review.tizen.org Git - profile/ivi/webkit-efl.git/log

Implement Function.prototype.bind
https://bugs.webkit.org/show_bug.cgi?id=26382

Reviewed by Sam Weinig.

Source/JavaScriptCore:

This patch provides a basic functional implementation
for Function.bind. It should (hopefully!) be fully
functionally correct, and the bound functions can be
called to quickly (since they are a subclass of
JSFunction, not InternalFunction), but we'll probably
want to follow up with some optimization work to keep
bound calls in JIT code.

* JavaScriptCore.JSVALUE32_64only.exp:
* JavaScriptCore.JSVALUE64only.exp:
* JavaScriptCore.exp:
* JavaScriptCore.xcodeproj/project.pbxproj:
* jit/JITStubs.cpp:
(JSC::JITThunks::hostFunctionStub):
* jit/JITStubs.h:
* jsc.cpp:
(GlobalObject::addFunction):
* runtime/CommonIdentifiers.h:
* runtime/ConstructData.h:
* runtime/Executable.h:
(JSC::NativeExecutable::NativeExecutable):
* runtime/FunctionPrototype.cpp:
(JSC::FunctionPrototype::addFunctionProperties):
(JSC::functionProtoFuncBind):
* runtime/FunctionPrototype.h:
* runtime/JSBoundFunction.cpp: Added.
(JSC::boundFunctionCall):
(JSC::boundFunctionConstruct):
(JSC::JSBoundFunction::create):
(JSC::JSBoundFunction::hasInstance):
(JSC::JSBoundFunction::getOwnPropertySlot):
(JSC::JSBoundFunction::getOwnPropertyDescriptor):
(JSC::JSBoundFunction::JSBoundFunction):
(JSC::JSBoundFunction::finishCreation):
* runtime/JSBoundFunction.h: Added.
(JSC::JSBoundFunction::targetFunction):
(JSC::JSBoundFunction::boundThis):
(JSC::JSBoundFunction::boundArgs):
(JSC::JSBoundFunction::createStructure):
* runtime/JSFunction.cpp:
(JSC::JSFunction::create):
(JSC::JSFunction::finishCreation):
(JSC::createDescriptorForThrowingProperty):
(JSC::JSFunction::getOwnPropertySlot):
* runtime/JSFunction.h:
* runtime/JSGlobalData.cpp:
(JSC::JSGlobalData::getHostFunction):
* runtime/JSGlobalData.h:
* runtime/JSGlobalObject.cpp:
(JSC::JSGlobalObject::reset):
(JSC::JSGlobalObject::visitChildren):
* runtime/JSGlobalObject.h:
(JSC::JSGlobalObject::boundFunctionStructure):
* runtime/Lookup.cpp:
(JSC::setUpStaticFunctionSlot):

Source/WebCore:

Test: fast/js/function-bind.html

* bindings/js/JSDOMBinding.cpp:
(WebCore::objectToStringFunctionGetter):
* bindings/js/JSDOMWindowCustom.cpp:
(WebCore::nonCachingStaticFunctionGetter):
* bindings/js/JSHistoryCustom.cpp:
(WebCore::nonCachingStaticBackFunctionGetter):
(WebCore::nonCachingStaticForwardFunctionGetter):
(WebCore::nonCachingStaticGoFunctionGetter):
* bindings/js/JSLocationCustom.cpp:
(WebCore::nonCachingStaticReplaceFunctionGetter):
(WebCore::nonCachingStaticReloadFunctionGetter):
(WebCore::nonCachingStaticAssignFunctionGetter):
- Function::create no longer requires functionStructure() to be passed.

LayoutTests:

We now pass Function.bind tests.

* fast/js/Object-getOwnPropertyNames-expected.txt:
* fast/js/basic-strict-mode-expected.txt:
* fast/js/function-bind-expected.txt: Added.
* fast/js/function-bind.html: Added.
* fast/js/mozilla/strict/15.3.4.5-expected.txt:
* fast/js/script-tests/function-bind.js: Added.
* ietestcenter/Javascript/15.2.3.3-4-38-expected.txt:
* ietestcenter/Javascript/15.3.4.5-0-1-expected.txt:
* ietestcenter/Javascript/15.3.4.5-0-2-expected.txt:
* ietestcenter/Javascript/15.3.4.5-13.b-1-expected.txt:
* ietestcenter/Javascript/15.3.4.5-13.b-2-expected.txt:
* ietestcenter/Javascript/15.3.4.5-13.b-3-expected.txt:
* ietestcenter/Javascript/15.3.4.5-13.b-4-expected.txt:
* ietestcenter/Javascript/15.3.4.5-13.b-5-expected.txt:
* ietestcenter/Javascript/15.3.4.5-15-1-expected.txt:
* ietestcenter/Javascript/15.3.4.5-15-2-expected.txt:
* ietestcenter/Javascript/15.3.4.5-16-1-expected.txt:
* ietestcenter/Javascript/15.3.4.5-2-1-expected.txt:
* ietestcenter/Javascript/15.3.4.5-2-2-expected.txt:
* ietestcenter/Javascript/15.3.4.5-2-3-expected.txt:
* ietestcenter/Javascript/15.3.4.5-2-4-expected.txt:
* ietestcenter/Javascript/15.3.4.5-2-5-expected.txt:
* ietestcenter/Javascript/15.3.4.5-2-6-expected.txt:
* ietestcenter/Javascript/15.3.4.5-2-7-expected.txt:
* ietestcenter/Javascript/15.3.4.5-2-8-expected.txt:
* ietestcenter/Javascript/15.3.4.5-2-9-expected.txt:
* ietestcenter/Javascript/15.3.4.5-8-1-expected.txt:
* ietestcenter/Javascript/15.3.4.5-8-2-expected.txt:
* ietestcenter/Javascript/15.3.4.5-9-1-expected.txt:
* ietestcenter/Javascript/15.3.4.5-9-2-expected.txt:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@95751 268f45cc-cd09-0410-ab3c-d52691b4dbfc

https://bugs.webkit.org/show_bug.cgi?id=68650

Make determineStartPosition smart enough to not skip over clean lines that have moved to a new
region with a different available content logical width. When this happens, we go ahead and treat
the line as dirty.

Reviewed by Sam Weinig.

Source/WebCore:

Covered by existing fast/regions tests.

* rendering/RenderBlock.cpp:
(WebCore::RenderBlock::layoutBlock):
Disable region fitting if we detect that the regions all have uniform widths. In this case we know
the content width can't vary, so there's no reason to waste time worrying about it.

(WebCore::RenderBlock::computeLogicalLocationForFloat):
Pull the content width checking code into computeLogicalLocationForFloat, since we've incorporated
region fitting into the fixed offsets now rather than the line functions.

(WebCore::RenderBlock::positionNewFloats):
Changed because more code moved from here into computeLogicalLocationForFloat.

(WebCore::RenderBlock::adjustForRegionFittingIfNeeded):
No changes. Just moved the function above the left offset function.

(WebCore::RenderBlock::logicalLeftOffsetForContent):
(WebCore::RenderBlock::logicalRightOffsetForContent):
Added. These new functions give the fixed left and right offsets for lines in a particular region.
When no regions are present or when all the regions have a uniform width, they are identical
to the versions of the functions that take no arguments.

(WebCore::RenderBlock::logicalRightOffsetForLine):
Changed to no longer do region fitting, since that's built into logicalRightOffsetForContent now.

(WebCore::RenderBlock::lineWidthForPaginatedLineChanged):
New function that takes a root line box and compares its cached old content width with the width
available at the line's new location.

* rendering/RenderBlock.h:
(WebCore::RenderBlock::logicalRightOffsetForLine):
(WebCore::RenderBlock::logicalLeftOffsetForLine):
Modified to call the logicalXXXOffsetForContent functions that take a block direction position.

(WebCore::RenderBlock::logicalRightOffsetForContent):
(WebCore::RenderBlock::logicalLeftOffsetForContent):
Added the new functions that take a position so that they can fit to the correct region.

(WebCore::RenderBlock::availableLogicalWidthForContent):
New helper function that calls right - left (similar to the corresponding line function that includes
floats).

* rendering/RenderBlockLineLayout.cpp:
(WebCore::RenderBlock::determineStartPosition):
Patched to check if a line has moved to a place with a different available content width. If so, the line
is dirtied so that line layout can re-run.

* rendering/RootInlineBox.cpp:
(WebCore::RootInlineBox::RootInlineBox):
(WebCore::RootInlineBox::alignBoxesInBlockDirection):
Sets the paginated line width from the current block logical height.

* rendering/RootInlineBox.h:
(WebCore::RootInlineBox::paginationStrut):
(WebCore::RootInlineBox::setPaginationStrut):
Fixed paginationStrut to be LayoutUnit instead of int.

(WebCore::RootInlineBox::paginatedLineWidth):
(WebCore::RootInlineBox::setPaginatedLineWidth):
Added paginatedLineWidth getter/setter for caching the content width for a given root line.

LayoutTests:

* platform/mac/fast/regions/webkit-flow-floats-inside-regions-bounds-expected.png:
* platform/mac/fast/regions/webkit-flow-floats-inside-regions-bounds-expected.txt:
* platform/mac/fast/regions/webkit-flow-inlines-inside-regions-bounds-expected.png:
* platform/mac/fast/regions/webkit-flow-inlines-inside-regions-bounds-expected.txt:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@95750 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Remove Widget::beforeMouseDown and Widget::afterMouseDown
https://bugs.webkit.org/show_bug.cgi?id=68570

Reviewed by Darin Adler.

These two member functions were originally added to avoid crashes due to removing NSViews while they were
being tracked by AppKit. Since they were added, we've moved away from NSViews for form controls, and the bugs in
AppKit that lead to crashes have been fixed.

This patch was rolled out in r95743 because it broke a bunch of tests. The patch accidentally removed setting back
m_sendingEventToSubview to false.

* page/mac/EventHandlerMac.mm:
(WebCore::EventHandler::passMouseDownEventToWidget):
* platform/Widget.h:
* platform/mac/WidgetMac.mm:
(WebCore::Widget::Widget):
(WebCore::Widget::removeFromSuperview):

* rendering/RenderWidget.cpp:
Change widgetHierarchyUpdateSuspendCount to be an unsigned integer instead of a size_t.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@95749 268f45cc-cd09-0410-ab3c-d52691b4dbfc

    [WK2] UIProcess should check that WebProcess isn't sending unexpected file: URLs to it
        https://bugs.webkit.org/show_bug.cgi?id=68573

        Reviewed by Anders Carlsson.

        Re-landing with a slightly less aggressive check.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@95747 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Platform-specific results for test introduced by r95726

* platform/gtk/fast/inline/left-right-center-inline-alignment-in-ltr-and-rtl-blocks-expected.txt: Added.
* platform/mac/fast/inline/left-right-center-inline-alignment-in-ltr-and-rtl-blocks-expected.txt: Added.
* platform/qt/fast/inline/left-right-center-inline-alignment-in-ltr-and-rtl-blocks-expected.txt: Added.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@95746 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Unreviewed, rolling out r95335 and r95645.
http://trac.webkit.org/changeset/95335
http://trac.webkit.org/changeset/95645
https://bugs.webkit.org/show_bug.cgi?id=68649

Wrong fix (Requested by rniwa on #webkit).

Patch by Sheriff Bot <webkit.review.bot@gmail.com> on 2011-09-22

Source/WebCore:

* editing/DeleteSelectionCommand.cpp:
(WebCore::DeleteSelectionCommand::saveTypingStyleState):
* editing/EditingStyle.cpp:
(WebCore::EditingStyle::removeStyleAddedByNode):
* editing/EditingStyle.h:
* editing/ReplaceSelectionCommand.cpp:
(WebCore::ReplaceSelectionCommand::removeRedundantStylesAndKeepStyleSpanInline):
* editing/markup.cpp:
(WebCore::StyledMarkupAccumulator::serializeNodes):
(WebCore::ancestorToRetainStructureAndAppearance):
* editing/markup.h:

LayoutTests:

* editing/deleting/merge-paragraph-from-address-expected.txt: Removed.
* editing/deleting/merge-paragraph-from-address.html: Removed.
* editing/deleting/merge-paragraph-from-h6-expected.txt: Removed.
* editing/deleting/merge-paragraph-from-h6-with-style-expected.txt: Removed.
* editing/deleting/merge-paragraph-from-h6-with-style.html: Removed.
* editing/deleting/merge-paragraph-from-h6.html: Removed.
* editing/deleting/merge-paragraph-from-listing-expected.txt: Removed.
* editing/deleting/merge-paragraph-from-listing.html: Removed.
* editing/deleting/merge-paragraph-from-p-with-style-expected.txt: Removed.
* editing/deleting/merge-paragraph-from-p-with-style.html: Removed.
* editing/deleting/merge-paragraph-into-h1-expected.txt: Removed.
* editing/deleting/merge-paragraph-into-h1-style-expected.txt: Removed.
* editing/deleting/merge-paragraph-into-h1-with-style-expected.txt: Removed.
* editing/deleting/merge-paragraph-into-h1-with-style.html: Removed.
* editing/deleting/merge-paragraph-into-h1.html: Removed.
* editing/deleting/merge-paragraph-into-pre-expected.txt: Removed.
* editing/deleting/merge-paragraph-into-pre.html: Removed.
* platform/mac/editing/deleting/merge-whitespace-pre-expected.png:
* platform/mac/editing/deleting/merge-whitespace-pre-expected.txt:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@95745 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Unreviewed, update expected results after r95726

* platform/chromium-win/css2.1/20110323/abspos-non-replaced-width-margin-000-expected.png:
* platform/chromium-win/css2.1/20110323/abspos-replaced-width-margin-000-expected.png:
* platform/gtk/css2.1/20110323/abspos-non-replaced-width-margin-000-expected.txt:
* platform/gtk/css2.1/20110323/abspos-replaced-width-margin-000-expected.txt:
* platform/qt/css2.1/20110323/abspos-non-replaced-width-margin-000-expected.txt:
* platform/qt/css2.1/20110323/abspos-replaced-width-margin-000-expected.txt:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@95744 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Revert r95673 since it caused hundreds of tests to fail on Leopard / Snow Leopard.

