WebKit-https.git
9 years agoPerf-o-matic should show min/max/stdev on graphs
rniwa@webkit.org [Tue, 7 Feb 2012 02:21:09 +0000 (02:21 +0000)]
Perf-o-matic should show min/max/stdev on graphs
https://bugs.webkit.org/show_bug.cgi?id=77702

Reviewed by Andreas Kling.

Include min/max/stdev in test runs when available to be shown on the perf-o-matic.

Also merge a change to config.js and increment the version number by 1 (unreviewed).

* Websites/webkit-perf.appspot.com/app.yaml:
* Websites/webkit-perf.appspot.com/js/config.js:
* Websites/webkit-perf.appspot.com/runs_handler.py:
(RunsHandler.get):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@106890 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoUnreviewed, rolling out r106883.
haraken@chromium.org [Tue, 7 Feb 2012 02:04:09 +0000 (02:04 +0000)]
Unreviewed, rolling out r106883.
http://trac.webkit.org/changeset/106883
https://bugs.webkit.org/show_bug.cgi?id=77852

build failure around Chromium V8 bindings

* bindings/scripts/CodeGenerator.pm:
(ShouldCheckEnums):
(GenerateCompileTimeCheckForEnumsIfNeeded):
* bindings/scripts/CodeGeneratorJS.pm:
(GenerateHeader):
(GenerateImplementation):
(NativeToJSValue):
* bindings/scripts/CodeGeneratorV8.pm:
(GenerateSingleBatchedAttribute):
(GenerateFunctionCallString):
(NativeToJSValue):
* bindings/scripts/test/JS/JSTestInterface.cpp:
(WebCore):
* bindings/scripts/test/JS/JSTestObj.cpp:
(WebCore):
* bindings/scripts/test/TestObj.idl:
* bindings/scripts/test/TestTypedArray.idl:
* bindings/scripts/test/V8/V8TestInterface.cpp:
(WebCore):
* bindings/scripts/test/V8/V8TestObj.cpp:
(WebCore):
* css/WebKitCSSFilterValue.idl:
* css/WebKitCSSTransformValue.idl:
* dom/DOMCoreException.idl:
* dom/Document.idl:
* dom/EventException.idl:
* dom/RangeException.idl:
* fileapi/FileException.idl:
* fileapi/OperationNotAllowedException.idl:
* html/HTMLFormElement.idl:
* html/canvas/Float32Array.idl:
* html/canvas/Float64Array.idl:
* html/canvas/Int16Array.idl:
* html/canvas/Int32Array.idl:
* html/canvas/Int8Array.idl:
* html/canvas/OESStandardDerivatives.idl:
* html/canvas/OESVertexArrayObject.idl:
* html/canvas/Uint16Array.idl:
* html/canvas/Uint32Array.idl:
* html/canvas/Uint8Array.idl:
* html/canvas/Uint8ClampedArray.idl:
* html/canvas/WebGLCompressedTextures.idl:
* html/canvas/WebGLDebugRendererInfo.idl:
* html/canvas/WebGLRenderingContext.idl:
* inspector/JavaScriptCallFrame.idl:
* loader/appcache/DOMApplicationCache.idl:
* page/Console.idl:
* page/DOMWindow.idl:
* page/Location.idl:
* storage/IDBCursor.idl:
* storage/IDBDatabaseException.idl:
* storage/IDBObjectStore.idl:
* storage/SQLException.idl:
* svg/SVGComponentTransferFunctionElement.idl:
* svg/SVGException.idl:
* svg/SVGFEBlendElement.idl:
* svg/SVGFEColorMatrixElement.idl:
* svg/SVGFECompositeElement.idl:
* svg/SVGFEConvolveMatrixElement.idl:
* svg/SVGFEDisplacementMapElement.idl:
* svg/SVGFEMorphologyElement.idl:
* svg/SVGFETurbulenceElement.idl:
* svg/SVGGradientElement.idl:
* xml/XMLHttpRequestException.idl:
* xml/XPathException.idl:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@106889 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoWebGL must allocate smaller drawing buffer when the allocation fails.
commit-queue@webkit.org [Tue, 7 Feb 2012 01:58:16 +0000 (01:58 +0000)]
WebGL must allocate smaller drawing buffer when the allocation fails.
https://bugs.webkit.org/show_bug.cgi?id=76654

Patch by Yongsheng Zhu <yongsheng.zhu@intel.com> on 2012-02-06
Reviewed by Kenneth Russell.

Test: fast/canvas/webgl/drawingbuffer-test.html

* platform/graphics/gpu/DrawingBuffer.cpp:
(WebCore):
(WebCore::DrawingBuffer::create):
(WebCore::DrawingBuffer::reset):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@106888 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoUnreviewed gardening.
jchaffraix@webkit.org [Tue, 7 Feb 2012 01:42:01 +0000 (01:42 +0000)]
Unreviewed gardening.

* DEPS: Rolling the DEPS to get the apple_webkit update
and fix the Chromium build.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@106887 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoPort some indexeddb layout tests to workers.
dgrogan@chromium.org [Tue, 7 Feb 2012 01:39:09 +0000 (01:39 +0000)]
Port some indexeddb layout tests to workers.
https://bugs.webkit.org/show_bug.cgi?id=77805

Reviewed by Tony Chang.

* storage/indexeddb/index-basics-workers-expected.txt: Added.
* storage/indexeddb/index-basics-workers.html: Added.
* storage/indexeddb/index-basics.html:
* storage/indexeddb/objectstore-basics-expected.txt:
* storage/indexeddb/objectstore-basics-workers-expected.txt: Added.
* storage/indexeddb/objectstore-basics-workers.html: Added.
* storage/indexeddb/objectstore-basics.html:
* storage/indexeddb/resources/index-basics.js: Copied from LayoutTests/storage/indexeddb/index-basics.html.
(test):
(setVersion):
(deleteExisting):
(addData):
(addData2):
(addData3):
(getData):
(getObjectData):
(getObjectData2):
(getObjectData3):
(getDataFail):
(getObjectDataFail):
(getObjectData4):
(openKeyCursor):
(cursor1Continue):
(cursor1Continue2):
(cursor1Continue3):
(openObjectCursor):
(cursor2Continue):
(cursor2Continue2):
(cursor2Continue3):
(last.try.get testFailed):
(last):
* storage/indexeddb/resources/objectstore-basics.js: Copied from LayoutTests/storage/indexeddb/objectstore-basics.html.
(test):
(openSuccess):
(setVersionSuccess):
(createIndex):
(testSetVersionAbort):
(createAnotherIndex):
(checkMetadata):
(addData):
(addDateSuccess):
(addSuccess):
(addAgainFailure):
(getSuccess):
(getSuccessDateKey):
(removeSuccess):
(removeSuccessButFalse):
(testPreConditions.request.onsuccess):
(testPreConditions):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@106886 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoUnreviewed. Rebaselined run-bindings-tests results.
haraken@chromium.org [Tue, 7 Feb 2012 01:37:43 +0000 (01:37 +0000)]
Unreviewed. Rebaselined run-bindings-tests results.

* bindings/scripts/test/V8/V8TestObj.cpp:
(WebCore::TestObjInternal::optionsObjectCallback):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@106885 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoCorrect usage of LayoutUnits in dom, editing, page, accessibility, inspector, and...
leviw@chromium.org [Tue, 7 Feb 2012 01:36:25 +0000 (01:36 +0000)]
Correct usage of LayoutUnits in dom, editing, page, accessibility, inspector, and loader
https://bugs.webkit.org/show_bug.cgi?id=77891

Reviewed by Darin Adler.

Updating the usage of LayoutUnits in the dom, editing, page, accessibility, inspector, and loader
directories to properly use LayoutUnits vs ints.

No new tests. No new behavior.

* accessibility/AccessibilityObject.h:
(WebCore::AccessibilityObject::boundsForVisiblePositionRange):
(WebCore::AccessibilityObject::doAXBoundsForRange):
* accessibility/AccessibilityRenderObject.cpp:
(WebCore::AccessibilityRenderObject::isOffScreen):
(WebCore::AccessibilityRenderObject::clickPoint):
(WebCore::AccessibilityRenderObject::boundsForVisiblePositionRange):
(WebCore::AccessibilityRenderObject::doAXBoundsForRange):
(WebCore::AccessibilityRenderObject::accessibilityImageMapHitTest):
* accessibility/AccessibilityRenderObject.h:
(AccessibilityRenderObject):
* accessibility/AccessibilitySlider.cpp:
(WebCore::AccessibilitySlider::elementAccessibilityHitTest):
* accessibility/AccessibilitySlider.h:
(AccessibilitySlider):
* accessibility/mac/WebAccessibilityObjectWrapper.mm:
(-[WebAccessibilityObjectWrapper accessibilityShowContextMenu]):
* css/CSSComputedStyleDeclaration.cpp:
(WebCore::computedTransform):
(WebCore::CSSComputedStyleDeclaration::getPropertyCSSValue):
* dom/Clipboard.h:
(WebCore::Clipboard::dragLocation):
(Clipboard):
* dom/DocumentMarkerController.cpp:
(WebCore::DocumentMarkerController::renderedRectsForMarkers):
* dom/DocumentMarkerController.h:
* dom/Element.cpp:
(WebCore::Element::screenRect):
* dom/Element.h:
(Element):
* editing/Editor.cpp:
(WebCore::Editor::rangeForPoint):
(WebCore::Editor::firstRectForRange):
* editing/RenderedPosition.cpp:
(WebCore::RenderedPosition::absoluteRect):
* editing/RenderedPosition.h:
(WebCore::RenderedPosition::absoluteRect):
(RenderedPosition):
* editing/VisiblePosition.h:
(VisiblePosition):
* inspector/DOMNodeHighlighter.cpp:
* loader/SubframeLoader.cpp:
(WebCore::SubframeLoader::createJavaAppletWidget):
* loader/SubframeLoader.h:
(SubframeLoader):
* page/EventHandler.cpp:
(WebCore::EventHandler::handleMousePressEvent):
* page/Frame.cpp:
(WebCore::Frame::nodeImage):
* page/win/FrameCGWin.cpp:
(WebCore::Frame::nodeImage):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@106884 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoRename [DontCheckEnums], [ReturnsNew], [DoNotCheckDomainSecurityOnGet],
haraken@chromium.org [Tue, 7 Feb 2012 01:33:27 +0000 (01:33 +0000)]
Rename [DontCheckEnums], [ReturnsNew], [DoNotCheckDomainSecurityOnGet],
[DoNotCheckDomainSecurityOnSet] and [ImplementationFunction] IDLs
https://bugs.webkit.org/show_bug.cgi?id=77852

Reviewed by Adam Barth.

This patch renames [DontCheckEnums], [ReturnsNew], [DoNotCheckDomainSecurityOnGet],
[DoNotCheckDomainSecurityOnSet] and [ImplementationFunction] IDLs for clarification.

[DontCheckEnums] => [DoNotCheckConstants]  (This IDL inserts assertions to check if a
constant value is equal to the expected constant value)
[ReturnsNew] => [ReturnNewObject] (For clarification)
[DoNotCheckDomainSecurityOnGet] => [DoNotCheckDomainSecurityOnGetter] (For naming consistency
with other [*Getter] IDLs)
[DoNotCheckDomainSecurityOnSet] => [DoNotCheckDomainSecurityOnSetter] (For naming consistency
with other [*Setter] IDLs)
[ImplementationFunction=] => [ImplementedAs=] (For clarification. This IDL specifies a method
name in implementation)

No tests. No change in behavior.

* bindings/scripts/CodeGenerator.pm:
(ShouldCheckEnums):
(GenerateCompileTimeCheckForEnumsIfNeeded):
* bindings/scripts/CodeGeneratorJS.pm:
(GenerateHeader):
(GenerateImplementation):
(NativeToJSValue):
* bindings/scripts/CodeGeneratorV8.pm:
(GenerateSingleBatchedAttribute):
(GenerateFunctionCallString):
(NativeToJSValue):

* bindings/scripts/test/TestObj.idl:
* bindings/scripts/test/TestTypedArray.idl:

* bindings/scripts/test/JS/JSTestInterface.cpp: Updated run-bindings-tests results.
(WebCore):
* bindings/scripts/test/JS/JSTestObj.cpp: Ditto.
(WebCore):
* bindings/scripts/test/V8/V8TestInterface.cpp: Ditto.
(WebCore):
* bindings/scripts/test/V8/V8TestObj.cpp: Ditto.
(WebCore):

* bindings/scripts/test/TestObj.idl:
* bindings/scripts/test/TestTypedArray.idl:
* css/WebKitCSSFilterValue.idl:
* css/WebKitCSSTransformValue.idl:
* dom/DOMCoreException.idl:
* dom/Document.idl:
* dom/EventException.idl:
* dom/RangeException.idl:
* fileapi/FileException.idl:
* fileapi/OperationNotAllowedException.idl:
* html/HTMLFormElement.idl:
* html/canvas/Float32Array.idl:
* html/canvas/Float64Array.idl:
* html/canvas/Int16Array.idl:
* html/canvas/Int32Array.idl:
* html/canvas/Int8Array.idl:
* html/canvas/OESStandardDerivatives.idl:
* html/canvas/OESVertexArrayObject.idl:
* html/canvas/Uint16Array.idl:
* html/canvas/Uint32Array.idl:
* html/canvas/Uint8Array.idl:
* html/canvas/Uint8ClampedArray.idl:
* html/canvas/WebGLCompressedTextures.idl:
* html/canvas/WebGLDebugRendererInfo.idl:
* html/canvas/WebGLRenderingContext.idl:
* inspector/JavaScriptCallFrame.idl:
* loader/appcache/DOMApplicationCache.idl:
* page/Console.idl:
* page/DOMWindow.idl:
* page/Location.idl:
* storage/IDBCursor.idl:
* storage/IDBDatabaseException.idl:
* storage/IDBObjectStore.idl:
* storage/SQLException.idl:
* svg/SVGComponentTransferFunctionElement.idl:
* svg/SVGException.idl:
* svg/SVGFEBlendElement.idl:
* svg/SVGFEColorMatrixElement.idl:
* svg/SVGFECompositeElement.idl:
* svg/SVGFEConvolveMatrixElement.idl:
* svg/SVGFEDisplacementMapElement.idl:
* svg/SVGFEMorphologyElement.idl:
* svg/SVGFETurbulenceElement.idl:
* svg/SVGGradientElement.idl:
* xml/XMLHttpRequestException.idl:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@106883 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoRenderSVGShape::strokeContains will fail for some strokes
commit-queue@webkit.org [Tue, 7 Feb 2012 01:30:50 +0000 (01:30 +0000)]
RenderSVGShape::strokeContains will fail for some strokes
https://bugs.webkit.org/show_bug.cgi?id=76931

Patch by Stephen Chenney <schenney@chromium.org> on 2012-02-06
Reviewed by Darin Adler.

Source/WebCore:

Adding support for rounded zero-length endcaps for SVG path hit
testing.

Tests: svg/hittest/zero-length-butt-cap-path.xhtml
       svg/hittest/zero-length-round-cap-path.xhtml
       svg/hittest/zero-length-square-cap-path.xhtml

* rendering/svg/RenderSVGShape.cpp:
(WebCore::RenderSVGShape::strokeContains):

LayoutTests:

New tests for hit testing against an SVG path of zero length with
endcaps.

* svg/hittest/zero-length-butt-cap-path-expected.txt: Added.
* svg/hittest/zero-length-butt-cap-path.xhtml: Added.
* svg/hittest/zero-length-round-cap-path-expected.txt: Added.
* svg/hittest/zero-length-round-cap-path.xhtml: Added.
* svg/hittest/zero-length-square-cap-path-expected.txt: Added.
* svg/hittest/zero-length-square-cap-path.xhtml: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@106882 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoAdd webGLErrorsToConsoleEnabled Settings flag
commit-queue@webkit.org [Tue, 7 Feb 2012 01:09:43 +0000 (01:09 +0000)]
Add webGLErrorsToConsoleEnabled Settings flag
https://bugs.webkit.org/show_bug.cgi?id=77696

Patch by Gregg Tavares <gman@google.com> on 2012-02-06
Reviewed by Kenneth Russell.

No new tests as there is no change in behavior.

* html/canvas/WebGLRenderingContext.cpp:
(WebCore):
(WebCore::WebGLRenderingContext::setupFlags):
* page/Settings.cpp:
(WebCore::Settings::Settings):
* page/Settings.h:
(Settings):
(WebCore::Settings::webGLErrorsToConsoleEnabled):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@106881 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoConvert PopupMenuClient::clientPadding over to new layout abstraction
eae@chromium.org [Tue, 7 Feb 2012 01:07:59 +0000 (01:07 +0000)]
Convert PopupMenuClient::clientPadding over to new layout abstraction
https://bugs.webkit.org/show_bug.cgi?id=77798

Reviewed by Eric Seidel.

Change PopupMenuClient and all classes inheriting from it to use the new
layout types for padding.

Source/WebCore:

No new tests.

