2011-04-11  Ryosuke Niwa  <rniwa@webkit.org>

        Reviewed by Tony Chang.

        [chromium] Crash in WebViewImpl::caretOrSelectionBounds
        https://bugs.webkit.org/show_bug.cgi?id=58269

        The bug was caused by caretOrSelectionBounds's incorrectly assuming
        SelectionController::toNormalizedRange to always return a non-null Range.

        Fixed the bug by adding a null pointer check. Also replaced calls to deprecatedNode
        by containerNode() and calls to SelectionController::start() and SelectionController::end()
        by calls to SelectionController::base() and SelectionController::extent() because
        selection extends from base to extent, not from start to end.

        Test: editing/selection/extend-over-file-input-by-drag-crash.html

        * src/WebViewImpl.cpp:
        (WebKit::WebViewImpl::caretOrSelectionBounds):
2011-04-11  Ryosuke Niwa  <rniwa@webkit.org>

        Reviewed by Tony Chang.

        [chromium] Crash in WebViewImpl::caretOrSelectionBounds
        https://bugs.webkit.org/show_bug.cgi?id=58269

        Added a test to ensure WebKit does not crash when selecting over a file input element.
        While the bug was specific to Chromium port, the test will be run on all ports because
        all other ports should not crash either.

        * editing/selection/extend-over-file-input-by-drag-crash-expected.txt: Added.
        * editing/selection/extend-over-file-input-by-drag-crash.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@83548 268f45cc-cd09-0410-ab3c-d52691b4dbfc

<rdar://problem/9251515> and https://bugs.webkit.org/show_bug.cgi?id=58072
URLs accessed as a result of pushState/replaceState should count as "visited" and be in the global history

Reviewed by Maciej Stachowiak.

Source/WebCore:

Test: fast/loader/stateobjects/state-url-sets-links-visited.html

* loader/HistoryController.cpp:
(WebCore::HistoryController::pushState): Notify the history client and add the visited link.
(WebCore::HistoryController::replaceState): Ditto.

LayoutTests:

* fast/loader/stateobjects/state-url-sets-links-visited-expected.txt: Added.
* fast/loader/stateobjects/state-url-sets-links-visited.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@83547 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2011-04-11  George Guo  <George.Guo@nokia.com>

        Reviewed by Eric Seidel.

        prepare-ChangeLog support email input but -h did not show the option
        https://bugs.webkit.org/show_bug.cgi?id=58164

        Add email option to -h to improve the usability

        * Scripts/prepare-ChangeLog:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@83546 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2011-04-11  Dimitri Glazkov  <dglazkov@chromium.org>

        Reviewed by Eric Carlson.

        Rename MediaControls to MediaControlRootElement.
        https://bugs.webkit.org/show_bug.cgi?id=58250

        * Scripts/do-webcore-rename: Documented the renaming.
2011-04-11  Dimitri Glazkov  <dglazkov@chromium.org>

        Reviewed by Eric Carlson.

        Rename MediaControls to MediaControlRootElement.
        https://bugs.webkit.org/show_bug.cgi?id=58250

        Mechanical move using do-webcore-rename.
2011-04-11  Dimitri Glazkov  <dglazkov@chromium.org>

        Reviewed by Eric Carlson.

        Rename MediaControls to MediaControlRootElement.
        https://bugs.webkit.org/show_bug.cgi?id=58250

        Mechanical move using do-webcore-rename.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@83545 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2011-04-11  Dean Jackson  <dino@apple.com>

        Reviewed by Simon Fraser.

        https://bugs.webkit.org/show_bug.cgi?id=58285
        CompositeAnimation is cleared while an AnimationBase is waiting for notification

        Occasionally hardware animations would trigger a state where
        they would tell the AnimationControllerPrivate that it should notify
        any waiting animations that they are ready to start, even though
        the RenderObject had gone away and hence the CompositeAnimation
        had been destroyed. The fix is to zero the pointer from AnimationBase
        when the CompositeAnimation destructor runs (or its animations are
        cleared) and test the validity of the CompositeAnimation before it
        is used in AnimationBase.

        * page/animation/AnimationBase.cpp:
        (WebCore::AnimationBase::updateStateMachine):
        (WebCore::AnimationBase::fireAnimationEventsIfNeeded):
        (WebCore::AnimationBase::updatePlayState):
        (WebCore::AnimationBase::freezeAtTime):
        (WebCore::AnimationBase::beginAnimationUpdateTime):
        - test that the CompositeAnimation exists
        * page/animation/AnimationBase.h:
        (WebCore::AnimationBase::clear):
        - rename clearRenderer() to clear() as it now also clears the CompositeAnimation
        * page/animation/CompositeAnimation.cpp:
        (WebCore::CompositeAnimation::clearRenderer):
        (WebCore::CompositeAnimation::updateKeyframeAnimations):
        - call clear() rather than clearRenderer()

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@83542 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2011-04-11  Justin Novosad  <junov@chromium.org>

        Reviewed by Kenneth Russell.

        [Chromium] Text anti-aliasing fails when rendering text with shadow
        https://bugs.webkit.org/show_bug.cgi?id=57584

        * platform/chromium/test_expectations.txt:
        Tempoaray test failures that will need to be rebaselined
2011-04-11  Justin Novosad  <junov@chromium.org>

        Reviewed by Kenneth Russell.

        [Chromium] Text anti-aliasing fails when rendering text with shadow
        https://bugs.webkit.org/show_bug.cgi?id=57584

        * platform/graphics/chromium/FontChromiumWin.cpp:
        (WebCore::drawGlyphsWin): new static function, code factored-out
        from WebCore::Font::drawGlyphs
        (WebCore::Font::drawGlyphs): may separate font rendering into two
        passes to allow foreground glyphs to be rendered with GDI when Skia
        is required for rendering shadows
        * platform/graphics/skia/SkiaFontWin.cpp:
        (WebCore::windowsCanHandleDrawTextShadow):
        Modified method so that it does not assume a shadow is present, and
        returns true if there is no shadow (GDI can draw 'no shadow')
        (WebCore::windowsCanHandleTextDrawing):
        Factored out into sub methods so that the the shadow and foregound
        glyph properties can be tested separately independently.
        (WebCore::windowsCanHandleTextDrawingWithoutShadow):
        New method returns true if the foreground glyph only has effect
        that can be draw with GDI
        (WebCore::paintSkiaText):
        * platform/graphics/skia/SkiaFontWin.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@83541 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2011-04-11  Ryosuke Niwa  <rniwa@webkit.org>

        Unreviewed, rolling out r83515.
        http://trac.webkit.org/changeset/83515
        https://bugs.webkit.org/show_bug.cgi?id=57178

        Broke GTK builds. nextOnLineExists and prevOnLineExists are still used in AccessibilityObjectWrapperAtk.cpp.

        * rendering/InlineBox.cpp:
        (WebCore::InlineBox::nextOnLineExists):
        (WebCore::InlineBox::prevOnLineExists):
        * rendering/InlineBox.h:
        (WebCore::InlineBox::InlineBox):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@83540 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2011-04-11  Alexis Menard  <alexis.menard@openbossa.org>

        Unreviewed build fix for Mac.

        * WebKit.xcodeproj/project.pbxproj:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@83539 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Reverting, seeing something else still broken.  Sorry.

* html/HTMLTrackElement.cpp:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@83538 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Fix the build.

* html/HTMLTrackElement.cpp: Case-sensitive, please.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@83537 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Unreviewed, update chromium test expectations.

Mark failed tests in chromium that could be caused by r83514 & r83518.

* platform/chromium/test_expectations.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@83536 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Fix the build.  Don't enable TRACK, whatever that is, in an inconsistent manner.

Different projects had different settings at the Xcode level, and the Xcode settings didn't match the preprocessor define.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@83535 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2011-04-11  Alexis Menard  <alexis.menard@openbossa.org>

        Unreviewed build fix for Mac.

        An idl file was added in the wrong place.

        * WebCore.xcodeproj/project.pbxproj:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@83534 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Fix Windows build.

* MiniBrowser/win/BrowserView.cpp:
(BrowserView::create):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@83533 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2011-04-11  Chris Marrin  <cmarrin@apple.com>

        Rubberstamped by Dan Bernstein.

        xhr.responseType = 'arraybuffer' works on Mac but not on Windows
        https://bugs.webkit.org/show_bug.cgi?id=50334

        Got rid of guards around arraybuffer logic in XHR. ArrayBuffer and friends
        are always turned on now, so there's not need for the guard. I reenabled the
        test that was Skipped on Windows and it works now.

        * bindings/js/JSXMLHttpRequestCustom.cpp:
        (WebCore::JSXMLHttpRequest::markChildren):
        (WebCore::JSXMLHttpRequest::send):
        (WebCore::JSXMLHttpRequest::response):
        * xml/XMLHttpRequest.cpp:
        (WebCore::XMLHttpRequest::responseArrayBuffer):
        (WebCore::XMLHttpRequest::setResponseType):
        (WebCore::XMLHttpRequest::send):
        (WebCore::XMLHttpRequest::clearResponseBuffers):
        (WebCore::XMLHttpRequest::didReceiveData):
        * xml/XMLHttpRequest.h:
        (WebCore::XMLHttpRequest::optionalResponseArrayBuffer):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@83532 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2011-04-11  Sam Weinig  <sam@webkit.org>

        Reviewed by Anders Carlsson.

        Move focus management out of WebKit via  the UIClient
        <rdar://problem/8784068>
        https://bugs.webkit.org/show_bug.cgi?id=58278

        * MiniBrowser/mac/BrowserWindowController.m:
        (-[BrowserWindowController awakeFromNib]):
        * WebKitTestRunner/TestController.cpp:
        (WTR::TestController::createOtherPage):
        (WTR::TestController::initialize):
        Add stubs for new UIClient functions.
2011-04-11  Sam Weinig  <sam@webkit.org>

        Reviewed by Anders Carlsson.

        Move focus management out of WebKit via  the UIClient
        <rdar://problem/8784068>
        https://bugs.webkit.org/show_bug.cgi?id=58278

        * UIProcess/API/C/WKPage.h:
        * UIProcess/API/mac/PageClientImpl.h:
        * UIProcess/API/mac/PageClientImpl.mm:
        * UIProcess/API/qt/qwkpage.cpp:
        (QWKPage::QWKPage):
        * UIProcess/API/qt/qwkpage_p.h:
        * UIProcess/PageClient.h:
        * UIProcess/WebPageProxy.cpp:
        (WebKit::WebPageProxy::setFocus):
        * UIProcess/WebUIClient.cpp:
        (WebKit::WebUIClient::focus):
        (WebKit::WebUIClient::unfocus):
        * UIProcess/WebUIClient.h:
        * UIProcess/gtk/WebView.cpp:
        * UIProcess/gtk/WebView.h:
        * UIProcess/win/WebView.h:
        Remove PageClient::setFocus() in favor of WebUIClient::focus and WebUIClient::unfocus.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@83531 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2011-02-03  Martin Robinson  <mrobinson@igalia.com>

        Reviewed by Eric Seidel.

        [GTK] The GTK+ DRT needs an implementation of the PlainTextController
        https://bugs.webkit.org/show_bug.cgi?id=53605

        * platform/gtk/Skipped: Unskipped test which is now passing.
2011-02-03  Martin Robinson  <mrobinson@igalia.com>

        Reviewed by Eric Seidel.

        [GTK] The GTK+ DRT needs an implementation of the PlainTextController
        https://bugs.webkit.org/show_bug.cgi?id=53605

        Added jsValueToDOMRange to DumpRenderTreeSupportGtk. This static method can convert
        a JSValueRef to a WebKitDOMRange. When it's possible to do this with the public
        API, we can remove this method.

        * WebCoreSupport/DumpRenderTreeSupportGtk.cpp:
        (DumpRenderTreeSupportGtk::jsValueToDOMRange): Added.
        * WebCoreSupport/DumpRenderTreeSupportGtk.h: Added declaration.
2011-02-03  Martin Robinson  <mrobinson@igalia.com>

        Reviewed by Eric Seidel.

        [GTK] The GTK+ DRT needs an implementation of the PlainTextController
        https://bugs.webkit.org/show_bug.cgi?id=53605

        Add an implementation of the PlainTextController for the GTK+ DRT.

        * DumpRenderTree/gtk/DumpRenderTree.cpp:
        (addControllerToWindow): Added this helper method.
        (webViewWindowObjectCleared): Use the new helper to add EventSender
        and the PlainTextController to the window object.
        * GNUmakefile.am: Added PlainTextController source files.
        * DumpRenderTree/gtk/PlainTextController.cpp: Added. This will rely
        on DumpRenderTreeSupportGtk to convert a JSValueRef into a WebKitDOMRange
        until that functionality exists somewhere in the WebKitGTK+ stack.
        * DumpRenderTree/gtk/PlainTextController.h: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@83530 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Fix the build.

* JavaScriptCore.xcodeproj/project.pbxproj: Headers used outside of JavaScriptCore need to be marked as private.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@83529 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2011-04-11  Stephen White  <senorblanco@chromium.org>

        Unreviewed; new chromium baselines.

        New chromium-mac-leopard baseline images for tests affected by r83490.

        * platform/chromium-mac-leopard/svg/custom/foreign-object-skew-expected.png:
        * platform/chromium-mac-leopard/svg/custom/scrolling-embedded-svg-file-image-repaint-problem-expected.checksum: Removed.
        * platform/chromium-mac-leopard/svg/custom/scrolling-embedded-svg-file-image-repaint-problem-expected.png:
        * platform/chromium-mac-leopard/svg/zoom/page/zoom-foreignObject-expected.checksum: Removed.
        * platform/chromium-mac-leopard/svg/zoom/page/zoom-foreignObject-expected.png: Removed.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@83528 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2011-04-11  Anna Cavender  <annacc@chromium.org>

        Reviewed by Eric Carlson.

        Setup ENABLE(TRACK) feature define
        https://bugs.webkit.org/show_bug.cgi?id=53556

        * configure.ac:
2011-04-11  Anna Cavender  <annacc@chromium.org>

        Reviewed by Eric Carlson.

        Setup ENABLE(TRACK) feature define
        https://bugs.webkit.org/show_bug.cgi?id=53556

        * Configurations/FeatureDefines.xcconfig:
2011-04-11  Anna Cavender  <annacc@chromium.org>

        Reviewed by Eric Carlson.

