auto merge of #20395 : huonw/rust/char-stab-2, r=aturon

cc #19260 

The casing transformations are left unstable (it is highly likely to be better to adopt the proper non-1-to-1 case mappings, per #20333) as are `is_xid_*`.

I've got a little todo list in the last commit of things I thought about/was told about that I haven't yet handled (I'd also like some feedback).

4 files changed, 136 insertions, 17 deletions

diff --git a/src/libunicode/lib.rs b/src/libunicode/lib.rs
index 72e9ce2bcaf..a3884d0c86e 100644
--- a/src/libunicode/lib.rs
+++ b/src/libunicode/lib.rs
@@ -44,9 +44,9 @@ mod u_str;
 // re-export char so that std et al see it correctly
 /// Character manipulation (`char` type, Unicode Scalar Value)
 ///
-/// This module  provides the `Char` and `UnicodeChar` traits, as well as their
-/// implementation for the primitive `char` type, in order to allow basic character
-/// manipulation.
+/// This module provides the `CharExt` trait, as well as its
+/// implementation for the primitive `char` type, in order to allow
+/// basic character manipulation.
 ///
 /// A `char` actually represents a
 /// *[Unicode Scalar Value](http://www.unicode.org/glossary/#unicode_scalar_value)*,
@@ -58,16 +58,14 @@ mod u_str;
 /// however the converse is not always true due to the above range limits
 /// and, as such, should be performed via the `from_u32` function..
 pub mod char {
-    pub use core::char::{MAX, from_u32};
-    pub use core::char::{from_digit};
-    pub use core::char::Char;
+    pub use core::char::{MAX, from_u32, from_digit};
 
     pub use normalize::{decompose_canonical, decompose_compatible, compose};
 
     pub use tables::normalization::canonical_combining_class;
     pub use tables::UNICODE_VERSION;
 
-    pub use u_char::UnicodeChar;
+    pub use u_char::CharExt;
 }
 
 pub mod str {
diff --git a/src/libunicode/tables.rs b/src/libunicode/tables.rs
index e3550810010..c755ea93184 100644
--- a/src/libunicode/tables.rs
+++ b/src/libunicode/tables.rs
@@ -13,7 +13,7 @@
 #![allow(missing_docs, non_upper_case_globals, non_snake_case)]
 
 /// The version of [Unicode](http://www.unicode.org/)
-/// that the `UnicodeChar` and `UnicodeStrPrelude` traits are based on.
+/// that the unicode parts of `CharExt` and `UnicodeStrPrelude` traits are based on.
 pub const UNICODE_VERSION: (uint, uint, uint) = (7, 0, 0);
 
 fn bsearch_range_table(c: char, r: &'static [(char,char)]) -> bool {
diff --git a/src/libunicode/u_char.rs b/src/libunicode/u_char.rs
index 9c356801604..5693c222de1 100644
--- a/src/libunicode/u_char.rs
+++ b/src/libunicode/u_char.rs
@@ -8,19 +8,102 @@
 // option. This file may not be copied, modified, or distributed
 // except according to those terms.
 
-//! Unicode-intensive `char` methods.
+//! Unicode-intensive `char` methods along with the `core` methods.
 //!
 //! These methods implement functionality for `char` that requires knowledge of
 //! Unicode definitions, including normalization, categorization, and display information.
 
+use core::char;
+use core::char::CharExt as C;
 use core::option::Option;
 use tables::{derived_property, property, general_category, conversions, charwidth};
 