* platform/PopupMenuClient.h:
(PopupMenuClient):
* rendering/RenderMenuList.h:
* rendering/RenderTextControlSingleLine.cpp:
(WebCore::RenderTextControlSingleLine::clientPaddingLeft):
(WebCore::RenderTextControlSingleLine::clientPaddingRight):
* rendering/RenderTextControlSingleLine.h:

Source/WebKit/chromium:

* src/AutofillPopupMenuClient.cpp:
(WebKit::AutofillPopupMenuClient::clientPaddingLeft):
(WebKit::AutofillPopupMenuClient::clientPaddingRight):
* src/AutofillPopupMenuClient.h:
(AutofillPopupMenuClient):
* tests/PopupMenuTest.cpp:
(WebKit::TestPopupMenuClient::clientPaddingLeft):
(WebKit::TestPopupMenuClient::clientPaddingRight):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@106880 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoValueToNumber and ValueToDouble nodes don't do anything and should be removed
fpizlo@apple.com [Tue, 7 Feb 2012 01:07:32 +0000 (01:07 +0000)]
ValueToNumber and ValueToDouble nodes don't do anything and should be removed
https://bugs.webkit.org/show_bug.cgi?id=77855
<rdar://problem/10811325>

Reviewed by Gavin Barraclough.

Removed ValueToNumber and ValueToDouble, because the only thing they were doing
was wasting registers.

This looks like a 1% win on V8 (with a 5% win on crypto) and a 2-3% win on Kraken,
mostly due to a >10% win on gaussian-blur. No win anywhere else.

* dfg/DFGAbstractState.cpp:
(JSC::DFG::AbstractState::execute):
* dfg/DFGByteCodeParser.cpp:
(JSC::DFG::ByteCodeParser::getToInt32):
(ByteCodeParser):
(JSC::DFG::ByteCodeParser::handleMinMax):
(JSC::DFG::ByteCodeParser::handleIntrinsic):
(JSC::DFG::ByteCodeParser::parseBlock):
* dfg/DFGNode.h:
(DFG):
(JSC::DFG::Node::hasArithNodeFlags):
* dfg/DFGPropagator.cpp:
(JSC::DFG::Propagator::propagateArithNodeFlags):
(JSC::DFG::Propagator::propagateNodePredictions):
(JSC::DFG::Propagator::vote):
(JSC::DFG::Propagator::doRoundOfDoubleVoting):
(Propagator):
(JSC::DFG::Propagator::fixupNode):
(JSC::DFG::Propagator::canonicalize):
* dfg/DFGSpeculativeJIT.cpp:
(JSC::DFG::SpeculativeJIT::computeValueRecoveryFor):
* dfg/DFGSpeculativeJIT32_64.cpp:
(JSC::DFG::SpeculativeJIT::compile):
* dfg/DFGSpeculativeJIT64.cpp:
(JSC::DFG::SpeculativeJIT::compile):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@106879 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoFix color animations by value
commit-queue@webkit.org [Tue, 7 Feb 2012 00:55:15 +0000 (00:55 +0000)]
Fix color animations by value
https://bugs.webkit.org/show_bug.cgi?id=77812

Patch by Philip Rogers <pdr@google.com> on 2012-02-06
Reviewed by Darin Adler.

Source/WebCore:

Test: svg/animations/animate-color-fill-from-by.html

* svg/SVGAnimatedColor.cpp:
(WebCore::SVGAnimatedColorAnimator::calculateFromAndByValues):

LayoutTests:

* svg/animations/animate-color-fill-from-by-expected.txt: Added.
* svg/animations/animate-color-fill-from-by.html: Added.
* svg/animations/script-tests/animate-color-fill-from-by.js: Added.
(checkFillColor):
(sample1):
(sample2):
(sample3):
(executeTest):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@106878 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoDo not try to chdir to the directory where the executable is in
kov@webkit.org [Tue, 7 Feb 2012 00:53:23 +0000 (00:53 +0000)]
Do not try to chdir to the directory where the executable is in
only if the path is absolute. This makes tests runnable from
anywhere.

Reviewed by Martin Robinson.

* tests/test_utils.c:
(testutils_relative_chdir):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@106877 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoREGRESSION (r104303): Clicking inline video controls gives the video a focus ring
jer.noble@apple.com [Tue, 7 Feb 2012 00:52:02 +0000 (00:52 +0000)]
REGRESSION (r104303): Clicking inline video controls gives the video a focus ring
https://bugs.webkit.org/show_bug.cgi?id=77288

Reviewed by Darin Adler.

Source/WebCore:

Test: media/video-mouse-focus.html

Do not allow media elements to be focused by mouse click by overriding isMouseFocusable
and returning false.

* html/HTMLMediaElement.cpp:
(WebCore::HTMLMediaElement::isMouseFocusable):
* html/HTMLMediaElement.h:

LayoutTests:

* media/video-mouse-focus-expected.txt: Added.
* media/video-mouse-focus.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@106876 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoSource/WebCore: Overlay scrollbars flash when window is simply activated
andersca@apple.com [Tue, 7 Feb 2012 00:43:26 +0000 (00:43 +0000)]
Source/WebCore: Overlay scrollbars flash when window is simply activated
https://bugs.webkit.org/show_bug.cgi?id=77911
<rdar://problem/10211995>

Reviewed by Kenneth Russell.

Add a new member function, FocusController::setContainingWindowIsVisible, and move the code
that calls ScrollableArea::contentAreaDidShow/ScrollableArea::contentAreaDidHide there, since
we only want to flash scrollers when the window becomes visible.

* WebCore.exp.in:
* page/FocusController.cpp:
(WebCore::FocusController::FocusController):
(WebCore::FocusController::setActive):
(WebCore::FocusController::setContainingWindowIsVisible):
(WebCore):
* page/FocusController.h:
(FocusController):
(WebCore::FocusController::containingWindowIsVisible):

* platform/mac/ScrollAnimatorMac.mm:
(-[WebScrollbarPainterControllerDelegate scrollerImpPair:setContentAreaNeedsDisplayInRect:]):
Call ScrollAnimatorMac::contentAreaWillPaint here, since that will trigger AppKit to flash the scrollers.

Source/WebKit/mac: Overlay scrollbars flash when window is simply activated
https://bugs.webkit.org/show_bug.cgi?id=77911
<rdar://problem/10211995>

Reviewed by Kenneth Russell.

* WebView/WebView.mm:
(-[WebView _windowWillOrderOnScreen:]):
(-[WebView _windowWillOrderOffScreen:]):
Call FocusController::setContainingWindowIsVisible.

Source/WebKit2: Overlay scrollbars flash when window is simply activated
https://bugs.webkit.org/show_bug.cgi?id=77911

Reviewed by Kenneth Russell.

* UIProcess/API/mac/WKView.mm:
(-[WKView _updateWindowVisibility]):
Use -[NSWindow isVisible] here, since we also want to consider the window hidden if the application itself is hidden.

(-[WKView _windowDidOrderOffScreen:]):
(-[WKView _windowDidOrderOnScreen:]):
Call -[WKView updateWindowVisibility].

* WebProcess/WebPage/WebPage.cpp:
(WebKit::WebPage::setWindowIsVisible):
Call FocusController::setContainingWindowIsVisible.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@106875 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agopostMessage leaks MemoryEvent object
commit-queue@webkit.org [Tue, 7 Feb 2012 00:41:31 +0000 (00:41 +0000)]
postMessage leaks MemoryEvent object
https://bugs.webkit.org/show_bug.cgi?id=77893

Patch by Greg Simon <gregsimon@chromium.org> on 2012-02-06
Reviewed by Adam Barth.

Changed from PassRefPtr().leakRef() to RefPtr.get() when
calling into FrameLoaderClient to stop +1 extra ref.

* page/DOMWindow.cpp:
(WebCore::DOMWindow::postMessageTimerFired):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@106874 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoWebGL conformance test misc/functions-returning-strings.html fails
commit-queue@webkit.org [Tue, 7 Feb 2012 00:38:43 +0000 (00:38 +0000)]
WebGL conformance test misc/functions-returning-strings.html fails
https://bugs.webkit.org/show_bug.cgi?id=77149

Source/WebCore:

Make sure that WebGL methods returning strings don't return null when
they run successfully.

Patch by Ehsan Akhgari <ehsan.akhgari@gmail.com> on 2012-02-06
Reviewed by Kenneth Russell.

Test: fast/canvas/webgl/functions-returning-strings.html

* html/canvas/WebGLRenderingContext.cpp:
(WebCore):
(WebCore::WebGLRenderingContext::getProgramInfoLog):
(WebCore::WebGLRenderingContext::getShaderInfoLog):
(WebCore::WebGLRenderingContext::getShaderSource):
(WebCore::WebGLRenderingContext::ensureNotNull):
* html/canvas/WebGLRenderingContext.h:
(WebGLRenderingContext):

LayoutTests:

Imported the WebGL conformance test related to this.

Patch by Ehsan Akhgari <ehsan.akhgari@gmail.com> on 2012-02-06
Reviewed by Kenneth Russell.

* fast/canvas/webgl/functions-returning-strings-expected.txt: Added.
* fast/canvas/webgl/functions-returning-strings.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@106873 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoRefactor Mac platform implementation of the Pasteboard class.
enrica@apple.com [Tue, 7 Feb 2012 00:34:18 +0000 (00:34 +0000)]
Refactor Mac platform implementation of the Pasteboard class.
https://bugs.webkit.org/show_bug.cgi?id=77567

The goal of this change is to remove the majority of the methods in
the class interface that are Mac specific.
writeSelectionForTypes has been left to support OS X services.
Some of the methods have been turned into static functions.
The method asURL was being used only by the DragData class and its
implementation has been moved there.
This is a first step in the direction of removing NSPasteboard access from
the WebProcess for WebKit2 (https://bugs.webkit.org/show_bug.cgi?id=77259)
leaving the WebKit1 behavior unchanged.

Reviewed by Alexey Proskuryakov.

No new tests. No changes in behavior.

* platform/Pasteboard.h: Removed most of the Mac specific methods.
* platform/mac/ClipboardMac.mm:
(WebCore::ClipboardMac::writeRange):
(WebCore::ClipboardMac::writeURL):
* platform/mac/DragDataMac.mm:
(WebCore::DragData::asURL): Moved code from PasteboardMac.mm. Removed FIXME
because we only want to handle the case of single file, otherwise the user
doesn't know which of the files has been chosen.
* platform/mac/PasteboardMac.mm:
(WebCore::writeURLForTypes):
(WebCore::Pasteboard::writeURL):
(WebCore::writeFileWrapperAsRTFDAttachment): Now a static function.
(WebCore::Pasteboard::writeImage):
(WebCore::documentFragmentWithImageResource): Ditto.
(WebCore::documentFragmentWithRTF): Ditto.
(WebCore::Pasteboard::documentFragment):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@106872 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago[chromium] Drop tiles completely outside of layer bounds when resizing to a smaller...
jamesr@google.com [Tue, 7 Feb 2012 00:25:16 +0000 (00:25 +0000)]
[chromium] Drop tiles completely outside of layer bounds when resizing to a smaller size
https://bugs.webkit.org/show_bug.cgi?id=77910

Reviewed by Kenneth Russell.

Source/WebCore:

When resizing a tiled layer to a smaller size, drop all tiles that lie completely outside the new layer bounds.
This avoids attempting to access out-of-bounds tiles when iterating over all tiles in the tiler, which triggers
ASSERT()s, as well as saves some memory.

New unit test added to TiledLayerChromiumTest.

* platform/graphics/chromium/TiledLayerChromium.cpp:
(WebCore::TiledLayerChromium::invalidateRect):
* platform/graphics/chromium/cc/CCLayerTilingData.cpp:
(WebCore::CCLayerTilingData::setBounds):

Source/WebKit/chromium:

Adds test for resizing a layer to cover fewer tiles. Test hits ASSERT()s without any code changes.

* tests/TiledLayerChromiumTest.cpp:
(::TEST):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@106870 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoRevert r106859. It was completely bogus.
rniwa@webkit.org [Tue, 7 Feb 2012 00:08:27 +0000 (00:08 +0000)]
Revert r106859. It was completely bogus.

* src/WebViewImpl.cpp:
(WebKit::WebViewImpl::doPixelReadbackToCanvas):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@106867 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago[PerformanceTests] Each Dromaeo test needs its landing html.
morrita@google.com [Mon, 6 Feb 2012 23:44:33 +0000 (23:44 +0000)]
[PerformanceTests] Each Dromaeo test needs its landing html.
https://bugs.webkit.org/show_bug.cgi?id=77504

Reviewed by Ryosuke Niwa.

PerformanceTests:

- Added landing pages for each test which are listed in MANIFEST.json
- Removed some tests which depends on the library whose license is incompatible to WebKit.
- Added local copy of corresponding library. And replace the references to the remote sources
  to the local copies.
- Removed old library files under Dromaeo/resources/dromaeo/web/lib/.
- Skipped Sunsupier and v8 test. Such JS centric test should have its own test instead of
  running as a part of Dromaeo.

* Dromaeo/cssquery-dojo.html: Added.
* Dromaeo/cssquery-jquery.html: Added.
* Dromaeo/cssquery-prototype.html: Added.
* Dromaeo/dom-attr.html: Added.
* Dromaeo/dom-modify.html: Added.
* Dromaeo/dom-traverse.html: Added.
* Dromaeo/dromaeo-3d-cube.html: Added.
* Dromaeo/dromaeo-core-eval.html: Added.
* Dromaeo/dromaeo-object-array.html: Added.
* Dromaeo/dromaeo-object-regexp.html: Added.
* Dromaeo/dromaeo-object-string.html: Added.
* Dromaeo/dromaeo-string-base64.html: Added.
* Dromaeo/jslib-attr-jquery.html: Added.
* Dromaeo/jslib-attr-prototype.html: Added.
* Dromaeo/jslib-event-jquery.html: Added.
* Dromaeo/jslib-event-prototype.html: Added.
* Dromaeo/jslib-modify-jquery.html: Added.
* Dromaeo/jslib-modify-prototype.html: Added.
* Dromaeo/jslib-style-jquery.html: Added.
* Dromaeo/jslib-style-prototype.html: Added.
* Dromaeo/jslib-traverse-jquery.html: Added.
* Dromaeo/jslib-traverse-prototype.html: Added.
* Dromaeo/resources/dromaeo/web/lib/dojo.js: Removed.
* Dromaeo/resources/dromaeo/web/lib/dojo-1.6.1.js: Added.
* Dromaeo/resources/dromaeo/web/lib/jquery.js: Removed.
* Dromaeo/resources/dromaeo/web/lib/jquery-1.6.4.js: Added.
* Dromaeo/resources/dromaeo/web/lib/mootools.js: Removed.
* Dromaeo/resources/dromaeo/web/lib/prototype.js: Removed.
* Dromaeo/resources/dromaeo/web/lib/prototype-1.7.js: Added.
* Dromaeo/resources/dromaeo/web/tests/cssquery-dojo.html:
* Dromaeo/resources/dromaeo/web/tests/cssquery-jquery.html:
* Dromaeo/resources/dromaeo/web/tests/cssquery-mootools.html: Removed.
* Dromaeo/resources/dromaeo/web/tests/cssquery-prototype.html:
* Dromaeo/resources/dromaeo/web/tests/jslib-attr-jquery.html:
* Dromaeo/resources/dromaeo/web/tests/jslib-attr-prototype.html:
* Dromaeo/resources/dromaeo/web/tests/jslib-event-jquery.html:
* Dromaeo/resources/dromaeo/web/tests/jslib-event-prototype.html:
* Dromaeo/resources/dromaeo/web/tests/jslib-modify-jquery.html:
* Dromaeo/resources/dromaeo/web/tests/jslib-modify-prototype.html:
* Dromaeo/resources/dromaeo/web/tests/jslib-style-jquery.html:
* Dromaeo/resources/dromaeo/web/tests/jslib-style-prototype.html:
* Dromaeo/resources/dromaeo/web/tests/jslib-traverse-jquery.html:
* Dromaeo/resources/dromaeo/web/tests/jslib-traverse-prototype.html:
* Dromaeo/sunspider-3d-morph.html: Added.
* Dromaeo/sunspider-3d-raytrace.html: Added.
* Dromaeo/sunspider-access-binary-trees.html: Added.
* Dromaeo/sunspider-access-fannkuch.html: Added.
* Dromaeo/sunspider-access-nbody.html: Added.
* Dromaeo/sunspider-access-nsieve.html: Added.
* Dromaeo/sunspider-bitops-3bit-bits-in-byte.html: Added.
* Dromaeo/sunspider-bitops-bits-in-byte.html: Added.
* Dromaeo/sunspider-bitops-bitwise-and.html: Added.
* Dromaeo/sunspider-bitops-nsieve-bits.html: Added.
* Dromaeo/sunspider-controlflow-recursive.html: Added.
* Dromaeo/sunspider-crypto-aes.html: Added.
* Dromaeo/sunspider-crypto-md5.html: Added.
* Dromaeo/sunspider-crypto-sha1.html: Added.
* Dromaeo/sunspider-date-format-tofte.html: Added.
* Dromaeo/sunspider-date-format-xparb.html: Added.
* Dromaeo/sunspider-math-cordic.html: Added.
* Dromaeo/sunspider-math-partial-sums.html: Added.
* Dromaeo/sunspider-math-spectral-norm.html: Added.
* Dromaeo/sunspider-regexp-dna.html: Added.
* Dromaeo/sunspider-string-fasta.html: Added.
* Dromaeo/sunspider-string-tagcloud.html: Added.
* Dromaeo/sunspider-string-unpack-code.html: Added.
* Dromaeo/sunspider-string-validate-input.html: Added.
* Dromaeo/v8-crypto.html: Added.
* Dromaeo/v8-deltablue.html: Added.
* Dromaeo/v8-earley-boyer.html: Added.
* Dromaeo/v8-raytrace.html: Added.
* Dromaeo/v8-richards.html: Added.
* Skipped:

Tools:

Added an ignorable pattern which happens in some Dromaeo tests.