        Setup ENABLE(TRACK) feature define + initial HTMLTrackElement
        https://bugs.webkit.org/show_bug.cgi?id=53556

        * WebKit.xcodeproj/project.pbxproj:
2011-04-11  Anna Cavender  <annacc@chromium.org>

        Reviewed by Eric Carlson.

        Setup ENABLE(TRACK) feature define + initial HTMLTrackElement
        https://bugs.webkit.org/show_bug.cgi?id=53556

        No new tests. No real functionality added just yet.

        * CMakeLists.txt:
        * CodeGenerators.pri:
        * Configurations/FeatureDefines.xcconfig:
        * DerivedSources.make:
        * GNUmakefile.am:
        * GNUmakefile.list.am:
        * WebCore.gypi:
        * WebCore.xcodeproj/project.pbxproj:
        * features.pri:
        * html/HTMLAttributeNames.in:
        * html/HTMLTagNames.in:
        * html/HTMLTrackElement.cpp: Added.
        (WebCore::HTMLTrackElement::HTMLTrackElement):
        (WebCore::HTMLTrackElement::create):
        (WebCore::HTMLTrackElement::insertedIntoTree):
        (WebCore::HTMLTrackElement::willRemove):
        (WebCore::HTMLTrackElement::src):
        (WebCore::HTMLTrackElement::setSrc):
        (WebCore::HTMLTrackElement::kind):
        (WebCore::HTMLTrackElement::setKind):
        (WebCore::HTMLTrackElement::srclang):
        (WebCore::HTMLTrackElement::setSrclang):
        (WebCore::HTMLTrackElement::label):
        (WebCore::HTMLTrackElement::setLabel):
        (WebCore::HTMLTrackElement::isDefault):
        (WebCore::HTMLTrackElement::setIsDefault):
        (WebCore::HTMLTrackElement::isURLAttribute):
        * html/HTMLTrackElement.h: Added.
        * html/HTMLTrackElement.idl: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@83527 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2011-04-11  Ben Taylor  <bentaylor.solx86@gmail.com>

        Reviewed by David Levin.

        https://bugs.webkit.org/show_bug.cgi?id=57535

        Better compile fix for Solaris 10/Sun Studio 12 CC
        needing <sys/time.h>

        * loader/icon/IconRecord.h:
        * page/Page.h:
        * platform/network/ResourceResponseBase.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@83526 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Add radar numbers to two ChangeLog entries. (Rubber-stamped by Brian Weinstein).

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@83525 268f45cc-cd09-0410-ab3c-d52691b4dbfc

    Reviewed by Maciej Stachowiak.

        WebKit2: Cannot use Ctrl-Delete as a custom keyboard shortcut
        https://bugs.webkit.org/show_bug.cgi?id=58265
        <rdar://problem/9221468>

        * WebProcess/WebPage/WebPage.cpp:
        (WebKit::WebPage::keyEvent): Added a comment explaining that doing work after DOM event
        dispatch isn't great.
        * WebProcess/WebPage/mac/WebPageMac.mm:
        (WebKit::WebPage::performNonEditingBehaviorForSelector): Added a comment explaining how this
        might be moved down to WebCore.
        (WebKit::WebPage::performDefaultBehaviorForKeyEvent): This is now empty, since both Space
        and Backspace behaviors are implemented in WebCore.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@83522 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2011-04-11  Matthew Delaney  <mdelaney@apple.com>

        Reviewed by Dan Bernstein.

        [CG] drawImage with a subimage bleeds pixel data at borders
        https://bugs.webkit.org/show_bug.cgi?id=58267

        Test: fast/canvas/DrawImageSinglePixelStretch.html

        * platform/graphics/cg/ImageCG.cpp:
        (WebCore::BitmapImage::draw): Any interpolation quality other than 'none' will graft out the subimage first now.
2011-04-11  Matthew Delaney  <mdelaney@apple.com>

        Reviewed by Dan Bernstein.

        [CG] drawImage with a subimage bleeds pixel data at borders
        https://bugs.webkit.org/show_bug.cgi?id=58267

        * fast/canvas/DrawImageSinglePixelStretch.html: Added test case.
        * fast/canvas/resources/orangePixels.gif: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@83521 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2011-04-11  Geoffrey Garen  <ggaren@apple.com>

        Try to fix a few builds.

        Updated a few more build configurations for file moves.

        * CMakeListsWinCE.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@83520 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Try to fix the Windows build.

* UIProcess/win/WebInspectorProxyWin.cpp:
(WebKit::WebInspectorProxy::platformCreateInspectorPage):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@83519 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2011-04-11  Chris Marrin  <cmarrin@apple.com>

        Reviewed by Simon Fraser.

        iframe/compositing propagation should also work for <object> and framesets
        https://bugs.webkit.org/show_bug.cgi?id=39037

        Added pixel tests to for composited framesets and objects.

        * compositing/framesets: Added.
        * compositing/framesets/composited-frame-alignment.html: Added.
        * compositing/framesets/composited-frame-alignment-expected.checksum: Added.
        * compositing/framesets/composited-frame-alignment-expected.png: Added.
        * compositing/framesets/composited-frame-alignment-expected.txt: Added.
        * compositing/framesets/resources: Added.
        * compositing/framesets/resources/composited-subframe.html: Added.
        * compositing/objects: Added.
        * compositing/objects/composited-object-alignment.html: Added.
        * compositing/objects/composited-object-alignment-expected.checksum: Added.
        * compositing/objects/composited-object-alignment-expected.png: Added.
        * compositing/objects/composited-object-alignment-expected.txt: Added.
        * compositing/objects/resources: Added.
        * compositing/objects/resources/composited-subframe.html: Added.
2011-04-11  Chris Marrin  <cmarrin@apple.com>

        Reviewed by Simon Fraser.

        iframe/compositing propagation should also work for <object> and framesets
        https://bugs.webkit.org/show_bug.cgi?id=39037

        Generalized the code that handles the parenting of composited <iframe> elements
        to include <frame> and <object> elements. This includes forcing the parent
        element to be composited and constructing the layer tree in the child document
        and properly parenting it. Much of the change is giving the function calls more
        generic names (from ...IFrame... to ...Frame...)

        Compositing tests that were done in RenderIFrame have been moved to RenderPart
        since this is the highest base class common to all 3 elements.

        I also renamed several ...IFrame... functions to
        ...Frame... and changed them so they work on a RenderPart for determining
        if the renderer is composited. One other detail is that <iframe> and
        <object> never share style since they might be composited. I had to add
        <frame> to that list.

        Tests: compositing/framesets/composited-frame-alignment.html
               compositing/objects/composited-object-alignment.html

        * css/CSSStyleSelector.cpp: Add <frame> to list of element types that should never share style since it might be composited.
        (WebCore::CSSStyleSelector::canShareStyleWithElement):
        * page/FrameView.cpp:
        (WebCore::FrameView::hasCompositedContentIncludingDescendants):
        (WebCore::FrameView::setIsOverlapped):
        * rendering/RenderIFrame.cpp: Moved methods to RenderPart
        * rendering/RenderIFrame.h:
        * rendering/RenderLayerBacking.cpp:
        (WebCore::RenderLayerBacking::updateAfterWidgetResize):
        (WebCore::RenderLayerBacking::updateGraphicsLayerConfiguration):
        (WebCore::RenderLayerBacking::paintingGoesToWindow):
        * rendering/RenderLayerCompositor.cpp:
        (WebCore::RenderLayerCompositor::updateBacking):
        (WebCore::RenderLayerCompositor::rebuildCompositingLayerTree):
        (WebCore::RenderLayerCompositor::frameContentsCompositor):
        (WebCore::RenderLayerCompositor::parentFrameContentLayers):
        (WebCore::RenderLayerCompositor::didMoveOnscreen):
        (WebCore::RenderLayerCompositor::allowsIndependentlyCompositedFrames):
        (WebCore::RenderLayerCompositor::shouldPropagateCompositingToEnclosingFrame):
        (WebCore::RenderLayerCompositor::enclosingFrameElement):
        (WebCore::RenderLayerCompositor::requiresCompositingLayer):
        (WebCore::RenderLayerCompositor::requiresCompositingForFrame):
        (WebCore::RenderLayerCompositor::requiresScrollLayer):
        (WebCore::RenderLayerCompositor::ensureRootPlatformLayer):
        (WebCore::RenderLayerCompositor::attachRootPlatformLayer):
        (WebCore::RenderLayerCompositor::detachRootPlatformLayer):
        * rendering/RenderLayerCompositor.h:
        * rendering/RenderPart.cpp: Functions moved from RenderIFrame
        (WebCore::RenderPart::requiresLayer):
        (WebCore::RenderPart::requiresAcceleratedCompositing):
        * rendering/RenderPart.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@83518 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Bug 58263 - Use EncodedValueDescriptor on both JSVALUE32_64, JSVALUE64

Reviewed by Sam Weinig.

The JSJITInterface already uses EncodedValueDescriptor to access the tag/payload
separately on JSVALUE64, even though EncodedValueDescriptor is not used in
JSVALUE64's implementation of JSValue. Remove the separate definition for m_ptr
on X86_64. Using the union allows us to remove a layer of makeImmediate()/
immedaiteValue() methods.

* dfg/DFGNonSpeculativeJIT.cpp:
(JSC::DFG::NonSpeculativeJIT::compile):
* dfg/DFGSpeculativeJIT.cpp:
(JSC::DFG::SpeculativeJIT::compile):
* jit/JITInlineMethods.h:
(JSC::JIT::emitTagAsBoolImmediate):
* jit/JITOpcodes.cpp:
(JSC::JIT::emit_op_not):
(JSC::JIT::emit_op_jeq_null):
(JSC::JIT::emit_op_jneq_null):
(JSC::JIT::emit_op_get_pnames):
(JSC::JIT::emit_op_eq_null):
(JSC::JIT::emit_op_neq_null):
(JSC::JIT::emitSlow_op_not):
* runtime/JSCell.h:
* runtime/JSValue.h:
* runtime/JSValueInlineMethods.h:
(JSC::JSValue::encode):
(JSC::JSValue::decode):
(JSC::JSValue::operator==):
(JSC::JSValue::operator!=):
(JSC::JSValue::JSValue):
(JSC::JSValue::operator bool):
(JSC::JSValue::asInt32):
(JSC::JSValue::isUndefinedOrNull):
(JSC::JSValue::isBoolean):
(JSC::JSValue::isCell):
(JSC::JSValue::isInt32):
(JSC::JSValue::asDouble):
(JSC::JSValue::isNumber):
(JSC::JSValue::asCell):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@83517 268f45cc-cd09-0410-ab3c-d52691b4dbfc

    Reviewed by Dan Bernstein.

        REGRESSION (r83081): Esc key no longer removes current Kotoeri text operation
        https://bugs.webkit.org/show_bug.cgi?id=58274
        <rdar://problem/9263683>

        * WebProcess/WebPage/mac/WebPageMac.mm: (WebKit::WebPage::getMarkedRange): Wrap the returned
        temporary in RefPtr.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@83516 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2011-04-11  Yael Aharon  <yael.aharon@nokia.com>

        Reviewed by Eric Seidel.

        InlineBox::prevOnline and InlineBox::prevOnlineExists() are confusing and should be renamed
        https://bugs.webkit.org/show_bug.cgi?id=57178

        Remove prevOnLineExists() and nextOnLineExists() as they are not used anymore.
        No new tests needed because this is cleanup only.

        * rendering/InlineBox.cpp:
        * rendering/InlineBox.h:
        (WebCore::InlineBox::InlineBox):
        (WebCore::InlineBox::setPrevOnLine):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@83515 268f45cc-cd09-0410-ab3c-d52691b4dbfc

https://bugs.webkit.org/show_bug.cgi?id=58261

Reviewed by Dan Bernstein.

Fix for layout regression on marketwatch.com. When I converted TrailingFloatsRootInlineBox over
to the new overflow model, I messed up the math and passed a bottom value where I should have
passed a height value. This patch fixes the code to be logically equivalent to the way it
was prior to my changes.

Source/WebCore:

Added fast/overflow/trailing-float-linebox.html.

* rendering/RenderBlockLineLayout.cpp:
(WebCore::RenderBlock::layoutInlineChildren):

LayoutTests:

* fast/overflow/trailing-float-linebox.html: Added.
* platform/mac/fast/overflow/trailing-float-linebox-expected.checksum: Added.
* platform/mac/fast/overflow/trailing-float-linebox-expected.png: Added.
* platform/mac/fast/overflow/trailing-float-linebox-expected.txt: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@83514 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2011-04-11  Anders Carlsson  <andersca@apple.com>

        Reviewed by Sam Weinig.

        Remove m_urlAtProcessExit from WebPageProxy
        https://bugs.webkit.org/show_bug.cgi?id=58275

        Get rid of m_urlAtProcessExit and replace some zeros with nullptrs.

        * UIProcess/WebPageProxy.cpp:
        (WebKit::WebPageProxy::processDidCrash):
        * UIProcess/WebPageProxy.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@83513 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2011-04-11  Alexis Menard  <alexis.menard@openbossa.org>

        Reviewed by Benjamin Poulain.

        [Qt] Private Q_SLOTS void orientationChanged() can't be in qwebframe.h public header file.
        https://bugs.webkit.org/show_bug.cgi?id=58251

        Qt coding conventions states that private slots should be located in private implementation of
        the class. This allows us to rename/delete the slots in the future without breaking anything.

        No new tests added, just a simple refactoring.

