WebKit Bugzilla
New
Browse
Log In
×
Sign in with GitHub
or
Remember my login
Create Account
·
Forgot Password
Forgotten password account recovery
RESOLVED FIXED
115765
Begin unraveling the mess that is QuotesData
https://bugs.webkit.org/show_bug.cgi?id=115765
Summary
Begin unraveling the mess that is QuotesData
Anders Carlsson
Reported
2013-05-07 15:25:15 PDT
Begin unraveling the mess that is QuoteData
Attachments
Patch
(11.26 KB, patch)
2013-05-07 15:29 PDT
,
Anders Carlsson
kling
: review+
Details
Formatted Diff
Diff
View All
Add attachment
proposed patch, testcase, etc.
Anders Carlsson
Comment 1
2013-05-07 15:29:43 PDT
Created
attachment 200987
[details]
Patch
Andreas Kling
Comment 2
2013-05-07 15:38:52 PDT
Comment on
attachment 200987
[details]
Patch View in context:
https://bugs.webkit.org/attachment.cgi?id=200987&action=review
r=me, but you'll need to unbreak the RenderStyle optimization.
> Source/WebCore/ChangeLog:3 > + Begin unraveling the mess that is QuoteData
QuoteData => QuotesData
> Source/WebCore/rendering/style/QuotesData.cpp:30 > + quotes.reserveCapacity(2);
You should use reserveInitialCapacity() here.
> Source/WebCore/rendering/style/QuotesData.h:35 > + static PassRefPtr<QuotesData> create(const Vector<std::pair<String, String> >& quotes);
This could also be QuotesData::adopt(Voctor<>&) and use swap() to avoid copying the data.
> Source/WebCore/rendering/style/RenderStyle.cpp:755 > - if (QuotesData::equals(rareInheritedData->quotes.get(), q.get())) > - return; > rareInheritedData.access()->quotes = q;
You are removing an optimization here! rareInheritedData.access() is DataRef<T>'s copy-on-write accessor, this means that setQuotes() will always detach from shared rareInheritedData.
Anders Carlsson
Comment 3
2013-05-07 16:06:26 PDT
Committed
r149700
: <
http://trac.webkit.org/changeset/149700
>
Roger Fong
Comment 4
2013-05-07 17:33:57 PDT
This caused a lot of test failures:
http://build.webkit.org/results/Apple%20MountainLion%20Release%20WK2%20(Tests)/r149700%20(8675)/results.html
Alexey Proskuryakov
Comment 5
2013-05-07 23:20:51 PDT
> This caused a lot of test failures:
What happened with this? Looks like tests got happier after
r149707
, but still not quite: <
http://build.webkit.org/results/Apple%20MountainLion%20Release%20WK2%20(Tests)/r149708%20(8680)/results.html
>.
Ryosuke Niwa
Comment 6
2013-05-07 23:54:58 PDT
fast/css-generated-content/close-quote-negative-depth.html is still failing:
http://webkit-test-results.appspot.com/dashboards/flakiness_dashboard.html#tests=fast%2Fcss-generated-content%2Fclose-quote-negative-depth.html
Alexey Proskuryakov
Comment 7
2013-05-08 00:14:29 PDT
Ryosuke filed
bug 115776
about that.
Note
You need to
log in
before you can comment on or make changes to this bug.
Top of Page
Format For Printing
XML
Clone This Bug