* page/mac/EventHandlerMac.mm:
(WebCore::EventHandler::passMouseDownEventToWidget):
* platform/Widget.h:
* platform/mac/WidgetMac.mm:
(WebCore::Widget::Widget):
(WebCore::Widget::removeFromSuperview):
(WebCore::Widget::beforeMouseDown):
(WebCore::Widget::afterMouseDown):
* rendering/RenderWidget.cpp:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@95743 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Implement get_scoped_var in the DFG
https://bugs.webkit.org/show_bug.cgi?id=68640

Reviewed by Gavin Barraclough.

Naive implementation of get_scoped_var in the DFG. Essentially this
is the bare minimum required to get correct behaviour, so there's no
load/store coalescing or type profiling involved, even though these
would be wins. No impact on SunSpider or V8.

* dfg/DFGByteCodeParser.cpp:
(JSC::DFG::ByteCodeParser::parseBlock):
* dfg/DFGCapabilities.h:
(JSC::DFG::canCompileOpcode):
* dfg/DFGNode.h:
(JSC::DFG::Node::hasVarNumber):
(JSC::DFG::Node::hasScopeChainDepth):
(JSC::DFG::Node::scopeChainDepth):
* dfg/DFGPropagator.cpp:
(JSC::DFG::Propagator::propagateNodePredictions):
* dfg/DFGSpeculativeJIT.cpp:
(JSC::DFG::SpeculativeJIT::compile):

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@95742 268f45cc-cd09-0410-ab3c-d52691b4dbfc

https://bugs.webkit.org/show_bug.cgi?id=68645

Fix the layout test fast/regions directory. Remove platform-specific results that were
incorrectly checked in to cross-platform locations and add missing pixel test results.

Reviewed by Beth Dakin.

* fast/regions/content-flowed-into-regions-dynamically-inserted-expected.txt: Removed.
* fast/regions/flows-dependency-dynamic-remove-expected.txt: Removed.
* fast/regions/flows-dependency-same-flow-expected.txt: Removed.
* fast/regions/render-region-renderer-expected.txt: Removed.
* fast/regions/webkit-flow-floats-inside-regions-bounds-expected.txt: Removed.
* fast/regions/webkit-flow-inlines-inside-regions-bounds-expected.txt: Removed.
* fast/regions/webkit-flow-inlines-inside-regions-bounds-vertical-expected.txt: Removed.
* fast/regions/webkit-flow-inlines-inside-regions-bounds-vertical-rl-expected.txt: Removed.
* fast/regions/webkit-flow-renderer-expected.txt: Removed.
* fast/regions/webkit-flow-renderer-layer-expected.txt: Removed.
* fast/regions/webkit-flow-renderer-nested-expected.txt: Removed.
* platform/mac/fast/regions/content-flowed-into-regions-dynamically-added-expected.png: Added.
* platform/mac/fast/regions/content-flowed-into-regions-dynamically-inserted-expected.txt: Added.
* platform/mac/fast/regions/content-flowed-into-regions-dynamically-removed-expected.png: Added.
* platform/mac/fast/regions/content-flowed-into-regions-expected.png: Added.
* platform/mac/fast/regions/content-flowed-into-regions-with-dyn-index-expected.png: Added.
* platform/mac/fast/regions/content-flowed-into-regions-with-index-dom-expected.png: Added.
* platform/mac/fast/regions/content-flowed-into-regions-with-index-expected.png: Added.
* platform/mac/fast/regions/flows-dependency-dynamic-remove-expected.png: Added.
* platform/mac/fast/regions/flows-dependency-dynamic-remove-expected.txt: Added.
* platform/mac/fast/regions/flows-dependency-same-flow-expected.png: Added.
* platform/mac/fast/regions/flows-dependency-same-flow-expected.txt: Added.
* platform/mac/fast/regions/no-split-line-box-expected.png: Added.
* platform/mac/fast/regions/render-region-renderer-expected.txt: Added.
* platform/mac/fast/regions/webkit-flow-floats-inside-regions-bounds-expected.png: Added.
* platform/mac/fast/regions/webkit-flow-floats-inside-regions-bounds-expected.txt: Added.
* platform/mac/fast/regions/webkit-flow-inlines-inside-regions-bounds-expected.png: Added.
* platform/mac/fast/regions/webkit-flow-inlines-inside-regions-bounds-expected.txt: Added.
* platform/mac/fast/regions/webkit-flow-inlines-inside-regions-bounds-vertical-expected.png: Added.
* platform/mac/fast/regions/webkit-flow-inlines-inside-regions-bounds-vertical-expected.txt: Added.
* platform/mac/fast/regions/webkit-flow-inlines-inside-regions-bounds-vertical-rl-expected.png: Added.
* platform/mac/fast/regions/webkit-flow-inlines-inside-regions-bounds-vertical-rl-expected.txt: Added.
* platform/mac/fast/regions/webkit-flow-renderer-expected.txt: Added.
* platform/mac/fast/regions/webkit-flow-renderer-layer-expected.txt: Added.
* platform/mac/fast/regions/webkit-flow-renderer-nested-expected.txt: Added.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@95741 268f45cc-cd09-0410-ab3c-d52691b4dbfc

https://bugs.webkit.org/show_bug.cgi?id=68638

Make RenderFlowThread cache whether or not it has regions of varying widths. This will
be relevant for performance as we begin adding code to do custom block painting and
layout based off regions not having the same width.

Reviewed by Dan Bernstein and Adam Roben.

* rendering/RenderFlowThread.cpp:
(WebCore::RenderFlowThread::RenderFlowThread):
(WebCore::RenderFlowThread::layout):
* rendering/RenderFlowThread.h:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@95740 268f45cc-cd09-0410-ab3c-d52691b4dbfc

FrameView::invalidateRect and FrameView::setFrameRect shouldn't take LayoutRects
https://bugs.webkit.org/show_bug.cgi?id=68639

Reviewed by David Hyatt.

FrameView::invalidateRect and FrameView::setFrameRect override Widget and ScrollView
member functions that take IntRects, and Widget coordinates are currently defined in terms of
IntRects (and probably always should since we want them to be pixel-aligned).

* page/FrameView.cpp:
(WebCore::FrameView::invalidateRect):
(WebCore::FrameView::setFrameRect):
* page/FrameView.h:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@95738 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Unreviewed build fix (fixes warning about too many parens).

* rendering/RenderBox.cpp:
(WebCore::RenderBox::computePositionedLogicalWidthReplaced):

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@95737 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Widget::frameRect shouldn't be virtual
https://bugs.webkit.org/show_bug.cgi?id=68637

Reviewed by Sam Weinig.

Nobody overrides Widget::frameRect and nobody should.

* platform/Widget.h:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@95736 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Remove FindSafari

It isn't used or built anymore.

Fixes <http://webkit.org/b/68628>

Reviewed by Darin Adler.

* FindSafari: Removed this directory and its contents.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@95735 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Remove FindSafari from all our .sln files

It isn't used anymore, so there's no point in building it.

Part of <http://webkit.org/b/68628> Remove FindSafari

Reviewed by Steve Falkenburg.

Source/JavaScriptCore:

* JavaScriptCore.vcproj/JavaScriptCore.sln:

Source/WebKit/win:

* WebKit.vcproj/WebKit.sln: Also relinearized the build order while I was at it.

Tools:

* DumpRenderTree/DumpRenderTree.sln:
* WebKitTestRunner/WebKitTestRunner.sln:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@95734 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Remove all uses of FindSafari

It isn't really needed anymore.

Fixes <http://webkit.org/b/68626>.

Reviewed by Darin Adler.

* Scripts/old-run-webkit-tests: Moved read/writeRegistryString from here to webkitdirs.pm.

* Scripts/run-webkit-nightly.cmd: Removed. This used to be used to launch nightly builds of
WebKit, but now we use WebKit.exe for that.

* Scripts/webkitdirs.pm:
(installedSafariPath): Changed to use readRegistryString instead of invoking FindSafari to
do the same thing.

(readRegistryString):
(writeRegistryString):
Moved here from old-run-webkit-tests.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@95733 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Use the "direct object" form of system() to run WebKit.exe

This prevents the shell from parsing (and possibly splitting, if it contains spaces) the
path to WebKit.exe.

Fixes <http://webkit.org/b/68623> run-safari doesn't work if there are spaces in the path to
%WEBKITOUTPUTDIR%

Reviewed by Steve Falkenburg.

* Scripts/webkitdirs.pm:
(runSafari):

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@95732 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Quote the WEBKITLIBRARIESDIR path before passing it to the shell/cygpath

Fixes <http://webkit.org/b/68621> build-webkit goes totally haywire on Windows if there are
spaces in the path to the WebKit source tree

Reviewed by John Sullivan.

* Scripts/webkitdirs.pm:
(setupCygwinEnv):

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@95731 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Add testRegExp.vcproj to WebKit.sln

Fixes <http://webkit.org/b/68403> Windows bots don't build testRegExp.vcproj

Reviewed by Brian Weinstein.

* WebKit.vcproj/WebKit.sln: Added testRegExp. Made it build just after jsc, just like it
does when built as part of JavaScriptCore.sln.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@95730 268f45cc-cd09-0410-ab3c-d52691b4dbfc

[Qt] Use same DPI for application font as rest of app in DRT and WTR

QApplication will initialize the default application font based
on the application DPI at construction time, but we then override
the application DPI using QX11Info (hard-coding it to 96 for
consistency). This hard-coding is not reflected in the application
font, so we explicitly have update the font ourselves.

The 6 test results that are updated were originally produced with
a DPI of 75, as this is the default fallback DPI when a QFont is
constructed before QApplication. This was wrong, and the results
are updated to match a DPI of 96.

Reviewed by Csaba Osztrogonác.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@95729 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Ref protect shaders in V8WebGLRenderingContext::getAttachedShadersCallback
https://bugs.webkit.org/show_bug.cgi?id=68630

Patch by Sergey Glazunov <serg.glazunov@gmail.com> on 2011-09-22
Reviewed by Adam Barth.

Source/WebCore:

Test: fast/canvas/webgl/shader-deleted-by-accessor.html

* bindings/js/JSWebGLRenderingContextCustom.cpp:
(WebCore::JSWebGLRenderingContext::getAttachedShaders):
* bindings/v8/custom/V8WebGLRenderingContextCustom.cpp:
(WebCore::V8WebGLRenderingContext::getAttachedShadersCallback):
* html/canvas/WebGLRenderingContext.cpp:
(WebCore::WebGLRenderingContext::getAttachedShaders):
* html/canvas/WebGLRenderingContext.h:

LayoutTests:

* fast/canvas/webgl/shader-deleted-by-accessor-expected.txt: Added.
* fast/canvas/webgl/shader-deleted-by-accessor.html: Added.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@95728 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Unreviewed, platform-specific results for r95721.

* platform/mac/fast/css/outline-narrowLine-expected.txt: Added.
* platform/gtk/fast/css/outline-narrowLine-expected.txt: Added.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@95727 268f45cc-cd09-0410-ab3c-d52691b4dbfc

CSS 2.1 failure: abspos-non-replaced-width-margin-000, abspos-replaced-width-margin-000
https://bugs.webkit.org/show_bug.cgi?id=47148

Reviewed by David Hyatt.

Source/WebCore:

* rendering/RenderBlock.h:
* rendering/RenderBlockLineLayout.cpp:
(WebCore::RenderBlock::startOffsetForLine): Return offset from right, rather than offset from left + logicalwidth
* rendering/RenderBlockLineLayout.cpp:
(WebCore::RenderBlock::startAlignedOffsetForLine):
      Return the correct alignment for RTL blocks too. Looks like neither FF or Opera do this correctly.
      Covered by new test left-right-center-inline-alignment-in-ltr-and-rtl-blocks.html
* rendering/RenderBox.cpp:
(WebCore::computeInlineStaticDistance):
    - The CSS 2.1 tests now require UAs to use the parent block's direction (rather than the
      containing block's direction) when computing static distance. The container block's direction
      is still used in the other width and offset related calculations.
    - In the context of this function, the logicalRight value is just an element that will be used later to calculate
      the correct logicalLeft position for the RTL block. So whereas an LTR block can substract containerBlock->borderLogicalLeft()
      in this function directly, an RTL block has to add it here so that will be later subtracted from availableSpace to get
      logicalLeft in computePositionedLogicalWidthUsing().