        * Api/qwebframe.cpp:
        (QWebFramePrivate::_q_orientationChanged):
        * Api/qwebframe.h:
        * Api/qwebframe_p.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@83512 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2011-04-11  Adrienne Walker  <enne@google.com>

        Unreviewed, rolling out r83500.
        http://trac.webkit.org/changeset/83500
        https://bugs.webkit.org/show_bug.cgi?id=57113

        Regresses huge-layer-rotated test

        * platform/chromium-gpu-linux/compositing/color-matching/image-color-matching-expected.checksum:
        * platform/chromium-gpu-linux/compositing/color-matching/image-color-matching-expected.png:
        * platform/chromium-gpu-linux/compositing/geometry/horizontal-scroll-composited-expected.checksum:
        * platform/chromium-gpu-linux/compositing/geometry/horizontal-scroll-composited-expected.png:
        * platform/chromium-gpu-linux/compositing/geometry/vertical-scroll-composited-expected.checksum:
        * platform/chromium-gpu-linux/compositing/geometry/vertical-scroll-composited-expected.png:
        * platform/chromium-gpu-linux/compositing/text-on-large-layer-expected.checksum:
        * platform/chromium-gpu-linux/compositing/text-on-large-layer-expected.png:
        * platform/chromium-gpu-mac/compositing/color-matching/image-color-matching-expected.checksum:
        * platform/chromium-gpu-mac/compositing/color-matching/image-color-matching-expected.png:
        * platform/chromium-gpu-mac/compositing/geometry/horizontal-scroll-composited-expected.checksum:
        * platform/chromium-gpu-mac/compositing/geometry/horizontal-scroll-composited-expected.png:
        * platform/chromium-gpu-mac/compositing/geometry/vertical-scroll-composited-expected.checksum:
        * platform/chromium-gpu-mac/compositing/geometry/vertical-scroll-composited-expected.png:
        * platform/chromium-gpu-mac/compositing/overflow/overflow-scroll-expected.png:
        * platform/chromium-gpu-mac/compositing/text-on-large-layer-expected.checksum:
        * platform/chromium-gpu-mac/compositing/text-on-large-layer-expected.png:
        * platform/chromium-gpu-win/compositing/color-matching/image-color-matching-expected.checksum:
        * platform/chromium-gpu-win/compositing/color-matching/image-color-matching-expected.png:
        * platform/chromium-gpu-win/compositing/geometry/horizontal-scroll-composited-expected.checksum:
        * platform/chromium-gpu-win/compositing/geometry/horizontal-scroll-composited-expected.png:
        * platform/chromium-gpu-win/compositing/geometry/vertical-scroll-composited-expected.checksum:
        * platform/chromium-gpu-win/compositing/geometry/vertical-scroll-composited-expected.png:
        * platform/chromium-gpu-win/compositing/overflow/overflow-scroll-expected.checksum:
        * platform/chromium-gpu-win/compositing/overflow/overflow-scroll-expected.png:
        * platform/chromium-gpu-win/compositing/text-on-large-layer-expected.checksum:
        * platform/chromium-gpu-win/compositing/text-on-large-layer-expected.png:
        * platform/chromium-gpu/compositing/tiling/huge-layer-img-expected.checksum: Removed.
        * platform/chromium-gpu/compositing/tiling/huge-layer-img-expected.png: Removed.
2011-04-11  Adrienne Walker  <enne@google.com>

        Unreviewed, rolling out r83500.
        http://trac.webkit.org/changeset/83500
        https://bugs.webkit.org/show_bug.cgi?id=57113

        Regresses huge-layer-rotated test

        * platform/graphics/chromium/ContentLayerChromium.cpp:
        (WebCore::ContentLayerChromium::ContentLayerChromium):
        (WebCore::ContentLayerChromium::~ContentLayerChromium):
        (WebCore::ContentLayerChromium::cleanupResources):
        (WebCore::ContentLayerChromium::requiresClippedUpdateRect):
        (WebCore::ContentLayerChromium::paintContentsIfDirty):
        (WebCore::ContentLayerChromium::resizeUploadBuffer):
        (WebCore::ContentLayerChromium::updateTextureIfNeeded):
        (WebCore::ContentLayerChromium::updateTexture):
        (WebCore::ContentLayerChromium::draw):
        (WebCore::ContentLayerChromium::updateCompositorResources):
        (WebCore::ContentLayerChromium::unreserveContentsTexture):
        (WebCore::ContentLayerChromium::bindContentsTexture):
        (WebCore::ContentLayerChromium::dumpLayerProperties):
        * platform/graphics/chromium/ContentLayerChromium.h:
        (WebCore::ContentLayerChromium::drawsContent):
        * platform/graphics/chromium/GraphicsLayerChromium.cpp:
        (WebCore::GraphicsLayerChromium::setMaskLayer):
        * platform/graphics/chromium/ImageLayerChromium.cpp:
        (WebCore::ImageLayerChromium::paintContentsIfDirty):
        (WebCore::ImageLayerChromium::updateTextureIfNeeded):
        * platform/graphics/chromium/ImageLayerChromium.h:
        * platform/graphics/chromium/LayerChromium.h:
        (WebCore::LayerChromium::draw):
        * platform/graphics/chromium/LayerRendererChromium.cpp:
        (WebCore::LayerRendererChromium::drawRootLayer):
        (WebCore::LayerRendererChromium::drawLayers):
        (WebCore::LayerRendererChromium::paintContentsRecursive):
        (WebCore::LayerRendererChromium::copyOffscreenTextureToDisplay):
        (WebCore::LayerRendererChromium::drawLayer):
        (WebCore::LayerRendererChromium::initializeSharedObjects):
        (WebCore::LayerRendererChromium::cleanupSharedObjects):
        * platform/graphics/chromium/LayerRendererChromium.h:
        (WebCore::LayerRendererChromium::contentLayerProgram):
        * platform/graphics/chromium/LayerTilerChromium.cpp:
        (WebCore::LayerTilerChromium::invalidateRect):
        (WebCore::LayerTilerChromium::update):
        (WebCore::LayerTilerChromium::updateFromPixels):
        (WebCore::LayerTilerChromium::draw):
        * platform/graphics/chromium/LayerTilerChromium.h:
        * platform/graphics/chromium/PlatformCanvas.cpp:
        (WebCore::PlatformCanvas::AutoLocker::AutoLocker):
        (WebCore::PlatformCanvas::Painter::Painter):
        * platform/graphics/chromium/PlatformCanvas.h:
        * platform/graphics/chromium/RenderSurfaceChromium.cpp:
        (WebCore::RenderSurfaceChromium::draw):
        * platform/graphics/chromium/RenderSurfaceChromium.h:
        * platform/graphics/chromium/cc/CCCanvasLayerImpl.cpp:
        (WebCore::CCCanvasLayerImpl::draw):
        * platform/graphics/chromium/cc/CCCanvasLayerImpl.h:
        * platform/graphics/chromium/cc/CCHeadsUpDisplay.cpp:
        (WebCore::CCHeadsUpDisplay::draw):
        * platform/graphics/chromium/cc/CCLayerImpl.cpp:
        (WebCore::CCLayerImpl::draw):
        * platform/graphics/chromium/cc/CCLayerImpl.h:
        * platform/graphics/chromium/cc/CCPluginLayerImpl.cpp:
        (WebCore::CCPluginLayerImpl::draw):
        * platform/graphics/chromium/cc/CCPluginLayerImpl.h:
        * platform/graphics/chromium/cc/CCVideoLayerImpl.cpp:
        (WebCore::CCVideoLayerImpl::draw):
        * platform/graphics/chromium/cc/CCVideoLayerImpl.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@83511 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2011-04-11  Geoffrey Garen  <ggaren@apple.com>

        Try to fix a few builds.

        Updated a few more build configurations for file moves.

        * CMakeListsEfl.txt:
        * wscript:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@83510 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2011-04-11  Anders Carlsson  <andersca@apple.com>

        Reviewed by Adam Roben.

        Remove the WebContext member variable from WebPageProxy
        https://bugs.webkit.org/show_bug.cgi?id=58271

        * UIProcess/API/C/WKPage.cpp:
        (WKPageGetContext):
        * UIProcess/WebPageProxy.cpp:
        (WebKit::WebPageProxy::create):
        (WebKit::WebPageProxy::WebPageProxy):
        (WebKit::WebPageProxy::process):
        (WebKit::WebPageProxy::reattachToWebProcess):
        (WebKit::WebPageProxy::canShowMIMEType):
        (WebKit::WebPageProxy::receivedPolicyDecision):
        (WebKit::WebPageProxy::didStartProvisionalLoadForFrame):
        (WebKit::WebPageProxy::didReceiveServerRedirectForProvisionalLoadForFrame):
        (WebKit::WebPageProxy::didFailProvisionalLoadForFrame):
        (WebKit::WebPageProxy::didCommitLoadForFrame):
        (WebKit::WebPageProxy::didFinishDocumentLoadForFrame):
        (WebKit::WebPageProxy::didFinishLoadForFrame):
        (WebKit::WebPageProxy::didFailLoadForFrame):
        (WebKit::WebPageProxy::didSameDocumentNavigationForFrame):
        (WebKit::WebPageProxy::didReceiveTitleForFrame):
        (WebKit::WebPageProxy::didFirstLayoutForFrame):
        (WebKit::WebPageProxy::didFirstVisuallyNonEmptyLayoutForFrame):
        (WebKit::WebPageProxy::didRemoveFrameFromHierarchy):
        (WebKit::WebPageProxy::didDisplayInsecureContentForFrame):
        (WebKit::WebPageProxy::didRunInsecureContentForFrame):
        (WebKit::WebPageProxy::decidePolicyForNavigationAction):
        (WebKit::WebPageProxy::decidePolicyForNewWindowAction):
        (WebKit::WebPageProxy::decidePolicyForResponse):
        (WebKit::WebPageProxy::unableToImplementPolicy):
        (WebKit::WebPageProxy::willSubmitForm):
        (WebKit::WebPageProxy::mouseDidMoveOverElement):
        (WebKit::WebPageProxy::showContextMenu):
        (WebKit::WebPageProxy::contextMenuItemSelected):
        * UIProcess/WebPageProxy.h:
        * UIProcess/WebProcessProxy.cpp:
        (WebKit::WebProcessProxy::createWebPage):
        * UIProcess/mac/WebInspectorProxyMac.mm:
        (WebKit::WebInspectorProxy::platformCreateInspectorPage):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@83509 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2011-04-11  Eric Carlson  <eric.carlson@apple.com>

        Reviewed by Adam Roben.

        Ignore context change callbacks when not on the main thread
        https://bugs.webkit.org/show_bug.cgi?id=58256
        <rdar://problem/9266090>

        * WebProcess/WebPage/ca/win/LayerTreeHostCAWin.cpp:
        (WebKit::LayerTreeHostCAWin::contextDidChangeCallback): Do nothing when not called
            on the main thread. This should only happen when no changes have actually
            been committed to the context, eg. when a video frame has been added to an image
            queue, so return without triggering animations etc.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@83508 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2011-04-11  Geoffrey Garen  <ggaren@apple.com>

        Build fix: Updated a file name.

        * CMakeLists.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@83507 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2011-04-11  Geoffrey Garen  <ggaren@apple.com>

        Rubber-stamped by Sam Weinig.

        Moved remaining heap implementation files to the heap folder.

        * Android.mk:
        * CMakeLists.txt:
        * GNUmakefile.list.am:
        * JavaScriptCore.gypi:
        * JavaScriptCore.pro:
        * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.vcproj:
        * JavaScriptCore.xcodeproj/project.pbxproj:
        * heap/ConservativeRoots.cpp: Copied from runtime/ConservativeSet.cpp.
        * heap/ConservativeRoots.h: Copied from runtime/ConservativeSet.h.
        * heap/Handle.h:
        * heap/Heap.cpp:
        * heap/MachineStackMarker.cpp: Copied from runtime/MachineStackMarker.cpp.
        * heap/MachineStackMarker.h: Copied from runtime/MachineStackMarker.h.
        * heap/MarkStack.cpp: Copied from runtime/MarkStack.cpp.
        * heap/MarkStack.h: Copied from runtime/MarkStack.h.
        * heap/MarkStackPosix.cpp: Copied from runtime/MarkStackPosix.cpp.
        * heap/MarkStackSymbian.cpp: Copied from runtime/MarkStackSymbian.cpp.
        * heap/MarkStackWin.cpp: Copied from runtime/MarkStackWin.cpp.
        * heap/MarkedBlock.cpp: Copied from runtime/MarkedBlock.cpp.
        * heap/MarkedBlock.h: Copied from runtime/MarkedBlock.h.
        * heap/MarkedSpace.cpp: Copied from runtime/MarkedSpace.cpp.
        * heap/MarkedSpace.h: Copied from runtime/MarkedSpace.h.
        * interpreter/RegisterFile.cpp:
        * runtime/ConservativeSet.cpp: Removed.
        * runtime/ConservativeSet.h: Removed.
        * runtime/MachineStackMarker.cpp: Removed.
        * runtime/MachineStackMarker.h: Removed.
        * runtime/MarkStack.cpp: Removed.
        * runtime/MarkStack.h: Removed.
        * runtime/MarkStackPosix.cpp: Removed.
        * runtime/MarkStackSymbian.cpp: Removed.
        * runtime/MarkStackWin.cpp: Removed.
        * runtime/MarkedBlock.cpp: Removed.
        * runtime/MarkedBlock.h: Removed.
        * runtime/MarkedSpace.cpp: Removed.
        * runtime/MarkedSpace.h: Removed.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@83506 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2011-04-11  Stephen White  <senorblanco@chromium.org>

        Unreviewed; new chromium baselines.

        New chromium-mac baseline images for tests affected by r83490.

        * platform/chromium-mac/fast/lists/003-vertical-expected.checksum: Removed.
        * platform/chromium-mac/fast/lists/003-vertical-expected.png:
        * platform/chromium-mac/fast/text/decorations-transformed-expected.png:
        * platform/chromium-mac/svg/custom/foreign-object-skew-expected.png:
        * platform/chromium-mac/svg/custom/scrolling-embedded-svg-file-image-repaint-problem-expected.checksum: Removed.
        * platform/chromium-mac/svg/custom/scrolling-embedded-svg-file-image-repaint-problem-expected.png:
        * platform/chromium-mac/svg/zoom/page/zoom-foreignObject-expected.checksum: Removed.
        * platform/chromium-mac/svg/zoom/page/zoom-foreignObject-expected.png:
        * platform/chromium-mac/svg/zoom/text/zoom-foreignObject-expected.png:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@83505 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Unreviewed.

Adding my IRC nick.

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

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@83504 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2011-04-11  Varun Jain  <varunjain@chromium.org>

        Reviewed by Darin Fisher.

        Resolving FIXME to make new WebWidget method pure virtual since all downstream classes have implemented the method.
        https://bugs.webkit.org/show_bug.cgi?id=58245

        *  Source/WebKit/chromium/public/WebWidget.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@83503 268f45cc-cd09-0410-ab3c-d52691b4dbfc

<rdar://problem/9267042> Assertion failure on launch in
LayerChangesFlusher::setHook().

https://bugs.webkit.org/show_bug.cgi?id=58270

Reviewed by Adam Roben.

