import FIREFOX_52_6_0esr_RELEASE from mozilla-esr52 hg repo

third_party/rust/idna/.cargo-checksum.json

@@ -0,0 +1 @@
+{"files":{".cargo-ok":"e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855","Cargo.toml":"d0aad6852ec947597bab20cd85ad15e014044cb7717c9a91f4b8da0081b7134d","src/IdnaMappingTable.txt":"5e9f5929130b713e698162ac5b60a99ccfb831606686b1c50777cd920b55dee2","src/lib.rs":"b8e85707a40e8472d2e90849a1e0e24a7442f3c4614a57a60125f87d11e985a4","src/make_uts46_mapping_table.py":"36fa77c443672f15872d60438f96b7302eae28ec506b60a892579debc79b8e39","src/punycode.rs":"df883ec00b35cab38f96992667eef0767d8587746bfdab0613a03e4c49a26c16","src/uts46.rs":"c879570c511f210565ac0dbce8a212be6503e62cd7ed698830804bb7729e98de","src/uts46_mapping_table.rs":"daa59e4b6399a738f73967b222b7dce6c9706a471d306330d77380c89089fa24","tests/IdnaTest.txt":"12e7e150b04a7a2cb1f9b72222174844342218807126e9dbc53069505a5f6000","tests/punycode.rs":"2f4086411c00b0641377afe81071e51a695110a0cce474287557738c07f74322","tests/punycode_tests.json":"3d4ac0cf25984c37b9ce197f5df680a0136f728fb8ec82bc76624e42139eb3a8","tests/tests.rs":"bb92e129dc5e17e9a86ec6062dd7b3f4c905c4af69e773d7c70efea177654c7b","tests/uts46.rs":"be9f928c60b88a8e277ddfb1769f09a8cd273e2e120e8450fb9b34f4dc852b37"},"package":"1053236e00ce4f668aeca4a769a09b3bf5a682d802abd6f3cb39374f6b162c11"}

third_party/rust/idna/Cargo.toml

@@ -0,0 +1,24 @@
+[package]
+name = "idna"
+version = "0.1.0"
+authors = ["Simon Sapin <simon.sapin@exyr.org>"]
+description = "IDNA (Internationalizing Domain Names in Applications) and Punycode."
+repository = "https://github.com/servo/rust-url/"
+license = "MIT/Apache-2.0"
+
+[lib]
+doctest = false
+test = false
+
+[[test]]
+name = "tests"
+harness = false
+
+[dev-dependencies]
+rustc-test = "0.1"
+rustc-serialize = "0.3"
+
+[dependencies]
+unicode-bidi = "0.2.3"
+unicode-normalization = "0.1.2"
+matches = "0.1"

third_party/rust/idna/src/lib.rs

@@ -0,0 +1,73 @@
+// Copyright 2016 Simon Sapin.
+//
+// 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 Rust crate implements IDNA
+//! [per the WHATWG URL Standard](https://url.spec.whatwg.org/#idna).
+//!
+//! It also exposes the underlying algorithms from [*Unicode IDNA Compatibility Processing*
+//! (Unicode Technical Standard #46)](http://www.unicode.org/reports/tr46/)
+//! and [Punycode (RFC 3492)](https://tools.ietf.org/html/rfc3492).
+//!
+//! Quoting from [UTS #46’s introduction](http://www.unicode.org/reports/tr46/#Introduction):
+//!
+//! > Initially, domain names were restricted to ASCII characters.
+//! > A system was introduced in 2003 for internationalized domain names (IDN).
+//! > This system is called Internationalizing Domain Names for Applications,
+//! > or IDNA2003 for short.
+//! > This mechanism supports IDNs by means of a client software transformation
+//! > into a format known as Punycode.
+//! > A revision of IDNA was approved in 2010 (IDNA2008).
+//! > This revision has a number of incompatibilities with IDNA2003.
+//! >
+//! > The incompatibilities force implementers of client software,
+//! > such as browsers and emailers,
+//! > to face difficult choices during the transition period
+//! > as registries shift from IDNA2003 to IDNA2008.
+//! > This document specifies a mechanism
+//! > that minimizes the impact of this transition for client software,
+//! > allowing client software to access domains that are valid under either system.
+
+#[macro_use] extern crate matches;
+extern crate unicode_bidi;
+extern crate unicode_normalization;
+
+pub mod punycode;
+pub mod uts46;
+
+/// The [domain to ASCII](https://url.spec.whatwg.org/#concept-domain-to-ascii) algorithm.
+///
+/// Return the ASCII representation a domain name,
+/// normalizing characters (upper-case to lower-case and other kinds of equivalence)
+/// and using Punycode as necessary.
+///
+/// This process may fail.
+pub fn domain_to_ascii(domain: &str) -> Result<String, uts46::Errors> {
+    uts46::to_ascii(domain, uts46::Flags {
+        use_std3_ascii_rules: false,
+        transitional_processing: true, // XXX: switch when Firefox does
+        verify_dns_length: false,
+    })
+}
+
+/// The [domain to Unicode](https://url.spec.whatwg.org/#concept-domain-to-unicode) algorithm.
+///
+/// Return the Unicode representation of a domain name,
+/// normalizing characters (upper-case to lower-case and other kinds of equivalence)
+/// and decoding Punycode as necessary.
+///
+/// This may indicate [syntax violations](https://url.spec.whatwg.org/#syntax-violation)
+/// but always returns a string for the mapped domain.
+pub fn domain_to_unicode(domain: &str) -> (String, Result<(), uts46::Errors>) {
+    uts46::to_unicode(domain, uts46::Flags {
+        use_std3_ascii_rules: false,
+
+        // Unused:
+        transitional_processing: true,
+        verify_dns_length: false,
+    })
+}