* Scripts/webkitpy/performance_tests/perftestsrunner.py:
(PerfTestsRunner):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@106865 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agozvmul incorrectly multiplies complex arrays on Windows.
crogers@google.com [Mon, 6 Feb 2012 23:23:43 +0000 (23:23 +0000)]
zvmul incorrectly multiplies complex arrays on Windows.
https://bugs.webkit.org/show_bug.cgi?id=77900

Reviewed by Kenneth Russell.

Source/WebCore:

* platform/audio/VectorMath.cpp:
(WebCore::VectorMath::zvmul):

LayoutTests:

* platform/chromium/test_expectations.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@106864 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoREGRESSION(r106756): 10% performance hit on DOM/Template.
kling@webkit.org [Mon, 6 Feb 2012 23:14:11 +0000 (23:14 +0000)]
REGRESSION(r106756): 10% performance hit on DOM/Template.
<http://webkit.org/b/77831>

Reviewed by Ryosuke Niwa.

Let the StylePropertySet used for element attribute style have the element as its parent.
This is accomplished by adding an m_parentIsElement bit to StylePropertySet and sharing
some of the internal logic with inline styles.

In the end, this means that CSSParser will now pick up the document's CSSValuePool when
parsing properties for attribute styles, which fixes the perf regression from r106756.

* css/StylePropertySet.cpp:
(WebCore::StylePropertySet::StylePropertySet):
(WebCore::StylePropertySet::contextStyleSheet):

    Find contextStyleSheet via the parentElement() when there is one.

(WebCore::StylePropertySet::setNeedsStyleRecalc):

    Always set FullStyleChange for attribute style mutations. We can probably use the
    same lighter invalidation as inline styles, but that's a topic for another patch.

* css/StylePropertySet.h:
(WebCore::StylePropertySet::createInline):
(WebCore::StylePropertySet::createAttributeStyle):
(WebCore::StylePropertySet::parentRuleInternal):
(WebCore::StylePropertySet::clearParentRule):
(StylePropertySet):
(WebCore::StylePropertySet::parentElement):
(WebCore::StylePropertySet::clearParentElement):

    Added m_parentIsElement bit and update assertions as appropriate to not just
    cover the inline style case. Added a createAttributeStyle() helper to create
    a StylePropertySet for use as Element::attributeStyle().

* dom/StyledElement.h:
* dom/ElementAttributeData.h:
* dom/ElementAttributeData.cpp:
(WebCore::ElementAttributeData::ensureAttributeStyle):

    Use StylePropertySet::createAttributeStyle().

* dom/StyledElement.cpp:
(WebCore::StyledElement::removeCSSProperties):
(WebCore::StyledElement::addCSSProperty):
(WebCore::StyledElement::addCSSImageProperty):

    Remove setNeedsStyleRecalc() calls since that is now handled automatically by
    StylePropertySet's mutation methods.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@106863 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoIn AppleWebKit, stop rebuilding IDLs that need not to be rebuilt
haraken@chromium.org [Mon, 6 Feb 2012 23:10:54 +0000 (23:10 +0000)]
In AppleWebKit, stop rebuilding IDLs that need not to be rebuilt
https://bugs.webkit.org/show_bug.cgi?id=77510

Reviewed by Adam Barth.

Currently, if any IDL file is updated, all IDL files are rebuilt.
This patch stops rebuilding IDL files which are not modified nor
whose supplemental dependencies are not changed.

The new build flow is as follows:

    supplemental.dep : $(ALL_IDLS)
        perl resolve-supplemental.pl ...

    %.idl-needs-rebuild : %.idl supplemental.dep
        perl update-idl-needs-rebuild.pl ...

    JS%.cpp : %.idl-needs-rebuild
        perl generate-bindings.pl ...

resolve-supplemental.pl generates the following supplemental.dep.
The number in () is the last access timestamp of the file.

    A.idl(1200)
    B.idl(1000) B-supplemental1.idl(800) B-supplemental2.idl(1200)
    C.idl(1000)

update-idl-needs-rebuild.pl for X.idl touches X.idl-needs-rebuild,
if X.idl-needs-rebuild is older than X.idl or the IDL files which are
supplementing X.idl. For example, if the timestamps of A.idl-needs-rebuild,
B.idl-needs-rebuild and C.idl-needs-rebuild are all 1000, then A.idl-needs-rebuild
and B.idl-needs-rebuild will be touched.

Even if no IDL files are modified, update-idl-needs-rebuild.pl can run for all IDL files,
but generate-bindings.pl won't run. If any IDL file is updated, resolve-supplemental.pl
will run once, update-idl-needs-rebuild.pl will run for all IDL files, and
generate-bindings.pl will run for the IDL files which are modified or whose
dependencies are changed.

No tests. I manually confirmed the followings:
- Touch Element.idl, then only Element.idl is rebuilt
- Touch DOMWindow.idl, then only DOMWindow.idl is rebuilt
- Touch DOMWindowWebAudio.idl, then only DOMWindow.idl and DOMWindowWebAudio.idl are rebuilt

* DerivedSources.make: Modified the build flow as described above.
* bindings/scripts/generate-bindings.pl: Modified to read an IDL file path from .idl-needs-rebuild.
* bindings/scripts/update-idl-needs-rebuild.pl: Added. Touches X.idl-needs-rebuild if
the X.idl-needs-rebuild is older than X.idl or the IDL files which are supplementing X.idl.
(touch):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@106862 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoUnreviewed, chromium expectations update.
commit-queue@webkit.org [Mon, 6 Feb 2012 23:10:17 +0000 (23:10 +0000)]
Unreviewed, chromium expectations update.
Layout Test http/tests/media/video-served-as-text.html is flaky on cr-win
https://bugs.webkit.org/show_bug.cgi?id=72904

Test is not flaky on XP.

Patch by Ami Fischman <fischman@chromium.org> on 2012-02-06

* platform/chromium/test_expectations.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@106861 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoWindows build fix
mdelaney@apple.com [Mon, 6 Feb 2012 23:08:52 +0000 (23:08 +0000)]
Windows build fix

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@106860 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoChromium build fix after r106836.
rniwa@webkit.org [Mon, 6 Feb 2012 23:02:17 +0000 (23:02 +0000)]
Chromium build fix after r106836.

* src/WebViewImpl.cpp:
(WebKit::WebViewImpl::doPixelReadbackToCanvas):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@106859 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoAudioBus need to support stereo->mono down mix in copyFrom sumFrom etc.
commit-queue@webkit.org [Mon, 6 Feb 2012 23:00:32 +0000 (23:00 +0000)]
AudioBus need to support stereo->mono down mix in copyFrom sumFrom etc.
https://bugs.webkit.org/show_bug.cgi?id=77609

Patch by Wei James <james.wei@intel.com> on 2012-02-06
Reviewed by Kenneth Russell.

Test: webaudio/stereo2mono-down-mixing.html

* platform/audio/AudioBus.cpp:
(WebCore):
(WebCore::AudioBus::copyFrom):
(WebCore::AudioBus::sumFrom):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@106858 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago[Chromium] storage/indexeddb/index-cursor.html timeouts on Mac Debug
jsbell@chromium.org [Mon, 6 Feb 2012 22:55:26 +0000 (22:55 +0000)]
[Chromium] storage/indexeddb/index-cursor.html timeouts on Mac Debug
https://bugs.webkit.org/show_bug.cgi?id=65862

No sign of this test timing out in recent history, and significant
performance work has been done since it was filed. Unmarking as SLOW/PASS.

Reviewed by Tony Chang.

* platform/chromium/test_expectations.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@106857 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago[filesystem/Chromium] crackFileSystemURL needs to use innerURL
ericu@chromium.org [Mon, 6 Feb 2012 22:54:41 +0000 (22:54 +0000)]
[filesystem/Chromium] crackFileSystemURL needs to use innerURL
https://bugs.webkit.org/show_bug.cgi?id=77881

Reviewed by Adam Barth.

* src/AsyncFileSystemChromium.cpp:
(WebCore::AsyncFileSystem::crackFileSystemURL):
Add the new code path, taken only if innerURL is present.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@106856 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoUnreviewed gardening.
jchaffraix@webkit.org [Mon, 6 Feb 2012 22:39:36 +0000 (22:39 +0000)]
Unreviewed gardening.

More font differences that need a rebaseline.

* platform/chromium-linux-x86/fast/canvas/check-stale-putImageData-expected.png: Removed.
* platform/chromium-linux/fast/canvas/check-stale-putImageData-expected.png: Added.
* platform/chromium-win/fast/canvas/check-stale-putImageData-expected.png: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@106855 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago[Chromium] rebaseline test on win/mac after enabling paint opaque tracking for draw...
commit-queue@webkit.org [Mon, 6 Feb 2012 22:35:17 +0000 (22:35 +0000)]
[Chromium] rebaseline test on win/mac after enabling paint opaque tracking for draw culling
https://bugs.webkit.org/show_bug.cgi?id=77497

Patch by Dana Jansens <danakj@chromium.org> on 2012-02-06
Reviewed by James Robinson.

* platform/chromium-win/fast/repaint/block-selection-gap-in-composited-layer-expected.png: Added.
* platform/chromium/test_expectations.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@106854 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agohttp/tests/security/mixedContent/insecure-plugin-in-iframe.html is flaky
abarth@webkit.org [Mon, 6 Feb 2012 22:32:52 +0000 (22:32 +0000)]
http/tests/security/mixedContent/insecure-plugin-in-iframe.html is flaky
https://bugs.webkit.org/show_bug.cgi?id=77901

Reviewed by Eric Seidel.

This test fails occationally because of the order of the frame load
callbacks.  Now that we log mixed content errors to the console, we
don't need to dump the frame load callbacks in this test.

* http/tests/security/mixedContent/insecure-plugin-in-iframe-expected.txt:
* http/tests/security/mixedContent/insecure-plugin-in-iframe.html:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@106853 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoSource/WebCore: Add RefPtrs for parent and sibling counter nodes
cdn@chromium.org [Mon, 6 Feb 2012 22:30:09 +0000 (22:30 +0000)]
Source/WebCore: Add RefPtrs for parent and sibling counter nodes
https://bugs.webkit.org/show_bug.cgi?id=75212

Reviewed by Adam Barth.

Test: fast/css/counters/reparent-table-children-with-counters-crash.html

* rendering/RenderCounter.cpp:
(WebCore::findPlaceForCounter):
(WebCore::makeCounterNode):
(WebCore::updateCounters):

LayoutTests: Test for crash when counter nodes are reparented outside of tables
https://bugs.webkit.org/show_bug.cgi?id=75212

Reviewed by Adam Barth.

* fast/css/counters/reparent-table-children-with-counters-crash-expected.txt: Added.
* fast/css/counters/reparent-table-children-with-counters-crash.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@106852 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoUnreviewed fix for the new API test that is failing -
kov@webkit.org [Mon, 6 Feb 2012 22:26:29 +0000 (22:26 +0000)]
Unreviewed fix for the new API test that is failing -
WEBKIT_INSPECTOR_PATH needs to be set, since we do not install our
builds.

* tests/testwebinspector.c:
(main):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@106851 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoFix for burning SL tree
mdelaney@apple.com [Mon, 6 Feb 2012 22:17:44 +0000 (22:17 +0000)]
Fix for burning SL tree

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@106850 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoAccording to the flakiness dashboard, these tests do appear to be
abarth@webkit.org [Mon, 6 Feb 2012 22:17:23 +0000 (22:17 +0000)]
According to the flakiness dashboard, these tests do appear to be
successfully de-flaked.

* platform/chromium/test_expectations.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@106849 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago[GTK] Add missing pango include dir to fix build
commit-queue@webkit.org [Mon, 6 Feb 2012 21:59:08 +0000 (21:59 +0000)]
[GTK] Add missing pango include dir to fix build
https://bugs.webkit.org/show_bug.cgi?id=77832

Patch by Kalev Lember <kalevlember@gmail.com> on 2012-02-06
Reviewed by Martin Robinson.

Source/WebCore:

* GNUmakefile.am: Added $(PANGO_CFLAGS) to libWebCore_la_CPPFLAGS.

Tools:

* GNUmakefile.am: Added $(PANGO_CFLAGS) to libWebCoreInternals_la_CPPFLAGS.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@106847 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoBefore accessing a frame's script controller in V8 bindings, first check that the...
jochen@chromium.org [Mon, 6 Feb 2012 21:55:47 +0000 (21:55 +0000)]
Before accessing a frame's script controller in V8 bindings, first check that the frame actually exists
https://bugs.webkit.org/show_bug.cgi?id=77370

Reviewed by Adam Barth.

I don't have a working reproduction of the crash yet. As soon as I have
one,  I will add new layout tests.

* bindings/v8/PageScriptDebugServer.cpp:
(WebCore::PageScriptDebugServer::addListener):
* bindings/v8/ScheduledAction.cpp:
(WebCore::ScheduledAction::execute):
* bindings/v8/ScriptCachedFrameData.cpp:
(WebCore::ScriptCachedFrameData::restore):
* page/DOMTimer.cpp:
(WebCore::DOMTimer::DOMTimer): remove temporary debug code

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@106846 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoUnreviewed rebaseline after r106836.
jchaffraix@webkit.org [Mon, 6 Feb 2012 21:39:27 +0000 (21:39 +0000)]
Unreviewed rebaseline after r106836.

The rebaseline is caused by a font difference.

* platform/chromium-linux-x86/fast/canvas/check-stale-putImageData-expected.png: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@106844 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoCorrecting the version of WKSI committed in r106836
mdelaney@apple.com [Mon, 6 Feb 2012 21:34:05 +0000 (21:34 +0000)]
Correcting the version of WKSI committed in r106836

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@106843 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoUnreviewed chromium test expectation change
jsbell@chromium.org [Mon, 6 Feb 2012 21:25:49 +0000 (21:25 +0000)]
Unreviewed chromium test expectation change

Unmarking storage/indexeddb/data-corruption.html as TIMEOUT. Was apparently
bulk-marked as TIMEOUT on Snow Leopard as part of WK54322 although it is
not mentioned in that issue. The test hasn't failed in recent history and
there has been substantial rework on the IndexedDB feature since the
expectation was set.

* platform/chromium/test_expectations.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@106842 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoSource/WebCore: Support detaching TextureManager from ManagedTexture
jamesr@google.com [Mon, 6 Feb 2012 21:17:49 +0000 (21:17 +0000)]
Source/WebCore: Support detaching TextureManager from ManagedTexture
https://bugs.webkit.org/show_bug.cgi?id=77655

Reviewed by Kenneth Russell.
Initial patch by Alok Priyadarshi.

TextureManager now holds references to the textures it manages.
This allows TextureManager to inform managed textures when it gets deleted
so that the texture that outlive the TextureManager can handle the situation gracefully.

Unit test in TextureManagerTest.cpp

* platform/graphics/chromium/ManagedTexture.cpp:
(WebCore::ManagedTexture::ManagedTexture):
(WebCore::ManagedTexture::~ManagedTexture):
(WebCore):
(WebCore::ManagedTexture::managerWillDie):
(WebCore::ManagedTexture::isValid):
(WebCore::ManagedTexture::reserve):
(WebCore::ManagedTexture::unreserve):
(WebCore::ManagedTexture::steal):
(WebCore::ManagedTexture::reset):
* platform/graphics/chromium/ManagedTexture.h:
(WebCore::ManagedTexture::manager):
(ManagedTexture):
(WebCore::ManagedTexture::isReserved):
* platform/graphics/chromium/TextureManager.cpp:
(WebCore::TextureManager::~TextureManager):
(WebCore):
(WebCore::TextureManager::setPreferredMemoryLimitBytes):
(WebCore::TextureManager::registerTexture):
(WebCore::TextureManager::unregisterTexture):
* platform/graphics/chromium/TextureManager.h:
(WebCore):
(TextureManager):

Source/WebKit/chromium: [chromium] Support detaching TextureManager from ManagedTexture
https://bugs.webkit.org/show_bug.cgi?id=77655

Adds unit test for destroying a TextureManager while ManagedTextures are still alive referring to it.

Reviewed by Kenneth Russell.

