Add initial support for a new formatting syntax

The new macro is available under the name ifmt! (only an intermediate name)

6 files changed, 1331 insertions, 2 deletions

diff --git a/src/libstd/either.rs b/src/libstd/either.rs
index cfaef550c6f..bb74d9b3ec4 100644
--- a/src/libstd/either.rs
+++ b/src/libstd/either.rs
@@ -24,7 +24,7 @@ use vec;
 use vec::{OwnedVector, ImmutableVector};
 
 /// `Either` is a type that represents one of two alternatives
-#[deriving(Clone, Eq)]
+#[deriving(Clone, Eq, IterBytes)]
 pub enum Either<L, R> {
     Left(L),
     Right(R)
diff --git a/src/libstd/fmt/mod.rs b/src/libstd/fmt/mod.rs
new file mode 100644
index 00000000000..2b8807b2291
--- /dev/null
+++ b/src/libstd/fmt/mod.rs
@@ -0,0 +1,368 @@
+// Copyright 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.
+
+use prelude::*;
+
+use cast;
+use int;
+use rt::io::Decorator;
+use rt::io::mem::MemWriter;
+use rt::io;
+use str;
+use sys;
+use uint;
+use util;
+use vec;
+
+pub mod parse;
+pub mod rt;
+
+/// A struct to represent both where to emit formatting strings to and how they
+/// should be formatted. A mutable version of this is passed to all formatting
+/// traits.
+pub struct Formatter<'self> {
+    /// Flags for formatting (packed version of rt::Flag)
+    flags: uint,
+    /// Character used as 'fill' whenever there is alignment
+    fill: char,
+    /// Boolean indication of whether the output should be left-aligned
+    alignleft: bool,
+    /// Optionally specified integer width that the output should be
+    width: Option<uint>,
+    /// Optionally specified precision for numeric types
+    precision: Option<uint>,
+
+    /// Output buffer.
+    buf: &'self mut io::Writer,
+
+    priv curarg: vec::VecIterator<'self, Argument<'self>>,
+    priv args: &'self [Argument<'self>],
+}
+
+/// This struct represents the generic "argument" which is taken by the Xprintf
+/// family of functions. It contains a function to format the given value. At
+/// compile time it is ensured that the function and the value have the correct
+/// types, and then this struct is used to canonicalize arguments to one type.
+pub struct Argument<'self> {
+    priv formatter: extern "Rust" fn(&util::Void, &mut Formatter),
+    priv value: &'self util::Void,
+}
+
+#[allow(missing_doc)]
+pub trait Bool { fn fmt(&Self, &mut Formatter); }
+#[allow(missing_doc)]
+pub trait Char { fn fmt(&Self, &mut Formatter); }
+#[allow(missing_doc)]
+pub trait Signed { fn fmt(&Self, &mut Formatter); }
+#[allow(missing_doc)]
+pub trait Unsigned { fn fmt(&Self, &mut Formatter); }
+#[allow(missing_doc)]
+pub trait Octal { fn fmt(&Self, &mut Formatter); }
+#[allow(missing_doc)]
+pub trait Binary { fn fmt(&Self, &mut Formatter); }
+#[allow(missing_doc)]
+pub trait LowerHex { fn fmt(&Self, &mut Formatter); }
+#[allow(missing_doc)]
+pub trait UpperHex { fn fmt(&Self, &mut Formatter); }
+#[allow(missing_doc)]
+pub trait String { fn fmt(&Self, &mut Formatter); }
+#[allow(missing_doc)]
+pub trait Poly { fn fmt(&Self, &mut Formatter); }
+#[allow(missing_doc)]
+pub trait Pointer { fn fmt(&Self, &mut Formatter); }
+
+/// The sprintf function takes a precompiled format string and a list of
+/// arguments, to return the resulting formatted string.
+///
+/// This is currently an unsafe function because the types of all arguments
+/// aren't verified by immediate callers of this function. This currently does
+/// not validate that the correct types of arguments are specified for each
+/// format specifier, nor that each argument itself contains the right function
+/// for formatting the right type value. Because of this, the function is marked
+/// as `unsafe` if this is being called manually.
+///
+/// Thankfully the rust compiler provides the macro `ifmt!` which will perform
+/// all of this validation at compile-time and provides a safe interface for
+/// invoking this function.
+///
+/// # Arguments
+///
+///   * fmts - the precompiled format string to emit.
+///   * args - the list of arguments to the format string. These are only the
+///            positional arguments (not named)
+///
+/// Note that this function assumes that there are enough arguments for the
+/// format string.