third_party/rust/idna/src/make_uts46_mapping_table.py

@@ -0,0 +1,56 @@
+# Copyright 2013-2014 Valentin Gosu.
+#
+# 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.
+
+# Run as: python make_uts46_mapping_table.py IdnaMappingTable.txt > uts46_mapping_table.rs
+# You can get the latest idna table from
+# http://www.unicode.org/Public/idna/latest/IdnaMappingTable.txt
+
+print('''\
+// Copyright 2013-2014 Valentin Gosu.
+//
+// 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.
+
+// Generated by make_idna_table.py
+
+static TABLE: &'static [Range] = &[
+''')
+
+txt = open("IdnaMappingTable.txt")
+
+def char(s):
+    return (unichr(int(s, 16))
+        .encode('utf8')
+        .replace('\\', '\\\\')
+        .replace('"', '\\"')
+        .replace('\0', '\\0'))
+
+for line in txt:
+    # remove comments
+    line, _, _ = line.partition('#')
+    # skip empty lines
+    if len(line.strip()) == 0:
+        continue
+    fields = line.split(';')
+    if fields[0].strip() == 'D800..DFFF':
+        continue  # Surrogates don't occur in Rust strings.
+    first, _, last = fields[0].strip().partition('..')
+    if not last:
+        last = first
+    mapping = fields[1].strip().replace('_', ' ').title().replace(' ', '')
+    if len(fields) > 2:
+        if fields[2].strip():
+            mapping += '("%s")' % ''.join(char(c) for c in fields[2].strip().split(' '))
+        elif mapping == "Deviation":
+            mapping += '("")'
+    print("    Range { from: '%s', to: '%s', mapping: %s }," % (char(first), char(last), mapping))
+
+print("];")

