Versioning.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@92905 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Unreviewed, rolling out r92880.
http://trac.webkit.org/changeset/92880
https://bugs.webkit.org/show_bug.cgi?id=66123

Breaks compile in VS2010 (Requested by jamesr_ on #webkit).

Patch by Sheriff Bot <webkit.review.bot@gmail.com> on 2011-08-11

* wtf/PassRefPtr.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@92903 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Fix sunspider-compare-results to default to the same suite version as the sunspider script.

Rubber-stamped by Maciej Stachowiak.

* sunspider-compare-results:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@92902 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Don't conditionalize the use of -fomit-frame-pointer on compiler version as
all of our supported compilers are now new enough to have the same, sane behavior.

Rubber-stamped by Sam Weinig.

* Configurations/JavaScriptCore.xcconfig:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@92901 268f45cc-cd09-0410-ab3c-d52691b4dbfc

[chromium] Defer managed texture creation and destruction until updateCompositorResources
https://bugs.webkit.org/show_bug.cgi?id=64772

Patch by James Robinson <jamesr@chromium.org> on 2011-08-11
Reviewed by Kenneth Russell.

This adds support for deferring GraphicsContext3D calls for texture creation and destruction for managed
textures and defers these calls until updateCompositorResources() for contents textures.  The primary benefit of
this change is that it decouples managing the texture budget for the next frame (which has to happen before
painting layer contents) from the actual GL calls that create/destroy textures.  That way, in the threaded
compositing world we can continue to use textures from the previous frame while software painting contents for
the next frame into software buffers or SkPictures.

Also renames LayerTexture to ManagedTexture to better reflect what it is.  Not all ManagedTextures we create are
necessarily associated with layers, but they are all managed by a TextureManager.

Covered by compositing/ layout tests.

* WebCore.gypi:
* platform/graphics/chromium/ContentLayerChromium.h:
* platform/graphics/chromium/ImageLayerChromium.cpp:
(WebCore::ImageLayerTextureUpdater::updateTextureRect):
* platform/graphics/chromium/LayerRendererChromium.cpp:
(WebCore::LayerRendererChromium::releaseTextures):
(WebCore::LayerRendererChromium::drawLayers):
(WebCore::LayerRendererChromium::updateLayers):
(WebCore::LayerRendererChromium::getOffscreenLayerTexture):
(WebCore::LayerRendererChromium::useRenderSurface):
(WebCore::LayerRendererChromium::initializeSharedObjects):
* platform/graphics/chromium/LayerRendererChromium.h:
* platform/graphics/chromium/LayerTextureUpdater.h:
* platform/graphics/chromium/LayerTextureUpdaterCanvas.cpp:
(WebCore::LayerTextureUpdaterBitmap::updateTextureRect):
(WebCore::LayerTextureUpdaterSkPicture::updateTextureRect):
* platform/graphics/chromium/LayerTextureUpdaterCanvas.h:
* platform/graphics/chromium/LayerTilerChromium.cpp:
(WebCore::LayerTilerChromium::getSingleTexture):
(WebCore::LayerTilerChromium::createTile):
(WebCore::LayerTilerChromium::updateRect):
(WebCore::LayerTilerChromium::drawTiles):
* platform/graphics/chromium/LayerTilerChromium.h:
(WebCore::LayerTilerChromium::Tile::Tile):
(WebCore::LayerTilerChromium::Tile::texture):
* platform/graphics/chromium/ManagedTexture.cpp: Renamed from Source/WebCore/platform/graphics/chromium/LayerTexture.cpp.
(WebCore::ManagedTexture::ManagedTexture):
(WebCore::ManagedTexture::~ManagedTexture):
(WebCore::ManagedTexture::isValid):
(WebCore::ManagedTexture::reserve):
(WebCore::ManagedTexture::unreserve):
(WebCore::ManagedTexture::bindTexture):
(WebCore::ManagedTexture::framebufferTexture2D):
* platform/graphics/chromium/ManagedTexture.h: Renamed from Source/WebCore/platform/graphics/chromium/LayerTexture.h.
(WebCore::ManagedTexture::create):
(WebCore::ManagedTexture::format):
(WebCore::ManagedTexture::isReserved):
* platform/graphics/chromium/RenderSurfaceChromium.h:
* platform/graphics/chromium/TextureManager.cpp:
(WebCore::TextureManager::TextureManager):
(WebCore::TextureManager::deleteEvictedTextures):
(WebCore::TextureManager::removeTexture):
(WebCore::TextureManager::allocateTexture):
(WebCore::TextureManager::requestTexture):
* platform/graphics/chromium/TextureManager.h:
(WebCore::TextureManager::create):
(WebCore::TextureManager::setAssociatedContextDebugOnly):
(WebCore::TextureManager::associatedContextDebugOnly):
* platform/graphics/chromium/TiledLayerChromium.h:
* platform/graphics/chromium/cc/CCHeadsUpDisplay.cpp:
(WebCore::CCHeadsUpDisplay::draw):
* platform/graphics/chromium/cc/CCHeadsUpDisplay.h:
* platform/graphics/chromium/cc/CCRenderSurface.cpp:
(WebCore::CCRenderSurface::prepareContentsTexture):
(WebCore::CCRenderSurface::drawSurface):
* platform/graphics/chromium/cc/CCRenderSurface.h:
(WebCore::CCRenderSurface::contentsTexture):
* platform/graphics/chromium/cc/CCTiledLayerImpl.cpp:
(WebCore::CCTiledLayerImpl::bindContentsTexture):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@92900 268f45cc-cd09-0410-ab3c-d52691b4dbfc

