WebKit-https.git
7 years agoAdd /usr/bin to PATH so PrettyPatch can find git.
wsiegrist@apple.com [Fri, 20 Jul 2012 18:43:10 +0000 (18:43 +0000)]
Add /usr/bin to PATH so PrettyPatch can find git.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@123245 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoUnreviewed GTK gardening, adding image baseline for
zandobersek@gmail.com [Fri, 20 Jul 2012 18:40:29 +0000 (18:40 +0000)]
Unreviewed GTK gardening, adding image baseline for
fast/images/paletted-png-with-color-profile.html.

* platform/gtk/TestExpectations:
* platform/gtk/fast/images/paletted-png-with-color-profile-expected.png: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@123244 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoREGRESSION(r122873): 15% regression on Dromaeo/dom-attr
rniwa@webkit.org [Fri, 20 Jul 2012 18:28:24 +0000 (18:28 +0000)]
REGRESSION(r122873): 15% regression on Dromaeo/dom-attr
https://bugs.webkit.org/show_bug.cgi?id=91827

Reviewed by Anders Carlsson.

Move shouldInvalidateNodeListCaches from Document.cpp to Node.cpp since it's only called
in Node::invalidateNodeListCachesInAncestors.

Test: PerformanceTests/Dromaeo/dom-attr.html.

* dom/Document.cpp:
(WebCore):
* dom/Node.cpp:
(WebCore::shouldInvalidateNodeListCachesForAttr): Extracted from shouldInvalidateNodeListCaches
to unroll the loop in shouldInvalidateNodeListCaches. Apparently gcc wasn't doing the right thing.
(WebCore::Document::shouldInvalidateNodeListCaches):
(WebCore::Document::invalidateNodeListCaches):
(WebCore::Node::invalidateNodeListCachesInAncestors):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@123243 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoSVG Filter Effect sub-region not applied for some filters
schenney@chromium.org [Fri, 20 Jul 2012 18:27:06 +0000 (18:27 +0000)]
SVG Filter Effect sub-region not applied for some filters
https://bugs.webkit.org/show_bug.cgi?id=89767

Reviewed by Dirk Schulze.

Unreviewed test expectations update.

Numerous results changed by one pixel due to the difference between
clipping and pre-applying the filter subregion. Some feGaussianBlur
and feDropShadow results changed more, but that was expected.

* platform/chromium-linux-x86/svg/batik/filters: Removed.
* platform/chromium-linux-x86/svg/filters/shadow-on-filter-expected.png:
* platform/chromium-linux-x86/svg/filters/shadow-on-rect-with-filter-expected.png:
* platform/chromium-linux/svg/W3C-SVG-1.1/filters-color-01-b-expected.png:
* platform/chromium-linux/svg/W3C-SVG-1.1/filters-comptran-01-b-expected.png:
* platform/chromium-linux/svg/W3C-SVG-1.1/filters-example-01-b-expected.png:
* platform/chromium-linux/svg/W3C-SVG-1.1/filters-gauss-01-b-expected.png:
* platform/chromium-linux/svg/W3C-SVG-1.1/filters-morph-01-f-expected.png:
* platform/chromium-linux/svg/batik/filters/feTile-expected.png:
* platform/chromium-linux/svg/batik/filters/filterRegions-expected.png:
* platform/chromium-linux/svg/batik/text/smallFonts-expected.png:
* platform/chromium-linux/svg/batik/text/textFeatures-expected.png:
* platform/chromium-linux/svg/custom/feComponentTransfer-Discrete-expected.png:
* platform/chromium-linux/svg/custom/feComponentTransfer-Gamma-expected.png:
* platform/chromium-linux/svg/custom/feComponentTransfer-Linear-expected.png:
* platform/chromium-linux/svg/custom/feComponentTransfer-Table-expected.png:
* platform/chromium-linux/svg/dynamic-updates/SVGFEGaussianBlurElement-dom-in-attr-expected.png:
* platform/chromium-linux/svg/dynamic-updates/SVGFEGaussianBlurElement-dom-stdDeviation-attr-expected.png:
* platform/chromium-linux/svg/dynamic-updates/SVGFEGaussianBlurElement-dom-stdDeviation-call-expected.png:
* platform/chromium-linux/svg/dynamic-updates/SVGFEGaussianBlurElement-svgdom-in-prop-expected.png:
* platform/chromium-linux/svg/filters/feDropShadow-expected.png:
* platform/chromium-linux/svg/filters/filter-on-tspan-expected.png:
* platform/chromium-linux/svg/filters/filterRes-expected.png:
* platform/chromium-linux/svg/filters/shadow-on-filter-expected.png:
* platform/chromium-linux/svg/filters/shadow-on-rect-with-filter-expected.png:
* platform/chromium-linux/svg/filters/subRegion-one-effect-expected.png:
* platform/chromium-linux/svg/filters/subRegion-two-effects-expected.png:
* platform/chromium-mac-snowleopard/svg/W3C-SVG-1.1/filters-morph-01-f-expected.png:
* platform/chromium-mac-snowleopard/svg/batik/filters/feTile-expected.png:
* platform/chromium-mac-snowleopard/svg/batik/filters/filterRegions-expected.png:
* platform/chromium-mac-snowleopard/svg/batik/text/smallFonts-expected.png:
* platform/chromium-mac-snowleopard/svg/batik/text/textFeatures-expected.png:
* platform/chromium-mac-snowleopard/svg/custom/feComponentTransfer-Discrete-expected.png:
* platform/chromium-mac-snowleopard/svg/custom/feComponentTransfer-Gamma-expected.png:
* platform/chromium-mac-snowleopard/svg/custom/feComponentTransfer-Linear-expected.png:
* platform/chromium-mac-snowleopard/svg/custom/feComponentTransfer-Table-expected.png:
* platform/chromium-mac-snowleopard/svg/dynamic-updates/SVGFEGaussianBlurElement-dom-in-attr-expected.png:
* platform/chromium-mac-snowleopard/svg/dynamic-updates/SVGFEGaussianBlurElement-dom-stdDeviation-attr-expected.png:
* platform/chromium-mac-snowleopard/svg/dynamic-updates/SVGFEGaussianBlurElement-dom-stdDeviation-call-expected.png:
* platform/chromium-mac-snowleopard/svg/dynamic-updates/SVGFEGaussianBlurElement-svgdom-in-prop-expected.png:
* platform/chromium-mac-snowleopard/svg/filters/shadow-on-filter-expected.png:
* platform/chromium-mac-snowleopard/svg/filters/shadow-on-rect-with-filter-expected.png:
* platform/chromium-mac/svg/W3C-SVG-1.1/filters-color-01-b-expected.png:
* platform/chromium-mac/svg/W3C-SVG-1.1/filters-comptran-01-b-expected.png:
* platform/chromium-mac/svg/W3C-SVG-1.1/filters-example-01-b-expected.png:
* platform/chromium-mac/svg/W3C-SVG-1.1/filters-gauss-01-b-expected.png:
* platform/chromium-mac/svg/W3C-SVG-1.1/filters-morph-01-f-expected.png:
* platform/chromium-mac/svg/batik/filters/feTile-expected.png:
* platform/chromium-mac/svg/batik/filters/filterRegions-expected.png:
* platform/chromium-mac/svg/batik/text/smallFonts-expected.png:
* platform/chromium-mac/svg/batik/text/textFeatures-expected.png:
* platform/chromium-mac/svg/custom/feComponentTransfer-Discrete-expected.png:
* platform/chromium-mac/svg/custom/feComponentTransfer-Gamma-expected.png:
* platform/chromium-mac/svg/custom/feComponentTransfer-Linear-expected.png:
* platform/chromium-mac/svg/custom/feComponentTransfer-Table-expected.png:
* platform/chromium-mac/svg/dynamic-updates/SVGFEGaussianBlurElement-dom-in-attr-expected.png:
* platform/chromium-mac/svg/dynamic-updates/SVGFEGaussianBlurElement-dom-stdDeviation-attr-expected.png:
* platform/chromium-mac/svg/dynamic-updates/SVGFEGaussianBlurElement-dom-stdDeviation-call-expected.png:
* platform/chromium-mac/svg/dynamic-updates/SVGFEGaussianBlurElement-svgdom-in-prop-expected.png:
* platform/chromium-mac/svg/filters/feDropShadow-expected.png:
* platform/chromium-mac/svg/filters/feDropShadow-subregion-expected.png: Added.
* platform/chromium-mac/svg/filters/feGaussianBlur-expected.png:
* platform/chromium-mac/svg/filters/feGaussianBlur-subregion-expected.png: Added.
* platform/chromium-mac/svg/filters/feImage-subregions-expected.png:
* platform/chromium-mac/svg/filters/feImage-subregions-preseveAspectRatio-none-expected.png:
* platform/chromium-mac/svg/filters/feImage-subregions-preseveAspectRatio-none-with-viewBox-expected.png:
* platform/chromium-mac/svg/filters/feOffset-expected.png:
* platform/chromium-mac/svg/filters/feTile-expected.png:
* platform/chromium-mac/svg/filters/filter-placement-issue-expected.png:
* platform/chromium-mac/svg/filters/filter-rounding-issues-expected.png:
* platform/chromium-mac/svg/filters/filterRes-expected.png:
* platform/chromium-mac/svg/filters/filterRes1-expected.png:
* platform/chromium-mac/svg/filters/filterRes3-expected.png:
* platform/chromium-mac/svg/filters/parent-children-with-same-filter-expected.png:
* platform/chromium-mac/svg/filters/shadow-on-filter-expected.png:
* platform/chromium-mac/svg/filters/shadow-on-rect-with-filter-expected.png:
* platform/chromium-mac/svg/filters/subRegion-one-effect-expected.png:
* platform/chromium-mac/svg/filters/subRegion-two-effects-expected.png:
* platform/chromium-win-xp/svg/filters/shadow-on-filter-expected.png:
* platform/chromium-win-xp/svg/filters/shadow-on-rect-with-filter-expected.png:
* platform/chromium-win/svg/W3C-SVG-1.1/filters-color-01-b-expected.png:
* platform/chromium-win/svg/W3C-SVG-1.1/filters-comptran-01-b-expected.png:
* platform/chromium-win/svg/W3C-SVG-1.1/filters-example-01-b-expected.png:
* platform/chromium-win/svg/W3C-SVG-1.1/filters-gauss-01-b-expected.png:
* platform/chromium-win/svg/W3C-SVG-1.1/filters-morph-01-f-expected.png:
* platform/chromium-win/svg/batik/filters/feTile-expected.png:
* platform/chromium-win/svg/batik/filters/filterRegions-expected.png:
* platform/chromium-win/svg/batik/text/smallFonts-expected.png:
* platform/chromium-win/svg/batik/text/textFeatures-expected.png:
* platform/chromium-win/svg/custom/feComponentTransfer-Discrete-expected.png:
* platform/chromium-win/svg/custom/feComponentTransfer-Gamma-expected.png:
* platform/chromium-win/svg/custom/feComponentTransfer-Linear-expected.png:
* platform/chromium-win/svg/custom/feComponentTransfer-Table-expected.png:
* platform/chromium-win/svg/dynamic-updates/SVGFEGaussianBlurElement-dom-in-attr-expected.png:
* platform/chromium-win/svg/dynamic-updates/SVGFEGaussianBlurElement-dom-stdDeviation-attr-expected.png:
* platform/chromium-win/svg/dynamic-updates/SVGFEGaussianBlurElement-dom-stdDeviation-call-expected.png:
* platform/chromium-win/svg/dynamic-updates/SVGFEGaussianBlurElement-svgdom-in-prop-expected.png:
* platform/chromium-win/svg/filters/feDropShadow-expected.png:
* platform/chromium-win/svg/filters/feDropShadow-subregion-expected.png: Added.
* platform/chromium-win/svg/filters/feGaussianBlur-expected.png:
* platform/chromium-win/svg/filters/feGaussianBlur-subregion-expected.png: Added.
* platform/chromium-win/svg/filters/feImage-subregions-expected.png:
* platform/chromium-win/svg/filters/feImage-subregions-preseveAspectRatio-none-expected.png:
* platform/chromium-win/svg/filters/feImage-subregions-preseveAspectRatio-none-with-viewBox-expected.png:
* platform/chromium-win/svg/filters/feOffset-expected.png:
* platform/chromium-win/svg/filters/feTile-expected.png:
* platform/chromium-win/svg/filters/filter-on-tspan-expected.png:
* platform/chromium-win/svg/filters/filter-placement-issue-expected.png:
* platform/chromium-win/svg/filters/filter-rounding-issues-expected.png:
* platform/chromium-win/svg/filters/filterRes-expected.png:
* platform/chromium-win/svg/filters/filterRes1-expected.png:
* platform/chromium-win/svg/filters/filterRes3-expected.png:
* platform/chromium-win/svg/filters/parent-children-with-same-filter-expected.png:
* platform/chromium-win/svg/filters/shadow-on-filter-expected.png:
* platform/chromium-win/svg/filters/shadow-on-rect-with-filter-expected.png:
* platform/chromium-win/svg/filters/subRegion-one-effect-expected.png:
* platform/chromium-win/svg/filters/subRegion-two-effects-expected.png:
* platform/chromium/TestExpectations:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@123242 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years ago[Chromium] Add WebView::zoomToFindInPageRect for Android
abarth@webkit.org [Fri, 20 Jul 2012 18:19:39 +0000 (18:19 +0000)]
[Chromium] Add WebView::zoomToFindInPageRect for Android
https://bugs.webkit.org/show_bug.cgi?id=91801

Reviewed by Tony Chang.

On Android, when you find in page, we scale the viewport to reveal what
you've found (because the screen is small). This function instructs the
WebView to zoom to display a given rect.

* public/WebView.h:
(WebView):
* src/WebViewImpl.cpp:
(WebKit::WebViewImpl::animateZoomAroundPoint):
    - This function is a bit more general than necessary for this patch
      because it is also used in other cases for animating a zoom
      around a given point (e.g., double-tap to zoom). These uses will
      be upstreamed in later patches.
(WebKit::WebViewImpl::zoomToFindInPageRect):
* src/WebViewImpl.h:
(WebViewImpl):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@123241 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years ago[chromium] Don't include WebCore headers in TestInterfaces so it's safe to include...
jochen@chromium.org [Fri, 20 Jul 2012 18:13:17 +0000 (18:13 +0000)]
[chromium] Don't include WebCore headers in TestInterfaces so it's safe to include from outside of WebCore
https://bugs.webkit.org/show_bug.cgi?id=91847

Reviewed by Adam Barth.

This allows for including the TestIntefaces.h from content shell
without introducing a dependency to WebCore/wtf

* DumpRenderTree/chromium/TestRunner/TestInterfaces.cpp:
(TestInterfaces::Internal):
(TestInterfaces::Internal::Internal):
(TestInterfaces::Internal::~Internal):
(TestInterfaces::Internal::bindTo):
(TestInterfaces::Internal::resetAll):
(TestInterfaces::TestInterfaces):
(TestInterfaces::~TestInterfaces):
(TestInterfaces::bindTo):
(TestInterfaces::resetAll):
* DumpRenderTree/chromium/TestRunner/TestInterfaces.h:
(TestInterfaces):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@123240 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoFix more -Wunused-private-field violations
thakis@chromium.org [Fri, 20 Jul 2012 18:11:10 +0000 (18:11 +0000)]
Fix more -Wunused-private-field violations
https://bugs.webkit.org/show_bug.cgi?id=91876

Reviewed by Adam Barth.

As of http://lists.cs.uiuc.edu/pipermail/cfe-commits/Week-of-Mon-20120716/061102.html
the warning finds more bugs. Fix them.

* html/FTPDirectoryDocument.cpp:
(FTPDirectoryDocumentParser):
(WebCore::FTPDirectoryDocumentParser::FTPDirectoryDocumentParser):
* html/parser/HTMLTreeBuilder.cpp:
(WebCore::HTMLTreeBuilder::HTMLTreeBuilder):
* html/parser/HTMLTreeBuilder.h:
(HTMLTreeBuilder):
* html/shadow/MediaControlRootElementChromium.cpp:
(WebCore::MediaControlRootElementChromium::MediaControlRootElementChromium):
* html/shadow/MediaControlRootElementChromium.h:
(MediaControlRootElementChromium):
* inspector/InspectorDOMStorageResource.cpp:
(WebCore::InspectorDOMStorageResource::InspectorDOMStorageResource):
* inspector/InspectorDOMStorageResource.h:
(InspectorDOMStorageResource):
* page/Settings.cpp:
(WebCore::Settings::Settings):
* page/Settings.h:
(Settings):
* platform/graphics/chromium/cc/CCThreadProxy.cpp:
(WebCore::CCThreadProxy::CCThreadProxy):
* platform/graphics/chromium/cc/CCThreadProxy.h:
(CCThreadProxy):
* rendering/svg/RenderSVGTextPath.cpp:
(WebCore::RenderSVGTextPath::RenderSVGTextPath):
* rendering/svg/RenderSVGTextPath.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@123239 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years ago[chromium] Unreviewed test expectations update.
jsbell@chromium.org [Fri, 20 Jul 2012 18:10:42 +0000 (18:10 +0000)]
[chromium] Unreviewed test expectations update.
Un-marking various tests as flaky on WIN following rollback of r121610
in http://trac.webkit.org/changeset/123110. Investigation documented in:
https://bugs.webkit.org/show_bug.cgi?id=90469

* platform/chromium/TestExpectations:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@123238 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years ago[Qt] Make WTR use the same color palette as DRT
caio.oliveira@openbossa.org [Fri, 20 Jul 2012 18:05:33 +0000 (18:05 +0000)]
[Qt] Make WTR use the same color palette as DRT
https://bugs.webkit.org/show_bug.cgi?id=91870

Reviewed by Noam Rosenthal.

Make sure that WebKitTestRunner and DumpRenderTree use the same palette: the
standard from QWindowsStyle.