third_party/rust/idna/src/punycode.rs

@@ -0,0 +1,213 @@
+// Copyright 2013 Simon Sapin.
+//
+// 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.
+
+//! Punycode ([RFC 3492](http://tools.ietf.org/html/rfc3492)) implementation.
+//!
+//! Since Punycode fundamentally works on unicode code points,
+//! `encode` and `decode` take and return slices and vectors of `char`.
+//! `encode_str` and `decode_to_string` provide convenience wrappers
+//! that convert from and to Rust’s UTF-8 based `str` and `String` types.
+
+use std::u32;
+use std::char;
+use std::ascii::AsciiExt;
+
+// Bootstring parameters for Punycode
+static BASE: u32 = 36;
+static T_MIN: u32 = 1;
+static T_MAX: u32 = 26;
+static SKEW: u32 = 38;
+static DAMP: u32 = 700;
+static INITIAL_BIAS: u32 = 72;
+static INITIAL_N: u32 = 0x80;
+static DELIMITER: char = '-';
+
+
+#[inline]
+fn adapt(mut delta: u32, num_points: u32, first_time: bool) -> u32 {
+    delta /= if first_time { DAMP } else { 2 };
+    delta += delta / num_points;
+    let mut k = 0;
+    while delta > ((BASE - T_MIN) * T_MAX) / 2 {
+        delta /= BASE - T_MIN;
+        k += BASE;
+    }
+    k + (((BASE - T_MIN + 1) * delta) / (delta + SKEW))
+}
+
+
+/// Convert Punycode to an Unicode `String`.
+///
+/// This is a convenience wrapper around `decode`.
+#[inline]
+pub fn decode_to_string(input: &str) -> Option<String> {
+    decode(input).map(|chars| chars.into_iter().collect())
+}
+
+
+/// Convert Punycode to Unicode.
+///
+/// Return None on malformed input or overflow.
+/// Overflow can only happen on inputs that take more than
+/// 63 encoded bytes, the DNS limit on domain name labels.
+pub fn decode(input: &str) -> Option<Vec<char>> {
+    // Handle "basic" (ASCII) code points.
+    // They are encoded as-is before the last delimiter, if any.
+    let (mut output, input) = match input.rfind(DELIMITER) {
+        None => (Vec::new(), input),
+        Some(position) => (
+            input[..position].chars().collect(),
+            if position > 0 { &input[position + 1..] } else { input }
+        )
+    };
+    let mut code_point = INITIAL_N;
+    let mut bias = INITIAL_BIAS;
+    let mut i = 0;
+    let mut iter = input.bytes();
+    loop {
+        let previous_i = i;
+        let mut weight = 1;
+        let mut k = BASE;
+        let mut byte = match iter.next() {
+            None => break,
+            Some(byte) => byte,
+        };
+        // Decode a generalized variable-length integer into delta,
+        // which gets added to i.
+        loop {
+            let digit = match byte {
+                byte @ b'0' ... b'9' => byte - b'0' + 26,
+                byte @ b'A' ... b'Z' => byte - b'A',
+                byte @ b'a' ... b'z' => byte - b'a',
+                _ => return None
+            } as u32;
+            if digit > (u32::MAX - i) / weight {
+                return None  // Overflow
+            }
+            i += digit * weight;
+            let t = if k <= bias { T_MIN }
+                    else if k >= bias + T_MAX { T_MAX }
+                    else { k - bias };
+            if digit < t {
+                break
+            }
+            if weight > u32::MAX / (BASE - t) {
+                return None  // Overflow
+            }
+            weight *= BASE - t;
+            k += BASE;
+            byte = match iter.next() {
+                None => return None,  // End of input before the end of this delta
+                Some(byte) => byte,
+            };
+        }
+        let length = output.len() as u32;
+        bias = adapt(i - previous_i, length + 1, previous_i == 0);
+        if i / (length + 1) > u32::MAX - code_point {
+            return None  // Overflow
+        }
+        // i was supposed to wrap around from length+1 to 0,
+        // incrementing code_point each time.
+        code_point += i / (length + 1);
+        i %= length + 1;
+        let c = match char::from_u32(code_point) {
+            Some(c) => c,
+            None => return None
+        };
+        output.insert(i as usize, c);
+        i += 1;
+    }
+    Some(output)
+}
+
+
+/// Convert an Unicode `str` to Punycode.
+///
+/// This is a convenience wrapper around `encode`.
+#[inline]
+pub fn encode_str(input: &str) -> Option<String> {
+    encode(&input.chars().collect::<Vec<char>>())
+}
+
+
+/// Convert Unicode to Punycode.
+///
+/// Return None on overflow, which can only happen on inputs that would take more than
+/// 63 encoded bytes, the DNS limit on domain name labels.
+pub fn encode(input: &[char]) -> Option<String> {
+    // Handle "basic" (ASCII) code points. They are encoded as-is.
+    let output_bytes = input.iter().filter_map(|&c|
+        if c.is_ascii() { Some(c as u8) } else { None }
+    ).collect();
+    let mut output = unsafe { String::from_utf8_unchecked(output_bytes) };
+    let basic_length = output.len() as u32;
+    if basic_length > 0 {
+        output.push_str("-")
+    }
+    let mut code_point = INITIAL_N;
+    let mut delta = 0;
+    let mut bias = INITIAL_BIAS;
+    let mut processed = basic_length;
+    let input_length = input.len() as u32;
+    while processed < input_length {
+        // All code points < code_point have been handled already.
+        // Find the next larger one.
+        let min_code_point = input.iter().map(|&c| c as u32)
+                                  .filter(|&c| c >= code_point).min().unwrap();
+        if min_code_point - code_point > (u32::MAX - delta) / (processed + 1) {
+            return None  // Overflow
+        }
+        // Increase delta to advance the decoder’s <code_point,i> state to <min_code_point,0>
+        delta += (min_code_point - code_point) * (processed + 1);
+        code_point = min_code_point;
+        for &c in input {
+            let c = c as u32;
+            if c < code_point {
+                delta += 1;
+                if delta == 0 {
+                    return None  // Overflow
+                }
+            }
+            if c == code_point {
+                // Represent delta as a generalized variable-length integer:
+                let mut q = delta;
+                let mut k = BASE;
+                loop {
+                    let t = if k <= bias { T_MIN }
+                            else if k >= bias + T_MAX { T_MAX }
+                            else { k - bias };
+                    if q < t {
+                        break
+                    }
+                    let value = t + ((q - t) % (BASE - t));
+                    value_to_digit(value, &mut output);
+                    q = (q - t) / (BASE - t);
+                    k += BASE;
+                }
+                value_to_digit(q, &mut output);
+                bias = adapt(delta, processed + 1, processed == basic_length);
+                delta = 0;
+                processed += 1;
+            }
+        }
+        delta += 1;
+        code_point += 1;
+    }
+    Some(output)
+}
+
+
+#[inline]
+fn value_to_digit(value: u32, output: &mut String) {
+    let code_point = match value {
+        0 ... 25 => value + 0x61,  // a..z
+        26 ... 35 => value - 26 + 0x30,  // 0..9
+        _ => panic!()
+    };
+    unsafe { output.as_mut_vec().push(code_point as u8) }
+}