(WebCore::RenderBox::computePositionedLogicalWidth):
   - Update comments and use the container block's direction except when calculating the static distance.
   - Remove the notion of a quirks mode in this function and for calculating the position of replaced and non-replaced
     blocks generally. This is in line with both FF4 and Opera - neither of which have a quirks mode for them.
(WebCore::RenderBox::computePositionedLogicalWidthUsing):
   - Update comments and use the container block's direction except when calculating the static distance.
(WebCore::RenderBox::computePositionedLogicalHeight):
   - Update comments.
(WebCore::RenderBox::computePositionedLogicalWidthReplaced):
   - Update comments and use the container block's direction except when calculating the static distance.
   - If the containing block is RTL, always assume that values are over-constrained.
(WebCore::RenderBox::computePositionedLogicalHeightReplaced):
   - Update comments.

LayoutTests:

* platform/chromium-linux/css2.1/20110323/abspos-non-replaced-width-margin-000-expected.png:
* platform/chromium-linux/css2.1/20110323/abspos-replaced-width-margin-000-expected.png:
  New results for CSS 2.1 reference tests. These match exactly the reference test results.

* fast/inline/left-right-center-inline-alignment-in-ltr-and-rtl-blocks.html: Added.
* platform/chromium-linux/fast/inline/left-right-center-inline-alignment-in-ltr-and-rtl-blocks-expected.png: Added.
* platform/chromium-linux/fast/inline/left-right-center-inline-alignment-in-ltr-and-rtl-blocks-expected.txt: Added.
  New test for left,right and center alignment of inline LTR and RTL blocks. FF and Opera don't pass this test.

* fast/block/positioning/absolute-position-direction-strict.html: Removed.
* fast/block/positioning/absolute-position-direction.html: Renamed from LayoutTests/fast/block/positioning/absolute-position-direction-quirk.html.
* platform/chromium-linux/fast/block/positioning/absolute-position-direction-expected.png: Renamed from LayoutTests/platform/chromium-linux/fast/block/positioning/absolute-position-direction-quirk-expected.png.
* platform/chromium-linux/fast/block/positioning/absolute-position-direction-strict-expected.png: Removed.
* platform/chromium-win/fast/block/positioning/absolute-position-direction-expected.png: Renamed from LayoutTests/platform/chromium-win/fast/block/positioning/absolute-position-direction-quirk-expected.png.
* platform/chromium-win/fast/block/positioning/absolute-position-direction-expected.txt: Renamed from LayoutTests/platform/chromium-win/fast/block/positioning/absolute-position-direction-quirk-expected.txt.
* platform/chromium-win/fast/block/positioning/absolute-position-direction-strict-expected.png: Removed.
* platform/chromium-win/fast/block/positioning/absolute-position-direction-strict-expected.txt: Removed.
* platform/gtk/fast/block/positioning/absolute-position-direction-expected.png: Renamed from LayoutTests/platform/gtk/fast/block/positioning/absolute-position-direction-quirk-expected.png.
* platform/gtk/fast/block/positioning/absolute-position-direction-expected.txt: Renamed from LayoutTests/platform/gtk/fast/block/positioning/absolute-position-direction-quirk-expected.txt.
* platform/gtk/fast/block/positioning/absolute-position-direction-strict-expected.png: Removed.
* platform/gtk/fast/block/positioning/absolute-position-direction-strict-expected.txt: Removed.
* platform/mac-leopard/fast/block/positioning/absolute-position-direction-expected.png: Renamed from LayoutTests/platform/mac-leopard/fast/block/positioning/absolute-position-direction-quirk-expected.png.
* platform/mac-leopard/fast/block/positioning/absolute-position-direction-strict-expected.png: Removed.
* platform/mac/css2.1/20110323/abspos-non-replaced-width-margin-000-expected.txt:
* platform/mac/css2.1/20110323/abspos-replaced-width-margin-000-expected.txt:
* platform/mac/fast/block/positioning/absolute-position-direction-expected.png: Renamed from LayoutTests/platform/mac/fast/block/positioning/absolute-position-direction-quirk-expected.png.
* platform/mac/fast/block/positioning/absolute-position-direction-expected.txt: Renamed from LayoutTests/platform/mac/fast/block/positioning/absolute-position-direction-quirk-expected.txt.
* platform/mac/fast/block/positioning/absolute-position-direction-strict-expected.png: Removed.
* platform/mac/fast/block/positioning/absolute-position-direction-strict-expected.txt: Removed.
* platform/qt/fast/block/positioning/absolute-position-direction-expected.png: Renamed from LayoutTests/platform/qt/fast/block/positioning/absolute-position-direction-quirk-expected.png.
* platform/qt/fast/block/positioning/absolute-position-direction-expected.txt: Renamed from LayoutTests/platform/qt/fast/block/positioning/absolute-position-direction-quirk-expected.txt.
* platform/qt/fast/block/positioning/absolute-position-direction-strict-expected.png: Removed.
* platform/qt/fast/block/positioning/absolute-position-direction-strict-expected.txt: Removed.
  These tests are superseded by abspos-non-replaced-width-margin-000.htm. Using the parent() block's
  direction to compute static distance is now a strict requirement for CSS 2.1, rather than an IE quirk.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@95726 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Get rid of Widget::setBoundsSize
https://bugs.webkit.org/show_bug.cgi?id=68491

Reviewed by Sam Weinig.

This basically reverts http://trac.webkit.org/changeset/79167 and http://trac.webkit.org/changeset/75897
which were added to make zoom gestures work with WebKit1, but since we don't support zoom gestures in WebKit1
anymore, it's better for code cleanliness to get rid of Widget::setBoundsSize and the associated code. While this
will unfortunately break transformed iframes in WebKit1, but Simon says that it's an acceptable tradeoff.

Source/WebCore:

* WebCore.exp.in:
* loader/FrameLoader.cpp:
(WebCore::FrameLoader::open):
* page/FrameView.cpp:
(WebCore::FrameView::create):
* platform/ScrollView.cpp:
(WebCore::ScrollView::visibleContentRect):
(WebCore::ScrollView::updateScrollbars):
(WebCore::ScrollView::wheelEvent):
* platform/ScrollView.h:
* platform/Widget.cpp:
* platform/Widget.h:
(WebCore::Widget::resize):
* platform/chromium/PopupContainer.cpp:
(WebCore::PopupContainer::refresh):
* platform/gtk/ScrollViewGtk.cpp:
(WebCore::ScrollView::visibleContentRect):
* platform/mac/WidgetMac.mm:
(WebCore::Widget::paint):
* rendering/RenderWidget.cpp:
(WebCore::RenderWidget::setWidgetGeometry):
(WebCore::RenderWidget::setWidget):
(WebCore::RenderWidget::updateWidgetPosition):
* rendering/RenderWidget.h:

Source/WebKit/mac:

* Plugins/Hosted/WebHostedNetscapePluginView.mm:
(-[WebHostedNetscapePluginView updateAndSetWindow]):
* WebView/WebFrameView.mm:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@95725 268f45cc-cd09-0410-ab3c-d52691b4dbfc

https://bugs.webkit.org/show_bug.cgi?id=68590

Floats pushed to next page, column or region don't reposition properly if the amount of
available logical width at the new position changes. Refactor the code so that we can
run the float placement algorithm again when this happens.

Source/WebCore:

Covered by an existing regions test that exposes the issue.

Reviewed by Adam Roben.

* rendering/RenderBlock.cpp:
(WebCore::RenderBlock::computeLogicalLocationForFloat):
(WebCore::RenderBlock::positionNewFloats):
* rendering/RenderBlock.h:

LayoutTests:

Reviewed by Adam Roben.

* fast/regions/webkit-flow-floats-inside-regions-bounds-expected.txt:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@95724 268f45cc-cd09-0410-ab3c-d52691b4dbfc

[Qt] Unreviewed. Add Qt specific expected results for css3/selectors3 tests.
All of them pass, they are compared to mac png results manually.

* platform/qt/Skipped:
* platform/qt/css3/selectors3/ [...]: Added.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@95723 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Source/WebCore: Fire TextInput events on speech input, but not set/add any inputMethod attribute.
https://bugs.webkit.org/show_bug.cgi?id=60451

Reviewed by Ryosuke Niwa.

Introduce code to dispatch TextInput events when text is introduced via
the speech input feature. Handle also the cases where speech input is
started programmatically from the LayoutTestController.

Test: fast/speech/input-ontextinput-event.html

* dom/TextEventInputType.h:
* html/shadow/TextControlInnerElements.cpp:
(WebCore::InputFieldSpeechButtonElement::setRecognitionResult):

LayoutTests: Testing text events triggered by speech input.
https://bugs.webkit.org/show_bug.cgi?id=60451

Reviewed by Ryosuke Niwa.

* fast/speech/input-ontextinput-event-expected.txt: Added.
* fast/speech/input-ontextinput-event.html: Added.
* fast/speech/input-text-speechstart.html:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@95722 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Patch by Jason Liu <jason.liu@torchmobile.com.cn> on 2011-09-20
Reviewed by Simon Fraser.

CSS outline property on a narrow element has rabbit ear artifacts
https://bugs.webkit.org/show_bug.cgi?id=36793

Patch by Jason Liu <jason.liu@torchmobile.com.cn>
Patch updated by Robert Hogan <robert@webkit.org>

Source/WebCore:

Test: fast/css/outline-narrowLine.html

* rendering/RenderInline.cpp:
(WebCore::RenderInline::paintOutlineForLine):

LayoutTests:

* fast/css/outline-narrowLine.html: Added.
* platform/qt/fast/css/outline-narrowLine-expected.txt: Added.
* platform/qt/fast/css/outline-narrowLine-expected.png: Added.
* platform/chromium-linux/fast/css/outline-narrowLine-expected.txt: Added.
* platform/chromium-linux/fast/css/outline-narrowLine-expected.png: Added.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@95721 268f45cc-cd09-0410-ab3c-d52691b4dbfc

[GTK]Popup drop-down menu contains extra empty spaces at beginning and
occupies whole screen when items in popup reaches the height of display device.
https://bugs.webkit.org/show_bug.cgi?id=68533

When implementing the GTK+ menu position function, set pushIn to false, which tell
GTK+ not create a menu that is as big as the entire screen.

Patch by Wajahat Siddiqui <mdwajahatali.siddiqui@motorola.com> on 2011-09-22
Reviewed by Martin Robinson.

* platform/gtk/GtkPopupMenu.cpp:
(WebCore::GtkPopupMenu::menuPositionFunction): Set pushIn to false in the menu position function.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@95720 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Unreviewed. Rolled DEPS.

Patch by Sheriff Bot <webkit.review.bot@gmail.com> on 2011-09-22

* DEPS:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@95719 268f45cc-cd09-0410-ab3c-d52691b4dbfc

[chromium] Rebaseline following r95715
https://bugs.webkit.org/show_bug.cgi?id=68619

Unreviewed, rebaseline.

* platform/chromium-cg-mac/fast/events/scale-and-scroll-iframe-body-expected.png: Added.
* platform/chromium-cg-mac/fast/events/scale-and-scroll-iframe-window-expected.png: Added.
* platform/chromium-mac/fast/events/scale-and-scroll-iframe-body-expected.png: Added.
* platform/chromium-mac/fast/events/scale-and-scroll-iframe-window-expected.png: Added.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@95718 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Define export symbol correctly due to rename of pageScaleFactor to
frameScaleFactor.

Not reviewed. Build fix.

* WebCore.exp.in:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@95717 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Web Inspector: Console is always scrolled to the top
https://bugs.webkit.org/show_bug.cgi?id=67483

Reviewed by Pavel Feldman.

* inspector/front-end/ConsoleView.js:
(WebInspector.ConsoleView.prototype.wasShown):
(WebInspector.ConsoleView.prototype.storeScrollPositions):
(WebInspector.ConsoleView.prototype.restoreScrollPositions):
(WebInspector.ConsoleView.prototype.onResize):
(WebInspector.ConsoleView.prototype._immediatelyScrollIntoView):
(WebInspector.ConsoleView.prototype._cancelScheduledScrollIntoView):
(WebInspector.ConsoleView.prototype._consoleCleared):
* inspector/front-end/Drawer.js:
(WebInspector.Drawer.prototype.hide):
(WebInspector.Drawer.prototype.resize):
(WebInspector.Drawer.prototype._startStatusBarDragging):
(WebInspector.Drawer.prototype._statusBarDragging):
* inspector/front-end/inspector.js:
(WebInspector.animateStyle):
(WebInspector.animateStyle.forceComplete):

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@95716 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Source/WebCore: https://bugs.webkit.org/show_bug.cgi?id=68081
Fix scroll in page scaling mode.

Reviewed by Simon Fraser.

The following things are done in this patch:
1. Rename Frame::pageScaleFactor() to Frame::frameScaleFactor() to better
   reflect its purpose.
2. Frame::frameScaleFactor() returns the scale factor of this frame with
   respect to the container. So for the main frame it'll return the scale
   factor of the page, inner frames will return 1.0
3. scrollBy(), scrollX() and scrollY() to take into account of the frame
   scale factor, causing incorrect coordinates reported to Javascript.