* WebKitTestRunner/InjectedBundle/qt/ActivateFontsQt.cpp:
(WTR::activateFonts): Moved code to platformInitialize since setting style is not
exactly font related.
* WebKitTestRunner/InjectedBundle/qt/InjectedBundleQt.cpp:
(WTR::InjectedBundle::platformInitialize): Force the usage of the style's
palette.  Because at this point QApplication was already instantiated, the
default palette was already initialized to be the system one, so we need to
explicitly set the palette to be the standard from style. This is not needed in
DRT because there we set the style before instantiating QApplication.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@123237 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoUnreviewed GTK gardening.
zandobersek@gmail.com [Fri, 20 Jul 2012 18:02:17 +0000 (18:02 +0000)]
Unreviewed GTK gardening.

Generating baselines for two input placeholder tests that fail. The image
baselines used are from Chromium Linux that clearly denote what's missing
in the GTK port's output for these two tests, ignoring other marginal
rendering differences.

* platform/gtk/TestExpectations:
* platform/gtk/fast/forms/input-placeholder-paint-order-expected.png: Added.
* platform/gtk/fast/forms/input-placeholder-paint-order-expected.txt: Added.
* platform/gtk/fast/forms/textarea: Added.
* platform/gtk/fast/forms/textarea/textarea-placeholder-paint-order-expected.png: Added.
* platform/gtk/fast/forms/textarea/textarea-placeholder-paint-order-expected.txt: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@123236 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoCodeGeneratorInspector.py is unnecessarily chatty
adamk@chromium.org [Fri, 20 Jul 2012 17:56:12 +0000 (17:56 +0000)]
CodeGeneratorInspector.py is unnecessarily chatty
https://bugs.webkit.org/show_bug.cgi?id=91758

Reviewed by Vsevolod Vlasov.

The code aded in r123091 included a "print" statement for each written
file. This makes for noisy build output, especially noticeable in the
Chromium ninja build (where build output takes up a single line of the
console).

If this print statement is generally useful, it should be
hidden behind a --verbose commandline option, as we do for the binding
generators.

* inspector/CodeGeneratorInspector.py:
(SmartOutput.close): Remove print statement.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@123233 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoBaseline align text in the flakiness dashboard toolbar
tony@chromium.org [Fri, 20 Jul 2012 17:53:03 +0000 (17:53 +0000)]
Baseline align text in the flakiness dashboard toolbar
https://bugs.webkit.org/show_bug.cgi?id=91784

Reviewed by Ojan Vafai.

Baseline align the flex items.

* TestResultServer/static-dashboards/flakiness_dashboard.css:
(.forms): Add box-align: baseline
(.forms > *): Use block display to force flex items rather than box.
(#tests-form): Add box-align: baseline
(#tests-form > *): Use block display to force flex items rather than box.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@123232 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoFix missing css1/text_properties/text_transform.html results (txt)
commit-queue@webkit.org [Fri, 20 Jul 2012 17:49:15 +0000 (17:49 +0000)]
Fix missing css1/text_properties/text_transform.html results (txt)
https://bugs.webkit.org/show_bug.cgi?id=91875

Expected txt result was added.

Patch by Luciano Wolf <luciano.wolf@openbossa.org> on 2012-07-20
Reviewed by Noam Rosenthal.

* platform/qt/css1/text_properties/text_transform-expected.txt: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@123230 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years ago[Qt] Added/updated missing fast/table results after rebaseline
commit-queue@webkit.org [Fri, 20 Jul 2012 17:47:47 +0000 (17:47 +0000)]
[Qt] Added/updated missing fast/table results after rebaseline
https://bugs.webkit.org/show_bug.cgi?id=91621

Patch by Bruno de Oliveira Abinader <bruno.abinader@basyskom.com> on 2012-07-20
Reviewed by Noam Rosenthal.

The following tests were failing after r122985:
  fast/table/003.html = TEXT
  fast/table/colspanMinWidth.html = TEXT
  fast/table/spanOverlapRepaint.html = TEXT
  fast/table/text-field-baseline.html = TEXT

This patch includes/updates the missing expectations.

* platform/qt-5.0-wk2/fast/table/003-expected.png:
* platform/qt-5.0-wk2/fast/table/003-expected.txt:
* platform/qt-5.0-wk2/fast/table/colspanMinWidth-expected.png: Added.
* platform/qt-5.0-wk2/fast/table/colspanMinWidth-expected.txt: Added.
* platform/qt-5.0-wk2/fast/table/spanOverlapRepaint-expected.txt: Added.
* platform/qt-5.0-wk2/fast/table/text-field-baseline-expected.png: Added.
* platform/qt-5.0-wk2/fast/table/text-field-baseline-expected.txt: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@123229 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoPort bugs.webkit.org to mod_perl for new hardware.
wsiegrist@apple.com [Fri, 20 Jul 2012 17:35:33 +0000 (17:35 +0000)]
Port bugs.webkit.org to mod_perl for new hardware.

Reviewed by David Kilzer.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@123228 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoUnreviewed, rolling out r123220.
philn@webkit.org [Fri, 20 Jul 2012 17:29:23 +0000 (17:29 +0000)]
Unreviewed, rolling out r123220.
http://trac.webkit.org/changeset/123220
https://bugs.webkit.org/show_bug.cgi?id=91880

broke Debug build (Requested by philn on #webkit).

Patch by Sheriff Bot <webkit.review.bot@gmail.com> on 2012-07-20

* Scripts/webkitdirs.pm:
(buildAutotoolsProject):
* gtk/jhbuild.modules:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@123227 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoUnreviewed GTK gardening, rebaselining a test after r117477.
zandobersek@gmail.com [Fri, 20 Jul 2012 17:25:52 +0000 (17:25 +0000)]
Unreviewed GTK gardening, rebaselining a test after r117477.

* platform/gtk/TestExpectations:
* platform/gtk/tables/mozilla/bugs/bug10296-1-expected.png:
* platform/gtk/tables/mozilla/bugs/bug10296-1-expected.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@123226 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years ago[chromium] Enable GPU-accelerated skia implementation of
senorblanco@chromium.org [Fri, 20 Jul 2012 17:13:58 +0000 (17:13 +0000)]
[chromium] Enable GPU-accelerated skia implementation of
feDiffuseLighting, feSpecularLighting.
https://bugs.webkit.org/show_bug.cgi?id=91865

Reviewed by Dirk Schulze.

Translate the light parameters into the appropriate
SkLightingImageFilter calls.  This path is only enabled when
the filters are created with the Accelerated flag set.

This will be covered by tests in svg/filters/* when they are
added to the virtual GPU tests.

* WebCore.gypi:
* platform/graphics/filters/FELighting.h:
(FELighting):
* platform/graphics/filters/skia/FELightingSkia.cpp: Added.
(WebCore):
(WebCore::FELighting::platformApplySkia):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@123225 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years ago[Gtk] Many tests revealed as passing after moving from Skipped to test_expectations.txt
zandobersek@gmail.com [Fri, 20 Jul 2012 17:09:54 +0000 (17:09 +0000)]
[Gtk] Many tests revealed as passing after moving from Skipped to test_expectations.txt
https://bugs.webkit.org/show_bug.cgi?id=85591

Unreviewed, managing a few more unexpected passes for the GTK port.

Remove a few expectations for tests that were passing, reclassify some failures as
architecture-specific and add PASS expectations for a few tests that pass even though
they test features that are currently not supported on the GTK port.

* platform/gtk/TestExpectations:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@123224 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoWeb Inspector: disable fake workers inspection.
pfeldman@chromium.org [Fri, 20 Jul 2012 16:50:40 +0000 (16:50 +0000)]
Web Inspector: disable fake workers inspection.
https://bugs.webkit.org/show_bug.cgi?id=91867

Reviewed by Vsevolod Vlasov.

'Fake workers' is unreliable error-prone way to debug workers. There are a
lot of frameworks supporting it. Inspector should not do that since it has a more
powerful native worker inspection capabilities. Port owners should enable it
in case they want to provide workers debugging.

* English.lproj/localizedStrings.js:
* WebCore.gypi:
* WebCore.order:
* WebCore.vcproj/WebCore.vcproj:
* inspector/InjectedScriptHost.cpp:
(WebCore::InjectedScriptHost::InjectedScriptHost):
* inspector/InjectedScriptHost.h:
(InjectedScriptHost):
* inspector/InjectedScriptHost.idl:
* inspector/Inspector.json:
* inspector/InspectorAgent.cpp:
(WebCore::InspectorAgent::didCommitLoad):
(WebCore::InspectorAgent::enable):
* inspector/InspectorAgent.h:
(WebCore):
(InspectorAgent):
* inspector/InspectorInstrumentation.cpp:
* inspector/InspectorInstrumentation.h:
(InspectorInstrumentation):
* inspector/front-end/InjectedFakeWorker.js: Removed.
* inspector/front-end/ScriptsPanel.js:
* inspector/front-end/WebKit.qrc:
* inspector/front-end/WorkersSidebarPane.js:
(WebInspector.WorkersSidebarPane):
* inspector/front-end/externs.js:
* inspector/front-end/inspector.html:
* inspector/front-end/inspector.js:
* workers/AbstractWorker.cpp:
(WebCore::AbstractWorker::contextDestroyed):
* workers/AbstractWorker.h:
(AbstractWorker):
* workers/SharedWorker.cpp:
(WebCore::SharedWorker::create):
* workers/Worker.cpp:
(WebCore::Worker::create):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@123223 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoUnreviewed GTK gardening, adding platform-specific baseline that's required
zandobersek@gmail.com [Fri, 20 Jul 2012 15:41:29 +0000 (15:41 +0000)]
Unreviewed GTK gardening, adding platform-specific baseline that's required
after changes in r123145.

* platform/gtk/http/tests/security/cross-frame-access-put-expected.txt: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@123222 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoRebaseline failing test svg/as-image/svg-non-integer-scaled-image.html
schenney@chromium.org [Fri, 20 Jul 2012 15:33:33 +0000 (15:33 +0000)]
Rebaseline failing test svg/as-image/svg-non-integer-scaled-image.html

Unreviewed test expectations update.

* platform/chromium/svg/as-image: Added.
* platform/chromium/svg/as-image/svg-non-integer-scaled-image-expected.txt: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@123221 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years ago[GTK][jhbuild] Switch to GStreamer 0.11 build
philn@webkit.org [Fri, 20 Jul 2012 15:12:54 +0000 (15:12 +0000)]
[GTK][jhbuild] Switch to GStreamer 0.11 build
https://bugs.webkit.org/show_bug.cgi?id=91727

Reviewed by Gustavo Noronha Silva.

Switch build-webkit --gtk to GStreamer 0.11 support and build the
necessary GStreamer git modules from JHBuild.

* Scripts/webkitdirs.pm:
(buildAutotoolsProject): Build WebKit with GStreamer 0.11 support.
* gtk/jhbuild.modules: GStreamer build support.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@123220 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoUnreviewed, rolling out r123182.
commit-queue@webkit.org [Fri, 20 Jul 2012 14:57:55 +0000 (14:57 +0000)]
Unreviewed, rolling out r123182.
http://trac.webkit.org/changeset/123182
https://bugs.webkit.org/show_bug.cgi?id=91863

Causing ASSERT crashes in tests accessibility/canvas.html and
accessibility/canvas-accessibilitynodeobject.html (Requested
by jernoble on #webkit).

Patch by Sheriff Bot <webkit.review.bot@gmail.com> on 2012-07-20

Source/WebCore:

* CMakeLists.txt:
* GNUmakefile.list.am:
* Target.pri:
* WebCore.gypi:
* accessibility/AXObjectCache.cpp:
(WebCore):
* accessibility/AXObjectCache.h:
(AXObjectCache):
* accessibility/AccessibilityARIAGrid.cpp:
(WebCore::AccessibilityARIAGrid::AccessibilityARIAGrid):
(WebCore::AccessibilityARIAGrid::~AccessibilityARIAGrid):
(WebCore):
(WebCore::AccessibilityARIAGrid::create):
* accessibility/AccessibilityARIAGrid.h:
(AccessibilityARIAGrid):
* accessibility/AccessibilityARIAGridCell.cpp:
(WebCore::AccessibilityARIAGridCell::create):
* accessibility/AccessibilityARIAGridRow.cpp:
(WebCore::AccessibilityARIAGridRow::create):
* accessibility/AccessibilityAllInOne.cpp:
* accessibility/AccessibilityList.cpp:
(WebCore::AccessibilityList::create):
* accessibility/AccessibilityListBox.cpp:
(WebCore::AccessibilityListBox::create):
* accessibility/AccessibilityMediaControls.cpp:
(WebCore::AccessibilityMediaControl::create):
(WebCore::AccessibilityMediaControlsContainer::create):
(WebCore::AccessibilityMediaTimeline::create):
(WebCore::AccessibilityMediaTimeDisplay::create):
* accessibility/AccessibilityMenuList.cpp:
* accessibility/AccessibilityMenuList.h:
(WebCore::AccessibilityMenuList::create):
* accessibility/AccessibilityNodeObject.cpp: Removed.
* accessibility/AccessibilityNodeObject.h: Removed.
* accessibility/AccessibilityObject.h:
(AccessibilityObject):
* accessibility/AccessibilityProgressIndicator.cpp:
(WebCore::AccessibilityProgressIndicator::create):
* accessibility/AccessibilityRenderObject.cpp:
(WebCore::AccessibilityRenderObject::AccessibilityRenderObject):
(WebCore::AccessibilityRenderObject::create):
(WebCore::AccessibilityRenderObject::detach):
(WebCore::AccessibilityRenderObject::remapAriaRoleDueToParent):
(WebCore):
(WebCore::AccessibilityRenderObject::determineAriaRoleAttribute):
(WebCore::AccessibilityRenderObject::ariaRoleAttribute):
(WebCore::AccessibilityRenderObject::updateAccessibilityRole):
(WebCore::AccessibilityRenderObject::canSetFocusAttribute):
(WebCore::AccessibilityRenderObject::childrenChanged):
(WebCore::AccessibilityRenderObject::canHaveChildren):
(WebCore::AccessibilityRenderObject::addChildren):
* accessibility/AccessibilityRenderObject.h:
(AccessibilityRenderObject):
(WebCore::AccessibilityRenderObject::setRenderer):
* accessibility/AccessibilitySlider.cpp:
(WebCore::AccessibilitySlider::create):
* accessibility/AccessibilityTable.cpp:
(WebCore::AccessibilityTable::AccessibilityTable):
(WebCore::AccessibilityTable::~AccessibilityTable):
(WebCore):
(WebCore::AccessibilityTable::create):
* accessibility/AccessibilityTable.h:
(AccessibilityTable):
* accessibility/AccessibilityTableCell.cpp:
(WebCore::AccessibilityTableCell::create):
* accessibility/AccessibilityTableRow.cpp:
(WebCore::AccessibilityTableRow::create):

LayoutTests:

* accessibility/canvas-accessibilitynodeobject-expected.txt: Removed.
* accessibility/canvas-accessibilitynodeobject.html: Removed.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@123219 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years ago[EFL] Proxy configuration should honor the no_proxy environment variable
commit-queue@webkit.org [Fri, 20 Jul 2012 14:39:35 +0000 (14:39 +0000)]
[EFL] Proxy configuration should honor the no_proxy environment variable
https://bugs.webkit.org/show_bug.cgi?id=91747

Patch by Christophe Dumez <christophe.dumez@intel.com> on 2012-07-20
Reviewed by Kenneth Rohde Christiansen.

Source/WebCore:

Implement a custom Proxy URI Resolver for libsoup
so that we can use it in EFL port. This custom
proxy resolver brings support for setting
exceptions so that the proxy is not used for the
specified hosts.

By default, the proxy is not used for localhost and
127.0.0.1.

No new tests, no behavior change for layout tests.

* PlatformEfl.cmake:
* platform/network/soup/ProxyResolverSoup.cpp: Added.
(soup_proxy_resolver_wk_init):
(soupProxyResolverWkFinalize):
(soupProxyResolverWkSetProperty):
(soupProxyResolverWkGetProperty):
(shouldBypassProxy):
(idle_return_proxy_uri):
(soupProxyResolverWkGetProxyURIAsync):
(soupProxyResolverWkGetProxyURISync):
(soup_proxy_resolver_wk_class_init):
(soup_proxy_resolver_wk_interface_init):
(soupProxyResolverWkNew):
* platform/network/soup/ProxyResolverSoup.h: Added.

Source/WebKit:

Add WebCore/platform/network/soup to INCLUDE paths.

* PlatformEfl.cmake:

Source/WebKit/efl:

Use the new custom proxy resolver from WebCore in
WebKit1-EFL's ewk_network_proxy_uri_set() / get(),
instead of the less flexible SOUP_SESSION_PROXY_URI
SoupSession property.

* ewk/ewk_network.cpp:
(ewk_network_proxy_uri_set):
(ewk_network_proxy_uri_get):

Source/WebKit2:

Use the new custom proxy resolver from WebCore in
WebKit2-EFL so that it is possible for the client
to set proxy exceptions via the "no_proxy"
environment variable.

By default, the proxy set in the "http_proxy"
environment variable will not be used for requests
to localhost or 127.0.0.1.

* WebProcess/efl/WebProcessMainEfl.cpp:
(WebKit::WebProcessMainEfl):

Tools:

Add ProxyResolverSoup to the list of classes that contain GObjects
in style checker script so that it does not complain about naming
convention issues that are mandated by GObject.

* Scripts/webkitpy/style/checker.py:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@123218 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoFix chromium TestExpectations file.
schenney@chromium.org [Fri, 20 Jul 2012 14:38:44 +0000 (14:38 +0000)]
Fix chromium TestExpectations file.

Unreviewed build fix.

* platform/chromium/TestExpectations:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@123217 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years ago[Qt][WK2] Add support for multi-select list
pierre.rossi@gmail.com [Fri, 20 Jul 2012 14:38:01 +0000 (14:38 +0000)]
[Qt][WK2] Add support for multi-select list
https://bugs.webkit.org/show_bug.cgi?id=85527

Patch by Dinu Jacob <dinu.jacob@nokia.com> on 2012-06-01
Reviewed by Kenneth Rohde Christiansen.

Added support for multi-select list:
- Added multi-selection flag to PlatformPopupMenuData to indicate whether to accept multiple selections or not.
- Added selected state to WebPopupItem.
- Modified WebPopupMenuQt to support multiple selections.