third_party/rust/idna/src/uts46.rs

@@ -0,0 +1,322 @@
+// Copyright 2013-2014 Valentin Gosu.
+//
+// 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.
+
+//! [*Unicode IDNA Compatibility Processing*
+//! (Unicode Technical Standard #46)](http://www.unicode.org/reports/tr46/)
+
+use self::Mapping::*;
+use punycode;
+use std::ascii::AsciiExt;
+use unicode_normalization::UnicodeNormalization;
+use unicode_normalization::char::is_combining_mark;
+use unicode_bidi::{BidiClass, bidi_class};
+
+include!("uts46_mapping_table.rs");
+
+#[derive(Debug)]
+enum Mapping {
+    Valid,
+    Ignored,
+    Mapped(&'static str),
+    Deviation(&'static str),
+    Disallowed,
+    DisallowedStd3Valid,
+    DisallowedStd3Mapped(&'static str),
+}
+
+struct Range {
+    from: char,
+    to: char,
+    mapping: Mapping,
+}
+
+fn find_char(codepoint: char) -> &'static Mapping {
+    let mut min = 0;
+    let mut max = TABLE.len() - 1;
+    while max > min {
+        let mid = (min + max) >> 1;
+        if codepoint > TABLE[mid].to {
+           min = mid;
+        } else if codepoint < TABLE[mid].from {
+            max = mid;
+        } else {
+            min = mid;
+            max = mid;
+        }
+    }
+    &TABLE[min].mapping
+}
+
+fn map_char(codepoint: char, flags: Flags, output: &mut String, errors: &mut Vec<Error>) {
+    match *find_char(codepoint) {
+        Mapping::Valid => output.push(codepoint),
+        Mapping::Ignored => {},
+        Mapping::Mapped(mapping) => output.push_str(mapping),
+        Mapping::Deviation(mapping) => {
+            if flags.transitional_processing {
+                output.push_str(mapping)
+            } else {
+                output.push(codepoint)
+            }
+        }
+        Mapping::Disallowed => {
+            errors.push(Error::DissallowedCharacter);
+            output.push(codepoint);
+        }
+        Mapping::DisallowedStd3Valid => {
+            if flags.use_std3_ascii_rules {
+                errors.push(Error::DissallowedByStd3AsciiRules);
+            }
+            output.push(codepoint)
+        }
+        Mapping::DisallowedStd3Mapped(mapping) => {
+            if flags.use_std3_ascii_rules {
+                errors.push(Error::DissallowedMappedInStd3);
+            }
+            output.push_str(mapping)
+        }
+    }
+}
+
+// http://tools.ietf.org/html/rfc5893#section-2
+fn passes_bidi(label: &str, transitional_processing: bool) -> bool {
+    let mut chars = label.chars();
+    let class = match chars.next() {
+        Some(c) => bidi_class(c),
+        None => return true, // empty string
+    };
+
+    if class == BidiClass::L
+       || (class == BidiClass::ON && transitional_processing) // starts with \u200D
+       || (class == BidiClass::ES && transitional_processing) // hack: 1.35.+33.49
+       || class == BidiClass::EN // hack: starts with number 0à.\u05D0
+    { // LTR
+        // Rule 5
+        loop {
+            match chars.next() {
+                Some(c) => {
+                    let c = bidi_class(c);
+                    if !matches!(c, BidiClass::L | BidiClass::EN |
+                                    BidiClass::ES | BidiClass::CS |
+                                    BidiClass::ET | BidiClass::ON |
+                                    BidiClass::BN | BidiClass::NSM) {
+                        return false;
+                    }
+                },
+                None => { break; },
+            }
+        }
+
+        // Rule 6
+        let mut rev_chars = label.chars().rev();
+        let mut last = rev_chars.next();
+        loop { // must end in L or EN followed by 0 or more NSM
+            match last {
+                Some(c) if bidi_class(c) == BidiClass::NSM => {
+                    last = rev_chars.next();
+                    continue;
+                }
+                _ => { break; },
+            }
+        }
+
+        // TODO: does not pass for àˇ.\u05D0
+        // match last {
+        //     Some(c) if bidi_class(c) == BidiClass::L
+        //             || bidi_class(c) == BidiClass::EN => {},
+        //     Some(c) => { return false; },
+        //     _ => {}
+        // }
+
+    } else if class == BidiClass::R || class == BidiClass::AL { // RTL
+        let mut found_en = false;
+        let mut found_an = false;
+
+        // Rule 2
+        loop {
+            match chars.next() {
+                Some(c) => {
+                    let char_class = bidi_class(c);
+
+                    if char_class == BidiClass::EN {
+                        found_en = true;
+                    }
+                    if char_class == BidiClass::AN {
+                        found_an = true;
+                    }
+
+                    if !matches!(char_class, BidiClass::R | BidiClass::AL |
+                                             BidiClass::AN | BidiClass::EN |
+                                             BidiClass::ES | BidiClass::CS |
+                                             BidiClass::ET | BidiClass::ON |
+                                             BidiClass::BN | BidiClass::NSM) {
+                        return false;
+                    }
+                },