Tests: fast/events/scale-and-scroll-body.html
       fast/events/scale-and-scroll-iframe-body.html
       fast/events/scale-and-scroll-iframe-window.html
       fast/events/scale-and-scroll-window.html

* html/HTMLBodyElement.cpp:
(WebCore::adjustForZoom):
(WebCore::HTMLBodyElement::setScrollLeft):
(WebCore::HTMLBodyElement::setScrollTop):
* loader/HistoryController.cpp:
(WebCore::HistoryController::saveScrollPositionAndViewStateToItem):
* page/DOMWindow.cpp:
(WebCore::DOMWindow::scrollX):
(WebCore::DOMWindow::scrollY):
(WebCore::DOMWindow::scrollTo):
* page/Frame.cpp:
(WebCore::Frame::frameScaleFactor): Return page's scale factor if this is
                                    main frame, otherwise 1.0
* page/Frame.h:
* page/FrameView.cpp:
(WebCore::FrameView::applyOverflowToViewport):
(WebCore::FrameView::scrollXForFixedPosition):
(WebCore::FrameView::scrollYForFixedPosition):

Source/WebKit2: https://bugs.webkit.org/show_bug.cgi?id=68081
Fix scroll in page scaling mode.

Reviewed by Simon Fraser.

* WebProcess/Plugins/PluginView.cpp: Rename pageScaleFactor to frameScaleFactor.
(WebKit::PluginView::handleEvent):
(WebKit::PluginView::viewGeometryDidChange):
(WebKit::PluginView::clipRectInWindowCoordinates):

LayoutTests: https://bugs.webkit.org/show_bug.cgi?id=68081
Add layout test for page scale and scrolling.

Reviewed by Simon Fraser.

Add tests to cover the following cases:
1. Scale and scroll window
2. Scale and scroll document.body
3. Scale and scroll iframe's contentWindow
4. Scale and scroll iframe's contentDocument.body

* fast/events/scale-and-scroll-body-expected.png: Added.
* fast/events/scale-and-scroll-body-expected.txt: Added.
* fast/events/scale-and-scroll-body.html: Added.
* fast/events/scale-and-scroll-iframe-body-expected.png: Added.
* fast/events/scale-and-scroll-iframe-body-expected.txt: Added.
* fast/events/scale-and-scroll-iframe-body.html: Added.
* fast/events/scale-and-scroll-iframe-window-expected.png: Added.
* fast/events/scale-and-scroll-iframe-window-expected.txt: Added.
* fast/events/scale-and-scroll-iframe-window.html: Added.
* fast/events/scale-and-scroll-window-expected.png: Added.
* fast/events/scale-and-scroll-window-expected.txt: Added.
* fast/events/scale-and-scroll-window.html: Added.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@95715 268f45cc-cd09-0410-ab3c-d52691b4dbfc

CSS: Implement 'font' property in CSSComputedStyle.
https://bugs.webkit.org/show_bug.cgi?id=66666

Patch by Andreas Kling <kling@webkit.org> on 2011-09-22
Reviewed by Antti Koivisto.

Source/WebCore:

Test: fast/css/getComputedStyle/computed-style-font.html

* css/CSSComputedStyleDeclaration.cpp:
(WebCore::fontFamilyFromStyle): Factored out of getPropertyCSSValue().
This function will always return a CSSValueList, as opposed to the previous
implementation which would return a CSSPrimitiveValue if there was only one family.
getCSSPropertyValue() will return the first element of the list for 1-element lists
to preserve the current behavior of getPropertyCSSValue('font-family').

(WebCore::lineHeightFromStyle): Factored out of getPropertyCSSValue().
(WebCore::fontSizeFromStyle): Ditto.
(WebCore::fontStyleFromStyle): Ditto.
(WebCore::fontVariantFromStyle): Ditto.
(WebCore::fontWeightFromStyle): Ditto.

(WebCore::CSSComputedStyleDeclaration::getPropertyCSSValue): Added
implementation of the computed 'font' property (CSSPropertyFont.)

LayoutTests:

* fast/css/getComputedStyle/computed-style-font-expected.txt: Added.
* fast/css/getComputedStyle/computed-style-font.html: Added.
* fast/css/getComputedStyle/script-tests/computed-style-font.js: Added.
(computedFont):
(computedFontCSSValue):

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@95714 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Remove useless empty WebKit namespace.

Patch by Gustavo Noronha Silva <gustavo.noronha@collabora.com> on 2011-09-22
Reviewed by Xan Lopez.

* webkit/webkitglobals.cpp:
(webkitInit):

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@95713 268f45cc-cd09-0410-ab3c-d52691b4dbfc

[Qt][WK2] Unreivewed gardening.

* platform/qt-wk2/Skipped: Skip new failing test introduced in r95645.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@95712 268f45cc-cd09-0410-ab3c-d52691b4dbfc

[Qt] 69 tests failing after trac.webkit.org/changeset/92375
https://bugs.webkit.org/show_bug.cgi?id=65721

Unreviewed gardening. Update Qt 4.8 specific expected files after r95706.

* platform/qt-4.8/ [...]: Updated.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@95711 268f45cc-cd09-0410-ab3c-d52691b4dbfc

[Qt] Redirection overflow errors have wrong error domain
https://bugs.webkit.org/show_bug.cgi?id=68612

Reviewed by Andreas Kling.

Source/WebCore:

Redirection overflow currently puts the URL hostname in the errorDomain field.
Now redirection overflow errors will be classified as "HTTP" errors.

* platform/network/qt/QNetworkReplyHandler.cpp:
(WebCore::QNetworkReplyHandler::redirect):

Source/WebKit2:

Redirection overflow currently puts the URL hostname in the errorDomain field
and therefore are treated as a QWebError::EngineError. From now on they
will be treated as QWebError::HttpError, since their error domain was modified
to "HTTP" in QNetworkReplyHandler::redirect().

* UIProcess/API/qt/qweberror.cpp:
(QWebError::type):

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@95710 268f45cc-cd09-0410-ab3c-d52691b4dbfc

[Qt] Unreviewed gardening. Add Qt specific expected files for new test introduced in r95671.

* platform/qt/fast/table/quote-text-around-iframe-expected.png: Added.
* platform/qt/fast/table/quote-text-around-iframe-expected.txt: Added.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@95709 268f45cc-cd09-0410-ab3c-d52691b4dbfc

REGRESSION(r95679): It broke everything.
https://bugs.webkit.org/show_bug.cgi?id=68611

Unreviewed rolling out r95679 and r95690, because it broke everything.

* UIProcess/API/mac/WKView.mm:
(maybeCreateSandboxExtensionFromPasteboard):
(-[WKView performDragOperation:]):
* UIProcess/WebContext.cpp:
(WebKit::WebContext::didPerformClientRedirect):
(WebKit::WebContext::didPerformServerRedirect):
(WebKit::WebContext::didUpdateHistoryTitle):
(WebKit::WebContext::getPluginPath):
* UIProcess/WebPageProxy.cpp:
(WebKit::WebPageProxy::reattachToWebProcessWithItem):
(WebKit::WebPageProxy::initializeSandboxExtensionHandle):
(WebKit::WebPageProxy::loadURL):
(WebKit::WebPageProxy::loadURLRequest):
(WebKit::WebPageProxy::loadHTMLString):
(WebKit::WebPageProxy::loadAlternateHTMLString):
(WebKit::WebPageProxy::goForward):
(WebKit::WebPageProxy::goBack):
(WebKit::WebPageProxy::goToBackForwardItem):
(WebKit::WebPageProxy::didStartProvisionalLoadForFrame):
(WebKit::WebPageProxy::didReceiveServerRedirectForProvisionalLoadForFrame):
(WebKit::WebPageProxy::didSameDocumentNavigationForFrame):
(WebKit::WebPageProxy::decidePolicyForNavigationAction):
(WebKit::WebPageProxy::decidePolicyForNewWindowAction):
(WebKit::WebPageProxy::decidePolicyForResponse):
(WebKit::WebPageProxy::didInitiateLoadForResource):
(WebKit::WebPageProxy::didSendRequestForResource):
(WebKit::WebPageProxy::didReceiveResponseForResource):
(WebKit::WebPageProxy::missingPluginButtonClicked):
* UIProcess/WebPageProxy.h:
* UIProcess/WebProcessProxy.cpp:
(WebKit::WebProcessProxy::WebProcessProxy):
(WebKit::WebProcessProxy::addBackForwardItem):
* UIProcess/WebProcessProxy.h:
* UIProcess/cf/WebPageProxyCF.cpp:
(WebKit::WebPageProxy::restoreFromSessionStateData):

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@95708 268f45cc-cd09-0410-ab3c-d52691b4dbfc

32-bit call code clobbers the function cell tag
https://bugs.webkit.org/show_bug.cgi?id=68606

Patch by Filip Pizlo <fpizlo@apple.com> on 2011-09-22
Reviewed by Csaba Osztrogonác.

This is a minimalistic fix: it simply emits code to restore the
cell tag on the slow path, if we know that we failed due to
emitCallIfNotType.

* jit/JITCall32_64.cpp:
(JSC::JIT::compileOpCallVarargsSlowCase):
(JSC::JIT::compileOpCallSlowCase):

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@95707 268f45cc-cd09-0410-ab3c-d52691b4dbfc

[Qt] 69 tests failing after trac.webkit.org/changeset/92375
https://bugs.webkit.org/show_bug.cgi?id=65721

Unreviewed gardening. Update Qt specific expected files.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@95706 268f45cc-cd09-0410-ab3c-d52691b4dbfc

[chromium] Update test expectation following r95685
https://bugs.webkit.org/show_bug.cgi?id=68604

Unreviewed, update test expectation.

* platform/chromium/test_expectations.txt:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@95705 268f45cc-cd09-0410-ab3c-d52691b4dbfc

[Qt] 11 tests failed after r95671 because the tests are updated.
https://bugs.webkit.org/show_bug.cgi?id=68596

Update Qt specific expected files.

* platform/qt/fast/css-generated-content/table-cell-before-content-expected.png:
* platform/qt/fast/css-generated-content/table-cell-before-content-expected.txt:
* platform/qt/fast/table/table-cell-after-child-in-block-expected.txt:
* platform/qt/fast/table/table-cell-before-after-content-around-table-block-expected.txt:
* platform/qt/fast/table/table-cell-before-after-content-around-table-expected.txt:
* platform/qt/fast/table/table-cell-before-child-in-block-expected.txt:
* platform/qt/fast/table/table-row-after-child-in-block-expected.txt:
* platform/qt/fast/table/table-row-before-after-content-around-block-expected.txt:
* platform/qt/fast/table/table-row-before-after-content-around-table-expected.txt:
* platform/qt/fast/table/table-row-before-child-in-block-expected.txt:
* platform/qt/fast/table/table-row-style-not-updated-with-after-content-expected.png: Added.
* platform/qt/fast/table/table-row-style-not-updated-with-after-content-expected.txt:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@95704 268f45cc-cd09-0410-ab3c-d52691b4dbfc

[Qt] 11 tests failed after r95671 because the tests are updated.
https://bugs.webkit.org/show_bug.cgi?id=68596

Unreviewed gardening. Typo fixes.

Patch by Oliver Varga <voliver@inf.u-szeged.hu> on 2011-09-22

* platform/qt/Skipped:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@95703 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Unreviewed, rolling out r95699.
http://trac.webkit.org/changeset/95699
https://bugs.webkit.org/show_bug.cgi?id=67417

Makes many chromium compositor tests crash

Source/WebCore:

* platform/CrossThreadCopier.h:
* platform/graphics/chromium/LayerRendererChromium.cpp:
(WebCore::LayerRendererChromium::~LayerRendererChromium):
* platform/graphics/chromium/cc/CCHeadsUpDisplay.cpp:
(WebCore::CCHeadsUpDisplay::enabled):
* platform/graphics/chromium/cc/CCLayerImpl.cpp:
(WebCore::CCLayerImpl::CCLayerImpl):
(WebCore::CCLayerImpl::~CCLayerImpl):
* platform/graphics/chromium/cc/CCLayerTreeHost.cpp:
(WebCore::CCLayerTreeHost::CCLayerTreeHost):
(WebCore::CCLayerTreeHost::commitTo):
(WebCore::CCLayerTreeHost::setNeedsRedraw):
* platform/graphics/chromium/cc/CCLayerTreeHost.h:
(WebCore::CCSettings::CCSettings):
* platform/graphics/chromium/cc/CCLayerTreeHostImpl.cpp:
(WebCore::CCLayerTreeHostImpl::CCLayerTreeHostImpl):
(WebCore::CCLayerTreeHostImpl::~CCLayerTreeHostImpl):
* platform/graphics/chromium/cc/CCProxy.h:
* platform/graphics/chromium/cc/CCSingleThreadProxy.cpp:
(WebCore::CCSingleThreadProxy::finishAllRendering):
(WebCore::CCSingleThreadProxy::setNeedsCommit):
(WebCore::CCSingleThreadProxy::commitIfNeeded):
* platform/graphics/chromium/cc/CCThreadProxy.cpp:
(WebCore::CCThreadProxy::CCThreadProxy):
(WebCore::CCThreadProxy::~CCThreadProxy):
(WebCore::CCThreadProxy::compositeAndReadback):
(WebCore::CCThreadProxy::finishAllRendering):
(WebCore::CCThreadProxy::isStarted):
(WebCore::CCThreadProxy::initializeLayerRenderer):
(WebCore::CCThreadProxy::setNeedsCommit):
(WebCore::CCThreadProxy::setNeedsCommitAndRedraw):
(WebCore::CCThreadProxy::setNeedsRedraw):
(WebCore::CCThreadProxy::start):
(WebCore::CCThreadProxy::stop):
(WebCore::CCThreadProxy::beginFrameAndCommitOnCCThread):
(WebCore::CCThreadProxy::beginFrameAndCommit):
(WebCore::CCThreadProxy::commitOnCCThread):
(WebCore::CCThreadProxy::drawLayersOnCCThread):
(WebCore::CCThreadProxy::setNeedsCommitOnCCThread):
(WebCore::CCThreadProxy::setNeedsCommitAndRedrawOnCCThread):
(WebCore::CCThreadProxy::setNeedsRedrawOnCCThread):
* platform/graphics/chromium/cc/CCThreadProxy.h:

Source/WebKit/chromium:

* tests/CCLayerTreeHostTest.cpp:
(WTF::CCLayerTreeHostTest::doBeginTest):
(WTF::TEST_F):
* tests/TreeSynchronizerTest.cpp:
(WebCore::TEST):

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@95702 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Unreviewed, GTK rebaseline after r95671.