* Shared/PlatformPopupMenuData.cpp:
(WebKit::PlatformPopupMenuData::PlatformPopupMenuData):
(WebKit::PlatformPopupMenuData::encode):
(WebKit::PlatformPopupMenuData::decode):
* Shared/PlatformPopupMenuData.h:
(PlatformPopupMenuData):
* Shared/WebPopupItem.cpp:
(WebKit::WebPopupItem::WebPopupItem):
(WebKit::WebPopupItem::encode):
(WebKit::WebPopupItem::decode):
* Shared/WebPopupItem.h:
* UIProcess/API/qt/tests/qmltests/WebView/tst_multiSelect.qml: Added.
* UIProcess/API/qt/tests/qmltests/common/multiselect.html: Added.
* UIProcess/WebPageProxy.h:
(WebPageProxy):
* UIProcess/WebPopupMenuProxy.h:
(Client):
* UIProcess/qt/WebPageProxyQt.cpp:
(WebKit::WebPageProxy::changeSelectedIndex):
(WebKit):
(WebKit::WebPageProxy::closePopupMenu):
* UIProcess/qt/WebPopupMenuProxyQt.cpp:
(WebKit::PopupMenuItemModel::multiple):
(PopupMenuItemModel):
(WebKit::PopupMenuItemModel::Item::Item):
(ItemSelectorContextObject):
(WebKit::ItemSelectorContextObject::allowMultiSelect):
(WebKit::ItemSelectorContextObject::reject):
(WebKit::ItemSelectorContextObject::dismiss):
(WebKit::ItemSelectorContextObject::ItemSelectorContextObject):
(WebKit):
(WebKit::ItemSelectorContextObject::onIndexUpdate):
(WebKit::ItemSelectorContextObject::accept):
(WebKit::PopupMenuItemModel::PopupMenuItemModel):
(WebKit::PopupMenuItemModel::select):
(WebKit::PopupMenuItemModel::toggleItem):
(WebKit::PopupMenuItemModel::buildItems):
(WebKit::WebPopupMenuProxyQt::showPopupMenu):
(WebKit::WebPopupMenuProxyQt::hidePopupMenu):
(WebKit::WebPopupMenuProxyQt::selectIndex):
(WebKit::WebPopupMenuProxyQt::createItem):
* UIProcess/qt/WebPopupMenuProxyQt.h:
(WebPopupMenuProxyQt):
* WebProcess/WebCoreSupport/WebPopupMenu.cpp:
(WebKit::WebPopupMenu::didChangeSelectedIndex):
(WebKit::WebPopupMenu::populateItems):
* WebProcess/WebCoreSupport/WebPopupMenu.h:
(WebPopupMenu):
* WebProcess/WebCoreSupport/qt/WebPopupMenuQt.cpp:
(WebKit::WebPopupMenu::setUpPlatformData):
* WebProcess/WebPage/WebPage.cpp:
(WebKit::WebPage::didChangeSelectedIndexForActivePopupMenu):
(WebKit):
(WebKit::WebPage::changeSelectedIndex):
* WebProcess/WebPage/WebPage.h:
(WebPage):
* WebProcess/WebPage/WebPage.messages.in:
* WebProcess/WebPage/qt/WebPageQt.cpp:
(WebKit::WebPage::selectedIndex):
(WebKit):
(WebKit::WebPage::hidePopupMenu):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@123216 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoRebaseline Chromium results after r121171
schenney@chromium.org [Fri, 20 Jul 2012 14:30:51 +0000 (14:30 +0000)]
Rebaseline Chromium results after r121171
https://bugs.webkit.org/show_bug.cgi?id=91474

Reviewed by Dan Bernstein.

Unreviewed test expectations update.

Values have changed by one pixel, which is expected given the description in the bug report.

* platform/chromium-linux/svg/zoom/page/zoom-svg-through-object-with-absolute-size-2-expected.png:
* platform/chromium-linux/svg/zoom/page/zoom-svg-through-object-with-absolute-size-2-expected.txt:
* platform/chromium-linux/svg/zoom/page/zoom-svg-through-object-with-auto-size-expected.png:
* platform/chromium-linux/svg/zoom/page/zoom-svg-through-object-with-auto-size-expected.txt: Added.
* platform/chromium-mac-snowleopard/svg/zoom/page/zoom-svg-through-object-with-absolute-size-2-expected.png:
* platform/chromium-mac/svg/as-image/svg-non-integer-scaled-image-expected.png:
* platform/chromium-mac/svg/zoom/page/zoom-svg-through-object-with-absolute-size-2-expected.png:
* platform/chromium-mac/svg/zoom/page/zoom-svg-through-object-with-absolute-size-2-expected.txt:
* platform/chromium-mac/svg/zoom/page/zoom-svg-through-object-with-auto-size-expected.png:
* platform/chromium-mac/svg/zoom/page/zoom-svg-through-object-with-auto-size-expected.txt:
* platform/chromium-win/svg/as-image/svg-non-integer-scaled-image-expected.png:
* platform/chromium-win/svg/zoom/page/zoom-svg-through-object-with-absolute-size-2-expected.png:
* platform/chromium-win/svg/zoom/page/zoom-svg-through-object-with-absolute-size-2-expected.txt:
* platform/chromium-win/svg/zoom/page/zoom-svg-through-object-with-auto-size-expected.png:
* platform/chromium-win/svg/zoom/page/zoom-svg-through-object-with-auto-size-expected.txt:
* platform/chromium/svg/as-image: Removed.
* platform/chromium/svg/as-image/svg-non-integer-scaled-image-expected.txt: Removed.
* platform/efl/svg/zoom/page/zoom-svg-through-object-with-auto-size-expected.txt: Removed.
* platform/gtk/svg/zoom/page/zoom-svg-through-object-with-auto-size-expected.txt: Removed.
* platform/mac/svg/zoom/page/zoom-svg-through-object-with-auto-size-expected.txt: Removed.
* svg/zoom/page/zoom-svg-through-object-with-auto-size-expected.txt: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@123215 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoColorChooserClient should expose the element's location
pierre.rossi@gmail.com [Fri, 20 Jul 2012 14:23:37 +0000 (14:23 +0000)]
ColorChooserClient should expose the element's location
https://bugs.webkit.org/show_bug.cgi?id=91767

Reviewed by Kent Tamura.

The rationale is that the chrome client could have an interest in placing the picker next to
the element for usability reasons.

No new tests as there is not much logic here: Just exposing a geometry information to WebKit.

* html/ColorInputType.cpp:
(WebCore::ColorInputType::elementRectRelativeToWindow):
(WebCore):
* html/ColorInputType.h:
(ColorInputType):
* platform/ColorChooserClient.h:
(ColorChooserClient):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@123214 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years ago[EFL] Fix build when protocol handler and custom scheme is disabled
commit-queue@webkit.org [Fri, 20 Jul 2012 14:11:17 +0000 (14:11 +0000)]
[EFL] Fix build when protocol handler and custom scheme is disabled
https://bugs.webkit.org/show_bug.cgi?id=91840

Patch by Thiago Marcos P. Santos <thiago.santos@intel.com> on 2012-07-20
Reviewed by Kenneth Rohde Christiansen.

Fixed namespace closing bracket positioning.

* WebCoreSupport/RegisterProtocolHandlerClientEfl.cpp:
(WebCore):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@123213 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years ago[WK2] WebIntents doesn't build if mutation observers is disabled
commit-queue@webkit.org [Fri, 20 Jul 2012 14:05:08 +0000 (14:05 +0000)]
[WK2] WebIntents doesn't build if mutation observers is disabled
https://bugs.webkit.org/show_bug.cgi?id=91839

Patch by Thiago Marcos P. Santos <thiago.santos@intel.com> on 2012-07-20
Reviewed by Kenneth Rohde Christiansen.

WebIntents was depending on a code protected by ENABLE(MUTATION_OBSERVERS).
Now the code is also enabled by ENABLE(WEB_INTENTS).

* bindings/js/JSDictionary.cpp:
(WebCore):
* bindings/js/JSDictionary.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@123212 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years ago[CSSRegions]Assert failure when layout positioned objects in regions
mihnea@adobe.com [Fri, 20 Jul 2012 13:59:48 +0000 (13:59 +0000)]
[CSSRegions]Assert failure when layout positioned objects in regions
https://bugs.webkit.org/show_bug.cgi?id=90792

Reviewed by Andreas Kling.

Before clamping the containing block (for an out-of-flow positioned element
inside a named flow) to the region, we have to test whether the region is not
null, which can happen when the named flow does not have attached regions.

Source/WebCore:

Tests: fast/regions/positioned-vrl-in-named-flow.html
       fast/regions/positioned-vrl-in-parent-named-flow.html
       fast/regions/positioned-with-vrl-parent-in-named-flow.html

* rendering/RenderBox.cpp:
(WebCore::RenderBox::containingBlockLogicalWidthForPositioned):
(WebCore::RenderBox::computePositionedLogicalWidth):
(WebCore::RenderBox::computePositionedLogicalHeight):

LayoutTests:

* fast/regions/positioned-vrl-in-named-flow-expected.txt: Added.
* fast/regions/positioned-vrl-in-named-flow.html: Added.
* fast/regions/positioned-vrl-in-parent-named-flow-expected.txt: Added.
* fast/regions/positioned-vrl-in-parent-named-flow.html: Added.
* fast/regions/positioned-with-vrl-parent-in-named-flow-expected.txt: Added.
* fast/regions/positioned-with-vrl-parent-in-named-flow.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@123211 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoSVG Filter Effect sub-region not applied for some filters
schenney@chromium.org [Fri, 20 Jul 2012 13:55:02 +0000 (13:55 +0000)]
SVG Filter Effect sub-region not applied for some filters
https://bugs.webkit.org/show_bug.cgi?id=89767

Reviewed by Dirk Schulze.

Source/WebCore:

The filter effect region for SVG feGaussian and feDropShadow filters
was incorrectly set, failing to account for the radius of the blur.
This patch fixes the problem and removes an unneccessary clip
operation when the filter result is blitted into the target. That clip
was clipping the shadow from webkit-shadow.

Tests: svg/filters/feDropShadow-subregion.svg
       svg/filters/feGaussianBlur-subregion.svg

* platform/graphics/filters/FEDropShadow.cpp:
(WebCore::FEDropShadow::determineAbsolutePaintRect): Moved the code
that clips the paint rect by the filter effect subregion to come after
the paint is modified by the filter radius.
* platform/graphics/filters/FEGaussianBlur.cpp:
(WebCore::FEGaussianBlur::determineAbsolutePaintRect):
Moved the code that clips the paint rect by the filter effect subregion to
come after the paint is modified by the filter radius.
* rendering/svg/RenderSVGResourceFilter.cpp:
(WebCore::RenderSVGResourceFilter::postApplyResource):
Removed a clip that is unnecessary and that was incorrectly clipping drop shadows.

LayoutTests:

The filter effect region for SVG feGaussian and feDropShadow filters was incorrectly set,
failing to account for the radius of the blur. This patch fixes the problem and removes an
unneccessary clip operation when the filter result is blitted into the target. That clip
was clipping the shadow from webkit-shadow.

* platform/chromium/TestExpectations: Added temp entries for affected
tests. Will rebaseline later (not closing the bug in the meantime).
* svg/filters/feDropShadow-subregion-expected.png: Added.
* svg/filters/feDropShadow-subregion-expected.txt: Added.
* svg/filters/feDropShadow-subregion.svg: This verifies the subregion for feDropShadow.
* svg/filters/feGaussianBlur-subregion-expected.png: Added.
* svg/filters/feGaussianBlur-subregion-expected.txt: Added.
* svg/filters/feGaussianBlur-subregion.svg: This verifies the subregion for feGaussianBlur.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@123210 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoUnreviewed. Skip regions inspector test on EFL.
mihnea@adobe.com [Fri, 20 Jul 2012 13:41:54 +0000 (13:41 +0000)]
Unreviewed. Skip regions inspector test on EFL.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@123209 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years ago[Qt][WK2] Add configure tests for Xrender and GLX.
zeno.albisser@nokia.com [Fri, 20 Jul 2012 13:26:40 +0000 (13:26 +0000)]
[Qt][WK2] Add configure tests for Xrender and GLX.
https://bugs.webkit.org/show_bug.cgi?id=91845

GraphicsSurface on linux requires Xcomposite, Xrender
and GLX. If any of these requirements is not met,
we should disable GraphicsSurface.

Reviewed by Kenneth Rohde Christiansen.

* qmake/config.tests/glx/glx.cpp: Added.
(main):
* qmake/config.tests/glx/glx.pro: Added.
* qmake/config.tests/libXrender/libXrender.cpp: Added.
(main):
* qmake/config.tests/libXrender/libXrender.pro: Added.
* qmake/configure.pri:
* qmake/mkspecs/features/features.prf:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@123208 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoWeb Inspector: Status bar icons are misaligned by 1px to the right
pfeldman@chromium.org [Fri, 20 Jul 2012 13:07:48 +0000 (13:07 +0000)]
Web Inspector: Status bar icons are misaligned by 1px to the right
https://bugs.webkit.org/show_bug.cgi?id=91757

Reviewed by Vsevolod Vlasov.

* inspector/front-end/inspector.css:
(.glyph):
(.glyph.shadow):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@123207 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years ago[Qt] http/tests/security/cross-frame-access-put.html test fail after r123145
kkristof@inf.u-szeged.hu [Fri, 20 Jul 2012 12:50:42 +0000 (12:50 +0000)]
[Qt] http/tests/security/cross-frame-access-put.html test fail after r123145
https://bugs.webkit.org/show_bug.cgi?id=91830

Unreviewed gardening.

Patch by Balazs Ankes <bank@inf.u-szeged.hu> on 2012-07-20

* platform/qt/http/tests/security/cross-frame-access-put-expected.txt: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@123206 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoWeb Inspector: Protocol Extension: add getNamedFlowCollection command
commit-queue@webkit.org [Fri, 20 Jul 2012 12:34:58 +0000 (12:34 +0000)]
Web Inspector: Protocol Extension: add getNamedFlowCollection command
https://bugs.webkit.org/show_bug.cgi?id=91607

Patch by Andrei Poenaru <poenaru@adobe.com> on 2012-07-20
Reviewed by Pavel Feldman.

Source/WebCore:

Extended the protocol with "getNamedFlowCollection" command.
This command returns the CSS Named Flows from the document.

Test: inspector/styles/protocol-getNamedFlowCollection-command.html

* dom/WebKitNamedFlowCollection.cpp:
(WebCore::WebKitNamedFlowCollection::namedFlowsNames):
(WebCore):
* dom/WebKitNamedFlowCollection.h:
(WebKitNamedFlowCollection):
* inspector/Inspector.json:
* inspector/InspectorCSSAgent.cpp:
(WebCore::InspectorCSSAgent::getNamedFlowCollection):
(WebCore):
* inspector/InspectorCSSAgent.h:
(InspectorCSSAgent):
* inspector/InspectorDOMAgent.cpp:
(WebCore::InspectorDOMAgent::assertDocument):
(WebCore):
* inspector/InspectorDOMAgent.h:
(InspectorDOMAgent):
* inspector/front-end/CSSStyleModel.js:
(WebInspector.CSSStyleModel.prototype.getNamedFlowCollectionAsync):

LayoutTests:

The test builds a webpage with 3 Named Flows in the main frame and 1 Named Flow in an iframe.
Because the getNamedFlowCollection command is executed with the main document node id as an argument,
it should return only the first 3 Named Flows.

* inspector/styles/protocol-getNamedFlowCollection-command-expected.txt: Added.
* inspector/styles/protocol-getNamedFlowCollection-command.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@123205 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years ago[EFL] Gardening of failing tests
commit-queue@webkit.org [Fri, 20 Jul 2012 12:29:40 +0000 (12:29 +0000)]
[EFL] Gardening of failing tests
https://bugs.webkit.org/show_bug.cgi?id=91838

Unreviewed EFL gardening. Add failing tests to the Test Expectations.

Patch by Sudarsana Nagineni <sudarsana.nagineni@linux.intel.com> on 2012-07-20

* platform/efl/TestExpectations:
* platform/efl/fast/forms/state-restore-skip-stateless-expected.txt: Removed
the platform specific expected after r123066.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@123204 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years ago[EFL][WK2][WTR] Implement EventSenderProxy
commit-queue@webkit.org [Fri, 20 Jul 2012 11:31:06 +0000 (11:31 +0000)]
[EFL][WK2][WTR] Implement EventSenderProxy
https://bugs.webkit.org/show_bug.cgi?id=91731

Patch by Kangil Han <kangil.han@samsung.com> on 2012-07-20
Reviewed by Kentaro Hara.

This is an initial EFL port implementation of EventSenderProxy.

* WebKitTestRunner/EventSenderProxy.h:
(WTR):
(EventSenderProxy):
* WebKitTestRunner/InjectedBundle/EventSendingController.h:
* WebKitTestRunner/PlatformEfl.cmake:
* WebKitTestRunner/TestController.cpp:
(WTR::TestController::TestController):
(WTR::TestController::didReceiveSynchronousMessageFromInjectedBundle):
* WebKitTestRunner/efl/EventSenderProxyEfl.cpp: Added.
(WTR):
(WTREvent):
(WTR::WTREvent::WTREvent):
(WTR::KeyEventInfo::KeyEventInfo):
(KeyEventInfo):
(WTR::evasMouseButton):
(WTR::setEvasModifiers):
(WTR::dispatchMouseDownEvent):
(WTR::dispatchMouseUpEvent):
(WTR::dispatchMouseMoveEvent):
(WTR::dispatchMouseScrollByEvent):
(WTR::keyPadName):
(WTR::keyName):
(WTR::EventSenderProxy::EventSenderProxy):
(WTR::EventSenderProxy::updateClickCountForButton):
(WTR::EventSenderProxy::dispatchEvent):
(WTR::EventSenderProxy::replaySavedEvents):
(WTR::EventSenderProxy::sendOrQueueEvent):
(WTR::EventSenderProxy::mouseDown):
(WTR::EventSenderProxy::mouseUp):
(WTR::EventSenderProxy::mouseMoveTo):
(WTR::EventSenderProxy::mouseScrollBy):
(WTR::EventSenderProxy::leapForward):
(WTR::EventSenderProxy::keyDown):
(WTR::EventSenderProxy::addTouchPoint):
(WTR::EventSenderProxy::updateTouchPoint):
(WTR::EventSenderProxy::setTouchModifier):
(WTR::EventSenderProxy::touchStart):
(WTR::EventSenderProxy::touchMove):
(WTR::EventSenderProxy::touchEnd):
(WTR::EventSenderProxy::touchCancel):
(WTR::EventSenderProxy::clearTouchPoints):
(WTR::EventSenderProxy::releaseTouchPoint):
(WTR::EventSenderProxy::cancelTouchPoint):
(WTR::EventSenderProxy::setTouchPointRadius):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@123203 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years ago[EFL] Check parameters of ewk APIs in ewk_security_origin
commit-queue@webkit.org [Fri, 20 Jul 2012 11:26:48 +0000 (11:26 +0000)]
[EFL] Check parameters of ewk APIs in ewk_security_origin
https://bugs.webkit.org/show_bug.cgi?id=91833

Patch by Kihong Kwon <kihong.kwon@samsung.com> on 2012-07-20
Reviewed by Kentaro Hara.