+pub unsafe fn sprintf(fmt: &[rt::Piece], args: &[Argument]) -> ~str {
+    let output = MemWriter::new();
+    {
+        let mut formatter = Formatter {
+            flags: 0,
+            width: None,
+            precision: None,
+            // FIXME(#8248): shouldn't need a transmute
+            buf: cast::transmute(&output as &io::Writer),
+            alignleft: false,
+            fill: ' ',
+            args: args,
+            curarg: args.iter(),
+        };
+        for piece in fmt.iter() {
+            formatter.run(piece, None);
+        }
+    }
+    return str::from_bytes_owned(output.inner());
+}
+
+impl<'self> Formatter<'self> {
+    fn run(&mut self, piece: &rt::Piece, cur: Option<&str>) {
+        let setcount = |slot: &mut Option<uint>, cnt: &parse::Count| {
+            match *cnt {
+                parse::CountIs(n) => { *slot = Some(n); }
+                parse::CountImplied => { *slot = None; }
+                parse::CountIsParam(i) => {
+                    let v = self.args[i].value;
+                    unsafe { *slot = Some(*(v as *util::Void as *uint)); }
+                }
+                parse::CountIsNextParam => {
+                    let v = self.curarg.next().unwrap().value;
+                    unsafe { *slot = Some(*(v as *util::Void as *uint)); }
+                }
+            }
+        };
+
+        match *piece {
+            rt::String(s) => { self.buf.write(s.as_bytes()); }
+            rt::CurrentArgument(()) => { self.buf.write(cur.unwrap().as_bytes()); }
+            rt::Argument(ref arg) => {
+                // Fill in the format parameters into the formatter
+                self.fill = arg.format.fill;
+                self.alignleft = arg.format.alignleft;
+                self.flags = arg.format.flags;
+                setcount(&mut self.width, &arg.format.width);
+                setcount(&mut self.precision, &arg.format.precision);
+
+                // Extract the correct argument
+                let value = match arg.position {
+                    rt::ArgumentNext => { *self.curarg.next().unwrap() }
+                    rt::ArgumentIs(i) => self.args[i],
+                };
+
+                // Then actually do some printing
+                match arg.method {
+                    None => { (value.formatter)(value.value, self); }
+                    Some(ref method) => { self.execute(*method, value); }
+                }
+            }
+        }
+    }
+
+    fn execute(&mut self, method: &rt::Method, arg: Argument) {
+        match *method {
+            // Pluralization is selection upon a numeric value specified as the
+            // parameter.
+            rt::Plural(offset, ref selectors, ref default) => {
+                // This is validated at compile-time to be a pointer to a
+                // '&uint' value.
+                let value: &uint = unsafe { cast::transmute(arg.value) };
+                let value = *value;
+
+                // First, attempt to match against explicit values without the
+                // offsetted value
+                for s in selectors.iter() {
+                    match s.selector {
+                        Right(val) if value == val => {
+                            return self.runplural(value, s.result);
+                        }
+                        _ => {}
+                    }
+                }
+
+                // Next, offset the value and attempt to match against the
+                // keyword selectors.
+                let value = value - match offset { Some(i) => i, None => 0 };
+                for s in selectors.iter() {
+                    let run = match s.selector {
+                        Left(parse::Zero) => value == 0,
+                        Left(parse::One) => value == 1,
+                        Left(parse::Two) => value == 2,
+
+                        // XXX: Few/Many should have a user-specified boundary
+                        //      One possible option would be in the function
+                        //      pointer of the 'arg: Argument' struct.
+                        Left(parse::Few) => value < 8,
+                        Left(parse::Many) => value >= 8,
+
+                        Right(*) => false
+                    };
+                    if run {
+                        return self.runplural(value, s.result);
+                    }
+                }
+
+                self.runplural(value, *default);
+            }
+
+            // Select is just a matching against the string specified.
+            rt::Select(ref selectors, ref default) => {
+                // This is validated at compile-time to be a pointer to a
+                // string slice,
+                let value: & &str = unsafe { cast::transmute(arg.value) };
+                let value = *value;
+
+                for s in selectors.iter() {
+                    if s.selector == value {
+                        for piece in s.result.iter() {
+                            self.run(piece, Some(value));
+                        }
+                        return;
+                    }
+                }
+                for piece in default.iter() {
+                    self.run(piece, Some(value));
+                }
+            }
+        }
+    }
+
+    fn runplural(&mut self, value: uint, pieces: &[rt::Piece]) {
+        do uint::to_str_bytes(value, 10) |buf| {
+            let valuestr = str::from_bytes_slice(buf);
+            for piece in pieces.iter() {
+                self.run(piece, Some(valuestr));
+            }
+        }
+    }
+}
+
+/// This is a function which calls are emitted to by the compiler itself to
+/// create the Argument structures that are passed into the `sprintf` function.
+#[doc(hidden)]
+pub fn argument<'a, T>(f: extern "Rust" fn(&T, &mut Formatter),
+                       t: &'a T) -> Argument<'a> {
+    unsafe {
+        Argument {
+            formatter: cast::transmute(f),
+            value: cast::transmute(t)
+        }
+    }
+}
+
+/// When the compiler determines that the type of an argument *must* be a string
+/// (such as for select), then it invokes this method.
+#[doc(hidden)]
+pub fn argumentstr<'a>(s: &'a &str) -> Argument<'a> {
+    argument(String::fmt, s)
+}
+
+/// When the compiler determines that the type of an argument *must* be a uint
+/// (such as for plural), then it invokes this method.