* tests/TextureManagerTest.cpp:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@106840 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoWheel event handler count not updated when adding handlers to the window
andersca@apple.com [Mon, 6 Feb 2012 21:05:22 +0000 (21:05 +0000)]
Wheel event handler count not updated when adding handlers to the window
https://bugs.webkit.org/show_bug.cgi?id=77895

Reviewed by Dan Bernstein.

Source/WebCore:

Test: fast/events/wheelevent-handler-count.html

* page/DOMWindow.cpp:
(WebCore::DOMWindow::addEventListener):
(WebCore::DOMWindow::removeEventListener):
Update the wheel event count.

* testing/Internals.cpp:
(WebCore::Internals::wheelEventHandlerCount):
* testing/Internals.h:
* testing/Internals.idl:
Add wheelEventHandlerCount to window.internals.

LayoutTests:

* fast/events/wheelevent-handler-count-expected.txt: Added.
* fast/events/wheelevent-handler-count.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@106839 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoSource/WebKit2: [GTK] Fix remaining errors in GTK+ WebKit2 API
commit-queue@webkit.org [Mon, 6 Feb 2012 21:01:28 +0000 (21:01 +0000)]
Source/WebKit2:  [GTK] Fix remaining errors in GTK+ WebKit2 API
https://bugs.webkit.org/show_bug.cgi?id=77890

Patch by Martin Robinson <mrobinson@igalia.com> on 2012-02-06
Reviewed by Gustavo Noronha Silva.

Fix some style issues in the GTK+ WebKit2 API section of the source code.

* UIProcess/API/gtk/WebKitPrivate.h:
* UIProcess/API/gtk/WebKitWebView.cpp:
* UIProcess/API/gtk/tests/TestMain.h:

Tools: [GTK] Fix remaining errors in GTK+ WebKit2 API
https://bugs.webkit.org/show_bug.cgi?id=77890

Patch by Martin Robinson <mrobinson@igalia.com> on 2012-02-06
Reviewed by Gustavo Noronha Silva.

Verify that WebKit2 GTK+ test files are not skipped during style checks.
Skip soup_server_new when checking for NULL versus 0 usage.

* Scripts/webkitpy/style/checker_unittest.py:
(CheckerDispatcherSkipTest.test_should_skip_with_warning):
* Scripts/webkitpy/style/checkers/cpp.py:
(check_for_null):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@106838 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agotoDataURL() uses stale data after putImageData()
mdelaney@apple.com [Mon, 6 Feb 2012 20:54:13 +0000 (20:54 +0000)]
toDataURL() uses stale data after putImageData()
https://bugs.webkit.org/show_bug.cgi?id=65767

Source/WebCore:

This patch fixes the issue we've encountered of getting back
stale copies of the CGContext of accelerated ImageBuffers who have seen
putImageData calls but have not been drawn into via the CG API.
This issue is fixed by modifying the way we implement putImageData
in ImageBufferCG to draw the bits wrapped in a CGImage while the CGContext
is in a state where the data will effectively be copied (as is needed for
implementing putImageData) instead of directly modifying the bits of the IOSurface.

Reviewed by Chris Marrin.

Test: fast/canvas/check-stale-putImageData, pixel test to check that the canvas is in fact painted.

* platform/graphics/cg/ImageBufferCG.cpp: Implement new way of putting image data.
* platform/graphics/ImageBuffer.h: Merged two previously separate put data calls
    into a single and more sensibly named 'putByteArray', since that's what it does!

* WebCore.exp.in: Added new WKSI call for use in ImageBufferCG.cpp
* platform/mac/WebCoreSystemInterface.h:
* platform/mac/WebCoreSystemInterface.mm:

Using new method name.
* html/canvas/CanvasRenderingContext2D.cpp:
* platform/graphics/ImageBuffer.cpp:
* platform/graphics/ShadowBlur.cpp:
* platform/graphics/filters/FEColorMatrix.cpp:
* platform/graphics/filters/FEDropShadow.cpp:
* platform/graphics/filters/FilterEffect.cpp:

Updated other ports' ImageBuffers to use new method.
* platform/graphics/cairo/ImageBufferCairo.cpp:
* platform/graphics/qt/ImageBufferQt.cpp:
* platform/graphics/skia/ImageBufferSkia.cpp:
* platform/graphics/wince/ImageBufferWinCE.cpp:
* platform/graphics/wx/ImageBufferWx.cpp:

Source/WebKit/chromium:

Reviewed by Chris Marrin.

* src/WebViewImpl.cpp: Updated method name.
(WebKit::WebViewImpl::doPixelReadbackToCanvas):

Source/WebKit/mac:

Reviewed by Chris Marrin.

* WebCoreSupport/WebSystemInterface.mm:

Source/WebKit2:

Reviewed by Chris Marrin.

* WebProcess/WebCoreSupport/mac/WebSystemInterface.mm:
(InitWebCoreSystemInterface):

WebKitLibraries:

Added WKCGContextResetClip for use in reseting clip for new putByteArray method.

Reviewed by Chris Marrin.

* WebKitSystemInterface.h: Added WKCGContextResetClip.
* libWebKitSystemInterfaceLeopard.a:
* libWebKitSystemInterfaceSnowLeopard.a:
* libWebKitSystemInterfaceLion.a:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@106836 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoProvide more attribute methods in Element
caio.oliveira@openbossa.org [Mon, 6 Feb 2012 20:42:40 +0000 (20:42 +0000)]
Provide more attribute methods in Element
https://bugs.webkit.org/show_bug.cgi?id=77800

Reviewed by Ryosuke Niwa.

Source/WebCore:

Reduce the clients of NamedNodeMap. This will make easier to change the
implementation of our attribute storage. The clients now use Element methods
that expose Attribute* accessors.

Instead of checking the existence of NamedNodeMap, clients can call
hasAttributes() or hasAttributesWithoutUpdate() (that skips updating invalid
style or animation svg attributes).

If there are attributes, they can be accessed by index via attributeCount() /
attributeItem(), as well as by QualifiedName. Those accessors assume there are an
attribute storage.

* css/SelectorChecker.cpp:
(WebCore::anyAttributeMatches):
(WebCore::SelectorChecker::checkOneSelector):
* css/SelectorChecker.h:
(WebCore::SelectorChecker::checkExactAttribute):
* dom/DatasetDOMStringMap.cpp:
(WebCore::DatasetDOMStringMap::getNames):
(WebCore::DatasetDOMStringMap::item):
(WebCore::DatasetDOMStringMap::contains):
* dom/Document.cpp:
(WebCore::Document::importNode): use setAttributesFromElement() instead of manually copying.
* dom/Element.h:
(Element):
(WebCore::Element::hasAttributesWithoutUpdate):
(WebCore):
(WebCore::Element::attributeCount):
(WebCore::Element::attributeItem):
(WebCore::Element::getAttributeItem):
(WebCore::Element::removeAttribute):
* dom/Node.cpp:
(WebCore::Node::isDefaultNamespace):
(WebCore::Node::lookupNamespaceURI):
(WebCore::Node::lookupNamespacePrefix):
(WebCore::Node::compareDocumentPosition):
* editing/ApplyStyleCommand.cpp:
(WebCore::hasNoAttributeOrOnlyStyleAttribute):
(WebCore::isEmptyFontTag):
* editing/EditingStyle.cpp:
(WebCore::EditingStyle::elementIsStyledSpanOrHTMLEquivalent):
* editing/InsertParagraphSeparatorCommand.cpp:
(WebCore::highestVisuallyEquivalentDivBelowRoot):
* editing/MarkupAccumulator.cpp:
(WebCore::MarkupAccumulator::appendElement):
* editing/markup.cpp:
(WebCore::completeURLs):
(WebCore::StyledMarkupAccumulator::appendElement):
* html/HTMLEmbedElement.cpp:
(WebCore::HTMLEmbedElement::parametersForPlugin):
* html/HTMLInputElement.cpp:
(WebCore::HTMLInputElement::updateType):
* html/HTMLObjectElement.cpp:
(WebCore::HTMLObjectElement::parametersForPlugin):
* html/HTMLParamElement.cpp:
(WebCore::HTMLParamElement::isURLAttribute):
* html/parser/HTMLFormattingElementList.cpp:
(WebCore::attributeCountWithoutUpdate):
(WebCore::HTMLFormattingElementList::tryToEnsureNoahsArkConditionQuickly):
(WebCore::HTMLFormattingElementList::ensureNoahsArkCondition): Store Attribute*
to avoid looking up the right element again by name.
* inspector/DOMEditor.cpp:
(WebCore::DOMEditor::innerPatchNode):
(WebCore::DOMEditor::createDigest):
* inspector/InspectorDOMAgent.cpp:
(WebCore::InspectorDOMAgent::setAttributesAsText):
(WebCore::InspectorDOMAgent::performSearch):
(WebCore::InspectorDOMAgent::buildArrayForElementAttributes):
* page/PageSerializer.cpp:
(WebCore::isCharsetSpecifyingNode):
* svg/SVGStyledElement.cpp:
(WebCore::SVGStyledElement::getPresentationAttribute):
* xml/XPathFunctions.cpp:
(WebCore::XPath::FunLang::evaluate):
* xml/XPathNodeSet.cpp:
(WebCore::XPath::NodeSet::traversalSort):
* xml/XPathStep.cpp:
(WebCore::XPath::Step::nodesInAxis):
* xml/parser/XMLDocumentParserLibxml2.cpp:
(WebCore::XMLDocumentParser::XMLDocumentParser):
* xml/parser/XMLTreeBuilder.cpp:
(WebCore::XMLTreeBuilder::XMLTreeBuilder):

Source/WebKit/chromium:

* src/WebPageSerializerImpl.cpp:
(WebKit::WebPageSerializerImpl::openTagToString): access attributes via Element interface.

Source/WebKit/qt:

* Api/qwebelement.cpp:
(QWebElement::attributeNames): access attributes via Element interface.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@106833 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agowebkit-patch apply-from-bug should be able to apply non-reviewed patches
eric@webkit.org [Mon, 6 Feb 2012 20:29:09 +0000 (20:29 +0000)]
webkit-patch apply-from-bug should be able to apply non-reviewed patches
https://bugs.webkit.org/show_bug.cgi?id=77883

Reviewed by Adam Barth.

This is very helpful if you use bugzilla as a replacement for git-stash.
Makes it simpler to re-apply the patch w/o having to look up the patch id.

* Scripts/webkitpy/tool/commands/download.py:
(ProcessBugsMixin._fetch_list_of_patches_to_process):
* Scripts/webkitpy/tool/commands/download_unittest.py:
(DownloadCommandsTest.test_apply_from_bug):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@106832 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago[GTK] Insta-crash when closing browser with inspector window opened
kov@webkit.org [Mon, 6 Feb 2012 20:12:38 +0000 (20:12 +0000)]
[GTK] Insta-crash when closing browser with inspector window opened
https://bugs.webkit.org/show_bug.cgi?id=50744

Reviewed by Martin Robinson.

Source/WebKit/gtk:

* GNUmakefile.am:
* WebCoreSupport/InspectorClientGtk.cpp:
(WebKit::InspectorClient::releaseFrontendPage): also let go of the
frontend client when releasing the frontend page;
(WebKit::InspectorFrontendClient::~InspectorFrontendClient(): Call releaseFrontendPage()
like other ports do instead of just disconnectFrontendClient().
(WebKit::InspectorFrontendClient::destroyInspectorWindow): Use GRefPtr for the
inspector ref.
(WebKit::InspectorFrontendClient::bringToFront): Adjust for GRefPtr usage.
(WebKit::InspectorFrontendClient::attachWindow): Ditto.
(WebKit::InspectorFrontendClient::detachWindow): Ditto.
(WebKit::InspectorFrontendClient::inspectedURLChanged): Ditto.
* WebCoreSupport/InspectorClientGtk.h:
(InspectorFrontendClient): Use GRefPtr for inspector member.
* tests/testwebinspector.c: Added. Test that killing a WebView with the inspector
open does not crash, and that closing and reinspecting works.
(quitLoop):
(consoleMessageCallback):
(inspectElementCallback):
(closeInspector):
(showInspector):
(test_webkit_web_inspector_close_and_inspect):
(test_webkit_web_inspector_destroy_inspected_web_view):
(main):

LayoutTests:

* platform/gtk/Skipped: unskip inspector/debugger/open-close-open.html,
which now passes.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@106831 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoIndexedDB: Raise exception during add/put call if autoIncrement key insertion will...
jsbell@chromium.org [Mon, 6 Feb 2012 20:00:24 +0000 (20:00 +0000)]
IndexedDB: Raise exception during add/put call if autoIncrement key insertion will fail
https://bugs.webkit.org/show_bug.cgi?id=77374

Source/WebCore:

If a put request will use a key generator, try inserting a dummy key during the sync
put() call to check if the key insertion will succeed so an exception can be raised
early, rather than deferring to the asynchronous task.

Reviewed by Tony Chang.

Test: storage/indexeddb/keypath-edges.html

* storage/IDBObjectStoreBackendImpl.cpp:
(WebCore::IDBObjectStoreBackendImpl::put):

LayoutTests:

Reviewed by Tony Chang.

* storage/indexeddb/keypath-edges-expected.txt:
* storage/indexeddb/keypath-edges.html:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@106830 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago[Chromium] Test fast/forms/listbox-clear-restore.html is flaky
jchaffraix@webkit.org [Mon, 6 Feb 2012 19:56:45 +0000 (19:56 +0000)]
[Chromium] Test fast/forms/listbox-clear-restore.html is flaky
https://bugs.webkit.org/show_bug.cgi?id=77889

Unreviewed gardening.

* platform/chromium/test_expectations.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@106829 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoAdd a null check for the gdkwindow, that will happen if the window
kov@webkit.org [Mon, 6 Feb 2012 19:54:43 +0000 (19:54 +0000)]
Add a null check for the gdkwindow, that will happen if the window
is not mapped.

Reviewed by Martin Robinson.

* platform/gtk/GtkUtilities.cpp:
(WebCore::convertWidgetPointToScreenPoint):
* WebCoreSupport/ChromeClientGtk.cpp:
(WebKit::ChromeClient::setCursor):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@106828 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoSource/WebCore: IndexedDB createObjectStore should throw if options arg is invalid
commit-queue@webkit.org [Mon, 6 Feb 2012 19:50:11 +0000 (19:50 +0000)]
Source/WebCore: IndexedDB createObjectStore should throw if options arg is invalid
Added logic to OptionsObject to determine if an invalid object was created.
Javascript bindings now detect invalid OptionsObject's and throw TypeError when found.
https://bugs.webkit.org/show_bug.cgi?id=58471

Patch by Eugene Girard <girard@chromium.org> on 2012-02-06
Reviewed by Adam Barth.

Test: storage/indexeddb/createObjectStore-bad-options.html

* bindings/scripts/CodeGeneratorV8.pm:
(GenerateParametersCheck):
* bindings/v8/OptionsObject.cpp:
(WebCore::OptionsObject::isObject):
* bindings/v8/OptionsObject.h:

LayoutTests: IndexedDB createObjectStore should throw if options arg is invalid (not an object).
Added unit tests to test invalid arguments.
storage/indexddb/transaction-basics was modified only to correct calls that would
now throw.
https://bugs.webkit.org/show_bug.cgi?id=58471

Patch by Eugene Girard <girard@chromium.org> on 2012-02-06
Reviewed by Adam Barth.

* storage/indexeddb/create-object-store-options-expected.txt:
* storage/indexeddb/create-object-store-options.html:
* storage/indexeddb/index-basics-expected.txt:
* storage/indexeddb/index-basics.html:
* storage/indexeddb/resources/shared.js:
(evalAndExpectExceptionClass):
* storage/indexeddb/transaction-basics-expected.txt:
* storage/indexeddb/transaction-basics.html:
* storage/indexeddb/tutorial.html:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@106827 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago[Qt][WK2] http/tests/loading/state-object-security-exception.html fails
jesus@webkit.org [Mon, 6 Feb 2012 19:37:58 +0000 (19:37 +0000)]
[Qt][WK2] http/tests/loading/state-object-security-exception.html fails
https://bugs.webkit.org/show_bug.cgi?id=62731

Unreviewed gardening.

Skipped 3 tests because they were still failing after http://trac.webkit.org/changeset/106812 .
Skipped tests were:
fast/history/form-submit-in-frame-via-onclick.html
fast/history/form-submit-in-frame.html
http/tests/loading/state-object-security-exception.html

* platform/qt-wk2/Skipped:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@106825 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agogcc build fix after r106749.
rniwa@webkit.org [Mon, 6 Feb 2012 19:37:12 +0000 (19:37 +0000)]
gcc build fix after r106749.

* UIProcess/WebPreferences.cpp:
(WebKit::WebPreferences::WebPreferences):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@106824 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoUnreviewed, rolling out r106797 and r106806.
jchaffraix@webkit.org [Mon, 6 Feb 2012 19:28:37 +0000 (19:28 +0000)]
Unreviewed, rolling out r106797 and r106806.
http://trac.webkit.org/changeset/106797
http://trac.webkit.org/changeset/106806
https://bugs.webkit.org/show_bug.cgi?id=77888

The change is still causing some crashes in webaudio/ on the
Chromium bots (Requested by jchaffraix on #webkit).

