produce the same result. +/// Types express that they can be borrowed as some type `T` by implementing +/// `Borrow<T>`, providing a reference to a `T` in the trait’s +/// [`borrow`] method. A type is free to borrow as several different types. +/// If it wishes to mutably borrow as the type – allowing the underlying data +/// to be modified, it can additionally implement [`BorrowMut<T>`]. /// -/// `Borrow` is very similar to, but different than, `AsRef`. See -/// [the book][book] for more. +/// Further, when providing implementations for additional traits, it needs +/// to be considered whether they should behave identical to those of the +/// underlying type as a consequence of acting as a representation of that +/// underlying type. Generic code typically uses `Borrow<T>` when it relies +/// on the identical behavior of these additional trait implementations. +/// These traits will likely appear as additional trait bounds. /// -/// [book]: ../../book/first-edition/borrow-and-asref.html +/// If generic code merely needs to work for all types that can +/// provide a reference to related type `T`, it is often better to use +/// [`AsRef<T>`] as more types can safely implement it. +/// +/// [`AsRef<T>`]: ../../std/convert/trait.AsRef.html +/// [`BorrowMut<T>`]: trait.BorrowMut.html +/// [`Box<T>`]: ../../std/boxed/struct.Box.html +/// [`Mutex<T>`]: ../../std/sync/struct.Mutex.html +/// [`Rc<T>`]: ../../std/rc/struct.Rc.html +/// [`str`]: ../../std/primitive.str.html +/// [`String`]: ../../std/string/struct.String.html +/// [`borrow`]: #tymethod.borrow +/// +/// # Examples +/// +/// As a data collection, [`HashMap<K, V>`] owns both keys and values. If +/// the key’s actual data is wrapped in a managing type of some kind, it +/// should, however, still be possible to search for a value using a +/// reference to the key’s data. For instance, if the key is a string, then +/// it is likely stored with the hash map as a [`String`], while it should +/// be possible to search using a [`&str`][`str`]. Thus, `insert` needs to +/// operate on a `String` while `get` needs to be able to use a `&str`. +/// +/// Slightly simplified, the relevant parts of `HashMap<K, V>` look like +/// this: +/// +/// ``` +/// use std::borrow::Borrow; +/// use std::hash::Hash; +/// +/// pub struct HashMap<K, V> { +/// # marker: ::std::marker::PhantomData<(K, V)>, +/// // fields omitted +/// } +/// +/// impl<K, V> HashMap<K, V> { +/// pub fn insert(&self, key: K, value: V) -> Option<V> +/// where K: Hash + Eq +/// { +/// # unimplemented!() +/// // ... +/// } +/// +/// pub fn get<Q>(&self, k: &Q) -> Option<&V> +/// where +/// K: Borrow<Q>, +/// Q: Hash + Eq + ?Sized +/// { +/// # unimplemented!() +/// // ... +/// } +/// } +/// ``` +/// +/// The entire hash map is generic over a key type `K`. Because these keys +/// are stored with the hash map, this type has to own the key’s data. +/// When inserting a key-value pair, the map is given such a `K` and needs +/// to find the correct hash bucket and check if the key is already present +/// based on that `K`. It therefore requires `K: Hash + Eq`. +/// +/// When searching for a value in the map, however, having to provide a +/// reference to a `K` as the key to search for would require to always +/// create such an owned value. For string keys, this would mean a `String` +/// value needs to be created just for the search for cases where only a +/// `str` is available. +/// +/// Instead, the `get` method is generic over the type of the underlying key +/// data, called `Q` in the method signature above. It states that `K` +/// borrows as a `Q` by requiring that `K: Borrow<Q>`. By additionally +/// requiring `Q: Hash + Eq`, it signals the requirement that `K` and `Q` +/// have implementations of the `Hash` and `Eq` traits that produce identical +/// results. +/// +/// The implementation of `get` relies in particular on identical +/// implementations of `Hash` by determining the key’s hash bucket by calling +/// `Hash::hash` on the `Q` value even though it inserted the key based on +/// the hash value calculated from the `K` value. +/// +/// As a consequence, the hash map breaks if a `K` wrapping a `Q` value +/// produces a different hash than `Q`. For instance, imagine you have a +/// type that wraps a string but compares ASCII letters ignoring their case: +/// +/// ``` +/// pub struct CaseInsensitiveString(String); +/// +/// impl PartialEq for CaseInsensitiveString { +/// fn eq(&self, other: &Self) -> bool { +/// self.0.eq_ignore_ascii_case(&other.0) +/// } +/// } +/// +/// impl Eq for CaseInsensitiveString { } +/// ``` +/// +/// Because two equal values need to produce the same hash value, the +/// implementation of `Hash` needs to ignore ASCII case, too: +/// +/// ``` +/// # use std::hash::{Hash, Hasher}; +/// # pub struct CaseInsensitiveString(String); +/// impl Hash for CaseInsensitiveString { +/// fn hash<H: Hasher>(&self, state: &mut H) { +/// for c in self.0.as_bytes() { +/// c.to_ascii_lowercase().hash(state) +/// } +/// } +/// } +/// ``` +/// +/// Can `CaseInsensitiveString` implement `Borrow<str>`? It certainly can +/// provide a reference to a string slice via its contained owned string. +/// But because its `Hash` implementation differs, it behaves differently +/// from `str` and therefore must not, in fact, implement `Borrow<str>`. +/// If it wants to allow others access to the underlying `str`, it can do +/// that via `AsRef<str>` which doesn’t carry any extra requirements. +/// +/// [`Hash`]: ../../std/hash/trait.Hash.html +/// [`HashMap<K, V>`]: ../../std/collections/struct.HashMap.html +/// [`String`]: ../../std/string/struct.String.html +/// [`str`]: ../../std/primitive.str.html #[stable(feature = "rust1", since = "1.0.0")] pub trait Borrow<Borrowed: ?Sized> { /// Immutably borrows from an owned value. @@ -59,7 +189,11 @@ pub trait Borrow<Borrowed: ?Sized> { /// A trait for mutably borrowing data. /// -/// Similar to `Borrow`, but for mutable borrows. +/// As a companion to [`Borrow<T>`] this trait allows a type to borrow as +/// an underlying type by providing a mutable reference. See [`Borrow<T>`] +/// for more information on borrowing as another type. +/// +/// [`Borrow<T>`]: trait.Borrow.html #[stable(feature = "rust1", since = "1.0.0")] pub trait BorrowMut<Borrowed: ?Sized> : Borrow<Borrowed> { /// Mutably borrows from an owned value. diff --git a/src/libcore/cell.rs b/src/libcore/cell.rs index c5375d1e00c..009aba5f598 100644 --- a/src/libcore/cell.rs +++ b/src/libcore/cell.rs @@ -10,6 +10,24 @@ //! Shareable mutable containers. //! +//! Rust memory safety is based on this rule: Given an object `T`, it is only possible to +//! have one of the following: +//! +//! - Having several immutable references (`&T`) to the object (also known as **aliasing**). +//! - Having one mutable reference (`&mut T`) to the object (also known as **mutability**). +//! +//! This is enforced by the Rust compiler. However, there are situations where this rule is not +//! flexible enough. Sometimes it is required to have multiple references to an object and yet +//! mutate it. +//! +//! Shareable mutable containers exist to permit mutability in a controlled manner, even in the +//! presence of aliasing. Both `Cell<T>` and `RefCell<T>` allows to do this in a single threaded +//! way. However, neither `Cell<T>` nor `RefCell<T>` are thread safe (they do not implement +//! `Sync`). If you need to do aliasing and mutation between multiple threads it is possible to +//! use [`Mutex`](../../std/sync/struct.Mutex.html), +//! [`RwLock`](../../std/sync/struct.RwLock.html) or +//! [`atomic`](../../core/sync/atomic/index.html) types. +//! //! Values of the `Cell<T>` and `RefCell<T>` types may be mutated through shared references (i.e. //! the common `&T` type), whereas most Rust types can only be mutated through unique (`&mut T`) //! references. We say that `Cell<T>` and `RefCell<T>` provide 'interior mutability', in contrast @@ -128,13 +146,12 @@ //! //! ``` //! #![feature(core_intrinsics)] -//! #![feature(shared)] //! use std::cell::Cell; -//! use std::ptr::Shared; +//! use std::ptr::NonNull; //! use std::intrinsics::abort; //! //! struct Rc<T: ?Sized> { -//! ptr: Shared<RcBox<T>> +//! ptr: NonNull<RcBox<T>> //! } //! //! struct RcBox<T: ?Sized> { @@ -218,7 +235,8 @@ use ptr; /// /// See the [module-level documentation](index.html) for more. #[stable(feature = "rust1", since = "1.0.0")] -pub struct Cell<T> { +#[repr(transparent)] +pub struct Cell<T: ?Sized> { value: UnsafeCell<T>, } @@ -239,13 +257,40 @@ impl<T:Copy> Cell<T> { pub fn get(&self) -> T { unsafe{ *self.value.get() } } + + /// Updates the contained value using a function and returns the new value. + /// + /// # Examples + /// + /// ``` + /// #![feature(cell_update)] + /// + /// use std::cell::Cell; + /// + /// let c = Cell::new(5); + /// let new = c.update(|x| x + 1); + /// + /// assert_eq!(new, 6); + /// assert_eq!(c.get(), 6); + /// ``` + #[inline] + #[unstable(feature = "cell_update", issue = "50186")] + pub fn update<F>(&self, f: F) -> T + where + F: FnOnce(T) -> T, + { + let old = self.get(); + let new = f(old); + self.set(new); + new + } } #[stable(feature = "rust1", since = "1.0.0")] -unsafe impl<T> Send for Cell<T> where T: Send {} +unsafe impl<T: ?Sized> Send for Cell<T> where T: Send {} #[stable(feature = "rust1", since = "1.0.0")] -impl<T> !Sync for Cell<T> {} +impl<T: ?Sized> !Sync for Cell<T> {} #[stable(feature = "rust1", since = "1.0.0")] impl<T:Copy> Clone for Cell<T> { @@ -336,46 +381,6 @@ impl<T> Cell<T> { } } - /// Returns a raw pointer to the underlying data in this cell. - /// - /// # Examples - /// - /// ``` - /// use std::cell::Cell; - /// - /// let c = Cell::new(5); - /// - /// let ptr = c.as_ptr(); - /// ``` - #[inline] - #[stable(feature = "cell_as_ptr", since = "1.12.0")] - pub fn as_ptr(&self) -> *mut T { - self.value.get() - } - - /// Returns a mutable reference to the underlying data. - /// - /// This call borrows `Cell` mutably (at compile-time) which guarantees - /// that we possess the only reference. - /// - /// # Examples - /// - /// ``` - /// use std::cell::Cell; - /// - /// let mut c = Cell::new(5); - /// *c.get_mut() += 1; - /// - /// assert_eq!(c.get(), 6); - /// ``` - #[inline] - #[stable(feature = "cell_get_mut", since = "1.11.0")] - pub fn get_mut(&mut self) -> &mut T { - unsafe { - &mut *self.value.get() - } - } - /// Sets the contained value. /// /// # Examples @@ -450,7 +455,71 @@ impl<T> Cell<T> { /// ``` #[stable(feature = "move_cell", since = "1.17.0")] pub fn into_inner(self) -> T { - unsafe { self.value.into_inner() } + self.value.into_inner() + } +} + +impl<T: ?Sized> Cell<T> { + /// Returns a raw pointer to the underlying data in this cell. + /// + /// # Examples + /// + /// ``` + /// use std::cell::Cell; + /// + /// let c = Cell::new(5); + /// + /// let ptr = c.as_ptr(); + /// ``` + #[inline] + #[stable(feature = "cell_as_ptr", since = "1.12.0")] + pub fn as_ptr(&self) -> *mut T { + self.value.get() + } + + /// Returns a mutable reference to the underlying data. + /// + /// This call borrows `Cell` mutably (at compile-time) which guarantees + /// that we possess the only reference. + /// + /// # Examples + /// + /// ``` + /// use std::cell::Cell; + /// + /// let mut c = Cell::new(5); + /// *c.get_mut() += 1; + /// + /// assert_eq!(c.get(), 6); + /// ``` + #[inline] + #[stable(feature = "cell_get_mut", since = "1.11.0")] + pub fn get_mut(&mut self) -> &mut T { + unsafe { + &mut *self.value.get() + } + } + + /// Returns a `&Cell<T>` from a `&mut T` + /// + /// # Examples + /// + /// ``` + /// #![feature(as_cell)] + /// use std::cell::Cell; + /// + /// let slice: &mut [i32] = &mut [1, 2, 3]; + /// let cell_slice: &Cell<[i32]> = Cell::from_mut(slice); + /// let slice_cell: &[Cell<i32>] = cell_slice.as_slice_of_cells(); + /// + /// assert_eq!(slice_cell.len(), 3); + /// ``` + #[inline] + #[unstable(feature = "as_cell", issue="43038")] + pub fn from_mut(t: &mut T) -> &Cell<T> { + unsafe { + &*(t as *mut T as *const Cell<T>) + } } } @@ -477,6 +546,29 @@ impl<T: Default> Cell<T> { #[unstable(feature = "coerce_unsized", issue = "27732")] impl<T: CoerceUnsized<U>, U> CoerceUnsized<Cell<U>> for Cell<T> {} +impl<T> Cell<[T]> { + /// Returns a `&[Cell<T>]` from a `&Cell<[T]>` + /// + /// # Examples + /// + /// ``` + /// #![feature(as_cell)] + /// use std::cell::Cell; + /// + /// let slice: &mut [i32] = &mut [1, 2, 3]; + /// let cell_slice: &Cell<[i32]> = Cell::from_mut(slice); + /// let slice_cell: &[Cell<i32>] = cell_slice.as_slice_of_cells(); + /// + /// assert_eq!(slice_cell.len(), 3); + /// ``` + #[unstable(feature = "as_cell", issue="43038")] + pub fn as_slice_of_cells(&self) -> &[Cell<T>] { + unsafe { + &*(self as *const Cell<[T]> as *const [Cell<T>]) + } + } +} + /// A mutable memory location with dynamically checked borrow rules /// /// See the [module-level documentation](index.html) for more. @@ -526,11 +618,31 @@ impl Display for BorrowMutError { } } -// Values [1, MAX-1] represent the number of `Ref` active -// (will not outgrow its range since `usize` is the size of the address space) -type BorrowFlag = usize; +// Positive values represent the number of `Ref` active. Negative values +// represent the number of `RefMut` active. Multiple `RefMut`s can only be +// active at a time if they refer to distinct, nonoverlapping components of a +// `RefCell` (e.g., different ranges of a slice). +// +// `Ref` and `RefMut` are both two words in size, and so there will likely never +// be enough `Ref`s or `RefMut`s in existence to overflow half of the `usize` +// range. Thus, a `BorrowFlag` will probably never overflow or underflow. +// However, this is not a guarantee, as a pathological program could repeatedly +// create and then mem::forget `Ref`s or `RefMut`s. Thus, all code must +// explicitly check for overflow and underflow in order to avoid unsafety, or at +// least behave correctly in the event that overflow or underflow happens (e.g., +// see BorrowRef::new). +type BorrowFlag = isize; const UNUSED: BorrowFlag = 0; -const WRITING: BorrowFlag = !0; + +#[inline(always)] +fn is_writing(x: BorrowFlag) -> bool { + x < UNUSED +} + +#[inline(always)] +fn is_reading(x: BorrowFlag) -> bool { + x > UNUSED +} impl<T> RefCell<T> { /// Creates a new `RefCell` containing `value`. @@ -569,7 +681,7 @@ impl<T> RefCell<T> { // compiler statically verifies that it is not currently borrowed. // Therefore the following assertion is just a `debug_assert!`. debug_assert!(self.borrow.get() == UNUSED); - unsafe { self.value.into_inner() } + self.value.into_inner() } /// Replaces the wrapped value with a new one, returning the old value, @@ -731,8 +843,9 @@ impl<T: ?Sized> RefCell<T> { /// Mutably borrows the wrapped value. /// - /// The borrow lasts until the returned `RefMut` exits scope. The value - /// cannot be borrowed while this borrow is active. + /// The borrow lasts until the returned `RefMut` or all `RefMut`s derived + /// from it exit scope. The value cannot be borrowed while this borrow is + /// active. /// /// # Panics /// @@ -774,8 +887,9 @@ impl<T: ?Sized> RefCell<T> { /// Mutably borrows the wrapped value, returning an error if the value is currently borrowed. /// - /// The borrow lasts until the returned `RefMut` exits scope. The value cannot be borrowed - /// while this borrow is active. + /// The borrow lasts until the returned `RefMut` or all `RefMut`s derived + /// from it exit scope. The value cannot be borrowed while this borrow is + /// active. /// /// This is the non-panicking variant of [`borrow_mut`](#method.borrow_mut). /// @@ -863,6 +977,9 @@ impl<T: ?Sized> !Sync for RefCell<T> {} #[stable(feature = "rust1", since = "1.0.0")] impl<T: Clone> Clone for RefCell<T> { + /// # Panics + /// + /// Panics if the value is currently mutably borrowed. #[inline] fn clone(&self) -> RefCell<T> { RefCell::new(self.borrow().clone()) @@ -880,6 +997,9 @@ impl<T:Default> Default for RefCell<T> { #[stable(feature = "rust1", since = "1.0.0")] impl<T: ?Sized + PartialEq> PartialEq for RefCell<T> { + /// # Panics + /// + /// Panics if the value in either `RefCell` is currently borrowed. #[inline] fn eq(&self, other: &RefCell<T>) -> bool { *self.borrow() == *other.borrow() @@ -891,26 +1011,41 @@ impl<T: ?Sized + Eq> Eq for RefCell<T> {} #[stable(feature = "cell_ord", since = "1.10.0")] impl<T: ?Sized + PartialOrd> PartialOrd for RefCell<T> { + /// # Panics + /// + /// Panics if the value in either `RefCell` is currently borrowed. #[inline] fn partial_cmp(&self, other: &RefCell<T>) -> Option<Ordering> { self.borrow().partial_cmp(&*other.borrow()) } + /// # Panics + /// + /// Panics if the value in either `RefCell` is currently borrowed. #[inline] fn lt(&self, other: &RefCell<T>) -> bool { *self.borrow() < *other.borrow() } + /// # Panics + /// + /// Panics if the value in either `RefCell` is currently borrowed. #[inline] fn le(&self, other: &RefCell<T>) -> bool { *self.borrow() <= *other.borrow() } + /// # Panics + /// + /// Panics if the value in either `RefCell` is currently borrowed. #[inline] fn gt(&self, other: &RefCell<T>) -> bool { *self.borrow() > *other.borrow() } + /// # Panics + /// + /// Panics if the value in either `RefCell` is currently borrowed. #[inline] fn ge(&self, other: &RefCell<T>) -> bool { *self.borrow() >= *other.borrow() @@ -919,6 +1054,9 @@ impl<T: ?Sized + PartialOrd> PartialOrd for RefCell<T> { #[stable(feature = "cell_ord", since = "1.10.0")] impl<T: ?Sized + Ord> Ord for RefCell<T> { + /// # Panics + /// + /// Panics if the value in either `RefCell` is currently borrowed. #[inline] fn cmp(&self, other: &RefCell<T>) -> Ordering { self.borrow().cmp(&*other.borrow()) @@ -942,12 +1080,14 @@ struct BorrowRef<'b> { impl<'b> BorrowRef<'b> { #[inline] fn new(borrow: &'b Cell<BorrowFlag>) -> Option<BorrowRef<'b>> { - match borrow.get() { - WRITING => None, - b => { - borrow.set(b + 1); - Some(BorrowRef { borrow: borrow }) - }, + let b = borrow.get(); + if is_writing(b) || b == isize::max_value() { + // If there's currently a writing borrow, or if incrementing the + // refcount would overflow into a writing borrow. + None + } else { + borrow.set(b + 1); + Some(BorrowRef { borrow }) } } } @@ -956,7 +1096,7 @@ impl<'b> Drop for BorrowRef<'b> { #[inline] fn drop(&mut self) { let borrow = self.borrow.get(); - debug_assert!(borrow != WRITING && borrow != UNUSED); + debug_assert!(is_reading(borrow)); self.borrow.set(borrow - 1); } } @@ -965,11 +1105,12 @@ impl<'b> Clone for BorrowRef<'b> { #[inline] fn clone(&self) -> BorrowRef<'b> { // Since this Ref exists, we know the borrow flag - // is not set to WRITING. + // is a reading borrow. let borrow = self.borrow.get(); - debug_assert!(borrow != UNUSED); - // Prevent the borrow counter from overflowing. - assert!(borrow != WRITING); + debug_assert!(is_reading(borrow)); + // Prevent the borrow counter from overflowing into + // a writing borrow. + assert!(borrow != isize::max_value()); self.borrow.set(borrow + 1); BorrowRef { borrow: self.borrow } } @@ -1041,6 +1182,37 @@ impl<'b, T: ?Sized> Ref<'b, T> { borrow: orig.borrow, } } + + /// Split a `Ref` into multiple `Ref`s for different components of the + /// borrowed data. + /// + /// The `RefCell` is already immutably borrowed, so this cannot fail. + /// + /// This is an associated function that needs to be used as + /// `Ref::map_split(...)`. A method would interfere with methods of the same + /// name on the contents of a `RefCell` used through `Deref`. + /// + /// # Examples + /// + /// ``` + /// #![feature(refcell_map_split)] + /// use std::cell::{Ref, RefCell}; + /// + /// let cell = RefCell::new([1, 2, 3, 4]); + /// let borrow = cell.borrow(); + /// let (begin, end) = Ref::map_split(borrow, |slice| slice.split_at(2)); + /// assert_eq!(*begin, [1, 2]); + /// assert_eq!(*end, [3, 4]); + /// ``` + #[unstable(feature = "refcell_map_split", issue = "51476")] + #[inline] + pub fn map_split<U: ?Sized, V: ?Sized, F>(orig: Ref<'b, T>, f: F) -> (Ref<'b, U>, Ref<'b, V>) + where F: FnOnce(&T) -> (&U, &V) + { + let (a, b) = f(orig.value); + let borrow = orig.borrow.clone(); + (Ref { value: a, borrow }, Ref { value: b, borrow: orig.borrow }) + } } #[unstable(feature = "coerce_unsized", issue = "27732")] @@ -1086,9 +1258,47 @@ impl<'b, T: ?Sized> RefMut<'b, T> { let RefMut { value, borrow } = orig; RefMut { value: f(value), - borrow: borrow, + borrow, } } + + /// Split a `RefMut` into multiple `RefMut`s for different components of the + /// borrowed data. + /// + /// The underlying `RefCell` will remain mutably borrowed until both + /// returned `RefMut`s go out of scope. + /// + /// The `RefCell` is already mutably borrowed, so this cannot fail. + /// + /// This is an associated function that needs to be used as + /// `RefMut::map_split(...)`. A method would interfere with methods of the + /// same name on the contents of a `RefCell` used through `Deref`. + /// + /// # Examples + /// + /// ``` + /// #![feature(refcell_map_split)] + /// use std::cell::{RefCell, RefMut}; + /// + /// let cell = RefCell::new([1, 2, 3, 4]); + /// let borrow = cell.borrow_mut(); + /// let (mut begin, mut end) = RefMut::map_split(borrow, |slice| slice.split_at_mut(2)); + /// assert_eq!(*begin, [1, 2]); + /// assert_eq!(*end, [3, 4]); + /// begin.copy_from_slice(&[4, 3]); + /// end.copy_from_slice(&[2, 1]); + /// ``` + #[unstable(feature = "refcell_map_split", issue = "51476")] + #[inline] + pub fn map_split<U: ?Sized, V: ?Sized, F>( + orig: RefMut<'b, T>, f: F + ) -> (RefMut<'b, U>, RefMut<'b, V>) + where F: FnOnce(&mut T) -> (&mut U, &mut V) + { + let (a, b) = f(orig.value); + let borrow = orig.borrow.clone(); + (RefMut { value: a, borrow }, RefMut { value: b, borrow: orig.borrow }) + } } struct BorrowRefMut<'b> { @@ -1099,22 +1309,41 @@ impl<'b> Drop for BorrowRefMut<'b> { #[inline] fn drop(&mut self) { let borrow = self.borrow.get(); - debug_assert!(borrow == WRITING); - self.borrow.set(UNUSED); + debug_assert!(is_writing(borrow)); + self.borrow.set(borrow + 1); } } impl<'b> BorrowRefMut<'b> { #[inline] fn new(borrow: &'b Cell<BorrowFlag>) -> Option<BorrowRefMut<'b>> { + // NOTE: Unlike BorrowRefMut::clone, new is called to create the initial + // mutable reference, and so there must currently be no existing + // references. Thus, while clone increments the mutable refcount, here + // we explicitly only allow going from UNUSED to UNUSED - 1. match borrow.get() { UNUSED => { - borrow.set(WRITING); - Some(BorrowRefMut { borrow: borrow }) + borrow.set(UNUSED - 1); + Some(BorrowRefMut { borrow }) }, _ => None, } } + + // Clone a `BorrowRefMut`. + // + // This is only valid if each `BorrowRefMut` is used to track a mutable + // reference to a distinct, nonoverlapping range of the original object. + // This isn't in a Clone impl so that code doesn't call this implicitly. + #[inline] + fn clone(&self) -> BorrowRefMut<'b> { + let borrow = self.borrow.get(); + debug_assert!(is_writing(borrow)); + // Prevent the borrow counter from underflowing. + assert!(borrow != isize::min_value()); + self.borrow.set(borrow - 1); + BorrowRefMut { borrow: self.borrow } + } } /// A wrapper type for a mutably borrowed value from a `RefCell<T>`. @@ -1161,21 +1390,43 @@ impl<'a, T: ?Sized + fmt::Display> fmt::Display for RefMut<'a, T> { /// The `UnsafeCell<T>` type is the only legal way to obtain aliasable data that is considered /// mutable. In general, transmuting an `&T` type into an `&mut T` is considered undefined behavior. /// -/// The compiler makes optimizations based on the knowledge that `&T` is not mutably aliased or -/// mutated, and that `&mut T` is unique. When building abstractions like `Cell`, `RefCell`, -/// `Mutex`, etc, you need to turn these optimizations off. `UnsafeCell` is the only legal way -/// to do this. When `UnsafeCell<T>` is immutably aliased, it is still safe to obtain a mutable -/// reference to its interior and/or to mutate it. However, it is up to the abstraction designer -/// to ensure that no two mutable references obtained this way are active at the same time, and -/// that there are no active mutable references or mutations when an immutable reference is obtained -/// from the cell. This is often done via runtime checks. +/// If you have a reference `&SomeStruct`, then normally in Rust all fields of `SomeStruct` are +/// immutable. The compiler makes optimizations based on the knowledge that `&T` is not mutably +/// aliased or mutated, and that `&mut T` is unique. `UnsafeCell<T>` is the only core language +/// feature to work around this restriction. All other types that allow internal mutability, such as +/// `Cell<T>` and `RefCell<T>`, use `UnsafeCell` to wrap their internal data. /// -/// Note that while mutating or mutably aliasing the contents of an `& UnsafeCell<T>` is -/// okay (provided you enforce the invariants some other way); it is still undefined behavior -/// to have multiple `&mut UnsafeCell<T>` aliases. +/// The `UnsafeCell` API itself is technically very simple: it gives you a raw pointer `*mut T` to +/// its contents. It is up to _you_ as the abstraction designer to use that raw pointer correctly. +/// +/// The precise Rust aliasing rules are somewhat in flux, but the main points are not contentious: +/// +/// - If you create a safe reference with lifetime `'a` (either a `&T` or `&mut T` +/// reference) that is accessible by safe code (for example, because you returned it), +/// then you must not access the data in any way that contradicts that reference for the +/// remainder of `'a`. For example, this means that if you take the `*mut T` from an +/// `UnsafeCell<T>` and cast it to an `&T`, then the data in `T` must remain immutable +/// (modulo any `UnsafeCell` data found within `T`, of course) until that reference's +/// lifetime expires. Similarly, if you create a `&mut T` reference that is released to +/// safe code, then you must not access the data within the `UnsafeCell` until that +/// reference expires. +/// +/// - At all times, you must avoid data races. If multiple threads have access to +/// the same `UnsafeCell`, then any writes must have a proper happens-before relation to all other +/// accesses (or use atomics). /// +/// To assist with proper design, the following scenarios are explicitly declared legal +/// for single-threaded code: /// -/// Types like `Cell<T>` and `RefCell<T>` use this type to wrap their internal data. +/// 1. A `&T` reference can be released to safe code and there it can co-exist with other `&T` +/// references, but not with a `&mut T` +/// +/// 2. A `&mut T` reference may be released to safe code provided neither other `&mut T` nor `&T` +/// co-exist with it. A `&mut T` must always be unique. +/// +/// Note that while mutating or mutably aliasing the contents of an `&UnsafeCell<T>` is +/// okay (provided you enforce the invariants some other way), it is still undefined behavior +/// to have multiple `&mut UnsafeCell<T>` aliases. /// /// # Examples /// @@ -1192,6 +1443,7 @@ impl<'a, T: ?Sized + fmt::Display> fmt::Display for RefMut<'a, T> { /// ``` #[lang = "unsafe_cell"] #[stable(feature = "rust1", since = "1.0.0")] +#[repr(transparent)] pub struct UnsafeCell<T: ?Sized> { value: T, } @@ -1215,16 +1467,11 @@ impl<T> UnsafeCell<T> { #[stable(feature = "rust1", since = "1.0.0")] #[inline] pub const fn new(value: T) -> UnsafeCell<T> { - UnsafeCell { value: value } + UnsafeCell { value } } /// Unwraps the value. /// - /// # Safety - /// - /// This function is unsafe because this thread or another thread may currently be - /// inspecting the inner value. - /// /// # Examples /// /// ``` @@ -1232,11 +1479,11 @@ impl<T> UnsafeCell<T> { /// /// let uc = UnsafeCell::new(5); /// - /// let five = unsafe { uc.into_inner() }; + /// let five = uc.into_inner(); /// ``` #[inline] #[stable(feature = "rust1", since = "1.0.0")] - pub unsafe fn into_inner(self) -> T { + pub fn into_inner(self) -> T { self.value } } @@ -1245,9 +1492,9 @@ impl<T: ?Sized> UnsafeCell<T> { /// Gets a mutable pointer to the wrapped value. /// /// This can be cast to a pointer of any kind. - /// Ensure that the access is unique when casting to - /// `&mut T`, and ensure that there are no mutations or mutable - /// aliases going on when casting to `&T` + /// Ensure that the access is unique (no active references, mutable or not) + /// when casting to `&mut T`, and ensure that there are no mutations + /// or mutable aliases going on when casting to `&T` /// /// # Examples /// @@ -1285,7 +1532,7 @@ impl<T: CoerceUnsized<U>, U> CoerceUnsized<UnsafeCell<U>> for UnsafeCell<T> {} #[allow(unused)] fn assert_coerce_unsized(a: UnsafeCell<&i32>, b: Cell<&i32>, c: RefCell<&i32>) { - let _: UnsafeCell<&Send> = a; - let _: Cell<&Send> = b; - let _: RefCell<&Send> = c; + let _: UnsafeCell<&dyn Send> = a; + let _: Cell<&dyn Send> = b; + let _: RefCell<&dyn Send> = c; } diff --git a/src/libcore/char.rs b/src/libcore/char.rs deleted file mode 100644 index e8b81db0706..00000000000 --- a/src/libcore/char.rs +++ /dev/null @@ -1,908 +0,0 @@ -// Copyright 2012-2014 The Rust Project Developers. See the COPYRIGHT -// file at the top-level directory of this distribution and at -// http://rust-lang.org/COPYRIGHT. -// -// Licensed under the Apache License, Version 2.0 <LICENSE-APACHE or -// http://www.apache.org/licenses/LICENSE-2.0> or the MIT license -// <LICENSE-MIT or http://opensource.org/licenses/MIT>, at your -// option. This file may not be copied, modified, or distributed -// except according to those terms. - -//! Character manipulation. -//! -//! For more details, see ::std_unicode::char (a.k.a. std::char) - -#![allow(non_snake_case)] -#![stable(feature = "core_char", since = "1.2.0")] - -use char_private::is_printable; -use convert::TryFrom; -use fmt::{self, Write}; -use slice; -use str::{from_utf8_unchecked_mut, FromStr}; -use iter::FusedIterator; -use mem::transmute; - -// UTF-8 ranges and tags for encoding characters -const TAG_CONT: u8 = 0b1000_0000; -const TAG_TWO_B: u8 = 0b1100_0000; -const TAG_THREE_B: u8 = 0b1110_0000; -const TAG_FOUR_B: u8 = 0b1111_0000; -const MAX_ONE_B: u32 = 0x80; -const MAX_TWO_B: u32 = 0x800; -const MAX_THREE_B: u32 = 0x10000; - -/* - Lu Uppercase_Letter an uppercase letter - Ll Lowercase_Letter a lowercase letter - Lt Titlecase_Letter a digraphic character, with first part uppercase - Lm Modifier_Letter a modifier letter - Lo Other_Letter other letters, including syllables and ideographs - Mn Nonspacing_Mark a nonspacing combining mark (zero advance width) - Mc Spacing_Mark a spacing combining mark (positive advance width) - Me Enclosing_Mark an enclosing combining mark - Nd Decimal_Number a decimal digit - Nl Letter_Number a letterlike numeric character - No Other_Number a numeric character of other type - Pc Connector_Punctuation a connecting punctuation mark, like a tie - Pd Dash_Punctuation a dash or hyphen punctuation mark - Ps Open_Punctuation an opening punctuation mark (of a pair) - Pe Close_Punctuation a closing punctuation mark (of a pair) - Pi Initial_Punctuation an initial quotation mark - Pf Final_Punctuation a final quotation mark - Po Other_Punctuation a punctuation mark of other type - Sm Math_Symbol a symbol of primarily mathematical use - Sc Currency_Symbol a currency sign - Sk Modifier_Symbol a non-letterlike modifier symbol - So Other_Symbol a symbol of other type - Zs Space_Separator a space character (of various non-zero widths) - Zl Line_Separator U+2028 LINE SEPARATOR only - Zp Paragraph_Separator U+2029 PARAGRAPH SEPARATOR only - Cc Control a C0 or C1 control code - Cf Format a format control character - Cs Surrogate a surrogate code point - Co Private_Use a private-use character - Cn Unassigned a reserved unassigned code point or a noncharacter -*/ - -/// The highest valid code point a `char` can have. -/// -/// A [`char`] is a [Unicode Scalar Value], which means that it is a [Code -/// Point], but only ones within a certain range. `MAX` is the highest valid -/// code point that's a valid [Unicode Scalar Value]. -/// -/// [`char`]: ../../std/primitive.char.html -/// [Unicode Scalar Value]: http://www.unicode.org/glossary/#unicode_scalar_value -/// [Code Point]: http://www.unicode.org/glossary/#code_point -#[stable(feature = "rust1", since = "1.0.0")] -pub const MAX: char = '\u{10ffff}'; - -/// Converts a `u32` to a `char`. -/// -/// Note that all [`char`]s are valid [`u32`]s, and can be casted to one with -/// [`as`]: -/// -/// ``` -/// let c = '💯'; -/// let i = c as u32; -/// -/// assert_eq!(128175, i); -/// ``` -/// -/// However, the reverse is not true: not all valid [`u32`]s are valid -/// [`char`]s. `from_u32()` will return `None` if the input is not a valid value -/// for a [`char`]. -/// -/// [`char`]: ../../std/primitive.char.html -/// [`u32`]: ../../std/primitive.u32.html -/// [`as`]: ../../book/first-edition/casting-between-types.html#as -/// -/// For an unsafe version of this function which ignores these checks, see -/// [`from_u32_unchecked`]. -/// -/// [`from_u32_unchecked`]: fn.from_u32_unchecked.html -/// -/// # Examples -/// -/// Basic usage: -/// -/// ``` -/// use std::char; -/// -/// let c = char::from_u32(0x2764); -/// -/// assert_eq!(Some('❤'), c); -/// ``` -/// -/// Returning `None` when the input is not a valid [`char`]: -/// -/// ``` -/// use std::char; -/// -/// let c = char::from_u32(0x110000); -/// -/// assert_eq!(None, c); -/// ``` -#[inline] -#[stable(feature = "rust1", since = "1.0.0")] -pub fn from_u32(i: u32) -> Option<char> { - char::try_from(i).ok() -} - -/// Converts a `u32` to a `char`, ignoring validity. -/// -/// Note that all [`char`]s are valid [`u32`]s, and can be casted to one with -/// [`as`]: -/// -/// ``` -/// let c = '💯'; -/// let i = c as u32; -/// -/// assert_eq!(128175, i); -/// ``` -/// -/// However, the reverse is not true: not all valid [`u32`]s are valid -/// [`char`]s. `from_u32_unchecked()` will ignore this, and blindly cast to -/// [`char`], possibly creating an invalid one. -/// -/// [`char`]: ../../std/primitive.char.html -/// [`u32`]: ../../std/primitive.u32.html -/// [`as`]: ../../book/first-edition/casting-between-types.html#as -/// -/// # Safety -/// -/// This function is unsafe, as it may construct invalid `char` values. -/// -/// For a safe version of this function, see the [`from_u32`] function. -/// -/// [`from_u32`]: fn.from_u32.html -/// -/// # Examples -/// -/// Basic usage: -/// -/// ``` -/// use std::char; -/// -/// let c = unsafe { char::from_u32_unchecked(0x2764) }; -/// -/// assert_eq!('❤', c); -/// ``` -#[inline] -#[stable(feature = "char_from_unchecked", since = "1.5.0")] -pub unsafe fn from_u32_unchecked(i: u32) -> char { - transmute(i) -} - -#[stable(feature = "char_convert", since = "1.13.0")] -impl From<char> for u32 { - #[inline] - fn from(c: char) -> Self { - c as u32 - } -} - -/// Maps a byte in 0x00...0xFF to a `char` whose code point has the same value, in U+0000 to U+00FF. -/// -/// Unicode is designed such that this effectively decodes bytes -/// with the character encoding that IANA calls ISO-8859-1. -/// This encoding is compatible with ASCII. -/// -/// Note that this is different from ISO/IEC 8859-1 a.k.a. ISO 8859-1 (with one less hyphen), -/// which leaves some "blanks", byte values that are not assigned to any character. -/// ISO-8859-1 (the IANA one) assigns them to the C0 and C1 control codes. -/// -/// Note that this is *also* different from Windows-1252 a.k.a. code page 1252, -/// which is a superset ISO/IEC 8859-1 that assigns some (not all!) blanks -/// to punctuation and various Latin characters. -/// -/// To confuse things further, [on the Web](https://encoding.spec.whatwg.org/) -/// `ascii`, `iso-8859-1`, and `windows-1252` are all aliases -/// for a superset of Windows-1252 that fills the remaining blanks with corresponding -/// C0 and C1 control codes. -#[stable(feature = "char_convert", since = "1.13.0")] -impl From<u8> for char { - #[inline] - fn from(i: u8) -> Self { - i as char - } -} - - -/// An error which can be returned when parsing a char. -#[stable(feature = "char_from_str", since = "1.20.0")] -#[derive(Clone, Debug)] -pub struct ParseCharError { - kind: CharErrorKind, -} - -impl ParseCharError { - #[unstable(feature = "char_error_internals", - reason = "this method should not be available publicly", - issue = "0")] - #[doc(hidden)] - pub fn __description(&self) -> &str { - match self.kind { - CharErrorKind::EmptyString => { - "cannot parse char from empty string" - }, - CharErrorKind::TooManyChars => "too many characters in string" - } - } -} - -#[derive(Copy, Clone, Debug, PartialEq, Eq)] -enum CharErrorKind { - EmptyString, - TooManyChars, -} - -#[stable(feature = "char_from_str", since = "1.20.0")] -impl fmt::Display for ParseCharError { - fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result { - self.__description().fmt(f) - } -} - - -#[stable(feature = "char_from_str", since = "1.20.0")] -impl FromStr for char { - type Err = ParseCharError; - - #[inline] - fn from_str(s: &str) -> Result<Self, Self::Err> { - let mut chars = s.chars(); - match (chars.next(), chars.next()) { - (None, _) => { - Err(ParseCharError { kind: CharErrorKind::EmptyString }) - }, - (Some(c), None) => Ok(c), - _ => { - Err(ParseCharError { kind: CharErrorKind::TooManyChars }) - } - } - } -} - - -#[unstable(feature = "try_from", issue = "33417")] -impl TryFrom<u32> for char { - type Error = CharTryFromError; - - #[inline] - fn try_from(i: u32) -> Result<Self, Self::Error> { - if (i > MAX as u32) || (i >= 0xD800 && i <= 0xDFFF) { - Err(CharTryFromError(())) - } else { - Ok(unsafe { from_u32_unchecked(i) }) - } - } -} - -/// The error type returned when a conversion from u32 to char fails. -#[unstable(feature = "try_from", issue = "33417")] -#[derive(Copy, Clone, Debug, PartialEq, Eq)] -pub struct CharTryFromError(()); - -#[unstable(feature = "try_from", issue = "33417")] -impl fmt::Display for CharTryFromError { - fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result { - "converted integer out of range for `char`".fmt(f) - } -} - -/// Converts a digit in the given radix to a `char`. -/// -/// A 'radix' here is sometimes also called a 'base'. A radix of two -/// indicates a binary number, a radix of ten, decimal, and a radix of -/// sixteen, hexadecimal, to give some common values. Arbitrary -/// radices are supported. -/// -/// `from_digit()` will return `None` if the input is not a digit in -/// the given radix. -/// -/// # Panics -/// -/// Panics if given a radix larger than 36. -/// -/// # Examples -/// -/// Basic usage: -/// -/// ``` -/// use std::char; -/// -/// let c = char::from_digit(4, 10); -/// -/// assert_eq!(Some('4'), c); -/// -/// // Decimal 11 is a single digit in base 16 -/// let c = char::from_digit(11, 16); -/// -/// assert_eq!(Some('b'), c); -/// ``` -/// -/// Returning `None` when the input is not a digit: -/// -/// ``` -/// use std::char; -/// -/// let c = char::from_digit(20, 10); -/// -/// assert_eq!(None, c); -/// ``` -/// -/// Passing a large radix, causing a panic: -/// -/// ``` -/// use std::thread; -/// use std::char; -/// -/// let result = thread::spawn(|| { -/// // this panics -/// let c = char::from_digit(1, 37); -/// }).join(); -/// -/// assert!(result.is_err()); -/// ``` -#[inline] -#[stable(feature = "rust1", since = "1.0.0")] -pub fn from_digit(num: u32, radix: u32) -> Option<char> { - if radix > 36 { - panic!("from_digit: radix is too high (maximum 36)"); - } - if num < radix { - let num = num as u8; - if num < 10 { - Some((b'0' + num) as char) - } else { - Some((b'a' + num - 10) as char) - } - } else { - None - } -} - -// NB: the stabilization and documentation for this trait is in -// unicode/char.rs, not here -#[allow(missing_docs)] // docs in libunicode/u_char.rs -#[doc(hidden)] -#[unstable(feature = "core_char_ext", - reason = "the stable interface is `impl char` in later crate", - issue = "32110")] -pub trait CharExt { - #[stable(feature = "core", since = "1.6.0")] - fn is_digit(self, radix: u32) -> bool; - #[stable(feature = "core", since = "1.6.0")] - fn to_digit(self, radix: u32) -> Option<u32>; - #[stable(feature = "core", since = "1.6.0")] - fn escape_unicode(self) -> EscapeUnicode; - #[stable(feature = "core", since = "1.6.0")] - fn escape_default(self) -> EscapeDefault; - #[stable(feature = "char_escape_debug", since = "1.20.0")] - fn escape_debug(self) -> EscapeDebug; - #[stable(feature = "core", since = "1.6.0")] - fn len_utf8(self) -> usize; - #[stable(feature = "core", since = "1.6.0")] - fn len_utf16(self) -> usize; - #[stable(feature = "unicode_encode_char", since = "1.15.0")] - fn encode_utf8(self, dst: &mut [u8]) -> &mut str; - #[stable(feature = "unicode_encode_char", since = "1.15.0")] - fn encode_utf16(self, dst: &mut [u16]) -> &mut [u16]; -} - -#[stable(feature = "core", since = "1.6.0")] -impl CharExt for char { - #[inline] - fn is_digit(self, radix: u32) -> bool { - self.to_digit(radix).is_some() - } - - #[inline] - fn to_digit(self, radix: u32) -> Option<u32> { - if radix > 36 { - panic!("to_digit: radix is too high (maximum 36)"); - } - let val = match self { - '0' ... '9' => self as u32 - '0' as u32, - 'a' ... 'z' => self as u32 - 'a' as u32 + 10, - 'A' ... 'Z' => self as u32 - 'A' as u32 + 10, - _ => return None, - }; - if val < radix { Some(val) } - else { None } - } - - #[inline] - fn escape_unicode(self) -> EscapeUnicode { - let c = self as u32; - - // or-ing 1 ensures that for c==0 the code computes that one - // digit should be printed and (which is the same) avoids the - // (31 - 32) underflow - let msb = 31 - (c | 1).leading_zeros(); - - // the index of the most significant hex digit - let ms_hex_digit = msb / 4; - EscapeUnicode { - c: self, - state: EscapeUnicodeState::Backslash, - hex_digit_idx: ms_hex_digit as usize, - } - } - - #[inline] - fn escape_default(self) -> EscapeDefault { - let init_state = match self { - '\t' => EscapeDefaultState::Backslash('t'), - '\r' => EscapeDefaultState::Backslash('r'), - '\n' => EscapeDefaultState::Backslash('n'), - '\\' | '\'' | '"' => EscapeDefaultState::Backslash(self), - '\x20' ... '\x7e' => EscapeDefaultState::Char(self), - _ => EscapeDefaultState::Unicode(self.escape_unicode()) - }; - EscapeDefault { state: init_state } - } - - #[inline] - fn escape_debug(self) -> EscapeDebug { - let init_state = match self { - '\t' => EscapeDefaultState::Backslash('t'), - '\r' => EscapeDefaultState::Backslash('r'), - '\n' => EscapeDefaultState::Backslash('n'), - '\\' | '\'' | '"' => EscapeDefaultState::Backslash(self), - c if is_printable(c) => EscapeDefaultState::Char(c), - c => EscapeDefaultState::Unicode(c.escape_unicode()), - }; - EscapeDebug(EscapeDefault { state: init_state }) - } - - #[inline] - fn len_utf8(self) -> usize { - let code = self as u32; - if code < MAX_ONE_B { - 1 - } else if code < MAX_TWO_B { - 2 - } else if code < MAX_THREE_B { - 3 - } else { - 4 - } - } - - #[inline] - fn len_utf16(self) -> usize { - let ch = self as u32; - if (ch & 0xFFFF) == ch { 1 } else { 2 } - } - - #[inline] - fn encode_utf8(self, dst: &mut [u8]) -> &mut str { - let code = self as u32; - unsafe { - let len = - if code < MAX_ONE_B && !dst.is_empty() { - *dst.get_unchecked_mut(0) = code as u8; - 1 - } else if code < MAX_TWO_B && dst.len() >= 2 { - *dst.get_unchecked_mut(0) = (code >> 6 & 0x1F) as u8 | TAG_TWO_B; - *dst.get_unchecked_mut(1) = (code & 0x3F) as u8 | TAG_CONT; - 2 - } else if code < MAX_THREE_B && dst.len() >= 3 { - *dst.get_unchecked_mut(0) = (code >> 12 & 0x0F) as u8 | TAG_THREE_B; - *dst.get_unchecked_mut(1) = (code >> 6 & 0x3F) as u8 | TAG_CONT; - *dst.get_unchecked_mut(2) = (code & 0x3F) as u8 | TAG_CONT; - 3 - } else if dst.len() >= 4 { - *dst.get_unchecked_mut(0) = (code >> 18 & 0x07) as u8 | TAG_FOUR_B; - *dst.get_unchecked_mut(1) = (code >> 12 & 0x3F) as u8 | TAG_CONT; - *dst.get_unchecked_mut(2) = (code >> 6 & 0x3F) as u8 | TAG_CONT; - *dst.get_unchecked_mut(3) = (code & 0x3F) as u8 | TAG_CONT; - 4 - } else { - panic!("encode_utf8: need {} bytes to encode U+{:X}, but the buffer has {}", - from_u32_unchecked(code).len_utf8(), - code, - dst.len()) - }; - from_utf8_unchecked_mut(dst.get_unchecked_mut(..len)) - } - } - - #[inline] - fn encode_utf16(self, dst: &mut [u16]) -> &mut [u16] { - let mut code = self as u32; - unsafe { - if (code & 0xFFFF) == code && !dst.is_empty() { - // The BMP falls through (assuming non-surrogate, as it should) - *dst.get_unchecked_mut(0) = code as u16; - slice::from_raw_parts_mut(dst.as_mut_ptr(), 1) - } else if dst.len() >= 2 { - // Supplementary planes break into surrogates. - code -= 0x1_0000; - *dst.get_unchecked_mut(0) = 0xD800 | ((code >> 10) as u16); - *dst.get_unchecked_mut(1) = 0xDC00 | ((code as u16) & 0x3FF); - slice::from_raw_parts_mut(dst.as_mut_ptr(), 2) - } else { - panic!("encode_utf16: need {} units to encode U+{:X}, but the buffer has {}", - from_u32_unchecked(code).len_utf16(), - code, - dst.len()) - } - } - } -} - -/// Returns an iterator that yields the hexadecimal Unicode escape of a -/// character, as `char`s. -/// -/// This `struct` is created by the [`escape_unicode`] method on [`char`]. See -/// its documentation for more. -/// -/// [`escape_unicode`]: ../../std/primitive.char.html#method.escape_unicode -/// [`char`]: ../../std/primitive.char.html -#[derive(Clone, Debug)] -#[stable(feature = "rust1", since = "1.0.0")] -pub struct EscapeUnicode { - c: char, - state: EscapeUnicodeState, - - // The index of the next hex digit to be printed (0 if none), - // i.e. the number of remaining hex digits to be printed; - // increasing from the least significant digit: 0x543210 - hex_digit_idx: usize, -} - -// The enum values are ordered so that their representation is the -// same as the remaining length (besides the hexadecimal digits). This -// likely makes `len()` a single load from memory) and inline-worth. -#[derive(Clone, Debug)] -enum EscapeUnicodeState { - Done, - RightBrace, - Value, - LeftBrace, - Type, - Backslash, -} - -#[stable(feature = "rust1", since = "1.0.0")] -impl Iterator for EscapeUnicode { - type Item = char; - - fn next(&mut self) -> Option<char> { - match self.state { - EscapeUnicodeState::Backslash => { - self.state = EscapeUnicodeState::Type; - Some('\\') - } - EscapeUnicodeState::Type => { - self.state = EscapeUnicodeState::LeftBrace; - Some('u') - } - EscapeUnicodeState::LeftBrace => { - self.state = EscapeUnicodeState::Value; - Some('{') - } - EscapeUnicodeState::Value => { - let hex_digit = ((self.c as u32) >> (self.hex_digit_idx * 4)) & 0xf; - let c = from_digit(hex_digit, 16).unwrap(); - if self.hex_digit_idx == 0 { - self.state = EscapeUnicodeState::RightBrace; - } else { - self.hex_digit_idx -= 1; - } - Some(c) - } - EscapeUnicodeState::RightBrace => { - self.state = EscapeUnicodeState::Done; - Some('}') - } - EscapeUnicodeState::Done => None, - } - } - - #[inline] - fn size_hint(&self) -> (usize, Option<usize>) { - let n = self.len(); - (n, Some(n)) - } - - #[inline] - fn count(self) -> usize { - self.len() - } - - fn last(self) -> Option<char> { - match self.state { - EscapeUnicodeState::Done => None, - - EscapeUnicodeState::RightBrace | - EscapeUnicodeState::Value | - EscapeUnicodeState::LeftBrace | - EscapeUnicodeState::Type | - EscapeUnicodeState::Backslash => Some('}'), - } - } -} - -#[stable(feature = "exact_size_escape", since = "1.11.0")] -impl ExactSizeIterator for EscapeUnicode { - #[inline] - fn len(&self) -> usize { - // The match is a single memory access with no branching - self.hex_digit_idx + match self.state { - EscapeUnicodeState::Done => 0, - EscapeUnicodeState::RightBrace => 1, - EscapeUnicodeState::Value => 2, - EscapeUnicodeState::LeftBrace => 3, - EscapeUnicodeState::Type => 4, - EscapeUnicodeState::Backslash => 5, - } - } -} - -#[unstable(feature = "fused", issue = "35602")] -impl FusedIterator for EscapeUnicode {} - -#[stable(feature = "char_struct_display", since = "1.16.0")] -impl fmt::Display for EscapeUnicode { - fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result { - for c in self.clone() { - f.write_char(c)?; - } - Ok(()) - } -} - -/// An iterator that yields the literal escape code of a `char`. -/// -/// This `struct` is created by the [`escape_default`] method on [`char`]. See -/// its documentation for more. -/// -/// [`escape_default`]: ../../std/primitive.char.html#method.escape_default -/// [`char`]: ../../std/primitive.char.html -#[derive(Clone, Debug)] -#[stable(feature = "rust1", since = "1.0.0")] -pub struct EscapeDefault { - state: EscapeDefaultState -} - -#[derive(Clone, Debug)] -enum EscapeDefaultState { - Done, - Char(char), - Backslash(char), - Unicode(EscapeUnicode), -} - -#[stable(feature = "rust1", since = "1.0.0")] -impl Iterator for EscapeDefault { - type Item = char; - - fn next(&mut self) -> Option<char> { - match self.state { - EscapeDefaultState::Backslash(c) => { - self.state = EscapeDefaultState::Char(c); - Some('\\') - } - EscapeDefaultState::Char(c) => { - self.state = EscapeDefaultState::Done; - Some(c) - } - EscapeDefaultState::Done => None, - EscapeDefaultState::Unicode(ref mut iter) => iter.next(), - } - } - - #[inline] - fn size_hint(&self) -> (usize, Option<usize>) { - let n = self.len(); - (n, Some(n)) - } - - #[inline] - fn count(self) -> usize { - self.len() - } - - fn nth(&mut self, n: usize) -> Option<char> { - match self.state { - EscapeDefaultState::Backslash(c) if n == 0 => { - self.state = EscapeDefaultState::Char(c); - Some('\\') - }, - EscapeDefaultState::Backslash(c) if n == 1 => { - self.state = EscapeDefaultState::Done; - Some(c) - }, - EscapeDefaultState::Backslash(_) => { - self.state = EscapeDefaultState::Done; - None - }, - EscapeDefaultState::Char(c) => { - self.state = EscapeDefaultState::Done; - - if n == 0 { - Some(c) - } else { - None - } - }, - EscapeDefaultState::Done => return None, - EscapeDefaultState::Unicode(ref mut i) => return i.nth(n), - } - } - - fn last(self) -> Option<char> { - match self.state { - EscapeDefaultState::Unicode(iter) => iter.last(), - EscapeDefaultState::Done => None, - EscapeDefaultState::Backslash(c) | EscapeDefaultState::Char(c) => Some(c), - } - } -} - -#[stable(feature = "exact_size_escape", since = "1.11.0")] -impl ExactSizeIterator for EscapeDefault { - fn len(&self) -> usize { - match self.state { - EscapeDefaultState::Done => 0, - EscapeDefaultState::Char(_) => 1, - EscapeDefaultState::Backslash(_) => 2, - EscapeDefaultState::Unicode(ref iter) => iter.len(), - } - } -} - -#[unstable(feature = "fused", issue = "35602")] -impl FusedIterator for EscapeDefault {} - -#[stable(feature = "char_struct_display", since = "1.16.0")] -impl fmt::Display for EscapeDefault { - fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result { - for c in self.clone() { - f.write_char(c)?; - } - Ok(()) - } -} - -/// An iterator that yields the literal escape code of a `char`. -/// -/// This `struct` is created by the [`escape_debug`] method on [`char`]. See its -/// documentation for more. -/// -/// [`escape_debug`]: ../../std/primitive.char.html#method.escape_debug -/// [`char`]: ../../std/primitive.char.html -#[stable(feature = "char_escape_debug", since = "1.20.0")] -#[derive(Clone, Debug)] -pub struct EscapeDebug(EscapeDefault); - -#[stable(feature = "char_escape_debug", since = "1.20.0")] -impl Iterator for EscapeDebug { - type Item = char; - fn next(&mut self) -> Option<char> { self.0.next() } - fn size_hint(&self) -> (usize, Option<usize>) { self.0.size_hint() } -} - -#[stable(feature = "char_escape_debug", since = "1.20.0")] -impl ExactSizeIterator for EscapeDebug { } - -#[unstable(feature = "fused", issue = "35602")] -impl FusedIterator for EscapeDebug {} - -#[stable(feature = "char_escape_debug", since = "1.20.0")] -impl fmt::Display for EscapeDebug { - fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result { - fmt::Display::fmt(&self.0, f) - } -} - - - -/// An iterator over an iterator of bytes of the characters the bytes represent -/// as UTF-8 -#[unstable(feature = "decode_utf8", issue = "33906")] -#[derive(Clone, Debug)] -pub struct DecodeUtf8<I: Iterator<Item = u8>>(::iter::Peekable<I>); - -/// Decodes an `Iterator` of bytes as UTF-8. -#[unstable(feature = "decode_utf8", issue = "33906")] -#[inline] -pub fn decode_utf8<I: IntoIterator<Item = u8>>(i: I) -> DecodeUtf8<I::IntoIter> { - DecodeUtf8(i.into_iter().peekable()) -} - -/// `<DecodeUtf8 as Iterator>::next` returns this for an invalid input sequence. -#[unstable(feature = "decode_utf8", issue = "33906")] -#[derive(PartialEq, Eq, Debug)] -pub struct InvalidSequence(()); - -#[unstable(feature = "decode_utf8", issue = "33906")] -impl<I: Iterator<Item = u8>> Iterator for DecodeUtf8<I> { - type Item = Result<char, InvalidSequence>; - #[inline] - - fn next(&mut self) -> Option<Result<char, InvalidSequence>> { - self.0.next().map(|first_byte| { - // Emit InvalidSequence according to - // Unicode §5.22 Best Practice for U+FFFD Substitution - // http://www.unicode.org/versions/Unicode9.0.0/ch05.pdf#G40630 - - // Roughly: consume at least one byte, - // then validate one byte at a time and stop before the first unexpected byte - // (which might be the valid start of the next byte sequence). - - let mut code_point; - macro_rules! first_byte { - ($mask: expr) => { - code_point = u32::from(first_byte & $mask) - } - } - macro_rules! continuation_byte { - () => { continuation_byte!(0x80...0xBF) }; - ($range: pat) => { - match self.0.peek() { - Some(&byte @ $range) => { - code_point = (code_point << 6) | u32::from(byte & 0b0011_1111); - self.0.next(); - } - _ => return Err(InvalidSequence(())) - } - } - } - - match first_byte { - 0x00...0x7F => { - first_byte!(0b1111_1111); - } - 0xC2...0xDF => { - first_byte!(0b0001_1111); - continuation_byte!(); - } - 0xE0 => { - first_byte!(0b0000_1111); - continuation_byte!(0xA0...0xBF); // 0x80...0x9F here are overlong - continuation_byte!(); - } - 0xE1...0xEC | 0xEE...0xEF => { - first_byte!(0b0000_1111); - continuation_byte!(); - continuation_byte!(); - } - 0xED => { - first_byte!(0b0000_1111); - continuation_byte!(0x80...0x9F); // 0xA0..0xBF here are surrogates - continuation_byte!(); - } - 0xF0 => { - first_byte!(0b0000_0111); - continuation_byte!(0x90...0xBF); // 0x80..0x8F here are overlong - continuation_byte!(); - continuation_byte!(); - } - 0xF1...0xF3 => { - first_byte!(0b0000_0111); - continuation_byte!(); - continuation_byte!(); - continuation_byte!(); - } - 0xF4 => { - first_byte!(0b0000_0111); - continuation_byte!(0x80...0x8F); // 0x90..0xBF here are beyond char::MAX - continuation_byte!(); - continuation_byte!(); - } - _ => return Err(InvalidSequence(())) // Illegal first byte, overlong, or beyond MAX - } - unsafe { - Ok(from_u32_unchecked(code_point)) - } - }) - } -} - -#[unstable(feature = "fused", issue = "35602")] -impl<I: FusedIterator<Item = u8>> FusedIterator for DecodeUtf8<I> {} diff --git a/src/libcore/char/convert.rs b/src/libcore/char/convert.rs new file mode 100644 index 00000000000..803a924eb3a --- /dev/null +++ b/src/libcore/char/convert.rs @@ -0,0 +1,304 @@ +// Copyright 2012-2014 The Rust Project Developers. See the COPYRIGHT +// file at the top-level directory of this distribution and at +// http://rust-lang.org/COPYRIGHT. +// +// Licensed under the Apache License, Version 2.0 <LICENSE-APACHE or +// http://www.apache.org/licenses/LICENSE-2.0> or the MIT license +// <LICENSE-MIT or http://opensource.org/licenses/MIT>, at your +// option. This file may not be copied, modified, or distributed +// except according to those terms. + +//! Character conversions. + +use convert::TryFrom; +use fmt; +use mem::transmute; +use str::FromStr; +use super::MAX; + +/// Converts a `u32` to a `char`. +/// +/// Note that all [`char`]s are valid [`u32`]s, and can be cast to one with +/// [`as`]: +/// +/// ``` +/// let c = '💯'; +/// let i = c as u32; +/// +/// assert_eq!(128175, i); +/// ``` +/// +/// However, the reverse is not true: not all valid [`u32`]s are valid +/// [`char`]s. `from_u32()` will return `None` if the input is not a valid value +/// for a [`char`]. +/// +/// [`char`]: ../../std/primitive.char.html +/// [`u32`]: ../../std/primitive.u32.html +/// [`as`]: ../../book/first-edition/casting-between-types.html#as +/// +/// For an unsafe version of this function which ignores these checks, see +/// [`from_u32_unchecked`]. +/// +/// [`from_u32_unchecked`]: fn.from_u32_unchecked.html +/// +/// # Examples +/// +/// Basic usage: +/// +/// ``` +/// use std::char; +/// +/// let c = char::from_u32(0x2764); +/// +/// assert_eq!(Some('❤'), c); +/// ``` +/// +/// Returning `None` when the input is not a valid [`char`]: +/// +/// ``` +/// use std::char; +/// +/// let c = char::from_u32(0x110000); +/// +/// assert_eq!(None, c); +/// ``` +#[inline] +#[stable(feature = "rust1", since = "1.0.0")] +pub fn from_u32(i: u32) -> Option<char> { + char::try_from(i).ok() +} + +/// Converts a `u32` to a `char`, ignoring validity. +/// +/// Note that all [`char`]s are valid [`u32`]s, and can be cast to one with +/// [`as`]: +/// +/// ``` +/// let c = '💯'; +/// let i = c as u32; +/// +/// assert_eq!(128175, i); +/// ``` +/// +/// However, the reverse is not true: not all valid [`u32`]s are valid +/// [`char`]s. `from_u32_unchecked()` will ignore this, and blindly cast to +/// [`char`], possibly creating an invalid one. +/// +/// [`char`]: ../../std/primitive.char.html +/// [`u32`]: ../../std/primitive.u32.html +/// [`as`]: ../../book/first-edition/casting-between-types.html#as +/// +/// # Safety +/// +/// This function is unsafe, as it may construct invalid `char` values. +/// +/// For a safe version of this function, see the [`from_u32`] function. +/// +/// [`from_u32`]: fn.from_u32.html +/// +/// # Examples +/// +/// Basic usage: +/// +/// ``` +/// use std::char; +/// +/// let c = unsafe { char::from_u32_unchecked(0x2764) }; +/// +/// assert_eq!('❤', c); +/// ``` +#[inline] +#[stable(feature = "char_from_unchecked", since = "1.5.0")] +pub unsafe fn from_u32_unchecked(i: u32) -> char { + transmute(i) +} + +#[stable(feature = "char_convert", since = "1.13.0")] +impl From<char> for u32 { + #[inline] + fn from(c: char) -> Self { + c as u32 + } +} + +/// Maps a byte in 0x00...0xFF to a `char` whose code point has the same value, in U+0000 to U+00FF. +/// +/// Unicode is designed such that this effectively decodes bytes +/// with the character encoding that IANA calls ISO-8859-1. +/// This encoding is compatible with ASCII. +/// +/// Note that this is different from ISO/IEC 8859-1 a.k.a. ISO 8859-1 (with one less hyphen), +/// which leaves some "blanks", byte values that are not assigned to any character. +/// ISO-8859-1 (the IANA one) assigns them to the C0 and C1 control codes. +/// +/// Note that this is *also* different from Windows-1252 a.k.a. code page 1252, +/// which is a superset ISO/IEC 8859-1 that assigns some (not all!) blanks +/// to punctuation and various Latin characters. +/// +/// To confuse things further, [on the Web](https://encoding.spec.whatwg.org/) +/// `ascii`, `iso-8859-1`, and `windows-1252` are all aliases +/// for a superset of Windows-1252 that fills the remaining blanks with corresponding +/// C0 and C1 control codes. +#[stable(feature = "char_convert", since = "1.13.0")] +impl From<u8> for char { + #[inline] + fn from(i: u8) -> Self { + i as char + } +} + + +/// An error which can be returned when parsing a char. +#[stable(feature = "char_from_str", since = "1.20.0")] +#[derive(Clone, Debug, PartialEq, Eq)] +pub struct ParseCharError { + kind: CharErrorKind, +} + +impl ParseCharError { + #[unstable(feature = "char_error_internals", + reason = "this method should not be available publicly", + issue = "0")] + #[doc(hidden)] + pub fn __description(&self) -> &str { + match self.kind { + CharErrorKind::EmptyString => { + "cannot parse char from empty string" + }, + CharErrorKind::TooManyChars => "too many characters in string" + } + } +} + +#[derive(Copy, Clone, Debug, PartialEq, Eq)] +enum CharErrorKind { + EmptyString, + TooManyChars, +} + +#[stable(feature = "char_from_str", since = "1.20.0")] +impl fmt::Display for ParseCharError { + fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result { + self.__description().fmt(f) + } +} + + +#[stable(feature = "char_from_str", since = "1.20.0")] +impl FromStr for char { + type Err = ParseCharError; + + #[inline] + fn from_str(s: &str) -> Result<Self, Self::Err> { + let mut chars = s.chars(); + match (chars.next(), chars.next()) { + (None, _) => { + Err(ParseCharError { kind: CharErrorKind::EmptyString }) + }, + (Some(c), None) => Ok(c), + _ => { + Err(ParseCharError { kind: CharErrorKind::TooManyChars }) + } + } + } +} + + +#[unstable(feature = "try_from", issue = "33417")] +impl TryFrom<u32> for char { + type Error = CharTryFromError; + + #[inline] + fn try_from(i: u32) -> Result<Self, Self::Error> { + if (i > MAX as u32) || (i >= 0xD800 && i <= 0xDFFF) { + Err(CharTryFromError(())) + } else { + Ok(unsafe { from_u32_unchecked(i) }) + } + } +} + +/// The error type returned when a conversion from u32 to char fails. +#[unstable(feature = "try_from", issue = "33417")] +#[derive(Copy, Clone, Debug, PartialEq, Eq)] +pub struct CharTryFromError(()); + +#[unstable(feature = "try_from", issue = "33417")] +impl fmt::Display for CharTryFromError { + fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result { + "converted integer out of range for `char`".fmt(f) + } +} + +/// Converts a digit in the given radix to a `char`. +/// +/// A 'radix' here is sometimes also called a 'base'. A radix of two +/// indicates a binary number, a radix of ten, decimal, and a radix of +/// sixteen, hexadecimal, to give some common values. Arbitrary +/// radices are supported. +/// +/// `from_digit()` will return `None` if the input is not a digit in +/// the given radix. +/// +/// # Panics +/// +/// Panics if given a radix larger than 36. +/// +/// # Examples +/// +/// Basic usage: +/// +/// ``` +/// use std::char; +/// +/// let c = char::from_digit(4, 10); +/// +/// assert_eq!(Some('4'), c); +/// +/// // Decimal 11 is a single digit in base 16 +/// let c = char::from_digit(11, 16); +/// +/// assert_eq!(Some('b'), c); +/// ``` +/// +/// Returning `None` when the input is not a digit: +/// +/// ``` +/// use std::char; +/// +/// let c = char::from_digit(20, 10); +/// +/// assert_eq!(None, c); +/// ``` +/// +/// Passing a large radix, causing a panic: +/// +/// ``` +/// use std::thread; +/// use std::char; +/// +/// let result = thread::spawn(|| { +/// // this panics +/// let c = char::from_digit(1, 37); +/// }).join(); +/// +/// assert!(result.is_err()); +/// ``` +#[inline] +#[stable(feature = "rust1", since = "1.0.0")] +pub fn from_digit(num: u32, radix: u32) -> Option<char> { + if radix > 36 { + panic!("from_digit: radix is too high (maximum 36)"); + } + if num < radix { + let num = num as u8; + if num < 10 { + Some((b'0' + num) as char) + } else { + Some((b'a' + num - 10) as char) + } + } else { + None + } +} + diff --git a/src/libcore/char/decode.rs b/src/libcore/char/decode.rs new file mode 100644 index 00000000000..cc52f048b89 --- /dev/null +++ b/src/libcore/char/decode.rs @@ -0,0 +1,143 @@ +// Copyright 2012-2014 The Rust Project Developers. See the COPYRIGHT +// file at the top-level directory of this distribution and at +// http://rust-lang.org/COPYRIGHT. +// +// Licensed under the Apache License, Version 2.0 <LICENSE-APACHE or +// http://www.apache.org/licenses/LICENSE-2.0> or the MIT license +// <LICENSE-MIT or http://opensource.org/licenses/MIT>, at your +// option. This file may not be copied, modified, or distributed +// except according to those terms. + +//! UTF-8 and UTF-16 decoding iterators + +use fmt; +use super::from_u32_unchecked; + +/// An iterator that decodes UTF-16 encoded code points from an iterator of `u16`s. +#[stable(feature = "decode_utf16", since = "1.9.0")] +#[derive(Clone, Debug)] +pub struct DecodeUtf16<I> + where I: Iterator<Item = u16> +{ + iter: I, + buf: Option<u16>, +} + +/// An error that can be returned when decoding UTF-16 code points. +#[stable(feature = "decode_utf16", since = "1.9.0")] +#[derive(Debug, Clone, Eq, PartialEq)] +pub struct DecodeUtf16Error { + code: u16, +} + +/// Create an iterator over the UTF-16 encoded code points in `iter`, +/// returning unpaired surrogates as `Err`s. +/// +/// # Examples +/// +/// Basic usage: +/// +/// ``` +/// use std::char::decode_utf16; +/// +/// fn main() { +/// // 𝄞mus<invalid>ic<invalid> +/// let v = [0xD834, 0xDD1E, 0x006d, 0x0075, +/// 0x0073, 0xDD1E, 0x0069, 0x0063, +/// 0xD834]; +/// +/// assert_eq!(decode_utf16(v.iter().cloned()) +/// .map(|r| r.map_err(|e| e.unpaired_surrogate())) +/// .collect::<Vec<_>>(), +/// vec![Ok('𝄞'), +/// Ok('m'), Ok('u'), Ok('s'), +/// Err(0xDD1E), +/// Ok('i'), Ok('c'), +/// Err(0xD834)]); +/// } +/// ``` +/// +/// A lossy decoder can be obtained by replacing `Err` results with the replacement character: +/// +/// ``` +/// use std::char::{decode_utf16, REPLACEMENT_CHARACTER}; +/// +/// fn main() { +/// // 𝄞mus<invalid>ic<invalid> +/// let v = [0xD834, 0xDD1E, 0x006d, 0x0075, +/// 0x0073, 0xDD1E, 0x0069, 0x0063, +/// 0xD834]; +/// +/// assert_eq!(decode_utf16(v.iter().cloned()) +/// .map(|r| r.unwrap_or(REPLACEMENT_CHARACTER)) +/// .collect::<String>(), +/// "𝄞mus�ic�"); +/// } +/// ``` +#[stable(feature = "decode_utf16", since = "1.9.0")] +#[inline] +pub fn decode_utf16<I: IntoIterator<Item = u16>>(iter: I) -> DecodeUtf16<I::IntoIter> { + DecodeUtf16 { + iter: iter.into_iter(), + buf: None, + } +} + +#[stable(feature = "decode_utf16", since = "1.9.0")] +impl<I: Iterator<Item = u16>> Iterator for DecodeUtf16<I> { + type Item = Result<char, DecodeUtf16Error>; + + fn next(&mut self) -> Option<Result<char, DecodeUtf16Error>> { + let u = match self.buf.take() { + Some(buf) => buf, + None => self.iter.next()? + }; + + if u < 0xD800 || 0xDFFF < u { + // not a surrogate + Some(Ok(unsafe { from_u32_unchecked(u as u32) })) + } else if u >= 0xDC00 { + // a trailing surrogate + Some(Err(DecodeUtf16Error { code: u })) + } else { + let u2 = match self.iter.next() { + Some(u2) => u2, + // eof + None => return Some(Err(DecodeUtf16Error { code: u })), + }; + if u2 < 0xDC00 || u2 > 0xDFFF { + // not a trailing surrogate so we're not a valid + // surrogate pair, so rewind to redecode u2 next time. + self.buf = Some(u2); + return Some(Err(DecodeUtf16Error { code: u })); + } + + // all ok, so lets decode it. + let c = (((u - 0xD800) as u32) << 10 | (u2 - 0xDC00) as u32) + 0x1_0000; + Some(Ok(unsafe { from_u32_unchecked(c) })) + } + } + + #[inline] + fn size_hint(&self) -> (usize, Option<usize>) { + let (low, high) = self.iter.size_hint(); + // we could be entirely valid surrogates (2 elements per + // char), or entirely non-surrogates (1 element per char) + (low / 2, high) + } +} + +impl DecodeUtf16Error { + /// Returns the unpaired surrogate which caused this error. + #[stable(feature = "decode_utf16", since = "1.9.0")] + pub fn unpaired_surrogate(&self) -> u16 { + self.code + } +} + +#[stable(feature = "decode_utf16", since = "1.9.0")] +impl fmt::Display for DecodeUtf16Error { + fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result { + write!(f, "unpaired surrogate found: {:x}", self.code) + } +} diff --git a/src/libcore/char/methods.rs b/src/libcore/char/methods.rs new file mode 100644 index 00000000000..64a17786b0a --- /dev/null +++ b/src/libcore/char/methods.rs @@ -0,0 +1,1395 @@ +// Copyright 2012-2014 The Rust Project Developers. See the COPYRIGHT +// file at the top-level directory of this distribution and at +// http://rust-lang.org/COPYRIGHT. +// +// Licensed under the Apache License, Version 2.0 <LICENSE-APACHE or +// http://www.apache.org/licenses/LICENSE-2.0> or the MIT license +// <LICENSE-MIT or http://opensource.org/licenses/MIT>, at your +// option. This file may not be copied, modified, or distributed +// except according to those terms. + +//! impl char {} + +use slice; +use str::from_utf8_unchecked_mut; +use super::*; +use unicode::printable::is_printable; +use unicode::tables::{conversions, derived_property, general_category, property}; + +#[lang = "char"] +impl char { + /// Checks if a `char` is a digit in the given radix. + /// + /// A 'radix' here is sometimes also called a 'base'. A radix of two + /// indicates a binary number, a radix of ten, decimal, and a radix of + /// sixteen, hexadecimal, to give some common values. Arbitrary + /// radices are supported. + /// + /// Compared to `is_numeric()`, this function only recognizes the characters + /// `0-9`, `a-z` and `A-Z`. + /// + /// 'Digit' is defined to be only the following characters: + /// + /// * `0-9` + /// * `a-z` + /// * `A-Z` + /// + /// For a more comprehensive understanding of 'digit', see [`is_numeric`][is_numeric]. + /// + /// [is_numeric]: #method.is_numeric + /// + /// # Panics + /// + /// Panics if given a radix larger than 36. + /// + /// # Examples + /// + /// Basic usage: + /// + /// ``` + /// assert!('1'.is_digit(10)); + /// assert!('f'.is_digit(16)); + /// assert!(!'f'.is_digit(10)); + /// ``` + /// + /// Passing a large radix, causing a panic: + /// + /// ``` + /// use std::thread; + /// + /// let result = thread::spawn(|| { + /// // this panics + /// '1'.is_digit(37); + /// }).join(); + /// + /// assert!(result.is_err()); + /// ``` + #[stable(feature = "rust1", since = "1.0.0")] + #[inline] + pub fn is_digit(self, radix: u32) -> bool { + self.to_digit(radix).is_some() + } + + /// Converts a `char` to a digit in the given radix. + /// + /// A 'radix' here is sometimes also called a 'base'. A radix of two + /// indicates a binary number, a radix of ten, decimal, and a radix of + /// sixteen, hexadecimal, to give some common values. Arbitrary + /// radices are supported. + /// + /// 'Digit' is defined to be only the following characters: + /// + /// * `0-9` + /// * `a-z` + /// * `A-Z` + /// + /// # Errors + /// + /// Returns `None` if the `char` does not refer to a digit in the given radix. + /// + /// # Panics + /// + /// Panics if given a radix larger than 36. + /// + /// # Examples + /// + /// Basic usage: + /// + /// ``` + /// assert_eq!('1'.to_digit(10), Some(1)); + /// assert_eq!('f'.to_digit(16), Some(15)); + /// ``` + /// + /// Passing a non-digit results in failure: + /// + /// ``` + /// assert_eq!('f'.to_digit(10), None); + /// assert_eq!('z'.to_digit(16), None); + /// ``` + /// + /// Passing a large radix, causing a panic: + /// + /// ``` + /// use std::thread; + /// + /// let result = thread::spawn(|| { + /// '1'.to_digit(37); + /// }).join(); + /// + /// assert!(result.is_err()); + /// ``` + #[stable(feature = "rust1", since = "1.0.0")] + #[inline] + pub fn to_digit(self, radix: u32) -> Option<u32> { + if radix > 36 { + panic!("to_digit: radix is too high (maximum 36)"); + } + let val = match self { + '0' ..= '9' => self as u32 - '0' as u32, + 'a' ..= 'z' => self as u32 - 'a' as u32 + 10, + 'A' ..= 'Z' => self as u32 - 'A' as u32 + 10, + _ => return None, + }; + if val < radix { Some(val) } + else { None } + } + + /// Returns an iterator that yields the hexadecimal Unicode escape of a + /// character as `char`s. + /// + /// This will escape characters with the Rust syntax of the form + /// `\u{NNNNNN}` where `NNNNNN` is a hexadecimal representation. + /// + /// # Examples + /// + /// As an iterator: + /// + /// ``` + /// for c in '❤'.escape_unicode() { + /// print!("{}", c); + /// } + /// println!(); + /// ``` + /// + /// Using `println!` directly: + /// + /// ``` + /// println!("{}", '❤'.escape_unicode()); + /// ``` + /// + /// Both are equivalent to: + /// + /// ``` + /// println!("\\u{{2764}}"); + /// ``` + /// + /// Using `to_string`: + /// + /// ``` + /// assert_eq!('❤'.escape_unicode().to_string(), "\\u{2764}"); + /// ``` + #[stable(feature = "rust1", since = "1.0.0")] + #[inline] + pub fn escape_unicode(self) -> EscapeUnicode { + let c = self as u32; + + // or-ing 1 ensures that for c==0 the code computes that one + // digit should be printed and (which is the same) avoids the + // (31 - 32) underflow + let msb = 31 - (c | 1).leading_zeros(); + + // the index of the most significant hex digit + let ms_hex_digit = msb / 4; + EscapeUnicode { + c: self, + state: EscapeUnicodeState::Backslash, + hex_digit_idx: ms_hex_digit as usize, + } + } + + /// An extended version of `escape_debug` that optionally permits escaping + /// Extended Grapheme codepoints. This allows us to format characters like + /// nonspacing marks better when they're at the start of a string. + #[doc(hidden)] + #[unstable(feature = "str_internals", issue = "0")] + #[inline] + pub fn escape_debug_ext(self, escape_grapheme_extended: bool) -> EscapeDebug { + let init_state = match self { + '\t' => EscapeDefaultState::Backslash('t'), + '\r' => EscapeDefaultState::Backslash('r'), + '\n' => EscapeDefaultState::Backslash('n'), + '\\' | '\'' | '"' => EscapeDefaultState::Backslash(self), + _ if escape_grapheme_extended && self.is_grapheme_extended() => { + EscapeDefaultState::Unicode(self.escape_unicode()) + } + _ if is_printable(self) => EscapeDefaultState::Char(self), + _ => EscapeDefaultState::Unicode(self.escape_unicode()), + }; + EscapeDebug(EscapeDefault { state: init_state }) + } + + /// Returns an iterator that yields the literal escape code of a character + /// as `char`s. + /// + /// This will escape the characters similar to the `Debug` implementations + /// of `str` or `char`. + /// + /// # Examples + /// + /// As an iterator: + /// + /// ``` + /// for c in '\n'.escape_debug() { + /// print!("{}", c); + /// } + /// println!(); + /// ``` + /// + /// Using `println!` directly: + /// + /// ``` + /// println!("{}", '\n'.escape_debug()); + /// ``` + /// + /// Both are equivalent to: + /// + /// ``` + /// println!("\\n"); + /// ``` + /// + /// Using `to_string`: + /// + /// ``` + /// assert_eq!('\n'.escape_debug().to_string(), "\\n"); + /// ``` + #[stable(feature = "char_escape_debug", since = "1.20.0")] + #[inline] + pub fn escape_debug(self) -> EscapeDebug { + self.escape_debug_ext(true) + } + + /// Returns an iterator that yields the literal escape code of a character + /// as `char`s. + /// + /// The default is chosen with a bias toward producing literals that are + /// legal in a variety of languages, including C++11 and similar C-family + /// languages. The exact rules are: + /// + /// * Tab is escaped as `\t`. + /// * Carriage return is escaped as `\r`. + /// * Line feed is escaped as `\n`. + /// * Single quote is escaped as `\'`. + /// * Double quote is escaped as `\"`. + /// * Backslash is escaped as `\\`. + /// * Any character in the 'printable ASCII' range `0x20` .. `0x7e` + /// inclusive is not escaped. + /// * All other characters are given hexadecimal Unicode escapes; see + /// [`escape_unicode`][escape_unicode]. + /// + /// [escape_unicode]: #method.escape_unicode + /// + /// # Examples + /// + /// As an iterator: + /// + /// ``` + /// for c in '"'.escape_default() { + /// print!("{}", c); + /// } + /// println!(); + /// ``` + /// + /// Using `println!` directly: + /// + /// ``` + /// println!("{}", '"'.escape_default()); + /// ``` + /// + /// + /// Both are equivalent to: + /// + /// ``` + /// println!("\\\""); + /// ``` + /// + /// Using `to_string`: + /// + /// ``` + /// assert_eq!('"'.escape_default().to_string(), "\\\""); + /// ``` + #[stable(feature = "rust1", since = "1.0.0")] + #[inline] + pub fn escape_default(self) -> EscapeDefault { + let init_state = match self { + '\t' => EscapeDefaultState::Backslash('t'), + '\r' => EscapeDefaultState::Backslash('r'), + '\n' => EscapeDefaultState::Backslash('n'), + '\\' | '\'' | '"' => EscapeDefaultState::Backslash(self), + '\x20' ..= '\x7e' => EscapeDefaultState::Char(self), + _ => EscapeDefaultState::Unicode(self.escape_unicode()) + }; + EscapeDefault { state: init_state } + } + + /// Returns the number of bytes this `char` would need if encoded in UTF-8. + /// + /// That number of bytes is always between 1 and 4, inclusive. + /// + /// # Examples + /// + /// Basic usage: + /// + /// ``` + /// let len = 'A'.len_utf8(); + /// assert_eq!(len, 1); + /// + /// let len = 'ß'.len_utf8(); + /// assert_eq!(len, 2); + /// + /// let len = 'ℝ'.len_utf8(); + /// assert_eq!(len, 3); + /// + /// let len = '💣'.len_utf8(); + /// assert_eq!(len, 4); + /// ``` + /// + /// The `&str` type guarantees that its contents are UTF-8, and so we can compare the length it + /// would take if each code point was represented as a `char` vs in the `&str` itself: + /// + /// ``` + /// // as chars + /// let eastern = '東'; + /// let capitol = '京'; + /// + /// // both can be represented as three bytes + /// assert_eq!(3, eastern.len_utf8()); + /// assert_eq!(3, capitol.len_utf8()); + /// + /// // as a &str, these two are encoded in UTF-8 + /// let tokyo = "東京"; + /// + /// let len = eastern.len_utf8() + capitol.len_utf8(); + /// + /// // we can see that they take six bytes total... + /// assert_eq!(6, tokyo.len()); + /// + /// // ... just like the &str + /// assert_eq!(len, tokyo.len()); + /// ``` + #[stable(feature = "rust1", since = "1.0.0")] + #[inline] + pub fn len_utf8(self) -> usize { + let code = self as u32; + if code < MAX_ONE_B { + 1 + } else if code < MAX_TWO_B { + 2 + } else if code < MAX_THREE_B { + 3 + } else { + 4 + } + } + + /// Returns the number of 16-bit code units this `char` would need if + /// encoded in UTF-16. + /// + /// See the documentation for [`len_utf8`] for more explanation of this + /// concept. This function is a mirror, but for UTF-16 instead of UTF-8. + /// + /// [`len_utf8`]: #method.len_utf8 + /// + /// # Examples + /// + /// Basic usage: + /// + /// ``` + /// let n = 'ß'.len_utf16(); + /// assert_eq!(n, 1); + /// + /// let len = '💣'.len_utf16(); + /// assert_eq!(len, 2); + /// ``` + #[stable(feature = "rust1", since = "1.0.0")] + #[inline] + pub fn len_utf16(self) -> usize { + let ch = self as u32; + if (ch & 0xFFFF) == ch { 1 } else { 2 } + } + + /// Encodes this character as UTF-8 into the provided byte buffer, + /// and then returns the subslice of the buffer that contains the encoded character. + /// + /// # Panics + /// + /// Panics if the buffer is not large enough. + /// A buffer of length four is large enough to encode any `char`. + /// + /// # Examples + /// + /// In both of these examples, 'ß' takes two bytes to encode. + /// + /// ``` + /// let mut b = [0; 2]; + /// + /// let result = 'ß'.encode_utf8(&mut b); + /// + /// assert_eq!(result, "ß"); + /// + /// assert_eq!(result.len(), 2); + /// ``` + /// + /// A buffer that's too small: + /// + /// ``` + /// use std::thread; + /// + /// let result = thread::spawn(|| { + /// let mut b = [0; 1]; + /// + /// // this panics + /// 'ß'.encode_utf8(&mut b); + /// }).join(); + /// + /// assert!(result.is_err()); + /// ``` + #[stable(feature = "unicode_encode_char", since = "1.15.0")] + #[inline] + pub fn encode_utf8(self, dst: &mut [u8]) -> &mut str { + let code = self as u32; + unsafe { + let len = + if code < MAX_ONE_B && !dst.is_empty() { + *dst.get_unchecked_mut(0) = code as u8; + 1 + } else if code < MAX_TWO_B && dst.len() >= 2 { + *dst.get_unchecked_mut(0) = (code >> 6 & 0x1F) as u8 | TAG_TWO_B; + *dst.get_unchecked_mut(1) = (code & 0x3F) as u8 | TAG_CONT; + 2 + } else if code < MAX_THREE_B && dst.len() >= 3 { + *dst.get_unchecked_mut(0) = (code >> 12 & 0x0F) as u8 | TAG_THREE_B; + *dst.get_unchecked_mut(1) = (code >> 6 & 0x3F) as u8 | TAG_CONT; + *dst.get_unchecked_mut(2) = (code & 0x3F) as u8 | TAG_CONT; + 3 + } else if dst.len() >= 4 { + *dst.get_unchecked_mut(0) = (code >> 18 & 0x07) as u8 | TAG_FOUR_B; + *dst.get_unchecked_mut(1) = (code >> 12 & 0x3F) as u8 | TAG_CONT; + *dst.get_unchecked_mut(2) = (code >> 6 & 0x3F) as u8 | TAG_CONT; + *dst.get_unchecked_mut(3) = (code & 0x3F) as u8 | TAG_CONT; + 4 + } else { + panic!("encode_utf8: need {} bytes to encode U+{:X}, but the buffer has {}", + from_u32_unchecked(code).len_utf8(), + code, + dst.len()) + }; + from_utf8_unchecked_mut(dst.get_unchecked_mut(..len)) + } + } + + /// Encodes this character as UTF-16 into the provided `u16` buffer, + /// and then returns the subslice of the buffer that contains the encoded character. + /// + /// # Panics + /// + /// Panics if the buffer is not large enough. + /// A buffer of length 2 is large enough to encode any `char`. + /// + /// # Examples + /// + /// In both of these examples, '𝕊' takes two `u16`s to encode. + /// + /// ``` + /// let mut b = [0; 2]; + /// + /// let result = '𝕊'.encode_utf16(&mut b); + /// + /// assert_eq!(result.len(), 2); + /// ``` + /// + /// A buffer that's too small: + /// + /// ``` + /// use std::thread; + /// + /// let result = thread::spawn(|| { + /// let mut b = [0; 1]; + /// + /// // this panics + /// '𝕊'.encode_utf16(&mut b); + /// }).join(); + /// + /// assert!(result.is_err()); + /// ``` + #[stable(feature = "unicode_encode_char", since = "1.15.0")] + #[inline] + pub fn encode_utf16(self, dst: &mut [u16]) -> &mut [u16] { + let mut code = self as u32; + unsafe { + if (code & 0xFFFF) == code && !dst.is_empty() { + // The BMP falls through (assuming non-surrogate, as it should) + *dst.get_unchecked_mut(0) = code as u16; + slice::from_raw_parts_mut(dst.as_mut_ptr(), 1) + } else if dst.len() >= 2 { + // Supplementary planes break into surrogates. + code -= 0x1_0000; + *dst.get_unchecked_mut(0) = 0xD800 | ((code >> 10) as u16); + *dst.get_unchecked_mut(1) = 0xDC00 | ((code as u16) & 0x3FF); + slice::from_raw_parts_mut(dst.as_mut_ptr(), 2) + } else { + panic!("encode_utf16: need {} units to encode U+{:X}, but the buffer has {}", + from_u32_unchecked(code).len_utf16(), + code, + dst.len()) + } + } + } + + /// Returns true if this `char` is an alphabetic code point, and false if not. + /// + /// # Examples + /// + /// Basic usage: + /// + /// ``` + /// assert!('a'.is_alphabetic()); + /// assert!('京'.is_alphabetic()); + /// + /// let c = '💝'; + /// // love is many things, but it is not alphabetic + /// assert!(!c.is_alphabetic()); + /// ``` + #[stable(feature = "rust1", since = "1.0.0")] + #[inline] + pub fn is_alphabetic(self) -> bool { + match self { + 'a'..='z' | 'A'..='Z' => true, + c if c > '\x7f' => derived_property::Alphabetic(c), + _ => false, + } + } + + /// Returns true if this `char` satisfies the 'XID_Start' Unicode property, and false + /// otherwise. + /// + /// 'XID_Start' is a Unicode Derived Property specified in + /// [UAX #31](http://unicode.org/reports/tr31/#NFKC_Modifications), + /// mostly similar to `ID_Start` but modified for closure under `NFKx`. + #[unstable(feature = "rustc_private", + reason = "mainly needed for compiler internals", + issue = "27812")] + #[inline] + pub fn is_xid_start(self) -> bool { + derived_property::XID_Start(self) + } + + /// Returns true if this `char` satisfies the 'XID_Continue' Unicode property, and false + /// otherwise. + /// + /// 'XID_Continue' is a Unicode Derived Property specified in + /// [UAX #31](http://unicode.org/reports/tr31/#NFKC_Modifications), + /// mostly similar to 'ID_Continue' but modified for closure under NFKx. + #[unstable(feature = "rustc_private", + reason = "mainly needed for compiler internals", + issue = "27812")] + #[inline] + pub fn is_xid_continue(self) -> bool { + derived_property::XID_Continue(self) + } + + /// Returns true if this `char` is lowercase, and false otherwise. + /// + /// 'Lowercase' is defined according to the terms of the Unicode Derived Core + /// Property `Lowercase`. + /// + /// # Examples + /// + /// Basic usage: + /// + /// ``` + /// assert!('a'.is_lowercase()); + /// assert!('δ'.is_lowercase()); + /// assert!(!'A'.is_lowercase()); + /// assert!(!'Δ'.is_lowercase()); + /// + /// // The various Chinese scripts do not have case, and so: + /// assert!(!'中'.is_lowercase()); + /// ``` + #[stable(feature = "rust1", since = "1.0.0")] + #[inline] + pub fn is_lowercase(self) -> bool { + match self { + 'a'..='z' => true, + c if c > '\x7f' => derived_property::Lowercase(c), + _ => false, + } + } + + /// Returns true if this `char` is uppercase, and false otherwise. + /// + /// 'Uppercase' is defined according to the terms of the Unicode Derived Core + /// Property `Uppercase`. + /// + /// # Examples + /// + /// Basic usage: + /// + /// ``` + /// assert!(!'a'.is_uppercase()); + /// assert!(!'δ'.is_uppercase()); + /// assert!('A'.is_uppercase()); + /// assert!('Δ'.is_uppercase()); + /// + /// // The various Chinese scripts do not have case, and so: + /// assert!(!'中'.is_uppercase()); + /// ``` + #[stable(feature = "rust1", since = "1.0.0")] + #[inline] + pub fn is_uppercase(self) -> bool { + match self { + 'A'..='Z' => true, + c if c > '\x7f' => derived_property::Uppercase(c), + _ => false, + } + } + + /// Returns true if this `char` is whitespace, and false otherwise. + /// + /// 'Whitespace' is defined according to the terms of the Unicode Derived Core + /// Property `White_Space`. + /// + /// # Examples + /// + /// Basic usage: + /// + /// ``` + /// assert!(' '.is_whitespace()); + /// + /// // a non-breaking space + /// assert!('\u{A0}'.is_whitespace()); + /// + /// assert!(!'越'.is_whitespace()); + /// ``` + #[stable(feature = "rust1", since = "1.0.0")] + #[inline] + pub fn is_whitespace(self) -> bool { + match self { + ' ' | '\x09'..='\x0d' => true, + c if c > '\x7f' => property::White_Space(c), + _ => false, + } + } + + /// Returns true if this `char` is alphanumeric, and false otherwise. + /// + /// 'Alphanumeric'-ness is defined in terms of the Unicode General Categories + /// 'Nd', 'Nl', 'No' and the Derived Core Property 'Alphabetic'. + /// + /// # Examples + /// + /// Basic usage: + /// + /// ``` + /// assert!('٣'.is_alphanumeric()); + /// assert!('7'.is_alphanumeric()); + /// assert!('৬'.is_alphanumeric()); + /// assert!('¾'.is_alphanumeric()); + /// assert!('①'.is_alphanumeric()); + /// assert!('K'.is_alphanumeric()); + /// assert!('و'.is_alphanumeric()); + /// assert!('藏'.is_alphanumeric()); + /// ``` + #[stable(feature = "rust1", since = "1.0.0")] + #[inline] + pub fn is_alphanumeric(self) -> bool { + self.is_alphabetic() || self.is_numeric() + } + + /// Returns true if this `char` is a control code point, and false otherwise. + /// + /// 'Control code point' is defined in terms of the Unicode General + /// Category `Cc`. + /// + /// # Examples + /// + /// Basic usage: + /// + /// ``` + /// // U+009C, STRING TERMINATOR + /// assert!(''.is_control()); + /// assert!(!'q'.is_control()); + /// ``` + #[stable(feature = "rust1", since = "1.0.0")] + #[inline] + pub fn is_control(self) -> bool { + general_category::Cc(self) + } + + /// Returns true if this `char` is an extended grapheme character, and false otherwise. + /// + /// 'Extended grapheme character' is defined in terms of the Unicode Shaping and Rendering + /// Category `Grapheme_Extend`. + #[inline] + pub(crate) fn is_grapheme_extended(self) -> bool { + derived_property::Grapheme_Extend(self) + } + + /// Returns true if this `char` is numeric, and false otherwise. + /// + /// 'Numeric'-ness is defined in terms of the Unicode General Categories + /// 'Nd', 'Nl', 'No'. + /// + /// # Examples + /// + /// Basic usage: + /// + /// ``` + /// assert!('٣'.is_numeric()); + /// assert!('7'.is_numeric()); + /// assert!('৬'.is_numeric()); + /// assert!('¾'.is_numeric()); + /// assert!('①'.is_numeric()); + /// assert!(!'K'.is_numeric()); + /// assert!(!'و'.is_numeric()); + /// assert!(!'藏'.is_numeric()); + /// ``` + #[stable(feature = "rust1", since = "1.0.0")] + #[inline] + pub fn is_numeric(self) -> bool { + match self { + '0'..='9' => true, + c if c > '\x7f' => general_category::N(c), + _ => false, + } + } + + /// Returns an iterator that yields the lowercase equivalent of a `char` + /// as one or more `char`s. + /// + /// If a character does not have a lowercase equivalent, the same character + /// will be returned back by the iterator. + /// + /// This performs complex unconditional mappings with no tailoring: it maps + /// one Unicode character to its lowercase equivalent according to the + /// [Unicode database] and the additional complex mappings + /// [`SpecialCasing.txt`]. Conditional mappings (based on context or + /// language) are not considered here. + /// + /// For a full reference, see [here][reference]. + /// + /// [Unicode database]: ftp://ftp.unicode.org/Public/UNIDATA/UnicodeData.txt + /// + /// [`SpecialCasing.txt`]: ftp://ftp.unicode.org/Public/UNIDATA/SpecialCasing.txt + /// + /// [reference]: http://www.unicode.org/versions/Unicode7.0.0/ch03.pdf#G33992 + /// + /// # Examples + /// + /// As an iterator: + /// + /// ``` + /// for c in 'İ'.to_lowercase() { + /// print!("{}", c); + /// } + /// println!(); + /// ``` + /// + /// Using `println!` directly: + /// + /// ``` + /// println!("{}", 'İ'.to_lowercase()); + /// ``` + /// + /// Both are equivalent to: + /// + /// ``` + /// println!("i\u{307}"); + /// ``` + /// + /// Using `to_string`: + /// + /// ``` + /// assert_eq!('C'.to_lowercase().to_string(), "c"); + /// + /// // Sometimes the result is more than one character: + /// assert_eq!('İ'.to_lowercase().to_string(), "i\u{307}"); + /// + /// // Characters that do not have both uppercase and lowercase + /// // convert into themselves. + /// assert_eq!('山'.to_lowercase().to_string(), "山"); + /// ``` + #[stable(feature = "rust1", since = "1.0.0")] + #[inline] + pub fn to_lowercase(self) -> ToLowercase { + ToLowercase(CaseMappingIter::new(conversions::to_lower(self))) + } + + /// Returns an iterator that yields the uppercase equivalent of a `char` + /// as one or more `char`s. + /// + /// If a character does not have an uppercase equivalent, the same character + /// will be returned back by the iterator. + /// + /// This performs complex unconditional mappings with no tailoring: it maps + /// one Unicode character to its uppercase equivalent according to the + /// [Unicode database] and the additional complex mappings + /// [`SpecialCasing.txt`]. Conditional mappings (based on context or + /// language) are not considered here. + /// + /// For a full reference, see [here][reference]. + /// + /// [Unicode database]: ftp://ftp.unicode.org/Public/UNIDATA/UnicodeData.txt + /// + /// [`SpecialCasing.txt`]: ftp://ftp.unicode.org/Public/UNIDATA/SpecialCasing.txt + /// + /// [reference]: http://www.unicode.org/versions/Unicode7.0.0/ch03.pdf#G33992 + /// + /// # Examples + /// + /// As an iterator: + /// + /// ``` + /// for c in 'ß'.to_uppercase() { + /// print!("{}", c); + /// } + /// println!(); + /// ``` + /// + /// Using `println!` directly: + /// + /// ``` + /// println!("{}", 'ß'.to_uppercase()); + /// ``` + /// + /// Both are equivalent to: + /// + /// ``` + /// println!("SS"); + /// ``` + /// + /// Using `to_string`: + /// + /// ``` + /// assert_eq!('c'.to_uppercase().to_string(), "C"); + /// + /// // Sometimes the result is more than one character: + /// assert_eq!('ß'.to_uppercase().to_string(), "SS"); + /// + /// // Characters that do not have both uppercase and lowercase + /// // convert into themselves. + /// assert_eq!('山'.to_uppercase().to_string(), "山"); + /// ``` + /// + /// # Note on locale + /// + /// In Turkish, the equivalent of 'i' in Latin has five forms instead of two: + /// + /// * 'Dotless': I / ı, sometimes written ï + /// * 'Dotted': İ / i + /// + /// Note that the lowercase dotted 'i' is the same as the Latin. Therefore: + /// + /// ``` + /// let upper_i = 'i'.to_uppercase().to_string(); + /// ``` + /// + /// The value of `upper_i` here relies on the language of the text: if we're + /// in `en-US`, it should be `"I"`, but if we're in `tr_TR`, it should + /// be `"İ"`. `to_uppercase()` does not take this into account, and so: + /// + /// ``` + /// let upper_i = 'i'.to_uppercase().to_string(); + /// + /// assert_eq!(upper_i, "I"); + /// ``` + /// + /// holds across languages. + #[stable(feature = "rust1", since = "1.0.0")] + #[inline] + pub fn to_uppercase(self) -> ToUppercase { + ToUppercase(CaseMappingIter::new(conversions::to_upper(self))) + } + + /// Checks if the value is within the ASCII range. + /// + /// # Examples + /// + /// ``` + /// let ascii = 'a'; + /// let non_ascii = '❤'; + /// + /// assert!(ascii.is_ascii()); + /// assert!(!non_ascii.is_ascii()); + /// ``` + #[stable(feature = "ascii_methods_on_intrinsics", since = "1.23.0")] + #[inline] + pub fn is_ascii(&self) -> bool { + *self as u32 <= 0x7F + } + + /// Makes a copy of the value in its ASCII upper case equivalent. + /// + /// ASCII letters 'a' to 'z' are mapped to 'A' to 'Z', + /// but non-ASCII letters are unchanged. + /// + /// To uppercase the value in-place, use [`make_ascii_uppercase`]. + /// + /// To uppercase ASCII characters in addition to non-ASCII characters, use + /// [`to_uppercase`]. + /// + /// # Examples + /// + /// ``` + /// let ascii = 'a'; + /// let non_ascii = '❤'; + /// + /// assert_eq!('A', ascii.to_ascii_uppercase()); + /// assert_eq!('❤', non_ascii.to_ascii_uppercase()); + /// ``` + /// + /// [`make_ascii_uppercase`]: #method.make_ascii_uppercase + /// [`to_uppercase`]: #method.to_uppercase + #[stable(feature = "ascii_methods_on_intrinsics", since = "1.23.0")] + #[inline] + pub fn to_ascii_uppercase(&self) -> char { + if self.is_ascii() { + (*self as u8).to_ascii_uppercase() as char + } else { + *self + } + } + + /// Makes a copy of the value in its ASCII lower case equivalent. + /// + /// ASCII letters 'A' to 'Z' are mapped to 'a' to 'z', + /// but non-ASCII letters are unchanged. + /// + /// To lowercase the value in-place, use [`make_ascii_lowercase`]. + /// + /// To lowercase ASCII characters in addition to non-ASCII characters, use + /// [`to_lowercase`]. + /// + /// # Examples + /// + /// ``` + /// let ascii = 'A'; + /// let non_ascii = '❤'; + /// + /// assert_eq!('a', ascii.to_ascii_lowercase()); + /// assert_eq!('❤', non_ascii.to_ascii_lowercase()); + /// ``` + /// + /// [`make_ascii_lowercase`]: #method.make_ascii_lowercase + /// [`to_lowercase`]: #method.to_lowercase + #[stable(feature = "ascii_methods_on_intrinsics", since = "1.23.0")] + #[inline] + pub fn to_ascii_lowercase(&self) -> char { + if self.is_ascii() { + (*self as u8).to_ascii_lowercase() as char + } else { + *self + } + } + + /// Checks that two values are an ASCII case-insensitive match. + /// + /// Equivalent to `to_ascii_lowercase(a) == to_ascii_lowercase(b)`. + /// + /// # Examples + /// + /// ``` + /// let upper_a = 'A'; + /// let lower_a = 'a'; + /// let lower_z = 'z'; + /// + /// assert!(upper_a.eq_ignore_ascii_case(&lower_a)); + /// assert!(upper_a.eq_ignore_ascii_case(&upper_a)); + /// assert!(!upper_a.eq_ignore_ascii_case(&lower_z)); + /// ``` + #[stable(feature = "ascii_methods_on_intrinsics", since = "1.23.0")] + #[inline] + pub fn eq_ignore_ascii_case(&self, other: &char) -> bool { + self.to_ascii_lowercase() == other.to_ascii_lowercase() + } + + /// Converts this type to its ASCII upper case equivalent in-place. + /// + /// ASCII letters 'a' to 'z' are mapped to 'A' to 'Z', + /// but non-ASCII letters are unchanged. + /// + /// To return a new uppercased value without modifying the existing one, use + /// [`to_ascii_uppercase`]. + /// + /// # Examples + /// + /// ``` + /// let mut ascii = 'a'; + /// + /// ascii.make_ascii_uppercase(); + /// + /// assert_eq!('A', ascii); + /// ``` + /// + /// [`to_ascii_uppercase`]: #method.to_ascii_uppercase + #[stable(feature = "ascii_methods_on_intrinsics", since = "1.23.0")] + #[inline] + pub fn make_ascii_uppercase(&mut self) { + *self = self.to_ascii_uppercase(); + } + + /// Converts this type to its ASCII lower case equivalent in-place. + /// + /// ASCII letters 'A' to 'Z' are mapped to 'a' to 'z', + /// but non-ASCII letters are unchanged. + /// + /// To return a new lowercased value without modifying the existing one, use + /// [`to_ascii_lowercase`]. + /// + /// # Examples + /// + /// ``` + /// let mut ascii = 'A'; + /// + /// ascii.make_ascii_lowercase(); + /// + /// assert_eq!('a', ascii); + /// ``` + /// + /// [`to_ascii_lowercase`]: #method.to_ascii_lowercase + #[stable(feature = "ascii_methods_on_intrinsics", since = "1.23.0")] + #[inline] + pub fn make_ascii_lowercase(&mut self) { + *self = self.to_ascii_lowercase(); + } + + /// Checks if the value is an ASCII alphabetic character: + /// + /// - U+0041 'A' ... U+005A 'Z', or + /// - U+0061 'a' ... U+007A 'z'. + /// + /// # Examples + /// + /// ``` + /// let uppercase_a = 'A'; + /// let uppercase_g = 'G'; + /// let a = 'a'; + /// let g = 'g'; + /// let zero = '0'; + /// let percent = '%'; + /// let space = ' '; + /// let lf = '\n'; + /// let esc: char = 0x1b_u8.into(); + /// + /// assert!(uppercase_a.is_ascii_alphabetic()); + /// assert!(uppercase_g.is_ascii_alphabetic()); + /// assert!(a.is_ascii_alphabetic()); + /// assert!(g.is_ascii_alphabetic()); + /// assert!(!zero.is_ascii_alphabetic()); + /// assert!(!percent.is_ascii_alphabetic()); + /// assert!(!space.is_ascii_alphabetic()); + /// assert!(!lf.is_ascii_alphabetic()); + /// assert!(!esc.is_ascii_alphabetic()); + /// ``` + #[stable(feature = "ascii_ctype_on_intrinsics", since = "1.24.0")] + #[inline] + pub fn is_ascii_alphabetic(&self) -> bool { + self.is_ascii() && (*self as u8).is_ascii_alphabetic() + } + + /// Checks if the value is an ASCII uppercase character: + /// U+0041 'A' ... U+005A 'Z'. + /// + /// # Examples + /// + /// ``` + /// let uppercase_a = 'A'; + /// let uppercase_g = 'G'; + /// let a = 'a'; + /// let g = 'g'; + /// let zero = '0'; + /// let percent = '%'; + /// let space = ' '; + /// let lf = '\n'; + /// let esc: char = 0x1b_u8.into(); + /// + /// assert!(uppercase_a.is_ascii_uppercase()); + /// assert!(uppercase_g.is_ascii_uppercase()); + /// assert!(!a.is_ascii_uppercase()); + /// assert!(!g.is_ascii_uppercase()); + /// assert!(!zero.is_ascii_uppercase()); + /// assert!(!percent.is_ascii_uppercase()); + /// assert!(!space.is_ascii_uppercase()); + /// assert!(!lf.is_ascii_uppercase()); + /// assert!(!esc.is_ascii_uppercase()); + /// ``` + #[stable(feature = "ascii_ctype_on_intrinsics", since = "1.24.0")] + #[inline] + pub fn is_ascii_uppercase(&self) -> bool { + self.is_ascii() && (*self as u8).is_ascii_uppercase() + } + + /// Checks if the value is an ASCII lowercase character: + /// U+0061 'a' ... U+007A 'z'. + /// + /// # Examples + /// + /// ``` + /// let uppercase_a = 'A'; + /// let uppercase_g = 'G'; + /// let a = 'a'; + /// let g = 'g'; + /// let zero = '0'; + /// let percent = '%'; + /// let space = ' '; + /// let lf = '\n'; + /// let esc: char = 0x1b_u8.into(); + /// + /// assert!(!uppercase_a.is_ascii_lowercase()); + /// assert!(!uppercase_g.is_ascii_lowercase()); + /// assert!(a.is_ascii_lowercase()); + /// assert!(g.is_ascii_lowercase()); + /// assert!(!zero.is_ascii_lowercase()); + /// assert!(!percent.is_ascii_lowercase()); + /// assert!(!space.is_ascii_lowercase()); + /// assert!(!lf.is_ascii_lowercase()); + /// assert!(!esc.is_ascii_lowercase()); + /// ``` + #[stable(feature = "ascii_ctype_on_intrinsics", since = "1.24.0")] + #[inline] + pub fn is_ascii_lowercase(&self) -> bool { + self.is_ascii() && (*self as u8).is_ascii_lowercase() + } + + /// Checks if the value is an ASCII alphanumeric character: + /// + /// - U+0041 'A' ... U+005A 'Z', or + /// - U+0061 'a' ... U+007A 'z', or + /// - U+0030 '0' ... U+0039 '9'. + /// + /// # Examples + /// + /// ``` + /// let uppercase_a = 'A'; + /// let uppercase_g = 'G'; + /// let a = 'a'; + /// let g = 'g'; + /// let zero = '0'; + /// let percent = '%'; + /// let space = ' '; + /// let lf = '\n'; + /// let esc: char = 0x1b_u8.into(); + /// + /// assert!(uppercase_a.is_ascii_alphanumeric()); + /// assert!(uppercase_g.is_ascii_alphanumeric()); + /// assert!(a.is_ascii_alphanumeric()); + /// assert!(g.is_ascii_alphanumeric()); + /// assert!(zero.is_ascii_alphanumeric()); + /// assert!(!percent.is_ascii_alphanumeric()); + /// assert!(!space.is_ascii_alphanumeric()); + /// assert!(!lf.is_ascii_alphanumeric()); + /// assert!(!esc.is_ascii_alphanumeric()); + /// ``` + #[stable(feature = "ascii_ctype_on_intrinsics", since = "1.24.0")] + #[inline] + pub fn is_ascii_alphanumeric(&self) -> bool { + self.is_ascii() && (*self as u8).is_ascii_alphanumeric() + } + + /// Checks if the value is an ASCII decimal digit: + /// U+0030 '0' ... U+0039 '9'. + /// + /// # Examples + /// + /// ``` + /// let uppercase_a = 'A'; + /// let uppercase_g = 'G'; + /// let a = 'a'; + /// let g = 'g'; + /// let zero = '0'; + /// let percent = '%'; + /// let space = ' '; + /// let lf = '\n'; + /// let esc: char = 0x1b_u8.into(); + /// + /// assert!(!uppercase_a.is_ascii_digit()); + /// assert!(!uppercase_g.is_ascii_digit()); + /// assert!(!a.is_ascii_digit()); + /// assert!(!g.is_ascii_digit()); + /// assert!(zero.is_ascii_digit()); + /// assert!(!percent.is_ascii_digit()); + /// assert!(!space.is_ascii_digit()); + /// assert!(!lf.is_ascii_digit()); + /// assert!(!esc.is_ascii_digit()); + /// ``` + #[stable(feature = "ascii_ctype_on_intrinsics", since = "1.24.0")] + #[inline] + pub fn is_ascii_digit(&self) -> bool { + self.is_ascii() && (*self as u8).is_ascii_digit() + } + + /// Checks if the value is an ASCII hexadecimal digit: + /// + /// - U+0030 '0' ... U+0039 '9', or + /// - U+0041 'A' ... U+0046 'F', or + /// - U+0061 'a' ... U+0066 'f'. + /// + /// # Examples + /// + /// ``` + /// let uppercase_a = 'A'; + /// let uppercase_g = 'G'; + /// let a = 'a'; + /// let g = 'g'; + /// let zero = '0'; + /// let percent = '%'; + /// let space = ' '; + /// let lf = '\n'; + /// let esc: char = 0x1b_u8.into(); + /// + /// assert!(uppercase_a.is_ascii_hexdigit()); + /// assert!(!uppercase_g.is_ascii_hexdigit()); + /// assert!(a.is_ascii_hexdigit()); + /// assert!(!g.is_ascii_hexdigit()); + /// assert!(zero.is_ascii_hexdigit()); + /// assert!(!percent.is_ascii_hexdigit()); + /// assert!(!space.is_ascii_hexdigit()); + /// assert!(!lf.is_ascii_hexdigit()); + /// assert!(!esc.is_ascii_hexdigit()); + /// ``` + #[stable(feature = "ascii_ctype_on_intrinsics", since = "1.24.0")] + #[inline] + pub fn is_ascii_hexdigit(&self) -> bool { + self.is_ascii() && (*self as u8).is_ascii_hexdigit() + } + + /// Checks if the value is an ASCII punctuation character: + /// + /// - U+0021 ... U+002F `! " # $ % & ' ( ) * + , - . /`, or + /// - U+003A ... U+0040 `: ; < = > ? @`, or + /// - U+005B ... U+0060 ``[ \ ] ^ _ ` ``, or + /// - U+007B ... U+007E `{ | } ~` + /// + /// # Examples + /// + /// ``` + /// let uppercase_a = 'A'; + /// let uppercase_g = 'G'; + /// let a = 'a'; + /// let g = 'g'; + /// let zero = '0'; + /// let percent = '%'; + /// let space = ' '; + /// let lf = '\n'; + /// let esc: char = 0x1b_u8.into(); + /// + /// assert!(!uppercase_a.is_ascii_punctuation()); + /// assert!(!uppercase_g.is_ascii_punctuation()); + /// assert!(!a.is_ascii_punctuation()); + /// assert!(!g.is_ascii_punctuation()); + /// assert!(!zero.is_ascii_punctuation()); + /// assert!(percent.is_ascii_punctuation()); + /// assert!(!space.is_ascii_punctuation()); + /// assert!(!lf.is_ascii_punctuation()); + /// assert!(!esc.is_ascii_punctuation()); + /// ``` + #[stable(feature = "ascii_ctype_on_intrinsics", since = "1.24.0")] + #[inline] + pub fn is_ascii_punctuation(&self) -> bool { + self.is_ascii() && (*self as u8).is_ascii_punctuation() + } + + /// Checks if the value is an ASCII graphic character: + /// U+0021 '!' ... U+007E '~'. + /// + /// # Examples + /// + /// ``` + /// let uppercase_a = 'A'; + /// let uppercase_g = 'G'; + /// let a = 'a'; + /// let g = 'g'; + /// let zero = '0'; + /// let percent = '%'; + /// let space = ' '; + /// let lf = '\n'; + /// let esc: char = 0x1b_u8.into(); + /// + /// assert!(uppercase_a.is_ascii_graphic()); + /// assert!(uppercase_g.is_ascii_graphic()); + /// assert!(a.is_ascii_graphic()); + /// assert!(g.is_ascii_graphic()); + /// assert!(zero.is_ascii_graphic()); + /// assert!(percent.is_ascii_graphic()); + /// assert!(!space.is_ascii_graphic()); + /// assert!(!lf.is_ascii_graphic()); + /// assert!(!esc.is_ascii_graphic()); + /// ``` + #[stable(feature = "ascii_ctype_on_intrinsics", since = "1.24.0")] + #[inline] + pub fn is_ascii_graphic(&self) -> bool { + self.is_ascii() && (*self as u8).is_ascii_graphic() + } + + /// Checks if the value is an ASCII whitespace character: + /// U+0020 SPACE, U+0009 HORIZONTAL TAB, U+000A LINE FEED, + /// U+000C FORM FEED, or U+000D CARRIAGE RETURN. + /// + /// Rust uses the WhatWG Infra Standard's [definition of ASCII + /// whitespace][infra-aw]. There are several other definitions in + /// wide use. For instance, [the POSIX locale][pct] includes + /// U+000B VERTICAL TAB as well as all the above characters, + /// but—from the very same specification—[the default rule for + /// "field splitting" in the Bourne shell][bfs] considers *only* + /// SPACE, HORIZONTAL TAB, and LINE FEED as whitespace. + /// + /// If you are writing a program that will process an existing + /// file format, check what that format's definition of whitespace is + /// before using this function. + /// + /// [infra-aw]: https://infra.spec.whatwg.org/#ascii-whitespace + /// [pct]: http://pubs.opengroup.org/onlinepubs/9699919799/basedefs/V1_chap07.html#tag_07_03_01 + /// [bfs]: http://pubs.opengroup.org/onlinepubs/9699919799/utilities/V3_chap02.html#tag_18_06_05 + /// + /// # Examples + /// + /// ``` + /// let uppercase_a = 'A'; + /// let uppercase_g = 'G'; + /// let a = 'a'; + /// let g = 'g'; + /// let zero = '0'; + /// let percent = '%'; + /// let space = ' '; + /// let lf = '\n'; + /// let esc: char = 0x1b_u8.into(); + /// + /// assert!(!uppercase_a.is_ascii_whitespace()); + /// assert!(!uppercase_g.is_ascii_whitespace()); + /// assert!(!a.is_ascii_whitespace()); + /// assert!(!g.is_ascii_whitespace()); + /// assert!(!zero.is_ascii_whitespace()); + /// assert!(!percent.is_ascii_whitespace()); + /// assert!(space.is_ascii_whitespace()); + /// assert!(lf.is_ascii_whitespace()); + /// assert!(!esc.is_ascii_whitespace()); + /// ``` + #[stable(feature = "ascii_ctype_on_intrinsics", since = "1.24.0")] + #[inline] + pub fn is_ascii_whitespace(&self) -> bool { + self.is_ascii() && (*self as u8).is_ascii_whitespace() + } + + /// Checks if the value is an ASCII control character: + /// U+0000 NUL ... U+001F UNIT SEPARATOR, or U+007F DELETE. + /// Note that most ASCII whitespace characters are control + /// characters, but SPACE is not. + /// + /// # Examples + /// + /// ``` + /// let uppercase_a = 'A'; + /// let uppercase_g = 'G'; + /// let a = 'a'; + /// let g = 'g'; + /// let zero = '0'; + /// let percent = '%'; + /// let space = ' '; + /// let lf = '\n'; + /// let esc: char = 0x1b_u8.into(); + /// + /// assert!(!uppercase_a.is_ascii_control()); + /// assert!(!uppercase_g.is_ascii_control()); + /// assert!(!a.is_ascii_control()); + /// assert!(!g.is_ascii_control()); + /// assert!(!zero.is_ascii_control()); + /// assert!(!percent.is_ascii_control()); + /// assert!(!space.is_ascii_control()); + /// assert!(lf.is_ascii_control()); + /// assert!(esc.is_ascii_control()); + /// ``` + #[stable(feature = "ascii_ctype_on_intrinsics", since = "1.24.0")] + #[inline] + pub fn is_ascii_control(&self) -> bool { + self.is_ascii() && (*self as u8).is_ascii_control() + } +} diff --git a/src/libcore/char/mod.rs b/src/libcore/char/mod.rs new file mode 100644 index 00000000000..7e1313747ee --- /dev/null +++ b/src/libcore/char/mod.rs @@ -0,0 +1,504 @@ +// Copyright 2012-2014 The Rust Project Developers. See the COPYRIGHT +// file at the top-level directory of this distribution and at +// http://rust-lang.org/COPYRIGHT. +// +// Licensed under the Apache License, Version 2.0 <LICENSE-APACHE or +// http://www.apache.org/licenses/LICENSE-2.0> or the MIT license +// <LICENSE-MIT or http://opensource.org/licenses/MIT>, at your +// option. This file may not be copied, modified, or distributed +// except according to those terms. + +//! A character type. +//! +//! The `char` type represents a single character. More specifically, since +//! 'character' isn't a well-defined concept in Unicode, `char` is a '[Unicode +//! scalar value]', which is similar to, but not the same as, a '[Unicode code +//! point]'. +//! +//! [Unicode scalar value]: http://www.unicode.org/glossary/#unicode_scalar_value +//! [Unicode code point]: http://www.unicode.org/glossary/#code_point +//! +//! This module exists for technical reasons, the primary documentation for +//! `char` is directly on [the `char` primitive type](../../std/primitive.char.html) +//! itself. +//! +//! This module is the home of the iterator implementations for the iterators +//! implemented on `char`, as well as some useful constants and conversion +//! functions that convert various types to `char`. + +#![allow(non_snake_case)] +#![stable(feature = "core_char", since = "1.2.0")] + +mod convert; +mod decode; +mod methods; + +// stable re-exports +#[stable(feature = "rust1", since = "1.0.0")] +pub use self::convert::{from_u32, from_digit}; +#[stable(feature = "char_from_unchecked", since = "1.5.0")] +pub use self::convert::from_u32_unchecked; +#[stable(feature = "char_from_str", since = "1.20.0")] +pub use self::convert::ParseCharError; +#[unstable(feature = "try_from", issue = "33417")] +pub use self::convert::CharTryFromError; +#[stable(feature = "decode_utf16", since = "1.9.0")] +pub use self::decode::{decode_utf16, DecodeUtf16, DecodeUtf16Error}; + +// unstable re-exports +#[unstable(feature = "unicode_version", issue = "49726")] +pub use unicode::tables::UNICODE_VERSION; +#[unstable(feature = "unicode_version", issue = "49726")] +pub use unicode::version::UnicodeVersion; + +use fmt::{self, Write}; +use iter::FusedIterator; + +// UTF-8 ranges and tags for encoding characters +const TAG_CONT: u8 = 0b1000_0000; +const TAG_TWO_B: u8 = 0b1100_0000; +const TAG_THREE_B: u8 = 0b1110_0000; +const TAG_FOUR_B: u8 = 0b1111_0000; +const MAX_ONE_B: u32 = 0x80; +const MAX_TWO_B: u32 = 0x800; +const MAX_THREE_B: u32 = 0x10000; + +/* + Lu Uppercase_Letter an uppercase letter + Ll Lowercase_Letter a lowercase letter + Lt Titlecase_Letter a digraphic character, with first part uppercase + Lm Modifier_Letter a modifier letter + Lo Other_Letter other letters, including syllables and ideographs + Mn Nonspacing_Mark a nonspacing combining mark (zero advance width) + Mc Spacing_Mark a spacing combining mark (positive advance width) + Me Enclosing_Mark an enclosing combining mark + Nd Decimal_Number a decimal digit + Nl Letter_Number a letterlike numeric character + No Other_Number a numeric character of other type + Pc Connector_Punctuation a connecting punctuation mark, like a tie + Pd Dash_Punctuation a dash or hyphen punctuation mark + Ps Open_Punctuation an opening punctuation mark (of a pair) + Pe Close_Punctuation a closing punctuation mark (of a pair) + Pi Initial_Punctuation an initial quotation mark + Pf Final_Punctuation a final quotation mark + Po Other_Punctuation a punctuation mark of other type + Sm Math_Symbol a symbol of primarily mathematical use + Sc Currency_Symbol a currency sign + Sk Modifier_Symbol a non-letterlike modifier symbol + So Other_Symbol a symbol of other type + Zs Space_Separator a space character (of various non-zero widths) + Zl Line_Separator U+2028 LINE SEPARATOR only + Zp Paragraph_Separator U+2029 PARAGRAPH SEPARATOR only + Cc Control a C0 or C1 control code + Cf Format a format control character + Cs Surrogate a surrogate code point + Co Private_Use a private-use character + Cn Unassigned a reserved unassigned code point or a noncharacter +*/ + +/// The highest valid code point a `char` can have. +/// +/// A [`char`] is a [Unicode Scalar Value], which means that it is a [Code +/// Point], but only ones within a certain range. `MAX` is the highest valid +/// code point that's a valid [Unicode Scalar Value]. +/// +/// [`char`]: ../../std/primitive.char.html +/// [Unicode Scalar Value]: http://www.unicode.org/glossary/#unicode_scalar_value +/// [Code Point]: http://www.unicode.org/glossary/#code_point +#[stable(feature = "rust1", since = "1.0.0")] +pub const MAX: char = '\u{10ffff}'; + +/// `U+FFFD REPLACEMENT CHARACTER` (�) is used in Unicode to represent a +/// decoding error. +/// +/// It can occur, for example, when giving ill-formed UTF-8 bytes to +/// [`String::from_utf8_lossy`](../../std/string/struct.String.html#method.from_utf8_lossy). +#[stable(feature = "decode_utf16", since = "1.9.0")] +pub const REPLACEMENT_CHARACTER: char = '\u{FFFD}'; + +/// Returns an iterator that yields the hexadecimal Unicode escape of a +/// character, as `char`s. +/// +/// This `struct` is created by the [`escape_unicode`] method on [`char`]. See +/// its documentation for more. +/// +/// [`escape_unicode`]: ../../std/primitive.char.html#method.escape_unicode +/// [`char`]: ../../std/primitive.char.html +#[derive(Clone, Debug)] +#[stable(feature = "rust1", since = "1.0.0")] +pub struct EscapeUnicode { + c: char, + state: EscapeUnicodeState, + + // The index of the next hex digit to be printed (0 if none), + // i.e. the number of remaining hex digits to be printed; + // increasing from the least significant digit: 0x543210 + hex_digit_idx: usize, +} + +// The enum values are ordered so that their representation is the +// same as the remaining length (besides the hexadecimal digits). This +// likely makes `len()` a single load from memory) and inline-worth. +#[derive(Clone, Debug)] +enum EscapeUnicodeState { + Done, + RightBrace, + Value, + LeftBrace, + Type, + Backslash, +} + +#[stable(feature = "rust1", since = "1.0.0")] +impl Iterator for EscapeUnicode { + type Item = char; + + fn next(&mut self) -> Option<char> { + match self.state { + EscapeUnicodeState::Backslash => { + self.state = EscapeUnicodeState::Type; + Some('\\') + } + EscapeUnicodeState::Type => { + self.state = EscapeUnicodeState::LeftBrace; + Some('u') + } + EscapeUnicodeState::LeftBrace => { + self.state = EscapeUnicodeState::Value; + Some('{') + } + EscapeUnicodeState::Value => { + let hex_digit = ((self.c as u32) >> (self.hex_digit_idx * 4)) & 0xf; + let c = from_digit(hex_digit, 16).unwrap(); + if self.hex_digit_idx == 0 { + self.state = EscapeUnicodeState::RightBrace; + } else { + self.hex_digit_idx -= 1; + } + Some(c) + } + EscapeUnicodeState::RightBrace => { + self.state = EscapeUnicodeState::Done; + Some('}') + } + EscapeUnicodeState::Done => None, + } + } + + #[inline] + fn size_hint(&self) -> (usize, Option<usize>) { + let n = self.len(); + (n, Some(n)) + } + + #[inline] + fn count(self) -> usize { + self.len() + } + + fn last(self) -> Option<char> { + match self.state { + EscapeUnicodeState::Done => None, + + EscapeUnicodeState::RightBrace | + EscapeUnicodeState::Value | + EscapeUnicodeState::LeftBrace | + EscapeUnicodeState::Type | + EscapeUnicodeState::Backslash => Some('}'), + } + } +} + +#[stable(feature = "exact_size_escape", since = "1.11.0")] +impl ExactSizeIterator for EscapeUnicode { + #[inline] + fn len(&self) -> usize { + // The match is a single memory access with no branching + self.hex_digit_idx + match self.state { + EscapeUnicodeState::Done => 0, + EscapeUnicodeState::RightBrace => 1, + EscapeUnicodeState::Value => 2, + EscapeUnicodeState::LeftBrace => 3, + EscapeUnicodeState::Type => 4, + EscapeUnicodeState::Backslash => 5, + } + } +} + +#[stable(feature = "fused", since = "1.26.0")] +impl FusedIterator for EscapeUnicode {} + +#[stable(feature = "char_struct_display", since = "1.16.0")] +impl fmt::Display for EscapeUnicode { + fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result { + for c in self.clone() { + f.write_char(c)?; + } + Ok(()) + } +} + +/// An iterator that yields the literal escape code of a `char`. +/// +/// This `struct` is created by the [`escape_default`] method on [`char`]. See +/// its documentation for more. +/// +/// [`escape_default`]: ../../std/primitive.char.html#method.escape_default +/// [`char`]: ../../std/primitive.char.html +#[derive(Clone, Debug)] +#[stable(feature = "rust1", since = "1.0.0")] +pub struct EscapeDefault { + state: EscapeDefaultState +} + +#[derive(Clone, Debug)] +enum EscapeDefaultState { + Done, + Char(char), + Backslash(char), + Unicode(EscapeUnicode), +} + +#[stable(feature = "rust1", since = "1.0.0")] +impl Iterator for EscapeDefault { + type Item = char; + + fn next(&mut self) -> Option<char> { + match self.state { + EscapeDefaultState::Backslash(c) => { + self.state = EscapeDefaultState::Char(c); + Some('\\') + } + EscapeDefaultState::Char(c) => { + self.state = EscapeDefaultState::Done; + Some(c) + } + EscapeDefaultState::Done => None, + EscapeDefaultState::Unicode(ref mut iter) => iter.next(), + } + } + + #[inline] + fn size_hint(&self) -> (usize, Option<usize>) { + let n = self.len(); + (n, Some(n)) + } + + #[inline] + fn count(self) -> usize { + self.len() + } + + fn nth(&mut self, n: usize) -> Option<char> { + match self.state { + EscapeDefaultState::Backslash(c) if n == 0 => { + self.state = EscapeDefaultState::Char(c); + Some('\\') + }, + EscapeDefaultState::Backslash(c) if n == 1 => { + self.state = EscapeDefaultState::Done; + Some(c) + }, + EscapeDefaultState::Backslash(_) => { + self.state = EscapeDefaultState::Done; + None + }, + EscapeDefaultState::Char(c) => { + self.state = EscapeDefaultState::Done; + + if n == 0 { + Some(c) + } else { + None + } + }, + EscapeDefaultState::Done => None, + EscapeDefaultState::Unicode(ref mut i) => i.nth(n), + } + } + + fn last(self) -> Option<char> { + match self.state { + EscapeDefaultState::Unicode(iter) => iter.last(), + EscapeDefaultState::Done => None, + EscapeDefaultState::Backslash(c) | EscapeDefaultState::Char(c) => Some(c), + } + } +} + +#[stable(feature = "exact_size_escape", since = "1.11.0")] +impl ExactSizeIterator for EscapeDefault { + fn len(&self) -> usize { + match self.state { + EscapeDefaultState::Done => 0, + EscapeDefaultState::Char(_) => 1, + EscapeDefaultState::Backslash(_) => 2, + EscapeDefaultState::Unicode(ref iter) => iter.len(), + } + } +} + +#[stable(feature = "fused", since = "1.26.0")] +impl FusedIterator for EscapeDefault {} + +#[stable(feature = "char_struct_display", since = "1.16.0")] +impl fmt::Display for EscapeDefault { + fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result { + for c in self.clone() { + f.write_char(c)?; + } + Ok(()) + } +} + +/// An iterator that yields the literal escape code of a `char`. +/// +/// This `struct` is created by the [`escape_debug`] method on [`char`]. See its +/// documentation for more. +/// +/// [`escape_debug`]: ../../std/primitive.char.html#method.escape_debug +/// [`char`]: ../../std/primitive.char.html +#[stable(feature = "char_escape_debug", since = "1.20.0")] +#[derive(Clone, Debug)] +pub struct EscapeDebug(EscapeDefault); + +#[stable(feature = "char_escape_debug", since = "1.20.0")] +impl Iterator for EscapeDebug { + type Item = char; + fn next(&mut self) -> Option<char> { self.0.next() } + fn size_hint(&self) -> (usize, Option<usize>) { self.0.size_hint() } +} + +#[stable(feature = "char_escape_debug", since = "1.20.0")] +impl ExactSizeIterator for EscapeDebug { } + +#[stable(feature = "fused", since = "1.26.0")] +impl FusedIterator for EscapeDebug {} + +#[stable(feature = "char_escape_debug", since = "1.20.0")] +impl fmt::Display for EscapeDebug { + fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result { + fmt::Display::fmt(&self.0, f) + } +} + +/// Returns an iterator that yields the lowercase equivalent of a `char`. +/// +/// This `struct` is created by the [`to_lowercase`] method on [`char`]. See +/// its documentation for more. +/// +/// [`to_lowercase`]: ../../std/primitive.char.html#method.to_lowercase +/// [`char`]: ../../std/primitive.char.html +#[stable(feature = "rust1", since = "1.0.0")] +#[derive(Debug, Clone)] +pub struct ToLowercase(CaseMappingIter); + +#[stable(feature = "rust1", since = "1.0.0")] +impl Iterator for ToLowercase { + type Item = char; + fn next(&mut self) -> Option<char> { + self.0.next() + } +} + +#[stable(feature = "fused", since = "1.26.0")] +impl FusedIterator for ToLowercase {} + +/// Returns an iterator that yields the uppercase equivalent of a `char`. +/// +/// This `struct` is created by the [`to_uppercase`] method on [`char`]. See +/// its documentation for more. +/// +/// [`to_uppercase`]: ../../std/primitive.char.html#method.to_uppercase +/// [`char`]: ../../std/primitive.char.html +#[stable(feature = "rust1", since = "1.0.0")] +#[derive(Debug, Clone)] +pub struct ToUppercase(CaseMappingIter); + +#[stable(feature = "rust1", since = "1.0.0")] +impl Iterator for ToUppercase { + type Item = char; + fn next(&mut self) -> Option<char> { + self.0.next() + } +} + +#[stable(feature = "fused", since = "1.26.0")] +impl FusedIterator for ToUppercase {} + +#[derive(Debug, Clone)] +enum CaseMappingIter { + Three(char, char, char), + Two(char, char), + One(char), + Zero, +} + +impl CaseMappingIter { + fn new(chars: [char; 3]) -> CaseMappingIter { + if chars[2] == '\0' { + if chars[1] == '\0' { + CaseMappingIter::One(chars[0]) // Including if chars[0] == '\0' + } else { + CaseMappingIter::Two(chars[0], chars[1]) + } + } else { + CaseMappingIter::Three(chars[0], chars[1], chars[2]) + } + } +} + +impl Iterator for CaseMappingIter { + type Item = char; + fn next(&mut self) -> Option<char> { + match *self { + CaseMappingIter::Three(a, b, c) => { + *self = CaseMappingIter::Two(b, c); + Some(a) + } + CaseMappingIter::Two(b, c) => { + *self = CaseMappingIter::One(c); + Some(b) + } + CaseMappingIter::One(c) => { + *self = CaseMappingIter::Zero; + Some(c) + } + CaseMappingIter::Zero => None, + } + } +} + +impl fmt::Display for CaseMappingIter { + fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result { + match *self { + CaseMappingIter::Three(a, b, c) => { + f.write_char(a)?; + f.write_char(b)?; + f.write_char(c) + } + CaseMappingIter::Two(b, c) => { + f.write_char(b)?; + f.write_char(c) + } + CaseMappingIter::One(c) => { + f.write_char(c) + } + CaseMappingIter::Zero => Ok(()), + } + } +} + +#[stable(feature = "char_struct_display", since = "1.16.0")] +impl fmt::Display for ToLowercase { + fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result { + fmt::Display::fmt(&self.0, f) + } +} + +#[stable(feature = "char_struct_display", since = "1.16.0")] +impl fmt::Display for ToUppercase { + fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result { + fmt::Display::fmt(&self.0, f) + } +} diff --git a/src/libcore/clone.rs b/src/libcore/clone.rs index 826420a0c00..3b15ba2b4ab 100644 --- a/src/libcore/clone.rs +++ b/src/libcore/clone.rs @@ -87,6 +87,23 @@ /// fn clone(&self) -> Stats { *self } /// } /// ``` +/// +/// ## Additional implementors +/// +/// In addition to the [implementors listed below][impls], +/// the following types also implement `Clone`: +/// +/// * Function item types (i.e. the distinct types defined for each function) +/// * Function pointer types (e.g. `fn() -> i32`) +/// * Array types, for all sizes, if the item type also implements `Clone` (e.g. `[i32; 123456]`) +/// * Tuple types, if each component also implements `Clone` (e.g. `()`, `(i32, bool)`) +/// * Closure types, if they capture no value from the environment +/// or if all such captured values implement `Clone` themselves. +/// Note that variables captured by shared reference always implement `Clone` +/// (even if the referent doesn't), +/// while variables captured by mutable reference never implement `Clone`. +/// +/// [impls]: #implementors #[stable(feature = "rust1", since = "1.0.0")] #[lang = "clone"] pub trait Clone : Sized { @@ -100,6 +117,7 @@ pub trait Clone : Sized { /// assert_eq!("Hello", hello.clone()); /// ``` #[stable(feature = "rust1", since = "1.0.0")] + #[must_use = "cloning is often expensive and is not expected to have side effects"] fn clone(&self) -> Self; /// Performs copy-assignment from `source`. @@ -130,3 +148,67 @@ pub struct AssertParamIsClone<T: Clone + ?Sized> { _field: ::marker::PhantomData reason = "deriving hack, should not be public", issue = "0")] pub struct AssertParamIsCopy<T: Copy + ?Sized> { _field: ::marker::PhantomData<T> } + +/// Implementations of `Clone` for primitive types. +/// +/// Implementations that cannot be described in Rust +/// are implemented in `SelectionContext::copy_clone_conditions()` in librustc. +mod impls { + + use super::Clone; + + macro_rules! impl_clone { + ($($t:ty)*) => { + $( + #[stable(feature = "rust1", since = "1.0.0")] + impl Clone for $t { + #[inline] + fn clone(&self) -> Self { + *self + } + } + )* + } + } + + impl_clone! { + usize u8 u16 u32 u64 u128 + isize i8 i16 i32 i64 i128 + f32 f64 + bool char + } + + #[unstable(feature = "never_type", issue = "35121")] + impl Clone for ! { + #[inline] + fn clone(&self) -> Self { + *self + } + } + + #[stable(feature = "rust1", since = "1.0.0")] + impl<T: ?Sized> Clone for *const T { + #[inline] + fn clone(&self) -> Self { + *self + } + } + + #[stable(feature = "rust1", since = "1.0.0")] + impl<T: ?Sized> Clone for *mut T { + #[inline] + fn clone(&self) -> Self { + *self + } + } + + // Shared references can be cloned, but mutable references *cannot*! + #[stable(feature = "rust1", since = "1.0.0")] + impl<'a, T: ?Sized> Clone for &'a T { + #[inline] + fn clone(&self) -> Self { + *self + } + } + +} diff --git a/src/libcore/cmp.rs b/src/libcore/cmp.rs index 266cae3c122..58d6c4f5e09 100644 --- a/src/libcore/cmp.rs +++ b/src/libcore/cmp.rs @@ -106,7 +106,12 @@ use self::Ordering::*; /// ``` #[lang = "eq"] #[stable(feature = "rust1", since = "1.0.0")] -#[rustc_on_unimplemented = "can't compare `{Self}` with `{Rhs}`"] +#[doc(alias = "==")] +#[doc(alias = "!=")] +#[rustc_on_unimplemented( + message="can't compare `{Self}` with `{Rhs}`", + label="no implementation for `{Self} == {Rhs}`", +)] pub trait PartialEq<Rhs: ?Sized = Self> { /// This method tests for `self` and `other` values to be equal, and is used /// by `==`. @@ -160,6 +165,8 @@ pub trait PartialEq<Rhs: ?Sized = Self> { /// } /// impl Eq for Book {} /// ``` +#[doc(alias = "==")] +#[doc(alias = "!=")] #[stable(feature = "rust1", since = "1.0.0")] pub trait Eq: PartialEq<Self> { // this method is used solely by #[deriving] to assert @@ -343,7 +350,7 @@ impl Ordering { /// v.sort_by_key(|&num| (num > 3, Reverse(num))); /// assert_eq!(v, vec![3, 2, 1, 6, 5, 4]); /// ``` -#[derive(PartialEq, Eq, Debug)] +#[derive(PartialEq, Eq, Debug, Copy, Clone, Default, Hash)] #[stable(feature = "reverse_cmp_key", since = "1.19.0")] pub struct Reverse<T>(#[stable(feature = "reverse_cmp_key", since = "1.19.0")] pub T); @@ -427,6 +434,11 @@ impl<T: Ord> Ord for Reverse<T> { /// } /// } /// ``` +#[lang = "ord"] +#[doc(alias = "<")] +#[doc(alias = ">")] +#[doc(alias = "<=")] +#[doc(alias = ">=")] #[stable(feature = "rust1", since = "1.0.0")] pub trait Ord: Eq + PartialOrd<Self> { /// This method returns an `Ordering` between `self` and `other`. @@ -457,6 +469,7 @@ pub trait Ord: Eq + PartialOrd<Self> { /// assert_eq!(2, 2.max(2)); /// ``` #[stable(feature = "ord_max_min", since = "1.21.0")] + #[inline] fn max(self, other: Self) -> Self where Self: Sized { if other >= self { other } else { self } @@ -473,6 +486,7 @@ pub trait Ord: Eq + PartialOrd<Self> { /// assert_eq!(2, 2.min(2)); /// ``` #[stable(feature = "ord_max_min", since = "1.21.0")] + #[inline] fn min(self, other: Self) -> Self where Self: Sized { if self <= other { self } else { other } @@ -596,9 +610,16 @@ impl PartialOrd for Ordering { /// assert_eq!(x < y, true); /// assert_eq!(x.lt(&y), true); /// ``` -#[lang = "ord"] +#[lang = "partial_ord"] #[stable(feature = "rust1", since = "1.0.0")] -#[rustc_on_unimplemented = "can't compare `{Self}` with `{Rhs}`"] +#[doc(alias = ">")] +#[doc(alias = "<")] +#[doc(alias = "<=")] +#[doc(alias = ">=")] +#[rustc_on_unimplemented( + message="can't compare `{Self}` with `{Rhs}`", + label="no implementation for `{Self} < {Rhs}` and `{Self} > {Rhs}`", +)] pub trait PartialOrd<Rhs: ?Sized = Self>: PartialEq<Rhs> { /// This method returns an ordering between `self` and `other` values if one exists. /// diff --git a/src/libcore/convert.rs b/src/libcore/convert.rs index 41eb1eed4e4..67cb010c6b4 100644 --- a/src/libcore/convert.rs +++ b/src/libcore/convert.rs @@ -48,9 +48,9 @@ #![stable(feature = "rust1", since = "1.0.0")] -use fmt; - -/// An identity function. Two things are important to note about this function: +/// An identity function. +/// +/// Two things are important to note about this function: /// /// - It is not always equivalent to a closure like `|x| x` since the /// closure may coerce `x` into a different type. @@ -62,80 +62,51 @@ use fmt; /// /// # Examples /// -/// Using `id` to do nothing among other interesting functions: +/// Using `identity` to do nothing among other interesting functions: /// /// ```rust /// #![feature(convert_id)] -/// use std::convert::id; +/// use std::convert::identity; /// /// fn manipulation(x: u32) -> u32 { /// // Let's assume that this function does something interesting. /// x + 1 /// } /// -/// let _arr = &[id, manipulation]; +/// let _arr = &[identity, manipulation]; /// ``` /// -/// Using `id` to get a function that changes nothing in a conditional: +/// Using `identity` to get a function that changes nothing in a conditional: /// /// ```rust /// #![feature(convert_id)] -/// use std::convert::id; +/// use std::convert::identity; /// /// # let condition = true; /// /// # fn manipulation(x: u32) -> u32 { x + 1 } /// -/// let do_stuff = if condition { manipulation } else { id }; +/// let do_stuff = if condition { manipulation } else { identity }; /// /// // do more interesting stuff.. /// /// let _results = do_stuff(42); /// ``` /// -/// Using `id` to concatenate an iterator of iterators: -/// -/// ```rust -/// #![feature(convert_id)] -/// use std::convert::id; -/// -/// let vec_vec = vec![vec![1, 3, 4], vec![5, 6]]; -/// let iter_iter = vec_vec.into_iter().map(Vec::into_iter); -/// let concatenated = iter_iter.flat_map(id).collect::<Vec<_>>(); -/// assert_eq!(vec![1, 3, 4, 5, 6], concatenated); -/// ``` -/// -/// Using `id` to keep the `Some` variants of an iterator of `Option<T>`: +/// Using `identity` to keep the `Some` variants of an iterator of `Option<T>`: /// /// ```rust /// #![feature(convert_id)] -/// use std::convert::id; +/// use std::convert::identity; /// /// let iter = vec![Some(1), None, Some(3)].into_iter(); -/// let filtered = iter.filter_map(id).collect::<Vec<_>>(); +/// let filtered = iter.filter_map(identity).collect::<Vec<_>>(); /// assert_eq!(vec![1, 3], filtered); /// ``` #[unstable(feature = "convert_id", issue = "0")] #[inline] -pub fn id<T>(x: T) -> T { x } +pub fn identity<T>(x: T) -> T { x } -/// A type used as the error type for implementations of fallible conversion -/// traits in cases where conversions cannot actually fail. -/// -/// Because `Infallible` has no variants, a value of this type can never exist. -/// It is used only to satisfy trait signatures that expect an error type, and -/// signals to both the compiler and the user that the error case is impossible. -#[unstable(feature = "try_from", issue = "33417")] -#[derive(Clone, Copy, Debug, Eq, Hash, Ord, PartialEq, PartialOrd)] -pub enum Infallible {} - -#[unstable(feature = "try_from", issue = "33417")] -impl fmt::Display for Infallible { - fn fmt(&self, _: &mut fmt::Formatter) -> fmt::Result { - match *self { - } - } -} /// A cheap reference-to-reference conversion. Used to convert a value to a /// reference value within generic code. /// @@ -151,9 +122,9 @@ impl fmt::Display for Infallible { /// /// The key difference between the two traits is the intention: /// -/// - Use `AsRef` when goal is to simply convert into a reference -/// - Use `Borrow` when goal is related to writing code that is agnostic to the -/// type of borrow and if is reference or value +/// - Use `AsRef` when the goal is to simply convert into a reference +/// - Use `Borrow` when the goal is related to writing code that is agnostic to +/// the type of borrow and whether it is a reference or value /// /// See [the book][book] for a more detailed comparison. /// @@ -210,9 +181,9 @@ pub trait AsRef<T: ?Sized> { /// /// # Generic Implementations /// -/// - `AsMut` auto-dereferences if the inner type is a reference or a mutable -/// reference (e.g.: `foo.as_ref()` will work the same if `foo` has type -/// `&mut Foo` or `&&mut Foo`) +/// - `AsMut` auto-dereferences if the inner type is a mutable reference +/// (e.g.: `foo.as_mut()` will work the same if `foo` has type `&mut Foo` +/// or `&mut &mut Foo`) /// /// # Examples /// @@ -451,7 +422,7 @@ impl<'a, T: ?Sized, U: ?Sized> AsRef<U> for &'a mut T where T: AsRef<U> } } -// FIXME (#23442): replace the above impls for &/&mut with the following more general one: +// FIXME (#45742): replace the above impls for &/&mut with the following more general one: // // As lifts over Deref // impl<D: ?Sized + Deref, U: ?Sized> AsRef<U> for D where D::Target: AsRef<U> { // fn as_ref(&self) -> &U { @@ -468,7 +439,7 @@ impl<'a, T: ?Sized, U: ?Sized> AsMut<U> for &'a mut T where T: AsMut<U> } } -// FIXME (#23442): replace the above impl for &mut with the following more general one: +// FIXME (#45742): replace the above impl for &mut with the following more general one: // // AsMut lifts over DerefMut // impl<D: ?Sized + Deref, U: ?Sized> AsMut<U> for D where D::Target: AsMut<U> { // fn as_mut(&mut self) -> &mut U { @@ -507,7 +478,7 @@ impl<T, U> TryInto<U> for T where U: TryFrom<T> // with an uninhabited error type. #[unstable(feature = "try_from", issue = "33417")] impl<T, U> TryFrom<U> for T where T: From<U> { - type Error = Infallible; + type Error = !; fn try_from(value: U) -> Result<Self, Self::Error> { Ok(T::from(value)) diff --git a/src/libcore/fmt/builders.rs b/src/libcore/fmt/builders.rs index a1f4c6995da..3c5f934d4d8 100644 --- a/src/libcore/fmt/builders.rs +++ b/src/libcore/fmt/builders.rs @@ -11,7 +11,7 @@ use fmt; struct PadAdapter<'a> { - buf: &'a mut (fmt::Write + 'a), + buf: &'a mut (dyn fmt::Write + 'a), on_newline: bool, } @@ -84,7 +84,7 @@ impl<'a> fmt::Write for PadAdapter<'a> { /// // prints "Foo { bar: 10, baz: "Hello World" }" /// println!("{:?}", Foo { bar: 10, baz: "Hello World".to_string() }); /// ``` -#[must_use] +#[must_use = "must eventually call `finish()` on Debug builders"] #[allow(missing_debug_implementations)] #[stable(feature = "debug_builders", since = "1.2.0")] pub struct DebugStruct<'a, 'b: 'a> { @@ -107,7 +107,7 @@ pub fn debug_struct_new<'a, 'b>(fmt: &'a mut fmt::Formatter<'b>, impl<'a, 'b: 'a> DebugStruct<'a, 'b> { /// Adds a new field to the generated struct output. #[stable(feature = "debug_builders", since = "1.2.0")] - pub fn field(&mut self, name: &str, value: &fmt::Debug) -> &mut DebugStruct<'a, 'b> { + pub fn field(&mut self, name: &str, value: &dyn fmt::Debug) -> &mut DebugStruct<'a, 'b> { self.result = self.result.and_then(|_| { let prefix = if self.has_fields { "," @@ -181,7 +181,7 @@ impl<'a, 'b: 'a> DebugStruct<'a, 'b> { /// // prints "Foo(10, "Hello World")" /// println!("{:?}", Foo(10, "Hello World".to_string())); /// ``` -#[must_use] +#[must_use = "must eventually call `finish()` on Debug builders"] #[allow(missing_debug_implementations)] #[stable(feature = "debug_builders", since = "1.2.0")] pub struct DebugTuple<'a, 'b: 'a> { @@ -204,7 +204,7 @@ pub fn debug_tuple_new<'a, 'b>(fmt: &'a mut fmt::Formatter<'b>, name: &str) -> D impl<'a, 'b: 'a> DebugTuple<'a, 'b> { /// Adds a new field to the generated tuple struct output. #[stable(feature = "debug_builders", since = "1.2.0")] - pub fn field(&mut self, value: &fmt::Debug) -> &mut DebugTuple<'a, 'b> { + pub fn field(&mut self, value: &dyn fmt::Debug) -> &mut DebugTuple<'a, 'b> { self.result = self.result.and_then(|_| { let (prefix, space) = if self.fields > 0 { (",", " ") @@ -258,7 +258,7 @@ struct DebugInner<'a, 'b: 'a> { } impl<'a, 'b: 'a> DebugInner<'a, 'b> { - fn entry(&mut self, entry: &fmt::Debug) { + fn entry(&mut self, entry: &dyn fmt::Debug) { self.result = self.result.and_then(|_| { if self.is_pretty() { let mut slot = None; @@ -319,7 +319,7 @@ impl<'a, 'b: 'a> DebugInner<'a, 'b> { /// // prints "{10, 11}" /// println!("{:?}", Foo(vec![10, 11])); /// ``` -#[must_use] +#[must_use = "must eventually call `finish()` on Debug builders"] #[allow(missing_debug_implementations)] #[stable(feature = "debug_builders", since = "1.2.0")] pub struct DebugSet<'a, 'b: 'a> { @@ -340,7 +340,7 @@ pub fn debug_set_new<'a, 'b>(fmt: &'a mut fmt::Formatter<'b>) -> DebugSet<'a, 'b impl<'a, 'b: 'a> DebugSet<'a, 'b> { /// Adds a new entry to the set output. #[stable(feature = "debug_builders", since = "1.2.0")] - pub fn entry(&mut self, entry: &fmt::Debug) -> &mut DebugSet<'a, 'b> { + pub fn entry(&mut self, entry: &dyn fmt::Debug) -> &mut DebugSet<'a, 'b> { self.inner.entry(entry); self } @@ -390,7 +390,7 @@ impl<'a, 'b: 'a> DebugSet<'a, 'b> { /// // prints "[10, 11]" /// println!("{:?}", Foo(vec![10, 11])); /// ``` -#[must_use] +#[must_use = "must eventually call `finish()` on Debug builders"] #[allow(missing_debug_implementations)] #[stable(feature = "debug_builders", since = "1.2.0")] pub struct DebugList<'a, 'b: 'a> { @@ -411,7 +411,7 @@ pub fn debug_list_new<'a, 'b>(fmt: &'a mut fmt::Formatter<'b>) -> DebugList<'a, impl<'a, 'b: 'a> DebugList<'a, 'b> { /// Adds a new entry to the list output. #[stable(feature = "debug_builders", since = "1.2.0")] - pub fn entry(&mut self, entry: &fmt::Debug) -> &mut DebugList<'a, 'b> { + pub fn entry(&mut self, entry: &dyn fmt::Debug) -> &mut DebugList<'a, 'b> { self.inner.entry(entry); self } @@ -461,7 +461,7 @@ impl<'a, 'b: 'a> DebugList<'a, 'b> { /// // prints "{"A": 10, "B": 11}" /// println!("{:?}", Foo(vec![("A".to_string(), 10), ("B".to_string(), 11)])); /// ``` -#[must_use] +#[must_use = "must eventually call `finish()` on Debug builders"] #[allow(missing_debug_implementations)] #[stable(feature = "debug_builders", since = "1.2.0")] pub struct DebugMap<'a, 'b: 'a> { @@ -482,7 +482,7 @@ pub fn debug_map_new<'a, 'b>(fmt: &'a mut fmt::Formatter<'b>) -> DebugMap<'a, 'b impl<'a, 'b: 'a> DebugMap<'a, 'b> { /// Adds a new entry to the map output. #[stable(feature = "debug_builders", since = "1.2.0")] - pub fn entry(&mut self, key: &fmt::Debug, value: &fmt::Debug) -> &mut DebugMap<'a, 'b> { + pub fn entry(&mut self, key: &dyn fmt::Debug, value: &dyn fmt::Debug) -> &mut DebugMap<'a, 'b> { self.result = self.result.and_then(|_| { if self.is_pretty() { let mut slot = None; diff --git a/src/libcore/fmt/mod.rs b/src/libcore/fmt/mod.rs index 65aacb23bd7..928f95e3ba2 100644 --- a/src/libcore/fmt/mod.rs +++ b/src/libcore/fmt/mod.rs @@ -25,18 +25,19 @@ mod float; mod num; mod builders; -#[unstable(feature = "fmt_flags_align", issue = "27726")] +#[stable(feature = "fmt_flags_align", since = "1.28.0")] /// Possible alignments returned by `Formatter::align` #[derive(Debug)] pub enum Alignment { + #[stable(feature = "fmt_flags_align", since = "1.28.0")] /// Indication that contents should be left-aligned. Left, + #[stable(feature = "fmt_flags_align", since = "1.28.0")] /// Indication that contents should be right-aligned. Right, + #[stable(feature = "fmt_flags_align", since = "1.28.0")] /// Indication that contents should be center-aligned. Center, - /// No alignment was requested. - Unknown, } #[stable(feature = "debug_builders", since = "1.2.0")] @@ -254,7 +255,7 @@ pub struct Formatter<'a> { width: Option<usize>, precision: Option<usize>, - buf: &'a mut (Write+'a), + buf: &'a mut (dyn Write+'a), curarg: slice::Iter<'a, ArgumentV1<'a>>, args: &'a [ArgumentV1<'a>], } @@ -271,7 +272,7 @@ struct Void { /// /// It was added after #45197 showed that one could share a `!Sync` /// object across threads by passing it into `format_args!`. - _oibit_remover: PhantomData<*mut Fn()>, + _oibit_remover: PhantomData<*mut dyn Fn()>, } /// This struct represents the generic "argument" which is taken by the Xprintf @@ -333,7 +334,7 @@ impl<'a> ArgumentV1<'a> { // flags available in the v1 format of format_args #[derive(Copy, Clone)] -enum FlagV1 { SignPlus, SignMinus, Alternate, SignAwareZeroPad, } +enum FlagV1 { SignPlus, SignMinus, Alternate, SignAwareZeroPad, DebugLowerHex, DebugUpperHex } impl<'a> Arguments<'a> { /// When using the format_args!() macro, this function is used to generate the @@ -401,10 +402,21 @@ impl<'a> Arguments<'a> { /// safely be done, so no constructors are given and the fields are private /// to prevent modification. /// -/// The [`format_args!`] macro will safely create an instance of this structure -/// and pass it to a function or closure, passed as the first argument. The -/// macro validates the format string at compile-time so usage of the [`write`] -/// and [`format`] functions can be safely performed. +/// The [`format_args!`] macro will safely create an instance of this structure. +/// The macro validates the format string at compile-time so usage of the +/// [`write`] and [`format`] functions can be safely performed. +/// +/// You can use the `Arguments<'a>` that [`format_args!`] returns in `Debug` +/// and `Display` contexts as seen below. The example also shows that `Debug` +/// and `Display` format to the same thing: the interpolated format string +/// in `format_args!`. +/// +/// ```rust +/// let debug = format!("{:?}", format_args!("{} foo {:?}", 1, 2)); +/// let display = format!("{}", format_args!("{} foo {:?}", 1, 2)); +/// assert_eq!("1 foo 2", display); +/// assert_eq!(display, debug); +/// ``` /// /// [`format_args!`]: ../../std/macro.format_args.html /// [`format`]: ../../std/fmt/fn.format.html @@ -530,9 +542,13 @@ impl<'a> Display for Arguments<'a> { /// } /// ``` #[stable(feature = "rust1", since = "1.0.0")] -#[rustc_on_unimplemented = "`{Self}` cannot be formatted using `:?`; if it is \ - defined in your crate, add `#[derive(Debug)]` or \ - manually implement it"] +#[rustc_on_unimplemented( + on(crate_local, label="`{Self}` cannot be formatted using `{{:?}}`", + note="add `#[derive(Debug)]` or manually implement `{Debug}`"), + message="`{Self}` doesn't implement `{Debug}`", + label="`{Self}` cannot be formatted using `{{:?}}` because it doesn't implement `{Debug}`", +)] +#[doc(alias = "{:?}")] #[lang = "debug_trait"] pub trait Debug { /// Formats the value using the given formatter. @@ -593,9 +609,13 @@ pub trait Debug { /// /// println!("The origin is: {}", origin); /// ``` -#[rustc_on_unimplemented = "`{Self}` cannot be formatted with the default \ - formatter; try using `:?` instead if you are using \ - a format string"] +#[rustc_on_unimplemented( + message="`{Self}` doesn't implement `{Display}`", + label="`{Self}` cannot be formatted with the default formatter", + note="in format strings you may be able to use `{{:?}}` \ + (or {{:#?}} for pretty-print) instead", +)] +#[doc(alias = "{}")] #[stable(feature = "rust1", since = "1.0.0")] pub trait Display { /// Formats the value using the given formatter. @@ -679,18 +699,16 @@ pub trait Octal { /// /// The `Binary` trait should format its output as a number in binary. /// -/// For primitive signed integers (`i8` to `i128`, and `isize`), +/// For primitive signed integers ([`i8`] to [`i128`], and [`isize`]), /// negative values are formatted as the two’s complement representation. /// /// The alternate flag, `#`, adds a `0b` in front of the output. /// /// For more information on formatters, see [the module-level documentation][module]. /// -/// [module]: ../../std/fmt/index.html -/// /// # Examples /// -/// Basic usage with `i32`: +/// Basic usage with [`i32`]: /// /// ``` /// let x = 42; // 42 is '101010' in binary @@ -720,6 +738,12 @@ pub trait Octal { /// /// println!("l as binary is: {:b}", l); /// ``` +/// +/// [module]: ../../std/fmt/index.html +/// [`i8`]: ../../std/primitive.i8.html +/// [`i128`]: ../../std/primitive.i128.html +/// [`isize`]: ../../std/primitive.isize.html +/// [`i32`]: ../../std/primitive.i32.html #[stable(feature = "rust1", since = "1.0.0")] pub trait Binary { /// Formats the value using the given formatter. @@ -996,7 +1020,7 @@ pub trait UpperExp { /// /// [`write!`]: ../../std/macro.write.html #[stable(feature = "rust1", since = "1.0.0")] -pub fn write(output: &mut Write, args: Arguments) -> Result { +pub fn write(output: &mut dyn Write, args: Arguments) -> Result { let mut formatter = Formatter { flags: 0, width: None, @@ -1038,7 +1062,7 @@ pub fn write(output: &mut Write, args: Arguments) -> Result { impl<'a> Formatter<'a> { fn wrap_buf<'b, 'c, F>(&'b mut self, wrap: F) -> Formatter<'c> - where 'b: 'c, F: FnOnce(&'b mut (Write+'b)) -> &'c mut (Write+'c) + where 'b: 'c, F: FnOnce(&'b mut (dyn Write+'b)) -> &'c mut (dyn Write+'c) { Formatter { // We want to change this @@ -1179,6 +1203,23 @@ impl<'a> Formatter<'a> { /// is longer than this length /// /// Notably this function ignores the `flag` parameters. + /// + /// # Examples + /// + /// ``` + /// use std::fmt; + /// + /// struct Foo; + /// + /// impl fmt::Display for Foo { + /// fn fmt(&self, formatter: &mut fmt::Formatter) -> fmt::Result { + /// formatter.pad("Foo") + /// } + /// } + /// + /// assert_eq!(&format!("{:<4}", Foo), "Foo "); + /// assert_eq!(&format!("{:0>4}", Foo), "0Foo"); + /// ``` #[stable(feature = "rust1", since = "1.0.0")] pub fn pad(&mut self, s: &str) -> Result { // Make sure there's a fast path up front @@ -1192,7 +1233,11 @@ impl<'a> Formatter<'a> { // truncation. However other flags like `fill`, `width` and `align` // must act as always. if let Some((i, _)) = s.char_indices().skip(max).next() { - &s[..i] + // LLVM here can't prove that `..i` won't panic `&s[..i]`, but + // we know that it can't panic. Use `get` + `unwrap_or` to avoid + // `unsafe` and otherwise don't emit any panic-related code + // here. + s.get(..i).unwrap_or(&s) } else { &s } @@ -1297,7 +1342,7 @@ impl<'a> Formatter<'a> { } fn write_formatted_parts(&mut self, formatted: &flt2dec::Formatted) -> Result { - fn write_bytes(buf: &mut Write, s: &[u8]) -> Result { + fn write_bytes(buf: &mut dyn Write, s: &[u8]) -> Result { buf.write_str(unsafe { str::from_utf8_unchecked(s) }) } @@ -1341,7 +1386,7 @@ impl<'a> Formatter<'a> { self.buf.write_str(data) } - /// Writes some formatted information into this instance + /// Writes some formatted information into this instance. #[stable(feature = "rust1", since = "1.0.0")] pub fn write_fmt(&mut self, fmt: Arguments) -> Result { write(self.buf, fmt) @@ -1354,48 +1399,243 @@ impl<'a> Formatter<'a> { or `sign_aware_zero_pad` methods instead")] pub fn flags(&self) -> u32 { self.flags } - /// Character used as 'fill' whenever there is alignment + /// Character used as 'fill' whenever there is alignment. + /// + /// # Examples + /// + /// ``` + /// use std::fmt; + /// + /// struct Foo; + /// + /// impl fmt::Display for Foo { + /// fn fmt(&self, formatter: &mut fmt::Formatter) -> fmt::Result { + /// let c = formatter.fill(); + /// if let Some(width) = formatter.width() { + /// for _ in 0..width { + /// write!(formatter, "{}", c)?; + /// } + /// Ok(()) + /// } else { + /// write!(formatter, "{}", c) + /// } + /// } + /// } + /// + /// // We set alignment to the left with ">". + /// assert_eq!(&format!("{:G>3}", Foo), "GGG"); + /// assert_eq!(&format!("{:t>6}", Foo), "tttttt"); + /// ``` #[stable(feature = "fmt_flags", since = "1.5.0")] pub fn fill(&self) -> char { self.fill } - /// Flag indicating what form of alignment was requested - #[unstable(feature = "fmt_flags_align", reason = "method was just created", - issue = "27726")] - pub fn align(&self) -> Alignment { + /// Flag indicating what form of alignment was requested. + /// + /// # Examples + /// + /// ``` + /// extern crate core; + /// + /// use std::fmt::{self, Alignment}; + /// + /// struct Foo; + /// + /// impl fmt::Display for Foo { + /// fn fmt(&self, formatter: &mut fmt::Formatter) -> fmt::Result { + /// let s = if let Some(s) = formatter.align() { + /// match s { + /// Alignment::Left => "left", + /// Alignment::Right => "right", + /// Alignment::Center => "center", + /// } + /// } else { + /// "into the void" + /// }; + /// write!(formatter, "{}", s) + /// } + /// } + /// + /// fn main() { + /// assert_eq!(&format!("{:<}", Foo), "left"); + /// assert_eq!(&format!("{:>}", Foo), "right"); + /// assert_eq!(&format!("{:^}", Foo), "center"); + /// assert_eq!(&format!("{}", Foo), "into the void"); + /// } + /// ``` + #[stable(feature = "fmt_flags_align", since = "1.28.0")] + pub fn align(&self) -> Option<Alignment> { match self.align { - rt::v1::Alignment::Left => Alignment::Left, - rt::v1::Alignment::Right => Alignment::Right, - rt::v1::Alignment::Center => Alignment::Center, - rt::v1::Alignment::Unknown => Alignment::Unknown, + rt::v1::Alignment::Left => Some(Alignment::Left), + rt::v1::Alignment::Right => Some(Alignment::Right), + rt::v1::Alignment::Center => Some(Alignment::Center), + rt::v1::Alignment::Unknown => None, } } - /// Optionally specified integer width that the output should be + /// Optionally specified integer width that the output should be. + /// + /// # Examples + /// + /// ``` + /// use std::fmt; + /// + /// struct Foo(i32); + /// + /// impl fmt::Display for Foo { + /// fn fmt(&self, formatter: &mut fmt::Formatter) -> fmt::Result { + /// if let Some(width) = formatter.width() { + /// // If we received a width, we use it + /// write!(formatter, "{:width$}", &format!("Foo({})", self.0), width = width) + /// } else { + /// // Otherwise we do nothing special + /// write!(formatter, "Foo({})", self.0) + /// } + /// } + /// } + /// + /// assert_eq!(&format!("{:10}", Foo(23)), "Foo(23) "); + /// assert_eq!(&format!("{}", Foo(23)), "Foo(23)"); + /// ``` #[stable(feature = "fmt_flags", since = "1.5.0")] pub fn width(&self) -> Option<usize> { self.width } - /// Optionally specified precision for numeric types + /// Optionally specified precision for numeric types. + /// + /// # Examples + /// + /// ``` + /// use std::fmt; + /// + /// struct Foo(f32); + /// + /// impl fmt::Display for Foo { + /// fn fmt(&self, formatter: &mut fmt::Formatter) -> fmt::Result { + /// if let Some(precision) = formatter.precision() { + /// // If we received a precision, we use it. + /// write!(formatter, "Foo({1:.*})", precision, self.0) + /// } else { + /// // Otherwise we default to 2. + /// write!(formatter, "Foo({:.2})", self.0) + /// } + /// } + /// } + /// + /// assert_eq!(&format!("{:.4}", Foo(23.2)), "Foo(23.2000)"); + /// assert_eq!(&format!("{}", Foo(23.2)), "Foo(23.20)"); + /// ``` #[stable(feature = "fmt_flags", since = "1.5.0")] pub fn precision(&self) -> Option<usize> { self.precision } /// Determines if the `+` flag was specified. + /// + /// # Examples + /// + /// ``` + /// use std::fmt; + /// + /// struct Foo(i32); + /// + /// impl fmt::Display for Foo { + /// fn fmt(&self, formatter: &mut fmt::Formatter) -> fmt::Result { + /// if formatter.sign_plus() { + /// write!(formatter, + /// "Foo({}{})", + /// if self.0 < 0 { '-' } else { '+' }, + /// self.0) + /// } else { + /// write!(formatter, "Foo({})", self.0) + /// } + /// } + /// } + /// + /// assert_eq!(&format!("{:+}", Foo(23)), "Foo(+23)"); + /// assert_eq!(&format!("{}", Foo(23)), "Foo(23)"); + /// ``` #[stable(feature = "fmt_flags", since = "1.5.0")] pub fn sign_plus(&self) -> bool { self.flags & (1 << FlagV1::SignPlus as u32) != 0 } /// Determines if the `-` flag was specified. + /// + /// # Examples + /// + /// ``` + /// use std::fmt; + /// + /// struct Foo(i32); + /// + /// impl fmt::Display for Foo { + /// fn fmt(&self, formatter: &mut fmt::Formatter) -> fmt::Result { + /// if formatter.sign_minus() { + /// // You want a minus sign? Have one! + /// write!(formatter, "-Foo({})", self.0) + /// } else { + /// write!(formatter, "Foo({})", self.0) + /// } + /// } + /// } + /// + /// assert_eq!(&format!("{:-}", Foo(23)), "-Foo(23)"); + /// assert_eq!(&format!("{}", Foo(23)), "Foo(23)"); + /// ``` #[stable(feature = "fmt_flags", since = "1.5.0")] pub fn sign_minus(&self) -> bool { self.flags & (1 << FlagV1::SignMinus as u32) != 0 } /// Determines if the `#` flag was specified. + /// + /// # Examples + /// + /// ``` + /// use std::fmt; + /// + /// struct Foo(i32); + /// + /// impl fmt::Display for Foo { + /// fn fmt(&self, formatter: &mut fmt::Formatter) -> fmt::Result { + /// if formatter.alternate() { + /// write!(formatter, "Foo({})", self.0) + /// } else { + /// write!(formatter, "{}", self.0) + /// } + /// } + /// } + /// + /// assert_eq!(&format!("{:#}", Foo(23)), "Foo(23)"); + /// assert_eq!(&format!("{}", Foo(23)), "23"); + /// ``` #[stable(feature = "fmt_flags", since = "1.5.0")] pub fn alternate(&self) -> bool { self.flags & (1 << FlagV1::Alternate as u32) != 0 } /// Determines if the `0` flag was specified. + /// + /// # Examples + /// + /// ``` + /// use std::fmt; + /// + /// struct Foo(i32); + /// + /// impl fmt::Display for Foo { + /// fn fmt(&self, formatter: &mut fmt::Formatter) -> fmt::Result { + /// assert!(formatter.sign_aware_zero_pad()); + /// assert_eq!(formatter.width(), Some(4)); + /// // We ignore the formatter's options. + /// write!(formatter, "{}", self.0) + /// } + /// } + /// + /// assert_eq!(&format!("{:04}", Foo(23)), "23"); + /// ``` #[stable(feature = "fmt_flags", since = "1.5.0")] pub fn sign_aware_zero_pad(&self) -> bool { self.flags & (1 << FlagV1::SignAwareZeroPad as u32) != 0 } + // FIXME: Decide what public API we want for these two flags. + // https://github.com/rust-lang/rust/issues/48584 + fn debug_lower_hex(&self) -> bool { self.flags & (1 << FlagV1::DebugLowerHex as u32) != 0 } + + fn debug_upper_hex(&self) -> bool { self.flags & (1 << FlagV1::DebugUpperHex as u32) != 0 } + /// Creates a [`DebugStruct`] builder designed to assist with creation of /// [`fmt::Debug`] implementations for structs. /// @@ -1406,10 +1646,12 @@ impl<'a> Formatter<'a> { /// /// ```rust /// use std::fmt; + /// use std::net::Ipv4Addr; /// /// struct Foo { /// bar: i32, /// baz: String, + /// addr: Ipv4Addr, /// } /// /// impl fmt::Debug for Foo { @@ -1417,12 +1659,19 @@ impl<'a> Formatter<'a> { /// fmt.debug_struct("Foo") /// .field("bar", &self.bar) /// .field("baz", &self.baz) + /// .field("addr", &format_args!("{}", self.addr)) /// .finish() /// } /// } /// - /// // prints "Foo { bar: 10, baz: "Hello World" }" - /// println!("{:?}", Foo { bar: 10, baz: "Hello World".to_string() }); + /// assert_eq!( + /// "Foo { bar: 10, baz: \"Hello World\", addr: 127.0.0.1 }", + /// format!("{:?}", Foo { + /// bar: 10, + /// baz: "Hello World".to_string(), + /// addr: Ipv4Addr::new(127, 0, 0, 1), + /// }) + /// ); /// ``` #[stable(feature = "debug_builders", since = "1.2.0")] pub fn debug_struct<'b>(&'b mut self, name: &str) -> DebugStruct<'b, 'a> { @@ -1436,20 +1685,24 @@ impl<'a> Formatter<'a> { /// /// ```rust /// use std::fmt; + /// use std::marker::PhantomData; /// - /// struct Foo(i32, String); + /// struct Foo<T>(i32, String, PhantomData<T>); /// - /// impl fmt::Debug for Foo { + /// impl<T> fmt::Debug for Foo<T> { /// fn fmt(&self, fmt: &mut fmt::Formatter) -> fmt::Result { /// fmt.debug_tuple("Foo") /// .field(&self.0) /// .field(&self.1) + /// .field(&format_args!("_")) /// .finish() /// } /// } /// - /// // prints "Foo(10, "Hello World")" - /// println!("{:?}", Foo(10, "Hello World".to_string())); + /// assert_eq!( + /// "Foo(10, \"Hello\", _)", + /// format!("{:?}", Foo(10, "Hello".to_string(), PhantomData::<u8>)) + /// ); /// ``` #[stable(feature = "debug_builders", since = "1.2.0")] pub fn debug_tuple<'b>(&'b mut self, name: &str) -> DebugTuple<'b, 'a> { @@ -1499,6 +1752,41 @@ impl<'a> Formatter<'a> { /// // prints "{10, 11}" /// println!("{:?}", Foo(vec![10, 11])); /// ``` + /// + /// [`format_args!`]: ../../std/macro.format_args.html + /// + /// In this more complex example, we use [`format_args!`] and `.debug_set()` + /// to build a list of match arms: + /// + /// ```rust + /// use std::fmt; + /// + /// struct Arm<'a, L: 'a, R: 'a>(&'a (L, R)); + /// struct Table<'a, K: 'a, V: 'a>(&'a [(K, V)], V); + /// + /// impl<'a, L, R> fmt::Debug for Arm<'a, L, R> + /// where + /// L: 'a + fmt::Debug, R: 'a + fmt::Debug + /// { + /// fn fmt(&self, fmt: &mut fmt::Formatter) -> fmt::Result { + /// L::fmt(&(self.0).0, fmt)?; + /// fmt.write_str(" => ")?; + /// R::fmt(&(self.0).1, fmt) + /// } + /// } + /// + /// impl<'a, K, V> fmt::Debug for Table<'a, K, V> + /// where + /// K: 'a + fmt::Debug, V: 'a + fmt::Debug + /// { + /// fn fmt(&self, fmt: &mut fmt::Formatter) -> fmt::Result { + /// fmt.debug_set() + /// .entries(self.0.iter().map(Arm)) + /// .entry(&Arm(&(format_args!("_"), &self.1))) + /// .finish() + /// } + /// } + /// ``` #[stable(feature = "debug_builders", since = "1.2.0")] pub fn debug_set<'b>(&'b mut self) -> DebugSet<'b, 'a> { builders::debug_set_new(self) @@ -1586,6 +1874,7 @@ impl Display for ! { #[stable(feature = "rust1", since = "1.0.0")] impl Debug for bool { + #[inline] fn fmt(&self, f: &mut Formatter) -> Result { Display::fmt(self, f) } @@ -1748,6 +2037,7 @@ impl<T: Debug> Debug for [T] { #[stable(feature = "rust1", since = "1.0.0")] impl Debug for () { + #[inline] fn fmt(&self, f: &mut Formatter) -> Result { f.pad("()") } diff --git a/src/libcore/fmt/num.rs b/src/libcore/fmt/num.rs index ee989854a37..51391fa50d5 100644 --- a/src/libcore/fmt/num.rs +++ b/src/libcore/fmt/num.rs @@ -49,15 +49,13 @@ doit! { i8 i16 i32 i64 i128 isize u8 u16 u32 u64 u128 usize } #[doc(hidden)] trait GenericRadix { /// The number of digits. - fn base(&self) -> u8; + const BASE: u8; /// A radix-specific prefix string. - fn prefix(&self) -> &'static str { - "" - } + const PREFIX: &'static str; /// Converts an integer to corresponding radix digit. - fn digit(&self, x: u8) -> u8; + fn digit(x: u8) -> u8; /// Format an integer using the radix using a formatter. fn fmt_int<T: Int>(&self, mut x: T, f: &mut fmt::Formatter) -> fmt::Result { @@ -65,16 +63,16 @@ trait GenericRadix { // characters for a base 2 number. let zero = T::zero(); let is_nonnegative = x >= zero; - let mut buf = [0; 128]; + let mut buf: [u8; 128] = unsafe { mem::uninitialized() }; let mut curr = buf.len(); - let base = T::from_u8(self.base()); + let base = T::from_u8(Self::BASE); if is_nonnegative { // Accumulate each digit of the number from the least significant // to the most significant figure. for byte in buf.iter_mut().rev() { - let n = x % base; // Get the current place value. - x = x / base; // Deaccumulate the number. - *byte = self.digit(n.to_u8()); // Store the digit in the buffer. + let n = x % base; // Get the current place value. + x = x / base; // Deaccumulate the number. + *byte = Self::digit(n.to_u8()); // Store the digit in the buffer. curr -= 1; if x == zero { // No more digits left to accumulate. @@ -84,9 +82,9 @@ trait GenericRadix { } else { // Do the same as above, but accounting for two's complement. for byte in buf.iter_mut().rev() { - let n = zero - (x % base); // Get the current place value. - x = x / base; // Deaccumulate the number. - *byte = self.digit(n.to_u8()); // Store the digit in the buffer. + let n = zero - (x % base); // Get the current place value. + x = x / base; // Deaccumulate the number. + *byte = Self::digit(n.to_u8()); // Store the digit in the buffer. curr -= 1; if x == zero { // No more digits left to accumulate. @@ -95,7 +93,7 @@ trait GenericRadix { } } let buf = unsafe { str::from_utf8_unchecked(&buf[curr..]) }; - f.pad_integral(is_nonnegative, self.prefix(), buf) + f.pad_integral(is_nonnegative, Self::PREFIX, buf) } } @@ -107,10 +105,6 @@ struct Binary; #[derive(Clone, PartialEq)] struct Octal; -/// A decimal (base 10) radix -#[derive(Clone, PartialEq)] -struct Decimal; - /// A hexadecimal (base 16) radix, formatted with lower-case characters #[derive(Clone, PartialEq)] struct LowerHex; @@ -122,25 +116,24 @@ struct UpperHex; macro_rules! radix { ($T:ident, $base:expr, $prefix:expr, $($x:pat => $conv:expr),+) => { impl GenericRadix for $T { - fn base(&self) -> u8 { $base } - fn prefix(&self) -> &'static str { $prefix } - fn digit(&self, x: u8) -> u8 { + const BASE: u8 = $base; + const PREFIX: &'static str = $prefix; + fn digit(x: u8) -> u8 { match x { $($x => $conv,)+ - x => panic!