+#[doc(hidden)]
+pub fn argumentuint<'a>(s: &'a uint) -> Argument<'a> {
+    argument(Unsigned::fmt, s)
+}
+
+// Implementations of the core formatting traits
+
+impl Bool for bool {
+    fn fmt(b: &bool, f: &mut Formatter) {
+        String::fmt(&(if *b {"true"} else {"false"}), f);
+    }
+}
+
+impl<'self> String for &'self str {
+    fn fmt(s: & &'self str, f: &mut Formatter) {
+        // XXX: formatting args
+        f.buf.write(s.as_bytes())
+    }
+}
+
+impl Char for char {
+    fn fmt(c: &char, f: &mut Formatter) {
+        // XXX: formatting args
+        // XXX: shouldn't require an allocation
+        let mut s = ~"";
+        s.push_char(*c);
+        f.buf.write(s.as_bytes());
+    }
+}
+
+impl Signed for int {
+    fn fmt(c: &int, f: &mut Formatter) {
+        // XXX: formatting args
+        do int::to_str_bytes(*c, 10) |buf| {
+            f.buf.write(buf);
+        }
+    }
+}
+
+impl Unsigned for uint {
+    fn fmt(c: &uint, f: &mut Formatter) {
+        // XXX: formatting args
+        do uint::to_str_bytes(*c, 10) |buf| {
+            f.buf.write(buf);
+        }
+    }
+}
+
+impl Octal for uint {
+    fn fmt(c: &uint, f: &mut Formatter) {
+        // XXX: formatting args
+        do uint::to_str_bytes(*c, 8) |buf| {
+            f.buf.write(buf);
+        }
+    }
+}
+
+impl LowerHex for uint {
+    fn fmt(c: &uint, f: &mut Formatter) {
+        // XXX: formatting args
+        do uint::to_str_bytes(*c, 16) |buf| {
+            f.buf.write(buf);
+        }
+    }
+}
+
+impl UpperHex for uint {
+    fn fmt(c: &uint, f: &mut Formatter) {
+        // XXX: formatting args
+        do uint::to_str_bytes(*c, 16) |buf| {
+            let mut local = [0u8, ..16];
+            for (l, &b) in local.mut_iter().zip(buf.iter()) {
+                *l = match b as char {
+                    'a' .. 'f' => (b - 'a' as u8) + 'A' as u8,
+                    _ => b,
+                };
+            }
+            f.buf.write(local.slice_to(buf.len()));
+        }
+    }
+}
+
+impl<T> Poly for T {
+    fn fmt(t: &T, f: &mut Formatter) {
+        // XXX: formatting args
+        let s = sys::log_str(t);
+        f.buf.write(s.as_bytes());
+    }
+}
+
+// n.b. use 'const' to get an implementation for both '*mut' and '*' at the same
+//      time.
+impl<T> Pointer for *const T {
+    fn fmt(t: &*const T, f: &mut Formatter) {
+        // XXX: formatting args
+        f.buf.write("0x".as_bytes());
+        LowerHex::fmt(&(*t as uint), f);
+    }
+}
+
+// If you expected tests to be here, look instead at the run-pass/ifmt.rs test,
+// it's a lot easier than creating all of the rt::Piece structures here.
diff --git a/src/libstd/fmt/parse.rs b/src/libstd/fmt/parse.rs
new file mode 100644
index 00000000000..673ea1d3fa8
--- /dev/null
+++ b/src/libstd/fmt/parse.rs
@@ -0,0 +1,896 @@
+// Copyright 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.
+
+use prelude::*;
+
+use char;
+use str;
+use iterator;
+
+condition! { pub parse_error: ~str -> (); }
+
+/// A piece is a portion of the format string which represents the next part to
+/// emit. These are emitted as a stream by the `Parser` class.
+#[deriving(Eq)]
+pub enum Piece<'self> {
+    /// A literal string which should directly be emitted
+    String(&'self str),
+    /// A back-reference to whatever the current argument is. This is used
+    /// inside of a method call to refer back to the original argument.
+    CurrentArgument,
+    /// This describes that formatting should process the next argument (as
+    /// specified inside) for emission.
+    Argument(Argument<'self>),
+}
+
+/// Representation of an argument specification.
+#[deriving(Eq)]
+pub struct Argument<'self> {
+    /// Where to find this argument
+    position: Position<'self>,
+    /// How to format the argument
+    format: FormatSpec<'self>,
+    /// If not `None`, what method to invoke on the argument
+    method: Option<~Method<'self>>
+}
+
+/// Specification for the formatting of an argument in the format string.
+#[deriving(Eq)]
+pub struct FormatSpec<'self> {
+    /// Optionally specified character to fill alignment with
+    fill: Option<char>,
+    /// Optionally specified alignment
+    align: Option<Alignment>,
+    /// Packed version of various flags provided
+    flags: uint,
+    /// The integer precision to use
+    precision: Count,
+    /// The string width requested for the resulting format
+    width: Count,
+    /// The descriptor string representing the name of the format desired for
+    /// this argument, this can be empty or any number of characters, although
+    /// it is required to be one word.
+    ty: &'self str
+}
+
+/// Enum describing where an argument for a format can be located.
+#[deriving(Eq)]
+pub enum Position<'self> {
+    ArgumentNext, ArgumentIs(uint), ArgumentNamed(&'self str)
+}
+
+/// Enum of alignments which are supoprted.
+#[deriving(Eq)]
+pub enum Alignment { AlignLeft, AlignRight }
+
+/// Various flags which can be applied to format strings, the meaning of these
+/// flags is defined by the formatters themselves.
+#[deriving(Eq)]
+pub enum Flag {
+    FlagSignPlus,
+    FlagSignMinus,
+    FlagAlternate,
+}
+
+/// A count is used for the precision and width parameters of an integer, and
+/// can reference either an argument or a literal integer.
+#[deriving(Eq)]
+pub enum Count {
+    CountIs(uint),
+    CountIsParam(uint),
+    CountIsNextParam,
+    CountImplied,
+}
+
+/// Enum describing all of the possible methods which the formatting language
+/// currently supports.
+#[deriving(Eq)]
+pub enum Method<'self> {
+    /// A plural method selects on an integer over a list of either integer or
+    /// keyword-defined clauses. The meaning of the keywords is defined by the
+    /// current locale.
+    ///
+    /// An offset is optionally present at the beginning which is used to match
+    /// against keywords, but it is not matched against the literal integers.