* platform/graphics/ca/win/LayerChangesFlusher.cpp:
(WebCore::LayerChangesFlusher::setHook):
MSDN says that the HINSTANCE passed in must be null if the thread ID
passed is a thread created by this process and the hook procedure is
within this process.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@83502 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2011-04-11  MORITA Hajime  <morrita@google.com>

        Unreviewed build fix.

        Changeset 83492 broke Leopard build.
        https://bugs.webkit.org/show_bug.cgi?id=58268

        * editing/Editor.cpp:
        (WebCore::Editor::markMisspellingsAfterTypingToWord):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@83501 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2011-03-25  Adrienne Walker  <enne@google.com>

        Reviewed by James Robinson.

        [chromium] Tile content and image layers
        https://bugs.webkit.org/show_bug.cgi?id=57113

        Rebaseline a number of image tests where slight filtering differences
        from using tiles caused test failures with exact pixel matching.
        None of these results are perceptually different.

        * platform/chromium-gpu-linux/compositing/color-matching/image-color-matching-expected.checksum:
        * platform/chromium-gpu-linux/compositing/color-matching/image-color-matching-expected.png:
        * platform/chromium-gpu-linux/compositing/geometry/horizontal-scroll-composited-expected.checksum:
        * platform/chromium-gpu-linux/compositing/geometry/horizontal-scroll-composited-expected.png:
        * platform/chromium-gpu-linux/compositing/geometry/vertical-scroll-composited-expected.checksum:
        * platform/chromium-gpu-linux/compositing/geometry/vertical-scroll-composited-expected.png:
        * platform/chromium-gpu-linux/compositing/text-on-large-layer-expected.checksum:
        * platform/chromium-gpu-linux/compositing/text-on-large-layer-expected.png:
        * platform/chromium-gpu-mac/compositing/color-matching/image-color-matching-expected.checksum:
        * platform/chromium-gpu-mac/compositing/color-matching/image-color-matching-expected.png:
        * platform/chromium-gpu-mac/compositing/geometry/horizontal-scroll-composited-expected.checksum:
        * platform/chromium-gpu-mac/compositing/geometry/horizontal-scroll-composited-expected.png:
        * platform/chromium-gpu-mac/compositing/geometry/vertical-scroll-composited-expected.checksum:
        * platform/chromium-gpu-mac/compositing/geometry/vertical-scroll-composited-expected.png:
        * platform/chromium-gpu-mac/compositing/overflow/overflow-scroll-expected.png:
        * platform/chromium-gpu-mac/compositing/text-on-large-layer-expected.checksum:
        * platform/chromium-gpu-mac/compositing/text-on-large-layer-expected.png:
        * platform/chromium-gpu-win/compositing/color-matching/image-color-matching-expected.checksum:
        * platform/chromium-gpu-win/compositing/color-matching/image-color-matching-expected.png:
        * platform/chromium-gpu-win/compositing/geometry/horizontal-scroll-composited-expected.checksum:
        * platform/chromium-gpu-win/compositing/geometry/horizontal-scroll-composited-expected.png:
        * platform/chromium-gpu-win/compositing/geometry/vertical-scroll-composited-expected.checksum:
        * platform/chromium-gpu-win/compositing/geometry/vertical-scroll-composited-expected.png:
        * platform/chromium-gpu-win/compositing/overflow/overflow-scroll-expected.checksum:
        * platform/chromium-gpu-win/compositing/overflow/overflow-scroll-expected.png:
        * platform/chromium-gpu-win/compositing/text-on-large-layer-expected.checksum:
        * platform/chromium-gpu-win/compositing/text-on-large-layer-expected.png:
        * platform/chromium-gpu/compositing/tiling/huge-layer-img-expected.checksum: Added.
        * platform/chromium-gpu/compositing/tiling/huge-layer-img-expected.png: Added.
2011-03-25  Adrienne Walker  <enne@google.com>

        Reviewed by James Robinson.

        [chromium] Tile content and image layers
        https://bugs.webkit.org/show_bug.cgi?id=57113

        Layers tile by default if any dimension is larger than 512.  Smaller
        layers are contained within a single texture but still use the tiler
        infrastructure so that there's only one code path.

        Remove large layer support from content layers.  Content layers no
        longer own a platform canvas--they own a tiler.  Refactor tiler to
        allow for better separation of update/upload/paint.  Add rect
        parameter to update and draw functions on layers for the layer-space
        rect of interest.  This is necessary to know which tiles need to be
        drawn.

        * platform/graphics/chromium/ContentLayerChromium.cpp:
        (WebCore::ContentLayerChromium::ContentLayerChromium):
        (WebCore::ContentLayerChromium::~ContentLayerChromium):
        (WebCore::ContentLayerPainter::ContentLayerPainter):
        (WebCore::ContentLayerPainter::paint):
        (WebCore::ContentLayerChromium::paintContentsIfDirty):
        (WebCore::ContentLayerChromium::setLayerRenderer):
        (WebCore::ContentLayerChromium::tilingTransform):
        (WebCore::ContentLayerChromium::visibleLayerRect):
        (WebCore::ContentLayerChromium::layerBounds):
        (WebCore::ContentLayerChromium::updateLayerSize):
        (WebCore::ContentLayerChromium::draw):
        (WebCore::ContentLayerChromium::createTilerIfNeeded):
        (WebCore::ContentLayerChromium::updateCompositorResources):
        (WebCore::ContentLayerChromium::setTilingOption):
        (WebCore::ContentLayerChromium::bindContentsTexture):
        (WebCore::ContentLayerChromium::unreserveContentsTexture):
        (WebCore::ContentLayerChromium::setIsMask):
        (WebCore::ContentLayerChromium::dumpLayerProperties):
        * platform/graphics/chromium/ContentLayerChromium.h:
        (WebCore::ContentLayerChromium::drawsContent):
        * platform/graphics/chromium/GraphicsLayerChromium.cpp:
        (WebCore::GraphicsLayerChromium::setMaskLayer):
        * platform/graphics/chromium/ImageLayerChromium.cpp:
        (WebCore::ImageLayerChromium::paintContentsIfDirty):
        (WebCore::ImageLayerChromium::updateCompositorResources):
        (WebCore::ImageLayerChromium::layerBounds):
        (WebCore::ImageLayerChromium::tilingTransform):
        * platform/graphics/chromium/ImageLayerChromium.h:
        * platform/graphics/chromium/LayerChromium.h:
        (WebCore::LayerChromium::invalidateRect):
        (WebCore::LayerChromium::paintContentsIfDirty):
        (WebCore::LayerChromium::setIsMask):
        (WebCore::LayerChromium::draw):
        * platform/graphics/chromium/LayerRendererChromium.cpp:
        (WebCore::LayerRendererChromium::drawRootLayer):
        (WebCore::LayerRendererChromium::drawLayers):
        (WebCore::LayerRendererChromium::paintContentsRecursive):
        (WebCore::LayerRendererChromium::copyOffscreenTextureToDisplay):
        (WebCore::LayerRendererChromium::drawLayer):
        (WebCore::LayerRendererChromium::initializeSharedObjects):
        (WebCore::LayerRendererChromium::cleanupSharedObjects):
        * platform/graphics/chromium/LayerRendererChromium.h:
        * platform/graphics/chromium/LayerTilerChromium.cpp:
        (WebCore::LayerTilerChromium::getSingleTexture):
        (WebCore::LayerTilerChromium::invalidateRect):
        (WebCore::LayerTilerChromium::update):
        (WebCore::LayerTilerChromium::uploadCanvas):
        (WebCore::LayerTilerChromium::updateFromPixels):
        (WebCore::LayerTilerChromium::draw):
        (WebCore::LayerTilerChromium::unreserveTextures):
        * platform/graphics/chromium/LayerTilerChromium.h:
        (WebCore::LayerTilerChromium::setLayerRenderer):
        (WebCore::LayerTilerChromium::skipsDraw):
        * platform/graphics/chromium/PlatformCanvas.cpp:
        (WebCore::PlatformCanvas::AutoLocker::AutoLocker):
        (WebCore::PlatformCanvas::Painter::Painter):
        * platform/graphics/chromium/PlatformCanvas.h:
        * platform/graphics/chromium/RenderSurfaceChromium.cpp:
        (WebCore::RenderSurfaceChromium::draw):
        * platform/graphics/chromium/RenderSurfaceChromium.h:
        * platform/graphics/chromium/cc/CCCanvasLayerImpl.cpp:
        (WebCore::CCCanvasLayerImpl::draw):
        * platform/graphics/chromium/cc/CCCanvasLayerImpl.h:
        * platform/graphics/chromium/cc/CCHeadsUpDisplay.cpp:
        (WebCore::CCHeadsUpDisplay::draw):
        * platform/graphics/chromium/cc/CCLayerImpl.cpp:
        (WebCore::CCLayerImpl::draw):
        * platform/graphics/chromium/cc/CCLayerImpl.h:
        * platform/graphics/chromium/cc/CCPluginLayerImpl.cpp:
        (WebCore::CCPluginLayerImpl::draw):
        * platform/graphics/chromium/cc/CCPluginLayerImpl.h:
        * platform/graphics/chromium/cc/CCVideoLayerImpl.cpp:
        (WebCore::CCVideoLayerImpl::draw):
        * platform/graphics/chromium/cc/CCVideoLayerImpl.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@83500 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2011-04-11  Anders Carlsson  <andersca@apple.com>

        Reviewed by Sam Weinig.

        First step towards simplifying WebPageProxy/WebProcessProxy/WebContext ownership
        https://bugs.webkit.org/show_bug.cgi?id=58266

        With this patch, the WKView holds a strong reference to a WebPageProxy. The
        WebPageProxy in turn holds a strong reference to its WebProcessProxy. Finally,
        The WebProcessProxy holds a strong reference to its WebContext.

        The WebContext holds a strong reference to the running WebProcessProxy which results
        in a reference cycle that's broken when the web process exits.

        The reason for is to avoid crashes where WebPageProxy::process() returns null if the web process
        has crashed but has not yet been relaunched.

        * UIProcess/WebContext.cpp:
        (WebKit::WebContext::disconnectProcess):
        Add comment.

        (WebKit::WebContext::createWebPage):
        Return a PassRefPtr.

        (WebKit::WebContext::relaunchProcessIfNecessary):
        Change this to return a WebPageProxy.

        * UIProcess/WebPageProxy.cpp:
        (WebKit::WebPageProxy::create):
        This now takes a PassRefPtr<WebProcessProxy>.

        (WebKit::WebPageProxy::WebPageProxy):
        Ditto.

        (WebKit::WebPageProxy::~WebPageProxy):
        Call close() if necessary.

        (WebKit::WebPageProxy::reattachToWebProcess):
        Replace the current process with the new process.

        * UIProcess/WebProcessProxy.cpp:
        (WebKit::WebProcessProxy::create):
        Take a PassRefPtr<WebContext>.

        (WebKit::WebProcessProxy::WebProcessProxy):
        Ditto.

        (WebKit::WebProcessProxy::webPage):
        Remove .get() now that the page map uses weak references.

        (WebKit::WebProcessProxy::createWebPage):
        This now returns the created web page proxy.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@83499 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Dispatch sent messages to windows owned by the web process when waiting a sync CoreIPC reply

On Windows, windowed plugins' HWNDs are created as children of the WKView's window. This
creates a cross-process window hierarchy, which in turn attaches the input states of the UI
process's and web process's main threads (as if ::AttachThreadInput has been called). Having
the input states attached means that changes to the input state (e.g., changing the focus
window) can result in synchronous window messages being sent between the processes. This can
result in deadlocks if the UI process changes the input state while handling a synchronous
CoreIPC message from the web process. Since the web process isn't running its message loop
while waiting for the reply, it never processes the messages Windows is sending it from the
UI process.

The solution taken in this patch is to continue to dispatch sent (not posted) messages to
windows created by the web process while waiting for a sync CoreIPC reply. Someday we can
hopefully reduce the number of cases in which the UI process modifies the thread's input
state while handling a synchronous message; see the bug for details.

Fixes <http://webkit.org/b/58239> <rdar://problem/8769302> REGRESSION (WebKit2): Deadlock
clicking Flash plugin

Reviewed by Anders Carlsson.

* Platform/CoreIPC/BinarySemaphore.h:
(CoreIPC::BinarySemaphore::event): Added. Simple getter to expose the underlying event
HANDLE.

* Platform/CoreIPC/Connection.cpp:
(CoreIPC::Connection::SyncMessageState::waitWhileDispatchingSentMessages): New Windows-only
function that is used instead of wait() so that sent messages will continue to be
dispatched.
(CoreIPC::Connection::waitForSyncReply): Use waitWhileDispatchingSentMessages instead of
wait on Windows. Our Client gives us the set of windows that need to have windows delivered
to them.

* Platform/CoreIPC/Connection.h: Added new
windowsToReceiveSentMessagesWhileWaitingForSyncReply function to Client.

* Platform/RunLoop.h: Added new dispatchSentMessagesUntil function to be used while waiting
for a sync CoreIPC reply.

* Platform/win/RunLoopWin.cpp:
(RunLoop::dispatchSentMessagesUntil): Added. If we have no windows to dispatch messages to,
then just wait on the semaphore. Otherwise spin a ::MsgWaitForMultipleObjectsEx loop to
detect when the semaphore is signaled, the timeout elapses, or sent messages are available,
and handle each case appropriately.

* UIProcess/WebProcessProxy.h: Added new CoreIPC::Connection::Client function.

* UIProcess/win/WebProcessProxyWin.cpp: Added.
(WebKit::WebProcessProxy::windowsToReceiveSentMessagesWhileWaitingForSyncReply): Just return
an empty Vector. The web process never modifies the thread's input state while responding to
a synchronous message, so we don't have anything to worry about here.

* WebProcess/WebProcess.h: Added new CoreIPC::Connection::Client function.

* WebProcess/win/WebProcessWin.cpp:
(WebKit::addWindowToVectorIfOwnedByCurrentThread): New helper function to be called by
::EnumThreadWindows/::EnumChildWindows. Does what it says.
(WebKit::WebProcess::windowsToReceiveSentMessagesWhileWaitingForSyncReply): Added. Returns
all top-level windows created by this thread, descendants of those windows created by this
thread, and descendants of WKViews' windows created by this thread.