Patch by Sheriff Bot <webkit.review.bot@gmail.com> on 2012-02-06

Source/WebCore:

* Modules/intents/IntentRequest.cpp:
(WebCore::IntentRequest::create):
* bindings/generic/ActiveDOMCallback.cpp:
(WebCore::ActiveDOMCallback::ActiveDOMCallback):
* dom/ActiveDOMObject.cpp:
(WebCore::ActiveDOMObject::ActiveDOMObject):
(WebCore::ActiveDOMObject::~ActiveDOMObject):
* dom/ActiveDOMObject.h:
(ActiveDOMObject):
* dom/DocumentEventQueue.cpp:
(WebCore::DocumentEventQueue::DocumentEventQueue):
* dom/ScriptExecutionContext.cpp:
(WebCore::ScriptExecutionContext::ScriptExecutionContext):
(WebCore::ScriptExecutionContext::~ScriptExecutionContext):
(WebCore::ScriptExecutionContext::canSuspendActiveDOMObjects):
(WebCore::ScriptExecutionContext::suspendActiveDOMObjects):
(WebCore::ScriptExecutionContext::resumeActiveDOMObjects):
(WebCore::ScriptExecutionContext::stopActiveDOMObjects):
* dom/ScriptExecutionContext.h:
(ScriptExecutionContext):
* fileapi/DOMFileSystem.cpp:
(WebCore::DOMFileSystem::create):
* fileapi/FileReader.cpp:
(WebCore):
* fileapi/FileReader.h:
(WebCore::FileReader::create):
* fileapi/FileWriter.cpp:
(WebCore):
* fileapi/FileWriter.h:
(WebCore::FileWriter::create):
* history/CachedFrame.cpp:
(WebCore::CachedFrame::CachedFrame):
* html/HTMLAudioElement.cpp:
(WebCore::HTMLAudioElement::create):
(WebCore::HTMLAudioElement::createForJSConstructor):
* html/HTMLMarqueeElement.cpp:
(WebCore::HTMLMarqueeElement::create):
* html/HTMLVideoElement.cpp:
(WebCore::HTMLVideoElement::create):
* mediastream/PeerConnection.cpp:
(WebCore::PeerConnection::create):
* notifications/Notification.cpp:
(WebCore::Notification::create):
* notifications/NotificationCenter.cpp:
* notifications/NotificationCenter.h:
(WebCore::NotificationCenter::create):
* page/DOMTimer.cpp:
(WebCore::DOMTimer::install):
(WebCore::DOMTimer::fired):
* page/EventSource.cpp:
(WebCore::EventSource::create):
* page/SuspendableTimer.cpp:
(WebCore::SuspendableTimer::SuspendableTimer):
* storage/IDBDatabase.cpp:
(WebCore::IDBDatabase::create):
* storage/IDBRequest.cpp:
(WebCore::IDBRequest::create):
* storage/IDBTransaction.cpp:
(WebCore::IDBTransaction::create):
* storage/IDBVersionChangeRequest.cpp:
(WebCore::IDBVersionChangeRequest::create):
* webaudio/AudioContext.cpp:
(WebCore::AudioContext::create):
* websockets/WebSocket.cpp:
* websockets/WebSocket.h:
(WebCore::WebSocket::create):
* workers/SharedWorker.cpp:
(WebCore::SharedWorker::create):
* workers/Worker.cpp:
(WebCore::Worker::create):
* xml/XMLHttpRequest.cpp:
(WebCore::XMLHttpRequest::create):

LayoutTests:

* fast/events/resources/suspend-subframe-1.html: Removed.
* fast/events/resources/suspend-subframe-2.html: Removed.
* fast/events/suspend-timers-expected.txt: Removed.
* fast/events/suspend-timers.html: Removed.
* platform/chromium/test_expectations.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@106823 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoMove style related functions from NamedNodeMap to ElementAttributeData
caio.oliveira@openbossa.org [Mon, 6 Feb 2012 18:35:54 +0000 (18:35 +0000)]
Move style related functions from NamedNodeMap to ElementAttributeData
https://bugs.webkit.org/show_bug.cgi?id=77879

Reviewed by Ryosuke Niwa.

This commit also removes the NamedNodeMap::mappedAttributeCount(),
which is a leftover from removal of mapped attributes.

* dom/ElementAttributeData.cpp:
(WebCore::ElementAttributeData::ensureInlineStyleDecl):
(WebCore):
(WebCore::ElementAttributeData::destroyInlineStyleDecl):
(WebCore::ElementAttributeData::ensureAttributeStyle):
* dom/ElementAttributeData.h:
(WebCore::ElementAttributeData::inlineStyleDecl):
(ElementAttributeData):
(WebCore::ElementAttributeData::attributeStyle):
* dom/NamedNodeMap.cpp:
* dom/NamedNodeMap.h:
* dom/StyledElement.h:
(StyledElement):
(WebCore::StyledElement::inlineStyleDecl):
(WebCore::StyledElement::ensureInlineStyleDecl):
(WebCore::StyledElement::attributeStyle):
(WebCore::StyledElement::ensureAttributeStyle):
(WebCore::StyledElement::destroyInlineStyleDecl):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@106819 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoCrash in SubframeLoader::loadSubframe.
inferno@chromium.org [Mon, 6 Feb 2012 18:23:01 +0000 (18:23 +0000)]
Crash in SubframeLoader::loadSubframe.
https://bugs.webkit.org/show_bug.cgi?id=77345

Reviewed by Nate Chapin.

Source/WebCore:

Mutation event when loading subframe can blow away the
main frame. Add a RefPtr to protect against that.

Test: fast/frames/subframe-load-crash-main.html

* loader/SubframeLoader.cpp:
(WebCore::SubframeLoader::loadSubframe):

LayoutTests:

* fast/frames/resources/subframe-load-crash.svg: Added.
* fast/frames/subframe-load-crash-main-expected.txt: Added.
* fast/frames/subframe-load-crash-main.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@106818 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago[GTK] WebKitWebView does a lot of work during size_allocate when not mapped
commit-queue@webkit.org [Mon, 6 Feb 2012 17:59:27 +0000 (17:59 +0000)]
[GTK] WebKitWebView does a lot of work during size_allocate when not mapped
https://bugs.webkit.org/show_bug.cgi?id=77743

Patch by Martin Robinson <mrobinson@igalia.com> on 2012-02-06
Reviewed by Gustavo Noronha Silva.

Instead of resizing the guts of a WebView when it's not mapped, wait
until it's mapped and do one resize. This prevents unmapped WebViews
from adding to a container's resize cost.

Source/WebKit/gtk:

* webkit/webkitwebview.cpp:
(resizeWebViewFromAllocation): Abstracted out this helper.
(webkit_web_view_size_allocate): Wait until map to resize umapped widgets.
(webkitWebViewMap): Added this vmethod implementation.
(webkit_web_view_class_init): Added vmethod.
* webkit/webkitwebviewprivate.h: Added new needsResizeOnMap member.

Source/WebKit2:

* UIProcess/API/gtk/WebKitWebViewBase.cpp:
(_WebKitWebViewBasePrivate): Added a new member needsResizeOnMap.
(resizeWebKitWebViewBaseFromAllocation): Abstracted out this helper.
(webkitWebViewBaseSizeAllocate): Wait until map to resize unmapped WebViews.
(webkitWebViewBaseMap): Added this vmethod implementation.
(webkit_web_view_base_class_init): Added vmethod.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@106816 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoInitial implementation of GraphicsContext3DOpenGLES.cpp
commit-queue@webkit.org [Mon, 6 Feb 2012 17:49:59 +0000 (17:49 +0000)]
Initial implementation of GraphicsContext3DOpenGLES.cpp
https://bugs.webkit.org/show_bug.cgi?id=76248

Patch by ChangSeok Oh <shivamidow@gmail.com> on 2012-02-06
Reviewed by Martin Robinson.

Implemented APIs in GraphicsContext3DOpenGLES.cpp according to the GLES spec. roughly.
But no way to run these codes right now, because they need an extra port
specific implementation to work. I plan to add these extra codes for the GTK port
in the next patch. And also this patch doesn't support anti-aliasing yet. Another bug
will deal with it.
Moved some APIs in GraphicsContext3DOpenGLES.cpp to GraphicsContext3DCommon.cpp.
It looks it could be shared between gl and gles.
Two helper functions are added to avoid code duplication in GraphicsContext3D.
Added a missing period at the end of comment lines.

No new tests required.
We'll be able to verify this patch by using the existing webgl test cases.

* platform/graphics/GraphicsContext3D.h: Add build flag to access stencilBuffer & depthBuffer for gles.
(WebCore):
* platform/graphics/opengl/GraphicsContext3DOpenGL.cpp:
(WebCore::GraphicsContext3D::reshapeFBOs): Helper function extracted from GC3D::reshape to resize regular & multisampled FBOs.
(WebCore::GraphicsContext3D::resolveMultisamplingIfNecessary): Helper function to resolve multisampling.
* platform/graphics/opengl/GraphicsContext3DOpenGLCommon.cpp:
(WebCore::GraphicsContext3D::validateAttributes): Add condition to disable antialiasing & packedDepthStencilExtension for GLES.
(WebCore::GraphicsContext3D::paintRenderingResultsToCanvas):
(WebCore::GraphicsContext3D::paintRenderingResultsToImageData):

Following APIs looked shareable so that moved into GC3DOpenGLCommon.cpp.
(WebCore::GraphicsContext3D::prepareTexture):
(WebCore):
(WebCore::GraphicsContext3D::readRenderingResults):
(WebCore::GraphicsContext3D::reshape):
(WebCore::GraphicsContext3D::bindFramebuffer):
(WebCore::GraphicsContext3D::copyTexImage2D):
(WebCore::GraphicsContext3D::copyTexSubImage2D):
(WebCore::GraphicsContext3D::getActiveUniform):
(WebCore::GraphicsContext3D::readPixels):

Added a missing period at the end of comment line.
(WebCore::GraphicsContext3D::compileShader):
(WebCore::GraphicsContext3D::getActiveAttrib):
(WebCore::GraphicsContext3D::uniform2fv):
(WebCore::GraphicsContext3D::uniform3fv):
(WebCore::GraphicsContext3D::uniform4fv):
(WebCore::GraphicsContext3D::uniform2iv):
(WebCore::GraphicsContext3D::uniform3iv):
(WebCore::GraphicsContext3D::uniform4iv):
(WebCore::GraphicsContext3D::uniformMatrix2fv):
(WebCore::GraphicsContext3D::uniformMatrix3fv):
(WebCore::GraphicsContext3D::uniformMatrix4fv):
(WebCore::GraphicsContext3D::texSubImage2D):

* platform/graphics/opengl/GraphicsContext3DOpenGLES.cpp:
(WebCore::GraphicsContext3D::reshapeFBOs): Same with the above.
(WebCore::GraphicsContext3D::resolveMultisamplingIfNecessary): Same with the above.

Brief explanation about what the differences are between gl and gles.
(WebCore::GraphicsContext3D::renderbufferStorage): Removed codes for converting GLES parameter to GL parameter.
(WebCore::GraphicsContext3D::getIntegerv): Removed codes that emulate GLES.
(WebCore::GraphicsContext3D::texImage2D): Removed codes for converting GLES parameter to GL parameter.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@106815 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago[webkitpy] cpp style-checker complains about readability/naming in GStreamer elements...
philn@webkit.org [Mon, 6 Feb 2012 17:33:55 +0000 (17:33 +0000)]
[webkitpy] cpp style-checker complains about readability/naming in GStreamer elements source files
https://bugs.webkit.org/show_bug.cgi?id=77866

Reviewed by Martin Robinson.

* Scripts/webkitpy/style/checker.py: Ignore readability/naming
errors in some gstreamer source code files defining new GStreamer elements.
* Scripts/webkitpy/style/checker_unittest.py:
(GlobalVariablesTest.test_path_rules_specifier):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@106814 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago[Qt][WK2] http/tests/loading/state-object-security-exception.html fails
jesus@webkit.org [Mon, 6 Feb 2012 17:05:27 +0000 (17:05 +0000)]
[Qt][WK2] http/tests/loading/state-object-security-exception.html fails
https://bugs.webkit.org/show_bug.cgi?id=62731

Rubber stamped by Csaba Osztrogonác.

Unskipping tests that were kept as skipped even after
https://bugs.webkit.org/show_bug.cgi?id=42333 was fixed.

* platform/wk2/Skipped:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@106812 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoWeb Inspector: [Meta] Allow emulation of touch events
apavlov@chromium.org [Mon, 6 Feb 2012 16:57:13 +0000 (16:57 +0000)]
Web Inspector: [Meta] Allow emulation of touch events
https://bugs.webkit.org/show_bug.cgi?id=77096

Reviewed by Pavel Feldman.

This is the inspector part of the change, core part implementation tracked in bug 77105.

* English.lproj/localizedStrings.js:
* inspector/Inspector.json:
* inspector/InspectorDOMAgent.cpp:
(DOMAgentState):
(WebCore::InspectorDOMAgent::clearFrontend):
(WebCore::InspectorDOMAgent::restore):
(WebCore):
(WebCore::InspectorDOMAgent::updateTouchEventEmulationInDocuments):
(WebCore::InspectorDOMAgent::setTouchEmulationEnabled):
* inspector/InspectorDOMAgent.h:
(InspectorDOMAgent):
* inspector/front-end/DOMAgent.js:
(WebInspector.DOMAgent.prototype._captureDOM):
(WebInspector.DOMAgent.prototype._emulateTouchEventsChanged):
* inspector/front-end/Settings.js:
(WebInspector.Settings):
* inspector/front-end/SettingsScreen.js:
(WebInspector.SettingsScreen):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@106811 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago[Blackberry] Non-supported about: operations never stops loading
commit-queue@webkit.org [Mon, 6 Feb 2012 16:39:23 +0000 (16:39 +0000)]
[Blackberry] Non-supported about: operations never stops loading
https://bugs.webkit.org/show_bug.cgi?id=76366

Patch by Chris Guan <chris.guan@torchmobile.com.cn> on 2012-02-06
Reviewed by Rob Buis.

If user typed a non-supported "about:" scheme such as "about:nonsupport",
the loadAboutURL() function in NetworkManger should recognize and handle it as
an error of invalid url.

* platform/network/blackberry/NetworkJob.cpp:
(WebCore::NetworkJob::loadAboutURL):
(WebCore::NetworkJob::handleAbout):
* platform/network/blackberry/NetworkJob.h:
(NetworkJob):
* platform/network/blackberry/NetworkManager.cpp:
(WebCore::NetworkManager::startJob):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@106810 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoWeb Inspector: Remove unused disabledComputedProperties from methods in StylesSidebarPane
apavlov@chromium.org [Mon, 6 Feb 2012 16:26:58 +0000 (16:26 +0000)]
Web Inspector: Remove unused disabledComputedProperties from methods in StylesSidebarPane
https://bugs.webkit.org/show_bug.cgi?id=77876

Reviewed by Pavel Feldman.

* inspector/front-end/StylesSidebarPane.js:
(WebInspector.StylesSidebarPane.prototype._refreshUpdate):
(WebInspector.StylesSidebarPane.prototype._rebuildUpdate):
(WebInspector.StylesSidebarPane.prototype._markUsedProperties):
(WebInspector.StylesSidebarPane.prototype._refreshSectionsForStyleRules):
(WebInspector.StylesSidebarPane.prototype._rebuildSectionsForStyleRules):
(WebInspector.StylePropertiesSection.prototype.onpopulate):
(WebInspector.ComputedStylePropertiesSection):
(WebInspector.ComputedStylePropertiesSection.prototype._isPropertyInherited):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@106809 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoWeb Inspector: Redesign summary view / retaining tree contents
loislo@chromium.org [Mon, 6 Feb 2012 16:20:23 +0000 (16:20 +0000)]
Web Inspector: Redesign summary view / retaining tree contents
https://bugs.webkit.org/show_bug.cgi?id=77870

Patch by Alexei Filippov <alexeif@chromium.org> on 2012-02-06
Reviewed by Pavel Feldman.

1. Make object IDs less contrast.
2. Put array indices in [].
3. Do not write type if it's just "Object".

Source/WebCore:

* inspector/front-end/DetailedHeapshotGridNodes.js:
(WebInspector.HeapSnapshotGenericObjectNode.prototype._createObjectCell):
(WebInspector.HeapSnapshotGenericObjectNode.prototype.get data):
(WebInspector.HeapSnapshotObjectNode.prototype._emptyData):
(WebInspector.HeapSnapshotObjectNode.prototype._enhanceData):
(WebInspector.HeapSnapshotObjectNode.prototype._prefixObjectCell):
(WebInspector.HeapSnapshotConstructorNode.prototype.get data):
* inspector/front-end/heapProfiler.css:
(.detailed-heapshot-view .console-formatted-id):
(.detailed-heapshot-view td.object-column span.grayed):

LayoutTests:

* inspector/profiler/detailed-heapshots-test.js:
(initialize_DetailedHeapshotTest.):
(initialize_DetailedHeapshotTest):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@106808 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago[Texmap][Qt] Avoid an image copy when uploading textures in WebKit1
noam.rosenthal@nokia.com [Mon, 6 Feb 2012 16:10:13 +0000 (16:10 +0000)]
[Texmap][Qt] Avoid an image copy when uploading textures in WebKit1
https://bugs.webkit.org/show_bug.cgi?id=77748

Reviewed by Kenneth Rohde Christiansen.