+    ///
+    /// The final element of this enum is the default "other" case which is
+    /// always required to be specified.
+    Plural(Option<uint>, ~[PluralArm<'self>], ~[Piece<'self>]),
+
+    /// A select method selects over a string. Each arm is a different string
+    /// which can be selected for.
+    ///
+    /// As with `Plural`, a default "other" case is required as well.
+    Select(~[SelectArm<'self>], ~[Piece<'self>]),
+}
+
+/// Structure representing one "arm" of the `plural` function.
+#[deriving(Eq)]
+pub struct PluralArm<'self> {
+    /// A selector can either be specified by a keyword or with an integer
+    /// literal.
+    selector: Either<PluralKeyword, uint>,
+    /// Array of pieces which are the format of this arm
+    result: ~[Piece<'self>],
+}
+
+/// Enum of the 5 CLDR plural keywords. There is one more, "other", but that is
+/// specially placed in the `Plural` variant of `Method`
+///
+/// http://www.icu-project.org/apiref/icu4c/classicu_1_1PluralRules.html
+#[deriving(Eq, IterBytes)]
+pub enum PluralKeyword {
+    Zero, One, Two, Few, Many
+}
+
+/// Structure representing one "arm" of the `select` function.
+#[deriving(Eq)]
+pub struct SelectArm<'self> {
+    /// String selector which guards this arm
+    selector: &'self str,
+    /// Array of pieces which are the format of this arm
+    result: ~[Piece<'self>],
+}
+
+/// The parser structure for interpreting the input format string. This is
+/// modelled as an iterator over `Piece` structures to form a stream of tokens
+/// being output.
+///
+/// This is a recursive-descent parser for the sake of simplicity, and if
+/// necessary there's probably lots of room for improvement performance-wise.
+pub struct Parser<'self> {
+    priv input: &'self str,
+    priv cur: str::CharOffsetIterator<'self>,
+}
+
+impl<'self> iterator::Iterator<Piece<'self>> for Parser<'self> {
+    fn next(&mut self) -> Option<Piece<'self>> {
+        match self.cur.clone().next() {
+            Some((_, '#')) => { self.cur.next(); Some(CurrentArgument) }
+            Some((_, '{')) => {
+                self.cur.next();
+                let ret = Some(Argument(self.argument()));
+                if !self.consume('}') {
+                    self.err(~"unterminated format string");
+                }
+                ret
+            }
+            Some((pos, '\\')) => {
+                self.cur.next();
+                self.escape(); // ensure it's a valid escape sequence
+                Some(String(self.string(pos + 1))) // skip the '\' character
+            }
+            Some((_, '}')) | None => { None }
+            Some((pos, _)) => {
+                Some(String(self.string(pos)))
+            }
+        }
+    }
+}
+
+impl<'self> Parser<'self> {
+    /// Creates a new parser for the given format string
+    pub fn new<'a>(s: &'a str) -> Parser<'a> {
+        Parser {
+            input: s,
+            cur: s.char_offset_iter(),
+        }
+    }
+
+    /// Notifies of an error. The message doesn't actually need to be of type
+    /// ~str, but I think it does when this eventually uses conditions so it
+    /// might as well start using it now.
+    fn err(&self, msg: ~str) {
+        parse_error::cond.raise(msg);
+    }
+
+    /// Optionally consumes the specified character. If the character is not at
+    /// the current position, then the current iterator isn't moved and false is
+    /// returned, otherwise the character is consumed and true is returned.
+    fn consume(&mut self, c: char) -> bool {
+        match self.cur.clone().next() {
+            Some((_, maybe)) if c == maybe => {
+                self.cur.next();
+                true
+            }
+            Some(*) | None => false,
+        }
+    }
+
+    /// Attempts to consume any amount of whitespace followed by a character
+    fn wsconsume(&mut self, c: char) -> bool {
+        self.ws(); self.consume(c)
+    }
+
+    /// Consumes all whitespace characters until the first non-whitespace
+    /// character
+    fn ws(&mut self) {
+        loop {
+            match self.cur.clone().next() {
+                Some((_, c)) if char::is_whitespace(c) => { self.cur.next(); }
+                Some(*) | None => { return }
+            }
+        }
+    }
+
+    /// Consumes an escape sequence, failing if there is not a valid character
+    /// to be escaped.
+    fn escape(&mut self) -> char {
+        match self.cur.next() {
+            Some((_, c @ '#')) | Some((_, c @ '{')) |
+            Some((_, c @ '\\')) | Some((_, c @ '}')) => { c }
+            Some((_, c)) => {
+                self.err(fmt!("invalid escape character `%c`", c));
+                c
+            }
+            None => {
+                self.err(~"expected an escape sequence, but format string was \
+                           terminated");
+                ' '
+            }
+        }
+    }
+
+    /// Parses all of a string which is to be considered a "raw literal" in a
+    /// format string. This is everything outside of the braces.
+    fn string(&mut self, start: uint) -> &'self str {
+        loop {
+            // we may not consume the character, so clone the iterator
+            match self.cur.clone().next() {
+                Some((pos, '\\')) | Some((pos, '#')) |
+                Some((pos, '}')) | Some((pos, '{')) => {
+                    return self.input.slice(start, pos);
+                }
+                Some(*) => { self.cur.next(); }
+                None => {
+                    self.cur.next();
+                    return self.input.slice(start, self.input.len());
+                }
+            }
+        }
+    }
+
+    /// Parses an Argument structure, or what's contained within braces inside
+    /// the format string
+    fn argument(&mut self) -> Argument<'self> {
+        Argument {
+            position: self.position(),
+            format: self.format(),
+            method: self.method(),
+        }
+    }
+
+    /// Parses a positional argument for a format. This could either be an
+    /// integer index of an argument, a named argument, or a blank string.