* platform/gtk/fast/css-generated-content/table-cell-before-content-expected.txt:
* platform/gtk/fast/table/quote-text-around-iframe-expected.png: Added.
* platform/gtk/fast/table/quote-text-around-iframe-expected.txt: Added.
* platform/gtk/fast/table/table-cell-after-child-in-block-expected.txt:
* platform/gtk/fast/table/table-cell-before-after-content-around-table-block-expected.txt:
* platform/gtk/fast/table/table-cell-before-after-content-around-table-expected.txt:
* platform/gtk/fast/table/table-cell-before-child-in-block-expected.txt:
* platform/gtk/fast/table/table-row-after-child-in-block-expected.txt:
* platform/gtk/fast/table/table-row-before-after-content-around-block-expected.txt:
* platform/gtk/fast/table/table-row-before-after-content-around-table-expected.txt:
* platform/gtk/fast/table/table-row-before-child-in-block-expected.txt:
* platform/gtk/fast/table/table-row-style-not-updated-with-after-content-expected.txt:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@95701 268f45cc-cd09-0410-ab3c-d52691b4dbfc

[Qt] 11 tests failed after r95671.
https://bugs.webkit.org/show_bug.cgi?id=68596

Unreviewed gardening.

Patch by Oliver Varga <voliver@inf.u-szeged.hu> on 2011-09-22

* platform/qt/Skipped:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@95700 268f45cc-cd09-0410-ab3c-d52691b4dbfc

[chromium] Make CCThreadProxy draw
https://bugs.webkit.org/show_bug.cgi?id=67417

Source/WebCore:

Update the CCThreadProxy to correctly implement the CCProxy
interface, do all the right committing and updating steps, and
draw a picture on the screen.

Patch by Nat Duca <nduca@chromium.org> on 2011-09-22
Reviewed by James Robinson.

* platform/graphics/IntRect.h:
* platform/graphics/chromium/LayerRendererChromium.cpp:
(WebCore::LayerRendererChromium::~LayerRendererChromium):
* platform/graphics/chromium/cc/CCHeadsUpDisplay.cpp:
(WebCore::CCHeadsUpDisplay::enabled):
* platform/graphics/chromium/cc/CCLayerImpl.cpp:
(WebCore::CCLayerImpl::CCLayerImpl):
(WebCore::CCLayerImpl::~CCLayerImpl):
* platform/graphics/chromium/cc/CCLayerTreeHost.cpp:
(WebCore::CCLayerTreeHost::CCLayerTreeHost):
(WebCore::CCLayerTreeHost::commitTo):
(WebCore::CCLayerTreeHost::commitComplete):
(WebCore::CCLayerTreeHost::setNeedsRedraw):
* platform/graphics/chromium/cc/CCLayerTreeHost.h:
* platform/graphics/chromium/cc/CCLayerTreeHostImpl.cpp:
(WebCore::CCLayerTreeHostImpl::CCLayerTreeHostImpl):
(WebCore::CCLayerTreeHostImpl::~CCLayerTreeHostImpl):
* platform/graphics/chromium/cc/CCSingleThreadProxy.cpp:
(WebCore::CCSingleThreadProxy::finishAllRendering):
(WebCore::CCSingleThreadProxy::setNeedsCommit):
(WebCore::CCSingleThreadProxy::commitIfNeeded):
* platform/graphics/chromium/cc/CCThreadProxy.cpp:
(WebCore::CCThreadProxy::CCThreadProxy):
(WebCore::CCThreadProxy::~CCThreadProxy):
(WebCore::CCThreadProxy::compositeAndReadback):
(WebCore::CCThreadProxy::drawLayersAndReadbackOnCCThread):
(WebCore::CCThreadProxy::finishAllRendering):
(WebCore::CCThreadProxy::isStarted):
(WebCore::CCThreadProxy::setNeedsCommit):
(WebCore::CCThreadProxy::setNeedsCommitAndRedraw):
(WebCore::CCThreadProxy::setNeedsRedraw):
(WebCore::CCThreadProxy::start):
(WebCore::CCThreadProxy::stop):
(WebCore::CCThreadProxy::finishAllRenderingOnCCThread):
(WebCore::CCThreadProxy::createBeginFrameAndCommitTaskOnCCThread):
(WebCore::CCThreadProxy::beginFrameAndCommit):
(WebCore::CCThreadProxy::commitOnCCThread):
(WebCore::CCThreadProxy::scheduleDrawTaskOnCCThread):
(WebCore::CCThreadProxy::drawLayersAndPresentOnCCThread):
(WebCore::CCThreadProxy::drawLayersOnCCThread):
(WebCore::CCThreadProxy::updateSchedulerStateOnCCThread):
* platform/graphics/chromium/cc/CCThreadProxy.h:

Source/WebKit/chromium:

Disable CCLayerTreeHostTest temporarily. Will re-enable
with https://bugs.webkit.org/show_bug.cgi?id=67418

Patch by Nat Duca <nduca@chromium.org> on 2011-09-22
Reviewed by James Robinson.

* tests/CCLayerTreeHostTest.cpp:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@95699 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Implement WebKit side of IDBFactory::getDatabaseNames
https://bugs.webkit.org/show_bug.cgi?id=68037

Patch by Joshua Bell <jsbell@chromium.org> on 2011-09-21
Reviewed by Tony Chang.

Interface changes and stub implementations for
new IndexedDB IDBFactory.getDatabaseNames. This is
part one of a two-sided patch.

* public/WebIDBCallbacks.h:
(WebKit::WebIDBCallbacks::onSuccess):
* public/WebIDBFactory.h:
(WebKit::WebIDBFactory::getDatabaseNames):

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@95698 268f45cc-cd09-0410-ab3c-d52691b4dbfc

https://bugs.webkit.org/show_bug.cgi?id=67415
Text drawn via -webkit-background-clip:text is blurry at device scale factors >1.0
-and corresponding-
<rdar://problem/10060379>

Reviewed by Darin Adler.

New helper function RenderBoxModelObject scales the mask image by the
deviceScaleFactor to get an image of the appropriate resolution. It also scales
the image's GraphicsContext so that the clip is set up on the same scale. Back in
paintFillLayerExtended() we still clip the image to the original maskRect to get
everything scaled back to the appropriate size.
* rendering/RenderBoxModelObject.cpp:
(WebCore::createDeviceScaledImageBuffer):
(WebCore::RenderBoxModelObject::paintFillLayerExtended):

Make the deviceScaleFactor convenience function just a namespace-level function in
Page rather than a static member or Page.
* page/Page.cpp:
(WebCore::deviceScaleFactor):
* page/Page.h:

Pre-existing callers of Page::deviceScaleFactor(Frame*) must now use
WebCore::deviceScaleFactor(Frame*)
* editing/DeleteButtonController.cpp:
(WebCore::DeleteButtonController::createDeletionUI):
* rendering/RenderImage.cpp:
(WebCore::RenderImage::imageSizeForError):
(WebCore::RenderImage::paintReplaced):
* rendering/RenderLayer.cpp:
(WebCore::RenderLayer::drawPlatformResizerImage):

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@95697 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Unreviewed. Chromium rebaselines for r95671.

* platform/chromium-linux/fast/table/quote-text-around-iframe-expected.png: Added.
* platform/chromium-mac/fast/table/quote-text-around-iframe-expected.png: Added.
* platform/chromium-win/fast/css-generated-content/table-cell-before-content-expected.txt:
* platform/chromium-win/fast/table/quote-text-around-iframe-expected.png: Added.
* platform/chromium-win/fast/table/quote-text-around-iframe-expected.txt: Added.
* platform/chromium-win/fast/table/table-cell-after-child-in-block-expected.txt:
* platform/chromium-win/fast/table/table-cell-before-child-in-block-expected.txt:
* platform/chromium-win/fast/table/table-row-after-child-in-block-expected.txt:
* platform/chromium-win/fast/table/table-row-before-child-in-block-expected.txt:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@95696 268f45cc-cd09-0410-ab3c-d52691b4dbfc

[CG] ImageBufferData::getData has an invariant comparison in the inner part of a loop which doesn't get optimized out
https://bugs.webkit.org/show_bug.cgi?id=68588
<rdar://problem/10164955>

Reviewed by Simon Fraser.

Factor the unmultiplied check out of the inner loop, resulting in a
speed bump in ImageBufferData::getData.

No new tests, performance improvement.

* platform/graphics/cg/ImageBufferDataCG.cpp:
(WebCore::ImageBufferData::getData):

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@95695 268f45cc-cd09-0410-ab3c-d52691b4dbfc

:hover selector fails when hovering over a child select element with size attribute
https://bugs.webkit.org/show_bug.cgi?id=67594

Patch by Sameer Patil <mkrp87@motorola.com> on 2011-09-21
Reviewed by Darin Adler.

Source/WebCore:

Test: fast/css/hover-affects-ancestor.html

This patch addresses the problem of deactivation of the ancestor element :hover rules
when hovered its contained elements.

* rendering/RenderLayer.cpp:
(WebCore::RenderLayer::updateHoverActiveState):

LayoutTests:

* fast/css/hover-affects-ancestor-expected.txt: Added.
* fast/css/hover-affects-ancestor.html: Added.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@95694 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Add missing addPtr->add32 mapping for X86.

Rubber stamped by Sam Weinig.

* assembler/MacroAssembler.h:
(JSC::MacroAssembler::addPtr):

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@95693 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Add missing addDouble for AbsoluteAddress to X86

Rubber stamped by Geoff Garen.

* assembler/MacroAssemblerX86.h:
(JSC::MacroAssemblerX86::addDouble):
* assembler/X86Assembler.h:
(JSC::X86Assembler::addsd_mr):
(JSC::X86Assembler::cvtsi2sd_rr):
(JSC::X86Assembler::cvtsi2sd_mr):

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@95692 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Rebaseline for bug 65583 (path based border radius drawing on skia) part 4
https://bugs.webkit.org/show_bug.cgi?id=68339

Patch by Ben Wells <benwells@chromium.org> on 2011-09-21
Reviewed by James Robinson.

* platform/chromium-linux/css1/box_properties/border_style-expected.png:
* platform/chromium-linux/css1/box_properties/border_style_inline-expected.png:
* platform/chromium-linux/css2.1/: 14 changes
* platform/chromium-linux/fast/borders/border-radius-inset-outset-expected.png: Added.
* platform/chromium-linux/fast/borders/border-styles-split-expected.png: Added.
* platform/chromium-linux/fast/writing-mode/border-styles-vertical-lr-expected.png: Added.
* platform/chromium-linux/fast/writing-mode/border-styles-vertical-rl-expected.png: Added.
* platform/chromium-win/css1/box_properties/border_style-expected.png:
* platform/chromium-win/css1/box_properties/border_style_inline-expected.png:
* platform/chromium-win/css2.1/: 14 changes
* platform/chromium-win/fast/borders/border-radius-inset-outset-expected.png: Added.
* platform/chromium-win/fast/borders/border-radius-inset-outset-expected.txt: Added.
* platform/chromium-win/fast/borders/border-styles-split-expected.png: Added.
* platform/chromium-win/fast/borders/border-styles-split-expected.txt: Added.
* platform/chromium-win/fast/writing-mode/border-styles-vertical-lr-expected.png: Added.
* platform/chromium-win/fast/writing-mode/border-styles-vertical-lr-expected.txt: Added.
* platform/chromium-win/fast/writing-mode/border-styles-vertical-rl-expected.png: Added.
* platform/chromium-win/fast/writing-mode/border-styles-vertical-rl-expected.txt: Added.
* platform/chromium/test_expectations.txt:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@95691 268f45cc-cd09-0410-ab3c-d52691b4dbfc

    [WK2] UIProcess should check that WebProcess isn't sending unexpected file: URLs to it
        https://bugs.webkit.org/show_bug.cgi?id=68573

        Unreviewed follow-up fix.

        * UIProcess/WebProcessProxy.cpp: (WebKit::WebProcessProxy::checkURLReceivedFromWebProcess):
        Just like we allow null URLs, also allow empty strings.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@95690 268f45cc-cd09-0410-ab3c-d52691b4dbfc

[Chromium] Protect message ports from being deleted in V8MessageEvent::portsAccessorGetter
https://bugs.webkit.org/show_bug.cgi?id=68584

Patch by Sergey Glazunov <serg.glazunov@gmail.com> on 2011-09-21
Reviewed by Adam Barth.