("number not in the range 0..{}: {}", self.base() - 1, x), + x => panic!("number not in the range 0..={}: {}", Self::BASE - 1, x), } } } } } -radix! { Binary, 2, "0b", x @ 0 ... 1 => b'0' + x } -radix! { Octal, 8, "0o", x @ 0 ... 7 => b'0' + x } -radix! { Decimal, 10, "", x @ 0 ... 9 => b'0' + x } -radix! { LowerHex, 16, "0x", x @ 0 ... 9 => b'0' + x, - x @ 10 ... 15 => b'a' + (x - 10) } -radix! { UpperHex, 16, "0x", x @ 0 ... 9 => b'0' + x, - x @ 10 ... 15 => b'A' + (x - 10) } +radix! { Binary, 2, "0b", x @ 0 ..= 1 => b'0' + x } +radix! { Octal, 8, "0o", x @ 0 ..= 7 => b'0' + x } +radix! { LowerHex, 16, "0x", x @ 0 ..= 9 => b'0' + x, + x @ 10 ..= 15 => b'a' + (x - 10) } +radix! { UpperHex, 16, "0x", x @ 0 ..= 9 => b'0' + x, + x @ 10 ..= 15 => b'A' + (x - 10) } macro_rules! int_base { ($Trait:ident for $T:ident as $U:ident -> $Radix:ident) => { @@ -157,8 +150,15 @@ macro_rules! debug { ($T:ident) => { #[stable(feature = "rust1", since = "1.0.0")] impl fmt::Debug for $T { + #[inline] fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result { - fmt::Display::fmt(self, f) + if f.debug_lower_hex() { + fmt::LowerHex::fmt(self, f) + } else if f.debug_upper_hex() { + fmt::UpperHex::fmt(self, f) + } else { + fmt::Display::fmt(self, f) + } } } } diff --git a/src/libcore/future/future.rs b/src/libcore/future/future.rs new file mode 100644 index 00000000000..10b4ca9b0b2 --- /dev/null +++ b/src/libcore/future/future.rs @@ -0,0 +1,112 @@ +// Copyright 2018 The Rust Project Developers. See the COPYRIGHT +// file at the top-level directory of this distribution and at +// http://rust-lang.org/COPYRIGHT. +// +// Licensed under the Apache License, Version 2.0 <LICENSE-APACHE or +// http://www.apache.org/licenses/LICENSE-2.0> or the MIT license +// <LICENSE-MIT or http://opensource.org/licenses/MIT>, at your +// option. This file may not be copied, modified, or distributed +// except according to those terms. + +#![unstable(feature = "futures_api", + reason = "futures in libcore are unstable", + issue = "50547")] + +use mem::PinMut; +use marker::Unpin; +use task::{self, Poll}; + +/// A future represents an asychronous computation. +/// +/// A future is a value that may not have finished computing yet. This kind of +/// "asynchronous value" makes it possible for a thread to continue doing useful +/// work while it waits for the value to become available. +/// +/// # The `poll` method +/// +/// The core method of future, `poll`, *attempts* to resolve the future into a +/// final value. This method does not block if the value is not ready. Instead, +/// the current task is scheduled to be woken up when it's possible to make +/// further progress by `poll`ing again. The wake up is performed using +/// `cx.waker()`, a handle for waking up the current task. +/// +/// When using a future, you generally won't call `poll` directly, but instead +/// `await!` the value. +pub trait Future { + /// The result of the `Future`. + type Output; + + /// Attempt to resolve the future to a final value, registering + /// the current task for wakeup if the value is not yet available. + /// + /// # Return value + /// + /// This function returns: + /// + /// - [`Poll::Pending`] if the future is not ready yet + /// - [`Poll::Ready(val)`] with the result `val` of this future if it + /// finished successfully. + /// + /// Once a future has finished, clients should not `poll` it again. + /// + /// When a future is not ready yet, `poll` returns + /// `Poll::Pending`. The future will *also* register the + /// interest of the current task in the value being produced. For example, + /// if the future represents the availability of data on a socket, then the + /// task is recorded so that when data arrives, it is woken up (via + /// [`cx.waker()`]). Once a task has been woken up, + /// it should attempt to `poll` the future again, which may or may not + /// produce a final value. + /// + /// Note that if `Pending` is returned it only means that the *current* task + /// (represented by the argument `cx`) will receive a notification. Tasks + /// from previous calls to `poll` will *not* receive notifications. + /// + /// # Runtime characteristics + /// + /// Futures alone are *inert*; they must be *actively* `poll`ed to make + /// progress, meaning that each time the current task is woken up, it should + /// actively re-`poll` pending futures that it still has an interest in. + /// + /// The `poll` function is not called repeatedly in a tight loop for + /// futures, but only whenever the future itself is ready, as signaled via + /// the `Waker` inside `task::Context`. If you're familiar with the + /// `poll(2)` or `select(2)` syscalls on Unix it's worth noting that futures + /// typically do *not* suffer the same problems of "all wakeups must poll + /// all events"; they are more like `epoll(4)`. + /// + /// An implementation of `poll` should strive to return quickly, and must + /// *never* block. Returning quickly prevents unnecessarily clogging up + /// threads or event loops. If it is known ahead of time that a call to + /// `poll` may end up taking awhile, the work should be offloaded to a + /// thread pool (or something similar) to ensure that `poll` can return + /// quickly. + /// + /// # Panics + /// + /// Once a future has completed (returned `Ready` from `poll`), + /// then any future calls to `poll` may panic, block forever, or otherwise + /// cause bad behavior. The `Future` trait itself provides no guarantees + /// about the behavior of `poll` after a future has completed. + /// + /// [`Poll::Pending`]: ../task/enum.Poll.html#variant.Pending + /// [`Poll::Ready(val)`]: ../task/enum.Poll.html#variant.Ready + /// [`cx.waker()`]: ../task/struct.Context.html#method.waker + fn poll(self: PinMut<Self>, cx: &mut task::Context) -> Poll<Self::Output>; +} + +impl<'a, F: ?Sized + Future + Unpin> Future for &'a mut F { + type Output = F::Output; + + fn poll(mut self: PinMut<Self>, cx: &mut task::Context) -> Poll<Self::Output> { + F::poll(PinMut::new(&mut **self), cx) + } +} + +impl<'a, F: ?Sized + Future> Future for PinMut<'a, F> { + type Output = F::Output; + + fn poll(mut self: PinMut<Self>, cx: &mut task::Context) -> Poll<Self::Output> { + F::poll((*self).reborrow(), cx) + } +} diff --git a/src/libcore/future/future_obj.rs b/src/libcore/future/future_obj.rs new file mode 100644 index 00000000000..6045fac2b4b --- /dev/null +++ b/src/libcore/future/future_obj.rs @@ -0,0 +1,182 @@ +// Copyright 2018 The Rust Project Developers. See the COPYRIGHT +// file at the top-level directory of this distribution and at +// http://rust-lang.org/COPYRIGHT. +// +// Licensed under the Apache License, Version 2.0 <LICENSE-APACHE or +// http://www.apache.org/licenses/LICENSE-2.0> or the MIT license +// <LICENSE-MIT or http://opensource.org/licenses/MIT>, at your +// option. This file may not be copied, modified, or distributed +// except according to those terms. + +#![unstable(feature = "futures_api", + reason = "futures in libcore are unstable", + issue = "50547")] + +use fmt; +use future::Future; +use marker::{PhantomData, Unpin}; +use mem::PinMut; +use task::{Context, Poll}; + +/// A custom trait object for polling futures, roughly akin to +/// `Box<dyn Future<Output = T> + 'a>`. +/// +/// This custom trait object was introduced for two reasons: +/// - Currently it is not possible to take `dyn Trait` by value and +/// `Box<dyn Trait>` is not available in no_std contexts. +/// - The `Future` trait is currently not object safe: The `Future::poll` +/// method makes uses the arbitrary self types feature and traits in which +/// this feature is used are currently not object safe due to current compiler +/// limitations. (See tracking issue for arbitray self types for more +/// information #44874) +pub struct LocalFutureObj<'a, T> { + ptr: *mut (), + poll_fn: unsafe fn(*mut (), &mut Context) -> Poll<T>, + drop_fn: unsafe fn(*mut ()), + _marker: PhantomData<&'a ()>, +} + +impl<'a, T> Unpin for LocalFutureObj<'a, T> {} + +impl<'a, T> LocalFutureObj<'a, T> { + /// Create a `LocalFutureObj` from a custom trait object representation. + #[inline] + pub fn new<F: UnsafeFutureObj<'a, T> + 'a>(f: F) -> LocalFutureObj<'a, T> { + LocalFutureObj { + ptr: f.into_raw(), + poll_fn: F::poll, + drop_fn: F::drop, + _marker: PhantomData, + } + } + + /// Converts the `LocalFutureObj` into a `FutureObj` + /// To make this operation safe one has to ensure that the `UnsafeFutureObj` + /// instance from which this `LocalFutureObj` was created actually + /// implements `Send`. + #[inline] + pub unsafe fn into_future_obj(self) -> FutureObj<'a, T> { + FutureObj(self) + } +} + +impl<'a, T> fmt::Debug for LocalFutureObj<'a, T> { + fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result { + f.debug_struct("LocalFutureObj") + .finish() + } +} + +impl<'a, T> From<FutureObj<'a, T>> for LocalFutureObj<'a, T> { + #[inline] + fn from(f: FutureObj<'a, T>) -> LocalFutureObj<'a, T> { + f.0 + } +} + +impl<'a, T> Future for LocalFutureObj<'a, T> { + type Output = T; + + #[inline] + fn poll(self: PinMut<Self>, cx: &mut Context) -> Poll<T> { + unsafe { + (self.poll_fn)(self.ptr, cx) + } + } +} + +impl<'a, T> Drop for LocalFutureObj<'a, T> { + fn drop(&mut self) { + unsafe { + (self.drop_fn)(self.ptr) + } + } +} + +/// A custom trait object for polling futures, roughly akin to +/// `Box<dyn Future<Output = T> + Send + 'a>`. +/// +/// This custom trait object was introduced for two reasons: +/// - Currently it is not possible to take `dyn Trait` by value and +/// `Box<dyn Trait>` is not available in no_std contexts. +/// - The `Future` trait is currently not object safe: The `Future::poll` +/// method makes uses the arbitrary self types feature and traits in which +/// this feature is used are currently not object safe due to current compiler +/// limitations. (See tracking issue for arbitray self types for more +/// information #44874) +pub struct FutureObj<'a, T>(LocalFutureObj<'a, T>); + +impl<'a, T> Unpin for FutureObj<'a, T> {} +unsafe impl<'a, T> Send for FutureObj<'a, T> {} + +impl<'a, T> FutureObj<'a, T> { + /// Create a `FutureObj` from a custom trait object representation. + #[inline] + pub fn new<F: UnsafeFutureObj<'a, T> + Send>(f: F) -> FutureObj<'a, T> { + FutureObj(LocalFutureObj::new(f)) + } +} + +impl<'a, T> fmt::Debug for FutureObj<'a, T> { + fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result { + f.debug_struct("FutureObj") + .finish() + } +} + +impl<'a, T> Future for FutureObj<'a, T> { + type Output = T; + + #[inline] + fn poll(self: PinMut<Self>, cx: &mut Context) -> Poll<T> { + let pinned_field = unsafe { PinMut::map_unchecked(self, |x| &mut x.0) }; + pinned_field.poll(cx) + } +} + +/// A custom implementation of a future trait object for `FutureObj`, providing +/// a hand-rolled vtable. +/// +/// This custom representation is typically used only in `no_std` contexts, +/// where the default `Box`-based implementation is not available. +/// +/// The implementor must guarantee that it is safe to call `poll` repeatedly (in +/// a non-concurrent fashion) with the result of `into_raw` until `drop` is +/// called. +pub unsafe trait UnsafeFutureObj<'a, T>: 'a { + /// Convert an owned instance into a (conceptually owned) void pointer. + fn into_raw(self) -> *mut (); + + /// Poll the future represented by the given void pointer. + /// + /// # Safety + /// + /// The trait implementor must guarantee that it is safe to repeatedly call + /// `poll` with the result of `into_raw` until `drop` is called; such calls + /// are not, however, allowed to race with each other or with calls to + /// `drop`. + unsafe fn poll(ptr: *mut (), cx: &mut Context) -> Poll<T>; + + /// Drops the future represented by the given void pointer. + /// + /// # Safety + /// + /// The trait implementor must guarantee that it is safe to call this + /// function once per `into_raw` invocation; that call cannot race with + /// other calls to `drop` or `poll`. + unsafe fn drop(ptr: *mut ()); +} + +unsafe impl<'a, T, F> UnsafeFutureObj<'a, T> for &'a mut F + where F: Future<Output = T> + Unpin + 'a +{ + fn into_raw(self) -> *mut () { + self as *mut F as *mut () + } + + unsafe fn poll(ptr: *mut (), cx: &mut Context) -> Poll<T> { + PinMut::new_unchecked(&mut *(ptr as *mut F)).poll(cx) + } + + unsafe fn drop(_ptr: *mut ()) {} +} diff --git a/src/libcore/future/mod.rs b/src/libcore/future/mod.rs new file mode 100644 index 00000000000..f9361a0f4e7 --- /dev/null +++ b/src/libcore/future/mod.rs @@ -0,0 +1,21 @@ +// Copyright 2018 The Rust Project Developers. See the COPYRIGHT +// file at the top-level directory of this distribution and at +// http://rust-lang.org/COPYRIGHT. +// +// Licensed under the Apache License, Version 2.0 <LICENSE-APACHE or +// http://www.apache.org/licenses/LICENSE-2.0> or the MIT license +// <LICENSE-MIT or http://opensource.org/licenses/MIT>, at your +// option. This file may not be copied, modified, or distributed +// except according to those terms. + +#![unstable(feature = "futures_api", + reason = "futures in libcore are unstable", + issue = "50547")] + +//! Asynchronous values. + +mod future; +pub use self::future::Future; + +mod future_obj; +pub use self::future_obj::{FutureObj, LocalFutureObj, UnsafeFutureObj}; diff --git a/src/libcore/hash/mod.rs b/src/libcore/hash/mod.rs index 15545a04b64..e7907e03444 100644 --- a/src/libcore/hash/mod.rs +++ b/src/libcore/hash/mod.rs @@ -99,9 +99,10 @@ use mem; #[allow(deprecated)] pub use self::sip::SipHasher; -#[unstable(feature = "sip_hash_13", issue = "34767")] +#[unstable(feature = "hashmap_internals", issue = "0")] #[allow(deprecated)] -pub use self::sip::{SipHasher13, SipHasher24}; +#[doc(hidden)] +pub use self::sip::SipHasher13; mod sip; @@ -307,7 +308,7 @@ pub trait Hasher { } /// Writes a single `u128` into this hasher. #[inline] - #[unstable(feature = "i128", issue = "35118")] + #[stable(feature = "i128", since = "1.26.0")] fn write_u128(&mut self, i: u128) { self.write(&unsafe { mem::transmute::<_, [u8; 16]>(i) }) } @@ -347,7 +348,7 @@ pub trait Hasher { } /// Writes a single `i128` into this hasher. #[inline] - #[unstable(feature = "i128", issue = "35118")] + #[stable(feature = "i128", since = "1.26.0")] fn write_i128(&mut self, i: i128) { self.write_u128(i as u128) } @@ -541,6 +542,16 @@ impl<H> Default for BuildHasherDefault<H> { } } +#[stable(since = "1.29.0", feature = "build_hasher_eq")] +impl<H> PartialEq for BuildHasherDefault<H> { + fn eq(&self, _other: &BuildHasherDefault<H>) -> bool { + true + } +} + +#[stable(since = "1.29.0", feature = "build_hasher_eq")] +impl<H> Eq for BuildHasherDefault<H> {} + ////////////////////////////////////////////////////////////////////////////// mod impls { @@ -602,6 +613,13 @@ mod impls { } } + #[stable(feature = "never_hash", since = "1.29.0")] + impl Hash for ! { + fn hash<H: Hasher>(&self, _: &mut H) { + *self + } + } + macro_rules! impl_hash_tuple { () => ( #[stable(feature = "rust1", since = "1.0.0")] diff --git a/src/libcore/hash/sip.rs b/src/libcore/hash/sip.rs index 4e4d9b3f1e2..e3bdecdc4b1 100644 --- a/src/libcore/hash/sip.rs +++ b/src/libcore/hash/sip.rs @@ -23,10 +23,11 @@ use mem; /// (eg. `collections::HashMap` uses it by default). /// /// See: <https://131002.net/siphash> -#[unstable(feature = "sip_hash_13", issue = "34767")] +#[unstable(feature = "hashmap_internals", issue = "0")] #[rustc_deprecated(since = "1.13.0", reason = "use `std::collections::hash_map::DefaultHasher` instead")] #[derive(Debug, Clone, Default)] +#[doc(hidden)] pub struct SipHasher13 { hasher: Hasher<Sip13Rounds>, } @@ -34,11 +35,11 @@ pub struct SipHasher13 { /// An implementation of SipHash 2-4. /// /// See: <https://131002.net/siphash/> -#[unstable(feature = "sip_hash_13", issue = "34767")] +#[unstable(feature = "hashmap_internals", issue = "0")] #[rustc_deprecated(since = "1.13.0", reason = "use `std::collections::hash_map::DefaultHasher` instead")] #[derive(Debug, Clone, Default)] -pub struct SipHasher24 { +struct SipHasher24 { hasher: Hasher<Sip24Rounds>, } @@ -156,14 +157,16 @@ impl SipHasher { #[rustc_deprecated(since = "1.13.0", reason = "use `std::collections::hash_map::DefaultHasher` instead")] pub fn new_with_keys(key0: u64, key1: u64) -> SipHasher { - SipHasher(SipHasher24::new_with_keys(key0, key1)) + SipHasher(SipHasher24 { + hasher: Hasher::new_with_keys(key0, key1) + }) } } impl SipHasher13 { /// Creates a new `SipHasher13` with the two initial keys set to 0. #[inline] - #[unstable(feature = "sip_hash_13", issue = "34767")] + #[unstable(feature = "hashmap_internals", issue = "0")] #[rustc_deprecated(since = "1.13.0", reason = "use `std::collections::hash_map::DefaultHasher` instead")] pub fn new() -> SipHasher13 { @@ -172,7 +175,7 @@ impl SipHasher13 { /// Creates a `SipHasher13` that is keyed off the provided keys. #[inline] - #[unstable(feature = "sip_hash_13", issue = "34767")] + #[unstable(feature = "hashmap_internals", issue = "0")] #[rustc_deprecated(since = "1.13.0", reason = "use `std::collections::hash_map::DefaultHasher` instead")] pub fn new_with_keys(key0: u64, key1: u64) -> SipHasher13 { @@ -182,28 +185,6 @@ impl SipHasher13 { } } -impl SipHasher24 { - /// Creates a new `SipHasher24` with the two initial keys set to 0. - #[inline] - #[unstable(feature = "sip_hash_13", issue = "34767")] - #[rustc_deprecated(since = "1.13.0", - reason = "use `std::collections::hash_map::DefaultHasher` instead")] - pub fn new() -> SipHasher24 { - SipHasher24::new_with_keys(0, 0) - } - - /// Creates a `SipHasher24` that is keyed off the provided keys. - #[inline] - #[unstable(feature = "sip_hash_13", issue = "34767")] - #[rustc_deprecated(since = "1.13.0", - reason = "use `std::collections::hash_map::DefaultHasher` instead")] - pub fn new_with_keys(key0: u64, key1: u64) -> SipHasher24 { - SipHasher24 { - hasher: Hasher::new_with_keys(key0, key1) - } - } -} - impl<S: Sip> Hasher<S> { #[inline] fn new_with_keys(key0: u64, key1: u64) -> Hasher<S> { @@ -271,16 +252,16 @@ impl<S: Sip> Hasher<S> { impl super::Hasher for SipHasher { #[inline] fn write(&mut self, msg: &[u8]) { - self.0.write(msg) + self.0.hasher.write(msg) } #[inline] fn finish(&self) -> u64 { - self.0.finish() + self.0.hasher.finish() } } -#[unstable(feature = "sip_hash_13", issue = "34767")] +#[unstable(feature = "hashmap_internals", issue = "0")] impl super::Hasher for SipHasher13 { #[inline] fn write(&mut self, msg: &[u8]) { @@ -293,19 +274,6 @@ impl super::Hasher for SipHasher13 { } } -#[unstable(feature = "sip_hash_13", issue = "34767")] -impl super::Hasher for SipHasher24 { - #[inline] - fn write(&mut self, msg: &[u8]) { - self.hasher.write(msg) - } - - #[inline] - fn finish(&self) -> u64 { - self.hasher.finish() - } -} - impl<S: Sip> super::Hasher for Hasher<S> { // see short_write comment for explanation #[inline] diff --git a/src/libcore/hint.rs b/src/libcore/hint.rs new file mode 100644 index 00000000000..f4e96e67b2c --- /dev/null +++ b/src/libcore/hint.rs @@ -0,0 +1,61 @@ +// Copyright 2018 The Rust Project Developers. See the COPYRIGHT +// file at the top-level directory of this distribution and at +// http://rust-lang.org/COPYRIGHT. +// +// Licensed under the Apache License, Version 2.0 <LICENSE-APACHE or +// http://www.apache.org/licenses/LICENSE-2.0> or the MIT license +// <LICENSE-MIT or http://opensource.org/licenses/MIT>, at your +// option. This file may not be copied, modified, or distributed +// except according to those terms. + +#![stable(feature = "core_hint", since = "1.27.0")] + +//! Hints to compiler that affects how code should be emitted or optimized. + +use intrinsics; + +/// Informs the compiler that this point in the code is not reachable, enabling +/// further optimizations. +/// +/// # Safety +/// +/// Reaching this function is completely *undefined behavior* (UB). In +/// particular, the compiler assumes that all UB must never happen, and +/// therefore will eliminate all branches that reach to a call to +/// `unreachable_unchecked()`. +/// +/// Like all instances of UB, if this assumption turns out to be wrong, i.e. the +/// `unreachable_unchecked()` call is actually reachable among all possible +/// control flow, the compiler will apply the wrong optimization strategy, and +/// may sometimes even corrupt seemingly unrelated code, causing +/// difficult-to-debug problems. +/// +/// Use this function only when you can prove that the code will never call it. +/// +/// The [`unreachable!()`] macro is the safe counterpart of this function, which +/// will panic instead when executed. +/// +/// [`unreachable!()`]: ../macro.unreachable.html +/// +/// # Example +/// +/// ``` +/// fn div_1(a: u32, b: u32) -> u32 { +/// use std::hint::unreachable_unchecked; +/// +/// // `b.saturating_add(1)` is always positive (not zero), +/// // hence `checked_div` will never return None. +/// // Therefore, the else branch is unreachable. +/// a.checked_div(b.saturating_add(1)) +/// .unwrap_or_else(|| unsafe { unreachable_unchecked() }) +/// } +/// +/// assert_eq!(div_1(7, 0), 7); +/// assert_eq!(div_1(9, 1), 4); +/// assert_eq!(div_1(11, std::u32::MAX), 0); +/// ``` +#[inline] +#[stable(feature = "unreachable", since = "1.27.0")] +pub unsafe fn unreachable_unchecked() -> ! { + intrinsics::unreachable() +} diff --git a/src/libcore/internal_macros.rs b/src/libcore/internal_macros.rs index cb215a38e53..db75f9bf210 100644 --- a/src/libcore/internal_macros.rs +++ b/src/libcore/internal_macros.rs @@ -86,4 +86,3 @@ macro_rules! forward_ref_op_assign { } } } - diff --git a/src/libcore/intrinsics.rs b/src/libcore/intrinsics.rs index a611dc02469..9ddf902349d 100644 --- a/src/libcore/intrinsics.rs +++ b/src/libcore/intrinsics.rs @@ -10,7 +10,7 @@ //! rustc compiler intrinsics. //! -//! The corresponding definitions are in librustc_trans/intrinsic.rs. +//! The corresponding definitions are in librustc_codegen_llvm/intrinsic.rs. //! //! # Volatiles //! @@ -638,6 +638,9 @@ extern "rust-intrinsic" { /// NB: This is very different from the `unreachable!()` macro: Unlike the /// macro, which panics when it is executed, it is *undefined behavior* to /// reach code marked with this function. + /// + /// The stabilized version of this intrinsic is + /// [`std::hint::unreachable_unchecked`](../../std/hint/fn.unreachable_unchecked.html). pub fn unreachable() -> !; /// Informs the optimizer that a condition is always true. @@ -714,7 +717,7 @@ extern "rust-intrinsic" { /// Reinterprets the bits of a value of one type as another type. /// /// Both types must have the same size. Neither the original, nor the result, - /// may be an [invalid value](../../nomicon/meet-safe-and-unsafe.html). + /// may be an [invalid value](../../nomicon/what-unsafe-does.html). /// /// `transmute` is semantically equivalent to a bitwise move of one type /// into another. It copies the bits from the source value into the @@ -992,7 +995,7 @@ extern "rust-intrinsic" { /// ptr::copy_nonoverlapping(y, x, 1); /// ptr::copy_nonoverlapping(&t, y, 1); /// - /// // y and t now point to the same thing, but we need to completely forget `tmp` + /// // y and t now point to the same thing, but we need to completely forget `t` /// // because it's no longer relevant. /// mem::forget(t); /// } @@ -1082,6 +1085,13 @@ extern "rust-intrinsic" { /// [`std::ptr::write_volatile`](../../std/ptr/fn.write_volatile.html). pub fn volatile_store<T>(dst: *mut T, val: T); + /// Perform a volatile load from the `src` pointer + /// The pointer is not required to be aligned. + pub fn unaligned_volatile_load<T>(src: *const T) -> T; + /// Perform a volatile store to the `dst` pointer. + /// The pointer is not required to be aligned. + pub fn unaligned_volatile_store<T>(dst: *mut T, val: T); + /// Returns the square root of an `f32` pub fn sqrtf32(x: f32) -> f32; /// Returns the square root of an `f64` @@ -1292,6 +1302,9 @@ extern "rust-intrinsic" { /// Reverses the bytes in an integer type `T`. pub fn bswap<T>(x: T) -> T; + /// Reverses the bits in an integer type `T`. + pub fn bitreverse<T>(x: T) -> T; + /// Performs checked integer addition. /// The stabilized versions of this intrinsic are available on the integer /// primitives via the `overflowing_add` method. For example, @@ -1310,6 +1323,10 @@ extern "rust-intrinsic" { /// [`std::u32::overflowing_mul`](../../std/primitive.u32.html#method.overflowing_mul) pub fn mul_with_overflow<T>(x: T, y: T) -> (T, bool); + /// Performs an exact division, resulting in undefined behavior where + /// `x % y != 0` or `y == 0` or `x == T::min_value() && y == -1` + pub fn exact_div<T>(x: T, y: T) -> T; + /// Performs an unchecked division, resulting in undefined behavior /// where y = 0 or x = `T::min_value()` and y = -1 pub fn unchecked_div<T>(x: T, y: T) -> T; @@ -1354,40 +1371,6 @@ extern "rust-intrinsic" { /// source as well as std's catch implementation. pub fn try(f: fn(*mut u8), data: *mut u8, local_ptr: *mut u8) -> i32; - /// Computes the byte offset that needs to be applied to `ptr` in order to - /// make it aligned to `align`. - /// If it is not possible to align `ptr`, the implementation returns - /// `usize::max_value()`. - /// - /// There are no guarantees whatsover that offsetting the pointer will not - /// overflow or go beyond the allocation that `ptr` points into. - /// It is up to the caller to ensure that the returned offset is correct - /// in all terms other than alignment. - /// - /// # Examples - /// - /// Accessing adjacent `u8` as `u16` - /// - /// ``` - /// # #![feature(core_intrinsics)] - /// # fn foo(n: usize) { - /// # use std::intrinsics::align_offset; - /// # use std::mem::align_of; - /// # unsafe { - /// let x = [5u8, 6u8, 7u8, 8u8, 9u8]; - /// let ptr = &x[n] as *const u8; - /// let offset = align_offset(ptr as *const (), align_of::<u16>()); - /// if offset < x.len() - n - 1 { - /// let u16_ptr = ptr.offset(offset as isize) as *const u16; - /// assert_ne!(*u16_ptr, 500); - /// } else { - /// // while the pointer can be aligned via `offset`, it would point - /// // outside the allocation - /// } - /// # } } - /// ``` - pub fn align_offset(ptr: *const (), align: usize) -> usize; - /// Emits a `!nontemporal` store according to LLVM (see their docs). /// Probably will never become stable. pub fn nontemporal_store<T>(ptr: *mut T, val: T); diff --git a/src/libcore/iter/iterator.rs b/src/libcore/iter/iterator.rs index 35cd7441c66..39185291053 100644 --- a/src/libcore/iter/iterator.rs +++ b/src/libcore/iter/iterator.rs @@ -11,13 +11,14 @@ use cmp::Ordering; use ops::Try; -use super::{AlwaysOk, LoopState}; -use super::{Chain, Cycle, Cloned, Enumerate, Filter, FilterMap, FlatMap, Fuse}; +use super::LoopState; +use super::{Chain, Cycle, Cloned, Enumerate, Filter, FilterMap, Fuse}; +use super::{Flatten, FlatMap, flatten_compat}; use super::{Inspect, Map, Peekable, Scan, Skip, SkipWhile, StepBy, Take, TakeWhile, Rev}; use super::{Zip, Sum, Product}; use super::{ChainState, FromIterator, ZipImpl}; -fn _assert_is_object_safe(_: &Iterator<Item=()>) {} +fn _assert_is_object_safe(_: &dyn Iterator<Item=()>) {} /// An interface for dealing with iterators. /// @@ -28,8 +29,13 @@ fn _assert_is_object_safe(_: &Iterator<Item=()>) {} /// [module-level documentation]: index.html /// [impl]: index.html#implementing-iterator #[stable(feature = "rust1", since = "1.0.0")] -#[rustc_on_unimplemented = "`{Self}` is not an iterator; maybe try calling \ - `.iter()` or a similar method"] +#[rustc_on_unimplemented( + on( + _Self="&str", + label="`{Self}` is not an iterator; try calling `.chars()` or `.bytes()`" + ), + label="`{Self}` is not an iterator; maybe try calling `.iter()` or a similar method" +)] #[doc(spotlight)] pub trait Iterator { /// The type of the elements being iterated over. @@ -163,7 +169,7 @@ pub trait Iterator { /// This function might panic if the iterator has more than [`usize::MAX`] /// elements. /// - /// [`usize::MAX`]: ../../std/isize/constant.MAX.html + /// [`usize::MAX`]: ../../std/usize/constant.MAX.html /// /// # Examples /// @@ -265,9 +271,30 @@ pub trait Iterator { /// Creates an iterator starting at the same point, but stepping by /// the given amount at each iteration. /// - /// Note that it will always return the first element of the iterator, + /// Note 1: The first element of the iterator will always be returned, /// regardless of the step given. /// + /// Note 2: The time at which ignored elements are pulled is not fixed. + /// `StepBy` behaves like the sequence `next(), nth(step-1), nth(step-1), …`, + /// but is also free to behave like the sequence + /// `advance_n_and_return_first(step), advance_n_and_return_first(step), …` + /// Which way is used may change for some iterators for performance reasons. + /// The second way will advance the iterator earlier and may consume more items. + /// + /// `advance_n_and_return_first` is the equivalent of: + /// ``` + /// fn advance_n_and_return_first<I>(iter: &mut I, total_step: usize) -> Option<I::Item> + /// where + /// I: Iterator, + /// { + /// let next = iter.next(); + /// if total_step > 1 { + /// iter.nth(total_step-2); + /// } + /// next + /// } + /// ``` + /// /// # Panics /// /// The method will panic if the given step is `0`. @@ -277,7 +304,6 @@ pub trait Iterator { /// Basic usage: /// /// ``` - /// #![feature(iterator_step_by)] /// let a = [0, 1, 2, 3, 4, 5]; /// let mut iter = a.into_iter().step_by(2); /// @@ -287,9 +313,7 @@ pub trait Iterator { /// assert_eq!(iter.next(), None); /// ``` #[inline] - #[unstable(feature = "iterator_step_by", - reason = "unstable replacement of Range::step_by", - issue = "27741")] + #[stable(feature = "iterator_step_by", since = "1.28.0")] fn step_by(self, step: usize) -> StepBy<Self> where Self: Sized { assert!(step != 0); StepBy{iter: self, step: step - 1, first_take: true} @@ -360,8 +384,9 @@ pub trait Iterator { /// /// In other words, it zips two iterators together, into a single one. /// - /// When either iterator returns [`None`], all further calls to [`next`] - /// will return [`None`]. + /// If either iterator returns [`None`], [`next`] from the zipped iterator + /// will return [`None`]. If the first iterator returns [`None`], `zip` will + /// short-circuit and `next` will not be called on the second iterator. /// /// # Examples /// @@ -482,7 +507,7 @@ pub trait Iterator { fn map<B, F>(self, f: F) -> Map<Self, F> where Self: Sized, F: FnMut(Self::Item) -> B, { - Map{iter: self, f: f} + Map { iter: self, f } } /// Calls a closure on each element of an iterator. @@ -593,7 +618,7 @@ pub trait Iterator { fn filter<P>(self, predicate: P) -> Filter<Self, P> where Self: Sized, P: FnMut(&Self::Item) -> bool, { - Filter{iter: self, predicate: predicate} + Filter {iter: self, predicate } } /// Creates an iterator that both filters and maps. @@ -650,7 +675,7 @@ pub trait Iterator { fn filter_map<B, F>(self, f: F) -> FilterMap<Self, F> where Self: Sized, F: FnMut(Self::Item) -> Option<B>, { - FilterMap { iter: self, f: f } + FilterMap { iter: self, f } } /// Creates an iterator which gives the current iteration count as well as @@ -803,7 +828,7 @@ pub trait Iterator { fn skip_while<P>(self, predicate: P) -> SkipWhile<Self, P> where Self: Sized, P: FnMut(&Self::Item) -> bool, { - SkipWhile{iter: self, flag: false, predicate: predicate} + SkipWhile { iter: self, flag: false, predicate } } /// Creates an iterator that yields elements based on a predicate. @@ -883,7 +908,7 @@ pub trait Iterator { fn take_while<P>(self, predicate: P) -> TakeWhile<Self, P> where Self: Sized, P: FnMut(&Self::Item) -> bool, { - TakeWhile{iter: self, flag: false, predicate: predicate} + TakeWhile { iter: self, flag: false, predicate } } /// Creates an iterator that skips the first `n` elements. @@ -905,7 +930,7 @@ pub trait Iterator { #[inline] #[stable(feature = "rust1", since = "1.0.0")] fn skip(self, n: usize) -> Skip<Self> where Self: Sized { - Skip{iter: self, n: n} + Skip { iter: self, n } } /// Creates an iterator that yields its first `n` elements. @@ -937,7 +962,7 @@ pub trait Iterator { #[inline] #[stable(feature = "rust1", since = "1.0.0")] fn take(self, n: usize) -> Take<Self> where Self: Sized, { - Take{iter: self, n: n} + Take { iter: self, n } } /// An iterator adaptor similar to [`fold`] that holds internal state and @@ -968,13 +993,13 @@ pub trait Iterator { /// // each iteration, we'll multiply the state by the element /// *state = *state * x; /// - /// // the value passed on to the next iteration - /// Some(*state) + /// // then, we'll yield the negation of the state + /// Some(-*state) /// }); /// - /// assert_eq!(iter.next(), Some(1)); - /// assert_eq!(iter.next(), Some(2)); - /// assert_eq!(iter.next(), Some(6)); + /// assert_eq!(iter.next(), Some(-1)); + /// assert_eq!(iter.next(), Some(-2)); + /// assert_eq!(iter.next(), Some(-6)); /// assert_eq!(iter.next(), None); /// ``` #[inline] @@ -982,7 +1007,7 @@ pub trait Iterator { fn scan<St, B, F>(self, initial_state: St, f: F) -> Scan<Self, St, F> where Self: Sized, F: FnMut(&mut St, Self::Item) -> Option<B>, { - Scan{iter: self, f: f, state: initial_state} + Scan { iter: self, f, state: initial_state } } /// Creates an iterator that works like map, but flattens nested structure. @@ -992,11 +1017,15 @@ pub trait Iterator { /// an extra layer of indirection. `flat_map()` will remove this extra layer /// on its own. /// + /// You can think of `flat_map(f)` as the semantic equivalent + /// of [`map`]ping, and then [`flatten`]ing as in `map(f).flatten()`. + /// /// Another way of thinking about `flat_map()`: [`map`]'s closure returns /// one item for each element, and `flat_map()`'s closure returns an /// iterator for each element. /// /// [`map`]: #method.map + /// [`flatten`]: #method.flatten /// /// # Examples /// @@ -1016,7 +1045,75 @@ pub trait Iterator { fn flat_map<U, F>(self, f: F) -> FlatMap<Self, U, F> where Self: Sized, U: IntoIterator, F: FnMut(Self::Item) -> U, { - FlatMap{iter: self, f: f, frontiter: None, backiter: None } + FlatMap { inner: flatten_compat(self.map(f)) } + } + + /// Creates an iterator that flattens nested structure. + /// + /// This is useful when you have an iterator of iterators or an iterator of + /// things that can be turned into iterators and you want to remove one + /// level of indirection. + /// + /// # Examples + /// + /// Basic usage: + /// + /// ``` + /// let data = vec![vec![1, 2, 3, 4], vec![5, 6]]; + /// let flattened = data.into_iter().flatten().collect::<Vec<u8>>(); + /// assert_eq!(flattened, &[1, 2, 3, 4, 5, 6]); + /// ``` + /// + /// Mapping and then flattening: + /// + /// ``` + /// let words = ["alpha", "beta", "gamma"]; + /// + /// // chars() returns an iterator + /// let merged: String = words.iter() + /// .map(|s| s.chars()) + /// .flatten() + /// .collect(); + /// assert_eq!(merged, "alphabetagamma"); + /// ``` + /// + /// You can also rewrite this in terms of [`flat_map()`], which is preferable + /// in this case since it conveys intent more clearly: + /// + /// ``` + /// let words = ["alpha", "beta", "gamma"]; + /// + /// // chars() returns an iterator + /// let merged: String = words.iter() + /// .flat_map(|s| s.chars()) + /// .collect(); + /// assert_eq!(merged, "alphabetagamma"); + /// ``` + /// + /// Flattening once only removes one level of nesting: + /// + /// ``` + /// let d3 = [[[1, 2], [3, 4]], [[5, 6], [7, 8]]]; + /// + /// let d2 = d3.iter().flatten().collect::<Vec<_>>(); + /// assert_eq!(d2, [&[1, 2], &[3, 4], &[5, 6], &[7, 8]]); + /// + /// let d1 = d3.iter().flatten().flatten().collect::<Vec<_>>(); + /// assert_eq!(d1, [&1, &2, &3, &4, &5, &6, &7, &8]); + /// ``` + /// + /// Here we see that `flatten()` does not perform a "deep" flatten. + /// Instead, only one level of nesting is removed. That is, if you + /// `flatten()` a three-dimensional array the result will be + /// two-dimensional and not one-dimensional. To get a one-dimensional + /// structure, you have to `flatten()` again. + /// + /// [`flat_map()`]: #method.flat_map + #[inline] + #[stable(feature = "iterator_flatten", since = "1.29.0")] + fn flatten(self) -> Flatten<Self> + where Self: Sized, Self::Item: IntoIterator { + Flatten { inner: flatten_compat(self) } } /// Creates an iterator which ends after the first [`None`]. @@ -1086,8 +1183,9 @@ pub trait Iterator { /// happening at various parts in the pipeline. To do that, insert /// a call to `inspect()`. /// - /// It's much more common for `inspect()` to be used as a debugging tool - /// than to exist in your final code, but never say never. + /// It's more common for `inspect()` to be used as a debugging tool than to + /// exist in your final code, but applications may find it useful in certain + /// situations when errors need to be logged before being discarded. /// /// # Examples /// @@ -1098,19 +1196,19 @@ pub trait Iterator { /// /// // this iterator sequence is complex. /// let sum = a.iter() - /// .cloned() - /// .filter(|&x| x % 2 == 0) - /// .fold(0, |sum, i| sum + i); + /// .cloned() + /// .filter(|x| x % 2 == 0) + /// .fold(0, |sum, i| sum + i); /// /// println!("{}", sum); /// /// // let's add some inspect() calls to investigate what's happening /// let sum = a.iter() - /// .cloned() - /// .inspect(|x| println!("about to filter: {}", x)) - /// .filter(|&x| x % 2 == 0) - /// .inspect(|x| println!("made it through filter: {}", x)) - /// .fold(0, |sum, i| sum + i); + /// .cloned() + /// .inspect(|x| println!("about to filter: {}", x)) + /// .filter(|x| x % 2 == 0) + /// .inspect(|x| println!("made it through filter: {}", x)) + /// .fold(0, |sum, i| sum + i); /// /// println!("{}", sum); /// ``` @@ -1118,6 +1216,7 @@ pub trait Iterator { /// This will print: /// /// ```text + /// 6 /// about to filter: 1 /// about to filter: 4 /// made it through filter: 4 @@ -1126,12 +1225,38 @@ pub trait Iterator { /// about to filter: 3 /// 6 /// ``` + /// + /// Logging errors before discarding them: + /// + /// ``` + /// let lines = ["1", "2", "a"]; + /// + /// let sum: i32 = lines + /// .iter() + /// .map(|line| line.parse::<i32>()) + /// .inspect(|num| { + /// if let Err(ref e) = *num { + /// println!("Parsing error: {}", e); + /// } + /// }) + /// .filter_map(Result::ok) + /// .sum(); + /// + /// println!("Sum: {}", sum); + /// ``` + /// + /// This will print: + /// + /// ```text + /// Parsing error: invalid digit found in string + /// Sum: 3 + /// ``` #[inline] #[stable(feature = "rust1", since = "1.0.0")] fn inspect<F>(self, f: F) -> Inspect<Self, F> where Self: Sized, F: FnMut(&Self::Item), { - Inspect{iter: self, f: f} + Inspect { iter: self, f } } /// Borrows an iterator, rather than consuming it. @@ -1148,8 +1273,7 @@ pub trait Iterator { /// /// let iter = a.into_iter(); /// - /// let sum: i32 = iter.take(5) - /// .fold(0, |acc, &i| acc + i ); + /// let sum: i32 = iter.take(5).fold(0, |acc, i| acc + i ); /// /// assert_eq!(sum, 6); /// @@ -1163,9 +1287,7 @@ pub trait Iterator { /// let mut iter = a.into_iter(); /// /// // instead, we add in a .by_ref() - /// let sum: i32 = iter.by_ref() - /// .take(2) - /// .fold(0, |acc, &i| acc + i ); + /// let sum: i32 = iter.by_ref().take(2).fold(0, |acc, i| acc + i ); /// /// assert_eq!(sum, 3); /// @@ -1222,9 +1344,7 @@ pub trait Iterator { /// /// let a = [1, 2, 3]; /// - /// let doubled: VecDeque<i32> = a.iter() - /// .map(|&x| x * 2) - /// .collect(); + /// let doubled: VecDeque<i32> = a.iter().map(|&x| x * 2).collect(); /// /// assert_eq!(2, doubled[0]); /// assert_eq!(4, doubled[1]); @@ -1236,9 +1356,7 @@ pub trait Iterator { /// ``` /// let a = [1, 2, 3]; /// - /// let doubled = a.iter() - /// .map(|&x| x * 2) - /// .collect::<Vec<i32>>(); + /// let doubled = a.iter().map(|x| x * 2).collect::<Vec<i32>>(); /// /// assert_eq!(vec![2, 4, 6], doubled); /// ``` @@ -1249,9 +1367,7 @@ pub trait Iterator { /// ``` /// let a = [1, 2, 3]; /// - /// let doubled = a.iter() - /// .map(|&x| x * 2) - /// .collect::<Vec<_>>(); + /// let doubled = a.iter().map(|x| x * 2).collect::<Vec<_>>(); /// /// assert_eq!(vec![2, 4, 6], doubled); /// ``` @@ -1262,9 +1378,9 @@ pub trait Iterator { /// let chars = ['g', 'd', 'k', 'k', 'n']; /// /// let hello: String = chars.iter() - /// .map(|&x| x as u8) - /// .map(|x| (x + 1) as char) - /// .collect(); + /// .map(|&x| x as u8) + /// .map(|x| (x + 1) as char) + /// .collect(); /// /// assert_eq!("hello", hello); /// ``` @@ -1294,6 +1410,7 @@ pub trait Iterator { /// [`Result`]: ../../std/result/enum.Result.html #[inline] #[stable(feature = "rust1", since = "1.0.0")] + #[must_use = "if you really need to exhaust the iterator, consider `.for_each(drop)` instead"] fn collect<B: FromIterator<Self::Item>>(self) -> B where Self: Sized { FromIterator::from_iter(self) } @@ -1311,8 +1428,9 @@ pub trait Iterator { /// ``` /// let a = [1, 2, 3]; /// - /// let (even, odd): (Vec<i32>, Vec<i32>) = a.into_iter() - /// .partition(|&n| n % 2 == 0); + /// let (even, odd): (Vec<i32>, Vec<i32>) = a + /// .into_iter() + /// .partition(|&n| n % 2 == 0); /// /// assert_eq!(even, vec![2]); /// assert_eq!(odd, vec![1, 3]); @@ -1361,9 +1479,9 @@ pub trait Iterator { /// /// In particular, try to have this call `try_fold()` on the internal parts /// from which this iterator is composed. If multiple calls are needed, - /// the `?` operator be convenient for chaining the accumulator value along, - /// but beware any invariants that need to be upheld before those early - /// returns. This is a `&mut self` method, so iteration needs to be + /// the `?` operator may be convenient for chaining the accumulator value + /// along, but beware any invariants that need to be upheld before those + /// early returns. This is a `&mut self` method, so iteration needs to be /// resumable after hitting an error here. /// /// # Examples @@ -1371,12 +1489,10 @@ pub trait Iterator { /// Basic usage: /// /// ``` - /// #![feature(iterator_try_fold)] /// let a = [1, 2, 3]; /// /// // the checked sum of all of the elements of the array - /// let sum = a.iter() - /// .try_fold(0i8, |acc, &x| acc.checked_add(x)); + /// let sum = a.iter().try_fold(0i8, |acc, &x| acc.checked_add(x)); /// /// assert_eq!(sum, Some(6)); /// ``` @@ -1384,7 +1500,6 @@ pub trait Iterator { /// Short-circuiting: /// /// ``` - /// #![feature(iterator_try_fold)] /// let a = [10, 20, 30, 100, 40, 50]; /// let mut it = a.iter(); /// @@ -1398,7 +1513,7 @@ pub trait Iterator { /// assert_eq!(it.next(), Some(&40)); /// ``` #[inline] - #[unstable(feature = "iterator_try_fold", issue = "45594")] + #[stable(feature = "iterator_try_fold", since = "1.27.0")] fn try_fold<B, F, R>(&mut self, init: B, mut f: F) -> R where Self: Sized, F: FnMut(B, Self::Item) -> R, R: Try<Ok=B> { @@ -1409,6 +1524,41 @@ pub trait Iterator { Try::from_ok(accum) } + /// An iterator method that applies a fallible function to each item in the + /// iterator, stopping at the first error and returning that error. + /// + /// This can also be thought of as the fallible form of [`for_each()`] + /// or as the stateless version of [`try_fold()`]. + /// + /// [`for_each()`]: #method.for_each + /// [`try_fold()`]: #method.try_fold + /// + /// # Examples + /// + /// ``` + /// use std::fs::rename; + /// use std::io::{stdout, Write}; + /// use std::path::Path; + /// + /// let data = ["no_tea.txt", "stale_bread.json", "torrential_rain.png"]; + /// + /// let res = data.iter().try_for_each(|x| writeln!(stdout(), "{}", x)); + /// assert!(res.is_ok()); + /// + /// let mut it = data.iter().cloned(); + /// let res = it.try_for_each(|x| rename(x, Path::new(x).with_extension("old"))); + /// assert!(res.is_err()); + /// // It short-circuited, so the remaining items are still in the iterator: + /// assert_eq!(it.next(), Some("stale_bread.json")); + /// ``` + #[inline] + #[stable(feature = "iterator_try_fold", since = "1.27.0")] + fn try_for_each<F, R>(&mut self, mut f: F) -> R where + Self: Sized, F: FnMut(Self::Item) -> R, R: Try<Ok=()> + { + self.try_fold((), move |(), x| f(x)) + } + /// An iterator method that applies a function, producing a single, final value. /// /// `fold()` takes two arguments: an initial value, and a closure with two @@ -1426,6 +1576,10 @@ pub trait Iterator { /// Folding is useful whenever you have a collection of something, and want /// to produce a single value from it. /// + /// Note: `fold()`, and similar methods that traverse the entire iterator, + /// may not terminate for infinite iterators, even on traits for which a + /// result is determinable in finite time. + /// /// # Examples /// /// Basic usage: @@ -1434,8 +1588,7 @@ pub trait Iterator { /// let a = [1, 2, 3]; /// /// // the sum of all of the elements of the array - /// let sum = a.iter() - /// .fold(0, |acc, &x| acc + x); + /// let sum = a.iter().fold(0, |acc, x| acc + x); /// /// assert_eq!(sum, 6); /// ``` @@ -1478,7 +1631,7 @@ pub trait Iterator { fn fold<B, F>(mut self, init: B, mut f: F) -> B where Self: Sized, F: FnMut(B, Self::Item) -> B, { - self.try_fold(init, move |acc, x| AlwaysOk(f(acc, x))).0 + self.try_fold(init, move |acc, x| Ok::<B, !>(f(acc, x))).unwrap() } /// Tests if every element of the iterator matches a predicate. @@ -1523,7 +1676,7 @@ pub trait Iterator { fn all<F>(&mut self, mut f: F) -> bool where Self: Sized, F: FnMut(Self::Item) -> bool { - self.try_fold((), move |(), x| { + self.try_for_each(move |x| { if f(x) { LoopState::Continue(()) } else { LoopState::Break(()) } }) == LoopState::Continue(()) @@ -1572,7 +1725,7 @@ pub trait Iterator { Self: Sized, F: FnMut(Self::Item) -> bool { - self.try_fold((), move |(), x| { + self.try_for_each(move |x| { if f(x) { LoopState::Break(()) } else { LoopState::Continue(()) } }) == LoopState::Break(()) @@ -1626,12 +1779,44 @@ pub trait Iterator { Self: Sized, P: FnMut(&Self::Item) -> bool, { - self.try_fold((), move |(), x| { + self.try_for_each(move |x| { if predicate(&x) { LoopState::Break(x) } else { LoopState::Continue(()) } }).break_value() } + /// Applies function to the elements of iterator and returns + /// the first non-none result. + /// + /// `iter.find_map(f)` is equivalent to `iter.filter_map(f).next()`. + /// + /// + /// # Examples + /// + /// ``` + /// #![feature(iterator_find_map)] + /// let a = ["lol", "NaN", "2", "5"]; + /// + /// let mut first_number = a.iter().find_map(|s| s.parse().ok()); + /// + /// assert_eq!(first_number, Some(2)); + /// ``` + #[inline] + #[unstable(feature = "iterator_find_map", + reason = "unstable new API", + issue = "49602")] + fn find_map<B, F>(&mut self, mut f: F) -> Option<B> where + Self: Sized, + F: FnMut(Self::Item) -> Option<B>, + { + self.try_for_each(move |x| { + match f(x) { + Some(x) => LoopState::Break(x), + None => LoopState::Continue(()), + } + }).break_value() + } + /// Searches for an element in an iterator, returning its index. /// /// `position()` takes a closure that returns `true` or `false`. It applies diff --git a/src/libcore/iter/mod.rs b/src/libcore/iter/mod.rs index 06c29b47bf9..ef3f4ced4f9 100644 --- a/src/libcore/iter/mod.rs +++ b/src/libcore/iter/mod.rs @@ -298,15 +298,31 @@ //! //! This will print the numbers `0` through `4`, each on their own line. //! +//! Bear in mind that methods on infinite iterators, even those for which a +//! result can be determined mathematically in finite time, may not terminate. +//! Specifically, methods such as [`min`], which in the general case require +//! traversing every element in the iterator, are likely not to return +//! successfully for any infinite iterators. +//! +//! ```no_run +//! let ones = std::iter::repeat(1); +//! let least = ones.min().unwrap(); // Oh no! An infinite loop! +//! // `ones.min()` causes an infinite loop, so we won't reach this point! +//! println!("The smallest number one is {}.", least); +//! ``` +//! //! [`take`]: trait.Iterator.html#method.take +//! [`min`]: trait.Iterator.html#method.min #![stable(feature = "rust1", since = "1.0.0")] use cmp; use fmt; use iter_private::TrustedRandomAccess; -use ops::Try; +use ops::{self, Try}; use usize; +use intrinsics; +use mem; #[stable(feature = "rust1", since = "1.0.0")] pub use self::iterator::Iterator; @@ -318,6 +334,8 @@ pub use self::range::Step; #[stable(feature = "rust1", since = "1.0.0")] pub use self::sources::{Repeat, repeat}; +#[stable(feature = "iterator_repeat_with", since = "1.28.0")] +pub use self::sources::{RepeatWith, repeat_with}; #[stable(feature = "iter_empty", since = "1.2.0")] pub use self::sources::{Empty, empty}; #[stable(feature = "iter_once", since = "1.2.0")] @@ -327,7 +345,7 @@ pub use self::sources::{Once, once}; pub use self::traits::{FromIterator, IntoIterator, DoubleEndedIterator, Extend}; #[stable(feature = "rust1", since = "1.0.0")] pub use self::traits::{ExactSizeIterator, Sum, Product}; -#[unstable(feature = "fused", issue = "35602")] +#[stable(feature = "fused", since = "1.26.0")] pub use self::traits::FusedIterator; #[unstable(feature = "trusted_len", issue = "37572")] pub use self::traits::TrustedLen; @@ -337,21 +355,6 @@ mod range; mod sources; mod traits; -/// Transparent newtype used to implement foo methods in terms of try_foo. -/// Important until #43278 is fixed; might be better as `Result<T, !>` later. -struct AlwaysOk<T>(pub T); - -impl<T> Try for AlwaysOk<T> { - type Ok = T; - type Error = !; - #[inline] - fn into_result(self) -> Result<Self::Ok, Self::Error> { Ok(self.0) } - #[inline] - fn from_error(v: Self::Error) -> Self { v } - #[inline] - fn from_ok(v: Self::Ok) -> Self { AlwaysOk(v) } -} - /// Used to make try_fold closures more like normal loops #[derive(PartialEq)] enum LoopState<C, B> { @@ -489,7 +492,7 @@ impl<I> ExactSizeIterator for Rev<I> } } -#[unstable(feature = "fused", issue = "35602")] +#[stable(feature = "fused", since = "1.26.0")] impl<I> FusedIterator for Rev<I> where I: FusedIterator + DoubleEndedIterator {} @@ -572,21 +575,21 @@ impl<'a, I, T: 'a> ExactSizeIterator for Cloned<I> } } -#[unstable(feature = "fused", issue = "35602")] +#[stable(feature = "fused", since = "1.26.0")] impl<'a, I, T: 'a> FusedIterator for Cloned<I> where I: FusedIterator<Item=&'a T>, T: Clone {} #[doc(hidden)] -default unsafe impl<'a, I, T: 'a> TrustedRandomAccess for Cloned<I> +unsafe impl<'a, I, T: 'a> TrustedRandomAccess for Cloned<I> where I: TrustedRandomAccess<Item=&'a T>, T: Clone { - unsafe fn get_unchecked(&mut self, i: usize) -> Self::Item { + default unsafe fn get_unchecked(&mut self, i: usize) -> Self::Item { self.it.get_unchecked(i).clone() } #[inline] - fn may_have_side_effect() -> bool { true } + default fn may_have_side_effect() -> bool { true } } #[doc(hidden)] @@ -645,7 +648,7 @@ impl<I> Iterator for Cycle<I> where I: Clone + Iterator { } } -#[unstable(feature = "fused", issue = "35602")] +#[stable(feature = "fused", since = "1.26.0")] impl<I> FusedIterator for Cycle<I> where I: Clone + Iterator {} /// An iterator for stepping iterators by a custom amount. @@ -656,9 +659,7 @@ impl<I> FusedIterator for Cycle<I> where I: Clone + Iterator {} /// [`step_by`]: trait.Iterator.html#method.step_by /// [`Iterator`]: trait.Iterator.html #[must_use = "iterator adaptors are lazy and do nothing unless consumed"] -#[unstable(feature = "iterator_step_by", - reason = "unstable replacement of Range::step_by", - issue = "27741")] +#[stable(feature = "iterator_step_by", since = "1.28.0")] #[derive(Clone, Debug)] pub struct StepBy<I> { iter: I, @@ -666,20 +667,13 @@ pub struct StepBy<I> { first_take: bool, } -#[unstable(feature = "iterator_step_by", - reason = "unstable replacement of Range::step_by", - issue = "27741")] +#[stable(feature = "iterator_step_by", since = "1.28.0")] impl<I> Iterator for StepBy<I> where I: Iterator { type Item = I::Item; #[inline] fn next(&mut self) -> Option<Self::Item> { - if self.first_take { - self.first_take = false; - self.iter.next() - } else { - self.iter.nth(self.step) - } + <Self as StepBySpecIterator>::spec_next(self) } #[inline] @@ -694,12 +688,125 @@ impl<I> Iterator for StepBy<I> where I: Iterator { (f(inner_hint.0), inner_hint.1.map(f)) } } + + #[inline] + fn nth(&mut self, mut n: usize) -> Option<Self::Item> { + if self.first_take { + self.first_take = false; + let first = self.iter.next(); + if n == 0 { + return first; + } + n -= 1; + } + // n and self.step are indices, we need to add 1 to get the amount of elements + // When calling `.nth`, we need to subtract 1 again to convert back to an index + // step + 1 can't overflow because `.step_by` sets `self.step` to `step - 1` + let mut step = self.step + 1; + // n + 1 could overflow + // thus, if n is usize::MAX, instead of adding one, we call .nth(step) + if n == usize::MAX { + self.iter.nth(step - 1); + } else { + n += 1; + } + + // overflow handling + loop { + let mul = n.checked_mul(step); + if unsafe { intrinsics::likely(mul.is_some()) } { + return self.iter.nth(mul.unwrap() - 1); + } + let div_n = usize::MAX / n; + let div_step = usize::MAX / step; + let nth_n = div_n * n; + let nth_step = div_step * step; + let nth = if nth_n > nth_step { + step -= div_n; + nth_n + } else { + n -= div_step; + nth_step + }; + self.iter.nth(nth - 1); + } + } +} + +// hidden trait for specializing iterator methods +// could be generalized but is currently only used for StepBy +trait StepBySpecIterator { + type Item; + fn spec_next(&mut self) -> Option<Self::Item>; +} + +impl<I> StepBySpecIterator for StepBy<I> +where + I: Iterator, +{ + type Item = I::Item; + + #[inline] + default fn spec_next(&mut self) -> Option<I::Item> { + if self.first_take { + self.first_take = false; + self.iter.next() + } else { + self.iter.nth(self.step) + } + } +} + +impl<T> StepBySpecIterator for StepBy<ops::Range<T>> +where + T: Step, +{ + #[inline] + fn spec_next(&mut self) -> Option<Self::Item> { + self.first_take = false; + if !(self.iter.start < self.iter.end) { + return None; + } + // add 1 to self.step to get original step size back + // it was decremented for the general case on construction + if let Some(n) = self.iter.start.add_usize(self.step+1) { + let next = mem::replace(&mut self.iter.start, n); + Some(next) + } else { + let last = self.iter.start.clone(); + self.iter.start = self.iter.end.clone(); + Some(last) + } + } +} + +impl<T> StepBySpecIterator for StepBy<ops::RangeInclusive<T>> +where + T: Step, +{ + #[inline] + fn spec_next(&mut self) -> Option<Self::Item> { + self.first_take = false; + self.iter.compute_is_empty(); + if self.iter.is_empty.unwrap_or_default() { + return None; + } + // add 1 to self.step to get original step size back + // it was decremented for the general case on construction + if let Some(n) = self.iter.start.add_usize(self.step+1) { + self.iter.is_empty = Some(!(n <= self.iter.end)); + let next = mem::replace(&mut self.iter.start, n); + Some(next) + } else { + let last = self.iter.start.clone(); + self.iter.is_empty = Some(true); + Some(last) + } + } } // StepBy can only make the iterator shorter, so the len will still fit. -#[unstable(feature = "iterator_step_by", - reason = "unstable replacement of Range::step_by", - issue = "27741")] +#[stable(feature = "iterator_step_by", since = "1.28.0")] impl<I> ExactSizeIterator for StepBy<I> where I: ExactSizeIterator {} /// An iterator that strings two iterators together. @@ -942,7 +1049,7 @@ impl<A, B> DoubleEndedIterator for Chain<A, B> where } // Note: *both* must be fused to handle double-ended iterators. -#[unstable(feature = "fused", issue = "35602")] +#[stable(feature = "fused", since = "1.26.0")] impl<A, B> FusedIterator for Chain<A, B> where A: FusedIterator, B: FusedIterator<Item=A::Item>, @@ -985,6 +1092,11 @@ impl<A, B> Iterator for Zip<A, B> where A: Iterator, B: Iterator fn size_hint(&self) -> (usize, Option<usize>) { ZipImpl::size_hint(self) } + + #[inline] + fn nth(&mut self, n: usize) -> Option<Self::Item> { + ZipImpl::nth(self, n) + } } #[stable(feature = "rust1", since = "1.0.0")] @@ -1005,6 +1117,14 @@ trait ZipImpl<A, B> { fn new(a: A, b: B) -> Self; fn next(&mut self) -> Option<Self::Item>; fn size_hint(&self) -> (usize, Option<usize>); + fn nth(&mut self, n: usize) -> Option<Self::Item>; + fn super_nth(&mut self, mut n: usize) -> Option<Self::Item> { + while let Some(x) = self.next() { + if n == 0 { return Some(x) } + n -= 1; + } + None + } fn next_back(&mut self) -> Option<Self::Item> where A: DoubleEndedIterator + ExactSizeIterator, B: DoubleEndedIterator + ExactSizeIterator; @@ -1035,6 +1155,11 @@ impl<A, B> ZipImpl<A, B> for Zip<A, B> } #[inline] + default fn nth(&mut self, n: usize) -> Option<Self::Item> { + self.super_nth(n) + } + + #[inline] default fn next_back(&mut self) -> Option<(A::Item, B::Item)> where A: DoubleEndedIterator + ExactSizeIterator, B: DoubleEndedIterator + ExactSizeIterator @@ -1115,6 +1240,24 @@ impl<A, B> ZipImpl<A, B> for Zip<A, B> } #[inline] + fn nth(&mut self, n: usize) -> Option<Self::Item> { + let delta = cmp::min(n, self.len - self.index); + let end = self.index + delta; + while self.index < end { + let i = self.index; + self.index += 1; + if A::may_have_side_effect() { + unsafe { self.a.get_unchecked(i); } + } + if B::may_have_side_effect() { + unsafe { self.b.get_unchecked(i); } + } + } + + self.super_nth(n - delta) + } + + #[inline] fn next_back(&mut self) -> Option<(A::Item, B::Item)> where A: DoubleEndedIterator + ExactSizeIterator, B: DoubleEndedIterator + ExactSizeIterator @@ -1166,7 +1309,7 @@ unsafe impl<A, B> TrustedRandomAccess for Zip<A, B> } } -#[unstable(feature = "fused", issue = "35602")] +#[stable(feature = "fused", since = "1.26.0")] impl<A, B> FusedIterator for Zip<A, B> where A: FusedIterator, B: FusedIterator, {} @@ -1308,7 +1451,7 @@ impl<B, I: ExactSizeIterator, F> ExactSizeIterator for Map<I, F> } } -#[unstable(feature = "fused", issue = "35602")] +#[stable(feature = "fused", since = "1.26.0")] impl<B, I: FusedIterator, F> FusedIterator for Map<I, F> where F: FnMut(I::Item) -> B {} @@ -1457,7 +1600,7 @@ impl<I: DoubleEndedIterator, P> DoubleEndedIterator for Filter<I, P> } } -#[unstable(feature = "fused", issue = "35602")] +#[stable(feature = "fused", since = "1.26.0")] impl<I: FusedIterator, P> FusedIterator for Filter<I, P> where P: FnMut(&I::Item) -> bool {} @@ -1567,7 +1710,7 @@ impl<B, I: DoubleEndedIterator, F> DoubleEndedIterator for FilterMap<I, F> } } -#[unstable(feature = "fused", issue = "35602")] +#[stable(feature = "fused", since = "1.26.0")] impl<B, I: FusedIterator, F> FusedIterator for FilterMap<I, F> where F: FnMut(I::Item) -> Option<B> {} @@ -1722,7 +1865,7 @@ unsafe impl<I> TrustedRandomAccess for Enumerate<I> } } -#[unstable(feature = "fused", issue = "35602")] +#[stable(feature = "fused", since = "1.26.0")] impl<I> FusedIterator for Enumerate<I> where I: FusedIterator {} #[unstable(feature = "trusted_len", issue = "37572")] @@ -1776,7 +1919,7 @@ impl<I: Iterator> Iterator for Peekable<I> { #[inline] fn nth(&mut self, n: usize) -> Option<I::Item> { - // FIXME(#6393): merge these when borrow-checking gets better. + // FIXME(#43234): merge these when borrow-checking gets better. if n == 0 { match self.peeked.take() { Some(v) => v, @@ -1842,7 +1985,7 @@ impl<I: Iterator> Iterator for Peekable<I> { #[stable(feature = "rust1", since = "1.0.0")] impl<I: ExactSizeIterator> ExactSizeIterator for Peekable<I> {} -#[unstable(feature = "fused", issue = "35602")] +#[stable(feature = "fused", since = "1.26.0")] impl<I: FusedIterator> FusedIterator for Peekable<I> {} impl<I: Iterator> Peekable<I> { @@ -1976,7 +2119,7 @@ impl<I: Iterator, P> Iterator for SkipWhile<I, P> } } -#[unstable(feature = "fused", issue = "35602")] +#[stable(feature = "fused", since = "1.26.0")] impl<I, P> FusedIterator for SkipWhile<I, P> where I: FusedIterator, P: FnMut(&I::Item) -> bool {} @@ -2055,7 +2198,7 @@ impl<I: Iterator, P> Iterator for TakeWhile<I, P> } } -#[unstable(feature = "fused", issue = "35602")] +#[stable(feature = "fused", since = "1.26.0")] impl<I, P> FusedIterator for TakeWhile<I, P> where I: FusedIterator, P: FnMut(&I::Item) -> bool {} @@ -2194,7 +2337,7 @@ impl<I> DoubleEndedIterator for Skip<I> where I: DoubleEndedIterator + ExactSize } } -#[unstable(feature = "fused", issue = "35602")] +#[stable(feature = "fused", since = "1.26.0")] impl<I> FusedIterator for Skip<I> where I: FusedIterator {} /// An iterator that only iterates over the first `n` iterations of `iter`. @@ -2275,9 +2418,12 @@ impl<I> Iterator for Take<I> where I: Iterator{ #[stable(feature = "rust1", since = "1.0.0")] impl<I> ExactSizeIterator for Take<I> where I: ExactSizeIterator {} -#[unstable(feature = "fused", issue = "35602")] +#[stable(feature = "fused", since = "1.26.0")] impl<I> FusedIterator for Take<I> where I: FusedIterator {} +#[unstable(feature = "trusted_len", issue = "37572")] +unsafe impl<I: TrustedLen> TrustedLen for Take<I> {} + /// An iterator to maintain state while iterating another iterator. /// /// This `struct` is created by the [`scan`] method on [`Iterator`]. See its @@ -2347,12 +2493,15 @@ impl<B, I, St, F> Iterator for Scan<I, St, F> where /// [`Iterator`]: trait.Iterator.html #[must_use = "iterator adaptors are lazy and do nothing unless consumed"] #[stable(feature = "rust1", since = "1.0.0")] -#[derive(Clone)] pub struct FlatMap<I, U: IntoIterator, F> { - iter: I, - f: F, - frontiter: Option<U::IntoIter>, - backiter: Option<U::IntoIter>, + inner: FlattenCompat<Map<I, F>, <U as IntoIterator>::IntoIter> +} + +#[stable(feature = "rust1", since = "1.0.0")] +impl<I: Clone, U: Clone + IntoIterator, F: Clone> Clone for FlatMap<I, U, F> + where <U as IntoIterator>::IntoIter: Clone +{ + fn clone(&self) -> Self { FlatMap { inner: self.inner.clone() } } } #[stable(feature = "core_impl_debug", since = "1.9.0")] @@ -2360,11 +2509,7 @@ impl<I: fmt::Debug, U: IntoIterator, F> fmt::Debug for FlatMap<I, U, F> where U::IntoIter: fmt::Debug { fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result { - f.debug_struct("FlatMap") - .field("iter", &self.iter) - .field("frontiter", &self.frontiter) - .field("backiter", &self.backiter) - .finish() + f.debug_struct("FlatMap").field("inner", &self.inner).finish() } } @@ -2375,16 +2520,172 @@ impl<I: Iterator, U: IntoIterator, F> Iterator for FlatMap<I, U, F> type Item = U::Item; #[inline] + fn next(&mut self) -> Option<U::Item> { self.inner.next() } + + #[inline] + fn size_hint(&self) -> (usize, Option<usize>) { self.inner.size_hint() } + + #[inline] + fn try_fold<Acc, Fold, R>(&mut self, init: Acc, fold: Fold) -> R where + Self: Sized, Fold: FnMut(Acc, Self::Item) -> R, R: Try<Ok=Acc> + { + self.inner.try_fold(init, fold) + } + + #[inline] + fn fold<Acc, Fold>(self, init: Acc, fold: Fold) -> Acc + where Fold: FnMut(Acc, Self::Item) -> Acc, + { + self.inner.fold(init, fold) + } +} + +#[stable(feature = "rust1", since = "1.0.0")] +impl<I: DoubleEndedIterator, U, F> DoubleEndedIterator for FlatMap<I, U, F> + where F: FnMut(I::Item) -> U, + U: IntoIterator, + U::IntoIter: DoubleEndedIterator +{ + #[inline] + fn next_back(&mut self) -> Option<U::Item> { self.inner.next_back() } + + #[inline] + fn try_rfold<Acc, Fold, R>(&mut self, init: Acc, fold: Fold) -> R where + Self: Sized, Fold: FnMut(Acc, Self::Item) -> R, R: Try<Ok=Acc> + { + self.inner.try_rfold(init, fold) + } + + #[inline] + fn rfold<Acc, Fold>(self, init: Acc, fold: Fold) -> Acc + where Fold: FnMut(Acc, Self::Item) -> Acc, + { + self.inner.rfold(init, fold) + } +} + +#[stable(feature = "fused", since = "1.26.0")] +impl<I, U, F> FusedIterator for FlatMap<I, U, F> + where I: FusedIterator, U: IntoIterator, F: FnMut(I::Item) -> U {} + +/// An iterator that flattens one level of nesting in an iterator of things +/// that can be turned into iterators. +/// +/// This `struct` is created by the [`flatten`] method on [`Iterator`]. See its +/// documentation for more. +/// +/// [`flatten`]: trait.Iterator.html#method.flatten +/// [`Iterator`]: trait.Iterator.html +#[must_use = "iterator adaptors are lazy and do nothing unless consumed"] +#[stable(feature = "iterator_flatten", since = "1.29.0")] +pub struct Flatten<I: Iterator> +where I::Item: IntoIterator { + inner: FlattenCompat<I, <I::Item as IntoIterator>::IntoIter>, +} + +#[stable(feature = "iterator_flatten", since = "1.29.0")] +impl<I, U> fmt::Debug for Flatten<I> + where I: Iterator + fmt::Debug, U: Iterator + fmt::Debug, + I::Item: IntoIterator<IntoIter = U, Item = U::Item>, +{ + fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result { + f.debug_struct("Flatten").field("inner", &self.inner).finish() + } +} + +#[stable(feature = "iterator_flatten", since = "1.29.0")] +impl<I, U> Clone for Flatten<I> + where I: Iterator + Clone, U: Iterator + Clone, + I::Item: IntoIterator<IntoIter = U, Item = U::Item>, +{ + fn clone(&self) -> Self { Flatten { inner: self.inner.clone() } } +} + +#[stable(feature = "iterator_flatten", since = "1.29.0")] +impl<I, U> Iterator for Flatten<I> + where I: Iterator, U: Iterator, + I::Item: IntoIterator<IntoIter = U, Item = U::Item> +{ + type Item = U::Item; + + #[inline] + fn next(&mut self) -> Option<U::Item> { self.inner.next() } + + #[inline] + fn size_hint(&self) -> (usize, Option<usize>) { self.inner.size_hint() } + + #[inline] + fn try_fold<Acc, Fold, R>(&mut self, init: Acc, fold: Fold) -> R where + Self: Sized, Fold: FnMut(Acc, Self::Item) -> R, R: Try<Ok=Acc> + { + self.inner.try_fold(init, fold) + } + + #[inline] + fn fold<Acc, Fold>(self, init: Acc, fold: Fold) -> Acc + where Fold: FnMut(Acc, Self::Item) -> Acc, + { + self.inner.fold(init, fold) + } +} + +#[stable(feature = "iterator_flatten", since = "1.29.0")] +impl<I, U> DoubleEndedIterator for Flatten<I> + where I: DoubleEndedIterator, U: DoubleEndedIterator, + I::Item: IntoIterator<IntoIter = U, Item = U::Item> +{ + #[inline] + fn next_back(&mut self) -> Option<U::Item> { self.inner.next_back() } + + #[inline] + fn try_rfold<Acc, Fold, R>(&mut self, init: Acc, fold: Fold) -> R where + Self: Sized, Fold: FnMut(Acc, Self::Item) -> R, R: Try<Ok=Acc> + { + self.inner.try_rfold(init, fold) + } + + #[inline] + fn rfold<Acc, Fold>(self, init: Acc, fold: Fold) -> Acc + where Fold: FnMut(Acc, Self::Item) -> Acc, + { + self.inner.rfold(init, fold) + } +} + +#[stable(feature = "iterator_flatten", since = "1.29.0")] +impl<I, U> FusedIterator for Flatten<I> + where I: FusedIterator, U: Iterator, + I::Item: IntoIterator<IntoIter = U, Item = U::Item> {} + +/// Adapts an iterator by flattening it, for use in `flatten()` and `flat_map()`. +fn flatten_compat<I, U>(iter: I) -> FlattenCompat<I, U> { + FlattenCompat { iter, frontiter: None, backiter: None } +} + +/// Real logic of both `Flatten` and `FlatMap` which simply delegate to +/// this type. +#[derive(Clone, Debug)] +struct FlattenCompat<I, U> { + iter: I, + frontiter: Option<U>, + backiter: Option<U>, +} + +impl<I, U> Iterator for FlattenCompat<I, U> + where I: Iterator, U: Iterator, + I::Item: IntoIterator<IntoIter = U, Item = U::Item> +{ + type Item = U::Item; + + #[inline] fn next(&mut self) -> Option<U::Item> { loop { if let Some(ref mut inner) = self.frontiter { - if let Some(x) = inner.by_ref().next() { - return Some(x) - } + if let elt@Some(_) = inner.next() { return elt } } - match self.iter.next().map(&mut self.f) { + match self.iter.next() { None => return self.backiter.as_mut().and_then(|it| it.next()), - next => self.frontiter = next.map(IntoIterator::into_iter), + Some(inner) => self.frontiter = Some(inner.into_iter()), } } } @@ -2410,10 +2711,9 @@ impl<I: Iterator, U: IntoIterator, F> Iterator for FlatMap<I, U, F> self.frontiter = None; { - let f = &mut self.f; let frontiter = &mut self.frontiter; init = self.iter.try_fold(init, |acc, x| { - let mut mid = f(x).into_iter(); + let mut mid = x.into_iter(); let r = mid.try_fold(acc, &mut fold); *frontiter = Some(mid); r @@ -2434,27 +2734,23 @@ impl<I: Iterator, U: IntoIterator, F> Iterator for FlatMap<I, U, F> where Fold: FnMut(Acc, Self::Item) -> Acc, { self.frontiter.into_iter() - .chain(self.iter.map(self.f).map(U::into_iter)) + .chain(self.iter.map(IntoIterator::into_iter)) .chain(self.backiter) .fold(init, |acc, iter| iter.fold(acc, &mut fold)) } } -#[stable(feature = "rust1", since = "1.0.0")] -impl<I: DoubleEndedIterator, U, F> DoubleEndedIterator for FlatMap<I, U, F> where - F: FnMut(I::Item) -> U, - U: IntoIterator, - U::IntoIter: DoubleEndedIterator +impl<I, U> DoubleEndedIterator for FlattenCompat<I, U> + where I: DoubleEndedIterator, U: DoubleEndedIterator, + I::Item: IntoIterator<IntoIter = U, Item = U::Item> { #[inline] fn next_back(&mut self) -> Option<U::Item> { loop { if let Some(ref mut inner) = self.backiter { - if let Some(y) = inner.next_back() { - return Some(y) - } + if let elt@Some(_) = inner.next_back() { return elt } } - match self.iter.next_back().map(&mut self.f) { + match self.iter.next_back() { None => return self.frontiter.as_mut().and_then(|it| it.next_back()), next => self.backiter = next.map(IntoIterator::into_iter), } @@ -2471,10 +2767,9 @@ impl<I: DoubleEndedIterator, U, F> DoubleEndedIterator for FlatMap<I, U, F> wher self.backiter = None; { - let f = &mut self.f; let backiter = &mut self.backiter; init = self.iter.try_rfold(init, |acc, x| { - let mut mid = f(x).into_iter(); + let mut mid = x.into_iter(); let r = mid.try_rfold(acc, &mut fold); *backiter = Some(mid); r @@ -2495,16 +2790,12 @@ impl<I: DoubleEndedIterator, U, F> DoubleEndedIterator for FlatMap<I, U, F> wher where Fold: FnMut(Acc, Self::Item) -> Acc, { self.frontiter.into_iter() - .chain(self.iter.map(self.f).map(U::into_iter)) + .chain(self.iter.map(IntoIterator::into_iter)) .chain(self.backiter) .rfold(init, |acc, iter| iter.rfold(acc, &mut fold)) } } -#[unstable(feature = "fused", issue = "35602")] -impl<I, U, F> FusedIterator for FlatMap<I, U, F> - where I: FusedIterator, U: IntoIterator, F: FnMut(I::Item) -> U {} - /// An iterator that yields `None` forever after the underlying iterator /// yields `None` once. /// @@ -2521,7 +2812,7 @@ pub struct Fuse<I> { done: bool } -#[unstable(feature = "fused", issue = "35602")] +#[stable(feature = "fused", since = "1.26.0")] impl<I> FusedIterator for Fuse<I> where I: Iterator {} #[stable(feature = "rust1", since = "1.0.0")] @@ -2652,7 +2943,7 @@ unsafe impl<I> TrustedRandomAccess for Fuse<I> } } -#[unstable(feature = "fused", issue = "35602")] +#[stable(feature = "fused", since = "1.26.0")] impl<I> Iterator for Fuse<I> where I: FusedIterator { #[inline] fn next(&mut self) -> Option<<I as Iterator>::Item> { @@ -2694,7 +2985,7 @@ impl<I> Iterator for Fuse<I> where I: FusedIterator { } } -#[unstable(feature = "fused", reason = "recently added", issue = "35602")] +#[stable(feature = "fused", since = "1.26.0")] impl<I> DoubleEndedIterator for Fuse<I> where I: DoubleEndedIterator + FusedIterator { @@ -2838,6 +3129,6 @@ impl<I: ExactSizeIterator, F> ExactSizeIterator for Inspect<I, F> } } -#[unstable(feature = "fused", issue = "35602")] +#[stable(feature = "fused", since = "1.26.0")] impl<I: FusedIterator, F> FusedIterator for Inspect<I, F> where F: FnMut(&I::Item) {} diff --git a/src/libcore/iter/range.rs b/src/libcore/iter/range.rs index 66a76a24df4..651c7a35d41 100644 --- a/src/libcore/iter/range.rs +++ b/src/libcore/iter/range.rs @@ -186,9 +186,7 @@ macro_rules! range_exact_iter_impl { macro_rules! range_incl_exact_iter_impl { ($($t:ty)*) => ($( - #[unstable(feature = "inclusive_range", - reason = "recently added, follows RFC", - issue = "28237")] + #[stable(feature = "inclusive_range", since = "1.26.0")] impl ExactSizeIterator for ops::RangeInclusive<$t> { } )*) } @@ -202,9 +200,7 @@ macro_rules! range_trusted_len_impl { macro_rules! range_incl_trusted_len_impl { ($($t:ty)*) => ($( - #[unstable(feature = "inclusive_range", - reason = "recently added, follows RFC", - issue = "28237")] + #[unstable(feature = "trusted_len", issue = "37572")] unsafe impl TrustedLen for ops::RangeInclusive<$t> { } )*) } @@ -295,7 +291,7 @@ impl<A: Step> DoubleEndedIterator for ops::Range<A> { } } -#[unstable(feature = "fused", issue = "35602")] +#[stable(feature = "fused", since = "1.26.0")] impl<A: Step> FusedIterator for ops::Range<A> {} #[stable(feature = "rust1", since = "1.0.0")] @@ -322,34 +318,35 @@ impl<A: Step> Iterator for ops::RangeFrom<A> { } } -#[unstable(feature = "fused", issue = "35602")] +#[stable(feature = "fused", since = "1.26.0")] impl<A: Step> FusedIterator for ops::RangeFrom<A> {} -#[unstable(feature = "inclusive_range", reason = "recently added, follows RFC", issue = "28237")] +#[unstable(feature = "trusted_len", issue = "37572")] +unsafe impl<A: Step> TrustedLen for ops::RangeFrom<A> {} + +#[stable(feature = "inclusive_range", since = "1.26.0")] impl<A: Step> Iterator for ops::RangeInclusive<A> { type Item = A; #[inline] fn next(&mut self) -> Option<A> { - use cmp::Ordering::*; - - match self.start.partial_cmp(&self.end) { - Some(Less) => { - let n = self.start.add_one(); - Some(mem::replace(&mut self.start, n)) - }, - Some(Equal) => { - let last = self.start.replace_one(); - self.end.replace_zero(); - Some(last) - }, - _ => None, + self.compute_is_empty(); + if self.is_empty.unwrap_or_default() { + return None; } + let is_iterating = self.start < self.end; + self.is_empty = Some(!is_iterating); + Some(if is_iterating { + let n = self.start.add_one(); + mem::replace(&mut self.start, n) + } else { + self.start.clone() + }) } #[inline] fn size_hint(&self) -> (usize, Option<usize>) { - if !(self.start <= self.end) { + if self.is_empty() { return (0, Some(0)); } @@ -361,25 +358,29 @@ impl<A: Step> Iterator for ops::RangeInclusive<A> { #[inline] fn nth(&mut self, n: usize) -> Option<A> { + self.compute_is_empty(); + if self.is_empty.unwrap_or_default() { + return None; + } + if let Some(plus_n) = self.start.add_usize(n) { use cmp::Ordering::*; match plus_n.partial_cmp(&self.end) { Some(Less) => { + self.is_empty = Some(false); self.start = plus_n.add_one(); return Some(plus_n) } Some(Equal) => { - self.start.replace_one(); - self.end.replace_zero(); + self.is_empty = Some(true); return Some(plus_n) } _ => {} } } - self.start.replace_one(); - self.end.replace_zero(); + self.is_empty = Some(true); None } @@ -399,26 +400,24 @@ impl<A: Step> Iterator for ops::RangeInclusive<A> { } } -#[unstable(feature = "inclusive_range", reason = "recently added, follows RFC", issue = "28237")] +#[stable(feature = "inclusive_range", since = "1.26.0")] impl<A: Step> DoubleEndedIterator for ops::RangeInclusive<A> { #[inline] fn next_back(&mut self) -> Option<A> { - use cmp::Ordering::*; - - match self.start.partial_cmp(&self.end) { - Some(Less) => { - let n = self.end.sub_one(); - Some(mem::replace(&mut self.end, n)) - }, - Some(Equal) => { - let last = self.end.replace_zero(); - self.start.replace_one(); - Some(last) - }, - _ => None, + self.compute_is_empty(); + if self.is_empty.unwrap_or_default() { + return None; } + let is_iterating = self.start < self.end; + self.is_empty = Some(!is_iterating); + Some(if is_iterating { + let n = self.end.sub_one(); + mem::replace(&mut self.end, n) + } else { + self.end.clone() + }) } } -#[unstable(feature = "fused", issue = "35602")] +#[stable(feature = "fused", since = "1.26.0")] impl<A: Step> FusedIterator for ops::RangeInclusive<A> {} diff --git a/src/libcore/iter/sources.rs b/src/libcore/iter/sources.rs index b405f35d5e4..d500cc99fa1 100644 --- a/src/libcore/iter/sources.rs +++ b/src/libcore/iter/sources.rs @@ -41,9 +41,12 @@ impl<A: Clone> DoubleEndedIterator for Repeat<A> { fn next_back(&mut self) -> Option<A> { Some(self.element.clone()) } } -#[unstable(feature = "fused", issue = "35602")] +#[stable(feature = "fused", since = "1.26.0")] impl<A: Clone> FusedIterator for Repeat<A> {} +#[unstable(feature = "trusted_len", issue = "37572")] +unsafe impl<A: Clone> TrustedLen for Repeat<A> {} + /// Creates a new iterator that endlessly repeats a single element. /// /// The `repeat()` function repeats a single value over and over and over and @@ -54,6 +57,12 @@ impl<A: Clone> FusedIterator for Repeat<A> {} /// /// [`take`]: trait.Iterator.html#method.take /// +/// If the element type of the iterator you need does not implement `Clone`, +/// or if you do not want to keep the repeated element in memory, you can +/// instead use the [`repeat_with`] function. +/// +/// [`repeat_with`]: fn.repeat_with.html +/// /// # Examples /// /// Basic usage: @@ -96,6 +105,103 @@ pub fn repeat<T: Clone>(elt: T) -> Repeat<T> { Repeat{element: elt} } +/// An iterator that repeats elements of type `A` endlessly by +/// applying the provided closure `F: FnMut() -> A`. +/// +/// This `struct` is created by the [`repeat_with`] function. +/// See its documentation for more. +/// +/// [`repeat_with`]: fn.repeat_with.html +#[derive(Copy, Clone, Debug)] +#[stable(feature = "iterator_repeat_with", since = "1.28.0")] +pub struct RepeatWith<F> { + repeater: F +} + +#[stable(feature = "iterator_repeat_with", since = "1.28.0")] +impl<A, F: FnMut() -> A> Iterator for RepeatWith<F> { + type Item = A; + + #[inline] + fn next(&mut self) -> Option<A> { Some((self.repeater)()) } + + #[inline] + fn size_hint(&self) -> (usize, Option<usize>) { (usize::MAX, None) } +} + +#[stable(feature = "iterator_repeat_with", since = "1.28.0")] +impl<A, F: FnMut() -> A> FusedIterator for RepeatWith<F> {} + +#[unstable(feature = "trusted_len", issue = "37572")] +unsafe impl<A, F: FnMut() -> A> TrustedLen for RepeatWith<F> {} + +/// Creates a new iterator that repeats elements of type `A` endlessly by +/// applying the provided closure, the repeater, `F: FnMut() -> A`. +/// +/// The `repeat_with()` function calls the repeater over and over and over and +/// over and over and 🔁. +/// +/// Infinite iterators like `repeat_with()` are often used with adapters like +/// [`take`], in order to make them finite. +/// +/// [`take`]: trait.Iterator.html#method.take +/// +/// If the element type of the iterator you need implements `Clone`, and +/// it is OK to keep the source element in memory, you should instead use +/// the [`repeat`] function. +/// +/// [`repeat`]: fn.repeat.html +/// +/// An iterator produced by `repeat_with()` is not a `DoubleEndedIterator`. +/// If you need `repeat_with()` to return a `DoubleEndedIterator`, +/// please open a GitHub issue explaining your use case. +/// +/// # Examples +/// +/// Basic usage: +/// +/// ``` +/// use std::iter; +/// +/// // let's assume we have some value of a type that is not `Clone` +/// // or which don't want to have in memory just yet because it is expensive: +/// #[derive(PartialEq, Debug)] +/// struct Expensive; +/// +/// // a particular value forever: +/// let mut things = iter::repeat_with(|| Expensive); +/// +/// assert_eq!(Some(Expensive), things.next()); +/// assert_eq!(Some(Expensive), things.next()); +/// assert_eq!(Some(Expensive), things.next()); +/// assert_eq!(Some(Expensive), things.next()); +/// assert_eq!(Some(Expensive), things.next()); +/// ``` +/// +/// Using mutation and going finite: +/// +/// ```rust +/// use std::iter; +/// +/// // From the zeroth to the third power of two: +/// let mut curr = 1; +/// let mut pow2 = iter::repeat_with(|| { let tmp = curr; curr *= 2; tmp }) +/// .take(4); +/// +/// assert_eq!(Some(1), pow2.next()); +/// assert_eq!(Some(2), pow2.next()); +/// assert_eq!(Some(4), pow2.next()); +/// assert_eq!(Some(8), pow2.next()); +/// +/// // ... and now we're done +/// assert_eq!(None, pow2.next()); +/// ``` +#[inline] +#[stable(feature = "iterator_repeat_with", since = "1.28.0")] +pub fn repeat_with<A, F: FnMut() -> A>(repeater: F) -> RepeatWith<F> { + RepeatWith { repeater } +} + /// An iterator that yields nothing. /// /// This `struct` is created by the [`empty`] function. See its documentation for more. @@ -141,7 +247,7 @@ impl<T> ExactSizeIterator for Empty<T> { #[unstable(feature = "trusted_len", issue = "37572")] unsafe impl<T> TrustedLen for Empty<T> {} -#[unstable(feature = "fused", issue = "35602")] +#[stable(feature = "fused", since = "1.26.0")] impl<T> FusedIterator for Empty<T> {} // not #[derive] because that adds a Clone bound on T, @@ -222,7 +328,7 @@ impl<T> ExactSizeIterator for Once<T> { #[unstable(feature = "trusted_len", issue = "37572")] unsafe impl<T> TrustedLen for Once<T> {} -#[unstable(feature = "fused", issue = "35602")] +#[stable(feature = "fused", since = "1.26.0")] impl<T> FusedIterator for Once<T> {} /// Creates an iterator that yields an element exactly once. diff --git a/src/libcore/iter/traits.rs b/src/libcore/iter/traits.rs index 11e668d228c..4b2c1aa551e 100644 --- a/src/libcore/iter/traits.rs +++ b/src/libcore/iter/traits.rs @@ -10,7 +10,7 @@ use ops::{Mul, Add, Try}; use num::Wrapping; -use super::{AlwaysOk, LoopState}; +use super::LoopState; /// Conversion from an `Iterator`. /// @@ -104,8 +104,11 @@ use super::{AlwaysOk, LoopState}; /// assert_eq!(c.0, vec![0, 1, 2, 3, 4]); /// ``` #[stable(feature = "rust1", since = "1.0.0")] -#[rustc_on_unimplemented="a collection of type `{Self}` cannot be \ - built from an iterator over elements of type `{A}`"] +#[rustc_on_unimplemented( + message="a collection of type `{Self}` cannot be built from an iterator \ + over elements of type `{A}`", + label="a collection of type `{Self}` cannot be built from `std::iter::Iterator<Item={A}>`", +)] pub trait FromIterator<A>: Sized { /// Creates a value from an iterator. /// @@ -352,6 +355,13 @@ pub trait Extend<A> { fn extend<T: IntoIterator<Item=A>>(&mut self, iter: T); } +#[stable(feature = "extend_for_unit", since = "1.28.0")] +impl Extend<()> for () { + fn extend<T: IntoIterator<Item = ()>>(&mut self, iter: T) { + iter.into_iter().for_each(drop) + } +} + /// An iterator able to yield elements from both ends. /// /// Something that implements `DoubleEndedIterator` has one extra capability @@ -427,7 +437,6 @@ pub trait DoubleEndedIterator: Iterator { /// Basic usage: /// /// ``` - /// #![feature(iterator_try_fold)] /// let a = ["1", "2", "3"]; /// let sum = a.iter() /// .map(|&s| s.parse::<i32>()) @@ -438,7 +447,6 @@ pub trait DoubleEndedIterator: Iterator { /// Short-circuiting: /// /// ``` - /// #![feature(iterator_try_fold)] /// let a = ["1", "rust", "3"]; /// let mut it = a.iter(); /// let sum = it @@ -452,7 +460,7 @@ pub trait DoubleEndedIterator: Iterator { /// assert_eq!(it.next_back(), Some(&"1")); /// ``` #[inline] - #[unstable(feature = "iterator_try_fold", issue = "45594")] + #[stable(feature = "iterator_try_fold", since = "1.27.0")] fn try_rfold<B, F, R>(&mut self, init: B, mut f: F) -> R where Self: Sized, F: FnMut(B, Self::Item) -> R, R: Try<Ok=B> { @@ -491,7 +499,6 @@ pub trait DoubleEndedIterator: Iterator { /// Basic usage: /// /// ``` - /// #![feature(iter_rfold)] /// let a = [1, 2, 3]; /// /// // the sum of all of the elements of a @@ -505,7 +512,6 @@ pub trait DoubleEndedIterator: Iterator { /// and continuing with each element from the back until the front: /// /// ``` - /// #![feature(iter_rfold)] /// let numbers = [1, 2, 3, 4, 5]; /// /// let zero = "0".to_string(); @@ -517,14 +523,14 @@ pub trait DoubleEndedIterator: Iterator { /// assert_eq!(result, "(1 + (2 + (3 + (4 + (5 + 0)))))"); /// ``` #[inline] - #[unstable(feature = "iter_rfold", issue = "44705")] + #[stable(feature = "iter_rfold", since = "1.27.0")] fn rfold<B, F>(mut self, accum: B, mut f: F) -> B where Self: Sized, F: FnMut(B, Self::Item) -> B, { - self.try_rfold(accum, move |acc, x| AlwaysOk(f(acc, x))).0 + self.try_rfold(accum, move |acc, x| Ok::<B, !>(f(acc, x))).unwrap() } - /// Searches for an element of an iterator from the right that satisfies a predicate. + /// Searches for an element of an iterator from the back that satisfies a predicate. /// /// `rfind()` takes a closure that returns `true` or `false`. It applies /// this closure to each element of the iterator, starting at the end, and if any @@ -547,8 +553,6 @@ pub trait DoubleEndedIterator: Iterator { /// Basic usage: /// /// ``` - /// #![feature(iter_rfind)] - /// /// let a = [1, 2, 3]; /// /// assert_eq!(a.iter().rfind(|&&x| x == 2), Some(&2)); @@ -559,8 +563,6 @@ pub trait DoubleEndedIterator: Iterator { /// Stopping at the first `true`: /// /// ``` - /// #![feature(iter_rfind)] - /// /// let a = [1, 2, 3]; /// /// let mut iter = a.iter(); @@ -571,7 +573,7 @@ pub trait DoubleEndedIterator: Iterator { /// assert_eq!(iter.next_back(), Some(&1)); /// ``` #[inline] - #[unstable(feature = "iter_rfind", issue = "39480")] + #[stable(feature = "iter_rfind", since = "1.27.0")] fn rfind<P>(&mut self, mut predicate: P) -> Option<Self::Item> where Self: Sized, P: FnMut(&Self::Item) -> bool @@ -595,7 +597,7 @@ impl<'a, I: DoubleEndedIterator + ?Sized> DoubleEndedIterator for &'a mut I { /// that information can be useful. For example, if you want to iterate /// backwards, a good start is to know where the end is. /// -/// When implementing an `ExactSizeIterator`, You must also implement +/// When implementing an `ExactSizeIterator`, you must also implement /// [`Iterator`]. When doing so, the implementation of [`size_hint`] *must* /// return the exact size of the iterator. /// @@ -706,7 +708,7 @@ pub trait ExactSizeIterator: Iterator { /// ``` /// #![feature(exact_size_is_empty)] /// - /// let mut one_element = 0..1; + /// let mut one_element = std::iter::once(0); /// assert!(!one_element.is_empty()); /// /// assert_eq!(one_element.next(), Some(0)); @@ -901,6 +903,15 @@ impl<I, T, E> Iterator for ResultShunt<I, E> None => None, } } + + fn size_hint(&self) -> (usize, Option<usize>) { + if self.error.is_some() { + (0, Some(0)) + } else { + let (_, upper) = self.iter.size_hint(); + (0, upper) + } + } } #[stable(feature = "iter_arith_traits_result", since="1.16.0")] @@ -959,10 +970,10 @@ impl<T, U, E> Product<Result<U, E>> for Result<T, E> /// [`None`]: ../../std/option/enum.Option.html#variant.None /// [`Iterator::fuse`]: ../../std/iter/trait.Iterator.html#method.fuse /// [`Fuse`]: ../../std/iter/struct.Fuse.html -#[unstable(feature = "fused", issue = "35602")] +#[stable(feature = "fused", since = "1.26.0")] pub trait FusedIterator: Iterator {} -#[unstable(feature = "fused", issue = "35602")] +#[stable(feature = "fused", since = "1.26.0")] impl<'a, I: FusedIterator + ?Sized> FusedIterator for &'a mut I {} /// An iterator that reports an accurate length using size_hint. @@ -970,9 +981,11 @@ impl<'a, I: FusedIterator + ?Sized> FusedIterator for &'a mut I {} /// The iterator reports a size hint where it is either exact /// (lower bound is equal to upper bound), or the upper bound is [`None`]. /// The upper bound must only be [`None`] if the actual iterator length is -/// larger than [`usize::MAX`]. +/// larger than [`usize::MAX`]. In that case, the lower bound must be +/// [`usize::MAX`], resulting in a [`.size_hint`] of `(usize::MAX, None)`. /// -/// The iterator must produce exactly the number of elements it reported. +/// The iterator must produce exactly the number of elements it reported +/// or diverge before reaching the end. /// /// # Safety /// diff --git a/src/libcore/lib.rs b/src/libcore/lib.rs index c49292481c3..055da19de60 100644 --- a/src/libcore/lib.rs +++ b/src/libcore/lib.rs @@ -41,7 +41,7 @@ //! dictate the panic message, the file at which panic was invoked, and the //! line and column inside the file. It is up to consumers of this core //! library to define this panic function; it is only required to never -//! return. This requires a `lang` attribute named `panic_fmt`. +//! return. This requires a `lang` attribute named `panic_impl`. //! //! * `rust_eh_personality` - is used by the failure mechanisms of the //! compiler. This is often mapped to GCC's personality function, but crates @@ -50,6 +50,15 @@ // Since libcore defines many fundamental lang items, all tests live in a // separate crate, libcoretest, to avoid bizarre issues. +// +// Here we explicitly #[cfg]-out this whole crate when testing. If we don't do +// this, both the generated test artifact and the linked libtest (which +// transitively includes libcore) will both define the same set of lang items, +// and this will cause the E0152 "duplicate lang item found" error. See +// discussion in #50466 for details. +// +// This cfg won't affect doc tests. +#![cfg(not(test))] #![stable(feature = "core", since = "1.6.0")] #![doc(html_logo_url = "https://www.rust-lang.org/logos/rust-logo-128x128-blk-v2.png", @@ -63,35 +72,56 @@ #![no_core] #![deny(missing_docs)] #![deny(missing_debug_implementations)] -#![deny(warnings)] #![feature(allow_internal_unstable)] +#![feature(arbitrary_self_types)] #![feature(asm)] #![feature(associated_type_defaults)] -#![feature(cfg_target_feature)] +#![feature(attr_literals)] #![feature(cfg_target_has_atomic)] #![feature(concat_idents)] #![feature(const_fn)] +#![feature(const_int_ops)] +#![feature(const_fn_union)] #![feature(convert_id)] #![feature(custom_attribute)] +#![feature(doc_cfg)] +#![feature(doc_spotlight)] +#![feature(extern_types)] #![feature(fundamental)] -#![feature(i128_type)] -#![feature(inclusive_range_syntax)] #![feature(intrinsics)] #![feature(lang_items)] +#![feature(link_llvm_intrinsics)] #![feature(never_type)] +#![cfg_attr(not(stage0), feature(nll))] +#![feature(exhaustive_patterns)] +#![feature(macro_at_most_once_rep)] #![feature(no_core)] #![feature(on_unimplemented)] #![feature(optin_builtin_traits)] #![feature(prelude_import)] #![feature(repr_simd, platform_intrinsics)] #![feature(rustc_attrs)] +#![feature(rustc_const_unstable)] +#![feature(simd_ffi)] #![feature(specialization)] #![feature(staged_api)] +#![feature(stmt_expr_attributes)] #![feature(unboxed_closures)] #![feature(untagged_unions)] #![feature(unwind_attributes)] -#![feature(doc_spotlight)] +#![feature(doc_alias)] +#![feature(mmx_target_feature)] +#![feature(tbm_target_feature)] +#![feature(sse4a_target_feature)] +#![feature(arm_target_feature)] +#![feature(powerpc_target_feature)] +#![feature(mips_target_feature)] +#![feature(aarch64_target_feature)] +#![feature(const_slice_len)] +#![feature(const_str_as_bytes)] +#![feature(const_str_len)] +#![feature(non_exhaustive)] #[prelude_import] #[allow(unused)] @@ -116,14 +146,14 @@ mod uint_macros; #[path = "num/i16.rs"] pub mod i16; #[path = "num/i32.rs"] pub mod i32; #[path = "num/i64.rs"] pub mod i64; -#[path = "num/i128.rs"] pub mod i128; +#[path = "num/i128.rs"] pub mod i128; #[path = "num/usize.rs"] pub mod usize; #[path = "num/u8.rs"] pub mod u8; #[path = "num/u16.rs"] pub mod u16; #[path = "num/u32.rs"] pub mod u32; #[path = "num/u64.rs"] pub mod u64; -#[path = "num/u128.rs"] pub mod u128; +#[path = "num/u128.rs"] pub mod u128; #[path = "num/f32.rs"] pub mod f32; #[path = "num/f64.rs"] pub mod f64; @@ -139,8 +169,8 @@ pub mod prelude; pub mod intrinsics; pub mod mem; -pub mod nonzero; pub mod ptr; +pub mod hint; /* Core language traits */ @@ -156,9 +186,11 @@ pub mod borrow; pub mod any; pub mod array; +pub mod ascii; pub mod sync; pub mod cell; pub mod char; +pub mod panic; pub mod panicking; pub mod iter; pub mod option; @@ -169,9 +201,49 @@ pub mod slice; pub mod str; pub mod hash; pub mod fmt; +pub mod time; + +pub mod unicode; + +/* Async */ +pub mod future; +pub mod task; + +/* Heap memory allocator trait */ +#[allow(missing_docs)] +pub mod alloc; // note: does not need to be public -mod char_private; mod iter_private; +mod nonzero; mod tuple; mod unit; + +// Pull in the the `coresimd` crate directly into libcore. This is where all the +// architecture-specific (and vendor-specific) intrinsics are defined. AKA +// things like SIMD and such. Note that the actual source for all this lies in a +// different repository, rust-lang-nursery/stdsimd. That's why the setup here is +// a bit wonky. +#[allow(unused_macros)] +macro_rules! test_v16 { ($item:item) => {}; } +#[allow(unused_macros)] +macro_rules! test_v32 { ($item:item) => {}; } +#[allow(unused_macros)] +macro_rules! test_v64 { ($item:item) => {}; } +#[allow(unused_macros)] +macro_rules! test_v128 { ($item:item) => {}; } +#[allow(unused_macros)] +macro_rules! test_v256 { ($item:item) => {}; } +#[allow(unused_macros)] +macro_rules! test_v512 { ($item:item) => {}; } +#[allow(unused_macros)] +macro_rules! vector_impl { ($([$f:ident, $($args:tt)*]),*) => { $($f!($($args)*);)* } } +#[path = "../stdsimd/coresimd/mod.rs"] +#[allow(missing_docs, missing_debug_implementations, dead_code, unused_imports)] +#[unstable(feature = "stdsimd", issue = "48556")] +#[cfg(not(stage0))] // allow changes to how stdsimd works in stage0 +mod coresimd; + +#[stable(feature = "simd_arch", since = "1.27.0")] +#[cfg(not(stage0))] +pub use coresimd::arch; diff --git a/src/libcore/macros.rs b/src/libcore/macros.rs index f00128a8147..5b3b2d16356 100644 --- a/src/libcore/macros.rs +++ b/src/libcore/macros.rs @@ -19,73 +19,15 @@ macro_rules! panic { ($msg:expr) => ({ $crate::panicking::panic(&($msg, file!(), line!(), __rust_unstable_column!())) }); - ($fmt:expr, $($arg:tt)*) => ({ + ($msg:expr,) => ( + panic!($msg) + ); + ($fmt:expr, $($arg:tt)+) => ({ $crate::panicking::panic_fmt(format_args!($fmt, $($arg)*), &(file!(), line!(), __rust_unstable_column!())) }); } -/// Ensure that a boolean expression is `true` at runtime. -/// -/// This will invoke the [`panic!`] macro if the provided expression cannot be -/// evaluated to `true` at runtime. -/// -/// # Uses -/// -/// Assertions are always checked in both debug and release builds, and cannot -/// be disabled. See [`debug_assert!`] for assertions that are not enabled in -/// release builds by default. -/// -/// Unsafe code relies on `assert!` to enforce run-time invariants that, if -/// violated could lead to unsafety. -/// -/// Other use-cases of `assert!` include [testing] and enforcing run-time -/// invariants in safe code (whose violation cannot result in unsafety). -/// -/// # Custom Messages -/// -/// This macro has a second form, where a custom panic message can -/// be provided with or without arguments for formatting. See [`std::fmt`] -/// for syntax for this form. -/// -/// [`panic!`]: macro.panic.html -/// [`debug_assert!`]: macro.debug_assert.html -/// [testing]: ../book/second-edition/ch11-01-writing-tests.html#checking-results-with-the-assert-macro -/// [`std::fmt`]: ../std/fmt/index.html -/// -/// # Examples -/// -/// ``` -/// // the panic message for these assertions is the stringified value of the -/// // expression given. -/// assert!(true); -/// -/// fn some_computation() -> bool { true } // a very simple function -/// -/// assert!(some_computation()); -/// -/// // assert with a custom message -/// let x = true; -/// assert!(x, "x wasn't true!"); -/// -/// let a = 3; let b = 27; -/// assert!(a + b == 30, "a = {}, b = {}", a, b); -/// ``` -#[macro_export] -#[stable(feature = "rust1", since = "1.0.0")] -macro_rules! assert { - ($cond:expr) => ( - if !$cond { - panic!(concat!("assertion failed: ", stringify!($cond))) - } - ); - ($cond:expr, $($arg:tt)+) => ( - if !$cond { - panic!($($arg)+) - } - ); -} - /// Asserts that two expressions are equal to each other (using [`PartialEq`]). /// /// On panic, this macro will print the values of the expressions with their @@ -327,7 +269,7 @@ macro_rules! debug_assert_ne { /// } /// } /// -/// // The prefered method of quick returning Errors +/// // The preferred method of quick returning Errors /// fn write_to_file_question() -> Result<(), MyError> { /// let mut file = File::create("my_best_friends.txt")?; /// file.write_all(b"This is a list of my best friends.")?; @@ -353,13 +295,15 @@ macro_rules! debug_assert_ne { /// ``` #[macro_export] #[stable(feature = "rust1", since = "1.0.0")] +#[doc(alias = "?")] macro_rules! try { ($expr:expr) => (match $expr { $crate::result::Result::Ok(val) => val, $crate::result::Result::Err(err) => { return $crate::result::Result::Err($crate::convert::From::from(err)) } - }) + }); + ($expr:expr,) => (try!($expr)); } /// Write formatted data into a buffer. @@ -452,15 +396,16 @@ macro_rules! write { /// ``` #[macro_export] #[stable(feature = "rust1", since = "1.0.0")] +#[allow_internal_unstable] macro_rules! writeln { ($dst:expr) => ( write!($dst, "\n") ); - ($dst:expr, $fmt:expr) => ( - write!($dst, concat!($fmt, "\n")) + ($dst:expr,) => ( + writeln!($dst) ); - ($dst:expr, $fmt:expr, $($arg:tt)*) => ( - write!($dst, concat!($fmt, "\n"), $($arg)*) + ($dst:expr, $($arg:tt)*) => ( + $dst.write_fmt(format_args_nl!($($arg)*)) ); } @@ -474,13 +419,13 @@ macro_rules! writeln { /// * Iterators that dynamically terminate. /// /// If the determination that the code is unreachable proves incorrect, the -/// program immediately terminates with a [`panic!`]. The function [`unreachable`], -/// which belongs to the [`std::intrinsics`] module, informs the compilier to +/// program immediately terminates with a [`panic!`]. The function [`unreachable_unchecked`], +/// which belongs to the [`std::hint`] module, informs the compilier to /// optimize the code out of the release version entirely. /// /// [`panic!`]: ../std/macro.panic.html -/// [`unreachable`]: ../std/intrinsics/fn.unreachable.html -/// [`std::intrinsics`]: ../std/intrinsics/index.html +/// [`unreachable_unchecked`]: ../std/hint/fn.unreachable_unchecked.html +/// [`std::hint`]: ../std/hint/index.html /// /// # Panics /// @@ -524,6 +469,9 @@ macro_rules! unreachable { ($msg:expr) => ({ unreachable!("{}", $msg) }); + ($msg:expr,) => ({ + unreachable!($msg) + }); ($fmt:expr, $($arg:tt)*) => ({ panic!(concat!("internal error: entered unreachable code: ", $fmt), $($arg)*) }); @@ -593,6 +541,7 @@ macro_rules! unimplemented { /// into libsyntax itself. /// /// For more information, see documentation for `std`'s macros. +#[cfg(dox)] mod builtin { /// Unconditionally causes compilation to fail with the given error message when encountered. @@ -601,9 +550,11 @@ mod builtin { /// /// [`std::compile_error!`]: ../std/macro.compile_error.html #[stable(feature = "compile_error_macro", since = "1.20.0")] - #[macro_export] - #[cfg(dox)] - macro_rules! compile_error { ($msg:expr) => ({ /* compiler built-in */ }) } + #[rustc_doc_only_macro] + macro_rules! compile_error { + ($msg:expr) => ({ /* compiler built-in */ }); + ($msg:expr,) => ({ /* compiler built-in */ }); + } /// The core macro for formatted string creation & output. /// @@ -611,8 +562,7 @@ mod builtin { /// /// [`std::format_args!`]: ../std/macro.format_args.html #[stable(feature = "rust1", since = "1.0.0")] - #[macro_export] - #[cfg(dox)] + #[rustc_doc_only_macro] macro_rules! format_args { ($fmt:expr) => ({ /* compiler built-in */ }); ($fmt:expr, $($args:tt)*) => ({ /* compiler built-in */ }); @@ -624,8 +574,7 @@ mod builtin { /// /// [`std::env!`]: ../std/macro.env.html #[stable(feature = "rust1", since = "1.0.0")] - #[macro_export] - #[cfg(dox)] + #[rustc_doc_only_macro] macro_rules! env { ($name:expr) => ({ /* compiler built-in */ }); ($name:expr,) => ({ /* compiler built-in */ }); @@ -637,9 +586,11 @@ mod builtin { /// /// [`std::option_env!`]: ../std/macro.option_env.html #[stable(feature = "rust1", since = "1.0.0")] - #[macro_export] - #[cfg(dox)] - macro_rules! option_env { ($name:expr) => ({ /* compiler built-in */ }) } + #[rustc_doc_only_macro] + macro_rules! option_env { + ($name:expr) => ({ /* compiler built-in */ }); + ($name:expr,) => ({ /* compiler built-in */ }); + } /// Concatenate identifiers into one identifier. /// @@ -647,11 +598,10 @@ mod builtin { /// /// [`std::concat_idents!`]: ../std/macro.concat_idents.html #[unstable(feature = "concat_idents_macro", issue = "29599")] - #[macro_export] - #[cfg(dox)] + #[rustc_doc_only_macro] macro_rules! concat_idents { - ($($e:ident),*) => ({ /* compiler built-in */ }); - ($($e:ident,)*) => ({ /* compiler built-in */ }); + ($($e:ident),+) => ({ /* compiler built-in */ }); + ($($e:ident,)+) => ({ /* compiler built-in */ }); } /// Concatenates literals into a static string slice. @@ -660,8 +610,7 @@ mod builtin { /// /// [`std::concat!`]: ../std/macro.concat.html #[stable(feature = "rust1", since = "1.0.0")] - #[macro_export] - #[cfg(dox)] + #[rustc_doc_only_macro] macro_rules! concat { ($($e:expr),*) => ({ /* compiler built-in */ }); ($($e:expr,)*) => ({ /* compiler built-in */ }); @@ -673,8 +622,7 @@ mod builtin { /// /// [`std::line!`]: ../std/macro.line.html #[stable(feature = "rust1", since = "1.0.0")] - #[macro_export] - #[cfg(dox)] + #[rustc_doc_only_macro] macro_rules! line { () => ({ /* compiler built-in */ }) } /// A macro which expands to the column number on which it was invoked. @@ -683,8 +631,7 @@ mod builtin { /// /// [`std::column!`]: ../std/macro.column.html #[stable(feature = "rust1", since = "1.0.0")] - #[macro_export] - #[cfg(dox)] + #[rustc_doc_only_macro] macro_rules! column { () => ({ /* compiler built-in */ }) } /// A macro which expands to the file name from which it was invoked. @@ -693,8 +640,7 @@ mod builtin { /// /// [`std::file!`]: ../std/macro.file.html #[stable(feature = "rust1", since = "1.0.0")] - #[macro_export] - #[cfg(dox)] + #[rustc_doc_only_macro] macro_rules! file { () => ({ /* compiler built-in */ }) } /// A macro which stringifies its arguments. @@ -703,8 +649,7 @@ mod builtin { /// /// [`std::stringify!`]: ../std/macro.stringify.html #[stable(feature = "rust1", since = "1.0.0")] - #[macro_export] - #[cfg(dox)] + #[rustc_doc_only_macro] macro_rules! stringify { ($($t:tt)*) => ({ /* compiler built-in */ }) } /// Includes a utf8-encoded file as a string. @@ -713,9 +658,11 @@ mod builtin { /// /// [`std::include_str!`]: ../std/macro.include_str.html #[stable(feature = "rust1", since = "1.0.0")] - #[macro_export] - #[cfg(dox)] - macro_rules! include_str { ($file:expr) => ({ /* compiler built-in */ }) } + #[rustc_doc_only_macro] + macro_rules! include_str { + ($file:expr) => ({ /* compiler built-in */ }); + ($file:expr,) => ({ /* compiler built-in */ }); + } /// Includes a file as a reference to a byte array. /// @@ -723,9 +670,11 @@ mod builtin { /// /// [`std::include_bytes!`]: ../std/macro.include_bytes.html #[stable(feature = "rust1", since = "1.0.0")] - #[macro_export] - #[cfg(dox)] - macro_rules! include_bytes { ($file:expr) => ({ /* compiler built-in */ }) } + #[rustc_doc_only_macro] + macro_rules! include_bytes { + ($file:expr) => ({ /* compiler built-in */ }); + ($file:expr,) => ({ /* compiler built-in */ }); + } /// Expands to a string that represents the current module path. /// @@ -733,8 +682,7 @@ mod builtin { /// /// [`std::module_path!`]: ../std/macro.module_path.html #[stable(feature = "rust1", since = "1.0.0")] - #[macro_export] - #[cfg(dox)] + #[rustc_doc_only_macro] macro_rules! module_path { () => ({ /* compiler built-in */ }) } /// Boolean evaluation of configuration flags, at compile-time. @@ -743,8 +691,7 @@ mod builtin { /// /// [`std::cfg!`]: ../std/macro.cfg.html #[stable(feature = "rust1", since = "1.0.0")] - #[macro_export] - #[cfg(dox)] + #[rustc_doc_only_macro] macro_rules! cfg { ($($cfg:tt)*) => ({ /* compiler built-in */ }) } /// Parse a file as an expression or an item according to the context. @@ -753,7 +700,22 @@ mod builtin { /// /// [`std::include!`]: ../std/macro.include.html #[stable(feature = "rust1", since = "1.0.0")] - #[macro_export] - #[cfg(dox)] - macro_rules! include { ($file:expr) => ({ /* compiler built-in */ }) } + #[rustc_doc_only_macro] + macro_rules! include { + ($file:expr) => ({ /* compiler built-in */ }); + ($file:expr,) => ({ /* compiler built-in */ }); + } + + /// Ensure that a boolean expression is `true` at runtime. + /// + /// For more information, see the documentation for [`std::assert!`]. + /// + /// [`std::assert!`]: ../std/macro.assert.html + #[rustc_doc_only_macro] + #[stable(feature = "rust1", since = "1.0.0")] + macro_rules! assert { + ($cond:expr) => ({ /* compiler built-in */ }); + ($cond:expr,) => ({ /* compiler built-in */ }); + ($cond:expr, $($arg:tt)+) => ({ /* compiler built-in */ }); + } } diff --git a/src/libcore/marker.rs b/src/libcore/marker.rs index 3032fb2de33..d18e167fc3f 100644 --- a/src/libcore/marker.rs +++ b/src/libcore/marker.rs @@ -39,7 +39,10 @@ use hash::Hasher; /// [arc]: ../../std/sync/struct.Arc.html /// [ub]: ../../reference/behavior-considered-undefined.html #[stable(feature = "rust1", since = "1.0.0")] -#[rustc_on_unimplemented = "`{Self}` cannot be sent between threads safely"] +#[rustc_on_unimplemented( + message="`{Self}` cannot be sent between threads safely", + label="`{Self}` cannot be sent between threads safely" +)] pub unsafe auto trait Send { // empty. } @@ -63,9 +66,13 @@ impl<T: ?Sized> !Send for *mut T { } /// struct BarUse(Bar<[i32]>); // OK /// ``` /// -/// The one exception is the implicit `Self` type of a trait, which does not -/// get an implicit `Sized` bound. This is because a `Sized` bound prevents -/// the trait from being used to form a [trait object]: +/// The one exception is the implicit `Self` type of a trait. A trait does not +/// have an implicit `Sized` bound as this is incompatible with [trait object]s +/// where, by definition, the trait needs to work with all possible implementors, +/// and thus could be any size. +/// +/// Although Rust will let you bind `Sized` to a trait, you won't +/// be able to use it to form a trait object later: /// /// ``` /// # #![allow(unused_variables)] @@ -84,7 +91,12 @@ impl<T: ?Sized> !Send for *mut T { } /// [trait object]: ../../book/first-edition/trait-objects.html #[stable(feature = "rust1", since = "1.0.0")] #[lang = "sized"] -#[rustc_on_unimplemented = "`{Self}` does not have a constant size known at compile-time"] +#[rustc_on_unimplemented( + message="the size for values of type `{Self}` cannot be known at compilation time", + label="doesn't have a size known at compile-time", + note="to learn more, visit <https://doc.rust-lang.org/book/second-edition/\ + ch19-04-advanced-types.html#dynamically-sized-types-and-the-sized-trait>", +)] #[fundamental] // for Default, for example, which requires that `[T]: !Default` be evaluatable pub trait Sized { // Empty. @@ -256,6 +268,21 @@ pub trait Unsize<T: ?Sized> { /// non-`Copy` in the future, it could be prudent to omit the `Copy` implementation now, to /// avoid a breaking API change. /// +/// ## Additional implementors +/// +/// In addition to the [implementors listed below][impls], +/// the following types also implement `Copy`: +/// +/// * Function item types (i.e. the distinct types defined for each function) +/// * Function pointer types (e.g. `fn() -> i32`) +/// * Array types, for all sizes, if the item type also implements `Copy` (e.g. `[i32; 123456]`) +/// * Tuple types, if each component also implements `Copy` (e.g. `()`, `(i32, bool)`) +/// * Closure types, if they capture no value from the environment +/// or if all such captured values implement `Copy` themselves. +/// Note that variables captured by shared reference always implement `Copy` +/// (even if the referent doesn't), +/// while variables captured by mutable reference never implement `Copy`. +/// /// [`Vec<T>`]: ../../std/vec/struct.Vec.html /// [`String`]: ../../std/string/struct.String.html /// [`Drop`]: ../../std/ops/trait.Drop.html @@ -263,6 +290,7 @@ pub trait Unsize<T: ?Sized> { /// [`Clone`]: ../clone/trait.Clone.html /// [`String`]: ../../std/string/struct.String.html /// [`i32`]: ../../std/primitive.i32.html +/// [impls]: #implementors #[stable(feature = "rust1", since = "1.0.0")] #[lang = "copy"] pub trait Copy : Clone { @@ -274,7 +302,7 @@ pub trait Copy : Clone { /// This trait is automatically implemented when the compiler determines /// it's appropriate. /// -/// The precise definition is: a type `T` is `Sync` if `&T` is +/// The precise definition is: a type `T` is `Sync` if and only if `&T` is /// [`Send`][send]. In other words, if there is no possibility of /// [undefined behavior][ub] (including data races) when passing /// `&T` references between threads. @@ -339,8 +367,21 @@ pub trait Copy : Clone { /// [transmute]: ../../std/mem/fn.transmute.html #[stable(feature = "rust1", since = "1.0.0")] #[lang = "sync"] -#[rustc_on_unimplemented = "`{Self}` cannot be shared between threads safely"] +#[rustc_on_unimplemented( + message="`{Self}` cannot be shared between threads safely", + label="`{Self}` cannot be shared between threads safely" +)] pub unsafe auto trait Sync { + // FIXME(estebank): once support to add notes in `rustc_on_unimplemented` + // lands in beta, and it has been extended to check whether a closure is + // anywhere in the requirement chain, extend it as such (#48534): + // ``` + // on( + // closure, + // note="`{Self}` cannot be shared safely, consider marking the closure `move`" + // ), + // ``` + // Empty } @@ -561,3 +602,70 @@ unsafe impl<T: ?Sized> Freeze for *const T {} unsafe impl<T: ?Sized> Freeze for *mut T {} unsafe impl<'a, T: ?Sized> Freeze for &'a T {} unsafe impl<'a, T: ?Sized> Freeze for &'a mut T {} + +/// Types which can be moved out of a `PinMut`. +/// +/// The `Unpin` trait is used to control the behavior of the [`PinMut`] type. If a +/// type implements `Unpin`, it is safe to move a value of that type out of the +/// `PinMut` pointer. +/// +/// This trait is automatically implemented for almost every type. +/// +/// [`PinMut`]: ../mem/struct.PinMut.html +#[unstable(feature = "pin", issue = "49150")] +pub auto trait Unpin {} + +/// A type which does not implement `Unpin`. +/// +/// If a type contains a `Pinned`, it will not implement `Unpin` by default. +#[unstable(feature = "pin", issue = "49150")] +#[derive(Debug, Copy, Clone, Eq, PartialEq, Ord, PartialOrd, Hash)] +pub struct Pinned; + +#[unstable(feature = "pin", issue = "49150")] +impl !Unpin for Pinned {} + +#[unstable(feature = "pin", issue = "49150")] +impl<'a, T: ?Sized + 'a> Unpin for &'a T {} + +#[unstable(feature = "pin", issue = "49150")] +impl<'a, T: ?Sized + 'a> Unpin for &'a mut T {} + +/// Implementations of `Copy` for primitive types. +/// +/// Implementations that cannot be described in Rust +/// are implemented in `SelectionContext::copy_clone_conditions()` in librustc. +mod copy_impls { + + use super::Copy; + + macro_rules! impl_copy { + ($($t:ty)*) => { + $( + #[stable(feature = "rust1", since = "1.0.0")] + impl Copy for $t {} + )* + } + } + + impl_copy! { + usize u8 u16 u32 u64 u128 + isize i8 i16 i32 i64 i128 + f32 f64 + bool char + } + + #[unstable(feature = "never_type", issue = "35121")] + impl Copy for ! {} + + #[stable(feature = "rust1", since = "1.0.0")] + impl<T: ?Sized> Copy for *const T {} + + #[stable(feature = "rust1", since = "1.0.0")] + impl<T: ?Sized> Copy for *mut T {} + + // Shared references can be copied, but mutable references *cannot*! + #[stable(feature = "rust1", since = "1.0.0")] + impl<'a, T: ?Sized> Copy for &'a T {} + +} diff --git a/src/libcore/mem.rs b/src/libcore/mem.rs index 93f6a0214d7..ea711c69393 100644 --- a/src/libcore/mem.rs +++ b/src/libcore/mem.rs @@ -18,11 +18,13 @@ use clone; use cmp; use fmt; +use future::{Future, UnsafeFutureObj}; use hash; use intrinsics; -use marker::{Copy, PhantomData, Sized}; +use marker::{Copy, PhantomData, Sized, Unpin, Unsize}; use ptr; -use ops::{Deref, DerefMut}; +use task::{Context, Poll}; +use ops::{Deref, DerefMut, CoerceUnsized}; #[stable(feature = "rust1", since = "1.0.0")] pub use intrinsics::transmute; @@ -189,14 +191,17 @@ pub fn forget<T>(t: T) { /// Type | size_of::\<Type>() /// ---- | --------------- /// () | 0 +/// bool | 1 /// u8 | 1 /// u16 | 2 /// u32 | 4 /// u64 | 8 +/// u128 | 16 /// i8 | 1 /// i16 | 2 /// i32 | 4 /// i64 | 8 +/// i128 | 16 /// f32 | 4 /// f64 | 8 /// char | 4 @@ -224,6 +229,8 @@ pub fn forget<T>(t: T) { /// 2. Round up the current size to the nearest multiple of the next field's [alignment]. /// /// Finally, round the size of the struct to the nearest multiple of its [alignment]. +/// The alignment of the struct is usually the largest alignment of all its +/// fields; this can be changed with the use of `repr(align(N))`. /// /// Unlike `C`, zero sized structs are not rounded up to one byte in size. /// @@ -278,7 +285,8 @@ pub fn forget<T>(t: T) { /// // The size of the second field is 2, so add 2 to the size. Size is 4. /// // The alignment of the third field is 1, so add 0 to the size for padding. Size is 4. /// // The size of the third field is 1, so add 1 to the size. Size is 5. -/// // Finally, the alignment of the struct is 2, so add 1 to the size for padding. Size is 6. +/// // Finally, the alignment of the struct is 2 (because the largest alignment amongst its +/// // fields is 2), so add 1 to the size for padding. Size is 6. /// assert_eq!(6, mem::size_of::<FieldStruct>()); /// /// #[repr(C)] @@ -630,12 +638,13 @@ pub unsafe fn uninitialized<T>() -> T { #[stable(feature = "rust1", since = "1.0.0")] pub fn swap<T>(x: &mut T, y: &mut T) { unsafe { - ptr::swap_nonoverlapping(x, y, 1); + ptr::swap_nonoverlapping_one(x, y); } } -/// Replaces the value at a mutable location with a new one, returning the old value, without -/// deinitializing either one. +/// Moves `src` into the referenced `dest`, returning the previous `dest` value. +/// +/// Neither value is dropped. /// /// # Examples /// @@ -834,7 +843,9 @@ pub unsafe fn transmute_copy<T, U>(src: &T) -> U { /// Opaque type representing the discriminant of an enum. /// -/// See the `discriminant` function in this module for more information. +/// See the [`discriminant`] function in this module for more information. +/// +/// [`discriminant`]: fn.discriminant.html #[stable(feature = "discriminant_value", since = "1.21.0")] pub struct Discriminant<T>(u64, PhantomData<fn() -> T>); @@ -907,7 +918,6 @@ pub fn discriminant<T>(v: &T) -> Discriminant<T> { } } - /// A wrapper to inhibit compiler from automatically calling `T`’s destructor. /// /// This wrapper is 0-cost. @@ -944,9 +954,12 @@ pub fn discriminant<T>(v: &T) -> Discriminant<T> { /// } /// ``` #[stable(feature = "manually_drop", since = "1.20.0")] -#[allow(unions_with_drop_fields)] -#[derive(Copy)] -pub union ManuallyDrop<T>{ value: T } +#[lang = "manually_drop"] +#[derive(Copy, Clone, Debug, Default, PartialEq, Eq, PartialOrd, Ord, Hash)] +#[repr(transparent)] +pub struct ManuallyDrop<T: ?Sized> { + value: T, +} impl<T> ManuallyDrop<T> { /// Wrap a value to be manually dropped. @@ -958,9 +971,10 @@ impl<T> ManuallyDrop<T> { /// ManuallyDrop::new(Box::new(())); /// ``` #[stable(feature = "manually_drop", since = "1.20.0")] + #[rustc_const_unstable(feature = "const_manually_drop_new")] #[inline] - pub fn new(value: T) -> ManuallyDrop<T> { - ManuallyDrop { value: value } + pub const fn new(value: T) -> ManuallyDrop<T> { + ManuallyDrop { value } } /// Extract the value from the ManuallyDrop container. @@ -975,11 +989,11 @@ impl<T> ManuallyDrop<T> { #[stable(feature = "manually_drop", since = "1.20.0")] #[inline] pub fn into_inner(slot: ManuallyDrop<T>) -> T { - unsafe { - slot.value - } + slot.value } +} +impl<T: ?Sized> ManuallyDrop<T> { /// Manually drops the contained value. /// /// # Safety @@ -995,112 +1009,161 @@ impl<T> ManuallyDrop<T> { } #[stable(feature = "manually_drop", since = "1.20.0")] -impl<T> Deref for ManuallyDrop<T> { +impl<T: ?Sized> Deref for ManuallyDrop<T> { type Target = T; #[inline] fn deref(&self) -> &Self::Target { - unsafe { - &self.value - } + &self.value } } #[stable(feature = "manually_drop", since = "1.20.0")] -impl<T> DerefMut for ManuallyDrop<T> { +impl<T: ?Sized> DerefMut for ManuallyDrop<T> { #[inline] fn deref_mut(&mut self) -> &mut Self::Target { - unsafe { - &mut self.value - } + &mut self.value } } -#[stable(feature = "manually_drop", since = "1.20.0")] -impl<T: ::fmt::Debug> ::fmt::Debug for ManuallyDrop<T> { - fn fmt(&self, fmt: &mut ::fmt::Formatter) -> ::fmt::Result { - unsafe { - fmt.debug_tuple("ManuallyDrop").field(&self.value).finish() - } - } +/// A pinned reference. +/// +/// A pinned reference is a lot like a mutable reference, except that it is not +/// safe to move a value out of a pinned reference unless the type of that +/// value implements the `Unpin` trait. +#[unstable(feature = "pin", issue = "49150")] +#[fundamental] +pub struct PinMut<'a, T: ?Sized + 'a> { + inner: &'a mut T, } -#[stable(feature = "manually_drop_impls", since = "1.22.0")] -impl<T: Clone> Clone for ManuallyDrop<T> { - fn clone(&self) -> Self { - ManuallyDrop::new(self.deref().clone()) +#[unstable(feature = "pin", issue = "49150")] +impl<'a, T: ?Sized + Unpin> PinMut<'a, T> { + /// Construct a new `PinMut` around a reference to some data of a type that + /// implements `Unpin`. + #[unstable(feature = "pin", issue = "49150")] + pub fn new(reference: &'a mut T) -> PinMut<'a, T> { + PinMut { inner: reference } } - fn clone_from(&mut self, source: &Self) { - self.deref_mut().clone_from(source); + /// Get a mutable reference to the data inside of this `PinMut`. + #[unstable(feature = "pin", issue = "49150")] + pub fn get_mut(this: PinMut<'a, T>) -> &'a mut T { + this.inner } } -#[stable(feature = "manually_drop_impls", since = "1.22.0")] -impl<T: Default> Default for ManuallyDrop<T> { - fn default() -> Self { - ManuallyDrop::new(Default::default()) - } -} -#[stable(feature = "manually_drop_impls", since = "1.22.0")] -impl<T: PartialEq> PartialEq for ManuallyDrop<T> { - fn eq(&self, other: &Self) -> bool { - self.deref().eq(other) +#[unstable(feature = "pin", issue = "49150")] +impl<'a, T: ?Sized> PinMut<'a, T> { + /// Construct a new `PinMut` around a reference to some data of a type that + /// may or may not implement `Unpin`. + /// + /// This constructor is unsafe because we do not know what will happen with + /// that data after the reference ends. If you cannot guarantee that the + /// data will never move again, calling this constructor is invalid. + #[unstable(feature = "pin", issue = "49150")] + pub unsafe fn new_unchecked(reference: &'a mut T) -> PinMut<'a, T> { + PinMut { inner: reference } } - fn ne(&self, other: &Self) -> bool { - self.deref().ne(other) + /// Reborrow a `PinMut` for a shorter lifetime. + /// + /// For example, `PinMut::get_mut(x.reborrow())` (unsafely) returns a + /// short-lived mutable reference reborrowing from `x`. + #[unstable(feature = "pin", issue = "49150")] + pub fn reborrow<'b>(&'b mut self) -> PinMut<'b, T> { + PinMut { inner: self.inner } } -} -#[stable(feature = "manually_drop_impls", since = "1.22.0")] -impl<T: Eq> Eq for ManuallyDrop<T> {} + /// Get a mutable reference to the data inside of this `PinMut`. + /// + /// This function is unsafe. You must guarantee that you will never move + /// the data out of the mutable reference you receive when you call this + /// function. + #[unstable(feature = "pin", issue = "49150")] + pub unsafe fn get_mut_unchecked(this: PinMut<'a, T>) -> &'a mut T { + this.inner + } -#[stable(feature = "manually_drop_impls", since = "1.22.0")] -impl<T: PartialOrd> PartialOrd for ManuallyDrop<T> { - fn partial_cmp(&self, other: &Self) -> Option<::cmp::Ordering> { - self.deref().partial_cmp(other) + /// Construct a new pin by mapping the interior value. + /// + /// For example, if you wanted to get a `PinMut` of a field of something, + /// you could use this to get access to that field in one line of code. + /// + /// This function is unsafe. You must guarantee that the data you return + /// will not move so long as the argument value does not move (for example, + /// because it is one of the fields of that value), and also that you do + /// not move out of the argument you receive to the interior function. + #[unstable(feature = "pin", issue = "49150")] + pub unsafe fn map_unchecked<U, F>(this: PinMut<'a, T>, f: F) -> PinMut<'a, U> where + F: FnOnce(&mut T) -> &mut U + { + PinMut { inner: f(this.inner) } } - fn lt(&self, other: &Self) -> bool { - self.deref().lt(other) + /// Assign a new value to the memory behind the pinned reference. + #[unstable(feature = "pin", issue = "49150")] + pub fn set(this: PinMut<'a, T>, value: T) + where T: Sized, + { + *this.inner = value; } +} + +#[unstable(feature = "pin", issue = "49150")] +impl<'a, T: ?Sized> Deref for PinMut<'a, T> { + type Target = T; - fn le(&self, other: &Self) -> bool { - self.deref().le(other) + fn deref(&self) -> &T { + &*self.inner } +} - fn gt(&self, other: &Self) -> bool { - self.deref().gt(other) +#[unstable(feature = "pin", issue = "49150")] +impl<'a, T: ?Sized + Unpin> DerefMut for PinMut<'a, T> { + fn deref_mut(&mut self) -> &mut T { + self.inner } +} - fn ge(&self, other: &Self) -> bool { - self.deref().ge(other) +#[unstable(feature = "pin", issue = "49150")] +impl<'a, T: fmt::Debug + ?Sized> fmt::Debug for PinMut<'a, T> { + fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result { + fmt::Debug::fmt(&**self, f) } } -#[stable(feature = "manually_drop_impls", since = "1.22.0")] -impl<T: Ord> Ord for ManuallyDrop<T> { - fn cmp(&self, other: &Self) -> ::cmp::Ordering { - self.deref().cmp(other) +#[unstable(feature = "pin", issue = "49150")] +impl<'a, T: fmt::Display + ?Sized> fmt::Display for PinMut<'a, T> { + fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result { + fmt::Display::fmt(&**self, f) } } -#[stable(feature = "manually_drop_impls", since = "1.22.0")] -impl<T: ::hash::Hash> ::hash::Hash for ManuallyDrop<T> { - fn hash<H: ::hash::Hasher>(&self, state: &mut H) { - self.deref().hash(state); +#[unstable(feature = "pin", issue = "49150")] +impl<'a, T: ?Sized> fmt::Pointer for PinMut<'a, T> { + fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result { + fmt::Pointer::fmt(&(&*self.inner as *const T), f) } } -/// Tells LLVM that this point in the code is not reachable, enabling further -/// optimizations. -/// -/// NB: This is very different from the `unreachable!()` macro: Unlike the -/// macro, which panics when it is executed, it is *undefined behavior* to -/// reach code marked with this function. -#[inline] -#[unstable(feature = "unreachable", issue = "43751")] -pub unsafe fn unreachable() -> ! { - intrinsics::unreachable() +#[unstable(feature = "pin", issue = "49150")] +impl<'a, T: ?Sized + Unsize<U>, U: ?Sized> CoerceUnsized<PinMut<'a, U>> for PinMut<'a, T> {} + +#[unstable(feature = "pin", issue = "49150")] +impl<'a, T: ?Sized> Unpin for PinMut<'a, T> {} + +#[unstable(feature = "futures_api", issue = "50547")] +unsafe impl<'a, T, F> UnsafeFutureObj<'a, T> for PinMut<'a, F> + where F: Future<Output = T> + 'a +{ + fn into_raw(self) -> *mut () { + unsafe { PinMut::get_mut_unchecked(self) as *mut F as *mut () } + } + + unsafe fn poll(ptr: *mut (), cx: &mut Context) -> Poll<T> { + PinMut::new_unchecked(&mut *(ptr as *mut F)).poll(cx) + } + + unsafe fn drop(_ptr: *mut ()) {} } diff --git a/src/libcore/nonzero.rs b/src/libcore/nonzero.rs index 2c966eb3b57..cc36ea7f713 100644 --- a/src/libcore/nonzero.rs +++ b/src/libcore/nonzero.rs @@ -9,105 +9,14 @@ // except according to those terms. //! Exposes the NonZero lang item which provides optimization hints. -#![unstable(feature = "nonzero", - reason = "needs an RFC to flesh out the design", - issue = "27730")] use ops::CoerceUnsized; -/// Unsafe trait to indicate what types are usable with the NonZero struct -pub unsafe trait Zeroable { - /// Whether this value is zero - fn is_zero(&self) -> bool; -} - -macro_rules! impl_zeroable_for_pointer_types { - ( $( $Ptr: ty )+ ) => { - $( - /// For fat pointers to be considered "zero", only the "data" part needs to be null. - unsafe impl<T: ?Sized> Zeroable for $Ptr { - #[inline] - fn is_zero(&self) -> bool { - (*self).is_null() - } - } - )+ - } -} - -macro_rules! impl_zeroable_for_integer_types { - ( $( $Int: ty )+ ) => { - $( - unsafe impl Zeroable for $Int { - #[inline] - fn is_zero(&self) -> bool { - *self == 0 - } - } - )+ - } -} - -impl_zeroable_for_pointer_types! { - *const T - *mut T -} - -impl_zeroable_for_integer_types! { - usize u8 u16 u32 u64 u128 - isize i8 i16 i32 i64 i128 -} - /// A wrapper type for raw pointers and integers that will never be /// NULL or 0 that might allow certain optimizations. #[lang = "non_zero"] -#[derive(Copy, Clone, Eq, PartialEq, Ord, PartialOrd, Debug, Hash)] -pub struct NonZero<T: Zeroable>(T); - -impl<T: Zeroable> NonZero<T> { - /// Creates an instance of NonZero with the provided value. - /// You must indeed ensure that the value is actually "non-zero". - #[unstable(feature = "nonzero", - reason = "needs an RFC to flesh out the design", - issue = "27730")] - #[inline] - pub const unsafe fn new_unchecked(inner: T) -> Self { - NonZero(inner) - } - - /// Creates an instance of NonZero with the provided value. - #[inline] - pub fn new(inner: T) -> Option<Self> { - if inner.is_zero() { - None - } else { - Some(NonZero(inner)) - } - } - - /// Gets the inner value. - pub fn get(self) -> T { - self.0 - } -} - -impl<T: Zeroable+CoerceUnsized<U>, U: Zeroable> CoerceUnsized<NonZero<U>> for NonZero<T> {} - -impl<'a, T: ?Sized> From<&'a mut T> for NonZero<*mut T> { - fn from(reference: &'a mut T) -> Self { - NonZero(reference) - } -} - -impl<'a, T: ?Sized> From<&'a mut T> for NonZero<*const T> { - fn from(reference: &'a mut T) -> Self { - let ptr: *mut T = reference; - NonZero(ptr) - } -} +#[derive(Copy, Clone, Eq, PartialEq, Ord, PartialOrd, Hash)] +#[repr(transparent)] +pub(crate) struct NonZero<T>(pub(crate) T); -impl<'a, T: ?Sized> From<&'a T> for NonZero<*const T> { - fn from(reference: &'a T) -> Self { - NonZero(reference) - } -} +impl<T: CoerceUnsized<U>, U> CoerceUnsized<NonZero<U>> for NonZero<T> {} diff --git a/src/libcore/num/dec2flt/parse.rs b/src/libcore/num/dec2flt/parse.rs index d20986faa0f..e7ed94d4d91 100644 --- a/src/libcore/num/dec2flt/parse.rs +++ b/src/libcore/num/dec2flt/parse.rs @@ -40,7 +40,7 @@ pub struct Decimal<'a> { impl<'a> Decimal<'a> { pub fn new(integral: &'a [u8], fractional: &'a [u8], exp: i64) -> Decimal<'a> { - Decimal { integral: integral, fractional: fractional, exp: exp } + Decimal { integral, fractional, exp } } } @@ -73,7 +73,8 @@ pub fn parse_decimal(s: &str) -> ParseResult { } Some(&b'.') => { let (fractional, s) = eat_digits(&s[1..]); - if integral.is_empty() && fractional.is_empty() && s.is_empty() { + if integral.is_empty() && fractional.is_empty() { + // We require at least a single digit before or after the point. return Invalid; } diff --git a/src/libcore/num/dec2flt/rawfp.rs b/src/libcore/num/dec2flt/rawfp.rs index 12960fed045..38f4e4687a9 100644 --- a/src/libcore/num/dec2flt/rawfp.rs +++ b/src/libcore/num/dec2flt/rawfp.rs @@ -27,14 +27,13 @@ //! Many functions in this module only handle normal numbers. The dec2flt routines conservatively //! take the universally-correct slow path (Algorithm M) for very small and very large numbers. //! That algorithm needs only next_float() which does handle subnormals and zeros. -use u32; use cmp::Ordering::{Less, Equal, Greater}; -use ops::{Mul, Div, Neg}; +use convert::{TryFrom, TryInto}; +use ops::{Add, Mul, Div, Neg}; use fmt::{Debug, LowerExp}; -use mem::transmute; use num::diy_float::Fp; use num::FpCategory::{Infinite, Zero, Subnormal, Normal, Nan}; -use num::Float; +use num::FpCategory; use num::dec2flt::num::{self, Big}; use num::dec2flt::table; @@ -46,7 +45,7 @@ pub struct Unpacked { impl Unpacked { pub fn new(sig: u64, k: i16) -> Self { - Unpacked { sig: sig, k: k } + Unpacked { sig, k } } } @@ -55,22 +54,32 @@ impl Unpacked { /// See the parent module's doc comment for why this is necessary. /// /// Should **never ever** be implemented for other types or be used outside the dec2flt module. -/// Inherits from `Float` because there is some overlap, but all the reused methods are trivial. -pub trait RawFloat : Float + Copy + Debug + LowerExp - + Mul<Output=Self> + Div<Output=Self> + Neg<Output=Self> +pub trait RawFloat + : Copy + + Debug + + LowerExp + + Mul<Output=Self> + + Div<Output=Self> + + Neg<Output=Self> { const INFINITY: Self; const NAN: Self; const ZERO: Self; - /// Returns the mantissa, exponent and sign as integers. - fn integer_decode(self) -> (u64, i16, i8); + /// Type used by `to_bits` and `from_bits`. + type Bits: Add<Output = Self::Bits> + From<u8> + TryFrom<u64>; + + /// Raw transmutation to integer. + fn to_bits(self) -> Self::Bits; - /// Get the raw binary representation of the float. - fn transmute(self) -> u64; + /// Raw transmutation from integer. + fn from_bits(v: Self::Bits) -> Self; - /// Transmute the raw binary representation into a float. - fn from_bits(bits: u64) -> Self; + /// Returns the category that this number falls into. + fn classify(self) -> FpCategory; + + /// Returns the mantissa, exponent and sign as integers. + fn integer_decode(self) -> (u64, i16, i8); /// Decode the float. fn unpack(self) -> Unpacked; @@ -149,6 +158,8 @@ macro_rules! other_constants { } impl RawFloat for f32 { + type Bits = u32; + const SIG_BITS: u8 = 24; const EXP_BITS: u8 = 8; const CEIL_LOG5_OF_MAX_SIG: i16 = 11; @@ -159,7 +170,7 @@ impl RawFloat for f32 { /// Returns the mantissa, exponent and sign as integers. fn integer_decode(self) -> (u64, i16, i8) { - let bits: u32 = unsafe { transmute(self) }; + let bits = self.to_bits(); let sign: i8 = if bits >> 31 == 0 { 1 } else { -1 }; let mut exponent: i16 = ((bits >> 23) & 0xff) as i16; let mantissa = if exponent == 0 { @@ -172,16 +183,6 @@ impl RawFloat for f32 { (mantissa as u64, exponent, sign) } - fn transmute(self) -> u64 { - let bits: u32 = unsafe { transmute(self) }; - bits as u64 - } - - fn from_bits(bits: u64) -> f32 { - assert!(bits < u32::MAX as u64, "f32::from_bits: too many bits"); - unsafe { transmute(bits as u32) } - } - fn unpack(self) -> Unpacked { let (sig, exp, _sig) = self.integer_decode(); Unpacked::new(sig, exp) @@ -196,10 +197,16 @@ impl RawFloat for f32 { fn short_fast_pow10(e: usize) -> Self { table::F32_SHORT_POWERS[e] } + + fn classify(self) -> FpCategory { self.classify() } + fn to_bits(self) -> Self::Bits { self.to_bits() } + fn from_bits(v: Self::Bits) -> Self { Self::from_bits(v) } } impl RawFloat for f64 { + type Bits = u64; + const SIG_BITS: u8 = 53; const EXP_BITS: u8 = 11; const CEIL_LOG5_OF_MAX_SIG: i16 = 23; @@ -210,7 +217,7 @@ impl RawFloat for f64 { /// Returns the mantissa, exponent and sign as integers. fn integer_decode(self) -> (u64, i16, i8) { - let bits: u64 = unsafe { transmute(self) }; + let bits = self.to_bits(); let sign: i8 = if bits >> 63 == 0 { 1 } else { -1 }; let mut exponent: i16 = ((bits >> 52) & 0x7ff) as i16; let mantissa = if exponent == 0 { @@ -223,15 +230,6 @@ impl RawFloat for f64 { (mantissa, exponent, sign) } - fn transmute(self) -> u64 { - let bits: u64 = unsafe { transmute(self) }; - bits - } - - fn from_bits(bits: u64) -> f64 { - unsafe { transmute(bits) } - } - fn unpack(self) -> Unpacked { let (sig, exp, _sig) = self.integer_decode(); Unpacked::new(sig, exp) @@ -246,6 +244,10 @@ impl RawFloat for f64 { fn short_fast_pow10(e: usize) -> Self { table::F64_SHORT_POWERS[e] } + + fn classify(self) -> FpCategory { self.classify() } + fn to_bits(self) -> Self::Bits { self.to_bits() } + fn from_bits(v: Self::Bits) -> Self { Self::from_bits(v) } } /// Convert an Fp to the closest machine float type. @@ -296,14 +298,14 @@ pub fn encode_normal<T: RawFloat>(x: Unpacked) -> T { "encode_normal: exponent out of range"); // Leave sign bit at 0 ("+"), our numbers are all positive let bits = (k_enc as u64) << T::EXPLICIT_SIG_BITS | sig_enc; - T::from_bits(bits) + T::from_bits(bits.try_into().unwrap_or_else(|_| unreachable!())) } /// Construct a subnormal. A mantissa of 0 is allowed and constructs zero. pub fn encode_subnormal<T: RawFloat>(significand: u64) -> T { assert!(significand < T::MIN_SIG, "encode_subnormal: not actually subnormal"); // Encoded exponent is 0, the sign bit is 0, so we just have to reinterpret the bits. - T::from_bits(significand) + T::from_bits(significand.try_into().unwrap_or_else(|_| unreachable!())) } /// Approximate a bignum with an Fp. Rounds within 0.5 ULP with half-to-even. @@ -315,13 +317,13 @@ pub fn big_to_fp(f: &Big) -> Fp { // We cut off all bits prior to the index `start`, i.e., we effectively right-shift by // an amount of `start`, so this is also the exponent we need. let e = start as i16; - let rounded_down = Fp { f: leading, e: e }.normalize(); + let rounded_down = Fp { f: leading, e }.normalize(); // Round (half-to-even) depending on the truncated bits. match num::compare_with_half_ulp(f, start) { Less => rounded_down, Equal if leading % 2 == 0 => rounded_down, Equal | Greater => match leading.checked_add(1) { - Some(f) => Fp { f: f, e: e }.normalize(), + Some(f) => Fp { f, e }.normalize(), None => Fp { f: 1 << 63, e: e + 1 }, } } @@ -363,8 +365,7 @@ pub fn next_float<T: RawFloat>(x: T) -> T { // too is exactly what we want! // Finally, f64::MAX + 1 = 7eff...f + 1 = 7ff0...0 = f64::INFINITY. Zero | Subnormal | Normal => { - let bits: u64 = x.transmute(); - T::from_bits(bits + 1) + T::from_bits(x.to_bits() + T::Bits::from(1u8)) } } } diff --git a/src/libcore/num/diy_float.rs b/src/libcore/num/diy_float.rs index 97bcba2f2ff..b0561da5934 100644 --- a/src/libcore/num/diy_float.rs +++ b/src/libcore/num/diy_float.rs @@ -42,7 +42,7 @@ impl Fp { let tmp = (bd >> 32) + (ad & MASK) + (bc & MASK) + (1 << 31) /* round */; let f = ac + (ad >> 32) + (bc >> 32) + (tmp >> 32); let e = self.e + other.e + 64; - Fp { f: f, e: e } + Fp { f, e } } /// Normalizes itself so that the resulting mantissa is at least `2^63`. @@ -74,7 +74,7 @@ impl Fp { e -= 1; } debug_assert!(f >= (1 >> 63)); - Fp { f: f, e: e } + Fp { f, e } } /// Normalizes itself to have the shared exponent. diff --git a/src/libcore/num/f32.rs b/src/libcore/num/f32.rs index 0dc58d61e49..577c823f9a0 100644 --- a/src/libcore/num/f32.rs +++ b/src/libcore/num/f32.rs @@ -11,16 +11,14 @@ //! This module provides constants which are specific to the implementation //! of the `f32` floating point data type. //! -//! Mathematically significant numbers are provided in the `consts` sub-module. -//! //! *[See also the `f32` primitive type](../../std/primitive.f32.html).* +//! +//! Mathematically significant numbers are provided in the `consts` sub-module. #![stable(feature = "rust1", since = "1.0.0")] -use intrinsics; use mem; -use num::Float; -use num::FpCategory as Fp; +use num::FpCategory; /// The radix or base of the internal representation of `f32`. #[stable(feature = "rust1", since = "1.0.0")] @@ -33,7 +31,11 @@ pub const MANTISSA_DIGITS: u32 = 24; #[stable(feature = "rust1", since = "1.0.0")] pub const DIGITS: u32 = 6; -/// Difference between `1.0` and the next largest representable number. +/// [Machine epsilon] value for `f32`. +/// +/// This is the difference between `1.0` and the next largest representable number. +/// +/// [Machine epsilon]: https://en.wikipedia.org/wiki/Machine_epsilon #[stable(feature = "rust1", since = "1.0.0")] pub const EPSILON: f32 = 1.19209290e-07_f32; @@ -128,10 +130,18 @@ pub mod consts { #[stable(feature = "rust1", since = "1.0.0")] pub const LOG2_E: f32 = 1.44269504088896340735992468100189214_f32; + /// log<sub>2</sub>(10) + #[unstable(feature = "extra_log_consts", issue = "50540")] + pub const LOG2_10: f32 = 3.32192809488736234787031942948939018_f32; + /// log<sub>10</sub>(e) #[stable(feature = "rust1", since = "1.0.0")] pub const LOG10_E: f32 = 0.434294481903251827651128918916605082_f32; + /// log<sub>10</sub>(2) + #[unstable(feature = "extra_log_consts", issue = "50540")] + pub const LOG10_2: f32 = 0.301029995663981195213738894724493027_f32; + /// ln(2) #[stable(feature = "rust1", since = "1.0.0")] pub const LN_2: f32 = 0.693147180559945309417232121458176568_f32; @@ -141,120 +151,227 @@ pub mod consts { pub const LN_10: f32 = 2.30258509299404568401799145468436421_f32; } -#[unstable(feature = "core_float", - reason = "stable interface is via `impl f{32,64}` in later crates", - issue = "32110")] -impl Float for f32 { - /// Returns `true` if the number is NaN. +#[lang = "f32"] +#[cfg(not(test))] +impl f32 { + /// Returns `true` if this value is `NaN` and false otherwise. + /// + /// ``` + /// use std::f32; + /// + /// let nan = f32::NAN; + /// let f = 7.0_f32; + /// + /// assert!(nan.is_nan()); + /// assert!(!f.is_nan()); + /// ``` + #[stable(feature = "rust1", since = "1.0.0")] #[inline] - fn is_nan(self) -> bool { + pub fn is_nan(self) -> bool { self != self } - /// Returns `true` if the number is infinite. + /// Returns `true` if this value is positive infinity or negative infinity and + /// false otherwise. + /// + /// ``` + /// use std::f32; + /// + /// let f = 7.0f32; + /// let inf = f32::INFINITY; + /// let neg_inf = f32::NEG_INFINITY; + /// let nan = f32::NAN; + /// + /// assert!(!f.is_infinite()); + /// assert!(!nan.is_infinite()); + /// + /// assert!(inf.is_infinite()); + /// assert!(neg_inf.is_infinite()); + /// ``` + #[stable(feature = "rust1", since = "1.0.0")] #[inline] - fn is_infinite(self) -> bool { + pub fn is_infinite(self) -> bool { self == INFINITY || self == NEG_INFINITY } - /// Returns `true` if the number is neither infinite or NaN. + /// Returns `true` if this number is neither infinite nor `NaN`. + /// + /// ``` + /// use std::f32; + /// + /// let f = 7.0f32; + /// let inf = f32::INFINITY; + /// let neg_inf = f32::NEG_INFINITY; + /// let nan = f32::NAN; + /// + /// assert!(f.is_finite()); + /// + /// assert!(!nan.is_finite()); + /// assert!(!inf.is_finite()); + /// assert!(!neg_inf.is_finite()); + /// ``` + #[stable(feature = "rust1", since = "1.0.0")] #[inline] - fn is_finite(self) -> bool { + pub fn is_finite(self) -> bool { !(self.is_nan() || self.is_infinite()) } - /// Returns `true` if the number is neither zero, infinite, subnormal or NaN. + /// Returns `true` if the number is neither zero, infinite, + /// [subnormal][subnormal], or `NaN`. + /// + /// ``` + /// use std::f32; + /// + /// let min = f32::MIN_POSITIVE; // 1.17549435e-38f32 + /// let max = f32::MAX; + /// let lower_than_min = 1.0e-40_f32; + /// let zero = 0.0_f32; + /// + /// assert!(min.is_normal()); + /// assert!(max.is_normal()); + /// + /// assert!(!zero.is_normal()); + /// assert!(!f32::NAN.is_normal()); + /// assert!(!f32::INFINITY.is_normal()); + /// // Values between `0` and `min` are Subnormal. + /// assert!(!lower_than_min.is_normal()); + /// ``` + /// [subnormal]: https://en.wikipedia.org/wiki/Denormal_number + #[stable(feature = "rust1", since = "1.0.0")] #[inline] - fn is_normal(self) -> bool { - self.classify() == Fp::Normal + pub fn is_normal(self) -> bool { + self.classify() == FpCategory::Normal } /// Returns the floating point category of the number. If only one property /// is going to be tested, it is generally faster to use the specific /// predicate instead. - fn classify(self) -> Fp { + /// + /// ``` + /// use std::num::FpCategory; + /// use std::f32; + /// + /// let num = 12.4_f32; + /// let inf = f32::INFINITY; + /// + /// assert_eq!(num.classify(), FpCategory::Normal); + /// assert_eq!(inf.classify(), FpCategory::Infinite); + /// ``` + #[stable(feature = "rust1", since = "1.0.0")] + pub fn classify(self) -> FpCategory { const EXP_MASK: u32 = 0x7f800000; const MAN_MASK: u32 = 0x007fffff; - let bits: u32 = unsafe { mem::transmute(self) }; + let bits = self.to_bits(); match (bits & MAN_MASK, bits & EXP_MASK) { - (0, 0) => Fp::Zero, - (_, 0) => Fp::Subnormal, - (0, EXP_MASK) => Fp::Infinite, - (_, EXP_MASK) => Fp::Nan, - _ => Fp::Normal, - } - } - - /// Computes the absolute value of `self`. Returns `Float::nan()` if the - /// number is `Float::nan()`. - #[inline] - fn abs(self) -> f32 { - unsafe { intrinsics::fabsf32(self) } - } - - /// Returns a number that represents the sign of `self`. - /// - /// - `1.0` if the number is positive, `+0.0` or `Float::infinity()` - /// - `-1.0` if the number is negative, `-0.0` or `Float::neg_infinity()` - /// - `Float::nan()` if the number is `Float::nan()` - #[inline] - fn signum(self) -> f32 { - if self.is_nan() { - NAN - } else { - unsafe { intrinsics::copysignf32(1.0, self) } + (0, 0) => FpCategory::Zero, + (_, 0) => FpCategory::Subnormal, + (0, EXP_MASK) => FpCategory::Infinite, + (_, EXP_MASK) => FpCategory::Nan, + _ => FpCategory::Normal, } } /// Returns `true` if and only if `self` has a positive sign, including `+0.0`, `NaN`s with /// positive sign bit and positive infinity. + /// + /// ``` + /// let f = 7.0_f32; + /// let g = -7.0_f32; + /// + /// assert!(f.is_sign_positive()); + /// assert!(!g.is_sign_positive()); + /// ``` + #[stable(feature = "rust1", since = "1.0.0")] #[inline] - fn is_sign_positive(self) -> bool { + pub fn is_sign_positive(self) -> bool { !self.is_sign_negative() } /// Returns `true` if and only if `self` has a negative sign, including `-0.0`, `NaN`s with /// negative sign bit and negative infinity. + /// + /// ``` + /// let f = 7.0f32; + /// let g = -7.0f32; + /// + /// assert!(!f.is_sign_negative()); + /// assert!(g.is_sign_negative()); + /// ``` + #[stable(feature = "rust1", since = "1.0.0")] #[inline] - fn is_sign_negative(self) -> bool { + pub fn is_sign_negative(self) -> bool { // IEEE754 says: isSignMinus(x) is true if and only if x has negative sign. isSignMinus // applies to zeros and NaNs as well. - #[repr(C)] - union F32Bytes { - f: f32, - b: u32 - } - unsafe { F32Bytes { f: self }.b & 0x8000_0000 != 0 } + self.to_bits() & 0x8000_0000 != 0 } - /// Returns the reciprocal (multiplicative inverse) of the number. + /// Takes the reciprocal (inverse) of a number, `1/x`. + /// + /// ``` + /// use std::f32; + /// + /// let x = 2.0_f32; + /// let abs_difference = (x.recip() - (1.0/x)).abs(); + /// + /// assert!(abs_difference <= f32::EPSILON); + /// ``` + #[stable(feature = "rust1", since = "1.0.0")] #[inline] - fn recip(self) -> f32 { + pub fn recip(self) -> f32 { 1.0 / self } + /// Converts radians to degrees. + /// + /// ``` + /// use std::f32::{self, consts}; + /// + /// let angle = consts::PI; + /// + /// let abs_difference = (angle.to_degrees() - 180.0).abs(); + /// + /// assert!(abs_difference <= f32::EPSILON); + /// ``` + #[stable(feature = "f32_deg_rad_conversions", since="1.7.0")] #[inline] - fn powi(self, n: i32) -> f32 { - unsafe { intrinsics::powif32(self, n) } - } - - /// Converts to degrees, assuming the number is in radians. - #[inline] - fn to_degrees(self) -> f32 { - self * (180.0f32 / consts::PI) + pub fn to_degrees(self) -> f32 { + // Use a constant for better precision. + const PIS_IN_180: f32 = 57.2957795130823208767981548141051703_f32; + self * PIS_IN_180 } - /// Converts to radians, assuming the number is in degrees. + /// Converts degrees to radians. + /// + /// ``` + /// use std::f32::{self, consts}; + /// + /// let angle = 180.0f32; + /// + /// let abs_difference = (angle.to_radians() - consts::PI).abs(); + /// + /// assert!(abs_difference <= f32::EPSILON); + /// ``` + #[stable(feature = "f32_deg_rad_conversions", since="1.7.0")] #[inline] - fn to_radians(self) -> f32 { + pub fn to_radians(self) -> f32 { let value: f32 = consts::PI; self * (value / 180.0f32) } /// Returns the maximum of the two numbers. + /// + /// ``` + /// let x = 1.0f32; + /// let y = 2.0f32; + /// + /// assert_eq!(x.max(y), y); + /// ``` + /// + /// If one of the arguments is NaN, then the other argument is returned. + #[stable(feature = "rust1", since = "1.0.0")] #[inline] - fn max(self, other: f32) -> f32 { + pub fn max(self, other: f32) -> f32 { // IEEE754 says: maxNum(x, y) is the canonicalized number y if x < y, x if y < x, the // canonicalized number if one operand is a number and the other a quiet NaN. Otherwise it // is either x or y, canonicalized (this means results might differ among implementations). @@ -267,8 +384,18 @@ impl Float for f32 { } /// Returns the minimum of the two numbers. + /// + /// ``` + /// let x = 1.0f32; + /// let y = 2.0f32; + /// + /// assert_eq!(x.min(y), x); + /// ``` + /// + /// If one of the arguments is NaN, then the other argument is returned. + #[stable(feature = "rust1", since = "1.0.0")] #[inline] - fn min(self, other: f32) -> f32 { + pub fn min(self, other: f32) -> f32 { // IEEE754 says: minNum(x, y) is the canonicalized number x if x < y, y if y < x, the // canonicalized number if one operand is a number and the other a quiet NaN. Otherwise it // is either x or y, canonicalized (this means results might differ among implementations). @@ -279,4 +406,72 @@ impl Float for f32 { // multiplying by 1.0. Should switch to the `canonicalize` when it works. (if other.is_nan() || self < other { self } else { other }) * 1.0 } + + /// Raw transmutation to `u32`. + /// + /// This is currently identical to `transmute::<f32, u32>(self)` on all platforms. + /// + /// See `from_bits` for some discussion of the portability of this operation + /// (there are almost no issues). + /// + /// Note that this function is distinct from `as` casting, which attempts to + /// preserve the *numeric* value, and not the bitwise value. + /// + /// # Examples + /// + /// ``` + /// assert_ne!((1f32).to_bits(), 1f32 as u32); // to_bits() is not casting! + /// assert_eq!((12.5f32).to_bits(), 0x41480000); + /// + /// ``` + #[stable(feature = "float_bits_conv", since = "1.20.0")] + #[inline] + pub fn to_bits(self) -> u32 { + unsafe { mem::transmute(self) } + } + + /// Raw transmutation from `u32`. + /// + /// This is currently identical to `transmute::<u32, f32>(v)` on all platforms. + /// It turns out this is incredibly portable, for two reasons: + /// + /// * Floats and Ints have the same endianness on all supported platforms. + /// * IEEE-754 very precisely specifies the bit layout of floats. + /// + /// However there is one caveat: prior to the 2008 version of IEEE-754, how + /// to interpret the NaN signaling bit wasn't actually specified. Most platforms + /// (notably x86 and ARM) picked the interpretation that was ultimately + /// standardized in 2008, but some didn't (notably MIPS). As a result, all + /// signaling NaNs on MIPS are quiet NaNs on x86, and vice-versa. + /// + /// Rather than trying to preserve signaling-ness cross-platform, this + /// implementation favours preserving the exact bits. This means that + /// any payloads encoded in NaNs will be preserved even if the result of + /// this method is sent over the network from an x86 machine to a MIPS one. + /// + /// If the results of this method are only manipulated by the same + /// architecture that produced them, then there is no portability concern. + /// + /// If the input isn't NaN, then there is no portability concern. + /// + /// If you don't care about signalingness (very likely), then there is no + /// portability concern. + /// + /// Note that this function is distinct from `as` casting, which attempts to + /// preserve the *numeric* value, and not the bitwise value. + /// + /// # Examples + /// + /// ``` + /// use std::f32; + /// let v = f32::from_bits(0x41480000); + /// let difference = (v - 12.5).abs(); + /// assert!(difference <= 1e-5); + /// ``` + #[stable(feature = "float_bits_conv", since = "1.20.0")] + #[inline] + pub fn from_bits(v: u32) -> Self { + // It turns out the safety issues with sNaN were overblown! Hooray! + unsafe { mem::transmute(v) } + } } diff --git a/src/libcore/num/f64.rs b/src/libcore/num/f64.rs index 0e76000efe9..b8e3dd6ed64 100644 --- a/src/libcore/num/f64.rs +++ b/src/libcore/num/f64.rs @@ -11,16 +11,14 @@ //! This module provides constants which are specific to the implementation //! of the `f64` floating point data type. //! -//! Mathematically significant numbers are provided in the `consts` sub-module. -//! //! *[See also the `f64` primitive type](../../std/primitive.f64.html).* +//! +//! Mathematically significant numbers are provided in the `consts` sub-module. #![stable(feature = "rust1", since = "1.0.0")] -use intrinsics; use mem; -use num::FpCategory as Fp; -use num::Float; +use num::FpCategory; /// The radix or base of the internal representation of `f64`. #[stable(feature = "rust1", since = "1.0.0")] @@ -33,7 +31,11 @@ pub const MANTISSA_DIGITS: u32 = 53; #[stable(feature = "rust1", since = "1.0.0")] pub const DIGITS: u32 = 15; -/// Difference between `1.0` and the next largest representable number. +/// [Machine epsilon] value for `f64`. +/// +/// This is the difference between `1.0` and the next largest representable number. +/// +/// [Machine epsilon]: https://en.wikipedia.org/wiki/Machine_epsilon #[stable(feature = "rust1", since = "1.0.0")] pub const EPSILON: f64 = 2.2204460492503131e-16_f64; @@ -124,10 +126,18 @@ pub mod consts { #[stable(feature = "rust1", since = "1.0.0")] pub const E: f64 = 2.71828182845904523536028747135266250_f64; + /// log<sub>2</sub>(10) + #[unstable(feature = "extra_log_consts", issue = "50540")] + pub const LOG2_10: f64 = 3.32192809488736234787031942948939018_f64; + /// log<sub>2</sub>(e) #[stable(feature = "rust1", since = "1.0.0")] pub const LOG2_E: f64 = 1.44269504088896340735992468100189214_f64; + /// log<sub>10</sub>(2) + #[unstable(feature = "extra_log_consts", issue = "50540")] + pub const LOG10_2: f64 = 0.301029995663981195213738894724493027_f64; + /// log<sub>10</sub>(e) #[stable(feature = "rust1", since = "1.0.0")] pub const LOG10_E: f64 = 0.434294481903251827651128918916605082_f64; @@ -141,118 +151,240 @@ pub mod consts { pub const LN_10: f64 = 2.30258509299404568401799145468436421_f64; } -#[unstable(feature = "core_float", - reason = "stable interface is via `impl f{32,64}` in later crates", - issue = "32110")] -impl Float for f64 { - /// Returns `true` if the number is NaN. +#[lang = "f64"] +#[cfg(not(test))] +impl f64 { + /// Returns `true` if this value is `NaN` and false otherwise. + /// + /// ``` + /// use std::f64; + /// + /// let nan = f64::NAN; + /// let f = 7.0_f64; + /// + /// assert!(nan.is_nan()); + /// assert!(!f.is_nan()); + /// ``` + #[stable(feature = "rust1", since = "1.0.0")] #[inline] - fn is_nan(self) -> bool { + pub fn is_nan(self) -> bool { self != self } - /// Returns `true` if the number is infinite. + /// Returns `true` if this value is positive infinity or negative infinity and + /// false otherwise. + /// + /// ``` + /// use std::f64; + /// + /// let f = 7.0f64; + /// let inf = f64::INFINITY; + /// let neg_inf = f64::NEG_INFINITY; + /// let nan = f64::NAN; + /// + /// assert!(!f.is_infinite()); + /// assert!(!nan.is_infinite()); + /// + /// assert!(inf.is_infinite()); + /// assert!(neg_inf.is_infinite()); + /// ``` + #[stable(feature = "rust1", since = "1.0.0")] #[inline] - fn is_infinite(self) -> bool { + pub fn is_infinite(self) -> bool { self == INFINITY || self == NEG_INFINITY } - /// Returns `true` if the number is neither infinite or NaN. + /// Returns `true` if this number is neither infinite nor `NaN`. + /// + /// ``` + /// use std::f64; + /// + /// let f = 7.0f64; + /// let inf: f64 = f64::INFINITY; + /// let neg_inf: f64 = f64::NEG_INFINITY; + /// let nan: f64 = f64::NAN; + /// + /// assert!(f.is_finite()); + /// + /// assert!(!nan.is_finite()); + /// assert!(!inf.is_finite()); + /// assert!(!neg_inf.is_finite()); + /// ``` + #[stable(feature = "rust1", since = "1.0.0")] #[inline] - fn is_finite(self) -> bool { + pub fn is_finite(self) -> bool { !(self.is_nan() || self.is_infinite()) } - /// Returns `true` if the number is neither zero, infinite, subnormal or NaN. + /// Returns `true` if the number is neither zero, infinite, + /// [subnormal][subnormal], or `NaN`. + /// + /// ``` + /// use std::f64; + /// + /// let min = f64::MIN_POSITIVE; // 2.2250738585072014e-308f64 + /// let max = f64::MAX; + /// let lower_than_min = 1.0e-308_f64; + /// let zero = 0.0f64; + /// + /// assert!(min.is_normal()); + /// assert!(max.is_normal()); + /// + /// assert!(!zero.is_normal()); + /// assert!(!f64::NAN.is_normal()); + /// assert!(!f64::INFINITY.is_normal()); + /// // Values between `0` and `min` are Subnormal. + /// assert!(!lower_than_min.is_normal()); + /// ``` + /// [subnormal]: https://en.wikipedia.org/wiki/Denormal_number + #[stable(feature = "rust1", since = "1.0.0")] #[inline] - fn is_normal(self) -> bool { - self.classify() == Fp::Normal + pub fn is_normal(self) -> bool { + self.classify() == FpCategory::Normal } /// Returns the floating point category of the number. If only one property /// is going to be tested, it is generally faster to use the specific /// predicate instead. - fn classify(self) -> Fp { + /// + /// ``` + /// use std::num::FpCategory; + /// use std::f64; + /// + /// let num = 12.4_f64; + /// let inf = f64::INFINITY; + /// + /// assert_eq!(num.classify(), FpCategory::Normal); + /// assert_eq!(inf.classify(), FpCategory::Infinite); + /// ``` + #[stable(feature = "rust1", since = "1.0.0")] + pub fn classify(self) -> FpCategory { const EXP_MASK: u64 = 0x7ff0000000000000; const MAN_MASK: u64 = 0x000fffffffffffff; - let bits: u64 = unsafe { mem::transmute(self) }; + let bits = self.to_bits(); match (bits & MAN_MASK, bits & EXP_MASK) { - (0, 0) => Fp::Zero, - (_, 0) => Fp::Subnormal, - (0, EXP_MASK) => Fp::Infinite, - (_, EXP_MASK) => Fp::Nan, - _ => Fp::Normal, + (0, 0) => FpCategory::Zero, + (_, 0) => FpCategory::Subnormal, + (0, EXP_MASK) => FpCategory::Infinite, + (_, EXP_MASK) => FpCategory::Nan, + _ => FpCategory::Normal, } } - /// Computes the absolute value of `self`. Returns `Float::nan()` if the - /// number is `Float::nan()`. - #[inline] - fn abs(self) -> f64 { - unsafe { intrinsics::fabsf64(self) } - } - - /// Returns a number that represents the sign of `self`. + /// Returns `true` if and only if `self` has a positive sign, including `+0.0`, `NaN`s with + /// positive sign bit and positive infinity. + /// + /// ``` + /// let f = 7.0_f64; + /// let g = -7.0_f64; /// - /// - `1.0` if the number is positive, `+0.0` or `Float::infinity()` - /// - `-1.0` if the number is negative, `-0.0` or `Float::neg_infinity()` - /// - `Float::nan()` if the number is `Float::nan()` + /// assert!(f.is_sign_positive()); + /// assert!(!g.is_sign_positive()); + /// ``` + #[stable(feature = "rust1", since = "1.0.0")] #[inline] - fn signum(self) -> f64 { - if self.is_nan() { - NAN - } else { - unsafe { intrinsics::copysignf64(1.0, self) } - } + pub fn is_sign_positive(self) -> bool { + !self.is_sign_negative() } - /// Returns `true` if and only if `self` has a positive sign, including `+0.0`, `NaN`s with - /// positive sign bit and positive infinity. + #[stable(feature = "rust1", since = "1.0.0")] + #[rustc_deprecated(since = "1.0.0", reason = "renamed to is_sign_positive")] #[inline] - fn is_sign_positive(self) -> bool { - !self.is_sign_negative() + #[doc(hidden)] + pub fn is_positive(self) -> bool { + self.is_sign_positive() } /// Returns `true` if and only if `self` has a negative sign, including `-0.0`, `NaN`s with /// negative sign bit and negative infinity. + /// + /// ``` + /// let f = 7.0_f64; + /// let g = -7.0_f64; + /// + /// assert!(!f.is_sign_negative()); + /// assert!(g.is_sign_negative()); + /// ``` + #[stable(feature = "rust1", since = "1.0.0")] #[inline] - fn is_sign_negative(self) -> bool { - #[repr(C)] - union F64Bytes { - f: f64, - b: u64 - } - unsafe { F64Bytes { f: self }.b & 0x8000_0000_0000_0000 != 0 } + pub fn is_sign_negative(self) -> bool { + self.to_bits() & 0x8000_0000_0000_0000 != 0 } - /// Returns the reciprocal (multiplicative inverse) of the number. + #[stable(feature = "rust1", since = "1.0.0")] + #[rustc_deprecated(since = "1.0.0", reason = "renamed to is_sign_negative")] #[inline] - fn recip(self) -> f64 { - 1.0 / self + #[doc(hidden)] + pub fn is_negative(self) -> bool { + self.is_sign_negative() } + /// Takes the reciprocal (inverse) of a number, `1/x`. + /// + /// ``` + /// let x = 2.0_f64; + /// let abs_difference = (x.recip() - (1.0/x)).abs(); + /// + /// assert!(abs_difference < 1e-10); + /// ``` + #[stable(feature = "rust1", since = "1.0.0")] #[inline] - fn powi(self, n: i32) -> f64 { - unsafe { intrinsics::powif64(self, n) } + pub fn recip(self) -> f64 { + 1.0 / self } - /// Converts to degrees, assuming the number is in radians. + /// Converts radians to degrees. + /// + /// ``` + /// use std::f64::consts; + /// + /// let angle = consts::PI; + /// + /// let abs_difference = (angle.to_degrees() - 180.0).abs(); + /// + /// assert!(abs_difference < 1e-10); + /// ``` + #[stable(feature = "rust1", since = "1.0.0")] #[inline] - fn to_degrees(self) -> f64 { + pub fn to_degrees(self) -> f64 { + // The division here is correctly rounded with respect to the true + // value of 180/π. (This differs from f32, where a constant must be + // used to ensure a correctly rounded result.) self * (180.0f64 / consts::PI) } - /// Converts to radians, assuming the number is in degrees. + /// Converts degrees to radians. + /// + /// ``` + /// use std::f64::consts; + /// + /// let angle = 180.0_f64; + /// + /// let abs_difference = (angle.to_radians() - consts::PI).abs(); + /// + /// assert!(abs_difference < 1e-10); + /// ``` + #[stable(feature = "rust1", since = "1.0.0")] #[inline] - fn to_radians(self) -> f64 { + pub fn to_radians(self) -> f64 { let value: f64 = consts::PI; self * (value / 180.0) } /// Returns the maximum of the two numbers. + /// + /// ``` + /// let x = 1.0_f64; + /// let y = 2.0_f64; + /// + /// assert_eq!