+                None => { break; },
+            }
+        }
+        // Rule 3
+        let mut rev_chars = label.chars().rev();
+        let mut last = rev_chars.next();
+        loop { // must end in L or EN followed by 0 or more NSM
+            match last {
+                Some(c) if bidi_class(c) == BidiClass::NSM => {
+                    last = rev_chars.next();
+                    continue;
+                }
+                _ => { break; },
+            }
+        }
+        match last {
+            Some(c) if matches!(bidi_class(c), BidiClass::R | BidiClass::AL |
+                                               BidiClass::EN | BidiClass::AN) => {},
+            _ => { return false; }
+        }
+
+        // Rule 4
+        if found_an && found_en {
+            return false;
+        }
+    } else {
+        // Rule 2: Should start with L or R/AL
+        return false;
+    }
+
+    return true;
+}
+
+/// http://www.unicode.org/reports/tr46/#Validity_Criteria
+fn validate(label: &str, flags: Flags, errors: &mut Vec<Error>) {
+    if label.nfc().ne(label.chars()) {
+        errors.push(Error::ValidityCriteria);
+    }
+
+    // Can not contain '.' since the input is from .split('.')
+    if {
+        let mut chars = label.chars().skip(2);
+        let third = chars.next();
+        let fourth = chars.next();
+        (third, fourth) == (Some('-'), Some('-'))
+    } || label.starts_with("-")
+        || label.ends_with("-")
+        || label.chars().next().map_or(false, is_combining_mark)
+        || label.chars().any(|c| match *find_char(c) {
+            Mapping::Valid => false,
+            Mapping::Deviation(_) => flags.transitional_processing,
+            Mapping::DisallowedStd3Valid => flags.use_std3_ascii_rules,
+            _ => true,
+        })
+        || !passes_bidi(label, flags.transitional_processing)
+    {
+        errors.push(Error::ValidityCriteria)
+    }
+}
+
+/// http://www.unicode.org/reports/tr46/#Processing
+fn processing(domain: &str, flags: Flags, errors: &mut Vec<Error>) -> String {
+    let mut mapped = String::new();
+    for c in domain.chars() {
+        map_char(c, flags, &mut mapped, errors)
+    }
+    let normalized: String = mapped.nfc().collect();
+    let mut validated = String::new();
+    for label in normalized.split('.') {
+        if validated.len() > 0 {
+            validated.push('.');
+        }
+        if label.starts_with("xn--") {
+            match punycode::decode_to_string(&label["xn--".len()..]) {
+                Some(decoded_label) => {
+                    let flags = Flags { transitional_processing: false, ..flags };
+                    validate(&decoded_label, flags, errors);
+                    validated.push_str(&decoded_label)
+                }
+                None => errors.push(Error::PunycodeError)
+            }
+        } else {
+            validate(label, flags, errors);
+            validated.push_str(label)
+        }
+    }
+    validated
+}
+
+#[derive(Copy, Clone)]
+pub struct Flags {
+   pub use_std3_ascii_rules: bool,
+   pub transitional_processing: bool,
+   pub verify_dns_length: bool,
+}
+
+#[derive(PartialEq, Eq, Clone, Copy, Debug)]
+enum Error {
+    PunycodeError,
+    ValidityCriteria,
+    DissallowedByStd3AsciiRules,
+    DissallowedMappedInStd3,
+    DissallowedCharacter,
+    TooLongForDns,
+}
+
+/// Errors recorded during UTS #46 processing.
+///
+/// This is opaque for now, only indicating the precense of at least one error.
+/// More details may be exposed in the future.
+#[derive(Debug)]
+pub struct Errors(Vec<Error>);
+
+/// http://www.unicode.org/reports/tr46/#ToASCII
+pub fn to_ascii(domain: &str, flags: Flags) -> Result<String, Errors> {
+    let mut errors = Vec::new();
+    let mut result = String::new();
+    for label in processing(domain, flags, &mut errors).split('.') {
+        if result.len() > 0 {
+            result.push('.');
+        }
+        if label.is_ascii() {
+            result.push_str(label);
+        } else {
+            match punycode::encode_str(label) {
+                Some(x) => {
+                    result.push_str("xn--");
+                    result.push_str(&x);
+                },
+                None => errors.push(Error::PunycodeError)
+            }
+        }
+    }
+
+    if flags.verify_dns_length {
+        let domain = if result.ends_with(".") { &result[..result.len()-1]  } else { &*result };
+        if domain.len() < 1 || domain.len() > 253 ||
+                domain.split('.').any(|label| label.len() < 1 || label.len() > 63) {
+            errors.push(Error::TooLongForDns)
+        }
+    }
+    if errors.is_empty() {
+        Ok(result)
+    } else {
+        Err(Errors(errors))
+    }
+}
+
+/// http://www.unicode.org/reports/tr46/#ToUnicode
+///
+/// Only `use_std3_ascii_rules` is used in `flags`.
+pub fn to_unicode(domain: &str, mut flags: Flags) -> (String, Result<(), Errors>) {
+    flags.transitional_processing = false;
+    let mut errors = Vec::new();
+    let domain = processing(domain, flags, &mut errors);
+    let errors = if errors.is_empty() {
+        Ok(())
+    } else {
+        Err(Errors(errors))
+    };
+    (domain, errors)
+}