Use QPixmap::buffer() API to get access to the QPixmap's pixels without implicit copies.

Instrumentation shows that the deep image copies created from TextureMapperGL are
eliminated.

* platform/graphics/opengl/TextureMapperGL.cpp:
(WebCore::BitmapTextureGL::updateContents):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@106807 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoLayoutTest failures on r106797
tonyg@chromium.org [Mon, 6 Feb 2012 16:05:53 +0000 (16:05 +0000)]
LayoutTest failures on r106797
https://bugs.webkit.org/show_bug.cgi?id=77868

Call suspendIfNeeded from subclass of IDBRequest, IDBVersionChangeRequest.

Patch by Allan Sandfeld Jensen <allan.jensen@nokia.com> on 2012-02-06
Reviewed by Simon Hausmann.

* storage/IDBVersionChangeRequest.cpp:
(WebCore::IDBVersionChangeRequest::create):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@106806 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago[GTK] Make sure html_dir exists before calling gtkdoc-rebase
carlosgc@webkit.org [Mon, 6 Feb 2012 14:54:38 +0000 (14:54 +0000)]
[GTK] Make sure html_dir exists before calling gtkdoc-rebase
https://bugs.webkit.org/show_bug.cgi?id=77869

Reviewed by Philippe Normand.

* gtk/gtkdoc.py:
(GTKDoc.rebase_installed_docs): Check html_dir exists.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@106805 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago[Qt] Implement ImageBuffer::copyImage(ImageBuffer::DontCopyBackingStore)
noam.rosenthal@nokia.com [Mon, 6 Feb 2012 14:18:14 +0000 (14:18 +0000)]
[Qt] Implement ImageBuffer::copyImage(ImageBuffer::DontCopyBackingStore)
https://bugs.webkit.org/show_bug.cgi?id=77689

Reviewed by Kenneth Rohde Christiansen.

Use StillImageQt::createForRendering when using DontCopyBackingStore.
Enable DontCopyBackingStore in TextureMapperNode.
This removes deep copies resulting from the use of ImageBuffer.

Instrumentation shows that deep image copies resulted from ImageBuffer are eliminated.

* platform/graphics/qt/ImageBufferQt.cpp:
(WebCore::ImageBuffer::copyImage):
* platform/graphics/texmap/TextureMapperNode.cpp:
(WebCore::TextureMapperNode::renderContent):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@106804 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoWebGestureEvent can not encode delta and area.
commit-queue@webkit.org [Mon, 6 Feb 2012 14:07:27 +0000 (14:07 +0000)]
WebGestureEvent can not encode delta and area.
https://bugs.webkit.org/show_bug.cgi?id=77728

Source/WebCore:

Add area field to PlatformGestureEvent.

Patch by Allan Sandfeld Jensen <allan.jensen@nokia.com> on 2012-02-06
Reviewed by Kenneth Rohde Christiansen.

* platform/PlatformGestureEvent.h:
(WebCore::PlatformGestureEvent::PlatformGestureEvent):
(WebCore::PlatformGestureEvent::area):

Source/WebKit2:

Support delta and area fields in WebGestureEvent, and send area with Qt tap gesture.

Patch by Allan Sandfeld Jensen <allan.jensen@nokia.com> on 2012-02-06
Reviewed by Kenneth Rohde Christiansen.

* Shared/WebEvent.h:
(WebKit::WebGestureEvent::area):
(WebKit::WebGestureEvent::delta):
* Shared/WebEventConversion.cpp:
(WebKit::WebKit2PlatformGestureEvent::WebKit2PlatformGestureEvent):
* Shared/WebGestureEvent.cpp:
(WebKit::WebGestureEvent::WebGestureEvent):
(WebKit::WebGestureEvent::encode):
(WebKit::WebGestureEvent::decode):
* UIProcess/qt/QtWebPageEventHandler.cpp:
(QtWebPageEventHandler::handleSingleTapEvent):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@106803 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago[BlackBerry]Use extension for a mimetype as the suggested extension
commit-queue@webkit.org [Mon, 6 Feb 2012 13:55:14 +0000 (13:55 +0000)]
[BlackBerry]Use extension for a mimetype as the suggested extension
if the url file doesn't have an extension.
https://bugs.webkit.org/show_bug.cgi?id=76779

Patch by Charles Wei <charles.wei@torchmobile.com.cn> on 2012-02-06
Reviewed by Antonio Gomes.

No new tests.

* platform/network/blackberry/NetworkJob.cpp:
(WebCore::NetworkJob::sendResponseIfNeeded):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@106802 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoSpeech input: fix fast/speech/bubble-position.html
hans@chromium.org [Mon, 6 Feb 2012 13:24:20 +0000 (13:24 +0000)]
Speech input: fix fast/speech/bubble-position.html
https://bugs.webkit.org/show_bug.cgi?id=77732

Reviewed by Tony Gentilcore.

There is no need for the test to check the pixel-exact position of the
rect, since that seems to vary a little between platforms.
The point of the test is to check that the rect is within the iframe,
so let's do that.

* fast/speech/bubble-position-expected.txt:
* fast/speech/bubble-position.html:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@106801 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago[Qt][WK2] Remove TiledDrawingArea and dependencies.
jocelyn.turcotte@nokia.com [Mon, 6 Feb 2012 13:11:50 +0000 (13:11 +0000)]
[Qt][WK2] Remove TiledDrawingArea and dependencies.
https://bugs.webkit.org/show_bug.cgi?id=71681

Reviewed by Kenneth Rohde Christiansen.

Tiled drawing surface functionality got converted to LayerTree.
TiledDrawingArea and its dependencies are not used by any port anymore.

* CMakeLists.txt:
* GNUmakefile.am:
* Shared/DrawingAreaInfo.h:
* Target.pri:
* UIProcess/API/qt/qquickwebpage.cpp:
(QQuickWebPagePrivate::QQuickWebPagePrivate):
* UIProcess/API/qt/qquickwebpage_p.h:
* UIProcess/API/qt/qquickwebpage_p_p.h:
(QQuickWebPagePrivate):
* UIProcess/DrawingAreaProxy.h:
* UIProcess/DrawingAreaProxy.messages.in:
* UIProcess/TiledDrawingAreaProxy.cpp: Removed.
* UIProcess/TiledDrawingAreaProxy.h: Removed.
* UIProcess/qt/QtSGTileNode.cpp: Removed.
* UIProcess/qt/QtSGTileNode.h: Removed.
* UIProcess/qt/QtSGUpdateQueue.cpp: Removed.
* UIProcess/qt/QtSGUpdateQueue.h: Removed.
* UIProcess/qt/TiledDrawingAreaProxyQt.cpp: Removed.
* WebProcess/WebPage/DrawingArea.cpp:
(WebKit::DrawingArea::create):
* WebProcess/WebPage/DrawingArea.h:
* WebProcess/WebPage/DrawingArea.messages.in:
* WebProcess/WebPage/TiledDrawingArea.cpp: Removed.
* WebProcess/WebPage/TiledDrawingArea.h: Removed.
* WebProcess/WebPage/WebPage.h:
(WebPage):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@106800 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago[Qt] Don't version QtWebKit.experimental.
jocelyn.turcotte@nokia.com [Mon, 6 Feb 2012 13:09:27 +0000 (13:09 +0000)]
[Qt] Don't version QtWebKit.experimental.
https://bugs.webkit.org/show_bug.cgi?id=77739

Reviewed by Tor Arne Vestbø.

Source/WebKit/qt:

We won't support previous versions of the experimental API.

* declarative/experimental/plugin.cpp:

Source/WebKit2:

* UIProcess/API/qt/tests/qmltests/DesktopBehavior/tst_itemSelector.qml:
* UIProcess/API/qt/tests/qmltests/DesktopBehavior/tst_messaging.qml:
* UIProcess/API/qt/tests/qmltests/DesktopBehavior/tst_navigationRequested.qml:
* UIProcess/API/qt/tests/qmltests/WebView/tst_applicationScheme.qml:
* UIProcess/API/qt/tests/qmltests/WebView/tst_download.qml:
* UIProcess/API/qt/tests/qmltests/WebView/tst_geopermission.qml:
* UIProcess/API/qt/tests/qmltests/WebView/tst_javaScriptDialogs.qml:
* UIProcess/API/qt/tests/qmltests/WebView/tst_navigationHistory.qml:
* UIProcess/API/qt/tests/qmltests/WebView/tst_preferences.qml:

Tools:

* MiniBrowser/qt/qml/BrowserWindow.qml:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@106799 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoRename [HasIndexGetter], [HasNameGetter] and [HasCustomIndexSetter] IDLs
haraken@chromium.org [Mon, 6 Feb 2012 12:56:07 +0000 (12:56 +0000)]
Rename [HasIndexGetter], [HasNameGetter] and [HasCustomIndexSetter] IDLs
https://bugs.webkit.org/show_bug.cgi?id=77848

Reviewed by Adam Barth.

For naming consistency with [CustomGetter] and [CustomSetter],
this patch renames the following IDLs:

[HasIndexGetter] => [IndexedGetter]  (Remove "Has". This IDL is for "indexed" properties
in the Web IDL: http://dev.w3.org/2006/webapi/WebIDL/#idl-indexed-properties)
[HasCustomIndexSetter] => [CustomIndexedSetter]  (Ditto.)
[HasNameGetter] => [NamedGetter] (Remove "Has". This IDL is for "named" properties
in the Web IDL: http://dev.w3.org/2006/webapi/WebIDL/#idl-named-properties)

No tests. No change in behavior.

* bindings/scripts/CodeGeneratorJS.pm:
(GenerateGetOwnPropertySlotBody):
(GenerateGetOwnPropertyDescriptorBody):
(GenerateHeader):
(GenerateImplementation):
* bindings/scripts/CodeGeneratorV8.pm:
(GenerateHeaderNamedAndIndexedPropertyAccessors):
(GenerateImplementationIndexer):
(GenerateImplementationNamedPropertyGetter):

* bindings/scripts/test/TestEventTarget.idl: No change in run-bindings-tests results.
* bindings/scripts/test/TestTypedArray.idl: Ditto.

* Modules/gamepad/GamepadList.idl:
* css/CSSRuleList.idl:
* css/CSSStyleDeclaration.idl:
* css/CSSValueList.idl:
* css/MediaList.idl:
* css/StyleSheetList.idl:
* css/WebKitCSSFilterValue.idl:
* css/WebKitCSSKeyframesRule.idl:
* css/WebKitCSSTransformValue.idl:
* dom/ClientRectList.idl:
* dom/DOMStringList.idl:
* dom/DOMStringMap.idl:
* dom/DataTransferItemList.idl:
* dom/NamedNodeMap.idl:
* dom/NodeList.idl:
* dom/TouchList.idl:
* fileapi/EntryArray.idl:
* fileapi/EntryArraySync.idl:
* fileapi/FileList.idl:
* html/DOMSettableTokenList.idl:
* html/DOMTokenList.idl:
* html/HTMLAllCollection.idl:
* html/HTMLCollection.idl:
* html/HTMLFormElement.idl:
* html/HTMLOptionsCollection.idl:
* html/HTMLPropertiesCollection.idl:
* html/HTMLSelectElement.idl:
* html/TextTrackCueList.idl:
* html/canvas/CanvasPixelArray.idl:
* html/canvas/Float32Array.idl:
* html/canvas/Float64Array.idl:
* html/canvas/Int16Array.idl:
* html/canvas/Int32Array.idl:
* html/canvas/Int8Array.idl:
* html/canvas/Uint16Array.idl:
* html/canvas/Uint32Array.idl:
* html/canvas/Uint8Array.idl:
* html/canvas/Uint8ClampedArray.idl:
* html/track/TextTrackList.idl:
* mediastream/MediaStreamList.idl:
* mediastream/MediaStreamTrackList.idl:
* page/SpeechInputResultList.idl:
* page/WebKitAnimationList.idl:
* plugins/DOMMimeTypeArray.idl:
* plugins/DOMPlugin.idl:
* plugins/DOMPluginArray.idl:
* storage/Storage.idl:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@106798 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoSource/WebCore: Ensure timers and other active DOM objects do not fire in suspended...
commit-queue@webkit.org [Mon, 6 Feb 2012 12:36:41 +0000 (12:36 +0000)]
Source/WebCore: Ensure timers and other active DOM objects do not fire in suspended documents.
https://bugs.webkit.org/show_bug.cgi?id=53733

ScriptExecutionContext now remembers it has suspended active DOM objects
and suspends all newly installed active DOM objects as well.

All create-calls active DOM objects now calls the post constructor method
suspendIfNeeded that updates the suspend state. It is post constructor
because the suspend/resume functions are virtual and thus can not be called
from constructors.

Patch by Allan Sandfeld Jensen <allan.jensen@nokia.com> on 2012-02-06
Reviewed by Mihai Parparita.

Test: fast/events/suspend-timers.html

* Modules/intents/IntentRequest.cpp:
(WebCore::IntentRequest::create):
* bindings/generic/ActiveDOMCallback.cpp:
(WebCore::ActiveDOMCallback::ActiveDOMCallback):
* dom/ActiveDOMObject.cpp:
(WebCore::ActiveDOMObject::ActiveDOMObject):
(WebCore::ActiveDOMObject::~ActiveDOMObject):
(WebCore::ActiveDOMObject::suspendIfNeeded):
* dom/ActiveDOMObject.h:
(WebCore::ActiveDOMObject::suspendIfNeededCalled):
* dom/DocumentEventQueue.cpp:
(WebCore::DocumentEventQueue::DocumentEventQueue):
* dom/ScriptExecutionContext.cpp:
(WebCore::ScriptExecutionContext::ScriptExecutionContext):
(WebCore::ScriptExecutionContext::~ScriptExecutionContext):
(WebCore::ScriptExecutionContext::canSuspendActiveDOMObjects):
(WebCore::ScriptExecutionContext::suspendActiveDOMObjects):
(WebCore::ScriptExecutionContext::resumeActiveDOMObjects):
(WebCore::ScriptExecutionContext::stopActiveDOMObjects):
(WebCore::ScriptExecutionContext::suspendActiveDOMObjectIfNeeded):
* dom/ScriptExecutionContext.h:
(WebCore::ScriptExecutionContext::activeDOMObjectsAreSuspended):
* fileapi/DOMFileSystem.cpp:
(WebCore::DOMFileSystem::create):
* fileapi/FileReader.cpp:
(WebCore::FileReader::create):
* fileapi/FileReader.h:
* fileapi/FileWriter.cpp:
(WebCore::FileWriter::create):
* fileapi/FileWriter.h:
* history/CachedFrame.cpp:
(WebCore::CachedFrame::CachedFrame):
* html/HTMLAudioElement.cpp:
(WebCore::HTMLAudioElement::create):
* html/HTMLMarqueeElement.cpp:
(WebCore::HTMLMarqueeElement::create):
* html/HTMLVideoElement.cpp:
(WebCore::HTMLVideoElement::create):
* mediastream/PeerConnection.cpp:
(WebCore::PeerConnection::create):
* notifications/Notification.cpp:
(WebCore::Notification::create):
* notifications/NotificationCenter.cpp:
(WebCore::NotificationCenter::create):
* notifications/NotificationCenter.h:
* page/DOMTimer.cpp:
(WebCore::DOMTimer::install):
(WebCore::DOMTimer::fired):
* page/EventSource.cpp:
(WebCore::EventSource::create):
* page/SuspendableTimer.cpp:
(WebCore::SuspendableTimer::SuspendableTimer):
* storage/IDBDatabase.cpp:
(WebCore::IDBDatabase::create):
* storage/IDBRequest.cpp:
(WebCore::IDBRequest::create):
* storage/IDBTransaction.cpp:
(WebCore::IDBTransaction::create):
* webaudio/AudioContext.cpp:
(WebCore::AudioContext::create):
* websockets/WebSocket.cpp:
(WebCore::WebSocket::create):
* websockets/WebSocket.h:
* workers/SharedWorker.cpp:
(WebCore::SharedWorker::create):
* workers/Worker.cpp:
(WebCore::Worker::create):
* xml/XMLHttpRequest.cpp:
(WebCore::XMLHttpRequest::create):

LayoutTests: Test that timers do not fire from subframes of suspended documents.
https://bugs.webkit.org/show_bug.cgi?id=53733

Patch by Allan Sandfeld Jensen <allan.jensen@nokia.com> on 2012-02-06
Reviewed by Mihai Parparita.