For preventing crash, check parameters of ewk APIs in the ewk_security_origin.cpp.

* ewk/ewk_security_origin.cpp:
(ewk_security_origin_port_get):
(ewk_security_origin_web_database_usage_get):
(ewk_security_origin_web_database_quota_get):
(ewk_security_origin_web_database_quota_set):
(ewk_security_origin_application_cache_quota_set):
(ewk_security_origin_application_cache_clear):
(ewk_security_origin_web_database_get_all):
(ewk_security_origin_free):
* ewk/ewk_security_origin.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@123202 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years ago[Qt] Unreviewed gardening after r123192. Skipped a new test.
kkristof@inf.u-szeged.hu [Fri, 20 Jul 2012 11:25:08 +0000 (11:25 +0000)]
[Qt] Unreviewed gardening after r123192. Skipped a new test.

Patch by János Badics <jbadics@inf.u-szeged.hu> on 2012-07-20

* platform/qt/Skipped:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@123201 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoUnreviewed, rolling out r123085.
gyuyoung.kim@samsung.com [Fri, 20 Jul 2012 10:28:43 +0000 (10:28 +0000)]
Unreviewed, rolling out r123085.
http://trac.webkit.org/changeset/123085
https://bugs.webkit.org/show_bug.cgi?id=91719

r123085 causes crashes on EFL layout test bot.

Source/WebKit2:

* UIProcess/efl/WebContextEfl.cpp:
(WebKit::WebContext::applicationCacheDirectory):

Tools:

* WebKitTestRunner/CMakeLists.txt:
* WebKitTestRunner/InjectedBundle/efl/LayoutTestControllerEfl.cpp:
(WTR::LayoutTestController::platformInitialize):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@123200 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years ago[EFL][WK2] Use "load,finished" signal in EWK2UnitTestBase::loadUrlSync() instead...
commit-queue@webkit.org [Fri, 20 Jul 2012 10:17:05 +0000 (10:17 +0000)]
[EFL][WK2] Use "load,finished" signal in EWK2UnitTestBase::loadUrlSync() instead of "load,progress"
https://bugs.webkit.org/show_bug.cgi?id=91721

Patch by Christophe Dumez <christophe.dumez@intel.com> on 2012-07-20
Reviewed by Kenneth Rohde Christiansen.

The purpose of EWK2UnitTestBase::loadUrlSync() is to load
a URL in the view and wait synchronously for the load to finish.
The current implementation uses the "load,progress" signal to
detect when the load is finished, which is inefficient because
it gets emitted several times.

It is better to wait for the "load,finished" signal which gets
emitted only once when the load is complete.

* UIProcess/API/efl/tests/UnitTestUtils/EWK2UnitTestBase.cpp:
(EWK2UnitTest::onLoadFinished):
(EWK2UnitTest::EWK2UnitTestBase::EWK2UnitTestBase):
(EWK2UnitTest::EWK2UnitTestBase::loadUrlSync):
* UIProcess/API/efl/tests/UnitTestUtils/EWK2UnitTestBase.h:
(EWK2UnitTestBase):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@123199 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoWeb Inspector: Timeline: forward compatibility for load.
commit-queue@webkit.org [Fri, 20 Jul 2012 09:41:35 +0000 (09:41 +0000)]
Web Inspector: Timeline: forward compatibility for load.
https://bugs.webkit.org/show_bug.cgi?id=91714

Patch by Eugene Klyuchnikov <eustas.big@gmail.com> on 2012-07-20
Reviewed by Pavel Feldman.

Accept records of unrecognized types and render them as "unknown".

Source/WebCore:

Test: inspector/timeline/timeline-load-incompatible.html

* English.lproj/localizedStrings.js: Added UI string.
* inspector/front-end/TimelinePresentationModel.js:
(WebInspector.TimelinePresentationModel.recordStyle): Add missing record
styles at runtime.
(WebInspector.TimelinePresentationModel.Record.prototype._getRecordDetails):
Add some common information for default case.

LayoutTests:

* inspector/timeline/timeline-load-incompatible-expected.txt: Added.
* inspector/timeline/timeline-load-incompatible.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@123198 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoAnother unreviewed Mac build fix against r123184.
morrita@google.com [Fri, 20 Jul 2012 09:30:07 +0000 (09:30 +0000)]
Another unreviewed Mac build fix against r123184.

* WebCore.exp.in:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@123197 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years ago[CSSRegions]Crash when flowing a region into itself
mihnea@adobe.com [Fri, 20 Jul 2012 09:23:56 +0000 (09:23 +0000)]
[CSSRegions]Crash when flowing a region into itself
https://bugs.webkit.org/show_bug.cgi?id=90289

Reviewed by Andreas Kling.

If a region is designed to receive itself as content through a named flow, we avoid
a circular dependency by not attaching the region to the desired named flow.
In such occasions, the region's attached flow thread object is null and
we have to check this pointer before using it in RenderRegion::styleDidChange.

Source/WebCore:

Test: fast/regions/flow-region-into-itself-crash.html

* rendering/RenderRegion.cpp:
(WebCore::RenderRegion::styleDidChange):

LayoutTests:

* fast/regions/flow-region-into-itself-crash-expected.txt: Added.
* fast/regions/flow-region-into-itself-crash.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@123196 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoUnreviewed Mac build fix against r123184.
morrita@google.com [Fri, 20 Jul 2012 09:13:10 +0000 (09:13 +0000)]
Unreviewed Mac build fix against r123184.

* WebCore.exp.in:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@123195 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoIf value for responseType defined as type that not supported, it should not throw...
kinuko@chromium.org [Fri, 20 Jul 2012 08:41:16 +0000 (08:41 +0000)]
If value for responseType defined as type that not supported, it should not throw an exception in XHR 2
https://bugs.webkit.org/show_bug.cgi?id=90976

Reviewed by Alexey Proskuryakov.

Source/WebCore:

http://www.w3.org/TR/XMLHttpRequest2/#the-responsetype-attribute
The spec does not say it should throw an exception when a non-supported
type is set, and other browsers do not throw it either.

Test: fast/xmlhttprequest/xmlhttprequest-responsetype-set-type.html

* xml/XMLHttpRequest.cpp:
(WebCore::XMLHttpRequest::setResponseType): Stop throwing an exception when non-supported type is given.

LayoutTests:

Adding tests to see if setting invalid responseType does not throw an exception and does not change the value.

* fast/xmlhttprequest/xmlhttprequest-responsetype-set-invalidtype-expected.txt: Added.
* fast/xmlhttprequest/xmlhttprequest-responsetype-set-invalidtype.html: Added.
* fast/xmlhttprequest/xmlhttprequest-responsetype-arraybuffer-expected.txt:
* fast/xmlhttprequest/xmlhttprequest-responsetype-arraybuffer.html: Removed a line which tries to set invalid type (since we now have a separate test)

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@123194 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoShow or hide <dialog> depending on the open attribute
commit-queue@webkit.org [Fri, 20 Jul 2012 08:28:37 +0000 (08:28 +0000)]
Show or hide <dialog> depending on the open attribute
https://bugs.webkit.org/show_bug.cgi?id=90931

Patch by Matt Falkenhagen <falken@chromium.org> on 2012-07-20
Reviewed by Kent Tamura.

Source/WebCore:

Test: fast/dom/HTMLDialogElement/dialog-open.html

* css/html.css: Add CSS for dialog. This is copied verbatim from the HTML5 spec: http://www.whatwg.org/specs/web-apps/current-work/#flow-content-1
(dialog:not([open])):
(dialog):
* html/HTMLDialogElement.cpp:
(WebCore::HTMLDialogElement::close): Set open to false, to hide the dialog.
(WebCore::HTMLDialogElement::show): Set open to true, to show the dialog.
(WebCore):
(WebCore::HTMLDialogElement::isPresentationAttribute): Make openAttr a presentation attribute, to work around bug 91058
* html/HTMLDialogElement.h:
(HTMLDialogElement):
* html/HTMLDialogElement.idl:

LayoutTests:

* fast/dom/HTMLDialogElement/dialog-open-expected.txt: Added.
* fast/dom/HTMLDialogElement/dialog-open.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@123193 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoNeeds pseudo id to point the inner element of an img elemnet for styling
shinyak@chromium.org [Fri, 20 Jul 2012 08:06:36 +0000 (08:06 +0000)]
Needs pseudo id to point the inner element of an img elemnet for styling
https://bugs.webkit.org/show_bug.cgi?id=91590

Reviewed by Hajime Morita.

Source/WebCore:

We introduce pseudo id to point the element in UserAgentShadowDOM.
This enables us to style the inner element of an img element shadow.
Please see also http://trac.webkit.org/wiki/ShadowDOMForReplacedElements

Test: fast/dom/shadow/shadowdom-for-image-with-pseudo-id.html

* html/shadow/ImageInnerElement.cpp:
(WebCore::ImageInnerElement::shadowPseudoId):
(WebCore):
* html/shadow/ImageInnerElement.h:
(ImageInnerElement):

LayoutTests:

Uses pseudo id to confirm the style is applied to the inner element of an img element shadow.

* fast/dom/shadow/shadowdom-for-image-with-pseudo-id-expected.html: Added.
* fast/dom/shadow/shadowdom-for-image-with-pseudo-id.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@123192 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoREGRESSION(r121420): Performance regression of form state saving for pages with multi...
tkent@chromium.org [Fri, 20 Jul 2012 08:03:59 +0000 (08:03 +0000)]
REGRESSION(r121420): Performance regression of form state saving for pages with multiple forms
https://bugs.webkit.org/show_bug.cgi?id=91804

Reviewed by Hajime Morita.

The complexity of FormKeyGenerator::formKey() was O(N) where N is the
number of form elements with an identical action URL, and formKey() is
called for every form. So, it's O(N^2). A page in www.reddit.com
contains hundreds of form elements with action="#". So FormController::
formElementsState() took a few seconds on a slow machine.

In order to avoid O(N^2) operation, storing a map from form signatures
to next index numbers, instead of storing existing formKey strings.

No new tests. Just a performance improvement.

* html/FormController.cpp:
(FormKeyGenerator): Remove m_existingKeys. Add a map from a form
signature string to the next index number.
(WebCore::formSignature): Returns a signature string for a form, without
an index number. This is like "actionURL [name1 name2 ]"
(WebCore::FormKeyGenerator::formKey):
Creates a formKey string by concatenating a formSignature and #n. N is
obtained from m_formSignatureToNextIndexMap in O(1).
(WebCore::FormKeyGenerator::willDeleteForm):
Remove the code for m_existingKeys.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@123191 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoUnreviewed. Set svn:mime-type to image/png for PNG files without svn:mime-type.
yosin@chromium.org [Fri, 20 Jul 2012 07:56:46 +0000 (07:56 +0000)]
Unreviewed. Set svn:mime-type to image/png for PNG files without svn:mime-type.

* platform/qt-5.0-wk2/fast/forms/number/input-appearance-number-rtl-expected.png: Added property svn:mime-type.
* platform/qt-5.0-wk2/fast/forms/number/input-appearance-spinbutton-disabled-readonly-expected.png: Added property svn:mime-type.
* platform/qt-5.0-wk2/fast/forms/number/input-appearance-spinbutton-layer-expected.png: Added property svn:mime-type.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@123190 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years ago[WTR] Add *explicit* keyword to constructor which has a parameter
gyuyoung.kim@samsung.com [Fri, 20 Jul 2012 07:48:19 +0000 (07:48 +0000)]
[WTR] Add *explicit* keyword to constructor which has a parameter
https://bugs.webkit.org/show_bug.cgi?id=91745

Reviewed by Adam Barth.

Add *explicit* keyword to contructor which has a parameter in order to avoid implicit type conversion.

* WebKitTestRunner/EventSenderProxy.h:
(EventSenderProxy):
* WebKitTestRunner/TestInvocation.h:
(TestInvocation):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@123189 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoFix crash in WebCore::HTMLInputElement::dataList
keishi@webkit.org [Fri, 20 Jul 2012 07:46:00 +0000 (07:46 +0000)]
Fix crash in WebCore::HTMLInputElement::dataList
https://bugs.webkit.org/show_bug.cgi?id=91818

Reviewed by Kent Tamura.

Source/WebCore:

Fix crash that occurs when setting slider appearance on a regular node.

Test: fast/forms/range/slider-appearance-crash.html

* rendering/RenderTheme.cpp:
(WebCore::RenderTheme::paintSliderTicks): Check if node is an input element.

LayoutTests:

* fast/forms/range/slider-appearance-crash-expected.txt: Added.
* fast/forms/range/slider-appearance-crash.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@123188 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years ago[Chromium] Fix an assertion failure in TextFieldDecorationElement::hostInput()
tkent@chromium.org [Fri, 20 Jul 2012 07:42:57 +0000 (07:42 +0000)]
[Chromium] Fix an assertion failure in TextFieldDecorationElement::hostInput()
https://bugs.webkit.org/show_bug.cgi?id=91824

Reviewed by Hajime Morita.

No new tests. This code is used only in Chromium browser.

* html/shadow/TextFieldDecorationElement.cpp:
(WebCore::TextFieldDecorationElement::hostInput):
Use shadowHost(), and allow to return 0.
(WebCore::TextFieldDecorationElement::defaultEventHandler):
Check 0.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@123187 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoUnreviewed. Set svn:mime-type to image/png for PNG files without svn:mime-type.
yosin@chromium.org [Fri, 20 Jul 2012 07:36:17 +0000 (07:36 +0000)]
Unreviewed. Set svn:mime-type to image/png for PNG files without svn:mime-type.

* platform/qt-5.0-wk2/fast/forms/number/input-appearance-number-rtl-expected.png: Added property svn:mime-type.
* platform/qt-5.0-wk2/fast/forms/number/input-appearance-spinbutton-disabled-readonly-expected.png: Added property svn:mime-type.
* platform/qt-5.0-wk2/fast/forms/number/input-appearance-spinbutton-layer-expected.png: Added property svn:mime-type.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@123186 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years ago[Qt] Compile WebCore without QtWidgets
commit-queue@webkit.org [Fri, 20 Jul 2012 06:48:11 +0000 (06:48 +0000)]
[Qt] Compile WebCore without QtWidgets
https://bugs.webkit.org/show_bug.cgi?id=91819

Patch by Simon Hausmann <simon.hausmann@nokia.com> on 2012-07-19
Reviewed by Kenneth Rohde Christiansen.

* WebCore.pri: QT += opengl is not needed for Qt 5, given that QtGui has OpenGL
support built in with the QOpenGL* classes.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@123185 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years ago[Refactoring] Replace Node's Document pointer with a TreeScope pointer
morrita@google.com [Fri, 20 Jul 2012 06:33:48 +0000 (06:33 +0000)]
[Refactoring] Replace Node's Document pointer with a TreeScope pointer
https://bugs.webkit.org/show_bug.cgi?id=59816

Reviewed by Ryosuke Niwa.

.:

* Source/autotools/symbols.filter: Added newly exported symbols.

Source/WebCore:

Before this change, Node::treeScope() fetches the TreeScope object
from ElementRareData. This approach has several shortcomings:

- rareData() call is slow due to a hashtable lookup.
- In shadow tree, each node has its tree scope in ElementRareData,
  that means the rare-data is no longer rare in that case.

This change gets rid of ElementRareData::m_treeScope by replacing
Node::m_document with Node::m_treeScope. And retrieves the
document of Node through m_treeScope.

Note that Node::document() is a hot function and naive
replacemennt of m_document with m_treeScope can hurt the
speed. This change employs some tricks to address it.

- This change introduces Node::InShadowTree flag, if the flag is off,
  that means m_treeScope is actually a document an can be returned as the result.
  this eliminates an extract dereference.
- Node::m_treeScope can be null. But we don't want to issue any extra
  conditional statement. So this change represents a null
  TreeScope as TreeScope::nullInstance(), which saves one conditional
  statement.

With these changes, the Node::document() slowdown is minimized to
unnoticeable size.

No new tests. Covered by existing tests.