+    fn position(&mut self) -> Position<'self> {
+        match self.integer() {
+            Some(i) => { ArgumentIs(i) }
+            None => {
+                match self.cur.clone().next() {
+                    Some((_, c)) if char::is_alphabetic(c) => {
+                        ArgumentNamed(self.word())
+                    }
+                    _ => ArgumentNext
+                }
+            }
+        }
+    }
+
+    /// Parses a format specifier at the current position, returning all of the
+    /// relevant information in the FormatSpec struct.
+    fn format(&mut self) -> FormatSpec<'self> {
+        let mut spec = FormatSpec {
+            fill: None,
+            align: None,
+            flags: 0,
+            precision: CountImplied,
+            width: CountImplied,
+            ty: self.input.slice(0, 0),
+        };
+        if !self.consume(':') { return spec }
+
+        // fill character
+        match self.cur.clone().next() {
+            Some((_, c)) => {
+                match self.cur.clone().skip(1).next() {
+                    Some((_, '>')) | Some((_, '<')) => {
+                        spec.fill = Some(c);
+                        self.cur.next();
+                    }
+                    Some(*) | None => {}
+                }
+            }
+            None => {}
+        }
+        // Alignment
+        if self.consume('<') {
+            spec.align = Some(AlignLeft);
+        } else if self.consume('>') {
+            spec.align = Some(AlignRight);
+        }
+        // Sign flags
+        if self.consume('+') {
+            spec.flags |= 1 << (FlagSignPlus as uint);
+        } else if self.consume('-') {
+            spec.flags |= 1 << (FlagSignMinus as uint);
+        }
+        // Alternate marker
+        if self.consume('#') {
+            spec.flags |= 1 << (FlagAlternate as uint);
+        }
+        // Width and precision
+        spec.width = self.count();
+        if self.consume('.') {
+            if self.consume('*') {
+                spec.precision = CountIsNextParam;
+            } else {
+                spec.precision = self.count();
+            }
+        }
+        // Finally the actual format specifier
+        spec.ty = self.word();
+        return spec;
+    }
+
+    /// Parses a method to be applied to the previously specified argument and
+    /// its format. The two current supported methods are 'plural' and 'select'
+    fn method(&mut self) -> Option<~Method<'self>> {
+        if !self.wsconsume(',') {
+            return None;
+        }
+        self.ws();
+        match self.word() {
+            "select" => {
+                if !self.wsconsume(',') {
+                    self.err(~"`select` must be followed by `,`");
+                }
+                Some(self.select())
+            }
+            "plural" => {
+                if !self.wsconsume(',') {
+                    self.err(~"`plural` must be followed by `,`");
+                }
+                Some(self.plural())
+            }
+            "" => {
+                self.err(~"expected method after comma");
+                return None;
+            }
+            method => {
+                self.err(fmt!("unknown method: `%s`", method));
+                return None;
+            }
+        }
+    }
+
+    /// Parses a 'select' statement (after the initial 'select' word)
+    fn select(&mut self) -> ~Method<'self> {
+        let mut other = None;
+        let mut arms = ~[];
+        // Consume arms one at a time
+        loop {
+            self.ws();
+            let selector = self.word();
+            if selector == "" {
+                self.err(~"cannot have an empty selector");
+                break
+            }
+            if !self.wsconsume('{') {
+                self.err(~"selector must be followed by `{`");
+            }
+            let pieces = self.collect();
+            if !self.wsconsume('}') {
+                self.err(~"selector case must be terminated by `}`");
+            }
+            if selector == "other" {
+                if !other.is_none() {
+                    self.err(~"multiple `other` statements in `select");
+                }
+                other = Some(pieces);
+            } else {
+                arms.push(SelectArm { selector: selector, result: pieces });
+            }
+            self.ws();
+            match self.cur.clone().next() {
+                Some((_, '}')) => { break }
+                Some(*) | None => {}
+            }
+        }
+        // The "other" selector must be present
+        let other = match other {
+            Some(arm) => { arm }
+            None => {
+                self.err(~"`select` statement must provide an `other` case");
+                ~[]
+            }
+        };
+        ~Select(arms, other)
+    }
+
+    /// Parses a 'plural' statement (after the initial 'plural' word)
+    fn plural(&mut self) -> ~Method<'self> {
+        let mut offset = None;
+        let mut other = None;
+        let mut arms = ~[];
+
+        // First, attempt to parse the 'offset:' field. We know the set of
+        // selector words which can appear in plural arms, and the only ones
+        // which start with 'o' are "other" and "offset", hence look two
+        // characters deep to see if we can consume the word "offset"
+        self.ws();
+        let mut it = self.cur.clone();
+        match it.next() {
+            Some((_, 'o')) => {
+                match it.next() {
+                    Some((_, 'f')) => {
+                        let word = self.word();
+                        if word != "offset" {
+                            self.err(fmt!("expected `offset`, found `%s`",
+                                          word));