* fast/events/resources/suspend-subframe-1.html: Added.
* fast/events/resources/suspend-subframe-2.html: Added.
* fast/events/suspend-timers-expected.txt: Added.
* fast/events/suspend-timers.html: Added.
* platform/chromium/test_expectations.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@106797 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoUnreviewed WinCE build fix after r106197.
paroga@webkit.org [Mon, 6 Feb 2012 12:33:47 +0000 (12:33 +0000)]
Unreviewed WinCE build fix after r106197.

* tools/CodeProfiling.cpp:
(JSC::CodeProfiling::notifyAllocator): getenv() isn't supported by WinCE. Don't call it.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@106796 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoAdding FFTFrameGStreamer.cpp that I forgot to commit
philn@webkit.org [Mon, 6 Feb 2012 11:18:00 +0000 (11:18 +0000)]
Adding FFTFrameGStreamer.cpp that I forgot to commit
in r106537. That new file was reviewed in https://bugs.webkit.org/show_bug.cgi?id=73545.

* platform/audio/gstreamer/FFTFrameGStreamer.cpp: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@106795 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoAdd <shadow> element, which is guarded by SHADOW_DOM flag.
hayato@chromium.org [Mon, 6 Feb 2012 10:39:29 +0000 (10:39 +0000)]
Add <shadow> element, which is guarded by SHADOW_DOM flag.
https://bugs.webkit.org/show_bug.cgi?id=76435

Reviewed by Dimitri Glazkov.

Test: fast/dom/shadow/shadow-element.html

Source/WebCore:

* CMakeLists.txt:
* DerivedSources.cpp:
* DerivedSources.make:
* DerivedSources.pri:
* GNUmakefile.list.am:
* Target.pri:
* WebCore.gypi:
* WebCore.vcproj/WebCore.vcproj:
* WebCore.xcodeproj/project.pbxproj:
* html/HTMLElementsAllInOne.cpp:
* html/HTMLTagNames.in:
* html/shadow/HTMLShadowElement.cpp: Added.
(WebCore):
(WebCore::HTMLShadowElement::HTMLShadowElement):
(WebCore::HTMLShadowElement::create):
(WebCore::HTMLShadowElement::~HTMLShadowElement):
* html/shadow/HTMLShadowElement.h: Added.
(WebCore):
(HTMLShadowElement):
* html/shadow/HTMLShadowElement.idl: Added.

LayoutTests:

* fast/dom/shadow/shadow-element-expected.txt: Added.
* fast/dom/shadow/shadow-element.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@106794 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago[GTK] Update NEWS and configure.ac for 1.7.5 release
carlosgc@webkit.org [Mon, 6 Feb 2012 10:26:05 +0000 (10:26 +0000)]
[GTK] Update NEWS and configure.ac for 1.7.5 release
https://bugs.webkit.org/show_bug.cgi?id=77859

Reviewed by Philippe Normand.

.:

* configure.ac: Bumped version number.

Source/WebKit/gtk:

* NEWS: Added release notes for 1.7.5.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@106793 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agohttps://bugs.webkit.org/show_bug.cgi?id=76995.
commit-queue@webkit.org [Mon, 6 Feb 2012 10:18:57 +0000 (10:18 +0000)]
https://bugs.webkit.org/show_bug.cgi?id=76995.
WebKit fails IETC :indeterminate and input type=radio test.

All input types should respect indeterminate property(getter and setter). This is true with other browsers.
Indeterminate appearance for radio input type is supported by IOS platform. Webkit need not support indeterminate appearance
for radio input type on other platforms.

Patch by Joe Thomas <joethomas@motorola.com> on 2012-02-06
Reviewed by Kent Tamura.

Source/WebCore:

Test: fast/forms/indeterminate-input-types.html

* html/CheckboxInputType.cpp:
(WebCore::CheckboxInputType::supportsIndeterminateAppearance): Checks indeterminate appearance is supported.
(WebCore):
* html/CheckboxInputType.h:
(CheckboxInputType):
* html/HTMLInputElement.cpp:
(WebCore::HTMLInputElement::setIndeterminate): Removed the check isCheckable().
(WebCore::HTMLInputElement::isIndeterminate): Checks indeterminate appearance is supported.
(WebCore):
* html/HTMLInputElement.h:
(HTMLInputElement):
* html/InputType.cpp:
(WebCore::InputType::supportsIndeterminateAppearance): Checks indeterminate appearance is supported.
(WebCore):
* html/InputType.h:
(InputType):
* html/RadioInputType.cpp:
(WebCore::RadioInputType::willDispatchClick): Indeterminate related code is moved to IOS specific.
(WebCore::RadioInputType::didDispatchClick): Ditto.
(WebCore::RadioInputType::supportsIndeterminateAppearance): Checks indeterminate appearance is supported.
(WebCore):
* html/RadioInputType.h:
(RadioInputType):

LayoutTests:

* fast/forms/indeterminate-input-types-expected.txt: Added.
* fast/forms/indeterminate-input-types.html: Added.
* fast/forms/indeterminate-radio.html:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@106792 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago[BlackBerry] Remove unused variable in PluginViewBlackBerry
commit-queue@webkit.org [Mon, 6 Feb 2012 10:02:03 +0000 (10:02 +0000)]
[BlackBerry] Remove unused variable in PluginViewBlackBerry
https://bugs.webkit.org/show_bug.cgi?id=77847

Patch by Robin Cao <robin.cao@torchmobile.com.cn> on 2012-02-06
Reviewed by Kentaro Hara.

No changes in behavior, so no new tests.

* plugins/blackberry/PluginViewBlackBerry.cpp:
(WebCore::PluginView::updatePluginWidget):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@106791 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago[BlackBerry] Remove unused variable in BackingStore
commit-queue@webkit.org [Mon, 6 Feb 2012 09:58:05 +0000 (09:58 +0000)]
[BlackBerry] Remove unused variable in BackingStore
https://bugs.webkit.org/show_bug.cgi?id=77851

Patch by Robin Cao <robin.cao@torchmobile.com.cn> on 2012-02-06
Reviewed by Kentaro Hara.

* blackberry/Api/BackingStore.cpp:
(BlackBerry::WebKit::BackingStorePrivate::blitContents):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@106790 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2012-02-06 Nikolas Zimmermann <nzimmermann@rim.com>
zimmermann@webkit.org [Mon, 6 Feb 2012 09:40:14 +0000 (09:40 +0000)]
2012-02-06  Nikolas Zimmermann  <nzimmermann@rim.com>

        Not reviewed. Switch to unix file endings and at a trailing newline - hopefully the bots will accept my follow up patch then.

        * svg/filters/invalidate-on-child-layout.svg:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@106789 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago[GTK] Fix several gtkdoc-fixxref warnings
carlosgc@webkit.org [Mon, 6 Feb 2012 09:29:34 +0000 (09:29 +0000)]
[GTK] Fix several gtkdoc-fixxref warnings
https://bugs.webkit.org/show_bug.cgi?id=77613

Reviewed by Martin Robinson.

Source/WebKit/gtk:

* docs/webkitgtk-docs.sgml: Remove webkitspellcheckerenchant
section.
* docs/webkitgtk-sections.txt: Remove webkitspellcheckerenchant
symbols and move WebKitWebNavigationReason from the private
section to the global section.
* webkit/webkitwebnavigationaction.h: Document
WebKitWebNavigationReason enum.

Source/WebKit2:

* UIProcess/API/gtk/WebKitWindowProperties.cpp: Fix a typo, a
colon was missing in WebKitWebView::ready-to-show.

Tools:

* gtk/generate-gtkdoc:
(get_webkit1_options): Add webkitspellcheckerenchant.* to the list
of ignored files, since it's private.
* gtk/gtkdoc.py:
(GTKDoc._ignored_files_basenames): Helper function that returns a
string with the list of ignored files basenames separated by a
spaces, as expected by several gtkdoc commands.
(GTKDoc._run_gtkdoc_scan): Use _ignored_files_basenames for
--ignore-headers option.
(GTKDoc._run_gtkdoc_mkdb): Use _ignored_files_basenames for
--ignore-files option.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@106788 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoWeb Inspector: Fix CallArgument type in Inspector.json
commit-queue@webkit.org [Mon, 6 Feb 2012 09:27:50 +0000 (09:27 +0000)]
Web Inspector: Fix CallArgument type in Inspector.json
https://bugs.webkit.org/show_bug.cgi?id=77839

Patch by Peter Rybin <peter.rybin@gmail.com> on 2012-02-06
Reviewed by Yury Semikhatsky.

Protocol description is fixed: field type is changed. No changes to
program code are necessary.

* inspector/Inspector-0.1.json:
* inspector/Inspector-1.0.json:
* inspector/Inspector.json:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@106787 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago[GTK] Fix xrefs after installing API documentation
carlosgc@webkit.org [Mon, 6 Feb 2012 09:23:08 +0000 (09:23 +0000)]
[GTK] Fix xrefs after installing API documentation
https://bugs.webkit.org/show_bug.cgi?id=77551

Reviewed by Martin Robinson.

* GNUmakefile.am: Call generate-gtkdoc --rebase after installing
api docs.
* gtk/common.py:
(prefix_of_pkg_config_file): Get the prefix variable of the given
pkg-config file.
(gtk_version_of_pkg_config_file): Get the gtk version required by
the given pkg-config file.
* gtk/generate-gtkdoc: Add --rebase command line option to rebase
installed documentation.
(get_gtkdoc_module_paths): Get paths where API doc is installed
for the dependencies of the given pkg-config file.
(get_common_xref_deps): Get API doc directories of dependencies
common to WebKit1 and WebKit2.
(get_webkit2_options): Add cross_reference_deps option.
(get_webkit1_options): Ditto
(rebase_installed_docs): Helper function to create a generator for
the given pkg-config file and options and call rebase_installed_docs.
* gtk/gtkdoc.py:
(GTKDoc.__init__): Initialize cross_reference_deps.
(GTKDoc._run_gtkdoc_fixxref): Add API doc directories of
dependencies.
(GTKDoc.rebase_installed_docs): Call gtkdoc-rebase to fix xref
links of installed documentation.
(PkgConfigGTKDoc.__init__): Get the prefix from the pkg-config
file.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@106786 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoLayout Test http/tests/security/xssAuditor/dom-write-innerHTML.html is flaky
abarth@webkit.org [Mon, 6 Feb 2012 09:22:26 +0000 (09:22 +0000)]
Layout Test http/tests/security/xssAuditor/dom-write-innerHTML.html is flaky
https://bugs.webkit.org/show_bug.cgi?id=67538

Reviewed by Kentaro Hara.

Try to fix the flakiness in this test by kicking off the <img> load
before the load event has fired for the inner frame.  That way, the
load event for the main frame will be delayed until the image finishes
loading (and its error handler has fired), removing the need for the
setTimeout.

* http/tests/security/xssAuditor/dom-write-innerHTML.html:
* http/tests/security/xssAuditor/resources/echo-dom-write-innerHTML.html:
* platform/chromium/test_expectations.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@106785 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoIncorrect statistics shown when running run-webkit-tests with --repeat-each or -...
sergio@webkit.org [Mon, 6 Feb 2012 09:07:20 +0000 (09:07 +0000)]
Incorrect statistics shown when running run-webkit-tests with --repeat-each or --iterations
https://bugs.webkit.org/show_bug.cgi?id=77672

Reviewed by Dirk Pranke.

Test repetitions must be taken into account when working out
the statistics shown by run-webkit-tests.

* Scripts/webkitpy/layout_tests/controllers/manager.py:
(Manager.prepare_lists_and_print_output):
(Manager._print_result_summary):
* Scripts/webkitpy/layout_tests/models/result_summary.py:
(ResultSummary.__init__):
(ResultSummary.add):
* Scripts/webkitpy/layout_tests/run_webkit_tests_integrationtest.py:
(MainTest.test_repeat_each_iterations_num_tests):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@106784 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoRemove JSObject defineGetter/defineSetter lookupGetter/lookupSetter
barraclough@apple.com [Mon, 6 Feb 2012 08:56:02 +0000 (08:56 +0000)]
Remove JSObject defineGetter/defineSetter lookupGetter/lookupSetter
https://bugs.webkit.org/show_bug.cgi?id=77451

Reviewed by Sam Weinig.

These can now all be implemented in terms of defineOwnProperty & getPropertyDescriptor.
Also remove initializeGetterSetterProperty, since this is equivalent to putDirectAccessor.

Source/JavaScriptCore:

* JavaScriptCore.exp:
* debugger/DebuggerActivation.cpp:
(JSC::DebuggerActivation::defineOwnProperty):
* debugger/DebuggerActivation.h:
(DebuggerActivation):
* runtime/ClassInfo.h:
(MethodTable):
(JSC):
* runtime/JSBoundFunction.cpp:
(JSC::JSBoundFunction::finishCreation):
* runtime/JSCell.cpp:
(JSC):
* runtime/JSCell.h:
(JSCell):
* runtime/JSFunction.cpp:
(JSC::JSFunction::getOwnPropertySlot):
(JSC::JSFunction::getOwnPropertyDescriptor):
* runtime/JSGlobalObject.cpp:
(JSC::JSGlobalObject::defineOwnProperty):
(JSC):
* runtime/JSGlobalObject.h:
(JSGlobalObject):
* runtime/JSObject.cpp:
(JSC):
* runtime/JSObject.h:
(JSObject):
* runtime/ObjectPrototype.cpp:
(JSC::objectProtoFuncDefineGetter):
(JSC::objectProtoFuncDefineSetter):
(JSC::objectProtoFuncLookupGetter):
(JSC::objectProtoFuncLookupSetter):

Source/WebCore:

* bindings/js/JSDOMWindowCustom.cpp:
(WebCore):
(WebCore::JSDOMWindow::defineOwnProperty):
* bindings/js/JSDOMWindowShell.cpp:
(WebCore):
* bindings/js/JSDOMWindowShell.h:
(JSDOMWindowShell):
* bindings/js/JSLocationCustom.cpp:
(WebCore::JSLocation::defineOwnProperty):
(WebCore::JSLocationPrototype::defineOwnProperty):
* bindings/scripts/CodeGeneratorJS.pm:
(GenerateHeader):
* page/DOMWindow.idl:
* page/Location.idl:

Source/WebKit/qt:

* Api/qwebframe.cpp:
(QWebFramePrivate::addQtSenderToGlobalObject):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@106783 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoSource/JavaScriptCore: Unreviewed. Fix make distcheck.
carlosgc@webkit.org [Mon, 6 Feb 2012 08:51:19 +0000 (08:51 +0000)]
Source/JavaScriptCore: Unreviewed. Fix make distcheck.

* GNUmakefile.list.am: Add missing files.

Source/WebCore: Unreviewed. Fix make distrcheck.

* GNUmakefile.am: Fix typo in inspector json file.
* GNUmakefile.list.am: Remove non existent file.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@106782 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2012-02-06 Yury Semikhatsky <yurys@chromium.org>
yurys@chromium.org [Mon, 6 Feb 2012 08:04:25 +0000 (08:04 +0000)]
2012-02-06  Yury Semikhatsky  <yurys@chromium.org>

        Unreviewed. Qt minimal compilation fix after r106777.

        * inspector/InspectorDatabaseAgent.h: added missing #if ENABLED guards.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@106780 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoWeb Inspector: REGRESSION detached dom nodes aren't highlighted
loislo@chromium.org [Mon, 6 Feb 2012 07:54:53 +0000 (07:54 +0000)]
Web Inspector: REGRESSION detached dom nodes aren't highlighted
https://bugs.webkit.org/show_bug.cgi?id=77829

This functionality was lost when I replaced (Native roots) with (Detached DOM trees) on v8 side.

Reviewed by Yury Semikhatsky.

* inspector/front-end/HeapSnapshot.js:
(WebInspector.HeapSnapshotNode.prototype.get isDetachedDOMTreesRoot):
(WebInspector.HeapSnapshot.prototype._markDetachedDOMTreeNodes):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@106779 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoWeb Inspector: get rid of artificial heap snapshot nodes from the retaining tree.
loislo@chromium.org [Mon, 6 Feb 2012 07:52:59 +0000 (07:52 +0000)]
Web Inspector: get rid of artificial heap snapshot nodes from the retaining tree.
https://bugs.webkit.org/show_bug.cgi?id=77850

Reviewed by Yury Semikhatsky.

* inspector/front-end/DetailedHeapshotGridNodes.js:
(WebInspector.HeapSnapshotObjectNode):
(WebInspector.HeapSnapshotObjectNode.prototype.updateHasChildren):
(WebInspector.HeapSnapshotObjectNode.prototype._prefixObjectCell):
* inspector/front-end/HeapSnapshot.js:
(WebInspector.HeapSnapshotNode.prototype.get isArtificial):
(WebInspector.HeapSnapshot.prototype._init):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@106778 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoWeb Inspector: CodeGeneratorInspector.py: switch domain agents to formal interfaces...
yurys@chromium.org [Mon, 6 Feb 2012 07:39:40 +0000 (07:39 +0000)]
Web Inspector: CodeGeneratorInspector.py: switch domain agents to formal interfaces in BackendDispatcher
https://bugs.webkit.org/show_bug.cgi?id=77444

Patch by Peter Rybin <peter.rybin@gmail.com> on 2012-02-05
Reviewed by Yury Semikhatsky.

