WebKit Bugzilla
New
Browse
Log In
×
Sign in with GitHub
or
Remember my login
Create Account
·
Forgot Password
Forgotten password account recovery
REOPENED
Bug 33979
-webkit-mask-box-image draws a box while loading
https://bugs.webkit.org/show_bug.cgi?id=33979
Summary
-webkit-mask-box-image draws a box while loading
Simon Fraser (smfr)
Reported
2010-01-21 17:00:45 PST
-webkit-mask-box-image draws a box while loading
Attachments
Patch
(21.09 KB, patch)
2010-01-21 17:06 PST
,
Simon Fraser (smfr)
no flags
Details
Formatted Diff
Diff
View All
Add attachment
proposed patch, testcase, etc.
Simon Fraser (smfr)
Comment 1
2010-01-21 17:06:27 PST
Created
attachment 47160
[details]
Patch
mitz
Comment 2
2010-01-21 17:16:55 PST
Comment on
attachment 47160
[details]
Patch Change looks fine. Resorting to nit-picking.
> + <
rdar://problem/7378662
> > + > + > + > + * http/tests/misc/resources/slow-png-load.pl: Added.
Too many newlines.
> \ No newline at end of file
Too few newlines.
> + * rendering/RenderBox.cpp: > + (WebCore::RenderBox::paintMaskImages): > + * rendering/style/FillLayer.cpp: > + (WebCore::FillLayer::areImagesLoaded): > + * rendering/style/FillLayer.h:
Explain what you did there?
> + bool areImagesLoaded() const;
I prefer imagesAreLoaded() or hasLoaded[All]Images().
Simon Fraser (smfr)
Comment 3
2010-01-21 17:28:30 PST
http://trac.webkit.org/changeset/53663
Robert Hogan
Comment 4
2011-05-29 10:52:10 PDT
This test is puzzling me. It fails on Chromium and Qt - and I suspect it really fails on gtk as you can get the same rendertree results regardless of whether you really pass or not ('pass' being defined as not displaying the blue background of the boxes). The problem, as I see it, is that the blue background always gets rendered before the mask in RenderBlock::paintObject() so it will always get painted. Whether the images are loaded or not in RenderBox::paintMaskImages() seems immaterial, since there is nothing there that will paint over the already-rendered background. The other thing I can't get my head around is that isLoaded() is always true for a StyleImage so the code added with this patch will always think the images have been loaded. Maybe something has changed since this patch and these are regressions of some sort. More likely I'm not understanding the code properly. Simon, could you take another look at it? Does it still pass manually on mac?
Simon Fraser (smfr)
Comment 5
2011-05-29 11:08:06 PDT
The pixel test is now failing, so this regressed at some point.
Darin Adler
Comment 6
2011-06-18 12:19:43 PDT
Comment on
attachment 47160
[details]
Patch Clearing flag on original patch since this bug has been re-broken and needs a new fix.
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