<webkit.org/b/66113> [CFNetwork] willCacheResponse() leaks CFCachedURLResponseRef if delegate changes CacheStoragePolicy

Reviewed by Joseph Pecoraro.

This first appeared in ToT WebKit r23462 (Windows merge).

* platform/network/cf/ResourceHandleCFNet.cpp:
(WebCore::willCacheResponse): Only retain cachedResponse if we
are returning the same object passed into the method, otherwise
we end up double-retaining the new object created.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@92897 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Rework some Makefile logic to remove a double-negative.

Reviewed by Jon Honeycutt.

* DumpRenderTree/mac/PerlSupport/Makefile:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@92896 268f45cc-cd09-0410-ab3c-d52691b4dbfc

[chromium] Make WebViewImpl point at CCLayerTreeHost and related separation
https://bugs.webkit.org/show_bug.cgi?id=65791

With this patch, LayerRendererChromium becomes increasingly responsible
for rendering, while render scheduling and tree hosting moves into the
CCLayerTreeHost.

Reviewed by James Robinson.

Source/WebCore:

* platform/graphics/chromium/ContentLayerChromium.cpp:
(WebCore::ContentLayerChromium::createTextureUpdaterIfNeeded):
* platform/graphics/chromium/LayerChromium.h:
* platform/graphics/chromium/LayerRendererChromium.cpp:
(WebCore::LayerRendererChromium::create):
(WebCore::LayerRendererChromium::LayerRendererChromium):
(WebCore::LayerRendererChromium::initialize):
(WebCore::LayerRendererChromium::releaseTextures):
(WebCore::LayerRendererChromium::updateRootLayerContents):
(WebCore::LayerRendererChromium::drawRootLayer):
(WebCore::LayerRendererChromium::invalidateRootLayerRect):
(WebCore::LayerRendererChromium::rootLayerChanged):
(WebCore::LayerRendererChromium::viewportChanged):
(WebCore::LayerRendererChromium::updateLayers):
(WebCore::LayerRendererChromium::drawLayers):
(WebCore::LayerRendererChromium::drawLayersInternal):
(WebCore::LayerRendererChromium::getFramebufferPixels):
(WebCore::LayerRendererChromium::getOffscreenLayerTexture):
(WebCore::LayerRendererChromium::copyOffscreenTextureToDisplay):
(WebCore::LayerRendererChromium::useRenderSurface):
(WebCore::LayerRendererChromium::setScissorToRect):
(WebCore::LayerRendererChromium::layerTreeAsText):
(WebCore::LayerRendererChromium::dumpRenderSurfaces):
* platform/graphics/chromium/LayerRendererChromium.h:
(WebCore::LayerRendererChromium::settings):
(WebCore::LayerRendererChromium::owner):
(WebCore::LayerRendererChromium::rootLayer):
(WebCore::LayerRendererChromium::skiaContext):
* platform/graphics/chromium/WebGLLayerChromium.cpp:
(WebCore::WebGLLayerChromium::setTextureUpdated):
* platform/graphics/chromium/cc/CCHeadsUpDisplay.cpp:
(WebCore::CCHeadsUpDisplay::draw):