third_party/rust/idna/tests/punycode.rs

@@ -0,0 +1,65 @@
+// Copyright 2013 Simon Sapin.
+//
+// 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 idna::punycode::{decode, encode_str};
+use rustc_serialize::json::{Json, Object};
+use test::TestFn;
+
+fn one_test(decoded: &str, encoded: &str) {
+    match decode(encoded) {
+        None => panic!("Decoding {} failed.", encoded),
+        Some(result) => {
+            let result = result.into_iter().collect::<String>();
+            assert!(result == decoded,
+                    format!("Incorrect decoding of \"{}\":\n   \"{}\"\n!= \"{}\"\n",
+                            encoded, result, decoded))
+        }
+    }
+
+    match encode_str(decoded) {
+        None => panic!("Encoding {} failed.", decoded),
+        Some(result) => {
+            assert!(result == encoded,
+                    format!("Incorrect encoding of \"{}\":\n   \"{}\"\n!= \"{}\"\n",
+                            decoded, result, encoded))
+        }
+    }
+}
+
+fn get_string<'a>(map: &'a Object, key: &str) -> &'a str {
+    match map.get(&key.to_string()) {
+        Some(&Json::String(ref s)) => s,
+        None => "",
+        _ => panic!(),
+    }
+}
+
+pub fn collect_tests<F: FnMut(String, TestFn)>(add_test: &mut F) {
+    match Json::from_str(include_str!("punycode_tests.json")) {
+        Ok(Json::Array(tests)) => for (i, test) in tests.into_iter().enumerate() {
+            match test {
+                Json::Object(o) => {
+                    let test_name = {
+                        let desc = get_string(&o, "description");
+                            if desc.is_empty() {
+                            format!("Punycode {}", i + 1)
+                        } else {
+                            format!("Punycode {}: {}", i + 1, desc)
+                        }
+                    };
+                    add_test(test_name, TestFn::dyn_test_fn(move || one_test(
+                        get_string(&o, "decoded"),
+                        get_string(&o, "encoded"),
+                    )))
+                }
+                _ => panic!(),
+            }
+        },
+        other => panic!("{:?}", other)
+    }
+}

