REGRESSION (r146588): Cannot correctly display Chinese SNS Renren

The layout of the main content area is not right

[Alexey Proskuryakov]

Could you please provide more information, such as a screenshot?

I just tried opening <http://www.renren.com> in Safari 6.0.3 and in current build on OS X 10.8.3, and both looked good without any difference.

Created attachment 194923 [details]
The nightly build screenshot.

I have circled some area where the layout is wrong.

Created attachment 194924 [details]
The correct layout.

These screenshots contains my personal information. I'd appreciate it if you could delete the above after you download them.

[Alexey Proskuryakov]

<rdar://problem/13501595>

[Mark Rowe (bdash)]

The content of attachment 194923 [details] has been deleted by
    Mark Rowe (bdash) <[e-mail]>
who provided the following reason:

As requested by the uploader.

The token used to delete this attachment was generated at 2013-03-25 17:31:19 PST.

[Mark Rowe (bdash)]

The content of attachment 194924 [details] has been deleted by
    Mark Rowe (bdash) <[e-mail]>
who provided the following reason:

As requested by the uploader.

The token used to delete this attachment was generated at 2013-03-25 17:31:36 PST.

[David Kilzer (:ddkilzer)]

A *.webarchive of the page with issues would likely be more useful than screenshots.

Can you save a *.webarchive of the page and send it to Alexey or me?

Created attachment 194996 [details]
.webarchive of Safari and WebKit

Here it is.

In current build, the bug haven't fixed yet. :)

I uploaded screenshots in another thread, not it seems unnecessary..

[Alexey Proskuryakov]

That's bug 111856. I'm not sure if it's the same issue, or a different one.

That's right.
These two threads point to the same issue on the renren.com ...for several days(3 or 5)...

(In reply to comment #11)
> That's bug 111856. I'm not sure if it's the same issue, or a different one.

[zalan]

CSS parsing doesn't recover from broken syntax.

after the following, broken, style declaration, 

div {
    ;*display:expression(function(){})
}

the next valid style declaration does not get applied

#foo {
    color: red;
}

see attached reduction case.

Apparently the changes in CSSGrammar.y.in (r146588) causes this regression.

bugzilla can't find ​[e-mail] (author) so assigned it to the reviewer of r146588

[zalan]

Created attachment 195270 [details]
test reduction

[zalan]

*** Bug 113388 has been marked as a duplicate of this bug. ***

[Sergey Ryazanov]

This is a problem similar to https://bugs.webkit.org/show_bug.cgi?id=112880. Patch from this bug fixes this issue as well as a few more tests that fail on WebKit before r146588.

[zalan]

Thanks for the update.

(In reply to comment #16)
> This is a problem similar to https://bugs.webkit.org/show_bug.cgi?id=112880. Patch from this bug fixes this issue as well as a few more tests that fail on WebKit before r146588.

[Sergey Ryazanov]

Created attachment 195301 [details]
Patch

Thank you for the update...but why can't I still get the correct layout? I was using 146499 which works properly...

(In reply to comment #18)
> Created an attachment (id=195301) [details]
> Patch

[zalan]

The patch hasnt been landed yet.
 
(In reply to comment #19)
> Thank you for the update...but why can't I still get the correct layout? I was using 146499 which works properly...
> 
> (In reply to comment #18)
> > Created an attachment (id=195301) [details] [details]
> > Patch

[WebKit Review Bot]

Comment on attachment 195301 [details]
Patch

Clearing flags on attachment: 195301

Committed r147028: <http://trac.webkit.org/changeset/147028>

[WebKit Review Bot]

All reviewed patches have been landed.  Closing bug.