+                        } else {
+                            if !self.consume(':') {
+                                self.err(~"`offset` must be followed by `:`");
+                            }
+                            match self.integer() {
+                                Some(i) => { offset = Some(i); }
+                                None => {
+                                    self.err(~"offset must be an integer");
+                                }
+                            }
+                        }
+                    }
+                    Some(*) | None => {}
+                }
+            }
+            Some(*) | None => {}
+        }
+
+        // Next, generate all the arms
+        loop {
+            let mut isother = false;
+            let selector = if self.wsconsume('=') {
+                match self.integer() {
+                    Some(i) => Right(i),
+                    None => {
+                        self.err(~"plural `=` selectors must be followed by an \
+                                   integer");
+                        Right(0)
+                    }
+                }
+            } else {
+                let word = self.word();
+                match word {
+                    "other" => { isother = true; Left(Zero) }
+                    "zero"  => Left(Zero),
+                    "one"   => Left(One),
+                    "two"   => Left(Two),
+                    "few"   => Left(Few),
+                    "many"  => Left(Many),
+                    word    => {
+                        self.err(fmt!("unexpected plural selector `%s`", word));
+                        if word == "" {
+                            break
+                        } else {
+                            Left(Zero)
+                        }
+                    }
+                }
+            };
+            if !self.wsconsume('{') {
+                self.err(~"selector must be followed by `{`");
+            }
+            let pieces = self.collect();
+            if !self.wsconsume('}') {
+                self.err(~"selector case must be terminated by `}`");
+            }
+            if isother {
+                if !other.is_none() {
+                    self.err(~"multiple `other` statements in `select");
+                }
+                other = Some(pieces);
+            } else {
+                arms.push(PluralArm { selector: selector, result: pieces });
+            }
+            self.ws();
+            match self.cur.clone().next() {
+                Some((_, '}')) => { break }
+                Some(*) | None => {}
+            }
+        }
+
+        let other = match other {
+            Some(arm) => { arm }
+            None => {
+                self.err(~"`plural` statement must provide an `other` case");
+                ~[]
+            }
+        };
+        ~Plural(offset, arms, other)
+    }
+
+    /// Parses a Count parameter at the current position. This does not check
+    /// for 'CountIsNextParam' because that is only used in precision, not
+    /// width.
+    fn count(&mut self) -> Count {
+        match self.integer() {
+            Some(i) => {
+                if self.consume('$') {
+                    CountIsParam(i)
+                } else {
+                    CountIs(i)
+                }
+            }
+            None => { CountImplied }
+        }
+    }
+
+    /// Parses a word starting at the current position. A word is considered to
+    /// be an alphabetic character followed by any number of alphanumeric
+    /// characters.
+    fn word(&mut self) -> &'self str {
+        let start = match self.cur.clone().next() {
+            Some((pos, c)) if char::is_alphabetic(c) => {
+                self.cur.next();
+                pos
+            }
+            Some(*) | None => { return self.input.slice(0, 0); }
+        };
+        let mut end;
+        loop {
+            match self.cur.clone().next() {
+                Some((_, c)) if char::is_alphanumeric(c) => {
+                    self.cur.next();
+                }
+                Some((pos, _)) => { end = pos; break }
+                None => { end = self.input.len(); break }
+            }
+        }
+        self.input.slice(start, end)
+    }
+
+    /// Optionally parses an integer at the current position. This doesn't deal
+    /// with overflow at all, it's just accumulating digits.
+    fn integer(&mut self) -> Option<uint> {
+        let mut cur = 0;
+        let mut found = false;
+        loop {
+            match self.cur.clone().next() {
+                Some((_, c)) => {
+                    match char::to_digit(c, 10) {
+                        Some(i) => {
+                            cur = cur * 10 + i;
+                            found = true;
+                            self.cur.next();
+                        }
+                        None => { break }
+                    }
+                }
+                None => { break }
+            }
+        }
+        if found {
+            return Some(cur);
+        } else {
+            return None;
+        }
+    }
+}
+
+#[cfg(test)]
+mod tests {
+    use super::*;
+    use prelude::*;
+    use realstd::fmt::{String};
+
+    fn same(fmt: &'static str, p: ~[Piece<'static>]) {
+        let mut parser = Parser::new(fmt);
+        assert_eq!(p, parser.collect());
+    }
+
+    fn fmtdflt() -> FormatSpec<'static> {
+        return FormatSpec {
+            fill: None,
+            align: None,
+            flags: 0,
+            precision: CountImplied,
+            width: CountImplied,
+            ty: "",
+        }
+    }
+
+    fn musterr(s: &str) {
+        Parser::new(s).next();
+    }
+
+    #[test]
+    fn simple() {
+        same("asdf", ~[String("asdf")]);
