604172 - at-import-009 in CSS 2.1 test suite fails

Status: RESOLVED FIXED

Description

[David Baron :dbaron: ⌚️UTC-7]

http://test.csswg.org/suites/css2.1/20100917/xhtml1/at-import-009.xht fails; there are actually two underlying bugs:
 (1) we're missing an UngetToken in error handling of parsing @charset rules
 (2) GatherMedia needs to stop at a block or semicolon in invalid cases

I have a patch for both in my patch queue, but I seem to recall that I wasn't quite happy with the patch for the second:

http://hg.mozilla.org/users/dbaron_mozilla.com/patches/raw-file/6038c95fb322/at-charset-unget-token
http://hg.mozilla.org/users/dbaron_mozilla.com/patches/raw-file/6038c95fb322/gather-media-stop-in-at-rule

Comment 1

[David Baron :dbaron: ⌚️UTC-7]

Created attachment 484548 [details] [diff] [review]
patch 1: missing UngetToken

Comment 2

[David Baron :dbaron: ⌚️UTC-7]

Created attachment 484550 [details] [diff] [review]
patch 2: use more stop characters in GatherMedia

In theory, stop characters are a bad approach for everything except skipping over garbage.  I thought about removing other uses, but that makes error handling hard.  However, including extra stop chars isn't harmful in the other cases (other than skipping over garbage), since they're not valid syntax.

Comment 3

[Boris Zbarsky [:bz] (if a patch has no decent message, automatic r-)]

Comment on attachment 484548 [details] [diff] [review]
patch 1: missing UngetToken

r=me

Comment 4

[Boris Zbarsky [:bz] (if a patch has no decent message, automatic r-)]

Comment on attachment 484550 [details] [diff] [review]
patch 2: use more stop characters in GatherMedia

r=me

Comment 5

[David Baron :dbaron: ⌚️UTC-7]

https://hg.mozilla.org/projects/birch/rev/35b1ff9a626e
https://hg.mozilla.org/projects/birch/rev/60cef935a89c

Comment 6

[David Baron :dbaron: ⌚️UTC-7]

https://hg.mozilla.org/mozilla-central/rev/35b1ff9a626e
https://hg.mozilla.org/mozilla-central/rev/60cef935a89c