(WebCore::CCHeadsUpDisplay::enabled):
* platform/graphics/chromium/cc/CCHeadsUpDisplay.h:
* platform/graphics/chromium/cc/CCLayerTreeHost.cpp:
(WebCore::CCLayerTreeHost::create):
(WebCore::CCLayerTreeHost::CCLayerTreeHost):
(WebCore::CCLayerTreeHost::initialize):
(WebCore::CCLayerTreeHost::animateAndLayout):
(WebCore::CCLayerTreeHost::createLayerTreeHostCommitter):
(WebCore::CCLayerTreeHost::createLayerTreeHostImpl):
(WebCore::CCLayerTreeHost::context):
(WebCore::CCLayerTreeHost::compositeAndReadback):
(WebCore::CCLayerTreeHost::createRootLayerPainter):
(WebCore::CCLayerTreeHost::finishAllRendering):
(WebCore::CCLayerTreeHost::invalidateRootLayerRect):
(WebCore::CCLayerTreeHost::setNeedsCommitAndRedraw):
(WebCore::CCLayerTreeHost::setNeedsRedraw):
(WebCore::CCLayerTreeHost::setRootLayer):
(WebCore::CCLayerTreeHost::setViewport):
(WebCore::CCLayerTreeHost::setVisible):
(WebCore::CCLayerTreeHost::doComposite):
(WebCore::CCLayerTreeHost::composite):
(WebCore::CCLayerTreeHost::reallocateRenderer):
* platform/graphics/chromium/cc/CCLayerTreeHost.h:
(WebCore::CCLayerTreeHost::animating):
(WebCore::CCLayerTreeHost::setAnimating):
(WebCore::CCLayerTreeHost::rootLayer):
(WebCore::CCLayerTreeHost::settings):
(WebCore::CCLayerTreeHost::viewportContentRect):
(WebCore::CCLayerTreeHost::viewportScrollPosition):
(WebCore::CCLayerTreeHost::viewportVisibleRect):
* platform/graphics/chromium/cc/CCLayerTreeHostImpl.cpp:
(WebCore::CCLayerTreeHostImpl::create):
(WebCore::CCLayerTreeHostImpl::CCLayerTreeHostImpl):
(WebCore::CCLayerTreeHostImpl::drawLayers):
(WebCore::CCLayerTreeHostImpl::drawLayersOnMainThread):
* platform/graphics/chromium/cc/CCLayerTreeHostImpl.h:
* platform/graphics/chromium/cc/CCLayerTreeHostImplProxy.cpp:
(WebCore::CCLayerTreeHostImplProxy::initImplOnCCThread):

Source/WebKit/chromium:

* src/WebViewImpl.cpp:
(WebKit::WebViewImpl::WebViewImpl):
(WebKit::WebViewImpl::resize):
(WebKit::WebViewImpl::animate):
(WebKit::WebViewImpl::doPixelReadbackToCanvas):
(WebKit::WebViewImpl::paint):
(WebKit::WebViewImpl::composite):
(WebKit::WebViewImpl::setRootPlatformLayer):
(WebKit::WebViewImpl::setRootLayerNeedsDisplay):
(WebKit::WebViewImpl::scrollRootLayerRect):
(WebKit::WebViewImpl::invalidateRootLayerRect):
(WebKit::getCompositorSettings):
(WebKit::WebViewImpl::setIsAcceleratedCompositingActive):
(WebKit::WebViewImpl::createRootLayerPainter):
(WebKit::WebViewImpl::animateAndLayout):
(WebKit::WebViewImpl::didRecreateGraphicsContext):
(WebKit::WebViewImpl::scheduleComposite):
(WebKit::WebViewImpl::updateLayerTreeViewport):
(WebKit::WebViewImpl::graphicsContext3D):
(WebKit::WebViewImpl::setVisibilityState):
* src/WebViewImpl.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@92895 268f45cc-cd09-0410-ab3c-d52691b4dbfc