* win/WebKit2.vcproj: Added WebProcessProxyWin.cpp.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@83498 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2011-04-11  Eric Carlson  <eric.carlson@apple.com>

        Reviewed by Simon Fraser.

        Ignore context change callbacks when not on the main thread
        https://bugs.webkit.org/show_bug.cgi?id=58256
        <rdar://problem/9266090>

        * platform/graphics/ca/win/WKCACFViewLayerTreeHost.cpp:
        (WebCore::WKCACFViewLayerTreeHost::contextDidChange): Do nothing when not called
            on the main thread. This should only happen when no changes have actually
            been committed to the context, eg. when a video frame has been added to an image
            queue, so return without triggering animations etc.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@83497 268f45cc-cd09-0410-ab3c-d52691b4dbfc

WebKit2: Implement TextChecker on Windows
https://bugs.webkit.org/show_bug.cgi?id=57862

Reviewed by Anders Carlsson.

Part 7: Implement getGuessesForWord, learnWord, and ignoreWord.

* UIProcess/API/C/win/WKTextChecker.cpp:
(WKTextCheckerChangeSpellingToWord):
* UIProcess/API/C/win/WKTextChecker.h:

* UIProcess/WebPageProxy.cpp:
(WebKit::WebPageProxy::changeSpellingToWord):
Make this function const since it does not modify the WebPageProxy and making it const
allows WebTextChecker::changeSpellingToWord to take a const WebPageProxy.
(WebKit::WebPageProxy::learnWord):
* UIProcess/WebPageProxy.h:

* UIProcess/TextChecker.h:
* UIProcess/win/TextCheckerWin.cpp:
(WebKit::TextChecker::getGuessesForWord):
Ask the TextCheckerClient for the guesses.
(WebKit::TextChecker::learnWord):
Tell the TextCheckerClient.
(WebKit::TextChecker::ignoreWord):
Ditto.
* UIProcess/gtk/TextCheckerGtk.cpp:
(WebKit::TextChecker::learnWord):
Add an unused param that is necessary for Windows.
* UIProcess/mac/TextCheckerMac.mm:
(WebKit::TextChecker::learnWord):
Ditto.
* UIProcess/qt/TextCheckerQt.cpp:
(WebKit::TextChecker::learnWord):
Ditto.

* UIProcess/win/WebTextChecker.cpp:
(WebKit::WebTextChecker::changeSpellingToWord):
Tell the page.
* UIProcess/win/WebTextChecker.h:

* UIProcess/win/WebTextCheckerClient.cpp:
(WebKit::WebTextCheckerClient::guessesForWord):
(WebKit::WebTextCheckerClient::learnWord):
(WebKit::WebTextCheckerClient::ignoreWord):
* UIProcess/win/WebTextCheckerClient.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@83496 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2011-04-11  Thouraya ANDOLSI  <thouraya.andolsi@st.com>

        Reviewed by Oliver Hunt.

        SH4 JIT SUPPORT.
        https://bugs.webkit.org/show_bug.cgi?id=44329

        Enable JIT support for SH4 platforms.

        * configure.ac:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@83495 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2011-04-11  Daniel Cheng  <dcheng@chromium.org>

        Reviewed by Dmitry Titov.

        [chromium] Implement image/png support in DataTransferItems
        https://bugs.webkit.org/show_bug.cgi?id=58106

        Add a reftest to verify that the various bits in the pipeline work as expected.

        * editing/pasteboard/data-transfer-items-image-png-expected.html: Added.
        * editing/pasteboard/data-transfer-items-image-png.html: Added.
2011-04-11  Daniel Cheng  <dcheng@chromium.org>

        Reviewed by Dmitry Titov.

        [chromium] Implement image/png support in DataTransferItems
        https://bugs.webkit.org/show_bug.cgi?id=58106

        This patch implements DataTransferItem::getAsFile() using Blobs. In
        addition, plumbing for image/png has been added so image paste will be
        supported as long as the renderer makes that data available.

        Tests: editing/pasteboard/data-transfer-items-image-png.html

        * dom/DataTransferItem.h:
        * dom/DataTransferItem.idl:
        * platform/chromium/ClipboardMimeTypes.cpp:
        * platform/chromium/ClipboardMimeTypes.h:
        * platform/chromium/DataTransferItemChromium.cpp:
        (WebCore::DataTransferItemChromium::getAsString):
        (WebCore::DataTransferItemChromium::getAsFile):
        * platform/chromium/DataTransferItemChromium.h:
        * platform/chromium/PlatformBridge.h:
2011-04-11  Daniel Cheng  <dcheng@chromium.org>

        Reviewed by Dmitry Titov.

        [chromium] Implement image/png support in DataTransferItems
        https://bugs.webkit.org/show_bug.cgi?id=58106

        Add glue for reading image data from clipboard.

        * DEPS:
        * src/PlatformBridge.cpp:
        (WebCore::PlatformBridge::clipboardReadImage):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@83494 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2011-04-11  Mario Sanchez Prada  <msanchez@igalia.com>

        Reviewed by Chris Fleizach.

        [Gtk] Implement support for Embedded Objects
        https://bugs.webkit.org/show_bug.cgi?id=52148

        Expose special OBJECT character for replaced elements, implementing
        AtkText and AtkHyperlink when required.

        * accessibility/AccessibilityRenderObject.cpp:
        (WebCore::textIteratorBehaviorForTextRange): New helper function,
        to return the right behavior, depending on the platform, so it
        ensures that object replacement characters get emitted for GTK.
        (WebCore::AccessibilityRenderObject::textUnderElement): Use the
        new helper function textIteratorBehaviorForTextRange.
        (WebCore::AccessibilityRenderObject::stringValue): Ditto.
        (WebCore::AccessibilityRenderObject::indexForVisiblePosition):
        Consider replaced elements when calculating range length in GTK.

        * accessibility/gtk/AccessibilityObjectWrapperAtk.cpp:
        (textForRenderer): Ouput the 'object replacement character' for
        replaced objects.
        (getSelectionOffsetsForObject): Consider replaced elements when
        calculating range length in GTK.
        (webkitAccessibleHypertextGetLink): Remove wrong extra check that
        were causing only links to be considered.
        (webkitAccessibleHypertextGetNLinks): Replace wrong 'isLink()'
        check with the right one, by checking that the right ATK interface
        is being implemented by the AtkObject.
        (getInterfaceMaskFromObject): Implement the Hyperlink interface
        both for links and replaced objects.
        (objectAndOffsetUnignored): Consider replaced elements when
        calculating range length in GTK.

        * accessibility/gtk/WebKitAccessibleHyperlink.cpp:
        (getRangeLengthForObject): Ensure spaces are used for replaced
        elements when calling to TextIterator::rangeLength().

        * editing/TextIterator.h: New value in the TextIteratorBehavior
        enumeration (TextIteratorEmitsObjectReplacementCharacters) and new
        private variable to consider that new option internally.
        * editing/TextIterator.cpp:
        (WebCore::TextIterator::TextIterator): Initialize the new private
        attribute m_emitsObjectReplacementCharacters in constructors.
        (WebCore::TextIterator::handleReplacedElement): Emit the 'object
        replacement character' when m_emitsObjectReplacementCharacters.
2011-04-11  Mario Sanchez Prada  <msanchez@igalia.com>

        Reviewed by Chris Fleizach.

        [Gtk] Implement support for Embedded Objects
        https://bugs.webkit.org/show_bug.cgi?id=52148

        New accessibility unit test for embedded objects.

        * tests/testatk.c:
        (testWebkitAtkEmbeddedObjects): New unit test.
        (main): Added the new unit test.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@83493 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2011-04-11  Jia Pu  <jpu@apple.com>

        Reviewed by Adele Peterson.

        Shouldn't carry out autocorrection when start typing a new word
        https://bugs.webkit.org/show_bug.cgi?id=58241
        <rdar://problem/9264736>

        See WebCore/ChangeLog for detail.

        * platform/mac-leopard/Skipped:
        * platform/mac-snowleopard/Skipped:
        * platform/mac-tiger/Skipped:
        * platform/mac/editing/spelling/autocorrection-at-beginning-of-word-1-expected.checksum: Added.
        * platform/mac/editing/spelling/autocorrection-at-beginning-of-word-1-expected.png: Added.
        * platform/mac/editing/spelling/autocorrection-at-beginning-of-word-1-expected.txt: Added.
        * platform/mac/editing/spelling/autocorrection-at-beginning-of-word-1.html: Added.
        * platform/mac/editing/spelling/autocorrection-at-beginning-of-word-2-expected.checksum: Added.
        * platform/mac/editing/spelling/autocorrection-at-beginning-of-word-2-expected.png: Added.
        * platform/mac/editing/spelling/autocorrection-at-beginning-of-word-2-expected.txt: Added.
        * platform/mac/editing/spelling/autocorrection-at-beginning-of-word-2.html: Added.
2011-04-11  Jia Pu  <jpu@apple.com>

        Reviewed by Adele Peterson.

        Shouldn't carry out autocorrection when start typing a new word
        https://bugs.webkit.org/show_bug.cgi?id=58241
        <rdar://problem/9264736>

        Tests: platform/mac/editing/spelling/autocorrection-at-beginning-of-word-1.html
               platform/mac/editing/spelling/autocorrection-at-beginning-of-word-2.html

        In TypingCommand::markMisspellingsAfterTyping(), we want to call Editor::markMisspellingsAfterTypingToWord()
        only when the typing command is an insertion command, and preceding word contains at least one non-whitespace
        character.

        * editing/Editor.cpp:
        (WebCore::Editor::markMisspellingsAfterTypingToWord):
        * editing/Editor.h:
        * editing/TypingCommand.cpp:
        (WebCore::TypingCommand::markMisspellingsAfterTyping):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@83492 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2011-04-11  Vsevolod Vlasov  <vsevik@chromium.org>

        Reviewed by Pavel Feldman.

        PerformanceTiming returns inconsistent values when timing is null.
        https://bugs.webkit.org/show_bug.cgi?id=58143

        Made PerformanceTiming returned values consistent when there is no timing object.

        * page/PerformanceTiming.cpp:
        (WebCore::PerformanceTiming::domainLookupStart):
        (WebCore::PerformanceTiming::domainLookupEnd):
        (WebCore::PerformanceTiming::connectStart):
        (WebCore::PerformanceTiming::connectEnd):
        (WebCore::PerformanceTiming::requestStart):
        (WebCore::PerformanceTiming::responseStart):
        (WebCore::PerformanceTiming::domLoading):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@83491 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2011-04-05  Stephen White  <senorblanco@chromium.org>

        Fix for jittering when animating a rotated image.
        https://bugs.webkit.org/show_bug.cgi?id=50775

        Make GraphicsContext::roundToDevicePixels a no-op on Chrome/Mac.

        * platform/graphics/cg/GraphicsContextCG.cpp:
        (WebCore::GraphicsContext::roundToDevicePixels):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@83490 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2011-04-11  Ryosuke Niwa  <rniwa@webkit.org>

        Reviewed by Antti Koivisto.

        Text inserted at the collapsed selection between two BRs following a non-wrapping text that forces overflow
        https://bugs.webkit.org/show_bug.cgi?id=47806

        Add a regression test since this bug does not reproduce anymore.
        The test is co-authored by James Robinson and Levi Weintraub.

        * fast/text/text-between-two-brs-in-nowrap-overflow-expected.txt: Added.
        * fast/text/text-between-two-brs-in-nowrap-overflow.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@83487 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2011-04-11  Anders Carlsson  <andersca@apple.com>

        Reviewed by Sam Weinig.

        Remove unused WebProcessProxy functions
        https://bugs.webkit.org/show_bug.cgi?id=58262

        * UIProcess/WebProcessProxy.cpp:
        * UIProcess/WebProcessProxy.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@83486 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2011-04-06  Xiaomei Ji  <xji@chromium.org>

        Reviewed by Ryosuke Niwa.

        Continue experiment with moving caret by word in visual order.
        https://bugs.webkit.org/show_bug.cgi?id=57806

        This is the 2nd patch, which adds implementation when caret is inside box
        (not at boundaries). If the word break is inside the same box and not at the boundaries
        either, the word break will be returned. If need to search the adjacent boxes for word
        breaks, then, only the cases implemented in bug 57336 work.

        * editing/visible_units.cpp:
        (WebCore::leftmostPositionInRTLBoxInLTRBlock):
        (WebCore::rightmostPositionInLTRBoxInRTLBlock):
        (WebCore::lastWordBreakInBox):
        (WebCore::positionIsVisuallyOrderedInBoxInBlockWithDifferentDirectionality):
        (WebCore::nextWordBreakInBoxInsideBlockWithDifferentDirectionality):
        (WebCore::WordBoundaryEntry::WordBoundaryEntry):
        (WebCore::collectWordBreaksInBoxInsideBlockWithSameDirectionality):
        (WebCore::collectWordBreaksInBoxInsideBlockWithDifferntDirectionality):
        (WebCore::greatestValueUnder):
        (WebCore::smallestOffsetAbove):
        (WebCore::positionIsInsideBox):
        (WebCore::positionBeforeNextWord):
        (WebCore::positionAfterPreviousWord):
        (WebCore::leftWordPosition):
        (WebCore::rightWordPosition):
2011-04-06  Xiaomei Ji  <xji@chromium.org>

        Reviewed by Ryosuke Niwa.

        Continue experiment with moving caret by word in visual order.
        https://bugs.webkit.org/show_bug.cgi?id=57806

        Return correct result when caret and the word break are inside a box (not at boundaries).
        Added test case to test word break at a random start point
        (not start from a word break).

        * editing/selection/move-by-word-visually-expected.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@83483 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Unreviewed, chromium expectations change.

Mark huge-column-count.html as missing in chromium.

* platform/chromium/test_expectations.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@83482 268f45cc-cd09-0410-ab3c-d52691b4dbfc

[Snow Leopard WebKit2 Release Tests] inspector/styles/styles-add-blank-property.html flakey.
https://bugs.webkit.org/show_bug.cgi?id=57638

Skip inspector/styles/styles-add-blank-property.html to get the bots green.

* platform/mac-wk2/Skipped:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@83481 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2011-04-11  Andreas Kling  <andreas.kling@nokia.com>

        Unreviewed Qt build fix, QT_CONFIG doesn't have uitools on bots.

