Syntax 0.8, part 7: \UHHHHHH and other changes #314
Conversation
stasm
force-pushed
the
zeroeight-part7
branch
from
4e6cfc1 to
b2a932e
Compare
|
This PR depends on #313 which hasn't been merged yet. @Pike, if you want to review this first, the compare view might be helpful. Thanks. |
Pike
left a comment
Pike
left a comment
There was a problem hiding this comment.
Just a nit about checking u and U twice, otherwise, nice, r=me.
| for (let i = 0; i < 4; i++) { | ||
| const ch = ps.takeHexDigit(); | ||
| getUnicodeEscapeSequence(ps, u, digits) { | ||
| ps.expectChar(u); |
There was a problem hiding this comment.
We already checked this in the switch statement, I'd just ps.next() here, and drop the u argument.
There was a problem hiding this comment.
I'd prefer to keep expectChar. In fact, I'm planning a clean up after 0.8 where we get rid of most of ps.next(). I always seem to forget what it's supposed to consume, something expectChar is very explicit about.
There was a problem hiding this comment.
Oh, and we need the u arg for the error message. That's the real reason why it's there.
For compatibility with the reference parser which used to use the /./
regex, the tooling parser used to recognize additional EOL characters in
Comments. This has since been fixed in the reference parser:
projectfluent/fluent@2d224cb
stasm
force-pushed
the
zeroeight-part7
branch
from
b2a932e to
c8ecf45
Compare
2 participants
Implement part 7 of #303.
Recognize \UHHHHHH as an escape sequence fluent#201, tooling parser, runtime parser
Change line endings in comments to line_end fluent#219, tooling parser
Forbid nested VariantLists fluent#220, tooling parser