third_party/rust/idna/tests/punycode_tests.json

@@ -0,0 +1,120 @@
+[
+{
+    "description": "These tests are copied from https://github.com/bestiejs/punycode.js/blob/master/tests/tests.js , used under the MIT license.",
+    "decoded": "",
+    "encoded": ""
+},
+{
+    "description": "a single basic code point",
+    "decoded": "Bach",
+    "encoded": "Bach-"
+},
+{
+    "description": "a single non-ASCII character",
+    "decoded": "\u00FC",
+    "encoded": "tda"
+},
+{
+    "description": "multiple non-ASCII characters",
+    "decoded": "\u00FC\u00EB\u00E4\u00F6\u2665",
+    "encoded": "4can8av2009b"
+},
+{
+    "description": "mix of ASCII and non-ASCII characters",
+    "decoded": "b\u00FCcher",
+    "encoded": "bcher-kva"
+},
+{
+    "description": "long string with both ASCII and non-ASCII characters",
+    "decoded": "Willst du die Bl\u00FCthe des fr\u00FChen, die Fr\u00FCchte des sp\u00E4teren Jahres",
+    "encoded": "Willst du die Blthe des frhen, die Frchte des spteren Jahres-x9e96lkal"
+},
+{
+    "description": "Arabic (Egyptian)",
+    "decoded": "\u0644\u064A\u0647\u0645\u0627\u0628\u062A\u0643\u0644\u0645\u0648\u0634\u0639\u0631\u0628\u064A\u061F",
+    "encoded": "egbpdaj6bu4bxfgehfvwxn"
+},
+{
+    "description": "Chinese (simplified)",
+    "decoded": "\u4ED6\u4EEC\u4E3A\u4EC0\u4E48\u4E0D\u8BF4\u4E2d\u6587",
+    "encoded": "ihqwcrb4cv8a8dqg056pqjye"
+},
+{
+    "description": "Chinese (traditional)",
+    "decoded": "\u4ED6\u5011\u7232\u4EC0\u9EBD\u4E0D\u8AAA\u4E2D\u6587",
+    "encoded": "ihqwctvzc91f659drss3x8bo0yb"
+},
+{
+    "description": "Czech",
+    "decoded": "Pro\u010Dprost\u011Bnemluv\u00ED\u010Desky",
+    "encoded": "Proprostnemluvesky-uyb24dma41a"
+},
+{
+    "description": "Hebrew",
+    "decoded": "\u05DC\u05DE\u05D4\u05D4\u05DD\u05E4\u05E9\u05D5\u05D8\u05DC\u05D0\u05DE\u05D3\u05D1\u05E8\u05D9\u05DD\u05E2\u05D1\u05E8\u05D9\u05EA",
+    "encoded": "4dbcagdahymbxekheh6e0a7fei0b"
+},
+{
+    "description": "Hindi (Devanagari)",
+    "decoded": "\u092F\u0939\u0932\u094B\u0917\u0939\u093F\u0928\u094D\u0926\u0940\u0915\u094D\u092F\u094B\u0902\u0928\u0939\u0940\u0902\u092C\u094B\u0932\u0938\u0915\u0924\u0947\u0939\u0948\u0902",
+    "encoded": "i1baa7eci9glrd9b2ae1bj0hfcgg6iyaf8o0a1dig0cd"
+},
+{
+    "description": "Japanese (kanji and hiragana)",
+    "decoded": "\u306A\u305C\u307F\u3093\u306A\u65E5\u672C\u8A9E\u3092\u8A71\u3057\u3066\u304F\u308C\u306A\u3044\u306E\u304B",
+    "encoded": "n8jok5ay5dzabd5bym9f0cm5685rrjetr6pdxa"
+},
+{
+    "description": "Korean (Hangul syllables)",
+    "decoded": "\uC138\uACC4\uC758\uBAA8\uB4E0\uC0AC\uB78C\uB4E4\uC774\uD55C\uAD6D\uC5B4\uB97C\uC774\uD574\uD55C\uB2E4\uBA74\uC5BC\uB9C8\uB098\uC88B\uC744\uAE4C",
+    "encoded": "989aomsvi5e83db1d2a355cv1e0vak1dwrv93d5xbh15a0dt30a5jpsd879ccm6fea98c"
+},
+{
+    "description": "Russian (Cyrillic)",
+    "decoded": "\u043F\u043E\u0447\u0435\u043C\u0443\u0436\u0435\u043E\u043D\u0438\u043D\u0435\u0433\u043E\u0432\u043E\u0440\u044F\u0442\u043F\u043E\u0440\u0443\u0441\u0441\u043A\u0438",
+    "encoded": "b1abfaaepdrnnbgefbadotcwatmq2g4l"
+},
+{
+    "description": "Spanish",
+    "decoded": "Porqu\u00E9nopuedensimplementehablarenEspa\u00F1ol",
+    "encoded": "PorqunopuedensimplementehablarenEspaol-fmd56a"
+},
+{
+    "description": "Vietnamese",
+    "decoded": "T\u1EA1isaoh\u1ECDkh\u00F4ngth\u1EC3ch\u1EC9n\u00F3iti\u1EBFngVi\u1EC7t",
+    "encoded": "TisaohkhngthchnitingVit-kjcr8268qyxafd2f1b9g"
+},
+{
+    "decoded": "3\u5E74B\u7D44\u91D1\u516B\u5148\u751F",
+    "encoded": "3B-ww4c5e180e575a65lsy2b"
+},
+{
+    "decoded": "\u5B89\u5BA4\u5948\u7F8E\u6075-with-SUPER-MONKEYS",
+    "encoded": "-with-SUPER-MONKEYS-pc58ag80a8qai00g7n9n"
+},
+{
+    "decoded": "Hello-Another-Way-\u305D\u308C\u305E\u308C\u306E\u5834\u6240",
+    "encoded": "Hello-Another-Way--fc4qua05auwb3674vfr0b"
+},
+{
+    "decoded": "\u3072\u3068\u3064\u5C4B\u6839\u306E\u4E0B2",
+    "encoded": "2-u9tlzr9756bt3uc0v"
+},
+{
+    "decoded": "Maji\u3067Koi\u3059\u308B5\u79D2\u524D",
+    "encoded": "MajiKoi5-783gue6qz075azm5e"
+},
+{
+    "decoded": "\u30D1\u30D5\u30A3\u30FCde\u30EB\u30F3\u30D0",
+    "encoded": "de-jg4avhby1noc0d"
+},
+{
+    "decoded": "\u305D\u306E\u30B9\u30D4\u30FC\u30C9\u3067",
+    "encoded": "d9juau41awczczp"
+},
+{
+    "description": "ASCII string that breaks the existing rules for host-name labels (It's not a realistic example for IDNA, because IDNA never encodes pure ASCII labels.)",
+    "decoded": "-> $1.00 <-",
+    "encoded": "-> $1.00 <--"
+}
+]