        * Source/WebKit.pri:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@83480 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2011-04-11  Mario Sanchez Prada  <msanchez@igalia.com>

        Reviewed by Chris Fleizach.

        [GTK] Unskip accessibility/input-slider.html and accessibility/media-element.html
        https://bugs.webkit.org/show_bug.cgi?id=58040

        Unskip passing tests and update expectations.

        * platform/gtk/Skipped: Unskip passing tests.
        * platform/gtk/accessibility/input-slider-expected.txt: Added.
        * platform/gtk/accessibility/input-slider.html: Copied from
        accessibility/input-slider.html and adjusted accordingly.
        * platform/gtk/accessibility/media-element-expected.txt: Added.
        * platform/mac/accessibility/input-slider.html: Moved from
        accessibility/input-slider.html, since it is platform-specific.
        * platform/win/Skipped: Unskipped accessibility/input-slider.html,
        as it's no longer a cross-platform test.
2011-04-11  Mario Sanchez Prada  <msanchez@igalia.com>

        Reviewed by Chris Fleizach.

        [GTK] Unskip accessibility/input-slider.html and accessibility/media-element.html
        https://bugs.webkit.org/show_bug.cgi?id=58040

        Don't expose objects of role SliderThumbRoles in GTK.

        * accessibility/AccessibilitySlider.cpp:
        (WebCore::AccessibilitySlider::addChildren): Allow the platform
        make a final decision before including children in the hierarchy.
        (WebCore::AccessibilitySliderThumb::accessibilityIsIgnored):
        Implemented by relying on accessibilityPlatformIncludesObject().

        * accessibility/gtk/AccessibilityObjectAtk.cpp:
        (WebCore::AccessibilityObject::accessibilityPlatformIncludesObject):
        Ignore accessibility objects with role SliderThumbRole.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@83479 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2011-04-11  Dirk Pranke  <dpranke@chromium.org>

        Unreviewed, expectations change.
        https://bugs.webkit.org/show_bug.cgi?id=58193

        Suppress a few remaining Vista SVG failures until they can be
        triaged and/or rebaselined.

        * platform/chromium/test_expectations.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@83478 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2011-04-11  Dirk Pranke  <dpranke@chromium.org>

        Reviewed by Adam Roben.

        new-run-webkit-tests: fix crash when running under windows cmd.exe
        https://bugs.webkit.org/show_bug.cgi?id=58197

        We would intermittently crash when running NRWT directly under
        cmd.exe on windows because we were sharing the stdin file
        descriptor between the python process and the http server.
        cmd.exe really didn't like that, and there was no reason to
        share the descriptor, so we now use a PIPE instead.

        * Scripts/webkitpy/layout_tests/port/http_server.py:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@83477 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2011-04-11  Dirk Pranke  <dpranke@chromium.org>

        Reviewed by Ojan Vafai.

        new-run-webkit-tests: add unit tests for Port.diff_image()
        https://bugs.webkit.org/show_bug.cgi?id=58196

        Add the unit tests for the fix in bug 58195.

        * Scripts/webkitpy/layout_tests/port/chromium_unittest.py:
        * Scripts/webkitpy/layout_tests/port/port_testcase.py:
        * Scripts/webkitpy/layout_tests/port/webkit_unittest.py:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@83476 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2011-04-11  Dirk Pranke  <dpranke@chromium.org>

        Reviewed by Tony Chang.

        new-run-webkit-tests: implement support for audio, take two
        https://bugs.webkit.org/show_bug.cgi?id=58195

        Attempt to re-land the fix for bug 58101 (which was initially
        landed in r83330, but rolled out). This is the same patch but
        fixes crashes in Port.diff_image() caused by the change for
        empty image files being None instead of ''.

        * Scripts/webkitpy/layout_tests/layout_package/json_layout_results_generator.py:
        * Scripts/webkitpy/layout_tests/layout_package/single_test_runner.py:
        * Scripts/webkitpy/layout_tests/layout_package/test_expectations.py:
        * Scripts/webkitpy/layout_tests/layout_package/test_failures.py:
        * Scripts/webkitpy/layout_tests/layout_package/test_result_writer.py:
        * Scripts/webkitpy/layout_tests/port/base.py:
        * Scripts/webkitpy/layout_tests/port/chromium.py:
        * Scripts/webkitpy/layout_tests/port/dryrun.py:
        * Scripts/webkitpy/layout_tests/port/mock_drt.py:
        * Scripts/webkitpy/layout_tests/port/mock_drt_unittest.py:
        * Scripts/webkitpy/layout_tests/port/test.py:
        * Scripts/webkitpy/layout_tests/port/webkit.py:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@83475 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2011-04-11  Anders Carlsson  <andersca@apple.com>

        Reviewed by Sam Weinig.

        Crash in WebPageProxy::countStringMatches
        https://bugs.webkit.org/show_bug.cgi?id=58255
        <rdar://problem/9243837>

        * UIProcess/WebPageProxy.cpp:
        (WebKit::WebPageProxy::countStringMatches):
        Return early if the page is not valid.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@83474 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2011-04-11  Alexis Menard  <alexis.menard@openbossa.org>

        Reviewed by Kenneth Rohde Christiansen.

        [Qt] Sub-Frame content is not updated when scrolling in certain circumstances
        https://bugs.webkit.org/show_bug.cgi?id=50373

        Make sure that we invalidate the backing store when using TILED_BACKING_STORE
        and the page contains sub frames. This bug appears only when frame flattening is
        disabled and scrollable subframes.

        Original patch from Thomas Thrainer.

        * WebCoreSupport/ChromeClientQt.cpp:
        (WebCore::ChromeClientQt::invalidateWindow):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@83473 268f45cc-cd09-0410-ab3c-d52691b4dbfc

WebKit2: links don’t update to look visited
<rdar://problem/8806254>
https://bugs.webkit.org/show_bug.cgi?id=58252

Reviewed by Anders Carlsson.

* WebProcess/WebProcess.cpp:
(WebKit::WebProcess::visitedLinkStateChanged):
(WebKit::WebProcess::allVisitedLinkStateChanged):
* WebProcess/WebProcess.h:
Remove use of vestigial sharedPageGroup and instead iterate set of
page groups in use by the process.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@83472 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2011-04-11  Simon Fraser  <simon.fraser@apple.com>

        Reviewed by Dan Bernstein.

        Divide by zero in calcColumnWidth
        https://bugs.webkit.org/show_bug.cgi?id=58230

        Test: fast/multicol/huge-column-count.html

        Make sure we have at least one column, to avoid divide by zero.

        * rendering/RenderBlock.cpp:
        (WebCore::RenderBlock::calcColumnWidth):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@83470 268f45cc-cd09-0410-ab3c-d52691b4dbfc

   [Qt][WK2]http/tests/loading/preload-append-scan.php fails
       https://bugs.webkit.org/show_bug.cgi?id=58253

* platform/qt-wk2/Skipped: Add http/tests/loading/preload-append-scan.php until fix.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@83467 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2011-04-11  Mario Sanchez Prada  <msanchez@igalia.com>

        Reviewed by Xan Lopez.

        [GTK] Implement AccessibilityUIElement::{row|column}IndexRange in DRT
        https://bugs.webkit.org/show_bug.cgi?id=57854

        Unskipped table-cell-spans.html and updated GTK expectations.

        * platform/gtk/Skipped: Unskipped test.
        * platform/gtk/accessibility/table-cell-spans-expected.txt: Added.
2011-04-11  Mario Sanchez Prada  <msanchez@igalia.com>

        Reviewed by Xan Lopez.

        [GTK] Implement AccessibilityUIElement::{row|column}IndexRange in DRT
        https://bugs.webkit.org/show_bug.cgi?id=57854

        Implement missing features in GTK's DRT.

        * DumpRenderTree/gtk/AccessibilityUIElementGtk.cpp:
        (indexRangeInTable): New helper function to get the range string
        for the current cell inside the parent table, either from the
        point of view of rows or columns.
        (AccessibilityUIElement::rowIndexRange): Implemented by relying on
        the new helper function indexRangeInTable().
        (AccessibilityUIElement::columnIndexRange): Ditto.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@83466 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2011-04-11  Mario Sanchez Prada  <msanchez@igalia.com>

        Reviewed by Chris Fleizach.

        [GTK] Implement AccessibilityUIElement::cellForColumnAndRow in DRT
        https://bugs.webkit.org/show_bug.cgi?id=57826

        Skipped aria-tables.html and table-cells.html layout tests.

        Some changes were also made inside the aria-tables.html layout
        test, in order to allow unskipping it in GTK while still keeping
        it working in other platforms, without needing platform specific
        tests, just different expected files.

        * accessibility/aria-tables.html: Changed the layout test so it
        just prints role names instead of comparing their actual values
        with the expected ones in one specific platform (Mac)

        * platform/gtk/Skipped: Unskipped aria-tables.html and table-cells.html.
        * platform/gtk/accessibility/aria-tables-expected.txt: Added new
        expectations for the GTK port.
        * platform/gtk/accessibility/table-cells-expected.txt: Ditto.

        * platform/mac-leopard/Skipped: Skipping aria-tables.html in
        mac-leopard, as I don't know what the expected output would be in
        this case, and since the expected file was invalid anyway.

        * platform/mac/accessibility/aria-tables-expected.txt: Updated to
        check role names, after the changes done in the layout test.
2011-04-11  Mario Sanchez Prada  <msanchez@igalia.com>

        Reviewed by Chris Fleizach.

        [GTK] Implement AccessibilityUIElement::cellForColumnAndRow in DRT
        https://bugs.webkit.org/show_bug.cgi?id=57826

        Implement missing feature in GTK's DRT.

        * DumpRenderTree/gtk/AccessibilityUIElementGtk.cpp:
        (AccessibilityUIElement::cellForColumnAndRow): Implemented, by
        relying on the analogous function from the AtkTable interface.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@83465 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Fix a bug where the context menu does not show in PDF documents if there is no
UI delegate or the UI delegate does not respond to the context menu selector.

https://webkit.org/b/57958

Reviewed by Darin Adler.

* WebView/WebView.mm:
(-[WebView _menuForElement:defaultItems:]): CallUIDelegate returns nil
if UIDelegate is nil or doesn't respond to the selector. So check that
here to distinguish between using defaultMenuItems or the delegate
really returning nil to say "no context menu".

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@83464 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Fix the extract-localizable-strings script errors that started after
WEB_UI_STRING was added to LocalizedStrings.h in WebCore.

* Scripts/extract-localizable-strings: Skip LocalizedStrings.h.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@83463 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Windows build fix.

* JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@83462 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2011-04-11  Tony Chang  <tony@chromium.org>

        Reviewed by Eric Seidel.

        fix crc computation on 64 bit machines
        https://bugs.webkit.org/show_bug.cgi?id=58243

        unsigned long is dependenct on CPU architecture, so use unsigned which is always 32 bits

        * DumpRenderTree/CyclicRedundancyCheck.cpp:
        (makeCrcTable):
        (computeCrc):
        * DumpRenderTree/CyclicRedundancyCheck.h:
        * DumpRenderTree/PixelDumpSupport.cpp:
        (appendIntToVector):
        (convertChecksumToPNGComment):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@83461 268f45cc-cd09-0410-ab3c-d52691b4dbfc

WebKit2: Windows 7 Gestures Window Bounce shouldn't require a sync message
https://bugs.webkit.org/show_bug.cgi?id=58167
<rdar://problem/9259813>

Reviewed by Adam Roben.

Instead of making GestureDidScroll sync, have WebPageWin call from WebProcess ->
UIProcess when the gesture causes the page to scroll to the beginning or the
end of the document.

* UIProcess/PageClient.h:
* UIProcess/WebPageProxy.cpp:
(WebKit::WebPageProxy::gestureDidScroll): Not a sync message anymore.
(WebKit::WebPageProxy::setGestureScrollingLimitReached): Tell the page client that the gesture
    scrolling limnit was reached.
* UIProcess/WebPageProxy.h:
* UIProcess/WebPageProxy.messages.in: Add a new message.
* UIProcess/win/WebView.cpp:
(WebKit::WebView::WebView): Initialize new variable.
(WebKit::WebView::onGesture): Use the state of the member variable, not the response from
    the sync message.
* UIProcess/win/WebView.h:
(WebKit::WebView::setGestureScrollingLimitReached):
* WebProcess/WebPage/WebPage.h:
* WebProcess/WebPage/WebPage.messages.in:
* WebProcess/WebPage/win/WebPageWin.cpp:
(WebKit::scrollbarAtTopOfBottomOrDocument): Returns whether or not the scrollbar is at the
    top or bottom of the document.
(WebKit::WebPage::gestureDidScroll): Track whether or not we started at the beginning
    or end of the document, and whether or not we ended at the beginning or end of the document,
    and send a message if the value changed.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@83460 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Bug 58198 - Clean up JSValue implementation for JSVALUE64

Reviewed by Sam Weinig.

Source/JavaScriptCore:

Remove JSNumberCell, JSImmediate, unify some methods between JSVALUE32_64/JSVALUE64

JSNumberCell.h largely just contained the constructors for JSValue on JSVALUE64,
which should not have been here.  JSImmediate mostly contained uncalled methods,
along with the internal implementation of the JSValue constructors split unnecessarily
across a number of layers of function calls. These could largely be merged back
together. Many methods and constructors from JSVALUE32_64 and JSVALUE64 can by unified.

The .cpp files were empty.

Moving all these methods into JSValue.h seems to be a repro measurable regression, so
I have kept these methods in a separate JSValueInlineMethods.h. Adding the 64-bit tag
values as static const members of JSValue also measures as a repro regression, so I
have made these #defines.