(x.max(y), y); + /// ``` + /// + /// If one of the arguments is NaN, then the other argument is returned. + #[stable(feature = "rust1", since = "1.0.0")] #[inline] - fn max(self, other: f64) -> f64 { + pub fn max(self, other: f64) -> f64 { // IEEE754 says: maxNum(x, y) is the canonicalized number y if x < y, x if y < x, the // canonicalized number if one operand is a number and the other a quiet NaN. Otherwise it // is either x or y, canonicalized (this means results might differ among implementations). @@ -265,8 +397,18 @@ impl Float for f64 { } /// Returns the minimum of the two numbers. + /// + /// ``` + /// let x = 1.0_f64; + /// let y = 2.0_f64; + /// + /// assert_eq!(x.min(y), x); + /// ``` + /// + /// If one of the arguments is NaN, then the other argument is returned. + #[stable(feature = "rust1", since = "1.0.0")] #[inline] - fn min(self, other: f64) -> f64 { + pub fn min(self, other: f64) -> f64 { // IEEE754 says: minNum(x, y) is the canonicalized number x if x < y, y if y < x, the // canonicalized number if one operand is a number and the other a quiet NaN. Otherwise it // is either x or y, canonicalized (this means results might differ among implementations). @@ -277,4 +419,72 @@ impl Float for f64 { // multiplying by 1.0. Should switch to the `canonicalize` when it works. (if other.is_nan() || self < other { self } else { other }) * 1.0 } + + /// Raw transmutation to `u64`. + /// + /// This is currently identical to `transmute::<f64, u64>(self)` on all platforms. + /// + /// See `from_bits` for some discussion of the portability of this operation + /// (there are almost no issues). + /// + /// Note that this function is distinct from `as` casting, which attempts to + /// preserve the *numeric* value, and not the bitwise value. + /// + /// # Examples + /// + /// ``` + /// assert!((1f64).to_bits() != 1f64 as u64); // to_bits() is not casting! + /// assert_eq!((12.5f64).to_bits(), 0x4029000000000000); + /// + /// ``` + #[stable(feature = "float_bits_conv", since = "1.20.0")] + #[inline] + pub fn to_bits(self) -> u64 { + unsafe { mem::transmute(self) } + } + + /// Raw transmutation from `u64`. + /// + /// This is currently identical to `transmute::<u64, f64>(v)` on all platforms. + /// It turns out this is incredibly portable, for two reasons: + /// + /// * Floats and Ints have the same endianness on all supported platforms. + /// * IEEE-754 very precisely specifies the bit layout of floats. + /// + /// However there is one caveat: prior to the 2008 version of IEEE-754, how + /// to interpret the NaN signaling bit wasn't actually specified. Most platforms + /// (notably x86 and ARM) picked the interpretation that was ultimately + /// standardized in 2008, but some didn't (notably MIPS). As a result, all + /// signaling NaNs on MIPS are quiet NaNs on x86, and vice-versa. + /// + /// Rather than trying to preserve signaling-ness cross-platform, this + /// implementation favours preserving the exact bits. This means that + /// any payloads encoded in NaNs will be preserved even if the result of + /// this method is sent over the network from an x86 machine to a MIPS one. + /// + /// If the results of this method are only manipulated by the same + /// architecture that produced them, then there is no portability concern. + /// + /// If the input isn't NaN, then there is no portability concern. + /// + /// If you don't care about signalingness (very likely), then there is no + /// portability concern. + /// + /// Note that this function is distinct from `as` casting, which attempts to + /// preserve the *numeric* value, and not the bitwise value. + /// + /// # Examples + /// + /// ``` + /// use std::f64; + /// let v = f64::from_bits(0x4029000000000000); + /// let difference = (v - 12.5).abs(); + /// assert!(difference <= 1e-5); + /// ``` + #[stable(feature = "float_bits_conv", since = "1.20.0")] + #[inline] + pub fn from_bits(v: u64) -> Self { + // It turns out the safety issues with sNaN were overblown! Hooray! + unsafe { mem::transmute(v) } + } } diff --git a/src/libcore/num/flt2dec/decoder.rs b/src/libcore/num/flt2dec/decoder.rs index b779eefce57..c34a56f288f 100644 --- a/src/libcore/num/flt2dec/decoder.rs +++ b/src/libcore/num/flt2dec/decoder.rs @@ -77,8 +77,8 @@ pub fn decode<T: DecodableFloat>(v: T) -> (/*negative?*/ bool, FullDecoded) { // neighbors: (mant - 2, exp) -- (mant, exp) -- (mant + 2, exp) // Float::integer_decode always preserves the exponent, // so the mantissa is scaled for subnormals. - FullDecoded::Finite(Decoded { mant: mant, minus: 1, plus: 1, - exp: exp, inclusive: even }) + FullDecoded::Finite(Decoded { mant, minus: 1, plus: 1, + exp, inclusive: even }) } FpCategory::Normal => { let minnorm = <T as DecodableFloat>::min_pos_norm_value().integer_decode(); diff --git a/src/libcore/num/flt2dec/mod.rs b/src/libcore/num/flt2dec/mod.rs index beaa6e140a6..21a2e72dac8 100644 --- a/src/libcore/num/flt2dec/mod.rs +++ b/src/libcore/num/flt2dec/mod.rs @@ -424,20 +424,20 @@ pub fn to_shortest_str<'a, T, F>(mut format_shortest: F, v: T, match full_decoded { FullDecoded::Nan => { parts[0] = Part::Copy(b"NaN"); - Formatted { sign: sign, parts: &parts[..1] } + Formatted { sign, parts: &parts[..1] } } FullDecoded::Infinite => { parts[0] = Part::Copy(b"inf"); - Formatted { sign: sign, parts: &parts[..1] } + Formatted { sign, parts: &parts[..1] } } FullDecoded::Zero => { if frac_digits > 0 { // [0.][0000] parts[0] = Part::Copy(b"0."); parts[1] = Part::Zero(frac_digits); - Formatted { sign: sign, parts: &parts[..2] } + Formatted { sign, parts: &parts[..2] } } else { parts[0] = Part::Copy(b"0"); - Formatted { sign: sign, parts: &parts[..1] } + Formatted { sign, parts: &parts[..1] } } } FullDecoded::Finite(ref decoded) => { @@ -480,11 +480,11 @@ pub fn to_shortest_exp_str<'a, T, F>(mut format_shortest: F, v: T, match full_decoded { FullDecoded::Nan => { parts[0] = Part::Copy(b"NaN"); - Formatted { sign: sign, parts: &parts[..1] } + Formatted { sign, parts: &parts[..1] } } FullDecoded::Infinite => { parts[0] = Part::Copy(b"inf"); - Formatted { sign: sign, parts: &parts[..1] } + Formatted { sign, parts: &parts[..1] } } FullDecoded::Zero => { parts[0] = if dec_bounds.0 <= 0 && 0 < dec_bounds.1 { @@ -492,7 +492,7 @@ pub fn to_shortest_exp_str<'a, T, F>(mut format_shortest: F, v: T, } else { Part::Copy(if upper { b"0E0" } else { b"0e0" }) }; - Formatted { sign: sign, parts: &parts[..1] } + Formatted { sign, parts: &parts[..1] } } FullDecoded::Finite(ref decoded) => { let (len, exp) = format_shortest(decoded, buf); @@ -502,7 +502,7 @@ pub fn to_shortest_exp_str<'a, T, F>(mut format_shortest: F, v: T, } else { digits_to_exp_str(&buf[..len], exp, 0, upper, parts) }; - Formatted { sign: sign, parts: parts } + Formatted { sign, parts } } } } @@ -558,21 +558,21 @@ pub fn to_exact_exp_str<'a, T, F>(mut format_exact: F, v: T, match full_decoded { FullDecoded::Nan => { parts[0] = Part::Copy(b"NaN"); - Formatted { sign: sign, parts: &parts[..1] } + Formatted { sign, parts: &parts[..1] } } FullDecoded::Infinite => { parts[0] = Part::Copy(b"inf"); - Formatted { sign: sign, parts: &parts[..1] } + Formatted { sign, parts: &parts[..1] } } FullDecoded::Zero => { if ndigits > 1 { // [0.][0000][e0] parts[0] = Part::Copy(b"0."); parts[1] = Part::Zero(ndigits - 1); parts[2] = Part::Copy(if upper { b"E0" } else { b"e0" }); - Formatted { sign: sign, parts: &parts[..3] } + Formatted { sign, parts: &parts[..3] } } else { parts[0] = Part::Copy(if upper { b"0E0" } else { b"0e0" }); - Formatted { sign: sign, parts: &parts[..1] } + Formatted { sign, parts: &parts[..1] } } } FullDecoded::Finite(ref decoded) => { @@ -613,20 +613,20 @@ pub fn to_exact_fixed_str<'a, T, F>(mut format_exact: F, v: T, match full_decoded { FullDecoded::Nan => { parts[0] = Part::Copy(b"NaN"); - Formatted { sign: sign, parts: &parts[..1] } + Formatted { sign, parts: &parts[..1] } } FullDecoded::Infinite => { parts[0] = Part::Copy(b"inf"); - Formatted { sign: sign, parts: &parts[..1] } + Formatted { sign, parts: &parts[..1] } } FullDecoded::Zero => { if frac_digits > 0 { // [0.][0000] parts[0] = Part::Copy(b"0."); parts[1] = Part::Zero(frac_digits); - Formatted { sign: sign, parts: &parts[..2] } + Formatted { sign, parts: &parts[..2] } } else { parts[0] = Part::Copy(b"0"); - Formatted { sign: sign, parts: &parts[..1] } + Formatted { sign, parts: &parts[..1] } } } FullDecoded::Finite(ref decoded) => { @@ -646,10 +646,10 @@ pub fn to_exact_fixed_str<'a, T, F>(mut format_exact: F, v: T, if frac_digits > 0 { // [0.][0000] parts[0] = Part::Copy(b"0."); parts[1] = Part::Zero(frac_digits); - Formatted { sign: sign, parts: &parts[..2] } + Formatted { sign, parts: &parts[..2] } } else { parts[0] = Part::Copy(b"0"); - Formatted { sign: sign, parts: &parts[..1] } + Formatted { sign, parts: &parts[..1] } } } else { Formatted { sign, diff --git a/src/libcore/num/flt2dec/strategy/dragon.rs b/src/libcore/num/flt2dec/strategy/dragon.rs index 6aa4f297e75..aa6a08cb205 100644 --- a/src/libcore/num/flt2dec/strategy/dragon.rs +++ b/src/libcore/num/flt2dec/strategy/dragon.rs @@ -8,12 +8,11 @@ // option. This file may not be copied, modified, or distributed // except according to those terms. -/*! -Almost direct (but slightly optimized) Rust translation of Figure 3 of [1]. - -[1] Burger, R. G. and Dybvig, R. K. 1996. Printing floating-point numbers - quickly and accurately. SIGPLAN Not. 31, 5 (May. 1996), 108-116. -*/ +//! Almost direct (but slightly optimized) Rust translation of Figure 3 of "Printing +//! Floating-Point Numbers Quickly and Accurately"[^1]. +//! +//! [^1]: Burger, R. G. and Dybvig, R. K. 1996. Printing floating-point numbers +//! quickly and accurately. SIGPLAN Not. 31, 5 (May. 1996), 108-116. use cmp::Ordering; diff --git a/src/libcore/num/flt2dec/strategy/grisu.rs b/src/libcore/num/flt2dec/strategy/grisu.rs index cf70a1978f5..effe073c381 100644 --- a/src/libcore/num/flt2dec/strategy/grisu.rs +++ b/src/libcore/num/flt2dec/strategy/grisu.rs @@ -8,13 +8,12 @@ // option. This file may not be copied, modified, or distributed // except according to those terms. -/*! -Rust adaptation of Grisu3 algorithm described in [1]. It uses about -1KB of precomputed table, and in turn, it's very quick for most inputs. - -[1] Florian Loitsch. 2010. Printing floating-point numbers quickly and - accurately with integers. SIGPLAN Not. 45, 6 (June 2010), 233-243. -*/ +//! Rust adaptation of the Grisu3 algorithm described in "Printing Floating-Point Numbers Quickly +//! and Accurately with Integers"[^1]. It uses about 1KB of precomputed table, and in turn, it's +//! very quick for most inputs. +//! +//! [^1]: Florian Loitsch. 2010. Printing floating-point numbers quickly and +//! accurately with integers. SIGPLAN Not. 45, 6 (June 2010), 233-243. use num::diy_float::Fp; use num::flt2dec::{Decoded, MAX_SIG_DIGITS, round_up}; @@ -130,7 +129,7 @@ pub fn cached_power(alpha: i16, gamma: i16) -> (i16, Fp) { let idx = ((gamma as i32) - offset) * range / domain; let (f, e, k) = CACHED_POW10[idx as usize]; debug_assert!(alpha <= e && e <= gamma); - (k, Fp { f: f, e: e }) + (k, Fp { f, e }) } /// Given `x > 0`, returns `(k, 10^k)` such that `10^k <= x < 10^(k+1)`. diff --git a/src/libcore/num/i128.rs b/src/libcore/num/i128.rs index 04354e2e33f..989376d1ac2 100644 --- a/src/libcore/num/i128.rs +++ b/src/libcore/num/i128.rs @@ -12,6 +12,6 @@ //! //! *[See also the `i128` primitive type](../../std/primitive.i128.html).* -#![unstable(feature = "i128", issue="35118")] +#![stable(feature = "i128", since = "1.26.0")] -int_module! { i128, #[unstable(feature = "i128", issue="35118")] } +int_module! { i128, #[stable(feature = "i128", since="1.26.0")] } diff --git a/src/libcore/num/mod.rs b/src/libcore/num/mod.rs index b5d24203b5e..37856dc5469 100644 --- a/src/libcore/num/mod.rs +++ b/src/libcore/num/mod.rs @@ -12,12 +12,93 @@ #![stable(feature = "rust1", since = "1.0.0")] -use convert::{Infallible, TryFrom}; +use convert::TryFrom; use fmt; use intrinsics; +use mem; +use nonzero::NonZero; use ops; use str::FromStr; +macro_rules! impl_nonzero_fmt { + ( ( $( $Trait: ident ),+ ) for $Ty: ident ) => { + $( + #[stable(feature = "nonzero", since = "1.28.0")] + impl fmt::$Trait for $Ty { + #[inline] + fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result { + self.get().fmt(f) + } + } + )+ + } +} + +macro_rules! nonzero_integers { + ( $( $Ty: ident($Int: ty); )+ ) => { + $( + /// An integer that is known not to equal zero. + /// + /// This enables some memory layout optimization. + /// For example, `Option<NonZeroU32>` is the same size as `u32`: + /// + /// ```rust + /// use std::mem::size_of; + /// assert_eq!(size_of::<Option<std::num::NonZeroU32>>(), size_of::<u32>()); + /// ``` + #[stable(feature = "nonzero", since = "1.28.0")] + #[derive(Copy, Clone, Eq, PartialEq, Ord, PartialOrd, Hash)] + #[repr(transparent)] + pub struct $Ty(NonZero<$Int>); + + impl $Ty { + /// Create a non-zero without checking the value. + /// + /// # Safety + /// + /// The value must not be zero. + #[stable(feature = "nonzero", since = "1.28.0")] + #[inline] + pub const unsafe fn new_unchecked(n: $Int) -> Self { + $Ty(NonZero(n)) + } + + /// Create a non-zero if the given value is not zero. + #[stable(feature = "nonzero", since = "1.28.0")] + #[inline] + pub fn new(n: $Int) -> Option<Self> { + if n != 0 { + Some($Ty(NonZero(n))) + } else { + None + } + } + + /// Returns the value as a primitive type. + #[stable(feature = "nonzero", since = "1.28.0")] + #[inline] + pub fn get(self) -> $Int { + self.0 .0 + } + + } + + impl_nonzero_fmt! { + (Debug, Display, Binary, Octal, LowerHex, UpperHex) for $Ty + } + )+ + } +} + +nonzero_integers! { + NonZeroU8(u8); + NonZeroU16(u16); + NonZeroU32(u32); + NonZeroU64(u64); + NonZeroU128(u128); + NonZeroUsize(usize); +} + /// Provides intentionally-wrapped arithmetic on `T`. /// /// Operations like `+` on `u32` values is intended to never overflow, @@ -43,6 +124,7 @@ use str::FromStr; /// ``` #[stable(feature = "rust1", since = "1.0.0")] #[derive(PartialEq, Eq, PartialOrd, Ord, Clone, Copy, Default, Hash)] +#[repr(transparent)] pub struct Wrapping<T>(#[stable(feature = "rust1", since = "1.0.0")] pub T); @@ -88,1715 +170,2669 @@ impl<T: fmt::UpperHex> fmt::UpperHex for Wrapping<T> { } } -mod wrapping; - // All these modules are technically private and only exposed for coretests: pub mod flt2dec; pub mod dec2flt; pub mod bignum; pub mod diy_float; +macro_rules! doc_comment { + ($x:expr, $($tt:tt)*) => { + #[doc = $x] + $($tt)* + }; +} + +mod wrapping; + // `Int` + `SignedInt` implemented for signed integers macro_rules! int_impl { - ($SelfT:ty, $ActualT:ident, $UnsignedT:ty, $BITS:expr) => { - /// Returns the smallest value that can be represented by this integer type. - /// - /// # Examples - /// - /// Basic usage: - /// - /// ``` - /// assert_eq!(i8::min_value(), -128); - /// ``` - #[stable(feature = "rust1", since = "1.0.0")] - #[inline] - pub const fn min_value() -> Self { - !0 ^ ((!0 as $UnsignedT) >> 1) as Self + ($SelfT:ty, $ActualT:ident, $UnsignedT:ty, $BITS:expr, $Min:expr, $Max:expr, $Feature:expr, + $EndFeature:expr, $rot:expr, $rot_op:expr, $rot_result:expr, $swap_op:expr, $swapped:expr, + $reversed:expr) => { + doc_comment! { + concat!("Returns the smallest value that can be represented by this integer type. + +# Examples + +Basic usage: + +``` +", $Feature, "assert_eq!(", stringify!($SelfT), "::min_value(), ", stringify!($Min), ");", +$EndFeature, " +```"), + #[stable(feature = "rust1", since = "1.0.0")] + #[inline] + pub const fn min_value() -> Self { + !0 ^ ((!0 as $UnsignedT) >> 1) as Self + } } - /// Returns the largest value that can be represented by this integer type. - /// - /// # Examples - /// - /// Basic usage: - /// - /// ``` - /// assert_eq!(i8::max_value(), 127); - /// ``` - #[stable(feature = "rust1", since = "1.0.0")] - #[inline] - pub const fn max_value() -> Self { - !Self::min_value() + doc_comment! { + concat!("Returns the largest value that can be represented by this integer type. + +# Examples + +Basic usage: + +``` +", $Feature, "assert_eq!(", stringify!($SelfT), "::max_value(), ", stringify!($Max), ");", +$EndFeature, " +```"), + #[stable(feature = "rust1", since = "1.0.0")] + #[inline] + pub const fn max_value() -> Self { + !Self::min_value() + } } - /// Converts a string slice in a given base to an integer. - /// - /// The string is expected to be an optional `+` or `-` sign - /// followed by digits. - /// Leading and trailing whitespace represent an error. - /// Digits are a subset of these characters, depending on `radix`: - /// - /// * `0-9` - /// * `a-z` - /// * `A-Z` - /// - /// # Panics - /// - /// This function panics if `radix` is not in the range from 2 to 36. - /// - /// # Examples - /// - /// Basic usage: - /// - /// ``` - /// assert_eq!(i32::from_str_radix("A", 16), Ok(10)); - /// ``` - #[stable(feature = "rust1", since = "1.0.0")] - pub fn from_str_radix(src: &str, radix: u32) -> Result<Self, ParseIntError> { - from_str_radix(src, radix) + doc_comment! { + concat!("Converts a string slice in a given base to an integer. + +The string is expected to be an optional `+` or `-` sign followed by digits. +Leading and trailing whitespace represent an error. Digits are a subset of these characters, +depending on `radix`: + + * `0-9` + * `a-z` + * `A-Z` + +# Panics + +This function panics if `radix` is not in the range from 2 to 36. + +# Examples + +Basic usage: + +``` +", $Feature, "assert_eq!(", stringify!($SelfT), "::from_str_radix(\"A\", 16), Ok(10));", +$EndFeature, " +```"), + #[stable(feature = "rust1", since = "1.0.0")] + pub fn from_str_radix(src: &str, radix: u32) -> Result<Self, ParseIntError> { + from_str_radix(src, radix) + } } - /// Returns the number of ones in the binary representation of `self`. - /// - /// # Examples - /// - /// Basic usage: - /// - /// ``` - /// let n = -0b1000_0000i8; - /// - /// assert_eq!(n.count_ones(), 1); - /// ``` - #[stable(feature = "rust1", since = "1.0.0")] - #[inline] - pub fn count_ones(self) -> u32 { (self as $UnsignedT).count_ones() } + doc_comment! { + concat!("Returns the number of ones in the binary representation of `self`. - /// Returns the number of zeros in the binary representation of `self`. - /// - /// # Examples - /// - /// Basic usage: - /// - /// ``` - /// let n = -0b1000_0000i8; - /// - /// assert_eq!(n.count_zeros(), 7); - /// ``` - #[stable(feature = "rust1", since = "1.0.0")] - #[inline] - pub fn count_zeros(self) -> u32 { - (!self).count_ones() +# Examples + +Basic usage: + +``` +", $Feature, "let n = 0b100_0000", stringify!($SelfT), "; + +assert_eq!(n.count_ones(), 1);", +$EndFeature, " +``` +"), + #[stable(feature = "rust1", since = "1.0.0")] + #[rustc_const_unstable(feature = "const_int_ops")] + #[inline] + pub const fn count_ones(self) -> u32 { (self as $UnsignedT).count_ones() } } - /// Returns the number of leading zeros in the binary representation - /// of `self`. - /// - /// # Examples - /// - /// Basic usage: - /// - /// ``` - /// let n = -1i16; - /// - /// assert_eq!(n.leading_zeros(), 0); - /// ``` - #[stable(feature = "rust1", since = "1.0.0")] - #[inline] - pub fn leading_zeros(self) -> u32 { - (self as $UnsignedT).leading_zeros() + doc_comment! { + concat!("Returns the number of zeros in the binary representation of `self`. + +# Examples + +Basic usage: + +``` +", $Feature, "assert_eq!(", stringify!($SelfT), "::max_value().count_zeros(), 1);", $EndFeature, " +```"), + #[stable(feature = "rust1", since = "1.0.0")] + #[rustc_const_unstable(feature = "const_int_ops")] + #[inline] + pub const fn count_zeros(self) -> u32 { + (!self).count_ones() + } } - /// Returns the number of trailing zeros in the binary representation - /// of `self`. - /// - /// # Examples - /// - /// Basic usage: - /// - /// ``` - /// let n = -4i8; - /// - /// assert_eq!(n.trailing_zeros(), 2); - /// ``` - #[stable(feature = "rust1", since = "1.0.0")] - #[inline] - pub fn trailing_zeros(self) -> u32 { - (self as $UnsignedT).trailing_zeros() + doc_comment! { + concat!("Returns the number of leading zeros in the binary representation of `self`. + +# Examples + +Basic usage: + +``` +", $Feature, "let n = -1", stringify!($SelfT), "; + +assert_eq!(n.leading_zeros(), 0);", +$EndFeature, " +```"), + #[stable(feature = "rust1", since = "1.0.0")] + #[rustc_const_unstable(feature = "const_int_ops")] + #[inline] + pub const fn leading_zeros(self) -> u32 { + (self as $UnsignedT).leading_zeros() + } } - /// Shifts the bits to the left by a specified amount, `n`, - /// wrapping the truncated bits to the end of the resulting integer. - /// - /// Please note this isn't the same operation as `<<`! - /// - /// # Examples - /// - /// Basic usage: - /// - /// ``` - /// let n = 0x0123456789ABCDEFi64; - /// let m = -0x76543210FEDCBA99i64; - /// - /// assert_eq!(n.rotate_left(32), m); - /// ``` - #[stable(feature = "rust1", since = "1.0.0")] - #[inline] - pub fn rotate_left(self, n: u32) -> Self { - (self as $UnsignedT).rotate_left(n) as Self + doc_comment! { + concat!("Returns the number of trailing zeros in the binary representation of `self`. + +# Examples + +Basic usage: + +``` +", $Feature, "let n = -4", stringify!($SelfT), "; + +assert_eq!(n.trailing_zeros(), 2);", +$EndFeature, " +```"), + #[stable(feature = "rust1", since = "1.0.0")] + #[rustc_const_unstable(feature = "const_int_ops")] + #[inline] + pub const fn trailing_zeros(self) -> u32 { + (self as $UnsignedT).trailing_zeros() + } } - /// Shifts the bits to the right by a specified amount, `n`, - /// wrapping the truncated bits to the beginning of the resulting - /// integer. - /// - /// Please note this isn't the same operation as `>>`! - /// - /// # Examples - /// - /// Basic usage: - /// - /// ``` - /// let n = 0x0123456789ABCDEFi64; - /// let m = -0xFEDCBA987654322i64; - /// - /// assert_eq!(n.rotate_right(4), m); - /// ``` - #[stable(feature = "rust1", since = "1.0.0")] - #[inline] - pub fn rotate_right(self, n: u32) -> Self { - (self as $UnsignedT).rotate_right(n) as Self + doc_comment! { + concat!("Shifts the bits to the left by a specified amount, `n`, +wrapping the truncated bits to the end of the resulting integer. + +Please note this isn't the same operation as `<<`! + +# Examples + +Basic usage: + +``` +let n = ", $rot_op, stringify!($SelfT), "; +let m = ", $rot_result, "; + +assert_eq!(n.rotate_left(", $rot, "), m); +```"), + #[stable(feature = "rust1", since = "1.0.0")] + #[inline] + pub fn rotate_left(self, n: u32) -> Self { + (self as $UnsignedT).rotate_left(n) as Self + } } - /// Reverses the byte order of the integer. - /// - /// # Examples - /// - /// Basic usage: - /// - /// ``` - /// let n: i16 = 0b0000000_01010101; - /// assert_eq!(n, 85); - /// - /// let m = n.swap_bytes(); - /// - /// assert_eq!(m, 0b01010101_00000000); - /// assert_eq!(m, 21760); - /// ``` - #[stable(feature = "rust1", since = "1.0.0")] - #[inline] - pub fn swap_bytes(self) -> Self { - (self as $UnsignedT).swap_bytes() as Self + doc_comment! { + concat!("Shifts the bits to the right by a specified amount, `n`, +wrapping the truncated bits to the beginning of the resulting +integer. + +Please note this isn't the same operation as `>>`! + +# Examples + +Basic usage: + +``` +let n = ", $rot_result, stringify!($SelfT), "; +let m = ", $rot_op, "; + +assert_eq!(n.rotate_right(", $rot, "), m); +```"), + #[stable(feature = "rust1", since = "1.0.0")] + #[inline] + pub fn rotate_right(self, n: u32) -> Self { + (self as $UnsignedT).rotate_right(n) as Self + } } + doc_comment! { + concat!("Reverses the byte order of the integer. - /// Converts an integer from big endian to the target's endianness. - /// - /// On big endian this is a no-op. On little endian the bytes are - /// swapped. - /// - /// # Examples - /// - /// Basic usage: - /// - /// ``` - /// let n = 0x0123456789ABCDEFi64; - /// - /// if cfg!(target_endian = "big") { - /// assert_eq!(i64::from_be(n), n) - /// } else { - /// assert_eq!(i64::from_be(n), n.swap_bytes()) - /// } - /// ``` - #[stable(feature = "rust1", since = "1.0.0")] - #[inline] - pub fn from_be(x: Self) -> Self { - if cfg!(target_endian = "big") { x } else { x.swap_bytes() } +# Examples + +Basic usage: + +``` +let n = ", $swap_op, stringify!($SelfT), "; + +let m = n.swap_bytes(); + +assert_eq!(m, ", $swapped, "); +```"), + #[stable(feature = "rust1", since = "1.0.0")] + #[rustc_const_unstable(feature = "const_int_ops")] + #[inline] + pub const fn swap_bytes(self) -> Self { + (self as $UnsignedT).swap_bytes() as Self + } } - /// Converts an integer from little endian to the target's endianness. - /// - /// On little endian this is a no-op. On big endian the bytes are - /// swapped. - /// - /// # Examples - /// - /// Basic usage: - /// - /// ``` - /// let n = 0x0123456789ABCDEFi64; - /// - /// if cfg!(target_endian = "little") { - /// assert_eq!(i64::from_le(n), n) - /// } else { - /// assert_eq!(i64::from_le(n), n.swap_bytes()) - /// } - /// ``` - #[stable(feature = "rust1", since = "1.0.0")] - #[inline] - pub fn from_le(x: Self) -> Self { - if cfg!(target_endian = "little") { x } else { x.swap_bytes() } + doc_comment! { + concat!("Reverses the bit pattern of the integer. + +# Examples + +Basic usage: + +``` +#![feature(reverse_bits)] + +let n = ", $swap_op, stringify!($SelfT), "; +let m = n.reverse_bits(); + +assert_eq!(m, ", $reversed, "); +```"), + #[unstable(feature = "reverse_bits", issue = "48763")] + #[inline] + pub fn reverse_bits(self) -> Self { + (self as $UnsignedT).reverse_bits() as Self + } } - /// Converts `self` to big endian from the target's endianness. - /// - /// On big endian this is a no-op. On little endian the bytes are - /// swapped. - /// - /// # Examples - /// - /// Basic usage: - /// - /// ``` - /// let n = 0x0123456789ABCDEFi64; - /// - /// if cfg!(target_endian = "big") { - /// assert_eq!(n.to_be(), n) - /// } else { - /// assert_eq!(n.to_be(), n.swap_bytes()) - /// } - /// ``` - #[stable(feature = "rust1", since = "1.0.0")] - #[inline] - pub fn to_be(self) -> Self { // or not to be? - if cfg!(target_endian = "big") { self } else { self.swap_bytes() } + doc_comment! { + concat!("Converts an integer from big endian to the target's endianness. + +On big endian this is a no-op. On little endian the bytes are swapped. + +# Examples + +Basic usage: + +``` +", $Feature, "let n = 0x1A", stringify!($SelfT), "; + +if cfg!(target_endian = \"big\") { + assert_eq!(", stringify!($SelfT), "::from_be(n), n) +} else { + assert_eq!(", stringify!($SelfT), "::from_be(n), n.swap_bytes()) +}", +$EndFeature, " +```"), + #[stable(feature = "rust1", since = "1.0.0")] + #[rustc_const_unstable(feature = "const_int_ops")] + #[inline] + pub const fn from_be(x: Self) -> Self { + #[cfg(target_endian = "big")] + { + x + } + #[cfg(not(target_endian = "big"))] + { + x.swap_bytes() + } + } } - /// Converts `self` to little endian from the target's endianness. - /// - /// On little endian this is a no-op. On big endian the bytes are - /// swapped. - /// - /// # Examples - /// - /// Basic usage: - /// - /// ``` - /// let n = 0x0123456789ABCDEFi64; - /// - /// if cfg!(target_endian = "little") { - /// assert_eq!(n.to_le(), n) - /// } else { - /// assert_eq!(n.to_le(), n.swap_bytes()) - /// } - /// ``` - #[stable(feature = "rust1", since = "1.0.0")] - #[inline] - pub fn to_le(self) -> Self { - if cfg!(target_endian = "little") { self } else { self.swap_bytes() } + doc_comment! { + concat!("Converts an integer from little endian to the target's endianness. + +On little endian this is a no-op. On big endian the bytes are swapped. + +# Examples + +Basic usage: + +``` +", $Feature, "let n = 0x1A", stringify!($SelfT), "; + +if cfg!(target_endian = \"little\") { + assert_eq!(", stringify!($SelfT), "::from_le(n), n) +} else { + assert_eq!(", stringify!($SelfT), "::from_le(n), n.swap_bytes()) +}", +$EndFeature, " +```"), + #[stable(feature = "rust1", since = "1.0.0")] + #[rustc_const_unstable(feature = "const_int_ops")] + #[inline] + pub const fn from_le(x: Self) -> Self { + #[cfg(target_endian = "little")] + { + x + } + #[cfg(not(target_endian = "little"))] + { + x.swap_bytes() + } + } } - /// Checked integer addition. Computes `self + rhs`, returning `None` - /// if overflow occurred. - /// - /// # Examples - /// - /// Basic usage: - /// - /// ``` - /// assert_eq!(7i16.checked_add(32760), Some(32767)); - /// assert_eq!(8i16.checked_add(32760), None); - /// ``` - #[stable(feature = "rust1", since = "1.0.0")] - #[inline] - pub fn checked_add(self, rhs: Self) -> Option<Self> { - let (a, b) = self.overflowing_add(rhs); - if b {None} else {Some(a)} + doc_comment! { + concat!("Converts `self` to big endian from the target's endianness. + +On big endian this is a no-op. On little endian the bytes are swapped. + +# Examples + +Basic usage: + +``` +", $Feature, "let n = 0x1A", stringify!($SelfT), "; + +if cfg!(target_endian = \"big\") { + assert_eq!(n.to_be(), n) +} else { + assert_eq!(n.to_be(), n.swap_bytes()) +}", +$EndFeature, " +```"), + #[stable(feature = "rust1", since = "1.0.0")] + #[rustc_const_unstable(feature = "const_int_ops")] + #[inline] + pub const fn to_be(self) -> Self { // or not to be? + #[cfg(target_endian = "big")] + { + self + } + #[cfg(not(target_endian = "big"))] + { + self.swap_bytes() + } + } } - /// Checked integer subtraction. Computes `self - rhs`, returning - /// `None` if overflow occurred. - /// - /// # Examples - /// - /// Basic usage: - /// - /// ``` - /// assert_eq!((-127i8).checked_sub(1), Some(-128)); - /// assert_eq!((-128i8).checked_sub(1), None); - /// ``` - #[stable(feature = "rust1", since = "1.0.0")] - #[inline] - pub fn checked_sub(self, rhs: Self) -> Option<Self> { - let (a, b) = self.overflowing_sub(rhs); - if b {None} else {Some(a)} + doc_comment! { + concat!("Converts `self` to little endian from the target's endianness. + +On little endian this is a no-op. On big endian the bytes are swapped. + +# Examples + +Basic usage: + +``` +", $Feature, "let n = 0x1A", stringify!($SelfT), "; + +if cfg!(target_endian = \"little\") { + assert_eq!(n.to_le(), n) +} else { + assert_eq!(n.to_le(), n.swap_bytes()) +}", +$EndFeature, " +```"), + #[stable(feature = "rust1", since = "1.0.0")] + #[rustc_const_unstable(feature = "const_int_ops")] + #[inline] + pub const fn to_le(self) -> Self { + #[cfg(target_endian = "little")] + { + self + } + #[cfg(not(target_endian = "little"))] + { + self.swap_bytes() + } + } } - /// Checked integer multiplication. Computes `self * rhs`, returning - /// `None` if overflow occurred. - /// - /// # Examples - /// - /// Basic usage: - /// - /// ``` - /// assert_eq!(6i8.checked_mul(21), Some(126)); - /// assert_eq!(6i8.checked_mul(22), None); - /// ``` - #[stable(feature = "rust1", since = "1.0.0")] - #[inline] - pub fn checked_mul(self, rhs: Self) -> Option<Self> { - let (a, b) = self.overflowing_mul(rhs); - if b {None} else {Some(a)} + doc_comment! { + concat!("Checked integer addition. Computes `self + rhs`, returning `None` +if overflow occurred. + +# Examples + +Basic usage: + +``` +", $Feature, "assert_eq!((", stringify!($SelfT), +"::max_value() - 2).checked_add(1), Some(", stringify!($SelfT), "::max_value() - 1)); +assert_eq!((", stringify!($SelfT), "::max_value() - 2).checked_add(3), None);", +$EndFeature, " +```"), + #[stable(feature = "rust1", since = "1.0.0")] + #[inline] + pub fn checked_add(self, rhs: Self) -> Option<Self> { + let (a, b) = self.overflowing_add(rhs); + if b {None} else {Some(a)} + } } - /// Checked integer division. Computes `self / rhs`, returning `None` - /// if `rhs == 0` or the division results in overflow. - /// - /// # Examples - /// - /// Basic usage: - /// - /// ``` - /// assert_eq!((-127i8).checked_div(-1), Some(127)); - /// assert_eq!((-128i8).checked_div(-1), None); - /// assert_eq!((1i8).checked_div(0), None); - /// ``` - #[stable(feature = "rust1", since = "1.0.0")] - #[inline] - pub fn checked_div(self, rhs: Self) -> Option<Self> { - if rhs == 0 || (self == Self::min_value() && rhs == -1) { - None - } else { - Some(unsafe { intrinsics::unchecked_div(self, rhs) }) + doc_comment! { + concat!("Checked integer subtraction. Computes `self - rhs`, returning `None` if +overflow occurred. + +# Examples + +Basic usage: + +``` +", $Feature, "assert_eq!((", stringify!($SelfT), +"::min_value() + 2).checked_sub(1), Some(", stringify!($SelfT), "::min_value() + 1)); +assert_eq!((", stringify!($SelfT), "::min_value() + 2).checked_sub(3), None);", +$EndFeature, " +```"), + #[stable(feature = "rust1", since = "1.0.0")] + #[inline] + pub fn checked_sub(self, rhs: Self) -> Option<Self> { + let (a, b) = self.overflowing_sub(rhs); + if b {None} else {Some(a)} } } - /// Checked integer remainder. Computes `self % rhs`, returning `None` - /// if `rhs == 0` or the division results in overflow. - /// - /// # Examples - /// - /// Basic usage: - /// - /// ``` - /// use std::i32; - /// - /// assert_eq!(5i32.checked_rem(2), Some(1)); - /// assert_eq!(5i32.checked_rem(0), None); - /// assert_eq!(i32::MIN.checked_rem(-1), None); - /// ``` - #[stable(feature = "wrapping", since = "1.7.0")] - #[inline] - pub fn checked_rem(self, rhs: Self) -> Option<Self> { - if rhs == 0 || (self == Self::min_value() && rhs == -1) { - None - } else { - Some(unsafe { intrinsics::unchecked_rem(self, rhs) }) + doc_comment! { + concat!("Checked integer multiplication. Computes `self * rhs`, returning `None` if +overflow occurred. + +# Examples + +Basic usage: + +``` +", $Feature, "assert_eq!(", stringify!($SelfT), +"::max_value().checked_mul(1), Some(", stringify!($SelfT), "::max_value())); +assert_eq!(", stringify!($SelfT), "::max_value().checked_mul(2), None);", +$EndFeature, " +```"), + #[stable(feature = "rust1", since = "1.0.0")] + #[inline] + pub fn checked_mul(self, rhs: Self) -> Option<Self> { + let (a, b) = self.overflowing_mul(rhs); + if b {None} else {Some(a)} } } - /// Checked negation. Computes `-self`, returning `None` if `self == - /// MIN`. - /// - /// # Examples - /// - /// Basic usage: - /// - /// ``` - /// use std::i32; - /// - /// assert_eq!(5i32.checked_neg(), Some(-5)); - /// assert_eq!(i32::MIN.checked_neg(), None); - /// ``` - #[stable(feature = "wrapping", since = "1.7.0")] - #[inline] - pub fn checked_neg(self) -> Option<Self> { - let (a, b) = self.overflowing_neg(); - if b {None} else {Some(a)} + doc_comment! { + concat!("Checked integer division. Computes `self / rhs`, returning `None` if `rhs == 0` +or the division results in overflow. + +# Examples + +Basic usage: + +``` +", $Feature, "assert_eq!((", stringify!($SelfT), +"::min_value() + 1).checked_div(-1), Some(", stringify!($Max), ")); +assert_eq!(", stringify!($SelfT), "::min_value().checked_div(-1), None); +assert_eq!((1", stringify!($SelfT), ").checked_div(0), None);", +$EndFeature, " +```"), + #[stable(feature = "rust1", since = "1.0.0")] + #[inline] + pub fn checked_div(self, rhs: Self) -> Option<Self> { + if rhs == 0 || (self == Self::min_value() && rhs == -1) { + None + } else { + Some(unsafe { intrinsics::unchecked_div(self, rhs) }) + } + } } - /// Checked shift left. Computes `self << rhs`, returning `None` - /// if `rhs` is larger than or equal to the number of bits in `self`. - /// - /// # Examples - /// - /// Basic usage: - /// - /// ``` - /// assert_eq!(0x10i32.checked_shl(4), Some(0x100)); - /// assert_eq!(0x10i32.checked_shl(33), None); - /// ``` - #[stable(feature = "wrapping", since = "1.7.0")] - #[inline] - pub fn checked_shl(self, rhs: u32) -> Option<Self> { - let (a, b) = self.overflowing_shl(rhs); - if b {None} else {Some(a)} + doc_comment! { + concat!("Checked Euclidean division. Computes `self.div_euc(rhs)`, +returning `None` if `rhs == 0` or the division results in overflow. + +# Examples + +Basic usage: + +``` +#![feature(euclidean_division)] +assert_eq!((", stringify!($SelfT), +"::min_value() + 1).checked_div_euc(-1), Some(", stringify!($Max), ")); +assert_eq!(", stringify!($SelfT), "::min_value().checked_div_euc(-1), None); +assert_eq!((1", stringify!($SelfT), ").checked_div_euc(0), None); +```"), + #[unstable(feature = "euclidean_division", issue = "49048")] + #[inline] + pub fn checked_div_euc(self, rhs: Self) -> Option<Self> { + if rhs == 0 || (self == Self::min_value() && rhs == -1) { + None + } else { + Some(self.div_euc(rhs)) + } + } } - /// Checked shift right. Computes `self >> rhs`, returning `None` - /// if `rhs` is larger than or equal to the number of bits in `self`. - /// - /// # Examples - /// - /// Basic usage: - /// - /// ``` - /// assert_eq!(0x10i32.checked_shr(4), Some(0x1)); - /// assert_eq!(0x10i32.checked_shr(33), None); - /// ``` - #[stable(feature = "wrapping", since = "1.7.0")] - #[inline] - pub fn checked_shr(self, rhs: u32) -> Option<Self> { - let (a, b) = self.overflowing_shr(rhs); - if b {None} else {Some(a)} + doc_comment! { + concat!("Checked integer remainder. Computes `self % rhs`, returning `None` if +`rhs == 0` or the division results in overflow. + +# Examples + +Basic usage: + +``` +", $Feature, "use std::", stringify!($SelfT), "; + +assert_eq!(5", stringify!($SelfT), ".checked_rem(2), Some(1)); +assert_eq!(5", stringify!($SelfT), ".checked_rem(0), None); +assert_eq!(", stringify!($SelfT), "::MIN.checked_rem(-1), None);", +$EndFeature, " +```"), + #[stable(feature = "wrapping", since = "1.7.0")] + #[inline] + pub fn checked_rem(self, rhs: Self) -> Option<Self> { + if rhs == 0 || (self == Self::min_value() && rhs == -1) { + None + } else { + Some(unsafe { intrinsics::unchecked_rem(self, rhs) }) + } + } } - /// Checked absolute value. Computes `self.abs()`, returning `None` if - /// `self == MIN`. - /// - /// # Examples - /// - /// Basic usage: - /// - /// ``` - /// use std::i32; - /// - /// assert_eq!((-5i32).checked_abs(), Some(5)); - /// assert_eq!(i32::MIN.checked_abs(), None); - /// ``` - #[stable(feature = "no_panic_abs", since = "1.13.0")] - #[inline] - pub fn checked_abs(self) -> Option<Self> { - if self.is_negative() { - self.checked_neg() - } else { - Some(self) + doc_comment! { + concat!("Checked Euclidean modulo. Computes `self.mod_euc(rhs)`, returning `None` if +`rhs == 0` or the division results in overflow. + +# Examples + +Basic usage: + +``` +#![feature(euclidean_division)] +use std::", stringify!($SelfT), "; + +assert_eq!(5", stringify!($SelfT), ".checked_mod_euc(2), Some(1)); +assert_eq!(5", stringify!($SelfT), ".checked_mod_euc(0), None); +assert_eq!(", stringify!($SelfT), "::MIN.checked_mod_euc(-1), None); +```"), + #[unstable(feature = "euclidean_division", issue = "49048")] + #[inline] + pub fn checked_mod_euc(self, rhs: Self) -> Option<Self> { + if rhs == 0 || (self == Self::min_value() && rhs == -1) { + None + } else { + Some(self.mod_euc(rhs)) + } } } - /// Saturating integer addition. Computes `self + rhs`, saturating at - /// the numeric bounds instead of overflowing. - /// - /// # Examples - /// - /// Basic usage: - /// - /// ``` - /// assert_eq!(100i8.saturating_add(1), 101); - /// assert_eq!(100i8.saturating_add(127), 127); - /// ``` - #[stable(feature = "rust1", since = "1.0.0")] - #[inline] - pub fn saturating_add(self, rhs: Self) -> Self { - match self.checked_add(rhs) { - Some(x) => x, - None if rhs >= 0 => Self::max_value(), - None => Self::min_value(), + doc_comment! { + concat!("Checked negation. Computes `-self`, returning `None` if `self == MIN`. + +# Examples + +Basic usage: + +``` +", $Feature, "use std::", stringify!($SelfT), "; + +assert_eq!(5", stringify!($SelfT), ".checked_neg(), Some(-5)); +assert_eq!(", stringify!($SelfT), "::MIN.checked_neg(), None);", +$EndFeature, " +```"), + #[stable(feature = "wrapping", since = "1.7.0")] + #[inline] + pub fn checked_neg(self) -> Option<Self> { + let (a, b) = self.overflowing_neg(); + if b {None} else {Some(a)} } } - /// Saturating integer subtraction. Computes `self - rhs`, saturating - /// at the numeric bounds instead of overflowing. - /// - /// # Examples - /// - /// Basic usage: - /// - /// ``` - /// assert_eq!(100i8.saturating_sub(127), -27); - /// assert_eq!((-100i8).saturating_sub(127), -128); - /// ``` - #[stable(feature = "rust1", since = "1.0.0")] - #[inline] - pub fn saturating_sub(self, rhs: Self) -> Self { - match self.checked_sub(rhs) { - Some(x) => x, - None if rhs >= 0 => Self::min_value(), - None => Self::max_value(), + doc_comment! { + concat!("Checked shift left. Computes `self << rhs`, returning `None` if `rhs` is larger +than or equal to the number of bits in `self`. + +# Examples + +Basic usage: + +``` +", $Feature, "assert_eq!(0x1", stringify!($SelfT), ".checked_shl(4), Some(0x10)); +assert_eq!(0x1", stringify!($SelfT), ".checked_shl(129), None);", +$EndFeature, " +```"), + #[stable(feature = "wrapping", since = "1.7.0")] + #[inline] + pub fn checked_shl(self, rhs: u32) -> Option<Self> { + let (a, b) = self.overflowing_shl(rhs); + if b {None} else {Some(a)} } } - /// Saturating integer multiplication. Computes `self * rhs`, - /// saturating at the numeric bounds instead of overflowing. - /// - /// # Examples - /// - /// Basic usage: - /// - /// ``` - /// use std::i32; - /// - /// assert_eq!(100i32.saturating_mul(127), 12700); - /// assert_eq!((1i32 << 23).saturating_mul(1 << 23), i32::MAX); - /// assert_eq!((-1i32 << 23).saturating_mul(1 << 23), i32::MIN); - /// ``` - #[stable(feature = "wrapping", since = "1.7.0")] - #[inline] - pub fn saturating_mul(self, rhs: Self) -> Self { - self.checked_mul(rhs).unwrap_or_else(|| { - if (self < 0 && rhs < 0) || (self > 0 && rhs > 0) { - Self::max_value() + doc_comment! { + concat!("Checked shift right. Computes `self >> rhs`, returning `None` if `rhs` is +larger than or equal to the number of bits in `self`. + +# Examples + +Basic usage: + +``` +", $Feature, "assert_eq!(0x10", stringify!($SelfT), ".checked_shr(4), Some(0x1)); +assert_eq!(0x10", stringify!($SelfT), ".checked_shr(128), None);", +$EndFeature, " +```"), + #[stable(feature = "wrapping", since = "1.7.0")] + #[inline] + pub fn checked_shr(self, rhs: u32) -> Option<Self> { + let (a, b) = self.overflowing_shr(rhs); + if b {None} else {Some(a)} + } + } + + doc_comment! { + concat!("Checked absolute value. Computes `self.abs()`, returning `None` if +`self == MIN`. + +# Examples + +Basic usage: + +``` +", $Feature, "use std::", stringify!($SelfT), "; + +assert_eq!((-5", stringify!($SelfT), ").checked_abs(), Some(5)); +assert_eq!(", stringify!($SelfT), "::MIN.checked_abs(), None);", +$EndFeature, " +```"), + #[stable(feature = "no_panic_abs", since = "1.13.0")] + #[inline] + pub fn checked_abs(self) -> Option<Self> { + if self.is_negative() { + self.checked_neg() } else { - Self::min_value() + Some(self) } - }) + } } - /// Wrapping (modular) addition. Computes `self + rhs`, - /// wrapping around at the boundary of the type. - /// - /// # Examples - /// - /// Basic usage: - /// - /// ``` - /// assert_eq!(100i8.wrapping_add(27), 127); - /// assert_eq!(100i8.wrapping_add(127), -29); - /// ``` - #[stable(feature = "rust1", since = "1.0.0")] - #[inline] - pub fn wrapping_add(self, rhs: Self) -> Self { - unsafe { - intrinsics::overflowing_add(self, rhs) + doc_comment! { + concat!("Checked exponentiation. Computes `self.pow(exp)`, returning `None` if +overflow occurred. + +# Examples + +Basic usage: + +``` +#![feature(no_panic_pow)] +", $Feature, "assert_eq!(8", stringify!($SelfT), ".checked_pow(2), Some(64)); +assert_eq!(", stringify!($SelfT), "::max_value().checked_pow(2), None);", +$EndFeature, " +```"), + + #[unstable(feature = "no_panic_pow", issue = "48320")] + #[inline] + pub fn checked_pow(self, mut exp: u32) -> Option<Self> { + let mut base = self; + let mut acc: Self = 1; + + while exp > 1 { + if (exp & 1) == 1 { + acc = acc.checked_mul(base)?; + } + exp /= 2; + base = base.checked_mul(base)?; + } + + // Deal with the final bit of the exponent separately, since + // squaring the base afterwards is not necessary and may cause a + // needless overflow. + if exp == 1 { + acc = acc.checked_mul(base)?; + } + + Some(acc) } } - /// Wrapping (modular) subtraction. Computes `self - rhs`, - /// wrapping around at the boundary of the type. - /// - /// # Examples - /// - /// Basic usage: - /// - /// ``` - /// assert_eq!(0i8.wrapping_sub(127), -127); - /// assert_eq!((-2i8).wrapping_sub(127), 127); - /// ``` - #[stable(feature = "rust1", since = "1.0.0")] - #[inline] - pub fn wrapping_sub(self, rhs: Self) -> Self { - unsafe { - intrinsics::overflowing_sub(self, rhs) + doc_comment! { + concat!("Saturating integer addition. Computes `self + rhs`, saturating at the numeric +bounds instead of overflowing. + +# Examples + +Basic usage: + +``` +", $Feature, "assert_eq!(100", stringify!($SelfT), ".saturating_add(1), 101); +assert_eq!(", stringify!($SelfT), "::max_value().saturating_add(100), ", stringify!($SelfT), +"::max_value());", +$EndFeature, " +```"), + #[stable(feature = "rust1", since = "1.0.0")] + #[inline] + pub fn saturating_add(self, rhs: Self) -> Self { + match self.checked_add(rhs) { + Some(x) => x, + None if rhs >= 0 => Self::max_value(), + None => Self::min_value(), + } } } - /// Wrapping (modular) multiplication. Computes `self * - /// rhs`, wrapping around at the boundary of the type. - /// - /// # Examples - /// - /// Basic usage: - /// - /// ``` - /// assert_eq!(10i8.wrapping_mul(12), 120); - /// assert_eq!(11i8.wrapping_mul(12), -124); - /// ``` - #[stable(feature = "rust1", since = "1.0.0")] - #[inline] - pub fn wrapping_mul(self, rhs: Self) -> Self { - unsafe { - intrinsics::overflowing_mul(self, rhs) + doc_comment! { + concat!("Saturating integer subtraction. Computes `self - rhs`, saturating at the +numeric bounds instead of overflowing. + +# Examples + +Basic usage: + +``` +", $Feature, "assert_eq!(100", stringify!($SelfT), ".saturating_sub(127), -27); +assert_eq!(", stringify!($SelfT), "::min_value().saturating_sub(100), ", stringify!($SelfT), +"::min_value());", +$EndFeature, " +```"), + #[stable(feature = "rust1", since = "1.0.0")] + #[inline] + pub fn saturating_sub(self, rhs: Self) -> Self { + match self.checked_sub(rhs) { + Some(x) => x, + None if rhs >= 0 => Self::min_value(), + None => Self::max_value(), + } } } - /// Wrapping (modular) division. Computes `self / rhs`, - /// wrapping around at the boundary of the type. - /// - /// The only case where such wrapping can occur is when one - /// divides `MIN / -1` on a signed type (where `MIN` is the - /// negative minimal value for the type); this is equivalent - /// to `-MIN`, a positive value that is too large to represent - /// in the type. In such a case, this function returns `MIN` - /// itself. - /// - /// # Panics - /// - /// This function will panic if `rhs` is 0. - /// - /// # Examples - /// - /// Basic usage: - /// - /// ``` - /// assert_eq!(100u8.wrapping_div(10), 10); - /// assert_eq!((-128i8).wrapping_div(-1), -128); - /// ``` - #[stable(feature = "num_wrapping", since = "1.2.0")] - #[inline] - pub fn wrapping_div(self, rhs: Self) -> Self { - self.overflowing_div(rhs).0 + doc_comment! { + concat!("Saturating integer multiplication. Computes `self * rhs`, saturating at the +numeric bounds instead of overflowing. + +# Examples + +Basic usage: + +``` +", $Feature, "use std::", stringify!($SelfT), "; + +assert_eq!(10", stringify!($SelfT), ".saturating_mul(12), 120); +assert_eq!(", stringify!($SelfT), "::MAX.saturating_mul(10), ", stringify!($SelfT), "::MAX); +assert_eq!(", stringify!($SelfT), "::MIN.saturating_mul(10), ", stringify!($SelfT), "::MIN);", +$EndFeature, " +```"), + #[stable(feature = "wrapping", since = "1.7.0")] + #[inline] + pub fn saturating_mul(self, rhs: Self) -> Self { + self.checked_mul(rhs).unwrap_or_else(|| { + if (self < 0 && rhs < 0) || (self > 0 && rhs > 0) { + Self::max_value() + } else { + Self::min_value() + } + }) + } } - /// Wrapping (modular) remainder. Computes `self % rhs`, - /// wrapping around at the boundary of the type. - /// - /// Such wrap-around never actually occurs mathematically; - /// implementation artifacts make `x % y` invalid for `MIN / - /// -1` on a signed type (where `MIN` is the negative - /// minimal value). In such a case, this function returns `0`. - /// - /// # Panics - /// - /// This function will panic if `rhs` is 0. - /// - /// # Examples - /// - /// Basic usage: - /// - /// ``` - /// assert_eq!(100i8.wrapping_rem(10), 0); - /// assert_eq!((-128i8).wrapping_rem(-1), 0); - /// ``` - #[stable(feature = "num_wrapping", since = "1.2.0")] - #[inline] - pub fn wrapping_rem(self, rhs: Self) -> Self { - self.overflowing_rem(rhs).0 + doc_comment! { + concat!("Saturating integer exponentiation. Computes `self.pow(exp)`, +saturating at the numeric bounds instead of overflowing. + +# Examples + +Basic usage: + +``` +#![feature(no_panic_pow)] +", $Feature, "use std::", stringify!($SelfT), "; + +assert_eq!((-4", stringify!($SelfT), ").saturating_pow(3), -64); +assert_eq!(", stringify!($SelfT), "::MIN.saturating_pow(2), ", stringify!($SelfT), "::MAX); +assert_eq!(", stringify!($SelfT), "::MIN.saturating_pow(3), ", stringify!($SelfT), "::MIN);", +$EndFeature, " +```"), + #[unstable(feature = "no_panic_pow", issue = "48320")] + #[inline] + pub fn saturating_pow(self, exp: u32) -> Self { + match self.checked_pow(exp) { + Some(x) => x, + None if self < 0 && exp % 2 == 1 => Self::min_value(), + None => Self::max_value(), + } + } } - /// Wrapping (modular) negation. Computes `-self`, - /// wrapping around at the boundary of the type. - /// - /// The only case where such wrapping can occur is when one - /// negates `MIN` on a signed type (where `MIN` is the - /// negative minimal value for the type); this is a positive - /// value that is too large to represent in the type. In such - /// a case, this function returns `MIN` itself. - /// - /// # Examples - /// - /// Basic usage: - /// - /// ``` - /// assert_eq!(100i8.wrapping_neg(), -100); - /// assert_eq!((-128i8).wrapping_neg(), -128); - /// ``` - #[stable(feature = "num_wrapping", since = "1.2.0")] - #[inline] - pub fn wrapping_neg(self) -> Self { - self.overflowing_neg().0 + doc_comment! { + concat!("Wrapping (modular) addition. Computes `self + rhs`, wrapping around at the +boundary of the type. + +# Examples + +Basic usage: + +``` +", $Feature, "assert_eq!(100", stringify!($SelfT), ".wrapping_add(27), 127); +assert_eq!(", stringify!($SelfT), "::max_value().wrapping_add(2), ", stringify!($SelfT), +"::min_value() + 1);", +$EndFeature, " +```"), + #[stable(feature = "rust1", since = "1.0.0")] + #[inline] + pub fn wrapping_add(self, rhs: Self) -> Self { + unsafe { + intrinsics::overflowing_add(self, rhs) + } + } } - /// Panic-free bitwise shift-left; yields `self << mask(rhs)`, - /// where `mask` removes any high-order bits of `rhs` that - /// would cause the shift to exceed the bitwidth of the type. - /// - /// Note that this is *not* the same as a rotate-left; the - /// RHS of a wrapping shift-left is restricted to the range - /// of the type, rather than the bits shifted out of the LHS - /// being returned to the other end. The primitive integer - /// types all implement a `rotate_left` function, which may - /// be what you want instead. - /// - /// # Examples - /// - /// Basic usage: - /// - /// ``` - /// assert_eq!((-1i8).wrapping_shl(7), -128); - /// assert_eq!((-1i8).wrapping_shl(8), -1); - /// ``` - #[stable(feature = "num_wrapping", since = "1.2.0")] - #[inline] - pub fn wrapping_shl(self, rhs: u32) -> Self { - unsafe { - intrinsics::unchecked_shl(self, (rhs & ($BITS - 1)) as $SelfT) + doc_comment! { + concat!("Wrapping (modular) subtraction. Computes `self - rhs`, wrapping around at the +boundary of the type. + +# Examples + +Basic usage: + +``` +", $Feature, "assert_eq!(0", stringify!($SelfT), ".wrapping_sub(127), -127); +assert_eq!((-2", stringify!($SelfT), ").wrapping_sub(", stringify!($SelfT), "::max_value()), ", +stringify!($SelfT), "::max_value());", +$EndFeature, " +```"), + #[stable(feature = "rust1", since = "1.0.0")] + #[inline] + pub fn wrapping_sub(self, rhs: Self) -> Self { + unsafe { + intrinsics::overflowing_sub(self, rhs) + } } } - /// Panic-free bitwise shift-right; yields `self >> mask(rhs)`, - /// where `mask` removes any high-order bits of `rhs` that - /// would cause the shift to exceed the bitwidth of the type. - /// - /// Note that this is *not* the same as a rotate-right; the - /// RHS of a wrapping shift-right is restricted to the range - /// of the type, rather than the bits shifted out of the LHS - /// being returned to the other end. The primitive integer - /// types all implement a `rotate_right` function, which may - /// be what you want instead. - /// - /// # Examples - /// - /// Basic usage: - /// - /// ``` - /// assert_eq!((-128i8).wrapping_shr(7), -1); - /// assert_eq!((-128i8).wrapping_shr(8), -128); - /// ``` - #[stable(feature = "num_wrapping", since = "1.2.0")] - #[inline] - pub fn wrapping_shr(self, rhs: u32) -> Self { - unsafe { - intrinsics::unchecked_shr(self, (rhs & ($BITS - 1)) as $SelfT) + doc_comment! { + concat!("Wrapping (modular) multiplication. Computes `self * rhs`, wrapping around at +the boundary of the type. + +# Examples + +Basic usage: + +``` +", $Feature, "assert_eq!(10", stringify!($SelfT), ".wrapping_mul(12), 120); +assert_eq!(11i8.wrapping_mul(12), -124);", +$EndFeature, " +```"), + #[stable(feature = "rust1", since = "1.0.0")] + #[inline] + pub fn wrapping_mul(self, rhs: Self) -> Self { + unsafe { + intrinsics::overflowing_mul(self, rhs) + } } } - /// Wrapping (modular) absolute value. Computes `self.abs()`, - /// wrapping around at the boundary of the type. - /// - /// The only case where such wrapping can occur is when one takes - /// the absolute value of the negative minimal value for the type - /// this is a positive value that is too large to represent in the - /// type. In such a case, this function returns `MIN` itself. - /// - /// # Examples - /// - /// Basic usage: - /// - /// ``` - /// assert_eq!(100i8.wrapping_abs(), 100); - /// assert_eq!((-100i8).wrapping_abs(), 100); - /// assert_eq!((-128i8).wrapping_abs(), -128); - /// assert_eq!((-128i8).wrapping_abs() as u8, 128); - /// ``` - #[stable(feature = "no_panic_abs", since = "1.13.0")] - #[inline] - pub fn wrapping_abs(self) -> Self { - if self.is_negative() { - self.wrapping_neg() - } else { - self + doc_comment! { + concat!("Wrapping (modular) division. Computes `self / rhs`, wrapping around at the +boundary of the type. + +The only case where such wrapping can occur is when one divides `MIN / -1` on a signed type (where +`MIN` is the negative minimal value for the type); this is equivalent to `-MIN`, a positive value +that is too large to represent in the type. In such a case, this function returns `MIN` itself. + +# Panics + +This function will panic if `rhs` is 0. + +# Examples + +Basic usage: + +``` +", $Feature, "assert_eq!(100", stringify!($SelfT), ".wrapping_div(10), 10); +assert_eq!((-128i8).wrapping_div(-1), -128);", +$EndFeature, " +```"), + #[stable(feature = "num_wrapping", since = "1.2.0")] + #[inline] + pub fn wrapping_div(self, rhs: Self) -> Self { + self.overflowing_div(rhs).0 } } - /// Calculates `self` + `rhs` - /// - /// Returns a tuple of the addition along with a boolean indicating - /// whether an arithmetic overflow would occur. If an overflow would - /// have occurred then the wrapped value is returned. - /// - /// # Examples - /// - /// Basic usage - /// - /// ``` - /// use std::i32; - /// - /// assert_eq!(5i32.overflowing_add(2), (7, false)); - /// assert_eq!(i32::MAX.overflowing_add(1), (i32::MIN, true)); - /// ``` - #[inline] - #[stable(feature = "wrapping", since = "1.7.0")] - pub fn overflowing_add(self, rhs: Self) -> (Self, bool) { - let (a, b) = unsafe { - intrinsics::add_with_overflow(self as $ActualT, - rhs as $ActualT) - }; - (a as Self, b) + doc_comment! { + concat!("Wrapping Euclidean division. Computes `self.div_euc(rhs)`, +wrapping around at the boundary of the type. + +Wrapping will only occur in `MIN / -1` on a signed type (where `MIN` is the negative minimal value +for the type). This is equivalent to `-MIN`, a positive value that is too large to represent in the +type. In this case, this method returns `MIN` itself. + +# Panics + +This function will panic if `rhs` is 0. + +# Examples + +Basic usage: + +``` +#![feature(euclidean_division)] +assert_eq!(100", stringify!($SelfT), ".wrapping_div_euc(10), 10); +assert_eq!((-128i8).wrapping_div_euc(-1), -128); +```"), + #[unstable(feature = "euclidean_division", issue = "49048")] + #[inline] + pub fn wrapping_div_euc(self, rhs: Self) -> Self { + self.overflowing_div_euc(rhs).0 + } } - /// Calculates `self` - `rhs` - /// - /// Returns a tuple of the subtraction along with a boolean indicating - /// whether an arithmetic overflow would occur. If an overflow would - /// have occurred then the wrapped value is returned. - /// - /// # Examples - /// - /// Basic usage - /// - /// ``` - /// use std::i32; - /// - /// assert_eq!(5i32.overflowing_sub(2), (3, false)); - /// assert_eq!(i32::MIN.overflowing_sub(1), (i32::MAX, true)); - /// ``` - #[inline] - #[stable(feature = "wrapping", since = "1.7.0")] - pub fn overflowing_sub(self, rhs: Self) -> (Self, bool) { - let (a, b) = unsafe { - intrinsics::sub_with_overflow(self as $ActualT, - rhs as $ActualT) - }; - (a as Self, b) + doc_comment! { + concat!("Wrapping (modular) remainder. Computes `self % rhs`, wrapping around at the +boundary of the type. + +Such wrap-around never actually occurs mathematically; implementation artifacts make `x % y` +invalid for `MIN / -1` on a signed type (where `MIN` is the negative minimal value). In such a case, +this function returns `0`. + +# Panics + +This function will panic if `rhs` is 0. + +# Examples + +Basic usage: + +``` +", $Feature, "assert_eq!(100", stringify!($SelfT), ".wrapping_rem(10), 0); +assert_eq!((-128i8).wrapping_rem(-1), 0);", +$EndFeature, " +```"), + #[stable(feature = "num_wrapping", since = "1.2.0")] + #[inline] + pub fn wrapping_rem(self, rhs: Self) -> Self { + self.overflowing_rem(rhs).0 + } } - /// Calculates the multiplication of `self` and `rhs`. - /// - /// Returns a tuple of the multiplication along with a boolean - /// indicating whether an arithmetic overflow would occur. If an - /// overflow would have occurred then the wrapped value is returned. - /// - /// # Examples - /// - /// Basic usage - /// - /// ``` - /// assert_eq!(5i32.overflowing_mul(2), (10, false)); - /// assert_eq!(1_000_000_000i32.overflowing_mul(10), (1410065408, true)); - /// ``` - #[inline] - #[stable(feature = "wrapping", since = "1.7.0")] - pub fn overflowing_mul(self, rhs: Self) -> (Self, bool) { - let (a, b) = unsafe { - intrinsics::mul_with_overflow(self as $ActualT, - rhs as $ActualT) - }; - (a as Self, b) + doc_comment! { + concat!("Wrapping Euclidean modulo. Computes `self.mod_euc(rhs)`, wrapping around at the +boundary of the type. + +Wrapping will only occur in `MIN % -1` on a signed type (where `MIN` is the negative minimal value +for the type). In this case, this method returns 0. + +# Panics + +This function will panic if `rhs` is 0. + +# Examples + +Basic usage: + +``` +#![feature(euclidean_division)] +assert_eq!(100", stringify!($SelfT), ".wrapping_mod_euc(10), 0); +assert_eq!((-128i8).wrapping_mod_euc(-1), 0); +```"), + #[unstable(feature = "euclidean_division", issue = "49048")] + #[inline] + pub fn wrapping_mod_euc(self, rhs: Self) -> Self { + self.overflowing_mod_euc(rhs).0 + } } - /// Calculates the divisor when `self` is divided by `rhs`. - /// - /// Returns a tuple of the divisor along with a boolean indicating - /// whether an arithmetic overflow would occur. If an overflow would - /// occur then self is returned. - /// - /// # Panics - /// - /// This function will panic if `rhs` is 0. - /// - /// # Examples - /// - /// Basic usage - /// - /// ``` - /// use std::i32; - /// - /// assert_eq!(5i32.overflowing_div(2), (2, false)); - /// assert_eq!(i32::MIN.overflowing_div(-1), (i32::MIN, true)); - /// ``` - #[inline] - #[stable(feature = "wrapping", since = "1.7.0")] - pub fn overflowing_div(self, rhs: Self) -> (Self, bool) { - if self == Self::min_value() && rhs == -1 { - (self, true) - } else { - (self / rhs, false) + doc_comment! { + concat!("Wrapping (modular) negation. Computes `-self`, wrapping around at the boundary +of the type. + +The only case where such wrapping can occur is when one negates `MIN` on a signed type (where `MIN` +is the negative minimal value for the type); this is a positive value that is too large to represent +in the type. In such a case, this function returns `MIN` itself. + +# Examples + +Basic usage: + +``` +", $Feature, "assert_eq!(100", stringify!($SelfT), ".wrapping_neg(), -100); +assert_eq!(", stringify!($SelfT), "::min_value().wrapping_neg(), ", stringify!($SelfT), +"::min_value());", +$EndFeature, " +```"), + #[stable(feature = "num_wrapping", since = "1.2.0")] + #[inline] + pub fn wrapping_neg(self) -> Self { + self.overflowing_neg().0 } } - /// Calculates the remainder when `self` is divided by `rhs`. - /// - /// Returns a tuple of the remainder after dividing along with a boolean - /// indicating whether an arithmetic overflow would occur. If an - /// overflow would occur then 0 is returned. - /// - /// # Panics - /// - /// This function will panic if `rhs` is 0. - /// - /// # Examples - /// - /// Basic usage - /// - /// ``` - /// use std::i32; - /// - /// assert_eq!(5i32.overflowing_rem(2), (1, false)); - /// assert_eq!(i32::MIN.overflowing_rem(-1), (0, true)); - /// ``` - #[inline] - #[stable(feature = "wrapping", since = "1.7.0")] - pub fn overflowing_rem(self, rhs: Self) -> (Self, bool) { - if self == Self::min_value() && rhs == -1 { - (0, true) - } else { - (self % rhs, false) + doc_comment! { + concat!("Panic-free bitwise shift-left; yields `self << mask(rhs)`, where `mask` removes +any high-order bits of `rhs` that would cause the shift to exceed the bitwidth of the type. + +Note that this is *not* the same as a rotate-left; the RHS of a wrapping shift-left is restricted to +the range of the type, rather than the bits shifted out of the LHS being returned to the other end. +The primitive integer types all implement a `rotate_left` function, which may be what you want +instead. + +# Examples + +Basic usage: + +``` +", $Feature, "assert_eq!((-1", stringify!($SelfT), ").wrapping_shl(7), -128); +assert_eq!((-1", stringify!($SelfT), ").wrapping_shl(128), -1);", +$EndFeature, " +```"), + #[stable(feature = "num_wrapping", since = "1.2.0")] + #[inline] + pub fn wrapping_shl(self, rhs: u32) -> Self { + unsafe { + intrinsics::unchecked_shl(self, (rhs & ($BITS - 1)) as $SelfT) + } } } - /// Negates self, overflowing if this is equal to the minimum value. - /// - /// Returns a tuple of the negated version of self along with a boolean - /// indicating whether an overflow happened. If `self` is the minimum - /// value (e.g. `i32::MIN` for values of type `i32`), then the minimum - /// value will be returned again and `true` will be returned for an - /// overflow happening. - /// - /// # Examples - /// - /// Basic usage - /// - /// ``` - /// use std::i32; - /// - /// assert_eq!(2i32.overflowing_neg(), (-2, false)); - /// assert_eq!(i32::MIN.overflowing_neg(), (i32::MIN, true)); - /// ``` - #[inline] - #[stable(feature = "wrapping", since = "1.7.0")] - pub fn overflowing_neg(self) -> (Self, bool) { - if self == Self::min_value() { - (Self::min_value(), true) - } else { - (-self, false) + doc_comment! { + concat!("Panic-free bitwise shift-right; yields `self >> mask(rhs)`, where `mask` +removes any high-order bits of `rhs` that would cause the shift to exceed the bitwidth of the type. + +Note that this is *not* the same as a rotate-right; the RHS of a wrapping shift-right is restricted +to the range of the type, rather than the bits shifted out of the LHS being returned to the other +end. The primitive integer types all implement a `rotate_right` function, which may be what you want +instead. + +# Examples + +Basic usage: + +``` +", $Feature, "assert_eq!((-128", stringify!($SelfT), ").wrapping_shr(7), -1); +assert_eq!((-128i16).wrapping_shr(64), -128);", +$EndFeature, " +```"), + #[stable(feature = "num_wrapping", since = "1.2.0")] + #[inline] + pub fn wrapping_shr(self, rhs: u32) -> Self { + unsafe { + intrinsics::unchecked_shr(self, (rhs & ($BITS - 1)) as $SelfT) + } } } - /// Shifts self left by `rhs` bits. - /// - /// Returns a tuple of the shifted version of self along with a boolean - /// indicating whether the shift value was larger than or equal to the - /// number of bits. If the shift value is too large, then value is - /// masked (N-1) where N is the number of bits, and this value is then - /// used to perform the shift. - /// - /// # Examples - /// - /// Basic usage - /// - /// ``` - /// assert_eq!(0x10i32.overflowing_shl(4), (0x100, false)); - /// assert_eq!(0x10i32.overflowing_shl(36), (0x100, true)); - /// ``` - #[inline] - #[stable(feature = "wrapping", since = "1.7.0")] - pub fn overflowing_shl(self, rhs: u32) -> (Self, bool) { - (self.wrapping_shl(rhs), (rhs > ($BITS - 1))) + doc_comment! { + concat!("Wrapping (modular) absolute value. Computes `self.abs()`, wrapping around at +the boundary of the type. + +The only case where such wrapping can occur is when one takes the absolute value of the negative +minimal value for the type this is a positive value that is too large to represent in the type. In +such a case, this function returns `MIN` itself. + +# Examples + +Basic usage: + +``` +", $Feature, "assert_eq!(100", stringify!($SelfT), ".wrapping_abs(), 100); +assert_eq!((-100", stringify!($SelfT), ").wrapping_abs(), 100); +assert_eq!(", stringify!($SelfT), "::min_value().wrapping_abs(), ", stringify!($SelfT), +"::min_value()); +assert_eq!((-128i8).wrapping_abs() as u8, 128);", +$EndFeature, " +```"), + #[stable(feature = "no_panic_abs", since = "1.13.0")] + #[inline] + pub fn wrapping_abs(self) -> Self { + if self.is_negative() { + self.wrapping_neg() + } else { + self + } + } } - /// Shifts self right by `rhs` bits. - /// - /// Returns a tuple of the shifted version of self along with a boolean - /// indicating whether the shift value was larger than or equal to the - /// number of bits. If the shift value is too large, then value is - /// masked (N-1) where N is the number of bits, and this value is then - /// used to perform the shift. - /// - /// # Examples - /// - /// Basic usage - /// - /// ``` - /// assert_eq!(0x10i32.overflowing_shr(4), (0x1, false)); - /// assert_eq!(0x10i32.overflowing_shr(36), (0x1, true)); - /// ``` - #[inline] - #[stable(feature = "wrapping", since = "1.7.0")] - pub fn overflowing_shr(self, rhs: u32) -> (Self, bool) { - (self.wrapping_shr(rhs), (rhs > ($BITS - 1))) + doc_comment! { + concat!("Wrapping (modular) exponentiation. Computes `self.pow(exp)`, +wrapping around at the boundary of the type. + +# Examples + +Basic usage: + +``` +#![feature(no_panic_pow)] +", $Feature, "assert_eq!(3", stringify!($SelfT), ".wrapping_pow(4), 81); +assert_eq!(3i8.wrapping_pow(5), -13); +assert_eq!(3i8.wrapping_pow(6), -39);", +$EndFeature, " +```"), + #[unstable(feature = "no_panic_pow", issue = "48320")] + #[inline] + pub fn wrapping_pow(self, mut exp: u32) -> Self { + let mut base = self; + let mut acc: Self = 1; + + while exp > 1 { + if (exp & 1) == 1 { + acc = acc.wrapping_mul(base); + } + exp /= 2; + base = base.wrapping_mul(base); + } + + // Deal with the final bit of the exponent separately, since + // squaring the base afterwards is not necessary and may cause a + // needless overflow. + if exp == 1 { + acc = acc.wrapping_mul(base); + } + + acc + } } - /// Computes the absolute value of `self`. - /// - /// Returns a tuple of the absolute version of self along with a - /// boolean indicating whether an overflow happened. If self is the - /// minimum value (e.g. i32::MIN for values of type i32), then the - /// minimum value will be returned again and true will be returned for - /// an overflow happening. + doc_comment! { + concat!("Calculates `self` + `rhs` + +Returns a tuple of the addition along with a boolean indicating whether an arithmetic overflow would +occur. If an overflow would have occurred then the wrapped value is returned. + +# Examples + +Basic usage: + +``` +", $Feature, "use std::", stringify!($SelfT), "; + +assert_eq!(5", stringify!($SelfT), ".overflowing_add(2), (7, false)); +assert_eq!(", stringify!($SelfT), "::MAX.overflowing_add(1), (", stringify!($SelfT), +"::MIN, true));", $EndFeature, " +```"), + #[inline] + #[stable(feature = "wrapping", since = "1.7.0")] + pub fn overflowing_add(self, rhs: Self) -> (Self, bool) { + let (a, b) = unsafe { + intrinsics::add_with_overflow(self as $ActualT, + rhs as $ActualT) + }; + (a as Self, b) + } + } + + doc_comment! { + concat!("Calculates `self` - `rhs` + +Returns a tuple of the subtraction along with a boolean indicating whether an arithmetic overflow +would occur. If an overflow would have occurred then the wrapped value is returned. + +# Examples + +Basic usage: + +``` +", $Feature, "use std::", stringify!($SelfT), "; + +assert_eq!(5", stringify!($SelfT), ".overflowing_sub(2), (3, false)); +assert_eq!(", stringify!($SelfT), "::MIN.overflowing_sub(1), (", stringify!($SelfT), +"::MAX, true));", $EndFeature, " +```"), + #[inline] + #[stable(feature = "wrapping", since = "1.7.0")] + pub fn overflowing_sub(self, rhs: Self) -> (Self, bool) { + let (a, b) = unsafe { + intrinsics::sub_with_overflow(self as $ActualT, + rhs as $ActualT) + }; + (a as Self, b) + } + } + + doc_comment! { + concat!("Calculates the multiplication of `self` and `rhs`. + +Returns a tuple of the multiplication along with a boolean indicating whether an arithmetic overflow +would occur. If an overflow would have occurred then the wrapped value is returned. + +# Examples + +Basic usage: + +``` +", $Feature, "assert_eq!(5", stringify!($SelfT), ".overflowing_mul(2), (10, false)); +assert_eq!(1_000_000_000i32.overflowing_mul(10), (1410065408, true));", +$EndFeature, " +```"), + #[inline] + #[stable(feature = "wrapping", since = "1.7.0")] + pub fn overflowing_mul(self, rhs: Self) -> (Self, bool) { + let (a, b) = unsafe { + intrinsics::mul_with_overflow(self as $ActualT, + rhs as $ActualT) + }; + (a as Self, b) + } + } + + doc_comment! { + concat!("Calculates the divisor when `self` is divided by `rhs`. + +Returns a tuple of the divisor along with a boolean indicating whether an arithmetic overflow would +occur. If an overflow would occur then self is returned. + +# Panics + +This function will panic if `rhs` is 0. + +# Examples + +Basic usage: + +``` +", $Feature, "use std::", stringify!($SelfT), "; + +assert_eq!(5", stringify!($SelfT), ".overflowing_div(2), (2, false)); +assert_eq!(", stringify!($SelfT), "::MIN.overflowing_div(-1), (", stringify!($SelfT), +"::MIN, true));", +$EndFeature, " +```"), + #[inline] + #[stable(feature = "wrapping", since = "1.7.0")] + pub fn overflowing_div(self, rhs: Self) -> (Self, bool) { + if self == Self::min_value() && rhs == -1 { + (self, true) + } else { + (self / rhs, false) + } + } + } + + doc_comment! { + concat!("Calculates the quotient of Euclidean division `self.div_euc(rhs)`. + +Returns a tuple of the divisor along with a boolean indicating whether an arithmetic overflow would +occur. If an overflow would occur then `self` is returned. + +# Panics + +This function will panic if `rhs` is 0. + +# Examples + +Basic usage: + +``` +#![feature(euclidean_division)] +use std::", stringify!($SelfT), "; + +assert_eq!(5", stringify!($SelfT), ".overflowing_div_euc(2), (2, false)); +assert_eq!(", stringify!($SelfT), "::MIN.overflowing_div_euc(-1), (", stringify!($SelfT), +"::MIN, true)); +```"), + #[inline] + #[unstable(feature = "euclidean_division", issue = "49048")] + pub fn overflowing_div_euc(self, rhs: Self) -> (Self, bool) { + if self == Self::min_value() && rhs == -1 { + (self, true) + } else { + (self.div_euc(rhs), false) + } + } + } + + doc_comment! { + concat!("Calculates the remainder when `self` is divided by `rhs`. + +Returns a tuple of the remainder after dividing along with a boolean indicating whether an +arithmetic overflow would occur. If an overflow would occur then 0 is returned. + +# Panics + +This function will panic if `rhs` is 0. + +# Examples + +Basic usage: + +``` +", $Feature, "use std::", stringify!($SelfT), "; + +assert_eq!(5", stringify!($SelfT), ".overflowing_rem(2), (1, false)); +assert_eq!(", stringify!($SelfT), "::MIN.overflowing_rem(-1), (0, true));", +$EndFeature, " +```"), + #[inline] + #[stable(feature = "wrapping", since = "1.7.0")] + pub fn overflowing_rem(self, rhs: Self) -> (Self, bool) { + if self == Self::min_value() && rhs == -1 { + (0, true) + } else { + (self % rhs, false) + } + } + } + + + doc_comment! { + concat!("Calculates the remainder `self.mod_euc(rhs)` by Euclidean division. + +Returns a tuple of the remainder after dividing along with a boolean indicating whether an +arithmetic overflow would occur. If an overflow would occur then 0 is returned. + +# Panics + +This function will panic if `rhs` is 0. + +# Examples + +Basic usage: + +``` +#![feature(euclidean_division)] +use std::", stringify!($SelfT), "; + +assert_eq!(5", stringify!($SelfT), ".overflowing_mod_euc(2), (1, false)); +assert_eq!(", stringify!($SelfT), "::MIN.overflowing_mod_euc(-1), (0, true)); +```"), + #[unstable(feature = "euclidean_division", issue = "49048")] + #[inline] + pub fn overflowing_mod_euc(self, rhs: Self) -> (Self, bool) { + if self == Self::min_value() && rhs == -1 { + (0, true) + } else { + (self.mod_euc(rhs), false) + } + } + } + + + doc_comment! { + concat!("Negates self, overflowing if this is equal to the minimum value. + +Returns a tuple of the negated version of self along with a boolean indicating whether an overflow +happened. If `self` is the minimum value (e.g. `i32::MIN` for values of type `i32`), then the +minimum value will be returned again and `true` will be returned for an overflow happening. + +# Examples + +Basic usage: + +``` +", $Feature, "use std::", stringify!($SelfT), "; + +assert_eq!(2", stringify!($SelfT), ".overflowing_neg(), (-2, false)); +assert_eq!(", stringify!($SelfT), "::MIN.overflowing_neg(), (", stringify!($SelfT), +"::MIN, true));", $EndFeature, " +```"), + #[inline] + #[stable(feature = "wrapping", since = "1.7.0")] + pub fn overflowing_neg(self) -> (Self, bool) { + if self == Self::min_value() { + (Self::min_value(), true) + } else { + (-self, false) + } + } + } + + doc_comment! { + concat!("Shifts self left by `rhs` bits. + +Returns a tuple of the shifted version of self along with a boolean indicating whether the shift +value was larger than or equal to the number of bits. If the shift value is too large, then value is +masked (N-1) where N is the number of bits, and this value is then used to perform the shift. + +# Examples + +Basic usage: + +``` +", $Feature, "assert_eq!(0x1", stringify!($SelfT),".overflowing_shl(4), (0x10, false)); +assert_eq!(0x1i32.overflowing_shl(36), (0x10, true));", +$EndFeature, " +```"), + #[inline] + #[stable(feature = "wrapping", since = "1.7.0")] + pub fn overflowing_shl(self, rhs: u32) -> (Self, bool) { + (self.wrapping_shl(rhs), (rhs > ($BITS - 1))) + } + } + + doc_comment! { + concat!("Shifts self right by `rhs` bits. + +Returns a tuple of the shifted version of self along with a boolean indicating whether the shift +value was larger than or equal to the number of bits. If the shift value is too large, then value is +masked (N-1) where N is the number of bits, and this value is then used to perform the shift. + +# Examples + +Basic usage: + +``` +", $Feature, "assert_eq!(0x10", stringify!($SelfT), ".overflowing_shr(4), (0x1, false)); +assert_eq!(0x10i32.overflowing_shr(36), (0x1, true));", +$EndFeature, " +```"), + #[inline] + #[stable(feature = "wrapping", since = "1.7.0")] + pub fn overflowing_shr(self, rhs: u32) -> (Self, bool) { + (self.wrapping_shr(rhs), (rhs > ($BITS - 1))) + } + } + + doc_comment! { + concat!("Computes the absolute value of `self`. + +Returns a tuple of the absolute version of self along with a boolean indicating whether an overflow +happened. If self is the minimum value (e.g. ", stringify!($SelfT), "::MIN for values of type + ", stringify!($SelfT), "), then the minimum value will be returned again and true will be returned +for an overflow happening. + +# Examples + +Basic usage: + +``` +", $Feature, "assert_eq!(10", stringify!($SelfT), ".overflowing_abs(), (10, false)); +assert_eq!((-10", stringify!($SelfT), ").overflowing_abs(), (10, false)); +assert_eq!((", stringify!($SelfT), "::min_value()).overflowing_abs(), (", stringify!($SelfT), +"::min_value(), true));", +$EndFeature, " +```"), + #[stable(feature = "no_panic_abs", since = "1.13.0")] + #[inline] + pub fn overflowing_abs(self) -> (Self, bool) { + if self.is_negative() { + self.overflowing_neg() + } else { + (self, false) + } + } + } + + doc_comment! { + concat!("Raises self to the power of `exp`, using exponentiation by squaring. + +Returns a tuple of the exponentiation along with a bool indicating +whether an overflow happened. + +# Examples + +Basic usage: + +``` +#![feature(no_panic_pow)] +", $Feature, "assert_eq!(3", stringify!($SelfT), ".overflowing_pow(4), (81, false)); +assert_eq!(3i8.overflowing_pow(5), (-13, true));", +$EndFeature, " +```"), + #[unstable(feature = "no_panic_pow", issue = "48320")] + #[inline] + pub fn overflowing_pow(self, mut exp: u32) -> (Self, bool) { + let mut base = self; + let mut acc: Self = 1; + let mut overflown = false; + // Scratch space for storing results of overflowing_mul. + let mut r; + + while exp > 1 { + if (exp & 1) == 1 { + r = acc.overflowing_mul(base); + acc = r.0; + overflown |= r.1; + } + exp /= 2; + r = base.overflowing_mul(base); + base = r.0; + overflown |= r.1; + } + + // Deal with the final bit of the exponent separately, since + // squaring the base afterwards is not necessary and may cause a + // needless overflow. + if exp == 1 { + r = acc.overflowing_mul(base); + acc = r.0; + overflown |= r.1; + } + + (acc, overflown) + } + } + + doc_comment! { + concat!("Raises self to the power of `exp`, using exponentiation by squaring. + +# Examples + +Basic usage: + +``` +", $Feature, "let x: ", stringify!($SelfT), " = 2; // or any other integer type + +assert_eq!(x.pow(5), 32);", +$EndFeature, " +```"), + #[stable(feature = "rust1", since = "1.0.0")] + #[inline] + #[rustc_inherit_overflow_checks] + pub fn pow(self, mut exp: u32) -> Self { + let mut base = self; + let mut acc = 1; + + while exp > 1 { + if (exp & 1) == 1 { + acc = acc * base; + } + exp /= 2; + base = base * base; + } + + // Deal with the final bit of the exponent separately, since + // squaring the base afterwards is not necessary and may cause a + // needless overflow. + if exp == 1 { + acc = acc * base; + } + + acc + } + } + + doc_comment! { + concat!("Calculates the quotient of Euclidean division of `self` by `rhs`. + +This computes the integer `n` such that `self = n * rhs + self.mod_euc(rhs)`. +In other words, the result is `self / rhs` rounded to the integer `n` +such that `self >= n * rhs`. + +# Panics + +This function will panic if `rhs` is 0. + +# Examples + +Basic usage: + +``` +#![feature(euclidean_division)] +let a: ", stringify!($SelfT), " = 7; // or any other integer type +let b = 4; + +assert_eq!(a.div_euc(b), 1); // 7 >= 4 * 1 +assert_eq!(a.div_euc(-b), -1); // 7 >= -4 * -1 +assert_eq!((-a).div_euc(b), -2); // -7 >= 4 * -2 +assert_eq!((-a).div_euc(-b), 2); // -7 >= -4 * 2 +```"), + #[unstable(feature = "euclidean_division", issue = "49048")] + #[inline] + #[rustc_inherit_overflow_checks] + pub fn div_euc(self, rhs: Self) -> Self { + let q = self / rhs; + if self % rhs < 0 { + return if rhs > 0 { q - 1 } else { q + 1 } + } + q + } + } + + + doc_comment! { + concat!("Calculates the remainder `self mod rhs` by Euclidean division. + +In particular, the result `n` satisfies `0 <= n < rhs.abs()`. + +# Panics + +This function will panic if `rhs` is 0. + +# Examples + +Basic usage: + +``` +#![feature(euclidean_division)] +let a: ", stringify!($SelfT), " = 7; // or any other integer type +let b = 4; + +assert_eq!(a.mod_euc(b), 3); +assert_eq!((-a).mod_euc(b), 1); +assert_eq!(a.mod_euc(-b), 3); +assert_eq!((-a).mod_euc(-b), 1); +```"), + #[unstable(feature = "euclidean_division", issue = "49048")] + #[inline] + #[rustc_inherit_overflow_checks] + pub fn mod_euc(self, rhs: Self) -> Self { + let r = self % rhs; + if r < 0 { + if rhs < 0 { + r - rhs + } else { + r + rhs + } + } else { + r + } + } + } + + doc_comment! { + concat!("Computes the absolute value of `self`. + +# Overflow behavior + +The absolute value of `", stringify!($SelfT), "::min_value()` cannot be represented as an +`", stringify!($SelfT), "`, and attempting to calculate it will cause an overflow. This means that +code in debug mode will trigger a panic on this case and optimized code will return `", +stringify!($SelfT), "::min_value()` without a panic. + +# Examples + +Basic usage: + +``` +", $Feature, "assert_eq!(10", stringify!($SelfT), ".abs(), 10); +assert_eq!((-10", stringify!($SelfT), ").abs(), 10);", +$EndFeature, " +```"), + #[stable(feature = "rust1", since = "1.0.0")] + #[inline] + #[rustc_inherit_overflow_checks] + pub fn abs(self) -> Self { + if self.is_negative() { + // Note that the #[inline] above means that the overflow + // semantics of this negation depend on the crate we're being + // inlined into. + -self + } else { + self + } + } + } + + doc_comment! { + concat!("Returns a number representing sign of `self`. + + - `0` if the number is zero + - `1` if the number is positive + - `-1` if the number is negative + +# Examples + +Basic usage: + +``` +", $Feature, "assert_eq!(10", stringify!($SelfT), ".signum(), 1); +assert_eq!(0", stringify!($SelfT), ".signum(), 0); +assert_eq!((-10", stringify!($SelfT), ").signum(), -1);", +$EndFeature, " +```"), + #[stable(feature = "rust1", since = "1.0.0")] + #[inline] + pub fn signum(self) -> Self { + match self { + n if n > 0 => 1, + 0 => 0, + _ => -1, + } + } + } + + doc_comment! { + concat!("Returns `true` if `self` is positive and `false` if the number is zero or +negative. + +# Examples + +Basic usage: + +``` +", $Feature, "assert!(10", stringify!($SelfT), ".is_positive()); +assert!(!(-10", stringify!($SelfT), ").is_positive());", +$EndFeature, " +```"), + #[stable(feature = "rust1", since = "1.0.0")] + #[inline] + pub fn is_positive(self) -> bool { self > 0 } + } + + doc_comment! { + concat!("Returns `true` if `self` is negative and `false` if the number is zero or +positive. + +# Examples + +Basic usage: + +``` +", $Feature, "assert!((-10", stringify!($SelfT), ").is_negative()); +assert!(!10", stringify!($SelfT), ".is_negative());", +$EndFeature, " +```"), + #[stable(feature = "rust1", since = "1.0.0")] + #[inline] + pub fn is_negative(self) -> bool { self < 0 } + } + + /// Return the memory representation of this integer as a byte array in + /// big-endian (network) byte order. /// /// # Examples /// - /// Basic usage: - /// /// ``` - /// assert_eq!(10i8.overflowing_abs(), (10,false)); - /// assert_eq!((-10i8).overflowing_abs(), (10,false)); - /// assert_eq!((-128i8).overflowing_abs(), (-128,true)); + /// #![feature(int_to_from_bytes)] + /// + /// let bytes = 0x12_34_56_78_i32.to_be_bytes(); + /// assert_eq!(bytes, [0x12, 0x34, 0x56, 0x78]); /// ``` - #[stable(feature = "no_panic_abs", since = "1.13.0")] + #[unstable(feature = "int_to_from_bytes", issue = "52963")] #[inline] - pub fn overflowing_abs(self) -> (Self, bool) { - if self.is_negative() { - self.overflowing_neg() - } else { - (self, false) - } + pub fn to_be_bytes(self) -> [u8; mem::size_of::<Self>()] { + self.to_be().to_ne_bytes() } - /// Raises self to the power of `exp`, using exponentiation by squaring. + /// Return the memory representation of this integer as a byte array in + /// little-endian byte order. /// /// # Examples /// - /// Basic usage: - /// /// ``` - /// let x: i32 = 2; // or any other integer type + /// #![feature(int_to_from_bytes)] /// - /// assert_eq!(x.pow(4), 16); + /// let bytes = 0x12_34_56_78_i32.to_le_bytes(); + /// assert_eq!(bytes, [0x78, 0x56, 0x34, 0x12]); /// ``` - #[stable(feature = "rust1", since = "1.0.0")] + #[unstable(feature = "int_to_from_bytes", issue = "52963")] #[inline] - #[rustc_inherit_overflow_checks] - pub fn pow(self, mut exp: u32) -> Self { - let mut base = self; - let mut acc = 1; - - while exp > 1 { - if (exp & 1) == 1 { - acc = acc * base; - } - exp /= 2; - base = base * base; - } - - // Deal with the final bit of the exponent separately, since - // squaring the base afterwards is not necessary and may cause a - // needless overflow. - if exp == 1 { - acc = acc * base; - } - - acc + pub fn to_le_bytes(self) -> [u8; mem::size_of::<Self>()] { + self.to_le().to_ne_bytes() } - /// Computes the absolute value of `self`. + /// Return the memory representation of this integer as a byte array in + /// native byte order. /// - /// # Overflow behavior + /// As the target platform's native endianness is used, portable code + /// should use [`to_be_bytes`] or [`to_le_bytes`], as appropriate, + /// instead. /// - /// The absolute value of `i32::min_value()` cannot be represented as an - /// `i32`, and attempting to calculate it will cause an overflow. This - /// means that code in debug mode will trigger a panic on this case and - /// optimized code will return `i32::min_value()` without a panic. + /// [`to_be_bytes`]: #method.to_be_bytes + /// [`to_le_bytes`]: #method.to_le_bytes /// /// # Examples /// - /// Basic usage: - /// /// ``` - /// assert_eq!(10i8.abs(), 10); - /// assert_eq!((-10i8).abs(), 10); + /// #![feature(int_to_from_bytes)] + /// + /// let bytes = i32::min_value().to_be().to_ne_bytes(); + /// assert_eq!(bytes, [0x80, 0, 0, 0]); /// ``` - #[stable(feature = "rust1", since = "1.0.0")] + #[unstable(feature = "int_to_from_bytes", issue = "52963")] #[inline] - #[rustc_inherit_overflow_checks] - pub fn abs(self) -> Self { - if self.is_negative() { - // Note that the #[inline] above means that the overflow - // semantics of this negation depend on the crate we're being - // inlined into. - -self - } else { - self - } + pub fn to_ne_bytes(self) -> [u8; mem::size_of::<Self>()] { + unsafe { mem::transmute(self) } } - /// Returns a number representing sign of `self`. - /// - /// - `0` if the number is zero - /// - `1` if the number is positive - /// - `-1` if the number is negative + /// Create an integer value from its representation as a byte array in + /// big endian. /// /// # Examples /// - /// Basic usage: - /// /// ``` - /// assert_eq!(10i8.signum(), 1); - /// assert_eq!(0i8.signum(), 0); - /// assert_eq!((-10i8).signum(), -1); + /// #![feature(int_to_from_bytes)] + /// + /// let int = i32::from_be_bytes([0x12, 0x34, 0x56, 0x78]); + /// assert_eq!(int, 0x12_34_56_78); /// ``` - #[stable(feature = "rust1", since = "1.0.0")] + #[unstable(feature = "int_to_from_bytes", issue = "52963")] #[inline] - pub fn signum(self) -> Self { - match self { - n if n > 0 => 1, - 0 => 0, - _ => -1, - } + pub fn from_be_bytes(bytes: [u8; mem::size_of::<Self>()]) -> Self { + Self::from_be(Self::from_ne_bytes(bytes)) } - /// Returns `true` if `self` is positive and `false` if the number - /// is zero or negative. + /// Create an integer value from its representation as a byte array in + /// little endian. /// /// # Examples /// - /// Basic usage: - /// /// ``` - /// assert!(10i8.is_positive()); - /// assert!(!(-10i8).is_positive()); + /// #![feature(int_to_from_bytes)] + /// + /// let int = i32::from_le_bytes([0x12, 0x34, 0x56, 0x78]); + /// assert_eq!(int, 0x78_56_34_12); /// ``` - #[stable(feature = "rust1", since = "1.0.0")] + #[unstable(feature = "int_to_from_bytes", issue = "52963")] #[inline] - pub fn is_positive(self) -> bool { self > 0 } + pub fn from_le_bytes(bytes: [u8; mem::size_of::<Self>()]) -> Self { + Self::from_le(Self::from_ne_bytes(bytes)) + } - /// Returns `true` if `self` is negative and `false` if the number - /// is zero or positive. + /// Create an integer value from its memory representation as a byte + /// array in native endianness. /// - /// # Examples + /// As the target platform's native endianness is used, portable code + /// likely wants to use [`from_be_bytes`] or [`from_le_bytes`], as + /// appropriate instead. /// - /// Basic usage: + /// [`from_be_bytes`]: #method.from_be_bytes + /// [`from_le_bytes`]: #method.from_le_bytes + /// + /// # Examples /// /// ``` - /// assert!((-10i8).is_negative()); - /// assert!(!10i8.is_negative()); + /// #![feature(int_to_from_bytes)] + /// + /// let int = i32::from_be(i32::from_ne_bytes([0x80, 0, 0, 0])); + /// assert_eq!(int, i32::min_value()); /// ``` - #[stable(feature = "rust1", since = "1.0.0")] + #[unstable(feature = "int_to_from_bytes", issue = "52963")] #[inline] - pub fn is_negative(self) -> bool { self < 0 } + pub fn from_ne_bytes(bytes: [u8; mem::size_of::<Self>()]) -> Self { + unsafe { mem::transmute(bytes) } + } } } #[lang = "i8"] impl i8 { - int_impl! { i8, i8, u8, 8 } + int_impl! { i8, i8, u8, 8, -128, 127, "", "", 2, "-0x7e", "0xa", "0x12", "0x12", "0x48" } } #[lang = "i16"] impl i16 { - int_impl! { i16, i16, u16, 16 } + int_impl! { i16, i16, u16, 16, -32768, 32767, "", "", 4, "-0x5ffd", "0x3a", "0x1234", "0x3412", + "0x2c48" } } #[lang = "i32"] impl i32 { - int_impl! { i32, i32, u32, 32 } + int_impl! { i32, i32, u32, 32, -2147483648, 2147483647, "", "", 8, "0x10000b3", "0xb301", + "0x12345678", "0x78563412", "0x1e6a2c48" } } #[lang = "i64"] impl i64 { - int_impl! { i64, i64, u64, 64 } + int_impl! { i64, i64, u64, 64, -9223372036854775808, 9223372036854775807, "", "", 12, + "0xaa00000000006e1", "0x6e10aa", "0x1234567890123456", "0x5634129078563412", + "0x6a2c48091e6a2c48" } } #[lang = "i128"] impl i128 { - int_impl! { i128, i128, u128, 128 } + int_impl! { i128, i128, u128, 128, -170141183460469231731687303715884105728, + 170141183460469231731687303715884105727, "", "", 16, + "0x13f40000000000000000000000004f76", "0x4f7613f4", "0x12345678901234567890123456789012", + "0x12907856341290785634129078563412", "0x48091e6a2c48091e6a2c48091e6a2c48" + } } #[cfg(target_pointer_width = "16")] #[lang = "isize"] impl isize { - int_impl! { isize, i16, u16, 16 } + int_impl! { isize, i16, u16, 16, -32768, 32767, "", "", 4, "-0x5ffd", "0x3a", "0x1234", + "0x3412", "0x2c48" } } #[cfg(target_pointer_width = "32")] #[lang = "isize"] impl isize { - int_impl! { isize, i32, u32, 32 } + int_impl! { isize, i32, u32, 32, -2147483648, 2147483647, "", "", 8, "0x10000b3", "0xb301", + "0x12345678", "0x78563412", "0x1e6a2c48" } } #[cfg(target_pointer_width = "64")] #[lang = "isize"] impl isize { - int_impl! { isize, i64, u64, 64 } + int_impl! { isize, i64, u64, 64, -9223372036854775808, 9223372036854775807, "", "", + 12, "0xaa00000000006e1", "0x6e10aa", "0x1234567890123456", "0x5634129078563412", + "0x6a2c48091e6a2c48" } +} + +// Emits the correct `cttz` call, depending on the size of the type. +macro_rules! uint_cttz_call { + // As of LLVM 3.6 the codegen for the zero-safe cttz8 intrinsic + // emits two conditional moves on x86_64. By promoting the value to + // u16 and setting bit 8, we get better code without any conditional + // operations. + // FIXME: There's a LLVM patch (http://reviews.llvm.org/D9284) + // pending, remove this workaround once LLVM generates better code + // for cttz8. + ($value:expr, 8) => { intrinsics::cttz($value as u16 | 0x100) }; + ($value:expr, $_BITS:expr) => { intrinsics::cttz($value) } } // `Int` + `UnsignedInt` implemented for unsigned integers macro_rules! uint_impl { - ($SelfT:ty, $ActualT:ty, $BITS:expr) => { - /// Returns the smallest value that can be represented by this integer type. - /// - /// # Examples - /// - /// Basic usage: - /// - /// ``` - /// assert_eq!(u8::min_value(), 0); - /// ``` - #[stable(feature = "rust1", since = "1.0.0")] - #[inline] - pub const fn min_value() -> Self { 0 } + ($SelfT:ty, $ActualT:ty, $BITS:expr, $MaxV:expr, $Feature:expr, $EndFeature:expr, + $rot:expr, $rot_op:expr, $rot_result:expr, $swap_op:expr, $swapped:expr, + $reversed:expr ) => { + doc_comment! { + concat!("Returns the smallest value that can be represented by this integer type. - /// Returns the largest value that can be represented by this integer type. - /// - /// # Examples - /// - /// Basic usage: - /// - /// ``` - /// assert_eq!(u8::max_value(), 255); - /// ``` - #[stable(feature = "rust1", since = "1.0.0")] - #[inline] - pub const fn max_value() -> Self { !0 } +# Examples - /// Converts a string slice in a given base to an integer. - /// - /// The string is expected to be an optional `+` sign - /// followed by digits. - /// Leading and trailing whitespace represent an error. - /// Digits are a subset of these characters, depending on `radix`: - /// - /// * `0-9` - /// * `a-z` - /// * `A-Z` - /// - /// # Panics - /// - /// This function panics if `radix` is not in the range from 2 to 36. - /// - /// # Examples - /// - /// Basic usage: - /// - /// ``` - /// assert_eq!(u32::from_str_radix("A", 16), Ok(10)); - /// ``` - #[stable(feature = "rust1", since = "1.0.0")] - pub fn from_str_radix(src: &str, radix: u32) -> Result<Self, ParseIntError> { - from_str_radix(src, radix) +Basic usage: + +``` +", $Feature, "assert_eq!(", stringify!($SelfT), "::min_value(), 0);", $EndFeature, " +```"), + #[stable(feature = "rust1", since = "1.0.0")] + #[inline] + pub const fn min_value() -> Self { 0 } } - /// Returns the number of ones in the binary representation of `self`. - /// - /// # Examples - /// - /// Basic usage: - /// - /// ``` - /// let n = 0b01001100u8; - /// - /// assert_eq!(n.count_ones(), 3); - /// ``` - #[stable(feature = "rust1", since = "1.0.0")] - #[inline] - pub fn count_ones(self) -> u32 { - unsafe { intrinsics::ctpop(self as $ActualT) as u32 } + doc_comment! { + concat!("Returns the largest value that can be represented by this integer type. + +# Examples + +Basic usage: + +``` +", $Feature, "assert_eq!(", stringify!($SelfT), "::max_value(), ", +stringify!($MaxV), ");", $EndFeature, " +```"), + #[stable(feature = "rust1", since = "1.0.0")] + #[inline] + pub const fn max_value() -> Self { !0 } } - /// Returns the number of zeros in the binary representation of `self`. - /// - /// # Examples - /// - /// Basic usage: - /// - /// ``` - /// let n = 0b01001100u8; - /// - /// assert_eq!(n.count_zeros(), 5); - /// ``` - #[stable(feature = "rust1", since = "1.0.0")] - #[inline] - pub fn count_zeros(self) -> u32 { - (!self).count_ones() + doc_comment! { + concat!("Converts a string slice in a given base to an integer. + +The string is expected to be an optional `+` sign +followed by digits. +Leading and trailing whitespace represent an error. +Digits are a subset of these characters, depending on `radix`: + +* `0-9` +* `a-z` +* `A-Z` + +# Panics + +This function panics if `radix` is not in the range from 2 to 36. + +# Examples + +Basic usage: + +``` +", $Feature, "assert_eq!(", stringify!($SelfT), "::from_str_radix(\"A\", 16), Ok(10));", +$EndFeature, " +```"), + #[stable(feature = "rust1", since = "1.0.0")] + pub fn from_str_radix(src: &str, radix: u32) -> Result<Self, ParseIntError> { + from_str_radix(src, radix) + } } - /// Returns the number of leading zeros in the binary representation - /// of `self`. - /// - /// # Examples - /// - /// Basic usage: - /// - /// ``` - /// let n = 0b0101000u16; - /// - /// assert_eq!(n.leading_zeros(), 10); - /// ``` - #[stable(feature = "rust1", since = "1.0.0")] - #[inline] - pub fn leading_zeros(self) -> u32 { - unsafe { intrinsics::ctlz(self as $ActualT) as u32 } + doc_comment! { + concat!("Returns the number of ones in the binary representation of `self`. + +# Examples + +Basic usage: + +``` +", $Feature, "let n = 0b01001100", stringify!($SelfT), "; + +assert_eq!(n.count_ones(), 3);", $EndFeature, " +```"), + #[stable(feature = "rust1", since = "1.0.0")] + #[rustc_const_unstable(feature = "const_int_ops")] + #[inline] + pub const fn count_ones(self) -> u32 { + unsafe { intrinsics::ctpop(self as $ActualT) as u32 } + } } - /// Returns the number of trailing zeros in the binary representation - /// of `self`. - /// - /// # Examples - /// - /// Basic usage: - /// - /// ``` - /// let n = 0b0101000u16; - /// - /// assert_eq!(n.trailing_zeros(), 3); - /// ``` - #[stable(feature = "rust1", since = "1.0.0")] - #[inline] - pub fn trailing_zeros(self) -> u32 { - // As of LLVM 3.6 the codegen for the zero-safe cttz8 intrinsic - // emits two conditional moves on x86_64. By promoting the value to - // u16 and setting bit 8, we get better code without any conditional - // operations. - // FIXME: There's a LLVM patch (http://reviews.llvm.org/D9284) - // pending, remove this workaround once LLVM generates better code - // for cttz8. - unsafe { - if $BITS == 8 { - intrinsics::cttz(self as u16 | 0x100) as u32 - } else { - intrinsics::cttz(self) as u32 + doc_comment! { + concat!("Returns the number of zeros in the binary representation of `self`. + +# Examples + +Basic usage: + +``` +", $Feature, "assert_eq!(", stringify!($SelfT), "::max_value().count_zeros(), 0);", $EndFeature, " +```"), + #[stable(feature = "rust1", since = "1.0.0")] + #[rustc_const_unstable(feature = "const_int_ops")] + #[inline] + pub const fn count_zeros(self) -> u32 { + (!self).count_ones() + } + } + + doc_comment! { + concat!("Returns the number of leading zeros in the binary representation of `self`. + +# Examples + +Basic usage: + +``` +", $Feature, "let n = ", stringify!($SelfT), "::max_value() >> 2; + +assert_eq!(n.leading_zeros(), 2);", $EndFeature, " +```"), + #[stable(feature = "rust1", since = "1.0.0")] + #[rustc_const_unstable(feature = "const_int_ops")] + #[inline] + pub const fn leading_zeros(self) -> u32 { + unsafe { intrinsics::ctlz(self as $ActualT) as u32 } + } + } + + doc_comment! { + concat!("Returns the number of trailing zeros in the binary representation +of `self`. + +# Examples + +Basic usage: + +``` +", $Feature, "let n = 0b0101000", stringify!($SelfT), "; + +assert_eq!(n.trailing_zeros(), 3);", $EndFeature, " +```"), + #[stable(feature = "rust1", since = "1.0.0")] + #[rustc_const_unstable(feature = "const_int_ops")] + #[inline] + pub const fn trailing_zeros(self) -> u32 { + unsafe { uint_cttz_call!(self, $BITS) as u32 } + } + } + + doc_comment! { + concat!("Shifts the bits to the left by a specified amount, `n`, +wrapping the truncated bits to the end of the resulting integer. + +Please note this isn't the same operation as `<<`! + +# Examples + +Basic usage: + +``` +let n = ", $rot_op, stringify!($SelfT), "; +let m = ", $rot_result, "; + +assert_eq!(n.rotate_left(", $rot, "), m); +```"), + #[stable(feature = "rust1", since = "1.0.0")] + #[inline] + pub fn rotate_left(self, n: u32) -> Self { + // Protect against undefined behaviour for over-long bit shifts + let n = n % $BITS; + (self << n) | (self >> (($BITS - n) % $BITS)) + } + } + + doc_comment! { + concat!("Shifts the bits to the right by a specified amount, `n`, +wrapping the truncated bits to the beginning of the resulting +integer. + +Please note this isn't the same operation as `>>`! + +# Examples + +Basic usage: + +``` +let n = ", $rot_result, stringify!($SelfT), "; +let m = ", $rot_op, "; + +assert_eq!(n.rotate_right(", $rot, "), m); +```"), + #[stable(feature = "rust1", since = "1.0.0")] + #[inline] + pub fn rotate_right(self, n: u32) -> Self { + // Protect against undefined behaviour for over-long bit shifts + let n = n % $BITS; + (self >> n) | (self << (($BITS - n) % $BITS)) + } + } + + doc_comment! { + concat!(" +Reverses the byte order of the integer. + +# Examples + +Basic usage: + +``` +let n = ", $swap_op, stringify!($SelfT), "; +let m = n.swap_bytes(); + +assert_eq!(m, ", $swapped, "); +```"), + #[stable(feature = "rust1", since = "1.0.0")] + #[rustc_const_unstable(feature = "const_int_ops")] + #[inline] + pub const fn swap_bytes(self) -> Self { + unsafe { intrinsics::bswap(self as $ActualT) as Self } + } + } + + doc_comment! { + concat!("Reverses the bit pattern of the integer. + +# Examples + +Basic usage: + +``` +#![feature(reverse_bits)] + +let n = ", $swap_op, stringify!($SelfT), "; +let m = n.reverse_bits(); + +assert_eq!(m, ", $reversed, "); +```"), + #[unstable(feature = "reverse_bits", issue = "48763")] + #[inline] + pub fn reverse_bits(self) -> Self { + unsafe { intrinsics::bitreverse(self as $ActualT) as Self } + } + } + + doc_comment! { + concat!("Converts an integer from big endian to the target's endianness. + +On big endian this is a no-op. On little endian the bytes are +swapped. + +# Examples + +Basic usage: + +``` +", $Feature, "let n = 0x1A", stringify!($SelfT), "; + +if cfg!(target_endian = \"big\") { + assert_eq!(", stringify!($SelfT), "::from_be(n), n) +} else { + assert_eq!(", stringify!($SelfT), "::from_be(n), n.swap_bytes()) +}", $EndFeature, " +```"), + #[stable(feature = "rust1", since = "1.0.0")] + #[rustc_const_unstable(feature = "const_int_ops")] + #[inline] + pub const fn from_be(x: Self) -> Self { + #[cfg(target_endian = "big")] + { + x + } + #[cfg(not(target_endian = "big"))] + { + x.swap_bytes() } } } - /// Shifts the bits to the left by a specified amount, `n`, - /// wrapping the truncated bits to the end of the resulting integer. - /// - /// Please note this isn't the same operation as `<<`! - /// - /// # Examples - /// - /// Basic usage: - /// - /// ``` - /// let n = 0x0123456789ABCDEFu64; - /// let m = 0x3456789ABCDEF012u64; - /// - /// assert_eq!(n.rotate_left(12), m); - /// ``` - #[stable(feature = "rust1", since = "1.0.0")] - #[inline] - pub fn rotate_left(self, n: u32) -> Self { - // Protect against undefined behaviour for over-long bit shifts - let n = n % $BITS; - (self << n) | (self >> (($BITS - n) % $BITS)) + doc_comment! { + concat!("Converts an integer from little endian to the target's endianness. + +On little endian this is a no-op. On big endian the bytes are +swapped. + +# Examples + +Basic usage: + +``` +", $Feature, "let n = 0x1A", stringify!($SelfT), "; + +if cfg!(target_endian = \"little\") { + assert_eq!(", stringify!($SelfT), "::from_le(n), n) +} else { + assert_eq!(", stringify!($SelfT), "::from_le(n), n.swap_bytes()) +}", $EndFeature, " +```"), + #[stable(feature = "rust1", since = "1.0.0")] + #[rustc_const_unstable(feature = "const_int_ops")] + #[inline] + pub const fn from_le(x: Self) -> Self { + #[cfg(target_endian = "little")] + { + x + } + #[cfg(not(target_endian = "little"))] + { + x.swap_bytes() + } + } } - /// Shifts the bits to the right by a specified amount, `n`, - /// wrapping the truncated bits to the beginning of the resulting - /// integer. - /// - /// Please note this isn't the same operation as `>>`! - /// - /// # Examples - /// - /// Basic usage: - /// - /// ``` - /// let n = 0x0123456789ABCDEFu64; - /// let m = 0xDEF0123456789ABCu64; - /// - /// assert_eq!(n.rotate_right(12), m); - /// ``` - #[stable(feature = "rust1", since = "1.0.0")] - #[inline] - pub fn rotate_right(self, n: u32) -> Self { - // Protect against undefined behaviour for over-long bit shifts - let n = n % $BITS; - (self >> n) | (self << (($BITS - n) % $BITS)) + doc_comment! { + concat!("Converts `self` to big endian from the target's endianness. + +On big endian this is a no-op. On little endian the bytes are +swapped. + +# Examples + +Basic usage: + +``` +", $Feature, "let n = 0x1A", stringify!($SelfT), "; + +if cfg!(target_endian = \"big\") { + assert_eq!(n.to_be(), n) +} else { + assert_eq!(n.to_be(), n.swap_bytes()) +}", $EndFeature, " +```"), + #[stable(feature = "rust1", since = "1.0.0")] + #[rustc_const_unstable(feature = "const_int_ops")] + #[inline] + pub const fn to_be(self) -> Self { // or not to be? + #[cfg(target_endian = "big")] + { + self + } + #[cfg(not(target_endian = "big"))] + { + self.swap_bytes() + } + } } - /// Reverses the byte order of the integer. - /// - /// # Examples - /// - /// Basic usage: - /// - /// ``` - /// let n: u16 = 0b0000000_01010101; - /// assert_eq!(n, 85); - /// - /// let m = n.swap_bytes(); - /// - /// assert_eq!(m, 0b01010101_00000000); - /// assert_eq!(m, 21760); - /// ``` - #[stable(feature = "rust1", since = "1.0.0")] - #[inline] - pub fn swap_bytes(self) -> Self { - unsafe { intrinsics::bswap(self as $ActualT) as Self } + doc_comment! { + concat!("Converts `self` to little endian from the target's endianness. + +On little endian this is a no-op. On big endian the bytes are +swapped. + +# Examples + +Basic usage: + +``` +", $Feature, "let n = 0x1A", stringify!($SelfT), "; + +if cfg!(target_endian = \"little\") { + assert_eq!(n.to_le(), n) +} else { + assert_eq!(n.to_le(), n.swap_bytes()) +}", $EndFeature, " +```"), + #[stable(feature = "rust1", since = "1.0.0")] + #[rustc_const_unstable(feature = "const_int_ops")] + #[inline] + pub const fn to_le(self) -> Self { + #[cfg(target_endian = "little")] + { + self + } + #[cfg(not(target_endian = "little"))] + { + self.swap_bytes() + } + } } - /// Converts an integer from big endian to the target's endianness. - /// - /// On big endian this is a no-op. On little endian the bytes are - /// swapped. - /// - /// # Examples - /// - /// Basic usage: - /// - /// ``` - /// let n = 0x0123456789ABCDEFu64; - /// - /// if cfg!(target_endian = "big") { - /// assert_eq!(u64::from_be(n), n) - /// } else { - /// assert_eq!(u64::from_be(n), n.swap_bytes()) - /// } - /// ``` - #[stable(feature = "rust1", since = "1.0.0")] - #[inline] - pub fn from_be(x: Self) -> Self { - if cfg!(target_endian = "big") { x } else { x.swap_bytes() } + doc_comment! { + concat!("Checked integer addition. Computes `self + rhs`, returning `None` +if overflow occurred. + +# Examples + +Basic usage: + +``` +", $Feature, "assert_eq!((", stringify!($SelfT), "::max_value() - 2).checked_add(1), ", +"Some(", stringify!($SelfT), "::max_value() - 1)); +assert_eq!((", stringify!($SelfT), "::max_value() - 2).checked_add(3), None);", $EndFeature, " +```"), + #[stable(feature = "rust1", since = "1.0.0")] + #[inline] + pub fn checked_add(self, rhs: Self) -> Option<Self> { + let (a, b) = self.overflowing_add(rhs); + if b {None} else {Some(a)} + } } - /// Converts an integer from little endian to the target's endianness. - /// - /// On little endian this is a no-op. On big endian the bytes are - /// swapped. - /// - /// # Examples - /// - /// Basic usage: - /// - /// ``` - /// let n = 0x0123456789ABCDEFu64; - /// - /// if cfg!(target_endian = "little") { - /// assert_eq!(u64::from_le(n), n) - /// } else { - /// assert_eq!(u64::from_le(n), n.swap_bytes()) - /// } - /// ``` - #[stable(feature = "rust1", since = "1.0.0")] - #[inline] - pub fn from_le(x: Self) -> Self { - if cfg!(target_endian = "little") { x } else { x.swap_bytes() } + doc_comment! { + concat!("Checked integer subtraction. Computes `self - rhs`, returning +`None` if overflow occurred. + +# Examples + +Basic usage: + +``` +", $Feature, "assert_eq!(1", stringify!($SelfT), ".checked_sub(1), Some(0)); +assert_eq!(0", stringify!($SelfT), ".checked_sub(1), None);", $EndFeature, " +```"), + #[stable(feature = "rust1", since = "1.0.0")] + #[inline] + pub fn checked_sub(self, rhs: Self) -> Option<Self> { + let (a, b) = self.overflowing_sub(rhs); + if b {None} else {Some(a)} + } } - /// Converts `self` to big endian from the target's endianness. - /// - /// On big endian this is a no-op. On little endian the bytes are - /// swapped. - /// - /// # Examples - /// - /// Basic usage: - /// - /// ``` - /// let n = 0x0123456789ABCDEFu64; - /// - /// if cfg!(target_endian = "big") { - /// assert_eq!(n.to_be(), n) - /// } else { - /// assert_eq!(n.to_be(), n.swap_bytes()) - /// } - /// ``` - #[stable(feature = "rust1", since = "1.0.0")] - #[inline] - pub fn to_be(self) -> Self { // or not to be? - if cfg!(target_endian = "big") { self } else { self.swap_bytes() } + doc_comment! { + concat!("Checked integer multiplication. Computes `self * rhs`, returning +`None` if overflow occurred. + +# Examples + +Basic usage: + +``` +", $Feature, "assert_eq!(5", stringify!($SelfT), ".checked_mul(1), Some(5)); +assert_eq!(", stringify!($SelfT), "::max_value().checked_mul(2), None);", $EndFeature, " +```"), + #[stable(feature = "rust1", since = "1.0.0")] + #[inline] + pub fn checked_mul(self, rhs: Self) -> Option<Self> { + let (a, b) = self.overflowing_mul(rhs); + if b {None} else {Some(a)} + } } - /// Converts `self` to little endian from the target's endianness. - /// - /// On little endian this is a no-op. On big endian the bytes are - /// swapped. - /// - /// # Examples - /// - /// Basic usage: - /// - /// ``` - /// let n = 0x0123456789ABCDEFu64; - /// - /// if cfg!(target_endian = "little") { - /// assert_eq!(n.to_le(), n) - /// } else { - /// assert_eq!(n.to_le(), n.swap_bytes()) - /// } - /// ``` - #[stable(feature = "rust1", since = "1.0.0")] - #[inline] - pub fn to_le(self) -> Self { - if cfg!(target_endian = "little") { self } else { self.swap_bytes() } + doc_comment! { + concat!("Checked integer division. Computes `self / rhs`, returning `None` +if `rhs == 0`. + +# Examples + +Basic usage: + +``` +", $Feature, "assert_eq!(128", stringify!($SelfT), ".checked_div(2), Some(64)); +assert_eq!(1", stringify!($SelfT), ".checked_div(0), None);", $EndFeature, " +```"), + #[stable(feature = "rust1", since = "1.0.0")] + #[inline] + pub fn checked_div(self, rhs: Self) -> Option<Self> { + match rhs { + 0 => None, + rhs => Some(unsafe { intrinsics::unchecked_div(self, rhs) }), + } + } } - /// Checked integer addition. Computes `self + rhs`, returning `None` - /// if overflow occurred. - /// - /// # Examples - /// - /// Basic usage: - /// - /// ``` - /// assert_eq!(5u16.checked_add(65530), Some(65535)); - /// assert_eq!(6u16.checked_add(65530), None); - /// ``` - #[stable(feature = "rust1", since = "1.0.0")] - #[inline] - pub fn checked_add(self, rhs: Self) -> Option<Self> { - let (a, b) = self.overflowing_add(rhs); - if b {None} else {Some(a)} + doc_comment! { + concat!("Checked Euclidean division. Computes `self.div_euc(rhs)`, returning `None` +if `rhs == 0`. + +# Examples + +Basic usage: + +``` +#![feature(euclidean_division)] +assert_eq!(128", stringify!($SelfT), ".checked_div(2), Some(64)); +assert_eq!(1", stringify!($SelfT), ".checked_div_euc(0), None); +```"), + #[unstable(feature = "euclidean_division", issue = "49048")] + #[inline] + pub fn checked_div_euc(self, rhs: Self) -> Option<Self> { + if rhs == 0 { + None + } else { + Some(self.div_euc(rhs)) + } + } } - /// Checked integer subtraction. Computes `self - rhs`, returning - /// `None` if overflow occurred. - /// - /// # Examples - /// - /// Basic usage: - /// - /// ``` - /// assert_eq!(1u8.checked_sub(1), Some(0)); - /// assert_eq!(0u8.checked_sub(1), None); - /// ``` - #[stable(feature = "rust1", since = "1.0.0")] - #[inline] - pub fn checked_sub(self, rhs: Self) -> Option<Self> { - let (a, b) = self.overflowing_sub(rhs); - if b {None} else {Some(a)} + + doc_comment! { + concat!("Checked integer remainder. Computes `self % rhs`, returning `None` +if `rhs == 0`. + +# Examples + +Basic usage: + +``` +", $Feature, "assert_eq!(5", stringify!($SelfT), ".checked_rem(2), Some(1)); +assert_eq!(5", stringify!($SelfT), ".checked_rem(0), None);", $EndFeature, " +```"), + #[stable(feature = "wrapping", since = "1.7.0")] + #[inline] + pub fn checked_rem(self, rhs: Self) -> Option<Self> { + if rhs == 0 { + None + } else { + Some(unsafe { intrinsics::unchecked_rem(self, rhs) }) + } + } } - /// Checked integer multiplication. Computes `self * rhs`, returning - /// `None` if overflow occurred. - /// - /// # Examples - /// - /// Basic usage: - /// - /// ``` - /// assert_eq!(5u8.checked_mul(51), Some(255)); - /// assert_eq!(5u8.checked_mul(52), None); - /// ``` - #[stable(feature = "rust1", since = "1.0.0")] - #[inline] - pub fn checked_mul(self, rhs: Self) -> Option<Self> { - let (a, b) = self.overflowing_mul(rhs); - if b {None} else {Some(a)} + doc_comment! { + concat!("Checked Euclidean modulo. Computes `self.mod_euc(rhs)`, returning `None` +if `rhs == 0`. + +# Examples + +Basic usage: + +``` +#![feature(euclidean_division)] +assert_eq!(5", stringify!($SelfT), ".checked_mod_euc(2), Some(1)); +assert_eq!(5", stringify!($SelfT), ".checked_mod_euc(0), None); +```"), + #[unstable(feature = "euclidean_division", issue = "49048")] + #[inline] + pub fn checked_mod_euc(self, rhs: Self) -> Option<Self> { + if rhs == 0 { + None + } else { + Some(self.mod_euc(rhs)) + } + } } - /// Checked integer division. Computes `self / rhs`, returning `None` - /// if `rhs == 0`. - /// - /// # Examples - /// - /// Basic usage: - /// - /// ``` - /// assert_eq!(128u8.checked_div(2), Some(64)); - /// assert_eq!(1u8.checked_div(0), None); - /// ``` - #[stable(feature = "rust1", since = "1.0.0")] - #[inline] - pub fn checked_div(self, rhs: Self) -> Option<Self> { - match rhs { - 0 => None, - rhs => Some(unsafe { intrinsics::unchecked_div(self, rhs) }), + doc_comment! { + concat!("Checked negation. Computes `-self`, returning `None` unless `self == +0`. + +Note that negating any positive integer will overflow. + +# Examples + +Basic usage: + +``` +", $Feature, "assert_eq!(0", stringify!($SelfT), ".checked_neg(), Some(0)); +assert_eq!(1", stringify!($SelfT), ".checked_neg(), None);", $EndFeature, " +```"), + #[stable(feature = "wrapping", since = "1.7.0")] + #[inline] + pub fn checked_neg(self) -> Option<Self> { + let (a, b) = self.overflowing_neg(); + if b {None} else {Some(a)} } } - /// Checked integer remainder. Computes `self % rhs`, returning `None` - /// if `rhs == 0`. - /// - /// # Examples - /// - /// Basic usage: - /// - /// ``` - /// assert_eq!(5u32.checked_rem(2), Some(1)); - /// assert_eq!(5u32.checked_rem(0), None); - /// ``` - #[stable(feature = "wrapping", since = "1.7.0")] - #[inline] - pub fn checked_rem(self, rhs: Self) -> Option<Self> { - if rhs == 0 { - None - } else { - Some(unsafe { intrinsics::unchecked_rem(self, rhs) }) + doc_comment! { + concat!("Checked shift left. Computes `self << rhs`, returning `None` +if `rhs` is larger than or equal to the number of bits in `self`. + +# Examples + +Basic usage: + +``` +", $Feature, "assert_eq!(0x1", stringify!($SelfT), ".checked_shl(4), Some(0x10)); +assert_eq!(0x10", stringify!($SelfT), ".checked_shl(129), None);", $EndFeature, " +```"), + #[stable(feature = "wrapping", since = "1.7.0")] + #[inline] + pub fn checked_shl(self, rhs: u32) -> Option<Self> { + let (a, b) = self.overflowing_shl(rhs); + if b {None} else {Some(a)} } } - /// Checked negation. Computes `-self`, returning `None` unless `self == - /// 0`. - /// - /// Note that negating any positive integer will overflow. - /// - /// # Examples - /// - /// Basic usage: - /// - /// ``` - /// assert_eq!(0u32.checked_neg(), Some(0)); - /// assert_eq!(1u32.checked_neg(), None); - /// ``` - #[stable(feature = "wrapping", since = "1.7.0")] - #[inline] - pub fn checked_neg(self) -> Option<Self> { - let (a, b) = self.overflowing_neg(); - if b {None} else {Some(a)} + doc_comment! { + concat!("Checked shift right. Computes `self >> rhs`, returning `None` +if `rhs` is larger than or equal to the number of bits in `self`. + +# Examples + +Basic usage: + +``` +", $Feature, "assert_eq!(0x10", stringify!($SelfT), ".checked_shr(4), Some(0x1)); +assert_eq!(0x10", stringify!($SelfT), ".checked_shr(129), None);", $EndFeature, " +```"), + #[stable(feature = "wrapping", since = "1.7.0")] + #[inline] + pub fn checked_shr(self, rhs: u32) -> Option<Self> { + let (a, b) = self.overflowing_shr(rhs); + if b {None} else {Some(a)} + } } - /// Checked shift left. Computes `self << rhs`, returning `None` - /// if `rhs` is larger than or equal to the number of bits in `self`. - /// - /// # Examples - /// - /// Basic usage: - /// - /// ``` - /// assert_eq!(0x10u32.checked_shl(4), Some(0x100)); - /// assert_eq!(0x10u32.checked_shl(33), None); - /// ``` - #[stable(feature = "wrapping", since = "1.7.0")] - #[inline] - pub fn checked_shl(self, rhs: u32) -> Option<Self> { - let (a, b) = self.overflowing_shl(rhs); - if b {None} else {Some(a)} + doc_comment! { + concat!("Checked exponentiation. Computes `self.pow(exp)`, returning `None` if +overflow occurred. + +# Examples + +Basic usage: + +``` +#![feature(no_panic_pow)] +", $Feature, "assert_eq!(2", stringify!($SelfT), ".checked_pow(5), Some(32)); +assert_eq!(", stringify!($SelfT), "::max_value().checked_pow(2), None);", $EndFeature, " +```"), + #[unstable(feature = "no_panic_pow", issue = "48320")] + #[inline] + pub fn checked_pow(self, mut exp: u32) -> Option<Self> { + let mut base = self; + let mut acc: Self = 1; + + while exp > 1 { + if (exp & 1) == 1 { + acc = acc.checked_mul(base)?; + } + exp /= 2; + base = base.checked_mul(base)?; + } + + // Deal with the final bit of the exponent separately, since + // squaring the base afterwards is not necessary and may cause a + // needless overflow. + if exp == 1 { + acc = acc.checked_mul(base)?; + } + + Some(acc) + } } - /// Checked shift right. Computes `self >> rhs`, returning `None` - /// if `rhs` is larger than or equal to the number of bits in `self`. - /// - /// # Examples - /// - /// Basic usage: - /// - /// ``` - /// assert_eq!(0x10u32.checked_shr(4), Some(0x1)); - /// assert_eq!(0x10u32.checked_shr(33), None); - /// ``` - #[stable(feature = "wrapping", since = "1.7.0")] - #[inline] - pub fn checked_shr(self, rhs: u32) -> Option<Self> { - let (a, b) = self.overflowing_shr(rhs); - if b {None} else {Some(a)} + doc_comment! { + concat!("Saturating integer addition. Computes `self + rhs`, saturating at +the numeric bounds instead of overflowing. + +# Examples + +Basic usage: + +``` +", $Feature, "assert_eq!(100", stringify!($SelfT), ".saturating_add(1), 101); +assert_eq!(200u8.saturating_add(127), 255);", $EndFeature, " +```"), + #[stable(feature = "rust1", since = "1.0.0")] + #[inline] + pub fn saturating_add(self, rhs: Self) -> Self { + match self.checked_add(rhs) { + Some(x) => x, + None => Self::max_value(), + } + } } - /// Saturating integer addition. Computes `self + rhs`, saturating at - /// the numeric bounds instead of overflowing. - /// - /// # Examples - /// - /// Basic usage: - /// - /// ``` - /// assert_eq!(100u8.saturating_add(1), 101); - /// assert_eq!(200u8.saturating_add(127), 255); - /// ``` - #[stable(feature = "rust1", since = "1.0.0")] - #[inline] - pub fn saturating_add(self, rhs: Self) -> Self { - match self.checked_add(rhs) { - Some(x) => x, - None => Self::max_value(), + doc_comment! { + concat!("Saturating integer subtraction. Computes `self - rhs`, saturating +at the numeric bounds instead of overflowing. + +# Examples + +Basic usage: + +``` +", $Feature, "assert_eq!(100", stringify!($SelfT), ".saturating_sub(27), 73); +assert_eq!(13", stringify!($SelfT), ".saturating_sub(127), 0);", $EndFeature, " +```"), + #[stable(feature = "rust1", since = "1.0.0")] + #[inline] + pub fn saturating_sub(self, rhs: Self) -> Self { + match self.checked_sub(rhs) { + Some(x) => x, + None => Self::min_value(), + } } } - /// Saturating integer subtraction. Computes `self - rhs`, saturating - /// at the numeric bounds instead of overflowing. - /// - /// # Examples - /// - /// Basic usage: - /// - /// ``` - /// assert_eq!(100u8.saturating_sub(27), 73); - /// assert_eq!(13u8.saturating_sub(127), 0); - /// ``` - #[stable(feature = "rust1", since = "1.0.0")] - #[inline] - pub fn saturating_sub(self, rhs: Self) -> Self { - match self.checked_sub(rhs) { - Some(x) => x, - None => Self::min_value(), + doc_comment! { + concat!("Saturating integer multiplication. Computes `self * rhs`, +saturating at the numeric bounds instead of overflowing. + +# Examples + +Basic usage: + +``` +", $Feature, "use std::", stringify!($SelfT), "; + +assert_eq!(2", stringify!($SelfT), ".saturating_mul(10), 20); +assert_eq!((", stringify!($SelfT), "::MAX).saturating_mul(10), ", stringify!($SelfT), +"::MAX);", $EndFeature, " +```"), + #[stable(feature = "wrapping", since = "1.7.0")] + #[inline] + pub fn saturating_mul(self, rhs: Self) -> Self { + self.checked_mul(rhs).unwrap_or(Self::max_value()) } } - /// Saturating integer multiplication. Computes `self * rhs`, - /// saturating at the numeric bounds instead of overflowing. - /// - /// # Examples - /// - /// Basic usage: - /// - /// ``` - /// use std::u32; - /// - /// assert_eq!(100u32.saturating_mul(127), 12700); - /// assert_eq!((1u32 << 23).saturating_mul(1 << 23), u32::MAX); - /// ``` - #[stable(feature = "wrapping", since = "1.7.0")] - #[inline] - pub fn saturating_mul(self, rhs: Self) -> Self { - self.checked_mul(rhs).unwrap_or(Self::max_value()) + doc_comment! { + concat!("Saturating integer exponentiation. Computes `self.pow(exp)`, +saturating at the numeric bounds instead of overflowing. + +# Examples + +Basic usage: + +``` +#![feature(no_panic_pow)] +", $Feature, "use std::", stringify!($SelfT), "; + +assert_eq!(4", stringify!($SelfT), ".saturating_pow(3), 64); +assert_eq!(", stringify!($SelfT), "::MAX.saturating_pow(2), ", stringify!($SelfT), "::MAX);", +$EndFeature, " +```"), + #[unstable(feature = "no_panic_pow", issue = "48320")] + #[inline] + pub fn saturating_pow(self, exp: u32) -> Self { + match self.checked_pow(exp) { + Some(x) => x, + None => Self::max_value(), + } + } } - /// Wrapping (modular) addition. Computes `self + rhs`, - /// wrapping around at the boundary of the type. - /// - /// # Examples - /// - /// Basic usage: - /// - /// ``` - /// assert_eq!(200u8.wrapping_add(55), 255); - /// assert_eq!(200u8.wrapping_add(155), 99); - /// ``` - #[stable(feature = "rust1", since = "1.0.0")] - #[inline] - pub fn wrapping_add(self, rhs: Self) -> Self { - unsafe { - intrinsics::overflowing_add(self, rhs) + doc_comment! { + concat!("Wrapping (modular) addition. Computes `self + rhs`, +wrapping around at the boundary of the type. + +# Examples + +Basic usage: + +``` +", $Feature, "assert_eq!(200", stringify!($SelfT), ".wrapping_add(55), 255); +assert_eq!(200", stringify!($SelfT), ".wrapping_add(", stringify!($SelfT), "::max_value()), 199);", +$EndFeature, " +```"), + #[stable(feature = "rust1", since = "1.0.0")] + #[inline] + pub fn wrapping_add(self, rhs: Self) -> Self { + unsafe { + intrinsics::overflowing_add(self, rhs) + } } } - /// Wrapping (modular) subtraction. Computes `self - rhs`, - /// wrapping around at the boundary of the type. - /// - /// # Examples - /// - /// Basic usage: - /// - /// ``` - /// assert_eq!(100u8.wrapping_sub(100), 0); - /// assert_eq!(100u8.wrapping_sub(155), 201); - /// ``` - #[stable(feature = "rust1", since = "1.0.0")] - #[inline] - pub fn wrapping_sub(self, rhs: Self) -> Self { - unsafe { - intrinsics::overflowing_sub(self, rhs) + doc_comment! { + concat!("Wrapping (modular) subtraction. Computes `self - rhs`, +wrapping around at the boundary of the type. + +# Examples + +Basic usage: + +``` +", $Feature, "assert_eq!(100", stringify!($SelfT), ".wrapping_sub(100), 0); +assert_eq!(100", stringify!($SelfT), ".wrapping_sub(", stringify!($SelfT), "::max_value()), 101);", +$EndFeature, " +```"), + #[stable(feature = "rust1", since = "1.0.0")] + #[inline] + pub fn wrapping_sub(self, rhs: Self) -> Self { + unsafe { + intrinsics::overflowing_sub(self, rhs) + } } } @@ -1807,6 +2843,9 @@ macro_rules! uint_impl { /// /// Basic usage: /// + /// Please note that this example is shared between integer types. + /// Which explains why `u8` is used here. + /// /// ``` /// assert_eq!(10u8.wrapping_mul(12), 120); /// assert_eq!(25u8.wrapping_mul(12), 44); @@ -1819,43 +2858,92 @@ macro_rules! uint_impl { } } - /// Wrapping (modular) division. Computes `self / rhs`. - /// Wrapped division on unsigned types is just normal division. - /// There's no way wrapping could ever happen. - /// This function exists, so that all operations - /// are accounted for in the wrapping operations. - /// - /// # Examples - /// - /// Basic usage: - /// - /// ``` - /// assert_eq!(100u8.wrapping_div(10), 10); - /// ``` - #[stable(feature = "num_wrapping", since = "1.2.0")] - #[inline] - pub fn wrapping_div(self, rhs: Self) -> Self { - self / rhs + doc_comment! { + concat!("Wrapping (modular) division. Computes `self / rhs`. +Wrapped division on unsigned types is just normal division. +There's no way wrapping could ever happen. +This function exists, so that all operations +are accounted for in the wrapping operations. + +# Examples + +Basic usage: + +``` +", $Feature, "assert_eq!(100", stringify!($SelfT), ".wrapping_div(10), 10);", $EndFeature, " +```"), + #[stable(feature = "num_wrapping", since = "1.2.0")] + #[inline] + pub fn wrapping_div(self, rhs: Self) -> Self { + self / rhs + } } - /// Wrapping (modular) remainder. Computes `self % rhs`. - /// Wrapped remainder calculation on unsigned types is - /// just the regular remainder calculation. - /// There's no way wrapping could ever happen. - /// This function exists, so that all operations - /// are accounted for in the wrapping operations. - /// - /// # Examples - /// - /// Basic usage: - /// - /// ``` - /// assert_eq!(100u8.wrapping_rem(10), 0); - /// ``` - #[stable(feature = "num_wrapping", since = "1.2.0")] - #[inline] - pub fn wrapping_rem(self, rhs: Self) -> Self { - self % rhs + doc_comment! { + concat!("Wrapping Euclidean division. Computes `self.div_euc(rhs)`. +Wrapped division on unsigned types is just normal division. +There's no way wrapping could ever happen. +This function exists, so that all operations +are accounted for in the wrapping operations. + +# Examples + +Basic usage: + +``` +#![feature(euclidean_division)] +assert_eq!(100", stringify!($SelfT), ".wrapping_div_euc(10), 10); +```"), + #[unstable(feature = "euclidean_division", issue = "49048")] + #[inline] + pub fn wrapping_div_euc(self, rhs: Self) -> Self { + self / rhs + } + } + + doc_comment! { + concat!("Wrapping (modular) remainder. Computes `self % rhs`. +Wrapped remainder calculation on unsigned types is +just the regular remainder calculation. +There's no way wrapping could ever happen. +This function exists, so that all operations +are accounted for in the wrapping operations. + +# Examples + +Basic usage: + +``` +", $Feature, "assert_eq!(100", stringify!($SelfT), ".wrapping_rem(10), 0);", $EndFeature, " +```"), + #[stable(feature = "num_wrapping", since = "1.2.0")] + #[inline] + pub fn wrapping_rem(self, rhs: Self) -> Self { + self % rhs + } + } + + doc_comment! { + concat!("Wrapping Euclidean modulo. Computes `self.mod_euc(rhs)`. +Wrapped modulo calculation on unsigned types is +just the regular remainder calculation. +There's no way wrapping could ever happen. +This function exists, so that all operations +are accounted for in the wrapping operations. + +# Examples + +Basic usage: + +``` +#![feature(euclidean_division)] +assert_eq!(100", stringify!($SelfT), ".wrapping_mod_euc(10), 0); +```"), + #[unstable(feature = "euclidean_division", issue = "49048")] + #[inline] + pub fn wrapping_mod_euc(self, rhs: Self) -> Self { + self % rhs + } } /// Wrapping (modular) negation. Computes `-self`, @@ -1872,11 +2960,12 @@ macro_rules! uint_impl { /// /// Basic usage: /// + /// Please note that this example is shared between integer types. + /// Which explains why `i8` is used here. + /// /// ``` - /// assert_eq!(100u8.wrapping_neg(), 156); - /// assert_eq!(0u8.wrapping_neg(), 0); - /// assert_eq!(180u8.wrapping_neg(), 76); - /// assert_eq!(180u8.wrapping_neg(), (127 + 1) - (180u8 - (127 + 1))); + /// assert_eq!(100i8.wrapping_neg(), -100); + /// assert_eq!((-128i8).wrapping_neg(), -128); /// ``` #[stable(feature = "num_wrapping", since = "1.2.0")] #[inline] @@ -1884,110 +2973,157 @@ macro_rules! uint_impl { self.overflowing_neg().0 } - /// Panic-free bitwise shift-left; yields `self << mask(rhs)`, - /// where `mask` removes any high-order bits of `rhs` that - /// would cause the shift to exceed the bitwidth of the type. - /// - /// Note that this is *not* the same as a rotate-left; the - /// RHS of a wrapping shift-left is restricted to the range - /// of the type, rather than the bits shifted out of the LHS - /// being returned to the other end. The primitive integer - /// types all implement a `rotate_left` function, which may - /// be what you want instead. - /// - /// # Examples - /// - /// Basic usage: - /// - /// ``` - /// assert_eq!(1u8.wrapping_shl(7), 128); - /// assert_eq!(1u8.wrapping_shl(8), 1); - /// ``` - #[stable(feature = "num_wrapping", since = "1.2.0")] - #[inline] - pub fn wrapping_shl(self, rhs: u32) -> Self { - unsafe { - intrinsics::unchecked_shl(self, (rhs & ($BITS - 1)) as $SelfT) + doc_comment! { + concat!("Panic-free bitwise shift-left; yields `self << mask(rhs)`, +where `mask` removes any high-order bits of `rhs` that +would cause the shift to exceed the bitwidth of the type. + +Note that this is *not* the same as a rotate-left; the +RHS of a wrapping shift-left is restricted to the range +of the type, rather than the bits shifted out of the LHS +being returned to the other end. The primitive integer +types all implement a `rotate_left` function, which may +be what you want instead. + +# Examples + +Basic usage: + +``` +", $Feature, "assert_eq!(1", stringify!($SelfT), ".wrapping_shl(7), 128); +assert_eq!(1", stringify!($SelfT), ".wrapping_shl(128), 1);", $EndFeature, " +```"), + #[stable(feature = "num_wrapping", since = "1.2.0")] + #[inline] + pub fn wrapping_shl(self, rhs: u32) -> Self { + unsafe { + intrinsics::unchecked_shl(self, (rhs & ($BITS - 1)) as $SelfT) + } } } - /// Panic-free bitwise shift-right; yields `self >> mask(rhs)`, - /// where `mask` removes any high-order bits of `rhs` that - /// would cause the shift to exceed the bitwidth of the type. - /// - /// Note that this is *not* the same as a rotate-right; the - /// RHS of a wrapping shift-right is restricted to the range - /// of the type, rather than the bits shifted out of the LHS - /// being returned to the other end. The primitive integer - /// types all implement a `rotate_right` function, which may - /// be what you want instead. - /// - /// # Examples - /// - /// Basic usage: - /// - /// ``` - /// assert_eq!(128u8.wrapping_shr(7), 1); - /// assert_eq!(128u8.wrapping_shr(8), 128); - /// ``` - #[stable(feature = "num_wrapping", since = "1.2.0")] - #[inline] - pub fn wrapping_shr(self, rhs: u32) -> Self { - unsafe { - intrinsics::unchecked_shr(self, (rhs & ($BITS - 1)) as $SelfT) + doc_comment! { + concat!("Panic-free bitwise shift-right; yields `self >> mask(rhs)`, +where `mask` removes any high-order bits of `rhs` that +would cause the shift to exceed the bitwidth of the type. + +Note that this is *not* the same as a rotate-right; the +RHS of a wrapping shift-right is restricted to the range +of the type, rather than the bits shifted out of the LHS +being returned to the other end. The primitive integer +types all implement a `rotate_right` function, which may +be what you want instead. + +# Examples + +Basic usage: + +``` +", $Feature, "assert_eq!(128", stringify!($SelfT), ".wrapping_shr(7), 1); +assert_eq!(128", stringify!($SelfT), ".wrapping_shr(128), 128);", $EndFeature, " +```"), + #[stable(feature = "num_wrapping", since = "1.2.0")] + #[inline] + pub fn wrapping_shr(self, rhs: u32) -> Self { + unsafe { + intrinsics::unchecked_shr(self, (rhs & ($BITS - 1)) as $SelfT) + } } } - /// Calculates `self` + `rhs` - /// - /// Returns a tuple of the addition along with a boolean indicating - /// whether an arithmetic overflow would occur. If an overflow would - /// have occurred then the wrapped value is returned. - /// - /// # Examples - /// - /// Basic usage - /// - /// ``` - /// use std::u32; - /// - /// assert_eq!(5u32.overflowing_add(2), (7, false)); - /// assert_eq!(u32::MAX.overflowing_add(1), (0, true)); - /// ``` - #[inline] - #[stable(feature = "wrapping", since = "1.7.0")] - pub fn overflowing_add(self, rhs: Self) -> (Self, bool) { - let (a, b) = unsafe { - intrinsics::add_with_overflow(self as $ActualT, - rhs as $ActualT) - }; - (a as Self, b) + doc_comment! { + concat!("Wrapping (modular) exponentiation. Computes `self.pow(exp)`, +wrapping around at the boundary of the type. + +# Examples + +Basic usage: + +``` +#![feature(no_panic_pow)] +", $Feature, "assert_eq!(3", stringify!($SelfT), ".wrapping_pow(5), 243); +assert_eq!(3u8.wrapping_pow(6), 217);", $EndFeature, " +```"), + #[unstable(feature = "no_panic_pow", issue = "48320")] + #[inline] + pub fn wrapping_pow(self, mut exp: u32) -> Self { + let mut base = self; + let mut acc: Self = 1; + + while exp > 1 { + if (exp & 1) == 1 { + acc = acc.wrapping_mul(base); + } + exp /= 2; + base = base.wrapping_mul(base); + } + + // Deal with the final bit of the exponent separately, since + // squaring the base afterwards is not necessary and may cause a + // needless overflow. + if exp == 1 { + acc = acc.wrapping_mul(base); + } + + acc + } } - /// Calculates `self` - `rhs` - /// - /// Returns a tuple of the subtraction along with a boolean indicating - /// whether an arithmetic overflow would occur. If an overflow would - /// have occurred then the wrapped value is returned. - /// - /// # Examples - /// - /// Basic usage - /// - /// ``` - /// use std::u32; - /// - /// assert_eq!(5u32.overflowing_sub(2), (3, false)); - /// assert_eq!(0u32.overflowing_sub(1), (u32::MAX, true)); - /// ``` - #[inline] - #[stable(feature = "wrapping", since = "1.7.0")] - pub fn overflowing_sub(self, rhs: Self) -> (Self, bool) { - let (a, b) = unsafe { - intrinsics::sub_with_overflow(self as $ActualT, - rhs as $ActualT) - }; - (a as Self, b) + doc_comment! { + concat!("Calculates `self` + `rhs` + +Returns a tuple of the addition along with a boolean indicating +whether an arithmetic overflow would occur. If an overflow would +have occurred then the wrapped value is returned. + +# Examples + +Basic usage + +``` +", $Feature, "use std::", stringify!($SelfT), "; + +assert_eq!(5", stringify!($SelfT), ".overflowing_add(2), (7, false)); +assert_eq!(", stringify!($SelfT), "::MAX.overflowing_add(1), (0, true));", $EndFeature, " +```"), + #[inline] + #[stable(feature = "wrapping", since = "1.7.0")] + pub fn overflowing_add(self, rhs: Self) -> (Self, bool) { + let (a, b) = unsafe { + intrinsics::add_with_overflow(self as $ActualT, + rhs as $ActualT) + }; + (a as Self, b) + } + } + + doc_comment! { + concat!("Calculates `self` - `rhs` + +Returns a tuple of the subtraction along with a boolean indicating +whether an arithmetic overflow would occur. If an overflow would +have occurred then the wrapped value is returned. + +# Examples + +Basic usage + +``` +", $Feature, "use std::", stringify!($SelfT), "; + +assert_eq!(5", stringify!($SelfT), ".overflowing_sub(2), (3, false)); +assert_eq!(0", stringify!($SelfT), ".overflowing_sub(1), (", stringify!($SelfT), "::MAX, true));", +$EndFeature, " +```"), + #[inline] + #[stable(feature = "wrapping", since = "1.7.0")] + pub fn overflowing_sub(self, rhs: Self) -> (Self, bool) { + let (a, b) = unsafe { + intrinsics::sub_with_overflow(self as $ActualT, + rhs as $ActualT) + }; + (a as Self, b) + } } /// Calculates the multiplication of `self` and `rhs`. @@ -1998,7 +3134,10 @@ macro_rules! uint_impl { /// /// # Examples /// - /// Basic usage + /// Basic usage: + /// + /// Please note that this example is shared between integer types. + /// Which explains why `u32` is used here. /// /// ``` /// assert_eq!(5u32.overflowing_mul(2), (10, false)); @@ -2014,129 +3153,243 @@ macro_rules! uint_impl { (a as Self, b) } - /// Calculates the divisor when `self` is divided by `rhs`. - /// - /// Returns a tuple of the divisor along with a boolean indicating - /// whether an arithmetic overflow would occur. Note that for unsigned - /// integers overflow never occurs, so the second value is always - /// `false`. - /// - /// # Panics - /// - /// This function will panic if `rhs` is 0. - /// - /// # Examples - /// - /// Basic usage - /// - /// ``` - /// assert_eq!(5u32.overflowing_div(2), (2, false)); - /// ``` - #[inline] - #[stable(feature = "wrapping", since = "1.7.0")] - pub fn overflowing_div(self, rhs: Self) -> (Self, bool) { - (self / rhs, false) + doc_comment! { + concat!("Calculates the divisor when `self` is divided by `rhs`. + +Returns a tuple of the divisor along with a boolean indicating +whether an arithmetic overflow would occur. Note that for unsigned +integers overflow never occurs, so the second value is always +`false`. + +# Panics + +This function will panic if `rhs` is 0. + +# Examples + +Basic usage + +``` +", $Feature, "assert_eq!(5", stringify!($SelfT), ".overflowing_div(2), (2, false));", $EndFeature, " +```"), + #[inline] + #[stable(feature = "wrapping", since = "1.7.0")] + pub fn overflowing_div(self, rhs: Self) -> (Self, bool) { + (self / rhs, false) + } } - /// Calculates the remainder when `self` is divided by `rhs`. - /// - /// Returns a tuple of the remainder after dividing along with a boolean - /// indicating whether an arithmetic overflow would occur. Note that for - /// unsigned integers overflow never occurs, so the second value is - /// always `false`. - /// - /// # Panics - /// - /// This function will panic if `rhs` is 0. - /// - /// # Examples - /// - /// Basic usage - /// - /// ``` - /// assert_eq!(5u32.overflowing_rem(2), (1, false)); - /// ``` - #[inline] - #[stable(feature = "wrapping", since = "1.7.0")] - pub fn overflowing_rem(self, rhs: Self) -> (Self, bool) { - (self % rhs, false) + doc_comment! { + concat!("Calculates the quotient of Euclidean division `self.div_euc(rhs)`. + +Returns a tuple of the divisor along with a boolean indicating +whether an arithmetic overflow would occur. Note that for unsigned +integers overflow never occurs, so the second value is always +`false`. + +# Panics + +This function will panic if `rhs` is 0. + +# Examples + +Basic usage + +``` +#![feature(euclidean_division)] +assert_eq!(5", stringify!($SelfT), ".overflowing_div_euc(2), (2, false)); +```"), + #[inline] + #[unstable(feature = "euclidean_division", issue = "49048")] + pub fn overflowing_div_euc(self, rhs: Self) -> (Self, bool) { + (self / rhs, false) + } } - /// Negates self in an overflowing fashion. - /// - /// Returns `!self + 1` using wrapping operations to return the value - /// that represents the negation of this unsigned value. Note that for - /// positive unsigned values overflow always occurs, but negating 0 does - /// not overflow. - /// - /// # Examples - /// - /// Basic usage - /// - /// ``` - /// assert_eq!(0u32.overflowing_neg(), (0, false)); - /// assert_eq!(2u32.overflowing_neg(), (-2i32 as u32, true)); - /// ``` - #[inline] - #[stable(feature = "wrapping", since = "1.7.0")] - pub fn overflowing_neg(self) -> (Self, bool) { - ((!self).wrapping_add(1), self != 0) + doc_comment! { + concat!("Calculates the remainder when `self` is divided by `rhs`. + +Returns a tuple of the remainder after dividing along with a boolean +indicating whether an arithmetic overflow would occur. Note that for +unsigned integers overflow never occurs, so the second value is +always `false`. + +# Panics + +This function will panic if `rhs` is 0. + +# Examples + +Basic usage + +``` +", $Feature, "assert_eq!(5", stringify!($SelfT), ".overflowing_rem(2), (1, false));", $EndFeature, " +```"), + #[inline] + #[stable(feature = "wrapping", since = "1.7.0")] + pub fn overflowing_rem(self, rhs: Self) -> (Self, bool) { + (self % rhs, false) + } } - /// Shifts self left by `rhs` bits. - /// - /// Returns a tuple of the shifted version of self along with a boolean - /// indicating whether the shift value was larger than or equal to the - /// number of bits. If the shift value is too large, then value is - /// masked (N-1) where N is the number of bits, and this value is then - /// used to perform the shift. - /// - /// # Examples - /// - /// Basic usage - /// - /// ``` - /// assert_eq!(0x10u32.overflowing_shl(4), (0x100, false)); - /// assert_eq!(0x10u32.overflowing_shl(36), (0x100, true)); - /// ``` - #[inline] - #[stable(feature = "wrapping", since = "1.7.0")] - pub fn overflowing_shl(self, rhs: u32) -> (Self, bool) { - (self.wrapping_shl(rhs), (rhs > ($BITS - 1))) + doc_comment! { + concat!("Calculates the remainder `self.mod_euc(rhs)` by Euclidean division. + +Returns a tuple of the modulo after dividing along with a boolean +indicating whether an arithmetic overflow would occur. Note that for +unsigned integers overflow never occurs, so the second value is +always `false`. + +# Panics + +This function will panic if `rhs` is 0. + +# Examples + +Basic usage + +``` +#![feature(euclidean_division)] +assert_eq!(5", stringify!($SelfT), ".overflowing_mod_euc(2), (1, false)); +```"), + #[inline] + #[unstable(feature = "euclidean_division", issue = "49048")] + pub fn overflowing_mod_euc(self, rhs: Self) -> (Self, bool) { + (self % rhs, false) + } } - /// Shifts self right by `rhs` bits. - /// - /// Returns a tuple of the shifted version of self along with a boolean - /// indicating whether the shift value was larger than or equal to the - /// number of bits. If the shift value is too large, then value is - /// masked (N-1) where N is the number of bits, and this value is then - /// used to perform the shift. - /// - /// # Examples - /// - /// Basic usage - /// - /// ``` - /// assert_eq!(0x10u32.overflowing_shr(4), (0x1, false)); - /// assert_eq!(0x10u32.overflowing_shr(36), (0x1, true)); - /// ``` - #[inline] - #[stable(feature = "wrapping", since = "1.7.0")] - pub fn overflowing_shr(self, rhs: u32) -> (Self, bool) { - (self.wrapping_shr(rhs), (rhs > ($BITS - 1))) + doc_comment! { + concat!("Negates self in an overflowing fashion. + +Returns `!self + 1` using wrapping operations to return the value +that represents the negation of this unsigned value. Note that for +positive unsigned values overflow always occurs, but negating 0 does +not overflow. +# Examples + +Basic usage + +``` +", $Feature, "assert_eq!(0", stringify!($SelfT), ".overflowing_neg(), (0, false)); +assert_eq!(2", stringify!($SelfT), ".overflowing_neg(), (-2i32 as ", stringify!($SelfT), +", true));", $EndFeature, " +```"), + #[inline] + #[stable(feature = "wrapping", since = "1.7.0")] + pub fn overflowing_neg(self) -> (Self, bool) { + ((!self).wrapping_add(1), self != 0) + } } - /// Raises self to the power of `exp`, using exponentiation by squaring. - /// - /// # Examples - /// - /// Basic usage: - /// - /// ``` - /// assert_eq!(2u32.pow(4), 16); - /// ``` + doc_comment! { + concat!("Shifts self left by `rhs` bits. + +Returns a tuple of the shifted version of self along with a boolean +indicating whether the shift value was larger than or equal to the +number of bits. If the shift value is too large, then value is +masked (N-1) where N is the number of bits, and this value is then +used to perform the shift. + +# Examples + +Basic usage + +``` +", $Feature, "assert_eq!(0x1", stringify!($SelfT), ".overflowing_shl(4), (0x10, false)); +assert_eq!(0x1", stringify!($SelfT), ".overflowing_shl(132), (0x10, true));", $EndFeature, " +```"), + #[inline] + #[stable(feature = "wrapping", since = "1.7.0")] + pub fn overflowing_shl(self, rhs: u32) -> (Self, bool) { + (self.wrapping_shl(rhs), (rhs > ($BITS - 1))) + } + } + + doc_comment! { + concat!("Shifts self right by `rhs` bits. + +Returns a tuple of the shifted version of self along with a boolean +indicating whether the shift value was larger than or equal to the +number of bits. If the shift value is too large, then value is +masked (N-1) where N is the number of bits, and this value is then +used to perform the shift. + +# Examples + +Basic usage + +``` +", $Feature, "assert_eq!(0x10", stringify!($SelfT), ".overflowing_shr(4), (0x1, false)); +assert_eq!(0x10", stringify!($SelfT), ".overflowing_shr(132), (0x1, true));", $EndFeature, " +```"), + #[inline] + #[stable(feature = "wrapping", since = "1.7.0")] + pub fn overflowing_shr(self, rhs: u32) -> (Self, bool) { + (self.wrapping_shr(rhs), (rhs > ($BITS - 1))) + } + } + + doc_comment! { + concat!("Raises self to the power of `exp`, using exponentiation by squaring. + +Returns a tuple of the exponentiation along with a bool indicating +whether an overflow happened. + +# Examples + +Basic usage: + +``` +#![feature(no_panic_pow)] +", $Feature, "assert_eq!(3", stringify!($SelfT), ".overflowing_pow(5), (243, false)); +assert_eq!(3u8.overflowing_pow(6), (217, true));", $EndFeature, " +```"), + #[unstable(feature = "no_panic_pow", issue = "48320")] + #[inline] + pub fn overflowing_pow(self, mut exp: u32) -> (Self, bool) { + let mut base = self; + let mut acc: Self = 1; + let mut overflown = false; + // Scratch space for storing results of overflowing_mul. + let mut r; + + while exp > 1 { + if (exp & 1) == 1 { + r = acc.overflowing_mul(base); + acc = r.0; + overflown |= r.1; + } + exp /= 2; + r = base.overflowing_mul(base); + base = r.0; + overflown |= r.1; + } + + // Deal with the final bit of the exponent separately, since + // squaring the base afterwards is not necessary and may cause a + // needless overflow. + if exp == 1 { + r = acc.overflowing_mul(base); + acc = r.0; + overflown |= r.1; + } + + (acc, overflown) + } + } + + doc_comment! { + concat!("Raises self to the power of `exp`, using exponentiation by squaring. + +# Examples + +Basic usage: + +``` +", $Feature, "assert_eq!(2", stringify!($SelfT), ".pow(5), 32);", $EndFeature, " +```"), #[stable(feature = "rust1", since = "1.0.0")] #[inline] #[rustc_inherit_overflow_checks] @@ -2161,21 +3414,67 @@ macro_rules! uint_impl { acc } + } - /// Returns `true` if and only if `self == 2^k` for some `k`. - /// - /// # Examples - /// - /// Basic usage: - /// - /// ``` - /// assert!(16u8.is_power_of_two()); - /// assert!(!10u8.is_power_of_two()); - /// ``` - #[stable(feature = "rust1", since = "1.0.0")] - #[inline] - pub fn is_power_of_two(self) -> bool { - (self.wrapping_sub(1)) & self == 0 && !(self == 0) + doc_comment! { + concat!("Performs Euclidean division. + +For unsigned types, this is just the same as `self / rhs`. + +# Examples + +Basic usage: + +``` +#![feature(euclidean_division)] +assert_eq!(7", stringify!($SelfT), ".div_euc(4), 1); // or any other integer type +```"), + #[unstable(feature = "euclidean_division", issue = "49048")] + #[inline] + #[rustc_inherit_overflow_checks] + pub fn div_euc(self, rhs: Self) -> Self { + self / rhs + } + } + + + doc_comment! { + concat!("Calculates the remainder `self mod rhs` by Euclidean division. + +For unsigned types, this is just the same as `self % rhs`. + +# Examples + +Basic usage: + +``` +#![feature(euclidean_division)] +assert_eq!(7", stringify!($SelfT), ".mod_euc(4), 3); // or any other integer type +```"), + #[unstable(feature = "euclidean_division", issue = "49048")] + #[inline] + #[rustc_inherit_overflow_checks] + pub fn mod_euc(self, rhs: Self) -> Self { + self % rhs + } + } + + doc_comment! { + concat!("Returns `true` if and only if `self == 2^k` for some `k`. + +# Examples + +Basic usage: + +``` +", $Feature, "assert!(16", stringify!($SelfT), ".is_power_of_two()); +assert!(!10", stringify!($SelfT), ".is_power_of_two());", $EndFeature, " +```"), + #[stable(feature = "rust1", since = "1.0.0")] + #[inline] + pub fn is_power_of_two(self) -> bool { + (self.wrapping_sub(1)) & self == 0 && !(self == 0) + } } // Returns one less than next power of two. @@ -2200,50 +3499,196 @@ macro_rules! uint_impl { <$SelfT>::max_value() >> z } - /// Returns the smallest power of two greater than or equal to `self`. + doc_comment! { + concat!("Returns the smallest power of two greater than or equal to `self`. + +When return value overflows (i.e. `self > (1 << (N-1))` for type +`uN`), it panics in debug mode and return value is wrapped to 0 in +release mode (the only situation in which method can return 0). + +# Examples + +Basic usage: + +``` +", $Feature, "assert_eq!(2", stringify!($SelfT), ".next_power_of_two(), 2); +assert_eq!(3", stringify!($SelfT), ".next_power_of_two(), 4);", $EndFeature, " +```"), + #[stable(feature = "rust1", since = "1.0.0")] + #[inline] + pub fn next_power_of_two(self) -> Self { + // Call the trait to get overflow checks + ops::Add::add(self.one_less_than_next_power_of_two(), 1) + } + } + + doc_comment! { + concat!("Returns the smallest power of two greater than or equal to `n`. If +the next power of two is greater than the type's maximum value, +`None` is returned, otherwise the power of two is wrapped in `Some`. + +# Examples + +Basic usage: + +``` +", $Feature, "assert_eq!(2", stringify!($SelfT), +".checked_next_power_of_two(), Some(2)); +assert_eq!(3", stringify!($SelfT), ".checked_next_power_of_two(), Some(4)); +assert_eq!(", stringify!($SelfT), "::max_value().checked_next_power_of_two(), None);", +$EndFeature, " +```"), + #[stable(feature = "rust1", since = "1.0.0")] + pub fn checked_next_power_of_two(self) -> Option<Self> { + self.one_less_than_next_power_of_two().checked_add(1) + } + } + + doc_comment! { + concat!("Returns the smallest power of two greater than or equal to `n`. If +the next power of two is greater than the type's maximum value, +the return value is wrapped to `0`. + +# Examples + +Basic usage: + +``` +#![feature(wrapping_next_power_of_two)] +", $Feature, " +assert_eq!(2", stringify!($SelfT), ".wrapping_next_power_of_two(), 2); +assert_eq!(3", stringify!($SelfT), ".wrapping_next_power_of_two(), 4); +assert_eq!(", stringify!($SelfT), "::max_value().wrapping_next_power_of_two(), 0);", +$EndFeature, " +```"), + #[unstable(feature = "wrapping_next_power_of_two", issue = "32463", + reason = "needs decision on wrapping behaviour")] + pub fn wrapping_next_power_of_two(self) -> Self { + self.one_less_than_next_power_of_two().wrapping_add(1) + } + } + + /// Return the memory representation of this integer as a byte array in + /// big-endian (network) byte order. /// - /// When return value overflows (i.e. `self > (1 << (N-1))` for type - /// `uN`), it panics in debug mode and return value is wrapped to 0 in - /// release mode (the only situation in which method can return 0). + /// # Examples + /// + /// ``` + /// #![feature(int_to_from_bytes)] + /// + /// let bytes = 0x12_34_56_78_i32.to_be_bytes(); + /// assert_eq!(bytes, [0x12, 0x34, 0x56, 0x78]); + /// ``` + #[unstable(feature = "int_to_from_bytes", issue = "52963")] + #[inline] + pub fn to_be_bytes(self) -> [u8; mem::size_of::<Self>()] { + self.to_be().to_ne_bytes() + } + + /// Return the memory representation of this integer as a byte array in + /// little-endian byte order. /// /// # Examples /// - /// Basic usage: + /// ``` + /// #![feature(int_to_from_bytes)] + /// + /// let bytes = 0x12_34_56_78_i32.to_le_bytes(); + /// assert_eq!(bytes, [0x78, 0x56, 0x34, 0x12]); + /// ``` + #[unstable(feature = "int_to_from_bytes", issue = "52963")] + #[inline] + pub fn to_le_bytes(self) -> [u8; mem::size_of::<Self>()] { + self.to_le().to_ne_bytes() + } + + /// Return the memory representation of this integer as a byte array in + /// native byte order. + /// + /// As the target platform's native endianness is used, portable code + /// should use [`to_be_bytes`] or [`to_le_bytes`], as appropriate, + /// instead. + /// + /// [`to_be_bytes`]: #method.to_be_bytes + /// [`to_le_bytes`]: #method.to_le_bytes + /// + /// # Examples /// /// ``` - /// assert_eq!(2u8.next_power_of_two(), 2); - /// assert_eq!(3u8.next_power_of_two(), 4); + /// #![feature(int_to_from_bytes)] + /// + /// let bytes = i32::min_value().to_be().to_ne_bytes(); + /// assert_eq!(bytes, [0x80, 0, 0, 0]); /// ``` - #[stable(feature = "rust1", since = "1.0.0")] + #[unstable(feature = "int_to_from_bytes", issue = "52963")] #[inline] - pub fn next_power_of_two(self) -> Self { - // Call the trait to get overflow checks - ops::Add::add(self.one_less_than_next_power_of_two(), 1) + pub fn to_ne_bytes(self) -> [u8; mem::size_of::<Self>()] { + unsafe { mem::transmute(self) } } - /// Returns the smallest power of two greater than or equal to `n`. If - /// the next power of two is greater than the type's maximum value, - /// `None` is returned, otherwise the power of two is wrapped in `Some`. + /// Create an integer value from its representation as a byte array in + /// big endian. /// /// # Examples /// - /// Basic usage: + /// ``` + /// #![feature(int_to_from_bytes)] /// + /// let int = i32::from_be_bytes([0x12, 0x34, 0x56, 0x78]); + /// assert_eq!(int, 0x12_34_56_78); /// ``` - /// assert_eq!(2u8.checked_next_power_of_two(), Some(2)); - /// assert_eq!(3u8.checked_next_power_of_two(), Some(4)); - /// assert_eq!(200u8.checked_next_power_of_two(), None); + #[unstable(feature = "int_to_from_bytes", issue = "52963")] + #[inline] + pub fn from_be_bytes(bytes: [u8; mem::size_of::<Self>()]) -> Self { + Self::from_be(Self::from_ne_bytes(bytes)) + } + + /// Create an integer value from its representation as a byte array in + /// little endian. + /// + /// # Examples + /// /// ``` - #[stable(feature = "rust1", since = "1.0.0")] - pub fn checked_next_power_of_two(self) -> Option<Self> { - self.one_less_than_next_power_of_two().checked_add(1) + /// #![feature(int_to_from_bytes)] + /// + /// let int = i32::from_le_bytes([0x12, 0x34, 0x56, 0x78]); + /// assert_eq!(int, 0x78_56_34_12); + /// ``` + #[unstable(feature = "int_to_from_bytes", issue = "52963")] + #[inline] + pub fn from_le_bytes(bytes: [u8; mem::size_of::<Self>()]) -> Self { + Self::from_le(Self::from_ne_bytes(bytes)) + } + + /// Create an integer value from its memory representation as a byte + /// array in native endianness. + /// + /// As the target platform's native endianness is used, portable code + /// likely wants to use [`from_be_bytes`] or [`from_le_bytes`], as + /// appropriate instead. + /// + /// [`from_be_bytes`]: #method.from_be_bytes + /// [`from_le_bytes`]: #method.from_le_bytes + /// + /// # Examples + /// + /// ``` + /// #![feature(int_to_from_bytes)] + /// + /// let int = i32::from_be(i32::from_ne_bytes([0x80, 0, 0, 0])); + /// assert_eq!(int, i32::min_value()); + /// ``` + #[unstable(feature = "int_to_from_bytes", issue = "52963")] + #[inline] + pub fn from_ne_bytes(bytes: [u8; mem::size_of::<Self>()]) -> Self { + unsafe { mem::transmute(bytes) } } } } #[lang = "u8"] impl u8 { - uint_impl! { u8, u8, 8 } + uint_impl! { u8, u8, 8, 255, "", "", 2, "0x82", "0xa", "0x12", "0x12", "0x48" } /// Checks if the value is within the ASCII range. @@ -2383,8 +3828,6 @@ impl u8 { /// # Examples /// /// ``` - /// #![feature(ascii_ctype)] - /// /// let uppercase_a = b'A'; /// let uppercase_g = b'G'; /// let a = b'a'; @@ -2421,8 +3864,6 @@ impl u8 { /// # Examples /// /// ``` - /// #![feature(ascii_ctype)] - /// /// let uppercase_a = b'A'; /// let uppercase_g = b'G'; /// let a = b'a'; @@ -2459,8 +3900,6 @@ impl u8 { /// # Examples /// /// ``` - /// #![feature(ascii_ctype)] - /// /// let uppercase_a = b'A'; /// let uppercase_g = b'G'; /// let a = b'a'; @@ -2500,8 +3939,6 @@ impl u8 { /// # Examples /// /// ``` - /// #![feature(ascii_ctype)] - /// /// let uppercase_a = b'A'; /// let uppercase_g = b'G'; /// let a = b'a'; @@ -2538,8 +3975,6 @@ impl u8 { /// # Examples /// /// ``` - /// #![feature(ascii_ctype)] - /// /// let uppercase_a = b'A'; /// let uppercase_g = b'G'; /// let a = b'a'; @@ -2579,8 +4014,6 @@ impl u8 { /// # Examples /// /// ``` - /// #![feature(ascii_ctype)] - /// /// let uppercase_a = b'A'; /// let uppercase_g = b'G'; /// let a = b'a'; @@ -2621,8 +4054,6 @@ impl u8 { /// # Examples /// /// ``` - /// #![feature(ascii_ctype)] - /// /// let uppercase_a = b'A'; /// let uppercase_g = b'G'; /// let a = b'a'; @@ -2654,13 +4085,11 @@ impl u8 { } /// Checks if the value is an ASCII graphic character: - /// U+0021 '@' ... U+007E '~'. + /// U+0021 '!' ... U+007E '~'. /// /// # Examples /// /// ``` - /// #![feature(ascii_ctype)] - /// /// let uppercase_a = b'A'; /// let uppercase_g = b'G'; /// let a = b'a'; @@ -2714,8 +4143,6 @@ impl u8 { /// # Examples /// /// ``` - /// #![feature(ascii_ctype)] - /// /// let uppercase_a = b'A'; /// let uppercase_g = b'G'; /// let a = b'a'; @@ -2754,8 +4181,6 @@ impl u8 { /// # Examples /// /// ``` - /// #![feature(ascii_ctype)] - /// /// let uppercase_a = b'A'; /// let uppercase_g = b'G'; /// let a = b'a'; @@ -2789,39 +4214,45 @@ impl u8 { #[lang = "u16"] impl u16 { - uint_impl! { u16, u16, 16 } + uint_impl! { u16, u16, 16, 65535, "", "", 4, "0xa003", "0x3a", "0x1234", "0x3412", "0x2c48" } } #[lang = "u32"] impl u32 { - uint_impl! { u32, u32, 32 } + uint_impl! { u32, u32, 32, 4294967295, "", "", 8, "0x10000b3", "0xb301", "0x12345678", + "0x78563412", "0x1e6a2c48" } } #[lang = "u64"] impl u64 { - uint_impl! { u64, u64, 64 } + uint_impl! { u64, u64, 64, 18446744073709551615, "", "", 12, "0xaa00000000006e1", "0x6e10aa", + "0x1234567890123456", "0x5634129078563412", "0x6a2c48091e6a2c48" } } #[lang = "u128"] impl u128 { - uint_impl! { u128, u128, 128 } + uint_impl! { u128, u128, 128, 340282366920938463463374607431768211455, "", "", 16, + "0x13f40000000000000000000000004f76", "0x4f7613f4", "0x12345678901234567890123456789012", + "0x12907856341290785634129078563412", "0x48091e6a2c48091e6a2c48091e6a2c48" } } #[cfg(target_pointer_width = "16")] #[lang = "usize"] impl usize { - uint_impl! { usize, u16, 16 } + uint_impl! { usize, u16, 16, 65536, "", "", 4, "0xa003", "0x3a", "0x1234", "0x3412", "0x2c48" } } #[cfg(target_pointer_width = "32")] #[lang = "usize"] impl usize { - uint_impl! { usize, u32, 32 } + uint_impl! { usize, u32, 32, 4294967295, "", "", 8, "0x10000b3", "0xb301", "0x12345678", + "0x78563412", "0x1e6a2c48" } } #[cfg(target_pointer_width = "64")] #[lang = "usize"] impl usize { - uint_impl! { usize, u64, 64 } + uint_impl! { usize, u64, 64, 18446744073709551615, "", "", 12, "0xaa00000000006e1", "0x6e10aa", + "0x1234567890123456", "0x5634129078563412", "0x6a2c48091e6a2c48" } } /// A classification of floating point numbers. @@ -2874,75 +4305,6 @@ pub enum FpCategory { Normal, } -/// A built-in floating point number. -#[doc(hidden)] -#[unstable(feature = "core_float", - reason = "stable interface is via `impl f{32,64}` in later crates", - issue = "32110")] -pub trait Float: Sized { - /// Returns `true` if this value is NaN and false otherwise. - #[stable(feature = "core", since = "1.6.0")] - fn is_nan(self) -> bool; - /// Returns `true` if this value is positive infinity or negative infinity and - /// false otherwise. - #[stable(feature = "core", since = "1.6.0")] - fn is_infinite(self) -> bool; - /// Returns `true` if this number is neither infinite nor NaN. - #[stable(feature = "core", since = "1.6.0")] - fn is_finite(self) -> bool; - /// Returns `true` if this number is neither zero, infinite, denormal, or NaN. - #[stable(feature = "core", since = "1.6.0")] - fn is_normal(self) -> bool; - /// Returns the category that this number falls into. - #[stable(feature = "core", since = "1.6.0")] - fn classify(self) -> FpCategory; - - /// Computes the absolute value of `self`. Returns `Float::nan()` if the - /// number is `Float::nan()`. - #[stable(feature = "core", since = "1.6.0")] - fn abs(self) -> Self; - /// Returns a number that represents the sign of `self`. - /// - /// - `1.0` if the number is positive, `+0.0` or `Float::infinity()` - /// - `-1.0` if the number is negative, `-0.0` or `Float::neg_infinity()` - /// - `Float::nan()` if the number is `Float::nan()` - #[stable(feature = "core", since = "1.6.0")] - fn signum(self) -> Self; - - /// Returns `true` if `self` is positive, including `+0.0` and - /// `Float::infinity()`. - #[stable(feature = "core", since = "1.6.0")] - fn is_sign_positive(self) -> bool; - /// Returns `true` if `self` is negative, including `-0.0` and - /// `Float::neg_infinity()`. - #[stable(feature = "core", since = "1.6.0")] - fn is_sign_negative(self) -> bool; - - /// Take the reciprocal (inverse) of a number, `1/x`. - #[stable(feature = "core", since = "1.6.0")] - fn recip(self) -> Self; - - /// Raise a number to an integer power. - /// - /// Using this function is generally faster than using `powf` - #[stable(feature = "core", since = "1.6.0")] - fn powi(self, n: i32) -> Self; - - /// Convert radians to degrees. - #[stable(feature = "deg_rad_conversions", since="1.7.0")] - fn to_degrees(self) -> Self; - /// Convert degrees to radians. - #[stable(feature = "deg_rad_conversions", since="1.7.0")] - fn to_radians(self) -> Self; - - /// Returns the maximum of the two numbers. - #[stable(feature = "core_float_min_max", since="1.20.0")] - fn max(self, other: Self) -> Self; - /// Returns the minimum of the two numbers. - #[stable(feature = "core_float_min_max", since="1.20.0")] - fn min(self, other: Self) -> Self; -} - macro_rules! from_str_radix_int_impl { ($($t:ty)*) => {$( #[stable(feature = "rust1", since = "1.0.0")] @@ -2980,10 +4342,9 @@ impl fmt::Display for TryFromIntError { } #[unstable(feature = "try_from", issue = "33417")] -impl From<Infallible> for TryFromIntError { - fn from(infallible: Infallible) -> TryFromIntError { - match infallible { - } +impl From<!