Merge pull request #241 from avdv/fix-240

Treat unicode hex digits case-insensitively
This commit is contained in:
Alex Crichton 2018-05-09 16:15:56 -05:00 committed by GitHub
commit 03d3e56461
3 changed files with 11 additions and 8 deletions

View file

@ -2,6 +2,7 @@ use std::borrow::Cow;
use std::char; use std::char;
use std::str; use std::str;
use std::string; use std::string;
use std::string::String as StdString;
use self::Token::*; use self::Token::*;
@ -369,19 +370,15 @@ impl<'a> Tokenizer<'a> {
} }
fn hex(&mut self, start: usize, i: usize, len: usize) -> Result<char, Error> { fn hex(&mut self, start: usize, i: usize, len: usize) -> Result<char, Error> {
let mut val = 0; let mut buf = StdString::with_capacity(len);
for _ in 0..len { for _ in 0..len {
match self.one() { match self.one() {
Some((_, ch)) if '0' <= ch && ch <= '9' => { Some((_, ch)) if ch as u32 <= 0x7F && ch.is_digit(16) => buf.push(ch),
val = val * 16 + (ch as u32 - '0' as u32);
}
Some((_, ch)) if 'A' <= ch && ch <= 'F' => {
val = val * 16 + (ch as u32 - 'A' as u32) + 10;
}
Some((i, ch)) => return Err(Error::InvalidHexEscape(i, ch)), Some((i, ch)) => return Err(Error::InvalidHexEscape(i, ch)),
None => return Err(Error::UnterminatedString(start)), None => return Err(Error::UnterminatedString(start)),
} }
} }
let val = u32::from_str_radix(&buf, 16).unwrap();
match char::from_u32(val) { match char::from_u32(val) {
Some(ch) => Ok(ch), Some(ch) => Ok(ch),
None => Err(Error::InvalidEscapeValue(i, val)), None => Err(Error::InvalidEscapeValue(i, val)),

View file

@ -1,5 +1,8 @@
{ {
"answer1": {"type": "string", "value": "\u000B"}, "answer1": {"type": "string", "value": "\u000B"},
"answer4": {"type": "string", "value": "\u03B4α"}, "answer4": {"type": "string", "value": "\u03B4α"},
"answer8": {"type": "string", "value": "\u03B4β"} "answer8": {"type": "string", "value": "\u03B4β"},
"answer9": {"type": "string", "value": "\uc0de"},
"answer10": {"type": "string", "value": "\u03B4α"},
"answer11": {"type": "string", "value": "\uABC1"}
} }

View file

@ -1,3 +1,6 @@
answer1 = "\u000B" answer1 = "\u000B"
answer4 = "\u03B4α" answer4 = "\u03B4α"
answer8 = "\U000003B4β" answer8 = "\U000003B4β"
answer9 = "\uc0de"
answer10 = "\u03b4α"
answer11 = "\U0000abc1"