* Android.mk:
* CMakeLists.txt:
* GNUmakefile.list.am:
* JavaScriptCore.exp:
* JavaScriptCore.gypi:
* JavaScriptCore.pro:
* JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.vcproj:
* JavaScriptCore.xcodeproj/project.pbxproj:
    - Removed JSImmediate.h, JSNumberCell.h.
* bytecompiler/BytecodeGenerator.cpp:
(JSC::BytecodeGenerator::emitLoad):
    - Removed class JSImmediate.
* dfg/DFGNonSpeculativeJIT.cpp:
(JSC::DFG::NonSpeculativeJIT::compile):
    - Removed class JSImmediate.
* dfg/DFGSpeculativeJIT.cpp:
(JSC::DFG::SpeculativeJIT::compile):
    - Removed class JSImmediate.
* jit/JITArithmetic.cpp:
(JSC::JIT::compileBinaryArithOpSlowCase):
    - Removed class JSImmediate.
* jit/JITInlineMethods.h:
(JSC::JIT::emitJumpIfJSCell):
(JSC::JIT::emitJumpIfNotJSCell):
(JSC::JIT::emitJumpIfImmediateInteger):
(JSC::JIT::emitJumpIfNotImmediateInteger):
(JSC::JIT::emitFastArithDeTagImmediate):
(JSC::JIT::emitFastArithDeTagImmediateJumpIfZero):
(JSC::JIT::emitFastArithReTagImmediate):
(JSC::JIT::emitTagAsBoolImmediate):
    - Removed class JSImmediate.
* jit/JITOpcodes.cpp:
(JSC::JIT::emit_op_not):
(JSC::JIT::emit_op_jeq_null):
(JSC::JIT::emit_op_jneq_null):
(JSC::JIT::emit_op_get_pnames):
(JSC::JIT::emit_op_eq_null):
(JSC::JIT::emit_op_neq_null):
(JSC::JIT::emitSlow_op_not):
    - Removed class JSImmediate.
* jit/JSInterfaceJIT.h:
    - Removed class JSImmediate.
* runtime/JSCell.h:
    - Removed JSImmediate.h, JSNumberCell.h.
* runtime/JSImmediate.cpp: Removed.
* runtime/JSImmediate.h: Removed.
* runtime/JSNumberCell.cpp: Removed.
* runtime/JSNumberCell.h: Removed.
    - Removed.
* runtime/JSObject.h:
    - Removed JSImmediate.h, JSNumberCell.h.
* runtime/JSString.h:
    - Removed JSImmediate.h, JSNumberCell.h.
* runtime/JSValue.h:
    - Added tags for JSVALUE64, moved out some JSVALUE32_64 methods, unified with JSVALUE64.
* runtime/JSValueInlineMethods.h: Added.
(JSC::JSValue::toInt32):
(JSC::JSValue::toUInt32):
(JSC::JSValue::isUInt32):
(JSC::JSValue::asUInt32):
(JSC::JSValue::uncheckedGetNumber):
(JSC::JSValue::toJSNumber):
(JSC::jsNaN):
(JSC::JSValue::getNumber):
(JSC::JSValue::getBoolean):
(JSC::JSValue::JSValue):
(JSC::JSValue::encode):
(JSC::JSValue::decode):
(JSC::JSValue::operator bool):
(JSC::JSValue::operator==):
(JSC::JSValue::operator!=):
(JSC::JSValue::isUndefined):
(JSC::JSValue::isNull):
(JSC::JSValue::isUndefinedOrNull):
(JSC::JSValue::isCell):
(JSC::JSValue::isInt32):
(JSC::JSValue::isDouble):
(JSC::JSValue::isTrue):
(JSC::JSValue::isFalse):
(JSC::JSValue::tag):
(JSC::JSValue::payload):
(JSC::JSValue::asInt32):
(JSC::JSValue::asDouble):
(JSC::JSValue::asCell):
(JSC::JSValue::isNumber):
(JSC::JSValue::isBoolean):
(JSC::JSValue::makeImmediate):
(JSC::JSValue::immediateValue):
(JSC::reinterpretDoubleToIntptr):
(JSC::reinterpretIntptrToDouble):
    - Methods moved here from JSImmediate.h/JSNumberCell.h/JSValue.h.
* runtime/Operations.h:
    - Removed JSImmediate.h, JSNumberCell.h.
* wtf/StdLibExtras.h:
    - Export bitwise_cast.

Source/WebCore:

JSNumberCell.h has been deprecated.

* ForwardingHeaders/runtime/JSNumberCell.h: Removed.
* bindings/scripts/CodeGeneratorJS.pm:
* bridge/c/c_instance.cpp:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@83459 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Unreviewed; roll out unintentional change to WebFullScreenController.mm.

* WebView/WebFullScreenController.mm:
(-[WebFullScreenController enterFullscreen:]):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@83458 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Build fix.

* html/canvas/CanvasRenderingContext2D.cpp:
(WebCore::CanvasRenderingContext2D::State::State):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@83457 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Document source preload scanned repeatedly
https://bugs.webkit.org/show_bug.cgi?id=58123

Reviewed by Tony Gentilcore.

Don't clear the preload scanner after execution resumes. This would lose the current
scanning point and lead to rescanning when the preload scanner would get reconstructed
due to main parser blocking again.

Instead clear the scanner only in the specific case of receiving new data while the main
parser has already reached the end of the current input.

Also switched to using isWaitingForScripts() instead of m_treeBuilder->isPaused() for consistency.

The case the clearing in resumeParsingAfterScriptExecution() was added for is covered by
http/tests/loading/preload-slow-loading.php.

* html/parser/HTMLDocumentParser.cpp:
(WebCore::HTMLDocumentParser::insert):
(WebCore::HTMLDocumentParser::append):
(WebCore::HTMLDocumentParser::resumeParsingAfterScriptExecution):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@83456 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Unreviewed, Updated chromium test expectations.

Mark aria-table-hierarchy.html as failed instead of missing after the
expected result file is added.

* platform/chromium/test_expectations.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@83455 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2011-04-04  Jer Noble  <jer.noble@apple.com>

        Reviewed by Maciej Stachowiak.

        WK2: PDF: Find in page
        https://bugs.webkit.org/show_bug.cgi?id=57765

        Support searching text within PDF documents.  Find requests must be routed from the
        WebPageProxy, through the PageClient, and to the WKView, where they can be passed to
        the PDFViewController.

        * UIProcess/API/mac/PDFViewController.h:
        * UIProcess/API/mac/PDFViewController.mm:
        (_PDFSelectionsAreEqual): Copied from WebPDFView.
        (-[WKPDFView _nextMatchFor:direction:caseSensitive:wrap:fromSelection:startInSelection:]): Copied from WebPDFView.
        (-[WKPDFView _countMatches:caseSensitive:]): Added.
        (WebKit::PDFViewController::findString): Added.
        (WebKit::PDFViewController::countStringMatches): Added.
        * UIProcess/PageClient.h:
        * UIProcess/API/mac/PageClientImpl.h:
        * UIProcess/API/mac/PageClientImpl.mm:
        (WebKit::PageClientImpl::findStringInCustomRepresentation): Added.
        (WebKit::PageClientImpl::countStringMatchesInCustomRepresentation): Added.
        * UIProcess/API/mac/WKViewInternal.h:
        * UIProcess/API/mac/WKView.mm:
        (-[WKView _findStringInCustomRepresentation:withFindOptions:maxMatchCount:]): Added.
        (-[WKView _countStringMatchesInCustomRepresentation:withFindOptions:maxMatchCount:]): Added.
        * UIProcess/WebPageProxy.h: Moved a number of find-related functions from private: to public: so
            they could be called from PDFViewController.
        * UIProcess/WebPageProxy.cpp:
        (WebKit::WebPageProxy::findString): Added.
        (WebKit::WebPageProxy::countStringMatches): Added.
        * UIProcess/win/WebView.cpp:
        (WebKit::WebView::findStringInCustomRepresentation): Added stub.
        (WebKit::WebView::countStringMatchesInCustomRepresentation): Ditto.
        * UIProcess/win/WebView.h:
        * UIProcess/API/qt/qwkpage_p.h:
        (QWKPagePrivate::findStringInCustomRepresentation): Added stub.
        (QWKPagePrivate::countStringMatchesInCustomRepresentation): Added stub.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@83454 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Build fix.

* html/canvas/CanvasRenderingContext2D.cpp:
(WebCore::CanvasRenderingContext2D::State::State):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@83453 268f45cc-cd09-0410-ab3c-d52691b4dbfc

WebKit2: Implement TextChecker on Windows
https://bugs.webkit.org/show_bug.cgi?id=57862

Reviewed by Anders Carlsson.

Part 6: Update the Spelling UI with the spelling and grammar mistakes.

* UIProcess/API/C/win/WKAPICastWin.h:
(WebKit::toAPI):
Make it possible to go from a WebCore::GrammarDetail to a WebGrammarDetail.

* UIProcess/API/C/win/WKGrammarDetail.cpp:
(WKGrammarDetailCreate):
(WKGrammarDetailGetLocation):
(WKGrammarDetailGetLength):
(WKGrammarDetailCopyGuesses):
(WKGrammarDetailCopyUserDescription):
* UIProcess/API/C/win/WKGrammarDetail.h:

* UIProcess/API/C/win/WKTextChecker.cpp:
(WKTextCheckerCheckSpelling):
Tell the WebTextChecker.
* UIProcess/API/C/win/WKTextChecker.h:

* UIProcess/WebPageProxy.cpp:
(WebKit::WebPageProxy::contextMenuItemSelected):
If the show/hide spelling UI is selected and the spelling UI is not showing, make sure to
advance to the next misspelling (in this case, the first).
This behavior matches that in WebCore and WKView.mm.
(WebKit::WebPageProxy::advanceToNextMisspelling):
Make this function const since it does not modify the WebPageProxy and making it const
allows WebTextChecker::checkSpelling to take a const WebPageProxy.
(WebKit::WebPageProxy::spellDocumentTag):
(WebKit::WebPageProxy::updateSpellingUIWithMisspelledWord):
(WebKit::WebPageProxy::updateSpellingUIWithGrammarString):
* UIProcess/WebPageProxy.h:

* UIProcess/TextChecker.h:
* UIProcess/win/TextCheckerWin.cpp:
(WebKit::TextChecker::uniqueSpellDocumentTag):
Tell the TextCheckerClient which WebPageProxy this tag is for.
(WebKit::TextChecker::updateSpellingUIWithMisspelledWord):
Tell the TextCheckerClient.
(WebKit::TextChecker::updateSpellingUIWithGrammarString):
Ditto.
* UIProcess/mac/TextCheckerMac.mm:
(WebKit::TextChecker::uniqueSpellDocumentTag):
Add the new argument that is unused by this port.
(WebKit::TextChecker::updateSpellingUIWithMisspelledWord):
Ditto.
(WebKit::TextChecker::updateSpellingUIWithGrammarString):
Ditto.
* UIProcess/gtk/TextCheckerGtk.cpp:
(WebKit::TextChecker::uniqueSpellDocumentTag):
Ditto.
(WebKit::TextChecker::updateSpellingUIWithMisspelledWord):
Ditto.
(WebKit::TextChecker::updateSpellingUIWithGrammarString):
Ditto.
* UIProcess/qt/TextCheckerQt.cpp:
(WebKit::TextChecker::uniqueSpellDocumentTag):
Ditto.
(WebKit::TextChecker::updateSpellingUIWithMisspelledWord):
Ditto.
(WebKit::TextChecker::updateSpellingUIWithGrammarString):
Ditto.

* UIProcess/win/WebGrammarDetail.cpp:
(WebKit::WebGrammarDetail::create):
(WebKit::WebGrammarDetail::WebGrammarDetail):
(WebKit::WebGrammarDetail::guesses):
* UIProcess/win/WebGrammarDetail.h:
(WebKit::WebGrammarDetail::location):
(WebKit::WebGrammarDetail::length):
(WebKit::WebGrammarDetail::userDescription):

* UIProcess/win/WebTextChecker.cpp:
(WebKit::WebTextChecker::checkSpelling):
Tell the page to advance to the next misspelling. This matches the WK1 WebView.cpp logic.

* UIProcess/win/WebTextChecker.h:
* UIProcess/win/WebTextCheckerClient.cpp:
(WebKit::WebTextCheckerClient::uniqueSpellDocumentTag):
(WebKit::WebTextCheckerClient::toggleSpellingUIIsShowing):
(WebKit::WebTextCheckerClient::updateSpellingUIWithMisspelledWord):
(WebKit::WebTextCheckerClient::updateSpellingUIWithGrammarString):
* UIProcess/win/WebTextCheckerClient.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@83452 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Assertion failure in CanvasRenderingContext2D::State::fontsNeedUpdate when invalidating the font cache after opening canvas/philip/tests/initial.reset.2dstate.html (occurs on Qt debug test bot)
https://bugs.webkit.org/show_bug.cgi?id=58229

Reviewed by Alexey Proskuryakov.

* html/canvas/CanvasRenderingContext2D.cpp:
(WebCore::CanvasRenderingContext2D::State::State): Added copy constructor, to register the copy with
the font selector if needed.
(WebCore::CanvasRenderingContext2D::State::operator=): Added assignment constructor, to handle
registration with the font selector as needed.
* html/canvas/CanvasRenderingContext2D.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@83451 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2011-04-11  Mario Sanchez Prada  <msanchez@igalia.com>

        Reviewed by Chris Fleizach.

        [GTK] ARIA tables not exposing cells as HTML tables do
        https://bugs.webkit.org/show_bug.cgi?id=57463

        Unskipped passing test and added GTK-specific expectations.

        * platform/gtk/Skipped: Unskip accessibility test.
        * platform/gtk/accessibility/aria-table-hierarchy-expected.txt: Added.
2011-04-11  Mario Sanchez Prada  <msanchez@igalia.com>

        Reviewed by Chris Fleizach.

        [GTK] ARIA tables not exposing cells as HTML tables do
        https://bugs.webkit.org/show_bug.cgi?id=57463

        Expose cells for ARIA grids consistently with HTML tables in GTK,
        keeping the same behaviour for the other platforms.

        * accessibility/AccessibilityARIAGrid.cpp:
        (WebCore::AccessibilityARIAGrid::addChild): Add the row's children
        to the accessibility hierarchy when accessibility objects for rows
        are ignoring accessibility, otherwise add the row.
        (WebCore::AccessibilityARIAGrid::addChildren): Do not check at
        this point whether every child of the table ignores or not
        accessibility, letting addChild() make the proper decision later.

        * accessibility/AccessibilityARIAGridCell.cpp:
        (WebCore::AccessibilityARIAGridCell::parentTable): Consider that
        rows could be ignoring accessibility, hence the parent could be
        retrieved in the first call to parentObjectUnignored().
        (WebCore::AccessibilityARIAGridCell::rowIndexRange): Ditto.
        (WebCore::AccessibilityARIAGridCell::columnIndexRange): Ditto..