WebCore::AVFWrapper fails to remove observer for kCACFContextNeedsFlushNotification
https://bugs.webkit.org/show_bug.cgi?id=66116

We need to pass 0 as the object parameter to CFNotificationCenterRemoveObserver() when removing
kCACFContextNeedsFlushNotification to match what we do when we registered for the same
notification with CFNotificationCenterAddObserver().

Reviewed by John Sullivan.

No new tests, uses existing media tests.

* platform/graphics/avfoundation/MediaPlayerPrivateAVFoundationCF.cpp:
(WebCore::AVFWrapper::disconnectAndDeleteAVFWrapper): Pass 0 as the object parameter to CFNotificationCenterRemoveObserver().
(WebCore::AVFWrapper::notificationCallback): Remove obsolete FIXME comment.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@92894 268f45cc-cd09-0410-ab3c-d52691b4dbfc

[chromium] Expose "min/max scroll size", "has horizontal/vertical scrollbar", "number of wheel handlers" to clients
https://bugs.webkit.org/show_bug.cgi?id=66094

Reviewed by James Robinson.

We want to implement two-finger-swipe for history like safari does on
os x lion. The way this works is:
* If a page has no horizontal scrollbars, or the page is all the way
  to the left, scrolling to the left will go back in history instead.
* Same for right/forward.

This is modelled after the WebKit2 functions
WKPageIsPinnedToLeftSide(), WKPageIsPinnedToRightSide(),
WKPageHasHorizontalScrollbar(), and
webPageProxy::willHandleHorizontalScrollEvents().

* public/WebFrame.h:
* public/WebViewClient.h:
(WebKit::WebViewClient::numWheelEventHandlersChanged):
* src/ChromeClientImpl.cpp:
(WebKit::ChromeClientImpl::numWheelEventHandlersChanged):
* src/ChromeClientImpl.h:
* src/WebFrameImpl.cpp:
(WebKit::WebFrameImpl::hasHorizontalScrollbar):
(WebKit::WebFrameImpl::hasVerticalScrollbar):
(WebKit::WebFrameImpl::isPinnedToLeft):
(WebKit::WebFrameImpl::isPinnedToRight):
* src/WebFrameImpl.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@92893 268f45cc-cd09-0410-ab3c-d52691b4dbfc

DFG JIT verbose mode does not report the generated types of nodes
https://bugs.webkit.org/show_bug.cgi?id=65830

Reviewed by Sam Weinig.

Added code that prints the type selected for each node's result.

* dfg/DFGGenerationInfo.h:
(JSC::DFG::dataFormatToString):
* dfg/DFGNonSpeculativeJIT.cpp:
(JSC::DFG::NonSpeculativeJIT::compile):
* dfg/DFGSpeculativeJIT.cpp:
(JSC::DFG::SpeculativeJIT::compile):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@92892 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Share code between isStyleSpanOrSpanWithOnlyStyleAttribute, isUnstyledStyleSpan,
isSpanWithoutAttributesOrUnstyleStyleSpan and replaceWithSpanOrRemoveIfWithoutAttributes
https://bugs.webkit.org/show_bug.cgi?id=66091

Reviewed by Tony Chang.

Source/WebCore:

Extracted common code as hasNoAttributeOrOnlyStyleAttribute. The only behavioral difference is that
replaceWithSpanOrRemoveIfWithoutAttributes will now remove elements with class="Apple-style-span",
for which I'm adding a test.

