tree 21a35a57690d7e7e0da5742cd0646a8de6d11304
parent 2ab5f5f1f4dcd6373db8edfa9ae3eb69855d2843
author Leon Senft <leonsenft@google.com> 1487902203 -0800
committer Leon Senft <leonsenft@google.com> 1487902203 -0800

Fixes parsing string with unicode-range sequences

Strings with character sequences that resemble `unicode-range` property values,
`U+0-10FFFF`, would be missing the `U+` sequence in the parsed output. This is
because the tokenizer doesn't tokenize strings as their own tokens, and midway
through a string it was trying to emit a unicode range token. We now check to
see if we're currently parsing a string before interpreting `U+` as the
beginning of a unicode range.