> for TryFromIntError { + fn from(never: !) -> TryFromIntError { + never } } @@ -2992,7 +4353,7 @@ macro_rules! try_from_unbounded { ($source:ty, $($target:ty),*) => {$( #[unstable(feature = "try_from", issue = "33417")] impl TryFrom<$source> for $target { - type Error = Infallible; + type Error = TryFromIntError; #[inline] fn try_from(value: $source) -> Result<Self, Self::Error> { @@ -3103,7 +4464,7 @@ try_from_lower_bounded!(isize, usize); #[cfg(target_pointer_width = "16")] mod ptr_try_from_impls { use super::TryFromIntError; - use convert::{Infallible, TryFrom}; + use convert::TryFrom; try_from_upper_bounded!(usize, u8); try_from_unbounded!(usize, u16, u32, u64, u128); @@ -3115,21 +4476,18 @@ mod ptr_try_from_impls { try_from_both_bounded!(isize, i8); try_from_unbounded!(isize, i16, i32, i64, i128); - rev!(try_from_unbounded, usize, u16); rev!(try_from_upper_bounded, usize, u32, u64, u128); rev!(try_from_lower_bounded, usize, i8, i16); rev!(try_from_both_bounded, usize, i32, i64, i128); - rev!(try_from_unbounded, isize, u8); rev!(try_from_upper_bounded, isize, u16, u32, u64, u128); - rev!(try_from_unbounded, isize, i16); rev!(try_from_both_bounded, isize, i32, i64, i128); } #[cfg(target_pointer_width = "32")] mod ptr_try_from_impls { use super::TryFromIntError; - use convert::{Infallible, TryFrom}; + use convert::TryFrom; try_from_upper_bounded!(usize, u8, u16); try_from_unbounded!(usize, u32, u64, u128); @@ -3141,21 +4499,21 @@ mod ptr_try_from_impls { try_from_both_bounded!(isize, i8, i16); try_from_unbounded!(isize, i32, i64, i128); - rev!(try_from_unbounded, usize, u16, u32); + rev!(try_from_unbounded, usize, u32); rev!(try_from_upper_bounded, usize, u64, u128); rev!(try_from_lower_bounded, usize, i8, i16, i32); rev!(try_from_both_bounded, usize, i64, i128); - rev!(try_from_unbounded, isize, u8, u16); + rev!(try_from_unbounded, isize, u16); rev!(try_from_upper_bounded, isize, u32, u64, u128); - rev!(try_from_unbounded, isize, i16, i32); + rev!(try_from_unbounded, isize, i32); rev!(try_from_both_bounded, isize, i64, i128); } #[cfg(target_pointer_width = "64")] mod ptr_try_from_impls { use super::TryFromIntError; - use convert::{Infallible, TryFrom}; + use convert::TryFrom; try_from_upper_bounded!(usize, u8, u16, u32); try_from_unbounded!(usize, u64, u128); @@ -3167,14 +4525,14 @@ mod ptr_try_from_impls { try_from_both_bounded!(isize, i8, i16, i32); try_from_unbounded!(isize, i64, i128); - rev!(try_from_unbounded, usize, u16, u32, u64); + rev!(try_from_unbounded, usize, u32, u64); rev!(try_from_upper_bounded, usize, u128); rev!(try_from_lower_bounded, usize, i8, i16, i32, i64); rev!(try_from_both_bounded, usize, i128); - rev!(try_from_unbounded, isize, u8, u16, u32); + rev!(try_from_unbounded, isize, u16, u32); rev!(try_from_upper_bounded, isize, u64, u128); - rev!(try_from_unbounded, isize, i16, i32, i64); + rev!(try_from_unbounded, isize, i32, i64); rev!(try_from_both_bounded, isize, i128); } @@ -3284,6 +4642,13 @@ fn from_str_radix<T: FromStrRadixHelper>(src: &str, radix: u32) -> Result<T, Par /// This error is used as the error type for the `from_str_radix()` functions /// on the primitive integer types, such as [`i8::from_str_radix`]. /// +/// # Potential causes +/// +/// Among other causes, `ParseIntError` can be thrown because of leading or trailing whitespace +/// in the string e.g. when it is obtained from the standard input. +/// Using the [`str.trim()`] method ensures that no whitespace remains before parsing. +/// +/// [`str.trim()`]: ../../std/primitive.str.html#method.trim /// [`i8::from_str_radix`]: ../../std/primitive.i8.html#method.from_str_radix #[derive(Debug, Clone, PartialEq, Eq)] #[stable(feature = "rust1", since = "1.0.0")] @@ -3329,54 +4694,108 @@ pub use num::dec2flt::ParseFloatError; // Conversions T -> T are covered by a blanket impl and therefore excluded // Some conversions from and to usize/isize are not implemented due to portability concerns macro_rules! impl_from { - ($Small: ty, $Large: ty, #[$attr:meta]) => { + ($Small: ty, $Large: ty, #[$attr:meta], $doc: expr) => { #[$attr] + #[doc = $doc] impl From<$Small> for $Large { #[inline] fn from(small: $Small) -> $Large { small as $Large } } + }; + ($Small: ty, $Large: ty, #[$attr:meta]) => { + impl_from!($Small, + $Large, + #[$attr], + concat!("Converts `", + stringify!($Small), + "` to `", + stringify!($Large), + "` losslessly.")); } } +macro_rules! impl_from_bool { + ($target: ty, #[$attr:meta]) => { + impl_from!(bool, $target, #[$attr], concat!("Converts a `bool` to a `", + stringify!($target), "`. The resulting value is `0` for `false` and `1` for `true` +values. + +# Examples + +``` +assert_eq!(", stringify!($target), "::from(true), 1); +assert_eq!(", stringify!($target), "::from(false), 0); +```")); + }; +} + +// Bool -> Any +impl_from_bool! { u8, #[stable(feature = "from_bool", since = "1.28.0")] } +impl_from_bool! { u16, #[stable(feature = "from_bool", since = "1.28.0")] } +impl_from_bool! { u32, #[stable(feature = "from_bool", since = "1.28.0")] } +impl_from_bool! { u64, #[stable(feature = "from_bool", since = "1.28.0")] } +impl_from_bool! { u128, #[stable(feature = "from_bool", since = "1.28.0")] } +impl_from_bool! { usize, #[stable(feature = "from_bool", since = "1.28.0")] } +impl_from_bool! { i8, #[stable(feature = "from_bool", since = "1.28.0")] } +impl_from_bool! { i16, #[stable(feature = "from_bool", since = "1.28.0")] } +impl_from_bool! { i32, #[stable(feature = "from_bool", since = "1.28.0")] } +impl_from_bool! { i64, #[stable(feature = "from_bool", since = "1.28.0")] } +impl_from_bool! { i128, #[stable(feature = "from_bool", since = "1.28.0")] } +impl_from_bool! { isize, #[stable(feature = "from_bool", since = "1.28.0")] } + // Unsigned -> Unsigned impl_from! { u8, u16, #[stable(feature = "lossless_int_conv", since = "1.5.0")] } impl_from! { u8, u32, #[stable(feature = "lossless_int_conv", since = "1.5.0")] } impl_from! { u8, u64, #[stable(feature = "lossless_int_conv", since = "1.5.0")] } -impl_from! { u8, u128, #[unstable(feature = "i128", issue = "35118")] } +impl_from! { u8, u128, #[stable(feature = "i128", since = "1.26.0")] } impl_from! { u8, usize, #[stable(feature = "lossless_int_conv", since = "1.5.0")] } impl_from! { u16, u32, #[stable(feature = "lossless_int_conv", since = "1.5.0")] } impl_from! { u16, u64, #[stable(feature = "lossless_int_conv", since = "1.5.0")] } -impl_from! { u16, u128, #[unstable(feature = "i128", issue = "35118")] } +impl_from! { u16, u128, #[stable(feature = "i128", since = "1.26.0")] } impl_from! { u32, u64, #[stable(feature = "lossless_int_conv", since = "1.5.0")] } -impl_from! { u32, u128, #[unstable(feature = "i128", issue = "35118")] } -impl_from! { u64, u128, #[unstable(feature = "i128", issue = "35118")] } +impl_from! { u32, u128, #[stable(feature = "i128", since = "1.26.0")] } +impl_from! { u64, u128, #[stable(feature = "i128", since = "1.26.0")] } // Signed -> Signed impl_from! { i8, i16, #[stable(feature = "lossless_int_conv", since = "1.5.0")] } impl_from! { i8, i32, #[stable(feature = "lossless_int_conv", since = "1.5.0")] } impl_from! { i8, i64, #[stable(feature = "lossless_int_conv", since = "1.5.0")] } -impl_from! { i8, i128, #[unstable(feature = "i128", issue = "35118")] } +impl_from! { i8, i128, #[stable(feature = "i128", since = "1.26.0")] } impl_from! { i8, isize, #[stable(feature = "lossless_int_conv", since = "1.5.0")] } impl_from! { i16, i32, #[stable(feature = "lossless_int_conv", since = "1.5.0")] } impl_from! { i16, i64, #[stable(feature = "lossless_int_conv", since = "1.5.0")] } -impl_from! { i16, i128, #[unstable(feature = "i128", issue = "35118")] } +impl_from! { i16, i128, #[stable(feature = "i128", since = "1.26.0")] } impl_from! { i32, i64, #[stable(feature = "lossless_int_conv", since = "1.5.0")] } -impl_from! { i32, i128, #[unstable(feature = "i128", issue = "35118")] } -impl_from! { i64, i128, #[unstable(feature = "i128", issue = "35118")] } +impl_from! { i32, i128, #[stable(feature = "i128", since = "1.26.0")] } +impl_from! { i64, i128, #[stable(feature = "i128", since = "1.26.0")] } // Unsigned -> Signed impl_from! { u8, i16, #[stable(feature = "lossless_int_conv", since = "1.5.0")] } impl_from! { u8, i32, #[stable(feature = "lossless_int_conv", since = "1.5.0")] } impl_from! { u8, i64, #[stable(feature = "lossless_int_conv", since = "1.5.0")] } -impl_from! { u8, i128, #[unstable(feature = "i128", issue = "35118")] } +impl_from! { u8, i128, #[stable(feature = "i128", since = "1.26.0")] } impl_from! { u16, i32, #[stable(feature = "lossless_int_conv", since = "1.5.0")] } impl_from! { u16, i64, #[stable(feature = "lossless_int_conv", since = "1.5.0")] } -impl_from! { u16, i128, #[unstable(feature = "i128", issue = "35118")] } +impl_from! { u16, i128, #[stable(feature = "i128", since = "1.26.0")] } impl_from! { u32, i64, #[stable(feature = "lossless_int_conv", since = "1.5.0")] } -impl_from! { u32, i128, #[unstable(feature = "i128", issue = "35118")] } -impl_from! { u64, i128, #[unstable(feature = "i128", issue = "35118")] } +impl_from! { u32, i128, #[stable(feature = "i128", since = "1.26.0")] } +impl_from! { u64, i128, #[stable(feature = "i128", since = "1.26.0")] } + +// The C99 standard defines bounds on INTPTR_MIN, INTPTR_MAX, and UINTPTR_MAX +// which imply that pointer-sized integers must be at least 16 bits: +// https://port70.net/~nsz/c/c99/n1256.html#7.18.2.4 +impl_from! { u16, usize, #[stable(feature = "lossless_iusize_conv", since = "1.26.0")] } +impl_from! { u8, isize, #[stable(feature = "lossless_iusize_conv", since = "1.26.0")] } +impl_from! { i16, isize, #[stable(feature = "lossless_iusize_conv", since = "1.26.0")] } + +// RISC-V defines the possibility of a 128-bit address space (RV128). + +// CHERI proposes 256-bit “capabilities”. Unclear if this would be relevant to usize/isize. +// https://www.cl.cam.ac.uk/research/security/ctsrd/pdfs/20171017a-cheri-poster.pdf +// http://www.csl.sri.com/users/neumann/2012resolve-cheri.pdf + // Note: integers can only be represented with full precision in a float if // they fit in the significand, which is 24 bits in f32 and 53 bits in f64. diff --git a/src/libcore/num/u128.rs b/src/libcore/num/u128.rs index 987ac3e0007..e8c783a1bb5 100644 --- a/src/libcore/num/u128.rs +++ b/src/libcore/num/u128.rs @@ -12,5 +12,5 @@ //! //! *[See also the `u128` primitive type](../../std/primitive.u128.html).* -#![unstable(feature = "i128", issue="35118")] -uint_module! { u128, #[unstable(feature = "i128", issue="35118")] } +#![stable(feature = "i128", since = "1.26.0")] +uint_module! { u128, #[stable(feature = "i128", since="1.26.0")] } diff --git a/src/libcore/num/wrapping.rs b/src/libcore/num/wrapping.rs index ae1b0b3ce11..1c826c2fa76 100644 --- a/src/libcore/num/wrapping.rs +++ b/src/libcore/num/wrapping.rs @@ -112,17 +112,19 @@ macro_rules! sh_impl_all { //sh_impl_unsigned! { $t, u16 } //sh_impl_unsigned! { $t, u32 } //sh_impl_unsigned! { $t, u64 } + //sh_impl_unsigned! { $t, u128 } sh_impl_unsigned! { $t, usize } //sh_impl_signed! { $t, i8 } //sh_impl_signed! { $t, i16 } //sh_impl_signed! { $t, i32 } //sh_impl_signed! { $t, i64 } + //sh_impl_signed! { $t, i128 } //sh_impl_signed! { $t, isize } )*) } -sh_impl_all! { u8 u16 u32 u64 usize i8 i16 i32 i64 isize } +sh_impl_all! { u8 u16 u32 u64 u128 usize i8 i16 i32 i64 i128 isize } // FIXME(30524): impl Op<T> for Wrapping<T>, impl OpAssign<T> for Wrapping<T> macro_rules! wrapping_impl { @@ -317,11 +319,580 @@ macro_rules! wrapping_impl { } forward_ref_unop! { impl Neg, neg for Wrapping<$t>, #[stable(feature = "wrapping_ref", since = "1.14.0")] } + )*) } wrapping_impl! { usize u8 u16 u32 u64 u128 isize i8 i16 i32 i64 i128 } +macro_rules! wrapping_int_impl { + ($($t:ty)*) => ($( + impl Wrapping<$t> { + doc_comment! { + concat!("Returns the smallest value that can be represented by this integer type. + +# Examples + +Basic usage: + +``` +#![feature(wrapping_int_impl)] +use std::num::Wrapping; + +assert_eq!(<Wrapping<", stringify!($t), ">>::min_value(), ", +"Wrapping(", stringify!($t), "::min_value())); +```"), + #[unstable(feature = "wrapping_int_impl", issue = "32463")] + #[inline] + pub const fn min_value() -> Self { + Wrapping(<$t>::min_value()) + } + } + + doc_comment! { + concat!("Returns the largest value that can be represented by this integer type. + +# Examples + +Basic usage: + +``` +#![feature(wrapping_int_impl)] +use std::num::Wrapping; + +assert_eq!(<Wrapping<", stringify!($t), ">>::max_value(), ", +"Wrapping(", stringify!($t), "::max_value())); +```"), + #[unstable(feature = "wrapping_int_impl", issue = "32463")] + #[inline] + pub const fn max_value() -> Self { + Wrapping(<$t>::max_value()) + } + } + + doc_comment! { + concat!("Returns the number of ones in the binary representation of `self`. + +# Examples + +Basic usage: + +``` +#![feature(wrapping_int_impl)] +use std::num::Wrapping; + +let n = Wrapping(0b01001100", stringify!($t), "); + +assert_eq!(n.count_ones(), 3); +```"), + #[inline] + #[unstable(feature = "wrapping_int_impl", issue = "32463")] + pub fn count_ones(self) -> u32 { + self.0.count_ones() + } + } + + doc_comment! { + concat!("Returns the number of zeros in the binary representation of `self`. + +# Examples + +Basic usage: + +``` +#![feature(wrapping_int_impl)] +use std::num::Wrapping; + +assert_eq!(Wrapping(!0", stringify!($t), ").count_zeros(), 0); +```"), + #[inline] + #[unstable(feature = "wrapping_int_impl", issue = "32463")] + pub fn count_zeros(self) -> u32 { + self.0.count_zeros() + } + } + + doc_comment! { + concat!("Returns the number of trailing zeros in the binary representation +of `self`. + +# Examples + +Basic usage: + +``` +#![feature(wrapping_int_impl)] +use std::num::Wrapping; + +let n = Wrapping(0b0101000", stringify!($t), "); + +assert_eq!(n.trailing_zeros(), 3); +```"), + #[inline] + #[unstable(feature = "wrapping_int_impl", issue = "32463")] + pub fn trailing_zeros(self) -> u32 { + self.0.trailing_zeros() + } + } + + /// Shifts the bits to the left by a specified amount, `n`, + /// wrapping the truncated bits to the end of the resulting + /// integer. + /// + /// Please note this isn't the same operation as `>>`! + /// + /// # Examples + /// + /// Basic usage: + /// + /// ``` + /// #![feature(wrapping_int_impl)] + /// use std::num::Wrapping; + /// + /// let n: Wrapping<i64> = Wrapping(0x0123456789ABCDEF); + /// let m: Wrapping<i64> = Wrapping(-0x76543210FEDCBA99); + /// + /// assert_eq!(n.rotate_left(32), m); + /// ``` + #[inline] + #[unstable(feature = "wrapping_int_impl", issue = "32463")] + pub fn rotate_left(self, n: u32) -> Self { + Wrapping(self.0.rotate_left(n)) + } + + /// Shifts the bits to the right by a specified amount, `n`, + /// wrapping the truncated bits to the beginning of the resulting + /// integer. + /// + /// Please note this isn't the same operation as `<<`! + /// + /// # Examples + /// + /// Basic usage: + /// + /// ``` + /// #![feature(wrapping_int_impl)] + /// use std::num::Wrapping; + /// + /// let n: Wrapping<i64> = Wrapping(0x0123456789ABCDEF); + /// let m: Wrapping<i64> = Wrapping(-0xFEDCBA987654322); + /// + /// assert_eq!(n.rotate_right(4), m); + /// ``` + #[inline] + #[unstable(feature = "wrapping_int_impl", issue = "32463")] + pub fn rotate_right(self, n: u32) -> Self { + Wrapping(self.0.rotate_right(n)) + } + + /// Reverses the byte order of the integer. + /// + /// # Examples + /// + /// Basic usage: + /// + /// ``` + /// #![feature(wrapping_int_impl)] + /// use std::num::Wrapping; + /// + /// let n: Wrapping<i16> = Wrapping(0b0000000_01010101); + /// assert_eq!(n, Wrapping(85)); + /// + /// let m = n.swap_bytes(); + /// + /// assert_eq!(m, Wrapping(0b01010101_00000000)); + /// assert_eq!(m, Wrapping(21760)); + /// ``` + #[inline] + #[unstable(feature = "wrapping_int_impl", issue = "32463")] + pub fn swap_bytes(self) -> Self { + Wrapping(self.0.swap_bytes()) + } + + /// Reverses the bit pattern of the integer. + /// + /// # Examples + /// + /// Please note that this example is shared between integer types. + /// Which explains why `i16` is used here. + /// + /// Basic usage: + /// + /// ``` + /// #![feature(reverse_bits)] + /// use std::num::Wrapping; + /// + /// let n = Wrapping(0b0000000_01010101i16); + /// assert_eq!(n, Wrapping(85)); + /// + /// let m = n.reverse_bits(); + /// + /// assert_eq!(m.0 as u16, 0b10101010_00000000); + /// assert_eq!(m, Wrapping(-22016)); + /// ``` + #[unstable(feature = "reverse_bits", issue = "48763")] + #[inline] + pub fn reverse_bits(self) -> Self { + Wrapping(self.0.reverse_bits()) + } + + doc_comment! { + concat!("Converts an integer from big endian to the target's endianness. + +On big endian this is a no-op. On little endian the bytes are +swapped. + +# Examples + +Basic usage: + +``` +#![feature(wrapping_int_impl)] +use std::num::Wrapping; + +let n = Wrapping(0x1A", stringify!($t), "); + +if cfg!(target_endian = \"big\") { + assert_eq!(<Wrapping<", stringify!($t), ">>::from_be(n), n) +} else { + assert_eq!(<Wrapping<", stringify!($t), ">>::from_be(n), n.swap_bytes()) +} +```"), + #[inline] + #[unstable(feature = "wrapping_int_impl", issue = "32463")] + pub fn from_be(x: Self) -> Self { + Wrapping(<$t>::from_be(x.0)) + } + } + + doc_comment! { + concat!("Converts an integer from little endian to the target's endianness. + +On little endian this is a no-op. On big endian the bytes are +swapped. + +# Examples + +Basic usage: + +``` +#![feature(wrapping_int_impl)] +use std::num::Wrapping; + +let n = Wrapping(0x1A", stringify!($t), "); + +if cfg!(target_endian = \"little\") { + assert_eq!(<Wrapping<", stringify!($t), ">>::from_le(n), n) +} else { + assert_eq!(<Wrapping<", stringify!($t), ">>::from_le(n), n.swap_bytes()) +} +```"), + #[inline] + #[unstable(feature = "wrapping_int_impl", issue = "32463")] + pub fn from_le(x: Self) -> Self { + Wrapping(<$t>::from_le(x.0)) + } + } + + doc_comment! { + concat!("Converts `self` to big endian from the target's endianness. + +On big endian this is a no-op. On little endian the bytes are +swapped. + +# Examples + +Basic usage: + +``` +#![feature(wrapping_int_impl)] +use std::num::Wrapping; + +let n = Wrapping(0x1A", stringify!($t), "); + +if cfg!(target_endian = \"big\") { + assert_eq!(n.to_be(), n) +} else { + assert_eq!(n.to_be(), n.swap_bytes()) +} +```"), + #[inline] + #[unstable(feature = "wrapping_int_impl", issue = "32463")] + pub fn to_be(self) -> Self { + Wrapping(self.0.to_be()) + } + } + + doc_comment! { + concat!("Converts `self` to little endian from the target's endianness. + +On little endian this is a no-op. On big endian the bytes are +swapped. + +# Examples + +Basic usage: + +``` +#![feature(wrapping_int_impl)] +use std::num::Wrapping; + +let n = Wrapping(0x1A", stringify!($t), "); + +if cfg!(target_endian = \"little\") { + assert_eq!(n.to_le(), n) +} else { + assert_eq!(n.to_le(), n.swap_bytes()) +} +```"), + #[inline] + #[unstable(feature = "wrapping_int_impl", issue = "32463")] + pub fn to_le(self) -> Self { + Wrapping(self.0.to_le()) + } + } + + doc_comment! { + concat!("Raises self to the power of `exp`, using exponentiation by squaring. + +# Examples + +Basic usage: + +``` +#![feature(wrapping_int_impl)] +use std::num::Wrapping; + +assert_eq!(Wrapping(3", stringify!($t), ").pow(4), Wrapping(81)); +``` + +Results that are too large are wrapped: + +``` +#![feature(wrapping_int_impl)] +use std::num::Wrapping; + +assert_eq!(Wrapping(3i8).pow(5), Wrapping(-13)); +assert_eq!(Wrapping(3i8).pow(6), Wrapping(-39)); +```"), + #[inline] + #[unstable(feature = "wrapping_int_impl", issue = "32463")] + pub fn pow(self, exp: u32) -> Self { + Wrapping(self.0.wrapping_pow(exp)) + } + } + } + )*) +} + +wrapping_int_impl! { usize u8 u16 u32 u64 u128 isize i8 i16 i32 i64 i128 } + +macro_rules! wrapping_int_impl_signed { + ($($t:ty)*) => ($( + impl Wrapping<$t> { + doc_comment! { + concat!("Returns the number of leading zeros in the binary representation of `self`. + +# Examples + +Basic usage: + +``` +#![feature(wrapping_int_impl)] +use std::num::Wrapping; + +let n = Wrapping(", stringify!($t), "::max_value()) >> 2; + +assert_eq!(n.leading_zeros(), 3); +```"), + #[inline] + #[unstable(feature = "wrapping_int_impl", issue = "32463")] + pub fn leading_zeros(self) -> u32 { + self.0.leading_zeros() + } + } + + doc_comment! { + concat!("Computes the absolute value of `self`, wrapping around at +the boundary of the type. + +The only case where such wrapping can occur is when one takes the absolute value of the negative +minimal value for the type this is a positive value that is too large to represent in the type. In +such a case, this function returns `MIN` itself. + +# Examples + +Basic usage: + +``` +#![feature(wrapping_int_impl)] +use std::num::Wrapping; + +assert_eq!(Wrapping(100", stringify!($t), ").abs(), Wrapping(100)); +assert_eq!(Wrapping(-100", stringify!($t), ").abs(), Wrapping(100)); +assert_eq!(Wrapping(", stringify!($t), "::min_value()).abs(), Wrapping(", stringify!($t), +"::min_value())); +assert_eq!(Wrapping(-128i8).abs().0 as u8, 128u8); +```"), + #[inline] + #[unstable(feature = "wrapping_int_impl", issue = "32463")] + pub fn abs(self) -> Wrapping<$t> { + Wrapping(self.0.wrapping_abs()) + } + } + + doc_comment! { + concat!("Returns a number representing sign of `self`. + + - `0` if the number is zero + - `1` if the number is positive + - `-1` if the number is negative + +# Examples + +Basic usage: + +``` +#![feature(wrapping_int_impl)] +use std::num::Wrapping; + +assert_eq!(Wrapping(10", stringify!($t), ").signum(), Wrapping(1)); +assert_eq!(Wrapping(0", stringify!($t), ").signum(), Wrapping(0)); +assert_eq!(Wrapping(-10", stringify!($t), ").signum(), Wrapping(-1)); +```"), + #[inline] + #[unstable(feature = "wrapping_int_impl", issue = "32463")] + pub fn signum(self) -> Wrapping<$t> { + Wrapping(self.0.signum()) + } + } + + doc_comment! { + concat!("Returns `true` if `self` is positive and `false` if the number is zero or +negative. + +# Examples + +Basic usage: + +``` +#![feature(wrapping_int_impl)] +use std::num::Wrapping; + +assert!(Wrapping(10", stringify!($t), ").is_positive()); +assert!(!Wrapping(-10", stringify!($t), ").is_positive()); +```"), + #[inline] + #[unstable(feature = "wrapping_int_impl", issue = "32463")] + pub fn is_positive(self) -> bool { + self.0.is_positive() + } + } + + doc_comment! { + concat!("Returns `true` if `self` is negative and `false` if the number is zero or +positive. + +# Examples + +Basic usage: + +``` +#![feature(wrapping_int_impl)] +use std::num::Wrapping; + +assert!(Wrapping(-10", stringify!($t), ").is_negative()); +assert!(!Wrapping(10", stringify!($t), ").is_negative()); +```"), + #[inline] + #[unstable(feature = "wrapping_int_impl", issue = "32463")] + pub fn is_negative(self) -> bool { + self.0.is_negative() + } + } + } + )*) +} + +wrapping_int_impl_signed! { isize i8 i16 i32 i64 i128 } + +macro_rules! wrapping_int_impl_unsigned { + ($($t:ty)*) => ($( + impl Wrapping<$t> { + doc_comment! { + concat!("Returns the number of leading zeros in the binary representation of `self`. + +# Examples + +Basic usage: + +``` +#![feature(wrapping_int_impl)] +use std::num::Wrapping; + +let n = Wrapping(", stringify!($t), "::max_value()) >> 2; + +assert_eq!(n.leading_zeros(), 2); +```"), + #[inline] + #[unstable(feature = "wrapping_int_impl", issue = "32463")] + pub fn leading_zeros(self) -> u32 { + self.0.leading_zeros() + } + } + + doc_comment! { + concat!("Returns `true` if and only if `self == 2^k` for some `k`. + +# Examples + +Basic usage: + +``` +#![feature(wrapping_int_impl)] +use std::num::Wrapping; + +assert!(Wrapping(16", stringify!($t), ").is_power_of_two()); +assert!(!Wrapping(10", stringify!($t), ").is_power_of_two()); +```"), + #[inline] + #[unstable(feature = "wrapping_int_impl", issue = "32463")] + pub fn is_power_of_two(self) -> bool { + self.0.is_power_of_two() + } + } + + doc_comment! { + concat!("Returns the smallest power of two greater than or equal to `self`. + +When return value overflows (i.e. `self > (1 << (N-1))` for type +`uN`), overflows to `2^N = 0`. + +# Examples + +Basic usage: + +``` +#![feature(wrapping_next_power_of_two)] +use std::num::Wrapping; + +assert_eq!(Wrapping(2", stringify!($t), ").next_power_of_two(), Wrapping(2)); +assert_eq!(Wrapping(3", stringify!($t), ").next_power_of_two(), Wrapping(4)); +assert_eq!(Wrapping(200_u8).next_power_of_two(), Wrapping(0)); +```"), + #[inline] + #[unstable(feature = "wrapping_next_power_of_two", issue = "32463", + reason = "needs decision on wrapping behaviour")] + pub fn next_power_of_two(self) -> Self { + Wrapping(self.0.wrapping_next_power_of_two()) + } + } + } + )*) +} + +wrapping_int_impl_unsigned! { usize u8 u16 u32 u64 u128 } + mod shift_max { #![allow(non_upper_case_globals)] @@ -347,11 +918,13 @@ mod shift_max { pub const i16: u32 = (1 << 4) - 1; pub const i32: u32 = (1 << 5) - 1; pub const i64: u32 = (1 << 6) - 1; + pub const i128: u32 = (1 << 7) - 1; pub use self::platform::isize; pub const u8: u32 = i8; pub const u16: u32 = i16; pub const u32: u32 = i32; pub const u64: u32 = i64; + pub const u128: u32 = i128; pub use self::platform::usize; } diff --git a/src/libcore/ops/arith.rs b/src/libcore/ops/arith.rs index 8b3d662a6db..a1bc5463f73 100644 --- a/src/libcore/ops/arith.rs +++ b/src/libcore/ops/arith.rs @@ -75,13 +75,26 @@ /// ``` #[lang = "add"] #[stable(feature = "rust1", since = "1.0.0")] -#[rustc_on_unimplemented = "no implementation for `{Self} + {RHS}`"] +#[rustc_on_unimplemented( + on( + all(_Self="{integer}", RHS="{float}"), + message="cannot add a float to an integer", + ), + on( + all(_Self="{float}", RHS="{integer}"), + message="cannot add an integer to a float", + ), + message="cannot add `{RHS}` to `{Self}`", + label="no implementation for `{Self} + {RHS}`", +)] +#[doc(alias = "+")] pub trait Add<RHS=Self> { /// The resulting type after applying the `+` operator. #[stable(feature = "rust1", since = "1.0.0")] type Output; /// Performs the `+` operation. + #[must_use] #[stable(feature = "rust1", since = "1.0.0")] fn add(self, rhs: RHS) -> Self::Output; } @@ -170,13 +183,16 @@ add_impl! { usize u8 u16 u32 u64 u128 isize i8 i16 i32 i64 i128 f32 f64 } /// ``` #[lang = "sub"] #[stable(feature = "rust1", since = "1.0.0")] -#[rustc_on_unimplemented = "no implementation for `{Self} - {RHS}`"] +#[rustc_on_unimplemented(message="cannot subtract `{RHS}` from `{Self}`", + label="no implementation for `{Self} - {RHS}`")] +#[doc(alias = "-")] pub trait Sub<RHS=Self> { /// The resulting type after applying the `-` operator. #[stable(feature = "rust1", since = "1.0.0")] type Output; /// Performs the `-` operation. + #[must_use] #[stable(feature = "rust1", since = "1.0.0")] fn sub(self, rhs: RHS) -> Self::Output; } @@ -287,13 +303,16 @@ sub_impl! { usize u8 u16 u32 u64 u128 isize i8 i16 i32 i64 i128 f32 f64 } /// ``` #[lang = "mul"] #[stable(feature = "rust1", since = "1.0.0")] -#[rustc_on_unimplemented = "no implementation for `{Self} * {RHS}`"] +#[rustc_on_unimplemented(message="cannot multiply `{RHS}` to `{Self}`", + label="no implementation for `{Self} * {RHS}`")] +#[doc(alias = "*")] pub trait Mul<RHS=Self> { /// The resulting type after applying the `*` operator. #[stable(feature = "rust1", since = "1.0.0")] type Output; /// Performs the `*` operation. + #[must_use] #[stable(feature = "rust1", since = "1.0.0")] fn mul(self, rhs: RHS) -> Self::Output; } @@ -408,13 +427,16 @@ mul_impl! { usize u8 u16 u32 u64 u128 isize i8 i16 i32 i64 i128 f32 f64 } /// ``` #[lang = "div"] #[stable(feature = "rust1", since = "1.0.0")] -#[rustc_on_unimplemented = "no implementation for `{Self} / {RHS}`"] +#[rustc_on_unimplemented(message="cannot divide `{Self}` by `{RHS}`", + label="no implementation for `{Self} / {RHS}`")] +#[doc(alias = "/")] pub trait Div<RHS=Self> { /// The resulting type after applying the `/` operator. #[stable(feature = "rust1", since = "1.0.0")] type Output; /// Performs the `/` operation. + #[must_use] #[stable(feature = "rust1", since = "1.0.0")] fn div(self, rhs: RHS) -> Self::Output; } @@ -490,13 +512,16 @@ div_impl_float! { f32 f64 } /// ``` #[lang = "rem"] #[stable(feature = "rust1", since = "1.0.0")] -#[rustc_on_unimplemented = "no implementation for `{Self} % {RHS}`"] +#[rustc_on_unimplemented(message="cannot mod `{Self}` by `{RHS}`", + label="no implementation for `{Self} % {RHS}`")] +#[doc(alias = "%")] pub trait Rem<RHS=Self> { /// The resulting type after applying the `%` operator. #[stable(feature = "rust1", since = "1.0.0")] type Output = Self; /// Performs the `%` operation. + #[must_use] #[stable(feature = "rust1", since = "1.0.0")] fn rem(self, rhs: RHS) -> Self::Output; } @@ -574,12 +599,14 @@ rem_impl_float! { f32 f64 } /// ``` #[lang = "neg"] #[stable(feature = "rust1", since = "1.0.0")] +#[doc(alias = "-")] pub trait Neg { /// The resulting type after applying the `-` operator. #[stable(feature = "rust1", since = "1.0.0")] type Output; /// Performs the unary `-` operation. + #[must_use] #[stable(feature = "rust1", since = "1.0.0")] fn neg(self) -> Self::Output; } @@ -647,7 +674,10 @@ neg_impl_numeric! { isize i8 i16 i32 i64 i128 f32 f64 } /// ``` #[lang = "add_assign"] #[stable(feature = "op_assign_traits", since = "1.8.0")] -#[rustc_on_unimplemented = "no implementation for `{Self} += {Rhs}`"] +#[rustc_on_unimplemented(message="cannot add-assign `{Rhs}` to `{Self}`", + label="no implementation for `{Self} += {Rhs}`")] +#[doc(alias = "+")] +#[doc(alias = "+=")] pub trait AddAssign<Rhs=Self> { /// Performs the `+=` operation. #[stable(feature = "op_assign_traits", since = "1.8.0")] @@ -700,7 +730,10 @@ add_assign_impl! { usize u8 u16 u32 u64 u128 isize i8 i16 i32 i64 i128 f32 f64 } /// ``` #[lang = "sub_assign"] #[stable(feature = "op_assign_traits", since = "1.8.0")] -#[rustc_on_unimplemented = "no implementation for `{Self} -= {Rhs}`"] +#[rustc_on_unimplemented(message="cannot subtract-assign `{Rhs}` from `{Self}`", + label="no implementation for `{Self} -= {Rhs}`")] +#[doc(alias = "-")] +#[doc(alias = "-=")] pub trait SubAssign<Rhs=Self> { /// Performs the `-=` operation. #[stable(feature = "op_assign_traits", since = "1.8.0")] @@ -744,7 +777,10 @@ sub_assign_impl! { usize u8 u16 u32 u64 u128 isize i8 i16 i32 i64 i128 f32 f64 } /// ``` #[lang = "mul_assign"] #[stable(feature = "op_assign_traits", since = "1.8.0")] -#[rustc_on_unimplemented = "no implementation for `{Self} *= {Rhs}`"] +#[rustc_on_unimplemented(message="cannot multiply-assign `{Rhs}` to `{Self}`", + label="no implementation for `{Self} *= {Rhs}`")] +#[doc(alias = "*")] +#[doc(alias = "*=")] pub trait MulAssign<Rhs=Self> { /// Performs the `*=` operation. #[stable(feature = "op_assign_traits", since = "1.8.0")] @@ -788,7 +824,10 @@ mul_assign_impl! { usize u8 u16 u32 u64 u128 isize i8 i16 i32 i64 i128 f32 f64 } /// ``` #[lang = "div_assign"] #[stable(feature = "op_assign_traits", since = "1.8.0")] -#[rustc_on_unimplemented = "no implementation for `{Self} /= {Rhs}`"] +#[rustc_on_unimplemented(message="cannot divide-assign `{Self}` by `{Rhs}`", + label="no implementation for `{Self} /= {Rhs}`")] +#[doc(alias = "/")] +#[doc(alias = "/=")] pub trait DivAssign<Rhs=Self> { /// Performs the `/=` operation. #[stable(feature = "op_assign_traits", since = "1.8.0")] @@ -835,7 +874,10 @@ div_assign_impl! { usize u8 u16 u32 u64 u128 isize i8 i16 i32 i64 i128 f32 f64 } /// ``` #[lang = "rem_assign"] #[stable(feature = "op_assign_traits", since = "1.8.0")] -#[rustc_on_unimplemented = "no implementation for `{Self} %= {Rhs}`"] +#[rustc_on_unimplemented(message="cannot mod-assign `{Self}` by `{Rhs}``", + label="no implementation for `{Self} %= {Rhs}`")] +#[doc(alias = "%")] +#[doc(alias = "%=")] pub trait RemAssign<Rhs=Self> { /// Performs the `%=` operation. #[stable(feature = "op_assign_traits", since = "1.8.0")] diff --git a/src/libcore/ops/bit.rs b/src/libcore/ops/bit.rs index 7ac5fc4debf..3900f365b0a 100644 --- a/src/libcore/ops/bit.rs +++ b/src/libcore/ops/bit.rs @@ -46,6 +46,7 @@ pub trait Not { type Output; /// Performs the unary `!` operation. + #[must_use] #[stable(feature = "rust1", since = "1.0.0")] fn not(self) -> Self::Output; } @@ -119,14 +120,17 @@ not_impl! { bool usize u8 u16 u32 u64 u128 isize i8 i16 i32 i64 i128 } /// assert_eq!(bv1 & bv2, expected); /// ``` #[lang = "bitand"] +#[doc(alias = "&")] #[stable(feature = "rust1", since = "1.0.0")] -#[rustc_on_unimplemented = "no implementation for `{Self} & {RHS}`"] +#[rustc_on_unimplemented(message="no implementation for `{Self} & {RHS}`", + label="no implementation for `{Self} & {RHS}`")] pub trait BitAnd<RHS=Self> { /// The resulting type after applying the `&` operator. #[stable(feature = "rust1", since = "1.0.0")] type Output; /// Performs the `&` operation. + #[must_use] #[stable(feature = "rust1", since = "1.0.0")] fn bitand(self, rhs: RHS) -> Self::Output; } @@ -200,14 +204,17 @@ bitand_impl! { bool usize u8 u16 u32 u64 u128 isize i8 i16 i32 i64 i128 } /// assert_eq!(bv1 | bv2, expected); /// ``` #[lang = "bitor"] +#[doc(alias = "|")] #[stable(feature = "rust1", since = "1.0.0")] -#[rustc_on_unimplemented = "no implementation for `{Self} | {RHS}`"] +#[rustc_on_unimplemented(message="no implementation for `{Self} | {RHS}`", + label="no implementation for `{Self} | {RHS}`")] pub trait BitOr<RHS=Self> { /// The resulting type after applying the `|` operator. #[stable(feature = "rust1", since = "1.0.0")] type Output; /// Performs the `|` operation. + #[must_use] #[stable(feature = "rust1", since = "1.0.0")] fn bitor(self, rhs: RHS) -> Self::Output; } @@ -284,14 +291,17 @@ bitor_impl! { bool usize u8 u16 u32 u64 u128 isize i8 i16 i32 i64 i128 } /// assert_eq!(bv1 ^ bv2, expected); /// ``` #[lang = "bitxor"] +#[doc(alias = "^")] #[stable(feature = "rust1", since = "1.0.0")] -#[rustc_on_unimplemented = "no implementation for `{Self} ^ {RHS}`"] +#[rustc_on_unimplemented(message="no implementation for `{Self} ^ {RHS}`", + label="no implementation for `{Self} ^ {RHS}`")] pub trait BitXor<RHS=Self> { /// The resulting type after applying the `^` operator. #[stable(feature = "rust1", since = "1.0.0")] type Output; /// Performs the `^` operation. + #[must_use] #[stable(feature = "rust1", since = "1.0.0")] fn bitxor(self, rhs: RHS) -> Self::Output; } @@ -312,7 +322,12 @@ macro_rules! bitxor_impl { bitxor_impl! { bool usize u8 u16 u32 u64 u128 isize i8 i16 i32 i64 i128 } -/// The left shift operator `<<`. +/// The left shift operator `<<`. Note that because this trait is implemented +/// for all integer types with multiple right-hand-side types, Rust's type +/// checker has special handling for `_ << _`, setting the result type for +/// integer operations to the type of the left-hand-side operand. This means +/// that though `a << b` and `a.shl(b)` are one and the same from an evaluation +/// standpoint, they are different when it comes to type inference. /// /// # Examples /// @@ -364,14 +379,17 @@ bitxor_impl! { bool usize u8 u16 u32 u64 u128 isize i8 i16 i32 i64 i128 } /// SpinVector { vec: vec![2, 3, 4, 0, 1] }); /// ``` #[lang = "shl"] +#[doc(alias = "<<")] #[stable(feature = "rust1", since = "1.0.0")] -#[rustc_on_unimplemented = "no implementation for `{Self} << {RHS}`"] -pub trait Shl<RHS> { +#[rustc_on_unimplemented(message="no implementation for `{Self} << {RHS}`", + label="no implementation for `{Self} << {RHS}`")] +pub trait Shl<RHS=Self> { /// The resulting type after applying the `<<` operator. #[stable(feature = "rust1", since = "1.0.0")] type Output; /// Performs the `<<` operation. + #[must_use] #[stable(feature = "rust1", since = "1.0.0")] fn shl(self, rhs: RHS) -> Self::Output; } @@ -413,7 +431,12 @@ macro_rules! shl_impl_all { shl_impl_all! { u8 u16 u32 u64 u128 usize i8 i16 i32 i64 isize i128 } -/// The right shift operator `>>`. +/// The right shift operator `>>`. Note that because this trait is implemented +/// for all integer types with multiple right-hand-side types, Rust's type +/// checker has special handling for `_ >> _`, setting the result type for +/// integer operations to the type of the left-hand-side operand. This means +/// that though `a >> b` and `a.shr(b)` are one and the same from an evaluation +/// standpoint, they are different when it comes to type inference. /// /// # Examples /// @@ -465,14 +488,17 @@ shl_impl_all! { u8 u16 u32 u64 u128 usize i8 i16 i32 i64 isize i128 } /// SpinVector { vec: vec![3, 4, 0, 1, 2] }); /// ``` #[lang = "shr"] +#[doc(alias = ">>")] #[stable(feature = "rust1", since = "1.0.0")] -#[rustc_on_unimplemented = "no implementation for `{Self} >> {RHS}`"] -pub trait Shr<RHS> { +#[rustc_on_unimplemented(message="no implementation for `{Self} >> {RHS}`", + label="no implementation for `{Self} >> {RHS}`")] +pub trait Shr<RHS=Self> { /// The resulting type after applying the `>>` operator. #[stable(feature = "rust1", since = "1.0.0")] type Output; /// Performs the `>>` operation. + #[must_use] #[stable(feature = "rust1", since = "1.0.0")] fn shr(self, rhs: RHS) -> Self::Output; } @@ -578,8 +604,10 @@ shr_impl_all! { u8 u16 u32 u64 u128 usize i8 i16 i32 i64 i128 isize } /// assert_eq!(bv, expected); /// ``` #[lang = "bitand_assign"] +#[doc(alias = "&=")] #[stable(feature = "op_assign_traits", since = "1.8.0")] -#[rustc_on_unimplemented = "no implementation for `{Self} &= {Rhs}`"] +#[rustc_on_unimplemented(message="no implementation for `{Self} &= {Rhs}`", + label="no implementation for `{Self} &= {Rhs}`")] pub trait BitAndAssign<Rhs=Self> { /// Performs the `&=` operation. #[stable(feature = "op_assign_traits", since = "1.8.0")] @@ -625,8 +653,10 @@ bitand_assign_impl! { bool usize u8 u16 u32 u64 u128 isize i8 i16 i32 i64 i128 } /// assert_eq!(prefs, PersonalPreferences { likes_cats: true, likes_dogs: true }); /// ``` #[lang = "bitor_assign"] +#[doc(alias = "|=")] #[stable(feature = "op_assign_traits", since = "1.8.0")] -#[rustc_on_unimplemented = "no implementation for `{Self} |= {Rhs}`"] +#[rustc_on_unimplemented(message="no implementation for `{Self} |= {Rhs}`", + label="no implementation for `{Self} |= {Rhs}`")] pub trait BitOrAssign<Rhs=Self> { /// Performs the `|=` operation. #[stable(feature = "op_assign_traits", since = "1.8.0")] @@ -672,8 +702,10 @@ bitor_assign_impl! { bool usize u8 u16 u32 u64 u128 isize i8 i16 i32 i64 i128 } /// assert_eq!(personality, Personality { has_soul: true, likes_knitting: false}); /// ``` #[lang = "bitxor_assign"] +#[doc(alias = "^=")] #[stable(feature = "op_assign_traits", since = "1.8.0")] -#[rustc_on_unimplemented = "no implementation for `{Self} ^= {Rhs}`"] +#[rustc_on_unimplemented(message="no implementation for `{Self} ^= {Rhs}`", + label="no implementation for `{Self} ^= {Rhs}`")] pub trait BitXorAssign<Rhs=Self> { /// Performs the `^=` operation. #[stable(feature = "op_assign_traits", since = "1.8.0")] @@ -717,9 +749,11 @@ bitxor_assign_impl! { bool usize u8 u16 u32 u64 u128 isize i8 i16 i32 i64 i128 } /// assert_eq!(scalar, Scalar(16)); /// ``` #[lang = "shl_assign"] +#[doc(alias = "<<=")] #[stable(feature = "op_assign_traits", since = "1.8.0")] -#[rustc_on_unimplemented = "no implementation for `{Self} <<= {Rhs}`"] -pub trait ShlAssign<Rhs> { +#[rustc_on_unimplemented(message="no implementation for `{Self} <<= {Rhs}`", + label="no implementation for `{Self} <<= {Rhs}`")] +pub trait ShlAssign<Rhs=Self> { /// Performs the `<<=` operation. #[stable(feature = "op_assign_traits", since = "1.8.0")] fn shl_assign(&mut self, rhs: Rhs); @@ -783,8 +817,10 @@ shl_assign_impl_all! { u8 u16 u32 u64 u128 usize i8 i16 i32 i64 i128 isize } /// assert_eq!(scalar, Scalar(4)); /// ``` #[lang = "shr_assign"] +#[doc(alias = ">>=")] #[stable(feature = "op_assign_traits", since = "1.8.0")] -#[rustc_on_unimplemented = "no implementation for `{Self} >>= {Rhs}`"] +#[rustc_on_unimplemented(message="no implementation for `{Self} >>= {Rhs}`", + label="no implementation for `{Self} >>= {Rhs}`")] pub trait ShrAssign<Rhs=Self> { /// Performs the `>>=` operation. #[stable(feature = "op_assign_traits", since = "1.8.0")] diff --git a/src/libcore/ops/deref.rs b/src/libcore/ops/deref.rs index 4ce0740130b..54eecc82e19 100644 --- a/src/libcore/ops/deref.rs +++ b/src/libcore/ops/deref.rs @@ -68,6 +68,8 @@ /// assert_eq!('a', *x); /// ``` #[lang = "deref"] +#[doc(alias = "*")] +#[doc(alias = "&*")] #[stable(feature = "rust1", since = "1.0.0")] pub trait Deref { /// The resulting type after dereferencing. @@ -75,6 +77,7 @@ pub trait Deref { type Target: ?Sized; /// Dereferences the value. + #[must_use] #[stable(feature = "rust1", since = "1.0.0")] fn deref(&self) -> &Self::Target; } @@ -162,6 +165,7 @@ impl<'a, T: ?Sized> Deref for &'a mut T { /// assert_eq!('b', *x); /// ``` #[lang = "deref_mut"] +#[doc(alias = "*")] #[stable(feature = "rust1", since = "1.0.0")] pub trait DerefMut: Deref { /// Mutably dereferences the value. diff --git a/src/libcore/ops/drop.rs b/src/libcore/ops/drop.rs index 70ab7b2f3b7..474f7e34c34 100644 --- a/src/libcore/ops/drop.rs +++ b/src/libcore/ops/drop.rs @@ -95,7 +95,7 @@ pub trait Drop { /// Executes the destructor for this type. /// - /// This method is called implilcitly when the value goes out of scope, + /// This method is called implicitly when the value goes out of scope, /// and cannot be called explicitly (this is compiler error [E0040]). /// However, the [`std::mem::drop`] function in the prelude can be /// used to call the argument's `Drop` implementation. diff --git a/src/libcore/ops/generator.rs b/src/libcore/ops/generator.rs index dc7669d195c..4b70c5398be 100644 --- a/src/libcore/ops/generator.rs +++ b/src/libcore/ops/generator.rs @@ -56,11 +56,11 @@ pub enum GeneratorState<Y, R> { /// return "foo" /// }; /// -/// match generator.resume() { +/// match unsafe { generator.resume() } { /// GeneratorState::Yielded(1) => {} /// _ => panic!("unexpected return from resume"), /// } -/// match generator.resume() { +/// match unsafe { generator.resume() } { /// GeneratorState::Complete("foo") => {} /// _ => panic!("unexpected return from resume"), /// } @@ -98,6 +98,10 @@ pub trait Generator { /// generator will continue executing until it either yields or returns, at /// which point this function will return. /// + /// The function is unsafe because it can be used on an immovable generator. + /// After such a call, the immovable generator must not move again, but + /// this is not enforced by the compiler. + /// /// # Return value /// /// The `GeneratorState` enum returned from this function indicates what @@ -116,7 +120,7 @@ pub trait Generator { /// been returned previously. While generator literals in the language are /// guaranteed to panic on resuming after `Complete`, this is not guaranteed /// for all implementations of the `Generator` trait. - fn resume(&mut self) -> GeneratorState<Self::Yield, Self::Return>; + unsafe fn resume(&mut self) -> GeneratorState<Self::Yield, Self::Return>; } #[unstable(feature = "generator_trait", issue = "43122")] @@ -125,7 +129,7 @@ impl<'a, T> Generator for &'a mut T { type Yield = T::Yield; type Return = T::Return; - fn resume(&mut self) -> GeneratorState<Self::Yield, Self::Return> { + unsafe fn resume(&mut self) -> GeneratorState<Self::Yield, Self::Return> { (**self).resume() } } diff --git a/src/libcore/ops/index.rs b/src/libcore/ops/index.rs index d65c0aba504..1ac80ecc96f 100644 --- a/src/libcore/ops/index.rs +++ b/src/libcore/ops/index.rs @@ -60,8 +60,14 @@ /// assert_eq!(nucleotide_count[Nucleotide::T], 12); /// ``` #[lang = "index"] -#[rustc_on_unimplemented = "the type `{Self}` cannot be indexed by `{Idx}`"] +#[rustc_on_unimplemented( + message="the type `{Self}` cannot be indexed by `{Idx}`", + label="`{Self}` cannot be indexed by `{Idx}`", +)] #[stable(feature = "rust1", since = "1.0.0")] +#[doc(alias = "]")] +#[doc(alias = "[")] +#[doc(alias = "[]")] pub trait Index<Idx: ?Sized> { /// The returned type after indexing. #[stable(feature = "rust1", since = "1.0.0")] @@ -144,8 +150,14 @@ pub trait Index<Idx: ?Sized> { /// balance[Side::Left] = Weight::Kilogram(3.0); /// ``` #[lang = "index_mut"] -#[rustc_on_unimplemented = "the type `{Self}` cannot be mutably indexed by `{Idx}`"] +#[rustc_on_unimplemented( + message="the type `{Self}` cannot be mutably indexed by `{Idx}`", + label="`{Self}` cannot be mutably indexed by `{Idx}`", +)] #[stable(feature = "rust1", since = "1.0.0")] +#[doc(alias = "[")] +#[doc(alias = "]")] +#[doc(alias = "[]")] pub trait IndexMut<Idx: ?Sized>: Index<Idx> { /// Performs the mutable indexing (`container[index]`) operation. #[stable(feature = "rust1", since = "1.0.0")] diff --git a/src/libcore/ops/mod.rs b/src/libcore/ops/mod.rs index 70ef4487334..ce4f45762de 100644 --- a/src/libcore/ops/mod.rs +++ b/src/libcore/ops/mod.rs @@ -161,7 +161,6 @@ mod drop; mod function; mod generator; mod index; -mod place; mod range; mod try; mod unsize; @@ -191,8 +190,8 @@ pub use self::index::{Index, IndexMut}; #[stable(feature = "rust1", since = "1.0.0")] pub use self::range::{Range, RangeFrom, RangeFull, RangeTo}; -#[unstable(feature = "inclusive_range", reason = "recently added, follows RFC", issue = "28237")] -pub use self::range::{RangeInclusive, RangeToInclusive}; +#[stable(feature = "inclusive_range", since = "1.26.0")] +pub use self::range::{RangeInclusive, RangeToInclusive, RangeBounds, Bound}; #[unstable(feature = "try_trait", issue = "42327")] pub use self::try::Try; @@ -200,8 +199,5 @@ pub use self::try::Try; #[unstable(feature = "generator_trait", issue = "43122")] pub use self::generator::{Generator, GeneratorState}; -#[unstable(feature = "placement_new_protocol", issue = "27779")] -pub use self::place::{Place, Placer, InPlace, Boxed, BoxPlace}; - #[unstable(feature = "coerce_unsized", issue = "27732")] pub use self::unsize::CoerceUnsized; diff --git a/src/libcore/ops/place.rs b/src/libcore/ops/place.rs deleted file mode 100644 index 9fb171e7b92..00000000000 --- a/src/libcore/ops/place.rs +++ /dev/null @@ -1,140 +0,0 @@ -// Copyright 2012 The Rust Project Developers. See the COPYRIGHT -// file at the top-level directory of this distribution and at -// http://rust-lang.org/COPYRIGHT. -// -// Licensed under the Apache License, Version 2.0 <LICENSE-APACHE or -// http://www.apache.org/licenses/LICENSE-2.0> or the MIT license -// <LICENSE-MIT or http://opensource.org/licenses/MIT>, at your -// option. This file may not be copied, modified, or distributed -// except according to those terms. - -/// Both `PLACE <- EXPR` and `box EXPR` desugar into expressions -/// that allocate an intermediate "place" that holds uninitialized -/// state. The desugaring evaluates EXPR, and writes the result at -/// the address returned by the `pointer` method of this trait. -/// -/// A `Place` can be thought of as a special representation for a -/// hypothetical `&uninit` reference (which Rust cannot currently -/// express directly). That is, it represents a pointer to -/// uninitialized storage. -/// -/// The client is responsible for two steps: First, initializing the -/// payload (it can access its address via `pointer`). Second, -/// converting the agent to an instance of the owning pointer, via the -/// appropriate `finalize` method (see the `InPlace`. -/// -/// If evaluating EXPR fails, then it is up to the destructor for the -/// implementation of Place to clean up any intermediate state -/// (e.g. deallocate box storage, pop a stack, etc). -#[unstable(feature = "placement_new_protocol", issue = "27779")] -pub trait Place<Data: ?Sized> { - /// Returns the address where the input value will be written. - /// Note that the data at this address is generally uninitialized, - /// and thus one should use `ptr::write` for initializing it. - fn pointer(&mut self) -> *mut Data; -} - -/// Interface to implementations of `PLACE <- EXPR`. -/// -/// `PLACE <- EXPR` effectively desugars into: -/// -/// ``` -/// # #![feature(placement_new_protocol, box_heap)] -/// # use std::ops::{Placer, Place, InPlace}; -/// # #[allow(non_snake_case)] -/// # fn main() { -/// # let PLACE = std::boxed::HEAP; -/// # let EXPR = 1; -/// let p = PLACE; -/// let mut place = Placer::make_place(p); -/// let raw_place = Place::pointer(&mut place); -/// let value = EXPR; -/// unsafe { -/// std::ptr::write(raw_place, value); -/// InPlace::finalize(place) -/// } -/// # ; } -/// ``` -/// -/// The type of `PLACE <- EXPR` is derived from the type of `PLACE`; -/// if the type of `PLACE` is `P`, then the final type of the whole -/// expression is `P::Place::Owner` (see the `InPlace` and `Boxed` -/// traits). -/// -/// Values for types implementing this trait usually are transient -/// intermediate values (e.g. the return value of `Vec::emplace_back`) -/// or `Copy`, since the `make_place` method takes `self` by value. -#[unstable(feature = "placement_new_protocol", issue = "27779")] -pub trait Placer<Data: ?Sized> { - /// `Place` is the intermediate agent guarding the - /// uninitialized state for `Data`. - type Place: InPlace<Data>; - - /// Creates a fresh place from `self`. - fn make_place(self) -> Self::Place; -} - -/// Specialization of `Place` trait supporting `PLACE <- EXPR`. -#[unstable(feature = "placement_new_protocol", issue = "27779")] -pub trait InPlace<Data: ?Sized>: Place<Data> { - /// `Owner` is the type of the end value of `PLACE <- EXPR` - /// - /// Note that when `PLACE <- EXPR` is solely used for - /// side-effecting an existing data-structure, - /// e.g. `Vec::emplace_back`, then `Owner` need not carry any - /// information at all (e.g. it can be the unit type `()` in that - /// case). - type Owner; - - /// Converts self into the final value, shifting - /// deallocation/cleanup responsibilities (if any remain), over to - /// the returned instance of `Owner` and forgetting self. - unsafe fn finalize(self) -> Self::Owner; -} - -/// Core trait for the `box EXPR` form. -/// -/// `box EXPR` effectively desugars into: -/// -/// ``` -/// # #![feature(placement_new_protocol)] -/// # use std::ops::{BoxPlace, Place, Boxed}; -/// # #[allow(non_snake_case)] -/// # fn main() { -/// # let EXPR = 1; -/// let mut place = BoxPlace::make_place(); -/// let raw_place = Place::pointer(&mut place); -/// let value = EXPR; -/// # let _: Box<_> = -/// unsafe { -/// ::std::ptr::write(raw_place, value); -/// Boxed::finalize(place) -/// } -/// # ; } -/// ``` -/// -/// The type of `box EXPR` is supplied from its surrounding -/// context; in the above expansion, the result type `T` is used -/// to determine which implementation of `Boxed` to use, and that -/// `<T as Boxed>` in turn dictates determines which -/// implementation of `BoxPlace` to use, namely: -/// `<<T as Boxed>::Place as BoxPlace>`. -#[unstable(feature = "placement_new_protocol", issue = "27779")] -pub trait Boxed { - /// The kind of data that is stored in this kind of box. - type Data; /* (`Data` unused b/c cannot yet express below bound.) */ - /// The place that will negotiate the storage of the data. - type Place: BoxPlace<Self::Data>; - - /// Converts filled place into final owning value, shifting - /// deallocation/cleanup responsibilities (if any remain), over to - /// returned instance of `Self` and forgetting `filled`. - unsafe fn finalize(filled: Self::Place) -> Self; -} - -/// Specialization of `Place` trait supporting `box EXPR`. -#[unstable(feature = "placement_new_protocol", issue = "27779")] -pub trait BoxPlace<Data: ?Sized> : Place<Data> { - /// Creates a globally fresh place. - fn make_place() -> Self; -} diff --git a/src/libcore/ops/range.rs b/src/libcore/ops/range.rs index 3f573f7c7eb..9c635678d7a 100644 --- a/src/libcore/ops/range.rs +++ b/src/libcore/ops/range.rs @@ -9,6 +9,7 @@ // except according to those terms. use fmt; +use hash::{Hash, Hasher}; /// An unbounded range (`..`). /// @@ -45,6 +46,7 @@ use fmt; /// [`IntoIterator`]: ../iter/trait.Iterator.html /// [`Iterator`]: ../iter/trait.IntoIterator.html /// [slicing index]: ../slice/trait.SliceIndex.html +#[doc(alias = "..")] #[derive(Copy, Clone, PartialEq, Eq, Hash)] #[stable(feature = "rust1", since = "1.0.0")] pub struct RangeFull; @@ -60,7 +62,7 @@ impl fmt::Debug for RangeFull { /// (`start..end`). /// /// The `Range` `start..end` contains all values with `x >= start` and -/// `x < end`. +/// `x < end`. It is empty unless `start < end`. /// /// # Examples /// @@ -68,12 +70,13 @@ impl fmt::Debug for RangeFull { /// assert_eq!((3..5), std::ops::Range { start: 3, end: 5 }); /// assert_eq!(3 + 4 + 5, (3..6).sum()); /// -/// let arr = [0, 1, 2, 3]; -/// assert_eq!(arr[ .. ], [0,1,2,3]); -/// assert_eq!(arr[ ..3], [0,1,2 ]); -/// assert_eq!(arr[1.. ], [ 1,2,3]); -/// assert_eq!(arr[1..3], [ 1,2 ]); // Range +/// let arr = ['a', 'b', 'c', 'd']; +/// assert_eq!(arr[ .. ], ['a', 'b', 'c', 'd']); +/// assert_eq!(arr[ ..3], ['a', 'b', 'c', ]); +/// assert_eq!(arr[1.. ], [ 'b', 'c', 'd']); +/// assert_eq!(arr[1..3], [ 'b', 'c' ]); // Range /// ``` +#[doc(alias = "..")] #[derive(Clone, PartialEq, Eq, Hash)] // not Copy -- see #27186 #[stable(feature = "rust1", since = "1.0.0")] pub struct Range<Idx> { @@ -92,7 +95,6 @@ impl<Idx: fmt::Debug> fmt::Debug for Range<Idx> { } } -#[unstable(feature = "range_contains", reason = "recently added as per RFC", issue = "32311")] impl<Idx: PartialOrd<Idx>> Range<Idx> { /// Returns `true` if `item` is contained in the range. /// @@ -101,16 +103,55 @@ impl<Idx: PartialOrd<Idx>> Range<Idx> { /// ``` /// #![feature(range_contains)] /// - /// assert!(!(3..5).contains(2)); - /// assert!( (3..5).contains(3)); - /// assert!( (3..5).contains(4)); - /// assert!(!(3..5).contains(5)); + /// use std::f32; + /// + /// assert!(!(3..5).contains(&2)); + /// assert!( (3..5).contains(&3)); + /// assert!( (3..5).contains(&4)); + /// assert!(!(3..5).contains(&5)); + /// + /// assert!(!(3..3).contains(&3)); + /// assert!(!(3..2).contains(&3)); /// - /// assert!(!(3..3).contains(3)); - /// assert!(!(3..2).contains(3)); + /// assert!( (0.0..1.0).contains(&0.5)); + /// assert!(!(0.0..1.0).contains(&f32::NAN)); + /// assert!(!(0.0..f32::NAN).contains(&0.5)); + /// assert!(!(f32::NAN..1.0).contains(&0.5)); /// ``` - pub fn contains(&self, item: Idx) -> bool { - (self.start <= item) && (item < self.end) + #[unstable(feature = "range_contains", reason = "recently added as per RFC", issue = "32311")] + pub fn contains<U>(&self, item: &U) -> bool + where + Idx: PartialOrd<U>, + U: ?Sized + PartialOrd<Idx>, + { + <Self as RangeBounds<Idx>>::contains(self, item) + } + + /// Returns `true` if the range contains no items. + /// + /// # Examples + /// + /// ``` + /// #![feature(range_is_empty)] + /// + /// assert!(!(3..5).is_empty()); + /// assert!( (3..3).is_empty()); + /// assert!( (3..2).is_empty()); + /// ``` + /// + /// The range is empty if either side is incomparable: + /// + /// ``` + /// #![feature(range_is_empty)] + /// + /// use std::f32::NAN; + /// assert!(!(3.0..5.0).is_empty()); + /// assert!( (3.0..NAN).is_empty()); + /// assert!( (NAN..5.0).is_empty()); + /// ``` + #[unstable(feature = "range_is_empty", reason = "recently added", issue = "48111")] + pub fn is_empty(&self) -> bool { + !(self.start < self.end) } } @@ -137,6 +178,7 @@ impl<Idx: PartialOrd<Idx>> Range<Idx> { /// ``` /// /// [`Iterator`]: ../iter/trait.IntoIterator.html +#[doc(alias = "..")] #[derive(Clone, PartialEq, Eq, Hash)] // not Copy -- see #27186 #[stable(feature = "rust1", since = "1.0.0")] pub struct RangeFrom<Idx> { @@ -152,7 +194,6 @@ impl<Idx: fmt::Debug> fmt::Debug for RangeFrom<Idx> { } } -#[unstable(feature = "range_contains", reason = "recently added as per RFC", issue = "32311")] impl<Idx: PartialOrd<Idx>> RangeFrom<Idx> { /// Returns `true` if `item` is contained in the range. /// @@ -161,12 +202,23 @@ impl<Idx: PartialOrd<Idx>> RangeFrom<Idx> { /// ``` /// #![feature(range_contains)] /// - /// assert!(!(3..).contains(2)); - /// assert!( (3..).contains(3)); - /// assert!( (3..).contains(1_000_000_000)); + /// use std::f32; + /// + /// assert!(!(3..).contains(&2)); + /// assert!( (3..).contains(&3)); + /// assert!( (3..).contains(&1_000_000_000)); + /// + /// assert!( (0.0..).contains(&0.5)); + /// assert!(!(0.0..).contains(&f32::NAN)); + /// assert!(!(f32::NAN..).contains(&0.5)); /// ``` - pub fn contains(&self, item: Idx) -> bool { - (self.start <= item) + #[unstable(feature = "range_contains", reason = "recently added as per RFC", issue = "32311")] + pub fn contains<U>(&self, item: &U) -> bool + where + Idx: PartialOrd<U>, + U: ?Sized + PartialOrd<Idx>, + { + <Self as RangeBounds<Idx>>::contains(self, item) } } @@ -208,6 +260,7 @@ impl<Idx: PartialOrd<Idx>> RangeFrom<Idx> { /// [`IntoIterator`]: ../iter/trait.Iterator.html /// [`Iterator`]: ../iter/trait.IntoIterator.html /// [slicing index]: ../slice/trait.SliceIndex.html +#[doc(alias = "..")] #[derive(Copy, Clone, PartialEq, Eq, Hash)] #[stable(feature = "rust1", since = "1.0.0")] pub struct RangeTo<Idx> { @@ -223,7 +276,6 @@ impl<Idx: fmt::Debug> fmt::Debug for RangeTo<Idx> { } } -#[unstable(feature = "range_contains", reason = "recently added as per RFC", issue = "32311")] impl<Idx: PartialOrd<Idx>> RangeTo<Idx> { /// Returns `true` if `item` is contained in the range. /// @@ -232,74 +284,267 @@ impl<Idx: PartialOrd<Idx>> RangeTo<Idx> { /// ``` /// #![feature(range_contains)] /// - /// assert!( (..5).contains(-1_000_000_000)); - /// assert!( (..5).contains(4)); - /// assert!(!(..5).contains(5)); + /// use std::f32; + /// + /// assert!( (..5).contains(&-1_000_000_000)); + /// assert!( (..5).contains(&4)); + /// assert!(!(..5).contains(&5)); + /// + /// assert!( (..1.0).contains(&0.5)); + /// assert!(!(..1.0).contains(&f32::NAN)); + /// assert!(!(..f32::NAN).contains(&0.5)); /// ``` - pub fn contains(&self, item: Idx) -> bool { - (item < self.end) + #[unstable(feature = "range_contains", reason = "recently added as per RFC", issue = "32311")] + pub fn contains<U>(&self, item: &U) -> bool + where + Idx: PartialOrd<U>, + U: ?Sized + PartialOrd<Idx>, + { + <Self as RangeBounds<Idx>>::contains(self, item) } } /// An range bounded inclusively below and above (`start..=end`). /// /// The `RangeInclusive` `start..=end` contains all values with `x >= start` -/// and `x <= end`. +/// and `x <= end`. It is empty unless `start <= end`. +/// +/// This iterator is [fused], but the specific values of `start` and `end` after +/// iteration has finished are **unspecified** other than that [`.is_empty()`] +/// will return `true` once no more values will be produced. +/// +/// [fused]: ../iter/trait.FusedIterator.html +/// [`.is_empty()`]: #method.is_empty /// /// # Examples /// /// ``` -/// #![feature(inclusive_range,inclusive_range_syntax)] -/// -/// assert_eq!((3..=5), std::ops::RangeInclusive { start: 3, end: 5 }); +/// assert_eq!((3..=5), std::ops::RangeInclusive::new(3, 5)); /// assert_eq!(3 + 4 + 5, (3..=5).sum()); /// /// let arr = [0, 1, 2, 3]; /// assert_eq!(arr[ ..=2], [0,1,2 ]); /// assert_eq!(arr[1..=2], [ 1,2 ]); // RangeInclusive /// ``` -#[derive(Clone, PartialEq, Eq, Hash)] // not Copy -- see #27186 -#[unstable(feature = "inclusive_range", reason = "recently added, follows RFC", issue = "28237")] +#[doc(alias = "..=")] +#[derive(Clone)] // not Copy -- see #27186 +#[stable(feature = "inclusive_range", since = "1.26.0")] pub struct RangeInclusive<Idx> { - /// The lower bound of the range (inclusive). - #[unstable(feature = "inclusive_range", - reason = "recently added, follows RFC", - issue = "28237")] - pub start: Idx, - /// The upper bound of the range (inclusive). - #[unstable(feature = "inclusive_range", - reason = "recently added, follows RFC", - issue = "28237")] - pub end: Idx, + pub(crate) start: Idx, + pub(crate) end: Idx, + pub(crate) is_empty: Option<bool>, + // This field is: + // - `None` when next() or next_back() was never called + // - `Some(false)` when `start <= end` assuming no overflow + // - `Some(true)` otherwise + // The field cannot be a simple `bool` because the `..=` constructor can + // accept non-PartialOrd types, also we want the constructor to be const. +} + +trait RangeInclusiveEquality: Sized { + fn canonicalized_is_empty(range: &RangeInclusive<Self>) -> bool; +} +impl<T> RangeInclusiveEquality for T { + #[inline] + default fn canonicalized_is_empty(range: &RangeInclusive<Self>) -> bool { + range.is_empty.unwrap_or_default() + } } +impl<T: PartialOrd> RangeInclusiveEquality for T { + #[inline] + fn canonicalized_is_empty(range: &RangeInclusive<Self>) -> bool { + range.is_empty() + } +} + +#[stable(feature = "inclusive_range", since = "1.26.0")] +impl<Idx: PartialEq> PartialEq for RangeInclusive<Idx> { + #[inline] + fn eq(&self, other: &Self) -> bool { + self.start == other.start && self.end == other.end + && RangeInclusiveEquality::canonicalized_is_empty(self) + == RangeInclusiveEquality::canonicalized_is_empty(other) + } +} + +#[stable(feature = "inclusive_range", since = "1.26.0")] +impl<Idx: Eq> Eq for RangeInclusive<Idx> {} + +#[stable(feature = "inclusive_range", since = "1.26.0")] +impl<Idx: Hash> Hash for RangeInclusive<Idx> { + fn hash<H: Hasher>(&self, state: &mut H) { + self.start.hash(state); + self.end.hash(state); + RangeInclusiveEquality::canonicalized_is_empty(self).hash(state); + } +} + +impl<Idx> RangeInclusive<Idx> { + /// Creates a new inclusive range. Equivalent to writing `start..=end`. + /// + /// # Examples + /// + /// ``` + /// use std::ops::RangeInclusive; + /// + /// assert_eq!(3..=5, RangeInclusive::new(3, 5)); + /// ``` + #[stable(feature = "inclusive_range_methods", since = "1.27.0")] + #[inline] + pub const fn new(start: Idx, end: Idx) -> Self { + Self { start, end, is_empty: None } + } -#[unstable(feature = "inclusive_range", reason = "recently added, follows RFC", issue = "28237")] + /// Returns the lower bound of the range (inclusive). + /// + /// When using an inclusive range for iteration, the values of `start()` and + /// [`end()`] are unspecified after the iteration ended. To determine + /// whether the inclusive range is empty, use the [`is_empty()`] method + /// instead of comparing `start() > end()`. + /// + /// Note: the value returned by this method is unspecified after the range + /// has been iterated to exhaustion. + /// + /// [`end()`]: #method.end + /// [`is_empty()`]: #method.is_empty + /// + /// # Examples + /// + /// ``` + /// assert_eq!((3..=5).start(), &3); + /// ``` + #[stable(feature = "inclusive_range_methods", since = "1.27.0")] + #[inline] + pub fn start(&self) -> &Idx { + &self.start + } + + /// Returns the upper bound of the range (inclusive). + /// + /// When using an inclusive range for iteration, the values of [`start()`] + /// and `end()` are unspecified after the iteration ended. To determine + /// whether the inclusive range is empty, use the [`is_empty()`] method + /// instead of comparing `start() > end()`. + /// + /// Note: the value returned by this method is unspecified after the range + /// has been iterated to exhaustion. + /// + /// [`start()`]: #method.start + /// [`is_empty()`]: #method.is_empty + /// + /// # Examples + /// + /// ``` + /// assert_eq!((3..=5).end(), &5); + /// ``` + #[stable(feature = "inclusive_range_methods", since = "1.27.0")] + #[inline] + pub fn end(&self) -> &Idx { + &self.end + } + + /// Destructures the `RangeInclusive` into (lower bound, upper (inclusive) bound). + /// + /// Note: the value returned by this method is unspecified after the range + /// has been iterated to exhaustion. + /// + /// # Examples + /// + /// ``` + /// assert_eq!((3..=5).into_inner(), (3, 5)); + /// ``` + #[stable(feature = "inclusive_range_methods", since = "1.27.0")] + #[inline] + pub fn into_inner(self) -> (Idx, Idx) { + (self.start, self.end) + } +} + +#[stable(feature = "inclusive_range", since = "1.26.0")] impl<Idx: fmt::Debug> fmt::Debug for RangeInclusive<Idx> { fn fmt(&self, fmt: &mut fmt::Formatter) -> fmt::Result { write!(fmt, "{:?}..={:?}", self.start, self.end) } } -#[unstable(feature = "range_contains", reason = "recently added as per RFC", issue = "32311")] impl<Idx: PartialOrd<Idx>> RangeInclusive<Idx> { /// Returns `true` if `item` is contained in the range. /// /// # Examples /// /// ``` - /// #![feature(range_contains,inclusive_range_syntax)] + /// #![feature(range_contains)] + /// + /// use std::f32; + /// + /// assert!(!(3..=5).contains(&2)); + /// assert!( (3..=5).contains(&3)); + /// assert!( (3..=5).contains(&4)); + /// assert!( (3..=5).contains(&5)); + /// assert!(!(3..=5).contains(&6)); + /// + /// assert!( (3..=3).contains(&3)); + /// assert!(!(3..=2).contains(&3)); + /// + /// assert!( (0.0..=1.0).contains(&1.0)); + /// assert!(!(0.0..=1.0).contains(&f32::NAN)); + /// assert!(!(0.0..=f32::NAN).contains(&0.0)); + /// assert!(!(f32::NAN..=1.0).contains(&1.0)); + /// ``` + #[unstable(feature = "range_contains", reason = "recently added as per RFC", issue = "32311")] + pub fn contains<U>(&self, item: &U) -> bool + where + Idx: PartialOrd<U>, + U: ?Sized + PartialOrd<Idx>, + { + <Self as RangeBounds<Idx>>::contains(self, item) + } + + /// Returns `true` if the range contains no items. + /// + /// # Examples + /// + /// ``` + /// #![feature(range_is_empty)] + /// + /// assert!(!(3..=5).is_empty()); + /// assert!(!(3..=3).is_empty()); + /// assert!( (3..=2).is_empty()); + /// ``` /// - /// assert!(!(3..=5).contains(2)); - /// assert!( (3..=5).contains(3)); - /// assert!( (3..=5).contains(4)); - /// assert!( (3..=5).contains(5)); - /// assert!(!(3..=5).contains(6)); + /// The range is empty if either side is incomparable: /// - /// assert!( (3..=3).contains(3)); - /// assert!(!(3..=2).contains(3)); /// ``` - pub fn contains(&self, item: Idx) -> bool { - self.start <= item && item <= self.end + /// #![feature(range_is_empty)] + /// + /// use std::f32::NAN; + /// assert!(!(3.0..=5.0).is_empty()); + /// assert!( (3.0..=NAN).is_empty()); + /// assert!( (NAN..=5.0).is_empty()); + /// ``` + /// + /// This method returns `true` after iteration has finished: + /// + /// ``` + /// #![feature(range_is_empty)] + /// + /// let mut r = 3..=5; + /// for _ in r.by_ref() {} + /// // Precise field values are unspecified here + /// assert!(r.is_empty()); + /// ``` + #[unstable(feature = "range_is_empty", reason = "recently added", issue = "48111")] + #[inline] + pub fn is_empty(&self) -> bool { + self.is_empty.unwrap_or_else(|| !(self.start <= self.end)) + } + + // If this range's `is_empty` is field is unknown (`None`), update it to be a concrete value. + #[inline] + pub(crate) fn compute_is_empty(&mut self) { + if self.is_empty.is_none() { + self.is_empty = Some(!(self.start <= self.end)); + } } } @@ -313,7 +558,6 @@ impl<Idx: PartialOrd<Idx>> RangeInclusive<Idx> { /// The `..=end` syntax is a `RangeToInclusive`: /// /// ``` -/// #![feature(inclusive_range,inclusive_range_syntax)] /// assert_eq!((..=5), std::ops::RangeToInclusive{ end: 5 }); /// ``` /// @@ -321,8 +565,6 @@ impl<Idx: PartialOrd<Idx>> RangeInclusive<Idx> { /// `for` loop directly. This won't compile: /// /// ```compile_fail,E0277 -/// #![feature(inclusive_range_syntax)] -/// /// // error[E0277]: the trait bound `std::ops::RangeToInclusive<{integer}>: /// // std::iter::Iterator` is not satisfied /// for i in ..=5 { @@ -334,8 +576,6 @@ impl<Idx: PartialOrd<Idx>> RangeInclusive<Idx> { /// array elements up to and including the index indicated by `end`. /// /// ``` -/// #![feature(inclusive_range_syntax)] -/// /// let arr = [0, 1, 2, 3]; /// assert_eq!(arr[ ..=2], [0,1,2 ]); // RangeToInclusive /// assert_eq!(arr[1..=2], [ 1,2 ]); @@ -344,17 +584,16 @@ impl<Idx: PartialOrd<Idx>> RangeInclusive<Idx> { /// [`IntoIterator`]: ../iter/trait.Iterator.html /// [`Iterator`]: ../iter/trait.IntoIterator.html /// [slicing index]: ../slice/trait.SliceIndex.html +#[doc(alias = "..=")] #[derive(Copy, Clone, PartialEq, Eq, Hash)] -#[unstable(feature = "inclusive_range", reason = "recently added, follows RFC", issue = "28237")] +#[stable(feature = "inclusive_range", since = "1.26.0")] pub struct RangeToInclusive<Idx> { /// The upper bound of the range (inclusive) - #[unstable(feature = "inclusive_range", - reason = "recently added, follows RFC", - issue = "28237")] + #[stable(feature = "inclusive_range", since = "1.26.0")] pub end: Idx, } -#[unstable(feature = "inclusive_range", reason = "recently added, follows RFC", issue = "28237")] +#[stable(feature = "inclusive_range", since = "1.26.0")] impl<Idx: fmt::Debug> fmt::Debug for RangeToInclusive<Idx> { fn fmt(&self, fmt: &mut fmt::Formatter) -> fmt::Result { write!(fmt, "..={:?}", self.end) @@ -368,16 +607,295 @@ impl<Idx: PartialOrd<Idx>> RangeToInclusive<Idx> { /// # Examples /// /// ``` - /// #![feature(range_contains,inclusive_range_syntax)] + /// #![feature(range_contains)] + /// + /// use std::f32; /// - /// assert!( (..=5).contains(-1_000_000_000)); - /// assert!( (..=5).contains(5)); - /// assert!(!(..=5).contains(6)); + /// assert!( (..=5).contains(&-1_000_000_000)); + /// assert!( (..=5).contains(&5)); + /// assert!(!(..=5).contains(&6)); + /// + /// assert!( (..=1.0).contains(&1.0)); + /// assert!(!(..=1.0).contains(&f32::NAN)); + /// assert!(!(..=f32::NAN).contains(&0.5)); /// ``` - pub fn contains(&self, item: Idx) -> bool { - (item <= self.end) + #[unstable(feature = "range_contains", reason = "recently added as per RFC", issue = "32311")] + pub fn contains<U>(&self, item: &U) -> bool + where + Idx: PartialOrd<U>, + U: ?Sized + PartialOrd<Idx>, + { + <Self as RangeBounds<Idx>>::contains(self, item) } } // RangeToInclusive<Idx> cannot impl From<RangeTo<Idx>> // because underflow would be possible with (..0).into() + +/// An endpoint of a range of keys. +/// +/// # Examples +/// +/// `Bound`s are range endpoints: +/// +/// ``` +/// use std::ops::Bound::*; +/// use std::ops::RangeBounds; +/// +/// assert_eq!((..100).start_bound(), Unbounded); +/// assert_eq!((1..12).start_bound(), Included(&1)); +/// assert_eq!((1..12).end_bound(), Excluded(&12)); +/// ``` +/// +/// Using a tuple of `Bound`s as an argument to [`BTreeMap::range`]. +/// Note that in most cases, it's better to use range syntax (`1..5`) instead. +/// +/// ``` +/// use std::collections::BTreeMap; +/// use std::ops::Bound::{Excluded, Included, Unbounded}; +/// +/// let mut map = BTreeMap::new(); +/// map.insert(3, "a"); +/// map.insert(5, "b"); +/// map.insert(8, "c"); +/// +/// for (key, value) in map.range((Excluded(3), Included(8))) { +/// println!("{}: {}", key, value); +/// } +/// +/// assert_eq!(Some((&3, &"a")), map.range((Unbounded, Included(5))).next()); +/// ``` +/// +/// [`BTreeMap::range`]: ../../std/collections/btree_map/struct.BTreeMap.html#method.range +#[stable(feature = "collections_bound", since = "1.17.0")] +#[derive(Clone, Copy, Debug, Hash, PartialEq, Eq)] +pub enum Bound<T> { + /// An inclusive bound. + #[stable(feature = "collections_bound", since = "1.17.0")] + Included(#[stable(feature = "collections_bound", since = "1.17.0")] T), + /// An exclusive bound. + #[stable(feature = "collections_bound", since = "1.17.0")] + Excluded(#[stable(feature = "collections_bound", since = "1.17.0")] T), + /// An infinite endpoint. Indicates that there is no bound in this direction. + #[stable(feature = "collections_bound", since = "1.17.0")] + Unbounded, +} + +#[stable(feature = "collections_range", since = "1.28.0")] +/// `RangeBounds` is implemented by Rust's built-in range types, produced +/// by range syntax like `..`, `a..`, `..b` or `c..d`. +pub trait RangeBounds<T: ?Sized> { + /// Start index bound. + /// + /// Returns the start value as a `Bound`. + /// + /// # Examples + /// + /// ``` + /// # fn main() { + /// use std::ops::Bound::*; + /// use std::ops::RangeBounds; + /// + /// assert_eq!((..10).start_bound(), Unbounded); + /// assert_eq!((3..10).start_bound(), Included(&3)); + /// # } + /// ``` + #[stable(feature = "collections_range", since = "1.28.0")] + fn start_bound(&self) -> Bound<&T>; + + /// End index bound. + /// + /// Returns the end value as a `Bound`. + /// + /// # Examples + /// + /// ``` + /// # fn main() { + /// use std::ops::Bound::*; + /// use std::ops::RangeBounds; + /// + /// assert_eq!((3..).end_bound(), Unbounded); + /// assert_eq!((3..10).end_bound(), Excluded(&10)); + /// # } + /// ``` + #[stable(feature = "collections_range", since = "1.28.0")] + fn end_bound(&self) -> Bound<&T>; + + + /// Returns `true` if `item` is contained in the range. + /// + /// # Examples + /// + /// ``` + /// #![feature(range_contains)] + /// + /// use std::f32; + /// + /// assert!( (3..5).contains(&4)); + /// assert!(!(3..5).contains(&2)); + /// + /// assert!( (0.0..1.0).contains(&0.5)); + /// assert!(!(0.0..1.0).contains(&f32::NAN)); + /// assert!(!(0.0..f32::NAN).contains(&0.5)); + /// assert!(!(f32::NAN..1.0).contains(&0.5)); + #[unstable(feature = "range_contains", reason = "recently added as per RFC", issue = "32311")] + fn contains<U>(&self, item: &U) -> bool + where + T: PartialOrd<U>, + U: ?Sized + PartialOrd<T>, + { + (match self.start_bound() { + Included(ref start) => *start <= item, + Excluded(ref start) => *start < item, + Unbounded => true, + }) + && + (match self.end_bound() { + Included(ref end) => item <= *end, + Excluded(ref end) => item < *end, + Unbounded => true, + }) + } +} + +use self::Bound::{Excluded, Included, Unbounded}; + +#[stable(feature = "collections_range", since = "1.28.0")] +impl<T: ?Sized> RangeBounds<T> for RangeFull { + fn start_bound(&self) -> Bound<&T> { + Unbounded + } + fn end_bound(&self) -> Bound<&T> { + Unbounded + } +} + +#[stable(feature = "collections_range", since = "1.28.0")] +impl<T> RangeBounds<T> for RangeFrom<T> { + fn start_bound(&self) -> Bound<&T> { + Included(&self.start) + } + fn end_bound(&self) -> Bound<&T> { + Unbounded + } +} + +#[stable(feature = "collections_range", since = "1.28.0")] +impl<T> RangeBounds<T> for RangeTo<T> { + fn start_bound(&self) -> Bound<&T> { + Unbounded + } + fn end_bound(&self) -> Bound<&T> { + Excluded(&self.end) + } +} + +#[stable(feature = "collections_range", since = "1.28.0")] +impl<T> RangeBounds<T> for Range<T> { + fn start_bound(&self) -> Bound<&T> { + Included(&self.start) + } + fn end_bound(&self) -> Bound<&T> { + Excluded(&self.end) + } +} + +#[stable(feature = "collections_range", since = "1.28.0")] +impl<T> RangeBounds<T> for RangeInclusive<T> { + fn start_bound(&self) -> Bound<&T> { + Included(&self.start) + } + fn end_bound(&self) -> Bound<&T> { + Included(&self.end) + } +} + +#[stable(feature = "collections_range", since = "1.28.0")] +impl<T> RangeBounds<T> for RangeToInclusive<T> { + fn start_bound(&self) -> Bound<&T> { + Unbounded + } + fn end_bound(&self) -> Bound<&T> { + Included(&self.end) + } +} + +#[stable(feature = "collections_range", since = "1.28.0")] +impl<T> RangeBounds<T> for (Bound<T>, Bound<T>) { + fn start_bound(&self) -> Bound<&T> { + match *self { + (Included(ref start), _) => Included(start), + (Excluded(ref start), _) => Excluded(start), + (Unbounded, _) => Unbounded, + } + } + + fn end_bound(&self) -> Bound<&T> { + match *self { + (_, Included(ref end)) => Included(end), + (_, Excluded(ref end)) => Excluded(end), + (_, Unbounded) => Unbounded, + } + } +} + +#[stable(feature = "collections_range", since = "1.28.0")] +impl<'a, T: ?Sized + 'a> RangeBounds<T> for (Bound<&'a T>, Bound<&'a T>) { + fn start_bound(&self) -> Bound<&T> { + self.0 + } + + fn end_bound(&self) -> Bound<&T> { + self.1 + } +} + +#[stable(feature = "collections_range", since = "1.28.0")] +impl<'a, T> RangeBounds<T> for RangeFrom<&'a T> { + fn start_bound(&self) -> Bound<&T> { + Included(self.start) + } + fn end_bound(&self) -> Bound<&T> { + Unbounded + } +} + +#[stable(feature = "collections_range", since = "1.28.0")] +impl<'a, T> RangeBounds<T> for RangeTo<&'a T> { + fn start_bound(&self) -> Bound<&T> { + Unbounded + } + fn end_bound(&self) -> Bound<&T> { + Excluded(self.end) + } +} + +#[stable(feature = "collections_range", since = "1.28.0")] +impl<'a, T> RangeBounds<T> for Range<&'a T> { + fn start_bound(&self) -> Bound<&T> { + Included(self.start) + } + fn end_bound(&self) -> Bound<&T> { + Excluded(self.end) + } +} + +#[stable(feature = "collections_range", since = "1.28.0")] +impl<'a, T> RangeBounds<T> for RangeInclusive<&'a T> { + fn start_bound(&self) -> Bound<&T> { + Included(self.start) + } + fn end_bound(&self) -> Bound<&T> { + Included(self.end) + } +} + +#[stable(feature = "collections_range", since = "1.28.0")] +impl<'a, T> RangeBounds<T> for RangeToInclusive<&'a T> { + fn start_bound(&self) -> Bound<&T> { + Unbounded + } + fn end_bound(&self) -> Bound<&T> { + Included(self.end) + } +} diff --git a/src/libcore/ops/try.rs b/src/libcore/ops/try.rs index 81e5cb5c350..4f2d30aa6a8 100644 --- a/src/libcore/ops/try.rs +++ b/src/libcore/ops/try.rs @@ -20,7 +20,7 @@ any(from_method="from_error", from_method="from_ok"), from_desugaring="?"), message="the `?` operator can only be used in a \ - function that returns `Result` \ + function that returns `Result` or `Option` \ (or another type that implements `{Try}`)", label="cannot use the `?` operator in a function that returns `{Self}`"), on(all(from_method="into_result", from_desugaring="?"), @@ -28,6 +28,7 @@ that implement `{Try}`", label="the `?` operator cannot be applied to type `{Self}`") )] +#[doc(alias = "?")] pub trait Try { /// The type of this value when viewed as successful. #[unstable(feature = "try_trait", issue = "42327")] diff --git a/src/libcore/ops/unsize.rs b/src/libcore/ops/unsize.rs index cd896859b16..da72f374842 100644 --- a/src/libcore/ops/unsize.rs +++ b/src/libcore/ops/unsize.rs @@ -13,7 +13,7 @@ use marker::Unsize; /// Trait that indicates that this is a pointer or a wrapper for one, /// where unsizing can be performed on the pointee. /// -/// See the [DST coercion RfC][dst-coerce] and [the nomicon entry on coercion][nomicon-coerce] +/// See the [DST coercion RFC][dst-coerce] and [the nomicon entry on coercion][nomicon-coerce] /// for more details. /// /// For builtin pointer types, pointers to `T` will coerce to pointers to `U` if `T: Unsize<U>` diff --git a/src/libcore/option.rs b/src/libcore/option.rs index 15181dab853..f743fbfd075 100644 --- a/src/libcore/option.rs +++ b/src/libcore/option.rs @@ -146,7 +146,8 @@ #![stable(feature = "rust1", since = "1.0.0")] use iter::{FromIterator, FusedIterator, TrustedLen}; -use {mem, ops}; +use {hint, mem, ops::{self, Deref}}; +use mem::PinMut; // Note that this is not a lang item per se, but it has a hidden dependency on // `Iterator`, which is one. The compiler assumes that the `next` method of @@ -233,11 +234,11 @@ impl<T> Option<T> { /// [`usize`]: ../../std/primitive.usize.html /// /// ``` - /// let num_as_str: Option<String> = Some("10".to_string()); + /// let text: Option<String> = Some("Hello, world!".to_string()); /// // First, cast `Option<String>` to `Option<&String>` with `as_ref`, - /// // then consume *that* with `map`, leaving `num_as_str` on the stack. - /// let num_as_int: Option<usize> = num_as_str.as_ref().map(|n| n.len()); - /// println!("still can print num_as_str: {:?}", num_as_str); + /// // then consume *that* with `map`, leaving `text` on the stack. + /// let text_length: Option<usize> = text.as_ref().map(|s| s.len()); + /// println!("still can print text: {:?}", text); /// ``` #[inline] #[stable(feature = "rust1", since = "1.0.0")] @@ -269,6 +270,15 @@ impl<T> Option<T> { } } + /// Converts from `Option<T>` to `Option<PinMut<'_, T>>` + #[inline] + #[unstable(feature = "pin", issue = "49150")] + pub fn as_pin_mut<'a>(self: PinMut<'a, Self>) -> Option<PinMut<'a, T>> { + unsafe { + PinMut::get_mut_unchecked(self).as_mut().map(|x| PinMut::new_unchecked(x)) + } + } + ///////////////////////////////////////////////////////////////////////// // Getting to contained values ///////////////////////////////////////////////////////////////////////// @@ -628,8 +638,6 @@ impl<T> Option<T> { /// # Examples /// /// ```rust - /// #![feature(option_filter)] - /// /// fn is_even(n: &i32) -> bool { /// n % 2 == 0 /// } @@ -639,7 +647,7 @@ impl<T> Option<T> { /// assert_eq!(Some(4).filter(is_even), Some(4)); /// ``` #[inline] - #[unstable(feature = "option_filter", issue = "45860")] + #[stable(feature = "option_filter", since = "1.27.0")] pub fn filter<P: FnOnce(&T) -> bool>(self, predicate: P) -> Self { if let Some(x) = self { if predicate(&x) { @@ -707,6 +715,42 @@ impl<T> Option<T> { } } + /// Returns [`Some`] if exactly one of `self`, `optb` is [`Some`], otherwise returns `None`. + /// + /// [`Some`]: #variant.Some + /// [`None`]: #variant.None + /// + /// # Examples + /// + /// ``` + /// #![feature(option_xor)] + /// + /// let x = Some(2); + /// let y: Option<u32> = None; + /// assert_eq!(x.xor(y), Some(2)); + /// + /// let x: Option<u32> = None; + /// let y = Some(2); + /// assert_eq!(x.xor(y), Some(2)); + /// + /// let x = Some(2); + /// let y = Some(2); + /// assert_eq!(x.xor(y), None); + /// + /// let x: Option<u32> = None; + /// let y: Option<u32> = None; + /// assert_eq!(x.xor(y), None); + /// ``` + #[inline] + #[unstable(feature = "option_xor", issue = "50512")] + pub fn xor(self, optb: Option<T>) -> Option<T> { + match (self, optb) { + (Some(a), None) => Some(a), + (None, Some(b)) => Some(b), + _ => None, + } + } + ///////////////////////////////////////////////////////////////////////// // Entry-like operations to insert if None and return a reference ///////////////////////////////////////////////////////////////////////// @@ -740,7 +784,7 @@ impl<T> Option<T> { match *self { Some(ref mut v) => v, - _ => unreachable!(), + None => unsafe { hint::unreachable_unchecked() }, } } @@ -773,7 +817,7 @@ impl<T> Option<T> { match *self { Some(ref mut v) => v, - _ => unreachable!(), + None => unsafe { hint::unreachable_unchecked() }, } } @@ -789,18 +833,47 @@ impl<T> Option<T> { /// /// ``` /// let mut x = Some(2); - /// x.take(); + /// let y = x.take(); /// assert_eq!(x, None); + /// assert_eq!(y, Some(2)); /// /// let mut x: Option<u32> = None; - /// x.take(); + /// let y = x.take(); /// assert_eq!(x, None); + /// assert_eq!(y, None); /// ``` #[inline] #[stable(feature = "rust1", since = "1.0.0")] pub fn take(&mut self) -> Option<T> { mem::replace(self, None) } + + /// Replaces the actual value in the option by the value given in parameter, + /// returning the old value if present, + /// leaving a [`Some`] in its place without deinitializing either one. + /// + /// [`Some`]: #variant.Some + /// + /// # Examples + /// + /// ``` + /// #![feature(option_replace)] + /// + /// let mut x = Some(2); + /// let old = x.replace(5); + /// assert_eq!(x, Some(5)); + /// assert_eq!(old, Some(2)); + /// + /// let mut x = None; + /// let old = x.replace(3); + /// assert_eq!(x, Some(3)); + /// assert_eq!(old, None); + /// ``` + #[inline] + #[unstable(feature = "option_replace", issue = "51998")] + pub fn replace(&mut self, value: T) -> Option<T> { + mem::replace(self, Some(value)) + } } impl<'a, T: Clone> Option<&'a T> { @@ -829,14 +902,13 @@ impl<'a, T: Clone> Option<&'a mut T> { /// # Examples /// /// ``` - /// #![feature(option_ref_mut_cloned)] /// let mut x = 12; /// let opt_x = Some(&mut x); /// assert_eq!(opt_x, Some(&mut 12)); /// let cloned = opt_x.cloned(); /// assert_eq!(cloned, Some(12)); /// ``` - #[unstable(feature = "option_ref_mut_cloned", issue = "43738")] + #[stable(since = "1.26.0", feature = "option_ref_mut_cloned")] pub fn cloned(self) -> Option<T> { self.map(|t| t.clone()) } @@ -881,6 +953,46 @@ impl<T: Default> Option<T> { } } +#[unstable(feature = "inner_deref", reason = "newly added", issue = "50264")] +impl<T: Deref> Option<T> { + /// Converts from `&Option<T>` to `Option<&T::Target>`. + /// + /// Leaves the original Option in-place, creating a new one with a reference + /// to the original one, additionally coercing the contents via `Deref`. + pub fn deref(&self) -> Option<&T::Target> { + self.as_ref().map(|t| t.deref()) + } +} + +impl<T, E> Option<Result<T, E>> { + /// Transposes an `Option` of a `Result` into a `Result` of an `Option`. + /// + /// `None` will be mapped to `Ok(None)`. + /// `Some(Ok(_))` and `Some(Err(_))` will be mapped to `Ok(Some(_))` and `Err(_)`. + /// + /// # Examples + /// + /// ``` + /// #![feature(transpose_result)] + /// + /// #[derive(Debug, Eq, PartialEq)] + /// struct SomeErr; + /// + /// let x: Result<Option<i32>, SomeErr> = Ok(Some(5)); + /// let y: Option<Result<i32, SomeErr>> = Some(Ok(5)); + /// assert_eq!(x, y.transpose()); + /// ``` + #[inline] + #[unstable(feature = "transpose_result", issue = "47338")] + pub fn transpose(self) -> Result<Option<T>, E> { + match self { + Some(Ok(x)) => Ok(Some(x)), + Some(Err(e)) => Err(e), + None => Ok(None), + } + } +} + // This is a separate function to reduce the code size of .expect() itself. #[inline(never)] #[cold] @@ -888,7 +1000,6 @@ fn expect_failed(msg: &str) -> ! { panic!("{}", msg) } - ///////////////////////////////////////////////////////////////////////////// // Trait implementations ///////////////////////////////////////////////////////////////////////////// @@ -1022,7 +1133,7 @@ impl<'a, A> DoubleEndedIterator for Iter<'a, A> { #[stable(feature = "rust1", since = "1.0.0")] impl<'a, A> ExactSizeIterator for Iter<'a, A> {} -#[unstable(feature = "fused", issue = "35602")] +#[stable(feature = "fused", since = "1.26.0")] impl<'a, A> FusedIterator for Iter<'a, A> {} #[unstable(feature = "trusted_len", issue = "37572")] @@ -1030,6 +1141,7 @@ unsafe impl<'a, A> TrustedLen for Iter<'a, A> {} #[stable(feature = "rust1", since = "1.0.0")] impl<'a, A> Clone for Iter<'a, A> { + #[inline] fn clone(&self) -> Iter<'a, A> { Iter { inner: self.inner.clone() } } @@ -1067,7 +1179,7 @@ impl<'a, A> DoubleEndedIterator for IterMut<'a, A> { #[stable(feature = "rust1", since = "1.0.0")] impl<'a, A> ExactSizeIterator for IterMut<'a, A> {} -#[unstable(feature = "fused", issue = "35602")] +#[stable(feature = "fused", since = "1.26.0")] impl<'a, A> FusedIterator for IterMut<'a, A> {} #[unstable(feature = "trusted_len", issue = "37572")] unsafe impl<'a, A> TrustedLen for IterMut<'a, A> {} @@ -1104,7 +1216,7 @@ impl<A> DoubleEndedIterator for IntoIter<A> { #[stable(feature = "rust1", since = "1.0.0")] impl<A> ExactSizeIterator for IntoIter<A> {} -#[unstable(feature = "fused", issue = "35602")] +#[stable(feature = "fused", since = "1.26.0")] impl<A> FusedIterator for IntoIter<A> {} #[unstable(feature = "trusted_len", issue = "37572")] @@ -1160,6 +1272,16 @@ impl<A, V: FromIterator<A>> FromIterator<Option<A>> for Option<V> { None => None, } } + + #[inline] + fn size_hint(&self) -> (usize, Option<usize>) { + if self.found_none { + (0, Some(0)) + } else { + let (_, upper) = self.iter.size_hint(); + (0, upper) + } + } } let mut adapter = Adapter { iter: iter.into_iter(), found_none: false }; @@ -1186,14 +1308,17 @@ impl<T> ops::Try for Option<T> { type Ok = T; type Error = NoneError; + #[inline] fn into_result(self) -> Result<T, NoneError> { self.ok_or(NoneError) } + #[inline] fn from_ok(v: T) -> Self { Some(v) } + #[inline] fn from_error(_: NoneError) -> Self { None } diff --git a/src/libcore/panic.rs b/src/libcore/panic.rs new file mode 100644 index 00000000000..17cac5aa0a0 --- /dev/null +++ b/src/libcore/panic.rs @@ -0,0 +1,275 @@ +// Copyright 2018 The Rust Project Developers. See the COPYRIGHT +// file at the top-level directory of this distribution and at +// http://rust-lang.org/COPYRIGHT. +// +// Licensed under the Apache License, Version 2.0 <LICENSE-APACHE or +// http://www.apache.org/licenses/LICENSE-2.0> or the MIT license +// <LICENSE-MIT or http://opensource.org/licenses/MIT>, at your +// option. This file may not be copied, modified, or distributed +// except according to those terms. + +//! Panic support in the standard library. + +#![unstable(feature = "core_panic_info", + reason = "newly available in libcore", + issue = "44489")] + +use any::Any; +use fmt; + +/// A struct providing information about a panic. +/// +/// `PanicInfo` structure is passed to a panic hook set by the [`set_hook`] +/// function. +/// +/// [`set_hook`]: ../../std/panic/fn.set_hook.html +/// +/// # Examples +/// +/// ```should_panic +/// use std::panic; +/// +/// panic::set_hook(Box::new(|panic_info| { +/// if let Some(s) = panic_info.payload().downcast_ref::<&str>() { +/// println!("panic occurred: {:?}", s); +/// } else { +/// println!("panic occurred"); +/// } +/// })); +/// +/// panic!("Normal panic"); +/// ``` +#[lang = "panic_info"] +#[stable(feature = "panic_hooks", since = "1.10.0")] +#[derive(Debug)] +pub struct PanicInfo<'a> { + payload: &'a (dyn Any + Send), + message: Option<&'a fmt::Arguments<'a>>, + location: Location<'a>, +} + +impl<'a> PanicInfo<'a> { + #![unstable(feature = "panic_internals", + reason = "internal details of the implementation of the `panic!` \ + and related macros", + issue = "0")] + #[doc(hidden)] + #[inline] + pub fn internal_constructor(message: Option<&'a fmt::Arguments<'a>>, + location: Location<'a>) + -> Self { + struct NoPayload; + PanicInfo { payload: &NoPayload, location, message } + } + + #[doc(hidden)] + #[inline] + pub fn set_payload(&mut self, info: &'a (dyn Any + Send)) { + self.payload = info; + } + + /// Returns the payload associated with the panic. + /// + /// This will commonly, but not always, be a `&'static str` or [`String`]. + /// + /// [`String`]: ../../std/string/struct.String.html + /// + /// # Examples + /// + /// ```should_panic + /// use std::panic; + /// + /// panic::set_hook(Box::new(|panic_info| { + /// println!("panic occurred: {:?}", panic_info.payload().downcast_ref::<&str>().unwrap()); + /// })); + /// + /// panic!("Normal panic"); + /// ``` + #[stable(feature = "panic_hooks", since = "1.10.0")] + pub fn payload(&self) -> &(dyn Any + Send) { + self.payload + } + + /// If the `panic!` macro from the `core` crate (not from `std`) + /// was used with a formatting string and some additional arguments, + /// returns that message ready to be used for example with [`fmt::write`] + /// + /// [`fmt::write`]: ../fmt/fn.write.html + #[unstable(feature = "panic_info_message", issue = "44489")] + pub fn message(&self) -> Option<&fmt::Arguments> { + self.message + } + + /// Returns information about the location from which the panic originated, + /// if available. + /// + /// This method will currently always return [`Some`], but this may change + /// in future versions. + /// + /// [`Some`]: ../../std/option/enum.Option.html#variant.Some + /// + /// # Examples + /// + /// ```should_panic + /// use std::panic; + /// + /// panic::set_hook(Box::new(|panic_info| { + /// if let Some(location) = panic_info.location() { + /// println!("panic occurred in file '{}' at line {}", location.file(), + /// location.line()); + /// } else { + /// println!("panic occurred but can't get location information..."); + /// } + /// })); + /// + /// panic!("Normal panic"); + /// ``` + #[stable(feature = "panic_hooks", since = "1.10.0")] + pub fn location(&self) -> Option<&Location> { + // NOTE: If this is changed to sometimes return None, + // deal with that case in std::panicking::default_hook and std::panicking::begin_panic_fmt. + Some(&self.location) + } +} + +#[stable(feature = "panic_hook_display", since = "1.26.0")] +impl<'a> fmt::Display for PanicInfo<'a> { + fn fmt(&self, formatter: &mut fmt::Formatter) -> fmt::Result { + formatter.write_str("panicked at ")?; + if let Some(message) = self.message { + write!(formatter, "'{}', ", message)? + } else if let Some(payload) = self.payload.downcast_ref::<&'static str>() { + write!(formatter, "'{}', ", payload)? + } + // NOTE: we cannot use downcast_ref::<String>() here + // since String is not available in libcore! + // The payload is a String when `std::panic!` is called with multiple arguments, + // but in that case the message is also available. + + self.location.fmt(formatter) + } +} + +/// A struct containing information about the location of a panic. +/// +/// This structure is created by the [`location`] method of [`PanicInfo`]. +/// +/// [`location`]: ../../std/panic/struct.PanicInfo.html#method.location +/// [`PanicInfo`]: ../../std/panic/struct.PanicInfo.html +/// +/// # Examples +/// +/// ```should_panic +/// use std::panic; +/// +/// panic::set_hook(Box::new(|panic_info| { +/// if let Some(location) = panic_info.location() { +/// println!("panic occurred in file '{}' at line {}", location.file(), location.line()); +/// } else { +/// println!("panic occurred but can't get location information..."); +/// } +/// })); +/// +/// panic!("Normal panic"); +/// ``` +#[derive(Debug)] +#[stable(feature = "panic_hooks", since = "1.10.0")] +pub struct Location<'a> { + file: &'a str, + line: u32, + col: u32, +} + +impl<'a> Location<'a> { + #![unstable(feature = "panic_internals", + reason = "internal details of the implementation of the `panic!` \ + and related macros", + issue = "0")] + #[doc(hidden)] + pub fn internal_constructor(file: &'a str, line: u32, col: u32) -> Self { + Location { file, line, col } + } + + /// Returns the name of the source file from which the panic originated. + /// + /// # Examples + /// + /// ```should_panic + /// use std::panic; + /// + /// panic::set_hook(Box::new(|panic_info| { + /// if let Some(location) = panic_info.location() { + /// println!("panic occurred in file '{}'", location.file()); + /// } else { + /// println!("panic occurred but can't get location information..."); + /// } + /// })); + /// + /// panic!("Normal panic"); + /// ``` + #[stable(feature = "panic_hooks", since = "1.10.0")] + pub fn file(&self) -> &str { + self.file + } + + /// Returns the line number from which the panic originated. + /// + /// # Examples + /// + /// ```should_panic + /// use std::panic; + /// + /// panic::set_hook(Box::new(|panic_info| { + /// if let Some(location) = panic_info.location() { + /// println!("panic occurred at line {}", location.line()); + /// } else { + /// println!("panic occurred but can't get location information..."); + /// } + /// })); + /// + /// panic!("Normal panic"); + /// ``` + #[stable(feature = "panic_hooks", since = "1.10.0")] + pub fn line(&self) -> u32 { + self.line + } + + /// Returns the column from which the panic originated. + /// + /// # Examples + /// + /// ```should_panic + /// use std::panic; + /// + /// panic::set_hook(Box::new(|panic_info| { + /// if let Some(location) = panic_info.location() { + /// println!("panic occurred at column {}", location.column()); + /// } else { + /// println!("panic occurred but can't get location information..."); + /// } + /// })); + /// + /// panic!("Normal panic"); + /// ``` + #[stable(feature = "panic_col", since = "1.25.0")] + pub fn column(&self) -> u32 { + self.col + } +} + +#[stable(feature = "panic_hook_display", since = "1.26.0")] +impl<'a> fmt::Display for Location<'a> { + fn fmt(&self, formatter: &mut fmt::Formatter) -> fmt::Result { + write!(formatter, "{}:{}:{}", self.file, self.line, self.col) + } +} + +/// An internal trait used by libstd to pass data from libstd to `panic_unwind` +/// and other panic runtimes. Not intended to be stabilized any time soon, do +/// not use. +#[unstable(feature = "std_internals", issue = "0")] +#[doc(hidden)] +pub unsafe trait BoxMeUp { + fn box_me_up(&mut self) -> *mut (dyn Any + Send); + fn get(&mut self) -> &(dyn Any + Send); +} diff --git a/src/libcore/panicking.rs b/src/libcore/panicking.rs index 4170d91e5fc..58407de9566 100644 --- a/src/libcore/panicking.rs +++ b/src/libcore/panicking.rs @@ -37,6 +37,7 @@ issue = "0")] use fmt; +use panic::{Location, PanicInfo}; #[cold] #[inline(never)] // this is the slow path, always #[lang = "panic"] @@ -61,12 +62,17 @@ fn panic_bounds_check(file_line_col: &(&'static str, u32, u32), #[cold] #[inline(never)] pub fn panic_fmt(fmt: fmt::Arguments, file_line_col: &(&'static str, u32, u32)) -> ! { - #[allow(improper_ctypes)] - extern { - #[lang = "panic_fmt"] - #[unwind] - fn panic_impl(fmt: fmt::Arguments, file: &'static str, line: u32, col: u32) -> !; + // NOTE This function never crosses the FFI boundary; it's a Rust-to-Rust call + #[allow(improper_ctypes)] // PanicInfo contains a trait object which is not FFI safe + extern "Rust" { + #[lang = "panic_impl"] + fn panic_impl(pi: &PanicInfo) -> !; } + let (file, line, col) = *file_line_col; - unsafe { panic_impl(fmt, file, line, col) } + let pi = PanicInfo::internal_constructor( + Some(&fmt), + Location::internal_constructor(file, line, col), + ); + unsafe { panic_impl(&pi) } } diff --git a/src/libcore/prelude/v1.rs b/src/libcore/prelude/v1.rs index 6f4f273a31f..7c5738e94e5 100644 --- a/src/libcore/prelude/v1.rs +++ b/src/libcore/prelude/v1.rs @@ -57,14 +57,3 @@ pub use option::Option::{self, Some, None}; #[stable(feature = "core_prelude", since = "1.4.0")] #[doc(no_inline)] pub use result::Result::{self, Ok, Err}; - -// Re-exported extension traits for primitive types -#[stable(feature = "core_prelude", since = "1.4.0")] -#[doc(no_inline)] -pub use slice::SliceExt; -#[stable(feature = "core_prelude", since = "1.4.0")] -#[doc(no_inline)] -pub use str::StrExt; -#[stable(feature = "core_prelude", since = "1.4.0")] -#[doc(no_inline)] -pub use char::CharExt; diff --git a/src/libcore/ptr.rs b/src/libcore/ptr.rs index 7f7246df8f2..61033e75112 100644 --- a/src/libcore/ptr.rs +++ b/src/libcore/ptr.rs @@ -166,8 +166,6 @@ pub unsafe fn swap<T>(x: *mut T, y: *mut T) { /// Basic usage: /// /// ``` -/// #![feature(swap_nonoverlapping)] -/// /// use std::ptr; /// /// let mut x = [1, 2, 3, 4]; @@ -181,7 +179,7 @@ pub unsafe fn swap<T>(x: *mut T, y: *mut T) { /// assert_eq!(y, [1, 2, 9]); /// ``` #[inline] -#[unstable(feature = "swap_nonoverlapping", issue = "42818")] +#[stable(feature = "swap_nonoverlapping", since = "1.27.0")] pub unsafe fn swap_nonoverlapping<T>(x: *mut T, y: *mut T, count: usize) { let x = x as *mut u8; let y = y as *mut u8; @@ -190,6 +188,19 @@ pub unsafe fn swap_nonoverlapping<T>(x: *mut T, y: *mut T, count: usize) { } #[inline] +pub(crate) unsafe fn swap_nonoverlapping_one<T>(x: *mut T, y: *mut T) { + // For types smaller than the block optimization below, + // just swap directly to avoid pessimizing codegen. + if mem::size_of::<T>() < 32 { + let z = read(x); + copy_nonoverlapping(y, x, 1); + write(y, z); + } else { + swap_nonoverlapping(x, y, 1); + } +} + +#[inline] unsafe fn swap_nonoverlapping_bytes(x: *mut u8, y: *mut u8, len: usize) { // The approach here is to utilize simd to swap x & y efficiently. Testing reveals // that swapping either 32 bytes or 64 bytes at a time is most efficient for intel @@ -241,8 +252,9 @@ unsafe fn swap_nonoverlapping_bytes(x: *mut u8, y: *mut u8, len: usize) { } } -/// Replaces the value at `dest` with `src`, returning the old -/// value, without dropping either. +/// Moves `src` into the pointed `dest`, returning the previous `dest` value. +/// +/// Neither value is dropped. /// /// # Safety /// @@ -436,6 +448,12 @@ pub unsafe fn write_unaligned<T>(dst: *mut T, src: T) { /// `write_bytes`, or `copy`). Note that `*src = foo` counts as a use /// because it will attempt to drop the value previously at `*src`. /// +/// Just like in C, whether an operation is volatile has no bearing whatsoever +/// on questions involving concurrent access from multiple threads. Volatile +/// accesses behave exactly like non-atomic accesses in that regard. In particular, +/// a race between a `read_volatile` and any write operation to the same location +/// is undefined behavior. +/// /// # Examples /// /// Basic usage: @@ -486,6 +504,12 @@ pub unsafe fn read_volatile<T>(src: *const T) -> T { /// This is appropriate for initializing uninitialized memory, or overwriting /// memory that has previously been `read` from. /// +/// Just like in C, whether an operation is volatile has no bearing whatsoever +/// on questions involving concurrent access from multiple threads. Volatile +/// accesses behave exactly like non-atomic accesses in that regard. In particular, +/// a race between a `write_volatile` and any other operation (reading or writing) +/// on the same location is undefined behavior. +/// /// # Examples /// /// Basic usage: @@ -579,7 +603,7 @@ impl<T: ?Sized> *const T { /// Behavior: /// /// * Both the starting and resulting pointer must be either in bounds or one - /// byte past the end of an allocated object. + /// byte past the end of *the same* allocated object. /// /// * The computed offset, **in bytes**, cannot overflow an `isize`. /// @@ -631,9 +655,15 @@ impl<T: ?Sized> *const T { /// /// The resulting pointer does not need to be in bounds, but it is /// potentially hazardous to dereference (which requires `unsafe`). + /// In particular, the resulting pointer may *not* be used to access a + /// different allocated object than the one `self` points to. In other + /// words, `x.wrapping_offset(y.wrapping_offset_from(x))` is + /// *not* the same as `y`, and dereferencing it is undefined behavior + /// unless `x` and `y` point into the same allocated object. /// /// Always use `.offset(count)` instead when possible, because `offset` - /// allows the compiler to optimize better. + /// allows the compiler to optimize better. If you need to cross object + /// boundaries, cast the pointer to an integer and do the arithmetic there. /// /// # Examples /// @@ -665,39 +695,119 @@ impl<T: ?Sized> *const T { /// Calculates the distance between two pointers. The returned value is in /// units of T: the distance in bytes is divided by `mem::size_of::<T>()`. /// - /// If the address different between the two pointers ia not a multiple of + /// This function is the inverse of [`offset`]. + /// + /// [`offset`]: #method.offset + /// [`wrapping_offset_from`]: #method.wrapping_offset_from + /// + /// # Safety + /// + /// If any of the following conditions are violated, the result is Undefined + /// Behavior: + /// + /// * Both the starting and other pointer must be either in bounds or one + /// byte past the end of the same allocated object. + /// + /// * The distance between the pointers, **in bytes**, cannot overflow an `isize`. + /// + /// * The distance between the pointers, in bytes, must be an exact multiple + /// of the size of `T`. + /// + /// * The distance being in bounds cannot rely on "wrapping around" the address space. + /// + /// The compiler and standard library generally try to ensure allocations + /// never reach a size where an offset is a concern. For instance, `Vec` + /// and `Box` ensure they never allocate more than `isize::MAX` bytes, so + /// `ptr_into_vec.offset_from(vec.as_ptr())` is always safe. + /// + /// Most platforms fundamentally can't even construct such an allocation. + /// For instance, no known 64-bit platform can ever serve a request + /// for 2<sup>63</sup> bytes due to page-table limitations or splitting the address space. + /// However, some 32-bit and 16-bit platforms may successfully serve a request for + /// more than `isize::MAX` bytes with things like Physical Address + /// Extension. As such, memory acquired directly from allocators or memory + /// mapped files *may* be too large to handle with this function. + /// + /// Consider using [`wrapping_offset_from`] instead if these constraints are + /// difficult to satisfy. The only advantage of this method is that it + /// enables more aggressive compiler optimizations. + /// + /// # Panics + /// + /// This function panics if `T` is a Zero-Sized Type ("ZST"). + /// + /// # Examples + /// + /// Basic usage: + /// + /// ``` + /// #![feature(ptr_offset_from)] + /// + /// let a = [0; 5]; + /// let ptr1: *const i32 = &a[1]; + /// let ptr2: *const i32 = &a[3]; + /// unsafe { + /// assert_eq!(ptr2.offset_from(ptr1), 2); + /// assert_eq!(ptr1.offset_from(ptr2), -2); + /// assert_eq!(ptr1.offset(2), ptr2); + /// assert_eq!(ptr2.offset(-2), ptr1); + /// } + /// ``` + #[unstable(feature = "ptr_offset_from", issue = "41079")] + #[inline] + pub unsafe fn offset_from(self, origin: *const T) -> isize where T: Sized { + let pointee_size = mem::size_of::<T>(); + assert!(0 < pointee_size && pointee_size <= isize::max_value() as usize); + + // This is the same sequence that Clang emits for pointer subtraction. + // It can be neither `nsw` nor `nuw` because the input is treated as + // unsigned but then the output is treated as signed, so neither works. + let d = isize::wrapping_sub(self as _, origin as _); + intrinsics::exact_div(d, pointee_size as _) + } + + /// Calculates the distance between two pointers. The returned value is in + /// units of T: the distance in bytes is divided by `mem::size_of::<T>()`. + /// + /// If the address different between the two pointers is not a multiple of /// `mem::size_of::<T>()` then the result of the division is rounded towards /// zero. /// - /// This function returns `None` if `T` is a zero-sized typed. + /// Though this method is safe for any two pointers, note that its result + /// will be mostly useless if the two pointers aren't into the same allocated + /// object, for example if they point to two different local variables. + /// + /// # Panics + /// + /// This function panics if `T` is a zero-sized type. /// /// # Examples /// /// Basic usage: /// /// ``` - /// #![feature(offset_to)] + /// #![feature(ptr_wrapping_offset_from)] /// - /// fn main() { - /// let a = [0; 5]; - /// let ptr1: *const i32 = &a[1]; - /// let ptr2: *const i32 = &a[3]; - /// assert_eq!(ptr1.offset_to(ptr2), Some(2)); - /// assert_eq!(ptr2.offset_to(ptr1), Some(-2)); - /// assert_eq!(unsafe { ptr1.offset(2) }, ptr2); - /// assert_eq!(unsafe { ptr2.offset(-2) }, ptr1); - /// } + /// let a = [0; 5]; + /// let ptr1: *const i32 = &a[1]; + /// let ptr2: *const i32 = &a[3]; + /// assert_eq!(ptr2.wrapping_offset_from(ptr1), 2); + /// assert_eq!(ptr1.wrapping_offset_from(ptr2), -2); + /// assert_eq!(ptr1.wrapping_offset(2), ptr2); + /// assert_eq!(ptr2.wrapping_offset(-2), ptr1); + /// + /// let ptr1: *const i32 = 3 as _; + /// let ptr2: *const i32 = 13 as _; + /// assert_eq!(ptr2.wrapping_offset_from(ptr1), 2); /// ``` - #[unstable(feature = "offset_to", issue = "41079")] + #[unstable(feature = "ptr_wrapping_offset_from", issue = "41079")] #[inline] - pub fn offset_to(self, other: *const T) -> Option<isize> where T: Sized { - let size = mem::size_of::<T>(); - if size == 0 { - None - } else { - let diff = (other as isize).wrapping_sub(self as isize); - Some(diff / size as isize) - } + pub fn wrapping_offset_from(self, origin: *const T) -> isize where T: Sized { + let pointee_size = mem::size_of::<T>(); + assert!(0 < pointee_size && pointee_size <= isize::max_value() as usize); + + let d = isize::wrapping_sub(self as _, origin as _); + d.wrapping_div(pointee_size as _) } /// Calculates the offset from a pointer (convenience for `.offset(count as isize)`). @@ -740,8 +850,6 @@ impl<T: ?Sized> *const T { /// Basic usage: /// /// ``` - /// #![feature(pointer_methods)] - /// /// let s: &str = "123"; /// let ptr: *const u8 = s.as_ptr(); /// @@ -750,7 +858,7 @@ impl<T: ?Sized> *const T { /// println!("{}", *ptr.add(2) as char); /// } /// ``` - #[unstable(feature = "pointer_methods", issue = "43941")] + #[stable(feature = "pointer_methods", since = "1.26.0")] #[inline] pub unsafe fn add(self, count: usize) -> Self where T: Sized, @@ -799,8 +907,6 @@ impl<T: ?Sized> *const T { /// Basic usage: /// /// ``` - /// #![feature(pointer_methods)] - /// /// let s: &str = "123"; /// /// unsafe { @@ -809,7 +915,7 @@ impl<T: ?Sized> *const T { /// println!("{}", *end.sub(2) as char); /// } /// ``` - #[unstable(feature = "pointer_methods", issue = "43941")] + #[stable(feature = "pointer_methods", since = "1.26.0")] #[inline] pub unsafe fn sub(self, count: usize) -> Self where T: Sized, @@ -836,8 +942,6 @@ impl<T: ?Sized> *const T { /// Basic usage: /// /// ``` - /// #![feature(pointer_methods)] - /// /// // Iterate using a raw pointer in increments of two elements /// let data = [1u8, 2, 3, 4, 5]; /// let mut ptr: *const u8 = data.as_ptr(); @@ -852,7 +956,7 @@ impl<T: ?Sized> *const T { /// ptr = ptr.wrapping_add(step); /// } /// ``` - #[unstable(feature = "pointer_methods", issue = "43941")] + #[stable(feature = "pointer_methods", since = "1.26.0")] #[inline] pub fn wrapping_add(self, count: usize) -> Self where T: Sized, @@ -879,8 +983,6 @@ impl<T: ?Sized> *const T { /// Basic usage: /// /// ``` - /// #![feature(pointer_methods)] - /// /// // Iterate using a raw pointer in increments of two elements (backwards) /// let data = [1u8, 2, 3, 4, 5]; /// let mut ptr: *const u8 = data.as_ptr(); @@ -895,7 +997,7 @@ impl<T: ?Sized> *const T { /// ptr = ptr.wrapping_sub(step); /// } /// ``` - #[unstable(feature = "pointer_methods", issue = "43941")] + #[stable(feature = "pointer_methods", since = "1.26.0")] #[inline] pub fn wrapping_sub(self, count: usize) -> Self where T: Sized, @@ -922,8 +1024,6 @@ impl<T: ?Sized> *const T { /// Basic usage: /// /// ``` - /// #![feature(pointer_methods)] - /// /// let x = 12; /// let y = &x as *const i32; /// @@ -931,7 +1031,7 @@ impl<T: ?Sized> *const T { /// assert_eq!(y.read(), 12); /// } /// ``` - #[unstable(feature = "pointer_methods", issue = "43941")] + #[stable(feature = "pointer_methods", since = "1.26.0")] #[inline] pub unsafe fn read(self) -> T where T: Sized, @@ -969,13 +1069,17 @@ impl<T: ?Sized> *const T { /// `write_bytes`, or `copy`). Note that `*self = foo` counts as a use /// because it will attempt to drop the value previously at `*self`. /// + /// Just like in C, whether an operation is volatile has no bearing whatsoever + /// on questions involving concurrent access from multiple threads. Volatile + /// accesses behave exactly like non-atomic accesses in that regard. In particular, + /// a race between a `read_volatile` and any write operation to the same location + /// is undefined behavior. + /// /// # Examples /// /// Basic usage: /// /// ``` - /// #![feature(pointer_methods)] - /// /// let x = 12; /// let y = &x as *const i32; /// @@ -983,7 +1087,7 @@ impl<T: ?Sized> *const T { /// assert_eq!(y.read_volatile(), 12); /// } /// ``` - #[unstable(feature = "pointer_methods", issue = "43941")] + #[stable(feature = "pointer_methods", since = "1.26.0")] #[inline] pub unsafe fn read_volatile(self) -> T where T: Sized, @@ -1010,8 +1114,6 @@ impl<T: ?Sized> *const T { /// Basic usage: /// /// ``` - /// #![feature(pointer_methods)] - /// /// let x = 12; /// let y = &x as *const i32; /// @@ -1019,7 +1121,7 @@ impl<T: ?Sized> *const T { /// assert_eq!(y.read_unaligned(), 12); /// } /// ``` - #[unstable(feature = "pointer_methods", issue = "43941")] + #[stable(feature = "pointer_methods", since = "1.26.0")] #[inline] pub unsafe fn read_unaligned(self) -> T where T: Sized, @@ -1038,16 +1140,14 @@ impl<T: ?Sized> *const T { /// /// Care must be taken with the ownership of `self` and `dest`. /// This method semantically moves the values of `self` into `dest`. - /// However it does not drop the contents of `self`, or prevent the contents - /// of `dest` from being dropped or used. + /// However it does not drop the contents of `dest`, or prevent the contents + /// of `self` from being dropped or used. /// /// # Examples /// /// Efficiently create a Rust vector from an unsafe buffer: /// /// ``` - /// #![feature(pointer_methods)] - /// /// # #[allow(dead_code)] /// unsafe fn from_buf_raw<T: Copy>(ptr: *const T, elts: usize) -> Vec<T> { /// let mut dst = Vec::with_capacity(elts); @@ -1056,7 +1156,7 @@ impl<T: ?Sized> *const T { /// dst /// } /// ``` - #[unstable(feature = "pointer_methods", issue = "43941")] + #[stable(feature = "pointer_methods", since = "1.26.0")] #[inline] pub unsafe fn copy_to(self, dest: *mut T, count: usize) where T: Sized, @@ -1085,8 +1185,6 @@ impl<T: ?Sized> *const T { /// Efficiently create a Rust vector from an unsafe buffer: /// /// ``` - /// #![feature(pointer_methods)] - /// /// # #[allow(dead_code)] /// unsafe fn from_buf_raw<T: Copy>(ptr: *const T, elts: usize) -> Vec<T> { /// let mut dst = Vec::with_capacity(elts); @@ -1095,7 +1193,7 @@ impl<T: ?Sized> *const T { /// dst /// } /// ``` - #[unstable(feature = "pointer_methods", issue = "43941")] + #[stable(feature = "pointer_methods", since = "1.26.0")] #[inline] pub unsafe fn copy_to_nonoverlapping(self, dest: *mut T, count: usize) where T: Sized, @@ -1103,15 +1201,22 @@ impl<T: ?Sized> *const T { copy_nonoverlapping(self, dest, count) } - /// Computes the byte offset that needs to be applied in order to - /// make the pointer aligned to `align`. + /// Computes the offset that needs to be applied to the pointer in order to make it aligned to + /// `align`. + /// /// If it is not possible to align the pointer, the implementation returns /// `usize::max_value()`. /// - /// There are no guarantees whatsover that offsetting the pointer will not - /// overflow or go beyond the allocation that the pointer points into. - /// It is up to the caller to ensure that the returned offset is correct - /// in all terms other than alignment. + /// The offset is expressed in number of `T` elements, and not bytes. The value returned can be + /// used with the `offset` or `offset_to` methods. + /// + /// There are no guarantees whatsover that offsetting the pointer will not overflow or go + /// beyond the allocation that the pointer points into. It is up to the caller to ensure that + /// the returned offset is correct in all terms other than alignment. + /// + /// # Panics + /// + /// The function panics if `align` is not a power-of-two. /// /// # Examples /// @@ -1135,13 +1240,17 @@ impl<T: ?Sized> *const T { /// # } } /// ``` #[unstable(feature = "align_offset", issue = "44488")] - pub fn align_offset(self, align: usize) -> usize { + pub fn align_offset(self, align: usize) -> usize where T: Sized { + if !align.is_power_of_two() { + panic!("align_offset: align is not a power-of-two"); + } unsafe { - intrinsics::align_offset(self as *const _, align) + align_offset(self, align) } } } + #[lang = "mut_ptr"] impl<T: ?Sized> *mut T { /// Returns `true` if the pointer is null. @@ -1215,7 +1324,7 @@ impl<T: ?Sized> *mut T { /// Behavior: /// /// * Both the starting and resulting pointer must be either in bounds or one - /// byte past the end of an allocated object. + /// byte past the end of *the same* allocated object. /// /// * The computed offset, **in bytes**, cannot overflow an `isize`. /// @@ -1266,9 +1375,15 @@ impl<T: ?Sized> *mut T { /// /// The resulting pointer does not need to be in bounds, but it is /// potentially hazardous to dereference (which requires `unsafe`). + /// In particular, the resulting pointer may *not* be used to access a + /// different allocated object than the one `self` points to. In other + /// words, `x.wrapping_offset(y.wrapping_offset_from(x))` is + /// *not* the same as `y`, and dereferencing it is undefined behavior + /// unless `x` and `y` point into the same allocated object. /// /// Always use `.offset(count)` instead when possible, because `offset` - /// allows the compiler to optimize better. + /// allows the compiler to optimize better. If you need to cross object + /// boundaries, cast the pointer to an integer and do the arithmetic there. /// /// # Examples /// @@ -1330,77 +1445,108 @@ impl<T: ?Sized> *mut T { /// Calculates the distance between two pointers. The returned value is in /// units of T: the distance in bytes is divided by `mem::size_of::<T>()`. /// - /// If the address different between the two pointers ia not a multiple of - /// `mem::size_of::<T>()` then the result of the division is rounded towards - /// zero. + /// This function is the inverse of [`offset`]. + /// + /// [`offset`]: #method.offset-1 + /// [`wrapping_offset_from`]: #method.wrapping_offset_from-1 + /// + /// # Safety + /// + /// If any of the following conditions are violated, the result is Undefined + /// Behavior: + /// + /// * Both the starting and other pointer must be either in bounds or one + /// byte past the end of the same allocated object. + /// + /// * The distance between the pointers, **in bytes**, cannot overflow an `isize`. + /// + /// * The distance between the pointers, in bytes, must be an exact multiple + /// of the size of `T`. + /// + /// * The distance being in bounds cannot rely on "wrapping around" the address space. + /// + /// The compiler and standard library generally try to ensure allocations + /// never reach a size where an offset is a concern. For instance, `Vec` + /// and `Box` ensure they never allocate more than `isize::MAX` bytes, so + /// `ptr_into_vec.offset_from(vec.as_ptr())` is always safe. + /// + /// Most platforms fundamentally can't even construct such an allocation. + /// For instance, no known 64-bit platform can ever serve a request + /// for 2<sup>63</sup> bytes due to page-table limitations or splitting the address space. + /// However, some 32-bit and 16-bit platforms may successfully serve a request for + /// more than `isize::MAX` bytes with things like Physical Address + /// Extension. As such, memory acquired directly from allocators or memory + /// mapped files *may* be too large to handle with this function. + /// + /// Consider using [`wrapping_offset_from`] instead if these constraints are + /// difficult to satisfy. The only advantage of this method is that it + /// enables more aggressive compiler optimizations. /// - /// This function returns `None` if `T` is a zero-sized typed. + /// # Panics + /// + /// This function panics if `T` is a Zero-Sized Type ("ZST"). /// /// # Examples /// /// Basic usage: /// /// ``` - /// #![feature(offset_to)] + /// #![feature(ptr_offset_from)] /// - /// fn main() { - /// let mut a = [0; 5]; - /// let ptr1: *mut i32 = &mut a[1]; - /// let ptr2: *mut i32 = &mut a[3]; - /// assert_eq!(ptr1.offset_to(ptr2), Some(2)); - /// assert_eq!(ptr2.offset_to(ptr1), Some(-2)); - /// assert_eq!(unsafe { ptr1.offset(2) }, ptr2); - /// assert_eq!(unsafe { ptr2.offset(-2) }, ptr1); + /// let mut a = [0; 5]; + /// let ptr1: *mut i32 = &mut a[1]; + /// let ptr2: *mut i32 = &mut a[3]; + /// unsafe { + /// assert_eq!(ptr2.offset_from(ptr1), 2); + /// assert_eq!(ptr1.offset_from(ptr2), -2); + /// assert_eq!(ptr1.offset(2), ptr2); + /// assert_eq!(ptr2.offset(-2), ptr1); /// } /// ``` - #[unstable(feature = "offset_to", issue = "41079")] + #[unstable(feature = "ptr_offset_from", issue = "41079")] #[inline] - pub fn offset_to(self, other: *const T) -> Option<isize> where T: Sized { - let size = mem::size_of::<T>(); - if size == 0 { - None - } else { - let diff = (other as isize).wrapping_sub(self as isize); - Some(diff / size as isize) - } + pub unsafe fn offset_from(self, origin: *const T) -> isize where T: Sized { + (self as *const T).offset_from(origin) } - /// Computes the byte offset that needs to be applied in order to - /// make the pointer aligned to `align`. - /// If it is not possible to align the pointer, the implementation returns - /// `usize::max_value()`. + /// Calculates the distance between two pointers. The returned value is in + /// units of T: the distance in bytes is divided by `mem::size_of::<T>()`. + /// + /// If the address different between the two pointers is not a multiple of + /// `mem::size_of::<T>()` then the result of the division is rounded towards + /// zero. /// - /// There are no guarantees whatsover that offsetting the pointer will not - /// overflow or go beyond the allocation that the pointer points into. - /// It is up to the caller to ensure that the returned offset is correct - /// in all terms other than alignment. + /// Though this method is safe for any two pointers, note that its result + /// will be mostly useless if the two pointers aren't into the same allocated + /// object, for example if they point to two different local variables. + /// + /// # Panics + /// + /// This function panics if `T` is a zero-sized type. /// /// # Examples /// - /// Accessing adjacent `u8` as `u16` + /// Basic usage: /// /// ``` - /// # #![feature(align_offset)] - /// # fn foo(n: usize) { - /// # use std::mem::align_of; - /// # unsafe { - /// let x = [5u8, 6u8, 7u8, 8u8, 9u8]; - /// let ptr = &x[n] as *const u8; - /// let offset = ptr.align_offset(align_of::<u16>()); - /// if offset < x.len() - n - 1 { - /// let u16_ptr = ptr.offset(offset as isize) as *const u16; - /// assert_ne!(*u16_ptr, 500); - /// } else { - /// // while the pointer can be aligned via `offset`, it would point - /// // outside the allocation - /// } - /// # } } + /// #![feature(ptr_wrapping_offset_from)] + /// + /// let mut a = [0; 5]; + /// let ptr1: *mut i32 = &mut a[1]; + /// let ptr2: *mut i32 = &mut a[3]; + /// assert_eq!(ptr2.wrapping_offset_from(ptr1), 2); + /// assert_eq!(ptr1.wrapping_offset_from(ptr2), -2); + /// assert_eq!(ptr1.wrapping_offset(2), ptr2); + /// assert_eq!(ptr2.wrapping_offset(-2), ptr1); + /// + /// let ptr1: *mut i32 = 3 as _; + /// let ptr2: *mut i32 = 13 as _; + /// assert_eq!(ptr2.wrapping_offset_from(ptr1), 2); /// ``` - #[unstable(feature = "align_offset", issue = "44488")] - pub fn align_offset(self, align: usize) -> usize { - unsafe { - intrinsics::align_offset(self as *const _, align) - } + #[unstable(feature = "ptr_wrapping_offset_from", issue = "41079")] + #[inline] + pub fn wrapping_offset_from(self, origin: *const T) -> isize where T: Sized { + (self as *const T).wrapping_offset_from(origin) } /// Calculates the offset from a pointer (convenience for `.offset(count as isize)`). @@ -1443,8 +1589,6 @@ impl<T: ?Sized> *mut T { /// Basic usage: /// /// ``` - /// #![feature(pointer_methods)] - /// /// let s: &str = "123"; /// let ptr: *const u8 = s.as_ptr(); /// @@ -1453,7 +1597,7 @@ impl<T: ?Sized> *mut T { /// println!("{}", *ptr.add(2) as char); /// } /// ``` - #[unstable(feature = "pointer_methods", issue = "43941")] + #[stable(feature = "pointer_methods", since = "1.26.0")] #[inline] pub unsafe fn add(self, count: usize) -> Self where T: Sized, @@ -1502,8 +1646,6 @@ impl<T: ?Sized> *mut T { /// Basic usage: /// /// ``` - /// #![feature(pointer_methods)] - /// /// let s: &str = "123"; /// /// unsafe { @@ -1512,7 +1654,7 @@ impl<T: ?Sized> *mut T { /// println!("{}", *end.sub(2) as char); /// } /// ``` - #[unstable(feature = "pointer_methods", issue = "43941")] + #[stable(feature = "pointer_methods", since = "1.26.0")] #[inline] pub unsafe fn sub(self, count: usize) -> Self where T: Sized, @@ -1539,8 +1681,6 @@ impl<T: ?Sized> *mut T { /// Basic usage: /// /// ``` - /// #![feature(pointer_methods)] - /// /// // Iterate using a raw pointer in increments of two elements /// let data = [1u8, 2, 3, 4, 5]; /// let mut ptr: *const u8 = data.as_ptr(); @@ -1555,7 +1695,7 @@ impl<T: ?Sized> *mut T { /// ptr = ptr.wrapping_add(step); /// } /// ``` - #[unstable(feature = "pointer_methods", issue = "43941")] + #[stable(feature = "pointer_methods", since = "1.26.0")] #[inline] pub fn wrapping_add(self, count: usize) -> Self where T: Sized, @@ -1582,8 +1722,6 @@ impl<T: ?Sized> *mut T { /// Basic usage: /// /// ``` - /// #![feature(pointer_methods)] - /// /// // Iterate using a raw pointer in increments of two elements (backwards) /// let data = [1u8, 2, 3, 4, 5]; /// let mut ptr: *const u8 = data.as_ptr(); @@ -1598,7 +1736,7 @@ impl<T: ?Sized> *mut T { /// ptr = ptr.wrapping_sub(step); /// } /// ``` - #[unstable(feature = "pointer_methods", issue = "43941")] + #[stable(feature = "pointer_methods", since = "1.26.0")] #[inline] pub fn wrapping_sub(self, count: usize) -> Self where T: Sized, @@ -1625,8 +1763,6 @@ impl<T: ?Sized> *mut T { /// Basic usage: /// /// ``` - /// #![feature(pointer_methods)] - /// /// let x = 12; /// let y = &x as *const i32; /// @@ -1634,7 +1770,7 @@ impl<T: ?Sized> *mut T { /// assert_eq!(y.read(), 12); /// } /// ``` - #[unstable(feature = "pointer_methods", issue = "43941")] + #[stable(feature = "pointer_methods", since = "1.26.0")] #[inline] pub unsafe fn read(self) -> T where T: Sized, @@ -1668,17 +1804,21 @@ impl<T: ?Sized> *mut T { /// Beyond accepting a raw pointer, this is unsafe because it semantically /// moves the value out of `self` without preventing further usage of `self`. /// If `T` is not `Copy`, then care must be taken to ensure that the value at - /// `src` is not used before the data is overwritten again (e.g. with `write`, + /// `self` is not used before the data is overwritten again (e.g. with `write`, /// `write_bytes`, or `copy`). Note that `*self = foo` counts as a use /// because it will attempt to drop the value previously at `*self`. /// + /// Just like in C, whether an operation is volatile has no bearing whatsoever + /// on questions involving concurrent access from multiple threads. Volatile + /// accesses behave exactly like non-atomic accesses in that regard. In particular, + /// a race between a `read_volatile` and any write operation to the same location + /// is undefined behavior. + /// /// # Examples /// /// Basic usage: /// /// ``` - /// #![feature(pointer_methods)] - /// /// let x = 12; /// let y = &x as *const i32; /// @@ -1686,7 +1826,7 @@ impl<T: ?Sized> *mut T { /// assert_eq!(y.read_volatile(), 12); /// } /// ``` - #[unstable(feature = "pointer_methods", issue = "43941")] + #[stable(feature = "pointer_methods", since = "1.26.0")] #[inline] pub unsafe fn read_volatile(self) -> T where T: Sized, @@ -1713,8 +1853,6 @@ impl<T: ?Sized> *mut T { /// Basic usage: /// /// ``` - /// #![feature(pointer_methods)] - /// /// let x = 12; /// let y = &x as *const i32; /// @@ -1722,7 +1860,7 @@ impl<T: ?Sized> *mut T { /// assert_eq!(y.read_unaligned(), 12); /// } /// ``` - #[unstable(feature = "pointer_methods", issue = "43941")] + #[stable(feature = "pointer_methods", since = "1.26.0")] #[inline] pub unsafe fn read_unaligned(self) -> T where T: Sized, @@ -1749,8 +1887,6 @@ impl<T: ?Sized> *mut T { /// Efficiently create a Rust vector from an unsafe buffer: /// /// ``` - /// #![feature(pointer_methods)] - /// /// # #[allow(dead_code)] /// unsafe fn from_buf_raw<T: Copy>(ptr: *const T, elts: usize) -> Vec<T> { /// let mut dst = Vec::with_capacity(elts); @@ -1759,7 +1895,7 @@ impl<T: ?Sized> *mut T { /// dst /// } /// ``` - #[unstable(feature = "pointer_methods", issue = "43941")] + #[stable(feature = "pointer_methods", since = "1.26.0")] #[inline] pub unsafe fn copy_to(self, dest: *mut T, count: usize) where T: Sized, @@ -1788,8 +1924,6 @@ impl<T: ?Sized> *mut T { /// Efficiently create a Rust vector from an unsafe buffer: /// /// ``` - /// #![feature(pointer_methods)] - /// /// # #[allow(dead_code)] /// unsafe fn from_buf_raw<T: Copy>(ptr: *const T, elts: usize) -> Vec<T> { /// let mut dst = Vec::with_capacity(elts); @@ -1798,7 +1932,7 @@ impl<T: ?Sized> *mut T { /// dst /// } /// ``` - #[unstable(feature = "pointer_methods", issue = "43941")] + #[stable(feature = "pointer_methods", since = "1.26.0")] #[inline] pub unsafe fn copy_to_nonoverlapping(self, dest: *mut T, count: usize) where T: Sized, @@ -1825,8 +1959,6 @@ impl<T: ?Sized> *mut T { /// Efficiently create a Rust vector from an unsafe buffer: /// /// ``` - /// #![feature(pointer_methods)] - /// /// # #[allow(dead_code)] /// unsafe fn from_buf_raw<T: Copy>(ptr: *const T, elts: usize) -> Vec<T> { /// let mut dst: Vec<T> = Vec::with_capacity(elts); @@ -1835,7 +1967,7 @@ impl<T: ?Sized> *mut T { /// dst /// } /// ``` - #[unstable(feature = "pointer_methods", issue = "43941")] + #[stable(feature = "pointer_methods", since = "1.26.0")] #[inline] pub unsafe fn copy_from(self, src: *const T, count: usize) where T: Sized, @@ -1864,8 +1996,6 @@ impl<T: ?Sized> *mut T { /// Efficiently create a Rust vector from an unsafe buffer: /// /// ``` - /// #![feature(pointer_methods)] - /// /// # #[allow(dead_code)] /// unsafe fn from_buf_raw<T: Copy>(ptr: *const T, elts: usize) -> Vec<T> { /// let mut dst: Vec<T> = Vec::with_capacity(elts); @@ -1874,7 +2004,7 @@ impl<T: ?Sized> *mut T { /// dst /// } /// ``` - #[unstable(feature = "pointer_methods", issue = "43941")] + #[stable(feature = "pointer_methods", since = "1.26.0")] #[inline] pub unsafe fn copy_from_nonoverlapping(self, src: *const T, count: usize) where T: Sized, @@ -1899,7 +2029,7 @@ impl<T: ?Sized> *mut T { /// /// This has all the same safety problems as `ptr::read` with respect to /// invalid pointers, types, and double drops. - #[unstable(feature = "pointer_methods", issue = "43941")] + #[stable(feature = "pointer_methods", since = "1.26.0")] #[inline] pub unsafe fn drop_in_place(self) { drop_in_place(self) @@ -1929,8 +2059,6 @@ impl<T: ?Sized> *mut T { /// Basic usage: /// /// ``` - /// #![feature(pointer_methods)] - /// /// let mut x = 0; /// let y = &mut x as *mut i32; /// let z = 12; @@ -1940,7 +2068,7 @@ impl<T: ?Sized> *mut T { /// assert_eq!(y.read(), 12); /// } /// ``` - #[unstable(feature = "pointer_methods", issue = "43941")] + #[stable(feature = "pointer_methods", since = "1.26.0")] #[inline] pub unsafe fn write(self, val: T) where T: Sized, @@ -1954,8 +2082,6 @@ impl<T: ?Sized> *mut T { /// # Examples /// /// ``` - /// #![feature(pointer_methods)] - /// /// let mut vec = vec![0; 4]; /// unsafe { /// let vec_ptr = vec.as_mut_ptr(); @@ -1963,7 +2089,7 @@ impl<T: ?Sized> *mut T { /// } /// assert_eq!(vec, [b'a', b'a', 0, 0]); /// ``` - #[unstable(feature = "pointer_methods", issue = "43941")] + #[stable(feature = "pointer_methods", since = "1.26.0")] #[inline] pub unsafe fn write_bytes(self, val: u8, count: usize) where T: Sized, @@ -2003,13 +2129,17 @@ impl<T: ?Sized> *mut T { /// This is appropriate for initializing uninitialized memory, or overwriting /// memory that has previously been `read` from. /// + /// Just like in C, whether an operation is volatile has no bearing whatsoever + /// on questions involving concurrent access from multiple threads. Volatile + /// accesses behave exactly like non-atomic accesses in that regard. In particular, + /// a race between a `write_volatile` and any other operation (reading or writing) + /// on the same location is undefined behavior. + /// /// # Examples /// /// Basic usage: /// /// ``` - /// #![feature(pointer_methods)] - /// /// let mut x = 0; /// let y = &mut x as *mut i32; /// let z = 12; @@ -2019,7 +2149,7 @@ impl<T: ?Sized> *mut T { /// assert_eq!(y.read_volatile(), 12); /// } /// ``` - #[unstable(feature = "pointer_methods", issue = "43941")] + #[stable(feature = "pointer_methods", since = "1.26.0")] #[inline] pub unsafe fn write_volatile(self, val: T) where T: Sized, @@ -2040,8 +2170,8 @@ impl<T: ?Sized> *mut T { /// allocations or resources, so care must be taken not to overwrite an object /// that should be dropped. /// - /// Additionally, it does not drop `src`. Semantically, `src` is moved into the - /// location pointed to by `dst`. + /// Additionally, it does not drop `self`. Semantically, `self` is moved into the + /// location pointed to by `val`. /// /// This is appropriate for initializing uninitialized memory, or overwriting /// memory that has previously been `read` from. @@ -2051,8 +2181,6 @@ impl<T: ?Sized> *mut T { /// Basic usage: /// /// ``` - /// #![feature(pointer_methods)] - /// /// let mut x = 0; /// let y = &mut x as *mut i32; /// let z = 12; @@ -2062,7 +2190,7 @@ impl<T: ?Sized> *mut T { /// assert_eq!(y.read_unaligned(), 12); /// } /// ``` - #[unstable(feature = "pointer_methods", issue = "43941")] + #[stable(feature = "pointer_methods", since = "1.26.0")] #[inline] pub unsafe fn write_unaligned(self, val: T) where T: Sized, @@ -2077,7 +2205,7 @@ impl<T: ?Sized> *mut T { /// /// This is only unsafe because it accepts a raw pointer. /// Otherwise, this operation is identical to `mem::replace`. - #[unstable(feature = "pointer_methods", issue = "43941")] + #[stable(feature = "pointer_methods", since = "1.26.0")] #[inline] pub unsafe fn replace(self, src: T) -> T where T: Sized, @@ -2095,15 +2223,193 @@ impl<T: ?Sized> *mut T { /// as arguments. /// /// Ensure that these pointers are valid before calling `swap`. - #[unstable(feature = "pointer_methods", issue = "43941")] + #[stable(feature = "pointer_methods", since = "1.26.0")] #[inline] pub unsafe fn swap(self, with: *mut T) where T: Sized, { swap(self, with) } + + /// Computes the offset that needs to be applied to the pointer in order to make it aligned to + /// `align`. + /// + /// If it is not possible to align the pointer, the implementation returns + /// `usize::max_value()`. + /// + /// The offset is expressed in number of `T` elements, and not bytes. The value returned can be + /// used with the `offset` or `offset_to` methods. + /// + /// There are no guarantees whatsover that offsetting the pointer will not overflow or go + /// beyond the allocation that the pointer points into. It is up to the caller to ensure that + /// the returned offset is correct in all terms other than alignment. + /// + /// # Panics + /// + /// The function panics if `align` is not a power-of-two. + /// + /// # Examples + /// + /// Accessing adjacent `u8` as `u16` + /// + /// ``` + /// # #![feature(align_offset)] + /// # fn foo(n: usize) { + /// # use std::mem::align_of; + /// # unsafe { + /// let x = [5u8, 6u8, 7u8, 8u8, 9u8]; + /// let ptr = &x[n] as *const u8; + /// let offset = ptr.align_offset(align_of::<u16>()); + /// if offset < x.len() - n - 1 { + /// let u16_ptr = ptr.offset(offset as isize) as *const u16; + /// assert_ne!(*u16_ptr, 500); + /// } else { + /// // while the pointer can be aligned via `offset`, it would point + /// // outside the allocation + /// } + /// # } } + /// ``` + #[unstable(feature = "align_offset", issue = "44488")] + pub fn align_offset(self, align: usize) -> usize where T: Sized { + if !align.is_power_of_two() { + panic!("align_offset: align is not a power-of-two"); + } + unsafe { + align_offset(self, align) + } + } +} + +/// Align pointer `p`. +/// +/// Calculate offset (in terms of elements of `stride` stride) that has to be applied +/// to pointer `p` so that pointer `p` would get aligned to `a`. +/// +/// Note: This implementation has been carefully tailored to not panic. It is UB for this to panic. +/// The only real change that can be made here is change of `INV_TABLE_MOD_16` and associated +/// constants. +/// +/// If we ever decide to make it possible to call the intrinsic with `a` that is not a +/// power-of-two, it will probably be more prudent to just change to a naive implementation rather +/// than trying to adapt this to accomodate that change. +/// +/// Any questions go to @nagisa. +#[lang="align_offset"] +pub(crate) unsafe fn align_offset<T: Sized>(p: *const T, a: usize) -> usize { + /// Calculate multiplicative modular inverse of `x` modulo `m`. + /// + /// This implementation is tailored for align_offset and has following preconditions: + /// + /// * `m` is a power-of-two; + /// * `x < m`; (if `x ≥ m`, pass in `x % m` instead) + /// + /// Implementation of this function shall not panic. Ever. + #[inline] + fn mod_inv(x: usize, m: usize) -> usize { + /// Multiplicative modular inverse table modulo 2⁴ = 16. + /// + /// Note, that this table does not contain values where inverse does not exist (i.e. for + /// `0⁻¹ mod 16`, `2⁻¹ mod 16`, etc.) + const INV_TABLE_MOD_16: [usize; 8] = [1, 11, 13, 7, 9, 3, 5, 15]; + /// Modulo for which the `INV_TABLE_MOD_16` is intended. + const INV_TABLE_MOD: usize = 16; + /// INV_TABLE_MOD² + const INV_TABLE_MOD_SQUARED: usize = INV_TABLE_MOD * INV_TABLE_MOD; + + let table_inverse = INV_TABLE_MOD_16[(x & (INV_TABLE_MOD - 1)) >> 1]; + if m <= INV_TABLE_MOD { + table_inverse & (m - 1) + } else { + // We iterate "up" using the following formula: + // + // $$ xy ≡ 1 (mod 2ⁿ) → xy (2 - xy) ≡ 1 (mod 2²ⁿ) $$ + // + // until 2²ⁿ ≥ m. Then we can reduce to our desired `m` by taking the result `mod m`. + let mut inverse = table_inverse; + let mut going_mod = INV_TABLE_MOD_SQUARED; + loop { + // y = y * (2 - xy) mod n + // + // Note, that we use wrapping operations here intentionally – the original formula + // uses e.g. subtraction `mod n`. It is entirely fine to do them `mod + // usize::max_value()` instead, because we take the result `mod n` at the end + // anyway. + inverse = inverse.wrapping_mul( + 2usize.wrapping_sub(x.wrapping_mul(inverse)) + ) & (going_mod - 1); + if going_mod > m { + return inverse & (m - 1); + } + going_mod = going_mod.wrapping_mul(going_mod); + } + } + } + + let stride = ::mem::size_of::<T>(); + let a_minus_one = a.wrapping_sub(1); + let pmoda = p as usize & a_minus_one; + + if pmoda == 0 { + // Already aligned. Yay! + return 0; + } + + if stride <= 1 { + return if stride == 0 { + // If the pointer is not aligned, and the element is zero-sized, then no amount of + // elements will ever align the pointer. + !0 + } else { + a.wrapping_sub(pmoda) + }; + } + + let smoda = stride & a_minus_one; + // a is power-of-two so cannot be 0. stride = 0 is handled above. + let gcdpow = intrinsics::cttz_nonzero(stride).min(intrinsics::cttz_nonzero(a)); + let gcd = 1usize << gcdpow; + + if gcd == 1 { + // This branch solves for the variable $o$ in following linear congruence equation: + // + // ⎰ p + o ≡ 0 (mod a) # $p + o$ must be aligned to specified alignment $a$ + // ⎱ o ≡ 0 (mod s) # offset $o$ must be a multiple of stride $s$ + // + // where + // + // * a, s are co-prime + // + // This gives us the formula below: + // + // o = (a - (p mod a)) * (s⁻¹ mod a) * s + // + // The first term is “the relative alignment of p to a”, the second term is “how does + // incrementing p by one s change the relative alignment of p”, the third term is + // translating change in units of s to a byte count. + // + // Furthermore, the result produced by this solution is not “minimal”, so it is necessary + // to take the result $o mod lcm(s, a)$. Since $s$ and $a$ are co-prime (i.e. $gcd(s, a) = + // 1$) and $lcm(s, a) = s * a / gcd(s, a)$, we can replace $lcm(s, a)$ with just a $s * a$. + // + // (Author note: we decided later on to express the offset in "elements" rather than bytes, + // which drops the multiplication by `s` on both sides of the modulo.) + return intrinsics::unchecked_rem(a.wrapping_sub(pmoda).wrapping_mul(mod_inv(smoda, a)), a); + } + + if p as usize & (gcd - 1) == 0 { + // This can be aligned, but `a` and `stride` are not co-prime, so a somewhat adapted + // formula is used. + let j = a.wrapping_sub(pmoda) >> gcdpow; + let k = smoda >> gcdpow; + return intrinsics::unchecked_rem(j.wrapping_mul(mod_inv(k, a)), a >> gcdpow); + } + + // Cannot be aligned at all. + usize::max_value() } + + // Equality for pointers #[stable(feature = "rust1", since = "1.0.0")] impl<T: ?Sized> PartialEq for *const T { @@ -2321,7 +2627,7 @@ impl<T: ?Sized> PartialOrd for *mut T { /// its owning Unique. /// /// If you're uncertain of whether it's correct to use `Unique` for your purposes, -/// consider using `Shared`, which has weaker semantics. +/// consider using `NonNull`, which has weaker semantics. /// /// Unlike `*mut T`, the pointer must always be non-null, even if the pointer /// is never dereferenced. This is so that enums may use this forbidden value @@ -2330,9 +2636,11 @@ impl<T: ?Sized> PartialOrd for *mut T { /// /// Unlike `*mut T`, `Unique<T>` is covariant over `T`. This should always be correct /// for any type which upholds Unique's aliasing requirements. -#[allow(missing_debug_implementations)] -#[unstable(feature = "unique", reason = "needs an RFC to flesh out design", - issue = "27730")] +#[unstable(feature = "ptr_internals", issue = "0", + reason = "use NonNull instead and consider PhantomData<T> \ + (if you also use #[may_dangle]), Send, and/or Sync")] +#[doc(hidden)] +#[repr(transparent)] pub struct Unique<T: ?Sized> { pointer: NonZero<*const T>, // NOTE: this marker has no consequences for variance, but is necessary @@ -2343,61 +2651,76 @@ pub struct Unique<T: ?Sized> { _marker: PhantomData<T>, } +#[unstable(feature = "ptr_internals", issue = "0")] +impl<T: ?Sized> fmt::Debug for Unique<T> { + fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result { + fmt::Pointer::fmt(&self.as_ptr(), f) + } +} + /// `Unique` pointers are `Send` if `T` is `Send` because the data they /// reference is unaliased. Note that this aliasing invariant is /// unenforced by the type system; the abstraction using the /// `Unique` must enforce it. -#[unstable(feature = "unique", issue = "27730")] +#[unstable(feature = "ptr_internals", issue = "0")] unsafe impl<T: Send + ?Sized> Send for Unique<T> { } /// `Unique` pointers are `Sync` if `T` is `Sync` because the data they /// reference is unaliased. Note that this aliasing invariant is /// unenforced by the type system; the abstraction using the /// `Unique` must enforce it. -#[unstable(feature = "unique", issue = "27730")] +#[unstable(feature = "ptr_internals", issue = "0")] unsafe impl<T: Sync + ?Sized> Sync for Unique<T> { } -#[unstable(feature = "unique", issue = "27730")] +#[unstable(feature = "ptr_internals", issue = "0")] impl<T: Sized> Unique<T> { /// Creates a new `Unique` that is dangling, but well-aligned. /// /// This is useful for initializing types which lazily allocate, like /// `Vec::new` does. - pub fn empty() -> Self { + /// + /// Note that the pointer value may potentially represent a valid pointer to + /// a `T`, which means this must not be used as a "not yet initialized" + /// sentinel value. Types that lazily allocate must track initialization by + /// some other means. + // FIXME: rename to dangling() to match NonNull? + pub const fn empty() -> Self { unsafe { - let ptr = mem::align_of::<T>() as *mut T; - Unique::new_unchecked(ptr) + Unique::new_unchecked(mem::align_of::<T>() as *mut T) } } } -#[unstable(feature = "unique", issue = "27730")] +#[unstable(feature = "ptr_internals", issue = "0")] impl<T: ?Sized> Unique<T> { /// Creates a new `Unique`. /// /// # Safety /// /// `ptr` must be non-null. - #[unstable(feature = "unique", issue = "27730")] pub const unsafe fn new_unchecked(ptr: *mut T) -> Self { - Unique { pointer: NonZero::new_unchecked(ptr), _marker: PhantomData } + Unique { pointer: NonZero(ptr as _), _marker: PhantomData } } /// Creates a new `Unique` if `ptr` is non-null. pub fn new(ptr: *mut T) -> Option<Self> { - NonZero::new(ptr as *const T).map(|nz| Unique { pointer: nz, _marker: PhantomData }) + if !ptr.is_null() { + Some(Unique { pointer: NonZero(ptr as _), _marker: PhantomData }) + } else { + None + } } /// Acquires the underlying `*mut` pointer. pub fn as_ptr(self) -> *mut T { - self.pointer.get() as *mut T + self.pointer.0 as *mut T } /// Dereferences the content. /// /// The resulting lifetime is bound to self so this behaves "as if" /// it were actually an instance of T that is getting borrowed. If a longer - /// (unbound) lifetime is needed, use `&*my_ptr.ptr()`. + /// (unbound) lifetime is needed, use `&*my_ptr.as_ptr()`. pub unsafe fn as_ref(&self) -> &T { &*self.as_ptr() } @@ -2406,43 +2729,50 @@ impl<T: ?Sized> Unique<T> { /// /// The resulting lifetime is bound to self so this behaves "as if" /// it were actually an instance of T that is getting borrowed. If a longer - /// (unbound) lifetime is needed, use `&mut *my_ptr.ptr()`. + /// (unbound) lifetime is needed, use `&mut *my_ptr.as_ptr()`. pub unsafe fn as_mut(&mut self) -> &mut T { &mut *self.as_ptr() } } -#[unstable(feature = "unique", issue = "27730")] +#[unstable(feature = "ptr_internals", issue = "0")] impl<T: ?Sized> Clone for Unique<T> { fn clone(&self) -> Self { *self } } -#[unstable(feature = "unique", issue = "27730")] +#[unstable(feature = "ptr_internals", issue = "0")] impl<T: ?Sized> Copy for Unique<T> { } -#[unstable(feature = "unique", issue = "27730")] +#[unstable(feature = "ptr_internals", issue = "0")] impl<T: ?Sized, U: ?Sized> CoerceUnsized<Unique<U>> for Unique<T> where T: Unsize<U> { } -#[unstable(feature = "unique", issue = "27730")] +#[unstable(feature = "ptr_internals", issue = "0")] impl<T: ?Sized> fmt::Pointer for Unique<T> { fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result { fmt::Pointer::fmt(&self.as_ptr(), f) } } -#[unstable(feature = "unique", issue = "27730")] +#[unstable(feature = "ptr_internals", issue = "0")] impl<'a, T: ?Sized> From<&'a mut T> for Unique<T> { fn from(reference: &'a mut T) -> Self { - Unique { pointer: NonZero::from(reference), _marker: PhantomData } + Unique { pointer: NonZero(reference as _), _marker: PhantomData } } } -#[unstable(feature = "unique", issue = "27730")] +#[unstable(feature = "ptr_internals", issue = "0")] impl<'a, T: ?Sized> From<&'a T> for Unique<T> { fn from(reference: &'a T) -> Self { - Unique { pointer: NonZero::from(reference), _marker: PhantomData } + Unique { pointer: NonZero(reference as _), _marker: PhantomData } + } +} + +#[unstable(feature = "ptr_internals", issue = "0")] +impl<'a, T: ?Sized> From<NonNull<T>> for Unique<T> { + fn from(p: NonNull<T>) -> Self { + Unique { pointer: p.pointer, _marker: PhantomData } } } @@ -2450,77 +2780,87 @@ impl<'a, T: ?Sized> From<&'a T> for Unique<T> { /// /// This is often the correct thing to use when building data structures using /// raw pointers, but is ultimately more dangerous to use because of its additional -/// properties. If you're not sure if you should use `Shared<T>`, just use `*mut T`! +/// properties. If you're not sure if you should use `NonNull<T>`, just use `*mut T`! /// /// Unlike `*mut T`, the pointer must always be non-null, even if the pointer /// is never dereferenced. This is so that enums may use this forbidden value -/// as a discriminant -- `Option<Shared<T>>` has the same size as `Shared<T>`. +/// as a discriminant -- `Option<NonNull<T>>` has the same size as `*mut T`. /// However the pointer may still dangle if it isn't dereferenced. /// -/// Unlike `*mut T`, `Shared<T>` is covariant over `T`. If this is incorrect +/// Unlike `*mut T`, `NonNull<T>` is covariant over `T`. If this is incorrect /// for your use case, you should include some PhantomData in your type to /// provide invariance, such as `PhantomData<Cell<T>>` or `PhantomData<&'a mut T>`. /// Usually this won't be necessary; covariance is correct for most safe abstractions, /// such as Box, Rc, Arc, Vec, and LinkedList. This is the case because they /// provide a public API that follows the normal shared XOR mutable rules of Rust. -#[allow(missing_debug_implementations)] -#[unstable(feature = "shared", reason = "needs an RFC to flesh out design", - issue = "27730")] -pub struct Shared<T: ?Sized> { +#[stable(feature = "nonnull", since = "1.25.0")] +#[repr(transparent)] +pub struct NonNull<T: ?Sized> { pointer: NonZero<*const T>, } -/// `Shared` pointers are not `Send` because the data they reference may be aliased. +/// `NonNull` pointers are not `Send` because the data they reference may be aliased. // NB: This impl is unnecessary, but should provide better error messages. -#[unstable(feature = "shared", issue = "27730")] -impl<T: ?Sized> !Send for Shared<T> { } +#[stable(feature = "nonnull", since = "1.25.0")] +impl<T: ?Sized> !Send for NonNull<T> { } -/// `Shared` pointers are not `Sync` because the data they reference may be aliased. +/// `NonNull` pointers are not `Sync` because the data they reference may be aliased. // NB: This impl is unnecessary, but should provide better error messages. -#[unstable(feature = "shared", issue = "27730")] -impl<T: ?Sized> !Sync for Shared<T> { } +#[stable(feature = "nonnull", since = "1.25.0")] +impl<T: ?Sized> !Sync for NonNull<T> { } -#[unstable(feature = "shared", issue = "27730")] -impl<T: Sized> Shared<T> { - /// Creates a new `Shared` that is dangling, but well-aligned. +impl<T: Sized> NonNull<T> { + /// Creates a new `NonNull` that is dangling, but well-aligned. /// /// This is useful for initializing types which lazily allocate, like /// `Vec::new` does. - pub fn empty() -> Self { + /// + /// Note that the pointer value may potentially represent a valid pointer to + /// a `T`, which means this must not be used as a "not yet initialized" + /// sentinel value. Types that lazily allocate must track initialization by + /// some other means. + #[stable(feature = "nonnull", since = "1.25.0")] + pub fn dangling() -> Self { unsafe { let ptr = mem::align_of::<T>() as *mut T; - Shared::new_unchecked(ptr) + NonNull::new_unchecked(ptr) } } } -#[unstable(feature = "shared", issue = "27730")] -impl<T: ?Sized> Shared<T> { - /// Creates a new `Shared`. +impl<T: ?Sized> NonNull<T> { + /// Creates a new `NonNull`. /// /// # Safety /// /// `ptr` must be non-null. - #[unstable(feature = "shared", issue = "27730")] + #[stable(feature = "nonnull", since = "1.25.0")] pub const unsafe fn new_unchecked(ptr: *mut T) -> Self { - Shared { pointer: NonZero::new_unchecked(ptr) } + NonNull { pointer: NonZero(ptr as _) } } - /// Creates a new `Shared` if `ptr` is non-null. + /// Creates a new `NonNull` if `ptr` is non-null. + #[stable(feature = "nonnull", since = "1.25.0")] pub fn new(ptr: *mut T) -> Option<Self> { - NonZero::new(ptr as *const T).map(|nz| Shared { pointer: nz }) + if !ptr.is_null() { + Some(NonNull { pointer: NonZero(ptr as _) }) + } else { + None + } } /// Acquires the underlying `*mut` pointer. + #[stable(feature = "nonnull", since = "1.25.0")] pub fn as_ptr(self) -> *mut T { - self.pointer.get() as *mut T + self.pointer.0 as *mut T } /// Dereferences the content. /// /// The resulting lifetime is bound to self so this behaves "as if" /// it were actually an instance of T that is getting borrowed. If a longer - /// (unbound) lifetime is needed, use `&*my_ptr.ptr()`. + /// (unbound) lifetime is needed, use `&*my_ptr.as_ptr()`. + #[stable(feature = "nonnull", since = "1.25.0")] pub unsafe fn as_ref(&self) -> &T { &*self.as_ptr() } @@ -2529,56 +2869,96 @@ impl<T: ?Sized> Shared<T> { /// /// The resulting lifetime is bound to self so this behaves "as if" /// it were actually an instance of T that is getting borrowed. If a longer - /// (unbound) lifetime is needed, use `&mut *my_ptr.ptr_mut()`. + /// (unbound) lifetime is needed, use `&mut *my_ptr.as_ptr()`. + #[stable(feature = "nonnull", since = "1.25.0")] pub unsafe fn as_mut(&mut self) -> &mut T { &mut *self.as_ptr() } - /// Acquires the underlying pointer as a `*mut` pointer. - #[rustc_deprecated(since = "1.19", reason = "renamed to `as_ptr` for ergonomics/consistency")] - #[unstable(feature = "shared", issue = "27730")] - pub unsafe fn as_mut_ptr(&self) -> *mut T { - self.as_ptr() + /// Cast to a pointer of another type + #[stable(feature = "nonnull_cast", since = "1.27.0")] + pub fn cast<U>(self) -> NonNull<U> { + unsafe { + NonNull::new_unchecked(self.as_ptr() as *mut U) + } } } -#[unstable(feature = "shared", issue = "27730")] -impl<T: ?Sized> Clone for Shared<T> { +#[stable(feature = "nonnull", since = "1.25.0")] +impl<T: ?Sized> Clone for NonNull<T> { fn clone(&self) -> Self { *self } } -#[unstable(feature = "shared", issue = "27730")] -impl<T: ?Sized> Copy for Shared<T> { } +#[stable(feature = "nonnull", since = "1.25.0")] +impl<T: ?Sized> Copy for NonNull<T> { } -#[unstable(feature = "shared", issue = "27730")] -impl<T: ?Sized, U: ?Sized> CoerceUnsized<Shared<U>> for Shared<T> where T: Unsize<U> { } +#[unstable(feature = "coerce_unsized", issue = "27732")] +impl<T: ?Sized, U: ?Sized> CoerceUnsized<NonNull<U>> for NonNull<T> where T: Unsize<U> { } + +#[stable(feature = "nonnull", since = "1.25.0")] +impl<T: ?Sized> fmt::Debug for NonNull<T> { + fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result { + fmt::Pointer::fmt(&self.as_ptr(), f) + } +} -#[unstable(feature = "shared", issue = "27730")] -impl<T: ?Sized> fmt::Pointer for Shared<T> { +#[stable(feature = "nonnull", since = "1.25.0")] +impl<T: ?Sized> fmt::Pointer for NonNull<T> { fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result { fmt::Pointer::fmt(&self.as_ptr(), f) } } -#[unstable(feature = "shared", issue = "27730")] -impl<T: ?Sized> From<Unique<T>> for Shared<T> { +#[stable(feature = "nonnull", since = "1.25.0")] +impl<T: ?Sized> Eq for NonNull<T> {} + +#[stable(feature = "nonnull", since = "1.25.0")] +impl<T: ?Sized> PartialEq for NonNull<T> { + fn eq(&self, other: &Self) -> bool { + self.as_ptr() == other.as_ptr() + } +} + +#[stable(feature = "nonnull", since = "1.25.0")] +impl<T: ?Sized> Ord for NonNull<T> { + fn cmp(&self, other: &Self) -> Ordering { + self.as_ptr().cmp(&other.as_ptr()) + } +} + +#[stable(feature = "nonnull", since = "1.25.0")] +impl<T: ?Sized> PartialOrd for NonNull<T> { + fn partial_cmp(&self, other: &Self) -> Option<Ordering> { + self.as_ptr().partial_cmp(&other.as_ptr()) + } +} + +#[stable(feature = "nonnull", since = "1.25.0")] +impl<T: ?Sized> hash::Hash for NonNull<T> { + fn hash<H: hash::Hasher>(&self, state: &mut H) { + self.as_ptr().hash(state) + } +} + +#[unstable(feature = "ptr_internals", issue = "0")] +impl<T: ?Sized> From<Unique<T>> for NonNull<T> { fn from(unique: Unique<T>) -> Self { - Shared { pointer: unique.pointer } + NonNull { pointer: unique.pointer } } } -#[unstable(feature = "shared", issue = "27730")] -impl<'a, T: ?Sized> From<&'a mut T> for Shared<T> { +#[stable(feature = "nonnull", since = "1.25.0")] +impl<'a, T: ?Sized> From<&'a mut T> for NonNull<T> { fn from(reference: &'a mut T) -> Self { - Shared { pointer: NonZero::from(reference) } + NonNull { pointer: NonZero(reference as _) } } } -#[unstable(feature = "shared", issue = "27730")] -impl<'a, T: ?Sized> From<&'a T> for Shared<T> { +#[stable(feature = "nonnull", since = "1.25.0")] +impl<'a, T: ?Sized> From<&'a T> for NonNull<T> { fn from(reference: &'a T) -> Self { - Shared { pointer: NonZero::from(reference) } + NonNull { pointer: NonZero(reference as _) } } } diff --git a/src/libcore/result.rs b/src/libcore/result.rs index 2ace3d2aee8..ac908342655 100644 --- a/src/libcore/result.rs +++ b/src/libcore/result.rs @@ -242,7 +242,7 @@ use fmt; use iter::{FromIterator, FusedIterator, TrustedLen}; -use ops; +use ops::{self, Deref}; /// `Result` is a type that represents either success ([`Ok`]) or failure ([`Err`]). /// @@ -251,7 +251,7 @@ use ops; /// [`Ok`]: enum.Result.html#variant.Ok /// [`Err`]: enum.Result.html#variant.Err #[derive(Clone, Copy, PartialEq, PartialOrd, Eq, Ord, Debug, Hash)] -#[must_use] +#[must_use = "this `Result` may be an `Err` variant, which should be handled"] #[stable(feature = "rust1", since = "1.0.0")] pub enum Result<T, E> { /// Contains the success value @@ -909,6 +909,73 @@ impl<T: Default, E> Result<T, E> { } } +#[unstable(feature = "inner_deref", reason = "newly added", issue = "50264")] +impl<T: Deref, E> Result<T, E> { + /// Converts from `&Result<T, E>` to `Result<&T::Target, &E>`. + /// + /// Leaves the original Result in-place, creating a new one with a reference + /// to the original one, additionally coercing the `Ok` arm of the Result via + /// `Deref`. + pub fn deref_ok(&self) -> Result<&T::Target, &E> { + self.as_ref().map(|t| t.deref()) + } +} + +#[unstable(feature = "inner_deref", reason = "newly added", issue = "50264")] +impl<T, E: Deref> Result<T, E> { + /// Converts from `&Result<T, E>` to `Result<&T, &E::Target>`. + /// + /// Leaves the original Result in-place, creating a new one with a reference + /// to the original one, additionally coercing the `Err` arm of the Result via + /// `Deref`. + pub fn deref_err(&self) -> Result<&T, &E::Target> + { + self.as_ref().map_err(|e| e.deref()) + } +} + +#[unstable(feature = "inner_deref", reason = "newly added", issue = "50264")] +impl<T: Deref, E: Deref> Result<T, E> { + /// Converts from `&Result<T, E>` to `Result<&T::Target, &E::Target>`. + /// + /// Leaves the original Result in-place, creating a new one with a reference + /// to the original one, additionally coercing both the `Ok` and `Err` arms + /// of the Result via `Deref`. + pub fn deref(&self) -> Result<&T::Target, &E::Target> + { + self.as_ref().map(|t| t.deref()).map_err(|e| e.deref()) + } +} + +impl<T, E> Result<Option<T>, E> { + /// Transposes a `Result` of an `Option` into an `Option` of a `Result`. + /// + /// `Ok(None)` will be mapped to `None`. + /// `Ok(Some(_))` and `Err(_)` will be mapped to `Some(Ok(_))` and `Some(Err(_))`. + /// + /// # Examples + /// + /// ``` + /// #![feature(transpose_result)] + /// + /// #[derive(Debug, Eq, PartialEq)] + /// struct SomeErr; + /// + /// let x: Result<Option<i32>, SomeErr> = Ok(Some(5)); + /// let y: Option<Result<i32, SomeErr>> = Some(Ok(5)); + /// assert_eq!