-/// Useful functions for Unicode characters.
-#[experimental = "pending prelude organization"]
-pub trait UnicodeChar {
+/// Functionality for manipulating `char`.
+#[stable]
+pub trait CharExt {
+    /// Checks if a `char` parses as a numeric digit in the given radix.
+    ///
+    /// Compared to `is_numeric()`, this function only recognizes the characters
+    /// `0-9`, `a-z` and `A-Z`.
+    ///
+    /// # Return value
+    ///
+    /// Returns `true` if `c` is a valid digit under `radix`, and `false`
+    /// otherwise.
+    ///
+    /// # Panics
+    ///
+    /// Panics if given a radix > 36.
+    #[unstable = "pending integer conventions"]
+    fn is_digit(self, radix: uint) -> bool;
+
+    /// Converts a character to the corresponding digit.
+    ///
+    /// # Return value
+    ///
+    /// If `c` is between '0' and '9', the corresponding value between 0 and
+    /// 9. If `c` is 'a' or 'A', 10. If `c` is 'b' or 'B', 11, etc. Returns
+    /// none if the character does not refer to a digit in the given radix.
+    ///
+    /// # Panics
+    ///
+    /// Panics if given a radix outside the range [0..36].
+    #[unstable = "pending integer conventions"]
+    fn to_digit(self, radix: uint) -> Option<uint>;
+
+    /// Returns an iterator that yields the hexadecimal Unicode escape
+    /// of a character, as `char`s.
+    ///
+    /// All characters are escaped with Rust syntax of the form `\\u{NNNN}`
+    /// where `NNNN` is the shortest hexadecimal representation of the code
+    /// point.
+    #[stable]
+    fn escape_unicode(self) -> char::EscapeUnicode;
+
+    /// Returns an iterator that yields the 'default' ASCII and
+    /// C++11-like literal escape 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, CR and LF are escaped as '\t', '\r' and '\n' respectively.
+    /// * Single-quote, double-quote and backslash chars are backslash-
+    ///   escaped.
+    /// * Any other chars in the range [0x20,0x7e] are not escaped.
+    /// * Any other chars are given hex Unicode escapes; see `escape_unicode`.
+    #[stable]
+    fn escape_default(self) -> char::EscapeDefault;
+
+    /// Returns the amount of bytes this character would need if encoded in
+    /// UTF-8.
+    #[stable]
+    fn len_utf8(self) -> uint;
+
+    /// Returns the amount of bytes this character would need if encoded in
+    /// UTF-16.
+    #[stable]
+    fn len_utf16(self) -> uint;
+
+    /// Encodes this character as UTF-8 into the provided byte buffer,
+    /// and then returns the number of bytes written.
+    ///
+    /// If the buffer is not large enough, nothing will be written into it
+    /// and a `None` will be returned.
+    #[unstable = "pending decision about Iterator/Writer/Reader"]
+    fn encode_utf8(self, dst: &mut [u8]) -> Option<uint>;
+
+    /// Encodes this character as UTF-16 into the provided `u16` buffer,
+    /// and then returns the number of `u16`s written.
+    ///
+    /// If the buffer is not large enough, nothing will be written into it
+    /// and a `None` will be returned.
+    #[unstable = "pending decision about Iterator/Writer/Reader"]
+    fn encode_utf16(self, dst: &mut [u16]) -> Option<uint>;
+
     /// Returns whether the specified character is considered a Unicode
     /// alphabetic code point.
+    #[stable]
     fn is_alphabetic(self) -> bool;
 
     /// Returns whether the specified character satisfies the 'XID_Start'
@@ -29,6 +112,7 @@ pub trait UnicodeChar {
     /// '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.
+    #[experimental = "mainly needed for compiler internals"]
     fn is_xid_start(self) -> bool;
 
     /// Returns whether the specified `char` satisfies the 'XID_Continue'
@@ -37,38 +121,45 @@ pub trait UnicodeChar {
     /// '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.
+    #[experimental = "mainly needed for compiler internals"]
     fn is_xid_continue(self) -> bool;
 
     /// Indicates whether a character is in lowercase.
     ///
     /// This is defined according to the terms of the Unicode Derived Core
     /// Property `Lowercase`.
+    #[stable]
     fn is_lowercase(self) -> bool;
 
     /// Indicates whether a character is in uppercase.
     ///
     /// This is defined according to the terms of the Unicode Derived Core
     /// Property `Uppercase`.
+    #[stable]
     fn is_uppercase(self) -> bool;
 
     /// Indicates whether a character is whitespace.
     ///
     /// Whitespace is defined in terms of the Unicode Property `White_Space`.
+    #[stable]
     fn is_whitespace(self) -> bool;
 
     /// Indicates whether a character is alphanumeric.
     ///
     /// Alphanumericness is defined in terms of the Unicode General Categories
     /// 'Nd', 'Nl', 'No' and the Derived Core Property 'Alphabetic'.
+    #[stable]
     fn is_alphanumeric(self) -> bool;
 
     /// Indicates whether a character is a control code point.
     ///
     /// Control code points are defined in terms of the Unicode General
     /// Category `Cc`.
+    #[stable]
     fn is_control(self) -> bool;
 
     /// Indicates whether the character is numeric (Nd, Nl, or No).
+    #[stable]
     fn is_numeric(self) -> bool;
 
     /// Converts a character to its lowercase equivalent.
@@ -80,6 +171,7 @@ pub trait UnicodeChar {
     ///
     /// Returns the lowercase equivalent of the character, or the character
     /// itself if no conversion is possible.
+    #[experimental = "pending case transformation decisions"]
     fn to_lowercase(self) -> char;
 
     /// Converts a character to its uppercase equivalent.
@@ -102,6 +194,7 @@ pub trait UnicodeChar {
     /// [`SpecialCasing`.txt`]: ftp://ftp.unicode.org/Public/UNIDATA/SpecialCasing.txt
     ///
     /// [2]: http://www.unicode.org/versions/Unicode4.0.0/ch03.pdf#G33992
+    #[experimental = "pending case transformation decisions"]
     fn to_uppercase(self) -> char;
 