Source/WebCore:

Test: fast/dom/message-port-deleted-by-accessor.html

* bindings/v8/custom/V8MessageEventCustom.cpp:
(WebCore::V8MessageEvent::portsAccessorGetter):

LayoutTests:

* fast/dom/message-port-deleted-by-accessor-expected.txt: Added.
* fast/dom/message-port-deleted-by-accessor.html: Added.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@95689 268f45cc-cd09-0410-ab3c-d52691b4dbfc

[chromium] Rebaselines for passing tests and expectation updates/narrowing.

* platform/chromium-cg-mac-leopard/editing/inserting/editable-inline-element-expected.png: Added.
* platform/chromium-mac-leopard/css2.1/t090204-display-change-01-b-ao-expected.png: Added.
* platform/chromium-mac-leopard/editing/inserting/editable-inline-element-expected.png: Added.
* platform/chromium-mac/css2.1/t090204-display-change-01-b-ao-expected.png:
* platform/chromium-mac/css2.1/t090204-display-change-01-b-ao-expected.txt: Removed.
* platform/chromium/test_expectations.txt:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@95688 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Build fix following fix for bug #68586.

* jit/JIT.cpp:
* jit/JITInlineMethods.h:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@95687 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Add back protection against the NSView going away while handling mouseDown
https://bugs.webkit.org/show_bug.cgi?id=68585

Reviewed by Darin Adler.

Add calls to RenderWidget::suspendWidgetHierarchyUpdates() and RenderWidget::resumeWidgetHierarchyUpdates(),
as a followup fix to https://bugs.webkit.org/show_bug.cgi?id=68570.

* page/mac/EventHandlerMac.mm:
(WebCore::EventHandler::passMouseDownEventToWidget):

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@95686 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Crash in RenderBox::paintMaskImages when GraphicsContext's painting is disabled
https://bugs.webkit.org/show_bug.cgi?id=68133

Reviewed by Darin Adler.

.:

* Source/autotools/symbols.filter: Added the mangled symbols needed for window.internals

Source/WebCore:

Tests: fast/css/webkit-mask-crash-fieldset-legend.html
       fast/css/webkit-mask-crash-figure.html
       fast/css/webkit-mask-crash-table.html
       fast/css/webkit-mask-crash-td-2.html
       fast/css/webkit-mask-crash-td.html

GraphicsContext::getCTM crashes if called with a GraphicsContext that has painting
disabled. RenderBox::paintMaskImages would thus crash if called in this condition.

This change just modifies the different GraphicsContext::getCTM method to bail early
if painting is disabled on the GraphicsContext. The rest of the change is exposing
paintControlTints that exposes this.

* WebCore.exp.in: Added symbols of the newly export window.internals function.

* page/FrameView.cpp:
(WebCore::FrameView::updateControlTints): Split this function in 2 so that
I can expose the internal paintControlTints.

(WebCore::FrameView::paintControlTints):
This is the one exposed to Internals as we want to be testable regardless of
whether the platform supports control tints.

* page/FrameView.h: Added paintControlTints.

* testing/Internals.cpp:
(WebCore::Internals::paintControlTints):
* testing/Internals.h:
* testing/Internals.idl:
Added a way to force a fake painting so that we can easily reproduce the bugs.

* platform/graphics/cairo/GraphicsContextCairo.cpp:
(WebCore::GraphicsContext::getCTM):
* platform/graphics/cg/GraphicsContextCG.cpp:
(WebCore::GraphicsContext::getCTM):
* platform/graphics/qt/GraphicsContextQt.cpp:
(WebCore::GraphicsContext::getCTM):
* platform/graphics/skia/GraphicsContextSkia.cpp:
(WebCore::GraphicsContext::getCTM):
* platform/graphics/wince/GraphicsContextWinCE.cpp:
(WebCore::GraphicsContext::getCTM):
* platform/graphics/wx/GraphicsContextWx.cpp:
(WebCore::GraphicsContext::getCTM):
Fixed all our back-end to exit early if painting is disabled.

Source/WebKit2:

* win/WebKit2.def:
* win/WebKit2CFLite.def:
Exported the new FrameView::paintControlTints function.

LayoutTests:

Those tests checks that we do not crash when calling internals.paintControlTints.

* platform/mac/Skipped: Skipped 2 tests as they are hitting an ASSERT unrelated to
this change on Mac.

* fast/css/webkit-mask-crash-fieldset-legend-expected.txt: Added.
* fast/css/webkit-mask-crash-fieldset-legend.html: Added.
* fast/css/webkit-mask-crash-figure-expected.txt: Added.
* fast/css/webkit-mask-crash-figure.html: Added.
* fast/css/webkit-mask-crash-table-expected.txt: Added.
* fast/css/webkit-mask-crash-table.html: Added.
* fast/css/webkit-mask-crash-td-2-expected.txt: Added.
* fast/css/webkit-mask-crash-td-2.html: Added.
* fast/css/webkit-mask-crash-td-expected.txt: Added.
* fast/css/webkit-mask-crash-td.html: Added.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@95685 268f45cc-cd09-0410-ab3c-d52691b4dbfc

DFG JIT should be able to compile op_throw
https://bugs.webkit.org/show_bug.cgi?id=68571

Patch by Filip Pizlo <fpizlo@apple.com> on 2011-09-21
Reviewed by Geoffrey Garen.

This compiles op_throw in the simplest way possible: it's an OSR
point back to the old JIT. This is a good step towards increasing
coverage, particularly on Kraken, but it's neutral because the
same functions that do throw also use some other unsupported
opcodes.

* dfg/DFGByteCodeParser.cpp:
(JSC::DFG::ByteCodeParser::parseBlock):
* dfg/DFGCapabilities.h:
(JSC::DFG::canCompileOpcode):
* dfg/DFGNode.h:
* dfg/DFGPropagator.cpp:
(JSC::DFG::Propagator::propagateNodePredictions):
* dfg/DFGSpeculativeJIT.cpp:
(JSC::DFG::SpeculativeJIT::compile):

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@95684 268f45cc-cd09-0410-ab3c-d52691b4dbfc

DFG JIT should be able to compile op_throw
https://bugs.webkit.org/show_bug.cgi?id=68571

Reviewed by Geoffrey Garen.

This compiles op_throw in the simplest way possible: it's an OSR
point back to the old JIT. This is a good step towards increasing
coverage, particularly on Kraken, but it's neutral because the
same functions that do throw also use some other unsupported
opcodes.

* dfg/DFGByteCodeParser.cpp:
(JSC::DFG::ByteCodeParser::parseBlock):
* dfg/DFGCapabilities.h:
(JSC::DFG::canCompileOpcode):
* dfg/DFGNode.h:
* dfg/DFGPropagator.cpp:
(JSC::DFG::Propagator::propagateNodePredictions):
* dfg/DFGSpeculativeJIT.cpp:
(JSC::DFG::SpeculativeJIT::compile):

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@95683 268f45cc-cd09-0410-ab3c-d52691b4dbfc

IndexedDB: compare strings without decoding
https://bugs.webkit.org/show_bug.cgi?id=68554

Patch by Joshua Bell <jsbell@chromium.org> on 2011-09-21
Reviewed by Tony Chang.

Resolves a FIXME in IndexedDB that was also identified as a hotspot
during profiling. Yields a small performance improvement.

* Source/WebCore/storage/IDBLevelDBCoding.cpp:
* Source/WebCore/storage/IDBLevelDBCoding.h:
* Source/WebKit/chromium/tests/IDBLevelDBCodingTest.cpp:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@95682 268f45cc-cd09-0410-ab3c-d52691b4dbfc

DFG should support continuous optimization
https://bugs.webkit.org/show_bug.cgi?id=68329

Reviewed by Geoffrey Garen.

This adds the ability to reoptimize a code block if speculation
failures happen frequently. 6% speed-up on Kraken, 1% slow-down
on V8, neutral on SunSpider.

* CMakeLists.txt:
* GNUmakefile.list.am:
* JavaScriptCore.pro:
* JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.vcproj:
* JavaScriptCore.vcproj/WTF/WTF.vcproj:
* JavaScriptCore.xcodeproj/project.pbxproj:
* bytecode/CodeBlock.cpp:
(JSC::CodeBlock::CodeBlock):
(JSC::ProgramCodeBlock::jettison):
(JSC::EvalCodeBlock::jettison):
(JSC::FunctionCodeBlock::jettison):
(JSC::CodeBlock::shouldOptimizeNow):
(JSC::CodeBlock::dumpValueProfiles):
* bytecode/CodeBlock.h:
* dfg/DFGByteCodeParser.cpp:
(JSC::DFG::ByteCodeParser::getStrongPrediction):
* dfg/DFGJITCompiler.cpp:
(JSC::DFG::JITCompiler::exitSpeculativeWithOSR):
(JSC::DFG::JITCompiler::compileEntry):
(JSC::DFG::JITCompiler::compileBody):
* dfg/DFGJITCompiler.h:
(JSC::DFG::JITCompiler::noticeOSREntry):
* dfg/DFGOSREntry.cpp:
(JSC::DFG::prepareOSREntry):
* dfg/DFGOSREntry.h:
(JSC::DFG::getOSREntryDataBytecodeIndex):
* dfg/DFGSpeculativeJIT.cpp:
(JSC::DFG::SpeculativeJIT::compile):
* heap/ConservativeRoots.cpp:
(JSC::ConservativeRoots::ConservativeRoots):
(JSC::ConservativeRoots::~ConservativeRoots):
(JSC::DummyMarkHook::mark):
(JSC::ConservativeRoots::genericAddPointer):
(JSC::ConservativeRoots::genericAddSpan):
(JSC::ConservativeRoots::add):
* heap/ConservativeRoots.h:
* heap/Heap.cpp:
(JSC::Heap::addJettisonCodeBlock):
(JSC::Heap::markRoots):
* heap/Heap.h:
* heap/JettisonedCodeBlocks.cpp: Added.
(JSC::JettisonedCodeBlocks::JettisonedCodeBlocks):
(JSC::JettisonedCodeBlocks::~JettisonedCodeBlocks):
(JSC::JettisonedCodeBlocks::addCodeBlock):
(JSC::JettisonedCodeBlocks::clearMarks):
(JSC::JettisonedCodeBlocks::deleteUnmarkedCodeBlocks):
(JSC::JettisonedCodeBlocks::traceCodeBlocks):
* heap/JettisonedCodeBlocks.h: Added.
(JSC::JettisonedCodeBlocks::mark):
* interpreter/RegisterFile.cpp:
(JSC::RegisterFile::gatherConservativeRoots):
* interpreter/RegisterFile.h:
* jit/JITStubs.cpp:
(JSC::DEFINE_STUB_FUNCTION):
* runtime/Executable.cpp:
(JSC::jettisonCodeBlock):
(JSC::EvalExecutable::jettisonOptimizedCode):
(JSC::ProgramExecutable::jettisonOptimizedCode):
(JSC::FunctionExecutable::jettisonOptimizedCodeForCall):
(JSC::FunctionExecutable::jettisonOptimizedCodeForConstruct):
* runtime/Executable.h:
(JSC::FunctionExecutable::jettisonOptimizedCodeFor):
* wtf/BitVector.h: Added.
(WTF::BitVector::BitVector):
(WTF::BitVector::~BitVector):
(WTF::BitVector::operator=):
(WTF::BitVector::size):
(WTF::BitVector::ensureSize):
(WTF::BitVector::resize):
(WTF::BitVector::clearAll):
(WTF::BitVector::get):
(WTF::BitVector::set):
(WTF::BitVector::clear):
(WTF::BitVector::bitsInPointer):
(WTF::BitVector::maxInlineBits):
(WTF::BitVector::byteCount):
(WTF::BitVector::makeInlineBits):
(WTF::BitVector::OutOfLineBits::numBits):
(WTF::BitVector::OutOfLineBits::numWords):
(WTF::BitVector::OutOfLineBits::bits):
(WTF::BitVector::OutOfLineBits::create):
(WTF::BitVector::OutOfLineBits::destroy):
(WTF::BitVector::OutOfLineBits::OutOfLineBits):
(WTF::BitVector::isInline):
(WTF::BitVector::outOfLineBits):
(WTF::BitVector::resizeOutOfLine):
(WTF::BitVector::bits):

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@95681 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Update committers.py with a new contributor contact
https://bugs.webkit.org/show_bug.cgi?id=68565

Reviewed by Eric Carlson.

* Scripts/webkitpy/common/config/committers.py:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@95680 268f45cc-cd09-0410-ab3c-d52691b4dbfc

    [WK2] UIProcess should check that WebProcess isn't sending unexpected file: URLs to it
        https://bugs.webkit.org/show_bug.cgi?id=68573

        Reviewed by Anders Carlsson.

        * UIProcess/API/mac/WKView.mm:
        (maybeCreateSandboxExtensionFromPasteboard): Return a boolean, telling the caller whether
        an extension actually needed to be created
        (-[WKView performDragOperation:]): Tell process proxy when the process is going to get
        universal file read sandbox extension.

        * UIProcess/WebContext.cpp:
        (WebKit::WebContext::didPerformClientRedirect): Check the URLs.
        (WebKit::WebContext::didPerformServerRedirect): Ditto.
        (WebKit::WebContext::didUpdateHistoryTitle): Ditto.
        (WebKit::WebContext::getPluginPath): Ditto. Also, properly parse the URL - we can never
        assume that a string coming from WebProcess is a ParsedURLString.