+        same("a\\{b", ~[String("a"), String("{b")]);
+        same("a\\#b", ~[String("a"), String("#b")]);
+        same("a\\}b", ~[String("a"), String("}b")]);
+        same("a\\}", ~[String("a"), String("}")]);
+        same("\\}", ~[String("}")]);
+    }
+
+    #[test] #[should_fail] fn invalid01() { musterr("{") }
+    #[test] #[should_fail] fn invalid02() { musterr("\\") }
+    #[test] #[should_fail] fn invalid03() { musterr("\\a") }
+    #[test] #[should_fail] fn invalid04() { musterr("{3a}") }
+    #[test] #[should_fail] fn invalid05() { musterr("{:|}") }
+    #[test] #[should_fail] fn invalid06() { musterr("{:>>>}") }
+
+    #[test]
+    fn format_nothing() {
+        same("{}", ~[Argument(Argument {
+            position: ArgumentNext,
+            format: fmtdflt(),
+            method: None,
+        })]);
+    }
+    #[test]
+    fn format_position() {
+        same("{3}", ~[Argument(Argument {
+            position: ArgumentIs(3),
+            format: fmtdflt(),
+            method: None,
+        })]);
+    }
+    #[test]
+    fn format_position_nothing_else() {
+        same("{3:}", ~[Argument(Argument {
+            position: ArgumentIs(3),
+            format: fmtdflt(),
+            method: None,
+        })]);
+    }
+    #[test]
+    fn format_type() {
+        same("{3:a}", ~[Argument(Argument {
+            position: ArgumentIs(3),
+            format: FormatSpec {
+                fill: None,
+                align: None,
+                flags: 0,
+                precision: CountImplied,
+                width: CountImplied,
+                ty: "a",
+            },
+            method: None,
+        })]);
+    }
+    #[test]
+    fn format_align_fill() {
+        same("{3:>}", ~[Argument(Argument {
+            position: ArgumentIs(3),
+            format: FormatSpec {
+                fill: None,
+                align: Some(AlignRight),
+                flags: 0,
+                precision: CountImplied,
+                width: CountImplied,
+                ty: "",
+            },
+            method: None,
+        })]);
+        same("{3:0<}", ~[Argument(Argument {
+            position: ArgumentIs(3),
+            format: FormatSpec {
+                fill: Some('0'),
+                align: Some(AlignLeft),
+                flags: 0,
+                precision: CountImplied,
+                width: CountImplied,
+                ty: "",
+            },
+            method: None,
+        })]);
+        same("{3:*<abcd}", ~[Argument(Argument {
+            position: ArgumentIs(3),
+            format: FormatSpec {
+                fill: Some('*'),
+                align: Some(AlignLeft),
+                flags: 0,
+                precision: CountImplied,
+                width: CountImplied,
+                ty: "abcd",
+            },
+            method: None,
+        })]);
+    }
+    #[test]
+    fn format_counts() {
+        same("{:10s}", ~[Argument(Argument {
+            position: ArgumentNext,
+            format: FormatSpec {
+                fill: None,
+                align: None,
+                flags: 0,
+                precision: CountImplied,
+                width: CountIs(10),
+                ty: "s",
+            },
+            method: None,
+        })]);
+        same("{:10$.10s}", ~[Argument(Argument {
+            position: ArgumentNext,
+            format: FormatSpec {
+                fill: None,
+                align: None,
+                flags: 0,
+                precision: CountIs(10),
+                width: CountIsParam(10),
+                ty: "s",
+            },
+            method: None,
+        })]);
+        same("{:.*s}", ~[Argument(Argument {
+            position: ArgumentNext,
+            format: FormatSpec {
+                fill: None,
+                align: None,
+                flags: 0,
+                precision: CountIsNextParam,
+                width: CountImplied,
+                ty: "s",
+            },
+            method: None,
+        })]);
+        same("{:.10$s}", ~[Argument(Argument {
+            position: ArgumentNext,
+            format: FormatSpec {
+                fill: None,
+                align: None,
+                flags: 0,
+                precision: CountIsParam(10),
+                width: CountImplied,
+                ty: "s",
+            },
+            method: None,
+        })]);
+    }
+    #[test]
+    fn format_flags() {
+        same("{:-}", ~[Argument(Argument {
+            position: ArgumentNext,
+            format: FormatSpec {
+                fill: None,
+                align: None,
+                flags: (1 << FlagSignMinus as uint),
+                precision: CountImplied,
+                width: CountImplied,
+                ty: "",
+            },
+            method: None,
+        })]);
+        same("{:+#}", ~[Argument(Argument {
+            position: ArgumentNext,
+            format: FormatSpec {
+                fill: None,
+                align: None,
+                flags: (1 << FlagSignPlus as uint) | (1 << FlagAlternate as uint),
+                precision: CountImplied,
+                width: CountImplied,
+                ty: "",
+            },
+            method: None,
+        })]);
+    }
+    #[test]
+    fn format_mixture() {
+        same("abcd {3:a} efg", ~[String("abcd "), Argument(Argument {
+            position: ArgumentIs(3),
+            format: FormatSpec {
+                fill: None,
+                align: None,
+                flags: 0,
+                precision: CountImplied,
+                width: CountImplied,
+                ty: "a",
+            },
+            method: None,
+        }), String(" efg")]);
+    }
+
+    #[test]
+    fn select_simple() {