     /// Returns this character's displayed width in columns, or `None` if it is a
@@ -117,8 +210,26 @@ pub trait UnicodeChar {
     fn width(self, is_cjk: bool) -> Option<uint>;
 }
 
-#[experimental = "pending prelude organization"]
-impl UnicodeChar for char {
+#[stable]
+impl CharExt for char {
+    #[unstable = "pending integer conventions"]
+    fn is_digit(self, radix: uint) -> bool { C::is_digit(self, radix) }
+    #[unstable = "pending integer conventions"]
+    fn to_digit(self, radix: uint) -> Option<uint> { C::to_digit(self, radix) }
+    #[stable]
+    fn escape_unicode(self) -> char::EscapeUnicode { C::escape_unicode(self) }
+    #[stable]
+    fn escape_default(self) -> char::EscapeDefault { C::escape_default(self) }
+    #[stable]
+    fn len_utf8(self) -> uint { C::len_utf8(self) }
+    #[stable]
+    fn len_utf16(self) -> uint { C::len_utf16(self) }
+    #[unstable = "pending decision about Iterator/Writer/Reader"]
+    fn encode_utf8(self, dst: &mut [u8]) -> Option<uint> { C::encode_utf8(self, dst) }
+    #[unstable = "pending decision about Iterator/Writer/Reader"]
+    fn encode_utf16(self, dst: &mut [u16]) -> Option<uint> { C::encode_utf16(self, dst) }
+
+    #[stable]
     fn is_alphabetic(self) -> bool {
         match self {
             'a' ... 'z' | 'A' ... 'Z' => true,
@@ -127,10 +238,13 @@ impl UnicodeChar for char {
         }
     }
 
+    #[experimental = "mainly needed for compiler internals"]
     fn is_xid_start(self) -> bool { derived_property::XID_Start(self) }
 
+    #[experimental = "mainly needed for compiler internals"]
     fn is_xid_continue(self) -> bool { derived_property::XID_Continue(self) }
 
+    #[stable]
     fn is_lowercase(self) -> bool {
         match self {
             'a' ... 'z' => true,
@@ -139,6 +253,7 @@ impl UnicodeChar for char {
         }
     }
 
+    #[stable]
     fn is_uppercase(self) -> bool {
         match self {
             'A' ... 'Z' => true,
@@ -147,6 +262,7 @@ impl UnicodeChar for char {
         }
     }
 
+    #[stable]
     fn is_whitespace(self) -> bool {
         match self {
             ' ' | '\x09' ... '\x0d' => true,
@@ -155,12 +271,15 @@ impl UnicodeChar for char {
         }
     }
 
+    #[stable]
     fn is_alphanumeric(self) -> bool {
         self.is_alphabetic() || self.is_numeric()
     }
 
+    #[stable]
     fn is_control(self) -> bool { general_category::Cc(self) }
 
+    #[stable]
     fn is_numeric(self) -> bool {
         match self {
             '0' ... '9' => true,
@@ -169,8 +288,10 @@ impl UnicodeChar for char {
         }
     }
 
+    #[experimental = "pending case transformation decisions"]
     fn to_lowercase(self) -> char { conversions::to_lower(self) }
 
+    #[experimental = "pending case transformation decisions"]
     fn to_uppercase(self) -> char { conversions::to_upper(self) }
 
     #[experimental = "needs expert opinion. is_cjk flag stands out as ugly"]
diff --git a/src/libunicode/u_str.rs b/src/libunicode/u_str.rs
index 1b0c4171134..90949437774 100644
--- a/src/libunicode/u_str.rs
+++ b/src/libunicode/u_str.rs
@@ -13,7 +13,7 @@
 //! Unicode-intensive string manipulations.
 //!
 //! This module provides functionality to `str` that requires the Unicode methods provided by the
-//! UnicodeChar trait.
+//! unicode parts of the CharExt trait.
 
 use self::GraphemeState::*;
 use core::prelude::*;
@@ -26,7 +26,7 @@ use core::num::Int;
 use core::slice;
 use core::str::Split;
 
-use u_char::UnicodeChar;
+use u_char::CharExt as UCharExt; // conflicts with core::prelude::CharExt
 use tables::grapheme::GraphemeCat;
 
 /// An iterator over the words of a string, separated by a sequence of whitespace
@@ -529,7 +529,7 @@ impl<I> Iterator for Utf16Encoder<I> where I: Iterator<Item=char> {
 
         let mut buf = [0u16; 2];
         self.chars.next().map(|ch| {
-            let n = ch.encode_utf16(buf.as_mut_slice()).unwrap_or(0);
+            let n = CharExt::encode_utf16(ch, buf.as_mut_slice()).unwrap_or(0);
             if n == 2 { self.extra = buf[1]; }
             buf[0]
         })