Test: editing/style/remove-styled-element-with-style-span.html

* editing/ApplyStyleCommand.cpp:
(WebCore::hasNoAttributeOrOnlyStyleAttribute):
(WebCore::isStyleSpanOrSpanWithOnlyStyleAttribute):
(WebCore::isUnstyledStyleSpan):
(WebCore::isSpanWithoutAttributesOrUnstyleStyleSpan):
(WebCore::ApplyStyleCommand::replaceWithSpanOrRemoveIfWithoutAttributes):

LayoutTests:

Added a test to ensure WebKit removes implicitly styled elements such as em, b, etc...
with class="Apple-style-span" when they're not necessary.

* editing/style/remove-styled-element-with-style-span-expected.txt: Added.
* editing/style/remove-styled-element-with-style-span.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@92891 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Source/WebCore: Add HTMLUnknownElement interface as defined in
http://www.w3.org/TR/html5/elements.html#elements-in-the-dom.
https://bugs.webkit.org/show_bug.cgi?id=41841

Patch by Tom Zakrajsek <tomz@codeaurora.org> on 2011-08-11
Reviewed by Adam Barth.

Test: fast/html/unknown-tag.html

* CMakeLists.txt:
* CodeGenerators.pri:
* DerivedSources.cpp:
* DerivedSources.make:
* GNUmakefile.list.am:
* WebCore.gypi:
* WebCore.vcproj/WebCore.vcproj:
* WebCore.xcodeproj/project.pbxproj:
* bindings/scripts/CodeGeneratorV8.pm:
(IsDOMNodeType):
* dom/make_names.pl:
(defaultParametersHash):
(buildConstructorMap):
(printJSElementIncludes):
(printElementIncludes):
(printWrapperFunctions):
(printWrapperFactoryCppFile):
* html/HTMLTagNames.in:
* html/HTMLUnknownElement.h: Added.
(WebCore::HTMLUnknownElement::create):
(WebCore::HTMLUnknownElement::HTMLUnknownElement):
* html/HTMLUnknownElement.idl: Added.
* mathml/mathtags.in:
* page/DOMWindow.idl:
* svg/svgtags.in:

LayoutTests: Created tests for HTMLUnknownElement interface.
https://bugs.webkit.org/show_bug.cgi?id=41841.

Patch by Tom Zakrajsek <tomz@codeaurora.org> on 2011-08-11
Reviewed by Adam Barth.

* fast/dom/prototype-inheritance-expected.txt:
* fast/html/unknown-tag.html: Added.
* platform/chromium/fast/dom/prototype-inheritance-expected.txt:
* platform/gtk/fast/dom/prototype-inheritance-expected.txt:
* platform/qt/fast/dom/prototype-inheritance-expected.txt:
* platform/win/fast/dom/prototype-inheritance-expected.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@92890 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Add a FIXME.

* TestWebKitAPI/TestsController.cpp:
(TestWebKitAPI::TestsController::TestsController):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@92889 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Remove dead code: borderInnerRect
https://bugs.webkit.org/show_bug.cgi?id=66087

Reviewed by Adam Barth.

Removing an old and currently unused function: RenderObject::borderInnerRect.

No new tests since I'm just pruning dead code.

* rendering/RenderObject.cpp:
* rendering/RenderObject.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@92888 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Initialize threading before running any tests
https://bugs.webkit.org/show_bug.cgi?id=66112

Reviewed by Adam Barth.

* TestWebKitAPI/TestsController.cpp:
(TestWebKitAPI::TestsController::TestsController):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@92887 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Add unexpected-passes.html to TestFailures for marking tests as passing
https://bugs.webkit.org/show_bug.cgi?id=66102

Reviewed by Dimitri Glazkov.

* BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/garden-o-matic.html:
* BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/main.js:
    - Move some code into the library so it can be shared.
* BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/model.js:
* BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/ui.js:
* BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/ui_unittests.js:
    - Add round-trip unit tests.
* BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/unexpected-passes.html: Added.
* BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/unexpected-passes.js: Added.
    - I'm not entirely sure this feature should remain a separate HTML
      file, but it seemed better than cluttering up the main HTML file
      with too much extra stuff.
* Scripts/webkitpy/tool/servers/gardeningserver.py:
* Scripts/webkitpy/tool/servers/gardeningserver_unittest.py:
    - The gardening server couldn't handle adding PASS expectations.  Now it can.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@92886 268f45cc-cd09-0410-ab3c-d52691b4dbfc

pfeldman says we need to rebaseline this test.  The massively complex
baseline fallback graph confused him when he created a platform/mac
fallback.

* platform/chromium-win/inspector/debugger: Added.
* platform/chromium-win/inspector/debugger/scripts-panel-expected.txt: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@92885 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Avoid adding the visible wash layer twice
https://bugs.webkit.org/show_bug.cgi?id=66098

Reviewed by Chris Marrin.

Fix two issues; only create a m_visibleTileWashLayer if we don't have one
already, and avoid adding multiple m_visibleTileWashLayers to the sublayers
array when flipping between tiled and non-tiled.

* platform/graphics/ca/GraphicsLayerCA.cpp:
(WebCore::GraphicsLayerCA::swapFromOrToTiledLayer):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@92884 268f45cc-cd09-0410-ab3c-d52691b4dbfc

The PageSerializer should always use absolute URL
https://bugs.webkit.org/show_bug.cgi?id=66006

Patch by Benjamin Poulain <benjamin@webkit.org> on 2011-08-11
Reviewed by Adam Barth.

Page serializer resolves every URL for storing the resources.
The MarkupAccumulator used there should use the same rules in order to
create a valid file.

No new tests because the test infrastructure for this
does not exist with layout tests and the Chromium unit test for
the serializer are disabled.

* page/PageSerializer.cpp:
(WebCore::SerializerMarkupAccumulator::SerializerMarkupAccumulator):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@92883 268f45cc-cd09-0410-ab3c-d52691b4dbfc

NRWT has wrong fallback order for Mac now that Lion exists
https://bugs.webkit.org/show_bug.cgi?id=66093

Reviewed by Adam Barth.

ORWT used a different system for fallback orders than NRWT did.
I moved win.py to match ORWT fallback orders in bug 64486.
This bug moves mac.py to match ORWT fallback order
and adds 'lion' as a supported OS version.

Because ChromiumMac shares OS version detection code with AppleMac
this also added support for 'lion' to chromium mac.  Hopefully that's a good thing.

* Scripts/webkitpy/layout_tests/port/chromium_mac.py:
* Scripts/webkitpy/layout_tests/port/chromium_mac_unittest.py:
* Scripts/webkitpy/layout_tests/port/mac.py:
* Scripts/webkitpy/layout_tests/port/mac_unittest.py:
* Scripts/webkitpy/layout_tests/port/win.py:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@92882 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Move some Snow Leopard test results to the right place.

* platform/mac-snowleopard/fast/loader/file-url-mimetypes-2-expected.txt: Removed.
* platform/mac-snowleopard/fast/loader/file-url-mimetypes-expected.txt: Removed.
* platform/mac-snowleopard/fast/text/international/Geeza-Pro-vertical-metrics-adjustment-expected.txt: Removed.
* platform/mac-snowleopard/fast/text/international/Geeza-Pro-vertical-metrics-adjustment.html: Removed.
* platform/mac-snowleopard/platform/mac/fast/loader/file-url-mimetypes-expected.txt: Copied from platform/mac-snowleopard/fast/loader/file-url-mimetypes-expected.txt.
* platform/mac-snowleopard/platform/mac/fast/text/international: Added.
* platform/mac-snowleopard/platform/mac/fast/text/international/Geeza-Pro-vertical-metrics-adjustment-expected.txt: Copied from platform/mac-snowleopard/fast/text/international/Geeza-Pro-vertical-metrics-adjustment-expected.txt.
* platform/mac-snowleopard/platform/mac/fast/text/international/Geeza-Pro-vertical-metrics-adjustment.html: Copied from platform/mac-snowleopard/fast/text/international/Geeza-Pro-vertical-metrics-adjustment.html.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@92881 268f45cc-cd09-0410-ab3c-d52691b4dbfc

nullptr can't be used for PassRefPtr
https://bugs.webkit.org/show_bug.cgi?id=66024

Patch by James Robinson <jamesr@chromium.org> on 2011-08-11
Reviewed by Anders Carlsson.