        * UIProcess/WebPageProxy.cpp:
        (WebKit::WebPageProxy::reattachToWebProcessWithItem): Tell process proxy when the process
        is going to get universal file read sandbox extension.
        (WebKit::WebPageProxy::maybeInitializeSandboxExtensionHandle): Changed to return a boolean,
        telling the caller whether an extension actually needed to be created.
        (WebKit::WebPageProxy::loadURL): Tell process proxy about extension.
        (WebKit::WebPageProxy::loadURLRequest): Ditto.
        (WebKit::WebPageProxy::loadHTMLString): Tell process proxy if a file URL was used as a base
        one for a string. In this case, WebKit2 assumes that WebProcess has access to a subdirectory,
        (typically, one where error page resources live), and can load from it.
        (WebKit::WebPageProxy::loadAlternateHTMLString): Ditto.
        (WebKit::WebPageProxy::goForward): Tell process proxy about extension.
        (WebKit::WebPageProxy::goBack): Tell process proxy about extension.
        (WebKit::WebPageProxy::goToBackForwardItem): Tell process proxy about extension.
        (WebKit::WebPageProxy::didStartProvisionalLoadForFrame): Check the URL.
        (WebKit::WebPageProxy::didReceiveServerRedirectForProvisionalLoadForFrame): Ditto.
        (WebKit::WebPageProxy::didSameDocumentNavigationForFrame): Ditto.
        (WebKit::WebPageProxy::decidePolicyForNavigationAction): Ditto.
        (WebKit::WebPageProxy::decidePolicyForNewWindowAction): Ditto.
        (WebKit::WebPageProxy::decidePolicyForResponse): Ditto.
        (WebKit::WebPageProxy::didInitiateLoadForResource): Ditto.
        (WebKit::WebPageProxy::didSendRequestForResource): Ditto.
        (WebKit::WebPageProxy::didReceiveResponseForResource): Ditto.
        (WebKit::WebPageProxy::missingPluginButtonClicked): Ditto.

        * UIProcess/WebPageProxy.h: Changed initializeSandboxExtensionHandle() to return a bool,
        and renamed to maybeInitializeSandboxExtensionHandle (matching WKView counterpart).

        * UIProcess/WebProcessProxy.cpp:
        (WebKit::WebProcessProxy::WebProcessProxy): Initialize m_mayHaveUniversalFileReadSandboxExtension.
        It's going to be true if we ever granted an extension for "/".
        (WebKit::WebProcessProxy::willLoadHTMLStringWithBaseURL): Remember the path, we should expect
        that WebProcess will load subresources from it.
        (WebKit::WebProcessProxy::checkURLReceivedFromWebProcess): Check that it's reasonable to expect
        WebProcess send us a URL like this.
        (WebKit::WebProcessProxy::addBackForwardItem): Check the URLs.

        * UIProcess/WebProcessProxy.h: Added data members remembering what to expect from this process.

        * UIProcess/cf/WebPageProxyCF.cpp: (WebKit::WebPageProxy::restoreFromSessionStateData):
        Tell process proxy when the process is going to get universal file read sandbox extension.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@95679 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Add settings to control the availability of the Web Audio API to WebKit and WebKit2.
https://bugs.webkit.org/show_bug.cgi?id=68382

Source/WebCore:

Reviewed by Darin Adler.

Only create a AudioContext object if the Web Audio feature is runtime-enabled in Settings.

No new tests, as this feature is not yet enabled by default.

* bindings/js/JSDOMWindowCustom.cpp:
(WebCore::JSDOMWindow::webkitAudioContext):

Source/WebKit/mac:

Reviewed by Darin Adler.

Initialize the webAudioEnabled preference to NO by default.

* WebView/WebPreferences.mm:
(+[WebPreferences initialize]):

Source/WebKit2:

Add support for the webAudioEnabled preference in WebKit2.

Reviewed by Darin Adler.

* Shared/WebPreferencesStore.h: Add WebAudioEnabled getter and setter macro.
* UIProcess/API/C/WKPreferences.cpp:
(WKPreferencesSetWebAudioEnabled): Added.
(WKPreferencesGetWebAudioEnabled): Added.
* UIProcess/API/C/WKPreferences.h:
* WebProcess/WebPage/WebPage.cpp:
(WebKit::WebPage::updatePreferences): Update the webAudioEnabled preference.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@95678 268f45cc-cd09-0410-ab3c-d52691b4dbfc

[chromium] Code cleanup in gesture recognizer unit tests
https://bugs.webkit.org/show_bug.cgi?id=68561

Fix some style nits in the gesture recognizer unit tests.

Patch by Robert Kroeger <rjkroege@chromium.org> on 2011-09-21
Reviewed by David Levin.

* tests/InnerGestureRecognizerTest.cpp:
(InspectableGestureRecognizerChromium::stubEdgeFunction):
(BuildablePlatformTouchPoint::BuildablePlatformTouchPoint):
(TEST_F):

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@95677 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Should support value profiling on CPU(X86)
https://bugs.webkit.org/show_bug.cgi?id=68575

Reviewed by Sam Weinig.

Fix verbose profiling in ToT (SlowCaseProfile had been
partially renamed to RareCaseProfile), add in-memory
bucket counter for CPU(X86), move JIT::m_canBeOptimized
out of the DFG_JIT ifdef.

* bytecode/CodeBlock.cpp:
(JSC::CodeBlock::resetRareCaseProfiles):
(JSC::CodeBlock::dumpValueProfiles):
* bytecode/CodeBlock.h:
* dfg/DFGByteCodeParser.cpp:
(JSC::DFG::ByteCodeParser::makeSafe):
* jit/JIT.cpp:
(JSC::JIT::privateCompileSlowCases):
(JSC::JIT::privateCompile):
* jit/JIT.h:
* jit/JITInlineMethods.h:
(JSC::JIT::emitValueProfilingSite):

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@95676 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Remove checks for Leopard-only Objective-C APIs
https://bugs.webkit.org/show_bug.cgi?id=68579

Reviewed by Sam Weinig.

We can always assume that OBJC_API_VERSION is at least 2, it's been 2 since Leopard.

../WebCore:

* bridge/objc/objc_class.mm:
(JSC::Bindings::ObjcClass::methodsNamed):
(JSC::Bindings::ObjcClass::fieldNamed):
* bridge/objc/objc_instance.mm:
(allocateAutoReleasePool):
* bridge/objc/objc_runtime.mm:
(JSC::Bindings::ObjcField::ObjcField):
* page/mac/EventHandlerMac.mm:

../WebKit/mac:

* Misc/WebNSObjectExtras.h:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@95675 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Fix resolve-Changelog's strange butchery of the LayoutTests/Changelog.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@95674 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Remove Widget::beforeMouseDown and Widget::afterMouseDown
https://bugs.webkit.org/show_bug.cgi?id=68570

Reviewed by Sam Weinig.

These two member functions were originally added to avoid crashes due to removing NSViews while they were
being tracked by AppKit. Since they were added, we've moved away from NSViews for form controls, and the bugs in
AppKit that lead to crashes have been fixed.

* page/mac/EventHandlerMac.mm:
(WebCore::EventHandler::passMouseDownEventToWidget):
* platform/Widget.h:
* platform/mac/WidgetMac.mm:
(WebCore::Widget::Widget):
(WebCore::Widget::removeFromSuperview):

* rendering/RenderWidget.cpp:
Change widgetHierarchyUpdateSuspendCount to be an unsigned integer instead of a size_t.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@95673 268f45cc-cd09-0410-ab3c-d52691b4dbfc

DFG does not support compiling functions as constructors
https://bugs.webkit.org/show_bug.cgi?id=68500

Reviewed by Oliver Hunt.

This adds support for compiling constructors to the DFG. It's a
1% speed-up on V8, mostly due to a 6% speed-up on early-boyer.
It's also a 13% win on access-binary-trees, but it's neutral in
the SunSpider and Kraken averages.

* dfg/DFGByteCodeParser.cpp:
(JSC::DFG::ByteCodeParser::parseBlock):
* dfg/DFGCapabilities.h:
(JSC::DFG::mightCompileFunctionForConstruct):
(JSC::DFG::canCompileOpcode):
* dfg/DFGNode.h:
* dfg/DFGOperations.cpp:
* dfg/DFGOperations.h:
* dfg/DFGPropagator.cpp:
(JSC::DFG::Propagator::propagateNodePredictions):
(JSC::DFG::Propagator::performNodeCSE):
* dfg/DFGSpeculativeJIT.cpp:
(JSC::DFG::SpeculativeJIT::compile):
* runtime/Executable.cpp:
(JSC::FunctionExecutable::compileOptimizedForConstruct):
(JSC::FunctionExecutable::compileForConstructInternal):
* runtime/Executable.h:
(JSC::FunctionExecutable::compileForConstruct):
(JSC::FunctionExecutable::compileFor):
(JSC::FunctionExecutable::compileOptimizedFor):

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@95672 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Source/WebCore: Not use anonymousContainer on beforeChild calculation
and let later code handle it. Fix style update on block-level
anonymous objects.
https://bugs.webkit.org/show_bug.cgi?id=67656

Reviewed by David Hyatt.

Test: fast/table/quote-text-around-iframe.html

* rendering/RenderTable.cpp:
(WebCore::RenderTable::addChild):
* rendering/RenderTableCell.cpp:
(WebCore::RenderTableCell::styleDidChange):
* rendering/RenderTableRow.cpp:
(WebCore::RenderTableRow::addChild):
* rendering/RenderObject.cpp:
(WebCore::RenderObject::propagateStyleToAnonymousChildren):

LayoutTests: Not use anonymousContainer on beforeChild calculation
and let later code handle it.
https://bugs.webkit.org/show_bug.cgi?id=67656

Reviewed by David Hyatt.

* fast/table/quote-text-around-iframe.html: Added.
* fast/table/table-row-style-not-updated-with-after-content-expected.txt:
* platform/mac/fast/table/quote-text-around-iframe-expected.png: Added.
* platform/mac/fast/table/quote-text-around-iframe-expected.txt: Added.
and lot of style propagation rebaselines.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@95671 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Source/WebCore: https://bugs.webkit.org/show_bug.cgi?id=68562

Pagination struts don't update properly when collapsing margin estimates are wrong.

Refactor the pagination adjustments that happen during block layout to allow for
a 3-pass layout system instead of just a 2-pass layout system. The basic idea is
to come up with an estimate that includes collapsing, clearance and pagination.

If the estimate only with collapsing and clearance was wrong, then we lay out again
using the correct collapsing margin values and clearance before applying pagination breaks.

After this second layout, the normal code that runs when unpaginated compares estimates
and lays out a third time to account for adjustments as a result of pagination movement.

Reviewed by Sam Weinig.

Added fast/regions/text-regions-split-small-pagination.html

* rendering/RenderBlock.cpp:
(WebCore::RenderBlock::estimateLogicalTopPosition):
(WebCore::RenderBlock::layoutBlockChild):
(WebCore::RenderBlock::adjustBlockChildForPagination):
* rendering/RenderBlock.h:

LayoutTests: https://bugs.webkit.org/show_bug.cgi?id=68562

Pagination struts don't update properly when collapsing margin estimates are wrong.

Add a new regions result to test the bug fix.

Reviewed by Sam Weinig.

* fast/regions/text-region-split-small-pagination.html: Added.
* platform/mac/fast/regions/text-region-split-small-pagination-expected.png: Added.
* platform/mac/fast/regions/text-region-split-small-pagination-expected.txt: Added.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@95670 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Remove old and busted context notification APIs
https://bugs.webkit.org/show_bug.cgi?id=68381

Reviewed by David Levin.

* public/WebFrameClient.h:
* src/FrameLoaderClientImpl.cpp:
(WebKit::FrameLoaderClientImpl::didCreateScriptContext):
(WebKit::FrameLoaderClientImpl::willReleaseScriptContext):

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@95669 268f45cc-cd09-0410-ab3c-d52691b4dbfc

[chromium] SharedGraphicsContext3D's context should not be destroyed during atexit
https://bugs.webkit.org/show_bug.cgi?id=68558

Reviewed by Stephen White.

The SharedGraphicsContext3D's internal context is lazily created and then leaked intentionally, but since the
last ref is in a function-static RefPtr<> it ends up being destroyed at process exit time. This is bad, since
by the time we get this far in process shutdown we can't actually cleanly run the destruction logic and we
crash. Since we are explicitly leaking this for the lifetime of the process there's no point in keeping it in a
RefPtr<>

* platform/graphics/gpu/SharedGraphicsContext3D.cpp:
(WebCore::SharedGraphicsContext3D::get):

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@95668 268f45cc-cd09-0410-ab3c-d52691b4dbfc

[Chromium] Protect the Frame in V8HTMLDocument::openCallback
https://bugs.webkit.org/show_bug.cgi?id=68555

Patch by Sergey Glazunov <serg.glazunov@gmail.com> on 2011-09-21
Reviewed by Nate Chapin.

Source/WebCore:

Test: fast/dom/frame-deleted-in-document-open.html

* bindings/v8/custom/V8HTMLDocumentCustom.cpp:
(WebCore::V8HTMLDocument::openCallback):

LayoutTests:

* fast/dom/frame-deleted-in-document-open-expected.txt: Added.
* fast/dom/frame-deleted-in-document-open.html: Added.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@95667 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Replace jsFunctionVPtr compares with a type check on the Structure.
https://bugs.webkit.org/show_bug.cgi?id=68557

Reviewed by Oliver Hunt.