(x.transpose(), y); + /// ``` + #[inline] + #[unstable(feature = "transpose_result", issue = "47338")] + pub fn transpose(self) -> Option<Result<T, E>> { + match self { + Ok(Some(x)) => Some(Ok(x)), + Ok(None) => None, + Err(e) => Some(Err(e)), + } + } +} + // This is a separate function to reduce the code size of the methods #[inline(never)] #[cold] @@ -1009,7 +1076,7 @@ impl<'a, T> DoubleEndedIterator for Iter<'a, T> { #[stable(feature = "rust1", since = "1.0.0")] impl<'a, T> ExactSizeIterator for Iter<'a, T> {} -#[unstable(feature = "fused", issue = "35602")] +#[stable(feature = "fused", since = "1.26.0")] impl<'a, T> FusedIterator for Iter<'a, T> {} #[unstable(feature = "trusted_len", issue = "37572")] @@ -1017,6 +1084,7 @@ unsafe impl<'a, A> TrustedLen for Iter<'a, A> {} #[stable(feature = "rust1", since = "1.0.0")] impl<'a, T> Clone for Iter<'a, T> { + #[inline] fn clone(&self) -> Iter<'a, T> { Iter { inner: self.inner } } } @@ -1053,7 +1121,7 @@ impl<'a, T> DoubleEndedIterator for IterMut<'a, T> { #[stable(feature = "rust1", since = "1.0.0")] impl<'a, T> ExactSizeIterator for IterMut<'a, T> {} -#[unstable(feature = "fused", issue = "35602")] +#[stable(feature = "fused", since = "1.26.0")] impl<'a, T> FusedIterator for IterMut<'a, T> {} #[unstable(feature = "trusted_len", issue = "37572")] @@ -1096,7 +1164,7 @@ impl<T> DoubleEndedIterator for IntoIter<T> { #[stable(feature = "rust1", since = "1.0.0")] impl<T> ExactSizeIterator for IntoIter<T> {} -#[unstable(feature = "fused", issue = "35602")] +#[stable(feature = "fused", since = "1.26.0")] impl<T> FusedIterator for IntoIter<T> {} #[unstable(feature = "trusted_len", issue = "37572")] @@ -1168,14 +1236,17 @@ impl<T,E> ops::Try for Result<T, E> { type Ok = T; type Error = E; + #[inline] fn into_result(self) -> Self { self } + #[inline] fn from_ok(v: T) -> Self { Ok(v) } + #[inline] fn from_error(v: E) -> Self { Err(v) } diff --git a/src/libcore/slice/memchr.rs b/src/libcore/slice/memchr.rs index 69c9cb37dcf..c9d3c7fea98 100644 --- a/src/libcore/slice/memchr.rs +++ b/src/libcore/slice/memchr.rs @@ -39,21 +39,10 @@ fn repeat_byte(b: u8) -> usize { (b as usize) << 8 | b as usize } -#[cfg(target_pointer_width = "32")] +#[cfg(not(target_pointer_width = "16"))] #[inline] fn repeat_byte(b: u8) -> usize { - let mut rep = (b as usize) << 8 | b as usize; - rep = rep << 16 | rep; - rep -} - -#[cfg(target_pointer_width = "64")] -#[inline] -fn repeat_byte(b: u8) -> usize { - let mut rep = (b as usize) << 8 | b as usize; - rep = rep << 16 | rep; - rep = rep << 32 | rep; - rep + (b as usize) * (::usize::MAX / 255) } /// Return the first index matching the byte `x` in `text`. @@ -111,27 +100,30 @@ pub fn memrchr(x: u8, text: &[u8]) -> Option<usize> { // - the first remaining bytes, < 2 word size let len = text.len(); let ptr = text.as_ptr(); - let usize_bytes = mem::size_of::<usize>(); - - // search to an aligned boundary - let end_align = (ptr as usize + len) & (usize_bytes - 1); - let mut offset; - if end_align > 0 { - offset = if end_align >= len { 0 } else { len - end_align }; - if let Some(index) = text[offset..].iter().rposition(|elt| *elt == x) { - return Some(offset + index); - } - } else { - offset = len; + type Chunk = usize; + + let (min_aligned_offset, max_aligned_offset) = { + // We call this just to obtain the length of the prefix and suffix. + // In the middle we always process two chunks at once. + let (prefix, _, suffix) = unsafe { text.align_to::<(Chunk, Chunk)>() }; + (prefix.len(), len - suffix.len()) + }; + + let mut offset = max_aligned_offset; + if let Some(index) = text[offset..].iter().rposition(|elt| *elt == x) { + return Some(offset + index); } - // search the body of the text + // search the body of the text, make sure we don't cross min_aligned_offset. + // offset is always aligned, so just testing `>` is sufficient and avoids possible + // overflow. let repeated_x = repeat_byte(x); + let chunk_bytes = mem::size_of::<Chunk>(); - while offset >= 2 * usize_bytes { + while offset > min_aligned_offset { unsafe { - let u = *(ptr.offset(offset as isize - 2 * usize_bytes as isize) as *const usize); - let v = *(ptr.offset(offset as isize - usize_bytes as isize) as *const usize); + let u = *(ptr.offset(offset as isize - 2 * chunk_bytes as isize) as *const Chunk); + let v = *(ptr.offset(offset as isize - chunk_bytes as isize) as *const Chunk); // break if there is a matching byte let zu = contains_zero_byte(u ^ repeated_x); @@ -140,91 +132,9 @@ pub fn memrchr(x: u8, text: &[u8]) -> Option<usize> { break; } } - offset -= 2 * usize_bytes; + offset -= 2 * chunk_bytes; } // find the byte before the point the body loop stopped text[..offset].iter().rposition(|elt| *elt == x) } - -// test fallback implementations on all platforms -#[test] -fn matches_one() { - assert_eq!(Some(0), memchr(b'a', b"a")); -} - -#[test] -fn matches_begin() { - assert_eq!(Some(0), memchr(b'a', b"aaaa")); -} - -#[test] -fn matches_end() { - assert_eq!(Some(4), memchr(b'z', b"aaaaz")); -} - -#[test] -fn matches_nul() { - assert_eq!(Some(4), memchr(b'\x00', b"aaaa\x00")); -} - -#[test] -fn matches_past_nul() { - assert_eq!(Some(5), memchr(b'z', b"aaaa\x00z")); -} - -#[test] -fn no_match_empty() { - assert_eq!(None, memchr(b'a', b"")); -} - -#[test] -fn no_match() { - assert_eq!(None, memchr(b'a', b"xyz")); -} - -#[test] -fn matches_one_reversed() { - assert_eq!(Some(0), memrchr(b'a', b"a")); -} - -#[test] -fn matches_begin_reversed() { - assert_eq!(Some(3), memrchr(b'a', b"aaaa")); -} - -#[test] -fn matches_end_reversed() { - assert_eq!(Some(0), memrchr(b'z', b"zaaaa")); -} - -#[test] -fn matches_nul_reversed() { - assert_eq!(Some(4), memrchr(b'\x00', b"aaaa\x00")); -} - -#[test] -fn matches_past_nul_reversed() { - assert_eq!(Some(0), memrchr(b'z', b"z\x00aaaa")); -} - -#[test] -fn no_match_empty_reversed() { - assert_eq!(None, memrchr(b'a', b"")); -} - -#[test] -fn no_match_reversed() { - assert_eq!(None, memrchr(b'a', b"xyz")); -} - -#[test] -fn each_alignment_reversed() { - let mut data = [1u8; 64]; - let needle = 2; - let pos = 40; - data[pos] = needle; - for start in 0..16 { - assert_eq!(Some(pos - start), memrchr(needle, &data[start..])); - } -} diff --git a/src/libcore/slice/mod.rs b/src/libcore/slice/mod.rs index 244bf476caf..a3960556d34 100644 --- a/src/libcore/slice/mod.rs +++ b/src/libcore/slice/mod.rs @@ -59,510 +59,389 @@ mod rotate; mod sort; #[repr(C)] -struct Repr<T> { - pub data: *const T, - pub len: usize, +union Repr<'a, T: 'a> { + rust: &'a [T], + rust_mut: &'a mut [T], + raw: FatPtr<T>, +} + +#[repr(C)] +struct FatPtr<T> { + data: *const T, + len: usize, } // // Extension traits // -/// Extension methods for slices. -#[unstable(feature = "core_slice_ext", - reason = "stable interface provided by `impl [T]` in later crates", - issue = "32110")] -#[allow(missing_docs)] // documented elsewhere -pub trait SliceExt { - type Item; - - #[stable(feature = "core", since = "1.6.0")] - fn split_at(&self, mid: usize) -> (&[Self::Item], &[Self::Item]); - - #[stable(feature = "core", since = "1.6.0")] - fn iter(&self) -> Iter<Self::Item>; - - #[stable(feature = "core", since = "1.6.0")] - fn split<P>(&self, pred: P) -> Split<Self::Item, P> - where P: FnMut(&Self::Item) -> bool; - - #[unstable(feature = "slice_rsplit", issue = "41020")] - fn rsplit<P>(&self, pred: P) -> RSplit<Self::Item, P> - where P: FnMut(&Self::Item) -> bool; - - #[stable(feature = "core", since = "1.6.0")] - fn splitn<P>(&self, n: usize, pred: P) -> SplitN<Self::Item, P> - where P: FnMut(&Self::Item) -> bool; - - #[stable(feature = "core", since = "1.6.0")] - fn rsplitn<P>(&self, n: usize, pred: P) -> RSplitN<Self::Item, P> - where P: FnMut(&Self::Item) -> bool; - - #[stable(feature = "core", since = "1.6.0")] - fn windows(&self, size: usize) -> Windows<Self::Item>; - - #[stable(feature = "core", since = "1.6.0")] - fn chunks(&self, size: usize) -> Chunks<Self::Item>; - - #[unstable(feature = "exact_chunks", issue = "47115")] - fn exact_chunks(&self, size: usize) -> ExactChunks<Self::Item>; - - #[stable(feature = "core", since = "1.6.0")] - fn get<I>(&self, index: I) -> Option<&I::Output> - where I: SliceIndex<Self>; - #[stable(feature = "core", since = "1.6.0")] - fn first(&self) -> Option<&Self::Item>; - - #[stable(feature = "core", since = "1.6.0")] - fn split_first(&self) -> Option<(&Self::Item, &[Self::Item])>; - - #[stable(feature = "core", since = "1.6.0")] - fn split_last(&self) -> Option<(&Self::Item, &[Self::Item])>; - - #[stable(feature = "core", since = "1.6.0")] - fn last(&self) -> Option<&Self::Item>; - - #[stable(feature = "core", since = "1.6.0")] - unsafe fn get_unchecked<I>(&self, index: I) -> &I::Output - where I: SliceIndex<Self>; - #[stable(feature = "core", since = "1.6.0")] - fn as_ptr(&self) -> *const Self::Item; - - #[stable(feature = "core", since = "1.6.0")] - fn binary_search(&self, x: &Self::Item) -> Result<usize, usize> - where Self::Item: Ord; - - #[stable(feature = "core", since = "1.6.0")] - fn binary_search_by<'a, F>(&'a self, f: F) -> Result<usize, usize> - where F: FnMut(&'a Self::Item) -> Ordering; - - #[stable(feature = "slice_binary_search_by_key", since = "1.10.0")] - fn binary_search_by_key<'a, B, F>(&'a self, b: &B, f: F) -> Result<usize, usize> - where F: FnMut(&'a Self::Item) -> B, - B: Ord; - - #[stable(feature = "core", since = "1.6.0")] - fn len(&self) -> usize; - - #[stable(feature = "core", since = "1.6.0")] - fn is_empty(&self) -> bool { self.len() == 0 } - - #[stable(feature = "core", since = "1.6.0")] - fn get_mut<I>(&mut self, index: I) -> Option<&mut I::Output> - where I: SliceIndex<Self>; - #[stable(feature = "core", since = "1.6.0")] - fn iter_mut(&mut self) -> IterMut<Self::Item>; - - #[stable(feature = "core", since = "1.6.0")] - fn first_mut(&mut self) -> Option<&mut Self::Item>; - - #[stable(feature = "core", since = "1.6.0")] - fn split_first_mut(&mut self) -> Option<(&mut Self::Item, &mut [Self::Item])>; - - #[stable(feature = "core", since = "1.6.0")] - fn split_last_mut(&mut self) -> Option<(&mut Self::Item, &mut [Self::Item])>; - - #[stable(feature = "core", since = "1.6.0")] - fn last_mut(&mut self) -> Option<&mut Self::Item>; - - #[stable(feature = "core", since = "1.6.0")] - fn split_mut<P>(&mut self, pred: P) -> SplitMut<Self::Item, P> - where P: FnMut(&Self::Item) -> bool; - - #[unstable(feature = "slice_rsplit", issue = "41020")] - fn rsplit_mut<P>(&mut self, pred: P) -> RSplitMut<Self::Item, P> - where P: FnMut(&Self::Item) -> bool; - - #[stable(feature = "core", since = "1.6.0")] - fn splitn_mut<P>(&mut self, n: usize, pred: P) -> SplitNMut<Self::Item, P> - where P: FnMut(&Self::Item) -> bool; - - #[stable(feature = "core", since = "1.6.0")] - fn rsplitn_mut<P>(&mut self, n: usize, pred: P) -> RSplitNMut<Self::Item, P> - where P: FnMut(&Self::Item) -> bool; - - #[stable(feature = "core", since = "1.6.0")] - fn chunks_mut(&mut self, chunk_size: usize) -> ChunksMut<Self::Item>; - - #[unstable(feature = "exact_chunks", issue = "47115")] - fn exact_chunks_mut(&mut self, size: usize) -> ExactChunksMut<Self::Item>; - - #[stable(feature = "core", since = "1.6.0")] - fn swap(&mut self, a: usize, b: usize); - - #[stable(feature = "core", since = "1.6.0")] - fn split_at_mut(&mut self, mid: usize) -> (&mut [Self::Item], &mut [Self::Item]); - - #[stable(feature = "core", since = "1.6.0")] - fn reverse(&mut self); - - #[stable(feature = "core", since = "1.6.0")] - unsafe fn get_unchecked_mut<I>(&mut self, index: I) -> &mut I::Output - where I: SliceIndex<Self>; - #[stable(feature = "core", since = "1.6.0")] - fn as_mut_ptr(&mut self) -> *mut Self::Item; - - #[stable(feature = "core", since = "1.6.0")] - fn contains(&self, x: &Self::Item) -> bool where Self::Item: PartialEq; - - #[stable(feature = "core", since = "1.6.0")] - fn starts_with(&self, needle: &[Self::Item]) -> bool where Self::Item: PartialEq; - - #[stable(feature = "core", since = "1.6.0")] - fn ends_with(&self, needle: &[Self::Item]) -> bool where Self::Item: PartialEq; - - #[unstable(feature = "slice_rotate", issue = "41891")] - fn rotate_left(&mut self, mid: usize); - - #[unstable(feature = "slice_rotate", issue = "41891")] - fn rotate_right(&mut self, k: usize); - - #[stable(feature = "clone_from_slice", since = "1.7.0")] - fn clone_from_slice(&mut self, src: &[Self::Item]) where Self::Item: Clone; - - #[stable(feature = "copy_from_slice", since = "1.9.0")] - fn copy_from_slice(&mut self, src: &[Self::Item]) where Self::Item: Copy; - - #[unstable(feature = "swap_with_slice", issue = "44030")] - fn swap_with_slice(&mut self, src: &mut [Self::Item]); - - #[stable(feature = "sort_unstable", since = "1.20.0")] - fn sort_unstable(&mut self) - where Self::Item: Ord; - - #[stable(feature = "sort_unstable", since = "1.20.0")] - fn sort_unstable_by<F>(&mut self, compare: F) - where F: FnMut(&Self::Item, &Self::Item) -> Ordering; - - #[stable(feature = "sort_unstable", since = "1.20.0")] - fn sort_unstable_by_key<B, F>(&mut self, f: F) - where F: FnMut(&Self::Item) -> B, - B: Ord; -} - -// Use macros to be generic over const/mut -macro_rules! slice_offset { - ($ptr:expr, $by:expr) => {{ - let ptr = $ptr; - if size_from_ptr(ptr) == 0 { - (ptr as *mut i8).wrapping_offset($by) as _ - } else { - ptr.offset($by) - } - }}; -} - -// make a &T from a *const T -macro_rules! make_ref { - ($ptr:expr) => {{ - let ptr = $ptr; - if size_from_ptr(ptr) == 0 { - // Use a non-null pointer value - &*(1 as *mut _) - } else { - &*ptr - } - }}; -} - -// make a &mut T from a *mut T -macro_rules! make_ref_mut { - ($ptr:expr) => {{ - let ptr = $ptr; - if size_from_ptr(ptr) == 0 { - // Use a non-null pointer value - &mut *(1 as *mut _) - } else { - &mut *ptr - } - }}; -} - -#[unstable(feature = "core_slice_ext", - reason = "stable interface provided by `impl [T]` in later crates", - issue = "32110")] -impl<T> SliceExt for [T] { - type Item = T; - - #[inline] - fn split_at(&self, mid: usize) -> (&[T], &[T]) { - (&self[..mid], &self[mid..]) - } - +#[lang = "slice"] +#[cfg(not(test))] +impl<T> [T] { + /// Returns the number of elements in the slice. + /// + /// # Examples + /// + /// ``` + /// let a = [1, 2, 3]; + /// assert_eq!(a.len(), 3); + /// ``` + #[stable(feature = "rust1", since = "1.0.0")] #[inline] - fn iter(&self) -> Iter<T> { + #[rustc_const_unstable(feature = "const_slice_len")] + pub const fn len(&self) -> usize { unsafe { - let p = if mem::size_of::<T>() == 0 { - 1 as *const _ - } else { - let p = self.as_ptr(); - assume(!p.is_null()); - p - }; - - Iter { - ptr: p, - end: slice_offset!(p, self.len() as isize), - _marker: marker::PhantomData - } + Repr { rust: self }.raw.len } } + /// Returns `true` if the slice has a length of 0. + /// + /// # Examples + /// + /// ``` + /// let a = [1, 2, 3]; + /// assert!(!a.is_empty()); + /// ``` + #[stable(feature = "rust1", since = "1.0.0")] #[inline] - fn split<P>(&self, pred: P) -> Split<T, P> - where P: FnMut(&T) -> bool - { - Split { - v: self, - pred, - finished: false - } - } - - #[inline] - fn rsplit<P>(&self, pred: P) -> RSplit<T, P> - where P: FnMut(&T) -> bool - { - RSplit { inner: self.split(pred) } - } - - #[inline] - fn splitn<P>(&self, n: usize, pred: P) -> SplitN<T, P> - where P: FnMut(&T) -> bool - { - SplitN { - inner: GenericSplitN { - iter: self.split(pred), - count: n - } - } - } - - #[inline] - fn rsplitn<P>(&self, n: usize, pred: P) -> RSplitN<T, P> - where P: FnMut(&T) -> bool - { - RSplitN { - inner: GenericSplitN { - iter: self.rsplit(pred), - count: n - } - } - } - - #[inline] - fn windows(&self, size: usize) -> Windows<T> { - assert!(size != 0); - Windows { v: self, size: size } - } - - #[inline] - fn chunks(&self, chunk_size: usize) -> Chunks<T> { - assert!(chunk_size != 0); - Chunks { v: self, chunk_size: chunk_size } - } - - #[inline] - fn exact_chunks(&self, chunk_size: usize) -> ExactChunks<T> { - assert!(chunk_size != 0); - let rem = self.len() % chunk_size; - let len = self.len() - rem; - ExactChunks { v: &self[..len], chunk_size: chunk_size} - } - - #[inline] - fn get<I>(&self, index: I) -> Option<&I::Output> - where I: SliceIndex<[T]> - { - index.get(self) + #[rustc_const_unstable(feature = "const_slice_len")] + pub const fn is_empty(&self) -> bool { + self.len() == 0 } + /// Returns the first element of the slice, or `None` if it is empty. + /// + /// # Examples + /// + /// ``` + /// let v = [10, 40, 30]; + /// assert_eq!(Some(&10), v.first()); + /// + /// let w: &[i32] = &[]; + /// assert_eq!(None, w.first()); + /// ``` + #[stable(feature = "rust1", since = "1.0.0")] #[inline] - fn first(&self) -> Option<&T> { + pub fn first(&self) -> Option<&T> { if self.is_empty() { None } else { Some(&self[0]) } } + /// Returns a mutable pointer to the first element of the slice, or `None` if it is empty. + /// + /// # Examples + /// + /// ``` + /// let x = &mut [0, 1, 2]; + /// + /// if let Some(first) = x.first_mut() { + /// *first = 5; + /// } + /// assert_eq!(x, &[5, 1, 2]); + /// ``` + #[stable(feature = "rust1", since = "1.0.0")] #[inline] - fn split_first(&self) -> Option<(&T, &[T])> { - if self.is_empty() { None } else { Some((&self[0], &self[1..])) } - } - - #[inline] - fn split_last(&self) -> Option<(&T, &[T])> { - let len = self.len(); - if len == 0 { None } else { Some((&self[len - 1], &self[..(len - 1)])) } - } - - #[inline] - fn last(&self) -> Option<&T> { - if self.is_empty() { None } else { Some(&self[self.len() - 1]) } - } - - #[inline] - unsafe fn get_unchecked<I>(&self, index: I) -> &I::Output - where I: SliceIndex<[T]> - { - index.get_unchecked(self) + pub fn first_mut(&mut self) -> Option<&mut T> { + if self.is_empty() { None } else { Some(&mut self[0]) } } + /// Returns the first and all the rest of the elements of the slice, or `None` if it is empty. + /// + /// # Examples + /// + /// ``` + /// let x = &[0, 1, 2]; + /// + /// if let Some((first, elements)) = x.split_first() { + /// assert_eq!(first, &0); + /// assert_eq!(elements, &[1, 2]); + /// } + /// ``` + #[stable(feature = "slice_splits", since = "1.5.0")] #[inline] - fn as_ptr(&self) -> *const T { - self as *const [T] as *const T - } - - fn binary_search_by<'a, F>(&'a self, mut f: F) -> Result<usize, usize> - where F: FnMut(&'a T) -> Ordering - { - let s = self; - let mut size = s.len(); - if size == 0 { - return Err(0); - } - let mut base = 0usize; - while size > 1 { - let half = size / 2; - let mid = base + half; - // mid is always in [0, size), that means mid is >= 0 and < size. - // mid >= 0: by definition - // mid < size: mid = size / 2 + size / 4 + size / 8 ... - let cmp = f(unsafe { s.get_unchecked(mid) }); - base = if cmp == Greater { base } else { mid }; - size -= half; - } - // base is always in [0, size) because base <= mid. - let cmp = f(unsafe { s.get_unchecked(base) }); - if cmp == Equal { Ok(base) } else { Err(base + (cmp == Less) as usize) } + pub fn split_first(&self) -> Option<(&T, &[T])> { + if self.is_empty() { None } else { Some((&self[0], &self[1..])) } } + /// Returns the first and all the rest of the elements of the slice, or `None` if it is empty. + /// + /// # Examples + /// + /// ``` + /// let x = &mut [0, 1, 2]; + /// + /// if let Some((first, elements)) = x.split_first_mut() { + /// *first = 3; + /// elements[0] = 4; + /// elements[1] = 5; + /// } + /// assert_eq!(x, &[3, 4, 5]); + /// ``` + #[stable(feature = "slice_splits", since = "1.5.0")] #[inline] - fn len(&self) -> usize { - unsafe { - mem::transmute::<&[T], Repr<T>>(self).len + pub fn split_first_mut(&mut self) -> Option<(&mut T, &mut [T])> { + if self.is_empty() { None } else { + let split = self.split_at_mut(1); + Some((&mut split.0[0], split.1)) } } + /// Returns the last and all the rest of the elements of the slice, or `None` if it is empty. + /// + /// # Examples + /// + /// ``` + /// let x = &[0, 1, 2]; + /// + /// if let Some((last, elements)) = x.split_last() { + /// assert_eq!(last, &2); + /// assert_eq!(elements, &[0, 1]); + /// } + /// ``` + #[stable(feature = "slice_splits", since = "1.5.0")] #[inline] - fn get_mut<I>(&mut self, index: I) -> Option<&mut I::Output> - where I: SliceIndex<[T]> - { - index.get_mut(self) + pub fn split_last(&self) -> Option<(&T, &[T])> { + let len = self.len(); + if len == 0 { None } else { Some((&self[len - 1], &self[..(len - 1)])) } } + /// Returns the last and all the rest of the elements of the slice, or `None` if it is empty. + /// + /// # Examples + /// + /// ``` + /// let x = &mut [0, 1, 2]; + /// + /// if let Some((last, elements)) = x.split_last_mut() { + /// *last = 3; + /// elements[0] = 4; + /// elements[1] = 5; + /// } + /// assert_eq!(x, &[4, 5, 3]); + /// ``` + #[stable(feature = "slice_splits", since = "1.5.0")] #[inline] - fn split_at_mut(&mut self, mid: usize) -> (&mut [T], &mut [T]) { + pub fn split_last_mut(&mut self) -> Option<(&mut T, &mut [T])> { let len = self.len(); - let ptr = self.as_mut_ptr(); - - unsafe { - assert!(mid <= len); - - (from_raw_parts_mut(ptr, mid), - from_raw_parts_mut(ptr.offset(mid as isize), len - mid)) + if len == 0 { None } else { + let split = self.split_at_mut(len - 1); + Some((&mut split.1[0], split.0)) } + } + /// Returns the last element of the slice, or `None` if it is empty. + /// + /// # Examples + /// + /// ``` + /// let v = [10, 40, 30]; + /// assert_eq!(Some(&30), v.last()); + /// + /// let w: &[i32] = &[]; + /// assert_eq!(None, w.last()); + /// ``` + #[stable(feature = "rust1", since = "1.0.0")] #[inline] - fn iter_mut(&mut self) -> IterMut<T> { - unsafe { - let p = if mem::size_of::<T>() == 0 { - 1 as *mut _ - } else { - let p = self.as_mut_ptr(); - assume(!p.is_null()); - p - }; - - IterMut { - ptr: p, - end: slice_offset!(p, self.len() as isize), - _marker: marker::PhantomData - } - } + pub fn last(&self) -> Option<&T> { + if self.is_empty() { None } else { Some(&self[self.len() - 1]) } } + /// Returns a mutable pointer to the last item in the slice. + /// + /// # Examples + /// + /// ``` + /// let x = &mut [0, 1, 2]; + /// + /// if let Some(last) = x.last_mut() { + /// *last = 10; + /// } + /// assert_eq!(x, &[0, 1, 10]); + /// ``` + #[stable(feature = "rust1", since = "1.0.0")] #[inline] - fn last_mut(&mut self) -> Option<&mut T> { + pub fn last_mut(&mut self) -> Option<&mut T> { let len = self.len(); if len == 0 { return None; } Some(&mut self[len - 1]) } + /// Returns a reference to an element or subslice depending on the type of + /// index. + /// + /// - If given a position, returns a reference to the element at that + /// position or `None` if out of bounds. + /// - If given a range, returns the subslice corresponding to that range, + /// or `None` if out of bounds. + /// + /// # Examples + /// + /// ``` + /// let v = [10, 40, 30]; + /// assert_eq!(Some(&40), v.get(1)); + /// assert_eq!(Some(&[10, 40][..]), v.get(0..2)); + /// assert_eq!(None, v.get(3)); + /// assert_eq!(None, v.get(0..4)); + /// ``` + #[stable(feature = "rust1", since = "1.0.0")] #[inline] - fn first_mut(&mut self) -> Option<&mut T> { - if self.is_empty() { None } else { Some(&mut self[0]) } - } - - #[inline] - fn split_first_mut(&mut self) -> Option<(&mut T, &mut [T])> { - if self.is_empty() { None } else { - let split = self.split_at_mut(1); - Some((&mut split.0[0], split.1)) - } - } - - #[inline] - fn split_last_mut(&mut self) -> Option<(&mut T, &mut [T])> { - let len = self.len(); - if len == 0 { None } else { - let split = self.split_at_mut(len - 1); - Some((&mut split.1[0], split.0)) - } - } - - #[inline] - fn split_mut<P>(&mut self, pred: P) -> SplitMut<T, P> - where P: FnMut(&T) -> bool + pub fn get<I>(&self, index: I) -> Option<&I::Output> + where I: SliceIndex<Self> { - SplitMut { v: self, pred: pred, finished: false } + index.get(self) } + /// Returns a mutable reference to an element or subslice depending on the + /// type of index (see [`get`]) or `None` if the index is out of bounds. + /// + /// [`get`]: #method.get + /// + /// # Examples + /// + /// ``` + /// let x = &mut [0, 1, 2]; + /// + /// if let Some(elem) = x.get_mut(1) { + /// *elem = 42; + /// } + /// assert_eq!(x, &[0, 42, 2]); + /// ``` + #[stable(feature = "rust1", since = "1.0.0")] #[inline] - fn rsplit_mut<P>(&mut self, pred: P) -> RSplitMut<T, P> - where P: FnMut(&T) -> bool + pub fn get_mut<I>(&mut self, index: I) -> Option<&mut I::Output> + where I: SliceIndex<Self> { - RSplitMut { inner: self.split_mut(pred) } + index.get_mut(self) } + /// Returns a reference to an element or subslice, without doing bounds + /// checking. + /// + /// This is generally not recommended, use with caution! For a safe + /// alternative see [`get`]. + /// + /// [`get`]: #method.get + /// + /// # Examples + /// + /// ``` + /// let x = &[1, 2, 4]; + /// + /// unsafe { + /// assert_eq!(x.get_unchecked(1), &2); + /// } + /// ``` + #[stable(feature = "rust1", since = "1.0.0")] #[inline] - fn splitn_mut<P>(&mut self, n: usize, pred: P) -> SplitNMut<T, P> - where P: FnMut(&T) -> bool + pub unsafe fn get_unchecked<I>(&self, index: I) -> &I::Output + where I: SliceIndex<Self> { - SplitNMut { - inner: GenericSplitN { - iter: self.split_mut(pred), - count: n - } - } + index.get_unchecked(self) } + /// Returns a mutable reference to an element or subslice, without doing + /// bounds checking. + /// + /// This is generally not recommended, use with caution! For a safe + /// alternative see [`get_mut`]. + /// + /// [`get_mut`]: #method.get_mut + /// + /// # Examples + /// + /// ``` + /// let x = &mut [1, 2, 4]; + /// + /// unsafe { + /// let elem = x.get_unchecked_mut(1); + /// *elem = 13; + /// } + /// assert_eq!(x, &[1, 13, 4]); + /// ``` + #[stable(feature = "rust1", since = "1.0.0")] #[inline] - fn rsplitn_mut<P>(&mut self, n: usize, pred: P) -> RSplitNMut<T, P> where - P: FnMut(&T) -> bool, + pub unsafe fn get_unchecked_mut<I>(&mut self, index: I) -> &mut I::Output + where I: SliceIndex<Self> { - RSplitNMut { - inner: GenericSplitN { - iter: self.rsplit_mut(pred), - count: n - } - } + index.get_unchecked_mut(self) } + /// Returns a raw pointer to the slice's buffer. + /// + /// The caller must ensure that the slice outlives the pointer this + /// function returns, or else it will end up pointing to garbage. + /// + /// Modifying the container referenced by this slice may cause its buffer + /// to be reallocated, which would also make any pointers to it invalid. + /// + /// # Examples + /// + /// ``` + /// let x = &[1, 2, 4]; + /// let x_ptr = x.as_ptr(); + /// + /// unsafe { + /// for i in 0..x.len() { + /// assert_eq!(x.get_unchecked(i), &*x_ptr.offset(i as isize)); + /// } + /// } + /// ``` + #[stable(feature = "rust1", since = "1.0.0")] #[inline] - fn chunks_mut(&mut self, chunk_size: usize) -> ChunksMut<T> { - assert!(chunk_size != 0); - ChunksMut { v: self, chunk_size: chunk_size } + #[rustc_const_unstable(feature = "const_slice_as_ptr")] + pub const fn as_ptr(&self) -> *const T { + self as *const [T] as *const T } + /// Returns an unsafe mutable pointer to the slice's buffer. + /// + /// The caller must ensure that the slice outlives the pointer this + /// function returns, or else it will end up pointing to garbage. + /// + /// Modifying the container referenced by this slice may cause its buffer + /// to be reallocated, which would also make any pointers to it invalid. + /// + /// # Examples + /// + /// ``` + /// let x = &mut [1, 2, 4]; + /// let x_ptr = x.as_mut_ptr(); + /// + /// unsafe { + /// for i in 0..x.len() { + /// *x_ptr.offset(i as isize) += 2; + /// } + /// } + /// assert_eq!(x, &[3, 4, 6]); + /// ``` + #[stable(feature = "rust1", since = "1.0.0")] #[inline] - fn exact_chunks_mut(&mut self, chunk_size: usize) -> ExactChunksMut<T> { - assert!(chunk_size != 0); - let rem = self.len() % chunk_size; - let len = self.len() - rem; - ExactChunksMut { v: &mut self[..len], chunk_size: chunk_size} + pub fn as_mut_ptr(&mut self) -> *mut T { + self as *mut [T] as *mut T } + /// Swaps two elements in the slice. + /// + /// # Arguments + /// + /// * a - The index of the first element + /// * b - The index of the second element + /// + /// # Panics + /// + /// Panics if `a` or `b` are out of bounds. + /// + /// # Examples + /// + /// ``` + /// let mut v = ["a", "b", "c", "d"]; + /// v.swap(1, 3); + /// assert!(v == ["a", "d", "c", "b"]); + /// ``` + #[stable(feature = "rust1", since = "1.0.0")] #[inline] - fn swap(&mut self, a: usize, b: usize) { + pub fn swap(&mut self, a: usize, b: usize) { unsafe { // Can't take two mutable loans from one vector, so instead just cast // them to their raw pointers to do the swap @@ -572,7 +451,18 @@ impl<T> SliceExt for [T] { } } - fn reverse(&mut self) { + /// Reverses the order of elements in the slice, in place. + /// + /// # Examples + /// + /// ``` + /// let mut v = [1, 2, 3]; + /// v.reverse(); + /// assert!(v == [3, 2, 1]); + /// ``` + #[stable(feature = "rust1", since = "1.0.0")] + #[inline] + pub fn reverse(&mut self) { let mut i: usize = 0; let ln = self.len(); @@ -635,42 +525,920 @@ impl<T> SliceExt for [T] { } } + /// Returns an iterator over the slice. + /// + /// # Examples + /// + /// ``` + /// let x = &[1, 2, 4]; + /// let mut iterator = x.iter(); + /// + /// assert_eq!(iterator.next(), Some(&1)); + /// assert_eq!(iterator.next(), Some(&2)); + /// assert_eq!(iterator.next(), Some(&4)); + /// assert_eq!(iterator.next(), None); + /// ``` + #[stable(feature = "rust1", since = "1.0.0")] + #[inline] + pub fn iter(&self) -> Iter<T> { + unsafe { + let ptr = self.as_ptr(); + assume(!ptr.is_null()); + + let end = if mem::size_of::<T>() == 0 { + (ptr as *const u8).wrapping_offset(self.len() as isize) as *const T + } else { + ptr.offset(self.len() as isize) + }; + + Iter { + ptr, + end, + _marker: marker::PhantomData + } + } + } + + /// Returns an iterator that allows modifying each value. + /// + /// # Examples + /// + /// ``` + /// let x = &mut [1, 2, 4]; + /// for elem in x.iter_mut() { + /// *elem += 2; + /// } + /// assert_eq!(x, &[3, 4, 6]); + /// ``` + #[stable(feature = "rust1", since = "1.0.0")] + #[inline] + pub fn iter_mut(&mut self) -> IterMut<T> { + unsafe { + let ptr = self.as_mut_ptr(); + assume(!ptr.is_null()); + + let end = if mem::size_of::<T>() == 0 { + (ptr as *mut u8).wrapping_offset(self.len() as isize) as *mut T + } else { + ptr.offset(self.len() as isize) + }; + + IterMut { + ptr, + end, + _marker: marker::PhantomData + } + } + } + + /// Returns an iterator over all contiguous windows of length + /// `size`. The windows overlap. If the slice is shorter than + /// `size`, the iterator returns no values. + /// + /// # Panics + /// + /// Panics if `size` is 0. + /// + /// # Examples + /// + /// ``` + /// let slice = ['r', 'u', 's', 't']; + /// let mut iter = slice.windows(2); + /// assert_eq!(iter.next().unwrap(), &['r', 'u']); + /// assert_eq!(iter.next().unwrap(), &['u', 's']); + /// assert_eq!(iter.next().unwrap(), &['s', 't']); + /// assert!(iter.next().is_none()); + /// ``` + /// + /// If the slice is shorter than `size`: + /// + /// ``` + /// let slice = ['f', 'o', 'o']; + /// let mut iter = slice.windows(4); + /// assert!(iter.next().is_none()); + /// ``` + #[stable(feature = "rust1", since = "1.0.0")] + #[inline] + pub fn windows(&self, size: usize) -> Windows<T> { + assert!(size != 0); + Windows { v: self, size } + } + + /// Returns an iterator over `chunk_size` elements of the slice at a + /// time. The chunks are slices and do not overlap. If `chunk_size` does + /// not divide the length of the slice, then the last chunk will + /// not have length `chunk_size`. + /// + /// See [`exact_chunks`] for a variant of this iterator that returns chunks + /// of always exactly `chunk_size` elements. + /// + /// # Panics + /// + /// Panics if `chunk_size` is 0. + /// + /// # Examples + /// + /// ``` + /// let slice = ['l', 'o', 'r', 'e', 'm']; + /// let mut iter = slice.chunks(2); + /// assert_eq!(iter.next().unwrap(), &['l', 'o']); + /// assert_eq!(iter.next().unwrap(), &['r', 'e']); + /// assert_eq!(iter.next().unwrap(), &['m']); + /// assert!(iter.next().is_none()); + /// ``` + /// + /// [`exact_chunks`]: #method.exact_chunks + #[stable(feature = "rust1", since = "1.0.0")] + #[inline] + pub fn chunks(&self, chunk_size: usize) -> Chunks<T> { + assert!(chunk_size != 0); + Chunks { v: self, chunk_size } + } + + /// Returns an iterator over `chunk_size` elements of the slice at a time. + /// The chunks are mutable slices, and do not overlap. If `chunk_size` does + /// not divide the length of the slice, then the last chunk will not + /// have length `chunk_size`. + /// + /// See [`exact_chunks_mut`] for a variant of this iterator that returns chunks + /// of always exactly `chunk_size` elements. + /// + /// # Panics + /// + /// Panics if `chunk_size` is 0. + /// + /// # Examples + /// + /// ``` + /// let v = &mut [0, 0, 0, 0, 0]; + /// let mut count = 1; + /// + /// for chunk in v.chunks_mut(2) { + /// for elem in chunk.iter_mut() { + /// *elem += count; + /// } + /// count += 1; + /// } + /// assert_eq!(v, &[1, 1, 2, 2, 3]); + /// ``` + /// + /// [`exact_chunks_mut`]: #method.exact_chunks_mut + #[stable(feature = "rust1", since = "1.0.0")] + #[inline] + pub fn chunks_mut(&mut self, chunk_size: usize) -> ChunksMut<T> { + assert!(chunk_size != 0); + ChunksMut { v: self, chunk_size } + } + + /// Returns an iterator over `chunk_size` elements of the slice at a + /// time. The chunks are slices and do not overlap. If `chunk_size` does + /// not divide the length of the slice, then the last up to `chunk_size-1` + /// elements will be omitted and can be retrieved from the `remainder` + /// function of the iterator. + /// + /// Due to each chunk having exactly `chunk_size` elements, the compiler + /// can often optimize the resulting code better than in the case of + /// [`chunks`]. + /// + /// # Panics + /// + /// Panics if `chunk_size` is 0. + /// + /// # Examples + /// + /// ``` + /// #![feature(exact_chunks)] + /// + /// let slice = ['l', 'o', 'r', 'e', 'm']; + /// let mut iter = slice.exact_chunks(2); + /// assert_eq!(iter.next().unwrap(), &['l', 'o']); + /// assert_eq!(iter.next().unwrap(), &['r', 'e']); + /// assert!(iter.next().is_none()); + /// ``` + /// + /// [`chunks`]: #method.chunks + #[unstable(feature = "exact_chunks", issue = "47115")] #[inline] - unsafe fn get_unchecked_mut<I>(&mut self, index: I) -> &mut I::Output - where I: SliceIndex<[T]> + pub fn exact_chunks(&self, chunk_size: usize) -> ExactChunks<T> { + assert!(chunk_size != 0); + let rem = self.len() % chunk_size; + let len = self.len() - rem; + let (fst, snd) = self.split_at(len); + ExactChunks { v: fst, rem: snd, chunk_size } + } + + /// Returns an iterator over `chunk_size` elements of the slice at a time. + /// The chunks are mutable slices, and do not overlap. If `chunk_size` does + /// not divide the length of the slice, then the last up to `chunk_size-1` + /// elements will be omitted and can be retrieved from the `into_remainder` + /// function of the iterator. + /// + /// Due to each chunk having exactly `chunk_size` elements, the compiler + /// can often optimize the resulting code better than in the case of + /// [`chunks_mut`]. + /// + /// # Panics + /// + /// Panics if `chunk_size` is 0. + /// + /// # Examples + /// + /// ``` + /// #![feature(exact_chunks)] + /// + /// let v = &mut [0, 0, 0, 0, 0]; + /// let mut count = 1; + /// + /// for chunk in v.exact_chunks_mut(2) { + /// for elem in chunk.iter_mut() { + /// *elem += count; + /// } + /// count += 1; + /// } + /// assert_eq!(v, &[1, 1, 2, 2, 0]); + /// ``` + /// + /// [`chunks_mut`]: #method.chunks_mut + #[unstable(feature = "exact_chunks", issue = "47115")] + #[inline] + pub fn exact_chunks_mut(&mut self, chunk_size: usize) -> ExactChunksMut<T> { + assert!(chunk_size != 0); + let rem = self.len() % chunk_size; + let len = self.len() - rem; + let (fst, snd) = self.split_at_mut(len); + ExactChunksMut { v: fst, rem: snd, chunk_size } + } + + /// Divides one slice into two at an index. + /// + /// The first will contain all indices from `[0, mid)` (excluding + /// the index `mid` itself) and the second will contain all + /// indices from `[mid, len)` (excluding the index `len` itself). + /// + /// # Panics + /// + /// Panics if `mid > len`. + /// + /// # Examples + /// + /// ``` + /// let v = [1, 2, 3, 4, 5, 6]; + /// + /// { + /// let (left, right) = v.split_at(0); + /// assert!(left == []); + /// assert!(right == [1, 2, 3, 4, 5, 6]); + /// } + /// + /// { + /// let (left, right) = v.split_at(2); + /// assert!(left == [1, 2]); + /// assert!(right == [3, 4, 5, 6]); + /// } + /// + /// { + /// let (left, right) = v.split_at(6); + /// assert!(left == [1, 2, 3, 4, 5, 6]); + /// assert!(right == []); + /// } + /// ``` + #[stable(feature = "rust1", since = "1.0.0")] + #[inline] + pub fn split_at(&self, mid: usize) -> (&[T], &[T]) { + (&self[..mid], &self[mid..]) + } + + /// Divides one mutable slice into two at an index. + /// + /// The first will contain all indices from `[0, mid)` (excluding + /// the index `mid` itself) and the second will contain all + /// indices from `[mid, len)` (excluding the index `len` itself). + /// + /// # Panics + /// + /// Panics if `mid > len`. + /// + /// # Examples + /// + /// ``` + /// let mut v = [1, 0, 3, 0, 5, 6]; + /// // scoped to restrict the lifetime of the borrows + /// { + /// let (left, right) = v.split_at_mut(2); + /// assert!(left == [1, 0]); + /// assert!(right == [3, 0, 5, 6]); + /// left[1] = 2; + /// right[1] = 4; + /// } + /// assert!(v == [1, 2, 3, 4, 5, 6]); + /// ``` + #[stable(feature = "rust1", since = "1.0.0")] + #[inline] + pub fn split_at_mut(&mut self, mid: usize) -> (&mut [T], &mut [T]) { + let len = self.len(); + let ptr = self.as_mut_ptr(); + + unsafe { + assert!(mid <= len); + + (from_raw_parts_mut(ptr, mid), + from_raw_parts_mut(ptr.offset(mid as isize), len - mid)) + } + } + + /// Returns an iterator over subslices separated by elements that match + /// `pred`. The matched element is not contained in the subslices. + /// + /// # Examples + /// + /// ``` + /// let slice = [10, 40, 33, 20]; + /// let mut iter = slice.split(|num| num % 3 == 0); + /// + /// assert_eq!(iter.next().unwrap(), &[10, 40]); + /// assert_eq!(iter.next().unwrap(), &[20]); + /// assert!(iter.next().is_none()); + /// ``` + /// + /// If the first element is matched, an empty slice will be the first item + /// returned by the iterator. Similarly, if the last element in the slice + /// is matched, an empty slice will be the last item returned by the + /// iterator: + /// + /// ``` + /// let slice = [10, 40, 33]; + /// let mut iter = slice.split(|num| num % 3 == 0); + /// + /// assert_eq!(iter.next().unwrap(), &[10, 40]); + /// assert_eq!(iter.next().unwrap(), &[]); + /// assert!(iter.next().is_none()); + /// ``` + /// + /// If two matched elements are directly adjacent, an empty slice will be + /// present between them: + /// + /// ``` + /// let slice = [10, 6, 33, 20]; + /// let mut iter = slice.split(|num| num % 3 == 0); + /// + /// assert_eq!(iter.next().unwrap(), &[10]); + /// assert_eq!(iter.next().unwrap(), &[]); + /// assert_eq!(iter.next().unwrap(), &[20]); + /// assert!(iter.next().is_none()); + /// ``` + #[stable(feature = "rust1", since = "1.0.0")] + #[inline] + pub fn split<F>(&self, pred: F) -> Split<T, F> + where F: FnMut(&T) -> bool { - index.get_unchecked_mut(self) + Split { + v: self, + pred, + finished: false + } } + /// Returns an iterator over mutable subslices separated by elements that + /// match `pred`. The matched element is not contained in the subslices. + /// + /// # Examples + /// + /// ``` + /// let mut v = [10, 40, 30, 20, 60, 50]; + /// + /// for group in v.split_mut(|num| *num % 3 == 0) { + /// group[0] = 1; + /// } + /// assert_eq!(v, [1, 40, 30, 1, 60, 1]); + /// ``` + #[stable(feature = "rust1", since = "1.0.0")] #[inline] - fn as_mut_ptr(&mut self) -> *mut T { - self as *mut [T] as *mut T + pub fn split_mut<F>(&mut self, pred: F) -> SplitMut<T, F> + where F: FnMut(&T) -> bool + { + SplitMut { v: self, pred, finished: false } } + /// Returns an iterator over subslices separated by elements that match + /// `pred`, starting at the end of the slice and working backwards. + /// The matched element is not contained in the subslices. + /// + /// # Examples + /// + /// ``` + /// let slice = [11, 22, 33, 0, 44, 55]; + /// let mut iter = slice.rsplit(|num| *num == 0); + /// + /// assert_eq!(iter.next().unwrap(), &[44, 55]); + /// assert_eq!(iter.next().unwrap(), &[11, 22, 33]); + /// assert_eq!(iter.next(), None); + /// ``` + /// + /// As with `split()`, if the first or last element is matched, an empty + /// slice will be the first (or last) item returned by the iterator. + /// + /// ``` + /// let v = &[0, 1, 1, 2, 3, 5, 8]; + /// let mut it = v.rsplit(|n| *n % 2 == 0); + /// assert_eq!(it.next().unwrap(), &[]); + /// assert_eq!(it.next().unwrap(), &[3, 5]); + /// assert_eq!(it.next().unwrap(), &[1, 1]); + /// assert_eq!(it.next().unwrap(), &[]); + /// assert_eq!(it.next(), None); + /// ``` + #[stable(feature = "slice_rsplit", since = "1.27.0")] #[inline] - fn contains(&self, x: &T) -> bool where T: PartialEq { - x.slice_contains(self) + pub fn rsplit<F>(&self, pred: F) -> RSplit<T, F> + where F: FnMut(&T) -> bool + { + RSplit { inner: self.split(pred) } } + /// Returns an iterator over mutable subslices separated by elements that + /// match `pred`, starting at the end of the slice and working + /// backwards. The matched element is not contained in the subslices. + /// + /// # Examples + /// + /// ``` + /// let mut v = [100, 400, 300, 200, 600, 500]; + /// + /// let mut count = 0; + /// for group in v.rsplit_mut(|num| *num % 3 == 0) { + /// count += 1; + /// group[0] = count; + /// } + /// assert_eq!(v, [3, 400, 300, 2, 600, 1]); + /// ``` + /// + #[stable(feature = "slice_rsplit", since = "1.27.0")] + #[inline] + pub fn rsplit_mut<F>(&mut self, pred: F) -> RSplitMut<T, F> + where F: FnMut(&T) -> bool + { + RSplitMut { inner: self.split_mut(pred) } + } + + /// Returns an iterator over subslices separated by elements that match + /// `pred`, limited to returning at most `n` items. The matched element is + /// not contained in the subslices. + /// + /// The last element returned, if any, will contain the remainder of the + /// slice. + /// + /// # Examples + /// + /// Print the slice split once by numbers divisible by 3 (i.e. `[10, 40]`, + /// `[20, 60, 50]`): + /// + /// ``` + /// let v = [10, 40, 30, 20, 60, 50]; + /// + /// for group in v.splitn(2, |num| *num % 3 == 0) { + /// println!("{:?}", group); + /// } + /// ``` + #[stable(feature = "rust1", since = "1.0.0")] #[inline] - fn starts_with(&self, needle: &[T]) -> bool where T: PartialEq { + pub fn splitn<F>(&self, n: usize, pred: F) -> SplitN<T, F> + where F: FnMut(&T) -> bool + { + SplitN { + inner: GenericSplitN { + iter: self.split(pred), + count: n + } + } + } + + /// Returns an iterator over subslices separated by elements that match + /// `pred`, limited to returning at most `n` items. The matched element is + /// not contained in the subslices. + /// + /// The last element returned, if any, will contain the remainder of the + /// slice. + /// + /// # Examples + /// + /// ``` + /// let mut v = [10, 40, 30, 20, 60, 50]; + /// + /// for group in v.splitn_mut(2, |num| *num % 3 == 0) { + /// group[0] = 1; + /// } + /// assert_eq!(v, [1, 40, 30, 1, 60, 50]); + /// ``` + #[stable(feature = "rust1", since = "1.0.0")] + #[inline] + pub fn splitn_mut<F>(&mut self, n: usize, pred: F) -> SplitNMut<T, F> + where F: FnMut(&T) -> bool + { + SplitNMut { + inner: GenericSplitN { + iter: self.split_mut(pred), + count: n + } + } + } + + /// Returns an iterator over subslices separated by elements that match + /// `pred` limited to returning at most `n` items. This starts at the end of + /// the slice and works backwards. The matched element is not contained in + /// the subslices. + /// + /// The last element returned, if any, will contain the remainder of the + /// slice. + /// + /// # Examples + /// + /// Print the slice split once, starting from the end, by numbers divisible + /// by 3 (i.e. `[50]`, `[10, 40, 30, 20]`): + /// + /// ``` + /// let v = [10, 40, 30, 20, 60, 50]; + /// + /// for group in v.rsplitn(2, |num| *num % 3 == 0) { + /// println!("{:?}", group); + /// } + /// ``` + #[stable(feature = "rust1", since = "1.0.0")] + #[inline] + pub fn rsplitn<F>(&self, n: usize, pred: F) -> RSplitN<T, F> + where F: FnMut(&T) -> bool + { + RSplitN { + inner: GenericSplitN { + iter: self.rsplit(pred), + count: n + } + } + } + + /// Returns an iterator over subslices separated by elements that match + /// `pred` limited to returning at most `n` items. This starts at the end of + /// the slice and works backwards. The matched element is not contained in + /// the subslices. + /// + /// The last element returned, if any, will contain the remainder of the + /// slice. + /// + /// # Examples + /// + /// ``` + /// let mut s = [10, 40, 30, 20, 60, 50]; + /// + /// for group in s.rsplitn_mut(2, |num| *num % 3 == 0) { + /// group[0] = 1; + /// } + /// assert_eq!(s, [1, 40, 30, 20, 60, 1]); + /// ``` + #[stable(feature = "rust1", since = "1.0.0")] + #[inline] + pub fn rsplitn_mut<F>(&mut self, n: usize, pred: F) -> RSplitNMut<T, F> + where F: FnMut(&T) -> bool + { + RSplitNMut { + inner: GenericSplitN { + iter: self.rsplit_mut(pred), + count: n + } + } + } + + /// Returns `true` if the slice contains an element with the given value. + /// + /// # Examples + /// + /// ``` + /// let v = [10, 40, 30]; + /// assert!(v.contains(&30)); + /// assert!(!v.contains(&50)); + /// ``` + #[stable(feature = "rust1", since = "1.0.0")] + pub fn contains(&self, x: &T) -> bool + where T: PartialEq + { + x.slice_contains(self) + } + + /// Returns `true` if `needle` is a prefix of the slice. + /// + /// # Examples + /// + /// ``` + /// let v = [10, 40, 30]; + /// assert!(v.starts_with(&[10])); + /// assert!(v.starts_with(&[10, 40])); + /// assert!(!v.starts_with(&[50])); + /// assert!(!v.starts_with(&[10, 50])); + /// ``` + /// + /// Always returns `true` if `needle` is an empty slice: + /// + /// ``` + /// let v = &[10, 40, 30]; + /// assert!(v.starts_with(&[])); + /// let v: &[u8] = &[]; + /// assert!(v.starts_with(&[])); + /// ``` + #[stable(feature = "rust1", since = "1.0.0")] + pub fn starts_with(&self, needle: &[T]) -> bool + where T: PartialEq + { let n = needle.len(); self.len() >= n && needle == &self[..n] } - #[inline] - fn ends_with(&self, needle: &[T]) -> bool where T: PartialEq { + /// Returns `true` if `needle` is a suffix of the slice. + /// + /// # Examples + /// + /// ``` + /// let v = [10, 40, 30]; + /// assert!(v.ends_with(&[30])); + /// assert!(v.ends_with(&[40, 30])); + /// assert!(!v.ends_with(&[50])); + /// assert!(!v.ends_with(&[50, 30])); + /// ``` + /// + /// Always returns `true` if `needle` is an empty slice: + /// + /// ``` + /// let v = &[10, 40, 30]; + /// assert!(v.ends_with(&[])); + /// let v: &[u8] = &[]; + /// assert!(v.ends_with(&[])); + /// ``` + #[stable(feature = "rust1", since = "1.0.0")] + pub fn ends_with(&self, needle: &[T]) -> bool + where T: PartialEq + { let (m, n) = (self.len(), needle.len()); m >= n && needle == &self[m-n..] } - fn binary_search(&self, x: &T) -> Result<usize, usize> + /// Binary searches this sorted slice for a given element. + /// + /// If the value is found then `Ok` is returned, containing the + /// index of the matching element; if the value is not found then + /// `Err` is returned, containing the index where a matching + /// element could be inserted while maintaining sorted order. + /// + /// # Examples + /// + /// Looks up a series of four elements. The first is found, with a + /// uniquely determined position; the second and third are not + /// found; the fourth could match any position in `[1, 4]`. + /// + /// ``` + /// let s = [0, 1, 1, 1, 1, 2, 3, 5, 8, 13, 21, 34, 55]; + /// + /// assert_eq!(s.binary_search(&13), Ok(9)); + /// assert_eq!(s.binary_search(&4), Err(7)); + /// assert_eq!(s.binary_search(&100), Err(13)); + /// let r = s.binary_search(&1); + /// assert!(match r { Ok(1..=4) => true, _ => false, }); + /// ``` + #[stable(feature = "rust1", since = "1.0.0")] + pub fn binary_search(&self, x: &T) -> Result<usize, usize> where T: Ord { self.binary_search_by(|p| p.cmp(x)) } - fn rotate_left(&mut self, mid: usize) { + /// Binary searches this sorted slice with a comparator function. + /// + /// The comparator function should implement an order consistent + /// with the sort order of the underlying slice, returning an + /// order code that indicates whether its argument is `Less`, + /// `Equal` or `Greater` the desired target. + /// + /// If a matching value is found then returns `Ok`, containing + /// the index for the matched element; if no match is found then + /// `Err` is returned, containing the index where a matching + /// element could be inserted while maintaining sorted order. + /// + /// # Examples + /// + /// Looks up a series of four elements. The first is found, with a + /// uniquely determined position; the second and third are not + /// found; the fourth could match any position in `[1, 4]`. + /// + /// ``` + /// let s = [0, 1, 1, 1, 1, 2, 3, 5, 8, 13, 21, 34, 55]; + /// + /// let seek = 13; + /// assert_eq!(s.binary_search_by(|probe| probe.cmp(&seek)), Ok(9)); + /// let seek = 4; + /// assert_eq!(s.binary_search_by(|probe| probe.cmp(&seek)), Err(7)); + /// let seek = 100; + /// assert_eq!(s.binary_search_by(|probe| probe.cmp(&seek)), Err(13)); + /// let seek = 1; + /// let r = s.binary_search_by(|probe| probe.cmp(&seek)); + /// assert!(match r { Ok(1..=4) => true, _ => false, }); + /// ``` + #[stable(feature = "rust1", since = "1.0.0")] + #[inline] + pub fn binary_search_by<'a, F>(&'a self, mut f: F) -> Result<usize, usize> + where F: FnMut(&'a T) -> Ordering + { + let s = self; + let mut size = s.len(); + if size == 0 { + return Err(0); + } + let mut base = 0usize; + while size > 1 { + let half = size / 2; + let mid = base + half; + // mid is always in [0, size), that means mid is >= 0 and < size. + // mid >= 0: by definition + // mid < size: mid = size / 2 + size / 4 + size / 8 ... + let cmp = f(unsafe { s.get_unchecked(mid) }); + base = if cmp == Greater { base } else { mid }; + size -= half; + } + // base is always in [0, size) because base <= mid. + let cmp = f(unsafe { s.get_unchecked(base) }); + if cmp == Equal { Ok(base) } else { Err(base + (cmp == Less) as usize) } + + } + + /// Binary searches this sorted slice with a key extraction function. + /// + /// Assumes that the slice is sorted by the key, for instance with + /// [`sort_by_key`] using the same key extraction function. + /// + /// If a matching value is found then returns `Ok`, containing the + /// index for the matched element; if no match is found then `Err` + /// is returned, containing the index where a matching element could + /// be inserted while maintaining sorted order. + /// + /// [`sort_by_key`]: #method.sort_by_key + /// + /// # Examples + /// + /// Looks up a series of four elements in a slice of pairs sorted by + /// their second elements. The first is found, with a uniquely + /// determined position; the second and third are not found; the + /// fourth could match any position in `[1, 4]`. + /// + /// ``` + /// let s = [(0, 0), (2, 1), (4, 1), (5, 1), (3, 1), + /// (1, 2), (2, 3), (4, 5), (5, 8), (3, 13), + /// (1, 21), (2, 34), (4, 55)]; + /// + /// assert_eq!(s.binary_search_by_key(&13, |&(a,b)| b), Ok(9)); + /// assert_eq!(s.binary_search_by_key(&4, |&(a,b)| b), Err(7)); + /// assert_eq!(s.binary_search_by_key(&100, |&(a,b)| b), Err(13)); + /// let r = s.binary_search_by_key(&1, |&(a,b)| b); + /// assert!(match r { Ok(1..=4) => true, _ => false, }); + /// ``` + #[stable(feature = "slice_binary_search_by_key", since = "1.10.0")] + #[inline] + pub fn binary_search_by_key<'a, B, F>(&'a self, b: &B, mut f: F) -> Result<usize, usize> + where F: FnMut(&'a T) -> B, + B: Ord + { + self.binary_search_by(|k| f(k).cmp(b)) + } + + /// Sorts the slice, but may not preserve the order of equal elements. + /// + /// This sort is unstable (i.e. may reorder equal elements), in-place (i.e. does not allocate), + /// and `O(n log n)` worst-case. + /// + /// # Current implementation + /// + /// The current algorithm is based on [pattern-defeating quicksort][pdqsort] by Orson Peters, + /// which combines the fast average case of randomized quicksort with the fast worst case of + /// heapsort, while achieving linear time on slices with certain patterns. It uses some + /// randomization to avoid degenerate cases, but with a fixed seed to always provide + /// deterministic behavior. + /// + /// It is typically faster than stable sorting, except in a few special cases, e.g. when the + /// slice consists of several concatenated sorted sequences. + /// + /// # Examples + /// + /// ``` + /// let mut v = [-5, 4, 1, -3, 2]; + /// + /// v.sort_unstable(); + /// assert!(v == [-5, -3, 1, 2, 4]); + /// ``` + /// + /// [pdqsort]: https://github.com/orlp/pdqsort + #[stable(feature = "sort_unstable", since = "1.20.0")] + #[inline] + pub fn sort_unstable(&mut self) + where T: Ord + { + sort::quicksort(self, |a, b| a.lt(b)); + } + + /// Sorts the slice with a comparator function, but may not preserve the order of equal + /// elements. + /// + /// This sort is unstable (i.e. may reorder equal elements), in-place (i.e. does not allocate), + /// and `O(n log n)` worst-case. + /// + /// # Current implementation + /// + /// The current algorithm is based on [pattern-defeating quicksort][pdqsort] by Orson Peters, + /// which combines the fast average case of randomized quicksort with the fast worst case of + /// heapsort, while achieving linear time on slices with certain patterns. It uses some + /// randomization to avoid degenerate cases, but with a fixed seed to always provide + /// deterministic behavior. + /// + /// It is typically faster than stable sorting, except in a few special cases, e.g. when the + /// slice consists of several concatenated sorted sequences. + /// + /// # Examples + /// + /// ``` + /// let mut v = [5, 4, 1, 3, 2]; + /// v.sort_unstable_by(|a, b| a.cmp(b)); + /// assert!(v == [1, 2, 3, 4, 5]); + /// + /// // reverse sorting + /// v.sort_unstable_by(|a, b| b.cmp(a)); + /// assert!(v == [5, 4, 3, 2, 1]); + /// ``` + /// + /// [pdqsort]: https://github.com/orlp/pdqsort + #[stable(feature = "sort_unstable", since = "1.20.0")] + #[inline] + pub fn sort_unstable_by<F>(&mut self, mut compare: F) + where F: FnMut(&T, &T) -> Ordering + { + sort::quicksort(self, |a, b| compare(a, b) == Ordering::Less); + } + + /// Sorts the slice with a key extraction function, but may not preserve the order of equal + /// elements. + /// + /// This sort is unstable (i.e. may reorder equal elements), in-place (i.e. does not allocate), + /// and `O(m n log(m n))` worst-case, where the key function is `O(m)`. + /// + /// # Current implementation + /// + /// The current algorithm is based on [pattern-defeating quicksort][pdqsort] by Orson Peters, + /// which combines the fast average case of randomized quicksort with the fast worst case of + /// heapsort, while achieving linear time on slices with certain patterns. It uses some + /// randomization to avoid degenerate cases, but with a fixed seed to always provide + /// deterministic behavior. + /// + /// # Examples + /// + /// ``` + /// let mut v = [-5i32, 4, 1, -3, 2]; + /// + /// v.sort_unstable_by_key(|k| k.abs()); + /// assert!(v == [1, 2, -3, 4, -5]); + /// ``` + /// + /// [pdqsort]: https://github.com/orlp/pdqsort + #[stable(feature = "sort_unstable", since = "1.20.0")] + #[inline] + pub fn sort_unstable_by_key<K, F>(&mut self, mut f: F) + where F: FnMut(&T) -> K, K: Ord + { + sort::quicksort(self, |a, b| f(a).lt(&f(b))); + } + + /// Rotates the slice in-place such that the first `mid` elements of the + /// slice move to the end while the last `self.len() - mid` elements move to + /// the front. After calling `rotate_left`, the element previously at index + /// `mid` will become the first element in the slice. + /// + /// # Panics + /// + /// This function will panic if `mid` is greater than the length of the + /// slice. Note that `mid == self.len()` does _not_ panic and is a no-op + /// rotation. + /// + /// # Complexity + /// + /// Takes linear (in `self.len()`) time. + /// + /// # Examples + /// + /// ``` + /// let mut a = ['a', 'b', 'c', 'd', 'e', 'f']; + /// a.rotate_left(2); + /// assert_eq!(a, ['c', 'd', 'e', 'f', 'a', 'b']); + /// ``` + /// + /// Rotating a subslice: + /// + /// ``` + /// let mut a = ['a', 'b', 'c', 'd', 'e', 'f']; + /// a[1..5].rotate_left(1); + /// assert_eq!(a, ['a', 'c', 'd', 'e', 'b', 'f']); + /// ``` + #[stable(feature = "slice_rotate", since = "1.26.0")] + pub fn rotate_left(&mut self, mid: usize) { assert!(mid <= self.len()); let k = self.len() - mid; @@ -680,7 +1448,38 @@ impl<T> SliceExt for [T] { } } - fn rotate_right(&mut self, k: usize) { + /// Rotates the slice in-place such that the first `self.len() - k` + /// elements of the slice move to the end while the last `k` elements move + /// to the front. After calling `rotate_right`, the element previously at + /// index `self.len() - k` will become the first element in the slice. + /// + /// # Panics + /// + /// This function will panic if `k` is greater than the length of the + /// slice. Note that `k == self.len()` does _not_ panic and is a no-op + /// rotation. + /// + /// # Complexity + /// + /// Takes linear (in `self.len()`) time. + /// + /// # Examples + /// + /// ``` + /// let mut a = ['a', 'b', 'c', 'd', 'e', 'f']; + /// a.rotate_right(2); + /// assert_eq!(a, ['e', 'f', 'a', 'b', 'c', 'd']); + /// ``` + /// + /// Rotate a subslice: + /// + /// ``` + /// let mut a = ['a', 'b', 'c', 'd', 'e', 'f']; + /// a[1..5].rotate_right(1); + /// assert_eq!(a, ['a', 'e', 'b', 'c', 'd', 'f']); + /// ``` + #[stable(feature = "slice_rotate", since = "1.26.0")] + pub fn rotate_right(&mut self, k: usize) { assert!(k <= self.len()); let mid = self.len() - k; @@ -690,8 +1489,63 @@ impl<T> SliceExt for [T] { } } - #[inline] - fn clone_from_slice(&mut self, src: &[T]) where T: Clone { + /// Copies the elements from `src` into `self`. + /// + /// The length of `src` must be the same as `self`. + /// + /// If `src` implements `Copy`, it can be more performant to use + /// [`copy_from_slice`]. + /// + /// # Panics + /// + /// This function will panic if the two slices have different lengths. + /// + /// # Examples + /// + /// Cloning two elements from a slice into another: + /// + /// ``` + /// let src = [1, 2, 3, 4]; + /// let mut dst = [0, 0]; + /// + /// // Because the slices have to be the same length, + /// // we slice the source slice from four elements + /// // to two. It will panic if we don't do this. + /// dst.clone_from_slice(&src[2..]); + /// + /// assert_eq!(src, [1, 2, 3, 4]); + /// assert_eq!(dst, [3, 4]); + /// ``` + /// + /// Rust enforces that there can only be one mutable reference with no + /// immutable references to a particular piece of data in a particular + /// scope. Because of this, attempting to use `clone_from_slice` on a + /// single slice will result in a compile failure: + /// + /// ```compile_fail + /// let mut slice = [1, 2, 3, 4, 5]; + /// + /// slice[..2].clone_from_slice(&slice[3..]); // compile fail! + /// ``` + /// + /// To work around this, we can use [`split_at_mut`] to create two distinct + /// sub-slices from a slice: + /// + /// ``` + /// let mut slice = [1, 2, 3, 4, 5]; + /// + /// { + /// let (left, right) = slice.split_at_mut(2); + /// left.clone_from_slice(&right[1..]); + /// } + /// + /// assert_eq!(slice, [4, 5, 3, 4, 5]); + /// ``` + /// + /// [`copy_from_slice`]: #method.copy_from_slice + /// [`split_at_mut`]: #method.split_at_mut + #[stable(feature = "clone_from_slice", since = "1.7.0")] + pub fn clone_from_slice(&mut self, src: &[T]) where T: Clone { assert!(self.len() == src.len(), "destination and source slices have different lengths"); // NOTE: We need to explicitly slice them to the same length @@ -702,57 +1556,355 @@ impl<T> SliceExt for [T] { for i in 0..len { self[i].clone_from(&src[i]); } + } - #[inline] - fn copy_from_slice(&mut self, src: &[T]) where T: Copy { - assert!(self.len() == src.len(), - "destination and source slices have different lengths"); + /// Copies all elements from `src` into `self`, using a memcpy. + /// + /// The length of `src` must be the same as `self`. + /// + /// If `src` does not implement `Copy`, use [`clone_from_slice`]. + /// + /// # Panics + /// + /// This function will panic if the two slices have different lengths. + /// + /// # Examples + /// + /// Copying two elements from a slice into another: + /// + /// ``` + /// let src = [1, 2, 3, 4]; + /// let mut dst = [0, 0]; + /// + /// // Because the slices have to be the same length, + /// // we slice the source slice from four elements + /// // to two. It will panic if we don't do this. + /// dst.copy_from_slice(&src[2..]); + /// + /// assert_eq!(src, [1, 2, 3, 4]); + /// assert_eq!(dst, [3, 4]); + /// ``` + /// + /// Rust enforces that there can only be one mutable reference with no + /// immutable references to a particular piece of data in a particular + /// scope. Because of this, attempting to use `copy_from_slice` on a + /// single slice will result in a compile failure: + /// + /// ```compile_fail + /// let mut slice = [1, 2, 3, 4, 5]; + /// + /// slice[..2].copy_from_slice(&slice[3..]); // compile fail! + /// ``` + /// + /// To work around this, we can use [`split_at_mut`] to create two distinct + /// sub-slices from a slice: + /// + /// ``` + /// let mut slice = [1, 2, 3, 4, 5]; + /// + /// { + /// let (left, right) = slice.split_at_mut(2); + /// left.copy_from_slice(&right[1..]); + /// } + /// + /// assert_eq!(slice, [4, 5, 3, 4, 5]); + /// ``` + /// + /// [`clone_from_slice`]: #method.clone_from_slice + /// [`split_at_mut`]: #method.split_at_mut + #[stable(feature = "copy_from_slice", since = "1.9.0")] + pub fn copy_from_slice(&mut self, src: &[T]) where T: Copy { + assert_eq!(self.len(), src.len(), + "destination and source slices have different lengths"); unsafe { ptr::copy_nonoverlapping( src.as_ptr(), self.as_mut_ptr(), self.len()); } } - #[inline] - fn swap_with_slice(&mut self, src: &mut [T]) { - assert!(self.len() == src.len(), + /// Swaps all elements in `self` with those in `other`. + /// + /// The length of `other` must be the same as `self`. + /// + /// # Panics + /// + /// This function will panic if the two slices have different lengths. + /// + /// # Example + /// + /// Swapping two elements across slices: + /// + /// ``` + /// let mut slice1 = [0, 0]; + /// let mut slice2 = [1, 2, 3, 4]; + /// + /// slice1.swap_with_slice(&mut slice2[2..]); + /// + /// assert_eq!(slice1, [3, 4]); + /// assert_eq!(slice2, [1, 2, 0, 0]); + /// ``` + /// + /// Rust enforces that there can only be one mutable reference to a + /// particular piece of data in a particular scope. Because of this, + /// attempting to use `swap_with_slice` on a single slice will result in + /// a compile failure: + /// + /// ```compile_fail + /// let mut slice = [1, 2, 3, 4, 5]; + /// slice[..2].swap_with_slice(&mut slice[3..]); // compile fail! + /// ``` + /// + /// To work around this, we can use [`split_at_mut`] to create two distinct + /// mutable sub-slices from a slice: + /// + /// ``` + /// let mut slice = [1, 2, 3, 4, 5]; + /// + /// { + /// let (left, right) = slice.split_at_mut(2); + /// left.swap_with_slice(&mut right[1..]); + /// } + /// + /// assert_eq!(slice, [4, 5, 3, 1, 2]); + /// ``` + /// + /// [`split_at_mut`]: #method.split_at_mut + #[stable(feature = "swap_with_slice", since = "1.27.0")] + pub fn swap_with_slice(&mut self, other: &mut [T]) { + assert!(self.len() == other.len(), "destination and source slices have different lengths"); unsafe { ptr::swap_nonoverlapping( - self.as_mut_ptr(), src.as_mut_ptr(), self.len()); + self.as_mut_ptr(), other.as_mut_ptr(), self.len()); } } + /// Function to calculate lenghts of the middle and trailing slice for `align_to{,_mut}`. + fn align_to_offsets<U>(&self) -> (usize, usize) { + // What we gonna do about `rest` is figure out what multiple of `U`s we can put in a + // lowest number of `T`s. And how many `T`s we need for each such "multiple". + // + // Consider for example T=u8 U=u16. Then we can put 1 U in 2 Ts. Simple. Now, consider + // for example a case where size_of::<T> = 16, size_of::<U> = 24. We can put 2 Us in + // place of every 3 Ts in the `rest` slice. A bit more complicated. + // + // Formula to calculate this is: + // + // Us = lcm(size_of::<T>, size_of::<U>) / size_of::<U> + // Ts = lcm(size_of::<T>, size_of::<U>) / size_of::<T> + // + // Expanded and simplified: + // + // Us = size_of::<T> / gcd(size_of::<T>, size_of::<U>) + // Ts = size_of::<U> / gcd(size_of::<T>, size_of::<U>) + // + // Luckily since all this is constant-evaluated... performance here matters not! + #[inline] + fn gcd(a: usize, b: usize) -> usize { + // iterative stein’s algorithm + // We should still make this `const fn` (and revert to recursive algorithm if we do) + // because relying on llvm to consteval all this is… well, it makes me + let (ctz_a, mut ctz_b) = unsafe { + if a == 0 { return b; } + if b == 0 { return a; } + (::intrinsics::cttz_nonzero(a), ::intrinsics::cttz_nonzero(b)) + }; + let k = ctz_a.min(ctz_b); + let mut a = a >> ctz_a; + let mut b = b; + loop { + // remove all factors of 2 from b + b >>= ctz_b; + if a > b { + ::mem::swap(&mut a, &mut b); + } + b = b - a; + unsafe { + if b == 0 { + break; + } + ctz_b = ::intrinsics::cttz_nonzero(b); + } + } + a << k + } + let gcd: usize = gcd(::mem::size_of::<T>(), ::mem::size_of::<U>()); + let ts: usize = ::mem::size_of::<U>() / gcd; + let us: usize = ::mem::size_of::<T>() / gcd; + + // Armed with this knowledge, we can find how many `U`s we can fit! + let us_len = self.len() / ts * us; + // And how many `T`s will be in the trailing slice! + let ts_len = self.len() % ts; + (us_len, ts_len) + } + + /// Transmute the slice to a slice of another type, ensuring aligment of the types is + /// maintained. + /// + /// This method splits the slice into three distinct slices: prefix, correctly aligned middle + /// slice of a new type, and the suffix slice. The middle slice will have the greatest length + /// possible for a given type and input slice. + /// + /// This method has no purpose when either input element `T` or output element `U` are + /// zero-sized and will return the original slice without splitting anything. + /// + /// # Unsafety + /// + /// This method is essentially a `transmute` with respect to the elements in the returned + /// middle slice, so all the usual caveats pertaining to `transmute::<T, U>` also apply here. + /// + /// # Examples + /// + /// Basic usage: + /// + /// ``` + /// # #![feature(slice_align_to)] + /// unsafe { + /// let bytes: [u8; 7] = [1, 2, 3, 4, 5, 6, 7]; + /// let (prefix, shorts, suffix) = bytes.align_to::<u16>(); + /// // less_efficient_algorithm_for_bytes(prefix); + /// // more_efficient_algorithm_for_aligned_shorts(shorts); + /// // less_efficient_algorithm_for_bytes(suffix); + /// } + /// ``` + #[unstable(feature = "slice_align_to", issue = "44488")] + pub unsafe fn align_to<U>(&self) -> (&[T], &[U], &[T]) { + // Note that most of this function will be constant-evaluated, + if ::mem::size_of::<U>() == 0 || ::mem::size_of::<T>() == 0 { + // handle ZSTs specially, which is – don't handle them at all. + return (self, &[], &[]); + } + + // First, find at what point do we split between the first and 2nd slice. Easy with + // ptr.align_offset. + let ptr = self.as_ptr(); + let offset = ::ptr::align_offset(ptr, ::mem::align_of::<U>()); + if offset > self.len() { + (self, &[], &[]) + } else { + let (left, rest) = self.split_at(offset); + // now `rest` is definitely aligned, so `from_raw_parts_mut` below is okay + let (us_len, ts_len) = rest.align_to_offsets::<U>(); + (left, + from_raw_parts(rest.as_ptr() as *const U, us_len), + from_raw_parts(rest.as_ptr().offset((rest.len() - ts_len) as isize), ts_len)) + } + } + + /// Transmute the slice to a slice of another type, ensuring aligment of the types is + /// maintained. + /// + /// This method splits the slice into three distinct slices: prefix, correctly aligned middle + /// slice of a new type, and the suffix slice. The middle slice will have the greatest length + /// possible for a given type and input slice. + /// + /// This method has no purpose when either input element `T` or output element `U` are + /// zero-sized and will return the original slice without splitting anything. + /// + /// # Unsafety + /// + /// This method is essentially a `transmute` with respect to the elements in the returned + /// middle slice, so all the usual caveats pertaining to `transmute::<T, U>` also apply here. + /// + /// # Examples + /// + /// Basic usage: + /// + /// ``` + /// # #![feature(slice_align_to)] + /// unsafe { + /// let mut bytes: [u8; 7] = [1, 2, 3, 4, 5, 6, 7]; + /// let (prefix, shorts, suffix) = bytes.align_to_mut::<u16>(); + /// // less_efficient_algorithm_for_bytes(prefix); + /// // more_efficient_algorithm_for_aligned_shorts(shorts); + /// // less_efficient_algorithm_for_bytes(suffix); + /// } + /// ``` + #[unstable(feature = "slice_align_to", issue = "44488")] + pub unsafe fn align_to_mut<U>(&mut self) -> (&mut [T], &mut [U], &mut [T]) { + // Note that most of this function will be constant-evaluated, + if ::mem::size_of::<U>() == 0 || ::mem::size_of::<T>() == 0 { + // handle ZSTs specially, which is – don't handle them at all. + return (self, &mut [], &mut []); + } + + // First, find at what point do we split between the first and 2nd slice. Easy with + // ptr.align_offset. + let ptr = self.as_ptr(); + let offset = ::ptr::align_offset(ptr, ::mem::align_of::<U>()); + if offset > self.len() { + (self, &mut [], &mut []) + } else { + let (left, rest) = self.split_at_mut(offset); + // now `rest` is definitely aligned, so `from_raw_parts_mut` below is okay + let (us_len, ts_len) = rest.align_to_offsets::<U>(); + let mut_ptr = rest.as_mut_ptr(); + (left, + from_raw_parts_mut(mut_ptr as *mut U, us_len), + from_raw_parts_mut(mut_ptr.offset((rest.len() - ts_len) as isize), ts_len)) + } + } +} + +#[lang = "slice_u8"] +#[cfg(not(test))] +impl [u8] { + /// Checks if all bytes in this slice are within the ASCII range. + #[stable(feature = "ascii_methods_on_intrinsics", since = "1.23.0")] #[inline] - fn binary_search_by_key<'a, B, F>(&'a self, b: &B, mut f: F) -> Result<usize, usize> - where F: FnMut(&'a Self::Item) -> B, - B: Ord - { - self.binary_search_by(|k| f(k).cmp(b)) + pub fn is_ascii(&self) -> bool { + self.iter().all(|b| b.is_ascii()) } + /// Checks that two slices are an ASCII case-insensitive match. + /// + /// Same as `to_ascii_lowercase(a) == to_ascii_lowercase(b)`, + /// but without allocating and copying temporaries. + #[stable(feature = "ascii_methods_on_intrinsics", since = "1.23.0")] #[inline] - fn sort_unstable(&mut self) - where Self::Item: Ord - { - sort::quicksort(self, |a, b| a.lt(b)); + pub fn eq_ignore_ascii_case(&self, other: &[u8]) -> bool { + self.len() == other.len() && + self.iter().zip(other).all(|(a, b)| { + a.eq_ignore_ascii_case(b) + }) } + /// Converts this slice to its ASCII upper case equivalent in-place. + /// + /// ASCII letters 'a' to 'z' are mapped to 'A' to 'Z', + /// but non-ASCII letters are unchanged. + /// + /// To return a new uppercased value without modifying the existing one, use + /// [`to_ascii_uppercase`]. + /// + /// [`to_ascii_uppercase`]: #method.to_ascii_uppercase + #[stable(feature = "ascii_methods_on_intrinsics", since = "1.23.0")] #[inline] - fn sort_unstable_by<F>(&mut self, mut compare: F) - where F: FnMut(&Self::Item, &Self::Item) -> Ordering - { - sort::quicksort(self, |a, b| compare(a, b) == Ordering::Less); + pub fn make_ascii_uppercase(&mut self) { + for byte in self { + byte.make_ascii_uppercase(); + } } + /// Converts this slice to its ASCII lower case equivalent in-place. + /// + /// ASCII letters 'A' to 'Z' are mapped to 'a' to 'z', + /// but non-ASCII letters are unchanged. + /// + /// To return a new lowercased value without modifying the existing one, use + /// [`to_ascii_lowercase`]. + /// + /// [`to_ascii_lowercase`]: #method.to_ascii_lowercase + #[stable(feature = "ascii_methods_on_intrinsics", since = "1.23.0")] #[inline] - fn sort_unstable_by_key<B, F>(&mut self, mut f: F) - where F: FnMut(&Self::Item) -> B, - B: Ord - { - sort::quicksort(self, |a, b| f(a).lt(&f(b))); + pub fn make_ascii_lowercase(&mut self) { + for byte in self { + byte.make_ascii_lowercase(); + } } + } #[stable(feature = "rust1", since = "1.0.0")] @@ -791,39 +1943,73 @@ fn slice_index_order_fail(index: usize, end: usize) -> ! { panic!("slice index starts at {} but ends at {}", index, end); } +#[inline(never)] +#[cold] +fn slice_index_overflow_fail() -> ! { + panic!("attempted to index slice up to maximum usize"); +} + +mod private_slice_index { + use super::ops; + #[stable(feature = "slice_get_slice", since = "1.28.0")] + pub trait Sealed {} + + #[stable(feature = "slice_get_slice", since = "1.28.0")] + impl Sealed for usize {} + #[stable(feature = "slice_get_slice", since = "1.28.0")] + impl Sealed for ops::Range<usize> {} + #[stable(feature = "slice_get_slice", since = "1.28.0")] + impl Sealed for ops::RangeTo<usize> {} + #[stable(feature = "slice_get_slice", since = "1.28.0")] + impl Sealed for ops::RangeFrom<usize> {} + #[stable(feature = "slice_get_slice", since = "1.28.0")] + impl Sealed for ops::RangeFull {} + #[stable(feature = "slice_get_slice", since = "1.28.0")] + impl Sealed for ops::RangeInclusive<usize> {} + #[stable(feature = "slice_get_slice", since = "1.28.0")] + impl Sealed for ops::RangeToInclusive<usize> {} +} + /// A helper trait used for indexing operations. -#[unstable(feature = "slice_get_slice", issue = "35729")] +#[stable(feature = "slice_get_slice", since = "1.28.0")] #[rustc_on_unimplemented = "slice indices are of type `usize` or ranges of `usize`"] -pub trait SliceIndex<T: ?Sized> { +pub trait SliceIndex<T: ?Sized>: private_slice_index::Sealed { /// The output type returned by methods. + #[stable(feature = "slice_get_slice", since = "1.28.0")] type Output: ?Sized; /// Returns a shared reference to the output at this location, if in /// bounds. + #[unstable(feature = "slice_index_methods", issue = "0")] fn get(self, slice: &T) -> Option<&Self::Output>; /// Returns a mutable reference to the output at this location, if in /// bounds. + #[unstable(feature = "slice_index_methods", issue = "0")] fn get_mut(self, slice: &mut T) -> Option<&mut Self::Output>; /// Returns a shared reference to the output at this location, without /// performing any bounds checking. + #[unstable(feature = "slice_index_methods", issue = "0")] unsafe fn get_unchecked(self, slice: &T) -> &Self::Output; /// Returns a mutable reference to the output at this location, without /// performing any bounds checking. + #[unstable(feature = "slice_index_methods", issue = "0")] unsafe fn get_unchecked_mut(self, slice: &mut T) -> &mut Self::Output; /// Returns a shared reference to the output at this location, panicking /// if out of bounds. + #[unstable(feature = "slice_index_methods", issue = "0")] fn index(self, slice: &T) -> &Self::Output; /// Returns a mutable reference to the output at this location, panicking /// if out of bounds. + #[unstable(feature = "slice_index_methods", issue = "0")] fn index_mut(self, slice: &mut T) -> &mut Self::Output; } -#[stable(feature = "slice-get-slice-impls", since = "1.15.0")] +#[stable(feature = "slice_get_slice_impls", since = "1.15.0")] impl<T> SliceIndex<[T]> for usize { type Output = T; @@ -872,7 +2058,7 @@ impl<T> SliceIndex<[T]> for usize { } } -#[stable(feature = "slice-get-slice-impls", since = "1.15.0")] +#[stable(feature = "slice_get_slice_impls", since = "1.15.0")] impl<T> SliceIndex<[T]> for ops::Range<usize> { type Output = [T]; @@ -933,7 +2119,7 @@ impl<T> SliceIndex<[T]> for ops::Range<usize> { } } -#[stable(feature = "slice-get-slice-impls", since = "1.15.0")] +#[stable(feature = "slice_get_slice_impls", since = "1.15.0")] impl<T> SliceIndex<[T]> for ops::RangeTo<usize> { type Output = [T]; @@ -968,7 +2154,7 @@ impl<T> SliceIndex<[T]> for ops::RangeTo<usize> { } } -#[stable(feature = "slice-get-slice-impls", since = "1.15.0")] +#[stable(feature = "slice_get_slice_impls", since = "1.15.0")] impl<T> SliceIndex<[T]> for ops::RangeFrom<usize> { type Output = [T]; @@ -1003,7 +2189,7 @@ impl<T> SliceIndex<[T]> for ops::RangeFrom<usize> { } } -#[stable(feature = "slice-get-slice-impls", since = "1.15.0")] +#[stable(feature = "slice_get_slice_impls", since = "1.15.0")] impl<T> SliceIndex<[T]> for ops::RangeFull { type Output = [T]; @@ -1039,48 +2225,46 @@ impl<T> SliceIndex<[T]> for ops::RangeFull { } -#[unstable(feature = "inclusive_range", reason = "recently added, follows RFC", issue = "28237")] +#[stable(feature = "inclusive_range", since = "1.26.0")] impl<T> SliceIndex<[T]> for ops::RangeInclusive<usize> { type Output = [T]; #[inline] fn get(self, slice: &[T]) -> Option<&[T]> { - if self.end == usize::max_value() { None } - else { (self.start..self.end + 1).get(slice) } + if *self.end() == usize::max_value() { None } + else { (*self.start()..self.end() + 1).get(slice) } } #[inline] fn get_mut(self, slice: &mut [T]) -> Option<&mut [T]> { - if self.end == usize::max_value() { None } - else { (self.start..self.end + 1).get_mut(slice) } + if *self.end() == usize::max_value() { None } + else { (*self.start()..self.end() + 1).get_mut(slice) } } #[inline] unsafe fn get_unchecked(self, slice: &[T]) -> &[T] { - (self.start..self.end + 1).get_unchecked(slice) + (*self.start()..self.end() + 1).get_unchecked(slice) } #[inline] unsafe fn get_unchecked_mut(self, slice: &mut [T]) -> &mut [T] { - (self.start..self.end + 1).get_unchecked_mut(slice) + (*self.start()..self.end() + 1).get_unchecked_mut(slice) } #[inline] fn index(self, slice: &[T]) -> &[T] { - assert!(self.end != usize::max_value(), - "attempted to index slice up to maximum usize"); - (self.start..self.end + 1).index(slice) + if *self.end() == usize::max_value() { slice_index_overflow_fail(); } + (*self.start()..self.end() + 1).index(slice) } #[inline] fn index_mut(self, slice: &mut [T]) -> &mut [T] { - assert!(self.end != usize::max_value(), - "attempted to index slice up to maximum usize"); - (self.start..self.end + 1).index_mut(slice) + if *self.end() == usize::max_value() { slice_index_overflow_fail(); } + (*self.start()..self.end() + 1).index_mut(slice) } } -#[unstable(feature = "inclusive_range", reason = "recently added, follows RFC", issue = "28237")] +#[stable(feature = "inclusive_range", since = "1.26.0")] impl<T> SliceIndex<[T]> for ops::RangeToInclusive<usize> { type Output = [T]; @@ -1155,14 +2339,88 @@ impl<'a, T> IntoIterator for &'a mut [T] { } } -#[inline] +// Macro helper functions +#[inline(always)] fn size_from_ptr<T>(_: *const T) -> usize { mem::size_of::<T>() } +// Inlining is_empty and len makes a huge performance difference +macro_rules! is_empty { + // The way we encode the length of a ZST iterator, this works both for ZST + // and non-ZST. + ($self: ident) => {$self.ptr == $self.end} +} +// To get rid of some bounds checks (see `position`), we compute the length in a somewhat +// unexpected way. (Tested by `codegen/slice-position-bounds-check`.) +macro_rules! len { + ($self: ident) => {{ + let start = $self.ptr; + let diff = ($self.end as usize).wrapping_sub(start as usize); + let size = size_from_ptr(start); + if size == 0 { + diff + } else { + // Using division instead of `offset_from` helps LLVM remove bounds checks + diff / size + } + }} +} + // The shared definition of the `Iter` and `IterMut` iterators macro_rules! iterator { - (struct $name:ident -> $ptr:ty, $elem:ty, $mkref:ident) => { + (struct $name:ident -> $ptr:ty, $elem:ty, $raw_mut:tt, $( $mut_:tt )*) => { + impl<'a, T> $name<'a, T> { + // Helper function for creating a slice from the iterator. + #[inline(always)] + fn make_slice(&self) -> &'a [T] { + unsafe { from_raw_parts(self.ptr, len!(self)) } + } + + // Helper function for moving the start of the iterator forwards by `offset` elements, + // returning the old start. + // Unsafe because the offset must be in-bounds or one-past-the-end. + #[inline(always)] + unsafe fn post_inc_start(&mut self, offset: isize) -> * $raw_mut T { + if mem::size_of::<T>() == 0 { + // This is *reducing* the length. `ptr` never changes with ZST. + self.end = (self.end as * $raw_mut u8).wrapping_offset(-offset) as * $raw_mut T; + self.ptr + } else { + let old = self.ptr; + self.ptr = self.ptr.offset(offset); + old + } + } + + // Helper function for moving the end of the iterator backwards by `offset` elements, + // returning the new end. + // Unsafe because the offset must be in-bounds or one-past-the-end. + #[inline(always)] + unsafe fn pre_dec_end(&mut self, offset: isize) -> * $raw_mut T { + if mem::size_of::<T>() == 0 { + self.end = (self.end as * $raw_mut u8).wrapping_offset(-offset) as * $raw_mut T; + self.ptr + } else { + self.end = self.end.offset(-offset); + self.end + } + } + } + + #[stable(feature = "rust1", since = "1.0.0")] + impl<'a, T> ExactSizeIterator for $name<'a, T> { + #[inline(always)] + fn len(&self) -> usize { + len!(self) + } + + #[inline(always)] + fn is_empty(&self) -> bool { + is_empty!(self) + } + } + #[stable(feature = "rust1", since = "1.0.0")] impl<'a, T> Iterator for $name<'a, T> { type Item = $elem; @@ -1171,33 +2429,48 @@ macro_rules! iterator { fn next(&mut self) -> Option<$elem> { // could be implemented with slices, but this avoids bounds checks unsafe { + assume(!self.ptr.is_null()); if mem::size_of::<T>() != 0 { - assume(!self.ptr.is_null()); assume(!self.end.is_null()); } - if self.ptr == self.end { + if is_empty!(self) { None } else { - Some($mkref!(self.ptr.post_inc())) + Some(& $( $mut_ )* *self.post_inc_start(1)) } } } #[inline] fn size_hint(&self) -> (usize, Option<usize>) { - let exact = ptrdistance(self.ptr, self.end); + let exact = len!(self); (exact, Some(exact)) } #[inline] fn count(self) -> usize { - self.len() + len!(self) } #[inline] fn nth(&mut self, n: usize) -> Option<$elem> { - // Call helper method. Can't put the definition here because mut versus const. - self.iter_nth(n) + if n >= len!(self) { + // This iterator is now empty. + if mem::size_of::<T>() == 0 { + // We have to do it this way as `ptr` may never be 0, but `end` + // could be (due to wrapping). + self.end = self.ptr; + } else { + self.ptr = self.end; + } + return None; + } + // We are in bounds. `offset` does the right thing even for ZSTs. + unsafe { + let elem = Some(& $( $mut_ )* *self.ptr.offset(n as isize)); + self.post_inc_start((n as isize).wrapping_add(1)); + elem + } } #[inline] @@ -1212,14 +2485,14 @@ macro_rules! iterator { // manual unrolling is needed when there are conditional exits from the loop let mut accum = init; unsafe { - while ptrdistance(self.ptr, self.end) >= 4 { - accum = f(accum, $mkref!(self.ptr.post_inc()))?; - accum = f(accum, $mkref!(self.ptr.post_inc()))?; - accum = f(accum, $mkref!(self.ptr.post_inc()))?; - accum = f(accum, $mkref!(self.ptr.post_inc()))?; + while len!(self) >= 4 { + accum = f(accum, & $( $mut_ )* *self.post_inc_start(1))?; + accum = f(accum, & $( $mut_ )* *self.post_inc_start(1))?; + accum = f(accum, & $( $mut_ )* *self.post_inc_start(1))?; + accum = f(accum, & $( $mut_ )* *self.post_inc_start(1))?; } - while self.ptr != self.end { - accum = f(accum, $mkref!(self.ptr.post_inc()))?; + while !is_empty!(self) { + accum = f(accum, & $( $mut_ )* *self.post_inc_start(1))?; } } Try::from_ok(accum) @@ -1244,8 +2517,8 @@ macro_rules! iterator { Self: Sized, P: FnMut(Self::Item) -> bool, { - // The addition might panic on overflow - let n = self.len(); + // The addition might panic on overflow. + let n = len!(self); self.try_fold(0, move |i, x| { if predicate(x) { Err(i) } else { Ok(i + 1) } @@ -1262,8 +2535,7 @@ macro_rules! iterator { Self: Sized + ExactSizeIterator + DoubleEndedIterator { // No need for an overflow check here, because `ExactSizeIterator` - // implies that the number of elements fits into a `usize`. - let n = self.len(); + let n = len!(self); self.try_rfold(n, move |i, x| { let i = i - 1; if predicate(x) { Err(i) } @@ -1282,14 +2554,14 @@ macro_rules! iterator { fn next_back(&mut self) -> Option<$elem> { // could be implemented with slices, but this avoids bounds checks unsafe { + assume(!self.ptr.is_null()); if mem::size_of::<T>() != 0 { - assume(!self.ptr.is_null()); assume(!self.end.is_null()); } - if self.end == self.ptr { + if is_empty!(self) { None } else { - Some($mkref!(self.end.pre_dec())) + Some(& $( $mut_ )* *self.pre_dec_end(1)) } } } @@ -1301,14 +2573,15 @@ macro_rules! iterator { // manual unrolling is needed when there are conditional exits from the loop let mut accum = init; unsafe { - while ptrdistance(self.ptr, self.end) >= 4 { - accum = f(accum, $mkref!(self.end.pre_dec()))?; - accum = f(accum, $mkref!(self.end.pre_dec()))?; - accum = f(accum, $mkref!(self.end.pre_dec()))?; - accum = f(accum, $mkref!(self.end.pre_dec()))?; + while len!(self) >= 4 { + accum = f(accum, & $( $mut_ )* *self.pre_dec_end(1))?; + accum = f(accum, & $( $mut_ )* *self.pre_dec_end(1))?; + accum = f(accum, & $( $mut_ )* *self.pre_dec_end(1))?; + accum = f(accum, & $( $mut_ )* *self.pre_dec_end(1))?; } - while self.ptr != self.end { - accum = f(accum, $mkref!(self.end.pre_dec()))?; + // inlining is_empty everywhere makes a huge performance difference + while !is_empty!(self) { + accum = f(accum, & $( $mut_ )* *self.pre_dec_end(1))?; } } Try::from_ok(accum) @@ -1327,35 +2600,13 @@ macro_rules! iterator { accum } } - } -} -macro_rules! make_slice { - ($start: expr, $end: expr) => {{ - let start = $start; - let diff = ($end as usize).wrapping_sub(start as usize); - if size_from_ptr(start) == 0 { - // use a non-null pointer value - unsafe { from_raw_parts(1 as *const _, diff) } - } else { - let len = diff / size_from_ptr(start); - unsafe { from_raw_parts(start, len) } - } - }} -} + #[stable(feature = "fused", since = "1.26.0")] + impl<'a, T> FusedIterator for $name<'a, T> {} -macro_rules! make_mut_slice { - ($start: expr, $end: expr) => {{ - let start = $start; - let diff = ($end as usize).wrapping_sub(start as usize); - if size_from_ptr(start) == 0 { - // use a non-null pointer value - unsafe { from_raw_parts_mut(1 as *mut _, diff) } - } else { - let len = diff / size_from_ptr(start); - unsafe { from_raw_parts_mut(start, len) } - } - }} + #[unstable(feature = "trusted_len", issue = "37572")] + unsafe impl<'a, T> TrustedLen for $name<'a, T> {} + } } /// Immutable slice iterator @@ -1381,7 +2632,9 @@ macro_rules! make_mut_slice { #[stable(feature = "rust1", since = "1.0.0")] pub struct Iter<'a, T: 'a> { ptr: *const T, - end: *const T, + end: *const T, // If T is a ZST, this is actually ptr+len. This encoding is picked so that + // ptr == end is a quick test for the Iterator being empty, that works + // for both ZST and non-ZST. _marker: marker::PhantomData<&'a T>, } @@ -1426,38 +2679,11 @@ impl<'a, T> Iter<'a, T> { /// ``` #[stable(feature = "iter_to_slice", since = "1.4.0")] pub fn as_slice(&self) -> &'a [T] { - make_slice!(self.ptr, self.end) - } - - // Helper function for Iter::nth - fn iter_nth(&mut self, n: usize) -> Option<&'a T> { - match self.as_slice().get(n) { - Some(elem_ref) => unsafe { - self.ptr = slice_offset!(self.ptr, (n as isize).wrapping_add(1)); - Some(elem_ref) - }, - None => { - self.ptr = self.end; - None - } - } + self.make_slice() } } -iterator!{struct Iter -> *const T, &'a T, make_ref} - -#[stable(feature = "rust1", since = "1.0.0")] -impl<'a, T> ExactSizeIterator for Iter<'a, T> { - fn is_empty(&self) -> bool { - self.ptr == self.end - } -} - -#[unstable(feature = "fused", issue = "35602")] -impl<'a, T> FusedIterator for Iter<'a, T> {} - -#[unstable(feature = "trusted_len", issue = "37572")] -unsafe impl<'a, T> TrustedLen for Iter<'a, T> {} +iterator!{struct Iter -> *const T, &'a T, const, /* no mut */} #[stable(feature = "rust1", since = "1.0.0")] impl<'a, T> Clone for Iter<'a, T> { @@ -1498,7 +2724,9 @@ impl<'a, T> AsRef<[T]> for Iter<'a, T> { #[stable(feature = "rust1", since = "1.0.0")] pub struct IterMut<'a, T: 'a> { ptr: *mut T, - end: *mut T, + end: *mut T, // If T is a ZST, this is actually ptr+len. This encoding is picked so that + // ptr == end is a quick test for the Iterator being empty, that works + // for both ZST and non-ZST. _marker: marker::PhantomData<&'a mut T>, } @@ -1506,7 +2734,7 @@ pub struct IterMut<'a, T: 'a> { impl<'a, T: 'a + fmt::Debug> fmt::Debug for IterMut<'a, T> { fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result { f.debug_tuple("IterMut") - .field(&make_slice!(self.ptr, self.end)) + .field(&self.make_slice()) .finish() } } @@ -1520,9 +2748,7 @@ impl<'a, T> IterMut<'a, T> { /// View the underlying data as a subslice of the original data. /// /// To avoid creating `&mut` references that alias, this is forced - /// to consume the iterator. Consider using the `Slice` and - /// `SliceMut` implementations for obtaining slices with more - /// restricted lifetimes that do not consume the iterator. + /// to consume the iterator. /// /// # Examples /// @@ -1554,83 +2780,11 @@ impl<'a, T> IterMut<'a, T> { /// ``` #[stable(feature = "iter_to_slice", since = "1.4.0")] pub fn into_slice(self) -> &'a mut [T] { - make_mut_slice!(self.ptr, self.end) - } - - // Helper function for IterMut::nth - fn iter_nth(&mut self, n: usize) -> Option<&'a mut T> { - match make_mut_slice!(self.ptr, self.end).get_mut(n) { - Some(elem_ref) => unsafe { - self.ptr = slice_offset!(self.ptr, (n as isize).wrapping_add(1)); - Some(elem_ref) - }, - None => { - self.ptr = self.end; - None - } - } + unsafe { from_raw_parts_mut(self.ptr, len!(self)) } } } -iterator!{struct IterMut -> *mut T, &'a mut T, make_ref_mut} - -#[stable(feature = "rust1", since = "1.0.0")] -impl<'a, T> ExactSizeIterator for IterMut<'a, T> { - fn is_empty(&self) -> bool { - self.ptr == self.end - } -} - -#[unstable(feature = "fused", issue = "35602")] -impl<'a, T> FusedIterator for IterMut<'a, T> {} - -#[unstable(feature = "trusted_len", issue = "37572")] -unsafe impl<'a, T> TrustedLen for IterMut<'a, T> {} - - -// Return the number of elements of `T` from `start` to `end`. -// Return the arithmetic difference if `T` is zero size. -#[inline(always)] -fn ptrdistance<T>(start: *const T, end: *const T) -> usize { - match start.offset_to(end) { - Some(x) => x as usize, - None => (end as usize).wrapping_sub(start as usize), - } -} - -// Extension methods for raw pointers, used by the iterators -trait PointerExt : Copy { - unsafe fn slice_offset(self, i: isize) -> Self; - - /// Increments `self` by 1, but returns the old value. - #[inline(always)] - unsafe fn post_inc(&mut self) -> Self { - let current = *self; - *self = self.slice_offset(1); - current - } - - /// Decrements `self` by 1, and returns the new value. - #[inline(always)] - unsafe fn pre_dec(&mut self) -> Self { - *self = self.slice_offset(-1); - *self - } -} - -impl<T> PointerExt for *const T { - #[inline(always)] - unsafe fn slice_offset(self, i: isize) -> Self { - slice_offset!(self, i) - } -} - -impl<T> PointerExt for *mut T { - #[inline(always)] - unsafe fn slice_offset(self, i: isize) -> Self { - slice_offset!(self, i) - } -} +iterator!{struct IterMut -> *mut T, &'a mut T, mut, mut} /// An internal abstraction over the splitting iterators, so that /// splitn, splitn_mut etc can be implemented once. @@ -1729,7 +2883,7 @@ impl<'a, T, P> SplitIter for Split<'a, T, P> where P: FnMut(&T) -> bool { } } -#[unstable(feature = "fused", issue = "35602")] +#[stable(feature = "fused", since = "1.26.0")] impl<'a, T, P> FusedIterator for Split<'a, T, P> where P: FnMut(&T) -> bool {} /// An iterator over the subslices of the vector which are separated @@ -1827,7 +2981,7 @@ impl<'a, T, P> DoubleEndedIterator for SplitMut<'a, T, P> where } } -#[unstable(feature = "fused", issue = "35602")] +#[stable(feature = "fused", since = "1.26.0")] impl<'a, T, P> FusedIterator for SplitMut<'a, T, P> where P: FnMut(&T) -> bool {} /// An iterator over subslices separated by elements that match a predicate @@ -1837,13 +2991,13 @@ impl<'a, T, P> FusedIterator for SplitMut<'a, T, P> where P: FnMut(&T) -> bool { /// /// [`rsplit`]: ../../std/primitive.slice.html#method.rsplit /// [slices]: ../../std/primitive.slice.html -#[unstable(feature = "slice_rsplit", issue = "41020")] +#[stable(feature = "slice_rsplit", since = "1.27.0")] #[derive(Clone)] // Is this correct, or does it incorrectly require `T: Clone`? pub struct RSplit<'a, T:'a, P> where P: FnMut(&T) -> bool { inner: Split<'a, T, P> } -#[unstable(feature = "slice_rsplit", issue = "41020")] +#[stable(feature = "slice_rsplit", since = "1.27.0")] impl<'a, T: 'a + fmt::Debug, P> fmt::Debug for RSplit<'a, T, P> where P: FnMut(&T) -> bool { fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result { f.debug_struct("RSplit") @@ -1853,7 +3007,7 @@ impl<'a, T: 'a + fmt::Debug, P> fmt::Debug for RSplit<'a, T, P> where P: FnMut(& } } -#[unstable(feature = "slice_rsplit", issue = "41020")] +#[stable(feature = "slice_rsplit", since = "1.27.0")] impl<'a, T, P> Iterator for RSplit<'a, T, P> where P: FnMut(&T) -> bool { type Item = &'a [T]; @@ -1868,7 +3022,7 @@ impl<'a, T, P> Iterator for RSplit<'a, T, P> where P: FnMut(&T) -> bool { } } -#[unstable(feature = "slice_rsplit", issue = "41020")] +#[stable(feature = "slice_rsplit", since = "1.27.0")] impl<'a, T, P> DoubleEndedIterator for RSplit<'a, T, P> where P: FnMut(&T) -> bool { #[inline] fn next_back(&mut self) -> Option<&'a [T]> { @@ -1876,7 +3030,7 @@ impl<'a, T, P> DoubleEndedIterator for RSplit<'a, T, P> where P: FnMut(&T) -> bo } } -#[unstable(feature = "slice_rsplit", issue = "41020")] +#[stable(feature = "slice_rsplit", since = "1.27.0")] impl<'a, T, P> SplitIter for RSplit<'a, T, P> where P: FnMut(&T) -> bool { #[inline] fn finish(&mut self) -> Option<&'a [T]> { @@ -1884,8 +3038,7 @@ impl<'a, T, P> SplitIter for RSplit<'a, T, P> where P: FnMut(&T) -> bool { } } -//#[unstable(feature = "fused", issue = "35602")] -#[unstable(feature = "slice_rsplit", issue = "41020")] +#[stable(feature = "slice_rsplit", since = "1.27.0")] impl<'a, T, P> FusedIterator for RSplit<'a, T, P> where P: FnMut(&T) -> bool {} /// An iterator over the subslices of the vector which are separated @@ -1895,12 +3048,12 @@ impl<'a, T, P> FusedIterator for RSplit<'a, T, P> where P: FnMut(&T) -> bool {} /// /// [`rsplit_mut`]: ../../std/primitive.slice.html#method.rsplit_mut /// [slices]: ../../std/primitive.slice.html -#[unstable(feature = "slice_rsplit", issue = "41020")] +#[stable(feature = "slice_rsplit", since = "1.27.0")] pub struct RSplitMut<'a, T:'a, P> where P: FnMut(&T) -> bool { inner: SplitMut<'a, T, P> } -#[unstable(feature = "slice_rsplit", issue = "41020")] +#[stable(feature = "slice_rsplit", since = "1.27.0")] impl<'a, T: 'a + fmt::Debug, P> fmt::Debug for RSplitMut<'a, T, P> where P: FnMut(&T) -> bool { fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result { f.debug_struct("RSplitMut") @@ -1910,7 +3063,7 @@ impl<'a, T: 'a + fmt::Debug, P> fmt::Debug for RSplitMut<'a, T, P> where P: FnMu } } -#[unstable(feature = "slice_rsplit", issue = "41020")] +#[stable(feature = "slice_rsplit", since = "1.27.0")] impl<'a, T, P> SplitIter for RSplitMut<'a, T, P> where P: FnMut(&T) -> bool { #[inline] fn finish(&mut self) -> Option<&'a mut [T]> { @@ -1918,7 +3071,7 @@ impl<'a, T, P> SplitIter for RSplitMut<'a, T, P> where P: FnMut(&T) -> bool { } } -#[unstable(feature = "slice_rsplit", issue = "41020")] +#[stable(feature = "slice_rsplit", since = "1.27.0")] impl<'a, T, P> Iterator for RSplitMut<'a, T, P> where P: FnMut(&T) -> bool { type Item = &'a mut [T]; @@ -1933,7 +3086,7 @@ impl<'a, T, P> Iterator for RSplitMut<'a, T, P> where P: FnMut(&T) -> bool { } } -#[unstable(feature = "slice_rsplit", issue = "41020")] +#[stable(feature = "slice_rsplit", since = "1.27.0")] impl<'a, T, P> DoubleEndedIterator for RSplitMut<'a, T, P> where P: FnMut(&T) -> bool, { @@ -1943,8 +3096,7 @@ impl<'a, T, P> DoubleEndedIterator for RSplitMut<'a, T, P> where } } -//#[unstable(feature = "fused", issue = "35602")] -#[unstable(feature = "slice_rsplit", issue = "41020")] +#[stable(feature = "slice_rsplit", since = "1.27.0")] impl<'a, T, P> FusedIterator for RSplitMut<'a, T, P> where P: FnMut(&T) -> bool {} /// An private iterator over subslices separated by elements that @@ -2080,7 +3232,7 @@ macro_rules! forward_iterator { } } - #[unstable(feature = "fused", issue = "35602")] + #[stable(feature = "fused", since = "1.26.0")] impl<'a, $elem, P> FusedIterator for $name<'a, $elem, P> where P: FnMut(&T) -> bool {} } @@ -2186,7 +3338,10 @@ impl<'a, T> DoubleEndedIterator for Windows<'a, T> { #[stable(feature = "rust1", since = "1.0.0")] impl<'a, T> ExactSizeIterator for Windows<'a, T> {} -#[unstable(feature = "fused", issue = "35602")] +#[unstable(feature = "trusted_len", issue = "37572")] +unsafe impl<'a, T> TrustedLen for Windows<'a, T> {} + +#[stable(feature = "fused", since = "1.26.0")] impl<'a, T> FusedIterator for Windows<'a, T> {} #[doc(hidden)] @@ -2305,7 +3460,10 @@ impl<'a, T> DoubleEndedIterator for Chunks<'a, T> { #[stable(feature = "rust1", since = "1.0.0")] impl<'a, T> ExactSizeIterator for Chunks<'a, T> {} -#[unstable(feature = "fused", issue = "35602")] +#[unstable(feature = "trusted_len", issue = "37572")] +unsafe impl<'a, T> TrustedLen for Chunks<'a, T> {} + +#[stable(feature = "fused", since = "1.26.0")] impl<'a, T> FusedIterator for Chunks<'a, T> {} #[doc(hidden)] @@ -2421,7 +3579,10 @@ impl<'a, T> DoubleEndedIterator for ChunksMut<'a, T> { #[stable(feature = "rust1", since = "1.0.0")] impl<'a, T> ExactSizeIterator for ChunksMut<'a, T> {} -#[unstable(feature = "fused", issue = "35602")] +#[unstable(feature = "trusted_len", issue = "37572")] +unsafe impl<'a, T> TrustedLen for ChunksMut<'a, T> {} + +#[stable(feature = "fused", since = "1.26.0")] impl<'a, T> FusedIterator for ChunksMut<'a, T> {} #[doc(hidden)] @@ -2441,25 +3602,39 @@ unsafe impl<'a, T> TrustedRandomAccess for ChunksMut<'a, T> { /// time). /// /// When the slice len is not evenly divided by the chunk size, the last -/// up to `chunk_size-1` elements will be omitted. +/// up to `chunk_size-1` elements will be omitted but can be retrieved from +/// the [`remainder`] function from the iterator. /// /// This struct is created by the [`exact_chunks`] method on [slices]. /// /// [`exact_chunks`]: ../../std/primitive.slice.html#method.exact_chunks +/// [`remainder`]: ../../std/slice/struct.ExactChunks.html#method.remainder /// [slices]: ../../std/primitive.slice.html #[derive(Debug)] #[unstable(feature = "exact_chunks", issue = "47115")] pub struct ExactChunks<'a, T:'a> { v: &'a [T], + rem: &'a [T], chunk_size: usize } +#[unstable(feature = "exact_chunks", issue = "47115")] +impl<'a, T> ExactChunks<'a, T> { + /// Return the remainder of the original slice that is not going to be + /// returned by the iterator. The returned slice has at most `chunk_size-1` + /// elements. + pub fn remainder(&self) -> &'a [T] { + self.rem + } +} + // FIXME(#26925) Remove in favor of `#[derive(Clone)]` #[unstable(feature = "exact_chunks", issue = "47115")] impl<'a, T> Clone for ExactChunks<'a, T> { fn clone(&self) -> ExactChunks<'a, T> { ExactChunks { v: self.v, + rem: self.rem, chunk_size: self.chunk_size, } } @@ -2531,7 +3706,10 @@ impl<'a, T> ExactSizeIterator for ExactChunks<'a, T> { } } -#[unstable(feature = "fused", issue = "35602")] +#[unstable(feature = "trusted_len", issue = "37572")] +unsafe impl<'a, T> TrustedLen for ExactChunks<'a, T> {} + +#[unstable(feature = "exact_chunks", issue = "47115")] impl<'a, T> FusedIterator for ExactChunks<'a, T> {} #[doc(hidden)] @@ -2544,21 +3722,36 @@ unsafe impl<'a, T> TrustedRandomAccess for ExactChunks<'a, T> { } /// An iterator over a slice in (non-overlapping) mutable chunks (`chunk_size` -/// elements at a time). When the slice len is not evenly divided by the chunk -/// size, the last up to `chunk_size-1` elements will be omitted. +/// elements at a time). +/// +/// When the slice len is not evenly divided by the chunk size, the last up to +/// `chunk_size-1` elements will be omitted but can be retrieved from the +/// [`into_remainder`] function from the iterator. /// /// This struct is created by the [`exact_chunks_mut`] method on [slices]. /// /// [`exact_chunks_mut`]: ../../std/primitive.slice.html#method.exact_chunks_mut +/// [`into_remainder`]: ../../std/slice/struct.ExactChunksMut.html#method.into_remainder /// [slices]: ../../std/primitive.slice.html #[derive(Debug)] #[unstable(feature = "exact_chunks", issue = "47115")] pub struct ExactChunksMut<'a, T:'a> { v: &'a mut [T], + rem: &'a mut [T], chunk_size: usize } #[unstable(feature = "exact_chunks", issue = "47115")] +impl<'a, T> ExactChunksMut<'a, T> { + /// Return the remainder of the original slice that is not going to be + /// returned by the iterator. The returned slice has at most `chunk_size-1` + /// elements. + pub fn into_remainder(self) -> &'a mut [T] { + self.rem + } +} + +#[unstable(feature = "exact_chunks", issue = "47115")] impl<'a, T> Iterator for ExactChunksMut<'a, T> { type Item = &'a mut [T]; @@ -2628,7 +3821,10 @@ impl<'a, T> ExactSizeIterator for ExactChunksMut<'a, T> { } } -#[unstable(feature = "fused", issue = "35602")] +#[unstable(feature = "trusted_len", issue = "37572")] +unsafe impl<'a, T> TrustedLen for ExactChunksMut<'a, T> {} + +#[unstable(feature = "exact_chunks", issue = "47115")] impl<'a, T> FusedIterator for ExactChunksMut<'a, T> {} #[doc(hidden)] @@ -2654,10 +3850,11 @@ unsafe impl<'a, T> TrustedRandomAccess for ExactChunksMut<'a, T> { /// valid for `len` elements, nor whether the lifetime inferred is a suitable /// lifetime for the returned slice. /// -/// `p` must be non-null, even for zero-length slices, because non-zero bits -/// are required to distinguish between a zero-length slice within `Some()` -/// from `None`. `p` can be a bogus non-dereferencable pointer, such as `0x1`, -/// for zero-length slices, though. +/// `data` must be non-null and aligned, even for zero-length slices. One +/// reason for this is that enum layout optimizations may rely on references +/// (including slices of any length) being aligned and non-null to distinguish +/// them from other data. You can obtain a pointer that is usable as `data` +/// for zero-length slices using [`NonNull::dangling()`]. /// /// # Caveat /// @@ -2672,17 +3869,19 @@ unsafe impl<'a, T> TrustedRandomAccess for ExactChunksMut<'a, T> { /// ``` /// use std::slice; /// -/// // manifest a slice out of thin air! -/// let ptr = 0x1234 as *const usize; -/// let amt = 10; -/// unsafe { -/// let slice = slice::from_raw_parts(ptr, amt); -/// } +/// // manifest a slice for a single element +/// let x = 42; +/// let ptr = &x as *const _; +/// let slice = unsafe { slice::from_raw_parts(ptr, 1) }; +/// assert_eq!(slice[0], 42); /// ``` +/// +/// [`NonNull::dangling()`]: ../../std/ptr/struct.NonNull.html#method.dangling #[inline] #[stable(feature = "rust1", since = "1.0.0")] -pub unsafe fn from_raw_parts<'a, T>(p: *const T, len: usize) -> &'a [T] { - mem::transmute(Repr { data: p, len: len }) +pub unsafe fn from_raw_parts<'a, T>(data: *const T, len: usize) -> &'a [T] { + debug_assert!(data as usize % mem::align_of::<T>() == 0, "attempt to create unaligned slice"); + Repr { raw: FatPtr { data, len } }.rust } /// Performs the same functionality as `from_raw_parts`, except that a mutable @@ -2690,16 +3889,17 @@ pub unsafe fn from_raw_parts<'a, T>(p: *const T, len: usize) -> &'a [T] { /// /// This function is unsafe for the same reasons as `from_raw_parts`, as well /// as not being able to provide a non-aliasing guarantee of the returned -/// mutable slice. `p` must be non-null even for zero-length slices as with -/// `from_raw_parts`. +/// mutable slice. `data` must be non-null and aligned even for zero-length +/// slices as with `from_raw_parts`. #[inline] #[stable(feature = "rust1", since = "1.0.0")] -pub unsafe fn from_raw_parts_mut<'a, T>(p: *mut T, len: usize) -> &'a mut [T] { - mem::transmute(Repr { data: p, len: len }) +pub unsafe fn from_raw_parts_mut<'a, T>(data: *mut T, len: usize) -> &'a mut [T] { + debug_assert!(data as usize % mem::align_of::<T>() == 0, "attempt to create unaligned slice"); + Repr { raw: FatPtr { data, len} }.rust_mut } /// Converts a reference to T into a slice of length 1 (without copying). -#[unstable(feature = "from_ref", issue = "45703")] +#[stable(feature = "from_ref", since = "1.28.0")] pub fn from_ref<T>(s: &T) -> &[T] { unsafe { from_raw_parts(s, 1) @@ -2707,8 +3907,8 @@ pub fn from_ref<T>(s: &T) -> &[T] { } /// Converts a reference to T into a slice of length 1 (without copying). -#[unstable(feature = "from_ref", issue = "45703")] -pub fn from_ref_mut<T>(s: &mut T) -> &mut [T] { +#[stable(feature = "from_ref", since = "1.28.0")] +pub fn from_mut<T>(s: &mut T) -> &mut [T] { unsafe { from_raw_parts_mut(s, 1) } diff --git a/src/libcore/slice/rotate.rs b/src/libcore/slice/rotate.rs index e4a4e33c172..28ef53ccb5c 100644 --- a/src/libcore/slice/rotate.rs +++ b/src/libcore/slice/rotate.rs @@ -48,7 +48,6 @@ impl<T> RawArray<T> { /// # Safety /// /// The specified range must be valid for reading and writing. -/// The type `T` must have non-zero size. /// /// # Algorithm /// @@ -73,6 +72,7 @@ pub unsafe fn ptr_rotate<T>(mut left: usize, mid: *mut T, mut right: usize) { loop { let delta = cmp::min(left, right); if delta <= RawArray::<T>::cap() { + // We will always hit this immediately for ZST. break; } diff --git a/src/libcore/str/lossy.rs b/src/libcore/str/lossy.rs new file mode 100644 index 00000000000..186d6adbc91 --- /dev/null +++ b/src/libcore/str/lossy.rs @@ -0,0 +1,212 @@ +// Copyright 2012-2017 The Rust Project Developers. See the COPYRIGHT +// file at the top-level directory of this distribution and at +// http://rust-lang.org/COPYRIGHT. +// +// Licensed under the Apache License, Version 2.0 <LICENSE-APACHE or +// http://www.apache.org/licenses/LICENSE-2.0> or the MIT license +// <LICENSE-MIT or http://opensource.org/licenses/MIT>, at your +// option. This file may not be copied, modified, or distributed +// except according to those terms. + +use char; +use str as core_str; +use fmt; +use fmt::Write; +use mem; + +/// Lossy UTF-8 string. +#[unstable(feature = "str_internals", issue = "0")] +pub struct Utf8Lossy { + bytes: [u8] +} + +impl Utf8Lossy { + pub fn from_str(s: &str) -> &Utf8Lossy { + Utf8Lossy::from_bytes(s.as_bytes()) + } + + pub fn from_bytes(bytes: &[u8]) -> &Utf8Lossy { + unsafe { mem::transmute(bytes) } + } + + pub fn chunks(&self) -> Utf8LossyChunksIter { + Utf8LossyChunksIter { source: &self.bytes } + } +} + + +/// Iterator over lossy UTF-8 string +#[unstable(feature = "str_internals", issue = "0")] +#[allow(missing_debug_implementations)] +pub struct Utf8LossyChunksIter<'a> { + source: &'a [u8], +} + +#[unstable(feature = "str_internals", issue = "0")] +#[derive(PartialEq, Eq, Debug)] +pub struct Utf8LossyChunk<'a> { + /// Sequence of valid chars. + /// Can be empty between broken UTF-8 chars. + pub valid: &'a str, + /// Single broken char, empty if none. + /// Empty iff iterator item is last. + pub broken: &'a [u8], +} + +impl<'a> Iterator for Utf8LossyChunksIter<'a> { + type Item = Utf8LossyChunk<'a>; + + fn next(&mut self) -> Option<Utf8LossyChunk<'a>> { + if self.source.len() == 0 { + return None; + } + + const TAG_CONT_U8: u8 = 128; + fn unsafe_get(xs: &[u8], i: usize) -> u8 { + unsafe { *xs.get_unchecked(i) } + } + fn safe_get(xs: &[u8], i: usize) -> u8 { + if i >= xs.len() { 0 } else { unsafe_get(xs, i) } + } + + let mut i = 0; + while i < self.source.len() { + let i_ = i; + + let byte = unsafe_get(self.source, i); + i += 1; + + if byte < 128 { + + } else { + let w = core_str::utf8_char_width(byte); + + macro_rules! error { () => ({ + unsafe { + let r = Utf8LossyChunk { + valid: core_str::from_utf8_unchecked(&self.source[0..i_]), + broken: &self.source[i_..i], + }; + self.source = &self.source[i..]; + return Some(r); + } + })} + + match w { + 2 => { + if safe_get(self.source, i) & 192 != TAG_CONT_U8 { + error!(); + } + i += 1; + } + 3 => { + match (byte, safe_get(self.source, i)) { + (0xE0, 0xA0 ..= 0xBF) => (), + (0xE1 ..= 0xEC, 0x80 ..= 0xBF) => (), + (0xED, 0x80 ..= 0x9F) => (), + (0xEE ..= 0xEF, 0x80 ..= 0xBF) => (), + _ => { + error!(); + } + } + i += 1; + if safe_get(self.source, i) & 192 != TAG_CONT_U8 { + error!(); + } + i += 1; + } + 4 => { + match (byte, safe_get(self.source, i)) { + (0xF0, 0x90 ..= 0xBF) => (), + (0xF1 ..= 0xF3, 0x80 ..= 0xBF) => (), + (0xF4, 0x80 ..= 0x8F) => (), + _ => { + error!(); + } + } + i += 1; + if safe_get(self.source, i) & 192 != TAG_CONT_U8 { + error!(); + } + i += 1; + if safe_get(self.source, i) & 192 != TAG_CONT_U8 { + error!(); + } + i += 1; + } + _ => { + error!(); + } + } + } + } + + let r = Utf8LossyChunk { + valid: unsafe { core_str::from_utf8_unchecked(self.source) }, + broken: &[], + }; + self.source = &[]; + Some(r) + } +} + + +impl fmt::Display for Utf8Lossy { + fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result { + // If we're the empty string then our iterator won't actually yield + // anything, so perform the formatting manually + if self.bytes.len() == 0 { + return "".fmt(f) + } + + for Utf8LossyChunk { valid, broken } in self.chunks() { + // If we successfully decoded the whole chunk as a valid string then + // we can return a direct formatting of the string which will also + // respect various formatting flags if possible. + if valid.len() == self.bytes.len() { + assert!(broken.is_empty()); + return valid.fmt(f) + } + + f.write_str(valid)?; + if !broken.is_empty() { + f.write_char(char::REPLACEMENT_CHARACTER)?; + } + } + Ok(()) + } +} + +impl fmt::Debug for Utf8Lossy { + fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result { + f.write_char('"')?; + + for Utf8LossyChunk { valid, broken } in self.chunks() { + + // Valid part. + // Here we partially parse UTF-8 again which is suboptimal. + { + let mut from = 0; + for (i, c) in valid.char_indices() { + let esc = c.escape_debug(); + // If char needs escaping, flush backlog so far and write, else skip + if esc.len() != 1 { + f.write_str(&valid[from..i])?; + for c in esc { + f.write_char(c)?; + } + from = i + c.len_utf8(); + } + } + f.write_str(&valid[from..])?; + } + + // Broken parts of string as hex escape. + for &b in broken { + write!(f, "\\x{:02x}", b)?; + } + } + + f.write_char('"') + } +} diff --git a/src/libcore/str/mod.rs b/src/libcore/str/mod.rs index 765b369e4b2..810d19df0c5 100644 --- a/src/libcore/str/mod.rs +++ b/src/libcore/str/mod.rs @@ -19,13 +19,17 @@ use self::pattern::{Searcher, ReverseSearcher, DoubleEndedSearcher}; use char; use fmt; -use iter::{Map, Cloned, FusedIterator, TrustedLen}; +use iter::{Map, Cloned, FusedIterator, TrustedLen, Filter}; use iter_private::TrustedRandomAccess; -use slice::{self, SliceIndex}; +use slice::{self, SliceIndex, Split as SliceSplit}; use mem; pub mod pattern; +#[unstable(feature = "str_internals", issue = "0")] +#[allow(missing_docs)] +pub mod lossy; + /// A trait to abstract the idea of creating a new instance of a type from a /// string. /// @@ -165,6 +169,37 @@ Section: Creating a string /// /// [`String`]: ../../std/string/struct.String.html#method.from_utf8 /// [`&str`]: ../../std/str/fn.from_utf8.html +/// +/// # Examples +/// +/// This error type’s methods can be used to create functionality +/// similar to `String::from_utf8_lossy` without allocating heap memory: +/// +/// ``` +/// fn from_utf8_lossy<F>(mut input: &[u8], mut push: F) where F: FnMut(&str) { +/// loop { +/// match ::std::str::from_utf8(input) { +/// Ok(valid) => { +/// push(valid); +/// break +/// } +/// Err(error) => { +/// let (valid, after_valid) = input.split_at(error.valid_up_to()); +/// unsafe { +/// push(::std::str::from_utf8_unchecked(valid)) +/// } +/// push("\u{FFFD}"); +/// +/// if let Some(invalid_sequence_length) = error.error_len() { +/// input = &after_valid[invalid_sequence_length..] +/// } else { +/// break +/// } +/// } +/// } +/// } +/// } +/// ``` #[derive(Copy, Eq, PartialEq, Clone, Debug)] #[stable(feature = "rust1", since = "1.0.0")] pub struct Utf8Error { @@ -209,7 +244,10 @@ impl Utf8Error { /// The length provided is that of the invalid byte sequence /// that starts at the index given by `valid_up_to()`. /// Decoding should resume after that sequence - /// (after inserting a U+FFFD REPLACEMENT CHARACTER) in case of lossy decoding. + /// (after inserting a [`U+FFFD REPLACEMENT CHARACTER`][U+FFFD]) in case of + /// lossy decoding. + /// + /// [U+FFFD]: ../../std/char/constant.REPLACEMENT_CHARACTER.html #[stable(feature = "utf8_error_error_len", since = "1.20.0")] pub fn error_len(&self) -> Option<usize> { self.error_len.map(|len| len as usize) @@ -341,37 +379,6 @@ pub fn from_utf8_mut(v: &mut [u8]) -> Result<&mut str, Utf8Error> { Ok(unsafe { from_utf8_unchecked_mut(v) }) } -/// Forms a str from a pointer and a length. -/// -/// The `len` argument is the number of bytes in the string. -/// -/// # Safety -/// -/// This function is unsafe as there is no guarantee that the given pointer is -/// valid for `len` bytes, nor whether the lifetime inferred is a suitable -/// lifetime for the returned str. -/// -/// The data must be valid UTF-8 -/// -/// `p` must be non-null, even for zero-length strs, because non-zero bits -/// are required to distinguish between a zero-length str within `Some()` -/// from `None`. `p` can be a bogus non-dereferencable pointer, such as `0x1`, -/// for zero-length strs, though. -/// -/// # Caveat -/// -/// The lifetime for the returned str is inferred from its usage. To -/// prevent accidental misuse, it's suggested to tie the lifetime to whichever -/// source lifetime is safe in the context, such as by providing a helper -/// function taking the lifetime of a host value for the str, or by explicit -/// annotation. -/// Performs the same functionality as `from_raw_parts`, except that a mutable -/// str is returned. -/// -unsafe fn from_raw_parts_mut<'a>(p: *mut u8, len: usize) -> &'a mut str { - from_utf8_unchecked_mut(slice::from_raw_parts_mut(p, len)) -} - /// Converts a slice of bytes to a string slice without checking /// that the string contains valid UTF-8. /// @@ -609,7 +616,7 @@ impl<'a> DoubleEndedIterator for Chars<'a> { } } -#[unstable(feature = "fused", issue = "35602")] +#[stable(feature = "fused", since = "1.26.0")] impl<'a> FusedIterator for Chars<'a> {} impl<'a> Chars<'a> { @@ -692,17 +699,14 @@ impl<'a> Iterator for CharIndices<'a> { impl<'a> DoubleEndedIterator for CharIndices<'a> { #[inline] fn next_back(&mut self) -> Option<(usize, char)> { - match self.iter.next_back() { - None => None, - Some(ch) => { - let index = self.front_offset + self.iter.iter.len(); - Some((index, ch)) - } - } + self.iter.next_back().map(|ch| { + let index = self.front_offset + self.iter.iter.len(); + (index, ch) + }) } } -#[unstable(feature = "fused", issue = "35602")] +#[stable(feature = "fused", since = "1.26.0")] impl<'a> FusedIterator for CharIndices<'a> {} impl<'a> CharIndices<'a> { @@ -817,7 +821,7 @@ impl<'a> ExactSizeIterator for Bytes<'a> { } } -#[unstable(feature = "fused", issue = "35602")] +#[stable(feature = "fused", since = "1.26.0")] impl<'a> FusedIterator for Bytes<'a> {} #[unstable(feature = "trusted_len", issue = "37572")] @@ -977,10 +981,10 @@ macro_rules! generate_pattern_iterators { } } - #[unstable(feature = "fused", issue = "35602")] + #[stable(feature = "fused", since = "1.26.0")] impl<'a, P: Pattern<'a>> FusedIterator for $forward_iterator<'a, P> {} - #[unstable(feature = "fused", issue = "35602")] + #[stable(feature = "fused", since = "1.26.0")] impl<'a, P: Pattern<'a>> FusedIterator for $reverse_iterator<'a, P> where P::Searcher: ReverseSearcher<'a> {} @@ -1051,7 +1055,7 @@ impl<'a, P: Pattern<'a>> SplitInternal<'a, P> { if !self.finished && (self.allow_trailing_empty || self.end - self.start > 0) { self.finished = true; unsafe { - let string = self.matcher.haystack().slice_unchecked(self.start, self.end); + let string = self.matcher.haystack().get_unchecked(self.start..self.end); Some(string) } } else { @@ -1066,7 +1070,7 @@ impl<'a, P: Pattern<'a>> SplitInternal<'a, P> { let haystack = self.matcher.haystack(); match self.matcher.next_match() { Some((a, b)) => unsafe { - let elt = haystack.slice_unchecked(self.start, a); + let elt = haystack.get_unchecked(self.start..a); self.start = b; Some(elt) }, @@ -1091,13 +1095,13 @@ impl<'a, P: Pattern<'a>> SplitInternal<'a, P> { let haystack = self.matcher.haystack(); match self.matcher.next_match_back() { Some((a, b)) => unsafe { - let elt = haystack.slice_unchecked(b, self.end); + let elt = haystack.get_unchecked(b..self.end); self.end = a; Some(elt) }, None => unsafe { self.finished = true; - Some(haystack.slice_unchecked(self.start, self.end)) + Some(haystack.get_unchecked(self.start..self.end)) }, } } @@ -1218,7 +1222,7 @@ impl<'a, P: Pattern<'a>> MatchIndicesInternal<'a, P> { #[inline] fn next(&mut self) -> Option<(usize, &'a str)> { self.0.next_match().map(|(start, end)| unsafe { - (start, self.0.haystack().slice_unchecked(start, end)) + (start, self.0.haystack().get_unchecked(start..end)) }) } @@ -1227,7 +1231,7 @@ impl<'a, P: Pattern<'a>> MatchIndicesInternal<'a, P> { where P::Searcher: ReverseSearcher<'a> { self.0.next_match_back().map(|(start, end)| unsafe { - (start, self.0.haystack().slice_unchecked(start, end)) + (start, self.0.haystack().get_unchecked(start..end)) }) } } @@ -1270,7 +1274,7 @@ impl<'a, P: Pattern<'a>> MatchesInternal<'a, P> { fn next(&mut self) -> Option<&'a str> { self.0.next_match().map(|(a, b)| unsafe { // Indices are known to be on utf8 boundaries - self.0.haystack().slice_unchecked(a, b) + self.0.haystack().get_unchecked(a..b) }) } @@ -1280,7 +1284,7 @@ impl<'a, P: Pattern<'a>> MatchesInternal<'a, P> { { self.0.next_match_back().map(|(a, b)| unsafe { // Indices are known to be on utf8 boundaries - self.0.haystack().slice_unchecked(a, b) + self.0.haystack().get_unchecked(a..b) }) } } @@ -1337,7 +1341,7 @@ impl<'a> DoubleEndedIterator for Lines<'a> { } } -#[unstable(feature = "fused", issue = "35602")] +#[stable(feature = "fused", since = "1.26.0")] impl<'a> FusedIterator for Lines<'a> {} /// Created with the method [`lines_any`]. @@ -1403,7 +1407,7 @@ impl<'a> DoubleEndedIterator for LinesAny<'a> { } } -#[unstable(feature = "fused", issue = "35602")] +#[stable(feature = "fused", since = "1.26.0")] #[allow(deprecated)] impl<'a> FusedIterator for LinesAny<'a> {} @@ -1480,10 +1484,10 @@ fn run_utf8_validation(v: &[u8]) -> Result<(), Utf8Error> { }, 3 => { match (first, next!()) { - (0xE0 , 0xA0 ... 0xBF) | - (0xE1 ... 0xEC, 0x80 ... 0xBF) | - (0xED , 0x80 ... 0x9F) | - (0xEE ... 0xEF, 0x80 ... 0xBF) => {} + (0xE0 , 0xA0 ..= 0xBF) | + (0xE1 ..= 0xEC, 0x80 ..= 0xBF) | + (0xED , 0x80 ..= 0x9F) | + (0xEE ..= 0xEF, 0x80 ..= 0xBF) => {} _ => err!(Some(1)) } if next!() & !CONT_MASK != TAG_CONT_U8 { @@ -1492,9 +1496,9 @@ fn run_utf8_validation(v: &[u8]) -> Result<(), Utf8Error> { } 4 => { match (first, next!()) { - (0xF0 , 0x90 ... 0xBF) | - (0xF1 ... 0xF3, 0x80 ... 0xBF) | - (0xF4 , 0x80 ... 0x8F) => {} + (0xF0 , 0x90 ..= 0xBF) | + (0xF1 ..= 0xF3, 0x80 ..= 0xBF) | + (0xF4 , 0x80 ..= 0x8F) => {} _ => err!(Some(1)) } if next!() & !CONT_MASK != TAG_CONT_U8 { @@ -1566,7 +1570,7 @@ static UTF8_CHAR_WIDTH: [u8; 256] = [ #[unstable(feature = "str_internals", issue = "0")] #[inline] pub fn utf8_char_width(b: u8) -> usize { - return UTF8_CHAR_WIDTH[b as usize] as usize; + UTF8_CHAR_WIDTH[b as usize] as usize } /// Mask of the value bits of a continuation byte. @@ -1779,9 +1783,7 @@ mod traits { } } - #[unstable(feature = "inclusive_range", - reason = "recently added, follows RFC", - issue = "28237")] + #[stable(feature = "inclusive_range", since = "1.26.0")] impl ops::Index<ops::RangeInclusive<usize>> for str { type Output = str; @@ -1791,9 +1793,7 @@ mod traits { } } - #[unstable(feature = "inclusive_range", - reason = "recently added, follows RFC", - issue = "28237")] + #[stable(feature = "inclusive_range", since = "1.26.0")] impl ops::Index<ops::RangeToInclusive<usize>> for str { type Output = str; @@ -1803,18 +1803,14 @@ mod traits { } } - #[unstable(feature = "inclusive_range", - reason = "recently added, follows RFC", - issue = "28237")] + #[stable(feature = "inclusive_range", since = "1.26.0")] impl ops::IndexMut<ops::RangeInclusive<usize>> for str { #[inline] fn index_mut(&mut self, index: ops::RangeInclusive<usize>) -> &mut str { index.index_mut(self) } } - #[unstable(feature = "inclusive_range", - reason = "recently added, follows RFC", - issue = "28237")] + #[stable(feature = "inclusive_range", since = "1.26.0")] impl ops::IndexMut<ops::RangeToInclusive<usize>> for str { #[inline] fn index_mut(&mut self, index: ops::RangeToInclusive<usize>) -> &mut str { @@ -1822,6 +1818,12 @@ mod traits { } } + #[inline(never)] + #[cold] + fn str_index_overflow_fail() -> ! { + panic!("attempted to index str up to maximum usize"); + } + #[stable(feature = "str_checked_slicing", since = "1.20.0")] impl SliceIndex<str> for ops::RangeFull { type Output = str; @@ -1997,202 +1999,73 @@ mod traits { } } - #[unstable(feature = "inclusive_range", - reason = "recently added, follows RFC", - issue = "28237")] + #[stable(feature = "inclusive_range", since = "1.26.0")] impl SliceIndex<str> for ops::RangeInclusive<usize> { type Output = str; #[inline] fn get(self, slice: &str) -> Option<&Self::Output> { - if let Some(end) = self.end.checked_add(1) { - (self.start..end).get(slice) - } else { - None - } + if *self.end() == usize::max_value() { None } + else { (*self.start()..self.end()+1).get(slice) } } #[inline] fn get_mut(self, slice: &mut str) -> Option<&mut Self::Output> { - if let Some(end) = self.end.checked_add(1) { - (self.start..end).get_mut(slice) - } else { - None - } + if *self.end() == usize::max_value() { None } + else { (*self.start()..self.end()+1).get_mut(slice) } } #[inline] unsafe fn get_unchecked(self, slice: &str) -> &Self::Output { - (self.start..self.end+1).get_unchecked(slice) + (*self.start()..self.end()+1).get_unchecked(slice) } #[inline] unsafe fn get_unchecked_mut(self, slice: &mut str) -> &mut Self::Output { - (self.start..self.end+1).get_unchecked_mut(slice) + (*self.start()..self.end()+1).get_unchecked_mut(slice) } #[inline] fn index(self, slice: &str) -> &Self::Output { - assert!(self.end != usize::max_value(), - "attempted to index str up to maximum usize"); - (self.start..self.end+1).index(slice) + if *self.end() == usize::max_value() { str_index_overflow_fail(); } + (*self.start()..self.end()+1).index(slice) } #[inline] fn index_mut(self, slice: &mut str) -> &mut Self::Output { - assert!(self.end != usize::max_value(), - "attempted to index str up to maximum usize"); - (self.start..self.end+1).index_mut(slice) + if *self.end() == usize::max_value() { str_index_overflow_fail(); } + (*self.start()..self.end()+1).index_mut(slice) } } - #[unstable(feature = "inclusive_range", - reason = "recently added, follows RFC", - issue = "28237")] + #[stable(feature = "inclusive_range", since = "1.26.0")] impl SliceIndex<str> for ops::RangeToInclusive<usize> { type Output = str; #[inline] fn get(self, slice: &str) -> Option<&Self::Output> { - if self.end < usize::max_value() && slice.is_char_boundary(self.end + 1) { - Some(unsafe { self.get_unchecked(slice) }) - } else { - None - } + if self.end == usize::max_value() { None } + else { (..self.end+1).get(slice) } } #[inline] fn get_mut(self, slice: &mut str) -> Option<&mut Self::Output> { - if self.end < usize::max_value() && slice.is_char_boundary(self.end + 1) { - Some(unsafe { self.get_unchecked_mut(slice) }) - } else { - None - } + if self.end == usize::max_value() { None } + else { (..self.end+1).get_mut(slice) } } #[inline] unsafe fn get_unchecked(self, slice: &str) -> &Self::Output { - let ptr = slice.as_ptr(); - super::from_utf8_unchecked(slice::from_raw_parts(ptr, self.end + 1)) + (..self.end+1).get_unchecked(slice) } #[inline] unsafe fn get_unchecked_mut(self, slice: &mut str) -> &mut Self::Output { - let ptr = slice.as_ptr(); - super::from_utf8_unchecked_mut(slice::from_raw_parts_mut(ptr as *mut u8, self.end + 1)) + (..self.end+1).get_unchecked_mut(slice) } #[inline] fn index(self, slice: &str) -> &Self::Output { - assert!(self.end != usize::max_value(), - "attempted to index str up to maximum usize"); - let end = self.end + 1; - self.get(slice).unwrap_or_else(|| super::slice_error_fail(slice, 0, end)) + if self.end == usize::max_value() { str_index_overflow_fail(); } + (..self.end+1).index(slice) } #[inline] fn index_mut(self, slice: &mut str) -> &mut Self::Output { - assert!(self.end != usize::max_value(), - "attempted to index str up to maximum usize"); - if slice.is_char_boundary(self.end) { - unsafe { self.get_unchecked_mut(slice) } - } else { - super::slice_error_fail(slice, 0, self.end + 1) - } + if self.end == usize::max_value() { str_index_overflow_fail(); } + (..self.end+1).index_mut(slice) } } - -} - - -/// Methods for string slices -#[allow(missing_docs)] -#[doc(hidden)] -#[unstable(feature = "core_str_ext", - reason = "stable interface provided by `impl str` in later crates", - issue = "32110")] -pub trait StrExt { - // NB there are no docs here are they're all located on the StrExt trait in - // liballoc, not here. - - #[stable(feature = "core", since = "1.6.0")] - fn contains<'a, P: Pattern<'a>>(&'a self, pat: P) -> bool; - #[stable(feature = "core", since = "1.6.0")] - fn chars(&self) -> Chars; - #[stable(feature = "core", since = "1.6.0")] - fn bytes(&self) -> Bytes; - #[stable(feature = "core", since = "1.6.0")] - fn char_indices(&self) -> CharIndices; - #[stable(feature = "core", since = "1.6.0")] - fn split<'a, P: Pattern<'a>>(&'a self, pat: P) -> Split<'a, P>; - #[stable(feature = "core", since = "1.6.0")] - fn rsplit<'a, P: Pattern<'a>>(&'a self, pat: P) -> RSplit<'a, P> - where P::Searcher: ReverseSearcher<'a>; - #[stable(feature = "core", since = "1.6.0")] - fn splitn<'a, P: Pattern<'a>>(&'a self, count: usize, pat: P) -> SplitN<'a, P>; - #[stable(feature = "core", since = "1.6.0")] - fn rsplitn<'a, P: Pattern<'a>>(&'a self, count: usize, pat: P) -> RSplitN<'a, P> - where P::Searcher: ReverseSearcher<'a>; - #[stable(feature = "core", since = "1.6.0")] - fn split_terminator<'a, P: Pattern<'a>>(&'a self, pat: P) -> SplitTerminator<'a, P>; - #[stable(feature = "core", since = "1.6.0")] - fn rsplit_terminator<'a, P: Pattern<'a>>(&'a self, pat: P) -> RSplitTerminator<'a, P> - where P::Searcher: ReverseSearcher<'a>; - #[stable(feature = "core", since = "1.6.0")] - fn matches<'a, P: Pattern<'a>>(&'a self, pat: P) -> Matches<'a, P>; - #[stable(feature = "core", since = "1.6.0")] - fn rmatches<'a, P: Pattern<'a>>(&'a self, pat: P) -> RMatches<'a, P> - where P::Searcher: ReverseSearcher<'a>; - #[stable(feature = "core", since = "1.6.0")] - fn match_indices<'a, P: Pattern<'a>>(&'a self, pat: P) -> MatchIndices<'a, P>; - #[stable(feature = "core", since = "1.6.0")] - fn rmatch_indices<'a, P: Pattern<'a>>(&'a self, pat: P) -> RMatchIndices<'a, P> - where P::Searcher: ReverseSearcher<'a>; - #[stable(feature = "core", since = "1.6.0")] - fn lines(&self) -> Lines; - #[stable(feature = "core", since = "1.6.0")] - #[rustc_deprecated(since = "1.6.0", reason = "use lines() instead now")] - #[allow(deprecated)] - fn lines_any(&self) -> LinesAny; - #[stable(feature = "str_checked_slicing", since = "1.20.0")] - fn get<I: SliceIndex<str>>(&self, i: I) -> Option<&I::Output>; - #[stable(feature = "str_checked_slicing", since = "1.20.0")] - fn get_mut<I: SliceIndex<str>>(&mut self, i: I) -> Option<&mut I::Output>; - #[stable(feature = "str_checked_slicing", since = "1.20.0")] - unsafe fn get_unchecked<I: SliceIndex<str>>(&self, i: I) -> &I::Output; - #[stable(feature = "str_checked_slicing", since = "1.20.0")] - unsafe fn get_unchecked_mut<I: SliceIndex<str>>(&mut self, i: I) -> &mut I::Output; - #[stable(feature = "core", since = "1.6.0")] - unsafe fn slice_unchecked(&self, begin: usize, end: usize) -> &str; - #[stable(feature = "core", since = "1.6.0")] - unsafe fn slice_mut_unchecked(&mut self, begin: usize, end: usize) -> &mut str; - #[stable(feature = "core", since = "1.6.0")] - fn starts_with<'a, P: Pattern<'a>>(&'a self, pat: P) -> bool; - #[stable(feature = "core", since = "1.6.0")] - fn ends_with<'a, P: Pattern<'a>>(&'a self, pat: P) -> bool - where P::Searcher: ReverseSearcher<'a>; - #[stable(feature = "core", since = "1.6.0")] - fn trim_matches<'a, P: Pattern<'a>>(&'a self, pat: P) -> &'a str - where P::Searcher: DoubleEndedSearcher<'a>; - #[stable(feature = "core", since = "1.6.0")] - fn trim_left_matches<'a, P: Pattern<'a>>(&'a self, pat: P) -> &'a str; - #[stable(feature = "core", since = "1.6.0")] - fn trim_right_matches<'a, P: Pattern<'a>>(&'a self, pat: P) -> &'a str - where P::Searcher: ReverseSearcher<'a>; - #[stable(feature = "is_char_boundary", since = "1.9.0")] - fn is_char_boundary(&self, index: usize) -> bool; - #[stable(feature = "core", since = "1.6.0")] - fn as_bytes(&self) -> &[u8]; - #[stable(feature = "str_mut_extras", since = "1.20.0")] - unsafe fn as_bytes_mut(&mut self) -> &mut [u8]; - #[stable(feature = "core", since = "1.6.0")] - fn find<'a, P: Pattern<'a>>(&'a self, pat: P) -> Option<usize>; - #[stable(feature = "core", since = "1.6.0")] - fn rfind<'a, P: Pattern<'a>>(&'a self, pat: P) -> Option<usize> - where P::Searcher: ReverseSearcher<'a>; - fn find_str<'a, P: Pattern<'a>>(&'a self, pat: P) -> Option<usize>; - #[stable(feature = "core", since = "1.6.0")] - fn split_at(&self, mid: usize) -> (&str, &str); - #[stable(feature = "core", since = "1.6.0")] - fn split_at_mut(&mut self, mid: usize) -> (&mut str, &mut str); - #[stable(feature = "core", since = "1.6.0")] - fn as_ptr(&self) -> *const u8; - #[stable(feature = "core", since = "1.6.0")] - fn len(&self) -> usize; - #[stable(feature = "core", since = "1.6.0")] - fn is_empty(&self) -> bool; - #[stable(feature = "core", since = "1.6.0")] - fn parse<T: FromStr>(&self) -> Result<T, T::Err>; } // truncate `&str` to length at most equal to `max` @@ -2239,154 +2112,1554 @@ fn slice_error_fail(s: &str, begin: usize, end: usize) -> ! { index, ch, char_range, s_trunc, ellipsis); } -#[stable(feature = "core", since = "1.6.0")] -impl StrExt for str { +#[lang = "str"] +#[cfg(not(test))] +impl str { + /// Returns the length of `self`. + /// + /// This length is in bytes, not [`char`]s or graphemes. In other words, + /// it may not be what a human considers the length of the string. + /// + /// # Examples + /// + /// Basic usage: + /// + /// ``` + /// let len = "foo".len(); + /// assert_eq!(3, len); + /// + /// let len = "ƒoo".len(); // fancy f! + /// assert_eq!(4, len); + /// ``` + #[stable(feature = "rust1", since = "1.0.0")] #[inline] - fn contains<'a, P: Pattern<'a>>(&'a self, pat: P) -> bool { - pat.is_contained_in(self) + #[rustc_const_unstable(feature = "const_str_len")] + pub const fn len(&self) -> usize { + self.as_bytes().len() } + /// Returns `true` if `self` has a length of zero bytes. + /// + /// # Examples + /// + /// Basic usage: + /// + /// ``` + /// let s = ""; + /// assert!(s.is_empty()); + /// + /// let s = "not empty"; + /// assert!(!s.is_empty()); + /// ``` #[inline] - fn chars(&self) -> Chars { - Chars{iter: self.as_bytes().iter()} + #[stable(feature = "rust1", since = "1.0.0")] + #[rustc_const_unstable(feature = "const_str_len")] + pub const fn is_empty(&self) -> bool { + self.len() == 0 } + /// Checks that `index`-th byte lies at the start and/or end of a + /// UTF-8 code point sequence. + /// + /// The start and end of the string (when `index == self.len()`) are + /// considered to be + /// boundaries. + /// + /// Returns `false` if `index` is greater than `self.len()`. + /// + /// # Examples + /// + /// ``` + /// let s = "Löwe 老虎 Léopard"; + /// assert!(s.is_char_boundary(0)); + /// // start of `老` + /// assert!(s.is_char_boundary(6)); + /// assert!(s.is_char_boundary(s.len())); + /// + /// // second byte of `ö` + /// assert!(!s.is_char_boundary(2)); + /// + /// // third byte of `老` + /// assert!(!s.is_char_boundary(8)); + /// ``` + #[stable(feature = "is_char_boundary", since = "1.9.0")] #[inline] - fn bytes(&self) -> Bytes { - Bytes(self.as_bytes().iter().cloned()) + pub fn is_char_boundary(&self, index: usize) -> bool { + // 0 and len are always ok. + // Test for 0 explicitly so that it can optimize out the check + // easily and skip reading string data for that case. + if index == 0 || index == self.len() { return true; } + match self.as_bytes().get(index) { + None => false, + // This is bit magic equivalent to: b < 128 || b >= 192 + Some(&b) => (b as i8) >= -0x40, + } } + /// Converts a string slice to a byte slice. To convert the byte slice back + /// into a string slice, use the [`str::from_utf8`] function. + /// + /// [`str::from_utf8`]: ./str/fn.from_utf8.html + /// + /// # Examples + /// + /// Basic usage: + /// + /// ``` + /// let bytes = "bors".as_bytes(); + /// assert_eq!(b"bors", bytes); + /// ``` + #[stable(feature = "rust1", since = "1.0.0")] + #[inline(always)] + #[rustc_const_unstable(feature="const_str_as_bytes")] + pub const fn as_bytes(&self) -> &[u8] { + union Slices<'a> { + str: &'a str, + slice: &'a [u8], + } + unsafe { Slices { str: self }.slice } + } + + /// Converts a mutable string slice to a mutable byte slice. To convert the + /// mutable byte slice back into a mutable string slice, use the + /// [`str::from_utf8_mut`] function. + /// + /// [`str::from_utf8_mut`]: ./str/fn.from_utf8_mut.html + /// + /// # Examples + /// + /// Basic usage: + /// + /// ``` + /// let mut s = String::from("Hello"); + /// let bytes = unsafe { s.as_bytes_mut() }; + /// + /// assert_eq!(b"Hello", bytes); + /// ``` + /// + /// Mutability: + /// + /// ``` + /// let mut s = String::from("🗻∈🌏"); + /// + /// unsafe { + /// let bytes = s.as_bytes_mut(); + /// + /// bytes[0] = 0xF0; + /// bytes[1] = 0x9F; + /// bytes[2] = 0x8D; + /// bytes[3] = 0x94; + /// } + /// + /// assert_eq!("🍔∈🌏", s); + /// ``` + #[stable(feature = "str_mut_extras", since = "1.20.0")] + #[inline(always)] + pub unsafe fn as_bytes_mut(&mut self) -> &mut [u8] { + &mut *(self as *mut str as *mut [u8]) + } + + /// Converts a string slice to a raw pointer. + /// + /// As string slices are a slice of bytes, the raw pointer points to a + /// [`u8`]. This pointer will be pointing to the first byte of the string + /// slice. + /// + /// [`u8`]: primitive.u8.html + /// + /// # Examples + /// + /// Basic usage: + /// + /// ``` + /// let s = "Hello"; + /// let ptr = s.as_ptr(); + /// ``` + #[stable(feature = "rust1", since = "1.0.0")] #[inline] - fn char_indices(&self) -> CharIndices { - CharIndices { front_offset: 0, iter: self.chars() } + #[rustc_const_unstable(feature = "const_str_as_ptr")] + pub const fn as_ptr(&self) -> *const u8 { + self as *const str as *const u8 } + /// Returns a subslice of `str`. + /// + /// This is the non-panicking alternative to indexing the `str`. Returns + /// [`None`] whenever equivalent indexing operation would panic. + /// + /// [`None`]: option/enum.Option.html#variant.None + /// + /// # Examples + /// + /// ``` + /// let v = String::from("🗻∈🌏"); + /// + /// assert_eq!(Some("🗻"), v.get(0..4)); + /// + /// // indices not on UTF-8 sequence boundaries + /// assert!(v.get(1..).is_none()); + /// assert!(v.get(..8).is_none()); + /// + /// // out of bounds + /// assert!(v.get(..42).is_none()); + /// ``` + #[stable(feature = "str_checked_slicing", since = "1.20.0")] #[inline] - fn split<'a, P: Pattern<'a>>(&'a self, pat: P) -> Split<'a, P> { - Split(SplitInternal { - start: 0, - end: self.len(), - matcher: pat.into_searcher(self), - allow_trailing_empty: true, - finished: false, - }) + pub fn get<I: SliceIndex<str>>(&self, i: I) -> Option<&I::Output> { + i.get(self) } + /// Returns a mutable subslice of `str`. + /// + /// This is the non-panicking alternative to indexing the `str`. Returns + /// [`None`] whenever equivalent indexing operation would panic. + /// + /// [`None`]: option/enum.Option.html#variant.None + /// + /// # Examples + /// + /// ``` + /// let mut v = String::from("hello"); + /// // correct length + /// assert!(v.get_mut(0..5).is_some()); + /// // out of bounds + /// assert!(v.get_mut(..42).is_none()); + /// assert_eq!(Some("he"), v.get_mut(0..2).map(|v| &*v)); + /// + /// assert_eq!("hello", v); + /// { + /// let s = v.get_mut(0..2); + /// let s = s.map(|s| { + /// s.make_ascii_uppercase(); + /// &*s + /// }); + /// assert_eq!(Some("HE"), s); + /// } + /// assert_eq!("HEllo", v); + /// ``` + #[stable(feature = "str_checked_slicing", since = "1.20.0")] #[inline] - fn rsplit<'a, P: Pattern<'a>>(&'a self, pat: P) -> RSplit<'a, P> - where P::Searcher: ReverseSearcher<'a> - { - RSplit(self.split(pat).0) + pub fn get_mut<I: SliceIndex<str>>(&mut self, i: I) -> Option<&mut I::Output> { + i.get_mut(self) } + /// Returns a unchecked subslice of `str`. + /// + /// This is the unchecked alternative to indexing the `str`. + /// + /// # Safety + /// + /// Callers of this function are responsible that these preconditions are + /// satisfied: + /// + /// * The starting index must come before the ending index; + /// * Indexes must be within bounds of the original slice; + /// * Indexes must lie on UTF-8 sequence boundaries. + /// + /// Failing that, the returned string slice may reference invalid memory or + /// violate the invariants communicated by the `str` type. + /// + /// # Examples + /// + /// ``` + /// let v = "🗻∈🌏"; + /// unsafe { + /// assert_eq!("🗻", v.get_unchecked(0..4)); + /// assert_eq!("∈", v.get_unchecked(4..7)); + /// assert_eq!("🌏", v.get_unchecked(7..11)); + /// } + /// ``` + #[stable(feature = "str_checked_slicing", since = "1.20.0")] #[inline] - fn splitn<'a, P: Pattern<'a>>(&'a self, count: usize, pat: P) -> SplitN<'a, P> { - SplitN(SplitNInternal { - iter: self.split(pat).0, - count, - }) + pub unsafe fn get_unchecked<I: SliceIndex<str>>(&self, i: I) -> &I::Output { + i.get_unchecked(self) } + /// Returns a mutable, unchecked subslice of `str`. + /// + /// This is the unchecked alternative to indexing the `str`. + /// + /// # Safety + /// + /// Callers of this function are responsible that these preconditions are + /// satisfied: + /// + /// * The starting index must come before the ending index; + /// * Indexes must be within bounds of the original slice; + /// * Indexes must lie on UTF-8 sequence boundaries. + /// + /// Failing that, the returned string slice may reference invalid memory or + /// violate the invariants communicated by the `str` type. + /// + /// # Examples + /// + /// ``` + /// let mut v = String::from("🗻∈🌏"); + /// unsafe { + /// assert_eq!("🗻", v.get_unchecked_mut(0..4)); + /// assert_eq!("∈", v.get_unchecked_mut(4..7)); + /// assert_eq!("🌏", v.get_unchecked_mut(7..11)); + /// } + /// ``` + #[stable(feature = "str_checked_slicing", since = "1.20.0")] #[inline] - fn rsplitn<'a, P: Pattern<'a>>(&'a self, count: usize, pat: P) -> RSplitN<'a, P> - where P::Searcher: ReverseSearcher<'a> - { - RSplitN(self.splitn(count, pat).0) + pub unsafe fn get_unchecked_mut<I: SliceIndex<str>>(&mut self, i: I) -> &mut I::Output { + i.get_unchecked_mut(self) } + /// Creates a string slice from another string slice, bypassing safety + /// checks. + /// + /// This is generally not recommended, use with caution! For a safe + /// alternative see [`str`] and [`Index`]. + /// + /// [`str`]: primitive.str.html + /// [`Index`]: ops/trait.Index.html + /// + /// This new slice goes from `begin` to `end`, including `begin` but + /// excluding `end`. + /// + /// To get a mutable string slice instead, see the + /// [`slice_mut_unchecked`] method. + /// + /// [`slice_mut_unchecked`]: #method.slice_mut_unchecked + /// + /// # Safety + /// + /// Callers of this function are responsible that three preconditions are + /// satisfied: + /// + /// * `begin` must come before `end`. + /// * `begin` and `end` must be byte positions within the string slice. + /// * `begin` and `end` must lie on UTF-8 sequence boundaries. + /// + /// # Examples + /// + /// Basic usage: + /// + /// ``` + /// let s = "Löwe 老虎 Léopard"; + /// + /// unsafe { + /// assert_eq!("Löwe 老虎 Léopard", s.slice_unchecked(0, 21)); + /// } + /// + /// let s = "Hello, world!"; + /// + /// unsafe { + /// assert_eq!("world", s.slice_unchecked(7, 12)); + /// } + /// ``` + #[stable(feature = "rust1", since = "1.0.0")] + #[rustc_deprecated(since = "1.29.0", reason = "use `get_unchecked(begin..end)` instead")] #[inline] - fn split_terminator<'a, P: Pattern<'a>>(&'a self, pat: P) -> SplitTerminator<'a, P> { - SplitTerminator(SplitInternal { - allow_trailing_empty: false, - ..self.split(pat).0 - }) + pub unsafe fn slice_unchecked(&self, begin: usize, end: usize) -> &str { + (begin..end).get_unchecked(self) } + /// Creates a string slice from another string slice, bypassing safety + /// checks. + /// This is generally not recommended, use with caution! For a safe + /// alternative see [`str`] and [`IndexMut`]. + /// + /// [`str`]: primitive.str.html + /// [`IndexMut`]: ops/trait.IndexMut.html + /// + /// This new slice goes from `begin` to `end`, including `begin` but + /// excluding `end`. + /// + /// To get an immutable string slice instead, see the + /// [`slice_unchecked`] method. + /// + /// [`slice_unchecked`]: #method.slice_unchecked + /// + /// # Safety + /// + /// Callers of this function are responsible that three preconditions are + /// satisfied: + /// + /// * `begin` must come before `end`. + /// * `begin` and `end` must be byte positions within the string slice. + /// * `begin` and `end` must lie on UTF-8 sequence boundaries. + #[stable(feature = "str_slice_mut", since = "1.5.0")] + #[rustc_deprecated(since = "1.29.0", reason = "use `get_unchecked_mut(begin..end)` instead")] #[inline] - fn rsplit_terminator<'a, P: Pattern<'a>>(&'a self, pat: P) -> RSplitTerminator<'a, P> - where P::Searcher: ReverseSearcher<'a> - { - RSplitTerminator(self.split_terminator(pat).0) + pub unsafe fn slice_mut_unchecked(&mut self, begin: usize, end: usize) -> &mut str { + (begin..end).get_unchecked_mut(self) } + /// Divide one string slice into two at an index. + /// + /// The argument, `mid`, should be a byte offset from the start of the + /// string. It must also be on the boundary of a UTF-8 code point. + /// + /// The two slices returned go from the start of the string slice to `mid`, + /// and from `mid` to the end of the string slice. + /// + /// To get mutable string slices instead, see the [`split_at_mut`] + /// method. + /// + /// [`split_at_mut`]: #method.split_at_mut + /// + /// # Panics + /// + /// Panics if `mid` is not on a UTF-8 code point boundary, or if it is + /// beyond the last code point of the string slice. + /// + /// # Examples + /// + /// Basic usage: + /// + /// ``` + /// let s = "Per Martin-Löf"; + /// + /// let (first, last) = s.split_at(3); + /// + /// assert_eq!("Per", first); + /// assert_eq!(" Martin-Löf", last); + /// ``` #[inline] - fn matches<'a, P: Pattern<'a>>(&'a self, pat: P) -> Matches<'a, P> { - Matches(MatchesInternal(pat.into_searcher(self))) + #[stable(feature = "str_split_at", since = "1.4.0")] + pub fn split_at(&self, mid: usize) -> (&str, &str) { + // is_char_boundary checks that the index is in [0, .len()] + if self.is_char_boundary(mid) { + unsafe { + (self.get_unchecked(0..mid), + self.get_unchecked(mid..self.len())) + } + } else { + slice_error_fail(self, 0, mid) + } } + /// Divide one mutable string slice into two at an index. + /// + /// The argument, `mid`, should be a byte offset from the start of the + /// string. It must also be on the boundary of a UTF-8 code point. + /// + /// The two slices returned go from the start of the string slice to `mid`, + /// and from `mid` to the end of the string slice. + /// + /// To get immutable string slices instead, see the [`split_at`] method. + /// + /// [`split_at`]: #method.split_at + /// + /// # Panics + /// + /// Panics if `mid` is not on a UTF-8 code point boundary, or if it is + /// beyond the last code point of the string slice. + /// + /// # Examples + /// + /// Basic usage: + /// + /// ``` + /// let mut s = "Per Martin-Löf".to_string(); + /// { + /// let (first, last) = s.split_at_mut(3); + /// first.make_ascii_uppercase(); + /// assert_eq!("PER", first); + /// assert_eq!(" Martin-Löf", last); + /// } + /// assert_eq!("PER Martin-Löf", s); + /// ``` #[inline] - fn rmatches<'a, P: Pattern<'a>>(&'a self, pat: P) -> RMatches<'a, P> - where P::Searcher: ReverseSearcher<'a> - { - RMatches(self.matches(pat).0) + #[stable(feature = "str_split_at", since = "1.4.0")] + pub fn split_at_mut(&mut self, mid: usize) -> (&mut str, &mut str) { + // is_char_boundary checks that the index is in [0, .len()] + if self.is_char_boundary(mid) { + let len = self.len(); + let ptr = self.as_ptr() as *mut u8; + unsafe { + (from_utf8_unchecked_mut(slice::from_raw_parts_mut(ptr, mid)), + from_utf8_unchecked_mut(slice::from_raw_parts_mut( + ptr.offset(mid as isize), + len - mid + ))) + } + } else { + slice_error_fail(self, 0, mid) + } } + /// Returns an iterator over the [`char`]s of a string slice. + /// + /// As a string slice consists of valid UTF-8, we can iterate through a + /// string slice by [`char`]. This method returns such an iterator. + /// + /// It's important to remember that [`char`] represents a Unicode Scalar + /// Value, and may not match your idea of what a 'character' is. Iteration + /// over grapheme clusters may be what you actually want. + /// + /// # Examples + /// + /// Basic usage: + /// + /// ``` + /// let word = "goodbye"; + /// + /// let count = word.chars().count(); + /// assert_eq!(7, count); + /// + /// let mut chars = word.chars(); + /// + /// assert_eq!(Some('g'), chars.next()); + /// assert_eq!(Some('o'), chars.next()); + /// assert_eq!(Some('o'), chars.next()); + /// assert_eq!(Some('d'), chars.next()); + /// assert_eq!(Some('b'), chars.next()); + /// assert_eq!(Some('y'), chars.next()); + /// assert_eq!(Some('e'), chars.next()); + /// + /// assert_eq!(None, chars.next()); + /// ``` + /// + /// Remember, [`char`]s may not match your human intuition about characters: + /// + /// ``` + /// let y = "y̆"; + /// + /// let mut chars = y.chars(); + /// + /// assert_eq!(Some('y'), chars.next()); // not 'y̆' + /// assert_eq!(Some('\u{0306}'), chars.next()); + /// + /// assert_eq!(None, chars.next()); + /// ``` + #[stable(feature = "rust1", since = "1.0.0")] #[inline] - fn match_indices<'a, P: Pattern<'a>>(&'a self, pat: P) -> MatchIndices<'a, P> { - MatchIndices(MatchIndicesInternal(pat.into_searcher(self))) + pub fn chars(&self) -> Chars { + Chars{iter: self.as_bytes().iter()} } + /// Returns an iterator over the [`char`]s of a string slice, and their + /// positions. + /// + /// As a string slice consists of valid UTF-8, we can iterate through a + /// string slice by [`char`]. This method returns an iterator of both + /// these [`char`]s, as well as their byte positions. + /// + /// The iterator yields tuples. The position is first, the [`char`] is + /// second. + /// + /// # Examples + /// + /// Basic usage: + /// + /// ``` + /// let word = "goodbye"; + /// + /// let count = word.char_indices().count(); + /// assert_eq!(7, count); + /// + /// let mut char_indices = word.char_indices(); + /// + /// assert_eq!(Some((0, 'g')), char_indices.next()); + /// assert_eq!(Some((1, 'o')), char_indices.next()); + /// assert_eq!(Some((2, 'o')), char_indices.next()); + /// assert_eq!(Some((3, 'd')), char_indices.next()); + /// assert_eq!(Some((4, 'b')), char_indices.next()); + /// assert_eq!(Some((5, 'y')), char_indices.next()); + /// assert_eq!(Some((6, 'e')), char_indices.next()); + /// + /// assert_eq!(None, char_indices.next()); + /// ``` + /// + /// Remember, [`char`]s may not match your human intuition about characters: + /// + /// ``` + /// let yes = "y̆es"; + /// + /// let mut char_indices = yes.char_indices(); + /// + /// assert_eq!(Some((0, 'y')), char_indices.next()); // not (0, 'y̆') + /// assert_eq!(Some((1, '\u{0306}')), char_indices.next()); + /// + /// // note the 3 here - the last character took up two bytes + /// assert_eq!(Some((3, 'e')), char_indices.next()); + /// assert_eq!(Some((4, 's')), char_indices.next()); + /// + /// assert_eq!(None, char_indices.next()); + /// ``` + #[stable(feature = "rust1", since = "1.0.0")] #[inline] - fn rmatch_indices<'a, P: Pattern<'a>>(&'a self, pat: P) -> RMatchIndices<'a, P> - where P::Searcher: ReverseSearcher<'a> - { - RMatchIndices(self.match_indices(pat).0) + pub fn char_indices(&self) -> CharIndices { + CharIndices { front_offset: 0, iter: self.chars() } } + + /// An iterator over the bytes of a string slice. + /// + /// As a string slice consists of a sequence of bytes, we can iterate + /// through a string slice by byte. This method returns such an iterator. + /// + /// # Examples + /// + /// Basic usage: + /// + /// ``` + /// let mut bytes = "bors".bytes(); + /// + /// assert_eq!(Some(b'b'), bytes.next()); + /// assert_eq!(Some(b'o'), bytes.next()); + /// assert_eq!(Some(b'r'), bytes.next()); + /// assert_eq!(Some(b's'), bytes.next()); + /// + /// assert_eq!(None, bytes.next()); + /// ``` + #[stable(feature = "rust1", since = "1.0.0")] #[inline] - fn lines(&self) -> Lines { + pub fn bytes(&self) -> Bytes { + Bytes(self.as_bytes().iter().cloned()) + } + + /// Split a string slice by whitespace. + /// + /// The iterator returned will return string slices that are sub-slices of + /// the original string slice, separated by any amount of whitespace. + /// + /// 'Whitespace' is defined according to the terms of the Unicode Derived + /// Core Property `White_Space`. If you only want to split on ASCII whitespace + /// instead, use [`split_ascii_whitespace`]. + /// + /// [`split_ascii_whitespace`]: #method.split_ascii_whitespace + /// + /// # Examples + /// + /// Basic usage: + /// + /// ``` + /// let mut iter = "A few words".split_whitespace(); + /// + /// assert_eq!(Some("A"), iter.next()); + /// assert_eq!(Some("few"), iter.next()); + /// assert_eq!(Some("words"), iter.next()); + /// + /// assert_eq!(None, iter.next()); + /// ``` + /// + /// All kinds of whitespace are considered: + /// + /// ``` + /// let mut iter = " Mary had\ta\u{2009}little \n\t lamb".split_whitespace(); + /// assert_eq!(Some("Mary"), iter.next()); + /// assert_eq!(Some("had"), iter.next()); + /// assert_eq!(Some("a"), iter.next()); + /// assert_eq!(Some("little"), iter.next()); + /// assert_eq!(Some("lamb"), iter.next()); + /// + /// assert_eq!(None, iter.next()); + /// ``` + #[stable(feature = "split_whitespace", since = "1.1.0")] + #[inline] + pub fn split_whitespace(&self) -> SplitWhitespace { + SplitWhitespace { inner: self.split(IsWhitespace).filter(IsNotEmpty) } + } + + /// Split a string slice by ASCII whitespace. + /// + /// The iterator returned will return string slices that are sub-slices of + /// the original string slice, separated by any amount of ASCII whitespace. + /// + /// To split by Unicode `Whitespace` instead, use [`split_whitespace`]. + /// + /// [`split_whitespace`]: #method.split_whitespace + /// + /// # Examples + /// + /// Basic usage: + /// + /// ``` + /// #![feature(split_ascii_whitespace)] + /// let mut iter = "A few words".split_ascii_whitespace(); + /// + /// assert_eq!(Some("A"), iter.next()); + /// assert_eq!(Some("few"), iter.next()); + /// assert_eq!(Some("words"), iter.next()); + /// + /// assert_eq!(None, iter.next()); + /// ``` + /// + /// All kinds of ASCII whitespace are considered: + /// + /// ``` + /// let mut iter = " Mary had\ta little \n\t lamb".split_whitespace(); + /// assert_eq!(Some("Mary"), iter.next()); + /// assert_eq!(Some("had"), iter.next()); + /// assert_eq!(Some("a"), iter.next()); + /// assert_eq!(Some("little"), iter.next()); + /// assert_eq!(Some("lamb"), iter.next()); + /// + /// assert_eq!(None, iter.next()); + /// ``` + #[unstable(feature = "split_ascii_whitespace", issue = "48656")] + #[inline] + pub fn split_ascii_whitespace(&self) -> SplitAsciiWhitespace { + let inner = self + .as_bytes() + .split(IsAsciiWhitespace) + .filter(IsNotEmpty) + .map(UnsafeBytesToStr); + SplitAsciiWhitespace { inner } + } + + /// An iterator over the lines of a string, as string slices. + /// + /// Lines are ended with either a newline (`\n`) or a carriage return with + /// a line feed (`\r\n`). + /// + /// The final line ending is optional. + /// + /// # Examples + /// + /// Basic usage: + /// + /// ``` + /// let text = "foo\r\nbar\n\nbaz\n"; + /// let mut lines = text.lines(); + /// + /// assert_eq!(Some("foo"), lines.next()); + /// assert_eq!(Some("bar"), lines.next()); + /// assert_eq!(Some(""), lines.next()); + /// assert_eq!(Some("baz"), lines.next()); + /// + /// assert_eq!(None, lines.next()); + /// ``` + /// + /// The final line ending isn't required: + /// + /// ``` + /// let text = "foo\nbar\n\r\nbaz"; + /// let mut lines = text.lines(); + /// + /// assert_eq!(Some("foo"), lines.next()); + /// assert_eq!(Some("bar"), lines.next()); + /// assert_eq!(Some(""), lines.next()); + /// assert_eq!(Some("baz"), lines.next()); + /// + /// assert_eq!(None, lines.next()); + /// ``` + #[stable(feature = "rust1", since = "1.0.0")] + #[inline] + pub fn lines(&self) -> Lines { Lines(self.split_terminator('\n').map(LinesAnyMap)) } + /// An iterator over the lines of a string. + #[stable(feature = "rust1", since = "1.0.0")] + #[rustc_deprecated(since = "1.4.0", reason = "use lines() instead now")] #[inline] #[allow(deprecated)] - fn lines_any(&self) -> LinesAny { + pub fn lines_any(&self) -> LinesAny { LinesAny(self.lines()) } + /// Returns an iterator of `u16` over the string encoded as UTF-16. + /// + /// # Examples + /// + /// Basic usage: + /// + /// ``` + /// let text = "Zażółć gęślą jaźń"; + /// + /// let utf8_len = text.len(); + /// let utf16_len = text.encode_utf16().count(); + /// + /// assert!(utf16_len <= utf8_len); + /// ``` + #[stable(feature = "encode_utf16", since = "1.8.0")] + pub fn encode_utf16(&self) -> EncodeUtf16 { + EncodeUtf16 { chars: self.chars(), extra: 0 } + } + + /// Returns `true` if the given pattern matches a sub-slice of + /// this string slice. + /// + /// Returns `false` if it does not. + /// + /// # Examples + /// + /// Basic usage: + /// + /// ``` + /// let bananas = "bananas"; + /// + /// assert!(bananas.contains("nana")); + /// assert!(!bananas.contains("apples")); + /// ``` + #[stable(feature = "rust1", since = "1.0.0")] #[inline] - fn get<I: SliceIndex<str>>(&self, i: I) -> Option<&I::Output> { - i.get(self) + pub fn contains<'a, P: Pattern<'a>>(&'a self, pat: P) -> bool { + pat.is_contained_in(self) + } + + /// Returns `true` if the given pattern matches a prefix of this + /// string slice. + /// + /// Returns `false` if it does not. + /// + /// # Examples + /// + /// Basic usage: + /// + /// ``` + /// let bananas = "bananas"; + /// + /// assert!(bananas.starts_with("bana")); + /// assert!(!bananas.starts_with("nana")); + /// ``` + #[stable(feature = "rust1", since = "1.0.0")] + pub fn starts_with<'a, P: Pattern<'a>>(&'a self, pat: P) -> bool { + pat.is_prefix_of(self) + } + + /// Returns `true` if the given pattern matches a suffix of this + /// string slice. + /// + /// Returns `false` if it does not. + /// + /// # Examples + /// + /// Basic usage: + /// + /// ``` + /// let bananas = "bananas"; + /// + /// assert!(bananas.ends_with("anas")); + /// assert!(!bananas.ends_with("nana")); + /// ``` + #[stable(feature = "rust1", since = "1.0.0")] + pub fn ends_with<'a, P: Pattern<'a>>(&'a self, pat: P) -> bool + where P::Searcher: ReverseSearcher<'a> + { + pat.is_suffix_of(self) } + /// Returns the byte index of the first character of this string slice that + /// matches the pattern. + /// + /// Returns [`None`] if the pattern doesn't match. + /// + /// The pattern can be a `&str`, [`char`], or a closure that determines if + /// a character matches. + /// + /// [`None`]: option/enum.Option.html#variant.None + /// + /// # Examples + /// + /// Simple patterns: + /// + /// ``` + /// let s = "Löwe 老虎 Léopard"; + /// + /// assert_eq!(s.find('L'), Some(0)); + /// assert_eq!(s.find('é'), Some(14)); + /// assert_eq!(s.find("Léopard"), Some(13)); + /// ``` + /// + /// More complex patterns using point-free style and closures: + /// + /// ``` + /// let s = "Löwe 老虎 Léopard"; + /// + /// assert_eq!(s.find(char::is_whitespace), Some(5)); + /// assert_eq!(s.find(char::is_lowercase), Some(1)); + /// assert_eq!(s.find(|c: char| c.is_whitespace() || c.is_lowercase()), Some(1)); + /// assert_eq!(s.find(|c: char| (c < 'o') && (c > 'a')), Some(4)); + /// ``` + /// + /// Not finding the pattern: + /// + /// ``` + /// let s = "Löwe 老虎 Léopard"; + /// let x: &[_] = &['1', '2']; + /// + /// assert_eq!(s.find(x), None); + /// ``` + #[stable(feature = "rust1", since = "1.0.0")] #[inline] - fn get_mut<I: SliceIndex<str>>(&mut self, i: I) -> Option<&mut I::Output> { - i.get_mut(self) + pub fn find<'a, P: Pattern<'a>>(&'a self, pat: P) -> Option<usize> { + pat.into_searcher(self).next_match().map(|(i, _)| i) } + /// Returns the byte index of the last character of this string slice that + /// matches the pattern. + /// + /// Returns [`None`] if the pattern doesn't match. + /// + /// The pattern can be a `&str`, [`char`], or a closure that determines if + /// a character matches. + /// + /// [`None`]: option/enum.Option.html#variant.None + /// + /// # Examples + /// + /// Simple patterns: + /// + /// ``` + /// let s = "Löwe 老虎 Léopard"; + /// + /// assert_eq!(s.rfind('L'), Some(13)); + /// assert_eq!(s.rfind('é'), Some(14)); + /// ``` + /// + /// More complex patterns with closures: + /// + /// ``` + /// let s = "Löwe 老虎 Léopard"; + /// + /// assert_eq!(s.rfind(char::is_whitespace), Some(12)); + /// assert_eq!