diff options
author | Claudio Bley <claudio.bley@gmail.com> | 2018-05-08 23:02:24 +0200 |
---|---|---|
committer | Claudio Bley <claudio.bley@gmail.com> | 2018-05-09 22:18:14 +0200 |
commit | 264d828654ff323595e28c8a1f08214481f40f9c (patch) | |
tree | 88a818734b16170ebd90176bb0e83fada510a5f7 /src | |
parent | 44bec613697e884df64634536554c611d3135b3b (diff) | |
download | milf-rs-264d828654ff323595e28c8a1f08214481f40f9c.tar.gz milf-rs-264d828654ff323595e28c8a1f08214481f40f9c.zip |
Treat unicode hex digits case-insensitively
In Rust >= 1.24.0 we could have used `char::is_ascii_hexdigit`, but to keep
compatiblity with older versions, `char::is_digit(16)` is used.
Fixes #240.
Diffstat (limited to 'src')
-rw-r--r-- | src/tokens.rs | 11 |
1 files changed, 4 insertions, 7 deletions
diff --git a/src/tokens.rs b/src/tokens.rs index bcabd94..3f47f02 100644 --- a/src/tokens.rs +++ b/src/tokens.rs @@ -2,6 +2,7 @@ use std::borrow::Cow; use std::char; use std::str; use std::string; +use std::string::String as StdString; use self::Token::*; @@ -369,19 +370,15 @@ impl<'a> Tokenizer<'a> { } 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 { match self.one() { - Some((_, ch)) if '0' <= ch && ch <= '9' => { - 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((_, ch)) if ch as u32 <= 0x7F && ch.is_digit(16) => buf.push(ch), Some((i, ch)) => return Err(Error::InvalidHexEscape(i, ch)), None => return Err(Error::UnterminatedString(start)), } } + let val = u32::from_str_radix(&buf, 16).unwrap(); match char::from_u32(val) { Some(ch) => Ok(ch), None => Err(Error::InvalidEscapeValue(i, val)), |