This will permit calls to still optimize to subclasses of JSFunction
that have the correct type (but a different C++ vptr).

This patch stops passing the globalData into numerous functions.

* dfg/DFGByteCodeParser.cpp:
(JSC::DFG::ByteCodeParser::parseBlock):
* dfg/DFGGraph.h:
(JSC::DFG::Graph::isFunctionConstant):
(JSC::DFG::Graph::valueOfFunctionConstant):
* dfg/DFGJITCompiler.h:
(JSC::DFG::JITCompiler::isFunctionConstant):
(JSC::DFG::JITCompiler::valueOfFunctionConstant):
* dfg/DFGOperations.cpp:
* interpreter/Interpreter.cpp:
(JSC::Interpreter::privateExecute):
* jit/JIT.h:
* jit/JITCall.cpp:
(JSC::JIT::compileOpCallVarargs):
(JSC::JIT::compileOpCallSlowCase):
* jit/JITCall32_64.cpp:
(JSC::JIT::compileOpCallVarargs):
(JSC::JIT::compileOpCallSlowCase):
* jit/JITInlineMethods.h:
(JSC::JIT::emitJumpIfNotType):
* jit/JITStubs.cpp:
(JSC::DEFINE_STUB_FUNCTION):
* runtime/Executable.h:
(JSC::isHostFunction):
* runtime/JSFunction.h:
(JSC::JSFunction::createStructure):
* runtime/JSObject.cpp:
(JSC::JSObject::put):
(JSC::JSObject::putWithAttributes):
* runtime/JSObject.h:
(JSC::getJSFunction):
(JSC::JSObject::putDirect):
(JSC::JSObject::putDirectWithoutTransition):
* runtime/JSType.h:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@95666 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Unreviewed. Remove unused ENABLE_AS_IMAGE flag after r95234 from
Options{Efl,WinCE}.cmake.

* Source/cmake/OptionsEfl.cmake:
* Source/cmake/OptionsWinCE.cmake:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@95665 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Clean up CSS Counter code
https://bugs.webkit.org/show_bug.cgi?id=68492

Reviewed by Tony Chang.

Source/WebCore:

Several cleanups/fixes:
- Make Counter::listStyle() return the actual list style string,
rather than empty string.
- Handle a specified list style of "none" correctly when generating
cssText and in Counter::listStyle().
- Use StringBuilder when generating cssText.

Test: fast/css/counters/getCounterValue.html

* css/CSSParser.cpp:
(WebCore::CSSParser::parseCounterContent):
Simplify processing (moving conversion to CSSStyleSelector),
pass a CSS_IDENT as the listStyle to Counter::create.
* css/CSSPrimitiveValue.cpp:
(WebCore::CSSPrimitiveValue::cssText):
Use StringBuilder.
* css/CSSStyleSelector.cpp:
(WebCore::CSSStyleSelector::applyProperty):
Convert from a CSS_IDENT to an EListStyleType.
* css/Counter.h:
(WebCore::Counter::listStyleIdent):
Renamed from listStyleNumber to match its new functionality.

LayoutTests:

Added tests for setting the list style to "none"
as well as tests for getCounterValue.

* fast/css/counters/counter-cssText-expected.txt:
* fast/css/counters/counter-cssText.html:
* fast/css/counters/getCounterValue-expected.txt: Added.
* fast/css/counters/getCounterValue.html: Added.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@95664 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Removed WTFTHREADDATA_MULTITHREADED, making it always true
https://bugs.webkit.org/show_bug.cgi?id=68549

Reviewed by Darin Adler.

Another part of making threads exist in WebKit.

* wtf/WTFThreadData.cpp:
* wtf/WTFThreadData.h:
(WTF::wtfThreadData):

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@95663 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2011-09-21  Chris Marrin  <cmarrin@apple.com>

        Check in WebCore.vcproj file unchanged but written like MSDev likes it

        Unreviewed.

        * WebCode.vcproj/WebCore.vcproj:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@95662 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Use the prefix "WK" instead of "Web" for Objective-C classes defined in WebKit2.

Reviewed by Anders Carlsson.

Renamed WebInspectorProxyObjCAdapter to WKWebInspectorProxyObjCAdapter.
Renamed WebInspectorWKView to WKWebInspectorWKView.
Renamed WebFindIndicatorWindowAnimation to WKFindIndicatorWindowAnimation.
Renamed WebFindIndicatorView to WKFindIndicatorView.
Renamed WebUserDataWrapper to WKUserDataWrapper.

Updated the build phase checking for inappropriate Objective-C class names to disallow the prefix
"Web" and only allow "WK".

* UIProcess/API/mac/FindIndicatorWindow.h:
* UIProcess/API/mac/FindIndicatorWindow.mm:
(WebKit::FindIndicatorWindow::setFindIndicator):
(WebKit::FindIndicatorWindow::startFadeOutTimerFired):
* UIProcess/WebInspectorProxy.h:
* UIProcess/mac/WebContextMenuProxyMac.mm:
(-[WKMenuTarget forwardContextMenuAction:]):
(WebKit::nsMenuItemVector):
* UIProcess/mac/WebInspectorProxyMac.mm:
(WebKit::WebInspectorProxy::platformCreateInspectorPage):
(WebKit::WebInspectorProxy::platformOpen):
* WebKit2.xcodeproj/project.pbxproj:
* WebProcess/FullScreen/mac/WebFullScreenManagerMac.mm:
(WebKit::WebFullScreenManagerMac::WebFullScreenManagerMac):

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@95661 268f45cc-cd09-0410-ab3c-d52691b4dbfc

[WK2] [Mac] Implement a more-complete MouseDown/MouseUp/MouseMoveTo functions for WebKit2 EventSender
https://bugs.webkit.org/show_bug.cgi?id=68108

This patch implements the event sender mouse events through WebKit2 UIProcess which is
closer to the real simulation than a WebProcess-only approach. The patch only supports Mac
platform as the first step so the existing code is still kept for other platforms.

Source/WebKit2:

Patch by Chang Shu <cshu@webkit.org> on 2011-09-21
Reviewed by Darin Adler.

* UIProcess/API/C/WKPage.cpp:
(WKPageSetShouldSendEventsSynchronously):
* UIProcess/API/C/WKPagePrivate.h:
* UIProcess/WebPageProxy.cpp:
(WebKit::WebPageProxy::WebPageProxy):
(WebKit::WebPageProxy::handleMouseEvent):
(WebKit::WebPageProxy::handleKeyboardEvent):
* UIProcess/WebPageProxy.h:
(WebKit::WebPageProxy::setShouldSendEventsSynchronously):
* WebProcess/WebPage/WebPage.cpp:
(WebKit::WebPage::mouseEventSyncForTesting):
* WebProcess/WebPage/WebPage.h:
* WebProcess/WebPage/WebPage.messages.in:

Tools:

Patch by Chang Shu <cshu@webkit.org> on 2011-09-21
Reviewed by Darin Adler.

* WebKitTestRunner/EventSenderProxy.h:
(WTR::EventSenderProxy::EventSenderProxy):
(WTR::EventSenderProxy::leapForward):
(WTR::EventSenderProxy::currentEventTime):
* WebKitTestRunner/InjectedBundle/EventSendingController.cpp:
(WTR::EventSendingController::EventSendingController):
(WTR::EventSendingController::mouseDown):
(WTR::EventSendingController::mouseUp):
(WTR::EventSendingController::mouseMoveTo):
(WTR::EventSendingController::leapForward):
(WTR::EventSendingController::keyDown):
* WebKitTestRunner/InjectedBundle/EventSendingController.h:
* WebKitTestRunner/TestController.cpp:
(WTR::TestController::didReceiveSynchronousMessageFromInjectedBundle):
* WebKitTestRunner/mac/EventSenderProxy.mm:
(WTR::eventTypeForMouseButtonAndAction):
(WTR::operator==):
(WTR::EventSenderProxy::updateClickCountForButton):
(WTR::EventSenderProxy::mouseDown):
(WTR::EventSenderProxy::mouseUp):
(WTR::EventSenderProxy::mouseMoveTo):
(WTR::EventSenderProxy::keyDown):
* WebKitTestRunner/qt/EventSenderProxyQt.cpp:
(WTR::EventSenderProxy::keyDown):
(WTR::EventSenderProxy::mouseDown):
(WTR::EventSenderProxy::mouseUp):
(WTR::EventSenderProxy::mouseMoveTo):

LayoutTests:

Unskip the passed tests.

Patch by Chang Shu <cshu@webkit.org> on 2011-09-21
Reviewed by Darin Adler.

* platform/mac-wk2/Skipped:
* platform/wk2/Skipped:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@95660 268f45cc-cd09-0410-ab3c-d52691b4dbfc

[EFL] Revert pointer operator coding style in ewk
https://bugs.webkit.org/show_bug.cgi?id=68231

Reviewed by Antonio Gomes.

The WebKit coding style mandates that "(...) in C++ code - Both pointer
types and reference types should be written with no space between the
type name and the * or &.". Until now EFL port was using the
coding-style of the library for platform code instead of WebKit's.

We are now trying to be more compliant with WebKit coding-style so it's
easier for other reviewers to review EFL code. The first step in this
direction is to move the '*' and '&' of pointers and references to the
right place.

This patch was generated with 'uncrustify' with the configure file
attached to the respective but report. There was one fix that had to be
manually made in order to keep the 'check-webkit-style' happy: a
function definition with a meaningless name. Other functions like this
exist but they will be fixed by a later patch.

Some nice side-effects of using uncrustify:

- Whitespace cleanup on comments
- Function parameters are aligned to the open parenthesis
- Remove tab characters

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@95659 268f45cc-cd09-0410-ab3c-d52691b4dbfc

[chromium] Optimize ImageBuffer constructor when accelerated
https://bugs.webkit.org/show_bug.cgi?id=68501

If the accelerated rendering mode is passed in to ImageBuffer, instead
of creating a bitmap PlatformCanvas and subsequently replacing its
device with an SkGpuDevice, create an SkCanvas and SkGpuDevice first,
and fall back to raster creation only failure. This saves on system
RAM for the backing store (and possibly GDI resources).

Reviewed by James Robinson.

Covered by existing tests.

* platform/graphics/gpu/SharedGraphicsContext3D.cpp:
(WebCore::SharedGraphicsContext3D::get):
* platform/graphics/gpu/SharedGraphicsContext3D.h:
Rename "SharedGraphicsContext3D::create()" to "get()" to more
accurately reflect its function.
* platform/graphics/skia/ImageBufferSkia.cpp:
(WebCore::createAcceleratedCanvas):
Refactor creation of accelerated ImageBuffer.
(WebCore::ImageBuffer::ImageBuffer):
Attempt creation of accelerated ImageBuffer first, and only fall back
to bitmap-backed canvas creation if it fails.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@95658 268f45cc-cd09-0410-ab3c-d52691b4dbfc

[EFL] Don't try to free user-provided list
https://bugs.webkit.org/show_bug.cgi?id=68356

Reviewed by Antonio Gomes.

A coding-style cleanup revealed that we were using EINA_LIST_FREE in a
wrong manner. If we intended to free the data, we should call free()
(or whatever function is required to destroy it). However not only did
it use the wrong EFL api but it also introduced a change in behavior,
freeing user-provided data.

* ewk/ewk_frame.cpp:
(ewk_frame_feed_touch_event): do not free user-provided data.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@95656 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Source/JavaScriptCore: JavaScriptCore Part of: Prevent the WebKit frameworks from defining inappropriately-named Objective-C classes
https://bugs.webkit.org/show_bug.cgi?id=68451

Reviewed by Darin Adler.

* JavaScriptCore.xcodeproj/project.pbxproj: Added a script build phase that invokes
check-for-inappropriate-objc-class-names, allowing only class names prefixed with "JS".

Source/WebCore: WebCore part of: Prevent the WebKit frameworks from defining inappropriately-named Objective-C classes
https://bugs.webkit.org/show_bug.cgi?id=68451

Reviewed by Darin Adler.

* WebCore.xcodeproj/project.pbxproj: Added a script build phase that invokes
check-for-inappropriate-objc-class-names, allowing only class names prefixed with "DOM" or "Web".

Source/WebKit: WebKit part of: Prevent the WebKit frameworks from defining inappropriately-named Objective-C classes
https://bugs.webkit.org/show_bug.cgi?id=68451

Reviewed by Darin Adler.

* WebKit.xcodeproj/project.pbxproj: Added a script build phase that invokes
check-for-inappropriate-objc-class-names, allowing only class names prefixed with "Web" or "_Web".

Source/WebKit2: WebKit2 part of: Prevent the WebKit frameworks from defining inappropriately-named Objective-C classes
https://bugs.webkit.org/show_bug.cgi?id=68451

Reviewed by Darin Adler.

* WebKit2.xcodeproj/project.pbxproj: Added a script build phase that invokes
check-for-inappropriate-objc-class-names, allowing only class names prefixed with "WK" or "Web".

Tools: Tools part of: Prevent the WebKit frameworks from defining inappropriately-named Objective-C classes
https://bugs.webkit.org/show_bug.cgi?id=68451

Reviewed by Darin Adler.

* Scripts/check-for-inappropriate-objc-class-names: Added. Checks for
Objective-C classes with names not having one of a list of prefixes
passed on the command line.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@95655 268f45cc-cd09-0410-ab3c-d52691b4dbfc