(s.rfind(char::is_lowercase), Some(20)); + /// ``` + /// + /// Not finding the pattern: + /// + /// ``` + /// let s = "Löwe 老虎 Léopard"; + /// let x: &[_] = &['1', '2']; + /// + /// assert_eq!(s.rfind(x), None); + /// ``` + #[stable(feature = "rust1", since = "1.0.0")] #[inline] - unsafe fn get_unchecked<I: SliceIndex<str>>(&self, i: I) -> &I::Output { - i.get_unchecked(self) + pub fn rfind<'a, P: Pattern<'a>>(&'a self, pat: P) -> Option<usize> + where P::Searcher: ReverseSearcher<'a> + { + pat.into_searcher(self).next_match_back().map(|(i, _)| i) } + /// An iterator over substrings of this string slice, separated by + /// characters matched by a pattern. + /// + /// The pattern can be a `&str`, [`char`], or a closure that determines the + /// split. + /// + /// # Iterator behavior + /// + /// The returned iterator will be a [`DoubleEndedIterator`] if the pattern + /// allows a reverse search and forward/reverse search yields the same + /// elements. This is true for, eg, [`char`] but not for `&str`. + /// + /// [`DoubleEndedIterator`]: iter/trait.DoubleEndedIterator.html + /// + /// If the pattern allows a reverse search but its results might differ + /// from a forward search, the [`rsplit`] method can be used. + /// + /// [`rsplit`]: #method.rsplit + /// + /// # Examples + /// + /// Simple patterns: + /// + /// ``` + /// let v: Vec<&str> = "Mary had a little lamb".split(' ').collect(); + /// assert_eq!(v, ["Mary", "had", "a", "little", "lamb"]); + /// + /// let v: Vec<&str> = "".split('X').collect(); + /// assert_eq!(v, [""]); + /// + /// let v: Vec<&str> = "lionXXtigerXleopard".split('X').collect(); + /// assert_eq!(v, ["lion", "", "tiger", "leopard"]); + /// + /// let v: Vec<&str> = "lion::tiger::leopard".split("::").collect(); + /// assert_eq!(v, ["lion", "tiger", "leopard"]); + /// + /// let v: Vec<&str> = "abc1def2ghi".split(char::is_numeric).collect(); + /// assert_eq!(v, ["abc", "def", "ghi"]); + /// + /// let v: Vec<&str> = "lionXtigerXleopard".split(char::is_uppercase).collect(); + /// assert_eq!(v, ["lion", "tiger", "leopard"]); + /// ``` + /// + /// A more complex pattern, using a closure: + /// + /// ``` + /// let v: Vec<&str> = "abc1defXghi".split(|c| c == '1' || c == 'X').collect(); + /// assert_eq!(v, ["abc", "def", "ghi"]); + /// ``` + /// + /// If a string contains multiple contiguous separators, you will end up + /// with empty strings in the output: + /// + /// ``` + /// let x = "||||a||b|c".to_string(); + /// let d: Vec<_> = x.split('|').collect(); + /// + /// assert_eq!(d, &["", "", "", "", "a", "", "b", "c"]); + /// ``` + /// + /// Contiguous separators are separated by the empty string. + /// + /// ``` + /// let x = "(///)".to_string(); + /// let d: Vec<_> = x.split('/').collect(); + /// + /// assert_eq!(d, &["(", "", "", ")"]); + /// ``` + /// + /// Separators at the start or end of a string are neighbored + /// by empty strings. + /// + /// ``` + /// let d: Vec<_> = "010".split("0").collect(); + /// assert_eq!(d, &["", "1", ""]); + /// ``` + /// + /// When the empty string is used as a separator, it separates + /// every character in the string, along with the beginning + /// and end of the string. + /// + /// ``` + /// let f: Vec<_> = "rust".split("").collect(); + /// assert_eq!(f, &["", "r", "u", "s", "t", ""]); + /// ``` + /// + /// Contiguous separators can lead to possibly surprising behavior + /// when whitespace is used as the separator. This code is correct: + /// + /// ``` + /// let x = " a b c".to_string(); + /// let d: Vec<_> = x.split(' ').collect(); + /// + /// assert_eq!(d, &["", "", "", "", "a", "", "b", "c"]); + /// ``` + /// + /// It does _not_ give you: + /// + /// ```,ignore + /// assert_eq!(d, &["a", "b", "c"]); + /// ``` + /// + /// Use [`split_whitespace`] for this behavior. + /// + /// [`split_whitespace`]: #method.split_whitespace + #[stable(feature = "rust1", since = "1.0.0")] #[inline] - unsafe fn get_unchecked_mut<I: SliceIndex<str>>(&mut self, i: I) -> &mut I::Output { - i.get_unchecked_mut(self) + pub fn split<'a, P: Pattern<'a>>(&'a self, pat: P) -> Split<'a, P> { + Split(SplitInternal { + start: 0, + end: self.len(), + matcher: pat.into_searcher(self), + allow_trailing_empty: true, + finished: false, + }) } + /// An iterator over substrings of the given string slice, separated by + /// characters matched by a pattern and yielded in reverse order. + /// + /// The pattern can be a `&str`, [`char`], or a closure that determines the + /// split. + /// + /// # Iterator behavior + /// + /// The returned iterator requires that the pattern supports a reverse + /// search, and it will be a [`DoubleEndedIterator`] if a forward/reverse + /// search yields the same elements. + /// + /// [`DoubleEndedIterator`]: iter/trait.DoubleEndedIterator.html + /// + /// For iterating from the front, the [`split`] method can be used. + /// + /// [`split`]: #method.split + /// + /// # Examples + /// + /// Simple patterns: + /// + /// ``` + /// let v: Vec<&str> = "Mary had a little lamb".rsplit(' ').collect(); + /// assert_eq!(v, ["lamb", "little", "a", "had", "Mary"]); + /// + /// let v: Vec<&str> = "".rsplit('X').collect(); + /// assert_eq!(v, [""]); + /// + /// let v: Vec<&str> = "lionXXtigerXleopard".rsplit('X').collect(); + /// assert_eq!(v, ["leopard", "tiger", "", "lion"]); + /// + /// let v: Vec<&str> = "lion::tiger::leopard".rsplit("::").collect(); + /// assert_eq!(v, ["leopard", "tiger", "lion"]); + /// ``` + /// + /// A more complex pattern, using a closure: + /// + /// ``` + /// let v: Vec<&str> = "abc1defXghi".rsplit(|c| c == '1' || c == 'X').collect(); + /// assert_eq!(v, ["ghi", "def", "abc"]); + /// ``` + #[stable(feature = "rust1", since = "1.0.0")] #[inline] - unsafe fn slice_unchecked(&self, begin: usize, end: usize) -> &str { - (begin..end).get_unchecked(self) + pub fn rsplit<'a, P: Pattern<'a>>(&'a self, pat: P) -> RSplit<'a, P> + where P::Searcher: ReverseSearcher<'a> + { + RSplit(self.split(pat).0) } + /// An iterator over substrings of the given string slice, separated by + /// characters matched by a pattern. + /// + /// The pattern can be a `&str`, [`char`], or a closure that determines the + /// split. + /// + /// Equivalent to [`split`], except that the trailing substring + /// is skipped if empty. + /// + /// [`split`]: #method.split + /// + /// This method can be used for string data that is _terminated_, + /// rather than _separated_ by a pattern. + /// + /// # Iterator behavior + /// + /// The returned iterator will be a [`DoubleEndedIterator`] if the pattern + /// allows a reverse search and forward/reverse search yields the same + /// elements. This is true for, eg, [`char`] but not for `&str`. + /// + /// [`DoubleEndedIterator`]: iter/trait.DoubleEndedIterator.html + /// + /// If the pattern allows a reverse search but its results might differ + /// from a forward search, the [`rsplit_terminator`] method can be used. + /// + /// [`rsplit_terminator`]: #method.rsplit_terminator + /// + /// # Examples + /// + /// Basic usage: + /// + /// ``` + /// let v: Vec<&str> = "A.B.".split_terminator('.').collect(); + /// assert_eq!(v, ["A", "B"]); + /// + /// let v: Vec<&str> = "A..B..".split_terminator(".").collect(); + /// assert_eq!(v, ["A", "", "B", ""]); + /// ``` + #[stable(feature = "rust1", since = "1.0.0")] #[inline] - unsafe fn slice_mut_unchecked(&mut self, begin: usize, end: usize) -> &mut str { - (begin..end).get_unchecked_mut(self) + pub fn split_terminator<'a, P: Pattern<'a>>(&'a self, pat: P) -> SplitTerminator<'a, P> { + SplitTerminator(SplitInternal { + allow_trailing_empty: false, + ..self.split(pat).0 + }) } + /// An iterator over substrings of `self`, separated by characters + /// matched by a pattern and yielded in reverse order. + /// + /// The pattern can be a simple `&str`, [`char`], or a closure that + /// determines the split. + /// Additional libraries might provide more complex patterns like + /// regular expressions. + /// + /// Equivalent to [`split`], except that the trailing substring is + /// skipped if empty. + /// + /// [`split`]: #method.split + /// + /// This method can be used for string data that is _terminated_, + /// rather than _separated_ by a pattern. + /// + /// # Iterator behavior + /// + /// The returned iterator requires that the pattern supports a + /// reverse search, and it will be double ended if a forward/reverse + /// search yields the same elements. + /// + /// For iterating from the front, the [`split_terminator`] method can be + /// used. + /// + /// [`split_terminator`]: #method.split_terminator + /// + /// # Examples + /// + /// ``` + /// let v: Vec<&str> = "A.B.".rsplit_terminator('.').collect(); + /// assert_eq!(v, ["B", "A"]); + /// + /// let v: Vec<&str> = "A..B..".rsplit_terminator(".").collect(); + /// assert_eq!(v, ["", "B", "", "A"]); + /// ``` + #[stable(feature = "rust1", since = "1.0.0")] #[inline] - fn starts_with<'a, P: Pattern<'a>>(&'a self, pat: P) -> bool { - pat.is_prefix_of(self) + pub fn rsplit_terminator<'a, P: Pattern<'a>>(&'a self, pat: P) -> RSplitTerminator<'a, P> + where P::Searcher: ReverseSearcher<'a> + { + RSplitTerminator(self.split_terminator(pat).0) } + /// An iterator over substrings of the given string slice, separated by a + /// pattern, restricted to returning at most `n` items. + /// + /// If `n` substrings are returned, the last substring (the `n`th substring) + /// will contain the remainder of the string. + /// + /// The pattern can be a `&str`, [`char`], or a closure that determines the + /// split. + /// + /// # Iterator behavior + /// + /// The returned iterator will not be double ended, because it is + /// not efficient to support. + /// + /// If the pattern allows a reverse search, the [`rsplitn`] method can be + /// used. + /// + /// [`rsplitn`]: #method.rsplitn + /// + /// # Examples + /// + /// Simple patterns: + /// + /// ``` + /// let v: Vec<&str> = "Mary had a little lambda".splitn(3, ' ').collect(); + /// assert_eq!(v, ["Mary", "had", "a little lambda"]); + /// + /// let v: Vec<&str> = "lionXXtigerXleopard".splitn(3, "X").collect(); + /// assert_eq!(v, ["lion", "", "tigerXleopard"]); + /// + /// let v: Vec<&str> = "abcXdef".splitn(1, 'X').collect(); + /// assert_eq!(v, ["abcXdef"]); + /// + /// let v: Vec<&str> = "".splitn(1, 'X').collect(); + /// assert_eq!(v, [""]); + /// ``` + /// + /// A more complex pattern, using a closure: + /// + /// ``` + /// let v: Vec<&str> = "abc1defXghi".splitn(2, |c| c == '1' || c == 'X').collect(); + /// assert_eq!(v, ["abc", "defXghi"]); + /// ``` + #[stable(feature = "rust1", since = "1.0.0")] #[inline] - fn ends_with<'a, P: Pattern<'a>>(&'a self, pat: P) -> bool + pub fn splitn<'a, P: Pattern<'a>>(&'a self, n: usize, pat: P) -> SplitN<'a, P> { + SplitN(SplitNInternal { + iter: self.split(pat).0, + count: n, + }) + } + + /// An iterator over substrings of this string slice, separated by a + /// pattern, starting from the end of the string, restricted to returning + /// at most `n` items. + /// + /// If `n` substrings are returned, the last substring (the `n`th substring) + /// will contain the remainder of the string. + /// + /// The pattern can be a `&str`, [`char`], or a closure that + /// determines the split. + /// + /// # Iterator behavior + /// + /// The returned iterator will not be double ended, because it is not + /// efficient to support. + /// + /// For splitting from the front, the [`splitn`] method can be used. + /// + /// [`splitn`]: #method.splitn + /// + /// # Examples + /// + /// Simple patterns: + /// + /// ``` + /// let v: Vec<&str> = "Mary had a little lamb".rsplitn(3, ' ').collect(); + /// assert_eq!(v, ["lamb", "little", "Mary had a"]); + /// + /// let v: Vec<&str> = "lionXXtigerXleopard".rsplitn(3, 'X').collect(); + /// assert_eq!(v, ["leopard", "tiger", "lionX"]); + /// + /// let v: Vec<&str> = "lion::tiger::leopard".rsplitn(2, "::").collect(); + /// assert_eq!(v, ["leopard", "lion::tiger"]); + /// ``` + /// + /// A more complex pattern, using a closure: + /// + /// ``` + /// let v: Vec<&str> = "abc1defXghi".rsplitn(2, |c| c == '1' || c == 'X').collect(); + /// assert_eq!(v, ["ghi", "abc1def"]); + /// ``` + #[stable(feature = "rust1", since = "1.0.0")] + #[inline] + pub fn rsplitn<'a, P: Pattern<'a>>(&'a self, n: usize, pat: P) -> RSplitN<'a, P> where P::Searcher: ReverseSearcher<'a> { - pat.is_suffix_of(self) + RSplitN(self.splitn(n, pat).0) + } + + /// An iterator over the disjoint matches of a pattern within the given string + /// slice. + /// + /// The pattern can be a `&str`, [`char`], or a closure that + /// determines if a character matches. + /// + /// # Iterator behavior + /// + /// The returned iterator will be a [`DoubleEndedIterator`] if the pattern + /// allows a reverse search and forward/reverse search yields the same + /// elements. This is true for, eg, [`char`] but not for `&str`. + /// + /// [`DoubleEndedIterator`]: iter/trait.DoubleEndedIterator.html + /// + /// If the pattern allows a reverse search but its results might differ + /// from a forward search, the [`rmatches`] method can be used. + /// + /// [`rmatches`]: #method.rmatches + /// + /// # Examples + /// + /// Basic usage: + /// + /// ``` + /// let v: Vec<&str> = "abcXXXabcYYYabc".matches("abc").collect(); + /// assert_eq!(v, ["abc", "abc", "abc"]); + /// + /// let v: Vec<&str> = "1abc2abc3".matches(char::is_numeric).collect(); + /// assert_eq!(v, ["1", "2", "3"]); + /// ``` + #[stable(feature = "str_matches", since = "1.2.0")] + #[inline] + pub fn matches<'a, P: Pattern<'a>>(&'a self, pat: P) -> Matches<'a, P> { + Matches(MatchesInternal(pat.into_searcher(self))) + } + + /// An iterator over the disjoint matches of a pattern within this string slice, + /// yielded in reverse order. + /// + /// The pattern can be a `&str`, [`char`], or a closure that determines if + /// a character matches. + /// + /// # Iterator behavior + /// + /// The returned iterator requires that the pattern supports a reverse + /// search, and it will be a [`DoubleEndedIterator`] if a forward/reverse + /// search yields the same elements. + /// + /// [`DoubleEndedIterator`]: iter/trait.DoubleEndedIterator.html + /// + /// For iterating from the front, the [`matches`] method can be used. + /// + /// [`matches`]: #method.matches + /// + /// # Examples + /// + /// Basic usage: + /// + /// ``` + /// let v: Vec<&str> = "abcXXXabcYYYabc".rmatches("abc").collect(); + /// assert_eq!(v, ["abc", "abc", "abc"]); + /// + /// let v: Vec<&str> = "1abc2abc3".rmatches(char::is_numeric).collect(); + /// assert_eq!(v, ["3", "2", "1"]); + /// ``` + #[stable(feature = "str_matches", since = "1.2.0")] + #[inline] + pub fn rmatches<'a, P: Pattern<'a>>(&'a self, pat: P) -> RMatches<'a, P> + where P::Searcher: ReverseSearcher<'a> + { + RMatches(self.matches(pat).0) + } + + /// An iterator over the disjoint matches of a pattern within this string + /// slice as well as the index that the match starts at. + /// + /// For matches of `pat` within `self` that overlap, only the indices + /// corresponding to the first match are returned. + /// + /// The pattern can be a `&str`, [`char`], or a closure that determines + /// if a character matches. + /// + /// # Iterator behavior + /// + /// The returned iterator will be a [`DoubleEndedIterator`] if the pattern + /// allows a reverse search and forward/reverse search yields the same + /// elements. This is true for, eg, [`char`] but not for `&str`. + /// + /// [`DoubleEndedIterator`]: iter/trait.DoubleEndedIterator.html + /// + /// If the pattern allows a reverse search but its results might differ + /// from a forward search, the [`rmatch_indices`] method can be used. + /// + /// [`rmatch_indices`]: #method.rmatch_indices + /// + /// # Examples + /// + /// Basic usage: + /// + /// ``` + /// let v: Vec<_> = "abcXXXabcYYYabc".match_indices("abc").collect(); + /// assert_eq!(v, [(0, "abc"), (6, "abc"), (12, "abc")]); + /// + /// let v: Vec<_> = "1abcabc2".match_indices("abc").collect(); + /// assert_eq!(v, [(1, "abc"), (4, "abc")]); + /// + /// let v: Vec<_> = "ababa".match_indices("aba").collect(); + /// assert_eq!(v, [(0, "aba")]); // only the first `aba` + /// ``` + #[stable(feature = "str_match_indices", since = "1.5.0")] + #[inline] + pub fn match_indices<'a, P: Pattern<'a>>(&'a self, pat: P) -> MatchIndices<'a, P> { + MatchIndices(MatchIndicesInternal(pat.into_searcher(self))) } + /// An iterator over the disjoint matches of a pattern within `self`, + /// yielded in reverse order along with the index of the match. + /// + /// For matches of `pat` within `self` that overlap, only the indices + /// corresponding to the last match are returned. + /// + /// The pattern can be a `&str`, [`char`], or a closure that determines if a + /// character matches. + /// + /// # Iterator behavior + /// + /// The returned iterator requires that the pattern supports a reverse + /// search, and it will be a [`DoubleEndedIterator`] if a forward/reverse + /// search yields the same elements. + /// + /// [`DoubleEndedIterator`]: iter/trait.DoubleEndedIterator.html + /// + /// For iterating from the front, the [`match_indices`] method can be used. + /// + /// [`match_indices`]: #method.match_indices + /// + /// # Examples + /// + /// Basic usage: + /// + /// ``` + /// let v: Vec<_> = "abcXXXabcYYYabc".rmatch_indices("abc").collect(); + /// assert_eq!(v, [(12, "abc"), (6, "abc"), (0, "abc")]); + /// + /// let v: Vec<_> = "1abcabc2".rmatch_indices("abc").collect(); + /// assert_eq!(v, [(4, "abc"), (1, "abc")]); + /// + /// let v: Vec<_> = "ababa".rmatch_indices("aba").collect(); + /// assert_eq!(v, [(2, "aba")]); // only the last `aba` + /// ``` + #[stable(feature = "str_match_indices", since = "1.5.0")] #[inline] - fn trim_matches<'a, P: Pattern<'a>>(&'a self, pat: P) -> &'a str + pub fn rmatch_indices<'a, P: Pattern<'a>>(&'a self, pat: P) -> RMatchIndices<'a, P> + where P::Searcher: ReverseSearcher<'a> + { + RMatchIndices(self.match_indices(pat).0) + } + + /// Returns a string slice with leading and trailing whitespace removed. + /// + /// 'Whitespace' is defined according to the terms of the Unicode Derived + /// Core Property `White_Space`. + /// + /// # Examples + /// + /// Basic usage: + /// + /// ``` + /// let s = " Hello\tworld\t"; + /// + /// assert_eq!("Hello\tworld", s.trim()); + /// ``` + #[stable(feature = "rust1", since = "1.0.0")] + pub fn trim(&self) -> &str { + self.trim_matches(|c: char| c.is_whitespace()) + } + + /// Returns a string slice with leading whitespace removed. + /// + /// 'Whitespace' is defined according to the terms of the Unicode Derived + /// Core Property `White_Space`. + /// + /// # Text directionality + /// + /// A string is a sequence of bytes. 'Left' in this context means the first + /// position of that byte string; for a language like Arabic or Hebrew + /// which are 'right to left' rather than 'left to right', this will be + /// the _right_ side, not the left. + /// + /// # Examples + /// + /// Basic usage: + /// + /// ``` + /// let s = " Hello\tworld\t"; + /// + /// assert_eq!("Hello\tworld\t", s.trim_left()); + /// ``` + /// + /// Directionality: + /// + /// ``` + /// let s = " English"; + /// assert!(Some('E') == s.trim_left().chars().next()); + /// + /// let s = " עברית"; + /// assert!(Some('ע') == s.trim_left().chars().next()); + /// ``` + #[stable(feature = "rust1", since = "1.0.0")] + pub fn trim_left(&self) -> &str { + self.trim_left_matches(|c: char| c.is_whitespace()) + } + + /// Returns a string slice with trailing whitespace removed. + /// + /// 'Whitespace' is defined according to the terms of the Unicode Derived + /// Core Property `White_Space`. + /// + /// # Text directionality + /// + /// A string is a sequence of bytes. 'Right' in this context means the last + /// position of that byte string; for a language like Arabic or Hebrew + /// which are 'right to left' rather than 'left to right', this will be + /// the _left_ side, not the right. + /// + /// # Examples + /// + /// Basic usage: + /// + /// ``` + /// let s = " Hello\tworld\t"; + /// + /// assert_eq!(" Hello\tworld", s.trim_right()); + /// ``` + /// + /// Directionality: + /// + /// ``` + /// let s = "English "; + /// assert!(Some('h') == s.trim_right().chars().rev().next()); + /// + /// let s = "עברית "; + /// assert!(Some('ת') == s.trim_right().chars().rev().next()); + /// ``` + #[stable(feature = "rust1", since = "1.0.0")] + pub fn trim_right(&self) -> &str { + self.trim_right_matches(|c: char| c.is_whitespace()) + } + + /// Returns a string slice with all prefixes and suffixes that match a + /// pattern repeatedly removed. + /// + /// The pattern can be a [`char`] or a closure that determines if a + /// character matches. + /// + /// # Examples + /// + /// Simple patterns: + /// + /// ``` + /// assert_eq!("11foo1bar11".trim_matches('1'), "foo1bar"); + /// assert_eq!("123foo1bar123".trim_matches(char::is_numeric), "foo1bar"); + /// + /// let x: &[_] = &['1', '2']; + /// assert_eq!("12foo1bar12".trim_matches(x), "foo1bar"); + /// ``` + /// + /// A more complex pattern, using a closure: + /// + /// ``` + /// assert_eq!("1foo1barXX".trim_matches(|c| c == '1' || c == 'X'), "foo1bar"); + /// ``` + #[stable(feature = "rust1", since = "1.0.0")] + pub fn trim_matches<'a, P: Pattern<'a>>(&'a self, pat: P) -> &'a str where P::Searcher: DoubleEndedSearcher<'a> { let mut i = 0; @@ -2402,12 +3675,36 @@ impl StrExt for str { } unsafe { // Searcher is known to return valid indices - self.slice_unchecked(i, j) + self.get_unchecked(i..j) } } - #[inline] - fn trim_left_matches<'a, P: Pattern<'a>>(&'a self, pat: P) -> &'a str { + /// Returns a string slice with all prefixes that match a pattern + /// repeatedly removed. + /// + /// The pattern can be a `&str`, [`char`], or a closure that determines if + /// a character matches. + /// + /// # Text directionality + /// + /// A string is a sequence of bytes. 'Left' in this context means the first + /// position of that byte string; for a language like Arabic or Hebrew + /// which are 'right to left' rather than 'left to right', this will be + /// the _right_ side, not the left. + /// + /// # Examples + /// + /// Basic usage: + /// + /// ``` + /// assert_eq!("11foo1bar11".trim_left_matches('1'), "foo1bar11"); + /// assert_eq!("123foo1bar123".trim_left_matches(char::is_numeric), "foo1bar123"); + /// + /// let x: &[_] = &['1', '2']; + /// assert_eq!("12foo1bar12".trim_left_matches(x), "foo1bar12"); + /// ``` + #[stable(feature = "rust1", since = "1.0.0")] + pub fn trim_left_matches<'a, P: Pattern<'a>>(&'a self, pat: P) -> &'a str { let mut i = self.len(); let mut matcher = pat.into_searcher(self); if let Some((a, _)) = matcher.next_reject() { @@ -2415,12 +3712,42 @@ impl StrExt for str { } unsafe { // Searcher is known to return valid indices - self.slice_unchecked(i, self.len()) + self.get_unchecked(i..self.len()) } } - #[inline] - fn trim_right_matches<'a, P: Pattern<'a>>(&'a self, pat: P) -> &'a str + /// Returns a string slice with all suffixes that match a pattern + /// repeatedly removed. + /// + /// The pattern can be a `&str`, [`char`], or a closure that + /// determines if a character matches. + /// + /// # Text directionality + /// + /// A string is a sequence of bytes. 'Right' in this context means the last + /// position of that byte string; for a language like Arabic or Hebrew + /// which are 'right to left' rather than 'left to right', this will be + /// the _left_ side, not the right. + /// + /// # Examples + /// + /// Simple patterns: + /// + /// ``` + /// assert_eq!("11foo1bar11".trim_right_matches('1'), "11foo1bar"); + /// assert_eq!("123foo1bar123".trim_right_matches(char::is_numeric), "123foo1bar"); + /// + /// let x: &[_] = &['1', '2']; + /// assert_eq!("12foo1bar12".trim_right_matches(x), "12foo1bar"); + /// ``` + /// + /// A more complex pattern, using a closure: + /// + /// ``` + /// assert_eq!("1fooX".trim_right_matches(|c| c == '1' || c == 'X'), "1foo"); + /// ``` + #[stable(feature = "rust1", since = "1.0.0")] + pub fn trim_right_matches<'a, P: Pattern<'a>>(&'a self, pat: P) -> &'a str where P::Searcher: ReverseSearcher<'a> { let mut j = 0; @@ -2430,101 +3757,376 @@ impl StrExt for str { } unsafe { // Searcher is known to return valid indices - self.slice_unchecked(0, j) + self.get_unchecked(0..j) } } + /// Parses this string slice into another type. + /// + /// Because `parse` is so general, it can cause problems with type + /// inference. As such, `parse` is one of the few times you'll see + /// the syntax affectionately known as the 'turbofish': `::<>`. This + /// helps the inference algorithm understand specifically which type + /// you're trying to parse into. + /// + /// `parse` can parse any type that implements the [`FromStr`] trait. + /// + /// [`FromStr`]: str/trait.FromStr.html + /// + /// # Errors + /// + /// Will return [`Err`] if it's not possible to parse this string slice into + /// the desired type. + /// + /// [`Err`]: str/trait.FromStr.html#associatedtype.Err + /// + /// # Examples + /// + /// Basic usage + /// + /// ``` + /// let four: u32 = "4".parse().unwrap(); + /// + /// assert_eq!(4, four); + /// ``` + /// + /// Using the 'turbofish' instead of annotating `four`: + /// + /// ``` + /// let four = "4".parse::<u32>(); + /// + /// assert_eq!(Ok(4), four); + /// ``` + /// + /// Failing to parse: + /// + /// ``` + /// let nope = "j".parse::<u32>(); + /// + /// assert!(nope.is_err()); + /// ``` #[inline] - fn is_char_boundary(&self, index: usize) -> bool { - // 0 and len are always ok. - // Test for 0 explicitly so that it can optimize out the check - // easily and skip reading string data for that case. - if index == 0 || index == self.len() { return true; } - match self.as_bytes().get(index) { - None => false, - // This is bit magic equivalent to: b < 128 || b >= 192 - Some(&b) => (b as i8) >= -0x40, - } + #[stable(feature = "rust1", since = "1.0.0")] + pub fn parse<F: FromStr>(&self) -> Result<F, F::Err> { + FromStr::from_str(self) } + /// Checks if all characters in this string are within the ASCII range. + /// + /// # Examples + /// + /// ``` + /// let ascii = "hello!\n"; + /// let non_ascii = "Grüße, Jürgen ❤"; + /// + /// assert!(ascii.is_ascii()); + /// assert!(!non_ascii.is_ascii()); + /// ``` + #[stable(feature = "ascii_methods_on_intrinsics", since = "1.23.0")] #[inline] - fn as_bytes(&self) -> &[u8] { - unsafe { &*(self as *const str as *const [u8]) } + pub fn is_ascii(&self) -> bool { + // We can treat each byte as character here: all multibyte characters + // start with a byte that is not in the ascii range, so we will stop + // there already. + self.bytes().all(|b| b.is_ascii()) } + /// Checks that two strings are an ASCII case-insensitive match. + /// + /// Same as `to_ascii_lowercase(a) == to_ascii_lowercase(b)`, + /// but without allocating and copying temporaries. + /// + /// # Examples + /// + /// ``` + /// assert!("Ferris".eq_ignore_ascii_case("FERRIS")); + /// assert!("Ferrös".eq_ignore_ascii_case("FERRöS")); + /// assert!(!"Ferrös".eq_ignore_ascii_case("FERRÖS")); + /// ``` + #[stable(feature = "ascii_methods_on_intrinsics", since = "1.23.0")] #[inline] - unsafe fn as_bytes_mut(&mut self) -> &mut [u8] { - &mut *(self as *mut str as *mut [u8]) + pub fn eq_ignore_ascii_case(&self, other: &str) -> bool { + self.as_bytes().eq_ignore_ascii_case(other.as_bytes()) } - fn find<'a, P: Pattern<'a>>(&'a self, pat: P) -> Option<usize> { - pat.into_searcher(self).next_match().map(|(i, _)| i) + /// Converts this string to its ASCII upper case equivalent in-place. + /// + /// ASCII letters 'a' to 'z' are mapped to 'A' to 'Z', + /// but non-ASCII letters are unchanged. + /// + /// To return a new uppercased value without modifying the existing one, use + /// [`to_ascii_uppercase`]. + /// + /// [`to_ascii_uppercase`]: #method.to_ascii_uppercase + #[stable(feature = "ascii_methods_on_intrinsics", since = "1.23.0")] + pub fn make_ascii_uppercase(&mut self) { + let me = unsafe { self.as_bytes_mut() }; + me.make_ascii_uppercase() } - fn rfind<'a, P: Pattern<'a>>(&'a self, pat: P) -> Option<usize> - where P::Searcher: ReverseSearcher<'a> - { - pat.into_searcher(self).next_match_back().map(|(i, _)| i) + /// Converts this string to its ASCII lower case equivalent in-place. + /// + /// ASCII letters 'A' to 'Z' are mapped to 'a' to 'z', + /// but non-ASCII letters are unchanged. + /// + /// To return a new lowercased value without modifying the existing one, use + /// [`to_ascii_lowercase`]. + /// + /// [`to_ascii_lowercase`]: #method.to_ascii_lowercase + #[stable(feature = "ascii_methods_on_intrinsics", since = "1.23.0")] + pub fn make_ascii_lowercase(&mut self) { + let me = unsafe { self.as_bytes_mut() }; + me.make_ascii_lowercase() + } +} + +#[stable(feature = "rust1", since = "1.0.0")] +impl AsRef<[u8]> for str { + #[inline] + fn as_ref(&self) -> &[u8] { + self.as_bytes() + } +} + +#[stable(feature = "rust1", since = "1.0.0")] +impl<'a> Default for &'a str { + /// Creates an empty str + fn default() -> &'a str { "" } +} + +#[stable(feature = "default_mut_str", since = "1.28.0")] +impl<'a> Default for &'a mut str { + /// Creates an empty mutable str + fn default() -> &'a mut str { unsafe { from_utf8_unchecked_mut(&mut []) } } +} + +/// An iterator over the non-whitespace substrings of a string, +/// separated by any amount of whitespace. +/// +/// This struct is created by the [`split_whitespace`] method on [`str`]. +/// See its documentation for more. +/// +/// [`split_whitespace`]: ../../std/primitive.str.html#method.split_whitespace +/// [`str`]: ../../std/primitive.str.html +#[stable(feature = "split_whitespace", since = "1.1.0")] +#[derive(Clone, Debug)] +pub struct SplitWhitespace<'a> { + inner: Filter<Split<'a, IsWhitespace>, IsNotEmpty>, +} + +/// An iterator over the non-ASCII-whitespace substrings of a string, +/// separated by any amount of ASCII whitespace. +/// +/// This struct is created by the [`split_ascii_whitespace`] method on [`str`]. +/// See its documentation for more. +/// +/// [`split_ascii_whitespace`]: ../../std/primitive.str.html#method.split_ascii_whitespace +/// [`str`]: ../../std/primitive.str.html +#[unstable(feature = "split_ascii_whitespace", issue = "48656")] +#[derive(Clone, Debug)] +pub struct SplitAsciiWhitespace<'a> { + inner: Map<Filter<SliceSplit<'a, u8, IsAsciiWhitespace>, IsNotEmpty>, UnsafeBytesToStr>, +} + +#[derive(Clone)] +struct IsWhitespace; + +impl FnOnce<(char, )> for IsWhitespace { + type Output = bool; + + #[inline] + extern "rust-call" fn call_once(mut self, arg: (char, )) -> bool { + self.call_mut(arg) } +} - fn find_str<'a, P: Pattern<'a>>(&'a self, pat: P) -> Option<usize> { - self.find(pat) +impl FnMut<(char, )> for IsWhitespace { + #[inline] + extern "rust-call" fn call_mut(&mut self, arg: (char, )) -> bool { + arg.0.is_whitespace() } +} + +#[derive(Clone)] +struct IsAsciiWhitespace; + +impl<'a> FnOnce<(&'a u8, )> for IsAsciiWhitespace { + type Output = bool; #[inline] - fn split_at(&self, mid: usize) -> (&str, &str) { - // is_char_boundary checks that the index is in [0, .len()] - if self.is_char_boundary(mid) { - unsafe { - (self.slice_unchecked(0, mid), - self.slice_unchecked(mid, self.len())) - } - } else { - slice_error_fail(self, 0, mid) - } + extern "rust-call" fn call_once(mut self, arg: (&u8, )) -> bool { + self.call_mut(arg) } +} - fn split_at_mut(&mut self, mid: usize) -> (&mut str, &mut str) { - // is_char_boundary checks that the index is in [0, .len()] - if self.is_char_boundary(mid) { - let len = self.len(); - let ptr = self.as_ptr() as *mut u8; - unsafe { - (from_raw_parts_mut(ptr, mid), - from_raw_parts_mut(ptr.offset(mid as isize), len - mid)) - } - } else { - slice_error_fail(self, 0, mid) - } +impl<'a> FnMut<(&'a u8, )> for IsAsciiWhitespace { + #[inline] + extern "rust-call" fn call_mut(&mut self, arg: (&u8, )) -> bool { + arg.0.is_ascii_whitespace() } +} + +#[derive(Clone)] +struct IsNotEmpty; + +impl<'a, 'b> FnOnce<(&'a &'b str, )> for IsNotEmpty { + type Output = bool; #[inline] - fn as_ptr(&self) -> *const u8 { - self as *const str as *const u8 + extern "rust-call" fn call_once(mut self, arg: (&'a &'b str, )) -> bool { + self.call_mut(arg) } +} +impl<'a, 'b> FnMut<(&'a &'b str, )> for IsNotEmpty { #[inline] - fn len(&self) -> usize { - self.as_bytes().len() + extern "rust-call" fn call_mut(&mut self, arg: (&'a &'b str, )) -> bool { + !arg.0.is_empty() } +} + +impl<'a, 'b> FnOnce<(&'a &'b [u8], )> for IsNotEmpty { + type Output = bool; #[inline] - fn is_empty(&self) -> bool { self.len() == 0 } + extern "rust-call" fn call_once(mut self, arg: (&'a &'b [u8], )) -> bool { + self.call_mut(arg) + } +} +impl<'a, 'b> FnMut<(&'a &'b [u8], )> for IsNotEmpty { #[inline] - fn parse<T: FromStr>(&self) -> Result<T, T::Err> { FromStr::from_str(self) } + extern "rust-call" fn call_mut(&mut self, arg: (&'a &'b [u8], )) -> bool { + !arg.0.is_empty() + } } -#[stable(feature = "rust1", since = "1.0.0")] -impl AsRef<[u8]> for str { +#[derive(Clone)] +struct UnsafeBytesToStr; + +impl<'a> FnOnce<(&'a [u8], )> for UnsafeBytesToStr { + type Output = &'a str; + #[inline] - fn as_ref(&self) -> &[u8] { - self.as_bytes() + extern "rust-call" fn call_once(mut self, arg: (&'a [u8], )) -> &'a str { + self.call_mut(arg) } } -#[stable(feature = "rust1", since = "1.0.0")] -impl<'a> Default for &'a str { - /// Creates an empty str - fn default() -> &'a str { "" } +impl<'a> FnMut<(&'a [u8], )> for UnsafeBytesToStr { + #[inline] + extern "rust-call" fn call_mut(&mut self, arg: (&'a [u8], )) -> &'a str { + unsafe { from_utf8_unchecked(arg.0) } + } +} + + +#[stable(feature = "split_whitespace", since = "1.1.0")] +impl<'a> Iterator for SplitWhitespace<'a> { + type Item = &'a str; + + #[inline] + fn next(&mut self) -> Option<&'a str> { + self.inner.next() + } + + #[inline] + fn size_hint(&self) -> (usize, Option<usize>) { + self.inner.size_hint() + } } + +#[stable(feature = "split_whitespace", since = "1.1.0")] +impl<'a> DoubleEndedIterator for SplitWhitespace<'a> { + #[inline] + fn next_back(&mut self) -> Option<&'a str> { + self.inner.next_back() + } +} + +#[stable(feature = "fused", since = "1.26.0")] +impl<'a> FusedIterator for SplitWhitespace<'a> {} + +#[unstable(feature = "split_ascii_whitespace", issue = "48656")] +impl<'a> Iterator for SplitAsciiWhitespace<'a> { + type Item = &'a str; + + #[inline] + fn next(&mut self) -> Option<&'a str> { + self.inner.next() + } + + #[inline] + fn size_hint(&self) -> (usize, Option<usize>) { + self.inner.size_hint() + } +} + +#[unstable(feature = "split_ascii_whitespace", issue = "48656")] +impl<'a> DoubleEndedIterator for SplitAsciiWhitespace<'a> { + #[inline] + fn next_back(&mut self) -> Option<&'a str> { + self.inner.next_back() + } +} + +#[unstable(feature = "split_ascii_whitespace", issue = "48656")] +impl<'a> FusedIterator for SplitAsciiWhitespace<'a> {} + +/// An iterator of [`u16`] over the string encoded as UTF-16. +/// +/// [`u16`]: ../../std/primitive.u16.html +/// +/// This struct is created by the [`encode_utf16`] method on [`str`]. +/// See its documentation for more. +/// +/// [`encode_utf16`]: ../../std/primitive.str.html#method.encode_utf16 +/// [`str`]: ../../std/primitive.str.html +#[derive(Clone)] +#[stable(feature = "encode_utf16", since = "1.8.0")] +pub struct EncodeUtf16<'a> { + chars: Chars<'a>, + extra: u16, +} + +#[stable(feature = "collection_debug", since = "1.17.0")] +impl<'a> fmt::Debug for EncodeUtf16<'a> { + fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result { + f.pad("EncodeUtf16 { .. }") + } +} + +#[stable(feature = "encode_utf16", since = "1.8.0")] +impl<'a> Iterator for EncodeUtf16<'a> { + type Item = u16; + + #[inline] + fn next(&mut self) -> Option<u16> { + if self.extra != 0 { + let tmp = self.extra; + self.extra = 0; + return Some(tmp); + } + + let mut buf = [0; 2]; + self.chars.next().map(|ch| { + let n = ch.encode_utf16(&mut buf).len(); + if n == 2 { + self.extra = buf[1]; + } + buf[0] + }) + } + + #[inline] + fn size_hint(&self) -> (usize, Option<usize>) { + let (low, high) = self.chars.size_hint(); + // every char gets either one u16 or two u16, + // so this iterator is between 1 or 2 times as + // long as the underlying iterator. + (low, high.and_then(|n| n.checked_mul(2))) + } +} + +#[stable(feature = "fused", since = "1.26.0")] +impl<'a> FusedIterator for EncodeUtf16<'a> {} diff --git a/src/libcore/str/pattern.rs b/src/libcore/str/pattern.rs index 089d691773a..5e63fa9ff35 100644 --- a/src/libcore/str/pattern.rs +++ b/src/libcore/str/pattern.rs @@ -258,7 +258,7 @@ pub struct CharSearcher<'a> { /// `finger` is the current byte index of the forward search. /// Imagine that it exists before the byte at its index, i.e. - /// haystack[finger] is the first byte of the slice we must inspect during + /// `haystack[finger]` is the first byte of the slice we must inspect during /// forward searching finger: usize, /// `finger_back` is the current byte index of the reverse search. @@ -324,7 +324,7 @@ unsafe impl<'a> Searcher<'a> for CharSearcher<'a> { // the second byte when searching for the third. // // However, this is totally okay. While we have the invariant that - // self.finger is on a UTF8 boundary, this invariant is not relid upon + // self.finger is on a UTF8 boundary, this invariant is not relied upon // within this method (it is relied upon in CharSearcher::next()). // // We only exit this method when we reach the end of the string, or if we @@ -354,7 +354,7 @@ unsafe impl<'a> ReverseSearcher<'a> for CharSearcher<'a> { #[inline] fn next_back(&mut self) -> SearchStep { let old_finger = self.finger_back; - let slice = unsafe { self.haystack.slice_unchecked(self.finger, old_finger) }; + let slice = unsafe { self.haystack.get_unchecked(self.finger..old_finger) }; let mut iter = slice.chars(); let old_len = iter.iter.len(); if let Some(ch) = iter.next_back() { diff --git a/src/libcore/sync/atomic.rs b/src/libcore/sync/atomic.rs index 3da9e9c87dd..617e067e078 100644 --- a/src/libcore/sync/atomic.rs +++ b/src/libcore/sync/atomic.rs @@ -29,7 +29,7 @@ //! //! [`Ordering`]: enum.Ordering.html //! -//! [1]: http://llvm.org/docs/LangRef.html#memory-model-for-concurrent-operations +//! [1]: https://llvm.org/docs/LangRef.html#memory-model-for-concurrent-operations //! [2]: ../../../nomicon/atomics.html //! //! Atomic variables are safe to share between threads (they implement [`Sync`]) @@ -97,9 +97,10 @@ use fmt; /// Save power or switch hyperthreads in a busy-wait spin-loop. /// /// This function is deliberately more primitive than -/// `std::thread::yield_now` and does not directly yield to the -/// system's scheduler. In some cases it might be useful to use a -/// combination of both functions. Careful benchmarking is advised. +/// [`std::thread::yield_now`](../../../std/thread/fn.yield_now.html) and +/// does not directly yield to the system's scheduler. +/// In some cases it might be useful to use a combination of both functions. +/// Careful benchmarking is advised. /// /// On some platforms this function may not do anything at all. #[inline] @@ -123,6 +124,7 @@ pub fn spin_loop_hint() { /// [`bool`]: ../../../std/primitive.bool.html #[cfg(target_has_atomic = "8")] #[stable(feature = "rust1", since = "1.0.0")] +#[repr(transparent)] pub struct AtomicBool { v: UnsafeCell<u8>, } @@ -146,6 +148,7 @@ unsafe impl Sync for AtomicBool {} /// This type has the same in-memory representation as a `*mut T`. #[cfg(target_has_atomic = "ptr")] #[stable(feature = "rust1", since = "1.0.0")] +#[repr(transparent)] pub struct AtomicPtr<T> { p: UnsafeCell<*mut T>, } @@ -175,50 +178,85 @@ unsafe impl<T> Sync for AtomicPtr<T> {} /// "relaxed" atomics allow all reorderings. /// /// Rust's memory orderings are [the same as -/// LLVM's](http://llvm.org/docs/LangRef.html#memory-model-for-concurrent-operations). +/// LLVM's](https://llvm.org/docs/LangRef.html#memory-model-for-concurrent-operations). /// /// For more information see the [nomicon]. /// /// [nomicon]: ../../../nomicon/atomics.html #[stable(feature = "rust1", since = "1.0.0")] #[derive(Copy, Clone, Debug)] +#[non_exhaustive] pub enum Ordering { /// No ordering constraints, only atomic operations. /// /// Corresponds to LLVM's [`Monotonic`] ordering. /// - /// [`Monotonic`]: http://llvm.org/docs/Atomics.html#monotonic + /// [`Monotonic`]: https://llvm.org/docs/Atomics.html#monotonic #[stable(feature = "rust1", since = "1.0.0")] Relaxed, - /// When coupled with a store, all previous writes become visible - /// to the other threads that perform a load with [`Acquire`] ordering - /// on the same value. + /// When coupled with a store, all previous operations become ordered + /// before any load of this value with [`Acquire`] (or stronger) ordering. + /// In particular, all previous writes become visible to all threads + /// that perform an [`Acquire`] (or stronger) load of this value. /// - /// [`Acquire`]: http://llvm.org/docs/Atomics.html#acquire + /// Notice that using this ordering for an operation that combines loads + /// and stores leads to a [`Relaxed`] load operation! + /// + /// This ordering is only applicable for operations that can perform a store. + /// + /// Corresponds to LLVM's [`Release`] ordering. + /// + /// [`Release`]: https://llvm.org/docs/Atomics.html#release + /// [`Acquire`]: https://llvm.org/docs/Atomics.html#acquire + /// [`Relaxed`]: https://llvm.org/docs/Atomics.html#monotonic #[stable(feature = "rust1", since = "1.0.0")] Release, - /// When coupled with a load, all subsequent loads will see data - /// written before a store with [`Release`] ordering on the same value - /// in other threads. + /// When coupled with a load, if the loaded value was written by a store operation with + /// [`Release`] (or stronger) ordering, then all subsequent operations + /// become ordered after that store. In particular, all subsequent loads will see data + /// written before the store. + /// + /// Notice that using this ordering for an operation that combines loads + /// and stores leads to a [`Relaxed`] store operation! /// - /// [`Release`]: http://llvm.org/docs/Atomics.html#release + /// This ordering is only applicable for operations that can perform a load. + /// + /// Corresponds to LLVM's [`Acquire`] ordering. + /// + /// [`Acquire`]: https://llvm.org/docs/Atomics.html#acquire + /// [`Release`]: https://llvm.org/docs/Atomics.html#release + /// [`Relaxed`]: https://llvm.org/docs/Atomics.html#monotonic #[stable(feature = "rust1", since = "1.0.0")] Acquire, - /// When coupled with a load, uses [`Acquire`] ordering, and with a store - /// [`Release`] ordering. + /// Has the effects of both [`Acquire`] and [`Release`] together: + /// For loads it uses [`Acquire`] ordering. For stores it uses the [`Release`] ordering. + /// + /// Notice that in the case of `compare_and_swap`, it is possible that the operation ends up + /// not performing any store and hence it has just `Acquire` ordering. However, + /// `AcqRel` will never perform [`Relaxed`] accesses. + /// + /// This ordering is only applicable for operations that combine both loads and stores. /// - /// [`Acquire`]: http://llvm.org/docs/Atomics.html#acquire - /// [`Release`]: http://llvm.org/docs/Atomics.html#release + /// Corresponds to LLVM's [`AcquireRelease`] ordering. + /// + /// [`AcquireRelease`]: https://llvm.org/docs/Atomics.html#acquirerelease + /// [`Acquire`]: https://llvm.org/docs/Atomics.html#acquire + /// [`Release`]: https://llvm.org/docs/Atomics.html#release + /// [`Relaxed`]: https://llvm.org/docs/Atomics.html#monotonic #[stable(feature = "rust1", since = "1.0.0")] AcqRel, - /// Like `AcqRel` with the additional guarantee that all threads see all + /// Like [`Acquire`]/[`Release`]/[`AcqRel`] (for load, store, and load-with-store + /// operations, respectively) with the additional guarantee that all threads see all /// sequentially consistent operations in the same order. + /// + /// Corresponds to LLVM's [`SequentiallyConsistent`] ordering. + /// + /// [`SequentiallyConsistent`]: https://llvm.org/docs/Atomics.html#sequentiallyconsistent + /// [`Acquire`]: https://llvm.org/docs/Atomics.html#acquire + /// [`Release`]: https://llvm.org/docs/Atomics.html#release + /// [`AcqRel`]: https://llvm.org/docs/Atomics.html#acquirerelease #[stable(feature = "rust1", since = "1.0.0")] SeqCst, - // Prevent exhaustive matching to allow for future extension - #[doc(hidden)] - #[unstable(feature = "future_atomic_orderings", issue = "0")] - __Nonexhaustive, } /// An [`AtomicBool`] initialized to `false`. @@ -285,21 +323,24 @@ impl AtomicBool { #[inline] #[stable(feature = "atomic_access", since = "1.15.0")] pub fn into_inner(self) -> bool { - unsafe { self.v.into_inner() != 0 } + self.v.into_inner() != 0 } /// Loads a value from the bool. /// /// `load` takes an [`Ordering`] argument which describes the memory ordering - /// of this operation. + /// of this operation. Possible values are [`SeqCst`], [`Acquire`] and [`Relaxed`]. /// /// # Panics /// /// Panics if `order` is [`Release`] or [`AcqRel`]. /// /// [`Ordering`]: enum.Ordering.html + /// [`Relaxed`]: enum.Ordering.html#variant.Relaxed /// [`Release`]: enum.Ordering.html#variant.Release + /// [`Acquire`]: enum.Ordering.html#variant.Acquire /// [`AcqRel`]: enum.Ordering.html#variant.AcqRel + /// [`SeqCst`]: enum.Ordering.html#variant.SeqCst /// /// # Examples /// @@ -319,9 +360,18 @@ impl AtomicBool { /// Stores a value into the bool. /// /// `store` takes an [`Ordering`] argument which describes the memory ordering - /// of this operation. + /// of this operation. Possible values are [`SeqCst`], [`Release`] and [`Relaxed`]. + /// + /// # Panics + /// + /// Panics if `order` is [`Acquire`] or [`AcqRel`]. /// /// [`Ordering`]: enum.Ordering.html + /// [`Relaxed`]: enum.Ordering.html#variant.Relaxed + /// [`Release`]: enum.Ordering.html#variant.Release + /// [`Acquire`]: enum.Ordering.html#variant.Acquire + /// [`AcqRel`]: enum.Ordering.html#variant.AcqRel + /// [`SeqCst`]: enum.Ordering.html#variant.SeqCst /// /// # Examples /// @@ -333,13 +383,6 @@ impl AtomicBool { /// some_bool.store(false, Ordering::Relaxed); /// assert_eq!(some_bool.load(Ordering::Relaxed), false); /// ``` - /// - /// # Panics - /// - /// Panics if `order` is [`Acquire`] or [`AcqRel`]. - /// - /// [`Acquire`]: enum.Ordering.html#variant.Acquire - /// [`AcqRel`]: enum.Ordering.html#variant.AcqRel #[inline] #[stable(feature = "rust1", since = "1.0.0")] pub fn store(&self, val: bool, order: Ordering) { @@ -351,9 +394,14 @@ impl AtomicBool { /// Stores a value into the bool, returning the previous value. /// /// `swap` takes an [`Ordering`] argument which describes the memory ordering - /// of this operation. + /// of this operation. All ordering modes are possible. Note that using + /// [`Acquire`] makes the store part of this operation [`Relaxed`], and + /// using [`Release`] makes the load part [`Relaxed`]. /// /// [`Ordering`]: enum.Ordering.html + /// [`Relaxed`]: enum.Ordering.html#variant.Relaxed + /// [`Release`]: enum.Ordering.html#variant.Release + /// [`Acquire`]: enum.Ordering.html#variant.Acquire /// /// # Examples /// @@ -367,6 +415,7 @@ impl AtomicBool { /// ``` #[inline] #[stable(feature = "rust1", since = "1.0.0")] + #[cfg(target_has_atomic = "cas")] pub fn swap(&self, val: bool, order: Ordering) -> bool { unsafe { atomic_swap(self.v.get(), val as u8, order) != 0 } } @@ -377,9 +426,16 @@ impl AtomicBool { /// was updated. /// /// `compare_and_swap` also takes an [`Ordering`] argument which describes the memory - /// ordering of this operation. + /// ordering of this operation. Notice that even when using [`AcqRel`], the operation + /// might fail and hence just perform an `Acquire` load, but not have `Release` semantics. + /// Using [`Acquire`] makes the store part of this operation [`Relaxed`] if it + /// happens, and using [`Release`] makes the load part [`Relaxed`]. /// /// [`Ordering`]: enum.Ordering.html + /// [`Relaxed`]: enum.Ordering.html#variant.Relaxed + /// [`Release`]: enum.Ordering.html#variant.Release + /// [`Acquire`]: enum.Ordering.html#variant.Acquire + /// [`AcqRel`]: enum.Ordering.html#variant.AcqRel /// [`bool`]: ../../../std/primitive.bool.html /// /// # Examples @@ -397,6 +453,7 @@ impl AtomicBool { /// ``` #[inline] #[stable(feature = "rust1", since = "1.0.0")] + #[cfg(target_has_atomic = "cas")] pub fn compare_and_swap(&self, current: bool, new: bool, order: Ordering) -> bool { match self.compare_exchange(current, new, order, strongest_failure_ordering(order)) { Ok(x) => x, @@ -412,13 +469,18 @@ impl AtomicBool { /// `compare_exchange` takes two [`Ordering`] arguments to describe the memory /// ordering of this operation. The first describes the required ordering if the /// operation succeeds while the second describes the required ordering when the - /// operation fails. The failure ordering can't be [`Release`] or [`AcqRel`] and must - /// be equivalent or weaker than the success ordering. + /// operation fails. Using [`Acquire`] as success ordering makes the store part + /// of this operation [`Relaxed`], and using [`Release`] makes the successful load + /// [`Relaxed`]. The failure ordering can only be [`SeqCst`], [`Acquire`] or [`Relaxed`] + /// and must be equivalent to or weaker than the success ordering. + /// /// /// [`bool`]: ../../../std/primitive.bool.html /// [`Ordering`]: enum.Ordering.html + /// [`Relaxed`]: enum.Ordering.html#variant.Relaxed /// [`Release`]: enum.Ordering.html#variant.Release - /// [`AcqRel`]: enum.Ordering.html#variant.Release + /// [`Acquire`]: enum.Ordering.html#variant.Acquire + /// [`SeqCst`]: enum.Ordering.html#variant.SeqCst /// /// # Examples /// @@ -442,6 +504,7 @@ impl AtomicBool { /// ``` #[inline] #[stable(feature = "extended_compare_and_swap", since = "1.10.0")] + #[cfg(target_has_atomic = "cas")] pub fn compare_exchange(&self, current: bool, new: bool, @@ -464,16 +527,20 @@ impl AtomicBool { /// previous value. /// /// `compare_exchange_weak` takes two [`Ordering`] arguments to describe the memory - /// ordering of this operation. The first describes the required ordering if the operation - /// succeeds while the second describes the required ordering when the operation fails. The - /// failure ordering can't be [`Release`] or [`AcqRel`] and must be equivalent or - /// weaker than the success ordering. + /// ordering of this operation. The first describes the required ordering if the + /// operation succeeds while the second describes the required ordering when the + /// operation fails. Using [`Acquire`] as success ordering makes the store part + /// of this operation [`Relaxed`], and using [`Release`] makes the successful load + /// [`Relaxed`]. The failure ordering can only be [`SeqCst`], [`Acquire`] or [`Relaxed`] + /// and must be equivalent to or weaker than the success ordering. /// /// [`bool`]: ../../../std/primitive.bool.html /// [`compare_exchange`]: #method.compare_exchange /// [`Ordering`]: enum.Ordering.html + /// [`Relaxed`]: enum.Ordering.html#variant.Relaxed /// [`Release`]: enum.Ordering.html#variant.Release - /// [`AcqRel`]: enum.Ordering.html#variant.Release + /// [`Acquire`]: enum.Ordering.html#variant.Acquire + /// [`SeqCst`]: enum.Ordering.html#variant.SeqCst /// /// # Examples /// @@ -514,6 +581,16 @@ impl AtomicBool { /// /// Returns the previous value. /// + /// `fetch_and` takes an [`Ordering`] argument which describes the memory ordering + /// of this operation. All ordering modes are possible. Note that using + /// [`Acquire`] makes the store part of this operation [`Relaxed`], and + /// using [`Release`] makes the load part [`Relaxed`]. + /// + /// [`Ordering`]: enum.Ordering.html + /// [`Relaxed`]: enum.Ordering.html#variant.Relaxed + /// [`Release`]: enum.Ordering.html#variant.Release + /// [`Acquire`]: enum.Ordering.html#variant.Acquire + /// /// # Examples /// /// ``` @@ -533,6 +610,7 @@ impl AtomicBool { /// ``` #[inline] #[stable(feature = "rust1", since = "1.0.0")] + #[cfg(target_has_atomic = "cas")] pub fn fetch_and(&self, val: bool, order: Ordering) -> bool { unsafe { atomic_and(self.v.get(), val as u8, order) != 0 } } @@ -544,6 +622,16 @@ impl AtomicBool { /// /// Returns the previous value. /// + /// `fetch_nand` takes an [`Ordering`] argument which describes the memory ordering + /// of this operation. All ordering modes are possible. Note that using + /// [`Acquire`] makes the store part of this operation [`Relaxed`], and + /// using [`Release`] makes the load part [`Relaxed`]. + /// + /// [`Ordering`]: enum.Ordering.html + /// [`Relaxed`]: enum.Ordering.html#variant.Relaxed + /// [`Release`]: enum.Ordering.html#variant.Release + /// [`Acquire`]: enum.Ordering.html#variant.Acquire + /// /// # Examples /// /// ``` @@ -564,6 +652,7 @@ impl AtomicBool { /// ``` #[inline] #[stable(feature = "rust1", since = "1.0.0")] + #[cfg(target_has_atomic = "cas")] pub fn fetch_nand(&self, val: bool, order: Ordering) -> bool { // We can't use atomic_nand here because it can result in a bool with // an invalid value. This happens because the atomic operation is done @@ -587,6 +676,16 @@ impl AtomicBool { /// /// Returns the previous value. /// + /// `fetch_or` takes an [`Ordering`] argument which describes the memory ordering + /// of this operation. All ordering modes are possible. Note that using + /// [`Acquire`] makes the store part of this operation [`Relaxed`], and + /// using [`Release`] makes the load part [`Relaxed`]. + /// + /// [`Ordering`]: enum.Ordering.html + /// [`Relaxed`]: enum.Ordering.html#variant.Relaxed + /// [`Release`]: enum.Ordering.html#variant.Release + /// [`Acquire`]: enum.Ordering.html#variant.Acquire + /// /// # Examples /// /// ``` @@ -606,6 +705,7 @@ impl AtomicBool { /// ``` #[inline] #[stable(feature = "rust1", since = "1.0.0")] + #[cfg(target_has_atomic = "cas")] pub fn fetch_or(&self, val: bool, order: Ordering) -> bool { unsafe { atomic_or(self.v.get(), val as u8, order) != 0 } } @@ -617,6 +717,16 @@ impl AtomicBool { /// /// Returns the previous value. /// + /// `fetch_xor` takes an [`Ordering`] argument which describes the memory ordering + /// of this operation. All ordering modes are possible. Note that using + /// [`Acquire`] makes the store part of this operation [`Relaxed`], and + /// using [`Release`] makes the load part [`Relaxed`]. + /// + /// [`Ordering`]: enum.Ordering.html + /// [`Relaxed`]: enum.Ordering.html#variant.Relaxed + /// [`Release`]: enum.Ordering.html#variant.Release + /// [`Acquire`]: enum.Ordering.html#variant.Acquire + /// /// # Examples /// /// ``` @@ -636,6 +746,7 @@ impl AtomicBool { /// ``` #[inline] #[stable(feature = "rust1", since = "1.0.0")] + #[cfg(target_has_atomic = "cas")] pub fn fetch_xor(&self, val: bool, order: Ordering) -> bool { unsafe { atomic_xor(self.v.get(), val as u8, order) != 0 } } @@ -695,21 +806,24 @@ impl<T> AtomicPtr<T> { #[inline] #[stable(feature = "atomic_access", since = "1.15.0")] pub fn into_inner(self) -> *mut T { - unsafe { self.p.into_inner() } + self.p.into_inner() } /// Loads a value from the pointer. /// /// `load` takes an [`Ordering`] argument which describes the memory ordering - /// of this operation. + /// of this operation. Possible values are [`SeqCst`], [`Acquire`] and [`Relaxed`]. /// /// # Panics /// /// Panics if `order` is [`Release`] or [`AcqRel`]. /// /// [`Ordering`]: enum.Ordering.html + /// [`Relaxed`]: enum.Ordering.html#variant.Relaxed /// [`Release`]: enum.Ordering.html#variant.Release + /// [`Acquire`]: enum.Ordering.html#variant.Acquire /// [`AcqRel`]: enum.Ordering.html#variant.AcqRel + /// [`SeqCst`]: enum.Ordering.html#variant.SeqCst /// /// # Examples /// @@ -730,9 +844,18 @@ impl<T> AtomicPtr<T> { /// Stores a value into the pointer. /// /// `store` takes an [`Ordering`] argument which describes the memory ordering - /// of this operation. + /// of this operation. Possible values are [`SeqCst`], [`Release`] and [`Relaxed`]. + /// + /// # Panics + /// + /// Panics if `order` is [`Acquire`] or [`AcqRel`]. /// /// [`Ordering`]: enum.Ordering.html + /// [`Relaxed`]: enum.Ordering.html#variant.Relaxed + /// [`Release`]: enum.Ordering.html#variant.Release + /// [`Acquire`]: enum.Ordering.html#variant.Acquire + /// [`AcqRel`]: enum.Ordering.html#variant.AcqRel + /// [`SeqCst`]: enum.Ordering.html#variant.SeqCst /// /// # Examples /// @@ -746,13 +869,6 @@ impl<T> AtomicPtr<T> { /// /// some_ptr.store(other_ptr, Ordering::Relaxed); /// ``` - /// - /// # Panics - /// - /// Panics if `order` is [`Acquire`] or [`AcqRel`]. - /// - /// [`Acquire`]: enum.Ordering.html#variant.Acquire - /// [`AcqRel`]: enum.Ordering.html#variant.AcqRel #[inline] #[stable(feature = "rust1", since = "1.0.0")] pub fn store(&self, ptr: *mut T, order: Ordering) { @@ -764,9 +880,14 @@ impl<T> AtomicPtr<T> { /// Stores a value into the pointer, returning the previous value. /// /// `swap` takes an [`Ordering`] argument which describes the memory ordering - /// of this operation. + /// of this operation. All ordering modes are possible. Note that using + /// [`Acquire`] makes the store part of this operation [`Relaxed`], and + /// using [`Release`] makes the load part [`Relaxed`]. /// /// [`Ordering`]: enum.Ordering.html + /// [`Relaxed`]: enum.Ordering.html#variant.Relaxed + /// [`Release`]: enum.Ordering.html#variant.Release + /// [`Acquire`]: enum.Ordering.html#variant.Acquire /// /// # Examples /// @@ -782,6 +903,7 @@ impl<T> AtomicPtr<T> { /// ``` #[inline] #[stable(feature = "rust1", since = "1.0.0")] + #[cfg(target_has_atomic = "cas")] pub fn swap(&self, ptr: *mut T, order: Ordering) -> *mut T { unsafe { atomic_swap(self.p.get() as *mut usize, ptr as usize, order) as *mut T } } @@ -792,9 +914,16 @@ impl<T> AtomicPtr<T> { /// was updated. /// /// `compare_and_swap` also takes an [`Ordering`] argument which describes the memory - /// ordering of this operation. + /// ordering of this operation. Notice that even when using [`AcqRel`], the operation + /// might fail and hence just perform an `Acquire` load, but not have `Release` semantics. + /// Using [`Acquire`] makes the store part of this operation [`Relaxed`] if it + /// happens, and using [`Release`] makes the load part [`Relaxed`]. /// /// [`Ordering`]: enum.Ordering.html + /// [`Relaxed`]: enum.Ordering.html#variant.Relaxed + /// [`Release`]: enum.Ordering.html#variant.Release + /// [`Acquire`]: enum.Ordering.html#variant.Acquire + /// [`AcqRel`]: enum.Ordering.html#variant.AcqRel /// /// # Examples /// @@ -811,6 +940,7 @@ impl<T> AtomicPtr<T> { /// ``` #[inline] #[stable(feature = "rust1", since = "1.0.0")] + #[cfg(target_has_atomic = "cas")] pub fn compare_and_swap(&self, current: *mut T, new: *mut T, order: Ordering) -> *mut T { match self.compare_exchange(current, new, order, strongest_failure_ordering(order)) { Ok(x) => x, @@ -824,14 +954,18 @@ impl<T> AtomicPtr<T> { /// the previous value. On success this value is guaranteed to be equal to `current`. /// /// `compare_exchange` takes two [`Ordering`] arguments to describe the memory - /// ordering of this operation. The first describes the required ordering if - /// the operation succeeds while the second describes the required ordering when - /// the operation fails. The failure ordering can't be [`Release`] or [`AcqRel`] - /// and must be equivalent or weaker than the success ordering. + /// ordering of this operation. The first describes the required ordering if the + /// operation succeeds while the second describes the required ordering when the + /// operation fails. Using [`Acquire`] as success ordering makes the store part + /// of this operation [`Relaxed`], and using [`Release`] makes the successful load + /// [`Relaxed`]. The failure ordering can only be [`SeqCst`], [`Acquire`] or [`Relaxed`] + /// and must be equivalent to or weaker than the success ordering. /// /// [`Ordering`]: enum.Ordering.html + /// [`Relaxed`]: enum.Ordering.html#variant.Relaxed /// [`Release`]: enum.Ordering.html#variant.Release - /// [`AcqRel`]: enum.Ordering.html#variant.AcqRel + /// [`Acquire`]: enum.Ordering.html#variant.Acquire + /// [`SeqCst`]: enum.Ordering.html#variant.SeqCst /// /// # Examples /// @@ -849,6 +983,7 @@ impl<T> AtomicPtr<T> { /// ``` #[inline] #[stable(feature = "extended_compare_and_swap", since = "1.10.0")] + #[cfg(target_has_atomic = "cas")] pub fn compare_exchange(&self, current: *mut T, new: *mut T, @@ -876,15 +1011,19 @@ impl<T> AtomicPtr<T> { /// previous value. /// /// `compare_exchange_weak` takes two [`Ordering`] arguments to describe the memory - /// ordering of this operation. The first describes the required ordering if the operation - /// succeeds while the second describes the required ordering when the operation fails. The - /// failure ordering can't be [`Release`] or [`AcqRel`] and must be equivalent or - /// weaker than the success ordering. + /// ordering of this operation. The first describes the required ordering if the + /// operation succeeds while the second describes the required ordering when the + /// operation fails. Using [`Acquire`] as success ordering makes the store part + /// of this operation [`Relaxed`], and using [`Release`] makes the successful load + /// [`Relaxed`]. The failure ordering can only be [`SeqCst`], [`Acquire`] or [`Relaxed`] + /// and must be equivalent to or weaker than the success ordering. /// /// [`compare_exchange`]: #method.compare_exchange /// [`Ordering`]: enum.Ordering.html + /// [`Relaxed`]: enum.Ordering.html#variant.Relaxed /// [`Release`]: enum.Ordering.html#variant.Release - /// [`AcqRel`]: enum.Ordering.html#variant.AcqRel + /// [`Acquire`]: enum.Ordering.html#variant.Acquire + /// [`SeqCst`]: enum.Ordering.html#variant.SeqCst /// /// # Examples /// @@ -945,7 +1084,10 @@ macro_rules! atomic_int { $stable_debug:meta, $stable_access:meta, $stable_from:meta, + $stable_nand:meta, $s_int_type:expr, $int_ref:expr, + $extra_feature:expr, + $min_fn:ident, $max_fn:ident, $int_type:ident $atomic_type:ident $atomic_init:ident) => { /// An integer type which can be safely shared between threads. /// @@ -957,13 +1099,9 @@ macro_rules! atomic_int { /// ). For more about the differences between atomic types and /// non-atomic types, please see the [module-level documentation]. /// - /// Please note that examples are shared between atomic variants of - /// primitive integer types, so it's normal that they are all - /// demonstrating [`AtomicIsize`]. - /// /// [module-level documentation]: index.html - /// [`AtomicIsize`]: struct.AtomicIsize.html #[$stable] + #[repr(transparent)] pub struct $atomic_type { v: UnsafeCell<$int_type>, } @@ -988,9 +1126,7 @@ macro_rules! atomic_int { #[$stable_debug] impl fmt::Debug for $atomic_type { fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result { - f.debug_tuple(stringify!($atomic_type)) - .field(&self.load(Ordering::SeqCst)) - .finish() + fmt::Debug::fmt(&self.load(Ordering::SeqCst), f) } } @@ -999,373 +1135,678 @@ macro_rules! atomic_int { unsafe impl Sync for $atomic_type {} impl $atomic_type { - /// Creates a new atomic integer. - /// - /// # Examples - /// - /// ``` - /// use std::sync::atomic::AtomicIsize; - /// - /// let atomic_forty_two = AtomicIsize::new(42); - /// ``` - #[inline] - #[$stable] - pub const fn new(v: $int_type) -> Self { - $atomic_type {v: UnsafeCell::new(v)} + doc_comment! { + concat!("Creates a new atomic integer. + +# Examples + +``` +", $extra_feature, "use std::sync::atomic::", stringify!($atomic_type), "; + +let atomic_forty_two = ", stringify!($atomic_type), "::new(42); +```"), + #[inline] + #[$stable] + pub const fn new(v: $int_type) -> Self { + $atomic_type {v: UnsafeCell::new(v)} + } } - /// Returns a mutable reference to the underlying integer. - /// - /// This is safe because the mutable reference guarantees that no other threads are - /// concurrently accessing the atomic data. - /// - /// # Examples - /// - /// ``` - /// use std::sync::atomic::{AtomicIsize, Ordering}; - /// - /// let mut some_isize = AtomicIsize::new(10); - /// assert_eq!(*some_isize.get_mut(), 10); - /// *some_isize.get_mut() = 5; - /// assert_eq!(some_isize.load(Ordering::SeqCst), 5); - /// ``` - #[inline] - #[$stable_access] - pub fn get_mut(&mut self) -> &mut $int_type { - unsafe { &mut *self.v.get() } + doc_comment! { + concat!("Returns a mutable reference to the underlying integer. + +This is safe because the mutable reference guarantees that no other threads are +concurrently accessing the atomic data. + +# Examples + +``` +", $extra_feature, "use std::sync::atomic::{", stringify!($atomic_type), ", Ordering}; + +let mut some_var = ", stringify!($atomic_type), "::new(10); +assert_eq!(*some_var.get_mut(), 10); +*some_var.get_mut() = 5; +assert_eq!(some_var.load(Ordering::SeqCst), 5); +```"), + #[inline] + #[$stable_access] + pub fn get_mut(&mut self) -> &mut $int_type { + unsafe { &mut *self.v.get() } + } } - /// Consumes the atomic and returns the contained value. - /// - /// This is safe because passing `self` by value guarantees that no other threads are - /// concurrently accessing the atomic data. - /// - /// # Examples - /// - /// ``` - /// use std::sync::atomic::AtomicIsize; - /// - /// let some_isize = AtomicIsize::new(5); - /// assert_eq!(some_isize.into_inner(), 5); - /// ``` - #[inline] - #[$stable_access] - pub fn into_inner(self) -> $int_type { - unsafe { self.v.into_inner() } + doc_comment! { + concat!("Consumes the atomic and returns the contained value. + +This is safe because passing `self` by value guarantees that no other threads are +concurrently accessing the atomic data. + +# Examples + +``` +", $extra_feature, "use std::sync::atomic::", stringify!($atomic_type), "; + +let some_var = ", stringify!($atomic_type), "::new(5); +assert_eq!(some_var.into_inner(), 5); +```"), + #[inline] + #[$stable_access] + pub fn into_inner(self) -> $int_type { + self.v.into_inner() + } } - /// Loads a value from the atomic integer. - /// - /// `load` takes an [`Ordering`] argument which describes the memory ordering of this - /// operation. - /// - /// # Panics - /// - /// Panics if `order` is [`Release`] or [`AcqRel`]. - /// - /// [`Ordering`]: enum.Ordering.html - /// [`Release`]: enum.Ordering.html#variant.Release - /// [`AcqRel`]: enum.Ordering.html#variant.AcqRel - /// - /// # Examples - /// - /// ``` - /// use std::sync::atomic::{AtomicIsize, Ordering}; - /// - /// let some_isize = AtomicIsize::new(5); - /// - /// assert_eq!(some_isize.load(Ordering::Relaxed), 5); - /// ``` - #[inline] - #[$stable] - pub fn load(&self, order: Ordering) -> $int_type { - unsafe { atomic_load(self.v.get(), order) } + doc_comment! { + concat!("Loads a value from the atomic integer. + +`load` takes an [`Ordering`] argument which describes the memory ordering of this operation. +Possible values are [`SeqCst`], [`Acquire`] and [`Relaxed`]. + +# Panics + +Panics if `order` is [`Release`] or [`AcqRel`]. + +[`Ordering`]: enum.Ordering.html +[`Relaxed`]: enum.Ordering.html#variant.Relaxed +[`Release`]: enum.Ordering.html#variant.Release +[`Acquire`]: enum.Ordering.html#variant.Acquire +[`AcqRel`]: enum.Ordering.html#variant.AcqRel +[`SeqCst`]: enum.Ordering.html#variant.SeqCst + +# Examples + +``` +", $extra_feature, "use std::sync::atomic::{", stringify!($atomic_type), ", Ordering}; + +let some_var = ", stringify!($atomic_type), "::new(5); + +assert_eq!(some_var.load(Ordering::Relaxed), 5); +```"), + #[inline] + #[$stable] + pub fn load(&self, order: Ordering) -> $int_type { + unsafe { atomic_load(self.v.get(), order) } + } } - /// Stores a value into the atomic integer. - /// - /// `store` takes an [`Ordering`] argument which describes the memory ordering of this - /// operation. - /// - /// [`Ordering`]: enum.Ordering.html - /// - /// # Examples - /// - /// ``` - /// use std::sync::atomic::{AtomicIsize, Ordering}; - /// - /// let some_isize = AtomicIsize::new(5); - /// - /// some_isize.store(10, Ordering::Relaxed); - /// assert_eq!(some_isize.load(Ordering::Relaxed), 10); - /// ``` - /// - /// # Panics - /// - /// Panics if `order` is [`Acquire`] or [`AcqRel`]. - /// - /// [`Acquire`]: enum.Ordering.html#variant.Acquire - /// [`AcqRel`]: enum.Ordering.html#variant.AcqRel - #[inline] - #[$stable] - pub fn store(&self, val: $int_type, order: Ordering) { - unsafe { atomic_store(self.v.get(), val, order); } + doc_comment! { + concat!("Stores a value into the atomic integer. + +`store` takes an [`Ordering`] argument which describes the memory ordering of this operation. + Possible values are [`SeqCst`], [`Release`] and [`Relaxed`]. + +# Panics + +Panics if `order` is [`Acquire`] or [`AcqRel`]. + +[`Ordering`]: enum.Ordering.html +[`Relaxed`]: enum.Ordering.html#variant.Relaxed +[`Release`]: enum.Ordering.html#variant.Release +[`Acquire`]: enum.Ordering.html#variant.Acquire +[`AcqRel`]: enum.Ordering.html#variant.AcqRel +[`SeqCst`]: enum.Ordering.html#variant.SeqCst + +# Examples + +``` +", $extra_feature, "use std::sync::atomic::{", stringify!($atomic_type), ", Ordering}; + +let some_var = ", stringify!($atomic_type), "::new(5); + +some_var.store(10, Ordering::Relaxed); +assert_eq!(some_var.load(Ordering::Relaxed), 10); +```"), + #[inline] + #[$stable] + pub fn store(&self, val: $int_type, order: Ordering) { + unsafe { atomic_store(self.v.get(), val, order); } + } } - /// Stores a value into the atomic integer, returning the previous value. - /// - /// `swap` takes an [`Ordering`] argument which describes the memory ordering of this - /// operation. - /// - /// [`Ordering`]: enum.Ordering.html - /// - /// # Examples - /// - /// ``` - /// use std::sync::atomic::{AtomicIsize, Ordering}; - /// - /// let some_isize = AtomicIsize::new(5); - /// - /// assert_eq!(some_isize.swap(10, Ordering::Relaxed), 5); - /// ``` - #[inline] - #[$stable] - pub fn swap(&self, val: $int_type, order: Ordering) -> $int_type { - unsafe { atomic_swap(self.v.get(), val, order) } + doc_comment! { + concat!("Stores a value into the atomic integer, returning the previous value. + +`swap` takes an [`Ordering`] argument which describes the memory ordering +of this operation. All ordering modes are possible. Note that using +[`Acquire`] makes the store part of this operation [`Relaxed`], and +using [`Release`] makes the load part [`Relaxed`]. + +[`Ordering`]: enum.Ordering.html +[`Relaxed`]: enum.Ordering.html#variant.Relaxed +[`Release`]: enum.Ordering.html#variant.Release +[`Acquire`]: enum.Ordering.html#variant.Acquire + +# Examples + +``` +", $extra_feature, "use std::sync::atomic::{", stringify!($atomic_type), ", Ordering}; + +let some_var = ", stringify!($atomic_type), "::new(5); + +assert_eq!(some_var.swap(10, Ordering::Relaxed), 5); +```"), + #[inline] + #[$stable] + #[cfg(target_has_atomic = "cas")] + pub fn swap(&self, val: $int_type, order: Ordering) -> $int_type { + unsafe { atomic_swap(self.v.get(), val, order) } + } } - /// Stores a value into the atomic integer if the current value is the same as the - /// `current` value. - /// - /// The return value is always the previous value. If it is equal to `current`, then the - /// value was updated. - /// - /// `compare_and_swap` also takes an [`Ordering`] argument which describes the memory - /// ordering of this operation. - /// - /// [`Ordering`]: enum.Ordering.html - /// - /// # Examples - /// - /// ``` - /// use std::sync::atomic::{AtomicIsize, Ordering}; - /// - /// let some_isize = AtomicIsize::new(5); - /// - /// assert_eq!(some_isize.compare_and_swap(5, 10, Ordering::Relaxed), 5); - /// assert_eq!(some_isize.load(Ordering::Relaxed), 10); - /// - /// assert_eq!(some_isize.compare_and_swap(6, 12, Ordering::Relaxed), 10); - /// assert_eq!(some_isize.load(Ordering::Relaxed), 10); - /// ``` - #[inline] - #[$stable] - pub fn compare_and_swap(&self, - current: $int_type, - new: $int_type, - order: Ordering) -> $int_type { - match self.compare_exchange(current, - new, - order, - strongest_failure_ordering(order)) { - Ok(x) => x, - Err(x) => x, + doc_comment! { + concat!("Stores a value into the atomic integer if the current value is the same as +the `current` value. + +The return value is always the previous value. If it is equal to `current`, then the +value was updated. + +`compare_and_swap` also takes an [`Ordering`] argument which describes the memory +ordering of this operation. Notice that even when using [`AcqRel`], the operation +might fail and hence just perform an `Acquire` load, but not have `Release` semantics. +Using [`Acquire`] makes the store part of this operation [`Relaxed`] if it +happens, and using [`Release`] makes the load part [`Relaxed`]. + +[`Ordering`]: enum.Ordering.html +[`Relaxed`]: enum.Ordering.html#variant.Relaxed +[`Release`]: enum.Ordering.html#variant.Release +[`Acquire`]: enum.Ordering.html#variant.Acquire +[`AcqRel`]: enum.Ordering.html#variant.AcqRel + +# Examples + +``` +", $extra_feature, "use std::sync::atomic::{", stringify!($atomic_type), ", Ordering}; + +let some_var = ", stringify!($atomic_type), "::new(5); + +assert_eq!(some_var.compare_and_swap(5, 10, Ordering::Relaxed), 5); +assert_eq!(some_var.load(Ordering::Relaxed), 10); + +assert_eq!(some_var.compare_and_swap(6, 12, Ordering::Relaxed), 10); +assert_eq!(some_var.load(Ordering::Relaxed), 10); +```"), + #[inline] + #[$stable] + #[cfg(target_has_atomic = "cas")] + pub fn compare_and_swap(&self, + current: $int_type, + new: $int_type, + order: Ordering) -> $int_type { + match self.compare_exchange(current, + new, + order, + strongest_failure_ordering(order)) { + Ok(x) => x, + Err(x) => x, + } } } - /// Stores a value into the atomic integer if the current value is the same as the - /// `current` value. - /// - /// The return value is a result indicating whether the new value was written and - /// containing the previous value. On success this value is guaranteed to be equal to - /// `current`. - /// - /// `compare_exchange` takes two [`Ordering`] arguments to describe the memory - /// ordering of this operation. The first describes the required ordering if - /// the operation succeeds while the second describes the required ordering when - /// the operation fails. The failure ordering can't be [`Release`] or [`AcqRel`] and - /// must be equivalent or weaker than the success ordering. - /// - /// [`Ordering`]: enum.Ordering.html - /// [`Release`]: enum.Ordering.html#variant.Release - /// [`AcqRel`]: enum.Ordering.html#variant.AcqRel - /// - /// # Examples - /// - /// ``` - /// use std::sync::atomic::{AtomicIsize, Ordering}; - /// - /// let some_isize = AtomicIsize::new(5); - /// - /// assert_eq!(some_isize.compare_exchange(5, 10, - /// Ordering::Acquire, - /// Ordering::Relaxed), - /// Ok(5)); - /// assert_eq!(some_isize.load(Ordering::Relaxed), 10); - /// - /// assert_eq!(some_isize.compare_exchange(6, 12, - /// Ordering::SeqCst, - /// Ordering::Acquire), - /// Err(10)); - /// assert_eq!(some_isize.load(Ordering::Relaxed), 10); - /// ``` - #[inline] - #[$stable_cxchg] - pub fn compare_exchange(&self, - current: $int_type, - new: $int_type, - success: Ordering, - failure: Ordering) -> Result<$int_type, $int_type> { - unsafe { atomic_compare_exchange(self.v.get(), current, new, success, failure) } + doc_comment! { + concat!("Stores a value into the atomic integer if the current value is the same as +the `current` value. + +The return value is a result indicating whether the new value was written and +containing the previous value. On success this value is guaranteed to be equal to +`current`. + +`compare_exchange` takes two [`Ordering`] arguments to describe the memory +ordering of this operation. The first describes the required ordering if the +operation succeeds while the second describes the required ordering when the +operation fails. Using [`Acquire`] as success ordering makes the store part +of this operation [`Relaxed`], and using [`Release`] makes the successful load +[`Relaxed`]. The failure ordering can only be [`SeqCst`], [`Acquire`] or [`Relaxed`] +and must be equivalent to or weaker than the success ordering. + +[`Ordering`]: enum.Ordering.html +[`Relaxed`]: enum.Ordering.html#variant.Relaxed +[`Release`]: enum.Ordering.html#variant.Release +[`Acquire`]: enum.Ordering.html#variant.Acquire +[`SeqCst`]: enum.Ordering.html#variant.SeqCst + +# Examples + +``` +", $extra_feature, "use std::sync::atomic::{", stringify!($atomic_type), ", Ordering}; + +let some_var = ", stringify!($atomic_type), "::new(5); + +assert_eq!(some_var.compare_exchange(5, 10, + Ordering::Acquire, + Ordering::Relaxed), + Ok(5)); +assert_eq!(some_var.load(Ordering::Relaxed), 10); + +assert_eq!(some_var.compare_exchange(6, 12, + Ordering::SeqCst, + Ordering::Acquire), + Err(10)); +assert_eq!(some_var.load(Ordering::Relaxed), 10); +```"), + #[inline] + #[$stable_cxchg] + #[cfg(target_has_atomic = "cas")] + pub fn compare_exchange(&self, + current: $int_type, + new: $int_type, + success: Ordering, + failure: Ordering) -> Result<$int_type, $int_type> { + unsafe { atomic_compare_exchange(self.v.get(), current, new, success, failure) } + } } - /// Stores a value into the atomic integer if the current value is the same as the - /// `current` value. - /// - /// Unlike [`compare_exchange`], this function is allowed to spuriously fail even - /// when the comparison succeeds, which can result in more efficient code on some - /// platforms. The return value is a result indicating whether the new value was - /// written and containing the previous value. - /// - /// `compare_exchange_weak` takes two [`Ordering`] arguments to describe the memory - /// ordering of this operation. The first describes the required ordering if the - /// operation succeeds while the second describes the required ordering when the - /// operation fails. The failure ordering can't be [`Release`] or [`AcqRel`] and - /// must be equivalent or weaker than the success ordering. - /// - /// [`compare_exchange`]: #method.compare_exchange - /// [`Ordering`]: enum.Ordering.html - /// [`Release`]: enum.Ordering.html#variant.Release - /// [`AcqRel`]: enum.Ordering.html#variant.AcqRel - /// - /// # Examples - /// - /// ``` - /// use std::sync::atomic::{AtomicIsize, Ordering}; - /// - /// let val = AtomicIsize::new(4); - /// - /// let mut old = val.load(Ordering::Relaxed); - /// loop { - /// let new = old * 2; - /// match val.compare_exchange_weak(old, new, Ordering::SeqCst, Ordering::Relaxed) { - /// Ok(_) => break, - /// Err(x) => old = x, - /// } - /// } - /// ``` - #[inline] - #[$stable_cxchg] - pub fn compare_exchange_weak(&self, - current: $int_type, - new: $int_type, - success: Ordering, - failure: Ordering) -> Result<$int_type, $int_type> { - unsafe { - atomic_compare_exchange_weak(self.v.get(), current, new, success, failure) + doc_comment! { + concat!("Stores a value into the atomic integer if the current value is the same as +the `current` value. + +Unlike [`compare_exchange`], this function is allowed to spuriously fail even +when the comparison succeeds, which can result in more efficient code on some +platforms. The return value is a result indicating whether the new value was +written and containing the previous value. + +`compare_exchange_weak` takes two [`Ordering`] arguments to describe the memory +ordering of this operation. The first describes the required ordering if the +operation succeeds while the second describes the required ordering when the +operation fails. Using [`Acquire`] as success ordering makes the store part +of this operation [`Relaxed`], and using [`Release`] makes the successful load +[`Relaxed`]. The failure ordering can only be [`SeqCst`], [`Acquire`] or [`Relaxed`] +and must be equivalent to or weaker than the success ordering. + +[`compare_exchange`]: #method.compare_exchange +[`Ordering`]: enum.Ordering.html +[`Relaxed`]: enum.Ordering.html#variant.Relaxed +[`Release`]: enum.Ordering.html#variant.Release +[`Acquire`]: enum.Ordering.html#variant.Acquire +[`SeqCst`]: enum.Ordering.html#variant.SeqCst + +# Examples + +``` +", $extra_feature, "use std::sync::atomic::{", stringify!($atomic_type), ", Ordering}; + +let val = ", stringify!($atomic_type), "::new(4); + +let mut old = val.load(Ordering::Relaxed); +loop { + let new = old * 2; + match val.compare_exchange_weak(old, new, Ordering::SeqCst, Ordering::Relaxed) { + Ok(_) => break, + Err(x) => old = x, + } +} +```"), + #[inline] + #[$stable_cxchg] + pub fn compare_exchange_weak(&self, + current: $int_type, + new: $int_type, + success: Ordering, + failure: Ordering) -> Result<$int_type, $int_type> { + unsafe { + atomic_compare_exchange_weak(self.v.get(), current, new, success, failure) + } } } - /// Adds to the current value, returning the previous value. - /// - /// This operation wraps around on overflow. - /// - /// # Examples - /// - /// ``` - /// use std::sync::atomic::{AtomicIsize, Ordering}; - /// - /// let foo = AtomicIsize::new(0); - /// assert_eq!(foo.fetch_add(10, Ordering::SeqCst), 0); - /// assert_eq!(foo.load(Ordering::SeqCst), 10); - /// ``` - #[inline] - #[$stable] - pub fn fetch_add(&self, val: $int_type, order: Ordering) -> $int_type { - unsafe { atomic_add(self.v.get(), val, order) } + doc_comment! { + concat!("Adds to the current value, returning the previous value. + +This operation wraps around on overflow. + +`fetch_add` takes an [`Ordering`] argument which describes the memory ordering +of this operation. All ordering modes are possible. Note that using +[`Acquire`] makes the store part of this operation [`Relaxed`], and +using [`Release`] makes the load part [`Relaxed`]. + +[`Ordering`]: enum.Ordering.html +[`Relaxed`]: enum.Ordering.html#variant.Relaxed +[`Release`]: enum.Ordering.html#variant.Release +[`Acquire`]: enum.Ordering.html#variant.Acquire + +# Examples + +``` +", $extra_feature, "use std::sync::atomic::{", stringify!($atomic_type), ", Ordering}; + +let foo = ", stringify!($atomic_type), "::new(0); +assert_eq!(foo.fetch_add(10, Ordering::SeqCst), 0); +assert_eq!(foo.load(Ordering::SeqCst), 10); +```"), + #[inline] + #[$stable] + pub fn fetch_add(&self, val: $int_type, order: Ordering) -> $int_type { + unsafe { atomic_add(self.v.get(), val, order) } + } } - /// Subtracts from the current value, returning the previous value. - /// - /// This operation wraps around on overflow. - /// - /// # Examples - /// - /// ``` - /// use std::sync::atomic::{AtomicIsize, Ordering}; - /// - /// let foo = AtomicIsize::new(0); - /// assert_eq!(foo.fetch_sub(10, Ordering::SeqCst), 0); - /// assert_eq!(foo.load(Ordering::SeqCst), -10); - /// ``` - #[inline] - #[$stable] - pub fn fetch_sub(&self, val: $int_type, order: Ordering) -> $int_type { - unsafe { atomic_sub(self.v.get(), val, order) } + doc_comment! { + concat!("Subtracts from the current value, returning the previous value. + +This operation wraps around on overflow. + +`fetch_sub` takes an [`Ordering`] argument which describes the memory ordering +of this operation. All ordering modes are possible. Note that using +[`Acquire`] makes the store part of this operation [`Relaxed`], and +using [`Release`] makes the load part [`Relaxed`]. + +[`Ordering`]: enum.Ordering.html +[`Relaxed`]: enum.Ordering.html#variant.Relaxed +[`Release`]: enum.Ordering.html#variant.Release +[`Acquire`]: enum.Ordering.html#variant.Acquire + +# Examples + +``` +", $extra_feature, "use std::sync::atomic::{", stringify!($atomic_type), ", Ordering}; + +let foo = ", stringify!($atomic_type), "::new(20); +assert_eq!(foo.fetch_sub(10, Ordering::SeqCst), 20); +assert_eq!(foo.load(Ordering::SeqCst), 10); +```"), + #[inline] + #[$stable] + pub fn fetch_sub(&self, val: $int_type, order: Ordering) -> $int_type { + unsafe { atomic_sub(self.v.get(), val, order) } + } } - /// Bitwise "and" with the current value. - /// - /// Performs a bitwise "and" operation on the current value and the argument `val`, and - /// sets the new value to the result. - /// - /// Returns the previous value. - /// - /// # Examples - /// - /// ``` - /// use std::sync::atomic::{AtomicIsize, Ordering}; - /// - /// let foo = AtomicIsize::new(0b101101); - /// assert_eq!(foo.fetch_and(0b110011, Ordering::SeqCst), 0b101101); - /// assert_eq!(foo.load(Ordering::SeqCst), 0b100001); - #[inline] - #[$stable] - pub fn fetch_and(&self, val: $int_type, order: Ordering) -> $int_type { - unsafe { atomic_and(self.v.get(), val, order) } + doc_comment! { + concat!("Bitwise \"and\" with the current value. + +Performs a bitwise \"and\" operation on the current value and the argument `val`, and +sets the new value to the result. + +Returns the previous value. + +`fetch_and` takes an [`Ordering`] argument which describes the memory ordering +of this operation. All ordering modes are possible. Note that using +[`Acquire`] makes the store part of this operation [`Relaxed`], and +using [`Release`] makes the load part [`Relaxed`]. + +[`Ordering`]: enum.Ordering.html +[`Relaxed`]: enum.Ordering.html#variant.Relaxed +[`Release`]: enum.Ordering.html#variant.Release +[`Acquire`]: enum.Ordering.html#variant.Acquire + +# Examples + +``` +", $extra_feature, "use std::sync::atomic::{", stringify!($atomic_type), ", Ordering}; + +let foo = ", stringify!($atomic_type), "::new(0b101101); +assert_eq!(foo.fetch_and(0b110011, Ordering::SeqCst), 0b101101); +assert_eq!(foo.load(Ordering::SeqCst), 0b100001); +```"), + #[inline] + #[$stable] + pub fn fetch_and(&self, val: $int_type, order: Ordering) -> $int_type { + unsafe { atomic_and(self.v.get(), val, order) } + } } - /// Bitwise "or" with the current value. - /// - /// Performs a bitwise "or" operation on the current value and the argument `val`, and - /// sets the new value to the result. - /// - /// Returns the previous value. - /// - /// # Examples - /// - /// ``` - /// use std::sync::atomic::{AtomicIsize, Ordering}; - /// - /// let foo = AtomicIsize::new(0b101101); - /// assert_eq!(foo.fetch_or(0b110011, Ordering::SeqCst), 0b101101); - /// assert_eq!(foo.load(Ordering::SeqCst), 0b111111); - #[inline] - #[$stable] - pub fn fetch_or(&self, val: $int_type, order: Ordering) -> $int_type { - unsafe { atomic_or(self.v.get(), val, order) } + doc_comment! { + concat!("Bitwise \"nand\" with the current value. + +Performs a bitwise \"nand\" operation on the current value and the argument `val`, and +sets the new value to the result. + +Returns the previous value. + +`fetch_nand` takes an [`Ordering`] argument which describes the memory ordering +of this operation. All ordering modes are possible. Note that using +[`Acquire`] makes the store part of this operation [`Relaxed`], and +using [`Release`] makes the load part [`Relaxed`]. + +[`Ordering`]: enum.Ordering.html +[`Relaxed`]: enum.Ordering.html#variant.Relaxed +[`Release`]: enum.Ordering.html#variant.Release +[`Acquire`]: enum.Ordering.html#variant.Acquire + +# Examples + +``` +", $extra_feature, " +use std::sync::atomic::{", stringify!($atomic_type), ", Ordering}; + +let foo = ", stringify!($atomic_type), "::new(0x13); +assert_eq!(foo.fetch_nand(0x31, Ordering::SeqCst), 0x13); +assert_eq!(foo.load(Ordering::SeqCst), !(0x13 & 0x31)); +```"), + #[inline] + #[$stable_nand] + pub fn fetch_nand(&self, val: $int_type, order: Ordering) -> $int_type { + unsafe { atomic_nand(self.v.get(), val, order) } + } } - /// Bitwise "xor" with the current value. - /// - /// Performs a bitwise "xor" operation on the current value and the argument `val`, and - /// sets the new value to the result. - /// - /// Returns the previous value. - /// - /// # Examples - /// - /// ``` - /// use std::sync::atomic::{AtomicIsize, Ordering}; - /// - /// let foo = AtomicIsize::new(0b101101); - /// assert_eq!(foo.fetch_xor(0b110011, Ordering::SeqCst), 0b101101); - /// assert_eq!(foo.load(Ordering::SeqCst), 0b011110); - #[inline] - #[$stable] - pub fn fetch_xor(&self, val: $int_type, order: Ordering) -> $int_type { - unsafe { atomic_xor(self.v.get(), val, order) } + doc_comment! { + concat!("Bitwise \"or\" with the current value. + +Performs a bitwise \"or\" operation on the current value and the argument `val`, and +sets the new value to the result. + +Returns the previous value. + +`fetch_or` takes an [`Ordering`] argument which describes the memory ordering +of this operation. All ordering modes are possible. Note that using +[`Acquire`] makes the store part of this operation [`Relaxed`], and +using [`Release`] makes the load part [`Relaxed`]. + +[`Ordering`]: enum.Ordering.html +[`Relaxed`]: enum.Ordering.html#variant.Relaxed +[`Release`]: enum.Ordering.html#variant.Release +[`Acquire`]: enum.Ordering.html#variant.Acquire + +# Examples + +``` +", $extra_feature, "use std::sync::atomic::{", stringify!($atomic_type), ", Ordering}; + +let foo = ", stringify!($atomic_type), "::new(0b101101); +assert_eq!(foo.fetch_or(0b110011, Ordering::SeqCst), 0b101101); +assert_eq!(foo.load(Ordering::SeqCst), 0b111111); +```"), + #[inline] + #[$stable] + pub fn fetch_or(&self, val: $int_type, order: Ordering) -> $int_type { + unsafe { atomic_or(self.v.get(), val, order) } + } } + + doc_comment! { + concat!("Bitwise \"xor\" with the current value. + +Performs a bitwise \"xor\" operation on the current value and the argument `val`, and +sets the new value to the result. + +Returns the previous value. + +`fetch_xor` takes an [`Ordering`] argument which describes the memory ordering +of this operation. All ordering modes are possible. Note that using +[`Acquire`] makes the store part of this operation [`Relaxed`], and +using [`Release`] makes the load part [`Relaxed`]. + +[`Ordering`]: enum.Ordering.html +[`Relaxed`]: enum.Ordering.html#variant.Relaxed +[`Release`]: enum.Ordering.html#variant.Release +[`Acquire`]: enum.Ordering.html#variant.Acquire + +# Examples + +``` +", $extra_feature, "use std::sync::atomic::{", stringify!($atomic_type), ", Ordering}; + +let foo = ", stringify!($atomic_type), "::new(0b101101); +assert_eq!(foo.fetch_xor(0b110011, Ordering::SeqCst), 0b101101); +assert_eq!(foo.load(Ordering::SeqCst), 0b011110); +```"), + #[inline] + #[$stable] + pub fn fetch_xor(&self, val: $int_type, order: Ordering) -> $int_type { + unsafe { atomic_xor(self.v.get(), val, order) } + } + } + + doc_comment! { + concat!("Fetches the value, and applies a function to it that returns an optional +new value. Returns a `Result` of `Ok(previous_value)` if the function returned `Some(_)`, else +`Err(previous_value)`. + +Note: This may call the function multiple times if the value has been changed from other threads in +the meantime, as long as the function returns `Some(_)`, but the function will have been applied +but once to the stored value. + +`fetch_update` takes two [`Ordering`] arguments to describe the memory +ordering of this operation. The first describes the required ordering for loads +and failed updates while the second describes the required ordering when the +operation finally succeeds. Beware that this is different from the two +modes in [`compare_exchange`]! + +Using [`Acquire`] as success ordering makes the store part +of this operation [`Relaxed`], and using [`Release`] makes the final successful load +[`Relaxed`]. The (failed) load ordering can only be [`SeqCst`], [`Acquire`] or [`Relaxed`] +and must be equivalent to or weaker than the success ordering. + +[`bool`]: ../../../std/primitive.bool.html +[`compare_exchange`]: #method.compare_exchange +[`Ordering`]: enum.Ordering.html +[`Relaxed`]: enum.Ordering.html#variant.Relaxed +[`Release`]: enum.Ordering.html#variant.Release +[`Acquire`]: enum.Ordering.html#variant.Acquire +[`SeqCst`]: enum.Ordering.html#variant.SeqCst + +# Examples + +```rust +#![feature(no_more_cas)] +", $extra_feature, "use std::sync::atomic::{", stringify!($atomic_type), ", Ordering}; + +let x = ", stringify!($atomic_type), "::new(7); +assert_eq!(x.fetch_update(|_| None, Ordering::SeqCst, Ordering::SeqCst), Err(7)); +assert_eq!(x.fetch_update(|x| Some(x + 1), Ordering::SeqCst, Ordering::SeqCst), Ok(7)); +assert_eq!(x.fetch_update(|x| Some(x + 1), Ordering::SeqCst, Ordering::SeqCst), Ok(8)); +assert_eq!(x.load(Ordering::SeqCst), 9); +```"), + #[inline] + #[unstable(feature = "no_more_cas", + reason = "no more CAS loops in user code", + issue = "48655")] + pub fn fetch_update<F>(&self, + mut f: F, + fetch_order: Ordering, + set_order: Ordering) -> Result<$int_type, $int_type> + where F: FnMut($int_type) -> Option<$int_type> { + let mut prev = self.load(fetch_order); + while let Some(next) = f(prev) { + match self.compare_exchange_weak(prev, next, set_order, fetch_order) { + x @ Ok(_) => return x, + Err(next_prev) => prev = next_prev + } + } + Err(prev) + } + } + + doc_comment! { + concat!("Maximum with the current value. + +Finds the maximum of the current value and the argument `val`, and +sets the new value to the result. + +Returns the previous value. + +`fetch_max` takes an [`Ordering`] argument which describes the memory ordering +of this operation. All ordering modes are possible. Note that using +[`Acquire`] makes the store part of this operation [`Relaxed`], and +using [`Release`] makes the load part [`Relaxed`]. + +[`Ordering`]: enum.Ordering.html +[`Relaxed`]: enum.Ordering.html#variant.Relaxed +[`Release`]: enum.Ordering.html#variant.Release +[`Acquire`]: enum.Ordering.html#variant.Acquire + +# Examples + +``` +#![feature(atomic_min_max)] +", $extra_feature, "use std::sync::atomic::{", stringify!($atomic_type), ", Ordering}; + +let foo = ", stringify!($atomic_type), "::new(23); +assert_eq!(foo.fetch_max(42, Ordering::SeqCst), 23); +assert_eq!(foo.load(Ordering::SeqCst), 42); +``` + +If you want to obtain the maximum value in one step, you can use the following: + +``` +#![feature(atomic_min_max)] +", $extra_feature, "use std::sync::atomic::{", stringify!($atomic_type), ", Ordering}; + +let foo = ", stringify!($atomic_type), "::new(23); +let bar = 42; +let max_foo = foo.fetch_max(bar, Ordering::SeqCst).max(bar); +assert!(max_foo == 42); +```"), + #[inline] + #[unstable(feature = "atomic_min_max", + reason = "easier and faster min/max than writing manual CAS loop", + issue = "48655")] + pub fn fetch_max(&self, val: $int_type, order: Ordering) -> $int_type { + unsafe { $max_fn(self.v.get(), val, order) } + } + } + + doc_comment! { + concat!("Minimum with the current value. + +Finds the minimum of the current value and the argument `val`, and +sets the new value to the result. + +Returns the previous value. + +`fetch_min` takes an [`Ordering`] argument which describes the memory ordering +of this operation. All ordering modes are possible. Note that using +[`Acquire`] makes the store part of this operation [`Relaxed`], and +using [`Release`] makes the load part [`Relaxed`]. + +[`Ordering`]: enum.Ordering.html +[`Relaxed`]: enum.Ordering.html#variant.Relaxed +[`Release`]: enum.Ordering.html#variant.Release +[`Acquire`]: enum.Ordering.html#variant.Acquire + +# Examples + +``` +#![feature(atomic_min_max)] +", $extra_feature, "use std::sync::atomic::{", stringify!($atomic_type), ", Ordering}; + +let foo = ", stringify!($atomic_type), "::new(23); +assert_eq!(foo.fetch_min(42, Ordering::Relaxed), 23); +assert_eq!(foo.load(Ordering::Relaxed), 23); +assert_eq!(foo.fetch_min(22, Ordering::Relaxed), 23); +assert_eq!(foo.load(Ordering::Relaxed), 22); +``` + +If you want to obtain the minimum value in one step, you can use the following: + +``` +#![feature(atomic_min_max)] +", $extra_feature, "use std::sync::atomic::{", stringify!($atomic_type), ", Ordering}; + +let foo = ", stringify!($atomic_type), "::new(23); +let bar = 12; +let min_foo = foo.fetch_min(bar, Ordering::SeqCst).min(bar); +assert_eq!(min_foo, 12); +```"), + #[inline] + #[unstable(feature = "atomic_min_max", + reason = "easier and faster min/max than writing manual CAS loop", + issue = "48655")] + pub fn fetch_min(&self, val: $int_type, order: Ordering) -> $int_type { + unsafe { $min_fn(self.v.get(), val, order) } + } + } + } } } @@ -1377,7 +1818,10 @@ atomic_int! { unstable(feature = "integer_atomics", issue = "32976"), unstable(feature = "integer_atomics", issue = "32976"), unstable(feature = "integer_atomics", issue = "32976"), + unstable(feature = "integer_atomics", issue = "32976"), "i8", "../../../std/primitive.i8.html", + "#![feature(integer_atomics)]\n\n", + atomic_min, atomic_max, i8 AtomicI8 ATOMIC_I8_INIT } #[cfg(target_has_atomic = "8")] @@ -1387,7 +1831,10 @@ atomic_int! { unstable(feature = "integer_atomics", issue = "32976"), unstable(feature = "integer_atomics", issue = "32976"), unstable(feature = "integer_atomics", issue = "32976"), + unstable(feature = "integer_atomics", issue = "32976"), "u8", "../../../std/primitive.u8.html", + "#![feature(integer_atomics)]\n\n", + atomic_umin, atomic_umax, u8 AtomicU8 ATOMIC_U8_INIT } #[cfg(target_has_atomic = "16")] @@ -1397,7 +1844,10 @@ atomic_int! { unstable(feature = "integer_atomics", issue = "32976"), unstable(feature = "integer_atomics", issue = "32976"), unstable(feature = "integer_atomics", issue = "32976"), + unstable(feature = "integer_atomics", issue = "32976"), "i16", "../../../std/primitive.i16.html", + "#![feature(integer_atomics)]\n\n", + atomic_min, atomic_max, i16 AtomicI16 ATOMIC_I16_INIT } #[cfg(target_has_atomic = "16")] @@ -1407,7 +1857,10 @@ atomic_int! { unstable(feature = "integer_atomics", issue = "32976"), unstable(feature = "integer_atomics", issue = "32976"), unstable(feature = "integer_atomics", issue = "32976"), + unstable(feature = "integer_atomics", issue = "32976"), "u16", "../../../std/primitive.u16.html", + "#![feature(integer_atomics)]\n\n", + atomic_umin, atomic_umax, u16 AtomicU16 ATOMIC_U16_INIT } #[cfg(target_has_atomic = "32")] @@ -1417,7 +1870,10 @@ atomic_int! { unstable(feature = "integer_atomics", issue = "32976"), unstable(feature = "integer_atomics", issue = "32976"), unstable(feature = "integer_atomics", issue = "32976"), + unstable(feature = "integer_atomics", issue = "32976"), "i32", "../../../std/primitive.i32.html", + "#![feature(integer_atomics)]\n\n", + atomic_min, atomic_max, i32 AtomicI32 ATOMIC_I32_INIT } #[cfg(target_has_atomic = "32")] @@ -1427,7 +1883,10 @@ atomic_int! { unstable(feature = "integer_atomics", issue = "32976"), unstable(feature = "integer_atomics", issue = "32976"), unstable(feature = "integer_atomics", issue = "32976"), + unstable(feature = "integer_atomics", issue = "32976"), "u32", "../../../std/primitive.u32.html", + "#![feature(integer_atomics)]\n\n", + atomic_umin, atomic_umax, u32 AtomicU32 ATOMIC_U32_INIT } #[cfg(target_has_atomic = "64")] @@ -1437,7 +1896,10 @@ atomic_int! { unstable(feature = "integer_atomics", issue = "32976"), unstable(feature = "integer_atomics", issue = "32976"), unstable(feature = "integer_atomics", issue = "32976"), + unstable(feature = "integer_atomics", issue = "32976"), "i64", "../../../std/primitive.i64.html", + "#![feature(integer_atomics)]\n\n", + atomic_min, atomic_max, i64 AtomicI64 ATOMIC_I64_INIT } #[cfg(target_has_atomic = "64")] @@ -1447,7 +1909,10 @@ atomic_int! { unstable(feature = "integer_atomics", issue = "32976"), unstable(feature = "integer_atomics", issue = "32976"), unstable(feature = "integer_atomics", issue = "32976"), + unstable(feature = "integer_atomics", issue = "32976"), "u64", "../../../std/primitive.u64.html", + "#![feature(integer_atomics)]\n\n", + atomic_umin, atomic_umax, u64 AtomicU64 ATOMIC_U64_INIT } #[cfg(target_has_atomic = "ptr")] @@ -1457,7 +1922,10 @@ atomic_int!{ stable(feature = "atomic_debug", since = "1.3.0"), stable(feature = "atomic_access", since = "1.15.0"), stable(feature = "atomic_from", since = "1.23.0"), + stable(feature = "atomic_nand", since = "1.27.0"), "isize", "../../../std/primitive.isize.html", + "", + atomic_min, atomic_max, isize AtomicIsize ATOMIC_ISIZE_INIT } #[cfg(target_has_atomic = "ptr")] @@ -1467,11 +1935,15 @@ atomic_int!{ stable(feature = "atomic_debug", since = "1.3.0"), stable(feature = "atomic_access", since = "1.15.0"), stable(feature = "atomic_from", since = "1.23.0"), + stable(feature = "atomic_nand", since = "1.27.0"), "usize", "../../../std/primitive.usize.html", + "", + atomic_umin, atomic_umax, usize AtomicUsize ATOMIC_USIZE_INIT } #[inline] +#[cfg(target_has_atomic = "cas")] fn strongest_failure_ordering(order: Ordering) -> Ordering { match order { Release => Relaxed, @@ -1479,7 +1951,6 @@ fn strongest_failure_ordering(order: Ordering) -> Ordering { SeqCst => SeqCst, Acquire => Acquire, AcqRel => Acquire, - __Nonexhaustive => __Nonexhaustive, } } @@ -1491,7 +1962,6 @@ unsafe fn atomic_store<T>(dst: *mut T, val: T, order: Ordering) { SeqCst => intrinsics::atomic_store(dst, val), Acquire => panic!("there is no such thing as an acquire store"), AcqRel => panic!("there is no such thing as an acquire/release store"), - __Nonexhaustive => panic!("invalid memory ordering"), } } @@ -1503,11 +1973,11 @@ unsafe fn atomic_load<T>(dst: *const T, order: Ordering) -> T { SeqCst => intrinsics::atomic_load(dst), Release => panic!("there is no such thing as a release load"), AcqRel => panic!("there is no such thing as an acquire/release load"), - __Nonexhaustive => panic!("invalid memory ordering"), } } #[inline] +#[cfg(target_has_atomic = "cas")] unsafe fn atomic_swap<T>(dst: *mut T, val: T, order: Ordering) -> T { match order { Acquire => intrinsics::atomic_xchg_acq(dst, val), @@ -1515,7 +1985,6 @@ unsafe fn atomic_swap<T>(dst: *mut T, val: T, order: Ordering) -> T { AcqRel => intrinsics::atomic_xchg_acqrel(dst, val), Relaxed => intrinsics::atomic_xchg_relaxed(dst, val), SeqCst => intrinsics::atomic_xchg(dst, val), - __Nonexhaustive => panic!("invalid memory ordering"), } } @@ -1528,7 +1997,6 @@ unsafe fn atomic_add<T>(dst: *mut T, val: T, order: Ordering) -> T { AcqRel => intrinsics::atomic_xadd_acqrel(dst, val), Relaxed => intrinsics::atomic_xadd_relaxed(dst, val), SeqCst => intrinsics::atomic_xadd(dst, val), - __Nonexhaustive => panic!("invalid memory ordering"), } } @@ -1541,11 +2009,11 @@ unsafe fn atomic_sub<T>(dst: *mut T, val: T, order: Ordering) -> T { AcqRel => intrinsics::atomic_xsub_acqrel(dst, val), Relaxed => intrinsics::atomic_xsub_relaxed(dst, val), SeqCst => intrinsics::atomic_xsub(dst, val), - __Nonexhaustive => panic!("invalid memory ordering"), } } #[inline] +#[cfg(target_has_atomic = "cas")] unsafe fn atomic_compare_exchange<T>(dst: *mut T, old: T, new: T, @@ -1562,8 +2030,6 @@ unsafe fn atomic_compare_exchange<T>(dst: *mut T, (AcqRel, Relaxed) => intrinsics::atomic_cxchg_acqrel_failrelaxed(dst, old, new), (SeqCst, Relaxed) => intrinsics::atomic_cxchg_failrelaxed(dst, old, new), (SeqCst, Acquire) => intrinsics::atomic_cxchg_failacq(dst, old, new), - (__Nonexhaustive, _) => panic!("invalid memory ordering"), - (_, __Nonexhaustive) => panic!("invalid memory ordering"), (_, AcqRel) => panic!("there is no such thing as an acquire/release failure ordering"), (_, Release) => panic!("there is no such thing as a release failure ordering"), _ => panic!("a failure ordering can't be stronger than a success ordering"), @@ -1588,8 +2054,6 @@ unsafe fn atomic_compare_exchange_weak<T>(dst: *mut T, (AcqRel, Relaxed) => intrinsics::atomic_cxchgweak_acqrel_failrelaxed(dst, old, new), (SeqCst, Relaxed) => intrinsics::atomic_cxchgweak_failrelaxed(dst, old, new), (SeqCst, Acquire) => intrinsics::atomic_cxchgweak_failacq(dst, old, new), - (__Nonexhaustive, _) => panic!("invalid memory ordering"), - (_, __Nonexhaustive) => panic!("invalid memory ordering"), (_, AcqRel) => panic!("there is no such thing as an acquire/release failure ordering"), (_, Release) => panic!("there is no such thing as a release failure ordering"), _ => panic!("a failure ordering can't be stronger than a success ordering"), @@ -1605,7 +2069,17 @@ unsafe fn atomic_and<T>(dst: *mut T, val: T, order: Ordering) -> T { AcqRel => intrinsics::atomic_and_acqrel(dst, val), Relaxed => intrinsics::atomic_and_relaxed(dst, val), SeqCst => intrinsics::atomic_and(dst, val), - __Nonexhaustive => panic!("invalid memory ordering"), + } +} + +#[inline] +unsafe fn atomic_nand<T>(dst: *mut T, val: T, order: Ordering) -> T { + match order { + Acquire => intrinsics::atomic_nand_acq(dst, val), + Release => intrinsics::atomic_nand_rel(dst, val), + AcqRel => intrinsics::atomic_nand_acqrel(dst, val), + Relaxed => intrinsics::atomic_nand_relaxed(dst, val), + SeqCst => intrinsics::atomic_nand(dst, val), } } @@ -1617,7 +2091,6 @@ unsafe fn atomic_or<T>(dst: *mut T, val: T, order: Ordering) -> T { AcqRel => intrinsics::atomic_or_acqrel(dst, val), Relaxed => intrinsics::atomic_or_relaxed(dst, val), SeqCst => intrinsics::atomic_or(dst, val), - __Nonexhaustive => panic!("invalid memory ordering"), } } @@ -1629,7 +2102,54 @@ unsafe fn atomic_xor<T>(dst: *mut T, val: T, order: Ordering) -> T { AcqRel => intrinsics::atomic_xor_acqrel(dst, val), Relaxed => intrinsics::atomic_xor_relaxed(dst, val), SeqCst => intrinsics::atomic_xor(dst, val), - __Nonexhaustive => panic!("invalid memory ordering"), + } +} + +/// returns the max value (signed comparison) +#[inline] +unsafe fn atomic_max<T>(dst: *mut T, val: T, order: Ordering) -> T { + match order { + Acquire => intrinsics::atomic_max_acq(dst, val), + Release => intrinsics::atomic_max_rel(dst, val), + AcqRel => intrinsics::atomic_max_acqrel(dst, val), + Relaxed => intrinsics::atomic_max_relaxed(dst, val), + SeqCst => intrinsics::atomic_max(dst, val), + } +} + +/// returns the min value (signed comparison) +#[inline] +unsafe fn atomic_min<T>(dst: *mut T, val: T, order: Ordering) -> T { + match order { + Acquire => intrinsics::atomic_min_acq(dst, val), + Release => intrinsics::atomic_min_rel(dst, val), + AcqRel => intrinsics::atomic_min_acqrel(dst, val), + Relaxed => intrinsics::atomic_min_relaxed(dst, val), + SeqCst => intrinsics::atomic_min(dst, val), + } +} + +/// returns the max value (signed comparison) +#[inline] +unsafe fn atomic_umax<T>(dst: *mut T, val: T, order: Ordering) -> T { + match order { + Acquire => intrinsics::atomic_umax_acq(dst, val), + Release => intrinsics::atomic_umax_rel(dst, val), + AcqRel => intrinsics::atomic_umax_acqrel(dst, val), + Relaxed => intrinsics::atomic_umax_relaxed(dst, val), + SeqCst => intrinsics::atomic_umax(dst, val), + } +} + +/// returns the min value (signed comparison) +#[inline] +unsafe fn atomic_umin<T>(dst: *mut T, val: T, order: Ordering) -> T { + match order { + Acquire => intrinsics::atomic_umin_acq(dst, val), + Release => intrinsics::atomic_umin_rel(dst, val), + AcqRel => intrinsics::atomic_umin_acqrel(dst, val), + Relaxed => intrinsics::atomic_umin_relaxed(dst, val), + SeqCst => intrinsics::atomic_umin(dst, val), } } @@ -1719,7 +2239,6 @@ pub fn fence(order: Ordering) { AcqRel => intrinsics::atomic_fence_acqrel(), SeqCst => intrinsics::atomic_fence(), Relaxed => panic!("there is no such thing as a relaxed fence"), - __Nonexhaustive => panic!("invalid memory ordering"), } } } @@ -1809,7 +2328,6 @@ pub fn compiler_fence(order: Ordering) { AcqRel => intrinsics::atomic_singlethreadfence_acqrel(), SeqCst => intrinsics::atomic_singlethreadfence(), Relaxed => panic!("there is no such thing as a relaxed compiler fence"), - __Nonexhaustive => panic!("invalid memory ordering"), } } } @@ -1819,7 +2337,7 @@ pub fn compiler_fence(order: Ordering) { #[stable(feature = "atomic_debug", since = "1.3.0")] impl fmt::Debug for AtomicBool { fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result { - f.debug_tuple("AtomicBool").field(&self.load(Ordering::SeqCst)).finish() + fmt::Debug::fmt(&self.load(Ordering::SeqCst), f) } } @@ -1827,7 +2345,7 @@ impl fmt::Debug for AtomicBool { #[stable(feature = "atomic_debug", since = "1.3.0")] impl<T> fmt::Debug for AtomicPtr<T> { fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result { - f.debug_tuple("AtomicPtr").field(&self.load(Ordering::SeqCst)).finish() + fmt::Debug::fmt(&self.load(Ordering::SeqCst), f) } } diff --git a/src/libcore/task/context.rs b/src/libcore/task/context.rs new file mode 100644 index 00000000000..5a29c8528ef --- /dev/null +++ b/src/libcore/task/context.rs @@ -0,0 +1,98 @@ +// Copyright 2018 The Rust Project Developers. See the COPYRIGHT +// file at the top-level directory of this distribution and at +// http://rust-lang.org/COPYRIGHT. +// +// Licensed under the Apache License, Version 2.0 <LICENSE-APACHE or +// http://www.apache.org/licenses/LICENSE-2.0> or the MIT license +// <LICENSE-MIT or http://opensource.org/licenses/MIT>, at your +// option. This file may not be copied, modified, or distributed +// except according to those terms. + +#![unstable(feature = "futures_api", + reason = "futures in libcore are unstable", + issue = "50547")] + +use fmt; +use super::{Spawn, Waker, LocalWaker}; + +/// Information about the currently-running task. +/// +/// Contexts are always tied to the stack, since they are set up specifically +/// when performing a single `poll` step on a task. +pub struct Context<'a> { + local_waker: &'a LocalWaker, + spawner: &'a mut dyn Spawn, +} + +impl<'a> fmt::Debug for Context<'a> { + fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result { + f.debug_struct("Context") + .finish() + } +} + +impl<'a> Context<'a> { + /// Create a new task `Context` with the provided `local_waker`, `waker`, + /// and `spawner`. + #[inline] + pub fn new( + local_waker: &'a LocalWaker, + spawner: &'a mut dyn Spawn, + ) -> Context<'a> { + Context { local_waker, spawner } + } + + /// Get the `LocalWaker` associated with the current task. + #[inline] + pub fn local_waker(&self) -> &'a LocalWaker { + self.local_waker + } + + /// Get the `Waker` associated with the current task. + #[inline] + pub fn waker(&self) -> &'a Waker { + unsafe { &*(self.local_waker as *const LocalWaker as *const Waker) } + } + + /// Get the spawner associated with this task. + /// + /// This method is useful primarily if you want to explicitly handle + /// spawn failures. + #[inline] + pub fn spawner(&mut self) -> &mut dyn Spawn { + self.spawner + } + + /// Produce a context like the current one, but using the given waker + /// instead. + /// + /// This advanced method is primarily used when building "internal + /// schedulers" within a task, where you want to provide some customized + /// wakeup logic. + #[inline] + pub fn with_waker<'b>( + &'b mut self, + local_waker: &'b LocalWaker, + ) -> Context<'b> { + Context { + local_waker, + spawner: self.spawner, + } + } + + /// Produce a context like the current one, but using the given spawner + /// instead. + /// + /// This advanced method is primarily used when building "internal + /// schedulers" within a task. + #[inline] + pub fn with_spawner<'b, Sp: Spawn>( + &'b mut self, + spawner: &'b mut Sp, + ) -> Context<'b> { + Context { + local_waker: self.local_waker, + spawner, + } + } +} diff --git a/src/libcore/task/mod.rs b/src/libcore/task/mod.rs new file mode 100644 index 00000000000..f51e5f7ce0e --- /dev/null +++ b/src/libcore/task/mod.rs @@ -0,0 +1,27 @@ +// Copyright 2018 The Rust Project Developers. See the COPYRIGHT +// file at the top-level directory of this distribution and at +// http://rust-lang.org/COPYRIGHT. +// +// Licensed under the Apache License, Version 2.0 <LICENSE-APACHE or +// http://www.apache.org/licenses/LICENSE-2.0> or the MIT license +// <LICENSE-MIT or http://opensource.org/licenses/MIT>, at your +// option. This file may not be copied, modified, or distributed +// except according to those terms. + +#![unstable(feature = "futures_api", + reason = "futures in libcore are unstable", + issue = "50547")] + +//! Types and Traits for working with asynchronous tasks. + +mod context; +pub use self::context::Context; + +mod spawn; +pub use self::spawn::{Spawn, SpawnErrorKind, SpawnObjError, SpawnLocalObjError}; + +mod poll; +pub use self::poll::Poll; + +mod wake; +pub use self::wake::{Waker, LocalWaker, UnsafeWake}; diff --git a/src/libcore/task/poll.rs b/src/libcore/task/poll.rs new file mode 100644 index 00000000000..fb027efc6dc --- /dev/null +++ b/src/libcore/task/poll.rs @@ -0,0 +1,137 @@ +// Copyright 2018 The Rust Project Developers. See the COPYRIGHT +// file at the top-level directory of this distribution and at +// http://rust-lang.org/COPYRIGHT. +// +// Licensed under the Apache License, Version 2.0 <LICENSE-APACHE or +// http://www.apache.org/licenses/LICENSE-2.0> or the MIT license +// <LICENSE-MIT or http://opensource.org/licenses/MIT>, at your +// option. This file may not be copied, modified, or distributed +// except according to those terms. + +#![unstable(feature = "futures_api", + reason = "futures in libcore are unstable", + issue = "50547")] + +use ops::Try; +use result::Result; + +/// Indicates whether a value is available or if the current task has been +/// scheduled to receive a wakeup instead. +#[derive(Copy, Clone, Debug, Eq, PartialEq, Ord, PartialOrd, Hash)] +pub enum Poll<T> { + /// Represents that a value is immediately ready. + Ready(T), + + /// Represents that a value is not ready yet. + /// + /// When a function returns `Pending`, the function *must* also + /// ensure that the current task is scheduled to be awoken when + /// progress can be made. + Pending, +} + +impl<T> Poll<T> { + /// Change the ready value of this `Poll` with the closure provided + pub fn map<U, F>(self, f: F) -> Poll<U> + where F: FnOnce(T) -> U + { + match self { + Poll::Ready(t) => Poll::Ready(f(t)), + Poll::Pending => Poll::Pending, + } + } + + /// Returns whether this is `Poll::Ready` + #[inline] + pub fn is_ready(&self) -> bool { + match *self { + Poll::Ready(_) => true, + Poll::Pending => false, + } + } + + /// Returns whether this is `Poll::Pending` + #[inline] + pub fn is_pending(&self) -> bool { + !self.is_ready() + } +} + +impl<T, E> Poll<Result<T, E>> { + /// Change the success value of this `Poll` with the closure provided + pub fn map_ok<U, F>(self, f: F) -> Poll<Result<U, E>> + where F: FnOnce(T) -> U + { + match self { + Poll::Ready(Ok(t)) => Poll::Ready(Ok(f(t))), + Poll::Ready(Err(e)) => Poll::Ready(Err(e)), + Poll::Pending => Poll::Pending, + } + } + + /// Change the error value of this `Poll` with the closure provided + pub fn map_err<U, F>(self, f: F) -> Poll<Result<T, U>> + where F: FnOnce(E) -> U + { + match self { + Poll::Ready(Ok(t)) => Poll::Ready(Ok(t)), + Poll::Ready(Err(e)) => Poll::Ready(Err(f(e))), + Poll::Pending => Poll::Pending, + } + } +} + +impl<T> From<T> for Poll<T> { + fn from(t: T) -> Poll<T> { + Poll::Ready(t) + } +} + +impl<T, E> Try for Poll<Result<T, E>> { + type Ok = Poll<T>; + type Error = E; + + #[inline] + fn into_result(self) -> Result<Self::Ok, Self::Error> { + match self { + Poll::Ready(Ok(x)) => Ok(Poll::Ready(x)), + Poll::Ready(Err(e)) => Err(e), + Poll::Pending => Ok(Poll::Pending), + } + } + + #[inline] + fn from_error(e: Self::Error) -> Self { + Poll::Ready(Err(e)) + } + + #[inline] + fn from_ok(x: Self::Ok) -> Self { + x.map(Ok) + } +} + +impl<T, E> Try for Poll<Option<Result<T, E>>> { + type Ok = Poll<Option<T>>; + type Error = E; + + #[inline] + fn into_result(self) -> Result<Self::Ok, Self::Error> { + match self { + Poll::Ready(Some(Ok(x))) => Ok(Poll::Ready(Some(x))), + Poll::Ready(Some(Err(e))) => Err(e), + Poll::Ready(None) => Ok(Poll::Ready(None)), + Poll::Pending => Ok(Poll::Pending), + } + } + + #[inline] + fn from_error(e: Self::Error) -> Self { + Poll::Ready(Some(Err(e))) + } + + #[inline] + fn from_ok(x: Self::Ok) -> Self { + x.map(|x| x.map(Ok)) + } +} diff --git a/src/libcore/task/spawn.rs b/src/libcore/task/spawn.rs new file mode 100644 index 00000000000..58ee85d232b --- /dev/null +++ b/src/libcore/task/spawn.rs @@ -0,0 +1,93 @@ +// Copyright 2018 The Rust Project Developers. See the COPYRIGHT +// file at the top-level directory of this distribution and at +// http://rust-lang.org/COPYRIGHT. +// +// Licensed under the Apache License, Version 2.0 <LICENSE-APACHE or +// http://www.apache.org/licenses/LICENSE-2.0> or the MIT license +// <LICENSE-MIT or http://opensource.org/licenses/MIT>, at your +// option. This file may not be copied, modified, or distributed +// except according to those terms. + +#![unstable(feature = "futures_api", + reason = "futures in libcore are unstable", + issue = "50547")] + +use fmt; +use future::{FutureObj, LocalFutureObj}; + +/// Spawns tasks that poll futures to completion onto its associated task +/// executor. +/// +/// The term "task" refers to a kind of lightweight "thread". Task executors +/// are responsible for scheduling the execution of tasks on operating system +/// threads. +pub trait Spawn { + /// Spawns a new task with the given future. The future will be polled until + /// completion. + /// + /// # Errors + /// + /// The executor may be unable to spawn tasks, either because it has + /// been shut down or is resource-constrained. + fn spawn_obj( + &mut self, + future: FutureObj<'static, ()>, + ) -> Result<(), SpawnObjError>; + + /// Determines whether the executor is able to spawn new tasks. + /// + /// # Returns + /// + /// An `Ok` return means the executor is *likely* (but not guaranteed) + /// to accept a subsequent spawn attempt. Likewise, an `Err` return + /// means that `spawn` is likely, but not guaranteed, to yield an error. + #[inline] + fn status(&self) -> Result<(), SpawnErrorKind> { + Ok(()) + } +} + +/// Provides the reason that an executor was unable to spawn. +pub struct SpawnErrorKind { + _hidden: (), +} + +impl fmt::Debug for SpawnErrorKind { + fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result { + f.debug_tuple("SpawnErrorKind") + .field(&"shutdown") + .finish() + } +} + +impl SpawnErrorKind { + /// Spawning is failing because the executor has been shut down. + pub fn shutdown() -> SpawnErrorKind { + SpawnErrorKind { _hidden: () } + } + + /// Check whether this error is the `shutdown` error. + pub fn is_shutdown(&self) -> bool { + true + } +} + +/// The result of a failed spawn +#[derive(Debug)] +pub struct SpawnObjError { + /// The kind of error + pub kind: SpawnErrorKind, + + /// The future for which spawning inside a task was attempted + pub future: FutureObj<'static, ()>, +} + +/// The result of a failed spawn +#[derive(Debug)] +pub struct SpawnLocalObjError { + /// The kind of error + pub kind: SpawnErrorKind, + + /// The future for which spawning inside a task was attempted + pub future: LocalFutureObj<'static, ()>, +} diff --git a/src/libcore/task/wake.rs b/src/libcore/task/wake.rs new file mode 100644 index 00000000000..d770536ef42 --- /dev/null +++ b/src/libcore/task/wake.rs @@ -0,0 +1,279 @@ +// Copyright 2018 The Rust Project Developers. See the COPYRIGHT +// file at the top-level directory of this distribution and at +// http://rust-lang.org/COPYRIGHT. +// +// Licensed under the Apache License, Version 2.0 <LICENSE-APACHE or +// http://www.apache.org/licenses/LICENSE-2.0> or the MIT license +// <LICENSE-MIT or http://opensource.org/licenses/MIT>, at your +// option. This file may not be copied, modified, or distributed +// except according to those terms. + +#![unstable(feature = "futures_api", + reason = "futures in libcore are unstable", + issue = "50547")] + +use {fmt, mem}; +use marker::Unpin; +use ptr::NonNull; + +/// A `Waker` is a handle for waking up a task by notifying its executor that it +/// is ready to be run. +/// +/// This handle contains a trait object pointing to an instance of the `UnsafeWake` +/// trait, allowing notifications to get routed through it. +#[repr(transparent)] +pub struct Waker { + inner: NonNull<dyn UnsafeWake>, +} + +impl Unpin for Waker {} +unsafe impl Send for Waker {} +unsafe impl Sync for Waker {} + +impl Waker { + /// Constructs a new `Waker` directly. + /// + /// Note that most code will not need to call this. Implementers of the + /// `UnsafeWake` trait will typically provide a wrapper that calls this + /// but you otherwise shouldn't call it directly. + /// + /// If you're working with the standard library then it's recommended to + /// use the `Waker::from` function instead which works with the safe + /// `Arc` type and the safe `Wake` trait. + #[inline] + pub unsafe fn new(inner: NonNull<dyn UnsafeWake>) -> Self { + Waker { inner } + } + + /// Wake up the task associated with this `Waker`. + #[inline] + pub fn wake(&self) { + unsafe { self.inner.as_ref().wake() } + } + + /// Returns whether or not this `Waker` and `other` awaken the same task. + /// + /// This function works on a best-effort basis, and may return false even + /// when the `Waker`s would awaken the same task. However, if this function + /// returns true, it is guaranteed that the `Waker`s will awaken the same + /// task. + /// + /// This function is primarily used for optimization purposes. + #[inline] + pub fn will_wake(&self, other: &Waker) -> bool { + self.inner == other.inner + } +} + +impl Clone for Waker { + #[inline] + fn clone(&self) -> Self { + unsafe { + self.inner.as_ref().clone_raw() + } + } +} + +impl fmt::Debug for Waker { + fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result { + f.debug_struct("Waker") + .finish() + } +} + +impl Drop for Waker { + #[inline] + fn drop(&mut self) { + unsafe { + self.inner.as_ref().drop_raw() + } + } +} + +/// A `LocalWaker` is a handle for waking up a task by notifying its executor that it +/// is ready to be run. +/// +/// This is similar to the `Waker` type, but cannot be sent across threads. +/// Task executors can use this type to implement more optimized singlethreaded wakeup +/// behavior. +#[repr(transparent)] +pub struct LocalWaker { + inner: NonNull<dyn UnsafeWake>, +} + +impl Unpin for LocalWaker {} +impl !Send for LocalWaker {} +impl !Sync for LocalWaker {} + +impl LocalWaker { + /// Constructs a new `LocalWaker` directly. + /// + /// Note that most code will not need to call this. Implementers of the + /// `UnsafeWake` trait will typically provide a wrapper that calls this + /// but you otherwise shouldn't call it directly. + /// + /// If you're working with the standard library then it's recommended to + /// use the `local_waker_from_nonlocal` or `local_waker` to convert a `Waker` + /// into a `LocalWaker`. + /// + /// For this function to be used safely, it must be sound to call `inner.wake_local()` + /// on the current thread. + #[inline] + pub unsafe fn new(inner: NonNull<dyn UnsafeWake>) -> Self { + LocalWaker { inner } + } + + /// Wake up the task associated with this `LocalWaker`. + #[inline] + pub fn wake(&self) { + unsafe { self.inner.as_ref().wake_local() } + } + + /// Returns whether or not this `LocalWaker` and `other` `LocalWaker` awaken the same task. + /// + /// This function works on a best-effort basis, and may return false even + /// when the `LocalWaker`s would awaken the same task. However, if this function + /// returns true, it is guaranteed that the `LocalWaker`s will awaken the same + /// task. + /// + /// This function is primarily used for optimization purposes. + #[inline] + pub fn will_wake(&self, other: &LocalWaker) -> bool { + self.inner == other.inner + } + + /// Returns whether or not this `LocalWaker` and `other` `Waker` awaken the same task. + /// + /// This function works on a best-effort basis, and may return false even + /// when the `Waker`s would awaken the same task. However, if this function + /// returns true, it is guaranteed that the `LocalWaker`s will awaken the same + /// task. + /// + /// This function is primarily used for optimization purposes. + #[inline] + pub fn will_wake_nonlocal(&self, other: &Waker) -> bool { + self.inner == other.inner + } +} + +impl From<LocalWaker> for Waker { + #[inline] + fn from(local_waker: LocalWaker) -> Self { + let inner = local_waker.inner; + mem::forget(local_waker); + Waker { inner } + } +} + +impl Clone for LocalWaker { + #[inline] + fn clone(&self) -> Self { + let waker = unsafe { self.inner.as_ref().clone_raw() }; + let inner = waker.inner; + mem::forget(waker); + LocalWaker { inner } + } +} + +impl fmt::Debug for LocalWaker { + fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result { + f.debug_struct("Waker") + .finish() + } +} + +impl Drop for LocalWaker { + #[inline] + fn drop(&mut self) { + unsafe { + self.inner.as_ref().drop_raw() + } + } +} + +/// An unsafe trait for implementing custom memory management for a `Waker` or `LocalWaker`. +/// +/// A `Waker` conceptually is a cloneable trait object for `Wake`, and is +/// most often essentially just `Arc<dyn Wake>`. However, in some contexts +/// (particularly `no_std`), it's desirable to avoid `Arc` in favor of some +/// custom memory management strategy. This trait is designed to allow for such +/// customization. +/// +/// When using `std`, a default implementation of the `UnsafeWake` trait is provided for +/// `Arc<T>` where `T: Wake`. +pub unsafe trait UnsafeWake: Send + Sync { + /// Creates a clone of this `UnsafeWake` and stores it behind a `Waker`. + /// + /// This function will create a new uniquely owned handle that under the + /// hood references the same notification instance. In other words calls + /// to `wake` on the returned handle should be equivalent to calls to + /// `wake` on this handle. + /// + /// # Unsafety + /// + /// This function is unsafe to call because it's asserting the `UnsafeWake` + /// value is in a consistent state, i.e. hasn't been dropped. + unsafe fn clone_raw(&self) -> Waker; + + /// Drops this instance of `UnsafeWake`, deallocating resources + /// associated with it. + /// + /// FIXME(cramertj) + /// This method is intended to have a signature such as: + /// + /// ```ignore (not-a-doctest) + /// fn drop_raw(self: *mut Self); + /// ``` + /// + /// Unfortunately in Rust today that signature is not object safe. + /// Nevertheless it's recommended to implement this function *as if* that + /// were its signature. As such it is not safe to call on an invalid + /// pointer, nor is the validity of the pointer guaranteed after this + /// function returns. + /// + /// # Unsafety + /// + /// This function is unsafe to call because it's asserting the `UnsafeWake` + /// value is in a consistent state, i.e. hasn't been dropped. + unsafe fn drop_raw(&self); + + /// Indicates that the associated task is ready to make progress and should + /// be `poll`ed. + /// + /// Executors generally maintain a queue of "ready" tasks; `wake` should place + /// the associated task onto this queue. + /// + /// # Panics + /// + /// Implementations should avoid panicking, but clients should also be prepared + /// for panics. + /// + /// # Unsafety + /// + /// This function is unsafe to call because it's asserting the `UnsafeWake` + /// value is in a consistent state, i.e. hasn't been dropped. + unsafe fn wake(&self); + + /// Indicates that the associated task is ready to make progress and should + /// be `poll`ed. This function is the same as `wake`, but can only be called + /// from the thread that this `UnsafeWake` is "local" to. This allows for + /// implementors to provide specialized wakeup behavior specific to the current + /// thread. This function is called by `LocalWaker::wake`. + /// + /// Executors generally maintain a queue of "ready" tasks; `wake_local` should place + /// the associated task onto this queue. + /// + /// # Panics + /// + /// Implementations should avoid panicking, but clients should also be prepared + /// for panics. + /// + /// # Unsafety + /// + /// This function is unsafe to call because it's asserting the `UnsafeWake` + /// value is in a consistent state, i.e. hasn't been dropped, and that the + /// `UnsafeWake` hasn't moved from the thread on which it was created. + unsafe fn wake_local(&self) { + self.wake() + } +} diff --git a/src/libcore/tests/any.rs b/src/libcore/tests/any.rs index 2d3e81aa131..a80bf939530 100644 --- a/src/libcore/tests/any.rs +++ b/src/libcore/tests/any.rs @@ -17,7 +17,7 @@ static TEST: &'static str = "Test"; #[test] fn any_referenced() { - let (a, b, c) = (&5 as &Any, &TEST as &Any, &Test as &Any); + let (a, b, c) = (&5 as &dyn Any, &TEST as &dyn Any, &Test as &dyn Any); assert!(a.is::<i32>()); assert!(!b.is::<i32>()); @@ -34,7 +34,11 @@ fn any_referenced() { #[test] fn any_owning() { - let (a, b, c) = (box 5_usize as Box<Any>, box TEST as Box<Any>, box Test as Box<Any>); + let (a, b, c) = ( + box 5_usize as Box<dyn Any>, + box TEST as Box<dyn Any>, + box Test as Box<dyn Any>, + ); assert!(a.is::<usize>()); assert!(!b.is::<usize>()); @@ -51,7 +55,7 @@ fn any_owning() { #[test] fn any_downcast_ref() { - let a = &5_usize as &Any; + let a = &5_usize as &dyn Any; match a.downcast_ref::<usize>() { Some(&5) => {} @@ -69,9 +73,9 @@ fn any_downcast_mut() { let mut a = 5_usize; let mut b: Box<_> = box 7_usize; - let a_r = &mut a as &mut Any; + let a_r = &mut a as &mut dyn Any; let tmp: &mut usize = &mut *b; - let b_r = tmp as &mut Any; + let b_r = tmp as &mut dyn Any; match a_r.downcast_mut::<usize>() { Some(x) => { @@ -113,7 +117,7 @@ fn any_downcast_mut() { #[test] fn any_fixed_vec() { let test = [0_usize; 8]; - let test = &test as &Any; + let test = &test as &dyn Any; assert!(test.is::<[usize; 8]>()); assert!(!test.is::<[usize; 10]>()); } diff --git a/src/libcore/tests/ascii.rs b/src/libcore/tests/ascii.rs new file mode 100644 index 00000000000..950222dbcfa --- /dev/null +++ b/src/libcore/tests/ascii.rs @@ -0,0 +1,357 @@ +// Copyright 2013-2014 The Rust Project Developers. See the COPYRIGHT +// file at the top-level directory of this distribution and at +// http://rust-lang.org/COPYRIGHT. +// +// Licensed under the Apache License, Version 2.0 <LICENSE-APACHE or +// http://www.apache.org/licenses/LICENSE-2.0> or the MIT license +// <LICENSE-MIT or http://opensource.org/licenses/MIT>, at your +// option. This file may not be copied, modified, or distributed +// except according to those terms. + +use core::char::from_u32; + +#[test] +fn test_is_ascii() { + assert!(b"".is_ascii()); + assert!(b"banana\0\x7F".is_ascii()); + assert!(b"banana\0\x7F".iter().all(|b| b.is_ascii())); + assert!(!b"Vi\xe1\xbb\x87t Nam".is_ascii()); + assert!(!b"Vi\xe1\xbb\x87t Nam".iter().all(|b| b.is_ascii())); + assert!(!b"\xe1\xbb\x87".iter().any(|b| b.is_ascii())); + + assert!("".is_ascii()); + assert!("banana\0\u{7F}".is_ascii()); + assert!("banana\0\u{7F}".chars().all(|c| c.is_ascii())); + assert!(!"ประเทศไทย中华Việt Nam".chars().all(|c| c.is_ascii())); + assert!(!"ประเทศไทย中华ệ ".chars().any(|c| c.is_ascii())); +} + +#[test] +fn test_to_ascii_uppercase() { + assert_eq!("url()URL()uRl()ürl".to_ascii_uppercase(), "URL()URL()URL()üRL"); + assert_eq!("hıKß".to_ascii_uppercase(), "HıKß"); + + for i in 0..501 { + let upper = if 'a' as u32 <= i && i <= 'z' as u32 { i + 'A' as u32 - 'a' as u32 } + else { i }; + assert_eq!((from_u32(i).unwrap()).to_string().to_ascii_uppercase(), + (from_u32(upper).unwrap()).to_string()); + } +} + +#[test] +fn test_to_ascii_lowercase() { + assert_eq!("url()URL()uRl()Ürl".to_ascii_lowercase(), "url()url()url()Ürl"); + // Dotted capital I, Kelvin sign, Sharp S. + assert_eq!("HİKß".to_ascii_lowercase(), "hİKß"); + + for i in 0..501 { + let lower = if 'A' as u32 <= i && i <= 'Z' as u32 { i + 'a' as u32 - 'A' as u32 } + else { i }; + assert_eq!((from_u32(i).unwrap()).to_string().to_ascii_lowercase(), + (from_u32(lower).unwrap()).to_string()); + } +} + +#[test] +fn test_make_ascii_lower_case() { + macro_rules! test { + ($from: expr, $to: expr) => { + { + let mut x = $from; + x.make_ascii_lowercase(); + assert_eq!(x, $to); + } + } + } + test!(b'A', b'a'); + test!(b'a', b'a'); + test!(b'!', b'!'); + test!('A', 'a'); + test!('À', 'À'); + test!('a', 'a'); + test!('!', '!'); + test!(b"H\xc3\x89".to_vec(), b"h\xc3\x89"); + test!("HİKß".to_string(), "hİKß"); +} + + +#[test] +fn test_make_ascii_upper_case() { + macro_rules! test { + ($from: expr, $to: expr) => { + { + let mut x = $from; + x.make_ascii_uppercase(); + assert_eq!(x, $to); + } + } + } + test!(b'a', b'A'); + test!(b'A', b'A'); + test!(b'!', b'!'); + test!('a', 'A'); + test!('à', 'à'); + test!('A', 'A'); + test!('!', '!'); + test!(b"h\xc3\xa9".to_vec(), b"H\xc3\xa9"); + test!("hıKß".to_string(), "HıKß"); + + let mut x = "Hello".to_string(); + x[..3].make_ascii_uppercase(); // Test IndexMut on String. + assert_eq!(x, "HELlo") +} + +#[test] +fn test_eq_ignore_ascii_case() { + assert!("url()URL()uRl()Ürl".eq_ignore_ascii_case("url()url()url()Ürl")); + assert!(!"Ürl".eq_ignore_ascii_case("ürl")); + // Dotted capital I, Kelvin sign, Sharp S. + assert!("HİKß".eq_ignore_ascii_case("hİKß")); + assert!(!"İ".eq_ignore_ascii_case("i")); + assert!(!"K".eq_ignore_ascii_case("k")); + assert!(!"ß".eq_ignore_ascii_case("s")); + + for i in 0..501 { + let lower = if 'A' as u32 <= i && i <= 'Z' as u32 { i + 'a' as u32 - 'A' as u32 } + else { i }; + assert!((from_u32(i).unwrap()).to_string().eq_ignore_ascii_case( + &from_u32(lower).unwrap().to_string())); + } +} + +#[test] +fn inference_works() { + let x = "a".to_string(); + x.eq_ignore_ascii_case("A"); +} + +// Shorthands used by the is_ascii_* tests. +macro_rules! assert_all { + ($what:ident, $($str:tt),+) => {{ + $( + for b in $str.chars() { + if !b.$what() { + panic!("expected {}({}) but it isn't", + stringify!($what), b); + } + } + for b in $str.as_bytes().iter() { + if !b.$what() { + panic!("expected {}(0x{:02x})) but it isn't", + stringify!($what), b); + } + } + )+ + }}; + ($what:ident, $($str:tt),+,) => (assert_all!($what,$($str),+)) +} +macro_rules! assert_none { + ($what:ident, $($str:tt),+) => {{ + $( + for b in $str.chars() { + if b.$what() { + panic!("expected not-{}({}) but it is", + stringify!($what), b); + } + } + for b in $str.as_bytes().iter() { + if b.$what() { + panic!("expected not-{}(0x{:02x})) but it is", + stringify!($what), b); + } + } + )* + }}; + ($what:ident, $($str:tt),+,) => (assert_none!($what,$($str),+)) +} + +#[test] +fn test_is_ascii_alphabetic() { + assert_all!(is_ascii_alphabetic, + "", + "abcdefghijklmnopqrstuvwxyz", + "ABCDEFGHIJKLMNOQPRSTUVWXYZ", + ); + assert_none!(is_ascii_alphabetic, + "0123456789", + "!\"#$%&'()*+,-./:;<=>?@[\\]^_`{|}~", + " \t\n\x0c\r", + "\x00\x01\x02\x03\x04\x05\x06\x07", + "\x08\x09\x0a\x0b\x0c\x0d\x0e\x0f", + "\x10\x11\x12\x13\x14\x15\x16\x17", + "\x18\x19\x1a\x1b\x1c\x1d\x1e\x1f", + "\x7f", + ); +} + +#[test] +fn test_is_ascii_uppercase() { + assert_all!(is_ascii_uppercase, + "", + "ABCDEFGHIJKLMNOQPRSTUVWXYZ", + ); + assert_none!(is_ascii_uppercase, + "abcdefghijklmnopqrstuvwxyz", + "0123456789", + "!\"#$%&'()*+,-./:;<=>?@[\\]^_`{|}~", + " \t\n\x0c\r", + "\x00\x01\x02\x03\x04\x05\x06\x07", + "\x08\x09\x0a\x0b\x0c\x0d\x0e\x0f", + "\x10\x11\x12\x13\x14\x15\x16\x17", + "\x18\x19\x1a\x1b\x1c\x1d\x1e\x1f", + "\x7f", + ); +} + +#[test] +fn test_is_ascii_lowercase() { + assert_all!(is_ascii_lowercase, + "abcdefghijklmnopqrstuvwxyz", + ); + assert_none!(is_ascii_lowercase, + "ABCDEFGHIJKLMNOQPRSTUVWXYZ", + "0123456789", + "!\"#$%&'()*+,-./:;<=>?@[\\]^_`{|}~", + " \t\n\x0c\r", + "\x00\x01\x02\x03\x04\x05\x06\x07", + "\x08\x09\x0a\x0b\x0c\x0d\x0e\x0f", + "\x10\x11\x12\x13\x14\x15\x16\x17", + "\x18\x19\x1a\x1b\x1c\x1d\x1e\x1f", + "\x7f", + ); +} + +#[test] +fn test_is_ascii_alphanumeric() { + assert_all!(is_ascii_alphanumeric, + "", + "abcdefghijklmnopqrstuvwxyz", + "ABCDEFGHIJKLMNOQPRSTUVWXYZ", + "0123456789", + ); + assert_none!(is_ascii_alphanumeric, + "!\"#$%&'()*+,-./:;<=>?@[\\]^_`{|}~", + " \t\n\x0c\r", + "\x00\x01\x02\x03\x04\x05\x06\x07", + "\x08\x09\x0a\x0b\x0c\x0d\x0e\x0f", + "\x10\x11\x12\x13\x14\x15\x16\x17", + "\x18\x19\x1a\x1b\x1c\x1d\x1e\x1f", + "\x7f", + ); +} + +#[test] +fn test_is_ascii_digit() { + assert_all!(is_ascii_digit, + "", + "0123456789", + ); + assert_none!(is_ascii_digit, + "abcdefghijklmnopqrstuvwxyz", + "ABCDEFGHIJKLMNOQPRSTUVWXYZ", + "!\"#$%&'()*+,-./:;<=>?@[\\]^_`{|}~", + " \t\n\x0c\r", + "\x00\x01\x02\x03\x04\x05\x06\x07", + "\x08\x09\x0a\x0b\x0c\x0d\x0e\x0f", + "\x10\x11\x12\x13\x14\x15\x16\x17", + "\x18\x19\x1a\x1b\x1c\x1d\x1e\x1f", + "\x7f", + ); +} + +#[test] +fn test_is_ascii_hexdigit() { + assert_all!(is_ascii_hexdigit, + "", + "0123456789", + "abcdefABCDEF", + ); + assert_none!(is_ascii_hexdigit, + "ghijklmnopqrstuvwxyz", + "GHIJKLMNOQPRSTUVWXYZ", + "!\"#$%&'()*+,-./:;<=>?@[\\]^_`{|}~", + " \t\n\x0c\r", + "\x00\x01\x02\x03\x04\x05\x06\x07", + "\x08\x09\x0a\x0b\x0c\x0d\x0e\x0f", + "\x10\x11\x12\x13\x14\x15\x16\x17", + "\x18\x19\x1a\x1b\x1c\x1d\x1e\x1f", + "\x7f", + ); +} + +#[test] +fn test_is_ascii_punctuation() { + assert_all!(is_ascii_punctuation, + "", + "!\"#$%&'()*+,-./:;<=>?@[\\]^_`{|}~", + ); + assert_none!(is_ascii_punctuation, + "abcdefghijklmnopqrstuvwxyz", + "ABCDEFGHIJKLMNOQPRSTUVWXYZ", + "0123456789", + " \t\n\x0c\r", + "\x00\x01\x02\x03\x04\x05\x06\x07", + "\x08\x09\x0a\x0b\x0c\x0d\x0e\x0f", + "\x10\x11\x12\x13\x14\x15\x16\x17", + "\x18\x19\x1a\x1b\x1c\x1d\x1e\x1f", + "\x7f", + ); +} + +#[test] +fn test_is_ascii_graphic() { + assert_all!(is_ascii_graphic, + "", + "abcdefghijklmnopqrstuvwxyz", + "ABCDEFGHIJKLMNOQPRSTUVWXYZ", + "0123456789", + "!\"#$%&'()*+,-./:;<=>?@[\\]^_`{|}~", + ); + assert_none!(is_ascii_graphic, + " \t\n\x0c\r", + "\x00\x01\x02\x03\x04\x05\x06\x07", + "\x08\x09\x0a\x0b\x0c\x0d\x0e\x0f", + "\x10\x11\x12\x13\x14\x15\x16\x17", + "\x18\x19\x1a\x1b\x1c\x1d\x1e\x1f", + "\x7f", + ); +} + +#[test] +fn test_is_ascii_whitespace() { + assert_all!(is_ascii_whitespace, + "", + " \t\n\x0c\r", + ); + assert_none!(is_ascii_whitespace, + "abcdefghijklmnopqrstuvwxyz", + "ABCDEFGHIJKLMNOQPRSTUVWXYZ", + "0123456789", + "!\"#$%&'()*+,-./:;<=>?@[\\]^_`{|}~", + "\x00\x01\x02\x03\x04\x05\x06\x07", + "\x08\x0b\x0e\x0f", + "\x10\x11\x12\x13\x14\x15\x16\x17", + "\x18\x19\x1a\x1b\x1c\x1d\x1e\x1f", + "\x7f", + ); +} + +#[test] +fn test_is_ascii_control() { + assert_all!(is_ascii_control, + "", + "\x00\x01\x02\x03\x04\x05\x06\x07", + "\x08\x09\x0a\x0b\x0c\x0d\x0e\x0f", + "\x10\x11\x12\x13\x14\x15\x16\x17", + "\x18\x19\x1a\x1b\x1c\x1d\x1e\x1f", + "\x7f", + ); + assert_none!(is_ascii_control, + "abcdefghijklmnopqrstuvwxyz", + "ABCDEFGHIJKLMNOQPRSTUVWXYZ", + "0123456789", + "!\"#$%&'()*+,-./:;<=>?@[\\]^_`{|}~", + " ", + ); +} diff --git a/src/libcore/tests/atomic.rs b/src/libcore/tests/atomic.rs index 9babe24a985..a3667b3f3fe 100644 --- a/src/libcore/tests/atomic.rs +++ b/src/libcore/tests/atomic.rs @@ -49,6 +49,13 @@ fn uint_and() { } #[test] +fn uint_nand() { + let x = AtomicUsize::new(0xf731); + assert_eq!(x.fetch_nand(0x137f, SeqCst), 0xf731); + assert_eq!(x.load(SeqCst), !(0xf731 & 0x137f)); +} + +#[test] fn uint_or() { let x = AtomicUsize::new(0xf731); assert_eq!(x.fetch_or(0x137f, SeqCst), 0xf731); @@ -70,6 +77,13 @@ fn int_and() { } #[test] +fn int_nand() { + let x = AtomicIsize::new(0xf731); + assert_eq!(x.fetch_nand(0x137f, SeqCst), 0xf731); + assert_eq!(x.load(SeqCst), !(0xf731 & 0x137f)); +} + +#[test] fn int_or() { let x = AtomicIsize::new(0xf731); assert_eq!(x.fetch_or(0x137f, SeqCst), 0xf731); @@ -90,8 +104,10 @@ static S_UINT: AtomicUsize = AtomicUsize::new(0); #[test] fn static_init() { - assert!(!S_FALSE.load(SeqCst)); - assert!(S_TRUE.load(SeqCst)); - assert!(S_INT.load(SeqCst) == 0); - assert!(S_UINT.load(SeqCst) == 0); + // Note that we're not really testing the mutability here but it's important + // on Android at the moment (#49775) + assert!(!S_FALSE.swap(true, SeqCst)); + assert!(S_TRUE.swap(false, SeqCst)); + assert!(S_INT.fetch_add(1, SeqCst) == 0); + assert!(S_UINT.fetch_add(1, SeqCst) == 0); } diff --git a/src/libcore/tests/cell.rs b/src/libcore/tests/cell.rs index cc0ef6a6f17..4b7243b9cfc 100644 --- a/src/libcore/tests/cell.rs +++ b/src/libcore/tests/cell.rs @@ -27,6 +27,17 @@ fn smoketest_cell() { } #[test] +fn cell_update() { + let x = Cell::new(10); + + assert_eq!(x.update(|x| x + 5), 15); + assert_eq!(x.get(), 15); + + assert_eq!(x.update(|x| x / 3), 5); + assert_eq!(x.get(), 5); +} + +#[test] fn cell_has_sensible_show() { let x = Cell::new("foo bar"); assert!(format!("{:?}", x).contains(x.get())); @@ -155,6 +166,64 @@ fn ref_map_does_not_update_flag() { } #[test] +fn ref_map_split_updates_flag() { + let x = RefCell::new([1, 2]); + { + let b1 = x.borrow(); + assert!(x.try_borrow().is_ok()); + assert!(x.try_borrow_mut().is_err()); + { + let (_b2, _b3) = Ref::map_split(b1, |slc| slc.split_at(1)); + assert!(x.try_borrow().is_ok()); + assert!(x.try_borrow_mut().is_err()); + } + assert!(x.try_borrow().is_ok()); + assert!(x.try_borrow_mut().is_ok()); + } + assert!(x.try_borrow().is_ok()); + assert!(x.try_borrow_mut().is_ok()); + + { + let b1 = x.borrow_mut(); + assert!(x.try_borrow().is_err()); + assert!(x.try_borrow_mut().is_err()); + { + let (_b2, _b3) = RefMut::map_split(b1, |slc| slc.split_at_mut(1)); + assert!(x.try_borrow().is_err()); + assert!(x.try_borrow_mut().is_err()); + drop(_b2); + assert!(x.try_borrow().is_err()); + assert!(x.try_borrow_mut().is_err()); + } + assert!(x.try_borrow().is_ok()); + assert!(x.try_borrow_mut().is_ok()); + } + assert!(x.try_borrow().is_ok()); + assert!(x.try_borrow_mut().is_ok()); +} + +#[test] +fn ref_map_split() { + let x = RefCell::new([1, 2]); + let (b1, b2) = Ref::map_split(x.borrow(), |slc| slc.split_at(1)); + assert_eq!(*b1, [1]); + assert_eq!(*b2, [2]); +} + +#[test] +fn ref_mut_map_split() { + let x = RefCell::new([1, 2]); + { + let (mut b1, mut b2) = RefMut::map_split(x.borrow_mut(), |slc| slc.split_at_mut(1)); + assert_eq!(*b1, [1]); + assert_eq!(*b2, [2]); + b1[0] = 2; + b2[0] = 1; + } + assert_eq!(*x.borrow(), [2, 1]); +} + +#[test] fn ref_map_accessor() { struct X(RefCell<(u32, char)>); impl X { diff --git a/src/libcore/tests/char.rs b/src/libcore/tests/char.rs index 4e10ceac878..3d99c8ea9e2 100644 --- a/src/libcore/tests/char.rs +++ b/src/libcore/tests/char.rs @@ -148,9 +148,10 @@ fn test_is_control() { } #[test] -fn test_is_digit() { +fn test_is_numeric() { assert!('2'.is_numeric()); assert!('7'.is_numeric()); + assert!('¾'.is_numeric()); assert!(!'c'.is_numeric()); assert!(!'i'.is_numeric()); assert!(!'z'.is_numeric()); @@ -181,6 +182,7 @@ fn test_escape_debug() { assert_eq!(string('\u{ff}'), "\u{ff}"); assert_eq!(string('\u{11b}'), "\u{11b}"); assert_eq!(string('\u{1d4b6}'), "\u{1d4b6}"); + assert_eq!(string('\u{301}'), "\\u{301}"); // combining character assert_eq!(string('\u{200b}'),"\\u{200b}"); // zero width space assert_eq!(string('\u{e000}'), "\\u{e000}"); // private use 1 assert_eq!(string('\u{100000}'), "\\u{100000}"); // private use 2 @@ -362,53 +364,3 @@ fn eu_iterator_specializations() { check('\u{12340}'); check('\u{10FFFF}'); } - -#[test] -fn test_decode_utf8() { - macro_rules! assert_decode_utf8 { - ($input_bytes: expr, $expected_str: expr) => { - let input_bytes: &[u8] = &$input_bytes; - let s = char::decode_utf8(input_bytes.iter().cloned()) - .map(|r_b| r_b.unwrap_or('\u{FFFD}')) - .collect::<String>(); - assert_eq!(s, $expected_str, - "input bytes: {:?}, expected str: {:?}, result: {:?}", - input_bytes, $expected_str, s); - assert_eq!(String::from_utf8_lossy(&$input_bytes), $expected_str); - } - } - - assert_decode_utf8!([], ""); - assert_decode_utf8!([0x41], "A"); - assert_decode_utf8!([0xC1, 0x81], "��"); - assert_decode_utf8!([0xE2, 0x99, 0xA5], "♥"); - assert_decode_utf8!([0xE2, 0x99, 0xA5, 0x41], "♥A"); - assert_decode_utf8!([0xE2, 0x99], "�"); - assert_decode_utf8!([0xE2, 0x99, 0x41], "�A"); - assert_decode_utf8!([0xC0], "�"); - assert_decode_utf8!([0xC0, 0x41], "�A"); - assert_decode_utf8!([0x80], "�"); - assert_decode_utf8!([0x80, 0x41], "�A"); - assert_decode_utf8!([0xFE], "�"); - assert_decode_utf8!([0xFE, 0x41], "�A"); - assert_decode_utf8!([0xFF], "�"); - assert_decode_utf8!([0xFF, 0x41], "�A"); - assert_decode_utf8!([0xC0, 0x80], "��"); - - // Surrogates - assert_decode_utf8!([0xED, 0x9F, 0xBF], "\u{D7FF}"); - assert_decode_utf8!([0xED, 0xA0, 0x80], "���"); - assert_decode_utf8!([0xED, 0xBF, 0x80], "���"); - assert_decode_utf8!([0xEE, 0x80, 0x80], "\u{E000}"); - - // char::MAX - assert_decode_utf8!([0xF4, 0x8F, 0xBF, 0xBF], "\u{10FFFF}"); - assert_decode_utf8!([0xF4, 0x8F, 0xBF, 0x41], "�A"); - assert_decode_utf8!([0xF4, 0x90, 0x80, 0x80], "����"); - - // 5 and 6 bytes sequence - // Part of the original design of UTF-8, - // but invalid now that UTF-8 is artificially restricted to match the range of UTF-16. - assert_decode_utf8!([0xF8, 0x80, 0x80, 0x80, 0x80], "�����"); - assert_decode_utf8!([0xFC, 0x80, 0x80, 0x80, 0x80, 0x80], "������"); -} diff --git a/src/libcore/tests/fmt/num.rs b/src/libcore/tests/fmt/num.rs index 4ddedd91004..bc205ec0582 100644 --- a/src/libcore/tests/fmt/num.rs +++ b/src/libcore/tests/fmt/num.rs @@ -150,3 +150,9 @@ fn test_format_int_twos_complement() { assert!(format!("{}", i32::MIN) == "-2147483648"); assert!(format!("{}", i64::MIN) == "-9223372036854775808"); } + +#[test] +fn test_format_debug_hex() { + assert!(format!("{:02x?}", b"Foo\0") == "[46, 6f, 6f, 00]"); + assert!(format!("{:02X?}", b"Foo\0") == "[46, 6F, 6F, 00]"); +} diff --git a/src/libcore/tests/hash/mod.rs b/src/libcore/tests/hash/mod.rs index 8716421b424..85c9d41b65b 100644 --- a/src/libcore/tests/hash/mod.rs +++ b/src/libcore/tests/hash/mod.rs @@ -128,7 +128,7 @@ fn test_custom_state() { fn test_indirect_hasher() { let mut hasher = MyHasher { hash: 0 }; { - let mut indirect_hasher: &mut Hasher = &mut hasher; + let mut indirect_hasher: &mut dyn Hasher = &mut hasher; 5u32.hash(&mut indirect_hasher); } assert_eq!(hasher.hash, 5); diff --git a/src/libcore/tests/hash/sip.rs b/src/libcore/tests/hash/sip.rs index c6dd41798f2..bad858011e9 100644 --- a/src/libcore/tests/hash/sip.rs +++ b/src/libcore/tests/hash/sip.rs @@ -11,7 +11,7 @@ #![allow(deprecated)] use core::hash::{Hash, Hasher}; -use core::hash::{SipHasher, SipHasher13, SipHasher24}; +use core::hash::{SipHasher, SipHasher13}; use core::{slice, mem}; // Hash just the bytes of the slice, without length prefix @@ -224,14 +224,14 @@ fn test_siphash_2_4() { let k1 = 0x_0f_0e_0d_0c_0b_0a_09_08; let mut buf = Vec::new(); let mut t = 0; - let mut state_inc = SipHasher24::new_with_keys(k0, k1); + let mut state_inc = SipHasher::new_with_keys(k0, k1); while t < 64 { let vec = u8to64_le!(vecs[t], 0); - let out = hash_with(SipHasher24::new_with_keys(k0, k1), &Bytes(&buf)); + let out = hash_with(SipHasher::new_with_keys(k0, k1), &Bytes(&buf)); assert_eq!(vec, out); - let full = hash_with(SipHasher24::new_with_keys(k0, k1), &Bytes(&buf)); + let full = hash_with(SipHasher::new_with_keys(k0, k1), &Bytes(&buf)); let i = state_inc.finish(); assert_eq!(full, i); @@ -322,13 +322,13 @@ fn test_hash_no_concat_alias() { #[test] fn test_write_short_works() { let test_usize = 0xd0c0b0a0usize; - let mut h1 = SipHasher24::new(); + let mut h1 = SipHasher::new(); h1.write_usize(test_usize); h1.write(b"bytes"); h1.write(b"string"); h1.write_u8(0xFFu8); h1.write_u8(0x01u8); - let mut h2 = SipHasher24::new(); + let mut h2 = SipHasher::new(); h2.write(unsafe { slice::from_raw_parts(&test_usize as *const _ as *const u8, mem::size_of::<usize>()) diff --git a/src/libcore/tests/intrinsics.rs b/src/libcore/tests/intrinsics.rs index 2b380abf63c..9f3cba26a62 100644 --- a/src/libcore/tests/intrinsics.rs +++ b/src/libcore/tests/intrinsics.rs @@ -22,7 +22,7 @@ fn test_typeid_sized_types() { #[test] fn test_typeid_unsized_types() { trait Z {} - struct X(str); struct Y(Z + 'static); + struct X(str); struct Y(dyn Z + 'static); assert_eq!(TypeId::of::<X>(), TypeId::of::<X>()); assert_eq!(TypeId::of::<Y>(), TypeId::of::<Y>()); diff --git a/src/libcore/tests/iter.rs b/src/libcore/tests/iter.rs index 8997cf9c6bf..72b115f8b5f 100644 --- a/src/libcore/tests/iter.rs +++ b/src/libcore/tests/iter.rs @@ -145,6 +145,43 @@ fn test_iterator_chain_find() { } #[test] +fn test_zip_nth() { + let xs = [0, 1, 2, 4, 5]; + let ys = [10, 11, 12]; + + let mut it = xs.iter().zip(&ys); + assert_eq!(it.nth(0), Some((&0, &10))); + assert_eq!(it.nth(1), Some((&2, &12))); + assert_eq!(it.nth(0), None); + + let mut it = xs.iter().zip(&ys); + assert_eq!(it.nth(3), None); + + let mut it = ys.iter().zip(&xs); + assert_eq!(it.nth(3), None); +} + +#[test] +fn test_zip_nth_side_effects() { + let mut a = Vec::new(); + let mut b = Vec::new(); + let value = [1, 2, 3, 4, 5, 6].iter().cloned() + .map(|n| { + a.push(n); + n * 10 + }) + .zip([2, 3, 4, 5, 6, 7, 8].iter().cloned().map(|n| { + b.push(n * 100); + n * 1000 + })) + .skip(1) + .nth(3); + assert_eq!(value, Some((50, 6000))); + assert_eq!(a, vec![1, 2, 3, 4, 5]); + assert_eq!(b, vec![200, 300, 400, 500, 600]); +} + +#[test] fn test_iterator_step_by() { // Identity let mut it = (0..).step_by(1).take(3); @@ -162,6 +199,68 @@ fn test_iterator_step_by() { } #[test] +fn test_iterator_step_by_nth() { + let mut it = (0..16).step_by(5); + assert_eq!(it.nth(0), Some(0)); + assert_eq!(it.nth(0), Some(5)); + assert_eq!(it.nth(0), Some(10)); + assert_eq!(it.nth(0), Some(15)); + assert_eq!(it.nth(0), None); + + let it = (0..18).step_by(5); + assert_eq!(it.clone().nth(0), Some(0)); + assert_eq!(it.clone().nth(1), Some(5)); + assert_eq!(it.clone().nth(2), Some(10)); + assert_eq!(it.clone().nth(3), Some(15)); + assert_eq!(it.clone().nth(4), None); + assert_eq!(it.clone().nth(42), None); +} + +#[test] +fn test_iterator_step_by_nth_overflow() { + #[cfg(target_pointer_width = "8")] + type Bigger = u16; + #[cfg(target_pointer_width = "16")] + type Bigger = u32; + #[cfg(target_pointer_width = "32")] + type Bigger = u64; + #[cfg(target_pointer_width = "64")] + type Bigger = u128; + + #[derive(Clone)] + struct Test(Bigger); + impl<'a> Iterator for &'a mut Test { + type Item = i32; + fn next(&mut self) -> Option<Self::Item> { Some(21) } + fn nth(&mut self, n: usize) -> Option<Self::Item> { + self.0 += n as Bigger + 1; + Some(42) + } + } + + let mut it = Test(0); + let root = usize::MAX >> (::std::mem::size_of::<usize>() * 8 / 2); + let n = root + 20; + (&mut it).step_by(n).nth(n); + assert_eq!(it.0, n as Bigger * n as Bigger); + + // large step + let mut it = Test(0); + (&mut it).step_by(usize::MAX).nth(5); + assert_eq!(it.0, (usize::MAX as Bigger) * 5); + + // n + 1 overflows + let mut it = Test(0); + (&mut it).step_by(2).nth(usize::MAX); + assert_eq!(it.0, (usize::MAX as Bigger) * 2); + + // n + 1 overflows + let mut it = Test(0); + (&mut it).step_by(1).nth(usize::MAX); + assert_eq!(it.0, (usize::MAX as Bigger) * 1); +} + +#[test] #[should_panic] fn test_iterator_step_by_zero() { let mut it = (0..).step_by(0); @@ -813,6 +912,44 @@ fn test_iterator_flat_map_fold() { } #[test] +fn test_iterator_flatten() { + let xs = [0, 3, 6]; + let ys = [0, 1, 2, 3, 4, 5, 6, 7, 8]; + let it = xs.iter().map(|&x| (x..).step_by(1).take(3)).flatten(); + let mut i = 0; + for x in it { + assert_eq!(x, ys[i]); + i += 1; + } + assert_eq!(i, ys.len()); +} + +/// Test `Flatten::fold` with items already picked off the front and back, +/// to make sure all parts of the `Flatten` are folded correctly. +#[test] +fn test_iterator_flatten_fold() { + let xs = [0, 3, 6]; + let ys = [1, 2, 3, 4, 5, 6, 7]; + let mut it = xs.iter().map(|&x| x..x+3).flatten(); + assert_eq!(it.next(), Some(0)); + assert_eq!(it.next_back(), Some(8)); + let i = it.fold(0, |i, x| { + assert_eq!(x, ys[i]); + i + 1 + }); + assert_eq!(i, ys.len()); + + let mut it = xs.iter().map(|&x| x..x+3).flatten(); + assert_eq!(it.next(), Some(0)); + assert_eq!(it.next_back(), Some(8)); + let i = it.rfold(ys.len(), |i, x| { + assert_eq!(x, ys[i - 1]); + i - 1 + }); + assert_eq!(i, 0); +} + +#[test] fn test_inspect() { let xs = [1, 2, 3, 4]; let mut n = 0; @@ -1010,6 +1147,33 @@ fn test_find() { } #[test] +fn test_find_map() { + let xs: &[isize] = &[]; + assert_eq!(xs.iter().find_map(half_if_even), None); + let xs: &[isize] = &[3, 5]; + assert_eq!(xs.iter().find_map(half_if_even), None); + let xs: &[isize] = &[4, 5]; + assert_eq!(xs.iter().find_map(half_if_even), Some(2)); + let xs: &[isize] = &[3, 6]; + assert_eq!(xs.iter().find_map(half_if_even), Some(3)); + + let xs: &[isize] = &[1, 2, 3, 4, 5, 6, 7]; + let mut iter = xs.iter(); + assert_eq!(iter.find_map(half_if_even), Some(1)); + assert_eq!(iter.find_map(half_if_even), Some(2)); + assert_eq!(iter.find_map(half_if_even), Some(3)); + assert_eq!(iter.next(), Some(&7)); + + fn half_if_even(x: &isize) -> Option<isize> { + if x % 2 == 0 { + Some(x / 2) + } else { + None + } + } +} + +#[test] fn test_position() { let v = &[1, 3, 9, 27, 103, 14, 11]; assert_eq!(v.iter().position(|x| *x & 1 == 0).unwrap(), 5); @@ -1226,6 +1390,23 @@ fn test_double_ended_flat_map() { } #[test] +fn test_double_ended_flatten() { + let u = [0,1]; + let v = [5,6,7,8]; + let mut it = u.iter().map(|x| &v[*x..v.len()]).flatten(); + assert_eq!(it.next_back().unwrap(), &8); + assert_eq!(it.next().unwrap(), &5); + assert_eq!(it.next_back().unwrap(), &7); + assert_eq!(it.next_back().unwrap(), &6); + assert_eq!(it.next_back().unwrap(), &8); + assert_eq!(it.next().unwrap(), &6); + assert_eq!(it.next_back().unwrap(), &7); + assert_eq!(it.next_back(), None); + assert_eq!(it.next(), None); + assert_eq!(it.next_back(), None); +} + +#[test] fn test_double_ended_range() { assert_eq!((11..14).rev().collect::<Vec<_>>(), [13, 12, 11]); for _ in (10..0).rev() { @@ -1261,23 +1442,84 @@ fn test_range() { } #[test] +fn test_range_exhaustion() { + let mut r = 10..10; + assert!(r.is_empty()); + assert_eq!(r.next(), None); + assert_eq!(r.next_back(), None); + assert_eq!(r, 10..10); + + let mut r = 10..12; + assert_eq!(r.next(), Some(10)); + assert_eq!(r.next(), Some(11)); + assert!(r.is_empty()); + assert_eq!(r, 12..12); + assert_eq!(r.next(), None); + + let mut r = 10..12; + assert_eq!(r.next_back(), Some(11)); + assert_eq!(r.next_back(), Some(10)); + assert!(r.is_empty()); + assert_eq!(r, 10..10); + assert_eq!(r.next_back(), None); + + let mut r = 100..10; + assert!(r.is_empty()); + assert_eq!(r.next(), None); + assert_eq!(r.next_back(), None); + assert_eq!(r, 100..10); +} + +#[test] fn test_range_inclusive_exhaustion() { let mut r = 10..=10; assert_eq!(r.next(), Some(10)); - assert_eq!(r, 1..=0); + assert!(r.is_empty()); + assert_eq!(r.next(), None); + assert_eq!(r.next(), None); let mut r = 10..=10; assert_eq!(r.next_back(), Some(10)); - assert_eq!(r, 1..=0); + assert!(r.is_empty()); + assert_eq!(r.next_back(), None); + + let mut r = 10..=12; + assert_eq!(r.next(), Some(10)); + assert_eq!(r.next(), Some(11)); + assert_eq!(r.next(), Some(12)); + assert!(r.is_empty()); + assert_eq!(r.next(), None); + + let mut r = 10..=12; + assert_eq!(r.next_back(), Some(12)); + assert_eq!(r.next_back(), Some(11)); + assert_eq!(r.next_back(), Some(10)); + assert!(r.is_empty()); + assert_eq!(r.next_back(), None); let mut r = 10..=12; assert_eq!(r.nth(2), Some(12)); - assert_eq!(r, 1..=0); + assert!(r.is_empty()); + assert_eq!(r.next(), None); let mut r = 10..=12; assert_eq!(r.nth(5), None); - assert_eq!(r, 1..=0); + assert!(r.is_empty()); + assert_eq!(r.next(), None); + let mut r = 100..=10; + assert_eq!(r.next(), None); + assert!(r.is_empty()); + assert_eq!(r.next(), None); + assert_eq!(r.next(), None); + assert_eq!(r, 100..=10); + + let mut r = 100..=10; + assert_eq!(r.next_back(), None); + assert!(r.is_empty()); + assert_eq!(r.next_back(), None); + assert_eq!(r.next_back(), None); + assert_eq!(r, 100..=10); } #[test] @@ -1309,6 +1551,29 @@ fn test_range_from_nth() { assert_eq!(r, 16..); assert_eq!(r.nth(10), Some(26)); assert_eq!(r, 27..); + + assert_eq!((0..).size_hint(), (usize::MAX, None)); +} + +fn is_trusted_len<I: TrustedLen>(_: I) {} + +#[test] +fn test_range_from_take() { + let mut it = (0..).take(3); + assert_eq!(it.next(), Some(0)); + assert_eq!(it.next(), Some(1)); + assert_eq!(it.next(), Some(2)); + assert_eq!(it.next(), None); + is_trusted_len((0..).take(3)); + assert_eq!((0..).take(3).size_hint(), (3, Some(3))); + assert_eq!((0..).take(0).size_hint(), (0, Some(0))); + assert_eq!((0..).take(usize::MAX).size_hint(), (usize::MAX, Some(usize::MAX))); +} + +#[test] +fn test_range_from_take_collect() { + let v: Vec<_> = (0..).take(3).collect(); + assert_eq!(v, vec![0, 1, 2]); } #[test] @@ -1324,9 +1589,10 @@ fn test_range_inclusive_nth() { assert_eq!(r.nth(2), Some(15)); assert_eq!(r, 16..=20); assert_eq!(r.is_empty(), false); + assert_eq!(ExactSizeIterator::is_empty(&r), false); assert_eq!(r.nth(10), None); assert_eq!(r.is_empty(), true); - assert_eq!(r, 1..=0); // We may not want to document/promise this detail + assert_eq!(ExactSizeIterator::is_empty(&r), true); } #[test] @@ -1353,6 +1619,14 @@ fn test_range_step() { } #[test] +fn test_range_inclusive_step() { + assert_eq!((0..=50).step_by(10).collect::<Vec<_>>(), [0, 10, 20, 30, 40, 50]); + assert_eq!((0..=5).step_by(1).collect::<Vec<_>>(), [0, 1, 2, 3, 4, 5]); + assert_eq!((200..=255u8).step_by(10).collect::<Vec<_>>(), [200, 210, 220, 230, 240, 250]); + assert_eq!((250..=255u8).step_by(1).collect::<Vec<_>>(), [250, 251, 252, 253, 254, 255]); +} + +#[test] fn test_range_last_max() { assert_eq!((0..20).last(), Some(19)); assert_eq!((-20..0).last(), Some(-1)); @@ -1398,11 +1672,84 @@ fn test_range_inclusive_min() { } #[test] +fn test_range_inclusive_folds() { + assert_eq!((1..=10).sum::<i32>(), 55); + assert_eq!((1..=10).rev().sum::<i32>(), 55); + + let mut it = 40..=50; + assert_eq!(it.try_fold(0, i8::checked_add), None); + assert_eq!(it, 44..=50); + assert_eq!(it.try_rfold(0, i8::checked_add), None); + assert_eq!(it, 44..=47); + + let mut it = 10..=20; + assert_eq!(it.try_fold(0, |a,b| Some(a+b)), Some(165)); + assert!(it.is_empty()); + + let mut it = 10..=20; + assert_eq!(it.try_rfold(0, |a,b| Some(a+b)), Some(165)); + assert!(it.is_empty()); +} + +#[test] fn test_repeat() { let mut it = repeat(42); assert_eq!(it.next(), Some(42)); assert_eq!(it.next(), Some(42)); assert_eq!(it.next(), Some(42)); + assert_eq!(repeat(42).size_hint(), (usize::MAX, None)); +} + +#[test] +fn test_repeat_take() { + let mut it = repeat(42).take(3); + assert_eq!(it.next(), Some(42)); + assert_eq!(it.next(), Some(42)); + assert_eq!(it.next(), Some(42)); + assert_eq!(it.next(), None); + is_trusted_len(repeat(42).take(3)); + assert_eq!(repeat(42).take(3).size_hint(), (3, Some(3))); + assert_eq!(repeat(42).take(0).size_hint(), (0, Some(0))); + assert_eq!(repeat(42).take(usize::MAX).size_hint(), (usize::MAX, Some(usize::MAX))); +} + +#[test] +fn test_repeat_take_collect() { + let v: Vec<_> = repeat(42).take(3).collect(); + assert_eq!(v, vec![42, 42, 42]); +} + +#[test] +fn test_repeat_with() { + #[derive(PartialEq, Debug)] + struct NotClone(usize); + let mut it = repeat_with(|| NotClone(42)); + assert_eq!(it.next(), Some(NotClone(42))); + assert_eq!(it.next(), Some(NotClone(42))); + assert_eq!(it.next(), Some(NotClone(42))); + assert_eq!(repeat_with(|| NotClone(42)).size_hint(), (usize::MAX, None)); +} + +#[test] +fn test_repeat_with_take() { + let mut it = repeat_with(|| 42).take(3); + assert_eq!