* dom/Document.cpp: Took care of connectio betwen TreeScope.
(WebCore::Document::Document):
(WebCore::Document::~Document):
(WebCore::Document::suggestedMIMEType):
* dom/Document.h:
(WebCore::Node::treeScope): Now just return m_treeScope, taking care of nullInstance() case.
(WebCore):
(WebCore::Node::setTreeScope): Now just sets m_treeScope.
(WebCore::Node::documentInternal): Extracted from document() to have ASSERT-free version.
(WebCore::Node::document): Re-implemented over treeScope() and the flag.
(WebCore::Node::isDocumentNode): Re-implemented using treeScope()
(WebCore::Node::Node):
* dom/Node.cpp:
(WebCore::Node::~Node):
(WebCore::Node::reportMemoryUsage):
* dom/Node.h:
(Node):
(WebCore::Node::inDocument):
(WebCore::Node::isInShadowTree): Rewrote to use InShadowTree flag.
* dom/NodeRareData.h:
(WebCore::NodeRareData::NodeRareData): Eliminated m_treeScope.
(NodeRareData):
* dom/ShadowRoot.cpp:
(WebCore::ShadowRoot::ShadowRoot):
* dom/TreeScope.cpp:
(WebCore::TreeScope::TreeScope):
(WebCore):
(WebCore::TreeScope::setParentTreeScope): Added.
(WebCore::TreeScope::isDocumentScope): Added.
(WebCore::TreeScope::nullInstance): Added.
* dom/TreeScope.h: Added m_parentTreeScope.
(WebCore):
(WebCore::TreeScope::rootDocument): Added.
(TreeScope):
* dom/TreeScopeAdopter.cpp: No longer calls setDocument()
(WebCore::TreeScopeAdopter::moveTreeToNewScope):
(WebCore::TreeScopeAdopter::moveNodeToNewDocument):
* editing/MoveSelectionCommand.cpp: Includes Document.h to find inlined Node functions
* editing/RemoveNodeCommand.cpp: Includes Document.h to find inlined Node functions
* editing/RemoveNodePreservingChildrenCommand.cpp: Includes Document.h to find inlined Node functions
* inspector/PageConsoleAgent.cpp: Includes Document.h to find inlined Node functions

Source/WebKit2:

* win/WebKit2.def: Added newly exported symbols.
* win/WebKit2CFLite.def: Ditto.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@123184 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoSVG not properly respecting max-width.
hyatt@apple.com [Fri, 20 Jul 2012 06:12:54 +0000 (06:12 +0000)]
SVG not properly respecting max-width.
https://bugs.webkit.org/show_bug.cgi?id=91474

Source/WebCore:

My previous checkin for 91474 accidentally inverted width and height in the division case for
the computation of height. When I fixed this inversion, I discovered that <object> elements
are in fact also broken with max-width handling, and that furthermore, trying to apply the same
max-width fix by calling RenderBox::computeReplacedLogicalWidth/Height failed because those methods
call intrinsicLogicalWidth()/Height(). Becuase m_intrinsicSize is out-of-date and does not reflect
the values we just obtained from the contentRenderer, we use the default 300x150 values for object
and fail to render.

In order to both fix SVG/<object> with max-width constraints and to keep <object> rendering correctly
even when there are no max-width constraints, I was forced to update the m_intrinsicSize immediately
in order to make sure that the RenderBox methods returned the right values when computing the width/height
constrained to max/min-width/height values.

Reviewed by Dan Bernstein.

Added two new tests in svg/as-image. One test covers non-rectangular images to test for the inversion
mistake I made. The second test applies a max-width to <object> and shows that we have never gotten
this right before. An existing test in svg/as-image/ already covers basic <object> use (and tests that
the intrinsic size of 300x150 is not used when an explicit non-percentage size is specified on the SVG
itself).

* rendering/RenderReplaced.cpp:
(WebCore::RenderReplaced::computeAspectRatioInformationForRenderBox):
Add a check to update m_intrinsicSize when we know it should apply, so that the calls to check against
min/max-width fetch this correct size.

* rendering/RenderReplaced.h:
(RenderReplaced):
Make m_intrinsicSize mutable because of the mutation that occurs during the method above. It may be
that we should re-evaluate whether all of these methods should be const, but this would impact RenderBox
methods as well, so I chose to hold off going down that rabbit hole.

LayoutTests:

Reviewed by Dan Bernstein.

* svg/as-image/svg-intrinsic-size-rectangular-expected.html: Added.
* svg/as-image/svg-intrinsic-size-rectangular.html: Added.
* svg/as-image/svg-object-intrinsic-size-expected.html: Added.
* svg/as-image/svg-object-intrinsic-size.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@123183 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoAX: Need AccessibilityObjects for nodes without renderers in canvas subtree
dmazzoni@google.com [Fri, 20 Jul 2012 06:06:04 +0000 (06:06 +0000)]
AX: Need AccessibilityObjects for nodes without renderers in canvas subtree
https://bugs.webkit.org/show_bug.cgi?id=87899

Reviewed by Chris Fleizach.

Source/WebCore:

Refactors AccessibilityRenderObject so that it inherits from a new class,
AccessibilityNodeObject, that can be constructed from a Node without a
renderer. Modifies AXObjectCache so that it automatically creates an
AccessibilityNodeObject for elements in a canvas subtree but not otherwise.
A new layout test verifies that this correctly exposes an accessibility
tree with appropriate roles for elements in a canvas subtree.

This patch does not try to complete the implementation of
AccessibilityNodeObject. Most AX methods are still unimplemented and need
to be migrated from AccessibilityRenderObject to AccessibilityNodeObject
in a future patch.

This patch also doesn't change anything outside of WebCore/accessibility, so
the rest of WebCore only calls AXObjectCache when there are changes to a
RenderObject, not to a Node. Accessible notifications on nodes without
renderers need to be implemented in a future patch.

Test: accessibility/canvas-accessibilitynodeobject.html

* CMakeLists.txt:
* GNUmakefile.list.am:
* Target.pri:
* WebCore.gypi:
* accessibility/AXObjectCache.cpp:
(WebCore::AXObjectCache::get):
(WebCore):
(WebCore::createFromNode):
(WebCore::AXObjectCache::getOrCreate):
(WebCore::AXObjectCache::remove):
* accessibility/AXObjectCache.h:
(AXObjectCache):
(WebCore::AXObjectCache::get):
(WebCore::AXObjectCache::getOrCreate):
(WebCore::AXObjectCache::remove):
* accessibility/AccessibilityARIAGrid.cpp:
(WebCore::AccessibilityARIAGrid::AccessibilityARIAGrid):
(WebCore):
(WebCore::AccessibilityARIAGrid::~AccessibilityARIAGrid):
(WebCore::AccessibilityARIAGrid::init):
(WebCore::AccessibilityARIAGrid::create):
* accessibility/AccessibilityARIAGrid.h:
(AccessibilityARIAGrid):
* accessibility/AccessibilityARIAGridCell.cpp:
(WebCore::AccessibilityARIAGridCell::create):
* accessibility/AccessibilityARIAGridRow.cpp:
(WebCore::AccessibilityARIAGridRow::create):
* accessibility/AccessibilityAllInOne.cpp:
* accessibility/AccessibilityList.cpp:
(WebCore::AccessibilityList::create):
* accessibility/AccessibilityListBox.cpp:
(WebCore::AccessibilityListBox::create):
* accessibility/AccessibilityMediaControls.cpp:
(WebCore::AccessibilityMediaControl::create):
(WebCore::AccessibilityMediaControlsContainer::create):
(WebCore::AccessibilityMediaTimeline::create):
(WebCore::AccessibilityMediaTimeDisplay::create):
* accessibility/AccessibilityMenuList.cpp:
(WebCore::AccessibilityMenuList::create):
(WebCore):
* accessibility/AccessibilityMenuList.h:
(AccessibilityMenuList):
* accessibility/AccessibilityNodeObject.cpp: Added.
(WebCore):
(WebCore::AccessibilityNodeObject::AccessibilityNodeObject):
(WebCore::AccessibilityNodeObject::~AccessibilityNodeObject):
(WebCore::AccessibilityNodeObject::init):
(WebCore::AccessibilityNodeObject::create):
(WebCore::AccessibilityNodeObject::detach):
(WebCore::AccessibilityNodeObject::childrenChanged):
(WebCore::AccessibilityNodeObject::updateAccessibilityRole):
(WebCore::AccessibilityNodeObject::firstChild):
(WebCore::AccessibilityNodeObject::lastChild):
(WebCore::AccessibilityNodeObject::previousSibling):
(WebCore::AccessibilityNodeObject::nextSibling):
(WebCore::AccessibilityNodeObject::parentObjectIfExists):
(WebCore::AccessibilityNodeObject::parentObject):
(WebCore::AccessibilityNodeObject::elementRect):
(WebCore::AccessibilityNodeObject::setNode):
(WebCore::AccessibilityNodeObject::document):
(WebCore::AccessibilityNodeObject::determineAccessibilityRole):
(WebCore::AccessibilityNodeObject::addChildren):
(WebCore::AccessibilityNodeObject::accessibilityIsIgnored):
(WebCore::AccessibilityNodeObject::canSetFocusAttribute):
(WebCore::AccessibilityNodeObject::determineAriaRoleAttribute):
(WebCore::AccessibilityNodeObject::ariaRoleAttribute):
(WebCore::AccessibilityNodeObject::remapAriaRoleDueToParent):
* accessibility/AccessibilityNodeObject.h: Added.
(WebCore):
(AccessibilityNodeObject):
(WebCore::AccessibilityNodeObject::isAccessibilityNodeObject):
(WebCore::AccessibilityNodeObject::node):
(WebCore::AccessibilityNodeObject::isDetached):
(WebCore::toAccessibilityNodeObject):
* accessibility/AccessibilityObject.h:
(WebCore::AccessibilityObject::isAccessibilityNodeObject):
* accessibility/AccessibilityProgressIndicator.cpp:
(WebCore::AccessibilityProgressIndicator::create):
* accessibility/AccessibilityRenderObject.cpp:
(WebCore::AccessibilityRenderObject::AccessibilityRenderObject):
(WebCore::AccessibilityRenderObject::init):
(WebCore):
(WebCore::AccessibilityRenderObject::create):
(WebCore::AccessibilityRenderObject::detach):
(WebCore::AccessibilityRenderObject::setRenderer):
(WebCore::AccessibilityRenderObject::canHaveChildren):
(WebCore::AccessibilityRenderObject::addCanvasChildren):
(WebCore::AccessibilityRenderObject::addChildren):
* accessibility/AccessibilityRenderObject.h:
(AccessibilityRenderObject):
* accessibility/AccessibilitySlider.cpp:
(WebCore::AccessibilitySlider::create):
* accessibility/AccessibilityTable.cpp:
(WebCore::AccessibilityTable::AccessibilityTable):
(WebCore):
(WebCore::AccessibilityTable::~AccessibilityTable):
(WebCore::AccessibilityTable::init):
(WebCore::AccessibilityTable::create):
* accessibility/AccessibilityTable.h:
(AccessibilityTable):
* accessibility/AccessibilityTableCell.cpp:
(WebCore::AccessibilityTableCell::create):
* accessibility/AccessibilityTableRow.cpp:
(WebCore::AccessibilityTableRow::create):

LayoutTests:

This new test adds a bunch of elements to a normal container and the same elements to
a canvas subtree, then it iterates over all of the accessible objects within each
container and makes sure they have identical roles. Because we know the nodes in the
canvas subtree don't have renderers, this successfully tests that AccessibilityNodeObjects
are getting created for these nodes.

* accessibility/canvas-accessibilitynodeobject-expected.txt: Added.
* accessibility/canvas-accessibilitynodeobject.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@123182 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoThe ATSUI-based complex text code is unused
mitz@apple.com [Fri, 20 Jul 2012 06:01:10 +0000 (06:01 +0000)]
The ATSUI-based complex text code is unused
https://bugs.webkit.org/show_bug.cgi?id=91816

Reviewed by Sam Weinig.

../WebCore:

* WebCore.gypi: Removed references to the files deleted in this change.
* WebCore.xcodeproj/project.pbxproj: Ditto.
* platform/graphics/FontCache.h:
(FontCache): Replaced USE(CORE_TEXT) with PLATFORM(MAC) || PLATFORM(CHROMIUM) && OS(DARWIN).
* platform/graphics/SimpleFontData.h:
(SimpleFontData): Removed USE(ATSUI)-only code, replaced USE(CORE_TEXT) with
(PLATFORM(CHROMIUM) && OS(DARWIN)).
* platform/graphics/mac/ComplexTextController.cpp: Removed USE(ATSUI)-only code. Moved
collectComplexTextRunsForCharacters() and the ComplexTextRun constructor out of this file.
* platform/graphics/mac/ComplexTextController.h: Removed USE(ATSUI)-only code and removed
#if USE(CORE_TEXT) because anyone including this header is using Core Text.
* platform/graphics/mac/ComplexTextControllerATSUI.cpp: Removed.
* platform/graphics/mac/ComplexTextControllerCoreText.mm:
(WebCore::ComplexTextController::ComplexTextRun::ComplexTextRun): Moved from
ComplexTextController.cpp into this file and merged with createTextRunFromFontDataCoreText().
(WebCore::ComplexTextController::collectComplexTextRunsForCharacters): Moved from
ComplexTextController.cpp into this file and merged with
collectComplexTextRunsForCharactersCoreText().
* platform/graphics/mac/SimpleFontDataATSUI.mm: Removed.
* platform/graphics/mac/SimpleFontDataMac.mm:
(WebCore::SimpleFontData::platformInit): Remove USE(ATSUI)-only code.
(WebCore::SimpleFontData::platformDestroy): Ditto.
* platform/text/mac/ShapeArabic.c: Removed.
* platform/text/mac/ShapeArabic.h: Removed.

../WTF:

Removed definitions of WTF_USE_ATSUI and WTF_USE_CORE_TEXT.

* wtf/Platform.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@123181 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoIntroduce Node::shadowHost()
tkent@chromium.org [Fri, 20 Jul 2012 05:54:31 +0000 (05:54 +0000)]
Introduce Node::shadowHost()
https://bugs.webkit.org/show_bug.cgi?id=91814

Reviewed by Hajime Morita.

Introduce Node::shadowHost(), which return the host element, or 0.
Node::shadowAncestorNode() is used to obtain a shadow host. However it
is confusing because it returns 'this' if this is not in a shadow tree.

Replaces one callsite of shadowAncestorNode() with shadowHost().

No behavior change.

* dom/Node.cpp:
(WebCore::Node::shadowHost): Added.
* dom/Node.h:
(Node): Declare shadowHost(), and add a comment to shadowAncestorNode().
* html/shadow/CalendarPickerElement.cpp:
(WebCore::CalendarPickerElement::hostInput):
Replace shadowAncestorNode() with shaodwHost().

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@123180 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years ago[BlackBerry] Make sure to send favicon when go back/forward
mary.wu@torchmobile.com.cn [Fri, 20 Jul 2012 05:33:33 +0000 (05:33 +0000)]
[BlackBerry] Make sure to send favicon when go back/forward
https://bugs.webkit.org/show_bug.cgi?id=91808

Reviewed by George Staikos.

When go back/forward, if the page was in page cache, it would have no chance
to send favicon. So we'll send it in commitLoad right after send the blank
icon.
RIM PR# 177495

* WebCoreSupport/FrameLoaderClientBlackBerry.cpp:
(WebCore::FrameLoaderClientBlackBerry::dispatchDidCommitLoad):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@123179 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoForm state restore: Need to identify a form by its content
tkent@chromium.org [Fri, 20 Jul 2012 04:48:17 +0000 (04:48 +0000)]
Form state restore: Need to identify a form by its content
https://bugs.webkit.org/show_bug.cgi?id=91209

Reviewed by Hajime Morita.

Source/WebCore:

Add names of the first two controls of a form to its formKey
string. By this change, we can correctly restore states to
reordered forms like webkit.org/b/91209#c0.

Tests: Added test cases to fast/forms/state-restore-per-form.html.

* html/FormController.cpp:
(WebCore::recordFormStructure):
Append at most two name attribute values.
(WebCore::createKey): Insert a string built by recordFromStructure().
(WebCore::formStateSignature): Bump the version.

LayoutTests:

* fast/forms/state-restore-broken-state-expected.txt:
Updated for the serialized format change.
* fast/forms/state-restore-per-form-expected.txt:
* fast/forms/state-restore-per-form.html:
- Add a control to #form2 before 'additionalControl'
 We can't identify this form if the second named control is changed.
- Add test cases of webkit.org/b/91209#c0.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@123178 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years ago[EFL] [WK2] Add methods to get/set a custom text encoding
commit-queue@webkit.org [Fri, 20 Jul 2012 04:31:58 +0000 (04:31 +0000)]
[EFL] [WK2] Add methods to get/set a custom text encoding
https://bugs.webkit.org/show_bug.cgi?id=90604

Patch by Sudarsana Nagineni <sudarsana.nagineni@linux.intel.com> on 2012-07-19
Reviewed by Kenneth Rohde Christiansen.

Source/WebKit2:

Add methods to get/set custom character encoding.

* UIProcess/API/efl/ewk_view.cpp:
(_Ewk_View_Private_Data):
(_ewk_view_priv_del):
(ewk_view_setting_encoding_custom_get):
(ewk_view_setting_encoding_custom_set):
* UIProcess/API/efl/ewk_view.h:

Tools:

Added support for setting custom character encoding in MiniBrowser.
Use 'F3' to set the custom character encoding.

* MiniBrowser/efl/main.c:
(on_key_down):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@123177 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoEnable CSS variables compile time flag in Chrome.
macpherson@chromium.org [Fri, 20 Jul 2012 04:29:39 +0000 (04:29 +0000)]
Enable CSS variables compile time flag in Chrome.
https://bugs.webkit.org/show_bug.cgi?id=91708

Reviewed by Dimitri Glazkov.

Flips the ENABLE_CSS_VARIABLES compile flag on for Chrome. There is also a run-time flag that is off by default,
so this should not have a user visible effect other than making the variables runtime flag visible at chrome://flags/

* features.gypi:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@123176 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoenable Web Audio for chromium android port
commit-queue@webkit.org [Fri, 20 Jul 2012 04:27:18 +0000 (04:27 +0000)]
enable Web Audio for chromium android port
https://bugs.webkit.org/show_bug.cgi?id=89428

Patch by Wei James <james.wei@intel.com> on 2012-07-19
Reviewed by Kenneth Russell.

Source/WebCore:

* WebCore.gyp/WebCore.gyp:
* platform/audio/FFTFrameStub.cpp:
(WebCore::FFTFrame::doFFT):

Source/WebKit/chromium:

* features.gypi:

Source/WTF:

* wtf/MathExtras.h:
(log2):
(log2f):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@123175 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years ago[Chromium] Suppress tests requiring a rebaseline.
commit-queue@webkit.org [Fri, 20 Jul 2012 04:27:13 +0000 (04:27 +0000)]
[Chromium] Suppress tests requiring a rebaseline.
Skia roll to 4683 changes gradient rendering, and rebaselines are required.
https://bugs.webkit.org/show_bug.cgi?id=91797

Patch by Jeff Timanus <twiz@chromium.org> on 2012-07-19
Reviewed by Kenneth Russell.