+        same("{, select, other { haha } }", ~[Argument(Argument{
+            position: ArgumentNext,
+            format: fmtdflt(),
+            method: Some(~Select(~[], ~[String(" haha ")]))
+        })]);
+        same("{1, select, other { haha } }", ~[Argument(Argument{
+            position: ArgumentIs(1),
+            format: fmtdflt(),
+            method: Some(~Select(~[], ~[String(" haha ")]))
+        })]);
+        same("{1, select, other {#} }", ~[Argument(Argument{
+            position: ArgumentIs(1),
+            format: fmtdflt(),
+            method: Some(~Select(~[], ~[CurrentArgument]))
+        })]);
+        same("{1, select, other {{2, select, other {lol}}} }", ~[Argument(Argument{
+            position: ArgumentIs(1),
+            format: fmtdflt(),
+            method: Some(~Select(~[], ~[Argument(Argument{
+                position: ArgumentIs(2),
+                format: fmtdflt(),
+                method: Some(~Select(~[], ~[String("lol")]))
+            })])) // wat
+        })]);
+    }
+
+    #[test]
+    fn select_cases() {
+        same("{1, select, a{1} b{2} c{3} other{4} }", ~[Argument(Argument{
+            position: ArgumentIs(1),
+            format: fmtdflt(),
+            method: Some(~Select(~[
+                SelectArm{ selector: "a", result: ~[String("1")] },
+                SelectArm{ selector: "b", result: ~[String("2")] },
+                SelectArm{ selector: "c", result: ~[String("3")] },
+            ], ~[String("4")]))
+        })]);
+    }
+
+    #[test] #[should_fail] fn badselect01() {
+        musterr("{select, }")
+    }
+    #[test] #[should_fail] fn badselect02() {
+        musterr("{1, select}")
+    }
+    #[test] #[should_fail] fn badselect03() {
+        musterr("{1, select, }")
+    }
+    #[test] #[should_fail] fn badselect04() {
+        musterr("{1, select, a {}}")
+    }
+    #[test] #[should_fail] fn badselect05() {
+        musterr("{1, select, other }}")
+    }
+    #[test] #[should_fail] fn badselect06() {
+        musterr("{1, select, other {}")
+    }
+    #[test] #[should_fail] fn badselect07() {
+        musterr("{select, other {}")
+    }
+    #[test] #[should_fail] fn badselect08() {
+        musterr("{1 select, other {}")
+    }
+    #[test] #[should_fail] fn badselect09() {
+        musterr("{:d select, other {}")
+    }
+    #[test] #[should_fail] fn badselect10() {
+        musterr("{1:d select, other {}")
+    }
+
+    #[test]
+    fn plural_simple() {
+        same("{, plural, other { haha } }", ~[Argument(Argument{
+            position: ArgumentNext,
+            format: fmtdflt(),
+            method: Some(~Plural(None, ~[], ~[String(" haha ")]))
+        })]);
+        same("{:, plural, other { haha } }", ~[Argument(Argument{
+            position: ArgumentNext,
+            format: fmtdflt(),
+            method: Some(~Plural(None, ~[], ~[String(" haha ")]))
+        })]);
+        same("{, plural, offset:1 =2{2} =3{3} many{yes} other{haha} }",
+        ~[Argument(Argument{
+            position: ArgumentNext,
+            format: fmtdflt(),
+            method: Some(~Plural(Some(1), ~[
+                PluralArm{ selector: Right(2), result: ~[String("2")] },
+                PluralArm{ selector: Right(3), result: ~[String("3")] },
+                PluralArm{ selector: Left(Many), result: ~[String("yes")] }
+            ], ~[String("haha")]))
+        })]);
+    }
+}
diff --git a/src/libstd/fmt/rt.rs b/src/libstd/fmt/rt.rs
new file mode 100644
index 00000000000..6feb1d7a848
--- /dev/null
+++ b/src/libstd/fmt/rt.rs
@@ -0,0 +1,62 @@
+// Copyright 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 is an internal module used by the ifmt! runtime. These structures are
+//! emitted to static arrays to precompile format strings ahead of time.
+//!
+//! These definitions are similar to their `ct` equivalents, but differ in that
+//! these can be statically allocated and are slightly optimized for the runtime
+
+#[allow(missing_doc)];
+#[doc(hidden)];
+
+use either::Either;
+use fmt::parse;
+use option::Option;
+
+pub enum Piece<'self> {
+    String(&'self str),
+    // FIXME(#8259): this shouldn't require the unit-value here
+    CurrentArgument(()),
+    Argument(Argument<'self>),
+}
+
+pub struct Argument<'self> {
+    position: Position,
+    format: FormatSpec,
+    method: Option<&'self Method<'self>>
+}
+
+pub struct FormatSpec {
+    fill: char,
+    alignleft: bool,
+    flags: uint,
+    precision: parse::Count,
+    width: parse::Count,
+}
+
+pub enum Position {
+    ArgumentNext, ArgumentIs(uint)
+}
+
+pub enum Method<'self> {
+    Plural(Option<uint>, &'self [PluralArm<'self>], &'self [Piece<'self>]),
+    Select(&'self [SelectArm<'self>], &'self [Piece<'self>]),
+}
+
+pub struct PluralArm<'self> {
+    selector: Either<parse::PluralKeyword, uint>,
+    result: &'self [Piece<'self>],
+}
+
+pub struct SelectArm<'self> {
+    selector: &'self str,
+    result: &'self [Piece<'self>],
+}
diff --git a/src/libstd/rt/io/mem.rs b/src/libstd/rt/io/mem.rs
index c93945a6a9a..277897e5d2e 100644
--- a/src/libstd/rt/io/mem.rs
+++ b/src/libstd/rt/io/mem.rs
@@ -26,7 +26,7 @@ pub struct MemWriter {
 }
 
 impl MemWriter {
-    pub fn new() -> MemWriter { MemWriter { buf: ~[] } }
+    pub fn new() -> MemWriter { MemWriter { buf: vec::with_capacity(128) } }
 }
 
 impl Writer for MemWriter {
diff --git a/src/libstd/std.rs b/src/libstd/std.rs
index 568709c89da..7000b56069d 100644
--- a/src/libstd/std.rs
+++ b/src/libstd/std.rs
@@ -177,6 +177,7 @@ pub mod rand;
 pub mod run;
 pub mod sys;
 pub mod cast;
+pub mod fmt;
 pub mod repr;
 pub mod cleanup;
 pub mod reflect;
@@ -216,4 +217,6 @@ mod std {
     pub use unstable;
     pub use str;
     pub use os;
+    pub use fmt;
+    pub use to_bytes;
 }