(it.next(), Some(42)); + assert_eq!(it.next(), Some(42)); + assert_eq!(it.next(), Some(42)); + assert_eq!(it.next(), None); + is_trusted_len(repeat_with(|| 42).take(3)); + assert_eq!(repeat_with(|| 42).take(3).size_hint(), (3, Some(3))); + assert_eq!(repeat_with(|| 42).take(0).size_hint(), (0, Some(0))); + assert_eq!(repeat_with(|| 42).take(usize::MAX).size_hint(), + (usize::MAX, Some(usize::MAX))); +} + +#[test] +fn test_repeat_with_take_collect() { + let mut curr = 1; + let v: Vec<_> = repeat_with(|| { let tmp = curr; curr *= 2; tmp }) + .take(5).collect(); + assert_eq!(v, vec![1, 2, 4, 8, 16]); } #[test] @@ -1746,3 +2093,54 @@ fn test_flat_map_try_folds() { assert_eq!(iter.try_rfold(0, i8::checked_add), None); assert_eq!(iter.next_back(), Some(35)); } + +#[test] +fn test_flatten_try_folds() { + let f = &|acc, x| i32::checked_add(acc*2/3, x); + let mr = &|x| (5*x)..(5*x + 5); + assert_eq!((0..10).map(mr).flatten().try_fold(7, f), (0..50).try_fold(7, f)); + assert_eq!((0..10).map(mr).flatten().try_rfold(7, f), (0..50).try_rfold(7, f)); + let mut iter = (0..10).map(mr).flatten(); + iter.next(); iter.next_back(); // have front and back iters in progress + assert_eq!(iter.try_rfold(7, f), (1..49).try_rfold(7, f)); + + let mut iter = (0..10).map(|x| (4*x)..(4*x + 4)).flatten(); + assert_eq!(iter.try_fold(0, i8::checked_add), None); + assert_eq!(iter.next(), Some(17)); + assert_eq!(iter.try_rfold(0, i8::checked_add), None); + assert_eq!(iter.next_back(), Some(35)); +} + +#[test] +fn test_functor_laws() { + // identity: + fn identity<T>(x: T) -> T { x } + assert_eq!((0..10).map(identity).sum::<usize>(), (0..10).sum()); + + // composition: + fn f(x: usize) -> usize { x + 3 } + fn g(x: usize) -> usize { x * 2 } + fn h(x: usize) -> usize { g(f(x)) } + assert_eq!((0..10).map(f).map(g).sum::<usize>(), (0..10).map(h).sum()); +} + +#[test] +fn test_monad_laws_left_identity() { + fn f(x: usize) -> impl Iterator<Item = usize> { + (0..10).map(move |y| x * y) + } + assert_eq!(once(42).flat_map(f.clone()).sum::<usize>(), f(42).sum()); +} + +#[test] +fn test_monad_laws_right_identity() { + assert_eq!((0..10).flat_map(|x| once(x)).sum::<usize>(), (0..10).sum()); +} + +#[test] +fn test_monad_laws_associativity() { + fn f(x: usize) -> impl Iterator<Item = usize> { 0..x } + fn g(x: usize) -> impl Iterator<Item = usize> { (0..x).rev() } + assert_eq!((0..10).flat_map(f).flat_map(g).sum::<usize>(), + (0..10).flat_map(|x| f(x).flat_map(g)).sum::<usize>()); +} diff --git a/src/libcore/tests/lib.rs b/src/libcore/tests/lib.rs index 2c0009569d7..fc92a5a2225 100644 --- a/src/libcore/tests/lib.rs +++ b/src/libcore/tests/lib.rs @@ -8,47 +8,47 @@ // option. This file may not be copied, modified, or distributed // except according to those terms. -#![deny(warnings)] - #![feature(box_syntax)] -#![feature(core_float)] +#![feature(cell_update)] #![feature(core_private_bignum)] #![feature(core_private_diy_float)] #![feature(dec2flt)] -#![feature(decode_utf8)] +#![feature(euclidean_division)] #![feature(exact_size_is_empty)] #![feature(fixed_size_array)] #![feature(flt2dec)] #![feature(fmt_internals)] -#![feature(iterator_step_by)] -#![feature(i128_type)] -#![feature(inclusive_range)] -#![feature(inclusive_range_syntax)] -#![feature(iterator_try_fold)] -#![feature(iter_rfind)] -#![feature(iter_rfold)] -#![feature(nonzero)] +#![feature(hashmap_internals)] #![feature(pattern)] +#![feature(range_is_empty)] #![feature(raw)] +#![feature(refcell_map_split)] #![feature(refcell_replace_swap)] -#![feature(sip_hash_13)] #![feature(slice_patterns)] -#![feature(slice_rotate)] #![feature(sort_internals)] #![feature(specialization)] #![feature(step_trait)] +#![feature(str_internals)] #![feature(test)] #![feature(trusted_len)] #![feature(try_from)] #![feature(try_trait)] -#![feature(unique)] #![feature(exact_chunks)] +#![feature(slice_align_to)] +#![feature(align_offset)] +#![feature(reverse_bits)] +#![feature(iterator_find_map)] +#![feature(inner_deref)] +#![feature(slice_internals)] +#![feature(option_replace)] extern crate core; extern crate test; +extern crate rand; mod any; mod array; +mod ascii; mod atomic; mod cell; mod char; @@ -58,6 +58,7 @@ mod fmt; mod hash; mod intrinsics; mod iter; +mod manually_drop; mod mem; mod nonzero; mod num; @@ -68,4 +69,6 @@ mod ptr; mod result; mod slice; mod str; +mod str_lossy; +mod time; mod tuple; diff --git a/src/libcore/tests/manually_drop.rs b/src/libcore/tests/manually_drop.rs new file mode 100644 index 00000000000..82dfb8d4c0b --- /dev/null +++ b/src/libcore/tests/manually_drop.rs @@ -0,0 +1,29 @@ +// Copyright 2018 The Rust Project Developers. See the COPYRIGHT +// file at the top-level directory of this distribution and at +// http://rust-lang.org/COPYRIGHT. +// +// Licensed under the Apache License, Version 2.0 <LICENSE-APACHE or +// http://www.apache.org/licenses/LICENSE-2.0> or the MIT license +// <LICENSE-MIT or http://opensource.org/licenses/MIT>, at your +// option. This file may not be copied, modified, or distributed +// except according to those terms. + +use core::mem::ManuallyDrop; + +#[test] +fn smoke() { + struct TypeWithDrop; + impl Drop for TypeWithDrop { + fn drop(&mut self) { + unreachable!("Should not get dropped"); + } + } + + let x = ManuallyDrop::new(TypeWithDrop); + drop(x); + + // also test unsizing + let x : Box<ManuallyDrop<[TypeWithDrop]>> = + Box::new(ManuallyDrop::new([TypeWithDrop, TypeWithDrop])); + drop(x); +} diff --git a/src/libcore/tests/mem.rs b/src/libcore/tests/mem.rs index f55a1c81463..714f2babbdf 100644 --- a/src/libcore/tests/mem.rs +++ b/src/libcore/tests/mem.rs @@ -109,11 +109,11 @@ fn test_transmute() { trait Foo { fn dummy(&self) { } } impl Foo for isize {} - let a = box 100isize as Box<Foo>; + let a = box 100isize as Box<dyn Foo>; unsafe { let x: ::core::raw::TraitObject = transmute(a); assert!(*(x.data as *const isize) == 100); - let _x: Box<Foo> = transmute(x); + let _x: Box<dyn Foo> = transmute(x); } unsafe { diff --git a/src/libcore/tests/nonzero.rs b/src/libcore/tests/nonzero.rs index a795dd57504..8d39298bac3 100644 --- a/src/libcore/tests/nonzero.rs +++ b/src/libcore/tests/nonzero.rs @@ -8,7 +8,7 @@ // option. This file may not be copied, modified, or distributed // except according to those terms. -use core::nonzero::NonZero; +use core::num::NonZeroU32; use core::option::Option; use core::option::Option::{Some, None}; use std::mem::size_of; @@ -16,28 +16,28 @@ use std::mem::size_of; #[test] fn test_create_nonzero_instance() { let _a = unsafe { - NonZero::new_unchecked(21) + NonZeroU32::new_unchecked(21) }; } #[test] fn test_size_nonzero_in_option() { - assert_eq!(size_of::<NonZero<u32>>(), size_of::<Option<NonZero<u32>>>()); + assert_eq!(size_of::<NonZeroU32>(), size_of::<Option<NonZeroU32>>()); } #[test] fn test_match_on_nonzero_option() { let a = Some(unsafe { - NonZero::new_unchecked(42) + NonZeroU32::new_unchecked(42) }); match a { Some(val) => assert_eq!(val.get(), 42), - None => panic!("unexpected None while matching on Some(NonZero(_))") + None => panic!("unexpected None while matching on Some(NonZeroU32(_))") } - match unsafe { Some(NonZero::new_unchecked(43)) } { + match unsafe { Some(NonZeroU32::new_unchecked(43)) } { Some(val) => assert_eq!(val.get(), 43), - None => panic!("unexpected None while matching on Some(NonZero(_))") + None => panic!("unexpected None while matching on Some(NonZeroU32(_))") } } @@ -98,3 +98,26 @@ fn test_match_option_string() { None => panic!("unexpected None while matching on Some(String { ... })") } } + +mod atom { + use core::num::NonZeroU32; + + #[derive(PartialEq, Eq)] + pub struct Atom { + index: NonZeroU32, // private + } + pub const FOO_ATOM: Atom = Atom { index: unsafe { NonZeroU32::new_unchecked(7) } }; +} + +macro_rules! atom { + ("foo") => { atom::FOO_ATOM } +} + +#[test] +fn test_match_nonzero_const_pattern() { + match atom!("foo") { + // Using as a pattern is supported by the compiler: + atom!("foo") => {} + _ => panic!("Expected the const item as a pattern to match.") + } +} diff --git a/src/libcore/tests/num/dec2flt/mod.rs b/src/libcore/tests/num/dec2flt/mod.rs index 9934e1dab96..17b2f59cd4d 100644 --- a/src/libcore/tests/num/dec2flt/mod.rs +++ b/src/libcore/tests/num/dec2flt/mod.rs @@ -102,6 +102,12 @@ fn lonely_dot() { } #[test] +fn exponentiated_dot() { + assert!(".e0".parse::<f32>().is_err()); + assert!(".e0".parse::<f64>().is_err()); +} + +#[test] fn lonely_sign() { assert!("+".parse::<f32>().is_err()); assert!("-".parse::<f64>().is_err()); diff --git a/src/libcore/tests/num/dec2flt/parse.rs b/src/libcore/tests/num/dec2flt/parse.rs index 09acf2bc517..3ad694e38ad 100644 --- a/src/libcore/tests/num/dec2flt/parse.rs +++ b/src/libcore/tests/num/dec2flt/parse.rs @@ -8,7 +8,6 @@ // option. This file may not be copied, modified, or distributed // except according to those terms. -use std::iter; use core::num::dec2flt::parse::{Decimal, parse_decimal}; use core::num::dec2flt::parse::ParseResult::{Valid, Invalid}; @@ -46,7 +45,7 @@ fn valid() { assert_eq!(parse_decimal("1.e300"), Valid(Decimal::new(b"1", b"", 300))); assert_eq!(parse_decimal(".1e300"), Valid(Decimal::new(b"", b"1", 300))); assert_eq!(parse_decimal("101e-33"), Valid(Decimal::new(b"101", b"", -33))); - let zeros: String = iter::repeat('0').take(25).collect(); + let zeros = "0".repeat(25); let s = format!("1.5e{}", zeros); assert_eq!(parse_decimal(&s), Valid(Decimal::new(b"1", b"5", 0))); } diff --git a/src/libcore/tests/num/flt2dec/mod.rs b/src/libcore/tests/num/flt2dec/mod.rs index ef0178815f9..04567e25e25 100644 --- a/src/libcore/tests/num/flt2dec/mod.rs +++ b/src/libcore/tests/num/flt2dec/mod.rs @@ -23,6 +23,7 @@ mod strategy { mod dragon; mod grisu; } +mod random; pub fn decode_finite<T: DecodableFloat>(v: T) -> Decoded { match decode(v).1 { diff --git a/src/libcore/tests/num/flt2dec/random.rs b/src/libcore/tests/num/flt2dec/random.rs new file mode 100644 index 00000000000..a1928657dab --- /dev/null +++ b/src/libcore/tests/num/flt2dec/random.rs @@ -0,0 +1,158 @@ +// Copyright 2017 The Rust Project Developers. See the COPYRIGHT +// file at the top-level directory of this distribution and at +// http://rust-lang.org/COPYRIGHT. +// +// Licensed under the Apache License, Version 2.0 <LICENSE-APACHE or +// http://www.apache.org/licenses/LICENSE-2.0> or the MIT license +// <LICENSE-MIT or http://opensource.org/licenses/MIT>, at your +// option. This file may not be copied, modified, or distributed +// except according to those terms. + +#![cfg(not(target_arch = "wasm32"))] + +use std::i16; +use std::str; + +use core::num::flt2dec::MAX_SIG_DIGITS; +use core::num::flt2dec::strategy::grisu::format_exact_opt; +use core::num::flt2dec::strategy::grisu::format_shortest_opt; +use core::num::flt2dec::{decode, DecodableFloat, FullDecoded, Decoded}; + +use rand::{self, Rand, XorShiftRng}; +use rand::distributions::{IndependentSample, Range}; + +pub fn decode_finite<T: DecodableFloat>(v: T) -> Decoded { + match decode(v).1 { + FullDecoded::Finite(decoded) => decoded, + full_decoded => panic!("expected finite, got {:?} instead", full_decoded) + } +} + + +fn iterate<F, G, V>(func: &str, k: usize, n: usize, mut f: F, mut g: G, mut v: V) -> (usize, usize) + where F: FnMut(&Decoded, &mut [u8]) -> Option<(usize, i16)>, + G: FnMut(&Decoded, &mut [u8]) -> (usize, i16), + V: FnMut(usize) -> Decoded { + assert!(k <= 1024); + + let mut npassed = 0; // f(x) = Some(g(x)) + let mut nignored = 0; // f(x) = None + + for i in 0..n { + if (i & 0xfffff) == 0 { + println!("in progress, {:x}/{:x} (ignored={} passed={} failed={})", + i, n, nignored, npassed, i - nignored - npassed); + } + + let decoded = v(i); + let mut buf1 = [0; 1024]; + if let Some((len1, e1)) = f(&decoded, &mut buf1[..k]) { + let mut buf2 = [0; 1024]; + let (len2, e2) = g(&decoded, &mut buf2[..k]); + if e1 == e2 && &buf1[..len1] == &buf2[..len2] { + npassed += 1; + } else { + println!("equivalence test failed, {:x}/{:x}: {:?} f(i)={}e{} g(i)={}e{}", + i, n, decoded, str::from_utf8(&buf1[..len1]).unwrap(), e1, + str::from_utf8(&buf2[..len2]).unwrap(), e2); + } + } else { + nignored += 1; + } + } + println!("{}({}): done, ignored={} passed={} failed={}", + func, k, nignored, npassed, n - nignored - npassed); + assert!(nignored + npassed == n, + "{}({}): {} out of {} values returns an incorrect value!", + func, k, n - nignored - npassed, n); + (npassed, nignored) +} + +pub fn f32_random_equivalence_test<F, G>(f: F, g: G, k: usize, n: usize) + where F: FnMut(&Decoded, &mut [u8]) -> Option<(usize, i16)>, + G: FnMut(&Decoded, &mut [u8]) -> (usize, i16) { + let mut rng: XorShiftRng = Rand::rand(&mut rand::thread_rng()); + let f32_range = Range::new(0x0000_0001u32, 0x7f80_0000); + iterate("f32_random_equivalence_test", k, n, f, g, |_| { + let x = f32::from_bits(f32_range.ind_sample(&mut rng)); + decode_finite(x) + }); +} + +pub fn f64_random_equivalence_test<F, G>(f: F, g: G, k: usize, n: usize) + where F: FnMut(&Decoded, &mut [u8]) -> Option<(usize, i16)>, + G: FnMut(&Decoded, &mut [u8]) -> (usize, i16) { + let mut rng: XorShiftRng = Rand::rand(&mut rand::thread_rng()); + let f64_range = Range::new(0x0000_0000_0000_0001u64, 0x7ff0_0000_0000_0000); + iterate("f64_random_equivalence_test", k, n, f, g, |_| { + let x = f64::from_bits(f64_range.ind_sample(&mut rng)); + decode_finite(x) + }); +} + +pub fn f32_exhaustive_equivalence_test<F, G>(f: F, g: G, k: usize) + where F: FnMut(&Decoded, &mut [u8]) -> Option<(usize, i16)>, + G: FnMut(&Decoded, &mut [u8]) -> (usize, i16) { + // we have only 2^23 * (2^8 - 1) - 1 = 2,139,095,039 positive finite f32 values, + // so why not simply testing all of them? + // + // this is of course very stressful (and thus should be behind an `#[ignore]` attribute), + // but with `-C opt-level=3 -C lto` this only takes about an hour or so. + + // iterate from 0x0000_0001 to 0x7f7f_ffff, i.e. all finite ranges + let (npassed, nignored) = iterate("f32_exhaustive_equivalence_test", + k, 0x7f7f_ffff, f, g, |i: usize| { + + let x = f32::from_bits(i as u32 + 1); + decode_finite(x) + }); + assert_eq!((npassed, nignored), (2121451881, 17643158)); +} + +#[test] +fn shortest_random_equivalence_test() { + use core::num::flt2dec::strategy::dragon::format_shortest as fallback; + f64_random_equivalence_test(format_shortest_opt, fallback, MAX_SIG_DIGITS, 10_000); + f32_random_equivalence_test(format_shortest_opt, fallback, MAX_SIG_DIGITS, 10_000); +} + +#[test] #[ignore] // it is too expensive +fn shortest_f32_exhaustive_equivalence_test() { + // it is hard to directly test the optimality of the output, but we can at least test if + // two different algorithms agree to each other. + // + // this reports the progress and the number of f32 values returned `None`. + // with `--nocapture` (and plenty of time and appropriate rustc flags), this should print: + // `done, ignored=17643158 passed=2121451881 failed=0`. + + use core::num::flt2dec::strategy::dragon::format_shortest as fallback; + f32_exhaustive_equivalence_test(format_shortest_opt, fallback, MAX_SIG_DIGITS); +} + +#[test] #[ignore] // it is too expensive +fn shortest_f64_hard_random_equivalence_test() { + // this again probably has to use appropriate rustc flags. + + use core::num::flt2dec::strategy::dragon::format_shortest as fallback; + f64_random_equivalence_test(format_shortest_opt, fallback, + MAX_SIG_DIGITS, 100_000_000); +} + +#[test] +fn exact_f32_random_equivalence_test() { + use core::num::flt2dec::strategy::dragon::format_exact as fallback; + for k in 1..21 { + f32_random_equivalence_test(|d, buf| format_exact_opt(d, buf, i16::MIN), + |d, buf| fallback(d, buf, i16::MIN), k, 1_000); + } +} + +#[test] +fn exact_f64_random_equivalence_test() { + use core::num::flt2dec::strategy::dragon::format_exact as fallback; + for k in 1..21 { + f64_random_equivalence_test(|d, buf| format_exact_opt(d, buf, i16::MIN), + |d, buf| fallback(d, buf, i16::MIN), k, 1_000); + } +} + diff --git a/src/libcore/tests/num/int_macros.rs b/src/libcore/tests/num/int_macros.rs index 8d791283ab8..71d2e794538 100644 --- a/src/libcore/tests/num/int_macros.rs +++ b/src/libcore/tests/num/int_macros.rs @@ -31,6 +31,11 @@ mod tests { } #[test] + fn test_mod_euc() { + assert!((-1 as $T).mod_euc(MIN) == MAX); + } + + #[test] pub fn test_abs() { assert!((1 as $T).abs() == 1 as $T); assert!((0 as $T).abs() == 0 as $T); diff --git a/src/libcore/tests/num/mod.rs b/src/libcore/tests/num/mod.rs index 587dcbe6d67..ab96d3126bb 100644 --- a/src/libcore/tests/num/mod.rs +++ b/src/libcore/tests/num/mod.rs @@ -143,6 +143,15 @@ fn test_infallible_try_from_int_error() { } macro_rules! test_impl_from { + ($fn_name:ident, bool, $target: ty) => { + #[test] + fn $fn_name() { + let one: $target = 1; + let zero: $target = 0; + assert_eq!(one, <$target>::from(true)); + assert_eq!(zero, <$target>::from(false)); + } + }; ($fn_name: ident, $Small: ty, $Large: ty) => { #[test] fn $fn_name() { @@ -182,6 +191,18 @@ test_impl_from! { test_u16i32, u16, i32 } test_impl_from! { test_u16i64, u16, i64 } test_impl_from! { test_u32i64, u32, i64 } +// Bool -> Integer +test_impl_from! { test_boolu8, bool, u8 } +test_impl_from! { test_boolu16, bool, u16 } +test_impl_from! { test_boolu32, bool, u32 } +test_impl_from! { test_boolu64, bool, u64 } +test_impl_from! { test_boolu128, bool, u128 } +test_impl_from! { test_booli8, bool, i8 } +test_impl_from! { test_booli16, bool, i16 } +test_impl_from! { test_booli32, bool, i32 } +test_impl_from! { test_booli64, bool, i64 } +test_impl_from! { test_booli128, bool, i128 } + // Signed -> Float test_impl_from! { test_i8f32, i8, f32 } test_impl_from! { test_i8f64, i8, f64 } @@ -635,51 +656,69 @@ assume_usize_width! { macro_rules! test_float { ($modname: ident, $fty: ty, $inf: expr, $neginf: expr, $nan: expr) => { mod $modname { - use core::num::Float; // FIXME(nagisa): these tests should test for sign of -0.0 #[test] fn min() { - assert_eq!(0.0.min(0.0), 0.0); - assert_eq!((-0.0).min(-0.0), -0.0); - assert_eq!(9.0.min(9.0), 9.0); - assert_eq!((-9.0).min(0.0), -9.0); - assert_eq!(0.0.min(9.0), 0.0); - assert_eq!((-0.0).min(-9.0), -9.0); - assert_eq!($inf.min(9.0), 9.0); - assert_eq!(9.0.min($inf), 9.0); - assert_eq!($inf.min(-9.0), -9.0); - assert_eq!((-9.0).min($inf), -9.0); - assert_eq!($neginf.min(9.0), $neginf); - assert_eq!(9.0.min($neginf), $neginf); - assert_eq!($neginf.min(-9.0), $neginf); - assert_eq!((-9.0).min($neginf), $neginf); - assert_eq!($nan.min(9.0), 9.0); - assert_eq!($nan.min(-9.0), -9.0); - assert_eq!(9.0.min($nan), 9.0); - assert_eq!((-9.0).min($nan), -9.0); - assert!($nan.min($nan).is_nan()); + assert_eq!((0.0 as $fty).min(0.0), 0.0); + assert_eq!((-0.0 as $fty).min(-0.0), -0.0); + assert_eq!((9.0 as $fty).min(9.0), 9.0); + assert_eq!((-9.0 as $fty).min(0.0), -9.0); + assert_eq!((0.0 as $fty).min(9.0), 0.0); + assert_eq!((-0.0 as $fty).min(-9.0), -9.0); + assert_eq!(($inf as $fty).min(9.0), 9.0); + assert_eq!((9.0 as $fty).min($inf), 9.0); + assert_eq!(($inf as $fty).min(-9.0), -9.0); + assert_eq!((-9.0 as $fty).min($inf), -9.0); + assert_eq!(($neginf as $fty).min(9.0), $neginf); + assert_eq!((9.0 as $fty).min($neginf), $neginf); + assert_eq!(($neginf as $fty).min(-9.0), $neginf); + assert_eq!((-9.0 as $fty).min($neginf), $neginf); + assert_eq!(($nan as $fty).min(9.0), 9.0); + assert_eq!(($nan as $fty).min(-9.0), -9.0); + assert_eq!((9.0 as $fty).min($nan), 9.0); + assert_eq!((-9.0 as $fty).min($nan), -9.0); + assert!(($nan as $fty).min($nan).is_nan()); } #[test] fn max() { - assert_eq!(0.0.max(0.0), 0.0); - assert_eq!((-0.0).max(-0.0), -0.0); - assert_eq!(9.0.max(9.0), 9.0); - assert_eq!((-9.0).max(0.0), 0.0); - assert_eq!(0.0.max(9.0), 9.0); - assert_eq!((-0.0).max(-9.0), -0.0); - assert_eq!($inf.max(9.0), $inf); - assert_eq!(9.0.max($inf), $inf); - assert_eq!($inf.max(-9.0), $inf); - assert_eq!((-9.0).max($inf), $inf); - assert_eq!($neginf.max(9.0), 9.0); - assert_eq!(9.0.max($neginf), 9.0); - assert_eq!($neginf.max(-9.0), -9.0); - assert_eq!((-9.0).max($neginf), -9.0); - assert_eq!($nan.max(9.0), 9.0); - assert_eq!($nan.max(-9.0), -9.0); - assert_eq!(9.0.max($nan), 9.0); - assert_eq!((-9.0).max($nan), -9.0); - assert!($nan.max($nan).is_nan()); + assert_eq!((0.0 as $fty).max(0.0), 0.0); + assert_eq!((-0.0 as $fty).max(-0.0), -0.0); + assert_eq!((9.0 as $fty).max(9.0), 9.0); + assert_eq!((-9.0 as $fty).max(0.0), 0.0); + assert_eq!((0.0 as $fty).max(9.0), 9.0); + assert_eq!((-0.0 as $fty).max(-9.0), -0.0); + assert_eq!(($inf as $fty).max(9.0), $inf); + assert_eq!((9.0 as $fty).max($inf), $inf); + assert_eq!(($inf as $fty).max(-9.0), $inf); + assert_eq!((-9.0 as $fty).max($inf), $inf); + assert_eq!(($neginf as $fty).max(9.0), 9.0); + assert_eq!((9.0 as $fty).max($neginf), 9.0); + assert_eq!(($neginf as $fty).max(-9.0), -9.0); + assert_eq!((-9.0 as $fty).max($neginf), -9.0); + assert_eq!(($nan as $fty).max(9.0), 9.0); + assert_eq!(($nan as $fty).max(-9.0), -9.0); + assert_eq!((9.0 as $fty).max($nan), 9.0); + assert_eq!((-9.0 as $fty).max($nan), -9.0); + assert!(($nan as $fty).max($nan).is_nan()); + } + #[test] + fn mod_euc() { + let a: $fty = 42.0; + assert!($inf.mod_euc(a).is_nan()); + assert_eq!(a.mod_euc($inf), a); + assert!(a.mod_euc($nan).is_nan()); + assert!($inf.mod_euc($inf).is_nan()); + assert!($inf.mod_euc($nan).is_nan()); + assert!($nan.mod_euc($inf).is_nan()); + } + #[test] + fn div_euc() { + let a: $fty = 42.0; + assert_eq!(a.div_euc($inf), 0.0); + assert!(a.div_euc($nan).is_nan()); + assert!($inf.div_euc($inf).is_nan()); + assert!($inf.div_euc($nan).is_nan()); + assert!($nan.div_euc($inf).is_nan()); } } } } diff --git a/src/libcore/tests/num/uint_macros.rs b/src/libcore/tests/num/uint_macros.rs index daa1cc3a7f4..ca6906f7310 100644 --- a/src/libcore/tests/num/uint_macros.rs +++ b/src/libcore/tests/num/uint_macros.rs @@ -98,6 +98,17 @@ mod tests { } #[test] + fn test_reverse_bits() { + assert_eq!(A.reverse_bits().reverse_bits(), A); + assert_eq!(B.reverse_bits().reverse_bits(), B); + assert_eq!(C.reverse_bits().reverse_bits(), C); + + // Swapping these should make no difference + assert_eq!(_0.reverse_bits(), _0); + assert_eq!(_1.reverse_bits(), _1); + } + + #[test] fn test_le() { assert_eq!($T::from_le(A.to_le()), A); assert_eq!($T::from_le(B.to_le()), B); diff --git a/src/libcore/tests/ops.rs b/src/libcore/tests/ops.rs index 9d2fa1abff6..d66193b1687 100644 --- a/src/libcore/tests/ops.rs +++ b/src/libcore/tests/ops.rs @@ -50,21 +50,45 @@ fn test_full_range() { #[test] fn test_range_inclusive() { - let mut r = RangeInclusive { start: 1i8, end: 2 }; + let mut r = RangeInclusive::new(1i8, 2); assert_eq!(r.next(), Some(1)); assert_eq!(r.next(), Some(2)); assert_eq!(r.next(), None); - r = RangeInclusive { start: 127i8, end: 127 }; + r = RangeInclusive::new(127i8, 127); assert_eq!(r.next(), Some(127)); assert_eq!(r.next(), None); - r = RangeInclusive { start: -128i8, end: -128 }; + r = RangeInclusive::new(-128i8, -128); assert_eq!(r.next_back(), Some(-128)); assert_eq!(r.next_back(), None); // degenerate - r = RangeInclusive { start: 1, end: -1 }; + r = RangeInclusive::new(1, -1); assert_eq!(r.size_hint(), (0, Some(0))); assert_eq!(r.next(), None); } + + +#[test] +fn test_range_is_empty() { + use core::f32::*; + + assert!(!(0.0 .. 10.0).is_empty()); + assert!( (-0.0 .. 0.0).is_empty()); + assert!( (10.0 .. 0.0).is_empty()); + + assert!(!(NEG_INFINITY .. INFINITY).is_empty()); + assert!( (EPSILON .. NAN).is_empty()); + assert!( (NAN .. EPSILON).is_empty()); + assert!( (NAN .. NAN).is_empty()); + + assert!(!(0.0 ..= 10.0).is_empty()); + assert!(!(-0.0 ..= 0.0).is_empty()); + assert!( (10.0 ..= 0.0).is_empty()); + + assert!(!(NEG_INFINITY ..= INFINITY).is_empty()); + assert!( (EPSILON ..= NAN).is_empty()); + assert!( (NAN ..= EPSILON).is_empty()); + assert!( (NAN ..= NAN).is_empty()); +} diff --git a/src/libcore/tests/option.rs b/src/libcore/tests/option.rs index 22109e28edd..1324ba2d9a9 100644 --- a/src/libcore/tests/option.rs +++ b/src/libcore/tests/option.rs @@ -240,7 +240,7 @@ fn test_collect() { assert!(v == None); // test that it does not take more elements than it needs - let mut functions: [Box<Fn() -> Option<()>>; 3] = + let mut functions: [Box<dyn Fn() -> Option<()>>; 3] = [box || Some(()), box || None, box || panic!()]; let v: Option<Vec<()>> = functions.iter_mut().map(|f| (*f)()).collect(); @@ -297,3 +297,35 @@ fn test_try() { } assert_eq!(try_option_err(), Err(NoneError)); } + +#[test] +fn test_option_deref() { + // Some: &Option<T: Deref>::Some(T) -> Option<&T::Deref::Target>::Some(&*T) + let ref_option = &Some(&42); + assert_eq!(ref_option.deref(), Some(&42)); + + let ref_option = &Some(String::from("a result")); + assert_eq!(ref_option.deref(), Some("a result")); + + let ref_option = &Some(vec![1, 2, 3, 4, 5]); + assert_eq!(ref_option.deref(), Some(&[1, 2, 3, 4, 5][..])); + + // None: &Option<T: Deref>>::None -> None + let ref_option: &Option<&i32> = &None; + assert_eq!(ref_option.deref(), None); +} + +#[test] +fn test_replace() { + let mut x = Some(2); + let old = x.replace(5); + + assert_eq!(x, Some(5)); + assert_eq!(old, Some(2)); + + let mut x = None; + let old = x.replace(3); + + assert_eq!(x, Some(3)); + assert_eq!(old, None); +} diff --git a/src/libcore/tests/ptr.rs b/src/libcore/tests/ptr.rs index 98436f0e1d1..92160910d8f 100644 --- a/src/libcore/tests/ptr.rs +++ b/src/libcore/tests/ptr.rs @@ -84,16 +84,16 @@ fn test_is_null() { assert!(nms.is_null()); // Pointers to unsized types -- trait objects - let ci: *const ToString = &3; + let ci: *const dyn ToString = &3; assert!(!ci.is_null()); - let mi: *mut ToString = &mut 3; + let mi: *mut dyn ToString = &mut 3; assert!(!mi.is_null()); - let nci: *const ToString = null::<isize>(); + let nci: *const dyn ToString = null::<isize>(); assert!(nci.is_null()); - let nmi: *mut ToString = null_mut::<isize>(); + let nmi: *mut dyn ToString = null_mut::<isize>(); assert!(nmi.is_null()); } @@ -140,16 +140,16 @@ fn test_as_ref() { assert_eq!(nms.as_ref(), None); // Pointers to unsized types -- trait objects - let ci: *const ToString = &3; + let ci: *const dyn ToString = &3; assert!(ci.as_ref().is_some()); - let mi: *mut ToString = &mut 3; + let mi: *mut dyn ToString = &mut 3; assert!(mi.as_ref().is_some()); - let nci: *const ToString = null::<isize>(); + let nci: *const dyn ToString = null::<isize>(); assert!(nci.as_ref().is_none()); - let nmi: *mut ToString = null_mut::<isize>(); + let nmi: *mut dyn ToString = null_mut::<isize>(); assert!(nmi.as_ref().is_none()); } } @@ -182,10 +182,10 @@ fn test_as_mut() { assert_eq!(nms.as_mut(), None); // Pointers to unsized types -- trait objects - let mi: *mut ToString = &mut 3; + let mi: *mut dyn ToString = &mut 3; assert!(mi.as_mut().is_some()); - let nmi: *mut ToString = null_mut::<isize>(); + let nmi: *mut dyn ToString = null_mut::<isize>(); assert!(nmi.as_mut().is_none()); } } @@ -249,9 +249,9 @@ fn test_set_memory() { } #[test] -fn test_unsized_unique() { +fn test_unsized_nonnull() { let xs: &[i32] = &[1, 2, 3]; - let ptr = unsafe { Unique::new_unchecked(xs as *const [i32] as *mut [i32]) }; + let ptr = unsafe { NonNull::new_unchecked(xs as *const [i32] as *mut [i32]) }; let ys = unsafe { ptr.as_ref() }; let zs: &[i32] = &[1, 2, 3]; assert!(ys == zs); @@ -296,3 +296,92 @@ fn write_unaligned_drop() { } DROPS.with(|d| assert_eq!(*d.borrow(), [0])); } + +#[test] +fn align_offset_zst() { + // For pointers of stride = 0, the pointer is already aligned or it cannot be aligned at + // all, because no amount of elements will align the pointer. + let mut p = 1; + while p < 1024 { + assert_eq!((p as *const ()).align_offset(p), 0); + if p != 1 { + assert_eq!(((p + 1) as *const ()).align_offset(p), !0); + } + p = (p + 1).next_power_of_two(); + } +} + +#[test] +fn align_offset_stride1() { + // For pointers of stride = 1, the pointer can always be aligned. The offset is equal to + // number of bytes. + let mut align = 1; + while align < 1024 { + for ptr in 1..2*align { + let expected = ptr % align; + let offset = if expected == 0 { 0 } else { align - expected }; + assert_eq!((ptr as *const u8).align_offset(align), offset, + "ptr = {}, align = {}, size = 1", ptr, align); + } + align = (align + 1).next_power_of_two(); + } +} + +#[test] +fn align_offset_weird_strides() { + #[repr(packed)] + struct A3(u16, u8); + struct A4(u32); + #[repr(packed)] + struct A5(u32, u8); + #[repr(packed)] + struct A6(u32, u16); + #[repr(packed)] + struct A7(u32, u16, u8); + #[repr(packed)] + struct A8(u32, u32); + #[repr(packed)] + struct A9(u32, u32, u8); + #[repr(packed)] + struct A10(u32, u32, u16); + + unsafe fn test_weird_stride<T>(ptr: *const T, align: usize) -> bool { + let numptr = ptr as usize; + let mut expected = usize::max_value(); + // Naive but definitely correct way to find the *first* aligned element of stride::<T>. + for el in 0..align { + if (numptr + el * ::std::mem::size_of::<T>()) % align == 0 { + expected = el; + break; + } + } + let got = ptr.align_offset(align); + if got != expected { + eprintln!("aligning {:p} (with stride of {}) to {}, expected {}, got {}", ptr, + ::std::mem::size_of::<T>(), align, expected, got); + return true; + } + return false; + } + + // For pointers of stride != 1, we verify the algorithm against the naivest possible + // implementation + let mut align = 1; + let mut x = false; + while align < 1024 { + for ptr in 1usize..4*align { + unsafe { + x |= test_weird_stride::<A3>(ptr as *const A3, align); + x |= test_weird_stride::<A4>(ptr as *const A4, align); + x |= test_weird_stride::<A5>(ptr as *const A5, align); + x |= test_weird_stride::<A6>(ptr as *const A6, align); + x |= test_weird_stride::<A7>(ptr as *const A7, align); + x |= test_weird_stride::<A8>(ptr as *const A8, align); + x |= test_weird_stride::<A9>(ptr as *const A9, align); + x |= test_weird_stride::<A10>(ptr as *const A10, align); + } + } + align = (align + 1).next_power_of_two(); + } + assert!(!x); +} diff --git a/src/libcore/tests/result.rs b/src/libcore/tests/result.rs index ce41bde8342..0c00992ffd8 100644 --- a/src/libcore/tests/result.rs +++ b/src/libcore/tests/result.rs @@ -81,7 +81,7 @@ fn test_collect() { assert!(v == Err(2)); // test that it does not take more elements than it needs - let mut functions: [Box<Fn() -> Result<(), isize>>; 3] = + let mut functions: [Box<dyn Fn() -> Result<(), isize>>; 3] = [box || Ok(()), box || Err(1), box || panic!()]; let v: Result<Vec<()>, isize> = functions.iter_mut().map(|f| (*f)()).collect(); @@ -220,14 +220,109 @@ fn test_try() { assert_eq!(try_result_none(), None); fn try_result_ok() -> Result<u8, u8> { - let val = Ok(1)?; + let result: Result<u8, u8> = Ok(1); + let val = result?; Ok(val) } assert_eq!(try_result_ok(), Ok(1)); fn try_result_err() -> Result<u8, u8> { - let val = Err(1)?; + let result: Result<u8, u8> = Err(1); + let val = result?; Ok(val) } assert_eq!(try_result_err(), Err(1)); } + +#[test] +fn test_result_deref() { + // &Result<T: Deref, E>::Ok(T).deref_ok() -> + // Result<&T::Deref::Target, &E>::Ok(&*T) + let ref_ok = &Result::Ok::<&i32, u8>(&42); + let expected_result = Result::Ok::<&i32, &u8>(&42); + assert_eq!(ref_ok.deref_ok(), expected_result); + + let ref_ok = &Result::Ok::<String, u32>(String::from("a result")); + let expected_result = Result::Ok::<&str, &u32>("a result"); + assert_eq!(ref_ok.deref_ok(), expected_result); + + let ref_ok = &Result::Ok::<Vec<i32>, u32>(vec![1, 2, 3, 4, 5]); + let expected_result = Result::Ok::<&[i32], &u32>(&[1, 2, 3, 4, 5][..]); + assert_eq!(ref_ok.deref_ok(), expected_result); + + // &Result<T: Deref, E: Deref>::Ok(T).deref() -> + // Result<&T::Deref::Target, &E::Deref::Target>::Ok(&*T) + let ref_ok = &Result::Ok::<&i32, &u8>(&42); + let expected_result = Result::Ok::<&i32, &u8>(&42); + assert_eq!(ref_ok.deref(), expected_result); + + let ref_ok = &Result::Ok::<String, &u32>(String::from("a result")); + let expected_result = Result::Ok::<&str, &u32>("a result"); + assert_eq!(ref_ok.deref(), expected_result); + + let ref_ok = &Result::Ok::<Vec<i32>, &u32>(vec![1, 2, 3, 4, 5]); + let expected_result = Result::Ok::<&[i32], &u32>(&[1, 2, 3, 4, 5][..]); + assert_eq!(ref_ok.deref(), expected_result); + + // &Result<T, E: Deref>::Err(T).deref_err() -> + // Result<&T, &E::Deref::Target>::Err(&*E) + let ref_err = &Result::Err::<u8, &i32>(&41); + let expected_result = Result::Err::<&u8, &i32>(&41); + assert_eq!(ref_err.deref_err(), expected_result); + + let ref_err = &Result::Err::<u32, String>(String::from("an error")); + let expected_result = Result::Err::<&u32, &str>("an error"); + assert_eq!(ref_err.deref_err(), expected_result); + + let ref_err = &Result::Err::<u32, Vec<i32>>(vec![5, 4, 3, 2, 1]); + let expected_result = Result::Err::<&u32, &[i32]>(&[5, 4, 3, 2, 1][..]); + assert_eq!(ref_err.deref_err(), expected_result); + + // &Result<T: Deref, E: Deref>::Err(T).deref_err() -> + // Result<&T, &E::Deref::Target>::Err(&*E) + let ref_err = &Result::Err::<&u8, &i32>(&41); + let expected_result = Result::Err::<&u8, &i32>(&41); + assert_eq!(ref_err.deref(), expected_result); + + let ref_err = &Result::Err::<&u32, String>(String::from("an error")); + let expected_result = Result::Err::<&u32, &str>("an error"); + assert_eq!(ref_err.deref(), expected_result); + + let ref_err = &Result::Err::<&u32, Vec<i32>>(vec![5, 4, 3, 2, 1]); + let expected_result = Result::Err::<&u32, &[i32]>(&[5, 4, 3, 2, 1][..]); + assert_eq!(ref_err.deref(), expected_result); + + // The following cases test calling deref_* with the wrong variant (i.e. + // `deref_ok()` with a `Result::Err()`, or `deref_err()` with a `Result::Ok()`. + // While unusual, these cases are supported to ensure that an `inner_deref` + // call can still be made even when one of the Result types does not implement + // `Deref` (for example, std::io::Error). + + // &Result<T, E: Deref>::Ok(T).deref_err() -> + // Result<&T, &E::Deref::Target>::Ok(&T) + let ref_ok = &Result::Ok::<i32, &u8>(42); + let expected_result = Result::Ok::<&i32, &u8>(&42); + assert_eq!(ref_ok.deref_err(), expected_result); + + let ref_ok = &Result::Ok::<&str, &u32>("a result"); + let expected_result = Result::Ok::<&&str, &u32>(&"a result"); + assert_eq!(ref_ok.deref_err(), expected_result); + + let ref_ok = &Result::Ok::<[i32; 5], &u32>([1, 2, 3, 4, 5]); + let expected_result = Result::Ok::<&[i32; 5], &u32>(&[1, 2, 3, 4, 5]); + assert_eq!(ref_ok.deref_err(), expected_result); + + // &Result<T: Deref, E>::Err(E).deref_ok() -> + // Result<&T::Deref::Target, &E>::Err(&E) + let ref_err = &Result::Err::<&u8, i32>(41); + let expected_result = Result::Err::<&u8, &i32>(&41); + assert_eq!(ref_err.deref_ok(), expected_result); + + let ref_err = &Result::Err::<&u32, &str>("an error"); + let expected_result = Result::Err::<&u32, &&str>(&"an error"); + assert_eq!(ref_err.deref_ok(), expected_result); + + let ref_err = &Result::Err::<&u32, [i32; 5]>([5, 4, 3, 2, 1]); + let expected_result = Result::Err::<&u32, &[i32; 5]>(&[5, 4, 3, 2, 1]); + assert_eq!(ref_err.deref_ok(), expected_result); +} diff --git a/src/libcore/tests/slice.rs b/src/libcore/tests/slice.rs index 13740b95802..b087ec81f59 100644 --- a/src/libcore/tests/slice.rs +++ b/src/libcore/tests/slice.rs @@ -10,7 +10,6 @@ use core::result::Result::{Ok, Err}; - #[test] fn test_position() { let b = [1, 2, 3, 5, 5]; @@ -61,8 +60,8 @@ fn test_binary_search() { assert_eq!(b.binary_search(&0), Err(0)); assert_eq!(b.binary_search(&1), Ok(0)); assert_eq!(b.binary_search(&2), Err(1)); - assert!(match b.binary_search(&3) { Ok(1...3) => true, _ => false }); - assert!(match b.binary_search(&3) { Ok(1...3) => true, _ => false }); + assert!(match b.binary_search(&3) { Ok(1..=3) => true, _ => false }); + assert!(match b.binary_search(&3) { Ok(1..=3) => true, _ => false }); assert_eq!(b.binary_search(&4), Err(4)); assert_eq!(b.binary_search(&5), Err(4)); assert_eq!(b.binary_search(&6), Err(4)); @@ -261,6 +260,13 @@ fn test_exact_chunks_last() { } #[test] +fn test_exact_chunks_remainder() { + let v: &[i32] = &[0, 1, 2, 3, 4]; + let c = v.exact_chunks(2); + assert_eq!(c.remainder(), &[4]); +} + +#[test] fn test_exact_chunks_zip() { let v1: &[i32] = &[0, 1, 2, 3, 4]; let v2: &[i32] = &[6, 7, 8, 9, 10]; @@ -312,6 +318,13 @@ fn test_exact_chunks_mut_last() { } #[test] +fn test_exact_chunks_mut_remainder() { + let v: &mut [i32] = &mut [0, 1, 2, 3, 4]; + let c = v.exact_chunks_mut(2); + assert_eq!(c.into_remainder(), &[4]); +} + +#[test] fn test_exact_chunks_mut_zip() { let v1: &mut [i32] = &mut [0, 1, 2, 3, 4]; let v2: &[i32] = &[6, 7, 8, 9, 10]; @@ -378,47 +391,349 @@ fn test_windows_zip() { } #[test] -fn get_range() { - let v: &[i32] = &[0, 1, 2, 3, 4, 5]; - assert_eq!(v.get(..), Some(&[0, 1, 2, 3, 4, 5][..])); - assert_eq!(v.get(..2), Some(&[0, 1][..])); - assert_eq!(v.get(2..), Some(&[2, 3, 4, 5][..])); - assert_eq!(v.get(1..4), Some(&[1, 2, 3][..])); - assert_eq!(v.get(7..), None); - assert_eq!(v.get(7..10), None); -} +#[allow(const_err)] +fn test_iter_ref_consistency() { + use std::fmt::Debug; -#[test] -fn get_mut_range() { - let v: &mut [i32] = &mut [0, 1, 2, 3, 4, 5]; - assert_eq!(v.get_mut(..), Some(&mut [0, 1, 2, 3, 4, 5][..])); - assert_eq!(v.get_mut(..2), Some(&mut [0, 1][..])); - assert_eq!(v.get_mut(2..), Some(&mut [2, 3, 4, 5][..])); - assert_eq!(v.get_mut(1..4), Some(&mut [1, 2, 3][..])); - assert_eq!(v.get_mut(7..), None); - assert_eq!(v.get_mut(7..10), None); -} + fn test<T : Copy + Debug + PartialEq>(x : T) { + let v : &[T] = &[x, x, x]; + let v_ptrs : [*const T; 3] = match v { + [ref v1, ref v2, ref v3] => [v1 as *const _, v2 as *const _, v3 as *const _], + _ => unreachable!() + }; + let len = v.len(); -#[test] -fn get_unchecked_range() { - unsafe { - let v: &[i32] = &[0, 1, 2, 3, 4, 5]; - assert_eq!(v.get_unchecked(..), &[0, 1, 2, 3, 4, 5][..]); - assert_eq!(v.get_unchecked(..2), &[0, 1][..]); - assert_eq!(v.get_unchecked(2..), &[2, 3, 4, 5][..]); - assert_eq!(v.get_unchecked(1..4), &[1, 2, 3][..]); + // nth(i) + for i in 0..len { + assert_eq!(&v[i] as *const _, v_ptrs[i]); // check the v_ptrs array, just to be sure + let nth = v.iter().nth(i).unwrap(); + assert_eq!(nth as *const _, v_ptrs[i]); + } + assert_eq!(v.iter().nth(len), None, "nth(len) should return None"); + + // stepping through with nth(0) + { + let mut it = v.iter(); + for i in 0..len { + let next = it.nth(0).unwrap(); + assert_eq!(next as *const _, v_ptrs[i]); + } + assert_eq!(it.nth(0), None); + } + + // next() + { + let mut it = v.iter(); + for i in 0..len { + let remaining = len - i; + assert_eq!(it.size_hint(), (remaining, Some(remaining))); + + let next = it.next().unwrap(); + assert_eq!(next as *const _, v_ptrs[i]); + } + assert_eq!(it.size_hint(), (0, Some(0))); + assert_eq!(it.next(), None, "The final call to next() should return None"); + } + + // next_back() + { + let mut it = v.iter(); + for i in 0..len { + let remaining = len - i; + assert_eq!(it.size_hint(), (remaining, Some(remaining))); + + let prev = it.next_back().unwrap(); + assert_eq!(prev as *const _, v_ptrs[remaining-1]); + } + assert_eq!(it.size_hint(), (0, Some(0))); + assert_eq!(it.next_back(), None, "The final call to next_back() should return None"); + } } -} -#[test] -fn get_unchecked_mut_range() { - unsafe { - let v: &mut [i32] = &mut [0, 1, 2, 3, 4, 5]; - assert_eq!(v.get_unchecked_mut(..), &mut [0, 1, 2, 3, 4, 5][..]); - assert_eq!(v.get_unchecked_mut(..2), &mut [0, 1][..]); - assert_eq!(v.get_unchecked_mut(2..), &mut[2, 3, 4, 5][..]); - assert_eq!(v.get_unchecked_mut(1..4), &mut [1, 2, 3][..]); + fn test_mut<T : Copy + Debug + PartialEq>(x : T) { + let v : &mut [T] = &mut [x, x, x]; + let v_ptrs : [*mut T; 3] = match v { + [ref v1, ref v2, ref v3] => + [v1 as *const _ as *mut _, v2 as *const _ as *mut _, v3 as *const _ as *mut _], + _ => unreachable!() + }; + let len = v.len(); + + // nth(i) + for i in 0..len { + assert_eq!(&mut v[i] as *mut _, v_ptrs[i]); // check the v_ptrs array, just to be sure + let nth = v.iter_mut().nth(i).unwrap(); + assert_eq!(nth as *mut _, v_ptrs[i]); + } + assert_eq!(v.iter().nth(len), None, "nth(len) should return None"); + + // stepping through with nth(0) + { + let mut it = v.iter(); + for i in 0..len { + let next = it.nth(0).unwrap(); + assert_eq!(next as *const _, v_ptrs[i]); + } + assert_eq!(it.nth(0), None); + } + + // next() + { + let mut it = v.iter_mut(); + for i in 0..len { + let remaining = len - i; + assert_eq!(it.size_hint(), (remaining, Some(remaining))); + + let next = it.next().unwrap(); + assert_eq!(next as *mut _, v_ptrs[i]); + } + assert_eq!(it.size_hint(), (0, Some(0))); + assert_eq!(it.next(), None, "The final call to next() should return None"); + } + + // next_back() + { + let mut it = v.iter_mut(); + for i in 0..len { + let remaining = len - i; + assert_eq!(it.size_hint(), (remaining, Some(remaining))); + + let prev = it.next_back().unwrap(); + assert_eq!(prev as *mut _, v_ptrs[remaining-1]); + } + assert_eq!(it.size_hint(), (0, Some(0))); + assert_eq!(it.next_back(), None, "The final call to next_back() should return None"); + } } + + // Make sure iterators and slice patterns yield consistent addresses for various types, + // including ZSTs. + test(0u32); + test(()); + test([0u32; 0]); // ZST with alignment > 0 + test_mut(0u32); + test_mut(()); + test_mut([0u32; 0]); // ZST with alignment > 0 +} + +// The current implementation of SliceIndex fails to handle methods +// orthogonally from range types; therefore, it is worth testing +// all of the indexing operations on each input. +mod slice_index { + // This checks all six indexing methods, given an input range that + // should succeed. (it is NOT suitable for testing invalid inputs) + macro_rules! assert_range_eq { + ($arr:expr, $range:expr, $expected:expr) + => { + let mut arr = $arr; + let mut expected = $expected; + { + let s: &[_] = &arr; + let expected: &[_] = &expected; + + assert_eq!(&s[$range], expected, "(in assertion for: index)"); + assert_eq!(s.get($range), Some(expected), "(in assertion for: get)"); + unsafe { + assert_eq!( + s.get_unchecked($range), expected, + "(in assertion for: get_unchecked)", + ); + } + } + { + let s: &mut [_] = &mut arr; + let expected: &mut [_] = &mut expected; + + assert_eq!( + &mut s[$range], expected, + "(in assertion for: index_mut)", + ); + assert_eq!( + s.get_mut($range), Some(&mut expected[..]), + "(in assertion for: get_mut)", + ); + unsafe { + assert_eq!( + s.get_unchecked_mut($range), expected, + "(in assertion for: get_unchecked_mut)", + ); + } + } + } + } + + // Make sure the macro can actually detect bugs, + // because if it can't, then what are we even doing here? + // + // (Be aware this only demonstrates the ability to detect bugs + // in the FIRST method that panics, as the macro is not designed + // to be used in `should_panic`) + #[test] + #[should_panic(expected = "out of range")] + fn assert_range_eq_can_fail_by_panic() { + assert_range_eq!([0, 1, 2], 0..5, [0, 1, 2]); + } + + // (Be aware this only demonstrates the ability to detect bugs + // in the FIRST method it calls, as the macro is not designed + // to be used in `should_panic`) + #[test] + #[should_panic(expected = "==")] + fn assert_range_eq_can_fail_by_inequality() { + assert_range_eq!([0, 1, 2], 0..2, [0, 1, 2]); + } + + // Test cases for bad index operations. + // + // This generates `should_panic` test cases for Index/IndexMut + // and `None` test cases for get/get_mut. + macro_rules! panic_cases { + ($( + // each test case needs a unique name to namespace the tests + in mod $case_name:ident { + data: $data:expr; + + // optional: + // + // one or more similar inputs for which data[input] succeeds, + // and the corresponding output as an array. This helps validate + // "critical points" where an input range straddles the boundary + // between valid and invalid. + // (such as the input `len..len`, which is just barely valid) + $( + good: data[$good:expr] == $output:expr; + )* + + bad: data[$bad:expr]; + message: $expect_msg:expr; + } + )*) => {$( + mod $case_name { + #[test] + fn pass() { + let mut v = $data; + + $( assert_range_eq!($data, $good, $output); )* + + { + let v: &[_] = &v; + assert_eq!(v.get($bad), None, "(in None assertion for get)"); + } + + { + let v: &mut [_] = &mut v; + assert_eq!(v.get_mut($bad), None, "(in None assertion for get_mut)"); + } + } + + #[test] + #[should_panic(expected = $expect_msg)] + fn index_fail() { + let v = $data; + let v: &[_] = &v; + let _v = &v[$bad]; + } + + #[test] + #[should_panic(expected = $expect_msg)] + fn index_mut_fail() { + let mut v = $data; + let v: &mut [_] = &mut v; + let _v = &mut v[$bad]; + } + } + )*}; + } + + #[test] + fn simple() { + let v = [0, 1, 2, 3, 4, 5]; + + assert_range_eq!(v, .., [0, 1, 2, 3, 4, 5]); + assert_range_eq!(v, ..2, [0, 1]); + assert_range_eq!(v, ..=1, [0, 1]); + assert_range_eq!(v, 2.., [2, 3, 4, 5]); + assert_range_eq!(v, 1..4, [1, 2, 3]); + assert_range_eq!(v, 1..=3, [1, 2, 3]); + } + + panic_cases! { + in mod rangefrom_len { + data: [0, 1, 2, 3, 4, 5]; + + good: data[6..] == []; + bad: data[7..]; + message: "but ends at"; // perhaps not ideal + } + + in mod rangeto_len { + data: [0, 1, 2, 3, 4, 5]; + + good: data[..6] == [0, 1, 2, 3, 4, 5]; + bad: data[..7]; + message: "out of range"; + } + + in mod rangetoinclusive_len { + data: [0, 1, 2, 3, 4, 5]; + + good: data[..=5] == [0, 1, 2, 3, 4, 5]; + bad: data[..=6]; + message: "out of range"; + } + + in mod range_len_len { + data: [0, 1, 2, 3, 4, 5]; + + good: data[6..6] == []; + bad: data[7..7]; + message: "out of range"; + } + + in mod rangeinclusive_len_len { + data: [0, 1, 2, 3, 4, 5]; + + good: data[6..=5] == []; + bad: data[7..=6]; + message: "out of range"; + } + } + + panic_cases! { + in mod range_neg_width { + data: [0, 1, 2, 3, 4, 5]; + + good: data[4..4] == []; + bad: data[4..3]; + message: "but ends at"; + } + + in mod rangeinclusive_neg_width { + data: [0, 1, 2, 3, 4, 5]; + + good: data[4..=3] == []; + bad: data[4..=2]; + message: "but ends at"; + } + } + + panic_cases! { + in mod rangeinclusive_overflow { + data: [0, 1]; + + // note: using 0 specifically ensures that the result of overflowing is 0..0, + // so that `get` doesn't simply return None for the wrong reason. + bad: data[0 ..= ::std::usize::MAX]; + message: "maximum usize"; + } + + in mod rangetoinclusive_overflow { + data: [0, 1]; + + bad: data[..= ::std::usize::MAX]; + message: "maximum usize"; + } + } // panic_cases! } #[test] @@ -481,3 +796,207 @@ fn test_rotate_right() { assert_eq!(a[(i + 42) % N], i); } } + +#[test] +#[cfg(not(target_arch = "wasm32"))] +fn sort_unstable() { + use core::cmp::Ordering::{Equal, Greater, Less}; + use core::slice::heapsort; + use rand::{Rng, XorShiftRng}; + + let mut v = [0; 600]; + let mut tmp = [0; 600]; + let mut rng = XorShiftRng::new_unseeded(); + + for len in (2..25).chain(500..510) { + let v = &mut v[0..len]; + let tmp = &mut tmp[0..len]; + + for &modulus in &[5, 10, 100, 1000] { + for _ in 0..100 { + for i in 0..len { + v[i] = rng.gen::<i32>() % modulus; + } + + // Sort in default order. + tmp.copy_from_slice(v); + tmp.sort_unstable(); + assert!(tmp.windows(2).all(|w| w[0] <= w[1])); + + // Sort in ascending order. + tmp.copy_from_slice(v); + tmp.sort_unstable_by(|a, b| a.cmp(b)); + assert!(tmp.windows(2).all(|w| w[0] <= w[1])); + + // Sort in descending order. + tmp.copy_from_slice(v); + tmp.sort_unstable_by(|a, b| b.cmp(a)); + assert!(tmp.windows(2).all(|w| w[0] >= w[1])); + + // Test heapsort using `<` operator. + tmp.copy_from_slice(v); + heapsort(tmp, |a, b| a < b); + assert!(tmp.windows(2).all(|w| w[0] <= w[1])); + + // Test heapsort using `>` operator. + tmp.copy_from_slice(v); + heapsort(tmp, |a, b| a > b); + assert!(tmp.windows(2).all(|w| w[0] >= w[1])); + } + } + } + + // Sort using a completely random comparison function. + // This will reorder the elements *somehow*, but won't panic. + for i in 0..v.len() { + v[i] = i as i32; + } + v.sort_unstable_by(|_, _| *rng.choose(&[Less, Equal, Greater]).unwrap()); + v.sort_unstable(); + for i in 0..v.len() { + assert_eq!(v[i], i as i32); + } + + // Should not panic. + [0i32; 0].sort_unstable(); + [(); 10].sort_unstable(); + [(); 100].sort_unstable(); + + let mut v = [0xDEADBEEFu64]; + v.sort_unstable(); + assert!(v == [0xDEADBEEF]); +} + +pub mod memchr { + use core::slice::memchr::{memchr, memrchr}; + + // test fallback implementations on all platforms + #[test] + fn matches_one() { + assert_eq!(Some(0), memchr(b'a', b"a")); + } + + #[test] + fn matches_begin() { + assert_eq!(Some(0), memchr(b'a', b"aaaa")); + } + + #[test] + fn matches_end() { + assert_eq!(Some(4), memchr(b'z', b"aaaaz")); + } + + #[test] + fn matches_nul() { + assert_eq!(Some(4), memchr(b'\x00', b"aaaa\x00")); + } + + #[test] + fn matches_past_nul() { + assert_eq!(Some(5), memchr(b'z', b"aaaa\x00z")); + } + + #[test] + fn no_match_empty() { + assert_eq!(None, memchr(b'a', b"")); + } + + #[test] + fn no_match() { + assert_eq!(None, memchr(b'a', b"xyz")); + } + + #[test] + fn matches_one_reversed() { + assert_eq!(Some(0), memrchr(b'a', b"a")); + } + + #[test] + fn matches_begin_reversed() { + assert_eq!(Some(3), memrchr(b'a', b"aaaa")); + } + + #[test] + fn matches_end_reversed() { + assert_eq!(Some(0), memrchr(b'z', b"zaaaa")); + } + + #[test] + fn matches_nul_reversed() { + assert_eq!(Some(4), memrchr(b'\x00', b"aaaa\x00")); + } + + #[test] + fn matches_past_nul_reversed() { + assert_eq!(Some(0), memrchr(b'z', b"z\x00aaaa")); + } + + #[test] + fn no_match_empty_reversed() { + assert_eq!(None, memrchr(b'a', b"")); + } + + #[test] + fn no_match_reversed() { + assert_eq!(None, memrchr(b'a', b"xyz")); + } + + #[test] + fn each_alignment_reversed() { + let mut data = [1u8; 64]; + let needle = 2; + let pos = 40; + data[pos] = needle; + for start in 0..16 { + assert_eq!(Some(pos - start), memrchr(needle, &data[start..])); + } + } +} + +#[test] +fn test_align_to_simple() { + let bytes = [1u8, 2, 3, 4, 5, 6, 7]; + let (prefix, aligned, suffix) = unsafe { bytes.align_to::<u16>() }; + assert_eq!(aligned.len(), 3); + assert!(prefix == [1] || suffix == [7]); + let expect1 = [1 << 8 | 2, 3 << 8 | 4, 5 << 8 | 6]; + let expect2 = [1 | 2 << 8, 3 | 4 << 8, 5 | 6 << 8]; + let expect3 = [2 << 8 | 3, 4 << 8 | 5, 6 << 8 | 7]; + let expect4 = [2 | 3 << 8, 4 | 5 << 8, 6 | 7 << 8]; + assert!(aligned == expect1 || aligned == expect2 || aligned == expect3 || aligned == expect4, + "aligned={:?} expected={:?} || {:?} || {:?} || {:?}", + aligned, expect1, expect2, expect3, expect4); +} + +#[test] +fn test_align_to_zst() { + let bytes = [1, 2, 3, 4, 5, 6, 7]; + let (prefix, aligned, suffix) = unsafe { bytes.align_to::<()>() }; + assert_eq!(aligned.len(), 0); + assert!(prefix == [1, 2, 3, 4, 5, 6, 7] || suffix == [1, 2, 3, 4, 5, 6, 7]); +} + +#[test] +fn test_align_to_non_trivial() { + #[repr(align(8))] struct U64(u64, u64); + #[repr(align(8))] struct U64U64U32(u64, u64, u32); + let data = [U64(1, 2), U64(3, 4), U64(5, 6), U64(7, 8), U64(9, 10), U64(11, 12), U64(13, 14), + U64(15, 16)]; + let (prefix, aligned, suffix) = unsafe { data.align_to::<U64U64U32>() }; + assert_eq!(aligned.len(), 4); + assert_eq!(prefix.len() + suffix.len(), 2); +} + +#[test] +fn test_align_to_empty_mid() { + use core::mem; + + // Make sure that we do not create empty unaligned slices for the mid part, even when the + // overall slice is too short to contain an aligned address. + let bytes = [1, 2, 3, 4, 5, 6, 7]; + type Chunk = u32; + for offset in 0..4 { + let (_, mid, _) = unsafe { bytes[offset..offset+1].align_to::<Chunk>() }; + assert_eq!(mid.as_ptr() as usize % mem::align_of::<Chunk>(), 0); + } +} diff --git a/src/libcore/tests/str.rs b/src/libcore/tests/str.rs index 08daafccc54..343c9596c53 100644 --- a/src/libcore/tests/str.rs +++ b/src/libcore/tests/str.rs @@ -8,4 +8,4 @@ // option. This file may not be copied, modified, or distributed // except according to those terms. -// All `str` tests live in collectionstests::str +// All `str` tests live in liballoc/tests diff --git a/src/libcore/tests/str_lossy.rs b/src/libcore/tests/str_lossy.rs new file mode 100644 index 00000000000..56ef3f070c1 --- /dev/null +++ b/src/libcore/tests/str_lossy.rs @@ -0,0 +1,91 @@ +// Copyright 2012-2017 The Rust Project Developers. See the COPYRIGHT +// file at the top-level directory of this distribution and at +// http://rust-lang.org/COPYRIGHT. +// +// Licensed under the Apache License, Version 2.0 <LICENSE-APACHE or +// http://www.apache.org/licenses/LICENSE-2.0> or the MIT license +// <LICENSE-MIT or http://opensource.org/licenses/MIT>, at your +// option. This file may not be copied, modified, or distributed +// except according to those terms. + +use core::str::lossy::*; + +#[test] +fn chunks() { + let mut iter = Utf8Lossy::from_bytes(b"hello").chunks(); + assert_eq!(Some(Utf8LossyChunk { valid: "hello", broken: b"", }), iter.next()); + assert_eq!(None, iter.next()); + + let mut iter = Utf8Lossy::from_bytes("ศไทย中华Việt Nam".as_bytes()).chunks(); + assert_eq!(Some(Utf8LossyChunk { valid: "ศไทย中华Việt Nam", broken: b"", }), iter.next()); + assert_eq!(None, iter.next()); + + let mut iter = Utf8Lossy::from_bytes(b"Hello\xC2 There\xFF Goodbye").chunks(); + assert_eq!(Some(Utf8LossyChunk { valid: "Hello", broken: b"\xC2", }), iter.next()); + assert_eq!(Some(Utf8LossyChunk { valid: " There", broken: b"\xFF", }), iter.next()); + assert_eq!(Some(Utf8LossyChunk { valid: " Goodbye", broken: b"", }), iter.next()); + assert_eq!(None, iter.next()); + + let mut iter = Utf8Lossy::from_bytes(b"Hello\xC0\x80 There\xE6\x83 Goodbye").chunks(); + assert_eq!(Some(Utf8LossyChunk { valid: "Hello", broken: b"\xC0", }), iter.next()); + assert_eq!(Some(Utf8LossyChunk { valid: "", broken: b"\x80", }), iter.next()); + assert_eq!(Some(Utf8LossyChunk { valid: " There", broken: b"\xE6\x83", }), iter.next()); + assert_eq!(Some(Utf8LossyChunk { valid: " Goodbye", broken: b"", }), iter.next()); + assert_eq!(None, iter.next()); + + let mut iter = Utf8Lossy::from_bytes(b"\xF5foo\xF5\x80bar").chunks(); + assert_eq!(Some(Utf8LossyChunk { valid: "", broken: b"\xF5", }), iter.next()); + assert_eq!(Some(Utf8LossyChunk { valid: "foo", broken: b"\xF5", }), iter.next()); + assert_eq!(Some(Utf8LossyChunk { valid: "", broken: b"\x80", }), iter.next()); + assert_eq!(Some(Utf8LossyChunk { valid: "bar", broken: b"", }), iter.next()); + assert_eq!(None, iter.next()); + + let mut iter = Utf8Lossy::from_bytes(b"\xF1foo\xF1\x80bar\xF1\x80\x80baz").chunks(); + assert_eq!(Some(Utf8LossyChunk { valid: "", broken: b"\xF1", }), iter.next()); + assert_eq!(Some(Utf8LossyChunk { valid: "foo", broken: b"\xF1\x80", }), iter.next()); + assert_eq!(Some(Utf8LossyChunk { valid: "bar", broken: b"\xF1\x80\x80", }), iter.next()); + assert_eq!(Some(Utf8LossyChunk { valid: "baz", broken: b"", }), iter.next()); + assert_eq!(None, iter.next()); + + let mut iter = Utf8Lossy::from_bytes(b"\xF4foo\xF4\x80bar\xF4\xBFbaz").chunks(); + assert_eq!(Some(Utf8LossyChunk { valid: "", broken: b"\xF4", }), iter.next()); + assert_eq!(Some(Utf8LossyChunk { valid: "foo", broken: b"\xF4\x80", }), iter.next()); + assert_eq!(Some(Utf8LossyChunk { valid: "bar", broken: b"\xF4", }), iter.next()); + assert_eq!(Some(Utf8LossyChunk { valid: "", broken: b"\xBF", }), iter.next()); + assert_eq!(Some(Utf8LossyChunk { valid: "baz", broken: b"", }), iter.next()); + assert_eq!(None, iter.next()); + + let mut iter = Utf8Lossy::from_bytes(b"\xF0\x80\x80\x80foo\xF0\x90\x80\x80bar").chunks(); + assert_eq!(Some(Utf8LossyChunk { valid: "", broken: b"\xF0", }), iter.next()); + assert_eq!(Some(Utf8LossyChunk { valid: "", broken: b"\x80", }), iter.next()); + assert_eq!(Some(Utf8LossyChunk { valid: "", broken: b"\x80", }), iter.next()); + assert_eq!(Some(Utf8LossyChunk { valid: "", broken: b"\x80", }), iter.next()); + assert_eq!(Some(Utf8LossyChunk { valid: "foo\u{10000}bar", broken: b"", }), iter.next()); + assert_eq!(None, iter.next()); + + // surrogates + let mut iter = Utf8Lossy::from_bytes(b"\xED\xA0\x80foo\xED\xBF\xBFbar").chunks(); + assert_eq!(Some(Utf8LossyChunk { valid: "", broken: b"\xED", }), iter.next()); + assert_eq!(Some(Utf8LossyChunk { valid: "", broken: b"\xA0", }), iter.next()); + assert_eq!(Some(Utf8LossyChunk { valid: "", broken: b"\x80", }), iter.next()); + assert_eq!(Some(Utf8LossyChunk { valid: "foo", broken: b"\xED", }), iter.next()); + assert_eq!(Some(Utf8LossyChunk { valid: "", broken: b"\xBF", }), iter.next()); + assert_eq!(Some(Utf8LossyChunk { valid: "", broken: b"\xBF", }), iter.next()); + assert_eq!(Some(Utf8LossyChunk { valid: "bar", broken: b"", }), iter.next()); + assert_eq!(None, iter.next()); +} + +#[test] +fn display() { + assert_eq!( + "Hello\u{FFFD}\u{FFFD} There\u{FFFD} Goodbye", + &Utf8Lossy::from_bytes(b"Hello\xC0\x80 There\xE6\x83 Goodbye").to_string()); +} + +#[test] +fn debug() { + assert_eq!( + "\"Hello\\xc0\\x80 There\\xe6\\x83 Goodbye\\u{10d4ea}\"", + &format!("{:?}", Utf8Lossy::from_bytes( + b"Hello\xC0\x80 There\xE6\x83 Goodbye\xf4\x8d\x93\xaa"))); +} diff --git a/src/libcore/tests/time.rs b/src/libcore/tests/time.rs new file mode 100644 index 00000000000..466f28f0ef0 --- /dev/null +++ b/src/libcore/tests/time.rs @@ -0,0 +1,304 @@ +// Copyright 2018 The Rust Project Developers. See the COPYRIGHT +// file at the top-level directory of this distribution and at +// http://rust-lang.org/COPYRIGHT. +// +// Licensed under the Apache License, Version 2.0 <LICENSE-APACHE or +// http://www.apache.org/licenses/LICENSE-2.0> or the MIT license +// <LICENSE-MIT or http://opensource.org/licenses/MIT>, at your +// option. This file may not be copied, modified, or distributed +// except according to those terms. + +use core::time::Duration; + +#[test] +fn creation() { + assert!(Duration::from_secs(1) != Duration::from_secs(0)); + assert_eq!(Duration::from_secs(1) + Duration::from_secs(2), + Duration::from_secs(3)); + assert_eq!(Duration::from_millis(10) + Duration::from_secs(4), + Duration::new(4, 10 * 1_000_000)); + assert_eq!(Duration::from_millis(4000), Duration::new(4, 0)); +} + +#[test] +fn secs() { + assert_eq!(Duration::new(0, 0).as_secs(), 0); + assert_eq!(Duration::new(0, 500_000_005).as_secs(), 0); + assert_eq!(Duration::new(0, 1_050_000_001).as_secs(), 1); + assert_eq!(Duration::from_secs(1).as_secs(), 1); + assert_eq!(Duration::from_millis(999).as_secs(), 0); + assert_eq!(Duration::from_millis(1001).as_secs(), 1); + assert_eq!(Duration::from_micros(999_999).as_secs(), 0); + assert_eq!(Duration::from_micros(1_000_001).as_secs(), 1); + assert_eq!(Duration::from_nanos(999_999_999).as_secs(), 0); + assert_eq!(Duration::from_nanos(1_000_000_001).as_secs(), 1); +} + +#[test] +fn millis() { + assert_eq!(Duration::new(0, 0).subsec_millis(), 0); + assert_eq!(Duration::new(0, 500_000_005).subsec_millis(), 500); + assert_eq!(Duration::new(0, 1_050_000_001).subsec_millis(), 50); + assert_eq!(Duration::from_secs(1).subsec_millis(), 0); + assert_eq!(Duration::from_millis(999).subsec_millis(), 999); + assert_eq!(Duration::from_millis(1001).subsec_millis(), 1); + assert_eq!(Duration::from_micros(999_999).subsec_millis(), 999); + assert_eq!(Duration::from_micros(1_001_000).subsec_millis(), 1); + assert_eq!(Duration::from_nanos(999_999_999).subsec_millis(), 999); + assert_eq!(Duration::from_nanos(1_001_000_000).subsec_millis(), 1); +} + +#[test] +fn micros() { + assert_eq!(Duration::new(0, 0).subsec_micros(), 0); + assert_eq!(Duration::new(0, 500_000_005).subsec_micros(), 500_000); + assert_eq!(Duration::new(0, 1_050_000_001).subsec_micros(), 50_000); + assert_eq!(Duration::from_secs(1).subsec_micros(), 0); + assert_eq!(Duration::from_millis(999).subsec_micros(), 999_000); + assert_eq!(Duration::from_millis(1001).subsec_micros(), 1_000); + assert_eq!(Duration::from_micros(999_999).subsec_micros(), 999_999); + assert_eq!(Duration::from_micros(1_000_001).subsec_micros(), 1); + assert_eq!(Duration::from_nanos(999_999_999).subsec_micros(), 999_999); + assert_eq!(Duration::from_nanos(1_000_001_000).subsec_micros(), 1); +} + +#[test] +fn nanos() { + assert_eq!(Duration::new(0, 0).subsec_nanos(), 0); + assert_eq!(Duration::new(0, 5).subsec_nanos(), 5); + assert_eq!(Duration::new(0, 1_000_000_001).subsec_nanos(), 1); + assert_eq!(Duration::from_secs(1).subsec_nanos(), 0); + assert_eq!(Duration::from_millis(999).subsec_nanos(), 999_000_000); + assert_eq!(Duration::from_millis(1001).subsec_nanos(), 1_000_000); + assert_eq!(Duration::from_micros(999_999).subsec_nanos(), 999_999_000); + assert_eq!(Duration::from_micros(1_000_001).subsec_nanos(), 1000); + assert_eq!(Duration::from_nanos(999_999_999).subsec_nanos(), 999_999_999); + assert_eq!(Duration::from_nanos(1_000_000_001).subsec_nanos(), 1); +} + +#[test] +fn add() { + assert_eq!(Duration::new(0, 0) + Duration::new(0, 1), + Duration::new(0, 1)); + assert_eq!(Duration::new(0, 500_000_000) + Duration::new(0, 500_000_001), + Duration::new(1, 1)); +} + +#[test] +fn checked_add() { + assert_eq!(Duration::new(0, 0).checked_add(Duration::new(0, 1)), + Some(Duration::new(0, 1))); + assert_eq!(Duration::new(0, 500_000_000).checked_add(Duration::new(0, 500_000_001)), + Some(Duration::new(1, 1))); + assert_eq!(Duration::new(1, 0).checked_add(Duration::new(::core::u64::MAX, 0)), None); +} + +#[test] +fn sub() { + assert_eq!(Duration::new(0, 1) - Duration::new(0, 0), + Duration::new(0, 1)); + assert_eq!(Duration::new(0, 500_000_001) - Duration::new(0, 500_000_000), + Duration::new(0, 1)); + assert_eq!(Duration::new(1, 0) - Duration::new(0, 1), + Duration::new(0, 999_999_999)); +} + +#[test] +fn checked_sub() { + let zero = Duration::new(0, 0); + let one_nano = Duration::new(0, 1); + let one_sec = Duration::new(1, 0); + assert_eq!(one_nano.checked_sub(zero), Some(Duration::new(0, 1))); + assert_eq!(one_sec.checked_sub(one_nano), + Some(Duration::new(0, 999_999_999))); + assert_eq!(zero.checked_sub(one_nano), None); + assert_eq!(zero.checked_sub(one_sec), None); +} + +#[test] +#[should_panic] +fn sub_bad1() { + let _ = Duration::new(0, 0) - Duration::new(0, 1); +} + +#[test] +#[should_panic] +fn sub_bad2() { + let _ = Duration::new(0, 0) - Duration::new(1, 0); +} + +#[test] +fn mul() { + assert_eq!(Duration::new(0, 1) * 2, Duration::new(0, 2)); + assert_eq!(Duration::new(1, 1) * 3, Duration::new(3, 3)); + assert_eq!(Duration::new(0, 500_000_001) * 4, Duration::new(2, 4)); + assert_eq!(Duration::new(0, 500_000_001) * 4000, + Duration::new(2000, 4000)); +} + +#[test] +fn checked_mul() { + assert_eq!(Duration::new(0, 1).checked_mul(2), Some(Duration::new(0, 2))); + assert_eq!(Duration::new(1, 1).checked_mul(3), Some(Duration::new(3, 3))); + assert_eq!(Duration::new(0, 500_000_001).checked_mul(4), Some(Duration::new(2, 4))); + assert_eq!(Duration::new(0, 500_000_001).checked_mul(4000), + Some(Duration::new(2000, 4000))); + assert_eq!(Duration::new(::core::u64::MAX - 1, 0).checked_mul(2), None); +} + +#[test] +fn div() { + assert_eq!(Duration::new(0, 1) / 2, Duration::new(0, 0)); + assert_eq!(Duration::new(1, 1) / 3, Duration::new(0, 333_333_333)); + assert_eq!(Duration::new(99, 999_999_000) / 100, + Duration::new(0, 999_999_990)); +} + +#[test] +fn checked_div() { + assert_eq!(Duration::new(2, 0).checked_div(2), Some(Duration::new(1, 0))); + assert_eq!(Duration::new(1, 0).checked_div(2), Some(Duration::new(0, 500_000_000))); + assert_eq!(Duration::new(2, 0).checked_div(0), None); +} + +#[test] +fn correct_sum() { + let durations = [ + Duration::new(1, 999_999_999), + Duration::new(2, 999_999_999), + Duration::new(0, 999_999_999), + Duration::new(0, 999_999_999), + Duration::new(0, 999_999_999), + Duration::new(5, 0), + ]; + let sum = durations.iter().sum::<Duration>(); + assert_eq!(sum, Duration::new(1+2+5+4, 1_000_000_000 - 5)); +} + +#[test] +fn debug_formatting_extreme_values() { + assert_eq!( + format!("{:?}", Duration::new(18_446_744_073_709_551_615, 123_456_789)), + "18446744073709551615.123456789s" + ); +} + +#[test] +fn debug_formatting_secs() { + assert_eq!(format!("{:?}", Duration::new(7, 000_000_000)), "7s"); + assert_eq!(format!("{:?}", Duration::new(7, 100_000_000)), "7.1s"); + assert_eq!(format!("{:?}", Duration::new(7, 000_010_000)), "7.00001s"); + assert_eq!(format!("{:?}", Duration::new(7, 000_000_001)), "7.000000001s"); + assert_eq!(format!("{:?}", Duration::new(7, 123_456_789)), "7.123456789s"); + + assert_eq!(format!("{:?}", Duration::new(88, 000_000_000)), "88s"); + assert_eq!(format!("{:?}", Duration::new(88, 100_000_000)), "88.1s"); + assert_eq!(format!("{:?}", Duration::new(88, 000_010_000)), "88.00001s"); + assert_eq!(format!("{:?}", Duration::new(88, 000_000_001)), "88.000000001s"); + assert_eq!(format!("{:?}", Duration::new(88, 123_456_789)), "88.123456789s"); + + assert_eq!(format!("{:?}", Duration::new(999, 000_000_000)), "999s"); + assert_eq!(format!("{:?}", Duration::new(999, 100_000_000)), "999.1s"); + assert_eq!(format!("{:?}", Duration::new(999, 000_010_000)), "999.00001s"); + assert_eq!(format!("{:?}", Duration::new(999, 000_000_001)), "999.000000001s"); + assert_eq!(format!("{:?}", Duration::new(999, 123_456_789)), "999.123456789s"); +} + +#[test] +fn debug_formatting_millis() { + assert_eq!(format!("{:?}", Duration::new(0, 7_000_000)), "7ms"); + assert_eq!(format!("{:?}", Duration::new(0, 7_100_000)), "7.1ms"); + assert_eq!(format!("{:?}", Duration::new(0, 7_000_001)), "7.000001ms"); + assert_eq!(format!("{:?}", Duration::new(0, 7_123_456)), "7.123456ms"); + + assert_eq!(format!("{:?}", Duration::new(0, 88_000_000)), "88ms"); + assert_eq!(format!("{:?}", Duration::new(0, 88_100_000)), "88.1ms"); + assert_eq!(format!("{:?}", Duration::new(0, 88_000_001)), "88.000001ms"); + assert_eq!(format!("{:?}", Duration::new(0, 88_123_456)), "88.123456ms"); + + assert_eq!(format!("{:?}", Duration::new(0, 999_000_000)), "999ms"); + assert_eq!(format!("{:?}", Duration::new(0, 999_100_000)), "999.1ms"); + assert_eq!(format!("{:?}", Duration::new(0, 999_000_001)), "999.000001ms"); + assert_eq!(format!("{:?}", Duration::new(0, 999_123_456)), "999.123456ms"); +} + +#[test] +fn debug_formatting_micros() { + assert_eq!(format!("{:?}", Duration::new(0, 7_000)), "7µs"); + assert_eq!(format!("{:?}", Duration::new(0, 7_100)), "7.1µs"); + assert_eq!(format!("{:?}", Duration::new(0, 7_001)), "7.001µs"); + assert_eq!(format!("{:?}", Duration::new(0, 7_123)), "7.123µs"); + + assert_eq!(format!("{:?}", Duration::new(0, 88_000)), "88µs"); + assert_eq!(format!("{:?}", Duration::new(0, 88_100)), "88.1µs"); + assert_eq!(format!("{:?}", Duration::new(0, 88_001)), "88.001µs"); + assert_eq!(format!("{:?}", Duration::new(0, 88_123)), "88.123µs"); + + assert_eq!(format!("{:?}", Duration::new(0, 999_000)), "999µs"); + assert_eq!(format!("{:?}", Duration::new(0, 999_100)), "999.1µs"); + assert_eq!(format!("{:?}", Duration::new(0, 999_001)), "999.001µs"); + assert_eq!(format!("{:?}", Duration::new(0, 999_123)), "999.123µs"); +} + +#[test] +fn debug_formatting_nanos() { + assert_eq!(format!("{:?}", Duration::new(0, 0)), "0ns"); + assert_eq!(format!("{:?}", Duration::new(0, 1)), "1ns"); + assert_eq!(format!("{:?}", Duration::new(0, 88)), "88ns"); + assert_eq!(format!("{:?}", Duration::new(0, 999)), "999ns"); +} + +#[test] +fn debug_formatting_precision_zero() { + assert_eq!(format!("{:.0?}", Duration::new(0, 0)), "0ns"); + assert_eq!(format!("{:.0?}", Duration::new(0, 123)), "123ns"); + + assert_eq!(format!("{:.0?}", Duration::new(0, 1_001)), "1µs"); + assert_eq!(format!("{:.0?}", Duration::new(0, 1_499)), "1µs"); + assert_eq!(format!("{:.0?}", Duration::new(0, 1_500)), "2µs"); + assert_eq!(format!("{:.0?}", Duration::new(0, 1_999)), "2µs"); + + assert_eq!(format!("{:.0?}", Duration::new(0, 1_000_001)), "1ms"); + assert_eq!(format!("{:.0?}", Duration::new(0, 1_499_999)), "1ms"); + assert_eq!(format!("{:.0?}", Duration::new(0, 1_500_000)), "2ms"); + assert_eq!(format!("{:.0?}", Duration::new(0, 1_999_999)), "2ms"); + + assert_eq!(format!("{:.0?}", Duration::new(1, 000_000_001)), "1s"); + assert_eq!(format!("{:.0?}", Duration::new(1, 499_999_999)), "1s"); + assert_eq!(format!("{:.0?}", Duration::new(1, 500_000_000)), "2s"); + assert_eq!(format!("{:.0?}", Duration::new(1, 999_999_999)), "2s"); +} + +#[test] +fn debug_formatting_precision_two() { + assert_eq!(format!("{:.2?}", Duration::new(0, 0)), "0.00ns"); + assert_eq!(format!("{:.2?}", Duration::new(0, 123)), "123.00ns"); + + assert_eq!(format!("{:.2?}", Duration::new(0, 1_000)), "1.00µs"); + assert_eq!(format!("{:.2?}", Duration::new(0, 7_001)), "7.00µs"); + assert_eq!(format!("{:.2?}", Duration::new(0, 7_100)), "7.10µs"); + assert_eq!(format!("{:.2?}", Duration::new(0, 7_109)), "7.11µs"); + assert_eq!(format!("{:.2?}", Duration::new(0, 7_199)), "7.20µs"); + assert_eq!(format!("{:.2?}", Duration::new(0, 1_999)), "2.00µs"); + + assert_eq!(format!("{:.2?}", Duration::new(0, 1_000_000)), "1.00ms"); + assert_eq!(format!("{:.2?}", Duration::new(0, 3_001_000)), "3.00ms"); + assert_eq!(format!("{:.2?}", Duration::new(0, 3_100_000)), "3.10ms"); + assert_eq!(format!("{:.2?}", Duration::new(0, 1_999_999)), "2.00ms"); + + assert_eq!(format!("{:.2?}", Duration::new(1, 000_000_000)), "1.00s"); + assert_eq!(format!("{:.2?}", Duration::new(4, 001_000_000)), "4.00s"); + assert_eq!(format!("{:.2?}", Duration::new(2, 100_000_000)), "2.10s"); + assert_eq!(format!("{:.2?}", Duration::new(2, 104_990_000)), "2.10s"); + assert_eq!(format!("{:.2?}", Duration::new(2, 105_000_000)), "2.11s"); + assert_eq!(format!("{:.2?}", Duration::new(8, 999_999_999)), "9.00s"); +} + +#[test] +fn debug_formatting_precision_high() { + assert_eq!(format!("{:.5?}", Duration::new(0, 23_678)), "23.67800µs"); + + assert_eq!(format!("{:.9?}", Duration::new(1, 000_000_000)), "1.000000000s"); + assert_eq!(format!("{:.10?}", Duration::new(4, 001_000_000)), "4.0010000000s"); + assert_eq!(format!("{:.20?}", Duration::new(4, 001_000_000)), "4.00100000000000000000s"); +} diff --git a/src/libcore/time.rs b/src/libcore/time.rs new file mode 100644 index 00000000000..b58920224eb --- /dev/null +++ b/src/libcore/time.rs @@ -0,0 +1,685 @@ +// Copyright 2012-2017 The Rust Project Developers. See the COPYRIGHT +// file at the top-level directory of this distribution and at +// http://rust-lang.org/COPYRIGHT. +// +// Licensed under the Apache License, Version 2.0 <LICENSE-APACHE or +// http://www.apache.org/licenses/LICENSE-2.0> or the MIT license +// <LICENSE-MIT or http://opensource.org/licenses/MIT>, at your +// option. This file may not be copied, modified, or distributed +// except according to those terms. +#![stable(feature = "duration_core", since = "1.25.0")] + +//! Temporal quantification. +//! +//! Example: +//! +//! ``` +//! use std::time::Duration; +//! +//! let five_seconds = Duration::new(5, 0); +//! // both declarations are equivalent +//! assert_eq!(Duration::new(5, 0), Duration::from_secs(5)); +//! ``` + +use fmt; +use iter::Sum; +use ops::{Add, Sub, Mul, Div, AddAssign, SubAssign, MulAssign, DivAssign}; + +const NANOS_PER_SEC: u32 = 1_000_000_000; +const NANOS_PER_MILLI: u32 = 1_000_000; +const NANOS_PER_MICRO: u32 = 1_000; +const MILLIS_PER_SEC: u64 = 1_000; +const MICROS_PER_SEC: u64 = 1_000_000; + +/// A `Duration` type to represent a span of time, typically used for system +/// timeouts. +/// +/// Each `Duration` is composed of a whole number of seconds and a fractional part +/// represented in nanoseconds. If the underlying system does not support +/// nanosecond-level precision, APIs binding a system timeout will typically round up +/// the number of nanoseconds. +/// +/// `Duration`s implement many common traits, including [`Add`], [`Sub`], and other +/// [`ops`] traits. +/// +/// [`Add`]: ../../std/ops/trait.Add.html +/// [`Sub`]: ../../std/ops/trait.Sub.html +/// [`ops`]: ../../std/ops/index.html +/// +/// # Examples +/// +/// ``` +/// use std::time::Duration; +/// +/// let five_seconds = Duration::new(5, 0); +/// let five_seconds_and_five_nanos = five_seconds + Duration::new(0, 5); +/// +/// assert_eq!(five_seconds_and_five_nanos.as_secs(), 5); +/// assert_eq!(five_seconds_and_five_nanos.subsec_nanos(), 5); +/// +/// let ten_millis = Duration::from_millis(10); +/// ``` +#[stable(feature = "duration", since = "1.3.0")] +#[derive(Clone, Copy, PartialEq, Eq, PartialOrd, Ord, Hash, Default)] +pub struct Duration { + secs: u64, + nanos: u32, // Always 0 <= nanos < NANOS_PER_SEC +} + +impl Duration { + /// Creates a new `Duration` from the specified number of whole seconds and + /// additional nanoseconds. + /// + /// If the number of nanoseconds is greater than 1 billion (the number of + /// nanoseconds in a second), then it will carry over into the seconds provided. + /// + /// # Panics + /// + /// This constructor will panic if the carry from the nanoseconds overflows + /// the seconds counter. + /// + /// # Examples + /// + /// ``` + /// use std::time::Duration; + /// + /// let five_seconds = Duration::new(5, 0); + /// ``` + #[stable(feature = "duration", since = "1.3.0")] + #[inline] + pub fn new(secs: u64, nanos: u32) -> Duration { + let secs = secs.checked_add((nanos / NANOS_PER_SEC) as u64) + .expect("overflow in Duration::new"); + let nanos = nanos % NANOS_PER_SEC; + Duration { secs, nanos } + } + + /// Creates a new `Duration` from the specified number of whole seconds. + /// + /// # Examples + /// + /// ``` + /// use std::time::Duration; + /// + /// let duration = Duration::from_secs(5); + /// + /// assert_eq!(5, duration.as_secs()); + /// assert_eq!(0, duration.subsec_nanos()); + /// ``` + #[stable(feature = "duration", since = "1.3.0")] + #[inline] + pub const fn from_secs(secs: u64) -> Duration { + Duration { secs, nanos: 0 } + } + + /// Creates a new `Duration` from the specified number of milliseconds. + /// + /// # Examples + /// + /// ``` + /// use std::time::Duration; + /// + /// let duration = Duration::from_millis(2569); + /// + /// assert_eq!(2, duration.as_secs()); + /// assert_eq!(569_000_000, duration.subsec_nanos()); + /// ``` + #[stable(feature = "duration", since = "1.3.0")] + #[inline] + pub const fn from_millis(millis: u64) -> Duration { + Duration { + secs: millis / MILLIS_PER_SEC, + nanos: ((millis % MILLIS_PER_SEC) as u32) * NANOS_PER_MILLI, + } + } + + /// Creates a new `Duration` from the specified number of microseconds. + /// + /// # Examples + /// + /// ``` + /// use std::time::Duration; + /// + /// let duration = Duration::from_micros(1_000_002); + /// + /// assert_eq!(1, duration.as_secs()); + /// assert_eq!(2000, duration.subsec_nanos()); + /// ``` + #[stable(feature = "duration_from_micros", since = "1.27.0")] + #[inline] + pub const fn from_micros(micros: u64) -> Duration { + Duration { + secs: micros / MICROS_PER_SEC, + nanos: ((micros % MICROS_PER_SEC) as u32) * NANOS_PER_MICRO, + } + } + + /// Creates a new `Duration` from the specified number of nanoseconds. + /// + /// # Examples + /// + /// ``` + /// use std::time::Duration; + /// + /// let duration = Duration::from_nanos(1_000_000_123); + /// + /// assert_eq!(1, duration.as_secs()); + /// assert_eq!(123, duration.subsec_nanos()); + /// ``` + #[stable(feature = "duration_extras", since = "1.27.0")] + #[inline] + pub const fn from_nanos(nanos: u64) -> Duration { + Duration { + secs: nanos / (NANOS_PER_SEC as u64), + nanos: (nanos % (NANOS_PER_SEC as u64)) as u32, + } + } + + /// Returns the number of _whole_ seconds contained by this `Duration`. + /// + /// The returned value does not include the fractional (nanosecond) part of the + /// duration, which can be obtained using [`subsec_nanos`]. + /// + /// # Examples + /// + /// ``` + /// use std::time::Duration; + /// + /// let duration = Duration::new(5, 730023852); + /// assert_eq!(duration.as_secs(), 5); + /// ``` + /// + /// To determine the total number of seconds represented by the `Duration`, + /// use `as_secs` in combination with [`subsec_nanos`]: + /// + /// ``` + /// use std::time::Duration; + /// + /// let duration = Duration::new(5, 730023852); + /// + /// assert_eq!(5.730023852, + /// duration.as_secs() as f64 + /// + duration.subsec_nanos() as f64 * 1e-9); + /// ``` + /// + /// [`subsec_nanos`]: #method.subsec_nanos + #[stable(feature = "duration", since = "1.3.0")] + #[rustc_const_unstable(feature="duration_getters")] + #[inline] + pub const fn as_secs(&self) -> u64 { self.secs } + + /// Returns the fractional part of this `Duration`, in whole milliseconds. + /// + /// This method does **not** return the length of the duration when + /// represented by milliseconds. The returned number always represents a + /// fractional portion of a second (i.e. it is less than one thousand). + /// + /// # Examples + /// + /// ``` + /// use std::time::Duration; + /// + /// let duration = Duration::from_millis(5432); + /// assert_eq!(duration.as_secs(), 5); + /// assert_eq!(duration.subsec_millis(), 432); + /// ``` + #[stable(feature = "duration_extras", since = "1.27.0")] + #[rustc_const_unstable(feature="duration_getters")] + #[inline] + pub const fn subsec_millis(&self) -> u32 { self.nanos / NANOS_PER_MILLI } + + /// Returns the fractional part of this `Duration`, in whole microseconds. + /// + /// This method does **not** return the length of the duration when + /// represented by microseconds. The returned number always represents a + /// fractional portion of a second (i.e. it is less than one million). + /// + /// # Examples + /// + /// ``` + /// use std::time::Duration; + /// + /// let duration = Duration::from_micros(1_234_567); + /// assert_eq!(duration.as_secs(), 1); + /// assert_eq!(duration.subsec_micros(), 234_567); + /// ``` + #[stable(feature = "duration_extras", since = "1.27.0")] + #[rustc_const_unstable(feature="duration_getters")] + #[inline] + pub const fn subsec_micros(&self) -> u32 { self.nanos / NANOS_PER_MICRO } + + /// Returns the fractional part of this `Duration`, in nanoseconds. + /// + /// This method does **not** return the length of the duration when + /// represented by nanoseconds. The returned number always represents a + /// fractional portion of a second (i.e. it is less than one billion). + /// + /// # Examples + /// + /// ``` + /// use std::time::Duration; + /// + /// let duration = Duration::from_millis(5010); + /// assert_eq!(duration.as_secs(), 5); + /// assert_eq!(duration.subsec_nanos(), 10_000_000); + /// ``` + #[stable(feature = "duration", since = "1.3.0")] + #[rustc_const_unstable(feature="duration_getters")] + #[inline] + pub const fn subsec_nanos(&self) -> u32 { self.nanos } + + /// Returns the total number of whole milliseconds contained by this `Duration`. + /// + /// # Examples + /// + /// ``` + /// # #![feature(duration_as_u128)] + /// use std::time::Duration; + /// + /// let duration = Duration::new(5, 730023852); + /// assert_eq!(duration.as_millis(), 5730); + /// ``` + #[unstable(feature = "duration_as_u128", issue = "50202")] + #[inline] + pub fn as_millis(&self) -> u128 { + self.secs as u128 * MILLIS_PER_SEC as u128 + (self.nanos / NANOS_PER_MILLI) as u128 + } + + /// Returns the total number of whole microseconds contained by this `Duration`. + /// + /// # Examples + /// + /// ``` + /// # #![feature(duration_as_u128)] + /// use std::time::Duration; + /// + /// let duration = Duration::new(5, 730023852); + /// assert_eq!(duration.as_micros(), 5730023); + /// ``` + #[unstable(feature = "duration_as_u128", issue = "50202")] + #[inline] + pub fn as_micros(&self) -> u128 { + self.secs as u128 * MICROS_PER_SEC as u128 + (self.nanos / NANOS_PER_MICRO) as u128 + } + + /// Returns the total number of nanoseconds contained by this `Duration`. + /// + /// # Examples + /// + /// ``` + /// # #![feature(duration_as_u128)] + /// use std::time::Duration; + /// + /// let duration = Duration::new(5, 730023852); + /// assert_eq!(duration.as_nanos(), 5730023852); + /// ``` + #[unstable(feature = "duration_as_u128", issue = "50202")] + #[inline] + pub fn as_nanos(&self) -> u128 { + self.secs as u128 * NANOS_PER_SEC as u128 + self.nanos as u128 + } + + /// Checked `Duration` addition. Computes `self + other`, returning [`None`] + /// if overflow occurred. + /// + /// [`None`]: ../../std/option/enum.Option.html#variant.None + /// + /// # Examples + /// + /// Basic usage: + /// + /// ``` + /// use std::time::Duration; + /// + /// assert_eq!(Duration::new(0, 0).checked_add(Duration::new(0, 1)), Some(Duration::new(0, 1))); + /// assert_eq!(Duration::new(1, 0).checked_add(Duration::new(std::u64::MAX, 0)), None); + /// ``` + #[stable(feature = "duration_checked_ops", since = "1.16.0")] + #[inline] + pub fn checked_add(self, rhs: Duration) -> Option<Duration> { + if let Some(mut secs) = self.secs.checked_add(rhs.secs) { + let mut nanos = self.nanos + rhs.nanos; + if nanos >= NANOS_PER_SEC { + nanos -= NANOS_PER_SEC; + if let Some(new_secs) = secs.checked_add(1) { + secs = new_secs; + } else { + return None; + } + } + debug_assert!(nanos < NANOS_PER_SEC); + Some(Duration { + secs, + nanos, + }) + } else { + None + } + } + + /// Checked `Duration` subtraction. Computes `self - other`, returning [`None`] + /// if the result would be negative or if overflow occurred. + /// + /// [`None`]: ../../std/option/enum.Option.html#variant.None + /// + /// # Examples + /// + /// Basic usage: + /// + /// ``` + /// use std::time::Duration; + /// + /// assert_eq!(Duration::new(0, 1).checked_sub(Duration::new(0, 0)), Some(Duration::new(0, 1))); + /// assert_eq!(Duration::new(0, 0).checked_sub(Duration::new(0, 1)), None); + /// ``` + #[stable(feature = "duration_checked_ops", since = "1.16.0")] + #[inline] + pub fn checked_sub(self, rhs: Duration) -> Option<Duration> { + if let Some(mut secs) = self.secs.checked_sub(rhs.secs) { + let nanos = if self.nanos >= rhs.nanos { + self.nanos - rhs.nanos + } else { + if let Some(sub_secs) = secs.checked_sub(1) { + secs = sub_secs; + self.nanos + NANOS_PER_SEC - rhs.nanos + } else { + return None; + } + }; + debug_assert!(nanos < NANOS_PER_SEC); + Some(Duration { secs, nanos }) + } else { + None + } + } + + /// Checked `Duration` multiplication. Computes `self * other`, returning + /// [`None`] if overflow occurred. + /// + /// [`None`]: ../../std/option/enum.Option.html#variant.None + /// + /// # Examples + /// + /// Basic usage: + /// + /// ``` + /// use std::time::Duration; + /// + /// assert_eq!(Duration::new(0, 500_000_001).checked_mul(2), Some(Duration::new(1, 2))); + /// assert_eq!(Duration::new(std::u64::MAX - 1, 0).checked_mul(2), None); + /// ``` + #[stable(feature = "duration_checked_ops", since = "1.16.0")] + #[inline] + pub fn checked_mul(self, rhs: u32) -> Option<Duration> { + // Multiply nanoseconds as u64, because it cannot overflow that way. + let total_nanos = self.nanos as u64 * rhs as u64; + let extra_secs = total_nanos / (NANOS_PER_SEC as u64); + let nanos = (total_nanos % (NANOS_PER_SEC as u64)) as u32; + if let Some(secs) = self.secs + .checked_mul(rhs as u64) + .and_then(|s| s.checked_add(extra_secs)) { + debug_assert!(nanos < NANOS_PER_SEC); + Some(Duration { + secs, + nanos, + }) + } else { + None + } + } + + /// Checked `Duration` division. Computes `self / other`, returning [`None`] + /// if `other == 0`. + /// + /// [`None`]: ../../std/option/enum.Option.html#variant.None + /// + /// # Examples + /// + /// Basic usage: + /// + /// ``` + /// use std::time::Duration; + /// + /// assert_eq!(Duration::new(2, 0).checked_div(2), Some(Duration::new(1, 0))); + /// assert_eq!(Duration::new(1, 0).checked_div(2), Some(Duration::new(0, 500_000_000))); + /// assert_eq!(Duration::new(2, 0).checked_div(0), None); + /// ``` + #[stable(feature = "duration_checked_ops", since = "1.16.0")] + #[inline] + pub fn checked_div(self, rhs: u32) -> Option<Duration> { + if rhs != 0 { + let secs = self.secs / (rhs as u64); + let carry = self.secs - secs * (rhs as u64); + let extra_nanos = carry * (NANOS_PER_SEC as u64) / (rhs as u64); + let nanos = self.nanos / rhs + (extra_nanos as u32); + debug_assert!(nanos < NANOS_PER_SEC); + Some(Duration { secs, nanos }) + } else { + None + } + } +} + +#[stable(feature = "duration", since = "1.3.0")] +impl Add for Duration { + type Output = Duration; + + fn add(self, rhs: Duration) -> Duration { + self.checked_add(rhs).expect("overflow when adding durations") + } +} + +#[stable(feature = "time_augmented_assignment", since = "1.9.0")] +impl AddAssign for Duration { + fn add_assign(&mut self, rhs: Duration) { + *self = *self + rhs; + } +} + +#[stable(feature = "duration", since = "1.3.0")] +impl Sub for Duration { + type Output = Duration; + + fn sub(self, rhs: Duration) -> Duration { + self.checked_sub(rhs).expect("overflow when subtracting durations") + } +} + +#[stable(feature = "time_augmented_assignment", since = "1.9.0")] +impl SubAssign for Duration { + fn sub_assign(&mut self, rhs: Duration) { + *self = *self - rhs; + } +} + +#[stable(feature = "duration", since = "1.3.0")] +impl Mul<u32> for Duration { + type Output = Duration; + + fn mul(self, rhs: u32) -> Duration { + self.checked_mul(rhs).expect("overflow when multiplying duration by scalar") + } +} + +#[stable(feature = "time_augmented_assignment", since = "1.9.0")] +impl MulAssign<u32> for Duration { + fn mul_assign(&mut self, rhs: u32) { + *self = *self * rhs; + } +} + +#[stable(feature = "duration", since = "1.3.0")] +impl Div<u32> for Duration { + type Output = Duration; + + fn div(self, rhs: u32) -> Duration { + self.checked_div(rhs).expect("divide by zero error when dividing duration by scalar") + } +} + +#[stable(feature = "time_augmented_assignment", since = "1.9.0")] +impl DivAssign<u32> for Duration { + fn div_assign(&mut self, rhs: u32) { + *self = *self / rhs; + } +} + +macro_rules! sum_durations { + ($iter:expr) => {{ + let mut total_secs: u64 = 0; + let mut total_nanos: u64 = 0; + + for entry in $iter { + total_secs = total_secs + .checked_add(entry.secs) + .expect("overflow in iter::sum over durations"); + total_nanos = match total_nanos.checked_add(entry.nanos as u64) { + Some(n) => n, + None => { + total_secs = total_secs + .checked_add(total_nanos / NANOS_PER_SEC as u64) + .expect("overflow in iter::sum over durations"); + (total_nanos % NANOS_PER_SEC as u64) + entry.nanos as u64 + } + }; + } + total_secs = total_secs + .checked_add(total_nanos / NANOS_PER_SEC as u64) + .expect("overflow in iter::sum over durations"); + total_nanos = total_nanos % NANOS_PER_SEC as u64; + Duration { + secs: total_secs, + nanos: total_nanos as u32, + } + }}; +} + +#[stable(feature = "duration_sum", since = "1.16.0")] +impl Sum for Duration { + fn sum<I: Iterator<Item=Duration>>(iter: I) -> Duration { + sum_durations!(iter) + } +} + +#[stable(feature = "duration_sum", since = "1.16.0")] +impl<'a> Sum<&'a Duration> for Duration { + fn sum<I: Iterator<Item=&'a Duration>>(iter: I) -> Duration { + sum_durations!(iter) + } +} + +#[stable(feature = "duration_debug_impl", since = "1.27.0")] +impl fmt::Debug for Duration { + fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result { + /// Formats a floating point number in decimal notation. + /// + /// The number is given as the `integer_part` and a fractional part. + /// The value of the fractional part is `fractional_part / divisor`. So + /// `integer_part` = 3, `fractional_part` = 12 and `divisor` = 100 + /// represents the number `3.012`. Trailing zeros are omitted. + /// + /// `divisor` must not be above 100_000_000. It also should be a power + /// of 10, everything else doesn't make sense. `fractional_part` has + /// to be less than `10 * divisor`! + fn fmt_decimal( + f: &mut fmt::Formatter, + mut integer_part: u64, + mut fractional_part: u32, + mut divisor: u32, + ) -> fmt::Result { + // Encode the fractional part into a temporary buffer. The buffer + // only need to hold 9 elements, because `fractional_part` has to + // be smaller than 10^9. The buffer is prefilled with '0' digits + // to simplify the code below. + let mut buf = [b'0'; 9]; + + // The next digit is written at this position + let mut pos = 0; + + // We keep writing digits into the buffer while there are non-zero + // digits left and we haven't written enough digits yet. + while fractional_part > 0 && pos < f.precision().unwrap_or(9) { + // Write new digit into the buffer + buf[pos] = b'0' + (fractional_part / divisor) as u8; + + fractional_part %= divisor; + divisor /= 10; + pos += 1; + } + + // If a precision < 9 was specified, there may be some non-zero + // digits left that weren't written into the buffer. In that case we + // need to perform rounding to match the semantics of printing + // normal floating point numbers. However, we only need to do work + // when rounding up. This happens if the first digit of the + // remaining ones is >= 5. + if fractional_part > 0 && fractional_part >= divisor * 5 { + // Round up the number contained in the buffer. We go through + // the buffer backwards and keep track of the carry. + let mut rev_pos = pos; + let mut carry = true; + while carry && rev_pos > 0 { + rev_pos -= 1; + + // If the digit in the buffer is not '9', we just need to + // increment it and can stop then (since we don't have a + // carry anymore). Otherwise, we set it to '0' (overflow) + // and continue. + if buf[rev_pos] < b'9' { + buf[rev_pos] += 1; + carry = false; + } else { + buf[rev_pos] = b'0'; + } + } + + // If we still have the carry bit set, that means that we set + // the whole buffer to '0's and need to increment the integer + // part. + if carry { + integer_part += 1; + } + } + + // Determine the end of the buffer: if precision is set, we just + // use as many digits from the buffer (capped to 9). If it isn't + // set, we only use all digits up to the last non-zero one. + let end = f.precision().map(|p| ::cmp::min(p, 9)).unwrap_or(pos); + + // If we haven't emitted a single fractional digit and the precision + // wasn't set to a non-zero value, we don't print the decimal point. + if end == 0 { + write!(f, "{}", integer_part) + } else { + // We are only writing ASCII digits into the buffer and it was + // initialized with '0's, so it contains valid UTF8. + let s = unsafe { + ::str::from_utf8_unchecked(&buf[..end]) + }; + + // If the user request a precision > 9, we pad '0's at the end. + let w = f.precision().unwrap_or(pos); + write!(f, "{}.{:0<width$}", integer_part, s, width = w) + } + } + + // Print leading '+' sign if requested + if f.sign_plus() { + write!(f, "+")?; + } + + if self.secs > 0 { + fmt_decimal(f, self.secs, self.nanos, 100_000_000)?; + f.write_str("s") + } else if self.nanos >= 1_000_000 { + fmt_decimal(f, self.nanos as u64 / 1_000_000, self.nanos % 1_000_000, 100_000)?; + f.write_str("ms") + } else if self.nanos >= 1_000 { + fmt_decimal(f, self.nanos as u64 / 1_000, self.nanos % 1_000, 100)?; + f.write_str("µs") + } else { + fmt_decimal(f, self.nanos as u64, 0, 1)?; + f.write_str("ns") + } + } +} diff --git a/src/libcore/unicode/bool_trie.rs b/src/libcore/unicode/bool_trie.rs new file mode 100644 index 00000000000..0e6437fded5 --- /dev/null +++ b/src/libcore/unicode/bool_trie.rs @@ -0,0 +1,76 @@ +// Copyright 2012-2014 The Rust Project Developers. See the COPYRIGHT +// file at the top-level directory of this distribution and at +// http://rust-lang.org/COPYRIGHT. +// +// Licensed under the Apache License, Version 2.0 <LICENSE-APACHE or +// http://www.apache.org/licenses/LICENSE-2.0> or the MIT license +// <LICENSE-MIT or http://opensource.org/licenses/MIT>, at your +// option. This file may not be copied, modified, or distributed +// except according to those terms. + +/// BoolTrie is a trie for representing a set of Unicode codepoints. It is +/// implemented with postfix compression (sharing of identical child nodes), +/// which gives both compact size and fast lookup. +/// +/// The space of Unicode codepoints is divided into 3 subareas, each +/// represented by a trie with different depth. In the first (0..0x800), there +/// is no trie structure at all; each u64 entry corresponds to a bitvector +/// effectively holding 64 bool values. +/// +/// In the second (0x800..0x10000), each child of the root node represents a +/// 64-wide subrange, but instead of storing the full 64-bit value of the leaf, +/// the trie stores an 8-bit index into a shared table of leaf values. This +/// exploits the fact that in reasonable sets, many such leaves can be shared. +/// +/// In the third (0x10000..0x110000), each child of the root node represents a +/// 4096-wide subrange, and the trie stores an 8-bit index into a 64-byte slice +/// of a child tree. Each of these 64 bytes represents an index into the table +/// of shared 64-bit leaf values. This exploits the sparse structure in the +/// non-BMP range of most Unicode sets. +pub struct BoolTrie { + // 0..0x800 (corresponding to 1 and 2 byte utf-8 sequences) + pub r1: [u64; 32], // leaves + + // 0x800..0x10000 (corresponding to 3 byte utf-8 sequences) + pub r2: [u8; 992], // first level + pub r3: &'static [u64], // leaves + + // 0x10000..0x110000 (corresponding to 4 byte utf-8 sequences) + pub r4: [u8; 256], // first level + pub r5: &'static [u8], // second level + pub r6: &'static [u64], // leaves +} +impl BoolTrie { + pub fn lookup(&self, c: char) -> bool { + let c = c as u32; + if c < 0x800 { + trie_range_leaf(c, self.r1[(c >> 6) as usize]) + } else if c < 0x10000 { + let child = self.r2[(c >> 6) as usize - 0x20]; + trie_range_leaf(c, self.r3[child as usize]) + } else { + let child = self.r4[(c >> 12) as usize - 0x10]; + let leaf = self.r5[((child as usize) << 6) + ((c >> 6) as usize & 0x3f)]; + trie_range_leaf(c, self.r6[leaf as usize]) + } + } +} + +pub struct SmallBoolTrie { + pub(crate) r1: &'static [u8], // first level + pub(crate) r2: &'static [u64], // leaves +} + +impl SmallBoolTrie { + pub fn lookup(&self, c: char) -> bool { + let c = c as u32; + match self.r1.get((c >> 6) as usize) { + Some(&child) => trie_range_leaf(c, self.r2[child as usize]), + None => false, + } + } +} + +fn trie_range_leaf(c: u32, bitmap_chunk: u64) -> bool { + ((bitmap_chunk >> (c & 63)) & 1) != 0 +} diff --git a/src/libcore/unicode/mod.rs b/src/libcore/unicode/mod.rs new file mode 100644 index 00000000000..e5cda880f88 --- /dev/null +++ b/src/libcore/unicode/mod.rs @@ -0,0 +1,30 @@ +// Copyright 2012-2014 The Rust Project Developers. See the COPYRIGHT +// file at the top-level directory of this distribution and at +// http://rust-lang.org/COPYRIGHT. +// +// Licensed under the Apache License, Version 2.0 <LICENSE-APACHE or +// http://www.apache.org/licenses/LICENSE-2.0> or the MIT license +// <LICENSE-MIT or http://opensource.org/licenses/MIT>, at your +// option. This file may not be copied, modified, or distributed +// except according to those terms. + +#![unstable(feature = "unicode_internals", issue = "0")] +#![allow(missing_docs)] + +mod bool_trie; +pub(crate) mod printable; +pub(crate) mod tables; +pub(crate) mod version; + +// For use in liballoc, not re-exported in libstd. +pub mod derived_property { + pub use unicode::tables::derived_property::{Case_Ignorable, Cased}; +} +pub mod conversions { + pub use unicode::tables::conversions::{to_lower, to_upper}; +} + +// For use in libsyntax +pub mod property { + pub use unicode::tables::property::Pattern_White_Space; +} diff --git a/src/libcore/unicode/printable.py b/src/libcore/unicode/printable.py new file mode 100644 index 00000000000..9410dafbbc3 --- /dev/null +++ b/src/libcore/unicode/printable.py @@ -0,0 +1,254 @@ +#!/usr/bin/env python +# +# Copyright 2011-2016 The Rust Project Developers. See the COPYRIGHT +# file at the top-level directory of this distribution and at +# http://rust-lang.org/COPYRIGHT. +# +# Licensed under the Apache License, Version 2.0 <LICENSE-APACHE or +# http://www.apache.org/licenses/LICENSE-2.0> or the MIT license +# <LICENSE-MIT or http://opensource.org/licenses/MIT>, at your +# option. This file may not be copied, modified, or distributed +# except according to those terms. + +# This script uses the following Unicode tables: +# - UnicodeData.txt + + +from collections import namedtuple +import csv +import os +import subprocess + +NUM_CODEPOINTS=0x110000 + +def to_ranges(iter): + current = None + for i in iter: + if current is None or i != current[1] or i in (0x10000, 0x20000): + if current is not None: + yield tuple(current) + current = [i, i + 1] + else: + current[1] += 1 + if current is not None: + yield tuple(current) + +def get_escaped(codepoints): + for c in codepoints: + if (c.class_ or "Cn") in "Cc Cf Cs Co Cn Zl Zp Zs".split() and c.value != ord(' '): + yield c.value + +def get_file(f): + try: + return open(os.path.basename(f)) + except FileNotFoundError: + subprocess.run(["curl", "-O", f], check=True) + return open(os.path.basename(f)) + +Codepoint = namedtuple('Codepoint', 'value class_') + +def get_codepoints(f): + r = csv.reader(f, delimiter=";") + prev_codepoint = 0 + class_first = None + for row in r: + codepoint = int(row[0], 16) + name = row[1] + class_ = row[2] + + if class_first is not None: + if not name.endswith("Last>"): + raise ValueError("Missing Last after First") + + for c in range(prev_codepoint + 1, codepoint): + yield Codepoint(c, class_first) + + class_first = None + if name.endswith("First>"): + class_first = class_ + + yield Codepoint(codepoint, class_) + prev_codepoint = codepoint + + if class_first != None: + raise ValueError("Missing Last after First") + + for c in range(prev_codepoint + 1, NUM_CODEPOINTS): + yield Codepoint(c, None) + +def compress_singletons(singletons): + uppers = [] # (upper, # items in lowers) + lowers = [] + + for i in singletons: + upper = i >> 8 + lower = i & 0xff + if len(uppers) == 0 or uppers[-1][0] != upper: + uppers.append((upper, 1)) + else: + upper, count = uppers[-1] + uppers[-1] = upper, count + 1 + lowers.append(lower) + + return uppers, lowers + +def compress_normal(normal): + # lengths 0x00..0x7f are encoded as 00, 01, ..., 7e, 7f + # lengths 0x80..0x7fff are encoded as 80 80, 80 81, ..., ff fe, ff ff + compressed = [] # [truelen, (truelenaux), falselen, (falselenaux)] + + prev_start = 0 + for start, count in normal: + truelen = start - prev_start + falselen = count + prev_start = start + count + + assert truelen < 0x8000 and falselen < 0x8000 + entry = [] + if truelen > 0x7f: + entry.append(0x80 | (truelen >> 8)) + entry.append(truelen & 0xff) + else: + entry.append(truelen & 0x7f) + if falselen > 0x7f: + entry.append(0x80 | (falselen >> 8)) + entry.append(falselen & 0xff) + else: + entry.append(falselen & 0x7f) + + compressed.append(entry) + + return compressed + +def print_singletons(uppers, lowers, uppersname, lowersname): + print("const {}: &'static [(u8, u8)] = &[".format(uppersname)) + for u, c in uppers: + print(" ({:#04x}, {}),".format(u, c)) + print("];") + print("const {}: &'static [u8] = &[".format(lowersname)) + for i in range(0, len(lowers), 8): + print(" {}".format(" ".join("{:#04x},".format(l) for l in lowers[i:i+8]))) + print("];") + +def print_normal(normal, normalname): + print("const {}: &'static [u8] = &[".format(normalname)) + for v in normal: + print(" {}".format(" ".join("{:#04x},".format(i) for i in v))) + print("];") + +def main(): + file = get_file("http://www.unicode.org/Public/UNIDATA/UnicodeData.txt") + + codepoints = get_codepoints(file) + + CUTOFF=0x10000 + singletons0 = [] + singletons1 = [] + normal0 = [] + normal1 = [] + extra = [] + + for a, b in to_ranges(get_escaped(codepoints)): + if a > 2 * CUTOFF: + extra.append((a, b - a)) + elif a == b - 1: + if a & CUTOFF: + singletons1.append(a & ~CUTOFF) + else: + singletons0.append(a) + elif a == b - 2: + if a & CUTOFF: + singletons1.append(a & ~CUTOFF) + singletons1.append((a + 1) & ~CUTOFF) + else: + singletons0.append(a) + singletons0.append(a + 1) + else: + if a >= 2 * CUTOFF: + extra.append((a, b - a)) + elif a & CUTOFF: + normal1.append((a & ~CUTOFF, b - a)) + else: + normal0.append((a, b - a)) + + singletons0u, singletons0l = compress_singletons(singletons0) + singletons1u, singletons1l = compress_singletons(singletons1) + normal0 = compress_normal(normal0) + normal1 = compress_normal(normal1) + + print("""\ +// Copyright 2012-2017 The Rust Project Developers. See the COPYRIGHT +// file at the top-level directory of this distribution and at +// http://rust-lang.org/COPYRIGHT. +// +// Licensed under the Apache License, Version 2.0 <LICENSE-APACHE or +// http://www.apache.org/licenses/LICENSE-2.0> or the MIT license +// <LICENSE-MIT or http://opensource.org/licenses/MIT>, at your +// option. This file may not be copied, modified, or distributed +// except according to those terms. + +// NOTE: The following code was generated by "src/libcore/unicode/printable.py", +// do not edit directly! + +fn check(x: u16, singletonuppers: &[(u8, u8)], singletonlowers: &[u8], + normal: &[u8]) -> bool { + let xupper = (x >> 8) as u8; + let mut lowerstart = 0; + for &(upper, lowercount) in singletonuppers { + let lowerend = lowerstart + lowercount as usize; + if xupper == upper { + for &lower in &singletonlowers[lowerstart..lowerend] { + if lower == x as u8 { + return false; + } + } + } else if xupper < upper { + break; + } + lowerstart = lowerend; + } + + let mut x = x as i32; + let mut normal = normal.iter().cloned(); + let mut current = true; + while let Some(v) = normal.next() { + let len = if v & 0x80 != 0 { + ((v & 0x7f) as i32) << 8 | normal.next().unwrap() as i32 + } else { + v as i32 + }; + x -= len; + if x < 0 { + break; + } + current = !current; + } + current +} + +pub(crate) fn is_printable(x: char) -> bool { + let x = x as u32; + let lower = x as u16; + if x < 0x10000 { + check(lower, SINGLETONS0U, SINGLETONS0L, NORMAL0) + } else if x < 0x20000 { + check(lower, SINGLETONS1U, SINGLETONS1L, NORMAL1) + } else {\ +""") + for a, b in extra: + print(" if 0x{:x} <= x && x < 0x{:x} {{".format(a, a + b)) + print(" return false;") + print(" }") + print("""\ + true + } +}\ +""") + print() + print_singletons(singletons0u, singletons0l, 'SINGLETONS0U', 'SINGLETONS0L') + print_singletons(singletons1u, singletons1l, 'SINGLETONS1U', 'SINGLETONS1L') + print_normal(normal0, 'NORMAL0') + print_normal(normal1, 'NORMAL1') + +if __name__ == '__main__': + main() diff --git a/src/libcore/char_private.rs b/src/libcore/unicode/printable.rs index e6803745ab5..519dd17bb9b 100644 --- a/src/libcore/char_private.rs +++ b/src/libcore/unicode/printable.rs @@ -8,7 +8,7 @@ // option. This file may not be copied, modified, or distributed // except according to those terms. -// NOTE: The following code was generated by "src/etc/char_private.py", +// NOTE: The following code was generated by "src/libcore/unicode/printable.py", // do not edit directly! fn check(x: u16, singletonuppers: &[(u8, u8)], singletonlowers: &[u8], @@ -83,14 +83,14 @@ pub(crate) fn is_printable(x: char) -> bool { const SINGLETONS0U: &'static [(u8, u8)] = &[ (0x00, 1), (0x03, 5), - (0x05, 8), + (0x05, 6), (0x06, 3), - (0x07, 4), + (0x07, 6), (0x08, 8), - (0x09, 16), - (0x0a, 27), + (0x09, 17), + (0x0a, 28), (0x0b, 25), - (0x0c, 22), + (0x0c, 20), (0x0d, 18), (0x0e, 22), (0x0f, 4), @@ -102,18 +102,17 @@ const SINGLETONS0U: &'static [(u8, u8)] = &[ (0x18, 2), (0x19, 3), (0x1a, 7), + (0x1c, 2), (0x1d, 1), (0x1f, 22), (0x20, 3), - (0x2b, 5), + (0x2b, 6), (0x2c, 2), (0x2d, 11), (0x2e, 1), (0x30, 3), - (0x31, 3), + (0x31, 2), (0x32, 2), - (0xa7, 1), - (0xa8, 2), (0xa9, 2), (0xaa, 4), (0xab, 8), @@ -125,19 +124,19 @@ const SINGLETONS0U: &'static [(u8, u8)] = &[ ]; const SINGLETONS0L: &'static [u8] = &[ 0xad, 0x78, 0x79, 0x8b, 0x8d, 0xa2, 0x30, 0x57, - 0x58, 0x60, 0x88, 0x8b, 0x8c, 0x90, 0x1c, 0x1d, - 0xdd, 0x0e, 0x0f, 0x4b, 0x4c, 0x2e, 0x2f, 0x3f, + 0x58, 0x8b, 0x8c, 0x90, 0x1c, 0x1d, 0xdd, 0x0e, + 0x0f, 0x4b, 0x4c, 0xfb, 0xfc, 0x2e, 0x2f, 0x3f, 0x5c, 0x5d, 0x5f, 0xb5, 0xe2, 0x84, 0x8d, 0x8e, 0x91, 0x92, 0xa9, 0xb1, 0xba, 0xbb, 0xc5, 0xc6, - 0xc9, 0xca, 0xde, 0xe4, 0xe5, 0x04, 0x11, 0x12, - 0x29, 0x31, 0x34, 0x37, 0x3a, 0x3b, 0x3d, 0x49, - 0x4a, 0x5d, 0x84, 0x8e, 0x92, 0xa9, 0xb1, 0xb4, - 0xba, 0xbb, 0xc6, 0xca, 0xce, 0xcf, 0xe4, 0xe5, - 0x00, 0x04, 0x0d, 0x0e, 0x11, 0x12, 0x29, 0x31, - 0x34, 0x3a, 0x3b, 0x45, 0x46, 0x49, 0x4a, 0x5e, - 0x64, 0x65, 0x84, 0x91, 0x9b, 0x9d, 0xc9, 0xce, - 0xcf, 0x04, 0x0d, 0x11, 0x29, 0x45, 0x49, 0x57, - 0x64, 0x65, 0x84, 0x8d, 0x91, 0xa9, 0xb4, 0xba, + 0xc9, 0xca, 0xde, 0xe4, 0xe5, 0xff, 0x00, 0x04, + 0x11, 0x12, 0x29, 0x31, 0x34, 0x37, 0x3a, 0x3b, + 0x3d, 0x49, 0x4a, 0x5d, 0x84, 0x8e, 0x92, 0xa9, + 0xb1, 0xb4, 0xba, 0xbb, 0xc6, 0xca, 0xce, 0xcf, + 0xe4, 0xe5, 0x00, 0x04, 0x0d, 0x0e, 0x11, 0x12, + 0x29, 0x31, 0x34, 0x3a, 0x3b, 0x45, 0x46, 0x49, + 0x4a, 0x5e, 0x64, 0x65, 0x84, 0x91, 0x9b, 0x9d, + 0xc9, 0xce, 0xcf, 0x0d, 0x11, 0x29, 0x45, 0x49, + 0x57, 0x64, 0x65, 0x8d, 0x91, 0xa9, 0xb4, 0xba, 0xbb, 0xc5, 0xc9, 0xdf, 0xe4, 0xe5, 0xf0, 0x04, 0x0d, 0x11, 0x45, 0x49, 0x64, 0x65, 0x80, 0x81, 0x84, 0xb2, 0xbc, 0xbe, 0xbf, 0xd5, 0xd7, 0xf0, @@ -150,18 +149,18 @@ const SINGLETONS0L: &'static [u8] = &[ 0x11, 0x16, 0x17, 0x5b, 0x5c, 0xf6, 0xf7, 0xfe, 0xff, 0x80, 0x0d, 0x6d, 0x71, 0xde, 0xdf, 0x0e, 0x0f, 0x1f, 0x6e, 0x6f, 0x1c, 0x1d, 0x5f, 0x7d, - 0x7e, 0xae, 0xaf, 0xfa, 0x16, 0x17, 0x1e, 0x1f, - 0x46, 0x47, 0x4e, 0x4f, 0x58, 0x5a, 0x5c, 0x5e, - 0x7e, 0x7f, 0xb5, 0xc5, 0xd4, 0xd5, 0xdc, 0xf0, - 0xf1, 0xf5, 0x72, 0x73, 0x8f, 0x74, 0x75, 0x96, - 0x97, 0xc9, 0x2f, 0x5f, 0x26, 0x2e, 0x2f, 0xa7, - 0xaf, 0xb7, 0xbf, 0xc7, 0xcf, 0xd7, 0xdf, 0x9a, - 0x40, 0x97, 0x98, 0x2f, 0x30, 0x8f, 0x1f, 0xff, - 0xaf, 0xfe, 0xff, 0xce, 0xff, 0x4e, 0x4f, 0x5a, - 0x5b, 0x07, 0x08, 0x0f, 0x10, 0x27, 0x2f, 0xee, - 0xef, 0x6e, 0x6f, 0x37, 0x3d, 0x3f, 0x42, 0x45, - 0x90, 0x91, 0xfe, 0xff, 0x53, 0x67, 0x75, 0xc8, - 0xc9, 0xd0, 0xd1, 0xd8, 0xd9, 0xe7, 0xfe, 0xff, + 0x7e, 0xae, 0xaf, 0xbb, 0xbc, 0xfa, 0x16, 0x17, + 0x1e, 0x1f, 0x46, 0x47, 0x4e, 0x4f, 0x58, 0x5a, + 0x5c, 0x5e, 0x7e, 0x7f, 0xb5, 0xc5, 0xd4, 0xd5, + 0xdc, 0xf0, 0xf1, 0xf5, 0x72, 0x73, 0x8f, 0x74, + 0x75, 0x96, 0x97, 0xc9, 0xff, 0x2f, 0x5f, 0x26, + 0x2e, 0x2f, 0xa7, 0xaf, 0xb7, 0xbf, 0xc7, 0xcf, + 0xd7, 0xdf, 0x9a, 0x40, 0x97, 0x98, 0x30, 0x8f, + 0x1f, 0xff, 0xce, 0xff, 0x4e, 0x4f, 0x5a, 0x5b, + 0x07, 0x08, 0x0f, 0x10, 0x27, 0x2f, 0xee, 0xef, + 0x6e, 0x6f, 0x37, 0x3d, 0x3f, 0x42, 0x45, 0x90, + 0x91, 0xfe, 0xff, 0x53, 0x67, 0x75, 0xc8, 0xc9, + 0xd0, 0xd1, 0xd8, 0xd9, 0xe7, 0xfe, 0xff, ]; const SINGLETONS1U: &'static [(u8, u8)] = &[ (0x00, 6), @@ -170,17 +169,18 @@ const SINGLETONS1U: &'static [(u8, u8)] = &[ (0x04, 2), (0x08, 8), (0x09, 2), - (0x0a, 3), + (0x0a, 5), (0x0b, 2), (0x10, 1), (0x11, 4), (0x12, 5), - (0x13, 18), + (0x13, 17), (0x14, 2), (0x15, 2), - (0x1a, 3), + (0x17, 2), + (0x1a, 2), (0x1c, 5), - (0x1d, 4), + (0x1d, 8), (0x24, 1), (0x6a, 3), (0x6b, 2), @@ -195,51 +195,49 @@ const SINGLETONS1U: &'static [(u8, u8)] = &[ (0xe8, 2), (0xee, 32), (0xf0, 4), - (0xf1, 1), - (0xf9, 1), + (0xf9, 4), ]; const SINGLETONS1L: &'static [u8] = &[ 0x0c, 0x27, 0x3b, 0x3e, 0x4e, 0x4f, 0x8f, 0x9e, 0x9e, 0x9f, 0x06, 0x07, 0x09, 0x36, 0x3d, 0x3e, - 0x56, 0xf3, 0xd0, 0xd1, 0x04, 0x14, 0x18, 0x56, - 0x57, 0xbd, 0x35, 0xce, 0xcf, 0xe0, 0x12, 0x87, - 0x89, 0x8e, 0x9e, 0x04, 0x0d, 0x0e, 0x11, 0x12, - 0x29, 0x31, 0x34, 0x3a, 0x3b, 0x45, 0x46, 0x49, - 0x4a, 0x4e, 0x4f, 0x64, 0x65, 0x5a, 0x5c, 0xb6, - 0xb7, 0x84, 0x85, 0x9d, 0x09, 0x37, 0x90, 0x91, - 0xa8, 0x07, 0x0a, 0x3b, 0x3e, 0x6f, 0x5f, 0xee, - 0xef, 0x5a, 0x62, 0x9a, 0x9b, 0x27, 0x28, 0x55, - 0x9d, 0xa0, 0xa1, 0xa3, 0xa4, 0xa7, 0xa8, 0xad, - 0xba, 0xbc, 0xc4, 0x06, 0x0b, 0x0c, 0x15, 0x1d, - 0x3a, 0x3f, 0x45, 0x51, 0xa6, 0xa7, 0xcc, 0xcd, - 0xa0, 0x07, 0x19, 0x1a, 0x22, 0x25, 0xc5, 0xc6, - 0x04, 0x20, 0x23, 0x25, 0x26, 0x28, 0x33, 0x38, - 0x3a, 0x48, 0x4a, 0x4c, 0x50, 0x53, 0x55, 0x56, - 0x58, 0x5a, 0x5c, 0x5e, 0x60, 0x63, 0x65, 0x66, - 0x6b, 0x73, 0x78, 0x7d, 0x7f, 0x8a, 0xa4, 0xaa, - 0xaf, 0xb0, 0xc0, 0xd0, 0x2f, 0x3f, + 0x56, 0xf3, 0xd0, 0xd1, 0x04, 0x14, 0x18, 0x36, + 0x37, 0x56, 0x57, 0xbd, 0x35, 0xce, 0xcf, 0xe0, + 0x12, 0x87, 0x89, 0x8e, 0x9e, 0x04, 0x0d, 0x0e, + 0x11, 0x12, 0x29, 0x31, 0x34, 0x3a, 0x45, 0x46, + 0x49, 0x4a, 0x4e, 0x4f, 0x64, 0x65, 0x5a, 0x5c, + 0xb6, 0xb7, 0x1b, 0x1c, 0x84, 0x85, 0x09, 0x37, + 0x90, 0x91, 0xa8, 0x07, 0x0a, 0x3b, 0x3e, 0x66, + 0x69, 0x8f, 0x92, 0x6f, 0x5f, 0xee, 0xef, 0x5a, + 0x62, 0x9a, 0x9b, 0x27, 0x28, 0x55, 0x9d, 0xa0, + 0xa1, 0xa3, 0xa4, 0xa7, 0xa8, 0xad, 0xba, 0xbc, + 0xc4, 0x06, 0x0b, 0x0c, 0x15, 0x1d, 0x3a, 0x3f, + 0x45, 0x51, 0xa6, 0xa7, 0xcc, 0xcd, 0xa0, 0x07, + 0x19, 0x1a, 0x22, 0x25, 0xc5, 0xc6, 0x04, 0x20, + 0x23, 0x25, 0x26, 0x28, 0x33, 0x38, 0x3a, 0x48, + 0x4a, 0x4c, 0x50, 0x53, 0x55, 0x56, 0x58, 0x5a, + 0x5c, 0x5e, 0x60, 0x63, 0x65, 0x66, 0x6b, 0x73, + 0x78, 0x7d, 0x7f, 0x8a, 0xa4, 0xaa, 0xaf, 0xb0, + 0xc0, 0xd0, 0x3f, 0x71, 0x72, 0x7b, ]; const NORMAL0: &'static [u8] = &[ 0x00, 0x20, 0x5f, 0x22, 0x82, 0xdf, 0x04, 0x82, 0x44, 0x08, - 0x1b, 0x05, - 0x05, 0x11, + 0x1b, 0x04, + 0x06, 0x11, 0x81, 0xac, 0x0e, - 0x3b, 0x05, - 0x6b, 0x35, - 0x1e, 0x16, - 0x80, 0xdf, 0x03, + 0x80, 0xab, 0x35, + 0x1e, 0x15, + 0x80, 0xe0, 0x03, 0x19, 0x08, 0x01, 0x04, - 0x22, 0x03, - 0x0a, 0x04, + 0x2f, 0x04, 0x34, 0x04, 0x07, 0x03, 0x01, 0x07, 0x06, 0x07, - 0x10, 0x0b, + 0x11, 0x0a, 0x50, 0x0f, 0x12, 0x07, 0x55, 0x08, @@ -286,7 +284,7 @@ const NORMAL0: &'static [u8] = &[ 0x6a, 0x06, 0x0a, 0x06, 0x1a, 0x06, - 0x58, 0x08, + 0x59, 0x07, 0x2b, 0x05, 0x46, 0x0a, 0x2c, 0x04, @@ -304,8 +302,8 @@ const NORMAL0: &'static [u8] = &[ 0x74, 0x08, 0x3c, 0x03, 0x0f, 0x03, - 0x3c, 0x37, - 0x08, 0x08, + 0x3c, 0x07, + 0x38, 0x08, 0x2a, 0x06, 0x82, 0xff, 0x11, 0x18, 0x08, @@ -316,15 +314,12 @@ const NORMAL0: &'static [u8] = &[ 0x80, 0x8c, 0x04, 0x82, 0x97, 0x19, 0x0b, 0x15, - 0x87, 0x5a, 0x03, - 0x16, 0x19, - 0x04, 0x10, - 0x80, 0xf4, 0x05, + 0x88, 0x94, 0x05, 0x2f, 0x05, 0x3b, 0x07, 0x02, 0x0e, 0x18, 0x09, - 0x80, 0xaa, 0x36, + 0x80, 0xaf, 0x31, 0x74, 0x0c, 0x80, 0xd6, 0x1a, 0x0c, 0x05, @@ -332,12 +327,12 @@ const NORMAL0: &'static [u8] = &[ 0x80, 0xb6, 0x05, 0x24, 0x0c, 0x9b, 0xc6, 0x0a, - 0xd2, 0x2b, 0x15, + 0xd2, 0x30, 0x10, 0x84, 0x8d, 0x03, 0x37, 0x09, 0x81, 0x5c, 0x14, 0x80, 0xb8, 0x08, - 0x80, 0xb8, 0x3f, + 0x80, 0xba, 0x3d, 0x35, 0x04, 0x0a, 0x06, 0x38, 0x08, @@ -402,9 +397,8 @@ const NORMAL1: &'static [u8] = &[ 0x01, 0x40, 0x38, 0x04, 0x4b, 0x05, - 0x28, 0x04, - 0x03, 0x04, - 0x09, 0x08, + 0x2f, 0x04, + 0x0a, 0x07, 0x09, 0x07, 0x40, 0x20, 0x27, 0x04, @@ -417,14 +411,17 @@ const NORMAL1: &'static [u8] = &[ 0x49, 0x37, 0x33, 0x0d, 0x33, 0x07, - 0x06, 0x81, 0x60, - 0x1f, 0x81, 0x81, + 0x2e, 0x08, + 0x0a, 0x81, 0x26, + 0x1f, 0x80, 0x81, + 0x28, 0x08, + 0x2a, 0x80, 0xa6, 0x4e, 0x04, 0x1e, 0x0f, 0x43, 0x0e, 0x19, 0x07, 0x0a, 0x06, - 0x44, 0x0c, + 0x47, 0x09, 0x27, 0x09, 0x75, 0x0b, 0x3f, 0x41, @@ -435,7 +432,7 @@ const NORMAL1: &'static [u8] = &[ 0x01, 0x05, 0x10, 0x03, 0x05, 0x80, 0x8b, - 0x5e, 0x22, + 0x5f, 0x21, 0x48, 0x08, 0x0a, 0x80, 0xa6, 0x5e, 0x22, @@ -444,9 +441,9 @@ const NORMAL1: &'static [u8] = &[ 0x0d, 0x13, 0x38, 0x08, 0x0a, 0x36, - 0x1a, 0x03, - 0x0f, 0x04, - 0x10, 0x81, 0x60, + 0x2c, 0x04, + 0x10, 0x80, 0xc0, + 0x3c, 0x64, 0x53, 0x0c, 0x01, 0x81, 0x00, 0x48, 0x08, @@ -457,7 +454,10 @@ const NORMAL1: &'static [u8] = &[ 0x47, 0x49, 0x37, 0x03, 0x0e, 0x08, - 0x0a, 0x82, 0xa6, + 0x0a, 0x06, + 0x39, 0x07, + 0x0a, 0x81, 0x36, + 0x19, 0x81, 0x07, 0x83, 0x9a, 0x66, 0x75, 0x0b, 0x80, 0xc4, 0x8a, 0xbc, @@ -469,12 +469,13 @@ const NORMAL1: &'static [u8] = &[ 0x26, 0x0a, 0x46, 0x0a, 0x28, 0x05, - 0x13, 0x83, 0x70, + 0x13, 0x82, 0xb0, + 0x5b, 0x65, 0x45, 0x0b, 0x2f, 0x10, 0x11, 0x40, 0x02, 0x1e, - 0x97, 0xed, 0x13, + 0x97, 0xf2, 0x0e, 0x82, 0xf3, 0xa5, 0x0d, 0x81, 0x1f, 0x51, 0x81, 0x8c, 0x89, 0x04, @@ -485,9 +486,10 @@ const NORMAL1: &'static [u8] = &[ 0x80, 0xf6, 0x0a, 0x73, 0x08, 0x6e, 0x17, - 0x46, 0x80, 0xba, + 0x46, 0x80, 0x9a, + 0x14, 0x0c, 0x57, 0x09, - 0x12, 0x80, 0x8e, + 0x19, 0x80, 0x87, 0x81, 0x47, 0x03, 0x85, 0x42, 0x0f, 0x15, 0x85, 0x50, @@ -495,7 +497,8 @@ const NORMAL1: &'static [u8] = &[ 0x80, 0xd7, 0x29, 0x4b, 0x05, 0x0a, 0x04, - 0x02, 0x84, 0xa0, + 0x02, 0x83, 0x11, + 0x44, 0x81, 0x4b, 0x3c, 0x06, 0x01, 0x04, 0x55, 0x05, @@ -514,18 +517,19 @@ const NORMAL1: &'static [u8] = &[ 0x06, 0x80, 0x9a, 0x83, 0xd5, 0x0b, 0x0d, 0x03, - 0x09, 0x07, + 0x0a, 0x06, 0x74, 0x0c, - 0x55, 0x2b, + 0x59, 0x27, 0x0c, 0x04, 0x38, 0x08, 0x0a, 0x06, 0x28, 0x08, 0x1e, 0x52, 0x0c, 0x04, - 0x3d, 0x03, - 0x1c, 0x14, - 0x18, 0x28, - 0x01, 0x0f, - 0x17, 0x86, 0x19, + 0x67, 0x03, + 0x29, 0x0d, + 0x0a, 0x06, + 0x03, 0x0d, + 0x30, 0x60, + 0x0e, 0x85, 0x92, ]; diff --git a/src/libcore/unicode/tables.rs b/src/libcore/unicode/tables.rs new file mode 100644 index 00000000000..3de855ac943 --- /dev/null +++ b/src/libcore/unicode/tables.rs @@ -0,0 +1,2601 @@ +// Copyright 2012-2018 The Rust Project Developers. See the COPYRIGHT +// file at the top-level directory of this distribution and at +// http://rust-lang.org/COPYRIGHT. +// +// Licensed under the Apache License, Version 2.0 <LICENSE-APACHE or +// http://www.apache.org/licenses/LICENSE-2.0> or the MIT license +// <LICENSE-MIT or http://opensource.org/licenses/MIT>, at your +// option. This file may not be copied, modified, or distributed +// except according to those terms. + +// NOTE: The following code was generated by "./unicode.py", do not edit directly + +#![allow(missing_docs, non_upper_case_globals, non_snake_case)] + +use unicode::version::UnicodeVersion; +use unicode::bool_trie::{BoolTrie, SmallBoolTrie}; + +/// The version of [Unicode](http://www.unicode.org/) that the Unicode parts of +/// `char` and `str` methods are based on. +#[unstable(feature = "unicode_version", issue = "49726")] +pub const UNICODE_VERSION: UnicodeVersion = UnicodeVersion { + major: 11, + minor: 0, + micro: 0, + _priv: (), +}; +pub mod general_category { + pub const Cc_table: &super::SmallBoolTrie = &super::SmallBoolTrie { + r1: &[ + 0, 1, 0 + ], + r2: &[ + 0x00000000ffffffff, 0x8000000000000000 + ], + }; + + pub fn Cc(c: char) -> bool { + Cc_table.lookup(c) + } + + pub const N_table: &super::BoolTrie = &super::BoolTrie { + r1: [ + 0x03ff000000000000, 0x0000000000000000, 0x720c000000000000, 0x0000000000000000, + 0x0000000000000000, 0x0000000000000000, 0x0000000000000000, 0x0000000000000000, + 0x0000000000000000, 0x0000000000000000, 0x0000000000000000, 0x0000000000000000, + 0x0000000000000000, 0x0000000000000000, 0x0000000000000000, 0x0000000000000000, + 0x0000000000000000, 0x0000000000000000, 0x0000000000000000, 0x0000000000000000, + 0x0000000000000000, 0x0000000000000000, 0x0000000000000000, 0x0000000000000000, + 0x0000000000000000, 0x000003ff00000000, 0x0000000000000000, 0x03ff000000000000, + 0x0000000000000000, 0x0000000000000000, 0x0000000000000000, 0x00000000000003ff + ], + r2: [ + 0, 0, 0, 0, 0, 1, 0, 2, 0, 1, 0, 1, 0, 3, 0, 4, 0, 5, 0, 1, 0, 6, 0, 1, 0, 7, 0, 7, 8, + 0, 0, 0, 0, 9, 7, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 10, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 11, 0, 0, 0, 12, 7, 0, 0, 0, 0, 13, 0, 14, 0, 0, 15, 0, 0, 7, 16, 0, 0, 15, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 17, 9, 0, 0, 18, 19, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 20, 21, 22, 0, 0, 0, 0, 0, 0, 0, 0, 0, 23, 24, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 25, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 26, 0, 0, 0, 0, 0, 27, 0, 28, + 29, 30, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 28, + 0, 0, 1, 0, 0, 0, 0, 31, 0, 0, 7, 9, 0, 0, 32, 0, 7, 0, 0, 0, 0, 0, 16, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 7, 0, 0, 0 + ], + r3: &[ + 0x0000000000000000, 0x0000ffc000000000, 0x03f0ffc000000000, 0x00fcffc000000000, + 0x0007ffc000000000, 0x7f00ffc000000000, 0x01ffffc07f000000, 0x0000000003ff0000, + 0x000fffff00000000, 0x00000000000003ff, 0x1ffffe0000000000, 0x0001c00000000000, + 0x03ff03ff00000000, 0x000000000000ffc0, 0x0000000007ff0000, 0x0000000003ff03ff, + 0x03ff000000000000, 0x03f1000000000000, 0xffffffffffff0000, 0x00000000000003e7, + 0xffffffff00000000, 0x000000000fffffff, 0xfffffc0000000000, 0xffc0000000000000, + 0x00000000000fffff, 0x2000000000000000, 0x070003fe00000080, 0x00000000003c0000, + 0x000003ff00000000, 0x00000000fffeff00, 0xfffe0000000003ff, 0x003f000000000000, + 0x03ff000003ff0000 + ], + r4: [ + 0, 1, 2, 3, 3, 3, 4, 3, 3, 3, 3, 3, 3, 5, 6, 7, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, + 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, + 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, + 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, + 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, + 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, + 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, + 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, + 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3 + ], + r5: &[ + 0, 0, 0, 0, 1, 2, 3, 0, 0, 0, 0, 4, 5, 6, 0, 7, 0, 0, 8, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 9, 10, 11, 12, 0, 13, 14, 0, 15, 16, 17, 0, 18, 19, 0, 0, 0, 0, 20, 21, 0, + 0, 0, 0, 22, 0, 0, 23, 24, 0, 0, 0, 25, 0, 21, 26, 0, 0, 27, 0, 0, 0, 21, 0, 0, 0, 0, 0, + 28, 0, 28, 0, 0, 0, 0, 0, 28, 0, 29, 30, 0, 0, 0, 0, 0, 0, 31, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 32, 0, 0, 0, 28, 8, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 33, 34, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 8, 0, 0, 0, 35, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 36, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 37, 0, 38, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 39, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 40, 0, 28, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 41, 42, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 43, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0 + ], + r6: &[ + 0x0000000000000000, 0x000fffffffffff80, 0x01ffffffffffffff, 0x0000000000000c00, + 0x0ffffffe00000000, 0x0000000f00000000, 0x0000000000000402, 0x00000000003e0000, + 0x000003ff00000000, 0xfe000000ff000000, 0x0000ff8000000000, 0xf800000000000000, + 0x000000000fc00000, 0x3000000000000000, 0xfffffffffffcffff, 0x60000000000001ff, + 0x00000000e0000000, 0x0000f80000000000, 0xff000000ff000000, 0x0000fe0000000000, + 0xfc00000000000000, 0x03ff000000000000, 0x7fffffff00000000, 0x0000007fe0000000, + 0x00000000001e0000, 0x0000fffffffc0000, 0xffc0000000000000, 0x001ffffe03ff0000, + 0x0000000003ff0000, 0x00000000000003ff, 0x0fff000000000000, 0x0007ffff00000000, + 0x00001fffffff0000, 0xffffffffffffffff, 0x00007fffffffffff, 0x00000003fbff0000, + 0x00000000007fffff, 0x000fffff00000000, 0x01ffffff00000000, 0xffffffffffffc000, + 0x000000000000ff80, 0xfffe000000000000, 0x001eefffffffffff, 0x0000000000001fff + ], + }; + + pub fn N(c: char) -> bool { + N_table.lookup(c) + } + +} + +pub mod derived_property { + pub const Alphabetic_table: &super::BoolTrie = &super::BoolTrie { + r1: [ + 0x0000000000000000, 0x07fffffe07fffffe, 0x0420040000000000, 0xff7fffffff7fffff, + 0xffffffffffffffff, 0xffffffffffffffff, 0xffffffffffffffff, 0xffffffffffffffff, + 0xffffffffffffffff, 0xffffffffffffffff, 0xffffffffffffffff, 0x0000501f0003ffc3, + 0x0000000000000000, 0xbcdf000000000020, 0xfffffffbffffd740, 0xffbfffffffffffff, + 0xffffffffffffffff, 0xffffffffffffffff, 0xfffffffffffffc03, 0xffffffffffffffff, + 0xfffeffffffffffff, 0xffffffff027fffff, 0xbfff0000000001ff, 0x000787ffffff00b6, + 0xffffffff07ff0000, 0xffffc000feffffff, 0xffffffffffffffff, 0x9c00e1fe1fefffff, + 0xffffffffffff0000, 0xffffffffffffe000, 0x0003ffffffffffff, 0x043007fffffffc00 + ], + r2: [ + 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, + 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 36, 36, 36, 36, 37, 38, 39, 40, 41, + 42, 43, 44, 36, 36, 36, 36, 36, 36, 36, 36, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, + 56, 57, 58, 59, 60, 61, 62, 31, 63, 64, 65, 66, 55, 67, 68, 69, 36, 36, 36, 70, 36, 36, + 36, 36, 71, 72, 73, 74, 31, 75, 76, 31, 77, 78, 79, 31, 31, 31, 31, 31, 31, 31, 31, 31, + 31, 31, 80, 81, 31, 31, 31, 31, 31, 31, 31, 31, 31, 31, 31, 31, 31, 31, 31, 31, 31, 31, + 31, 31, 31, 31, 31, 31, 31, 31, 31, 31, 82, 83, 36, 84, 85, 86, 87, 88, 89, 31, 31, 31, + 31, 31, 31, 31, 90, 44, 91, 92, 93, 36, 94, 95, 31, 31, 31, 31, 31, 31, 31, 31, 36, 36, + 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, + 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, + 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, + 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, + 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 55, 31, 36, 36, 36, 36, 36, 36, 36, 36, + 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, + 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, + 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, + 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, + 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, + 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, + 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, + 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, + 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, + 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, + 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, + 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, + 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, + 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, + 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 96, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, + 36, 36, 36, 36, 36, 36, 36, 36, 97, 98, 36, 36, 36, 36, 99, 100, 36, 96, 101, 36, 102, + 103, 104, 105, 36, 106, 107, 108, 109, 110, 111, 112, 113, 114, 115, 116, 36, 117, 36, + 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, + 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, + 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, + 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, + 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, + 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, + 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, + 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 36, 118, 119, + 31, 31, 31, 31, 31, 31, 31, 31, 31, 31, 31, 31, 31, 31, 31, 31, 31, 31, 31, 31, 31, 31, + 31, 31, 31, 31, 31, 31, 31, 31, 31, 31, 31, 31, 31, 31, 31, 31, 31, 31, 31, 31, 31, 31, + 31, 31, 31, 31, 31, 31, 31, 31, 31, 31, 31, 31, 31, 31, 31, 31, 31, 31, 31, 31, 31, 31, + 31, 31, 31, 31, 31, 31, 31, 31, 31, 31, 31, 31, 31, 31, 31, 31, 31, 31, 31, 31, 31, 31, + 31, 31, 31, 31, 31, 31, 31, 31, 31, 31, 31, 31, 31, 31, 31, 31, 31, 31, 31, 31, 31, 31, + 31, 31, 31, 31, 31, 31, 31, 31, 31, 31, 31, 31, 31, 31, 31, 31, 31, 31, 31, 31, 31, 31, + 36, 36, 36, 36, 36, 120, 36, 121, 122, 123, 124, 125, 36, 36, 36, 36, 126, 127, 128, + 129, 31, 130, 36, 131, 132, 133, 113, 134 + ], + r3: &[ + 0x00001ffffcffffff, 0x000007ff01ffffff, 0x3fdfffff00000000, 0xffff03f8fff00000, + 0xefffffffffffffff, 0xfffe000fffe1dfff, 0xe3c5fdfffff99fef, 0x1003000fb080599f, + 0xc36dfdfffff987ee, 0x003f00005e021987, 0xe3edfdfffffbbfee, 0x1e00000f00011bbf, + 0xe3edfdfffff99fee, 0x0002000fb0c0199f, 0xc3ffc718d63dc7ec, 0x0000000000811dc7, + 0xe3fffdfffffddfef, 0x0000000f07601ddf, 0xe3effdfffffddfef, 0x0006000f40601ddf, + 0xe7fffffffffddfef, 0xfc00000f80f05ddf, 0x2ffbfffffc7fffec, 0x000c0000ff5f807f, + 0x07fffffffffffffe, 0x000000000000207f, 0x3bffecaefef02596, 0x00000000f000205f, + 0x0000000000000001, 0xfffe1ffffffffeff, 0x1ffffffffeffff03, 0x0000000000000000, + 0xf97fffffffffffff, 0xffffc1e7ffff0000, 0xffffffff3000407f, 0xf7ffffffffff20bf, + 0xffffffffffffffff, 0xffffffff3d7f3dff, 0x7f3dffffffff3dff, 0xffffffffff7fff3d, + 0xffffffffff3dffff, 0x0000000087ffffff, 0xffffffff0000ffff, 0x3f3fffffffffffff, + 0xfffffffffffffffe, 0xffff9fffffffffff, 0xffffffff07fffffe, 0x01ffc7ffffffffff, + 0x000fffff000fdfff, 0x000ddfff000fffff, 0xffcfffffffffffff, 0x00000000108001ff, + 0xffffffff00000000, 0x01ffffffffffffff, 0xffff07ffffffffff, 0x003fffffffffffff, + 0x01ff0fff7fffffff, 0x001f3fffffff0000, 0xffff0fffffffffff, 0x00000000000003ff, + 0xffffffff0fffffff, 0x001ffffe7fffffff, 0x0000008000000000, 0xffefffffffffffff, + 0x0000000000000fef, 0xfc00f3ffffffffff, 0x0003ffbfffffffff, 0x3ffffffffc00e000, + 0xe7ffffffffff01ff, 0x006fde0000000000, 0x001fff8000000000, 0xffffffff3f3fffff, + 0x3fffffffaaff3f3f, 0x5fdfffffffffffff, 0x1fdc1fff0fcf1fdc, 0x8002000000000000, + 0x000000001fff0000, 0xf3ffbd503e2ffc84, 0xffffffff000043e0, 0x00000000000001ff, + 0xffc0000000000000, 0x000003ffffffffff, 0xffff7fffffffffff, 0xffffffff7fffffff, + 0x000c781fffffffff, 0xffff20bfffffffff, 0x000080ffffffffff, 0x7f7f7f7f007fffff, + 0xffffffff7f7f7f7f, 0x0000800000000000, 0x1f3e03fe000000e0, 0xfffffffee07fffff, + 0xf7ffffffffffffff, 0xfffeffffffffffe0, 0x07ffffff00007fff, 0xffff000000000000, + 0x0000ffffffffffff, 0x0000000000001fff, 0x3fffffffffff0000, 0x00000c00ffff1fff, + 0x8ff07fffffffffff, 0xfffffffcff800000, 0x03fffffffffff9ff, 0xff80000000000000, + 0x000000fffffff7bb, 0x000fffffffffffff, 0x68fc00000000002f, 0xffff07fffffffc00, + 0x1fffffff0007ffff, 0xfff7ffffffffffff, 0x7c00ffdf00008000, 0x007fffffffffffff, + 0xc47fffff00003fff, 0x7fffffffffffffff, 0x003cffff38000005, 0xffff7f7f007e7e7e, + 0xffff003ff7ffffff, 0x000007ffffffffff, 0xffff000fffffffff, 0x0ffffffffffff87f, + 0xffff3fffffffffff, 0x0000000003ffffff, 0x5f7ffdffe0f8007f, 0xffffffffffffffdb, + 0x0003ffffffffffff, 0xfffffffffff80000, 0x3fffffffffffffff, 0xffffffffffff0000, + 0xfffffffffffcffff, 0x0fff0000000000ff, 0xffdf000000000000, 0x1fffffffffffffff, + 0x07fffffe00000000, 0xffffffc007fffffe, 0x000000001cfcfcfc + ], + r4: [ + 0, 1, 2, 3, 4, 5, 6, 7, 8, 5, 5, 9, 5, 10, 11, 12, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 13, 14, + 15, 7, 16, 17, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, + 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, + 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, + 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, + 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, + 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, + 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, + 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5 + ], + r5: &[ + 0, 1, 2, 3, 4, 5, 4, 4, 4, 4, 6, 7, 8, 9, 10, 11, 2, 2, 12, 13, 14, 15, 4, 4, 2, 2, 2, + 2, 16, 17, 4, 4, 18, 19, 20, 21, 22, 4, 23, 4, 24, 25, 26, 27, 28, 29, 30, 4, 2, 31, 32, + 32, 33, 4, 4, 4, 4, 4, 4, 4, 34, 35, 4, 4, 2, 35, 36, 37, 32, 38, 2, 39, 40, 4, 41, 42, + 43, 44, 4, 4, 2, 45, 2, 46, 4, 4, 47, 48, 49, 50, 28, 4, 51, 4, 4, 4, 52, 4, 53, 54, 4, + 4, 4, 4, 55, 56, 57, 52, 4, 4, 4, 4, 58, 59, 60, 4, 61, 62, 63, 4, 4, 4, 4, 64, 4, 4, 4, + 4, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 65, 4, 2, 66, 2, 2, 2, 67, 4, 4, 4, 4, 4, + 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, + 4, 4, 4, 4, 4, 4, 4, 4, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 66, 4, 4, 4, 4, + 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, + 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, + 4, 2, 2, 2, 2, 2, 2, 2, 2, 2, 68, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, + 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, + 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, + 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, + 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, + 2, 2, 2, 2, 2, 2, 2, 2, 52, 20, 4, 69, 16, 70, 71, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 2, 4, + 4, 2, 72, 73, 74, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, + 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, + 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, + 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 75, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 32, 4, 4, + 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 2, 2, 2, 2, 20, 76, 2, 2, 2, 2, 2, + 77, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, + 4, 4, 4, 4, 4, 4, 4, 4, 2, 78, 79, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, + 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 2, 80, 81, 82, 83, 84, 2, 2, 2, 2, 85, 86, 87, 88, 89, + 90, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, + 4, 4, 4, 4, 91, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, + 4, 4, 4, 4, 4, 4, 4, 2, 2, 2, 92, 2, 93, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, + 4, 4, 94, 95, 96, 4, 4, 4, 4, 4, 4, 4, 4, 4, 76, 97, 98, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, + 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, + 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, + 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 99, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, + 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, + 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 5, 2, 2, 2, 10, 2, + 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, + 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, + 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, + 2, 100, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, + 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 101, 4, 4, 4, + 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, + 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 2, 2, 2, 2, 2, 2, 2, 2, 102, 4, 4, 4, 4, + 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4 + ], + r6: &[ + 0xb7ffff7fffffefff, 0x000000003fff3fff, 0xffffffffffffffff, 0x07ffffffffffffff, + 0x0000000000000000, 0x001fffffffffffff, 0xffffffff1fffffff, 0x000000000001ffff, + 0xffffe000ffffffff, 0x07ffffffffff07ff, 0xffffffff3fffffff, 0x00000000003eff0f, + 0xffff00003fffffff, 0x0fffffffff0fffff, 0xffff00ffffffffff, 0x0000000fffffffff, + 0x007fffffffffffff, 0x000000ff003fffff, 0x91bffffffffffd3f, 0x007fffff003fffff, + 0x000000007fffffff, 0x0037ffff00000000, 0x03ffffff003fffff, 0xc0ffffffffffffff, + 0x003ffffffeeff06f, 0x1fffffff00000000, 0x000000001fffffff, 0x0000001ffffffeff, + 0x003fffffffffffff, 0x0007ffff003fffff, 0x000000000003ffff, 0x00000000000001ff, + 0x0007ffffffffffff, 0x000000ffffffffff, 0xffff00801fffffff, 0x000000000000003f, + 0x01fffffffffffffc, 0x000001ffffff0000, 0x0047ffffffff0070, 0x000000001400001e, + 0x409ffffffffbffff, 0xffff01ffbfffbd7f, 0x000001ffffffffff, 0xe3edfdfffff99fef, + 0x0000000fe081199f, 0x00000000000007bb, 0x00000000000000b3, 0x7f3fffffffffffff, + 0x000000003f000000, 0x7fffffffffffffff, 0x0000000000000011, 0x000007ffe7ffffff, + 0x01ffffffffffffff, 0xffffffff00000000, 0x80000000ffffffff, 0x7fe7ffffffffffff, + 0xffffffffffff0000, 0x0000000020ffffcf, 0x7f7ffffffffffdff, 0xfffc000000000001, + 0x007ffefffffcffff, 0xb47ffffffffffb7f, 0xfffffdbf000000cb, 0x00000000017b7fff, + 0x007fffff00000000, 0x0000000003ffffff, 0x00007fffffffffff, 0x000000000000000f, + 0x000000000000007f, 0x00003fffffff0000, 0xe0fffff80000000f, 0x000000000000ffff, + 0x7fffffffffff001f, 0x00000000fff80000, 0x0000000300000000, 0x0003ffffffffffff, + 0xffff000000000000, 0x0fffffffffffffff, 0x1fff07ffffffffff, 0x0000000043ff01ff, + 0xffffffffffdfffff, 0xebffde64dfffffff, 0xffffffffffffffef, 0x7bffffffdfdfe7bf, + 0xfffffffffffdfc5f, 0xffffff3fffffffff, 0xf7fffffff7fffffd, 0xffdfffffffdfffff, + 0xffff7fffffff7fff, 0xfffffdfffffffdff, 0x0000000000000ff7, 0x000007dbf9ffff7f, + 0x000000000000001f, 0x000000000000008f, 0x0af7fe96ffffffef, 0x5ef7f796aa96ea84, + 0x0ffffbee0ffffbff, 0xffff03ffffff03ff, 0x00000000000003ff, 0x00000000007fffff, + 0xffff0003ffffffff, 0x00000001ffffffff, 0x000000003fffffff + ], + }; + + pub fn Alphabetic(c: char) -> bool { + Alphabetic_table.lookup(c) + } + + pub const Case_Ignorable_table: &super::BoolTrie = &super::BoolTrie { + r1: [ + 0x0400408000000000, 0x0000000140000000, 0x0190a10000000000, 0x0000000000000000, + 0x0000000000000000, 0x0000000000000000, 0x0000000000000000, 0x0000000000000000, + 0x0000000000000000, 0x0000000000000000, 0xffff000000000000, 0xffffffffffffffff, + 0xffffffffffffffff, 0x0430ffffffffffff, 0x00000000000000b0, 0x0000000000000000, + 0x0000000000000000, 0x0000000000000000, 0x00000000000003f8, 0x0000000000000000, + 0x0000000000000000, 0x0000000002000000, 0xbffffffffffe0000, 0x00100000000000b6, + 0x0000000017ff003f, 0x00010000fffff801, 0x0000000000000000, 0x00003dffbfc00000, + 0xffff000000028000, 0x00000000000007ff, 0x0001ffc000000000, 0x243ff80000000000 + ], + r2: [ + 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 8, 10, 11, 12, 13, 14, 15, 16, 11, 17, 18, 19, 2, 20, 21, + 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 2, 2, 2, 2, 2, 2, 2, 2, 2, 33, 2, 2, 2, 2, + 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 34, 35, 36, 37, 38, 39, 40, 2, 41, 2, 2, 2, 42, 43, 44, 2, + 45, 46, 47, 48, 49, 50, 2, 51, 52, 53, 54, 55, 2, 2, 2, 2, 2, 2, 56, 57, 58, 59, 60, 61, + 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, + 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 62, 2, 63, 2, 64, 2, 65, 66, 2, 2, 2, 2, + 2, 2, 2, 67, 2, 68, 69, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, + 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, + 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, + 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, + 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, + 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, + 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, + 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, + 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, + 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, + 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, + 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, + 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, + 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, + 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, + 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 70, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, + 2, 2, 2, 2, 2, 2, 2, 50, 2, 2, 2, 2, 71, 72, 73, 74, 75, 76, 77, 78, 79, 2, 2, 80, 81, + 82, 83, 84, 85, 86, 87, 88, 2, 89, 2, 90, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, + 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, + 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, + 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, + 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, + 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, + 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, + 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, + 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, + 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, + 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, + 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 91, 2, 92, 93, 2, 2, 2, 2, 2, 2, 2, 2, 94, 95, 2, 96, + 97, 98, 99, 100 + ], + r3: &[ + 0x00003fffffc00000, 0x000000000e000000, 0x0000000000000000, 0xfffffffffff80000, + 0x1400000000000007, 0x0002000c00fe21fe, 0x1000000000000002, 0x4000000c0000201e, + 0x1000000000000006, 0x0023000000023986, 0xfc00000c000021be, 0x9000000000000002, + 0x0000000c0040201e, 0x0000000000000004, 0x0000000000002001, 0xc000000000000011, + 0x0000000c00603dc1, 0x0000000c00003040, 0x1800000000000003, 0x0000000c0000201e, + 0x00000000005c0400, 0x07f2000000000000, 0x0000000000007fc0, 0x1bf2000000000000, + 0x0000000000003f40, 0x02a0000003000000, 0x7ffe000000000000, 0x1ffffffffeffe0df, + 0x0000000000000040, 0x66fde00000000000, 0x001e0001c3000000, 0x0000000020002064, + 0x1000000000000000, 0x00000000e0000000, 0x001c0000001c0000, 0x000c0000000c0000, + 0x3fb0000000000000, 0x00000000208ffe40, 0x0000000000007800, 0x0000000000000008, + 0x0000020000000060, 0x0e04018700000000, 0x0000000009800000, 0x9ff81fe57f400000, + 0x7fff008000000000, 0x17d000000000000f, 0x000ff80000000004, 0x00003b3c00000003, + 0x0003a34000000000, 0x00cff00000000000, 0x3f00000000000000, 0x031021fdfff70000, + 0xfffff00000000000, 0x010007ffffffffff, 0xfffffffff8000000, 0xfbffffffffffffff, + 0xa000000000000000, 0x6000e000e000e003, 0x00007c900300f800, 0x8002ffdf00000000, + 0x000000001fff0000, 0x0001ffffffff0000, 0x3000000000000000, 0x0003800000000000, + 0x8000800000000000, 0xffffffff00000000, 0x0000800000000000, 0x083e3c0000000020, + 0x000000007e000000, 0x7000000000000000, 0x0000000000200000, 0x0000000000001000, + 0xbff7800000000000, 0x00000000f0000000, 0x0003000000000000, 0x00000003ffffffff, + 0x0001000000000000, 0x0000000000000700, 0x0300000000000000, 0x0000006000000844, + 0x8003ffff00000030, 0x00003fc000000000, 0x000000000003ff80, 0x13c8000000000007, + 0x0000006000008000, 0x00667e0000000000, 0x1001000000001008, 0xc19d000000000000, + 0x0058300020000002, 0x00000000f8000000, 0x0000212000000000, 0x0000000040000000, + 0xfffc000000000000, 0x0000000000000003, 0x0000ffff0008ffff, 0x0000000000240000, + 0x8000000000000000, 0x4000000004004080, 0x0001000000000001, 0x00000000c0000000, + 0x0e00000800000000 + ], + r4: [ + 0, 1, 2, 2, 2, 2, 3, 2, 2, 2, 2, 4, 2, 5, 6, 7, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, + 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, + 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, + 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, + 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, + 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, + 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, + 2, 2, 2, 2, 2, 8, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, + 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2 + ], + r5: &[ + 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 2, 0, 3, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 4, 0, 0, 5, 0, 0, 0, 0, 0, 0, 0, 0, 6, 0, 0, 0, 0, 0, + 0, 0, 0, 7, 0, 0, 8, 9, 10, 11, 12, 13, 14, 15, 16, 0, 0, 17, 18, 19, 0, 0, 20, 21, 22, + 23, 0, 0, 24, 25, 26, 27, 28, 0, 29, 0, 0, 0, 30, 0, 0, 0, 0, 0, 0, 0, 31, 32, 33, 0, 0, + 0, 0, 0, 34, 0, 35, 0, 36, 37, 38, 0, 0, 0, 0, 39, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 40, 41, 42, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 43, 44, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 45, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 46, 47, 0, 0, 48, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 49, 50, 51, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 52, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 53, 0, 54, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 55, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 56, 57, 0, 0, 57, 57, 57, 58, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0 + ], + r6: &[ + 0x0000000000000000, 0x2000000000000000, 0x0000000100000000, 0x07c0000000000000, + 0x870000000000f06e, 0x0000006000000000, 0x000000f000000000, 0x000000000001ffc0, + 0xff00000000000002, 0x800000000000007f, 0x2678000000000003, 0x0000000000002000, + 0x001fef8000000007, 0x0008000000000000, 0x7fc0000000000003, 0x0000000000001e00, + 0x40d3800000000000, 0x000007f880000000, 0x1800000000000003, 0x001f1fc000000001, + 0xff00000000000000, 0x000000004000005c, 0x85f8000000000000, 0x000000000000000d, + 0xb03c000000000000, 0x0000000030000001, 0xa7f8000000000000, 0x0000000000000001, + 0x00bf280000000000, 0x00000fbce0000000, 0x06ff800000000000, 0x79f80000000007fe, + 0x000000000e7e0080, 0x00000000037ffc00, 0xbf7f000000000000, 0x006dfcfffffc0000, + 0xb47e000000000000, 0x00000000000000bf, 0x0000000000a30000, 0x0018000000000000, + 0x001f000000000000, 0x007f000000000000, 0x000000000000000f, 0x00000000ffff8000, + 0x0000000300000000, 0x0000000f60000000, 0xfff8038000000000, 0x00003c0000000fe7, + 0x000000000000001c, 0xf87fffffffffffff, 0x00201fffffffffff, 0x0000fffef8000010, + 0x000007dbf9ffff7f, 0x00000000007f0000, 0x00000000000007f0, 0xf800000000000000, + 0xffffffff00000002, 0xffffffffffffffff, 0x0000ffffffffffff + ], + }; + + pub fn Case_Ignorable(c: char) -> bool { + Case_Ignorable_table.lookup(c) + } + + pub const Cased_table: &super::BoolTrie = &super::BoolTrie { + r1: [ + 0x0000000000000000, 0x07fffffe07fffffe, 0x0420040000000000, 0xff7fffffff7fffff, + 0xffffffffffffffff, 0xffffffffffffffff, 0xf7ffffffffffffff, 0xfffffffffffffff0, + 0xffffffffffffffff, 0xffffffffffffffff, 0x01ffffffffefffff, 0x0000001f00000003, + 0x0000000000000000, 0xbccf000000000020, 0xfffffffbffffd740, 0xffbfffffffffffff, + 0xffffffffffffffff, 0xffffffffffffffff, 0xfffffffffffffc03, 0xffffffffffffffff, + 0xfffeffffffffffff, 0xffffffff007fffff, 0x00000000000001ff, 0x0000000000000000, + 0x0000000000000000, 0x0000000000000000, 0x0000000000000000, 0x0000000000000000, + 0x0000000000000000, 0x0000000000000000, 0x0000000000000000, 0x0000000000000000 + ], + r2: [ + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 1, 2, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 3, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 4, 0, 5, 5, 5, + 0, 5, 5, 5, 5, 6, 7, 8, 9, 0, 10, 11, 0, 12, 13, 14, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 15, 16, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 17, 18, 5, 19, 20, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 21, 22, + 0, 23, 5, 24, 25, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 26, 27, 5, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 28, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 29, 30, 0, 0 + ], + r3: &[ + 0x0000000000000000, 0xffffffff00000000, 0xe7ffffffffff20bf, 0x3f3fffffffffffff, + 0xe7ffffffffff01ff, 0xffffffffffffffff, 0xffffffff3f3fffff, 0x3fffffffaaff3f3f, + 0x5fdfffffffffffff, 0x1fdc1fff0fcf1fdc, 0x8002000000000000, 0x000000001fff0000, + 0xf21fbd503e2ffc84, 0xffffffff000043e0, 0x0000000000000018, 0xffc0000000000000, + 0x000003ffffffffff, 0xffff7fffffffffff, 0xffffffff7fffffff, 0x000c781fffffffff, + 0x000020bfffffffff, 0x00003fffffffffff, 0x000000003fffffff, 0xfffffffc00000000, + 0x03ffffffffff78ff, 0x0700000000000000, 0xffff000000000000, 0xffff003ff7ffffff, + 0x0000000000f8007f, 0x07fffffe00000000, 0x0000000007fffffe + ], + r4: [ + 0, 1, 2, 2, 2, 2, 3, 2, 2, 2, 2, 2, 2, 4, 5, 6, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, + 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, + 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, + 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, + 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, + 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, + 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, + 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, + 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2 + ], + r5: &[ + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 2, 3, 4, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 5, 5, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 6, 7, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 8, 9, 10, 11, 12, 1, 1, 1, 1, 13, 14, 15, 16, 17, + 18, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 19, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 3, 20, 21, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0 + ], + r6: &[ + 0x0000000000000000, 0xffffffffffffffff, 0x000000000000ffff, 0xffff000000000000, + 0x0fffffffff0fffff, 0x0007ffffffffffff, 0xffffffff00000000, 0x00000000ffffffff, + 0xffffffffffdfffff, 0xebffde64dfffffff, 0xffffffffffffffef, 0x7bffffffdfdfe7bf, + 0xfffffffffffdfc5f, 0xffffff3fffffffff, 0xf7fffffff7fffffd, 0xffdfffffffdfffff, + 0xffff7fffffff7fff, 0xfffffdfffffffdff, 0x0000000000000ff7, 0x000000000000000f, + 0xffff03ffffff03ff, 0x00000000000003ff + ], + }; + + pub fn Cased(c: char) -> bool { + Cased_table.lookup(c) + } + + pub const Grapheme_Extend_table: &super::BoolTrie = &super::BoolTrie { + r1: [ + 0x0000000000000000, 0x0000000000000000, 0x0000000000000000, 0x0000000000000000, + 0x0000000000000000, 0x0000000000000000, 0x0000000000000000, 0x0000000000000000, + 0x0000000000000000, 0x0000000000000000, 0x0000000000000000, 0x0000000000000000, + 0xffffffffffffffff, 0x0000ffffffffffff, 0x0000000000000000, 0x0000000000000000, + 0x0000000000000000, 0x0000000000000000, 0x00000000000003f8, 0x0000000000000000, + 0x0000000000000000, 0x0000000000000000, 0xbffffffffffe0000, 0x00000000000000b6, + 0x0000000007ff0000, 0x00010000fffff800, 0x0000000000000000, 0x00003d9f9fc00000, + 0xffff000000020000, 0x00000000000007ff, 0x0001ffc000000000, 0x200ff80000000000 + ], + r2: [ + 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 8, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 2, 21, 22, + 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 33, 2, 2, 2, 2, 2, + 2, 2, 2, 2, 2, 2, 2, 2, 2, 34, 35, 36, 37, 38, 2, 39, 2, 40, 2, 2, 2, 41, 42, 43, 2, 44, + 45, 46, 47, 48, 2, 2, 49, 2, 2, 2, 50, 2, 2, 2, 2, 2, 2, 2, 2, 51, 2, 2, 52, 2, 2, 2, 2, + 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, + 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 53, 2, 54, 2, 55, 2, 2, 2, 2, 2, 2, 2, 2, 56, + 2, 57, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, + 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, + 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, + 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, + 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, + 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, + 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, + 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, + 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, + 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, + 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, + 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, + 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, + 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, + 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, + 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, + 2, 2, 2, 2, 2, 2, 2, 2, 58, 59, 60, 2, 2, 2, 2, 61, 2, 2, 62, 63, 64, 65, 66, 67, 68, + 69, 70, 2, 2, 2, 71, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, + 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, + 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, + 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, + 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, + 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, + 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, + 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, + 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, + 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, + 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, + 2, 2, 2, 2, 72, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 73, 2, 2, 2, 2, 2, 59, 2 + ], + r3: &[ + 0x00003eeffbc00000, 0x000000000e000000, 0x0000000000000000, 0xfffffffbfff80000, + 0x1400000000000007, 0x0000000c00fe21fe, 0x5000000000000002, 0x4000000c0080201e, + 0x1000000000000006, 0x0023000000023986, 0xfc00000c000021be, 0xd000000000000002, + 0x0000000c00c0201e, 0x4000000000000004, 0x0000000000802001, 0xc000000000000011, + 0x0000000c00603dc1, 0x9000000000000002, 0x0000000c00603044, 0x5800000000000003, + 0x0000000c0080201e, 0x00000000805c8400, 0x07f2000000000000, 0x0000000000007f80, + 0x1bf2000000000000, 0x0000000000003f00, 0x02a0000003000000, 0x7ffe000000000000, + 0x1ffffffffeffe0df, 0x0000000000000040, 0x66fde00000000000, 0x001e0001c3000000, + 0x0000000020002064, 0x00000000e0000000, 0x001c0000001c0000, 0x000c0000000c0000, + 0x3fb0000000000000, 0x00000000200ffe40, 0x0000000000003800, 0x0000020000000060, + 0x0e04018700000000, 0x0000000009800000, 0x9ff81fe57f400000, 0x7fff000000000000, + 0x17d000000000000f, 0x000ff80000000004, 0x00003b3c00000003, 0x0003a34000000000, + 0x00cff00000000000, 0x031021fdfff70000, 0xfbffffffffffffff, 0x0000000000001000, + 0x0001ffffffff0000, 0x0003800000000000, 0x8000000000000000, 0xffffffff00000000, + 0x0000fc0000000000, 0x0000000006000000, 0x3ff7800000000000, 0x00000000c0000000, + 0x0003000000000000, 0x0000006000000844, 0x8003ffff00000030, 0x00003fc000000000, + 0x000000000003ff80, 0x13c8000000000007, 0x0000002000000000, 0x00667e0000000000, + 0x1000000000001008, 0xc19d000000000000, 0x0040300000000002, 0x0000212000000000, + 0x0000000040000000, 0x0000ffff0000ffff + ], + r4: [ + 0, 1, 2, 2, 2, 2, 3, 2, 2, 2, 2, 4, 2, 5, 6, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, + 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, + 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, + 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, + 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, + 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, + 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, + 2, 2, 2, 2, 2, 7, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, + 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2 + ], + r5: &[ + 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 2, 0, 3, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 4, 0, 0, 5, 0, 0, 0, 0, 0, 0, 0, 0, 6, 0, 0, 0, 0, 0, + 0, 0, 0, 7, 0, 0, 8, 9, 10, 0, 11, 12, 13, 14, 15, 0, 0, 16, 17, 18, 0, 0, 19, 20, 21, + 22, 0, 0, 23, 24, 25, 26, 27, 0, 28, 0, 0, 0, 29, 0, 0, 0, 0, 0, 0, 0, 30, 31, 32, 0, 0, + 0, 0, 0, 33, 0, 34, 0, 35, 36, 37, 0, 0, 0, 0, 38, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 39, 40, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 41, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 42, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 43, 44, 0, 0, 45, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 46, 47, 48, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 49, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 50, 0, 51, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 52, 53, 0, 0, + 53, 53, 53, 54, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0 + ], + r6: &[ + 0x0000000000000000, 0x2000000000000000, 0x0000000100000000, 0x07c0000000000000, + 0x870000000000f06e, 0x0000006000000000, 0x000000f000000000, 0x000000000001ffc0, + 0xff00000000000002, 0x800000000000007f, 0x0678000000000003, 0x001fef8000000007, + 0x0008000000000000, 0x7fc0000000000003, 0x0000000000001e00, 0x40d3800000000000, + 0x000007f880000000, 0x5800000000000003, 0x001f1fc000800001, 0xff00000000000000, + 0x000000004000005c, 0xa5f9000000000000, 0x000000000000000d, 0xb03c800000000000, + 0x0000000030000001, 0xa7f8000000000000, 0x0000000000000001, 0x00bf280000000000, + 0x00000fbce0000000, 0x06ff800000000000, 0x79f80000000007fe, 0x000000000e7e0080, + 0x00000000037ffc00, 0xbf7f000000000000, 0x006dfcfffffc0000, 0xb47e000000000000, + 0x00000000000000bf, 0x0000000000a30000, 0x0018000000000000, 0x001f000000000000, + 0x007f000000000000, 0x0000000000078000, 0x0000000060000000, 0xf807c3a000000000, + 0x00003c0000000fe7, 0x000000000000001c, 0xf87fffffffffffff, 0x00201fffffffffff, + 0x0000fffef8000010, 0x000007dbf9ffff7f, 0x00000000007f0000, 0x00000000000007f0, + 0xffffffff00000000, 0xffffffffffffffff, 0x0000ffffffffffff + ], + }; + + pub fn Grapheme_Extend(c: char) -> bool { + Grapheme_Extend_table.lookup(c) + } + + pub const Lowercase_table: &super::BoolTrie = &super::BoolTrie { + r1: [ + 0x0000000000000000, 0x07fffffe00000000, 0x0420040000000000, 0xff7fffff80000000, + 0x55aaaaaaaaaaaaaa, 0xd4aaaaaaaaaaab55, 0xe6512d2a4e243129, 0xaa29aaaab5555240, + 0x93faaaaaaaaaaaaa, 0xffffffffffffaa85, 0x01ffffffffefffff, 0x0000001f00000003, + 0x0000000000000000, 0x3c8a000000000020, 0xfffff00000010000, 0x192faaaaaae37fff, + 0xffff000000000000, 0xaaaaaaaaffffffff, 0xaaaaaaaaaaaaa802, 0xaaaaaaaaaaaad554, + 0x0000aaaaaaaaaaaa, 0xffffffff00000000, 0x00000000000001ff, 0x0000000000000000, + 0x0000000000000000, 0x0000000000000000, 0x0000000000000000, 0x0000000000000000, + 0x0000000000000000, 0x0000000000000000, 0x0000000000000000, 0x0000000000000000 + ], + r2: [ + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 2, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 3, 0, 4, 4, 4, + 0, 5, 5, 6, 5, 7, 8, 9, 10, 0, 11, 12, 0, 13, 14, 15, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 16, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 17, 18, 5, 19, 20, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 21, 22, + 0, 23, 24, 25, 26, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 17, 27, 4, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 28, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 29, 0, 0 + ], + r3: &[ + 0x0000000000000000, 0xe7ffffffffff0000, 0x3f00000000000000, 0x00000000000001ff, + 0xffffffffffffffff, 0xaaaaaaaaaaaaaaaa, 0xaaaaaaaabfeaaaaa, 0x00ff00ff003f00ff, + 0x3fff00ff00ff003f, 0x40df00ff00ff00ff, 0x00dc00ff00cf00dc, 0x8002000000000000, + 0x000000001fff0000, 0x321080000008c400, 0xffff0000000043c0, 0x0000000000000010, + 0x000003ffffff0000, 0xffff000000000000, 0x3fda15627fffffff, 0x0008501aaaaaaaaa, + 0x000020bfffffffff, 0x00002aaaaaaaaaaa, 0x000000003aaaaaaa, 0xaaabaaa800000000, + 0x95ffaaaaaaaaaaaa, 0x02a082aaaaba50aa, 0x0700000000000000, 0xffff003ff7ffffff, + 0x0000000000f8007f, 0x0000000007fffffe + ], + r4: [ + 0, 1, 2, 2, 2, 2, 3, 2, 2, 2, 2, 2, 2, 4, 5, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, + 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, + 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, + 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, + 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, + 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, + 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, + 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, + 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2 + ], + r5: &[ + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 2, 0, 3, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 4, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 5, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 6, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, + 21, 22, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 23, 24, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0 + ], + r6: &[ + 0x0000000000000000, 0xffffff0000000000, 0x000000000000ffff, 0x0fffffffff000000, + 0x0007ffffffffffff, 0x00000000ffffffff, 0xffffffff00000000, 0x000ffffffc000000, + 0x000000ffffdfc000, 0xebc000000ffffffc, 0xfffffc000000ffef, 0x00ffffffc000000f, + 0x00000ffffffc0000, 0xfc000000ffffffc0, 0xffffc000000fffff, 0x0ffffffc000000ff, + 0x0000ffffffc00000, 0x0000003ffffffc00, 0xf0000003f7fffffc, 0xffc000000fdfffff, + 0xffff0000003f7fff, 0xfffffc000000fdff, 0x0000000000000bf7, 0xfffffffc00000000, + 0x000000000000000f + ], + }; + + pub fn Lowercase(c: char) -> bool { + Lowercase_table.lookup(c) + } + + pub const Uppercase_table: &super::BoolTrie = &super::BoolTrie { + r1: [ + 0x0000000000000000, 0x0000000007fffffe, 0x0000000000000000, 0x000000007f7fffff, + 0xaa55555555555555, 0x2b555555555554aa, 0x11aed2d5b1dbced6, 0x55d255554aaaa490, + 0x6c05555555555555, 0x000000000000557a, 0x0000000000000000, 0x0000000000000000, + 0x0000000000000000, 0x8045000000000000, 0x00000ffbfffed740, 0xe6905555551c8000, + 0x0000ffffffffffff, 0x5555555500000000, 0x5555555555555401, 0x5555555555552aab, + 0xfffe555555555555, 0x00000000007fffff, 0x0000000000000000, 0x0000000000000000, + 0x0000000000000000, 0x0000000000000000, 0x0000000000000000, 0x0000000000000000, + 0x0000000000000000, 0x0000000000000000, 0x0000000000000000, 0x0000000000000000 + ], + r2: [ + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 1, 2, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 3, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 4, 0, 0, 0, 0, + 0, 5, 5, 6, 5, 7, 8, 9, 10, 0, 0, 0, 0, 11, 12, 13, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 14, + 15, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 16, 17, 5, 18, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 19, 20, 0, + 21, 22, 23, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 24, 0, 0, 0 + ], + r3: &[ + 0x0000000000000000, 0xffffffff00000000, 0x00000000000020bf, 0x003fffffffffffff, + 0xe7ffffffffff0000, 0x5555555555555555, 0x5555555540155555, 0xff00ff003f00ff00, + 0x0000ff00aa003f00, 0x0f00000000000000, 0x0f001f000f000f00, 0xc00f3d503e273884, + 0x0000ffff00000020, 0x0000000000000008, 0xffc0000000000000, 0x000000000000ffff, + 0x00007fffffffffff, 0xc025ea9d00000000, 0x0004280555555555, 0x0000155555555555, + 0x0000000005555555, 0x5554555400000000, 0x6a00555555555555, 0x015f7d5555452855, + 0x07fffffe00000000 + ], + r4: [ + 0, 1, 2, 2, 2, 2, 3, 2, 2, 2, 2, 2, 2, 4, 5, 6, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, + 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, + 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, + 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, + 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, + 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, + 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, + 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, + 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2 + ], + r5: &[ + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 2, 3, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 4, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 5, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 6, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, + 21, 22, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 23, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 2, 24, 25, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0 + ], + r6: &[ + 0x0000000000000000, 0x000000ffffffffff, 0xffff000000000000, 0x00000000000fffff, + 0x0007ffffffffffff, 0xffffffff00000000, 0x00000000ffffffff, 0xfff0000003ffffff, + 0xffffff0000003fff, 0x003fde64d0000003, 0x000003ffffff0000, 0x7b0000001fdfe7b0, + 0xfffff0000001fc5f, 0x03ffffff0000003f, 0x00003ffffff00000, 0xf0000003ffffff00, + 0xffff0000003fffff, 0xffffff00000003ff, 0x07fffffc00000001, 0x001ffffff0000000, + 0x00007fffffc00000, 0x000001ffffff0000, 0x0000000000000400, 0x00000003ffffffff, + 0xffff03ffffff03ff, 0x00000000000003ff + ], + }; + + pub fn Uppercase(c: char) -> bool { + Uppercase_table.lookup(c) + } + + pub const XID_Continue_table: &super::BoolTrie = &super::BoolTrie { + r1: [ + 0x03ff000000000000, 0x07fffffe87fffffe, 0x04a0040000000000, 0xff7fffffff7fffff, + 0xffffffffffffffff, 0xffffffffffffffff, 0xffffffffffffffff, 0xffffffffffffffff, + 0xffffffffffffffff, 0xffffffffffffffff, 0xffffffffffffffff, 0x0000501f0003ffc3, + 0xffffffffffffffff, 0xb8dfffffffffffff, 0xfffffffbffffd7c0, 0xffbfffffffffffff, + 0xffffffffffffffff, 0xffffffffffffffff, 0xfffffffffffffcfb, 0xffffffffffffffff, + 0xfffeffffffffffff, 0xffffffff027fffff, 0xbffffffffffe01ff, 0x000787ffffff00b6, + 0xffffffff07ff0000, 0xffffc3ffffffffff, 0xffffffffffffffff, 0x9ffffdff9fefffff, + 0xffffffffffff0000, 0xffffffffffffe7ff, 0x0003ffffffffffff, 0x243fffffffffffff + ], + r2: [ + 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, + 24, 25, 26, 27, 28, 29, 30, 31, 4, 32, 33, 34, 4, 4, 4, 4, 4, 35, 36, 37, 38, 