* platform/chromium/TestExpectations:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@123174 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agowebkitpy: executive_unittest still failing when run in parallel
dpranke@chromium.org [Fri, 20 Jul 2012 04:23:05 +0000 (04:23 +0000)]
webkitpy: executive_unittest still failing when run in parallel
https://bugs.webkit.org/show_bug.cgi?id=91795

Reviewed by Adam Barth.

It turns out that test_kill_process and test_kill_all are both
launching the same process, and so if they're run at the same
time, test_kill_all interferes w/ test_kill_process.

Merging the two tests solves the issue :).

* Scripts/webkitpy/common/system/executive_unittest.py:
(ExecutiveTest.test_kill_process):
(ExecutiveTest._assert_windows_image_name):
(ExecutiveTest):
(ExecutiveTest.test_windows_image_name):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@123173 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years ago[BlackBerry] Suspend when there's no target buffer until an external compositor is set
jpetsovits@rim.com [Fri, 20 Jul 2012 04:08:16 +0000 (04:08 +0000)]
[BlackBerry] Suspend when there's no target buffer until an external compositor is set
https://bugs.webkit.org/show_bug.cgi?id=91686
RIM PR 174365

Reviewed by Antonio Gomes.

If we don't have a client window (i.e. rendering to
GL directly) and a WebPageCompositor is only set
after a rendering operation, then we'll try to render
to BackingStorePrivate::buffer() which doesn't exist
at this point. That's bad, and gets us various
assertions and possibly worse.

Fix it by starting in a screen-suspended state and only
resuming screen and backingstore once a compositor is
actually set.

So, in effect, with this patch applied, the sequence
of events will look like this:

1) WebPage & BackingStore are initialize, neither window
   nor compositor exists, therefore buffer() returns 0.
   createSurface() therefore suspends screen and
   backingstore.
2) loadURL() or loadData() is called, web page is
   fully loaded, however we don't try to render because
   we're still suspended, still have no target buffer.
3) A WebPageCompositor is being set from outside.
   At the beginning of WebPage::setCompositor() we still
   don't have a buffer() so there's nothing to suspend,
   however, after the sync call to setCompositorHelper()
   the compositor is set so buffer() will return a
   nonzero value, causing us to resume at this point.

Using the existence of a target buffer to determine
whether or not to enable rendering or keep it suspended
seems like a good idea, and the implementation (while
not quite perfect yet) is a step forward from before.

* Api/BackingStore.cpp:
(BlackBerry::WebKit::BackingStorePrivate::createSurfaces):
* Api/WebPage.cpp:
(BlackBerry::WebKit::WebPagePrivate::setCompositor):
(BlackBerry::WebKit::WebPagePrivate::setCompositorHelper):
* Api/WebPage_p.h:
(WebPagePrivate):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@123172 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoSVG images broken when max-width specified.
hyatt@apple.com [Fri, 20 Jul 2012 03:59:03 +0000 (03:59 +0000)]
SVG images broken when max-width specified.
https://bugs.webkit.org/show_bug.cgi?id=91474

Source/WebCore:

SVG images were computing intrinsic dimensions when width and height were auto that did not
respect min-max width/height. Normal images had code that applied these constraints properly.
Looking at the code before the check-in that broke things, these constraints used to be
applied to all images regardless of type via calcAspectRatioLogicalWidth/Height.

This patch leaves the new function structure in place but converts the code to be more like
it was prior to the introduction of the regression. Instead of raw intrinsic sizes being
used in the SVG case, now all image types get the intrinsic sizes constrained when doing
width/height computations.

Reviewed by Dan Bernstein.

Test: svg/as-image/svg-intrinsic-size.html

* rendering/RenderReplaced.cpp:
(WebCore::RenderReplaced::computeAspectRatioInformationForRenderBox):
Rename computeIntrinsicRatioInformationForRenderBox to computeAspectRatioInformationForRenderBox.
Also rename the intrinsicSize reference to constrainedSize to reflect the fact that the size
is not necessarily the intrinsic size any longer but instead a size where both axes have been
constrained based off the max-min values of the opposite axes.

Move the constraining code out of RenderReplaced::computeIntrinsicRatioInformation into this
function so that the SVG code path appies the constraints as well. The movement of this code
is what fixes the bug.

(WebCore::RenderReplaced::computeIntrinsicRatioInformation):
Changed to remove the code that constrains the returned size, since it is shifting to
computeAspectRatioInformationForRenderBox instead.

(WebCore::RenderReplaced::computeReplacedLogicalWidth):
(WebCore::RenderReplaced::computeReplacedLogicalHeight):
* rendering/RenderReplaced.h:
(RenderReplaced):
Patch the name of the reference passed in to computeReplacedLogicalWidth/Height to be
constrainedSize instead of intrinsicSize, so that it is more obvious that the returned
result is not just the intrinsic size of the image.

LayoutTests:

Reviewed by Dan Bernstein.

* svg/as-image/svg-intrinsic-size-expected.html: Added.
* svg/as-image/svg-intrinsic-size.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@123171 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoUnreviewed. Fix build break because of changing parameter type of
gyuyoung.kim@samsung.com [Fri, 20 Jul 2012 03:57:56 +0000 (03:57 +0000)]
Unreviewed. Fix build break because of changing parameter type of
vibration virtual function on EFL port.

* WebProcess/WebCoreSupport/WebVibrationClient.cpp:
(WebKit::WebVibrationClient::vibrate):
* WebProcess/WebCoreSupport/WebVibrationClient.h:
(WebVibrationClient):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@123170 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoDFG cell checks should be hoisted
fpizlo@apple.com [Fri, 20 Jul 2012 03:50:02 +0000 (03:50 +0000)]
DFG cell checks should be hoisted
https://bugs.webkit.org/show_bug.cgi?id=91717

Reviewed by Geoffrey Garen.

The DFG has always had the policy of hoisting array and integer checks to
the point of variable assignment. Eventually, we added doubles and booleans
to the mix. But cells should really be part of this as well, particularly
for 32-bit where accessing a known-type variable is dramatically cheaper
than accessing a variable whose types is only predicted but otherwise
unproven.

This appears to be a definite speed-up for V8 on 32-bit, a possible speed-up
for Kraken, and a possible slow-down for V8 on 64-bit (around 0.2% if at
all). Any slow-downs can, and should, be addressed by making the hoisting
logic cognizant of variables that are never used in a manner that requires
type checks, and by sinking argument checks to the point(s) of first use.

To make this work I had to change some OSR machinery, and special-case the
type predictions of the 'this' argument for constructors. OSR exit normally
assumes that arguments are boxed, which happens to be true because the
type prediction used for check hoisting is LUB'd with the type of the
argument that was passed in - so either the arguments are always stored to
with the full tag+payload, or if only the payload is stored then the tag
matches whatever the caller would have set. But not so with the 'this'
argument for constructors, which is not initialized by the caller. We
could make this more precise by having argument types for OSR be inferred
using similar machinery to other locals, but I figured that for this patch
I should use the surgical fix.

* assembler/MacroAssemblerX86_64.h:
(JSC::MacroAssemblerX86_64::branchTestPtr):
(MacroAssemblerX86_64):
* assembler/X86Assembler.h:
(JSC::X86Assembler::testq_rm):
(X86Assembler):
* dfg/DFGAbstractState.cpp:
(JSC::DFG::AbstractState::initialize):
(JSC::DFG::AbstractState::execute):
* dfg/DFGDriver.cpp:
(JSC::DFG::compile):
* dfg/DFGGraph.h:
(JSC::DFG::Graph::isCreatedThisArgument):
(Graph):
* dfg/DFGSpeculativeJIT.cpp:
(JSC::DFG::SpeculativeJIT::checkArgumentTypes):
* dfg/DFGSpeculativeJIT32_64.cpp:
(JSC::DFG::SpeculativeJIT::compile):
* dfg/DFGSpeculativeJIT64.cpp:
(JSC::DFG::SpeculativeJIT::compile):
* dfg/DFGValueSource.h:
(JSC::DFG::ValueSource::forSpeculation):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@123169 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years ago[CSS2.1] Anonymous tables should be inline/block-level based off their parent
jchaffraix@webkit.org [Fri, 20 Jul 2012 01:31:55 +0000 (01:31 +0000)]
[CSS2.1] Anonymous tables should be inline/block-level based off their parent
https://bugs.webkit.org/show_bug.cgi?id=15365

Another unreviewed rebaseline after r123159 (gtk).

* fast/dynamic/insert-before-table-part-in-continuation-expected.txt: Added.
* platform/efl/fast/dynamic/insert-before-table-part-in-continuation-expected.txt: Removed.
* platform/gtk/fast/dynamic/insert-before-table-part-in-continuation-expected.txt: Removed.
* platform/gtk/tables/mozilla/bugs/bug3037-1-expected.txt: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@123168 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoFix test after r123145.
arv@chromium.org [Fri, 20 Jul 2012 00:22:06 +0000 (00:22 +0000)]
Fix test after r123145.

Unreviewed.

* fast/js/var-declarations-shadowing-expected.txt:
* fast/js/var-declarations-shadowing.html:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@123167 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoUnreviewed, reverting http://trac.webkit.org/changeset/123149.
dimich@chromium.org [Fri, 20 Jul 2012 00:21:16 +0000 (00:21 +0000)]
Unreviewed, reverting trac.webkit.org/changeset/123149.
It broke an IDB test on Chromium page cycler.

* bindings/v8/SerializedScriptValue.cpp:
(WebCore::SerializedScriptValue::SerializedScriptValue):
(WebCore):
* bindings/v8/SerializedScriptValue.h:
* dom/MessageEvent.cpp:
(WebCore::MessageEvent::MessageEvent):
(WebCore::MessageEvent::initMessageEvent):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@123166 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years ago[CSS2.1] Anonymous tables should be inline/block-level based off their parent
jchaffraix@webkit.org [Fri, 20 Jul 2012 00:20:55 +0000 (00:20 +0000)]
[CSS2.1] Anonymous tables should be inline/block-level based off their parent
https://bugs.webkit.org/show_bug.cgi?id=15365

Unreviewed gardening after r123159.

* mathml/msub-anonymous-child-render-crash-expected.txt:
MathML wraps msub / mpub in inline-blocks so the table wrapper is now inline too which leads to this
baseline change.

* platform/chromium-mac-snowleopard/fast/dynamic/insert-before-table-part-in-continuation-expected.png:
* platform/chromium-mac/fast/dynamic/insert-before-table-part-in-continuation-expected.png:
* platform/chromium-win/fast/dynamic/insert-before-table-part-in-continuation-expected.png:
* platform/efl/fast/dynamic/insert-before-table-part-in-continuation-expected.txt:
* platform/efl/tables/mozilla/bugs/bug3037-1-expected.txt: Removed.
* platform/gtk/tables/mozilla/bugs/bug3037-1-expected.txt: Removed.
* platform/mac-wk2/fast/dynamic: Added.
* platform/mac-wk2/fast/dynamic/insert-before-table-part-in-continuation-expected.txt: Added.
* platform/mac-wk2/mathml: Added.
* platform/mac-wk2/mathml/msub-anonymous-child-render-crash-expected.txt: Added.
* platform/mac-wk2/tables: Added.
* platform/mac-wk2/tables/mozilla: Added.
* platform/mac-wk2/tables/mozilla/bugs: Added.
* platform/mac-wk2/tables/mozilla/bugs/bug3037-1-expected.txt: Added.
* platform/mac/fast/dynamic/insert-before-table-part-in-continuation-expected.txt:
* platform/mac/tables/mozilla/bugs/bug3037-1-expected.txt:
* tables/mozilla/bugs/bug3037-1-expected.txt: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@123165 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoFast path of storage resize should be removed from property storage reallocation...
fpizlo@apple.com [Fri, 20 Jul 2012 00:04:51 +0000 (00:04 +0000)]
Fast path of storage resize should be removed from property storage reallocation, since it is only useful for arrays
https://bugs.webkit.org/show_bug.cgi?id=91796

Reviewed by Geoffrey Garen.

* dfg/DFGRepatch.cpp:
(JSC::DFG::emitPutTransitionStub):
* dfg/DFGSpeculativeJIT.cpp:
(JSC::DFG::SpeculativeJIT::compileReallocatePropertyStorage):
* runtime/JSObject.cpp:
(JSC::JSObject::growOutOfLineStorage):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@123164 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoUnreviewed gardening. Land chromium win and chromium-linux baselines for
tony@chromium.org [Thu, 19 Jul 2012 23:43:52 +0000 (23:43 +0000)]
Unreviewed gardening.  Land chromium win and chromium-linux baselines for
fast/backgrounds/size/zero.html after r121296.

* fast/backgrounds/size/zero-expected.txt: Renamed from LayoutTests/platform/efl/fast/backgrounds/size/zero-expected.txt.
* platform/chromium-linux/fast/backgrounds/size/zero-expected.png:
* platform/chromium-win-xp/fast/backgrounds/size/zero-expected.png: Added.
* platform/chromium-win/fast/backgrounds/size/zero-expected.png:
* platform/chromium-win/fast/backgrounds/size/zero-expected.txt:
* platform/chromium/TestExpectations:
* platform/gtk/fast/backgrounds/size/zero-expected.txt: Removed.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@123163 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years ago[Chromium] disable warning on Windows from gcc-only pragma in Assertions.cpp"
commit-queue@webkit.org [Thu, 19 Jul 2012 23:40:54 +0000 (23:40 +0000)]
[Chromium] disable warning on Windows from gcc-only pragma in Assertions.cpp"
https://bugs.webkit.org/show_bug.cgi?id=91776

Patch by Scott Graham <scottmg@chromium.org> on 2012-07-19
Reviewed by Ryosuke Niwa.

Avoids:
    ...\source\wtf\wtf\assertions.cpp(31) : warning C4068: unknown pragma

* WTF.gyp/WTF.gyp:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@123162 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years ago[chromium][Mac] Switch the MACOSX_DEPLOYMENT_TARGET to 10.6
commit-queue@webkit.org [Thu, 19 Jul 2012 23:32:04 +0000 (23:32 +0000)]
[chromium][Mac] Switch the MACOSX_DEPLOYMENT_TARGET to 10.6
https://bugs.webkit.org/show_bug.cgi?id=91752

Patch by Robert Sesek <rsesek@chromium.org> on 2012-07-19
Reviewed by Eric Seidel.

When building CHROMIUM && DARWIN, do not enable ATSUI support.

* wtf/Platform.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@123161 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years ago[Chromium] Support external autofill popups
abarth@webkit.org [Thu, 19 Jul 2012 23:27:36 +0000 (23:27 +0000)]
[Chromium] Support external autofill popups
https://bugs.webkit.org/show_bug.cgi?id=91651

Reviewed by Darin Fisher.

This looks like the last piece needed to support external autofill
popups. This function just lets the embedder select an index
(presumably because the user picked an index from the external autofill
popup).

* public/WebView.h:
(WebView):
* src/WebViewImpl.cpp:
(WebKit::WebViewImpl::selectAutofillSuggestionAtIndex):
(WebKit):
* src/WebViewImpl.h:
(WebViewImpl):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@123160 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoSource/WebCore: [CSS2.1] Anonymous tables should be inline/block-level based off...
jchaffraix@webkit.org [Thu, 19 Jul 2012 23:10:50 +0000 (23:10 +0000)]
Source/WebCore: [CSS2.1] Anonymous tables should be inline/block-level based off their parent
https://bugs.webkit.org/show_bug.cgi?id=15365

Reviewed by Abhishek Arya.

This change makes us match bullet 3 "Generate missing parents" of section 17.2.1 http://www.w3.org/TR/CSS2/tables.html#anonymous-boxes":
"If a table is contained by an inline element, then the anonymous tables should be inline-tables."

Patch based on an earlier change by Kang-Hao (Kenny) Lu <kennyluck@csail.mit.edu>.

Tests: css2.1/20110323/inline-table-002b.html
       fast/dynamic/insert-before-table-part-in-continuation.html
       fast/table/inline-table-in-continuation.html
       fast/table/inline-table-with-caption.html

* rendering/RenderTable.cpp:
(WebCore::RenderTable::createAnonymousWithParentRenderer):
Added a check to determine if we should be inline or block based on our parent to match CSS 2.1.

* rendering/RenderInline.cpp:
(WebCore::RenderInline::addChildIgnoringContinuation):
Changed the child addition logic to not wrap a table part in an inline context. This change works
as intended as we will call RenderObject::addChild which will create an anonymous inline table that
will be added under |this| instead of the table part. As the table is inline, it doesn't need to be
wrapped when we recursively call RenderInline::addChild.

Source/WebKit2: [WK2] Add Vibration API integration to WebContext and WebPage
https://bugs.webkit.org/show_bug.cgi?id=91081

Reviewed by Anders Carlsson.

Integrate Vibration API to WebPage, WebContext and properly route
messages to the WebVibrationProxy.

* UIProcess/API/C/WKContext.cpp:
(WKContextGetVibration):
* UIProcess/API/C/WKContext.h:
* UIProcess/WebContext.cpp:
(WebKit::WebContext::WebContext):
(WebKit::WebContext::~WebContext):
(WebKit::WebContext::disconnectProcess):
(WebKit::WebContext::didReceiveMessage):
* UIProcess/WebContext.h:
(WebKit):
(WebContext):
(WebKit::WebContext::vibrationProxy):
* UIProcess/WebProcessProxy.cpp:
(WebKit::WebProcessProxy::didReceiveMessage):
* WebProcess/WebPage/WebPage.cpp:
(WebKit::WebPage::WebPage):

Tools: [EFL] [WK2] Add a callback to handle delete request in MiniBrowser
https://bugs.webkit.org/show_bug.cgi?id=91483

Reviewed by Dirk Pranke.

Add a callback to handle delete request and quit the main loop when
window is asked to be deleted so that all the resources allocated
for the window will be freed properly.