        * accessibility/gtk/AccessibilityObjectWrapperAtk.cpp:
        (atkRole): Map ColumnHeader and RowHeader to ATK_ROLE_TABLE_CELL.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@83450 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2011-04-11  Dimitri Glazkov  <dglazkov@chromium.org>

        [GTK] Update expectations after r83445.

        * platform/gtk/fast/layers/video-layer-expected.txt: Updated.
        * platform/gtk/media/video-empty-source-expected.txt: Updated.
        * platform/gtk/media/video-no-audio-expected.txt: Updated.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@83448 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2011-04-11  Thouraya ANDOLSI  <thouraya.andolsi@st.com>

        Reviewed by Oliver Hunt.

        SH4 JIT SUPPORT.
        https://bugs.webkit.org/show_bug.cgi?id=44329

        Add JIT remaining part for SH4 platforms.

        * assembler/MacroAssemblerSH4.h:
        * jit/JIT.h:
        * jit/JITInlineMethods.h:
        * jit/JITOpcodes32_64.cpp:
        * jit/JITPropertyAccess32_64.cpp:
        * jit/JITStubs.cpp:
        * jit/JITStubs.h:
        * jit/JSInterfaceJIT.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@83447 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2011-04-11  Sergio Villar Senin  <svillar@igalia.com>

        Reviewed by Martin Robinson.

        [GTK] Fix make distcheck
        https://bugs.webkit.org/show_bug.cgi?id=58224

        Removed a file that is no longer in the source tree.

        * GNUmakefile.list.am:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@83446 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2011-04-11  Dimitri Glazkov  <dglazkov@chromium.org>

        Reviewed by Martin Robinson.

        REGRESSION(r83397): Most GTK media controls are hidden when no valid source is specified
        https://bugs.webkit.org/show_bug.cgi?id=58204

        Covered by existing tests.

        * platform/gtk/RenderThemeGtk.h:
        (WebCore::RenderThemeGtk::hasOwnDisabledStateHandlingFor): Let GTK handle
            the the state of media controls.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@83445 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2011-04-11  Dimitri Glazkov  <dglazkov@chromium.org>

        Remove obsolete comment, which was mistakenly landed in r83397.

        Though potentially a work of art, the comment lost its functional meaning a long time ago.

        * html/shadow/MediaControls.cpp:
        (WebCore::MediaControls::create): Removed comment.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@83444 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2011-04-05  Hans Wennborg  <hans@chromium.org>

        Reviewed by Steve Block.

        IndexedDB: Introduce skeleton for LevelDB backend
        https://bugs.webkit.org/show_bug.cgi?id=57827

        No new tests: no new functionality (yet).

        * WebCore.gyp/WebCore.gyp:
            Add dependency on leveldb when ENABLE_LEVELDB=1 in feature_defines.
        * WebCore.gypi:
            Add new files.
        * storage/IDBBackingStore.h:
        (WebCore::IDBBackingStore::ObjectStoreRecordIdentifier::~ObjectStoreRecordIdentifier):
            Introduce abstract type ObjectStoreRecordIdentifier. The SQLite
            backend uses integers to refer to a specific row in a table, but
            the LevelDB backend will use something else.
        * storage/IDBDatabaseBackendImpl.cpp:
        (WebCore::IDBDatabaseBackendImpl::createObjectStore):
        (WebCore::IDBDatabaseBackendImpl::createObjectStoreInternal):
        (WebCore::IDBDatabaseBackendImpl::deleteObjectStoreInternal):
        (WebCore::IDBDatabaseBackendImpl::loadObjectStores):
            Change functions to pass along the database id for operations that
            concern object stores. We want to be able to group object stores
            ids per database.
        * storage/IDBFactoryBackendImpl.cpp:
        (WebCore::IDBFactoryBackendImpl::open):
            Decide which backend to use based on the backingStoreType
            parameter.
        * storage/IDBIndexBackendImpl.cpp:
        (WebCore::IDBIndexBackendImpl::IDBIndexBackendImpl):
        (WebCore::IDBIndexBackendImpl::openCursorInternal):
        (WebCore::IDBIndexBackendImpl::getInternal):
        (WebCore::IDBIndexBackendImpl::addingKeyAllowed):
        * storage/IDBIndexBackendImpl.h:
        (WebCore::IDBIndexBackendImpl::create):
            Pass database and object store id to backend functions concerning
            indexes.
        * storage/IDBLevelDBBackingStore.cpp: Added.
        (WebCore::IDBLevelDBBackingStore::IDBLevelDBBackingStore):
        (WebCore::IDBLevelDBBackingStore::~IDBLevelDBBackingStore):
        (WebCore::IDBLevelDBBackingStore::open):
        (WebCore::IDBLevelDBBackingStore::extractIDBDatabaseMetaData):
        (WebCore::IDBLevelDBBackingStore::setIDBDatabaseMetaData):
        (WebCore::IDBLevelDBBackingStore::getObjectStores):
        (WebCore::IDBLevelDBBackingStore::createObjectStore):
        (WebCore::IDBLevelDBBackingStore::deleteObjectStore):
        (WebCore::IDBLevelDBBackingStore::getObjectStoreRecord):
        (WebCore::IDBLevelDBBackingStore::putObjectStoreRecord):
        (WebCore::IDBLevelDBBackingStore::clearObjectStore):
        (WebCore::IDBLevelDBBackingStore::deleteObjectStoreRecord):
        (WebCore::IDBLevelDBBackingStore::nextAutoIncrementNumber):
        (WebCore::IDBLevelDBBackingStore::keyExistsInObjectStore):
        (WebCore::IDBLevelDBBackingStore::forEachObjectStoreRecord):
        (WebCore::IDBLevelDBBackingStore::getIndexes):
        (WebCore::IDBLevelDBBackingStore::createIndex):
        (WebCore::IDBLevelDBBackingStore::deleteIndex):
        (WebCore::IDBLevelDBBackingStore::putIndexDataForRecord):
        (WebCore::IDBLevelDBBackingStore::deleteIndexDataForRecord):
        (WebCore::IDBLevelDBBackingStore::getObjectViaIndex):
        (WebCore::IDBLevelDBBackingStore::getPrimaryKeyViaIndex):
        (WebCore::IDBLevelDBBackingStore::keyExistsInIndex):
        (WebCore::IDBLevelDBBackingStore::openObjectStoreCursor):
        (WebCore::IDBLevelDBBackingStore::openIndexKeyCursor):
        (WebCore::IDBLevelDBBackingStore::openIndexCursor):
        (WebCore::IDBLevelDBBackingStore::createTransaction):
        * storage/IDBLevelDBBackingStore.h: Added.
            Add an empty implementation of the LevelDB backend.
        * storage/IDBObjectStoreBackendImpl.cpp:
            Keep track of database id, use the new RecordIdentifier.
        (WebCore::IDBObjectStoreBackendImpl::IDBObjectStoreBackendImpl):
        (WebCore::IDBObjectStoreBackendImpl::getInternal):
        (WebCore::IDBObjectStoreBackendImpl::putInternal):
        (WebCore::IDBObjectStoreBackendImpl::deleteInternal):
        (WebCore::IDBObjectStoreBackendImpl::clearInternal):
        (WebCore::populateIndex):
        (WebCore::IDBObjectStoreBackendImpl::createIndex):
        (WebCore::IDBObjectStoreBackendImpl::createIndexInternal):
        (WebCore::IDBObjectStoreBackendImpl::deleteIndexInternal):
        (WebCore::IDBObjectStoreBackendImpl::openCursorInternal):
        (WebCore::IDBObjectStoreBackendImpl::loadIndexes):
        (WebCore::IDBObjectStoreBackendImpl::genAutoIncrementKey):
        * storage/IDBObjectStoreBackendImpl.h:
        (WebCore::IDBObjectStoreBackendImpl::create):
        (WebCore::IDBObjectStoreBackendImpl::databaseId):
        * storage/IDBSQLiteBackingStore.cpp:
        (WebCore::IDBSQLiteBackingStore::open):
        (WebCore::IDBSQLiteBackingStore::createObjectStore):
        (WebCore::IDBSQLiteBackingStore::deleteObjectStore):
        (WebCore::IDBSQLiteBackingStore::getObjectStoreRecord):
        (WebCore::IDBSQLiteBackingStore::putObjectStoreRecord):
        (WebCore::IDBSQLiteBackingStore::clearObjectStore):
        (WebCore::IDBSQLiteBackingStore::deleteObjectStoreRecord):
        (WebCore::IDBSQLiteBackingStore::nextAutoIncrementNumber):
        (WebCore::IDBSQLiteBackingStore::keyExistsInObjectStore):
        (WebCore::IDBSQLiteBackingStore::forEachObjectStoreRecord):
        (WebCore::IDBSQLiteBackingStore::getIndexes):
        (WebCore::IDBSQLiteBackingStore::createIndex):
        (WebCore::IDBSQLiteBackingStore::deleteIndex):
        (WebCore::IDBSQLiteBackingStore::putIndexDataForRecord):
        (WebCore::IDBSQLiteBackingStore::deleteIndexDataForRecord):
        (WebCore::IDBSQLiteBackingStore::getObjectViaIndex):
        (WebCore::IDBSQLiteBackingStore::getPrimaryKeyViaIndex):
        (WebCore::IDBSQLiteBackingStore::keyExistsInIndex):
        (WebCore::IDBSQLiteBackingStore::openObjectStoreCursor):
        (WebCore::IDBSQLiteBackingStore::openIndexKeyCursor):
        (WebCore::IDBSQLiteBackingStore::openIndexCursor):
        * storage/IDBSQLiteBackingStore.h:
2011-04-05  Hans Wennborg  <hans@chromium.org>

        Reviewed by Steve Block.

        IndexedDB: Introduce skeleton for LevelDB backend
        https://bugs.webkit.org/show_bug.cgi?id=57827

        * features.gypi:
            Add ENABLE_LEVELDB=0.
        * public/WebIDBFactory.h:
        (WebKit::WebIDBFactory::open):
            Remove pre-roll open() function.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@83443 268f45cc-cd09-0410-ab3c-d52691b4dbfc

[Qt] HTML5 Drag and Drop demos not working
https://bugs.webkit.org/show_bug.cgi?id=56486

Reviewed by Andreas Kling.

Handling the drop with JavaScript was not working with Qt because if the action is ignored
in response to DragEnter, no further events are sent to the view.

Drag and drop is defined and used differently by webpages. The drag move events are determining
what action should take place. To adopt this behavior for Qt, we always accept drag enter events
on the widget.

* Api/qwebpage.cpp:
(QWebPagePrivate::dragEnterEvent):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@83442 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2011-04-11  Alexander Pavlov  <apavlov@chromium.org>

        Reviewed by Yury Semikhatsky.

        Web Inspector: Duplicate CSS properties are reported for non-lowercase property names in the Styles sidebar
        https://bugs.webkit.org/show_bug.cgi?id=58226

        * inspector/styles/get-set-stylesheet-text-expected.txt:
        * inspector/styles/get-set-stylesheet-text.html:
2011-04-11  Alexander Pavlov  <apavlov@chromium.org>

        Reviewed by Yury Semikhatsky.

        Web Inspector: Duplicate CSS properties are reported for non-lowercase property names in the Styles sidebar
        https://bugs.webkit.org/show_bug.cgi?id=58226

        * inspector/InspectorStyleSheet.cpp:
        (WebCore::InspectorStyle::populateAllProperties):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@83441 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2011-04-11  Andras Becsi  <abecsi@webkit.org>

        Reviewed by Andreas Kling.

        [Qt] REGRESSION(83122): tst_QWebElement::style() fails
        https://bugs.webkit.org/show_bug.cgi?id=58032

        According to the documentation of QWebElement the styleProperty method should
        not respect style inheritance and other CSS rules for the InlineStyle enum.

        r83122 fixed this behaviour.

        * tests/qwebelement/tst_qwebelement.cpp:
        (tst_QWebElement::style): Fix the expected color for QWebElement::InlineStyle.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@83440 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2011-04-11  Andreas Kling  <andreas.kling@nokia.com>

        Build fix after r83436.

        * WebCoreSupport/FrameLoaderClientQt.cpp:
        (WebCore::FrameLoaderClientQt::dispatchDidReceiveResponse):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@83439 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2011-04-07  Pavel Podivilov  <podivilov@chromium.org>

        Reviewed by Pavel Feldman.

        Web Inspector: get rid of source frame delegates for resources panel.
        https://bugs.webkit.org/show_bug.cgi?id=58041

        Use TextViewer configured with a domain-specific delegate to show/edit resource contents in resources panel.

        * inspector/front-end/ResourceView.js:
        (WebInspector.ResourceView.createResourceView):
        (WebInspector.ResourceView.resourceViewTypeMatchesResource):
        (WebInspector.ResourceSourceFrame):
        (WebInspector.ResourceSourceFrame.prototype.get resource):
        (WebInspector.ResourceSourceFrame.prototype.contentEditable):
        (WebInspector.ResourceSourceFrame.prototype._requestContent):
        (WebInspector.CSSSourceFrame):
        (WebInspector.CSSSourceFrame.prototype.contentEditable):
        (WebInspector.CSSSourceFrame.prototype._editContent):
        (WebInspector.CSSSourceFrame.prototype._editContent.handleInfos):
        (WebInspector.CSSSourceFrame.prototype._saveStyleSheet):
        * inspector/front-end/SourceFrame.js:
        (WebInspector.SourceFrame.prototype._ensureContentLoaded):
        (WebInspector.SourceFrame.prototype._requestContent):
        (WebInspector.SourceFrame.prototype._handleSave):
        (WebInspector.SourceFrame.prototype._editContent):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@83438 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2011-04-11  Alexis Menard  <alexis.menard@openbossa.org>

        Reviewed by Andreas Kling.

        [Qt] Regression : r83051 Oxygen's lineedits are not rendered properly
        https://bugs.webkit.org/show_bug.cgi?id=58076

        State_Sunken is more generic than pressed. It is used by items such as frames or
        line edits because they are "sunken" frames (e.g. QLineEdit). It can be required
        by some style like Oxygen. Therefore only in the mobile theme we check if the object
        is pressed.

        * platform/qt/RenderThemeQt.cpp:
        (WebCore::RenderThemeQt::paintTextField):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@83437 268f45cc-cd09-0410-ab3c-d52691b4dbfc