39, 40, + 41, 42, 4, 4, 4, 4, 4, 4, 4, 4, 43, 44, 45, 46, 47, 4, 48, 49, 50, 51, 52, 53, 54, 55, + 56, 57, 58, 59, 60, 4, 61, 4, 62, 63, 64, 65, 66, 4, 4, 4, 67, 4, 4, 4, 4, 68, 69, 70, + 71, 72, 73, 74, 75, 76, 77, 78, 60, 60, 60, 60, 60, 60, 60, 60, 60, 60, 60, 60, 60, 60, + 60, 60, 60, 60, 60, 60, 60, 60, 60, 60, 60, 60, 60, 60, 60, 60, 60, 60, 60, 60, 60, 60, + 60, 60, 60, 60, 60, 79, 80, 4, 81, 82, 83, 84, 85, 60, 60, 60, 60, 60, 60, 60, 60, 86, + 42, 87, 88, 89, 4, 90, 91, 60, 60, 60, 60, 60, 60, 60, 60, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, + 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, + 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, + 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, + 4, 4, 4, 4, 4, 52, 60, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, + 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, + 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, + 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, + 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, + 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, + 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, + 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, + 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, + 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, + 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, + 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 92, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, + 4, 4, 4, 4, 4, 93, 94, 4, 4, 4, 4, 95, 96, 4, 97, 98, 4, 99, 100, 101, 62, 4, 102, 103, + 104, 4, 105, 106, 107, 4, 108, 109, 110, 4, 111, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, + 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, + 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, + 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, + 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, + 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, + 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 112, 113, 60, 60, 60, 60, 60, 60, 60, + 60, 60, 60, 60, 60, 60, 60, 60, 60, 60, 60, 60, 60, 60, 60, 60, 60, 60, 60, 60, 60, 60, + 60, 60, 60, 60, 60, 60, 60, 60, 60, 60, 60, 60, 60, 60, 60, 60, 60, 60, 60, 60, 60, 60, + 60, 60, 60, 60, 60, 60, 60, 60, 60, 60, 60, 60, 60, 60, 60, 60, 60, 60, 60, 60, 60, 60, + 60, 60, 60, 60, 60, 60, 60, 60, 60, 60, 60, 60, 60, 60, 60, 60, 60, 60, 60, 60, 60, 60, + 60, 60, 60, 60, 60, 60, 60, 60, 60, 60, 60, 60, 60, 60, 60, 60, 60, 60, 60, 60, 60, 60, + 60, 60, 60, 60, 60, 60, 60, 60, 60, 60, 60, 60, 60, 60, 60, 4, 4, 4, 4, 4, 103, 4, 114, + 115, 116, 97, 117, 4, 118, 4, 4, 119, 120, 121, 122, 123, 124, 4, 125, 126, 127, 128, + 129 + ], + r3: &[ + 0x00003fffffffffff, 0x000007ff0fffffff, 0x3fdfffff00000000, 0xfffffffbfff80000, + 0xffffffffffffffff, 0xfffeffcfffffffff, 0xf3c5fdfffff99fef, 0x5003ffcfb080799f, + 0xd36dfdfffff987ee, 0x003fffc05e023987, 0xf3edfdfffffbbfee, 0xfe00ffcf00013bbf, + 0xf3edfdfffff99fee, 0x0002ffcfb0c0399f, 0xc3ffc718d63dc7ec, 0x0000ffc000813dc7, + 0xe3fffdfffffddfff, 0x0000ffcf07603ddf, 0xf3effdfffffddfef, 0x0006ffcf40603ddf, + 0xfffffffffffddfef, 0xfc00ffcf80f07ddf, 0x2ffbfffffc7fffec, 0x000cffc0ff5f847f, + 0x07fffffffffffffe, 0x0000000003ff7fff, 0x3bffecaefef02596, 0x00000000f3ff3f5f, + 0xc2a003ff03000001, 0xfffe1ffffffffeff, 0x1ffffffffeffffdf, 0x0000000000000040, + 0xffffffffffff03ff, 0xffffffff3fffffff, 0xf7ffffffffff20bf, 0xffffffff3d7f3dff, + 0x7f3dffffffff3dff, 0xffffffffff7fff3d, 0xffffffffff3dffff, 0x0003fe00e7ffffff, + 0xffffffff0000ffff, 0x3f3fffffffffffff, 0xfffffffffffffffe, 0xffff9fffffffffff, + 0xffffffff07fffffe, 0x01ffc7ffffffffff, 0x001fffff001fdfff, 0x000ddfff000fffff, + 0x000003ff308fffff, 0xffffffff03ff3800, 0x01ffffffffffffff, 0xffff07ffffffffff, + 0x003fffffffffffff, 0x0fff0fff7fffffff, 0x001f3fffffffffc0, 0xffff0fffffffffff, + 0x0000000007ff03ff, 0xffffffff0fffffff, 0x9fffffff7fffffff, 0x3fff008003ff03ff, + 0x0000000000000000, 0x000ff80003ff0fff, 0x000fffffffffffff, 0x00ffffffffffffff, + 0x3fffffffffffe3ff, 0xe7ffffffffff01ff, 0x03fffffffff70000, 0xfbffffffffffffff, + 0xffffffff3f3fffff, 0x3fffffffaaff3f3f, 0x5fdfffffffffffff, 0x1fdc1fff0fcf1fdc, + 0x8000000000000000, 0x8002000000100001, 0x000000001fff0000, 0x0001ffe21fff0000, + 0xf3fffd503f2ffc84, 0xffffffff000043e0, 0x00000000000001ff, 0xffff7fffffffffff, + 0xffffffff7fffffff, 0x000ff81fffffffff, 0xffff20bfffffffff, 0x800080ffffffffff, + 0x7f7f7f7f007fffff, 0xffffffff7f7f7f7f, 0x1f3efffe000000e0, 0xfffffffee67fffff, + 0xf7ffffffffffffff, 0xfffeffffffffffe0, 0x07ffffff00007fff, 0xffff000000000000, + 0x0000ffffffffffff, 0x0000000000001fff, 0x3fffffffffff0000, 0x00000fffffff1fff, + 0xbff0ffffffffffff, 0x0003ffffffffffff, 0xfffffffcff800000, 0x03fffffffffff9ff, + 0xff80000000000000, 0x000000ffffffffff, 0xe8ffffff03ff003f, 0xffff3fffffffffff, + 0x1fffffff000fffff, 0x7fffffff03ff8001, 0x007fffffffffffff, 0xfc7fffff03ff3fff, + 0x007cffff38000007, 0xffff7f7f007e7e7e, 0xffff003ff7ffffff, 0x03ff37ffffffffff, + 0xffff000fffffffff, 0x0ffffffffffff87f, 0x0000000003ffffff, 0x5f7ffdffe0f8007f, + 0xffffffffffffffdb, 0xfffffffffff80000, 0xfffffff03fffffff, 0x3fffffffffffffff, + 0xffffffffffff0000, 0xfffffffffffcffff, 0x03ff0000000000ff, 0x0018ffff0000ffff, + 0xaa8a00000000e000, 0x1fffffffffffffff, 0x87fffffe03ff0000, 0xffffffc007fffffe, + 0x7fffffffffffffff, 0x000000001cfcfcfc + ], + r4: [ + 0, 1, 2, 3, 4, 5, 6, 7, 8, 5, 5, 9, 5, 10, 11, 5, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 12, 13, + 14, 7, 15, 16, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, + 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, + 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, + 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, + 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, + 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, + 5, 5, 5, 5, 5, 5, 5, 17, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, + 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5 + ], + r5: &[ + 0, 1, 2, 3, 4, 5, 4, 6, 4, 4, 7, 8, 9, 10, 11, 12, 2, 2, 13, 14, 15, 16, 4, 4, 2, 2, 2, + 2, 17, 18, 4, 4, 19, 20, 21, 22, 23, 4, 24, 4, 25, 26, 27, 28, 29, 30, 31, 4, 2, 32, 33, + 33, 34, 4, 4, 4, 4, 4, 4, 4, 35, 36, 4, 4, 2, 37, 3, 38, 39, 40, 2, 41, 42, 4, 43, 44, + 45, 46, 4, 4, 2, 47, 2, 48, 4, 4, 49, 50, 2, 51, 52, 53, 54, 4, 4, 4, 3, 4, 55, 56, 4, + 4, 4, 4, 57, 58, 59, 60, 4, 4, 4, 4, 61, 62, 63, 4, 64, 65, 66, 4, 4, 4, 4, 67, 4, 4, 4, + 4, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 68, 4, 2, 69, 2, 2, 2, 70, 4, 4, 4, 4, 4, + 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, + 4, 4, 4, 4, 4, 4, 4, 4, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 69, 4, 4, 4, 4, + 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, + 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, + 4, 2, 2, 2, 2, 2, 2, 2, 2, 2, 71, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, + 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, + 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, + 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, + 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, + 2, 2, 2, 2, 2, 2, 2, 2, 60, 72, 4, 73, 17, 74, 75, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 2, 4, + 4, 2, 76, 77, 78, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, + 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, + 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, + 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 79, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 33, 4, 4, + 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 2, 2, 2, 2, 21, 80, 2, 2, 2, 2, 2, + 81, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, + 4, 4, 4, 4, 4, 4, 4, 4, 2, 82, 83, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, + 84, 85, 4, 4, 86, 4, 4, 4, 4, 4, 4, 2, 87, 88, 89, 90, 91, 2, 2, 2, 2, 92, 93, 94, 95, + 96, 97, 4, 4, 4, 4, 4, 4, 4, 4, 98, 99, 100, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, + 4, 4, 4, 4, 4, 4, 4, 101, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, + 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 2, 2, 2, 102, 2, 103, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, + 4, 4, 4, 4, 4, 4, 104, 105, 106, 4, 4, 4, 4, 4, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, + 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 107, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, + 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, + 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 5, 2, 2, 2, 11, 2, 2, 2, + 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, + 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, + 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 108, + 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, + 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 109, 4, 4, 4, 4, 4, + 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, + 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 2, 2, 2, 2, 2, 2, 2, 2, 110, 4, 4, 4, 4, 4, 4, + 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 2, 2, 2, 111, 4, 4, 4, 4, + 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, + 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4 + ], + r6: &[ + 0xb7ffff7fffffefff, 0x000000003fff3fff, 0xffffffffffffffff, 0x07ffffffffffffff, + 0x0000000000000000, 0x001fffffffffffff, 0x2000000000000000, 0xffffffff1fffffff, + 0x000000010001ffff, 0xffffe000ffffffff, 0x07ffffffffff07ff, 0xffffffff3fffffff, + 0x00000000003eff0f, 0xffff03ff3fffffff, 0x0fffffffff0fffff, 0xffff00ffffffffff, + 0x0000000fffffffff, 0x007fffffffffffff, 0x000000ff003fffff, 0x91bffffffffffd3f, + 0x007fffff003fffff, 0x000000007fffffff, 0x0037ffff00000000, 0x03ffffff003fffff, + 0xc0ffffffffffffff, 0x873ffffffeeff06f, 0x1fffffff00000000, 0x000000001fffffff, + 0x0000007ffffffeff, 0x003fffffffffffff, 0x0007ffff003fffff, 0x000000000003ffff, + 0x00000000000001ff, 0x0007ffffffffffff, 0x03ff00ffffffffff, 0xffff00801fffffff, + 0x000000000001ffff, 0x8000ffc00000007f, 0x03ff01ffffff0000, 0xffdfffffffffffff, + 0x004fffffffff0070, 0x0000000017ff1e1f, 0x40fffffffffbffff, 0xffff01ffbfffbd7f, + 0x03ff07ffffffffff, 0xfbedfdfffff99fef, 0x001f1fcfe081399f, 0x0000000043ff07ff, + 0x0000000003ff00bf, 0xff3fffffffffffff, 0x000000003f000001, 0x0000000003ff0011, + 0x00ffffffffffffff, 0x00000000000003ff, 0x03ff0fffe7ffffff, 0xffffffff00000000, + 0x800003ffffffffff, 0x7fffffffffffffff, 0xffffffffffff0080, 0x0000000023ffffcf, + 0x01ffffffffffffff, 0xff7ffffffffffdff, 0xfffc000003ff0001, 0x007ffefffffcffff, + 0xb47ffffffffffb7f, 0xfffffdbf03ff00ff, 0x000003ff01fb7fff, 0x007fffff00000000, + 0x0000000003ffffff, 0x00007fffffffffff, 0x000000000000000f, 0x000000000000007f, + 0x000003ff7fffffff, 0x001f3fffffff0000, 0xe0fffff803ff000f, 0x000000000000ffff, + 0x7fffffffffff001f, 0x00000000ffff8000, 0x0000000300000000, 0x0003ffffffffffff, + 0xffff000000000000, 0x0fffffffffffffff, 0x1fff07ffffffffff, 0x0000000063ff01ff, + 0xf807e3e000000000, 0x00003c0000000fe7, 0x000000000000001c, 0xffffffffffdfffff, + 0xebffde64dfffffff, 0xffffffffffffffef, 0x7bffffffdfdfe7bf, 0xfffffffffffdfc5f, + 0xffffff3fffffffff, 0xf7fffffff7fffffd, 0xffdfffffffdfffff, 0xffff7fffffff7fff, + 0xfffffdfffffffdff, 0xffffffffffffcff7, 0xf87fffffffffffff, 0x00201fffffffffff, + 0x0000fffef8000010, 0x000007dbf9ffff7f, 0x00000000007f001f, 0x0000000003ff07ff, + 0x0af7fe96ffffffef, 0x5ef7f796aa96ea84, 0x0ffffbee0ffffbff, 0x00000000007fffff, + 0xffff0003ffffffff, 0x00000001ffffffff, 0x000000003fffffff, 0x0000ffffffffffff + ], + }; + + pub fn XID_Continue(c: char) -> bool { + XID_Continue_table.lookup(c) + } + + pub const XID_Start_table: &super::BoolTrie = &super::BoolTrie { + r1: [ + 0x0000000000000000, 0x07fffffe07fffffe, 0x0420040000000000, 0xff7fffffff7fffff, + 0xffffffffffffffff, 0xffffffffffffffff, 0xffffffffffffffff, 0xffffffffffffffff, + 0xffffffffffffffff, 0xffffffffffffffff, 0xffffffffffffffff, 0x0000501f0003ffc3, + 0x0000000000000000, 0xb8df000000000000, 0xfffffffbffffd740, 0xffbfffffffffffff, + 0xffffffffffffffff, 0xffffffffffffffff, 0xfffffffffffffc03, 0xffffffffffffffff, + 0xfffeffffffffffff, 0xffffffff027fffff, 0x00000000000001ff, 0x000787ffffff0000, + 0xffffffff00000000, 0xfffec000000007ff, 0xffffffffffffffff, 0x9c00c060002fffff, + 0x0000fffffffd0000, 0xffffffffffffe000, 0x0002003fffffffff, 0x043007fffffffc00 + ], + r2: [ + 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, + 24, 23, 25, 26, 27, 28, 29, 3, 30, 31, 32, 33, 34, 34, 34, 34, 34, 35, 36, 37, 38, 39, + 40, 41, 42, 34, 34, 34, 34, 34, 34, 34, 34, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, + 54, 55, 56, 57, 58, 59, 60, 3, 61, 62, 63, 64, 65, 66, 67, 68, 34, 34, 34, 3, 34, 34, + 34, 34, 69, 70, 71, 72, 3, 73, 74, 3, 75, 76, 77, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, + 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 78, + 79, 34, 80, 81, 82, 83, 84, 3, 3, 3, 3, 3, 3, 3, 3, 85, 42, 86, 87, 88, 34, 89, 90, 3, + 3, 3, 3, 3, 3, 3, 3, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, + 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, + 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, + 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, + 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 53, 3, 34, + 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, + 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, + 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, + 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, + 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, + 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, + 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, + 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, + 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, + 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, + 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, + 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, + 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, + 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, + 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 91, 34, 34, 34, + 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 92, 93, 34, 34, 34, 34, 94, + 95, 96, 91, 97, 34, 98, 99, 100, 48, 101, 102, 103, 104, 105, 106, 107, 108, 109, 110, + 111, 112, 34, 113, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, + 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, + 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, + 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, + 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, + 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, + 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, + 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, + 34, 34, 34, 114, 115, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, + 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, + 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, + 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, + 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 34, 34, 34, 34, 34, + 116, 34, 117, 118, 119, 120, 121, 34, 122, 34, 34, 123, 124, 125, 126, 3, 127, 34, 128, + 129, 130, 131, 132 + ], + r3: &[ + 0x00000110043fffff, 0x000007ff01ffffff, 0x3fdfffff00000000, 0x0000000000000000, + 0x23fffffffffffff0, 0xfffe0003ff010000, 0x23c5fdfffff99fe1, 0x10030003b0004000, + 0x036dfdfffff987e0, 0x001c00005e000000, 0x23edfdfffffbbfe0, 0x0200000300010000, + 0x23edfdfffff99fe0, 0x00020003b0000000, 0x03ffc718d63dc7e8, 0x0000000000010000, + 0x23fffdfffffddfe0, 0x0000000307000000, 0x23effdfffffddfe1, 0x0006000340000000, + 0x27fffffffffddfe0, 0xfc00000380704000, 0x2ffbfffffc7fffe0, 0x000000000000007f, + 0x0005fffffffffffe, 0x2005ecaefef02596, 0x00000000f000005f, 0x0000000000000001, + 0x00001ffffffffeff, 0x0000000000001f00, 0x800007ffffffffff, 0xffe1c0623c3f0000, + 0xffffffff00004003, 0xf7ffffffffff20bf, 0xffffffffffffffff, 0xffffffff3d7f3dff, + 0x7f3dffffffff3dff, 0xffffffffff7fff3d, 0xffffffffff3dffff, 0x0000000007ffffff, + 0xffffffff0000ffff, 0x3f3fffffffffffff, 0xfffffffffffffffe, 0xffff9fffffffffff, + 0xffffffff07fffffe, 0x01ffc7ffffffffff, 0x0003ffff0003dfff, 0x0001dfff0003ffff, + 0x000fffffffffffff, 0x0000000010800000, 0xffffffff00000000, 0x01ffffffffffffff, + 0xffff05ffffffffff, 0x003fffffffffffff, 0x000000007fffffff, 0x001f3fffffff0000, + 0xffff0fffffffffff, 0x00000000000003ff, 0xffffffff007fffff, 0x00000000001fffff, + 0x0000008000000000, 0x000fffffffffffe0, 0x0000000000000fe0, 0xfc00c001fffffff8, + 0x0000003fffffffff, 0x0000000fffffffff, 0x3ffffffffc00e000, 0xe7ffffffffff01ff, + 0x0063de0000000000, 0xffffffff3f3fffff, 0x3fffffffaaff3f3f, 0x5fdfffffffffffff, + 0x1fdc1fff0fcf1fdc, 0x8002000000000000, 0x000000001fff0000, 0xf3fffd503f2ffc84, + 0xffffffff000043e0, 0x00000000000001ff, 0xffff7fffffffffff, 0xffffffff7fffffff, + 0x000c781fffffffff, 0xffff20bfffffffff, 0x000080ffffffffff, 0x7f7f7f7f007fffff, + 0x000000007f7f7f7f, 0x1f3e03fe000000e0, 0xfffffffee07fffff, 0xf7ffffffffffffff, + 0xfffeffffffffffe0, 0x07ffffff00007fff, 0xffff000000000000, 0x0000ffffffffffff, + 0x0000000000001fff, 0x3fffffffffff0000, 0x00000c00ffff1fff, 0x80007fffffffffff, + 0xffffffff3fffffff, 0xfffffffcff800000, 0x03fffffffffff9ff, 0xff80000000000000, + 0x00000007fffff7bb, 0x000ffffffffffffc, 0x68fc000000000000, 0xffff003ffffffc00, + 0x1fffffff0000007f, 0x0007fffffffffff0, 0x7c00ffdf00008000, 0x000001ffffffffff, + 0xc47fffff00000ff7, 0x3e62ffffffffffff, 0x001c07ff38000005, 0xffff7f7f007e7e7e, + 0xffff003ff7ffffff, 0x00000007ffffffff, 0xffff000fffffffff, 0x0ffffffffffff87f, + 0xffff3fffffffffff, 0x0000000003ffffff, 0x5f7ffdffa0f8007f, 0xffffffffffffffdb, + 0x0003ffffffffffff, 0xfffffffffff80000, 0xfffffff03fffffff, 0x3fffffffffffffff, + 0xffffffffffff0000, 0xfffffffffffcffff, 0x03ff0000000000ff, 0xaa8a000000000000, + 0x1fffffffffffffff, 0x07fffffe00000000, 0xffffffc007fffffe, 0x7fffffff3fffffff, + 0x000000001cfcfcfc + ], + r4: [ + 0, 1, 2, 3, 4, 5, 6, 7, 8, 5, 5, 9, 5, 10, 11, 5, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 12, 13, + 14, 7, 15, 16, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, + 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, + 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, + 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, + 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, + 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, + 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, + 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5 + ], + r5: &[ + 0, 1, 2, 3, 4, 5, 4, 4, 4, 4, 6, 7, 8, 9, 10, 11, 2, 2, 12, 13, 14, 15, 4, 4, 2, 2, 2, + 2, 16, 17, 4, 4, 18, 19, 20, 21, 22, 4, 23, 4, 24, 25, 26, 27, 28, 29, 30, 4, 2, 31, 32, + 32, 15, 4, 4, 4, 4, 4, 4, 4, 33, 34, 4, 4, 35, 4, 36, 37, 38, 39, 40, 41, 42, 4, 43, 20, + 44, 45, 4, 4, 5, 46, 47, 48, 4, 4, 49, 50, 47, 51, 52, 4, 53, 4, 4, 4, 54, 4, 55, 56, 4, + 4, 4, 4, 57, 58, 59, 60, 4, 4, 4, 4, 61, 62, 63, 4, 64, 65, 66, 4, 4, 4, 4, 67, 4, 4, 4, + 4, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 68, 4, 2, 49, 2, 2, 2, 69, 4, 4, 4, 4, 4, + 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, + 4, 4, 4, 4, 4, 4, 4, 4, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 49, 4, 4, 4, 4, + 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, + 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, + 4, 2, 2, 2, 2, 2, 2, 2, 2, 2, 70, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, + 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, + 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, + 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, + 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, + 2, 2, 2, 2, 2, 2, 2, 2, 60, 20, 4, 71, 47, 72, 63, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 2, 4, + 4, 2, 73, 74, 75, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, + 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, + 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, + 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 76, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 32, 4, 4, + 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 2, 2, 2, 2, 20, 77, 2, 2, 2, 2, 2, + 78, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, + 4, 4, 4, 4, 4, 4, 4, 4, 2, 79, 80, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, + 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 2, 81, 82, 83, 84, 85, 2, 2, 2, 2, 86, 87, 88, 89, 90, + 91, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, + 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, + 4, 4, 4, 4, 4, 4, 4, 2, 2, 2, 92, 2, 69, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, + 4, 4, 93, 94, 95, 4, 4, 4, 4, 4, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, + 2, 2, 2, 2, 2, 2, 2, 2, 2, 96, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, + 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, + 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 5, 2, 2, 2, 10, 2, 2, 2, 2, 2, 2, 2, 2, + 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, + 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, + 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 97, 2, 2, 2, 2, 2, + 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, + 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 98, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, + 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, + 4, 4, 4, 4, 4, 4, 4, 4, 4, 2, 2, 2, 2, 2, 2, 2, 2, 99, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, + 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4 + ], + r6: &[ + 0xb7ffff7fffffefff, 0x000000003fff3fff, 0xffffffffffffffff, 0x07ffffffffffffff, + 0x0000000000000000, 0x001fffffffffffff, 0xffffffff1fffffff, 0x000000000001ffff, + 0xffffe000ffffffff, 0x003fffffffff07ff, 0xffffffff3fffffff, 0x00000000003eff0f, + 0xffff00003fffffff, 0x0fffffffff0fffff, 0xffff00ffffffffff, 0x0000000fffffffff, + 0x007fffffffffffff, 0x000000ff003fffff, 0x91bffffffffffd3f, 0x007fffff003fffff, + 0x000000007fffffff, 0x0037ffff00000000, 0x03ffffff003fffff, 0xc0ffffffffffffff, + 0x003ffffffeef0001, 0x1fffffff00000000, 0x000000001fffffff, 0x0000001ffffffeff, + 0x003fffffffffffff, 0x0007ffff003fffff, 0x000000000003ffff, 0x00000000000001ff, + 0x0007ffffffffffff, 0xffff00801fffffff, 0x000000000000003f, 0x00fffffffffffff8, + 0x0000fffffffffff8, 0x000001ffffff0000, 0x0000007ffffffff8, 0x0047ffffffff0010, + 0x0007fffffffffff8, 0x000000001400001e, 0x00000ffffffbffff, 0xffff01ffbfffbd7f, + 0x23edfdfffff99fe0, 0x00000003e0010000, 0x0000000000000780, 0x0000ffffffffffff, + 0x00000000000000b0, 0x00007fffffffffff, 0x000000000f000000, 0x0000000000000010, + 0x000007ffffffffff, 0x0000000007ffffff, 0x00000fffffffffff, 0xffffffff00000000, + 0x80000000ffffffff, 0x0407fffffffff801, 0xfffffffff0010000, 0x00000000200003cf, + 0x01ffffffffffffff, 0x00007ffffffffdff, 0xfffc000000000001, 0x000000000000ffff, + 0x0001fffffffffb7f, 0xfffffdbf00000040, 0x00000000010003ff, 0x0007ffff00000000, + 0x0000000003ffffff, 0x000000000000000f, 0x000000000000007f, 0x00003fffffff0000, + 0xe0fffff80000000f, 0x000000000001001f, 0x00000000fff80000, 0x0000000300000000, + 0x0003ffffffffffff, 0xffff000000000000, 0x0fffffffffffffff, 0x1fff07ffffffffff, + 0x0000000003ff01ff, 0xffffffffffdfffff, 0xebffde64dfffffff, 0xffffffffffffffef, + 0x7bffffffdfdfe7bf, 0xfffffffffffdfc5f, 0xffffff3fffffffff, 0xf7fffffff7fffffd, + 0xffdfffffffdfffff, 0xffff7fffffff7fff, 0xfffffdfffffffdff, 0x0000000000000ff7, + 0x000000000000001f, 0x0af7fe96ffffffef, 0x5ef7f796aa96ea84, 0x0ffffbee0ffffbff, + 0x00000000007fffff, 0xffff0003ffffffff, 0x00000001ffffffff, 0x000000003fffffff + ], + }; + + pub fn XID_Start(c: char) -> bool { + XID_Start_table.lookup(c) + } + +} + +pub mod property { + pub const Pattern_White_Space_table: &super::SmallBoolTrie = &super::SmallBoolTrie { + r1: &[ + 0, 1, 2, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, + 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, + 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, + 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, + 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 3 + ], + r2: &[ + 0x0000000100003e00, 0x0000000000000000, 0x0000000000000020, 0x000003000000c000 + ], + }; + + pub fn Pattern_White_Space(c: char) -> bool { + Pattern_White_Space_table.lookup(c) + } + + pub const White_Space_table: &super::SmallBoolTrie = &super::SmallBoolTrie { + r1: &[ + 0, 1, 2, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, + 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, + 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, + 1, 1, 1, 3, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, + 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 4, 5, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, + 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, + 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 3 + ], + r2: &[ + 0x0000000100003e00, 0x0000000000000000, 0x0000000100000020, 0x0000000000000001, + 0x00008300000007ff, 0x0000000080000000 + ], + }; + + pub fn White_Space(c: char) -> bool { + White_Space_table.lookup(c) + } + +} + +pub mod conversions { + pub fn to_lower(c: char) -> [char; 3] { + match bsearch_case_table(c, to_lowercase_table) { + None => [c, '\0', '\0'], + Some(index) => to_lowercase_table[index].1, + } + } + + pub fn to_upper(c: char) -> [char; 3] { + match bsearch_case_table(c, to_uppercase_table) { + None => [c, '\0', '\0'], + Some(index) => to_uppercase_table[index].1, + } + } + + fn bsearch_case_table(c: char, table: &[(char, [char; 3])]) -> Option<usize> { + table.binary_search_by(|&(key, _)| key.cmp(&c)).ok() + } + + const to_lowercase_table: &[(char, [char; 3])] = &[ + ('\u{41}', ['\u{61}', '\0', '\0']), ('\u{42}', ['\u{62}', '\0', '\0']), ('\u{43}', + ['\u{63}', '\0', '\0']), ('\u{44}', ['\u{64}', '\0', '\0']), ('\u{45}', ['\u{65}', '\0', + '\0']), ('\u{46}', ['\u{66}', '\0', '\0']), ('\u{47}', ['\u{67}', '\0', '\0']), ('\u{48}', + ['\u{68}', '\0', '\0']), ('\u{49}', ['\u{69}', '\0', '\0']), ('\u{4a}', ['\u{6a}', '\0', + '\0']), ('\u{4b}', ['\u{6b}', '\0', '\0']), ('\u{4c}', ['\u{6c}', '\0', '\0']), ('\u{4d}', + ['\u{6d}', '\0', '\0']), ('\u{4e}', ['\u{6e}', '\0', '\0']), ('\u{4f}', ['\u{6f}', '\0', + '\0']), ('\u{50}', ['\u{70}', '\0', '\0']), ('\u{51}', ['\u{71}', '\0', '\0']), ('\u{52}', + ['\u{72}', '\0', '\0']), ('\u{53}', ['\u{73}', '\0', '\0']), ('\u{54}', ['\u{74}', '\0', + '\0']), ('\u{55}', ['\u{75}', '\0', '\0']), ('\u{56}', ['\u{76}', '\0', '\0']), ('\u{57}', + ['\u{77}', '\0', '\0']), ('\u{58}', ['\u{78}', '\0', '\0']), ('\u{59}', ['\u{79}', '\0', + '\0']), ('\u{5a}', ['\u{7a}', '\0', '\0']), ('\u{c0}', ['\u{e0}', '\0', '\0']), ('\u{c1}', + ['\u{e1}', '\0', '\0']), ('\u{c2}', ['\u{e2}', '\0', '\0']), ('\u{c3}', ['\u{e3}', '\0', + '\0']), ('\u{c4}', ['\u{e4}', '\0', '\0']), ('\u{c5}', ['\u{e5}', '\0', '\0']), ('\u{c6}', + ['\u{e6}', '\0', '\0']), ('\u{c7}', ['\u{e7}', '\0', '\0']), ('\u{c8}', ['\u{e8}', '\0', + '\0']), ('\u{c9}', ['\u{e9}', '\0', '\0']), ('\u{ca}', ['\u{ea}', '\0', '\0']), ('\u{cb}', + ['\u{eb}', '\0', '\0']), ('\u{cc}', ['\u{ec}', '\0', '\0']), ('\u{cd}', ['\u{ed}', '\0', + '\0']), ('\u{ce}', ['\u{ee}', '\0', '\0']), ('\u{cf}', ['\u{ef}', '\0', '\0']), ('\u{d0}', + ['\u{f0}', '\0', '\0']), ('\u{d1}', ['\u{f1}', '\0', '\0']), ('\u{d2}', ['\u{f2}', '\0', + '\0']), ('\u{d3}', ['\u{f3}', '\0', '\0']), ('\u{d4}', ['\u{f4}', '\0', '\0']), ('\u{d5}', + ['\u{f5}', '\0', '\0']), ('\u{d6}', ['\u{f6}', '\0', '\0']), ('\u{d8}', ['\u{f8}', '\0', + '\0']), ('\u{d9}', ['\u{f9}', '\0', '\0']), ('\u{da}', ['\u{fa}', '\0', '\0']), ('\u{db}', + ['\u{fb}', '\0', '\0']), ('\u{dc}', ['\u{fc}', '\0', '\0']), ('\u{dd}', ['\u{fd}', '\0', + '\0']), ('\u{de}', ['\u{fe}', '\0', '\0']), ('\u{100}', ['\u{101}', '\0', '\0']), + ('\u{102}', ['\u{103}', '\0', '\0']), ('\u{104}', ['\u{105}', '\0', '\0']), ('\u{106}', + ['\u{107}', '\0', '\0']), ('\u{108}', ['\u{109}', '\0', '\0']), ('\u{10a}', ['\u{10b}', + '\0', '\0']), ('\u{10c}', ['\u{10d}', '\0', '\0']), ('\u{10e}', ['\u{10f}', '\0', '\0']), + ('\u{110}', ['\u{111}', '\0', '\0']), ('\u{112}', ['\u{113}', '\0', '\0']), ('\u{114}', + ['\u{115}', '\0', '\0']), ('\u{116}', ['\u{117}', '\0', '\0']), ('\u{118}', ['\u{119}', + '\0', '\0']), ('\u{11a}', ['\u{11b}', '\0', '\0']), ('\u{11c}', ['\u{11d}', '\0', '\0']), + ('\u{11e}', ['\u{11f}', '\0', '\0']), ('\u{120}', ['\u{121}', '\0', '\0']), ('\u{122}', + ['\u{123}', '\0', '\0']), ('\u{124}', ['\u{125}', '\0', '\0']), ('\u{126}', ['\u{127}', + '\0', '\0']), ('\u{128}', ['\u{129}', '\0', '\0']), ('\u{12a}', ['\u{12b}', '\0', '\0']), + ('\u{12c}', ['\u{12d}', '\0', '\0']), ('\u{12e}', ['\u{12f}', '\0', '\0']), ('\u{130}', + ['\u{69}', '\u{307}', '\0']), ('\u{132}', ['\u{133}', '\0', '\0']), ('\u{134}', ['\u{135}', + '\0', '\0']), ('\u{136}', ['\u{137}', '\0', '\0']), ('\u{139}', ['\u{13a}', '\0', '\0']), + ('\u{13b}', ['\u{13c}', '\0', '\0']), ('\u{13d}', ['\u{13e}', '\0', '\0']), ('\u{13f}', + ['\u{140}', '\0', '\0']), ('\u{141}', ['\u{142}', '\0', '\0']), ('\u{143}', ['\u{144}', + '\0', '\0']), ('\u{145}', ['\u{146}', '\0', '\0']), ('\u{147}', ['\u{148}', '\0', '\0']), + ('\u{14a}', ['\u{14b}', '\0', '\0']), ('\u{14c}', ['\u{14d}', '\0', '\0']), ('\u{14e}', + ['\u{14f}', '\0', '\0']), ('\u{150}', ['\u{151}', '\0', '\0']), ('\u{152}', ['\u{153}', + '\0', '\0']), ('\u{154}', ['\u{155}', '\0', '\0']), ('\u{156}', ['\u{157}', '\0', '\0']), + ('\u{158}', ['\u{159}', '\0', '\0']), ('\u{15a}', ['\u{15b}', '\0', '\0']), ('\u{15c}', + ['\u{15d}', '\0', '\0']), ('\u{15e}', ['\u{15f}', '\0', '\0']), ('\u{160}', ['\u{161}', + '\0', '\0']), ('\u{162}', ['\u{163}', '\0', '\0']), ('\u{164}', ['\u{165}', '\0', '\0']), + ('\u{166}', ['\u{167}', '\0', '\0']), ('\u{168}', ['\u{169}', '\0', '\0']), ('\u{16a}', + ['\u{16b}', '\0', '\0']), ('\u{16c}', ['\u{16d}', '\0', '\0']), ('\u{16e}', ['\u{16f}', + '\0', '\0']), ('\u{170}', ['\u{171}', '\0', '\0']), ('\u{172}', ['\u{173}', '\0', '\0']), + ('\u{174}', ['\u{175}', '\0', '\0']), ('\u{176}', ['\u{177}', '\0', '\0']), ('\u{178}', + ['\u{ff}', '\0', '\0']), ('\u{179}', ['\u{17a}', '\0', '\0']), ('\u{17b}', ['\u{17c}', '\0', + '\0']), ('\u{17d}', ['\u{17e}', '\0', '\0']), ('\u{181}', ['\u{253}', '\0', '\0']), + ('\u{182}', ['\u{183}', '\0', '\0']), ('\u{184}', ['\u{185}', '\0', '\0']), ('\u{186}', + ['\u{254}', '\0', '\0']), ('\u{187}', ['\u{188}', '\0', '\0']), ('\u{189}', ['\u{256}', + '\0', '\0']), ('\u{18a}', ['\u{257}', '\0', '\0']), ('\u{18b}', ['\u{18c}', '\0', '\0']), + ('\u{18e}', ['\u{1dd}', '\0', '\0']), ('\u{18f}', ['\u{259}', '\0', '\0']), ('\u{190}', + ['\u{25b}', '\0', '\0']), ('\u{191}', ['\u{192}', '\0', '\0']), ('\u{193}', ['\u{260}', + '\0', '\0']), ('\u{194}', ['\u{263}', '\0', '\0']), ('\u{196}', ['\u{269}', '\0', '\0']), + ('\u{197}', ['\u{268}', '\0', '\0']), ('\u{198}', ['\u{199}', '\0', '\0']), ('\u{19c}', + ['\u{26f}', '\0', '\0']), ('\u{19d}', ['\u{272}', '\0', '\0']), ('\u{19f}', ['\u{275}', + '\0', '\0']), ('\u{1a0}', ['\u{1a1}', '\0', '\0']), ('\u{1a2}', ['\u{1a3}', '\0', '\0']), + ('\u{1a4}', ['\u{1a5}', '\0', '\0']), ('\u{1a6}', ['\u{280}', '\0', '\0']), ('\u{1a7}', + ['\u{1a8}', '\0', '\0']), ('\u{1a9}', ['\u{283}', '\0', '\0']), ('\u{1ac}', ['\u{1ad}', + '\0', '\0']), ('\u{1ae}', ['\u{288}', '\0', '\0']), ('\u{1af}', ['\u{1b0}', '\0', '\0']), + ('\u{1b1}', ['\u{28a}', '\0', '\0']), ('\u{1b2}', ['\u{28b}', '\0', '\0']), ('\u{1b3}', + ['\u{1b4}', '\0', '\0']), ('\u{1b5}', ['\u{1b6}', '\0', '\0']), ('\u{1b7}', ['\u{292}', + '\0', '\0']), ('\u{1b8}', ['\u{1b9}', '\0', '\0']), ('\u{1bc}', ['\u{1bd}', '\0', '\0']), + ('\u{1c4}', ['\u{1c6}', '\0', '\0']), ('\u{1c5}', ['\u{1c6}', '\0', '\0']), ('\u{1c7}', + ['\u{1c9}', '\0', '\0']), ('\u{1c8}', ['\u{1c9}', '\0', '\0']), ('\u{1ca}', ['\u{1cc}', + '\0', '\0']), ('\u{1cb}', ['\u{1cc}', '\0', '\0']), ('\u{1cd}', ['\u{1ce}', '\0', '\0']), + ('\u{1cf}', ['\u{1d0}', '\0', '\0']), ('\u{1d1}', ['\u{1d2}', '\0', '\0']), ('\u{1d3}', + ['\u{1d4}', '\0', '\0']), ('\u{1d5}', ['\u{1d6}', '\0', '\0']), ('\u{1d7}', ['\u{1d8}', + '\0', '\0']), ('\u{1d9}', ['\u{1da}', '\0', '\0']), ('\u{1db}', ['\u{1dc}', '\0', '\0']), + ('\u{1de}', ['\u{1df}', '\0', '\0']), ('\u{1e0}', ['\u{1e1}', '\0', '\0']), ('\u{1e2}', + ['\u{1e3}', '\0', '\0']), ('\u{1e4}', ['\u{1e5}', '\0', '\0']), ('\u{1e6}', ['\u{1e7}', + '\0', '\0']), ('\u{1e8}', ['\u{1e9}', '\0', '\0']), ('\u{1ea}', ['\u{1eb}', '\0', '\0']), + ('\u{1ec}', ['\u{1ed}', '\0', '\0']), ('\u{1ee}', ['\u{1ef}', '\0', '\0']), ('\u{1f1}', + ['\u{1f3}', '\0', '\0']), ('\u{1f2}', ['\u{1f3}', '\0', '\0']), ('\u{1f4}', ['\u{1f5}', + '\0', '\0']), ('\u{1f6}', ['\u{195}', '\0', '\0']), ('\u{1f7}', ['\u{1bf}', '\0', '\0']), + ('\u{1f8}', ['\u{1f9}', '\0', '\0']), ('\u{1fa}', ['\u{1fb}', '\0', '\0']), ('\u{1fc}', + ['\u{1fd}', '\0', '\0']), ('\u{1fe}', ['\u{1ff}', '\0', '\0']), ('\u{200}', ['\u{201}', + '\0', '\0']), ('\u{202}', ['\u{203}', '\0', '\0']), ('\u{204}', ['\u{205}', '\0', '\0']), + ('\u{206}', ['\u{207}', '\0', '\0']), ('\u{208}', ['\u{209}', '\0', '\0']), ('\u{20a}', + ['\u{20b}', '\0', '\0']), ('\u{20c}', ['\u{20d}', '\0', '\0']), ('\u{20e}', ['\u{20f}', + '\0', '\0']), ('\u{210}', ['\u{211}', '\0', '\0']), ('\u{212}', ['\u{213}', '\0', '\0']), + ('\u{214}', ['\u{215}', '\0', '\0']), ('\u{216}', ['\u{217}', '\0', '\0']), ('\u{218}', + ['\u{219}', '\0', '\0']), ('\u{21a}', ['\u{21b}', '\0', '\0']), ('\u{21c}', ['\u{21d}', + '\0', '\0']), ('\u{21e}', ['\u{21f}', '\0', '\0']), ('\u{220}', ['\u{19e}', '\0', '\0']), + ('\u{222}', ['\u{223}', '\0', '\0']), ('\u{224}', ['\u{225}', '\0', '\0']), ('\u{226}', + ['\u{227}', '\0', '\0']), ('\u{228}', ['\u{229}', '\0', '\0']), ('\u{22a}', ['\u{22b}', + '\0', '\0']), ('\u{22c}', ['\u{22d}', '\0', '\0']), ('\u{22e}', ['\u{22f}', '\0', '\0']), + ('\u{230}', ['\u{231}', '\0', '\0']), ('\u{232}', ['\u{233}', '\0', '\0']), ('\u{23a}', + ['\u{2c65}', '\0', '\0']), ('\u{23b}', ['\u{23c}', '\0', '\0']), ('\u{23d}', ['\u{19a}', + '\0', '\0']), ('\u{23e}', ['\u{2c66}', '\0', '\0']), ('\u{241}', ['\u{242}', '\0', '\0']), + ('\u{243}', ['\u{180}', '\0', '\0']), ('\u{244}', ['\u{289}', '\0', '\0']), ('\u{245}', + ['\u{28c}', '\0', '\0']), ('\u{246}', ['\u{247}', '\0', '\0']), ('\u{248}', ['\u{249}', + '\0', '\0']), ('\u{24a}', ['\u{24b}', '\0', '\0']), ('\u{24c}', ['\u{24d}', '\0', '\0']), + ('\u{24e}', ['\u{24f}', '\0', '\0']), ('\u{370}', ['\u{371}', '\0', '\0']), ('\u{372}', + ['\u{373}', '\0', '\0']), ('\u{376}', ['\u{377}', '\0', '\0']), ('\u{37f}', ['\u{3f3}', + '\0', '\0']), ('\u{386}', ['\u{3ac}', '\0', '\0']), ('\u{388}', ['\u{3ad}', '\0', '\0']), + ('\u{389}', ['\u{3ae}', '\0', '\0']), ('\u{38a}', ['\u{3af}', '\0', '\0']), ('\u{38c}', + ['\u{3cc}', '\0', '\0']), ('\u{38e}', ['\u{3cd}', '\0', '\0']), ('\u{38f}', ['\u{3ce}', + '\0', '\0']), ('\u{391}', ['\u{3b1}', '\0', '\0']), ('\u{392}', ['\u{3b2}', '\0', '\0']), + ('\u{393}', ['\u{3b3}', '\0', '\0']), ('\u{394}', ['\u{3b4}', '\0', '\0']), ('\u{395}', + ['\u{3b5}', '\0', '\0']), ('\u{396}', ['\u{3b6}', '\0', '\0']), ('\u{397}', ['\u{3b7}', + '\0', '\0']), ('\u{398}', ['\u{3b8}', '\0', '\0']), ('\u{399}', ['\u{3b9}', '\0', '\0']), + ('\u{39a}', ['\u{3ba}', '\0', '\0']), ('\u{39b}', ['\u{3bb}', '\0', '\0']), ('\u{39c}', + ['\u{3bc}', '\0', '\0']), ('\u{39d}', ['\u{3bd}', '\0', '\0']), ('\u{39e}', ['\u{3be}', + '\0', '\0']), ('\u{39f}', ['\u{3bf}', '\0', '\0']), ('\u{3a0}', ['\u{3c0}', '\0', '\0']), + ('\u{3a1}', ['\u{3c1}', '\0', '\0']), ('\u{3a3}', ['\u{3c3}', '\0', '\0']), ('\u{3a4}', + ['\u{3c4}', '\0', '\0']), ('\u{3a5}', ['\u{3c5}', '\0', '\0']), ('\u{3a6}', ['\u{3c6}', + '\0', '\0']), ('\u{3a7}', ['\u{3c7}', '\0', '\0']), ('\u{3a8}', ['\u{3c8}', '\0', '\0']), + ('\u{3a9}', ['\u{3c9}', '\0', '\0']), ('\u{3aa}', ['\u{3ca}', '\0', '\0']), ('\u{3ab}', + ['\u{3cb}', '\0', '\0']), ('\u{3cf}', ['\u{3d7}', '\0', '\0']), ('\u{3d8}', ['\u{3d9}', + '\0', '\0']), ('\u{3da}', ['\u{3db}', '\0', '\0']), ('\u{3dc}', ['\u{3dd}', '\0', '\0']), + ('\u{3de}', ['\u{3df}', '\0', '\0']), ('\u{3e0}', ['\u{3e1}', '\0', '\0']), ('\u{3e2}', + ['\u{3e3}', '\0', '\0']), ('\u{3e4}', ['\u{3e5}', '\0', '\0']), ('\u{3e6}', ['\u{3e7}', + '\0', '\0']), ('\u{3e8}', ['\u{3e9}', '\0', '\0']), ('\u{3ea}', ['\u{3eb}', '\0', '\0']), + ('\u{3ec}', ['\u{3ed}', '\0', '\0']), ('\u{3ee}', ['\u{3ef}', '\0', '\0']), ('\u{3f4}', + ['\u{3b8}', '\0', '\0']), ('\u{3f7}', ['\u{3f8}', '\0', '\0']), ('\u{3f9}', ['\u{3f2}', + '\0', '\0']), ('\u{3fa}', ['\u{3fb}', '\0', '\0']), ('\u{3fd}', ['\u{37b}', '\0', '\0']), + ('\u{3fe}', ['\u{37c}', '\0', '\0']), ('\u{3ff}', ['\u{37d}', '\0', '\0']), ('\u{400}', + ['\u{450}', '\0', '\0']), ('\u{401}', ['\u{451}', '\0', '\0']), ('\u{402}', ['\u{452}', + '\0', '\0']), ('\u{403}', ['\u{453}', '\0', '\0']), ('\u{404}', ['\u{454}', '\0', '\0']), + ('\u{405}', ['\u{455}', '\0', '\0']), ('\u{406}', ['\u{456}', '\0', '\0']), ('\u{407}', + ['\u{457}', '\0', '\0']), ('\u{408}', ['\u{458}', '\0', '\0']), ('\u{409}', ['\u{459}', + '\0', '\0']), ('\u{40a}', ['\u{45a}', '\0', '\0']), ('\u{40b}', ['\u{45b}', '\0', '\0']), + ('\u{40c}', ['\u{45c}', '\0', '\0']), ('\u{40d}', ['\u{45d}', '\0', '\0']), ('\u{40e}', + ['\u{45e}', '\0', '\0']), ('\u{40f}', ['\u{45f}', '\0', '\0']), ('\u{410}', ['\u{430}', + '\0', '\0']), ('\u{411}', ['\u{431}', '\0', '\0']), ('\u{412}', ['\u{432}', '\0', '\0']), + ('\u{413}', ['\u{433}', '\0', '\0']), ('\u{414}', ['\u{434}', '\0', '\0']), ('\u{415}', + ['\u{435}', '\0', '\0']), ('\u{416}', ['\u{436}', '\0', '\0']), ('\u{417}', ['\u{437}', + '\0', '\0']), ('\u{418}', ['\u{438}', '\0', '\0']), ('\u{419}', ['\u{439}', '\0', '\0']), + ('\u{41a}', ['\u{43a}', '\0', '\0']), ('\u{41b}', ['\u{43b}', '\0', '\0']), ('\u{41c}', + ['\u{43c}', '\0', '\0']), ('\u{41d}', ['\u{43d}', '\0', '\0']), ('\u{41e}', ['\u{43e}', + '\0', '\0']), ('\u{41f}', ['\u{43f}', '\0', '\0']), ('\u{420}', ['\u{440}', '\0', '\0']), + ('\u{421}', ['\u{441}', '\0', '\0']), ('\u{422}', ['\u{442}', '\0', '\0']), ('\u{423}', + ['\u{443}', '\0', '\0']), ('\u{424}', ['\u{444}', '\0', '\0']), ('\u{425}', ['\u{445}', + '\0', '\0']), ('\u{426}', ['\u{446}', '\0', '\0']), ('\u{427}', ['\u{447}', '\0', '\0']), + ('\u{428}', ['\u{448}', '\0', '\0']), ('\u{429}', ['\u{449}', '\0', '\0']), ('\u{42a}', + ['\u{44a}', '\0', '\0']), ('\u{42b}', ['\u{44b}', '\0', '\0']), ('\u{42c}', ['\u{44c}', + '\0', '\0']), ('\u{42d}', ['\u{44d}', '\0', '\0']), ('\u{42e}', ['\u{44e}', '\0', '\0']), + ('\u{42f}', ['\u{44f}', '\0', '\0']), ('\u{460}', ['\u{461}', '\0', '\0']), ('\u{462}', + ['\u{463}', '\0', '\0']), ('\u{464}', ['\u{465}', '\0', '\0']), ('\u{466}', ['\u{467}', + '\0', '\0']), ('\u{468}', ['\u{469}', '\0', '\0']), ('\u{46a}', ['\u{46b}', '\0', '\0']), + ('\u{46c}', ['\u{46d}', '\0', '\0']), ('\u{46e}', ['\u{46f}', '\0', '\0']), ('\u{470}', + ['\u{471}', '\0', '\0']), ('\u{472}', ['\u{473}', '\0', '\0']), ('\u{474}', ['\u{475}', + '\0', '\0']), ('\u{476}', ['\u{477}', '\0', '\0']), ('\u{478}', ['\u{479}', '\0', '\0']), + ('\u{47a}', ['\u{47b}', '\0', '\0']), ('\u{47c}', ['\u{47d}', '\0', '\0']), ('\u{47e}', + ['\u{47f}', '\0', '\0']), ('\u{480}', ['\u{481}', '\0', '\0']), ('\u{48a}', ['\u{48b}', + '\0', '\0']), ('\u{48c}', ['\u{48d}', '\0', '\0']), ('\u{48e}', ['\u{48f}', '\0', '\0']), + ('\u{490}', ['\u{491}', '\0', '\0']), ('\u{492}', ['\u{493}', '\0', '\0']), ('\u{494}', + ['\u{495}', '\0', '\0']), ('\u{496}', ['\u{497}', '\0', '\0']), ('\u{498}', ['\u{499}', + '\0', '\0']), ('\u{49a}', ['\u{49b}', '\0', '\0']), ('\u{49c}', ['\u{49d}', '\0', '\0']), + ('\u{49e}', ['\u{49f}', '\0', '\0']), ('\u{4a0}', ['\u{4a1}', '\0', '\0']), ('\u{4a2}', + ['\u{4a3}', '\0', '\0']), ('\u{4a4}', ['\u{4a5}', '\0', '\0']), ('\u{4a6}', ['\u{4a7}', + '\0', '\0']), ('\u{4a8}', ['\u{4a9}', '\0', '\0']), ('\u{4aa}', ['\u{4ab}', '\0', '\0']), + ('\u{4ac}', ['\u{4ad}', '\0', '\0']), ('\u{4ae}', ['\u{4af}', '\0', '\0']), ('\u{4b0}', + ['\u{4b1}', '\0', '\0']), ('\u{4b2}', ['\u{4b3}', '\0', '\0']), ('\u{4b4}', ['\u{4b5}', + '\0', '\0']), ('\u{4b6}', ['\u{4b7}', '\0', '\0']), ('\u{4b8}', ['\u{4b9}', '\0', '\0']), + ('\u{4ba}', ['\u{4bb}', '\0', '\0']), ('\u{4bc}', ['\u{4bd}', '\0', '\0']), ('\u{4be}', + ['\u{4bf}', '\0', '\0']), ('\u{4c0}', ['\u{4cf}', '\0', '\0']), ('\u{4c1}', ['\u{4c2}', + '\0', '\0']), ('\u{4c3}', ['\u{4c4}', '\0', '\0']), ('\u{4c5}', ['\u{4c6}', '\0', '\0']), + ('\u{4c7}', ['\u{4c8}', '\0', '\0']), ('\u{4c9}', ['\u{4ca}', '\0', '\0']), ('\u{4cb}', + ['\u{4cc}', '\0', '\0']), ('\u{4cd}', ['\u{4ce}', '\0', '\0']), ('\u{4d0}', ['\u{4d1}', + '\0', '\0']), ('\u{4d2}', ['\u{4d3}', '\0', '\0']), ('\u{4d4}', ['\u{4d5}', '\0', '\0']), + ('\u{4d6}', ['\u{4d7}', '\0', '\0']), ('\u{4d8}', ['\u{4d9}', '\0', '\0']), ('\u{4da}', + ['\u{4db}', '\0', '\0']), ('\u{4dc}', ['\u{4dd}', '\0', '\0']), ('\u{4de}', ['\u{4df}', + '\0', '\0']), ('\u{4e0}', ['\u{4e1}', '\0', '\0']), ('\u{4e2}', ['\u{4e3}', '\0', '\0']), + ('\u{4e4}', ['\u{4e5}', '\0', '\0']), ('\u{4e6}', ['\u{4e7}', '\0', '\0']), ('\u{4e8}', + ['\u{4e9}', '\0', '\0']), ('\u{4ea}', ['\u{4eb}', '\0', '\0']), ('\u{4ec}', ['\u{4ed}', + '\0', '\0']), ('\u{4ee}', ['\u{4ef}', '\0', '\0']), ('\u{4f0}', ['\u{4f1}', '\0', '\0']), + ('\u{4f2}', ['\u{4f3}', '\0', '\0']), ('\u{4f4}', ['\u{4f5}', '\0', '\0']), ('\u{4f6}', + ['\u{4f7}', '\0', '\0']), ('\u{4f8}', ['\u{4f9}', '\0', '\0']), ('\u{4fa}', ['\u{4fb}', + '\0', '\0']), ('\u{4fc}', ['\u{4fd}', '\0', '\0']), ('\u{4fe}', ['\u{4ff}', '\0', '\0']), + ('\u{500}', ['\u{501}', '\0', '\0']), ('\u{502}', ['\u{503}', '\0', '\0']), ('\u{504}', + ['\u{505}', '\0', '\0']), ('\u{506}', ['\u{507}', '\0', '\0']), ('\u{508}', ['\u{509}', + '\0', '\0']), ('\u{50a}', ['\u{50b}', '\0', '\0']), ('\u{50c}', ['\u{50d}', '\0', '\0']), + ('\u{50e}', ['\u{50f}', '\0', '\0']), ('\u{510}', ['\u{511}', '\0', '\0']), ('\u{512}', + ['\u{513}', '\0', '\0']), ('\u{514}', ['\u{515}', '\0', '\0']), ('\u{516}', ['\u{517}', + '\0', '\0']), ('\u{518}', ['\u{519}', '\0', '\0']), ('\u{51a}', ['\u{51b}', '\0', '\0']), + ('\u{51c}', ['\u{51d}', '\0', '\0']), ('\u{51e}', ['\u{51f}', '\0', '\0']), ('\u{520}', + ['\u{521}', '\0', '\0']), ('\u{522}', ['\u{523}', '\0', '\0']), ('\u{524}', ['\u{525}', + '\0', '\0']), ('\u{526}', ['\u{527}', '\0', '\0']), ('\u{528}', ['\u{529}', '\0', '\0']), + ('\u{52a}', ['\u{52b}', '\0', '\0']), ('\u{52c}', ['\u{52d}', '\0', '\0']), ('\u{52e}', + ['\u{52f}', '\0', '\0']), ('\u{531}', ['\u{561}', '\0', '\0']), ('\u{532}', ['\u{562}', + '\0', '\0']), ('\u{533}', ['\u{563}', '\0', '\0']), ('\u{534}', ['\u{564}', '\0', '\0']), + ('\u{535}', ['\u{565}', '\0', '\0']), ('\u{536}', ['\u{566}', '\0', '\0']), ('\u{537}', + ['\u{567}', '\0', '\0']), ('\u{538}', ['\u{568}', '\0', '\0']), ('\u{539}', ['\u{569}', + '\0', '\0']), ('\u{53a}', ['\u{56a}', '\0', '\0']), ('\u{53b}', ['\u{56b}', '\0', '\0']), + ('\u{53c}', ['\u{56c}', '\0', '\0']), ('\u{53d}', ['\u{56d}', '\0', '\0']), ('\u{53e}', + ['\u{56e}', '\0', '\0']), ('\u{53f}', ['\u{56f}', '\0', '\0']), ('\u{540}', ['\u{570}', + '\0', '\0']), ('\u{541}', ['\u{571}', '\0', '\0']), ('\u{542}', ['\u{572}', '\0', '\0']), + ('\u{543}', ['\u{573}', '\0', '\0']), ('\u{544}', ['\u{574}', '\0', '\0']), ('\u{545}', + ['\u{575}', '\0', '\0']), ('\u{546}', ['\u{576}', '\0', '\0']), ('\u{547}', ['\u{577}', + '\0', '\0']), ('\u{548}', ['\u{578}', '\0', '\0']), ('\u{549}', ['\u{579}', '\0', '\0']), + ('\u{54a}', ['\u{57a}', '\0', '\0']), ('\u{54b}', ['\u{57b}', '\0', '\0']), ('\u{54c}', + ['\u{57c}', '\0', '\0']), ('\u{54d}', ['\u{57d}', '\0', '\0']), ('\u{54e}', ['\u{57e}', + '\0', '\0']), ('\u{54f}', ['\u{57f}', '\0', '\0']), ('\u{550}', ['\u{580}', '\0', '\0']), + ('\u{551}', ['\u{581}', '\0', '\0']), ('\u{552}', ['\u{582}', '\0', '\0']), ('\u{553}', + ['\u{583}', '\0', '\0']), ('\u{554}', ['\u{584}', '\0', '\0']), ('\u{555}', ['\u{585}', + '\0', '\0']), ('\u{556}', ['\u{586}', '\0', '\0']), ('\u{10a0}', ['\u{2d00}', '\0', '\0']), + ('\u{10a1}', ['\u{2d01}', '\0', '\0']), ('\u{10a2}', ['\u{2d02}', '\0', '\0']), ('\u{10a3}', + ['\u{2d03}', '\0', '\0']), ('\u{10a4}', ['\u{2d04}', '\0', '\0']), ('\u{10a5}', ['\u{2d05}', + '\0', '\0']), ('\u{10a6}', ['\u{2d06}', '\0', '\0']), ('\u{10a7}', ['\u{2d07}', '\0', + '\0']), ('\u{10a8}', ['\u{2d08}', '\0', '\0']), ('\u{10a9}', ['\u{2d09}', '\0', '\0']), + ('\u{10aa}', ['\u{2d0a}', '\0', '\0']), ('\u{10ab}', ['\u{2d0b}', '\0', '\0']), ('\u{10ac}', + ['\u{2d0c}', '\0', '\0']), ('\u{10ad}', ['\u{2d0d}', '\0', '\0']), ('\u{10ae}', ['\u{2d0e}', + '\0', '\0']), ('\u{10af}', ['\u{2d0f}', '\0', '\0']), ('\u{10b0}', ['\u{2d10}', '\0', + '\0']), ('\u{10b1}', ['\u{2d11}', '\0', '\0']), ('\u{10b2}', ['\u{2d12}', '\0', '\0']), + ('\u{10b3}', ['\u{2d13}', '\0', '\0']), ('\u{10b4}', ['\u{2d14}', '\0', '\0']), ('\u{10b5}', + ['\u{2d15}', '\0', '\0']), ('\u{10b6}', ['\u{2d16}', '\0', '\0']), ('\u{10b7}', ['\u{2d17}', + '\0', '\0']), ('\u{10b8}', ['\u{2d18}', '\0', '\0']), ('\u{10b9}', ['\u{2d19}', '\0', + '\0']), ('\u{10ba}', ['\u{2d1a}', '\0', '\0']), ('\u{10bb}', ['\u{2d1b}', '\0', '\0']), + ('\u{10bc}', ['\u{2d1c}', '\0', '\0']), ('\u{10bd}', ['\u{2d1d}', '\0', '\0']), ('\u{10be}', + ['\u{2d1e}', '\0', '\0']), ('\u{10bf}', ['\u{2d1f}', '\0', '\0']), ('\u{10c0}', ['\u{2d20}', + '\0', '\0']), ('\u{10c1}', ['\u{2d21}', '\0', '\0']), ('\u{10c2}', ['\u{2d22}', '\0', + '\0']), ('\u{10c3}', ['\u{2d23}', '\0', '\0']), ('\u{10c4}', ['\u{2d24}', '\0', '\0']), + ('\u{10c5}', ['\u{2d25}', '\0', '\0']), ('\u{10c7}', ['\u{2d27}', '\0', '\0']), ('\u{10cd}', + ['\u{2d2d}', '\0', '\0']), ('\u{13a0}', ['\u{ab70}', '\0', '\0']), ('\u{13a1}', ['\u{ab71}', + '\0', '\0']), ('\u{13a2}', ['\u{ab72}', '\0', '\0']), ('\u{13a3}', ['\u{ab73}', '\0', + '\0']), ('\u{13a4}', ['\u{ab74}', '\0', '\0']), ('\u{13a5}', ['\u{ab75}', '\0', '\0']), + ('\u{13a6}', ['\u{ab76}', '\0', '\0']), ('\u{13a7}', ['\u{ab77}', '\0', '\0']), ('\u{13a8}', + ['\u{ab78}', '\0', '\0']), ('\u{13a9}', ['\u{ab79}', '\0', '\0']), ('\u{13aa}', ['\u{ab7a}', + '\0', '\0']), ('\u{13ab}', ['\u{ab7b}', '\0', '\0']), ('\u{13ac}', ['\u{ab7c}', '\0', + '\0']), ('\u{13ad}', ['\u{ab7d}', '\0', '\0']), ('\u{13ae}', ['\u{ab7e}', '\0', '\0']), + ('\u{13af}', ['\u{ab7f}', '\0', '\0']), ('\u{13b0}', ['\u{ab80}', '\0', '\0']), ('\u{13b1}', + ['\u{ab81}', '\0', '\0']), ('\u{13b2}', ['\u{ab82}', '\0', '\0']), ('\u{13b3}', ['\u{ab83}', + '\0', '\0']), ('\u{13b4}', ['\u{ab84}', '\0', '\0']), ('\u{13b5}', ['\u{ab85}', '\0', + '\0']), ('\u{13b6}', ['\u{ab86}', '\0', '\0']), ('\u{13b7}', ['\u{ab87}', '\0', '\0']), + ('\u{13b8}', ['\u{ab88}', '\0', '\0']), ('\u{13b9}', ['\u{ab89}', '\0', '\0']), ('\u{13ba}', + ['\u{ab8a}', '\0', '\0']), ('\u{13bb}', ['\u{ab8b}', '\0', '\0']), ('\u{13bc}', ['\u{ab8c}', + '\0', '\0']), ('\u{13bd}', ['\u{ab8d}', '\0', '\0']), ('\u{13be}', ['\u{ab8e}', '\0', + '\0']), ('\u{13bf}', ['\u{ab8f}', '\0', '\0']), ('\u{13c0}', ['\u{ab90}', '\0', '\0']), + ('\u{13c1}', ['\u{ab91}', '\0', '\0']), ('\u{13c2}', ['\u{ab92}', '\0', '\0']), ('\u{13c3}', + ['\u{ab93}', '\0', '\0']), ('\u{13c4}', ['\u{ab94}', '\0', '\0']), ('\u{13c5}', ['\u{ab95}', + '\0', '\0']), ('\u{13c6}', ['\u{ab96}', '\0', '\0']), ('\u{13c7}', ['\u{ab97}', '\0', + '\0']), ('\u{13c8}', ['\u{ab98}', '\0', '\0']), ('\u{13c9}', ['\u{ab99}', '\0', '\0']), + ('\u{13ca}', ['\u{ab9a}', '\0', '\0']), ('\u{13cb}', ['\u{ab9b}', '\0', '\0']), ('\u{13cc}', + ['\u{ab9c}', '\0', '\0']), ('\u{13cd}', ['\u{ab9d}', '\0', '\0']), ('\u{13ce}', ['\u{ab9e}', + '\0', '\0']), ('\u{13cf}', ['\u{ab9f}', '\0', '\0']), ('\u{13d0}', ['\u{aba0}', '\0', + '\0']), ('\u{13d1}', ['\u{aba1}', '\0', '\0']), ('\u{13d2}', ['\u{aba2}', '\0', '\0']), + ('\u{13d3}', ['\u{aba3}', '\0', '\0']), ('\u{13d4}', ['\u{aba4}', '\0', '\0']), ('\u{13d5}', + ['\u{aba5}', '\0', '\0']), ('\u{13d6}', ['\u{aba6}', '\0', '\0']), ('\u{13d7}', ['\u{aba7}', + '\0', '\0']), ('\u{13d8}', ['\u{aba8}', '\0', '\0']), ('\u{13d9}', ['\u{aba9}', '\0', + '\0']), ('\u{13da}', ['\u{abaa}', '\0', '\0']), ('\u{13db}', ['\u{abab}', '\0', '\0']), + ('\u{13dc}', ['\u{abac}', '\0', '\0']), ('\u{13dd}', ['\u{abad}', '\0', '\0']), ('\u{13de}', + ['\u{abae}', '\0', '\0']), ('\u{13df}', ['\u{abaf}', '\0', '\0']), ('\u{13e0}', ['\u{abb0}', + '\0', '\0']), ('\u{13e1}', ['\u{abb1}', '\0', '\0']), ('\u{13e2}', ['\u{abb2}', '\0', + '\0']), ('\u{13e3}', ['\u{abb3}', '\0', '\0']), ('\u{13e4}', ['\u{abb4}', '\0', '\0']), + ('\u{13e5}', ['\u{abb5}', '\0', '\0']), ('\u{13e6}', ['\u{abb6}', '\0', '\0']), ('\u{13e7}', + ['\u{abb7}', '\0', '\0']), ('\u{13e8}', ['\u{abb8}', '\0', '\0']), ('\u{13e9}', ['\u{abb9}', + '\0', '\0']), ('\u{13ea}', ['\u{abba}', '\0', '\0']), ('\u{13eb}', ['\u{abbb}', '\0', + '\0']), ('\u{13ec}', ['\u{abbc}', '\0', '\0']), ('\u{13ed}', ['\u{abbd}', '\0', '\0']), + ('\u{13ee}', ['\u{abbe}', '\0', '\0']), ('\u{13ef}', ['\u{abbf}', '\0', '\0']), ('\u{13f0}', + ['\u{13f8}', '\0', '\0']), ('\u{13f1}', ['\u{13f9}', '\0', '\0']), ('\u{13f2}', ['\u{13fa}', + '\0', '\0']), ('\u{13f3}', ['\u{13fb}', '\0', '\0']), ('\u{13f4}', ['\u{13fc}', '\0', + '\0']), ('\u{13f5}', ['\u{13fd}', '\0', '\0']), ('\u{1c90}', ['\u{10d0}', '\0', '\0']), + ('\u{1c91}', ['\u{10d1}', '\0', '\0']), ('\u{1c92}', ['\u{10d2}', '\0', '\0']), ('\u{1c93}', + ['\u{10d3}', '\0', '\0']), ('\u{1c94}', ['\u{10d4}', '\0', '\0']), ('\u{1c95}', ['\u{10d5}', + '\0', '\0']), ('\u{1c96}', ['\u{10d6}', '\0', '\0']), ('\u{1c97}', ['\u{10d7}', '\0', + '\0']), ('\u{1c98}', ['\u{10d8}', '\0', '\0']), ('\u{1c99}', ['\u{10d9}', '\0', '\0']), + ('\u{1c9a}', ['\u{10da}', '\0', '\0']), ('\u{1c9b}', ['\u{10db}', '\0', '\0']), ('\u{1c9c}', + ['\u{10dc}', '\0', '\0']), ('\u{1c9d}', ['\u{10dd}', '\0', '\0']), ('\u{1c9e}', ['\u{10de}', + '\0', '\0']), ('\u{1c9f}', ['\u{10df}', '\0', '\0']), ('\u{1ca0}', ['\u{10e0}', '\0', + '\0']), ('\u{1ca1}', ['\u{10e1}', '\0', '\0']), ('\u{1ca2}', ['\u{10e2}', '\0', '\0']), + ('\u{1ca3}', ['\u{10e3}', '\0', '\0']), ('\u{1ca4}', ['\u{10e4}', '\0', '\0']), ('\u{1ca5}', + ['\u{10e5}', '\0', '\0']), ('\u{1ca6}', ['\u{10e6}', '\0', '\0']), ('\u{1ca7}', ['\u{10e7}', + '\0', '\0']), ('\u{1ca8}', ['\u{10e8}', '\0', '\0']), ('\u{1ca9}', ['\u{10e9}', '\0', + '\0']), ('\u{1caa}', ['\u{10ea}', '\0', '\0']), ('\u{1cab}', ['\u{10eb}', '\0', '\0']), + ('\u{1cac}', ['\u{10ec}', '\0', '\0']), ('\u{1cad}', ['\u{10ed}', '\0', '\0']), ('\u{1cae}', + ['\u{10ee}', '\0', '\0']), ('\u{1caf}', ['\u{10ef}', '\0', '\0']), ('\u{1cb0}', ['\u{10f0}', + '\0', '\0']), ('\u{1cb1}', ['\u{10f1}', '\0', '\0']), ('\u{1cb2}', ['\u{10f2}', '\0', + '\0']), ('\u{1cb3}', ['\u{10f3}', '\0', '\0']), ('\u{1cb4}', ['\u{10f4}', '\0', '\0']), + ('\u{1cb5}', ['\u{10f5}', '\0', '\0']), ('\u{1cb6}', ['\u{10f6}', '\0', '\0']), ('\u{1cb7}', + ['\u{10f7}', '\0', '\0']), ('\u{1cb8}', ['\u{10f8}', '\0', '\0']), ('\u{1cb9}', ['\u{10f9}', + '\0', '\0']), ('\u{1cba}', ['\u{10fa}', '\0', '\0']), ('\u{1cbd}', ['\u{10fd}', '\0', + '\0']), ('\u{1cbe}', ['\u{10fe}', '\0', '\0']), ('\u{1cbf}', ['\u{10ff}', '\0', '\0']), + ('\u{1e00}', ['\u{1e01}', '\0', '\0']), ('\u{1e02}', ['\u{1e03}', '\0', '\0']), ('\u{1e04}', + ['\u{1e05}', '\0', '\0']), ('\u{1e06}', ['\u{1e07}', '\0', '\0']), ('\u{1e08}', ['\u{1e09}', + '\0', '\0']), ('\u{1e0a}', ['\u{1e0b}', '\0', '\0']), ('\u{1e0c}', ['\u{1e0d}', '\0', + '\0']), ('\u{1e0e}', ['\u{1e0f}', '\0', '\0']), ('\u{1e10}', ['\u{1e11}', '\0', '\0']), + ('\u{1e12}', ['\u{1e13}', '\0', '\0']), ('\u{1e14}', ['\u{1e15}', '\0', '\0']), ('\u{1e16}', + ['\u{1e17}', '\0', '\0']), ('\u{1e18}', ['\u{1e19}', '\0', '\0']), ('\u{1e1a}', ['\u{1e1b}', + '\0', '\0']), ('\u{1e1c}', ['\u{1e1d}', '\0', '\0']), ('\u{1e1e}', ['\u{1e1f}', '\0', + '\0']), ('\u{1e20}', ['\u{1e21}', '\0', '\0']), ('\u{1e22}', ['\u{1e23}', '\0', '\0']), + ('\u{1e24}', ['\u{1e25}', '\0', '\0']), ('\u{1e26}', ['\u{1e27}', '\0', '\0']), ('\u{1e28}', + ['\u{1e29}', '\0', '\0']), ('\u{1e2a}', ['\u{1e2b}', '\0', '\0']), ('\u{1e2c}', ['\u{1e2d}', + '\0', '\0']), ('\u{1e2e}', ['\u{1e2f}', '\0', '\0']), ('\u{1e30}', ['\u{1e31}', '\0', + '\0']), ('\u{1e32}', ['\u{1e33}', '\0', '\0']), ('\u{1e34}', ['\u{1e35}', '\0', '\0']), + ('\u{1e36}', ['\u{1e37}', '\0', '\0']), ('\u{1e38}', ['\u{1e39}', '\0', '\0']), ('\u{1e3a}', + ['\u{1e3b}', '\0', '\0']), ('\u{1e3c}', ['\u{1e3d}', '\0', '\0']), ('\u{1e3e}', ['\u{1e3f}', + '\0', '\0']), ('\u{1e40}', ['\u{1e41}', '\0', '\0']), ('\u{1e42}', ['\u{1e43}', '\0', + '\0']), ('\u{1e44}', ['\u{1e45}', '\0', '\0']), ('\u{1e46}', ['\u{1e47}', '\0', '\0']), + ('\u{1e48}', ['\u{1e49}', '\0', '\0']), ('\u{1e4a}', ['\u{1e4b}', '\0', '\0']), ('\u{1e4c}', + ['\u{1e4d}', '\0', '\0']), ('\u{1e4e}', ['\u{1e4f}', '\0', '\0']), ('\u{1e50}', ['\u{1e51}', + '\0', '\0']), ('\u{1e52}', ['\u{1e53}', '\0', '\0']), ('\u{1e54}', ['\u{1e55}', '\0', + '\0']), ('\u{1e56}', ['\u{1e57}', '\0', '\0']), ('\u{1e58}', ['\u{1e59}', '\0', '\0']), + ('\u{1e5a}', ['\u{1e5b}', '\0', '\0']), ('\u{1e5c}', ['\u{1e5d}', '\0', '\0']), ('\u{1e5e}', + ['\u{1e5f}', '\0', '\0']), ('\u{1e60}', ['\u{1e61}', '\0', '\0']), ('\u{1e62}', ['\u{1e63}', + '\0', '\0']), ('\u{1e64}', ['\u{1e65}', '\0', '\0']), ('\u{1e66}', ['\u{1e67}', '\0', + '\0']), ('\u{1e68}', ['\u{1e69}', '\0', '\0']), ('\u{1e6a}', ['\u{1e6b}', '\0', '\0']), + ('\u{1e6c}', ['\u{1e6d}', '\0', '\0']), ('\u{1e6e}', ['\u{1e6f}', '\0', '\0']), ('\u{1e70}', + ['\u{1e71}', '\0', '\0']), ('\u{1e72}', ['\u{1e73}', '\0', '\0']), ('\u{1e74}', ['\u{1e75}', + '\0', '\0']), ('\u{1e76}', ['\u{1e77}', '\0', '\0']), ('\u{1e78}', ['\u{1e79}', '\0', + '\0']), ('\u{1e7a}', ['\u{1e7b}', '\0', '\0']), ('\u{1e7c}', ['\u{1e7d}', '\0', '\0']), + ('\u{1e7e}', ['\u{1e7f}', '\0', '\0']), ('\u{1e80}', ['\u{1e81}', '\0', '\0']), ('\u{1e82}', + ['\u{1e83}', '\0', '\0']), ('\u{1e84}', ['\u{1e85}', '\0', '\0']), ('\u{1e86}', ['\u{1e87}', + '\0', '\0']), ('\u{1e88}', ['\u{1e89}', '\0', '\0']), ('\u{1e8a}', ['\u{1e8b}', '\0', + '\0']), ('\u{1e8c}', ['\u{1e8d}', '\0', '\0']), ('\u{1e8e}', ['\u{1e8f}', '\0', '\0']), + ('\u{1e90}', ['\u{1e91}', '\0', '\0']), ('\u{1e92}', ['\u{1e93}', '\0', '\0']), ('\u{1e94}', + ['\u{1e95}', '\0', '\0']), ('\u{1e9e}', ['\u{df}', '\0', '\0']), ('\u{1ea0}', ['\u{1ea1}', + '\0', '\0']), ('\u{1ea2}', ['\u{1ea3}', '\0', '\0']), ('\u{1ea4}', ['\u{1ea5}', '\0', + '\0']), ('\u{1ea6}', ['\u{1ea7}', '\0', '\0']), ('\u{1ea8}', ['\u{1ea9}', '\0', '\0']), + ('\u{1eaa}', ['\u{1eab}', '\0', '\0']), ('\u{1eac}', ['\u{1ead}', '\0', '\0']), ('\u{1eae}', + ['\u{1eaf}', '\0', '\0']), ('\u{1eb0}', ['\u{1eb1}', '\0', '\0']), ('\u{1eb2}', ['\u{1eb3}', + '\0', '\0']), ('\u{1eb4}', ['\u{1eb5}', '\0', '\0']), ('\u{1eb6}', ['\u{1eb7}', '\0', + '\0']), ('\u{1eb8}', ['\u{1eb9}', '\0', '\0']), ('\u{1eba}', ['\u{1ebb}', '\0', '\0']), + ('\u{1ebc}', ['\u{1ebd}', '\0', '\0']), ('\u{1ebe}', ['\u{1ebf}', '\0', '\0']), ('\u{1ec0}', + ['\u{1ec1}', '\0', '\0']), ('\u{1ec2}', ['\u{1ec3}', '\0', '\0']), ('\u{1ec4}', ['\u{1ec5}', + '\0', '\0']), ('\u{1ec6}', ['\u{1ec7}', '\0', '\0']), ('\u{1ec8}', ['\u{1ec9}', '\0', + '\0']), ('\u{1eca}', ['\u{1ecb}', '\0', '\0']), ('\u{1ecc}', ['\u{1ecd}', '\0', '\0']), + ('\u{1ece}', ['\u{1ecf}', '\0', '\0']), ('\u{1ed0}', ['\u{1ed1}', '\0', '\0']), ('\u{1ed2}', + ['\u{1ed3}', '\0', '\0']), ('\u{1ed4}', ['\u{1ed5}', '\0', '\0']), ('\u{1ed6}', ['\u{1ed7}', + '\0', '\0']), ('\u{1ed8}', ['\u{1ed9}', '\0', '\0']), ('\u{1eda}', ['\u{1edb}', '\0', + '\0']), ('\u{1edc}', ['\u{1edd}', '\0', '\0']), ('\u{1ede}', ['\u{1edf}', '\0', '\0']), + ('\u{1ee0}', ['\u{1ee1}', '\0', '\0']), ('\u{1ee2}', ['\u{1ee3}', '\0', '\0']), ('\u{1ee4}', + ['\u{1ee5}', '\0', '\0']), ('\u{1ee6}', ['\u{1ee7}', '\0', '\0']), ('\u{1ee8}', ['\u{1ee9}', + '\0', '\0']), ('\u{1eea}', ['\u{1eeb}', '\0', '\0']), ('\u{1eec}', ['\u{1eed}', '\0', + '\0']), ('\u{1eee}', ['\u{1eef}', '\0', '\0']), ('\u{1ef0}', ['\u{1ef1}', '\0', '\0']), + ('\u{1ef2}', ['\u{1ef3}', '\0', '\0']), ('\u{1ef4}', ['\u{1ef5}', '\0', '\0']), ('\u{1ef6}', + ['\u{1ef7}', '\0', '\0']), ('\u{1ef8}', ['\u{1ef9}', '\0', '\0']), ('\u{1efa}', ['\u{1efb}', + '\0', '\0']), ('\u{1efc}', ['\u{1efd}', '\0', '\0']), ('\u{1efe}', ['\u{1eff}', '\0', + '\0']), ('\u{1f08}', ['\u{1f00}', '\0', '\0']), ('\u{1f09}', ['\u{1f01}', '\0', '\0']), + ('\u{1f0a}', ['\u{1f02}', '\0', '\0']), ('\u{1f0b}', ['\u{1f03}', '\0', '\0']), ('\u{1f0c}', + ['\u{1f04}', '\0', '\0']), ('\u{1f0d}', ['\u{1f05}', '\0', '\0']), ('\u{1f0e}', ['\u{1f06}', + '\0', '\0']), ('\u{1f0f}', ['\u{1f07}', '\0', '\0']), ('\u{1f18}', ['\u{1f10}', '\0', + '\0']), ('\u{1f19}', ['\u{1f11}', '\0', '\0']), ('\u{1f1a}', ['\u{1f12}', '\0', '\0']), + ('\u{1f1b}', ['\u{1f13}', '\0', '\0']), ('\u{1f1c}', ['\u{1f14}', '\0', '\0']), ('\u{1f1d}', + ['\u{1f15}', '\0', '\0']), ('\u{1f28}', ['\u{1f20}', '\0', '\0']), ('\u{1f29}', ['\u{1f21}', + '\0', '\0']), ('\u{1f2a}', ['\u{1f22}', '\0', '\0']), ('\u{1f2b}', ['\u{1f23}', '\0', + '\0']), ('\u{1f2c}', ['\u{1f24}', '\0', '\0']), ('\u{1f2d}', ['\u{1f25}', '\0', '\0']), + ('\u{1f2e}', ['\u{1f26}', '\0', '\0']), ('\u{1f2f}', ['\u{1f27}', '\0', '\0']), ('\u{1f38}', + ['\u{1f30}', '\0', '\0']), ('\u{1f39}', ['\u{1f31}', '\0', '\0']), ('\u{1f3a}', ['\u{1f32}', + '\0', '\0']), ('\u{1f3b}', ['\u{1f33}', '\0', '\0']), ('\u{1f3c}', ['\u{1f34}', '\0', + '\0']), ('\u{1f3d}', ['\u{1f35}', '\0', '\0']), ('\u{1f3e}', ['\u{1f36}', '\0', '\0']), + ('\u{1f3f}', ['\u{1f37}', '\0', '\0']), ('\u{1f48}', ['\u{1f40}', '\0', '\0']), ('\u{1f49}', + ['\u{1f41}', '\0', '\0']), ('\u{1f4a}', ['\u{1f42}', '\0', '\0']), ('\u{1f4b}', ['\u{1f43}', + '\0', '\0']), ('\u{1f4c}', ['\u{1f44}', '\0', '\0']), ('\u{1f4d}', ['\u{1f45}', '\0', + '\0']), ('\u{1f59}', ['\u{1f51}', '\0', '\0']), ('\u{1f5b}', ['\u{1f53}', '\0', '\0']), + ('\u{1f5d}', ['\u{1f55}', '\0', '\0']), ('\u{1f5f}', ['\u{1f57}', '\0', '\0']), ('\u{1f68}', + ['\u{1f60}', '\0', '\0']), ('\u{1f69}', ['\u{1f61}', '\0', '\0']), ('\u{1f6a}', ['\u{1f62}', + '\0', '\0']), ('\u{1f6b}', ['\u{1f63}', '\0', '\0']), ('\u{1f6c}', ['\u{1f64}', '\0', + '\0']), ('\u{1f6d}', ['\u{1f65}', '\0', '\0']), ('\u{1f6e}', ['\u{1f66}', '\0', '\0']), + ('\u{1f6f}', ['\u{1f67}', '\0', '\0']), ('\u{1f88}', ['\u{1f80}', '\0', '\0']), ('\u{1f89}', + ['\u{1f81}', '\0', '\0']), ('\u{1f8a}', ['\u{1f82}', '\0', '\0']), ('\u{1f8b}', ['\u{1f83}', + '\0', '\0']), ('\u{1f8c}', ['\u{1f84}', '\0', '\0']), ('\u{1f8d}', ['\u{1f85}', '\0', + '\0']), ('\u{1f8e}', ['\u{1f86}', '\0', '\0']), ('\u{1f8f}', ['\u{1f87}', '\0', '\0']), + ('\u{1f98}', ['\u{1f90}', '\0', '\0']), ('\u{1f99}', ['\u{1f91}', '\0', '\0']), ('\u{1f9a}', + ['\u{1f92}', '\0', '\0']), ('\u{1f9b}', ['\u{1f93}', '\0', '\0']), ('\u{1f9c}', ['\u{1f94}', + '\0', '\0']), ('\u{1f9d}', ['\u{1f95}', '\0', '\0']), ('\u{1f9e}', ['\u{1f96}', '\0', + '\0']), ('\u{1f9f}', ['\u{1f97}', '\0', '\0']), ('\u{1fa8}', ['\u{1fa0}', '\0', '\0']), + ('\u{1fa9}', ['\u{1fa1}', '\0', '\0']), ('\u{1faa}', ['\u{1fa2}', '\0', '\0']), ('\u{1fab}', + ['\u{1fa3}', '\0', '\0']), ('\u{1fac}', ['\u{1fa4}', '\0', '\0']), ('\u{1fad}', ['\u{1fa5}', + '\0', '\0']), ('\u{1fae}', ['\u{1fa6}', '\0', '\0']), ('\u{1faf}', ['\u{1fa7}', '\0', + '\0']), ('\u{1fb8}', ['\u{1fb0}', '\0', '\0']), ('\u{1fb9}', ['\u{1fb1}', '\0', '\0']), + ('\u{1fba}', ['\u{1f70}', '\0', '\0']), ('\u{1fbb}', ['\u{1f71}', '\0', '\0']), ('\u{1fbc}', + ['\u{1fb3}', '\0', '\0']), ('\u{1fc8}', ['\u{1f72}', '\0', '\0']), ('\u{1fc9}', ['\u{1f73}', + '\0', '\0']), ('\u{1fca}', ['\u{1f74}', '\0', '\0']), ('\u{1fcb}', ['\u{1f75}', '\0', + '\0']), ('\u{1fcc}', ['\u{1fc3}', '\0', '\0']), ('\u{1fd8}', ['\u{1fd0}', '\0', '\0']), + ('\u{1fd9}', ['\u{1fd1}', '\0', '\0']), ('\u{1fda}', ['\u{1f76}', '\0', '\0']), ('\u{1fdb}', + ['\u{1f77}', '\0', '\0']), ('\u{1fe8}', ['\u{1fe0}', '\0', '\0']), ('\u{1fe9}', ['\u{1fe1}', + '\0', '\0']), ('\u{1fea}', ['\u{1f7a}', '\0', '\0']), ('\u{1feb}', ['\u{1f7b}', '\0', + '\0']), ('\u{1fec}', ['\u{1fe5}', '\0', '\0']), ('\u{1ff8}', ['\u{1f78}', '\0', '\0']), + ('\u{1ff9}', ['\u{1f79}', '\0', '\0']), ('\u{1ffa}', ['\u{1f7c}', '\0', '\0']), ('\u{1ffb}', + ['\u{1f7d}', '\0', '\0']), ('\u{1ffc}', ['\u{1ff3}', '\0', '\0']), ('\u{2126}', ['\u{3c9}', + '\0', '\0']), ('\u{212a}', ['\u{6b}', '\0', '\0']), ('\u{212b}', ['\u{e5}', '\0', '\0']), + ('\u{2132}', ['\u{214e}', '\0', '\0']), ('\u{2160}', ['\u{2170}', '\0', '\0']), ('\u{2161}', + ['\u{2171}', '\0', '\0']), ('\u{2162}', ['\u{2172}', '\0', '\0']), ('\u{2163}', ['\u{2173}', + '\0', '\0']), ('\u{2164}', ['\u{2174}', '\0', '\0']), ('\u{2165}', ['\u{2175}', '\0', + '\0']), ('\u{2166}', ['\u{2176}', '\0', '\0']), ('\u{2167}', ['\u{2177}', '\0', '\0']), + ('\u{2168}', ['\u{2178}', '\0', '\0']), ('\u{2169}', ['\u{2179}', '\0', '\0']), ('\u{216a}', + ['\u{217a}', '\0', '\0']), ('\u{216b}', ['\u{217b}', '\0', '\0']), ('\u{216c}', ['\u{217c}', + '\0', '\0']), ('\u{216d}', ['\u{217d}', '\0', '\0']), ('\u{216e}', ['\u{217e}', '\0', + '\0']), ('\u{216f}', ['\u{217f}', '\0', '\0']), ('\u{2183}', ['\u{2184}', '\0', '\0']), + ('\u{24b6}', ['\u{24d0}', '\0', '\0']), ('\u{24b7}', ['\u{24d1}', '\0', '\0']), ('\u{24b8}', + ['\u{24d2}', '\0', '\0']), ('\u{24b9}', ['\u{24d3}', '\0', '\0']), ('\u{24ba}', ['\u{24d4}', + '\0', '\0']), ('\u{24bb}', ['\u{24d5}', '\0', '\0']), ('\u{24bc}', ['\u{24d6}', '\0', + '\0']), ('\u{24bd}', ['\u{24d7}', '\0', '\0']), ('\u{24be}', ['\u{24d8}', '\0', '\0']), + ('\u{24bf}', ['\u{24d9}', '\0', '\0']), ('\u{24c0}', ['\u{24da}', '\0', '\0']), ('\u{24c1}', + ['\u{24db}', '\0', '\0']), ('\u{24c2}', ['\u{24dc}', '\0', '\0']), ('\u{24c3}', ['\u{24dd}', + '\0', '\0']), ('\u{24c4}', ['\u{24de}', '\0', '\0']), ('\u{24c5}', ['\u{24df}', '\0', + '\0']), ('\u{24c6}', ['\u{24e0}', '\0', '\0']), ('\u{24c7}', ['\u{24e1}', '\0', '\0']), + ('\u{24c8}', ['\u{24e2}', '\0', '\0']), ('\u{24c9}', ['\u{24e3}', '\0', '\0']), ('\u{24ca}', + ['\u{24e4}', '\0', '\0']), ('\u{24cb}', ['\u{24e5}', '\0', '\0']), ('\u{24cc}', ['\u{24e6}', + '\0', '\0']), ('\u{24cd}', ['\u{24e7}', '\0', '\0']), ('\u{24ce}', ['\u{24e8}', '\0', + '\0']), ('\u{24cf}', ['\u{24e9}', '\0', '\0']), ('\u{2c00}', ['\u{2c30}', '\0', '\0']), + ('\u{2c01}', ['\u{2c31}', '\0', '\0']), ('\u{2c02}', ['\u{2c32}', '\0', '\0']), ('\u{2c03}', + ['\u{2c33}', '\0', '\0']), ('\u{2c04}', ['\u{2c34}', '\0', '\0']), ('\u{2c05}', ['\u{2c35}', + '\0', '\0']), ('\u{2c06}', ['\u{2c36}', '\0', '\0']), ('\u{2c07}', ['\u{2c37}', '\0', + '\0']), ('\u{2c08}', ['\u{2c38}', '\0', '\0']), ('\u{2c09}', ['\u{2c39}', '\0', '\0']), + ('\u{2c0a}', ['\u{2c3a}', '\0', '\0']), ('\u{2c0b}', ['\u{2c3b}', '\0', '\0']), ('\u{2c0c}', + ['\u{2c3c}', '\0', '\0']), ('\u{2c0d}', ['\u{2c3d}', '\0', '\0']), ('\u{2c0e}', ['\u{2c3e}', + '\0', '\0']), ('\u{2c0f}', ['\u{2c3f}', '\0', '\0']), ('\u{2c10}', ['\u{2c40}', '\0', + '\0']), ('\u{2c11}', ['\u{2c41}', '\0', '\0']), ('\u{2c12}', ['\u{2c42}', '\0', '\0']), + ('\u{2c13}', ['\u{2c43}', '\0', '\0']), ('\u{2c14}', ['\u{2c44}', '\0', '\0']), ('\u{2c15}', + ['\u{2c45}', '\0', '\0']), ('\u{2c16}', ['\u{2c46}', '\0', '\0']), ('\u{2c17}', ['\u{2c47}', + '\0', '\0']), ('\u{2c18}', ['\u{2c48}', '\0', '\0']), ('\u{2c19}', ['\u{2c49}', '\0', + '\0']), ('\u{2c1a}', ['\u{2c4a}', '\0', '\0']), ('\u{2c1b}', ['\u{2c4b}', '\0', '\0']), + ('\u{2c1c}', ['\u{2c4c}', '\0', '\0']), ('\u{2c1d}', ['\u{2c4d}', '\0', '\0']), ('\u{2c1e}', + ['\u{2c4e}', '\0', '\0']), ('\u{2c1f}', ['\u{2c4f}', '\0', '\0']), ('\u{2c20}', ['\u{2c50}', + '\0', '\0']), ('\u{2c21}', ['\u{2c51}', '\0', '\0']), ('\u{2c22}', ['\u{2c52}', '\0', + '\0']), ('\u{2c23}', ['\u{2c53}', '\0', '\0']), ('\u{2c24}', ['\u{2c54}', '\0', '\0']), + ('\u{2c25}', ['\u{2c55}', '\0', '\0']), ('\u{2c26}', ['\u{2c56}', '\0', '\0']), ('\u{2c27}', + ['\u{2c57}', '\0', '\0']), ('\u{2c28}', ['\u{2c58}', '\0', '\0']), ('\u{2c29}', ['\u{2c59}', + '\0', '\0']), ('\u{2c2a}', ['\u{2c5a}', '\0', '\0']), ('\u{2c2b}', ['\u{2c5b}', '\0', + '\0']), ('\u{2c2c}', ['\u{2c5c}', '\0', '\0']), ('\u{2c2d}', ['\u{2c5d}', '\0', '\0']), + ('\u{2c2e}', ['\u{2c5e}', '\0', '\0']), ('\u{2c60}', ['\u{2c61}', '\0', '\0']), ('\u{2c62}', + ['\u{26b}', '\0', '\0']), ('\u{2c63}', ['\u{1d7d}', '\0', '\0']), ('\u{2c64}', ['\u{27d}', + '\0', '\0']), ('\u{2c67}', ['\u{2c68}', '\0', '\0']), ('\u{2c69}', ['\u{2c6a}', '\0', + '\0']), ('\u{2c6b}', ['\u{2c6c}', '\0', '\0']), ('\u{2c6d}', ['\u{251}', '\0', '\0']), + ('\u{2c6e}', ['\u{271}', '\0', '\0']), ('\u{2c6f}', ['\u{250}', '\0', '\0']), ('\u{2c70}', + ['\u{252}', '\0', '\0']), ('\u{2c72}', ['\u{2c73}', '\0', '\0']), ('\u{2c75}', ['\u{2c76}', + '\0', '\0']), ('\u{2c7e}', ['\u{23f}', '\0', '\0']), ('\u{2c7f}', ['\u{240}', '\0', '\0']), + ('\u{2c80}', ['\u{2c81}', '\0', '\0']), ('\u{2c82}', ['\u{2c83}', '\0', '\0']), ('\u{2c84}', + ['\u{2c85}', '\0', '\0']), ('\u{2c86}', ['\u{2c87}', '\0', '\0']), ('\u{2c88}', ['\u{2c89}', + '\0', '\0']), ('\u{2c8a}', ['\u{2c8b}', '\0', '\0']), ('\u{2c8c}', ['\u{2c8d}', '\0', + '\0']), ('\u{2c8e}', ['\u{2c8f}', '\0', '\0']), ('\u{2c90}', ['\u{2c91}', '\0', '\0']), + ('\u{2c92}', ['\u{2c93}', '\0', '\0']), ('\u{2c94}', ['\u{2c95}', '\0', '\0']), ('\u{2c96}', + ['\u{2c97}', '\0', '\0']), ('\u{2c98}', ['\u{2c99}', '\0', '\0']), ('\u{2c9a}', ['\u{2c9b}', + '\0', '\0']), ('\u{2c9c}', ['\u{2c9d}', '\0', '\0']), ('\u{2c9e}', ['\u{2c9f}', '\0', + '\0']), ('\u{2ca0}', ['\u{2ca1}', '\0', '\0']), ('\u{2ca2}', ['\u{2ca3}', '\0', '\0']), + ('\u{2ca4}', ['\u{2ca5}', '\0', '\0']), ('\u{2ca6}', ['\u{2ca7}', '\0', '\0']), ('\u{2ca8}', + ['\u{2ca9}', '\0', '\0']), ('\u{2caa}', ['\u{2cab}', '\0', '\0']), ('\u{2cac}', ['\u{2cad}', + '\0', '\0']), ('\u{2cae}', ['\u{2caf}', '\0', '\0']), ('\u{2cb0}', ['\u{2cb1}', '\0', + '\0']), ('\u{2cb2}', ['\u{2cb3}', '\0', '\0']), ('\u{2cb4}', ['\u{2cb5}', '\0', '\0']), + ('\u{2cb6}', ['\u{2cb7}', '\0', '\0']), ('\u{2cb8}', ['\u{2cb9}', '\0', '\0']), ('\u{2cba}', + ['\u{2cbb}', '\0', '\0']), ('\u{2cbc}', ['\u{2cbd}', '\0', '\0']), ('\u{2cbe}', ['\u{2cbf}', + '\0', '\0']), ('\u{2cc0}', ['\u{2cc1}', '\0', '\0']), ('\u{2cc2}', ['\u{2cc3}', '\0', + '\0']), ('\u{2cc4}', ['\u{2cc5}', '\0', '\0']), ('\u{2cc6}', ['\u{2cc7}', '\0', '\0']), + ('\u{2cc8}', ['\u{2cc9}', '\0', '\0']), ('\u{2cca}', ['\u{2ccb}', '\0', '\0']), ('\u{2ccc}', + ['\u{2ccd}', '\0', '\0']), ('\u{2cce}', ['\u{2ccf}', '\0', '\0']), ('\u{2cd0}', ['\u{2cd1}', + '\0', '\0']), ('\u{2cd2}', ['\u{2cd3}', '\0', '\0']), ('\u{2cd4}', ['\u{2cd5}', '\0', + '\0']), ('\u{2cd6}', ['\u{2cd7}', '\0', '\0']), ('\u{2cd8}', ['\u{2cd9}', '\0', '\0']), + ('\u{2cda}', ['\u{2cdb}', '\0', '\0']), ('\u{2cdc}', ['\u{2cdd}', '\0', '\0']), ('\u{2cde}', + ['\u{2cdf}', '\0', '\0']), ('\u{2ce0}', ['\u{2ce1}', '\0', '\0']), ('\u{2ce2}', ['\u{2ce3}', + '\0', '\0']), ('\u{2ceb}', ['\u{2cec}', '\0', '\0']), ('\u{2ced}', ['\u{2cee}', '\0', + '\0']), ('\u{2cf2}', ['\u{2cf3}', '\0', '\0']), ('\u{a640}', ['\u{a641}', '\0', '\0']), + ('\u{a642}', ['\u{a643}', '\0', '\0']), ('\u{a644}', ['\u{a645}', '\0', '\0']), ('\u{a646}', + ['\u{a647}', '\0', '\0']), ('\u{a648}', ['\u{a649}', '\0', '\0']), ('\u{a64a}', ['\u{a64b}', + '\0', '\0']), ('\u{a64c}', ['\u{a64d}', '\0', '\0']), ('\u{a64e}', ['\u{a64f}', '\0', + '\0']), ('\u{a650}', ['\u{a651}', '\0', '\0']), ('\u{a652}', ['\u{a653}', '\0', '\0']), + ('\u{a654}', ['\u{a655}', '\0', '\0']), ('\u{a656}', ['\u{a657}', '\0', '\0']), ('\u{a658}', + ['\u{a659}', '\0', '\0']), ('\u{a65a}', ['\u{a65b}', '\0', '\0']), ('\u{a65c}', ['\u{a65d}', + '\0', '\0']), ('\u{a65e}', ['\u{a65f}', '\0', '\0']), ('\u{a660}', ['\u{a661}', '\0', + '\0']), ('\u{a662}', ['\u{a663}', '\0', '\0']), ('\u{a664}', ['\u{a665}', '\0', '\0']), + ('\u{a666}', ['\u{a667}', '\0', '\0']), ('\u{a668}', ['\u{a669}', '\0', '\0']), ('\u{a66a}', + ['\u{a66b}', '\0', '\0']), ('\u{a66c}', ['\u{a66d}', '\0', '\0']), ('\u{a680}', ['\u{a681}', + '\0', '\0']), ('\u{a682}', ['\u{a683}', '\0', '\0']), ('\u{a684}', ['\u{a685}', '\0', + '\0']), ('\u{a686}', ['\u{a687}', '\0', '\0']), ('\u{a688}', ['\u{a689}', '\0', '\0']), + ('\u{a68a}', ['\u{a68b}', '\0', '\0']), ('\u{a68c}', ['\u{a68d}', '\0', '\0']), ('\u{a68e}', + ['\u{a68f}', '\0', '\0']), ('\u{a690}', ['\u{a691}', '\0', '\0']), ('\u{a692}', ['\u{a693}', + '\0', '\0']), ('\u{a694}', ['\u{a695}', '\0', '\0']), ('\u{a696}', ['\u{a697}', '\0', + '\0']), ('\u{a698}', ['\u{a699}', '\0', '\0']), ('\u{a69a}', ['\u{a69b}', '\0', '\0']), + ('\u{a722}', ['\u{a723}', '\0', '\0']), ('\u{a724}', ['\u{a725}', '\0', '\0']), ('\u{a726}', + ['\u{a727}', '\0', '\0']), ('\u{a728}', ['\u{a729}', '\0', '\0']), ('\u{a72a}', ['\u{a72b}', + '\0', '\0']), ('\u{a72c}', ['\u{a72d}', '\0', '\0']), ('\u{a72e}', ['\u{a72f}', '\0', + '\0']), ('\u{a732}', ['\u{a733}', '\0', '\0']), ('\u{a734}', ['\u{a735}', '\0', '\0']), + ('\u{a736}', ['\u{a737}', '\0', '\0']), ('\u{a738}', ['\u{a739}', '\0', '\0']), ('\u{a73a}', + ['\u{a73b}', '\0', '\0']), ('\u{a73c}', ['\u{a73d}', '\0', '\0']), ('\u{a73e}', ['\u{a73f}', + '\0', '\0']), ('\u{a740}', ['\u{a741}', '\0', '\0']), ('\u{a742}', ['\u{a743}', '\0', + '\0']), ('\u{a744}', ['\u{a745}', '\0', '\0']), ('\u{a746}', ['\u{a747}', '\0', '\0']), + ('\u{a748}', ['\u{a749}', '\0', '\0']), ('\u{a74a}', ['\u{a74b}', '\0', '\0']), ('\u{a74c}', + ['\u{a74d}', '\0', '\0']), ('\u{a74e}', ['\u{a74f}', '\0', '\0']), ('\u{a750}', ['\u{a751}', + '\0', '\0']), ('\u{a752}', ['\u{a753}', '\0', '\0']), ('\u{a754}', ['\u{a755}', '\0', + '\0']), ('\u{a756}', ['\u{a757}', '\0', '\0']), ('\u{a758}', ['\u{a759}', '\0', '\0']), + ('\u{a75a}', ['\u{a75b}', '\0', '\0']), ('\u{a75c}', ['\u{a75d}', '\0', '\0']), ('\u{a75e}', + ['\u{a75f}', '\0', '\0']), ('\u{a760}', ['\u{a761}', '\0', '\0']), ('\u{a762}', ['\u{a763}', + '\0', '\0']), ('\u{a764}', ['\u{a765}', '\0', '\0']), ('\u{a766}', ['\u{a767}', '\0', + '\0']), ('\u{a768}', ['\u{a769}', '\0', '\0']), ('\u{a76a}', ['\u{a76b}', '\0', '\0']), + ('\u{a76c}', ['\u{a76d}', '\0', '\0']), ('\u{a76e}', ['\u{a76f}', '\0', '\0']), ('\u{a779}', + ['\u{a77a}', '\0', '\0']), ('\u{a77b}', ['\u{a77c}', '\0', '\0']), ('\u{a77d}', ['\u{1d79}', + '\0', '\0']), ('\u{a77e}', ['\u{a77f}', '\0', '\0']), ('\u{a780}', ['\u{a781}', '\0', + '\0']), ('\u{a782}', ['\u{a783}', '\0', '\0']), ('\u{a784}', ['\u{a785}', '\0', '\0']), + ('\u{a786}', ['\u{a787}', '\0', '\0']), ('\u{a78b}', ['\u{a78c}', '\0', '\0']), ('\u{a78d}', + ['\u{265}', '\0', '\0']), ('\u{a790}', ['\u{a791}', '\0', '\0']), ('\u{a792}', ['\u{a793}', + '\0', '\0']), ('\u{a796}', ['\u{a797}', '\0', '\0']), ('\u{a798}', ['\u{a799}', '\0', + '\0']), ('\u{a79a}', ['\u{a79b}', '\0', '\0']), ('\u{a79c}', ['\u{a79d}', '\0', '\0']), + ('\u{a79e}', ['\u{a79f}', '\0', '\0']), ('\u{a7a0}', ['\u{a7a1}', '\0', '\0']), ('\u{a7a2}', + ['\u{a7a3}', '\0', '\0']), ('\u{a7a4}', ['\u{a7a5}', '\0', '\0']), ('\u{a7a6}', ['\u{a7a7}', + '\0', '\0']), ('\u{a7a8}', ['\u{a7a9}', '\0', '\0']), ('\u{a7aa}', ['\u{266}', '\0', '\0']), + ('\u{a7ab}', ['\u{25c}', '\0', '\0']), ('\u{a7ac}', ['\u{261}', '\0', '\0']), ('\u{a7ad}', + ['\u{26c}', '\0', '\0']), ('\u{a7ae}', ['\u{26a}', '\0', '\0']), ('\u{a7b0}', ['\u{29e}', + '\0', '\0']), ('\u{a7b1}', ['\u{287}', '\0', '\0']), ('\u{a7b2}', ['\u{29d}', '\0', '\0']), + ('\u{a7b3}', ['\u{ab53}', '\0', '\0']), ('\u{a7b4}', ['\u{a7b5}', '\0', '\0']), ('\u{a7b6}', + ['\u{a7b7}', '\0', '\0']), ('\u{a7b8}', ['\u{a7b9}', '\0', '\0']), ('\u{ff21}', ['\u{ff41}', + '\0', '\0']), ('\u{ff22}', ['\u{ff42}', '\0', '\0']), ('\u{ff23}', ['\u{ff43}', '\0', + '\0']), ('\u{ff24}', ['\u{ff44}', '\0', '\0']), ('\u{ff25}', ['\u{ff45}', '\0', '\0']), + ('\u{ff26}', ['\u{ff46}', '\0', '\0']), ('\u{ff27}', ['\u{ff47}', '\0', '\0']), ('\u{ff28}', + ['\u{ff48}', '\0', '\0']), ('\u{ff29}', ['\u{ff49}', '\0', '\0']), ('\u{ff2a}', ['\u{ff4a}', + '\0', '\0']), ('\u{ff2b}', ['\u{ff4b}', '\0', '\0']), ('\u{ff2c}', ['\u{ff4c}', '\0', + '\0']), ('\u{ff2d}', ['\u{ff4d}', '\0', '\0']), ('\u{ff2e}', ['\u{ff4e}', '\0', '\0']), + ('\u{ff2f}', ['\u{ff4f}', '\0', '\0']), ('\u{ff30}', ['\u{ff50}', '\0', '\0']), ('\u{ff31}', + ['\u{ff51}', '\0', '\0']), ('\u{ff32}', ['\u{ff52}', '\0', '\0']), ('\u{ff33}', ['\u{ff53}', + '\0', '\0']), ('\u{ff34}', ['\u{ff54}', '\0', '\0']), ('\u{ff35}', ['\u{ff55}', '\0', + '\0']), ('\u{ff36}', ['\u{ff56}', '\0', '\0']), ('\u{ff37}', ['\u{ff57}', '\0', '\0']), + ('\u{ff38}', ['\u{ff58}', '\0', '\0']), ('\u{ff39}', ['\u{ff59}', '\0', '\0']), ('\u{ff3a}', + ['\u{ff5a}', '\0', '\0']), ('\u{10400}', ['\u{10428}', '\0', '\0']), ('\u{10401}', + ['\u{10429}', '\0', '\0']), ('\u{10402}', ['\u{1042a}', '\0', '\0']), ('\u{10403}', + ['\u{1042b}', '\0', '\0']), ('\u{10404}', ['\u{1042c}', '\0', '\0']), ('\u{10405}', + ['\u{1042d}', '\0', '\0']), ('\u{10406}', ['\u{1042e}', '\0', '\0']), ('\u{10407}', + ['\u{1042f}', '\0', '\0']), ('\u{10408}', ['\u{10430}', '\0', '\0']), ('\u{10409}', + ['\u{10431}', '\0', '\0']), ('\u{1040a}', ['\u{10432}', '\0', '\0']), ('\u{1040b}', + ['\u{10433}', '\0', '\0']), ('\u{1040c}', ['\u{10434}', '\0', '\0']), ('\u{1040d}', + ['\u{10435}', '\0', '\0']), ('\u{1040e}', ['\u{10436}', '\0', '\0']), ('\u{1040f}', + ['\u{10437}', '\0', '\0']), ('\u{10410}', ['\u{10438}', '\0', '\0']), ('\u{10411}', + ['\u{10439}', '\0', '\0']), ('\u{10412}', ['\u{1043a}', '\0', '\0']), ('\u{10413}', + ['\u{1043b}', '\0', '\0']), ('\u{10414}', ['\u{1043c}', '\0', '\0']), ('\u{10415}', + ['\u{1043d}', '\0', '\0']), ('\u{10416}', ['\u{1043e}', '\0', '\0']), ('\u{10417}', + ['\u{1043f}', '\0', '\0']), ('\u{10418}', ['\u{10440}', '\0', '\0']), ('\u{10419}', + ['\u{10441}', '\0', '\0']), ('\u{1041a}', ['\u{10442}', '\0', '\0']), ('\u{1041b}', + ['\u{10443}', '\0', '\0']), ('\u{1041c}', ['\u{10444}', '\0', '\0']), ('\u{1041d}', + ['\u{10445}', '\0', '\0']), ('\u{1041e}', ['\u{10446}', '\0', '\0']), ('\u{1041f}', + ['\u{10447}', '\0', '\0']), ('\u{10420}', ['\u{10448}', '\0', '\0']), ('\u{10421}', + ['\u{10449}', '\0', '\0']), ('\u{10422}', ['\u{1044a}', '\0', '\0']), ('\u{10423}', + ['\u{1044b}', '\0', '\0']), ('\u{10424}', ['\u{1044c}', '\0', '\0']), ('\u{10425}', + ['\u{1044d}', '\0', '\0']), ('\u{10426}', ['\u{1044e}', '\0', '\0']), ('\u{10427}', + ['\u{1044f}', '\0', '\0']), ('\u{104b0}', ['\u{104d8}', '\0', '\0']), ('\u{104b1}', + ['\u{104d9}', '\0', '\0']), ('\u{104b2}', ['\u{104da}', '\0', '\0']), ('\u{104b3}', + ['\u{104db}', '\0', '\0']), ('\u{104b4}', ['\u{104dc}', '\0', '\0']), ('\u{104b5}', + ['\u{104dd}', '\0', '\0']), ('\u{104b6}', ['\u{104de}', '\0', '\0']), ('\u{104b7}', + ['\u{104df}', '\0', '\0']), ('\u{104b8}', ['\u{104e0}', '\0', '\0']), ('\u{104b9}', + ['\u{104e1}', '\0', '\0']), ('\u{104ba}', ['\u{104e2}', '\0', '\0']), ('\u{104bb}', + ['\u{104e3}', '\0', '\0']), ('\u{104bc}', ['\u{104e4}', '\0', '\0']), ('\u{104bd}', + ['\u{104e5}', '\0', '\0']), ('\u{104be}', ['\u{104e6}', '\0', '\0']), ('\u{104bf}', + ['\u{104e7}', '\0', '\0']), ('\u{104c0}', ['\u{104e8}', '\0', '\0']), ('\u{104c1}', + ['\u{104e9}', '\0', '\0']), ('\u{104c2}', ['\u{104ea}', '\0', '\0']), ('\u{104c3}', + ['\u{104eb}', '\0', '\0']), ('\u{104c4}', ['\u{104ec}', '\0', '\0']), ('\u{104c5}', + ['\u{104ed}', '\0', '\0']), ('\u{104c6}', ['\u{104ee}', '\0', '\0']), ('\u{104c7}', + ['\u{104ef}', '\0', '\0']), ('\u{104c8}', ['\u{104f0}', '\0', '\0']), ('\u{104c9}', + ['\u{104f1}', '\0', '\0']), ('\u{104ca}', ['\u{104f2}', '\0', '\0']), ('\u{104cb}', + ['\u{104f3}', '\0', '\0']), ('\u{104cc}', ['\u{104f4}', '\0', '\0']), ('\u{104cd}', + ['\u{104f5}', '\0', '\0']), ('\u{104ce}', ['\u{104f6}', '\0', '\0']), ('\u{104cf}', + ['\u{104f7}', '\0', '\0']), ('\u{104d0}', ['\u{104f8}', '\0', '\0']), ('\u{104d1}', + ['\u{104f9}', '\0', '\0']), ('\u{104d2}', ['\u{104fa}', '\0', '\0']), ('\u{104d3}', + ['\u{104fb}', '\0', '\0']), ('\u{10c80}', ['\u{10cc0}', '\0', '\0']), ('\u{10c81}', + ['\u{10cc1}', '\0', '\0']), ('\u{10c82}', ['\u{10cc2}', '\0', '\0']), ('\u{10c83}', + ['\u{10cc3}', '\0', '\0']), ('\u{10c84}', ['\u{10cc4}', '\0', '\0']), ('\u{10c85}', + ['\u{10cc5}', '\0', '\0']), ('\u{10c86}', ['\u{10cc6}', '\0', '\0']), ('\u{10c87}', + ['\u{10cc7}', '\0', '\0']), ('\u{10c88}', ['\u{10cc8}', '\0', '\0']), ('\u{10c89}', + ['\u{10cc9}', '\0', '\0']), ('\u{10c8a}', ['\u{10cca}', '\0', '\0']), ('\u{10c8b}', + ['\u{10ccb}', '\0', '\0']), ('\u{10c8c}', ['\u{10ccc}', '\0', '\0']), ('\u{10c8d}', + ['\u{10ccd}', '\0', '\0']), ('\u{10c8e}', ['\u{10cce}', '\0', '\0']), ('\u{10c8f}', + ['\u{10ccf}', '\0', '\0']), ('\u{10c90}', ['\u{10cd0}', '\0', '\0']), ('\u{10c91}', + ['\u{10cd1}', '\0', '\0']), ('\u{10c92}', ['\u{10cd2}', '\0', '\0']), ('\u{10c93}', + ['\u{10cd3}', '\0', '\0']), ('\u{10c94}', ['\u{10cd4}', '\0', '\0']), ('\u{10c95}', + ['\u{10cd5}', '\0', '\0']), ('\u{10c96}', ['\u{10cd6}', '\0', '\0']), ('\u{10c97}', + ['\u{10cd7}', '\0', '\0']), ('\u{10c98}', ['\u{10cd8}', '\0', '\0']), ('\u{10c99}', + ['\u{10cd9}', '\0', '\0']), ('\u{10c9a}', ['\u{10cda}', '\0', '\0']), ('\u{10c9b}', + ['\u{10cdb}', '\0', '\0']), ('\u{10c9c}', ['\u{10cdc}', '\0', '\0']), ('\u{10c9d}', + ['\u{10cdd}', '\0', '\0']), ('\u{10c9e}', ['\u{10cde}', '\0', '\0']), ('\u{10c9f}', + ['\u{10cdf}', '\0', '\0']), ('\u{10ca0}', ['\u{10ce0}', '\0', '\0']), ('\u{10ca1}', + ['\u{10ce1}', '\0', '\0']), ('\u{10ca2}', ['\u{10ce2}', '\0', '\0']), ('\u{10ca3}', + ['\u{10ce3}', '\0', '\0']), ('\u{10ca4}', ['\u{10ce4}', '\0', '\0']), ('\u{10ca5}', + ['\u{10ce5}', '\0', '\0']), ('\u{10ca6}', ['\u{10ce6}', '\0', '\0']), ('\u{10ca7}', + ['\u{10ce7}', '\0', '\0']), ('\u{10ca8}', ['\u{10ce8}', '\0', '\0']), ('\u{10ca9}', + ['\u{10ce9}', '\0', '\0']), ('\u{10caa}', ['\u{10cea}', '\0', '\0']), ('\u{10cab}', + ['\u{10ceb}', '\0', '\0']), ('\u{10cac}', ['\u{10cec}', '\0', '\0']), ('\u{10cad}', + ['\u{10ced}', '\0', '\0']), ('\u{10cae}', ['\u{10cee}', '\0', '\0']), ('\u{10caf}', + ['\u{10cef}', '\0', '\0']), ('\u{10cb0}', ['\u{10cf0}', '\0', '\0']), ('\u{10cb1}', + ['\u{10cf1}', '\0', '\0']), ('\u{10cb2}', ['\u{10cf2}', '\0', '\0']), ('\u{118a0}', + ['\u{118c0}', '\0', '\0']), ('\u{118a1}', ['\u{118c1}', '\0', '\0']), ('\u{118a2}', + ['\u{118c2}', '\0', '\0']), ('\u{118a3}', ['\u{118c3}', '\0', '\0']), ('\u{118a4}', + ['\u{118c4}', '\0', '\0']), ('\u{118a5}', ['\u{118c5}', '\0', '\0']), ('\u{118a6}', + ['\u{118c6}', '\0', '\0']), ('\u{118a7}', ['\u{118c7}', '\0', '\0']), ('\u{118a8}', + ['\u{118c8}', '\0', '\0']), ('\u{118a9}', ['\u{118c9}', '\0', '\0']), ('\u{118aa}', + ['\u{118ca}', '\0', '\0']), ('\u{118ab}', ['\u{118cb}', '\0', '\0']), ('\u{118ac}', + ['\u{118cc}', '\0', '\0']), ('\u{118ad}', ['\u{118cd}', '\0', '\0']), ('\u{118ae}', + ['\u{118ce}', '\0', '\0']), ('\u{118af}', ['\u{118cf}', '\0', '\0']), ('\u{118b0}', + ['\u{118d0}', '\0', '\0']), ('\u{118b1}', ['\u{118d1}', '\0', '\0']), ('\u{118b2}', + ['\u{118d2}', '\0', '\0']), ('\u{118b3}', ['\u{118d3}', '\0', '\0']), ('\u{118b4}', + ['\u{118d4}', '\0', '\0']), ('\u{118b5}', ['\u{118d5}', '\0', '\0']), ('\u{118b6}', + ['\u{118d6}', '\0', '\0']), ('\u{118b7}', ['\u{118d7}', '\0', '\0']), ('\u{118b8}', + ['\u{118d8}', '\0', '\0']), ('\u{118b9}', ['\u{118d9}', '\0', '\0']), ('\u{118ba}', + ['\u{118da}', '\0', '\0']), ('\u{118bb}', ['\u{118db}', '\0', '\0']), ('\u{118bc}', + ['\u{118dc}', '\0', '\0']), ('\u{118bd}', ['\u{118dd}', '\0', '\0']), ('\u{118be}', + ['\u{118de}', '\0', '\0']), ('\u{118bf}', ['\u{118df}', '\0', '\0']), ('\u{16e40}', + ['\u{16e60}', '\0', '\0']), ('\u{16e41}', ['\u{16e61}', '\0', '\0']), ('\u{16e42}', + ['\u{16e62}', '\0', '\0']), ('\u{16e43}', ['\u{16e63}', '\0', '\0']), ('\u{16e44}', + ['\u{16e64}', '\0', '\0']), ('\u{16e45}', ['\u{16e65}', '\0', '\0']), ('\u{16e46}', + ['\u{16e66}', '\0', '\0']), ('\u{16e47}', ['\u{16e67}', '\0', '\0']), ('\u{16e48}', + ['\u{16e68}', '\0', '\0']), ('\u{16e49}', ['\u{16e69}', '\0', '\0']), ('\u{16e4a}', + ['\u{16e6a}', '\0', '\0']), ('\u{16e4b}', ['\u{16e6b}', '\0', '\0']), ('\u{16e4c}', + ['\u{16e6c}', '\0', '\0']), ('\u{16e4d}', ['\u{16e6d}', '\0', '\0']), ('\u{16e4e}', + ['\u{16e6e}', '\0', '\0']), ('\u{16e4f}', ['\u{16e6f}', '\0', '\0']), ('\u{16e50}', + ['\u{16e70}', '\0', '\0']), ('\u{16e51}', ['\u{16e71}', '\0', '\0']), ('\u{16e52}', + ['\u{16e72}', '\0', '\0']), ('\u{16e53}', ['\u{16e73}', '\0', '\0']), ('\u{16e54}', + ['\u{16e74}', '\0', '\0']), ('\u{16e55}', ['\u{16e75}', '\0', '\0']), ('\u{16e56}', + ['\u{16e76}', '\0', '\0']), ('\u{16e57}', ['\u{16e77}', '\0', '\0']), ('\u{16e58}', + ['\u{16e78}', '\0', '\0']), ('\u{16e59}', ['\u{16e79}', '\0', '\0']), ('\u{16e5a}', + ['\u{16e7a}', '\0', '\0']), ('\u{16e5b}', ['\u{16e7b}', '\0', '\0']), ('\u{16e5c}', + ['\u{16e7c}', '\0', '\0']), ('\u{16e5d}', ['\u{16e7d}', '\0', '\0']), ('\u{16e5e}', + ['\u{16e7e}', '\0', '\0']), ('\u{16e5f}', ['\u{16e7f}', '\0', '\0']), ('\u{1e900}', + ['\u{1e922}', '\0', '\0']), ('\u{1e901}', ['\u{1e923}', '\0', '\0']), ('\u{1e902}', + ['\u{1e924}', '\0', '\0']), ('\u{1e903}', ['\u{1e925}', '\0', '\0']), ('\u{1e904}', + ['\u{1e926}', '\0', '\0']), ('\u{1e905}', ['\u{1e927}', '\0', '\0']), ('\u{1e906}', + ['\u{1e928}', '\0', '\0']), ('\u{1e907}', ['\u{1e929}', '\0', '\0']), ('\u{1e908}', + ['\u{1e92a}', '\0', '\0']), ('\u{1e909}', ['\u{1e92b}', '\0', '\0']), ('\u{1e90a}', + ['\u{1e92c}', '\0', '\0']), ('\u{1e90b}', ['\u{1e92d}', '\0', '\0']), ('\u{1e90c}', + ['\u{1e92e}', '\0', '\0']), ('\u{1e90d}', ['\u{1e92f}', '\0', '\0']), ('\u{1e90e}', + ['\u{1e930}', '\0', '\0']), ('\u{1e90f}', ['\u{1e931}', '\0', '\0']), ('\u{1e910}', + ['\u{1e932}', '\0', '\0']), ('\u{1e911}', ['\u{1e933}', '\0', '\0']), ('\u{1e912}', + ['\u{1e934}', '\0', '\0']), ('\u{1e913}', ['\u{1e935}', '\0', '\0']), ('\u{1e914}', + ['\u{1e936}', '\0', '\0']), ('\u{1e915}', ['\u{1e937}', '\0', '\0']), ('\u{1e916}', + ['\u{1e938}', '\0', '\0']), ('\u{1e917}', ['\u{1e939}', '\0', '\0']), ('\u{1e918}', + ['\u{1e93a}', '\0', '\0']), ('\u{1e919}', ['\u{1e93b}', '\0', '\0']), ('\u{1e91a}', + ['\u{1e93c}', '\0', '\0']), ('\u{1e91b}', ['\u{1e93d}', '\0', '\0']), ('\u{1e91c}', + ['\u{1e93e}', '\0', '\0']), ('\u{1e91d}', ['\u{1e93f}', '\0', '\0']), ('\u{1e91e}', + ['\u{1e940}', '\0', '\0']), ('\u{1e91f}', ['\u{1e941}', '\0', '\0']), ('\u{1e920}', + ['\u{1e942}', '\0', '\0']), ('\u{1e921}', ['\u{1e943}', '\0', '\0']) + ]; + + const to_uppercase_table: &[(char, [char; 3])] = &[ + ('\u{61}', ['\u{41}', '\0', '\0']), ('\u{62}', ['\u{42}', '\0', '\0']), ('\u{63}', + ['\u{43}', '\0', '\0']), ('\u{64}', ['\u{44}', '\0', '\0']), ('\u{65}', ['\u{45}', '\0', + '\0']), ('\u{66}', ['\u{46}', '\0', '\0']), ('\u{67}', ['\u{47}', '\0', '\0']), ('\u{68}', + ['\u{48}', '\0', '\0']), ('\u{69}', ['\u{49}', '\0', '\0']), ('\u{6a}', ['\u{4a}', '\0', + '\0']), ('\u{6b}', ['\u{4b}', '\0', '\0']), ('\u{6c}', ['\u{4c}', '\0', '\0']), ('\u{6d}', + ['\u{4d}', '\0', '\0']), ('\u{6e}', ['\u{4e}', '\0', '\0']), ('\u{6f}', ['\u{4f}', '\0', + '\0']), ('\u{70}', ['\u{50}', '\0', '\0']), ('\u{71}', ['\u{51}', '\0', '\0']), ('\u{72}', + ['\u{52}', '\0', '\0']), ('\u{73}', ['\u{53}', '\0', '\0']), ('\u{74}', ['\u{54}', '\0', + '\0']), ('\u{75}', ['\u{55}', '\0', '\0']), ('\u{76}', ['\u{56}', '\0', '\0']), ('\u{77}', + ['\u{57}', '\0', '\0']), ('\u{78}', ['\u{58}', '\0', '\0']), ('\u{79}', ['\u{59}', '\0', + '\0']), ('\u{7a}', ['\u{5a}', '\0', '\0']), ('\u{b5}', ['\u{39c}', '\0', '\0']), ('\u{df}', + ['\u{53}', '\u{53}', '\0']), ('\u{e0}', ['\u{c0}', '\0', '\0']), ('\u{e1}', ['\u{c1}', '\0', + '\0']), ('\u{e2}', ['\u{c2}', '\0', '\0']), ('\u{e3}', ['\u{c3}', '\0', '\0']), ('\u{e4}', + ['\u{c4}', '\0', '\0']), ('\u{e5}', ['\u{c5}', '\0', '\0']), ('\u{e6}', ['\u{c6}', '\0', + '\0']), ('\u{e7}', ['\u{c7}', '\0', '\0']), ('\u{e8}', ['\u{c8}', '\0', '\0']), ('\u{e9}', + ['\u{c9}', '\0', '\0']), ('\u{ea}', ['\u{ca}', '\0', '\0']), ('\u{eb}', ['\u{cb}', '\0', + '\0']), ('\u{ec}', ['\u{cc}', '\0', '\0']), ('\u{ed}', ['\u{cd}', '\0', '\0']), ('\u{ee}', + ['\u{ce}', '\0', '\0']), ('\u{ef}', ['\u{cf}', '\0', '\0']), ('\u{f0}', ['\u{d0}', '\0', + '\0']), ('\u{f1}', ['\u{d1}', '\0', '\0']), ('\u{f2}', ['\u{d2}', '\0', '\0']), ('\u{f3}', + ['\u{d3}', '\0', '\0']), ('\u{f4}', ['\u{d4}', '\0', '\0']), ('\u{f5}', ['\u{d5}', '\0', + '\0']), ('\u{f6}', ['\u{d6}', '\0', '\0']), ('\u{f8}', ['\u{d8}', '\0', '\0']), ('\u{f9}', + ['\u{d9}', '\0', '\0']), ('\u{fa}', ['\u{da}', '\0', '\0']), ('\u{fb}', ['\u{db}', '\0', + '\0']), ('\u{fc}', ['\u{dc}', '\0', '\0']), ('\u{fd}', ['\u{dd}', '\0', '\0']), ('\u{fe}', + ['\u{de}', '\0', '\0']), ('\u{ff}', ['\u{178}', '\0', '\0']), ('\u{101}', ['\u{100}', '\0', + '\0']), ('\u{103}', ['\u{102}', '\0', '\0']), ('\u{105}', ['\u{104}', '\0', '\0']), + ('\u{107}', ['\u{106}', '\0', '\0']), ('\u{109}', ['\u{108}', '\0', '\0']), ('\u{10b}', + ['\u{10a}', '\0', '\0']), ('\u{10d}', ['\u{10c}', '\0', '\0']), ('\u{10f}', ['\u{10e}', + '\0', '\0']), ('\u{111}', ['\u{110}', '\0', '\0']), ('\u{113}', ['\u{112}', '\0', '\0']), + ('\u{115}', ['\u{114}', '\0', '\0']), ('\u{117}', ['\u{116}', '\0', '\0']), ('\u{119}', + ['\u{118}', '\0', '\0']), ('\u{11b}', ['\u{11a}', '\0', '\0']), ('\u{11d}', ['\u{11c}', + '\0', '\0']), ('\u{11f}', ['\u{11e}', '\0', '\0']), ('\u{121}', ['\u{120}', '\0', '\0']), + ('\u{123}', ['\u{122}', '\0', '\0']), ('\u{125}', ['\u{124}', '\0', '\0']), ('\u{127}', + ['\u{126}', '\0', '\0']), ('\u{129}', ['\u{128}', '\0', '\0']), ('\u{12b}', ['\u{12a}', + '\0', '\0']), ('\u{12d}', ['\u{12c}', '\0', '\0']), ('\u{12f}', ['\u{12e}', '\0', '\0']), + ('\u{131}', ['\u{49}', '\0', '\0']), ('\u{133}', ['\u{132}', '\0', '\0']), ('\u{135}', + ['\u{134}', '\0', '\0']), ('\u{137}', ['\u{136}', '\0', '\0']), ('\u{13a}', ['\u{139}', + '\0', '\0']), ('\u{13c}', ['\u{13b}', '\0', '\0']), ('\u{13e}', ['\u{13d}', '\0', '\0']), + ('\u{140}', ['\u{13f}', '\0', '\0']), ('\u{142}', ['\u{141}', '\0', '\0']), ('\u{144}', + ['\u{143}', '\0', '\0']), ('\u{146}', ['\u{145}', '\0', '\0']), ('\u{148}', ['\u{147}', + '\0', '\0']), ('\u{149}', ['\u{2bc}', '\u{4e}', '\0']), ('\u{14b}', ['\u{14a}', '\0', + '\0']), ('\u{14d}', ['\u{14c}', '\0', '\0']), ('\u{14f}', ['\u{14e}', '\0', '\0']), + ('\u{151}', ['\u{150}', '\0', '\0']), ('\u{153}', ['\u{152}', '\0', '\0']), ('\u{155}', + ['\u{154}', '\0', '\0']), ('\u{157}', ['\u{156}', '\0', '\0']), ('\u{159}', ['\u{158}', + '\0', '\0']), ('\u{15b}', ['\u{15a}', '\0', '\0']), ('\u{15d}', ['\u{15c}', '\0', '\0']), + ('\u{15f}', ['\u{15e}', '\0', '\0']), ('\u{161}', ['\u{160}', '\0', '\0']), ('\u{163}', + ['\u{162}', '\0', '\0']), ('\u{165}', ['\u{164}', '\0', '\0']), ('\u{167}', ['\u{166}', + '\0', '\0']), ('\u{169}', ['\u{168}', '\0', '\0']), ('\u{16b}', ['\u{16a}', '\0', '\0']), + ('\u{16d}', ['\u{16c}', '\0', '\0']), ('\u{16f}', ['\u{16e}', '\0', '\0']), ('\u{171}', + ['\u{170}', '\0', '\0']), ('\u{173}', ['\u{172}', '\0', '\0']), ('\u{175}', ['\u{174}', + '\0', '\0']), ('\u{177}', ['\u{176}', '\0', '\0']), ('\u{17a}', ['\u{179}', '\0', '\0']), + ('\u{17c}', ['\u{17b}', '\0', '\0']), ('\u{17e}', ['\u{17d}', '\0', '\0']), ('\u{17f}', + ['\u{53}', '\0', '\0']), ('\u{180}', ['\u{243}', '\0', '\0']), ('\u{183}', ['\u{182}', '\0', + '\0']), ('\u{185}', ['\u{184}', '\0', '\0']), ('\u{188}', ['\u{187}', '\0', '\0']), + ('\u{18c}', ['\u{18b}', '\0', '\0']), ('\u{192}', ['\u{191}', '\0', '\0']), ('\u{195}', + ['\u{1f6}', '\0', '\0']), ('\u{199}', ['\u{198}', '\0', '\0']), ('\u{19a}', ['\u{23d}', + '\0', '\0']), ('\u{19e}', ['\u{220}', '\0', '\0']), ('\u{1a1}', ['\u{1a0}', '\0', '\0']), + ('\u{1a3}', ['\u{1a2}', '\0', '\0']), ('\u{1a5}', ['\u{1a4}', '\0', '\0']), ('\u{1a8}', + ['\u{1a7}', '\0', '\0']), ('\u{1ad}', ['\u{1ac}', '\0', '\0']), ('\u{1b0}', ['\u{1af}', + '\0', '\0']), ('\u{1b4}', ['\u{1b3}', '\0', '\0']), ('\u{1b6}', ['\u{1b5}', '\0', '\0']), + ('\u{1b9}', ['\u{1b8}', '\0', '\0']), ('\u{1bd}', ['\u{1bc}', '\0', '\0']), ('\u{1bf}', + ['\u{1f7}', '\0', '\0']), ('\u{1c5}', ['\u{1c4}', '\0', '\0']), ('\u{1c6}', ['\u{1c4}', + '\0', '\0']), ('\u{1c8}', ['\u{1c7}', '\0', '\0']), ('\u{1c9}', ['\u{1c7}', '\0', '\0']), + ('\u{1cb}', ['\u{1ca}', '\0', '\0']), ('\u{1cc}', ['\u{1ca}', '\0', '\0']), ('\u{1ce}', + ['\u{1cd}', '\0', '\0']), ('\u{1d0}', ['\u{1cf}', '\0', '\0']), ('\u{1d2}', ['\u{1d1}', + '\0', '\0']), ('\u{1d4}', ['\u{1d3}', '\0', '\0']), ('\u{1d6}', ['\u{1d5}', '\0', '\0']), + ('\u{1d8}', ['\u{1d7}', '\0', '\0']), ('\u{1da}', ['\u{1d9}', '\0', '\0']), ('\u{1dc}', + ['\u{1db}', '\0', '\0']), ('\u{1dd}', ['\u{18e}', '\0', '\0']), ('\u{1df}', ['\u{1de}', + '\0', '\0']), ('\u{1e1}', ['\u{1e0}', '\0', '\0']), ('\u{1e3}', ['\u{1e2}', '\0', '\0']), + ('\u{1e5}', ['\u{1e4}', '\0', '\0']), ('\u{1e7}', ['\u{1e6}', '\0', '\0']), ('\u{1e9}', + ['\u{1e8}', '\0', '\0']), ('\u{1eb}', ['\u{1ea}', '\0', '\0']), ('\u{1ed}', ['\u{1ec}', + '\0', '\0']), ('\u{1ef}', ['\u{1ee}', '\0', '\0']), ('\u{1f0}', ['\u{4a}', '\u{30c}', + '\0']), ('\u{1f2}', ['\u{1f1}', '\0', '\0']), ('\u{1f3}', ['\u{1f1}', '\0', '\0']), + ('\u{1f5}', ['\u{1f4}', '\0', '\0']), ('\u{1f9}', ['\u{1f8}', '\0', '\0']), ('\u{1fb}', + ['\u{1fa}', '\0', '\0']), ('\u{1fd}', ['\u{1fc}', '\0', '\0']), ('\u{1ff}', ['\u{1fe}', + '\0', '\0']), ('\u{201}', ['\u{200}', '\0', '\0']), ('\u{203}', ['\u{202}', '\0', '\0']), + ('\u{205}', ['\u{204}', '\0', '\0']), ('\u{207}', ['\u{206}', '\0', '\0']), ('\u{209}', + ['\u{208}', '\0', '\0']), ('\u{20b}', ['\u{20a}', '\0', '\0']), ('\u{20d}', ['\u{20c}', + '\0', '\0']), ('\u{20f}', ['\u{20e}', '\0', '\0']), ('\u{211}', ['\u{210}', '\0', '\0']), + ('\u{213}', ['\u{212}', '\0', '\0']), ('\u{215}', ['\u{214}', '\0', '\0']), ('\u{217}', + ['\u{216}', '\0', '\0']), ('\u{219}', ['\u{218}', '\0', '\0']), ('\u{21b}', ['\u{21a}', + '\0', '\0']), ('\u{21d}', ['\u{21c}', '\0', '\0']), ('\u{21f}', ['\u{21e}', '\0', '\0']), + ('\u{223}', ['\u{222}', '\0', '\0']), ('\u{225}', ['\u{224}', '\0', '\0']), ('\u{227}', + ['\u{226}', '\0', '\0']), ('\u{229}', ['\u{228}', '\0', '\0']), ('\u{22b}', ['\u{22a}', + '\0', '\0']), ('\u{22d}', ['\u{22c}', '\0', '\0']), ('\u{22f}', ['\u{22e}', '\0', '\0']), + ('\u{231}', ['\u{230}', '\0', '\0']), ('\u{233}', ['\u{232}', '\0', '\0']), ('\u{23c}', + ['\u{23b}', '\0', '\0']), ('\u{23f}', ['\u{2c7e}', '\0', '\0']), ('\u{240}', ['\u{2c7f}', + '\0', '\0']), ('\u{242}', ['\u{241}', '\0', '\0']), ('\u{247}', ['\u{246}', '\0', '\0']), + ('\u{249}', ['\u{248}', '\0', '\0']), ('\u{24b}', ['\u{24a}', '\0', '\0']), ('\u{24d}', + ['\u{24c}', '\0', '\0']), ('\u{24f}', ['\u{24e}', '\0', '\0']), ('\u{250}', ['\u{2c6f}', + '\0', '\0']), ('\u{251}', ['\u{2c6d}', '\0', '\0']), ('\u{252}', ['\u{2c70}', '\0', '\0']), + ('\u{253}', ['\u{181}', '\0', '\0']), ('\u{254}', ['\u{186}', '\0', '\0']), ('\u{256}', + ['\u{189}', '\0', '\0']), ('\u{257}', ['\u{18a}', '\0', '\0']), ('\u{259}', ['\u{18f}', + '\0', '\0']), ('\u{25b}', ['\u{190}', '\0', '\0']), ('\u{25c}', ['\u{a7ab}', '\0', '\0']), + ('\u{260}', ['\u{193}', '\0', '\0']), ('\u{261}', ['\u{a7ac}', '\0', '\0']), ('\u{263}', + ['\u{194}', '\0', '\0']), ('\u{265}', ['\u{a78d}', '\0', '\0']), ('\u{266}', ['\u{a7aa}', + '\0', '\0']), ('\u{268}', ['\u{197}', '\0', '\0']), ('\u{269}', ['\u{196}', '\0', '\0']), + ('\u{26a}', ['\u{a7ae}', '\0', '\0']), ('\u{26b}', ['\u{2c62}', '\0', '\0']), ('\u{26c}', + ['\u{a7ad}', '\0', '\0']), ('\u{26f}', ['\u{19c}', '\0', '\0']), ('\u{271}', ['\u{2c6e}', + '\0', '\0']), ('\u{272}', ['\u{19d}', '\0', '\0']), ('\u{275}', ['\u{19f}', '\0', '\0']), + ('\u{27d}', ['\u{2c64}', '\0', '\0']), ('\u{280}', ['\u{1a6}', '\0', '\0']), ('\u{283}', + ['\u{1a9}', '\0', '\0']), ('\u{287}', ['\u{a7b1}', '\0', '\0']), ('\u{288}', ['\u{1ae}', + '\0', '\0']), ('\u{289}', ['\u{244}', '\0', '\0']), ('\u{28a}', ['\u{1b1}', '\0', '\0']), + ('\u{28b}', ['\u{1b2}', '\0', '\0']), ('\u{28c}', ['\u{245}', '\0', '\0']), ('\u{292}', + ['\u{1b7}', '\0', '\0']), ('\u{29d}', ['\u{a7b2}', '\0', '\0']), ('\u{29e}', ['\u{a7b0}', + '\0', '\0']), ('\u{345}', ['\u{399}', '\0', '\0']), ('\u{371}', ['\u{370}', '\0', '\0']), + ('\u{373}', ['\u{372}', '\0', '\0']), ('\u{377}', ['\u{376}', '\0', '\0']), ('\u{37b}', + ['\u{3fd}', '\0', '\0']), ('\u{37c}', ['\u{3fe}', '\0', '\0']), ('\u{37d}', ['\u{3ff}', + '\0', '\0']), ('\u{390}', ['\u{399}', '\u{308}', '\u{301}']), ('\u{3ac}', ['\u{386}', '\0', + '\0']), ('\u{3ad}', ['\u{388}', '\0', '\0']), ('\u{3ae}', ['\u{389}', '\0', '\0']), + ('\u{3af}', ['\u{38a}', '\0', '\0']), ('\u{3b0}', ['\u{3a5}', '\u{308}', '\u{301}']), + ('\u{3b1}', ['\u{391}', '\0', '\0']), ('\u{3b2}', ['\u{392}', '\0', '\0']), ('\u{3b3}', + ['\u{393}', '\0', '\0']), ('\u{3b4}', ['\u{394}', '\0', '\0']), ('\u{3b5}', ['\u{395}', + '\0', '\0']), ('\u{3b6}', ['\u{396}', '\0', '\0']), ('\u{3b7}', ['\u{397}', '\0', '\0']), + ('\u{3b8}', ['\u{398}', '\0', '\0']), ('\u{3b9}', ['\u{399}', '\0', '\0']), ('\u{3ba}', + ['\u{39a}', '\0', '\0']), ('\u{3bb}', ['\u{39b}', '\0', '\0']), ('\u{3bc}', ['\u{39c}', + '\0', '\0']), ('\u{3bd}', ['\u{39d}', '\0', '\0']), ('\u{3be}', ['\u{39e}', '\0', '\0']), + ('\u{3bf}', ['\u{39f}', '\0', '\0']), ('\u{3c0}', ['\u{3a0}', '\0', '\0']), ('\u{3c1}', + ['\u{3a1}', '\0', '\0']), ('\u{3c2}', ['\u{3a3}', '\0', '\0']), ('\u{3c3}', ['\u{3a3}', + '\0', '\0']), ('\u{3c4}', ['\u{3a4}', '\0', '\0']), ('\u{3c5}', ['\u{3a5}', '\0', '\0']), + ('\u{3c6}', ['\u{3a6}', '\0', '\0']), ('\u{3c7}', ['\u{3a7}', '\0', '\0']), ('\u{3c8}', + ['\u{3a8}', '\0', '\0']), ('\u{3c9}', ['\u{3a9}', '\0', '\0']), ('\u{3ca}', ['\u{3aa}', + '\0', '\0']), ('\u{3cb}', ['\u{3ab}', '\0', '\0']), ('\u{3cc}', ['\u{38c}', '\0', '\0']), + ('\u{3cd}', ['\u{38e}', '\0', '\0']), ('\u{3ce}', ['\u{38f}', '\0', '\0']), ('\u{3d0}', + ['\u{392}', '\0', '\0']), ('\u{3d1}', ['\u{398}', '\0', '\0']), ('\u{3d5}', ['\u{3a6}', + '\0', '\0']), ('\u{3d6}', ['\u{3a0}', '\0', '\0']), ('\u{3d7}', ['\u{3cf}', '\0', '\0']), + ('\u{3d9}', ['\u{3d8}', '\0', '\0']), ('\u{3db}', ['\u{3da}', '\0', '\0']), ('\u{3dd}', + ['\u{3dc}', '\0', '\0']), ('\u{3df}', ['\u{3de}', '\0', '\0']), ('\u{3e1}', ['\u{3e0}', + '\0', '\0']), ('\u{3e3}', ['\u{3e2}', '\0', '\0']), ('\u{3e5}', ['\u{3e4}', '\0', '\0']), + ('\u{3e7}', ['\u{3e6}', '\0', '\0']), ('\u{3e9}', ['\u{3e8}', '\0', '\0']), ('\u{3eb}', + ['\u{3ea}', '\0', '\0']), ('\u{3ed}', ['\u{3ec}', '\0', '\0']), ('\u{3ef}', ['\u{3ee}', + '\0', '\0']), ('\u{3f0}', ['\u{39a}', '\0', '\0']), ('\u{3f1}', ['\u{3a1}', '\0', '\0']), + ('\u{3f2}', ['\u{3f9}', '\0', '\0']), ('\u{3f3}', ['\u{37f}', '\0', '\0']), ('\u{3f5}', + ['\u{395}', '\0', '\0']), ('\u{3f8}', ['\u{3f7}', '\0', '\0']), ('\u{3fb}', ['\u{3fa}', + '\0', '\0']), ('\u{430}', ['\u{410}', '\0', '\0']), ('\u{431}', ['\u{411}', '\0', '\0']), + ('\u{432}', ['\u{412}', '\0', '\0']), ('\u{433}', ['\u{413}', '\0', '\0']), ('\u{434}', + ['\u{414}', '\0', '\0']), ('\u{435}', ['\u{415}', '\0', '\0']), ('\u{436}', ['\u{416}', + '\0', '\0']), ('\u{437}', ['\u{417}', '\0', '\0']), ('\u{438}', ['\u{418}', '\0', '\0']), + ('\u{439}', ['\u{419}', '\0', '\0']), ('\u{43a}', ['\u{41a}', '\0', '\0']), ('\u{43b}', + ['\u{41b}', '\0', '\0']), ('\u{43c}', ['\u{41c}', '\0', '\0']), ('\u{43d}', ['\u{41d}', + '\0', '\0']), ('\u{43e}', ['\u{41e}', '\0', '\0']), ('\u{43f}', ['\u{41f}', '\0', '\0']), + ('\u{440}', ['\u{420}', '\0', '\0']), ('\u{441}', ['\u{421}', '\0', '\0']), ('\u{442}', + ['\u{422}', '\0', '\0']), ('\u{443}', ['\u{423}', '\0', '\0']), ('\u{444}', ['\u{424}', + '\0', '\0']), ('\u{445}', ['\u{425}', '\0', '\0']), ('\u{446}', ['\u{426}', '\0', '\0']), + ('\u{447}', ['\u{427}', '\0', '\0']), ('\u{448}', ['\u{428}', '\0', '\0']), ('\u{449}', + ['\u{429}', '\0', '\0']), ('\u{44a}', ['\u{42a}', '\0', '\0']), ('\u{44b}', ['\u{42b}', + '\0', '\0']), ('\u{44c}', ['\u{42c}', '\0', '\0']), ('\u{44d}', ['\u{42d}', '\0', '\0']), + ('\u{44e}', ['\u{42e}', '\0', '\0']), ('\u{44f}', ['\u{42f}', '\0', '\0']), ('\u{450}', + ['\u{400}', '\0', '\0']), ('\u{451}', ['\u{401}', '\0', '\0']), ('\u{452}', ['\u{402}', + '\0', '\0']), ('\u{453}', ['\u{403}', '\0', '\0']), ('\u{454}', ['\u{404}', '\0', '\0']), + ('\u{455}', ['\u{405}', '\0', '\0']), ('\u{456}', ['\u{406}', '\0', '\0']), ('\u{457}', + ['\u{407}', '\0', '\0']), ('\u{458}', ['\u{408}', '\0', '\0']), ('\u{459}', ['\u{409}', + '\0', '\0']), ('\u{45a}', ['\u{40a}', '\0', '\0']), ('\u{45b}', ['\u{40b}', '\0', '\0']), + ('\u{45c}', ['\u{40c}', '\0', '\0']), ('\u{45d}', ['\u{40d}', '\0', '\0']), ('\u{45e}', + ['\u{40e}', '\0', '\0']), ('\u{45f}', ['\u{40f}', '\0', '\0']), ('\u{461}', ['\u{460}', + '\0', '\0']), ('\u{463}', ['\u{462}', '\0', '\0']), ('\u{465}', ['\u{464}', '\0', '\0']), + ('\u{467}', ['\u{466}', '\0', '\0']), ('\u{469}', ['\u{468}', '\0', '\0']), ('\u{46b}', + ['\u{46a}', '\0', '\0']), ('\u{46d}', ['\u{46c}', '\0', '\0']), ('\u{46f}', ['\u{46e}', + '\0', '\0']), ('\u{471}', ['\u{470}', '\0', '\0']), ('\u{473}', ['\u{472}', '\0', '\0']), + ('\u{475}', ['\u{474}', '\0', '\0']), ('\u{477}', ['\u{476}', '\0', '\0']), ('\u{479}', + ['\u{478}', '\0', '\0']), ('\u{47b}', ['\u{47a}', '\0', '\0']), ('\u{47d}', ['\u{47c}', + '\0', '\0']), ('\u{47f}', ['\u{47e}', '\0', '\0']), ('\u{481}', ['\u{480}', '\0', '\0']), + ('\u{48b}', ['\u{48a}', '\0', '\0']), ('\u{48d}', ['\u{48c}', '\0', '\0']), ('\u{48f}', + ['\u{48e}', '\0', '\0']), ('\u{491}', ['\u{490}', '\0', '\0']), ('\u{493}', ['\u{492}', + '\0', '\0']), ('\u{495}', ['\u{494}', '\0', '\0']), ('\u{497}', ['\u{496}', '\0', '\0']), + ('\u{499}', ['\u{498}', '\0', '\0']), ('\u{49b}', ['\u{49a}', '\0', '\0']), ('\u{49d}', + ['\u{49c}', '\0', '\0']), ('\u{49f}', ['\u{49e}', '\0', '\0']), ('\u{4a1}', ['\u{4a0}', + '\0', '\0']), ('\u{4a3}', ['\u{4a2}', '\0', '\0']), ('\u{4a5}', ['\u{4a4}', '\0', '\0']), + ('\u{4a7}', ['\u{4a6}', '\0', '\0']), ('\u{4a9}', ['\u{4a8}', '\0', '\0']), ('\u{4ab}', + ['\u{4aa}', '\0', '\0']), ('\u{4ad}', ['\u{4ac}', '\0', '\0']), ('\u{4af}', ['\u{4ae}', + '\0', '\0']), ('\u{4b1}', ['\u{4b0}', '\0', '\0']), ('\u{4b3}', ['\u{4b2}', '\0', '\0']), + ('\u{4b5}', ['\u{4b4}', '\0', '\0']), ('\u{4b7}', ['\u{4b6}', '\0', '\0']), ('\u{4b9}', + ['\u{4b8}', '\0', '\0']), ('\u{4bb}', ['\u{4ba}', '\0', '\0']), ('\u{4bd}', ['\u{4bc}', + '\0', '\0']), ('\u{4bf}', ['\u{4be}', '\0', '\0']), ('\u{4c2}', ['\u{4c1}', '\0', '\0']), + ('\u{4c4}', ['\u{4c3}', '\0', '\0']), ('\u{4c6}', ['\u{4c5}', '\0', '\0']), ('\u{4c8}', + ['\u{4c7}', '\0', '\0']), ('\u{4ca}', ['\u{4c9}', '\0', '\0']), ('\u{4cc}', ['\u{4cb}', + '\0', '\0']), ('\u{4ce}', ['\u{4cd}', '\0', '\0']), ('\u{4cf}', ['\u{4c0}', '\0', '\0']), + ('\u{4d1}', ['\u{4d0}', '\0', '\0']), ('\u{4d3}', ['\u{4d2}', '\0', '\0']), ('\u{4d5}', + ['\u{4d4}', '\0', '\0']), ('\u{4d7}', ['\u{4d6}', '\0', '\0']), ('\u{4d9}', ['\u{4d8}', + '\0', '\0']), ('\u{4db}', ['\u{4da}', '\0', '\0']), ('\u{4dd}', ['\u{4dc}', '\0', '\0']), + ('\u{4df}', ['\u{4de}', '\0', '\0']), ('\u{4e1}', ['\u{4e0}', '\0', '\0']), ('\u{4e3}', + ['\u{4e2}', '\0', '\0']), ('\u{4e5}', ['\u{4e4}', '\0', '\0']), ('\u{4e7}', ['\u{4e6}', + '\0', '\0']), ('\u{4e9}', ['\u{4e8}', '\0', '\0']), ('\u{4eb}', ['\u{4ea}', '\0', '\0']), + ('\u{4ed}', ['\u{4ec}', '\0', '\0']), ('\u{4ef}', ['\u{4ee}', '\0', '\0']), ('\u{4f1}', + ['\u{4f0}', '\0', '\0']), ('\u{4f3}', ['\u{4f2}', '\0', '\0']), ('\u{4f5}', ['\u{4f4}', + '\0', '\0']), ('\u{4f7}', ['\u{4f6}', '\0', '\0']), ('\u{4f9}', ['\u{4f8}', '\0', '\0']), + ('\u{4fb}', ['\u{4fa}', '\0', '\0']), ('\u{4fd}', ['\u{4fc}', '\0', '\0']), ('\u{4ff}', + ['\u{4fe}', '\0', '\0']), ('\u{501}', ['\u{500}', '\0', '\0']), ('\u{503}', ['\u{502}', + '\0', '\0']), ('\u{505}', ['\u{504}', '\0', '\0']), ('\u{507}', ['\u{506}', '\0', '\0']), + ('\u{509}', ['\u{508}', '\0', '\0']), ('\u{50b}', ['\u{50a}', '\0', '\0']), ('\u{50d}', + ['\u{50c}', '\0', '\0']), ('\u{50f}', ['\u{50e}', '\0', '\0']), ('\u{511}', ['\u{510}', + '\0', '\0']), ('\u{513}', ['\u{512}', '\0', '\0']), ('\u{515}', ['\u{514}', '\0', '\0']), + ('\u{517}', ['\u{516}', '\0', '\0']), ('\u{519}', ['\u{518}', '\0', '\0']), ('\u{51b}', + ['\u{51a}', '\0', '\0']), ('\u{51d}', ['\u{51c}', '\0', '\0']), ('\u{51f}', ['\u{51e}', + '\0', '\0']), ('\u{521}', ['\u{520}', '\0', '\0']), ('\u{523}', ['\u{522}', '\0', '\0']), + ('\u{525}', ['\u{524}', '\0', '\0']), ('\u{527}', ['\u{526}', '\0', '\0']), ('\u{529}', + ['\u{528}', '\0', '\0']), ('\u{52b}', ['\u{52a}', '\0', '\0']), ('\u{52d}', ['\u{52c}', + '\0', '\0']), ('\u{52f}', ['\u{52e}', '\0', '\0']), ('\u{561}', ['\u{531}', '\0', '\0']), + ('\u{562}', ['\u{532}', '\0', '\0']), ('\u{563}', ['\u{533}', '\0', '\0']), ('\u{564}', + ['\u{534}', '\0', '\0']), ('\u{565}', ['\u{535}', '\0', '\0']), ('\u{566}', ['\u{536}', + '\0', '\0']), ('\u{567}', ['\u{537}', '\0', '\0']), ('\u{568}', ['\u{538}', '\0', '\0']), + ('\u{569}', ['\u{539}', '\0', '\0']), ('\u{56a}', ['\u{53a}', '\0', '\0']), ('\u{56b}', + ['\u{53b}', '\0', '\0']), ('\u{56c}', ['\u{53c}', '\0', '\0']), ('\u{56d}', ['\u{53d}', + '\0', '\0']), ('\u{56e}', ['\u{53e}', '\0', '\0']), ('\u{56f}', ['\u{53f}', '\0', '\0']), + ('\u{570}', ['\u{540}', '\0', '\0']), ('\u{571}', ['\u{541}', '\0', '\0']), ('\u{572}', + ['\u{542}', '\0', '\0']), ('\u{573}', ['\u{543}', '\0', '\0']), ('\u{574}', ['\u{544}', + '\0', '\0']), ('\u{575}', ['\u{545}', '\0', '\0']), ('\u{576}', ['\u{546}', '\0', '\0']), + ('\u{577}', ['\u{547}', '\0', '\0']), ('\u{578}', ['\u{548}', '\0', '\0']), ('\u{579}', + ['\u{549}', '\0', '\0']), ('\u{57a}', ['\u{54a}', '\0', '\0']), ('\u{57b}', ['\u{54b}', + '\0', '\0']), ('\u{57c}', ['\u{54c}', '\0', '\0']), ('\u{57d}', ['\u{54d}', '\0', '\0']), + ('\u{57e}', ['\u{54e}', '\0', '\0']), ('\u{57f}', ['\u{54f}', '\0', '\0']), ('\u{580}', + ['\u{550}', '\0', '\0']), ('\u{581}', ['\u{551}', '\0', '\0']), ('\u{582}', ['\u{552}', + '\0', '\0']), ('\u{583}', ['\u{553}', '\0', '\0']), ('\u{584}', ['\u{554}', '\0', '\0']), + ('\u{585}', ['\u{555}', '\0', '\0']), ('\u{586}', ['\u{556}', '\0', '\0']), ('\u{587}', + ['\u{535}', '\u{552}', '\0']), ('\u{10d0}', ['\u{1c90}', '\0', '\0']), ('\u{10d1}', + ['\u{1c91}', '\0', '\0']), ('\u{10d2}', ['\u{1c92}', '\0', '\0']), ('\u{10d3}', ['\u{1c93}', + '\0', '\0']), ('\u{10d4}', ['\u{1c94}', '\0', '\0']), ('\u{10d5}', ['\u{1c95}', '\0', + '\0']), ('\u{10d6}', ['\u{1c96}', '\0', '\0']), ('\u{10d7}', ['\u{1c97}', '\0', '\0']), + ('\u{10d8}', ['\u{1c98}', '\0', '\0']), ('\u{10d9}', ['\u{1c99}', '\0', '\0']), ('\u{10da}', + ['\u{1c9a}', '\0', '\0']), ('\u{10db}', ['\u{1c9b}', '\0', '\0']), ('\u{10dc}', ['\u{1c9c}', + '\0', '\0']), ('\u{10dd}', ['\u{1c9d}', '\0', '\0']), ('\u{10de}', ['\u{1c9e}', '\0', + '\0']), ('\u{10df}', ['\u{1c9f}', '\0', '\0']), ('\u{10e0}', ['\u{1ca0}', '\0', '\0']), + ('\u{10e1}', ['\u{1ca1}', '\0', '\0']), ('\u{10e2}', ['\u{1ca2}', '\0', '\0']), ('\u{10e3}', + ['\u{1ca3}', '\0', '\0']), ('\u{10e4}', ['\u{1ca4}', '\0', '\0']), ('\u{10e5}', ['\u{1ca5}', + '\0', '\0']), ('\u{10e6}', ['\u{1ca6}', '\0', '\0']), ('\u{10e7}', ['\u{1ca7}', '\0', + '\0']), ('\u{10e8}', ['\u{1ca8}', '\0', '\0']), ('\u{10e9}', ['\u{1ca9}', '\0', '\0']), + ('\u{10ea}', ['\u{1caa}', '\0', '\0']), ('\u{10eb}', ['\u{1cab}', '\0', '\0']), ('\u{10ec}', + ['\u{1cac}', '\0', '\0']), ('\u{10ed}', ['\u{1cad}', '\0', '\0']), ('\u{10ee}', ['\u{1cae}', + '\0', '\0']), ('\u{10ef}', ['\u{1caf}', '\0', '\0']), ('\u{10f0}', ['\u{1cb0}', '\0', + '\0']), ('\u{10f1}', ['\u{1cb1}', '\0', '\0']), ('\u{10f2}', ['\u{1cb2}', '\0', '\0']), + ('\u{10f3}', ['\u{1cb3}', '\0', '\0']), ('\u{10f4}', ['\u{1cb4}', '\0', '\0']), ('\u{10f5}', + ['\u{1cb5}', '\0', '\0']), ('\u{10f6}', ['\u{1cb6}', '\0', '\0']), ('\u{10f7}', ['\u{1cb7}', + '\0', '\0']), ('\u{10f8}', ['\u{1cb8}', '\0', '\0']), ('\u{10f9}', ['\u{1cb9}', '\0', + '\0']), ('\u{10fa}', ['\u{1cba}', '\0', '\0']), ('\u{10fd}', ['\u{1cbd}', '\0', '\0']), + ('\u{10fe}', ['\u{1cbe}', '\0', '\0']), ('\u{10ff}', ['\u{1cbf}', '\0', '\0']), ('\u{13f8}', + ['\u{13f0}', '\0', '\0']), ('\u{13f9}', ['\u{13f1}', '\0', '\0']), ('\u{13fa}', ['\u{13f2}', + '\0', '\0']), ('\u{13fb}', ['\u{13f3}', '\0', '\0']), ('\u{13fc}', ['\u{13f4}', '\0', + '\0']), ('\u{13fd}', ['\u{13f5}', '\0', '\0']), ('\u{1c80}', ['\u{412}', '\0', '\0']), + ('\u{1c81}', ['\u{414}', '\0', '\0']), ('\u{1c82}', ['\u{41e}', '\0', '\0']), ('\u{1c83}', + ['\u{421}', '\0', '\0']), ('\u{1c84}', ['\u{422}', '\0', '\0']), ('\u{1c85}', ['\u{422}', + '\0', '\0']), ('\u{1c86}', ['\u{42a}', '\0', '\0']), ('\u{1c87}', ['\u{462}', '\0', '\0']), + ('\u{1c88}', ['\u{a64a}', '\0', '\0']), ('\u{1d79}', ['\u{a77d}', '\0', '\0']), ('\u{1d7d}', + ['\u{2c63}', '\0', '\0']), ('\u{1e01}', ['\u{1e00}', '\0', '\0']), ('\u{1e03}', ['\u{1e02}', + '\0', '\0']), ('\u{1e05}', ['\u{1e04}', '\0', '\0']), ('\u{1e07}', ['\u{1e06}', '\0', + '\0']), ('\u{1e09}', ['\u{1e08}', '\0', '\0']), ('\u{1e0b}', ['\u{1e0a}', '\0', '\0']), + ('\u{1e0d}', ['\u{1e0c}', '\0', '\0']), ('\u{1e0f}', ['\u{1e0e}', '\0', '\0']), ('\u{1e11}', + ['\u{1e10}', '\0', '\0']), ('\u{1e13}', ['\u{1e12}', '\0', '\0']), ('\u{1e15}', ['\u{1e14}', + '\0', '\0']), ('\u{1e17}', ['\u{1e16}', '\0', '\0']), ('\u{1e19}', ['\u{1e18}', '\0', + '\0']), ('\u{1e1b}', ['\u{1e1a}', '\0', '\0']), ('\u{1e1d}', ['\u{1e1c}', '\0', '\0']), + ('\u{1e1f}', ['\u{1e1e}', '\0', '\0']), ('\u{1e21}', ['\u{1e20}', '\0', '\0']), ('\u{1e23}', + ['\u{1e22}', '\0', '\0']), ('\u{1e25}', ['\u{1e24}', '\0', '\0']), ('\u{1e27}', ['\u{1e26}', + '\0', '\0']), ('\u{1e29}', ['\u{1e28}', '\0', '\0']), ('\u{1e2b}', ['\u{1e2a}', '\0', + '\0']), ('\u{1e2d}', ['\u{1e2c}', '\0', '\0']), ('\u{1e2f}', ['\u{1e2e}', '\0', '\0']), + ('\u{1e31}', ['\u{1e30}', '\0', '\0']), ('\u{1e33}', ['\u{1e32}', '\0', '\0']), ('\u{1e35}', + ['\u{1e34}', '\0', '\0']), ('\u{1e37}', ['\u{1e36}', '\0', '\0']), ('\u{1e39}', ['\u{1e38}', + '\0', '\0']), ('\u{1e3b}', ['\u{1e3a}', '\0', '\0']), ('\u{1e3d}', ['\u{1e3c}', '\0', + '\0']), ('\u{1e3f}', ['\u{1e3e}', '\0', '\0']), ('\u{1e41}', ['\u{1e40}', '\0', '\0']), + ('\u{1e43}', ['\u{1e42}', '\0', '\0']), ('\u{1e45}', ['\u{1e44}', '\0', '\0']), ('\u{1e47}', + ['\u{1e46}', '\0', '\0']), ('\u{1e49}', ['\u{1e48}', '\0', '\0']), ('\u{1e4b}', ['\u{1e4a}', + '\0', '\0']), ('\u{1e4d}', ['\u{1e4c}', '\0', '\0']), ('\u{1e4f}', ['\u{1e4e}', '\0', + '\0']), ('\u{1e51}', ['\u{1e50}', '\0', '\0']), ('\u{1e53}', ['\u{1e52}', '\0', '\0']), + ('\u{1e55}', ['\u{1e54}', '\0', '\0']), ('\u{1e57}', ['\u{1e56}', '\0', '\0']), ('\u{1e59}', + ['\u{1e58}', '\0', '\0']), ('\u{1e5b}', ['\u{1e5a}', '\0', '\0']), ('\u{1e5d}', ['\u{1e5c}', + '\0', '\0']), ('\u{1e5f}', ['\u{1e5e}', '\0', '\0']), ('\u{1e61}', ['\u{1e60}', '\0', + '\0']), ('\u{1e63}', ['\u{1e62}', '\0', '\0']), ('\u{1e65}', ['\u{1e64}', '\0', '\0']), + ('\u{1e67}', ['\u{1e66}', '\0', '\0']), ('\u{1e69}', ['\u{1e68}', '\0', '\0']), ('\u{1e6b}', + ['\u{1e6a}', '\0', '\0']), ('\u{1e6d}', ['\u{1e6c}', '\0', '\0']), ('\u{1e6f}', ['\u{1e6e}', + '\0', '\0']), ('\u{1e71}', ['\u{1e70}', '\0', '\0']), ('\u{1e73}', ['\u{1e72}', '\0', + '\0']), ('\u{1e75}', ['\u{1e74}', '\0', '\0']), ('\u{1e77}', ['\u{1e76}', '\0', '\0']), + ('\u{1e79}', ['\u{1e78}', '\0', '\0']), ('\u{1e7b}', ['\u{1e7a}', '\0', '\0']), ('\u{1e7d}', + ['\u{1e7c}', '\0', '\0']), ('\u{1e7f}', ['\u{1e7e}', '\0', '\0']), ('\u{1e81}', ['\u{1e80}', + '\0', '\0']), ('\u{1e83}', ['\u{1e82}', '\0', '\0']), ('\u{1e85}', ['\u{1e84}', '\0', + '\0']), ('\u{1e87}', ['\u{1e86}', '\0', '\0']), ('\u{1e89}', ['\u{1e88}', '\0', '\0']), + ('\u{1e8b}', ['\u{1e8a}', '\0', '\0']), ('\u{1e8d}', ['\u{1e8c}', '\0', '\0']), ('\u{1e8f}', + ['\u{1e8e}', '\0', '\0']), ('\u{1e91}', ['\u{1e90}', '\0', '\0']), ('\u{1e93}', ['\u{1e92}', + '\0', '\0']), ('\u{1e95}', ['\u{1e94}', '\0', '\0']), ('\u{1e96}', ['\u{48}', '\u{331}', + '\0']), ('\u{1e97}', ['\u{54}', '\u{308}', '\0']), ('\u{1e98}', ['\u{57}', '\u{30a}', + '\0']), ('\u{1e99}', ['\u{59}', '\u{30a}', '\0']), ('\u{1e9a}', ['\u{41}', '\u{2be}', + '\0']), ('\u{1e9b}', ['\u{1e60}', '\0', '\0']), ('\u{1ea1}', ['\u{1ea0}', '\0', '\0']), + ('\u{1ea3}', ['\u{1ea2}', '\0', '\0']), ('\u{1ea5}', ['\u{1ea4}', '\0', '\0']), ('\u{1ea7}', + ['\u{1ea6}', '\0', '\0']), ('\u{1ea9}', ['\u{1ea8}', '\0', '\0']), ('\u{1eab}', ['\u{1eaa}', + '\0', '\0']), ('\u{1ead}', ['\u{1eac}', '\0', '\0']), ('\u{1eaf}', ['\u{1eae}', '\0', + '\0']), ('\u{1eb1}', ['\u{1eb0}', '\0', '\0']), ('\u{1eb3}', ['\u{1eb2}', '\0', '\0']), + ('\u{1eb5}', ['\u{1eb4}', '\0', '\0']), ('\u{1eb7}', ['\u{1eb6}', '\0', '\0']), ('\u{1eb9}', + ['\u{1eb8}', '\0', '\0']), ('\u{1ebb}', ['\u{1eba}', '\0', '\0']), ('\u{1ebd}', ['\u{1ebc}', + '\0', '\0']), ('\u{1ebf}', ['\u{1ebe}', '\0', '\0']), ('\u{1ec1}', ['\u{1ec0}', '\0', + '\0']), ('\u{1ec3}', ['\u{1ec2}', '\0', '\0']), ('\u{1ec5}', ['\u{1ec4}', '\0', '\0']), + ('\u{1ec7}', ['\u{1ec6}', '\0', '\0']), ('\u{1ec9}', ['\u{1ec8}', '\0', '\0']), ('\u{1ecb}', + ['\u{1eca}', '\0', '\0']), ('\u{1ecd}', ['\u{1ecc}', '\0', '\0']), ('\u{1ecf}', ['\u{1ece}', + '\0', '\0']), ('\u{1ed1}', ['\u{1ed0}', '\0', '\0']), ('\u{1ed3}', ['\u{1ed2}', '\0', + '\0']), ('\u{1ed5}', ['\u{1ed4}', '\0', '\0']), ('\u{1ed7}', ['\u{1ed6}', '\0', '\0']), + ('\u{1ed9}', ['\u{1ed8}', '\0', '\0']), ('\u{1edb}', ['\u{1eda}', '\0', '\0']), ('\u{1edd}', + ['\u{1edc}', '\0', '\0']), ('\u{1edf}', ['\u{1ede}', '\0', '\0']), ('\u{1ee1}', ['\u{1ee0}', + '\0', '\0']), ('\u{1ee3}', ['\u{1ee2}', '\0', '\0']), ('\u{1ee5}', ['\u{1ee4}', '\0', + '\0']), ('\u{1ee7}', ['\u{1ee6}', '\0', '\0']), ('\u{1ee9}', ['\u{1ee8}', '\0', '\0']), + ('\u{1eeb}', ['\u{1eea}', '\0', '\0']), ('\u{1eed}', ['\u{1eec}', '\0', '\0']), ('\u{1eef}', + ['\u{1eee}', '\0', '\0']), ('\u{1ef1}', ['\u{1ef0}', '\0', '\0']), ('\u{1ef3}', ['\u{1ef2}', + '\0', '\0']), ('\u{1ef5}', ['\u{1ef4}', '\0', '\0']), ('\u{1ef7}', ['\u{1ef6}', '\0', + '\0']), ('\u{1ef9}', ['\u{1ef8}', '\0', '\0']), ('\u{1efb}', ['\u{1efa}', '\0', '\0']), + ('\u{1efd}', ['\u{1efc}', '\0', '\0']), ('\u{1eff}', ['\u{1efe}', '\0', '\0']), ('\u{1f00}', + ['\u{1f08}', '\0', '\0']), ('\u{1f01}', ['\u{1f09}', '\0', '\0']), ('\u{1f02}', ['\u{1f0a}', + '\0', '\0']), ('\u{1f03}', ['\u{1f0b}', '\0', '\0']), ('\u{1f04}', ['\u{1f0c}', '\0', + '\0']), ('\u{1f05}', ['\u{1f0d}', '\0', '\0']), ('\u{1f06}', ['\u{1f0e}', '\0', '\0']), + ('\u{1f07}', ['\u{1f0f}', '\0', '\0']), ('\u{1f10}', ['\u{1f18}', '\0', '\0']), ('\u{1f11}', + ['\u{1f19}', '\0', '\0']), ('\u{1f12}', ['\u{1f1a}', '\0', '\0']), ('\u{1f13}', ['\u{1f1b}', + '\0', '\0']), ('\u{1f14}', ['\u{1f1c}', '\0', '\0']), ('\u{1f15}', ['\u{1f1d}', '\0', + '\0']), ('\u{1f20}', ['\u{1f28}', '\0', '\0']), ('\u{1f21}', ['\u{1f29}', '\0', '\0']), + ('\u{1f22}', ['\u{1f2a}', '\0', '\0']), ('\u{1f23}', ['\u{1f2b}', '\0', '\0']), ('\u{1f24}', + ['\u{1f2c}', '\0', '\0']), ('\u{1f25}', ['\u{1f2d}', '\0', '\0']), ('\u{1f26}', ['\u{1f2e}', + '\0', '\0']), ('\u{1f27}', ['\u{1f2f}', '\0', '\0']), ('\u{1f30}', ['\u{1f38}', '\0', + '\0']), ('\u{1f31}', ['\u{1f39}', '\0', '\0']), ('\u{1f32}', ['\u{1f3a}', '\0', '\0']), + ('\u{1f33}', ['\u{1f3b}', '\0', '\0']), ('\u{1f34}', ['\u{1f3c}', '\0', '\0']), ('\u{1f35}', + ['\u{1f3d}', '\0', '\0']), ('\u{1f36}', ['\u{1f3e}', '\0', '\0']), ('\u{1f37}', ['\u{1f3f}', + '\0', '\0']), ('\u{1f40}', ['\u{1f48}', '\0', '\0']), ('\u{1f41}', ['\u{1f49}', '\0', + '\0']), ('\u{1f42}', ['\u{1f4a}', '\0', '\0']), ('\u{1f43}', ['\u{1f4b}', '\0', '\0']), + ('\u{1f44}', ['\u{1f4c}', '\0', '\0']), ('\u{1f45}', ['\u{1f4d}', '\0', '\0']), ('\u{1f50}', + ['\u{3a5}', '\u{313}', '\0']), ('\u{1f51}', ['\u{1f59}', '\0', '\0']), ('\u{1f52}', + ['\u{3a5}', '\u{313}', '\u{300}']), ('\u{1f53}', ['\u{1f5b}', '\0', '\0']), ('\u{1f54}', + ['\u{3a5}', '\u{313}', '\u{301}']), ('\u{1f55}', ['\u{1f5d}', '\0', '\0']), ('\u{1f56}', + ['\u{3a5}', '\u{313}', '\u{342}']), ('\u{1f57}', ['\u{1f5f}', '\0', '\0']), ('\u{1f60}', + ['\u{1f68}', '\0', '\0']), ('\u{1f61}', ['\u{1f69}', '\0', '\0']), ('\u{1f62}', ['\u{1f6a}', + '\0', '\0']), ('\u{1f63}', ['\u{1f6b}', '\0', '\0']), ('\u{1f64}', ['\u{1f6c}', '\0', + '\0']), ('\u{1f65}', ['\u{1f6d}', '\0', '\0']), ('\u{1f66}', ['\u{1f6e}', '\0', '\0']), + ('\u{1f67}', ['\u{1f6f}', '\0', '\0']), ('\u{1f70}', ['\u{1fba}', '\0', '\0']), ('\u{1f71}', + ['\u{1fbb}', '\0', '\0']), ('\u{1f72}', ['\u{1fc8}', '\0', '\0']), ('\u{1f73}', ['\u{1fc9}', + '\0', '\0']), ('\u{1f74}', ['\u{1fca}', '\0', '\0']), ('\u{1f75}', ['\u{1fcb}', '\0', + '\0']), ('\u{1f76}', ['\u{1fda}', '\0', '\0']), ('\u{1f77}', ['\u{1fdb}', '\0', '\0']), + ('\u{1f78}', ['\u{1ff8}', '\0', '\0']), ('\u{1f79}', ['\u{1ff9}', '\0', '\0']), ('\u{1f7a}', + ['\u{1fea}', '\0', '\0']), ('\u{1f7b}', ['\u{1feb}', '\0', '\0']), ('\u{1f7c}', ['\u{1ffa}', + '\0', '\0']), ('\u{1f7d}', ['\u{1ffb}', '\0', '\0']), ('\u{1f80}', ['\u{1f08}', '\u{399}', + '\0']), ('\u{1f81}', ['\u{1f09}', '\u{399}', '\0']), ('\u{1f82}', ['\u{1f0a}', '\u{399}', + '\0']), ('\u{1f83}', ['\u{1f0b}', '\u{399}', '\0']), ('\u{1f84}', ['\u{1f0c}', '\u{399}', + '\0']), ('\u{1f85}', ['\u{1f0d}', '\u{399}', '\0']), ('\u{1f86}', ['\u{1f0e}', '\u{399}', + '\0']), ('\u{1f87}', ['\u{1f0f}', '\u{399}', '\0']), ('\u{1f88}', ['\u{1f08}', '\u{399}', + '\0']), ('\u{1f89}', ['\u{1f09}', '\u{399}', '\0']), ('\u{1f8a}', ['\u{1f0a}', '\u{399}', + '\0']), ('\u{1f8b}', ['\u{1f0b}', '\u{399}', '\0']), ('\u{1f8c}', ['\u{1f0c}', '\u{399}', + '\0']), ('\u{1f8d}', ['\u{1f0d}', '\u{399}', '\0']), ('\u{1f8e}', ['\u{1f0e}', '\u{399}', + '\0']), ('\u{1f8f}', ['\u{1f0f}', '\u{399}', '\0']), ('\u{1f90}', ['\u{1f28}', '\u{399}', + '\0']), ('\u{1f91}', ['\u{1f29}', '\u{399}', '\0']), ('\u{1f92}', ['\u{1f2a}', '\u{399}', + '\0']), ('\u{1f93}', ['\u{1f2b}', '\u{399}', '\0']), ('\u{1f94}', ['\u{1f2c}', '\u{399}', + '\0']), ('\u{1f95}', ['\u{1f2d}', '\u{399}', '\0']), ('\u{1f96}', ['\u{1f2e}', '\u{399}', + '\0']), ('\u{1f97}', ['\u{1f2f}', '\u{399}', '\0']), ('\u{1f98}', ['\u{1f28}', '\u{399}', + '\0']), ('\u{1f99}', ['\u{1f29}', '\u{399}', '\0']), ('\u{1f9a}', ['\u{1f2a}', '\u{399}', + '\0']), ('\u{1f9b}', ['\u{1f2b}', '\u{399}', '\0']), ('\u{1f9c}', ['\u{1f2c}', '\u{399}', + '\0']), ('\u{1f9d}', ['\u{1f2d}', '\u{399}', '\0']), ('\u{1f9e}', ['\u{1f2e}', '\u{399}', + '\0']), ('\u{1f9f}', ['\u{1f2f}', '\u{399}', '\0']), ('\u{1fa0}', ['\u{1f68}', '\u{399}', + '\0']), ('\u{1fa1}', ['\u{1f69}', '\u{399}', '\0']), ('\u{1fa2}', ['\u{1f6a}', '\u{399}', + '\0']), ('\u{1fa3}', ['\u{1f6b}', '\u{399}', '\0']), ('\u{1fa4}', ['\u{1f6c}', '\u{399}', + '\0']), ('\u{1fa5}', ['\u{1f6d}', '\u{399}', '\0']), ('\u{1fa6}', ['\u{1f6e}', '\u{399}', + '\0']), ('\u{1fa7}', ['\u{1f6f}', '\u{399}', '\0']), ('\u{1fa8}', ['\u{1f68}', '\u{399}', + '\0']), ('\u{1fa9}', ['\u{1f69}', '\u{399}', '\0']), ('\u{1faa}', ['\u{1f6a}', '\u{399}', + '\0']), ('\u{1fab}', ['\u{1f6b}', '\u{399}', '\0']), ('\u{1fac}', ['\u{1f6c}', '\u{399}', + '\0']), ('\u{1fad}', ['\u{1f6d}', '\u{399}', '\0']), ('\u{1fae}', ['\u{1f6e}', '\u{399}', + '\0']), ('\u{1faf}', ['\u{1f6f}', '\u{399}', '\0']), ('\u{1fb0}', ['\u{1fb8}', '\0', '\0']), + ('\u{1fb1}', ['\u{1fb9}', '\0', '\0']), ('\u{1fb2}', ['\u{1fba}', '\u{399}', '\0']), + ('\u{1fb3}', ['\u{391}', '\u{399}', '\0']), ('\u{1fb4}', ['\u{386}', '\u{399}', '\0']), + ('\u{1fb6}', ['\u{391}', '\u{342}', '\0']), ('\u{1fb7}', ['\u{391}', '\u{342}', '\u{399}']), + ('\u{1fbc}', ['\u{391}', '\u{399}', '\0']), ('\u{1fbe}', ['\u{399}', '\0', '\0']), + ('\u{1fc2}', ['\u{1fca}', '\u{399}', '\0']), ('\u{1fc3}', ['\u{397}', '\u{399}', '\0']), + ('\u{1fc4}', ['\u{389}', '\u{399}', '\0']), ('\u{1fc6}', ['\u{397}', '\u{342}', '\0']), + ('\u{1fc7}', ['\u{397}', '\u{342}', '\u{399}']), ('\u{1fcc}', ['\u{397}', '\u{399}', '\0']), + ('\u{1fd0}', ['\u{1fd8}', '\0', '\0']), ('\u{1fd1}', ['\u{1fd9}', '\0', '\0']), ('\u{1fd2}', + ['\u{399}', '\u{308}', '\u{300}']), ('\u{1fd3}', ['\u{399}', '\u{308}', '\u{301}']), + ('\u{1fd6}', ['\u{399}', '\u{342}', '\0']), ('\u{1fd7}', ['\u{399}', '\u{308}', '\u{342}']), + ('\u{1fe0}', ['\u{1fe8}', '\0', '\0']), ('\u{1fe1}', ['\u{1fe9}', '\0', '\0']), ('\u{1fe2}', + ['\u{3a5}', '\u{308}', '\u{300}']), ('\u{1fe3}', ['\u{3a5}', '\u{308}', '\u{301}']), + ('\u{1fe4}', ['\u{3a1}', '\u{313}', '\0']), ('\u{1fe5}', ['\u{1fec}', '\0', '\0']), + ('\u{1fe6}', ['\u{3a5}', '\u{342}', '\0']), ('\u{1fe7}', ['\u{3a5}', '\u{308}', '\u{342}']), + ('\u{1ff2}', ['\u{1ffa}', '\u{399}', '\0']), ('\u{1ff3}', ['\u{3a9}', '\u{399}', '\0']), + ('\u{1ff4}', ['\u{38f}', '\u{399}', '\0']), ('\u{1ff6}', ['\u{3a9}', '\u{342}', '\0']), + ('\u{1ff7}', ['\u{3a9}', '\u{342}', '\u{399}']), ('\u{1ffc}', ['\u{3a9}', '\u{399}', '\0']), + ('\u{214e}', ['\u{2132}', '\0', '\0']), ('\u{2170}', ['\u{2160}', '\0', '\0']), ('\u{2171}', + ['\u{2161}', '\0', '\0']), ('\u{2172}', ['\u{2162}', '\0', '\0']), ('\u{2173}', ['\u{2163}', + '\0', '\0']), ('\u{2174}', ['\u{2164}', '\0', '\0']), ('\u{2175}', ['\u{2165}', '\0', + '\0']), ('\u{2176}', ['\u{2166}', '\0', '\0']), ('\u{2177}', ['\u{2167}', '\0', '\0']), + ('\u{2178}', ['\u{2168}', '\0', '\0']), ('\u{2179}', ['\u{2169}', '\0', '\0']), ('\u{217a}', + ['\u{216a}', '\0', '\0']), ('\u{217b}', ['\u{216b}', '\0', '\0']), ('\u{217c}', ['\u{216c}', + '\0', '\0']), ('\u{217d}', ['\u{216d}', '\0', '\0']), ('\u{217e}', ['\u{216e}', '\0', + '\0']), ('\u{217f}', ['\u{216f}', '\0', '\0']), ('\u{2184}', ['\u{2183}', '\0', '\0']), + ('\u{24d0}', ['\u{24b6}', '\0', '\0']), ('\u{24d1}', ['\u{24b7}', '\0', '\0']), ('\u{24d2}', + ['\u{24b8}', '\0', '\0']), ('\u{24d3}', ['\u{24b9}', '\0', '\0']), ('\u{24d4}', ['\u{24ba}', + '\0', '\0']), ('\u{24d5}', ['\u{24bb}', '\0', '\0']), ('\u{24d6}', ['\u{24bc}', '\0', + '\0']), ('\u{24d7}', ['\u{24bd}', '\0', '\0']), ('\u{24d8}', ['\u{24be}', '\0', '\0']), + ('\u{24d9}', ['\u{24bf}', '\0', '\0']), ('\u{24da}', ['\u{24c0}', '\0', '\0']), ('\u{24db}', + ['\u{24c1}', '\0', '\0']), ('\u{24dc}', ['\u{24c2}', '\0', '\0']), ('\u{24dd}', ['\u{24c3}', + '\0', '\0']), ('\u{24de}', ['\u{24c4}', '\0', '\0']), ('\u{24df}', ['\u{24c5}', '\0', + '\0']), ('\u{24e0}', ['\u{24c6}', '\0', '\0']), ('\u{24e1}', ['\u{24c7}', '\0', '\0']), + ('\u{24e2}', ['\u{24c8}', '\0', '\0']), ('\u{24e3}', ['\u{24c9}', '\0', '\0']), ('\u{24e4}', + ['\u{24ca}', '\0', '\0']), ('\u{24e5}', ['\u{24cb}', '\0', '\0']), ('\u{24e6}', ['\u{24cc}', + '\0', '\0']), ('\u{24e7}', ['\u{24cd}', '\0', '\0']), ('\u{24e8}', ['\u{24ce}', '\0', + '\0']), ('\u{24e9}', ['\u{24cf}', '\0', '\0']), ('\u{2c30}', ['\u{2c00}', '\0', '\0']), + ('\u{2c31}', ['\u{2c01}', '\0', '\0']), ('\u{2c32}', ['\u{2c02}', '\0', '\0']), ('\u{2c33}', + ['\u{2c03}', '\0', '\0']), ('\u{2c34}', ['\u{2c04}', '\0', '\0']), ('\u{2c35}', ['\u{2c05}', + '\0', '\0']), ('\u{2c36}', ['\u{2c06}', '\0', '\0']), ('\u{2c37}', ['\u{2c07}', '\0', + '\0']), ('\u{2c38}', ['\u{2c08}', '\0', '\0']), ('\u{2c39}', ['\u{2c09}', '\0', '\0']), + ('\u{2c3a}', ['\u{2c0a}', '\0', '\0']), ('\u{2c3b}', ['\u{2c0b}', '\0', '\0']), ('\u{2c3c}', + ['\u{2c0c}', '\0', '\0']), ('\u{2c3d}', ['\u{2c0d}', '\0', '\0']), ('\u{2c3e}', ['\u{2c0e}', + '\0', '\0']), ('\u{2c3f}', ['\u{2c0f}', '\0', '\0']), ('\u{2c40}', ['\u{2c10}', '\0', + '\0']), ('\u{2c41}', ['\u{2c11}', '\0', '\0']), ('\u{2c42}', ['\u{2c12}', '\0', '\0']), + ('\u{2c43}', ['\u{2c13}', '\0', '\0']), ('\u{2c44}', ['\u{2c14}', '\0', '\0']), ('\u{2c45}', + ['\u{2c15}', '\0', '\0']), ('\u{2c46}', ['\u{2c16}', '\0', '\0']), ('\u{2c47}', ['\u{2c17}', + '\0', '\0']), ('\u{2c48}', ['\u{2c18}', '\0', '\0']), ('\u{2c49}', ['\u{2c19}', '\0', + '\0']), ('\u{2c4a}', ['\u{2c1a}', '\0', '\0']), ('\u{2c4b}', ['\u{2c1b}', '\0', '\0']), + ('\u{2c4c}', ['\u{2c1c}', '\0', '\0']), ('\u{2c4d}', ['\u{2c1d}', '\0', '\0']), ('\u{2c4e}', + ['\u{2c1e}', '\0', '\0']), ('\u{2c4f}', ['\u{2c1f}', '\0', '\0']), ('\u{2c50}', ['\u{2c20}', + '\0', '\0']), ('\u{2c51}', ['\u{2c21}', '\0', '\0']), ('\u{2c52}', ['\u{2c22}', '\0', + '\0']), ('\u{2c53}', ['\u{2c23}', '\0', '\0']), ('\u{2c54}', ['\u{2c24}', '\0', '\0']), + ('\u{2c55}', ['\u{2c25}', '\0', '\0']), ('\u{2c56}', ['\u{2c26}', '\0', '\0']), ('\u{2c57}', + ['\u{2c27}', '\0', '\0']), ('\u{2c58}', ['\u{2c28}', '\0', '\0']), ('\u{2c59}', ['\u{2c29}', + '\0', '\0']), ('\u{2c5a}', ['\u{2c2a}', '\0', '\0']), ('\u{2c5b}', ['\u{2c2b}', '\0', + '\0']), ('\u{2c5c}', ['\u{2c2c}', '\0', '\0']), ('\u{2c5d}', ['\u{2c2d}', '\0', '\0']), + ('\u{2c5e}', ['\u{2c2e}', '\0', '\0']), ('\u{2c61}', ['\u{2c60}', '\0', '\0']), ('\u{2c65}', + ['\u{23a}', '\0', '\0']), ('\u{2c66}', ['\u{23e}', '\0', '\0']), ('\u{2c68}', ['\u{2c67}', + '\0', '\0']), ('\u{2c6a}', ['\u{2c69}', '\0', '\0']), ('\u{2c6c}', ['\u{2c6b}', '\0', + '\0']), ('\u{2c73}', ['\u{2c72}', '\0', '\0']), ('\u{2c76}', ['\u{2c75}', '\0', '\0']), + ('\u{2c81}', ['\u{2c80}', '\0', '\0']), ('\u{2c83}', ['\u{2c82}', '\0', '\0']), ('\u{2c85}', + ['\u{2c84}', '\0', '\0']), ('\u{2c87}', ['\u{2c86}', '\0', '\0']), ('\u{2c89}', ['\u{2c88}', + '\0', '\0']), ('\u{2c8b}', ['\u{2c8a}', '\0', '\0']), ('\u{2c8d}', ['\u{2c8c}', '\0', + '\0']), ('\u{2c8f}', ['\u{2c8e}', '\0', '\0']), ('\u{2c91}', ['\u{2c90}', '\0', '\0']), + ('\u{2c93}', ['\u{2c92}', '\0', '\0']), ('\u{2c95}', ['\u{2c94}', '\0', '\0']), ('\u{2c97}', + ['\u{2c96}', '\0', '\0']), ('\u{2c99}', ['\u{2c98}', '\0', '\0']), ('\u{2c9b}', ['\u{2c9a}', + '\0', '\0']), ('\u{2c9d}', ['\u{2c9c}', '\0', '\0']), ('\u{2c9f}', ['\u{2c9e}', '\0', + '\0']), ('\u{2ca1}', ['\u{2ca0}', '\0', '\0']), ('\u{2ca3}', ['\u{2ca2}', '\0', '\0']), + ('\u{2ca5}', ['\u{2ca4}', '\0', '\0']), ('\u{2ca7}', ['\u{2ca6}', '\0', '\0']), ('\u{2ca9}', + ['\u{2ca8}', '\0', '\0']), ('\u{2cab}', ['\u{2caa}', '\0', '\0']), ('\u{2cad}', ['\u{2cac}', + '\0', '\0']), ('\u{2caf}', ['\u{2cae}', '\0', '\0']), ('\u{2cb1}', ['\u{2cb0}', '\0', + '\0']), ('\u{2cb3}', ['\u{2cb2}', '\0', '\0']), ('\u{2cb5}', ['\u{2cb4}', '\0', '\0']), + ('\u{2cb7}', ['\u{2cb6}', '\0', '\0']), ('\u{2cb9}', ['\u{2cb8}', '\0', '\0']), ('\u{2cbb}', + ['\u{2cba}', '\0', '\0']), ('\u{2cbd}', ['\u{2cbc}', '\0', '\0']), ('\u{2cbf}', ['\u{2cbe}', + '\0', '\0']), ('\u{2cc1}', ['\u{2cc0}', '\0', '\0']), ('\u{2cc3}', ['\u{2cc2}', '\0', + '\0']), ('\u{2cc5}', ['\u{2cc4}', '\0', '\0']), ('\u{2cc7}', ['\u{2cc6}', '\0', '\0']), + ('\u{2cc9}', ['\u{2cc8}', '\0', '\0']), ('\u{2ccb}', ['\u{2cca}', '\0', '\0']), ('\u{2ccd}', + ['\u{2ccc}', '\0', '\0']), ('\u{2ccf}', ['\u{2cce}', '\0', '\0']), ('\u{2cd1}', ['\u{2cd0}', + '\0', '\0']), ('\u{2cd3}', ['\u{2cd2}', '\0', '\0']), ('\u{2cd5}', ['\u{2cd4}', '\0', + '\0']), ('\u{2cd7}', ['\u{2cd6}', '\0', '\0']), ('\u{2cd9}', ['\u{2cd8}', '\0', '\0']), + ('\u{2cdb}', ['\u{2cda}', '\0', '\0']), ('\u{2cdd}', ['\u{2cdc}', '\0', '\0']), ('\u{2cdf}', + ['\u{2cde}', '\0', '\0']), ('\u{2ce1}', ['\u{2ce0}', '\0', '\0']), ('\u{2ce3}', ['\u{2ce2}', + '\0', '\0']), ('\u{2cec}', ['\u{2ceb}', '\0', '\0']), ('\u{2cee}', ['\u{2ced}', '\0', + '\0']), ('\u{2cf3}', ['\u{2cf2}', '\0', '\0']), ('\u{2d00}', ['\u{10a0}', '\0', '\0']), + ('\u{2d01}', ['\u{10a1}', '\0', '\0']), ('\u{2d02}', ['\u{10a2}', '\0', '\0']), ('\u{2d03}', + ['\u{10a3}', '\0', '\0']), ('\u{2d04}', ['\u{10a4}', '\0', '\0']), ('\u{2d05}', ['\u{10a5}', + '\0', '\0']), ('\u{2d06}', ['\u{10a6}', '\0', '\0']), ('\u{2d07}', ['\u{10a7}', '\0', + '\0']), ('\u{2d08}', ['\u{10a8}', '\0', '\0']), ('\u{2d09}', ['\u{10a9}', '\0', '\0']), + ('\u{2d0a}', ['\u{10aa}', '\0', '\0']), ('\u{2d0b}', ['\u{10ab}', '\0', '\0']), ('\u{2d0c}', + ['\u{10ac}', '\0', '\0']), ('\u{2d0d}', ['\u{10ad}', '\0', '\0']), ('\u{2d0e}', ['\u{10ae}', + '\0', '\0']), ('\u{2d0f}', ['\u{10af}', '\0', '\0']), ('\u{2d10}', ['\u{10b0}', '\0', + '\0']), ('\u{2d11}', ['\u{10b1}', '\0', '\0']), ('\u{2d12}', ['\u{10b2}', '\0', '\0']), + ('\u{2d13}', ['\u{10b3}', '\0', '\0']), ('\u{2d14}', ['\u{10b4}', '\0', '\0']), ('\u{2d15}', + ['\u{10b5}', '\0', '\0']), ('\u{2d16}', ['\u{10b6}', '\0', '\0']), ('\u{2d17}', ['\u{10b7}', + '\0', '\0']), ('\u{2d18}', ['\u{10b8}', '\0', '\0']), ('\u{2d19}', ['\u{10b9}', '\0', + '\0']), ('\u{2d1a}', ['\u{10ba}', '\0', '\0']), ('\u{2d1b}', ['\u{10bb}', '\0', '\0']), + ('\u{2d1c}', ['\u{10bc}', '\0', '\0']), ('\u{2d1d}', ['\u{10bd}', '\0', '\0']), ('\u{2d1e}', + ['\u{10be}', '\0', '\0']), ('\u{2d1f}', ['\u{10bf}', '\0', '\0']), ('\u{2d20}', ['\u{10c0}', + '\0', '\0']), ('\u{2d21}', ['\u{10c1}', '\0', '\0']), ('\u{2d22}', ['\u{10c2}', '\0', + '\0']), ('\u{2d23}', ['\u{10c3}', '\0', '\0']), ('\u{2d24}', ['\u{10c4}', '\0', '\0']), + ('\u{2d25}', ['\u{10c5}', '\0', '\0']), ('\u{2d27}', ['\u{10c7}', '\0', '\0']), ('\u{2d2d}', + ['\u{10cd}', '\0', '\0']), ('\u{a641}', ['\u{a640}', '\0', '\0']), ('\u{a643}', ['\u{a642}', + '\0', '\0']), ('\u{a645}', ['\u{a644}', '\0', '\0']), ('\u{a647}', ['\u{a646}', '\0', + '\0']), ('\u{a649}', ['\u{a648}', '\0', '\0']), ('\u{a64b}', ['\u{a64a}', '\0', '\0']), + ('\u{a64d}', ['\u{a64c}', '\0', '\0']), ('\u{a64f}', ['\u{a64e}', '\0', '\0']), ('\u{a651}', + ['\u{a650}', '\0', '\0']), ('\u{a653}', ['\u{a652}', '\0', '\0']), ('\u{a655}', ['\u{a654}', + '\0', '\0']), ('\u{a657}', ['\u{a656}', '\0', '\0']), ('\u{a659}', ['\u{a658}', '\0', + '\0']), ('\u{a65b}', ['\u{a65a}', '\0', '\0']), ('\u{a65d}', ['\u{a65c}', '\0', '\0']), + ('\u{a65f}', ['\u{a65e}', '\0', '\0']), ('\u{a661}', ['\u{a660}', '\0', '\0']), ('\u{a663}', + ['\u{a662}', '\0', '\0']), ('\u{a665}', ['\u{a664}', '\0', '\0']), ('\u{a667}', ['\u{a666}', + '\0', '\0']), ('\u{a669}', ['\u{a668}', '\0', '\0']), ('\u{a66b}', ['\u{a66a}', '\0', + '\0']), ('\u{a66d}', ['\u{a66c}', '\0', '\0']), ('\u{a681}', ['\u{a680}', '\0', '\0']), + ('\u{a683}', ['\u{a682}', '\0', '\0']), ('\u{a685}', ['\u{a684}', '\0', '\0']), ('\u{a687}', + ['\u{a686}', '\0', '\0']), ('\u{a689}', ['\u{a688}', '\0', '\0']), ('\u{a68b}', ['\u{a68a}', + '\0', '\0']), ('\u{a68d}', ['\u{a68c}', '\0', '\0']), ('\u{a68f}', ['\u{a68e}', '\0', + '\0']), ('\u{a691}', ['\u{a690}', '\0', '\0']), ('\u{a693}', ['\u{a692}', '\0', '\0']), + ('\u{a695}', ['\u{a694}', '\0', '\0']), ('\u{a697}', ['\u{a696}', '\0', '\0']), ('\u{a699}', + ['\u{a698}', '\0', '\0']), ('\u{a69b}', ['\u{a69a}', '\0', '\0']), ('\u{a723}', ['\u{a722}', + '\0', '\0']), ('\u{a725}', ['\u{a724}', '\0', '\0']), ('\u{a727}', ['\u{a726}', '\0', + '\0']), ('\u{a729}', ['\u{a728}', '\0', '\0']), ('\u{a72b}', ['\u{a72a}', '\0', '\0']), + ('\u{a72d}', ['\u{a72c}', '\0', '\0']), ('\u{a72f}', ['\u{a72e}', '\0', '\0']), ('\u{a733}', + ['\u{a732}', '\0', '\0']), ('\u{a735}', ['\u{a734}', '\0', '\0']), ('\u{a737}', ['\u{a736}', + '\0', '\0']), ('\u{a739}', ['\u{a738}', '\0', '\0']), ('\u{a73b}', ['\u{a73a}', '\0', + '\0']), ('\u{a73d}', ['\u{a73c}', '\0', '\0']), ('\u{a73f}', ['\u{a73e}', '\0', '\0']), + ('\u{a741}', ['\u{a740}', '\0', '\0']), ('\u{a743}', ['\u{a742}', '\0', '\0']), ('\u{a745}', + ['\u{a744}', '\0', '\0']), ('\u{a747}', ['\u{a746}', '\0', '\0']), ('\u{a749}', ['\u{a748}', + '\0', '\0']), ('\u{a74b}', ['\u{a74a}', '\0', '\0']), ('\u{a74d}', ['\u{a74c}', '\0', + '\0']), ('\u{a74f}', ['\u{a74e}', '\0', '\0']), ('\u{a751}', ['\u{a750}', '\0', '\0']), + ('\u{a753}', ['\u{a752}', '\0', '\0']), ('\u{a755}', ['\u{a754}', '\0', '\0']), ('\u{a757}', + ['\u{a756}', '\0', '\0']), ('\u{a759}', ['\u{a758}', '\0', '\0']), ('\u{a75b}', ['\u{a75a}', + '\0', '\0']), ('\u{a75d}', ['\u{a75c}', '\0', '\0']), ('\u{a75f}', ['\u{a75e}', '\0', + '\0']), ('\u{a761}', ['\u{a760}', '\0', '\0']), ('\u{a763}', ['\u{a762}', '\0', '\0']), + ('\u{a765}', ['\u{a764}', '\0', '\0']), ('\u{a767}', ['\u{a766}', '\0', '\0']), ('\u{a769}', + ['\u{a768}', '\0', '\0']), ('\u{a76b}', ['\u{a76a}', '\0', '\0']), ('\u{a76d}', ['\u{a76c}', + '\0', '\0']), ('\u{a76f}', ['\u{a76e}', '\0', '\0']), ('\u{a77a}', ['\u{a779}', '\0', + '\0']), ('\u{a77c}', ['\u{a77b}', '\0', '\0']), ('\u{a77f}', ['\u{a77e}', '\0', '\0']), + ('\u{a781}', ['\u{a780}', '\0', '\0']), ('\u{a783}', ['\u{a782}', '\0', '\0']), ('\u{a785}', + ['\u{a784}', '\0', '\0']), ('\u{a787}', ['\u{a786}', '\0', '\0']), ('\u{a78c}', ['\u{a78b}', + '\0', '\0']), ('\u{a791}', ['\u{a790}', '\0', '\0']), ('\u{a793}', ['\u{a792}', '\0', + '\0']), ('\u{a797}', ['\u{a796}', '\0', '\0']), ('\u{a799}', ['\u{a798}', '\0', '\0']), + ('\u{a79b}', ['\u{a79a}', '\0', '\0']), ('\u{a79d}', ['\u{a79c}', '\0', '\0']), ('\u{a79f}', + ['\u{a79e}', '\0', '\0']), ('\u{a7a1}', ['\u{a7a0}', '\0', '\0']), ('\u{a7a3}', ['\u{a7a2}', + '\0', '\0']), ('\u{a7a5}', ['\u{a7a4}', '\0', '\0']), ('\u{a7a7}', ['\u{a7a6}', '\0', + '\0']), ('\u{a7a9}', ['\u{a7a8}', '\0', '\0']), ('\u{a7b5}', ['\u{a7b4}', '\0', '\0']), + ('\u{a7b7}', ['\u{a7b6}', '\0', '\0']), ('\u{a7b9}', ['\u{a7b8}', '\0', '\0']), ('\u{ab53}', + ['\u{a7b3}', '\0', '\0']), ('\u{ab70}', ['\u{13a0}', '\0', '\0']), ('\u{ab71}', ['\u{13a1}', + '\0', '\0']), ('\u{ab72}', ['\u{13a2}', '\0', '\0']), ('\u{ab73}', ['\u{13a3}', '\0', + '\0']), ('\u{ab74}', ['\u{13a4}', '\0', '\0']), ('\u{ab75}', ['\u{13a5}', '\0', '\0']), + ('\u{ab76}', ['\u{13a6}', '\0', '\0']), ('\u{ab77}', ['\u{13a7}', '\0', '\0']), ('\u{ab78}', + ['\u{13a8}', '\0', '\0']), ('\u{ab79}', ['\u{13a9}', '\0', '\0']), ('\u{ab7a}', ['\u{13aa}', + '\0', '\0']), ('\u{ab7b}', ['\u{13ab}', '\0', '\0']), ('\u{ab7c}', ['\u{13ac}', '\0', + '\0']), ('\u{ab7d}', ['\u{13ad}', '\0', '\0']), ('\u{ab7e}', ['\u{13ae}', '\0', '\0']), + ('\u{ab7f}', ['\u{13af}', '\0', '\0']), ('\u{ab80}', ['\u{13b0}', '\0', '\0']), ('\u{ab81}', + ['\u{13b1}', '\0', '\0']), ('\u{ab82}', ['\u{13b2}', '\0', '\0']), ('\u{ab83}', ['\u{13b3}', + '\0', '\0']), ('\u{ab84}', ['\u{13b4}', '\0', '\0']), ('\u{ab85}', ['\u{13b5}', '\0', + '\0']), ('\u{ab86}', ['\u{13b6}', '\0', '\0']), ('\u{ab87}', ['\u{13b7}', '\0', '\0']), + ('\u{ab88}', ['\u{13b8}', '\0', '\0']), ('\u{ab89}', ['\u{13b9}', '\0', '\0']), ('\u{ab8a}', + ['\u{13ba}', '\0', '\0']), ('\u{ab8b}', ['\u{13bb}', '\0', '\0']), ('\u{ab8c}', ['\u{13bc}', + '\0', '\0']), ('\u{ab8d}', ['\u{13bd}', '\0', '\0']), ('\u{ab8e}', ['\u{13be}', '\0', + '\0']), ('\u{ab8f}', ['\u{13bf}', '\0', '\0']), ('\u{ab90}', ['\u{13c0}', '\0', '\0']), + ('\u{ab91}', ['\u{13c1}', '\0', '\0']), ('\u{ab92}', ['\u{13c2}', '\0', '\0']), ('\u{ab93}', + ['\u{13c3}', '\0', '\0']), ('\u{ab94}', ['\u{13c4}', '\0', '\0']), ('\u{ab95}', ['\u{13c5}', + '\0', '\0']), ('\u{ab96}', ['\u{13c6}', '\0', '\0']), ('\u{ab97}', ['\u{13c7}', '\0', + '\0']), ('\u{ab98}', ['\u{13c8}', '\0', '\0']), ('\u{ab99}', ['\u{13c9}', '\0', '\0']), + ('\u{ab9a}', ['\u{13ca}', '\0', '\0']), ('\u{ab9b}', ['\u{13cb}', '\0', '\0']), ('\u{ab9c}', + ['\u{13cc}', '\0', '\0']), ('\u{ab9d}', ['\u{13cd}', '\0', '\0']), ('\u{ab9e}', ['\u{13ce}', + '\0', '\0']), ('\u{ab9f}', ['\u{13cf}', '\0', '\0']), ('\u{aba0}', ['\u{13d0}', '\0', + '\0']), ('\u{aba1}', ['\u{13d1}', '\0', '\0']), ('\u{aba2}', ['\u{13d2}', '\0', '\0']), + ('\u{aba3}', ['\u{13d3}', '\0', '\0']), ('\u{aba4}', ['\u{13d4}', '\0', '\0']), ('\u{aba5}', + ['\u{13d5}', '\0', '\0']), ('\u{aba6}', ['\u{13d6}', '\0', '\0']), ('\u{aba7}', ['\u{13d7}', + '\0', '\0']), ('\u{aba8}', ['\u{13d8}', '\0', '\0']), ('\u{aba9}', ['\u{13d9}', '\0', + '\0']), ('\u{abaa}', ['\u{13da}', '\0', '\0']), ('\u{abab}', ['\u{13db}', '\0', '\0']), + ('\u{abac}', ['\u{13dc}', '\0', '\0']), ('\u{abad}', ['\u{13dd}', '\0', '\0']), ('\u{abae}', + ['\u{13de}', '\0', '\0']), ('\u{abaf}', ['\u{13df}', '\0', '\0']), ('\u{abb0}', ['\u{13e0}', + '\0', '\0']), ('\u{abb1}', ['\u{13e1}', '\0', '\0']), ('\u{abb2}', ['\u{13e2}', '\0', + '\0']), ('\u{abb3}', ['\u{13e3}', '\0', '\0']), ('\u{abb4}', ['\u{13e4}', '\0', '\0']), + ('\u{abb5}', ['\u{13e5}', '\0', '\0']), ('\u{abb6}', ['\u{13e6}', '\0', '\0']), ('\u{abb7}', + ['\u{13e7}', '\0', '\0']), ('\u{abb8}', ['\u{13e8}', '\0', '\0']), ('\u{abb9}', ['\u{13e9}', + '\0', '\0']), ('\u{abba}', ['\u{13ea}', '\0', '\0']), ('\u{abbb}', ['\u{13eb}', '\0', + '\0']), ('\u{abbc}', ['\u{13ec}', '\0', '\0']), ('\u{abbd}', ['\u{13ed}', '\0', '\0']), + ('\u{abbe}', ['\u{13ee}', '\0', '\0']), ('\u{abbf}', ['\u{13ef}', '\0', '\0']), ('\u{fb00}', + ['\u{46}', '\u{46}', '\0']), ('\u{fb01}', ['\u{46}', '\u{49}', '\0']), ('\u{fb02}', + ['\u{46}', '\u{4c}', '\0']), ('\u{fb03}', ['\u{46}', '\u{46}', '\u{49}']), ('\u{fb04}', + ['\u{46}', '\u{46}', '\u{4c}']), ('\u{fb05}', ['\u{53}', '\u{54}', '\0']), ('\u{fb06}', + ['\u{53}', '\u{54}', '\0']), ('\u{fb13}', ['\u{544}', '\u{546}', '\0']), ('\u{fb14}', + ['\u{544}', '\u{535}', '\0']), ('\u{fb15}', ['\u{544}', '\u{53b}', '\0']), ('\u{fb16}', + ['\u{54e}', '\u{546}', '\0']), ('\u{fb17}', ['\u{544}', '\u{53d}', '\0']), ('\u{ff41}', + ['\u{ff21}', '\0', '\0']), ('\u{ff42}', ['\u{ff22}', '\0', '\0']), ('\u{ff43}', ['\u{ff23}', + '\0', '\0']), ('\u{ff44}', ['\u{ff24}', '\0', '\0']), ('\u{ff45}', ['\u{ff25}', '\0', + '\0']), ('\u{ff46}', ['\u{ff26}', '\0', '\0']), ('\u{ff47}', ['\u{ff27}', '\0', '\0']), + ('\u{ff48}', ['\u{ff28}', '\0', '\0']), ('\u{ff49}', ['\u{ff29}', '\0', '\0']), ('\u{ff4a}', + ['\u{ff2a}', '\0', '\0']), ('\u{ff4b}', ['\u{ff2b}', '\0', '\0']), ('\u{ff4c}', ['\u{ff2c}', + '\0', '\0']), ('\u{ff4d}', ['\u{ff2d}', '\0', '\0']), ('\u{ff4e}', ['\u{ff2e}', '\0', + '\0']), ('\u{ff4f}', ['\u{ff2f}', '\0', '\0']), ('\u{ff50}', ['\u{ff30}', '\0', '\0']), + ('\u{ff51}', ['\u{ff31}', '\0', '\0']), ('\u{ff52}', ['\u{ff32}', '\0', '\0']), ('\u{ff53}', + ['\u{ff33}', '\0', '\0']), ('\u{ff54}', ['\u{ff34}', '\0', '\0']), ('\u{ff55}', ['\u{ff35}', + '\0', '\0']), ('\u{ff56}', ['\u{ff36}', '\0', '\0']), ('\u{ff57}', ['\u{ff37}', '\0', + '\0']), ('\u{ff58}', ['\u{ff38}', '\0', '\0']), ('\u{ff59}', ['\u{ff39}', '\0', '\0']), + ('\u{ff5a}', ['\u{ff3a}', '\0', '\0']), ('\u{10428}', ['\u{10400}', '\0', '\0']), + ('\u{10429}', ['\u{10401}', '\0', '\0']), ('\u{1042a}', ['\u{10402}', '\0', '\0']), + ('\u{1042b}', ['\u{10403}', '\0', '\0']), ('\u{1042c}', ['\u{10404}', '\0', '\0']), + ('\u{1042d}', ['\u{10405}', '\0', '\0']), ('\u{1042e}', ['\u{10406}', '\0', '\0']), + ('\u{1042f}', ['\u{10407}', '\0', '\0']), ('\u{10430}', ['\u{10408}', '\0', '\0']), + ('\u{10431}', ['\u{10409}', '\0', '\0']), ('\u{10432}', ['\u{1040a}', '\0', '\0']), + ('\u{10433}', ['\u{1040b}', '\0', '\0']), ('\u{10434}', ['\u{1040c}', '\0', '\0']), + ('\u{10435}', ['\u{1040d}', '\0', '\0']), ('\u{10436}', ['\u{1040e}', '\0', '\0']), + ('\u{10437}', ['\u{1040f}', '\0', '\0']), ('\u{10438}', ['\u{10410}', '\0', '\0']), + ('\u{10439}', ['\u{10411}', '\0', '\0']), ('\u{1043a}', ['\u{10412}', '\0', '\0']), + ('\u{1043b}', ['\u{10413}', '\0', '\0']), ('\u{1043c}', ['\u{10414}', '\0', '\0']), + ('\u{1043d}', ['\u{10415}', '\0', '\0']), ('\u{1043e}', ['\u{10416}', '\0', '\0']), + ('\u{1043f}', ['\u{10417}', '\0', '\0']), ('\u{10440}', ['\u{10418}', '\0', '\0']), + ('\u{10441}', ['\u{10419}', '\0', '\0']), ('\u{10442}', ['\u{1041a}', '\0', '\0']), + ('\u{10443}', ['\u{1041b}', '\0', '\0']), ('\u{10444}', ['\u{1041c}', '\0', '\0']), + ('\u{10445}', ['\u{1041d}', '\0', '\0']), ('\u{10446}', ['\u{1041e}', '\0', '\0']), + ('\u{10447}', ['\u{1041f}', '\0', '\0']), ('\u{10448}', ['\u{10420}', '\0', '\0']), + ('\u{10449}', ['\u{10421}', '\0', '\0']), ('\u{1044a}', ['\u{10422}', '\0', '\0']), + ('\u{1044b}', ['\u{10423}', '\0', '\0']), ('\u{1044c}', ['\u{10424}', '\0', '\0']), + ('\u{1044d}', ['\u{10425}', '\0', '\0']), ('\u{1044e}', ['\u{10426}', '\0', '\0']), + ('\u{1044f}', ['\u{10427}', '\0', '\0']), ('\u{104d8}', ['\u{104b0}', '\0', '\0']), + ('\u{104d9}', ['\u{104b1}', '\0', '\0']), ('\u{104da}', ['\u{104b2}', '\0', '\0']), + ('\u{104db}', ['\u{104b3}', '\0', '\0']), ('\u{104dc}', ['\u{104b4}', '\0', '\0']), + ('\u{104dd}', ['\u{104b5}', '\0', '\0']), ('\u{104de}', ['\u{104b6}', '\0', '\0']), + ('\u{104df}', ['\u{104b7}', '\0', '\0']), ('\u{104e0}', ['\u{104b8}', '\0', '\0']), + ('\u{104e1}', ['\u{104b9}', '\0', '\0']), ('\u{104e2}', ['\u{104ba}', '\0', '\0']), + ('\u{104e3}', ['\u{104bb}', '\0', '\0']), ('\u{104e4}', ['\u{104bc}', '\0', '\0']), + ('\u{104e5}', ['\u{104bd}', '\0', '\0']), ('\u{104e6}', ['\u{104be}', '\0', '\0']), + ('\u{104e7}', ['\u{104bf}', '\0', '\0']), ('\u{104e8}', ['\u{104c0}', '\0', '\0']), + ('\u{104e9}', ['\u{104c1}', '\0', '\0']), ('\u{104ea}', ['\u{104c2}', '\0', '\0']), + ('\u{104eb}', ['\u{104c3}', '\0', '\0']), ('\u{104ec}', ['\u{104c4}', '\0', '\0']), + ('\u{104ed}', ['\u{104c5}', '\0', '\0']), ('\u{104ee}', ['\u{104c6}', '\0', '\0']), + ('\u{104ef}', ['\u{104c7}', '\0', '\0']), ('\u{104f0}', ['\u{104c8}', '\0', '\0']), + ('\u{104f1}', ['\u{104c9}', '\0', '\0']), ('\u{104f2}', ['\u{104ca}', '\0', '\0']), + ('\u{104f3}', ['\u{104cb}', '\0', '\0']), ('\u{104f4}', ['\u{104cc}', '\0', '\0']), + ('\u{104f5}', ['\u{104cd}', '\0', '\0']), ('\u{104f6}', ['\u{104ce}', '\0', '\0']), + ('\u{104f7}', ['\u{104cf}', '\0', '\0']), ('\u{104f8}', ['\u{104d0}', '\0', '\0']), + ('\u{104f9}', ['\u{104d1}', '\0', '\0']), ('\u{104fa}', ['\u{104d2}', '\0', '\0']), + ('\u{104fb}', ['\u{104d3}', '\0', '\0']), ('\u{10cc0}', ['\u{10c80}', '\0', '\0']), + ('\u{10cc1}', ['\u{10c81}', '\0', '\0']), ('\u{10cc2}', ['\u{10c82}', '\0', '\0']), + ('\u{10cc3}', ['\u{10c83}', '\0', '\0']), ('\u{10cc4}', ['\u{10c84}', '\0', '\0']), + ('\u{10cc5}', ['\u{10c85}', '\0', '\0']), ('\u{10cc6}', ['\u{10c86}', '\0', '\0']), + ('\u{10cc7}', ['\u{10c87}', '\0', '\0']), ('\u{10cc8}', ['\u{10c88}', '\0', '\0']), + ('\u{10cc9}', ['\u{10c89}', '\0', '\0']), ('\u{10cca}', ['\u{10c8a}', '\0', '\0']), + ('\u{10ccb}', ['\u{10c8b}', '\0', '\0']), ('\u{10ccc}', ['\u{10c8c}', '\0', '\0']), + ('\u{10ccd}', ['\u{10c8d}', '\0', '\0']), ('\u{10cce}', ['\u{10c8e}', '\0', '\0']), + ('\u{10ccf}', ['\u{10c8f}', '\0', '\0']), ('\u{10cd0}', ['\u{10c90}', '\0', '\0']), + ('\u{10cd1}', ['\u{10c91}', '\0', '\0']), ('\u{10cd2}', ['\u{10c92}', '\0', '\0']), + ('\u{10cd3}', ['\u{10c93}', '\0', '\0']), ('\u{10cd4}', ['\u{10c94}', '\0', '\0']), + ('\u{10cd5}', ['\u{10c95}', '\0', '\0']), ('\u{10cd6}', ['\u{10c96}', '\0', '\0']), + ('\u{10cd7}', ['\u{10c97}', '\0', '\0']), ('\u{10cd8}', ['\u{10c98}', '\0', '\0']), + ('\u{10cd9}', ['\u{10c99}', '\0', '\0']), ('\u{10cda}', ['\u{10c9a}', '\0', '\0']), + ('\u{10cdb}', ['\u{10c9b}', '\0', '\0']), ('\u{10cdc}', ['\u{10c9c}', '\0', '\0']), + ('\u{10cdd}', ['\u{10c9d}', '\0', '\0']), ('\u{10cde}', ['\u{10c9e}', '\0', '\0']), + ('\u{10cdf}', ['\u{10c9f}', '\0', '\0']), ('\u{10ce0}', ['\u{10ca0}', '\0', '\0']), + ('\u{10ce1}', ['\u{10ca1}', '\0', '\0']), ('\u{10ce2}', ['\u{10ca2}', '\0', '\0']), + ('\u{10ce3}', ['\u{10ca3}', '\0', '\0']), ('\u{10ce4}', ['\u{10ca4}', '\0', '\0']), + ('\u{10ce5}', ['\u{10ca5}', '\0', '\0']), ('\u{10ce6}', ['\u{10ca6}', '\0', '\0']), + ('\u{10ce7}', ['\u{10ca7}', '\0', '\0']), ('\u{10ce8}', ['\u{10ca8}', '\0', '\0']), + ('\u{10ce9}', ['\u{10ca9}', '\0', '\0']), ('\u{10cea}', ['\u{10caa}', '\0', '\0']), + ('\u{10ceb}', ['\u{10cab}', '\0', '\0']), ('\u{10cec}', ['\u{10cac}', '\0', '\0']), + ('\u{10ced}', ['\u{10cad}', '\0', '\0']), ('\u{10cee}', ['\u{10cae}', '\0', '\0']), + ('\u{10cef}', ['\u{10caf}', '\0', '\0']), ('\u{10cf0}', ['\u{10cb0}', '\0', '\0']), + ('\u{10cf1}', ['\u{10cb1}', '\0', '\0']), ('\u{10cf2}', ['\u{10cb2}', '\0', '\0']), + ('\u{118c0}', ['\u{118a0}', '\0', '\0']), ('\u{118c1}', ['\u{118a1}', '\0', '\0']), + ('\u{118c2}', ['\u{118a2}', '\0', '\0']), ('\u{118c3}', ['\u{118a3}', '\0', '\0']), + ('\u{118c4}', ['\u{118a4}', '\0', '\0']), ('\u{118c5}', ['\u{118a5}', '\0', '\0']), + ('\u{118c6}', ['\u{118a6}', '\0', '\0']), ('\u{118c7}', ['\u{118a7}', '\0', '\0']), + ('\u{118c8}', ['\u{118a8}', '\0', '\0']), ('\u{118c9}', ['\u{118a9}', '\0', '\0']), + ('\u{118ca}', ['\u{118aa}', '\0', '\0']), ('\u{118cb}', ['\u{118ab}', '\0', '\0']), + ('\u{118cc}', ['\u{118ac}', '\0', '\0']), ('\u{118cd}', ['\u{118ad}', '\0', '\0']), + ('\u{118ce}', ['\u{118ae}', '\0', '\0']), ('\u{118cf}', ['\u{118af}', '\0', '\0']), + ('\u{118d0}', ['\u{118b0}', '\0', '\0']), ('\u{118d1}', ['\u{118b1}', '\0', '\0']), + ('\u{118d2}', ['\u{118b2}', '\0', '\0']), ('\u{118d3}', ['\u{118b3}', '\0', '\0']), + ('\u{118d4}', ['\u{118b4}', '\0', '\0']), ('\u{118d5}', ['\u{118b5}', '\0', '\0']), + ('\u{118d6}', ['\u{118b6}', '\0', '\0']), ('\u{118d7}', ['\u{118b7}', '\0', '\0']), + ('\u{118d8}', ['\u{118b8}', '\0', '\0']), ('\u{118d9}', ['\u{118b9}', '\0', '\0']), + ('\u{118da}', ['\u{118ba}', '\0', '\0']), ('\u{118db}', ['\u{118bb}', '\0', '\0']), + ('\u{118dc}', ['\u{118bc}', '\0', '\0']), ('\u{118dd}', ['\u{118bd}', '\0', '\0']), + ('\u{118de}', ['\u{118be}', '\0', '\0']), ('\u{118df}', ['\u{118bf}', '\0', '\0']), + ('\u{16e60}', ['\u{16e40}', '\0', '\0']), ('\u{16e61}', ['\u{16e41}', '\0', '\0']), + ('\u{16e62}', ['\u{16e42}', '\0', '\0']), ('\u{16e63}', ['\u{16e43}', '\0', '\0']), + ('\u{16e64}', ['\u{16e44}', '\0', '\0']), ('\u{16e65}', ['\u{16e45}', '\0', '\0']), + ('\u{16e66}', ['\u{16e46}', '\0', '\0']), ('\u{16e67}', ['\u{16e47}', '\0', '\0']), + ('\u{16e68}', ['\u{16e48}', '\0', '\0']), ('\u{16e69}', ['\u{16e49}', '\0', '\0']), + ('\u{16e6a}', ['\u{16e4a}', '\0', '\0']), ('\u{16e6b}', ['\u{16e4b}', '\0', '\0']), + ('\u{16e6c}', ['\u{16e4c}', '\0', '\0']), ('\u{16e6d}', ['\u{16e4d}', '\0', '\0']), + ('\u{16e6e}', ['\u{16e4e}', '\0', '\0']), ('\u{16e6f}', ['\u{16e4f}', '\0', '\0']), + ('\u{16e70}', ['\u{16e50}', '\0', '\0']), ('\u{16e71}', ['\u{16e51}', '\0', '\0']), + ('\u{16e72}', ['\u{16e52}', '\0', '\0']), ('\u{16e73}', ['\u{16e53}', '\0', '\0']), + ('\u{16e74}', ['\u{16e54}', '\0', '\0']), ('\u{16e75}', ['\u{16e55}', '\0', '\0']), + ('\u{16e76}', ['\u{16e56}', '\0', '\0']), ('\u{16e77}', ['\u{16e57}', '\0', '\0']), + ('\u{16e78}', ['\u{16e58}', '\0', '\0']), ('\u{16e79}', ['\u{16e59}', '\0', '\0']), + ('\u{16e7a}', ['\u{16e5a}', '\0', '\0']), ('\u{16e7b}', ['\u{16e5b}', '\0', '\0']), + ('\u{16e7c}', ['\u{16e5c}', '\0', '\0']), ('\u{16e7d}', ['\u{16e5d}', '\0', '\0']), + ('\u{16e7e}', ['\u{16e5e}', '\0', '\0']), ('\u{16e7f}', ['\u{16e5f}', '\0', '\0']), + ('\u{1e922}', ['\u{1e900}', '\0', '\0']), ('\u{1e923}', ['\u{1e901}', '\0', '\0']), + ('\u{1e924}', ['\u{1e902}', '\0', '\0']), ('\u{1e925}', ['\u{1e903}', '\0', '\0']), + ('\u{1e926}', ['\u{1e904}', '\0', '\0']), ('\u{1e927}', ['\u{1e905}', '\0', '\0']), + ('\u{1e928}', ['\u{1e906}', '\0', '\0']), ('\u{1e929}', ['\u{1e907}', '\0', '\0']), + ('\u{1e92a}', ['\u{1e908}', '\0', '\0']), ('\u{1e92b}', ['\u{1e909}', '\0', '\0']), + ('\u{1e92c}', ['\u{1e90a}', '\0', '\0']), ('\u{1e92d}', ['\u{1e90b}', '\0', '\0']), + ('\u{1e92e}', ['\u{1e90c}', '\0', '\0']), ('\u{1e92f}', ['\u{1e90d}', '\0', '\0']), + ('\u{1e930}', ['\u{1e90e}', '\0', '\0']), ('\u{1e931}', ['\u{1e90f}', '\0', '\0']), + ('\u{1e932}', ['\u{1e910}', '\0', '\0']), ('\u{1e933}', ['\u{1e911}', '\0', '\0']), + ('\u{1e934}', ['\u{1e912}', '\0', '\0']), ('\u{1e935}', ['\u{1e913}', '\0', '\0']), + ('\u{1e936}', ['\u{1e914}', '\0', '\0']), ('\u{1e937}', ['\u{1e915}', '\0', '\0']), + ('\u{1e938}', ['\u{1e916}', '\0', '\0']), ('\u{1e939}', ['\u{1e917}', '\0', '\0']), + ('\u{1e93a}', ['\u{1e918}', '\0', '\0']), ('\u{1e93b}', ['\u{1e919}', '\0', '\0']), + ('\u{1e93c}', ['\u{1e91a}', '\0', '\0']), ('\u{1e93d}', ['\u{1e91b}', '\0', '\0']), + ('\u{1e93e}', ['\u{1e91c}', '\0', '\0']), ('\u{1e93f}', ['\u{1e91d}', '\0', '\0']), + ('\u{1e940}', ['\u{1e91e}', '\0', '\0']), ('\u{1e941}', ['\u{1e91f}', '\0', '\0']), + ('\u{1e942}', ['\u{1e920}', '\0', '\0']), ('\u{1e943}', ['\u{1e921}', '\0', '\0']) + ]; + +} + diff --git a/src/libcore/unicode/unicode.py b/src/libcore/unicode/unicode.py new file mode 100755 index 00000000000..28a1e01805e --- /dev/null +++ b/src/libcore/unicode/unicode.py @@ -0,0 +1,506 @@ +#!/usr/bin/env python +# +# Copyright 2011-2013 The Rust Project Developers. See the COPYRIGHT +# file at the top-level directory of this distribution and at +# http://rust-lang.org/COPYRIGHT. +# +# Licensed under the Apache License, Version 2.0 <LICENSE-APACHE or +# http://www.apache.org/licenses/LICENSE-2.0> or the MIT license +# <LICENSE-MIT or http://opensource.org/licenses/MIT>, at your +# option. This file may not be copied, modified, or distributed +# except according to those terms. + +# This script uses the following Unicode tables: +# - DerivedCoreProperties.txt +# - DerivedNormalizationProps.txt +# - EastAsianWidth.txt +# - auxiliary/GraphemeBreakProperty.txt +# - PropList.txt +# - ReadMe.txt +# - Scripts.txt +# - UnicodeData.txt +# +# Since this should not require frequent updates, we just store this +# out-of-line and check the tables.rs file into git. + +import fileinput, re, os, sys, operator, math, datetime + +# The directory in which this file resides. +fdir = os.path.dirname(os.path.realpath(__file__)) + "/" + +preamble = '''// Copyright 2012-{year} The Rust Project Developers. See the COPYRIGHT +// file at the top-level directory of this distribution and at +// http://rust-lang.org/COPYRIGHT. +// +// Licensed under the Apache License, Version 2.0 <LICENSE-APACHE or +// http://www.apache.org/licenses/LICENSE-2.0> or the MIT license +// <LICENSE-MIT or http://opensource.org/licenses/MIT>, at your +// option. This file may not be copied, modified, or distributed +// except according to those terms. + +// NOTE: The following code was generated by "./unicode.py", do not edit directly + +#![allow(missing_docs, non_upper_case_globals, non_snake_case)] + +use unicode::version::UnicodeVersion; +use unicode::bool_trie::{{BoolTrie, SmallBoolTrie}}; +'''.format(year = datetime.datetime.now().year) + +# Mapping taken from Table 12 from: +# http://www.unicode.org/reports/tr44/#General_Category_Values +expanded_categories = { + 'Lu': ['LC', 'L'], 'Ll': ['LC', 'L'], 'Lt': ['LC', 'L'], + 'Lm': ['L'], 'Lo': ['L'], + 'Mn': ['M'], 'Mc': ['M'], 'Me': ['M'], + 'Nd': ['N'], 'Nl': ['N'], 'No': ['N'], + 'Pc': ['P'], 'Pd': ['P'], 'Ps': ['P'], 'Pe': ['P'], + 'Pi': ['P'], 'Pf': ['P'], 'Po': ['P'], + 'Sm': ['S'], 'Sc': ['S'], 'Sk': ['S'], 'So': ['S'], + 'Zs': ['Z'], 'Zl': ['Z'], 'Zp': ['Z'], + 'Cc': ['C'], 'Cf': ['C'], 'Cs': ['C'], 'Co': ['C'], 'Cn': ['C'], +} + +# these are the surrogate codepoints, which are not valid rust characters +surrogate_codepoints = (0xd800, 0xdfff) + +def fetch(f): + path = fdir + os.path.basename(f) + if not os.path.exists(path): + os.system("curl -o {0}{1} ftp://ftp.unicode.org/Public/UNIDATA/{1}".format(fdir, f)) + + if not os.path.exists(path): + sys.stderr.write("cannot load %s" % f) + exit(1) + +def is_surrogate(n): + return surrogate_codepoints[0] <= n <= surrogate_codepoints[1] + +def load_unicode_data(f): + fetch(f) + gencats = {} + to_lower = {} + to_upper = {} + to_title = {} + combines = {} + canon_decomp = {} + compat_decomp = {} + + udict = {} + range_start = -1 + for line in fileinput.input(fdir + f): + data = line.split(';') + if len(data) != 15: + continue + cp = int(data[0], 16) + if is_surrogate(cp): + continue + if range_start >= 0: + for i in range(range_start, cp): + udict[i] = data + range_start = -1 + if data[1].endswith(", First>"): + range_start = cp + continue + udict[cp] = data + + for code in udict: + (code_org, name, gencat, combine, bidi, + decomp, deci, digit, num, mirror, + old, iso, upcase, lowcase, titlecase) = udict[code] + + # generate char to char direct common and simple conversions + # uppercase to lowercase + if lowcase != "" and code_org != lowcase: + to_lower[code] = (int(lowcase, 16), 0, 0) + + # lowercase to uppercase + if upcase != "" and code_org != upcase: + to_upper[code] = (int(upcase, 16), 0, 0) + + # title case + if titlecase.strip() != "" and code_org != titlecase: + to_title[code] = (int(titlecase, 16), 0, 0) + + # store decomposition, if given + if decomp != "": + if decomp.startswith('<'): + seq = [] + for i in decomp.split()[1:]: + seq.append(int(i, 16)) + compat_decomp[code] = seq + else: + seq = [] + for i in decomp.split(): + seq.append(int(i, 16)) + canon_decomp[code] = seq + + # place letter in categories as appropriate + for cat in [gencat, "Assigned"] + expanded_categories.get(gencat, []): + if cat not in gencats: + gencats[cat] = [] + gencats[cat].append(code) + + # record combining class, if any + if combine != "0": + if combine not in combines: + combines[combine] = [] + combines[combine].append(code) + + # generate Not_Assigned from Assigned + gencats["Cn"] = gen_unassigned(gencats["Assigned"]) + # Assigned is not a real category + del(gencats["Assigned"]) + # Other contains Not_Assigned + gencats["C"].extend(gencats["Cn"]) + gencats = group_cats(gencats) + combines = to_combines(group_cats(combines)) + + return (canon_decomp, compat_decomp, gencats, combines, to_upper, to_lower, to_title) + +def load_special_casing(f, to_upper, to_lower, to_title): + fetch(f) + for line in fileinput.input(fdir + f): + data = line.split('#')[0].split(';') + if len(data) == 5: + code, lower, title, upper, _comment = data + elif len(data) == 6: + code, lower, title, upper, condition, _comment = data + if condition.strip(): # Only keep unconditional mappins + continue + else: + continue + code = code.strip() + lower = lower.strip() + title = title.strip() + upper = upper.strip() + key = int(code, 16) + for (map_, values) in [(to_lower, lower), (to_upper, upper), (to_title, title)]: + if values != code: + values = [int(i, 16) for i in values.split()] + for _ in range(len(values), 3): + values.append(0) + assert len(values) == 3 + map_[key] = values + +def group_cats(cats): + cats_out = {} + for cat in cats: + cats_out[cat] = group_cat(cats[cat]) + return cats_out + +def group_cat(cat): + cat_out = [] + letters = sorted(set(cat)) + cur_start = letters.pop(0) + cur_end = cur_start + for letter in letters: + assert letter > cur_end, \ + "cur_end: %s, letter: %s" % (hex(cur_end), hex(letter)) + if letter == cur_end + 1: + cur_end = letter + else: + cat_out.append((cur_start, cur_end)) + cur_start = cur_end = letter + cat_out.append((cur_start, cur_end)) + return cat_out + +def ungroup_cat(cat): + cat_out = [] + for (lo, hi) in cat: + while lo <= hi: + cat_out.append(lo) + lo += 1 + return cat_out + +def gen_unassigned(assigned): + assigned = set(assigned) + return ([i for i in range(0, 0xd800) if i not in assigned] + + [i for i in range(0xe000, 0x110000) if i not in assigned]) + +def to_combines(combs): + combs_out = [] + for comb in combs: + for (lo, hi) in combs[comb]: + combs_out.append((lo, hi, comb)) + combs_out.sort(key=lambda comb: comb[0]) + return combs_out + +def format_table_content(f, content, indent): + line = " "*indent + first = True + for chunk in content.split(","): + if len(line) + len(chunk) < 98: + if first: + line += chunk + else: + line += ", " + chunk + first = False + else: + f.write(line + ",\n") + line = " "*indent + chunk + f.write(line) + +def load_properties(f, interestingprops): + fetch(f) + props = {} + re1 = re.compile("^ *([0-9A-F]+) *; *(\w+)") + re2 = re.compile("^ *([0-9A-F]+)\.\.([0-9A-F]+) *; *(\w+)") + + for line in fileinput.input(fdir + os.path.basename(f)): + prop = None + d_lo = 0 + d_hi = 0 + m = re1.match(line) + if m: + d_lo = m.group(1) + d_hi = m.group(1) + prop = m.group(2) + else: + m = re2.match(line) + if m: + d_lo = m.group(1) + d_hi = m.group(2) + prop = m.group(3) + else: + continue + if interestingprops and prop not in interestingprops: + continue + d_lo = int(d_lo, 16) + d_hi = int(d_hi, 16) + if prop not in props: + props[prop] = [] + props[prop].append((d_lo, d_hi)) + + # optimize if possible + for prop in props: + props[prop] = group_cat(ungroup_cat(props[prop])) + + return props + +def escape_char(c): + return "'\\u{%x}'" % c if c != 0 else "'\\0'" + +def emit_table(f, name, t_data, t_type = "&[(char, char)]", is_pub=True, + pfun=lambda x: "(%s,%s)" % (escape_char(x[0]), escape_char(x[1]))): + pub_string = "" + if is_pub: + pub_string = "pub " + f.write(" %sconst %s: %s = &[\n" % (pub_string, name, t_type)) + data = "" + first = True + for dat in t_data: + if not first: + data += "," + first = False + data += pfun(dat) + format_table_content(f, data, 8) + f.write("\n ];\n\n") + +def compute_trie(rawdata, chunksize): + root = [] + childmap = {} + child_data = [] + for i in range(len(rawdata) // chunksize): + data = rawdata[i * chunksize: (i + 1) * chunksize] + child = '|'.join(map(str, data)) + if child not in childmap: + childmap[child] = len(childmap) + child_data.extend(data) + root.append(childmap[child]) + return (root, child_data) + +def emit_bool_trie(f, name, t_data, is_pub=True): + CHUNK = 64 + rawdata = [False] * 0x110000 + for (lo, hi) in t_data: + for cp in range(lo, hi + 1): + rawdata[cp] = True + + # convert to bitmap chunks of 64 bits each + chunks = [] + for i in range(0x110000 // CHUNK): + chunk = 0 + for j in range(64): + if rawdata[i * 64 + j]: + chunk |= 1 << j + chunks.append(chunk) + + pub_string = "" + if is_pub: + pub_string = "pub " + f.write(" %sconst %s: &super::BoolTrie = &super::BoolTrie {\n" % (pub_string, name)) + f.write(" r1: [\n") + data = ','.join('0x%016x' % chunk for chunk in chunks[0:0x800 // CHUNK]) + format_table_content(f, data, 12) + f.write("\n ],\n") + + # 0x800..0x10000 trie + (r2, r3) = compute_trie(chunks[0x800 // CHUNK : 0x10000 // CHUNK], 64 // CHUNK) + f.write(" r2: [\n") + data = ','.join(str(node) for node in r2) + format_table_content(f, data, 12) + f.write("\n ],\n") + f.write(" r3: &[\n") + data = ','.join('0x%016x' % chunk for chunk in r3) + format_table_content(f, data, 12) + f.write("\n ],\n") + + # 0x10000..0x110000 trie + (mid, r6) = compute_trie(chunks[0x10000 // CHUNK : 0x110000 // CHUNK], 64 // CHUNK) + (r4, r5) = compute_trie(mid, 64) + f.write(" r4: [\n") + data = ','.join(str(node) for node in r4) + format_table_content(f, data, 12) + f.write("\n ],\n") + f.write(" r5: &[\n") + data = ','.join(str(node) for node in r5) + format_table_content(f, data, 12) + f.write("\n ],\n") + f.write(" r6: &[\n") + data = ','.join('0x%016x' % chunk for chunk in r6) + format_table_content(f, data, 12) + f.write("\n ],\n") + + f.write(" };\n\n") + +def emit_small_bool_trie(f, name, t_data, is_pub=True): + last_chunk = max(hi // 64 for (lo, hi) in t_data) + n_chunks = last_chunk + 1 + chunks = [0] * n_chunks + for (lo, hi) in t_data: + for cp in range(lo, hi + 1): + if cp // 64 >= len(chunks): + print(cp, cp // 64, len(chunks), lo, hi) + chunks[cp // 64] |= 1 << (cp & 63) + + pub_string = "" + if is_pub: + pub_string = "pub " + f.write(" %sconst %s: &super::SmallBoolTrie = &super::SmallBoolTrie {\n" + % (pub_string, name)) + + (r1, r2) = compute_trie(chunks, 1) + + f.write(" r1: &[\n") + data = ','.join(str(node) for node in r1) + format_table_content(f, data, 12) + f.write("\n ],\n") + + f.write(" r2: &[\n") + data = ','.join('0x%016x' % node for node in r2) + format_table_content(f, data, 12) + f.write("\n ],\n") + + f.write(" };\n\n") + +def emit_property_module(f, mod, tbl, emit): + f.write("pub mod %s {\n" % mod) + for cat in sorted(emit): + if cat in ["Cc", "White_Space", "Pattern_White_Space"]: + emit_small_bool_trie(f, "%s_table" % cat, tbl[cat]) + f.write(" pub fn %s(c: char) -> bool {\n" % cat) + f.write(" %s_table.lookup(c)\n" % cat) + f.write(" }\n\n") + else: + emit_bool_trie(f, "%s_table" % cat, tbl[cat]) + f.write(" pub fn %s(c: char) -> bool {\n" % cat) + f.write(" %s_table.lookup(c)\n" % cat) + f.write(" }\n\n") + f.write("}\n\n") + +def emit_conversions_module(f, to_upper, to_lower, to_title): + f.write("pub mod conversions {") + f.write(""" + pub fn to_lower(c: char) -> [char; 3] { + match bsearch_case_table(c, to_lowercase_table) { + None => [c, '\\0', '\\0'], + Some(index) => to_lowercase_table[index].1, + } + } + + pub fn to_upper(c: char) -> [char; 3] { + match bsearch_case_table(c, to_uppercase_table) { + None => [c, '\\0', '\\0'], + Some(index) => to_uppercase_table[index].1, + } + } + + fn bsearch_case_table(c: char, table: &[(char, [char; 3])]) -> Option<usize> { + table.binary_search_by(|&(key, _)| key.cmp(&c)).ok() + } + +""") + t_type = "&[(char, [char; 3])]" + pfun = lambda x: "(%s,[%s,%s,%s])" % ( + escape_char(x[0]), escape_char(x[1][0]), escape_char(x[1][1]), escape_char(x[1][2])) + emit_table(f, "to_lowercase_table", + sorted(to_lower.items(), key=operator.itemgetter(0)), + is_pub=False, t_type = t_type, pfun=pfun) + emit_table(f, "to_uppercase_table", + sorted(to_upper.items(), key=operator.itemgetter(0)), + is_pub=False, t_type = t_type, pfun=pfun) + f.write("}\n\n") + +def emit_norm_module(f, canon, compat, combine, norm_props): + canon_keys = sorted(canon.keys()) + + compat_keys = sorted(compat.keys()) + + canon_comp = {} + comp_exclusions = norm_props["Full_Composition_Exclusion"] + for char in canon_keys: + if any(lo <= char <= hi for lo, hi in comp_exclusions): + continue + decomp = canon[char] + if len(decomp) == 2: + if decomp[0] not in canon_comp: + canon_comp[decomp[0]] = [] + canon_comp[decomp[0]].append( (decomp[1], char) ) + canon_comp_keys = sorted(canon_comp.keys()) + +if __name__ == "__main__": + r = fdir + "tables.rs" + if os.path.exists(r): + os.remove(r) + with open(r, "w") as rf: + # write the file's preamble + rf.write(preamble) + + # download and parse all the data + fetch("ReadMe.txt") + with open(fdir + "ReadMe.txt") as readme: + pattern = "for Version (\d+)\.(\d+)\.(\d+) of the Unicode" + unicode_version = re.search(pattern, readme.read()).groups() + rf.write(""" +/// The version of [Unicode](http://www.unicode.org/) that the Unicode parts of +/// `char` and `str` methods are based on. +#[unstable(feature = "unicode_version", issue = "49726")] +pub const UNICODE_VERSION: UnicodeVersion = UnicodeVersion { + major: %s, + minor: %s, + micro: %s, + _priv: (), +}; +""" % unicode_version) + (canon_decomp, compat_decomp, gencats, combines, + to_upper, to_lower, to_title) = load_unicode_data("UnicodeData.txt") + load_special_casing("SpecialCasing.txt", to_upper, to_lower, to_title) + want_derived = ["XID_Start", "XID_Continue", "Alphabetic", "Lowercase", "Uppercase", + "Cased", "Case_Ignorable", "Grapheme_Extend"] + derived = load_properties("DerivedCoreProperties.txt", want_derived) + scripts = load_properties("Scripts.txt", []) + props = load_properties("PropList.txt", + ["White_Space", "Join_Control", "Noncharacter_Code_Point", "Pattern_White_Space"]) + norm_props = load_properties("DerivedNormalizationProps.txt", + ["Full_Composition_Exclusion"]) + + # category tables + for (name, cat, pfuns) in ("general_category", gencats, ["N", "Cc"]), \ + ("derived_property", derived, want_derived), \ + ("property", props, ["White_Space", "Pattern_White_Space"]): + emit_property_module(rf, name, cat, pfuns) + + # normalizations and conversions module + emit_norm_module(rf, canon_decomp, compat_decomp, combines, norm_props) + emit_conversions_module(rf, to_upper, to_lower, to_title) + print("Regenerated tables.rs.") diff --git a/src/libcore/unicode/version.rs b/src/libcore/unicode/version.rs new file mode 100644 index 00000000000..59ebf5f5012 --- /dev/null +++ b/src/libcore/unicode/version.rs @@ -0,0 +1,28 @@ +// Copyright 2012-2014 The Rust Project Developers. See the COPYRIGHT +// file at the top-level directory of this distribution and at +// http://rust-lang.org/COPYRIGHT. +// +// Licensed under the Apache License, Version 2.0 <LICENSE-APACHE or +// http://www.apache.org/licenses/LICENSE-2.0> or the MIT license +// <LICENSE-MIT or http://opensource.org/licenses/MIT>, at your +// option. This file may not be copied, modified, or distributed +// except according to those terms. + +/// Represents a Unicode Version. +/// +/// See also: <http://www.unicode.org/versions/> +#[derive(Clone, Copy, Debug, Eq, Ord, PartialEq, PartialOrd)] +#[unstable(feature = "unicode_version", issue = "49726")] +pub struct UnicodeVersion { + /// Major version. + pub major: u32, + + /// Minor version. + pub minor: u32, + + /// Micro (or Update) version. + pub micro: u32, + + // Private field to keep struct expandable. + pub(crate) _priv: (), +} |