All agent code is switched to formal interfaces. Actual agent class
names and includes are dropped from InspectorBackendDispatcher.h.

* inspector/CodeGeneratorInspector.py:
(DomainNameFixes.get_fixed_data):
(Generator.go):
* inspector/InspectorAgent.h:
* inspector/InspectorApplicationCacheAgent.h:
* inspector/InspectorCSSAgent.cpp:
(WebCore::InspectorCSSAgent::getMatchedStylesForNode):
* inspector/InspectorCSSAgent.h:
(InspectorCSSAgent):
* inspector/InspectorConsoleAgent.h:
* inspector/InspectorDOMAgent.cpp:
(WebCore::InspectorDOMAgent::highlightNode):
* inspector/InspectorDOMAgent.h:
(InspectorDOMAgent):
* inspector/InspectorDOMDebuggerAgent.h:
* inspector/InspectorDOMStorageAgent.h:
* inspector/InspectorDatabaseAgent.h:
* inspector/InspectorDebuggerAgent.cpp:
(WebCore::InspectorDebuggerAgent::setBreakpoint):
(WebCore::InspectorDebuggerAgent::continueToLocation):
* inspector/InspectorDebuggerAgent.h:
(InspectorDebuggerAgent):
* inspector/InspectorFileSystemAgent.h:
* inspector/InspectorIndexedDBAgent.h:
* inspector/InspectorMemoryAgent.h:
* inspector/InspectorPageAgent.h:
* inspector/InspectorProfilerAgent.cpp:
(WebCore::InspectorProfilerAgent::getProfile):
(WebCore::InspectorProfilerAgent::removeProfile):
* inspector/InspectorProfilerAgent.h:
(InspectorProfilerAgent):
* inspector/InspectorResourceAgent.cpp:
(WebCore::InspectorResourceAgent::setExtraHTTPHeaders):
* inspector/InspectorResourceAgent.h:
(InspectorResourceAgent):
* inspector/InspectorRuntimeAgent.h:
* inspector/InspectorTimelineAgent.cpp:
(WebCore::InspectorTimelineAgent::start):
* inspector/InspectorTimelineAgent.h:
(InspectorTimelineAgent):
* inspector/InspectorWorkerAgent.cpp:
(WebCore::InspectorWorkerAgent::sendMessageToWorker):
* inspector/InspectorWorkerAgent.h:
(InspectorWorkerAgent):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@106777 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoRename [TreatNullAs=EmptyString] to [TreatNullAs=NullString],
haraken@chromium.org [Mon, 6 Feb 2012 07:06:00 +0000 (07:06 +0000)]
Rename [TreatNullAs=EmptyString] to [TreatNullAs=NullString],
[TreatUndefinedAs=EmptyString] to [TreatUndefinedAs=NullString]
https://bugs.webkit.org/show_bug.cgi?id=77611

Reviewed by Adam Barth.

While the Web IDL spec requires [TreatNullAs=EmptyString] and [TreatUndefinedAs=EmptyString],
the current WebKit treats them as (not an empty string but) a null string.
To avoid confusion, this patch renames [TreatNullAs=EmptyString] to [TreatNullAs=NullString],
and [TreatUndefinedAs=EmptyString] to [TreatUndefinedAs=NullString].
(Eventually we should fix WebKit so that it uses an empty string and then
rename them to [TreatNullAs=EmptyString] and [TreatUndefinedAs=EmptyString].)

No tests. No change in behavior.

* bindings/scripts/CodeGeneratorJS.pm: Added FIXME.
(JSValueToNative):
* bindings/scripts/CodeGeneratorV8.pm: Ditto.
(GetNativeTypeFromSignature):

* bindings/scripts/test/TestObj.idl:
* css/CSSCharsetRule.idl:
* css/CSSPageRule.idl:
* css/CSSRule.idl:
* css/CSSStyleDeclaration.idl:
* css/CSSStyleRule.idl:
* css/CSSValue.idl:
* css/MediaList.idl:
* css/WebKitCSSKeyframesRule.idl:
* dom/Attr.idl:
* dom/CharacterData.idl:
* dom/DOMImplementation.idl:
* dom/Document.idl:
* dom/Element.idl:
* dom/NamedNodeMap.idl:
* dom/Node.idl:
* dom/ProcessingInstruction.idl:
* dom/ShadowRoot.idl:
* fileapi/Blob.idl:
* fileapi/DirectoryEntry.idl:
* fileapi/DirectoryEntrySync.idl:
* fileapi/Entry.idl:
* fileapi/EntrySync.idl:
* fileapi/WebKitBlobBuilder.idl:
* html/HTMLAnchorElement.idl:
* html/HTMLButtonElement.idl:
* html/HTMLCanvasElement.idl:
* html/HTMLDocument.idl:
* html/HTMLElement.idl:
* html/HTMLFieldSetElement.idl:
* html/HTMLFormElement.idl:
* html/HTMLFrameElement.idl:
* html/HTMLInputElement.idl:
* html/HTMLKeygenElement.idl:
* html/HTMLMediaElement.idl:
* html/HTMLObjectElement.idl:
* html/HTMLOutputElement.idl:
* html/HTMLScriptElement.idl:
* html/HTMLSelectElement.idl:
* html/HTMLTextAreaElement.idl:
* html/HTMLTitleElement.idl:
* html/canvas/CanvasRenderingContext2D.idl:
* page/Console.idl:
* page/DOMWindow.idl:
* storage/StorageEvent.idl:
* svg/SVGAngle.idl:
* svg/SVGElement.idl:
* svg/SVGLength.idl:
* svg/SVGScriptElement.idl:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@106776 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoDFG's child references from one node to another should have room for type information
fpizlo@apple.com [Mon, 6 Feb 2012 06:44:24 +0000 (06:44 +0000)]
DFG's child references from one node to another should have room for type information
https://bugs.webkit.org/show_bug.cgi?id=77797

Reviewed by Oliver Hunt.

The DFG::Node::child fields now contain both a DFG::NodeIndex (which is just an unsigned)
and a DFG::UseKind (which is currently an effectively empty enum). They are encapsulated
together as a DFG::NodeUse, which can in most cases still be used as an index (for
example DFG::Graph, AbstractState, and SpeculativeJIT all accept NodeUse in most places
where they really want a NodeIndex).

The NodeUse stores both the index and the UseKind without bloating the memory usage of
DFG::Node, since we really don't need full 32 bits for the NodeIndex (a DFG::Node is
roughly 11 words, so if we assume that we never want to use more than 1GB to DFG compile
something - likely a sensible assumption! - then we will only be able to have room for
about 24 million nodes, which means we only need about 24.5 bits for the node index).
Currently the DFG::NodeUse allocates 4 bits for the UseKind and 28 bits for the index,
but stores the index as a signed number to make NoNode work naturally. Hence we really
just have 27 bits for the index.

This is performance-neutral on all benchmarks we track.

* JavaScriptCore.xcodeproj/project.pbxproj:
* dfg/DFGAbstractState.h:
(JSC::DFG::AbstractState::forNode):
(AbstractState):
* dfg/DFGByteCodeParser.cpp:
(JSC::DFG::ByteCodeParser::getLocal):
(JSC::DFG::ByteCodeParser::getArgument):
(JSC::DFG::ByteCodeParser::toInt32):
(JSC::DFG::ByteCodeParser::addVarArgChild):
(JSC::DFG::ByteCodeParser::processPhiStack):
* dfg/DFGCommon.h:
* dfg/DFGGraph.cpp:
(JSC::DFG::Graph::dump):
(DFG):
* dfg/DFGGraph.h:
(Graph):
(JSC::DFG::Graph::operator[]):
(JSC::DFG::Graph::at):
(JSC::DFG::Graph::ref):
(JSC::DFG::Graph::deref):
(JSC::DFG::Graph::clearAndDerefChild1):
(JSC::DFG::Graph::clearAndDerefChild2):
(JSC::DFG::Graph::clearAndDerefChild3):
* dfg/DFGJITCompiler.h:
(JSC::DFG::JITCompiler::getPrediction):
* dfg/DFGNode.h:
(JSC::DFG::Node::Node):
(JSC::DFG::Node::child1):
(JSC::DFG::Node::child1Unchecked):
(JSC::DFG::Node::child2):
(JSC::DFG::Node::child3):
(JSC::DFG::Node::firstChild):
(JSC::DFG::Node::numChildren):
(JSC::DFG::Node::dumpChildren):
(Node):
* dfg/DFGNodeReferenceBlob.h: Added.
(DFG):
(NodeReferenceBlob):
(JSC::DFG::NodeReferenceBlob::NodeReferenceBlob):
(JSC::DFG::NodeReferenceBlob::child):
(JSC::DFG::NodeReferenceBlob::child1):
(JSC::DFG::NodeReferenceBlob::child2):
(JSC::DFG::NodeReferenceBlob::child3):
(JSC::DFG::NodeReferenceBlob::child1Unchecked):
(JSC::DFG::NodeReferenceBlob::initialize):
(JSC::DFG::NodeReferenceBlob::firstChild):
(JSC::DFG::NodeReferenceBlob::setFirstChild):
(JSC::DFG::NodeReferenceBlob::numChildren):
(JSC::DFG::NodeReferenceBlob::setNumChildren):
* dfg/DFGNodeUse.h: Added.
(DFG):
(NodeUse):
(JSC::DFG::NodeUse::NodeUse):
(JSC::DFG::NodeUse::indexUnchecked):
(JSC::DFG::NodeUse::index):
(JSC::DFG::NodeUse::setIndex):
(JSC::DFG::NodeUse::useKind):
(JSC::DFG::NodeUse::setUseKind):
(JSC::DFG::NodeUse::isSet):
(JSC::DFG::NodeUse::operator!):
(JSC::DFG::NodeUse::operator==):
(JSC::DFG::NodeUse::operator!=):
(JSC::DFG::NodeUse::shift):
(JSC::DFG::NodeUse::makeWord):
(JSC::DFG::operator==):
(JSC::DFG::operator!=):
* dfg/DFGPropagator.cpp:
(JSC::DFG::Propagator::propagateArithNodeFlags):
(JSC::DFG::Propagator::vote):
(JSC::DFG::Propagator::toDouble):
(JSC::DFG::Propagator::fixupNode):
(JSC::DFG::Propagator::canonicalize):
(JSC::DFG::Propagator::startIndex):
(JSC::DFG::Propagator::globalVarLoadElimination):
(JSC::DFG::Propagator::getByValLoadElimination):
(JSC::DFG::Propagator::getByOffsetLoadElimination):
(JSC::DFG::Propagator::performSubstitution):
(JSC::DFG::Propagator::performNodeCSE):
* dfg/DFGScoreBoard.h:
(JSC::DFG::ScoreBoard::use):
* dfg/DFGSpeculativeJIT.cpp:
(JSC::DFG::SpeculativeJIT::useChildren):
(JSC::DFG::SpeculativeJIT::writeBarrier):
(JSC::DFG::SpeculativeJIT::nonSpeculativeStrictEq):
(JSC::DFG::SpeculativeJIT::compilePeepHoleObjectEquality):
(JSC::DFG::SpeculativeJIT::compilePeepHoleIntegerBranch):
(JSC::DFG::SpeculativeJIT::compileMovHint):
(JSC::DFG::SpeculativeJIT::computeValueRecoveryFor):
(JSC::DFG::SpeculativeJIT::compilePutByValForByteArray):
(JSC::DFG::SpeculativeJIT::compilePutByValForIntTypedArray):
(JSC::DFG::SpeculativeJIT::compilePutByValForFloatTypedArray):
(JSC::DFG::SpeculativeJIT::compileSoftModulo):
(JSC::DFG::SpeculativeJIT::compileAdd):
(JSC::DFG::SpeculativeJIT::compileArithSub):
(JSC::DFG::SpeculativeJIT::compileStrictEqForConstant):
(JSC::DFG::SpeculativeJIT::compileStrictEq):
* dfg/DFGSpeculativeJIT.h:
(JSC::DFG::SpeculativeJIT::at):
(JSC::DFG::SpeculativeJIT::canReuse):
(JSC::DFG::SpeculativeJIT::use):
(SpeculativeJIT):
(JSC::DFG::SpeculativeJIT::detectPeepHoleBranch):
(JSC::DFG::SpeculativeJIT::speculationCheck):
(JSC::DFG::SpeculativeJIT::terminateSpeculativeExecution):
(JSC::DFG::IntegerOperand::IntegerOperand):
(JSC::DFG::DoubleOperand::DoubleOperand):
(JSC::DFG::JSValueOperand::JSValueOperand):
(JSC::DFG::StorageOperand::StorageOperand):
(JSC::DFG::SpeculateIntegerOperand::SpeculateIntegerOperand):
(JSC::DFG::SpeculateStrictInt32Operand::SpeculateStrictInt32Operand):
(JSC::DFG::SpeculateDoubleOperand::SpeculateDoubleOperand):
(JSC::DFG::SpeculateCellOperand::SpeculateCellOperand):
(JSC::DFG::SpeculateBooleanOperand::SpeculateBooleanOperand):
* dfg/DFGSpeculativeJIT32_64.cpp:
(JSC::DFG::SpeculativeJIT::nonSpeculativeValueToNumber):
(JSC::DFG::SpeculativeJIT::nonSpeculativeValueToInt32):
(JSC::DFG::SpeculativeJIT::cachedPutById):
(JSC::DFG::SpeculativeJIT::nonSpeculativeNonPeepholeCompareNull):
(JSC::DFG::SpeculativeJIT::nonSpeculativePeepholeBranchNull):
(JSC::DFG::SpeculativeJIT::nonSpeculativeCompareNull):
(JSC::DFG::SpeculativeJIT::nonSpeculativePeepholeBranch):
(JSC::DFG::SpeculativeJIT::nonSpeculativeNonPeepholeCompare):
(JSC::DFG::SpeculativeJIT::nonSpeculativePeepholeStrictEq):
(JSC::DFG::SpeculativeJIT::nonSpeculativeNonPeepholeStrictEq):
(JSC::DFG::SpeculativeJIT::emitCall):
(JSC::DFG::SpeculativeJIT::compileValueAdd):
(JSC::DFG::SpeculativeJIT::compileObjectOrOtherLogicalNot):
(JSC::DFG::SpeculativeJIT::compileLogicalNot):
(JSC::DFG::SpeculativeJIT::emitObjectOrOtherBranch):
(JSC::DFG::SpeculativeJIT::emitBranch):
(JSC::DFG::SpeculativeJIT::compile):
* dfg/DFGSpeculativeJIT64.cpp:
(JSC::DFG::SpeculativeJIT::nonSpeculativeValueToNumber):
(JSC::DFG::SpeculativeJIT::nonSpeculativeValueToInt32):
(JSC::DFG::SpeculativeJIT::cachedPutById):
(JSC::DFG::SpeculativeJIT::nonSpeculativeNonPeepholeCompareNull):
(JSC::DFG::SpeculativeJIT::nonSpeculativePeepholeBranchNull):
(JSC::DFG::SpeculativeJIT::nonSpeculativeCompareNull):
(JSC::DFG::SpeculativeJIT::nonSpeculativePeepholeBranch):
(JSC::DFG::SpeculativeJIT::nonSpeculativeNonPeepholeCompare):
(JSC::DFG::SpeculativeJIT::nonSpeculativePeepholeStrictEq):
(JSC::DFG::SpeculativeJIT::nonSpeculativeNonPeepholeStrictEq):
(JSC::DFG::SpeculativeJIT::emitCall):
(JSC::DFG::SpeculativeJIT::compileObjectEquality):
(JSC::DFG::SpeculativeJIT::compileValueAdd):
(JSC::DFG::SpeculativeJIT::compileObjectOrOtherLogicalNot):
(JSC::DFG::SpeculativeJIT::compileLogicalNot):
(JSC::DFG::SpeculativeJIT::emitObjectOrOtherBranch):
(JSC::DFG::SpeculativeJIT::emitBranch):
(JSC::DFG::SpeculativeJIT::compile):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@106775 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoxss-DENIED-from-data-url-in-foreign-domain-subframe.html is flaky
abarth@webkit.org [Mon, 6 Feb 2012 05:55:10 +0000 (05:55 +0000)]
xss-DENIED-from-data-url-in-foreign-domain-subframe.html is flaky
https://bugs.webkit.org/show_bug.cgi?id=77849

Reviewed by Kentaro Hara.

This test doesn't seem to be sequenced correctly w.r.t. the load events
of the various frames involved.  This patch re-orders some of the
triggers in the hopes of making the test no longer flaky.

* http/tests/security/dataURL/resources/foreign-domain-data-url-accessor-iframe.html:
* http/tests/security/dataURL/xss-DENIED-from-data-url-in-foreign-domain-subframe-expected.txt:
* http/tests/security/dataURL/xss-DENIED-from-data-url-in-foreign-domain-subframe.html:
* platform/chromium/http/tests/security/dataURL/xss-DENIED-from-data-url-in-foreign-domain-subframe-expected.txt:
* platform/chromium/test_expectations.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@106774 268f45cc-cd09-0410-ab3c-d52691b4dbfc