third_party/rust/idna/tests/tests.rs

@@ -0,0 +1,25 @@
+extern crate idna;
+extern crate rustc_serialize;
+extern crate test;
+
+mod punycode;
+mod uts46;
+
+fn main() {
+    let mut tests = Vec::new();
+    {
+        let mut add_test = |name, run| {
+            tests.push(test::TestDescAndFn {
+                desc: test::TestDesc {
+                    name: test::DynTestName(name),
+                    ignore: false,
+                    should_panic: test::ShouldPanic::No,
+                },
+                testfn: run,
+            })
+        };
+        punycode::collect_tests(&mut add_test);
+        uts46::collect_tests(&mut add_test);
+    }
+    test::test_main(&std::env::args().collect::<Vec<_>>(), tests)
+}

third_party/rust/idna/tests/uts46.rs

@@ -0,0 +1,117 @@
+// Copyright 2013-2014 Valentin Gosu.
+//
+// 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 std::char;
+use idna::uts46;
+use test::TestFn;
+
+pub fn collect_tests<F: FnMut(String, TestFn)>(add_test: &mut F) {
+    // http://www.unicode.org/Public/idna/latest/IdnaTest.txt
+    for (i, line) in include_str!("IdnaTest.txt").lines().enumerate() {
+        if line == "" || line.starts_with("#") {
+            continue
+        }
+        // Remove comments
+        let mut line = match line.find("#") {
+            Some(index) => &line[0..index],
+            None => line
+        };
+
+        let mut expected_failure = false;
+        if line.starts_with("XFAIL") {
+            expected_failure = true;
+            line = &line[5..line.len()];
+        };
+
+        let mut pieces = line.split(';').map(|x| x.trim()).collect::<Vec<&str>>();
+
+        let test_type = pieces.remove(0);
+        let original = pieces.remove(0);
+        let source = unescape(original);
+        let to_unicode = pieces.remove(0);
+        let to_ascii = pieces.remove(0);
+        let nv8 = if pieces.len() > 0 { pieces.remove(0) } else { "" };
+
+        if expected_failure {
+            continue;
+        }
+
+        let test_name = format!("UTS #46 line {}", i + 1);
+        add_test(test_name, TestFn::dyn_test_fn(move || {
+            let result = uts46::to_ascii(&source, uts46::Flags {
+                use_std3_ascii_rules: true,
+                transitional_processing: test_type == "T",
+                verify_dns_length: true,
+            });
+
+            if to_ascii.starts_with("[") {
+                if to_ascii.starts_with("[C") {
+                    // http://unicode.org/reports/tr46/#Deviations
+                    // applications that perform IDNA2008 lookup are not required to check
+                    // for these contexts
+                    return;
+                }
+                let res = result.ok();
+                assert!(res == None, "Expected error. result: {} | original: {} | source: {}",
+                        res.unwrap(), original, source);
+                return;
+            }
+
+            let to_ascii = if to_ascii.len() > 0 {
+                to_ascii.to_string()
+            } else {
+                if to_unicode.len() > 0 {
+                    to_unicode.to_string()
+                } else {
+                    source.clone()
+                }
+            };
+
+            if nv8 == "NV8" {
+                // This result isn't valid under IDNA2008. Skip it
+                return;
+            }
+
+            assert!(result.is_ok(), "Couldn't parse {} | original: {} | error: {:?}",
+                    source, original, result.err());
+            let output = result.ok().unwrap();
+            assert!(output == to_ascii, "result: {} | expected: {} | original: {} | source: {}",
+                    output, to_ascii, original, source);
+        }))
+    }
+}
+
+fn unescape(input: &str) -> String {
+    let mut output = String::new();
+    let mut chars = input.chars();
+    loop {
+        match chars.next() {
+            None => return output,
+            Some(c) =>
+                if c == '\\' {
+                    match chars.next().unwrap() {
+                        '\\' => output.push('\\'),
+                        'u' => {
+                            let c1 = chars.next().unwrap().to_digit(16).unwrap();
+                            let c2 = chars.next().unwrap().to_digit(16).unwrap();
+                            let c3 = chars.next().unwrap().to_digit(16).unwrap();
+                            let c4 = chars.next().unwrap().to_digit(16).unwrap();
+                            match char::from_u32((((c1 * 16 + c2) * 16 + c3) * 16 + c4))
+                            {
+                                Some(c) => output.push(c),
+                                None => { output.push_str(&format!("\\u{:X}{:X}{:X}{:X}",c1,c2,c3,c4)); }
+                            };
+                        }
+                        _ => panic!("Invalid test data input"),
+                    }
+                } else {
+                    output.push(c);
+                }
+        }
+    }
+}