* MiniBrowser/efl/main.c:
(closeWindow):
(browserCreate):

LayoutTests: [CSS2.1] Anonymous tables should be inline/block-level based off their parent
https://bugs.webkit.org/show_bug.cgi?id=15365

Reviewed by Abhishek Arya.

* css2.1/20110323/inline-table-002b-expected.htm: Added.
* css2.1/20110323/inline-table-002b.html: Added.

* fast/table/inline-table-in-continuation-expected.html: Added.
* fast/table/inline-table-in-continuation.html: Added.
Added a test for an anonymous inline table in a continuation.

* fast/table/inline-table-with-caption-expected.html: Added.
* fast/table/inline-table-with-caption.html: Added.
Added a test for a table-caption in inline-context.

* platform/chromium-win/tables/mozilla/bugs/bug3037-1-expected.txt:
This is a progression: we now properly make the table inline and don't wrap it in anonymous blocks.

* platform/chromium-linux/fast/dynamic/insert-before-table-part-in-continuation-expected.png:
* platform/chromium-win/fast/dynamic/insert-before-table-part-in-continuation-expected.txt:
This is a progression too. However we still have some differences with Firefox and Opera.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@123159 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years ago[WK2] Add Vibration API integration to WebContext and WebPage
commit-queue@webkit.org [Thu, 19 Jul 2012 22:57:28 +0000 (22:57 +0000)]
[WK2] Add Vibration API integration to WebContext and WebPage
https://bugs.webkit.org/show_bug.cgi?id=91081

Patch by Sudarsana Nagineni <sudarsana.nagineni@linux.intel.com> on 2012-07-19
Reviewed by Anders Carlsson.

Integrate Vibration API to WebPage, WebContext and properly route
messages to the WebVibrationProxy.

* UIProcess/API/C/WKContext.cpp:
(WKContextGetVibration):
* UIProcess/API/C/WKContext.h:
* UIProcess/WebContext.cpp:
(WebKit::WebContext::WebContext):
(WebKit::WebContext::~WebContext):
(WebKit::WebContext::disconnectProcess):
(WebKit::WebContext::didReceiveMessage):
* UIProcess/WebContext.h:
(WebKit):
(WebContext):
(WebKit::WebContext::vibrationProxy):
* UIProcess/WebProcessProxy.cpp:
(WebKit::WebProcessProxy::didReceiveMessage):
* WebProcess/WebPage/WebPage.cpp:
(WebKit::WebPage::WebPage):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@123158 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoSource/WebCore: WebKit should pass <title> element value as the default description...
cfleizach@apple.com [Thu, 19 Jul 2012 22:55:33 +0000 (22:55 +0000)]
Source/WebCore: WebKit should pass <title> element value as the default description for web views
https://bugs.webkit.org/show_bug.cgi?id=91763

Reviewed by Anders Carlsson.

Update the logic for determining the description for the AXWebArea to include the document's title.

Test: platform/mac/accessibility/document-title-used-for-description.html

* accessibility/AccessibilityRenderObject.cpp:
(WebCore::AccessibilityRenderObject::webAreaAccessibilityDescription):
   Move web area description handling into a separate method.
(WebCore::AccessibilityRenderObject::accessibilityDescription):
* accessibility/AccessibilityRenderObject.h:
(AccessibilityRenderObject):

LayoutTests: WebKit should pass <title> element value as the default AXTitle for web views
https://bugs.webkit.org/show_bug.cgi?id=91763

Reviewed by Anders Carlsson.

* platform/mac/accessibility/document-title-used-for-description-expected.txt: Added.
* platform/mac/accessibility/document-title-used-for-description.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@123157 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years ago[Qt] Added/updated missing fast/forms results after rebaseline
commit-queue@webkit.org [Thu, 19 Jul 2012 22:40:46 +0000 (22:40 +0000)]
[Qt] Added/updated missing fast/forms results after rebaseline
https://bugs.webkit.org/show_bug.cgi?id=91504

Patch by Bruno de Oliveira Abinader <bruno.abinader@basyskom.com> on 2012-07-19
Reviewed by Noam Rosenthal.

The following tests were failing after r122879:
  fast/forms/HTMLOptionElement_label06.html = TEXT
  fast/forms/HTMLOptionElement_label07.html = TEXT
  fast/forms/basic-inputs.html = TEXT
  fast/forms/basic-textareas-quirks.html = TEXT
  fast/forms/box-shadow-override.html = TEXT
  fast/forms/control-restrict-line-height.html = TEXT
  fast/forms/encoding-test.html = TEXT
  fast/forms/fieldset-align.html = TEXT
  fast/forms/form-element-geometry.html = TEXT
  fast/forms/input-appearance-bkcolor.html = TEXT
  fast/forms/input-appearance-default-bkcolor.html = TEXT
  fast/forms/input-appearance-disabled.html = TEXT
  fast/forms/input-appearance-focus.html = TEXT
  fast/forms/input-appearance-preventDefault.html = TEXT
  fast/forms/input-appearance-readonly.html = TEXT
  fast/forms/input-appearance-selection.html = TEXT
  fast/forms/input-appearance-visibility.html = TEXT
  fast/forms/input-baseline.html = TEXT
  fast/forms/input-disabled-color.html = TEXT
  fast/forms/input-double-click-selection-gap-bug.html = TEXT
  fast/forms/input-placeholder-visibility-1.html = TEXT
  fast/forms/input-placeholder-visibility-3.html = TEXT
  fast/forms/input-readonly-dimmed.html = TEXT
  fast/forms/input-readonly-empty.html = TEXT
  fast/forms/input-spaces.html = TEXT
  fast/forms/input-text-click-inside.html = TEXT
  fast/forms/input-text-double-click.html = TEXT
  fast/forms/input-text-option-delete.html = TEXT
  fast/forms/input-text-self-emptying-click.html = TEXT
  fast/forms/input-text-word-wrap.html = TEXT
  fast/forms/input-width.html = TEXT
  fast/forms/menulist-separator-painting.html = TEXT
  fast/forms/number/input-appearance-number-rtl.html = TEXT
  fast/forms/number/input-appearance-spinbutton-disabled-readonly.html = TEXT
  fast/forms/number/input-appearance-spinbutton-layer.html = TEXT
  fast/forms/placeholder-position.html = TEXT
  fast/forms/placeholder-pseudo-style.html = TEXT
  fast/forms/search-cancel-button-style-sharing.html = TEXT
  fast/forms/search-display-none-cancel-button.html = TEXT
  fast/forms/search-rtl.html = TEXT
  fast/forms/search-vertical-alignment.html = TEXT
  fast/forms/select-background-none.html = TEXT
  fast/forms/select-baseline.html = TEXT
  fast/forms/selectlist-minsize.html = TEXT
  fast/forms/text-style-color.html = TEXT
  fast/forms/textarea-align.html = TEXT
  fast/forms/textarea-placeholder-pseudo-style.html = TEXT
  fast/forms/textarea-placeholder-visibility-1.html = TEXT
  fast/forms/textarea-placeholder-visibility-2.html = TEXT
  fast/forms/textarea-setinnerhtml.html = TEXT
  fast/forms/textfield-focus-ring.html = TEXT
  fast/forms/textfield-outline.html = TEXT
  fast/forms/textfield-overflow.html = TEXT

This patch includes/updates the missing expectations.

* platform/qt-5.0-wk2/fast/forms/HTMLOptionElement_label06-expected.png:
* platform/qt-5.0-wk2/fast/forms/HTMLOptionElement_label06-expected.txt:
* platform/qt-5.0-wk2/fast/forms/HTMLOptionElement_label07-expected.png:
* platform/qt-5.0-wk2/fast/forms/HTMLOptionElement_label07-expected.txt:
* platform/qt-5.0-wk2/fast/forms/basic-inputs-expected.png: Added.
* platform/qt-5.0-wk2/fast/forms/basic-inputs-expected.txt: Added.
* platform/qt-5.0-wk2/fast/forms/basic-textareas-quirks-expected.png: Added.
* platform/qt-5.0-wk2/fast/forms/basic-textareas-quirks-expected.txt: Added.
* platform/qt-5.0-wk2/fast/forms/box-shadow-override-expected.png:
* platform/qt-5.0-wk2/fast/forms/box-shadow-override-expected.txt:
* platform/qt-5.0-wk2/fast/forms/control-restrict-line-height-expected.png: Added.
* platform/qt-5.0-wk2/fast/forms/control-restrict-line-height-expected.txt: Added.
* platform/qt-5.0-wk2/fast/forms/encoding-test-expected.txt: Added.
* platform/qt-5.0-wk2/fast/forms/fieldset-align-expected.png: Added.
* platform/qt-5.0-wk2/fast/forms/fieldset-align-expected.txt: Added.
* platform/qt-5.0-wk2/fast/forms/form-element-geometry-expected.png:
* platform/qt-5.0-wk2/fast/forms/form-element-geometry-expected.txt:
* platform/qt-5.0-wk2/fast/forms/input-appearance-bkcolor-expected.txt: Added.
* platform/qt-5.0-wk2/fast/forms/input-appearance-default-bkcolor-expected.png: Added.
* platform/qt-5.0-wk2/fast/forms/input-appearance-default-bkcolor-expected.txt: Added.
* platform/qt-5.0-wk2/fast/forms/input-appearance-disabled-expected.png: Added.
* platform/qt-5.0-wk2/fast/forms/input-appearance-disabled-expected.txt: Added.
* platform/qt-5.0-wk2/fast/forms/input-appearance-focus-expected.png: Added.
* platform/qt-5.0-wk2/fast/forms/input-appearance-focus-expected.txt: Added.
* platform/qt-5.0-wk2/fast/forms/input-appearance-preventDefault-expected.txt: Added.
* platform/qt-5.0-wk2/fast/forms/input-appearance-readonly-expected.png: Added.
* platform/qt-5.0-wk2/fast/forms/input-appearance-readonly-expected.txt: Added.
* platform/qt-5.0-wk2/fast/forms/input-appearance-selection-expected.png: Added.
* platform/qt-5.0-wk2/fast/forms/input-appearance-selection-expected.txt: Added.
* platform/qt-5.0-wk2/fast/forms/input-appearance-visibility-expected.txt: Added.
* platform/qt-5.0-wk2/fast/forms/input-baseline-expected.png: Added.
* platform/qt-5.0-wk2/fast/forms/input-baseline-expected.txt: Added.
* platform/qt-5.0-wk2/fast/forms/input-disabled-color-expected.png:
* platform/qt-5.0-wk2/fast/forms/input-disabled-color-expected.txt:
* platform/qt-5.0-wk2/fast/forms/input-double-click-selection-gap-bug-expected.png: Added.
* platform/qt-5.0-wk2/fast/forms/input-double-click-selection-gap-bug-expected.txt: Added.
* platform/qt-5.0-wk2/fast/forms/input-placeholder-visibility-1-expected.txt: Added.
* platform/qt-5.0-wk2/fast/forms/input-placeholder-visibility-3-expected.txt: Added.
* platform/qt-5.0-wk2/fast/forms/input-readonly-dimmed-expected.txt: Added.
* platform/qt-5.0-wk2/fast/forms/input-readonly-empty-expected.txt: Added.
* platform/qt-5.0-wk2/fast/forms/input-spaces-expected.txt: Added.
* platform/qt-5.0-wk2/fast/forms/input-text-click-inside-expected.txt: Added.
* platform/qt-5.0-wk2/fast/forms/input-text-double-click-expected.txt: Added.
* platform/qt-5.0-wk2/fast/forms/input-text-option-delete-expected.txt: Added.
* platform/qt-5.0-wk2/fast/forms/input-text-self-emptying-click-expected.txt: Added.
* platform/qt-5.0-wk2/fast/forms/input-text-word-wrap-expected.txt: Added.
* platform/qt-5.0-wk2/fast/forms/input-width-expected.txt: Copied from LayoutTests/platform/qt-5.0-wk2/fast/forms/textarea-setinnerhtml-expected.txt.
* platform/qt-5.0-wk2/fast/forms/menulist-separator-painting-expected.txt:
* platform/qt-5.0-wk2/fast/forms/number/input-appearance-number-rtl-expected.png: Added.
* platform/qt-5.0-wk2/fast/forms/number/input-appearance-number-rtl-expected.txt: Added.
* platform/qt-5.0-wk2/fast/forms/number/input-appearance-spinbutton-disabled-readonly-expected.png: Added.
* platform/qt-5.0-wk2/fast/forms/number/input-appearance-spinbutton-disabled-readonly-expected.txt: Added.
* platform/qt-5.0-wk2/fast/forms/number/input-appearance-spinbutton-layer-expected.png: Added.
* platform/qt-5.0-wk2/fast/forms/number/input-appearance-spinbutton-layer-expected.txt: Added.
* platform/qt-5.0-wk2/fast/forms/placeholder-position-expected.png: Added.
* platform/qt-5.0-wk2/fast/forms/placeholder-position-expected.txt: Added.
* platform/qt-5.0-wk2/fast/forms/placeholder-pseudo-style-expected.png:
* platform/qt-5.0-wk2/fast/forms/placeholder-pseudo-style-expected.txt:
* platform/qt-5.0-wk2/fast/forms/search-cancel-button-style-sharing-expected.txt: Added.
* platform/qt-5.0-wk2/fast/forms/search-display-none-cancel-button-expected.txt: Added.
* platform/qt-5.0-wk2/fast/forms/search-rtl-expected.png: Added.
* platform/qt-5.0-wk2/fast/forms/search-rtl-expected.txt:
* platform/qt-5.0-wk2/fast/forms/search-vertical-alignment-expected.png: Added.
* platform/qt-5.0-wk2/fast/forms/search-vertical-alignment-expected.txt: Added.
* platform/qt-5.0-wk2/fast/forms/select-background-none-expected.txt:
* platform/qt-5.0-wk2/fast/forms/select-baseline-expected.png:
* platform/qt-5.0-wk2/fast/forms/select-baseline-expected.txt:
* platform/qt-5.0-wk2/fast/forms/selectlist-minsize-expected.txt:
* platform/qt-5.0-wk2/fast/forms/text-style-color-expected.png: Added.
* platform/qt-5.0-wk2/fast/forms/text-style-color-expected.txt: Added.
* platform/qt-5.0-wk2/fast/forms/textarea-align-expected.png:
* platform/qt-5.0-wk2/fast/forms/textarea-align-expected.txt:
* platform/qt-5.0-wk2/fast/forms/textarea-placeholder-pseudo-style-expected.png: Added.
* platform/qt-5.0-wk2/fast/forms/textarea-placeholder-pseudo-style-expected.txt: Added.
* platform/qt-5.0-wk2/fast/forms/textarea-placeholder-visibility-1-expected.txt: Added.
* platform/qt-5.0-wk2/fast/forms/textarea-placeholder-visibility-2-expected.txt: Added.
* platform/qt-5.0-wk2/fast/forms/textarea-setinnerhtml-expected.txt:
* platform/qt-5.0-wk2/fast/forms/textfield-focus-ring-expected.txt: Added.
* platform/qt-5.0-wk2/fast/forms/textfield-outline-expected.txt: Added.
* platform/qt-5.0-wk2/fast/forms/textfield-overflow-expected.txt: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@123156 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years ago[EFL] [WK2] Add a callback to handle delete request in MiniBrowser
commit-queue@webkit.org [Thu, 19 Jul 2012 22:35:42 +0000 (22:35 +0000)]
[EFL] [WK2] Add a callback to handle delete request in MiniBrowser
https://bugs.webkit.org/show_bug.cgi?id=91483

Patch by Sudarsana Nagineni <sudarsana.nagineni@linux.intel.com> on 2012-07-19
Reviewed by Dirk Pranke.

Add a callback to handle delete request and quit the main loop when
window is asked to be deleted so that all the resources allocated
for the window will be freed properly.

* MiniBrowser/efl/main.c:
(closeWindow):
(browserCreate):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@123155 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years ago[Chromium] Textures drawn during occlusion are incorrectly re-used when unoccluded.
commit-queue@webkit.org [Thu, 19 Jul 2012 22:34:41 +0000 (22:34 +0000)]
[Chromium] Textures drawn during occlusion are incorrectly re-used when unoccluded.
https://bugs.webkit.org/show_bug.cgi?id=91537

Patch by Zeev Lieber <zlieber@chromium.org> on 2012-07-19
Reviewed by Adrienne Walker.

Source/WebCore:

Detecting if any external occlusion occurs when drawing a
texture. If external occlusion happens, the texture is marked as
"incomplete" in LayerRendererChromium, and is not being considered
for re-use.

Minor refactoring of CCQuadCuller - extracted interface
(CCQuadSink) and united append() and appendSurface() methods.

Added unit tests to verify this fix; also added test to try
clipping instead of occlusion.

* platform/graphics/chromium/LayerRendererChromium.cpp:
(LayerRendererChromium::CachedTexture):
(WebCore::LayerRendererChromium::CachedTexture::create):
(WebCore::LayerRendererChromium::CachedTexture::~CachedTexture):
(WebCore::LayerRendererChromium::CachedTexture::isComplete):
(WebCore::LayerRendererChromium::CachedTexture::setIsComplete):
(WebCore::LayerRendererChromium::CachedTexture::CachedTexture):
(WebCore):
(WebCore::LayerRendererChromium::decideRenderPassAllocationsForFrame):
(WebCore::LayerRendererChromium::haveCachedResourcesForRenderPassId):
(WebCore::LayerRendererChromium::drawRenderPassQuad):
* platform/graphics/chromium/LayerRendererChromium.h:
(LayerRendererChromium):
* platform/graphics/chromium/cc/CCIOSurfaceLayerImpl.cpp:
(WebCore::CCIOSurfaceLayerImpl::appendQuads):
* platform/graphics/chromium/cc/CCIOSurfaceLayerImpl.h:
(CCIOSurfaceLayerImpl):
* platform/graphics/chromium/cc/CCLayerImpl.cpp:
(WebCore::CCLayerImpl::appendDebugBorderQuad):
* platform/graphics/chromium/cc/CCLayerImpl.h:
(WebCore):
(WebCore::CCLayerImpl::appendQuads):
(CCLayerImpl):
* platform/graphics/chromium/cc/CCOcclusionTracker.cpp:
(WebCore::::unoccludedContentRect):
(WebCore::::unoccludedContributingSurfaceContentRect):
(WebCore):
* platform/graphics/chromium/cc/CCOcclusionTracker.h:
(CCOcclusionTrackerBase):
* platform/graphics/chromium/cc/CCQuadCuller.cpp:
(WebCore::CCQuadCuller::CCQuadCuller):
(WebCore::CCQuadCuller::append):
* platform/graphics/chromium/cc/CCQuadCuller.h:
(WebCore):
(CCQuadCuller):
(WebCore::CCQuadCuller::~CCQuadCuller):
(WebCore::CCQuadCuller::hasExternalOcclusion):
* platform/graphics/chromium/cc/CCQuadSink.h:
(WebKit):
(WebCore):
(CCQuadSink):
(WebCore::CCQuadSink::~CCQuadSink):
* platform/graphics/chromium/cc/CCRenderPass.cpp:
(WebCore::CCRenderPass::CCRenderPass):
(WebCore::CCRenderPass::appendQuadsForLayer):
(WebCore::CCRenderPass::appendQuadsForRenderSurfaceLayer):
* platform/graphics/chromium/cc/CCRenderPassDrawQuad.cpp:
(WebCore::CCRenderPassDrawQuad::create):
(WebCore::CCRenderPassDrawQuad::CCRenderPassDrawQuad):
* platform/graphics/chromium/cc/CCRenderPassDrawQuad.h:
(CCRenderPassDrawQuad):
(WebCore::CCRenderPassDrawQuad::hasExternalOcclusion):
* platform/graphics/chromium/cc/CCRenderSurface.cpp:
(WebCore::CCRenderSurface::CCRenderSurface):
(WebCore::CCRenderSurface::appendQuads):
* platform/graphics/chromium/cc/CCRenderSurface.h:
(WebCore):
(WebCore::CCRenderSurface::setHasExternalOcclusion):
(WebCore::CCRenderSurface::hasExternalOcclusion):
(CCRenderSurface):
* platform/graphics/chromium/cc/CCScrollbarLayerImpl.cpp:
(WebCore::CCScrollbarLayerImpl::appendQuads):
* platform/graphics/chromium/cc/CCScrollbarLayerImpl.h:
(CCScrollbarLayerImpl):
* platform/graphics/chromium/cc/CCSolidColorLayerImpl.cpp:
(WebCore::CCSolidColorLayerImpl::appendQuads):
* platform/graphics/chromium/cc/CCSolidColorLayerImpl.h:
(CCSolidColorLayerImpl):
* platform/graphics/chromium/cc/CCTextureLayerImpl.cpp:
(WebCore::CCTextureLayerImpl::appendQuads):
* platform/graphics/chromium/cc/CCTextureLayerImpl.h:
(CCTextureLayerImpl):
* platform/graphics/chromium/cc/CCTiledLayerImpl.cpp:
(WebCore::CCTiledLayerImpl::appendQuads):
* platform/graphics/chromium/cc/CCTiledLayerImpl.h:
(CCTiledLayerImpl):
* platform/graphics/chromium/cc/CCVideoLayerImpl.cpp:
(WebCore::CCVideoLayerImpl::appendQuads):
* platform/graphics/chromium/cc/CCVideoLayerImpl.h:
(CCVideoLayerImpl):

Source/WebKit/chromium:

Added unit tests to test texture caching with clipping and
occlusion. Removed some minor code duplication when initializing
layers for various scenarios.

MockCCQuadCuller can now implement CCQuadSink.

* tests/CCLayerTreeHostImplTest.cpp:
* tests/CCQuadCullerTest.cpp:
* tests/MockCCQuadCuller.h:
(WebCore::MockCCQuadCuller::MockCCQuadCuller):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@123154 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years ago[Qt] [Part 6] platform/qt/css1 tests needs update after rebaseline and new testfonts
commit-queue@webkit.org [Thu, 19 Jul 2012 22:18:54 +0000 (22:18 +0000)]
[Qt] [Part 6] platform/qt/css1 tests needs update after rebaseline and new testfonts
https://bugs.webkit.org/show_bug.cgi?id=91142

Batch of expected result files reviewed. It comprises
/LayoutTests/platform/qt/css1/text_properties directory

Patch by Luciano Wolf <luciano.wolf@openbossa.org> on 2012-07-19
Reviewed by Simon Hausmann.

* platform/qt-5.0-wk1/Skipped:
* platform/qt-5.0-wk2/Skipped:
* platform/qt-5.0/css1/text_properties/letter_spacing-expected.txt: Removed.
* platform/qt-5.0/css1/text_properties/text_transform-expected.txt: Removed.
* platform/qt/css1/text_properties/letter_spacing-expected.png:
* platform/qt/css1/text_properties/letter_spacing-expected.txt:
* platform/qt/css1/text_properties/line_height-expected.png:
* platform/qt/css1/text_properties/line_height-expected.txt:
* platform/qt/css1/text_properties/text_align-expected.png:
* platform/qt/css1/text_properties/text_align-expected.txt:
* platform/qt/css1/text_properties/text_decoration-expected.png:
* platform/qt/css1/text_properties/text_decoration-expected.txt:
* platform/qt/css1/text_properties/text_indent-expected.png:
* platform/qt/css1/text_properties/text_indent-expected.txt:
* platform/qt/css1/text_properties/text_transform-expected.png:
* platform/qt/css1/text_properties/vertical_align-expected.png:
* platform/qt/css1/text_properties/vertical_align-expected.txt:
* platform/qt/css1/text_properties/word_spacing-expected.png:
* platform/qt/css1/text_properties/word_spacing-expected.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@123153 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years ago[chromium] Unreviewed gardening. Rebaseline some tests after r115846.
tony@chromium.org [Thu, 19 Jul 2012 22:11:20 +0000 (22:11 +0000)]
[chromium] Unreviewed gardening.  Rebaseline some tests after r115846.

* platform/chromium-mac-snowleopard/fast/repaint/inline-relative-positioned-expected.png: Removed.
* platform/chromium-mac-snowleopard/fast/repaint/lines-with-layout-delta-expected.png:
* platform/chromium-mac-snowleopard/fast/repaint/overflow-clip-subtree-layout-expected.png: Removed.
* platform/chromium-mac-snowleopard/fast/repaint/repaint-resized-overflow-expected.png:
* platform/chromium-mac-snowleopard/fast/repaint/subtree-layoutstate-transform-expected.png:
* platform/chromium-mac-snowleopard/fast/repaint/subtree-root-clip-2-expected.png: Removed.
* platform/chromium-mac-snowleopard/fast/repaint/subtree-root-clip-expected.png: Removed.
* platform/chromium/TestExpectations:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@123152 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoKeep overlap testing on for descendants of an animating element
simon.fraser@apple.com [Thu, 19 Jul 2012 22:02:47 +0000 (22:02 +0000)]
Keep overlap testing on for descendants of an animating element
https://bugs.webkit.org/show_bug.cgi?id=91787

Reviewed by Beth Dakin.

Source/WebCore:

RenderLayerCompositor::computeCompositingRequirements() would turn off
overlap testing for descendants of a layer that has a 3D transform,
or has an animating transform. This could cause compositing changes
inside an animating element when transitions start and stop, which
is undesirable.

Instead, we can keep overlap testing on for descendants of that
animating element, but we have to turn it off for layers later
in z-order.

Also enhance the "IndirectCompositingReason" flags so that in
the compositing log output, we can distinguish between layers that
were composited as a result of overlap testing from those compositing
for stacking reasons.

Test: compositing/layer-creation/animation-overlap-with-children.html

* rendering/RenderLayer.h: New IndirectCompositingReason value to
distinguish overlap from stacking.
* rendering/RenderLayerCompositor.cpp:
(WebCore::RenderLayerCompositor::computeCompositingRequirements):
Use an IndirectCompositingReason enum rather than a bool to track whether
the layer starts as composited.
Remove the if (layer->has3DTransform() || isRunningAcceleratedTransformAnimation(layer->renderer()))
check that turned off overlap testing for descendants of this layer.
Use IndirectCompositingForStacking (somewhat arbitrarily) for reflection layers.
Turn overlap testing off for layers later in the z-order.
(WebCore::RenderLayerCompositor::requiresOwnBackingStore): Take IndirectCompositingForStacking into
account.
(WebCore::RenderLayerCompositor::reasonForCompositing): Improve the logging output.

LayoutTests:

Test with two layers inside an animating container; one should be composited
because of overlap, the other should not be.

* compositing/layer-creation/animation-overlap-with-children-expected.txt: Added.
* compositing/layer-creation/animation-overlap-with-children.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@123151 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years ago[Qt] Enable CSS shaders in Qt (software mode)
noam.rosenthal@nokia.com [Thu, 19 Jul 2012 21:54:19 +0000 (21:54 +0000)]
[Qt] Enable CSS shaders in Qt (software mode)
https://bugs.webkit.org/show_bug.cgi?id=85140

Reviewed by Simon Hausmann.

Source/WebCore:

Added missing files to the Qt build.

No new tests, WebGL is still disabled by default in the bots.

* Target.pri:

Source/WebKit/qt:

Enable CSS Shaders in settings when WebGL is enabled.

* Api/qwebsettings.cpp:
(QWebSettingsPrivate::apply):

Source/WebKit2:

Add CSSCustomFilterEnabled to QWebPreferences. It's disabled by default.

* UIProcess/API/qt/qwebpreferences.cpp:
(QWebPreferencesPrivate::testAttribute):
(QWebPreferencesPrivate::setAttribute):
* UIProcess/API/qt/qwebpreferences_p_p.h:

Tools:

Enable CSS_SHADERS when WEBGL and CSS_FILTERS are enabled.

* qmake/mkspecs/features/features.prf:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@123150 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years ago[Chromium] Out of Memory is observed when a large object is passed to a Web Worker.
dimich@chromium.org [Thu, 19 Jul 2012 21:40:37 +0000 (21:40 +0000)]
[Chromium] Out of Memory is observed when a large object is passed to a Web Worker.
https://bugs.webkit.org/show_bug.cgi?id=91535.

Reviewed by Dave Levin.

No new tests because of no new behavior and lack of mechanism to monitor used memory.

* bindings/v8/SerializedScriptValue.cpp:
(WebCore::SerializedScriptValue::SerializedScriptValue):
(WebCore::SerializedScriptValue::registerMemoryAllocatedWithCurrentScriptContext):
(WebCore):
(WebCore::SerializedScriptValue::~SerializedScriptValue):
* bindings/v8/SerializedScriptValue.h:
(SerializedScriptValue):
Added a method to SerializedScriptValue to add memory pressure on GC proportional to the size of memory allocated in serialized data.

* dom/MessageEvent.cpp:
(WebCore::MessageEvent::MessageEvent):
(WebCore::MessageEvent::initMessageEvent):
Call the new method of SerializedScriptValue during creation of MessageEvent (presumably in the thread which contains the right V8 context).

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@123149 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years ago[Qt] [Part 5] platform/qt/css1 tests needs update after rebaseline and new testfonts
commit-queue@webkit.org [Thu, 19 Jul 2012 20:55:02 +0000 (20:55 +0000)]
[Qt] [Part 5] platform/qt/css1 tests needs update after rebaseline and new testfonts
https://bugs.webkit.org/show_bug.cgi?id=91140

Batch of expected result files reviewed. It comprises
/LayoutTests/platform/qt/css1/[pseudo, units] directories

Patch by Luciano Wolf <luciano.wolf@openbossa.org> on 2012-07-19
Reviewed by Simon Hausmann.

* platform/qt-5.0-wk1/Skipped:
* platform/qt-5.0-wk2/Skipped:
* platform/qt/css1/pseudo/anchor-expected.png:
* platform/qt/css1/pseudo/anchor-expected.txt:
* platform/qt/css1/pseudo/firstletter-expected.png:
* platform/qt/css1/pseudo/firstletter-expected.txt:
* platform/qt/css1/pseudo/firstline-expected.png:
* platform/qt/css1/pseudo/firstline-expected.txt:
* platform/qt/css1/pseudo/multiple_pseudo_elements-expected.png:
* platform/qt/css1/pseudo/multiple_pseudo_elements-expected.txt:
* platform/qt/css1/pseudo/pseudo_elements_in_selectors-expected.png:
* platform/qt/css1/pseudo/pseudo_elements_in_selectors-expected.txt:
* platform/qt/css1/units/color_units-expected.png:
* platform/qt/css1/units/color_units-expected.txt:
* platform/qt/css1/units/length_units-expected.png:
* platform/qt/css1/units/length_units-expected.txt:
* platform/qt/css1/units/percentage_units-expected.png:
* platform/qt/css1/units/percentage_units-expected.txt:
* platform/qt/css1/units/urls-expected.png:
* platform/qt/css1/units/urls-expected.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@123148 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoBug fixes and enhancements for OfflineASM annotation system.
commit-queue@webkit.org [Thu, 19 Jul 2012 20:53:22 +0000 (20:53 +0000)]
Bug fixes and enhancements for OfflineASM annotation system.
https://bugs.webkit.org/show_bug.cgi?id=91690

Patch by Mark Lam <mark.lam@apple.com> on 2012-07-19
Reviewed by Filip Pizlo.

* offlineasm/armv7.rb: added default handling of Instruction lower().
* offlineasm/asm.rb: added more support for annotations and more pretty printing.
* offlineasm/ast.rb: added more support for annotations.
* offlineasm/config.rb: added $preferredCommentStartColumn, simplified $enableInstrAnnotations.
* offlineasm/parser.rb: added more support for annotations.
* offlineasm/transform.rb: added more support for annotations.
* offlineasm/x86.rb: added default handling of Instruction lower().

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@123147 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoIn flipped blocks writing modes, no flipping occurs when mapping RenderText’s local...
mitz@apple.com [Thu, 19 Jul 2012 20:46:03 +0000 (20:46 +0000)]
In flipped blocks writing modes, no flipping occurs when mapping RenderText’s local coordinates to absolute
https://bugs.webkit.org/show_bug.cgi?id=91780

Reviewed by Anders Carlsson.

Source/WebCore:

Test: fast/writing-mode/flipped-blocks-text-map-local-to-container.html

When RenderObject::mapLocalToContainer() was called on a RenderText with ApplyContainerFlip,
it would not flip (if the container was not a box) but it would always pass
DoNotApplyContainerFlip when recurring to the parent. This meant that no one applied the flip.

* rendering/RenderInline.cpp:
(WebCore::RenderInline::mapLocalToContainer): Made the setting of applyContainerFlip to
false unconditional on the container actually being flipped.
* rendering/RenderObject.cpp:
(WebCore::RenderObject::mapLocalToContainer): Rather than always passing
DoNotApplyContainerFlip when recurring to the parent, changed this function to pass through
the value of applyContainerFlip it was called with, unless it applied the flip itself.

LayoutTests:

* fast/writing-mode/flipped-blocks-text-map-local-to-container-expected.txt: Added.
* fast/writing-mode/flipped-blocks-text-map-local-to-container.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@123146 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoWindow top should not be replaceable
arv@chromium.org [Thu, 19 Jul 2012 20:34:56 +0000 (20:34 +0000)]
Window top should not be replaceable
https://bugs.webkit.org/show_bug.cgi?id=91755

Reviewed by Adam Barth.

Window top is readonly and unforgable and should not be replaceable.

http://www.whatwg.org/specs/web-apps/current-work/multipage/browsers.html#the-window-object

This change brings JSC inline with all other browsers.

Source/WebCore:

No new tests. Modified existing tests.

* page/DOMWindow.idl:

LayoutTests:

* fast/dom/Window/get-set-properties-expected.txt:
* fast/dom/Window/get-set-properties.html:
* fast/dom/Window/window-property-shadowing-expected.txt:
* fast/dom/Window/window-property-shadowing.html:
* platform/chromium/fast/dom/Window/get-set-properties-expected.txt: Removed.
* platform/chromium/fast/dom/Window/window-property-shadowing-expected.txt: Removed.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@123145 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years ago[Qt] Added/updated missing fast/text results after rebaseline
commit-queue@webkit.org [Thu, 19 Jul 2012 20:20:31 +0000 (20:20 +0000)]
[Qt] Added/updated missing fast/text results after rebaseline
https://bugs.webkit.org/show_bug.cgi?id=91570

Patch by Bruno de Oliveira Abinader <bruno.abinader@basyskom.com> on 2012-07-19
Reviewed by Noam Rosenthal.

The following tests were failing after r122971:
  fast/text/international/unicode-bidi-plaintext-in-textarea.html = TEXT
  fast/text/textIteratorNilRenderer.html = TEXT
  fast/text/whitespace/030.html = TEXT

This patch includes/updates the missing expectations.

* platform/qt-5.0-wk2/fast/text/international/unicode-bidi-plaintext-in-textarea-expected.png: Added.
* platform/qt-5.0-wk2/fast/text/international/unicode-bidi-plaintext-in-textarea-expected.txt: Added.
* platform/qt-5.0-wk2/fast/text/textIteratorNilRenderer-expected.png: Added.
* platform/qt-5.0-wk2/fast/text/textIteratorNilRenderer-expected.txt: Added.
* platform/qt-5.0/fast/text/international/unicode-bidi-plaintext-in-textarea-expected.png: Added.
* platform/qt/fast/text/whitespace/030-expected.png:
* platform/qt/fast/text/whitespace/030-expected.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@123144 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agotest-webkitpy: fix executive_unittest nits, including a missing 'errno' import
dpranke@chromium.org [Thu, 19 Jul 2012 20:19:43 +0000 (20:19 +0000)]
test-webkitpy: fix executive_unittest nits, including a missing 'errno' import
https://bugs.webkit.org/show_bug.cgi?id=91770

Reviewed by Adam Barth.

The missing errno import is causing a flaky failure when run in
parallel on the bots.

* Scripts/webkitpy/common/system/executive_unittest.py:
(ExecutiveTest.assert_interpreter_for_content):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@123143 268f45cc-cd09-0410-ab3c-d52691b4dbfc