* wtf/PassRefPtr.h:
(WTF::PassRefPtr::PassRefPtr):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@92880 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Record that these tests are now passing in these configurations.

* platform/chromium/test_expectations.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@92879 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2011-08-11  Pavel Feldman  <pfeldman@chromium.org>

        Not reviewed: inspector's scripts-panel test baseline for mac added.

        * platform/mac/inspector/debugger/scripts-panel-expected.txt: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@92878 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Update test results now that we use absolute URLs.

* BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/checkout_unittests.js:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@92876 268f45cc-cd09-0410-ab3c-d52691b4dbfc

garden-o-matic should be able to determine when tests unexpectedly pass
https://bugs.webkit.org/show_bug.cgi?id=66092

Reviewed by Dimitri Glazkov.

This patch just contains the library methods for extracting this
information.  I'll add some UI in a subsequent patch.

* BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/results.js:
* BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/results_unittests.js:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@92875 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Enable fast path scrolling for composited iframes
https://bugs.webkit.org/show_bug.cgi?id=65056

Reviewed by James Robinson.

Source/WebCore:

iframes with composited layers for content should be able to do fast
path scrolling because they draw into their own backing. To enable
this, useSlowRepaints is modified to return true only for the cases
that apply to composited layers. The fast path scrolling function is
modified to send the invalidation to the backing itself rather than to
the window.

Renamed a number of "slow repaint" functions to be the more accurate
"cannot blit to window" as it will not always force slow path
scrolling.

* page/FrameView.cpp:
(WebCore::FrameView::reset):
(WebCore::FrameView::useSlowRepaints):
(WebCore::FrameView::useSlowRepaintsIfNotOverlapped):
(WebCore::FrameView::contentsInCompositedLayer):
(WebCore::FrameView::setCannotBlitToWindow):
(WebCore::FrameView::scrollContentsFastPath):
(WebCore::FrameView::scrollContentsSlowPath):
* page/FrameView.h:
* rendering/RenderLayer.cpp:
(WebCore::RenderLayer::cannotBlitToWindow):
* rendering/RenderLayer.h:
* rendering/RenderView.cpp:
(WebCore::RenderView::paintBoxDecorations):

LayoutTests:

Added test of iframe overlapping another iframe.  Mac gets its own
incorrect expectations due to bug 63282.

* compositing/iframes/overlapped-iframe-iframe-expected.txt: Added.
* compositing/iframes/overlapped-iframe-iframe.html: Added.
* platform/chromium-gpu/compositing/iframes/overlapped-iframe-iframe-expected.txt: Added.
* platform/mac/compositing/iframes/overlapped-iframe-iframe-expected.txt: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@92874 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Support creating compositing layers for scrollable frames and iframes
https://bugs.webkit.org/show_bug.cgi?id=55257

Reviewed by James Robinson.

Source/WebCore:

When force compositing mode is turned on, iframes with overflow will
become composited (in order to enable faster scrolling).

This patch was originally by Daniel Sievers <sievers@chromium.org>.

Tests: platform/chromium/compositing/force-compositing-mode/no-overflow-iframe-layer.html
       platform/chromium/compositing/force-compositing-mode/overflow-hidden-iframe-layer.html
       platform/chromium/compositing/force-compositing-mode/overflow-iframe-enter-compositing.html
       platform/chromium/compositing/force-compositing-mode/overflow-iframe-layer.html
       platform/chromium/compositing/force-compositing-mode/overflow-iframe-leave-compositing.html

* rendering/RenderLayerCompositor.cpp:
(WebCore::RenderLayerCompositor::cacheAcceleratedCompositingFlags):
(WebCore::RenderLayerCompositor::requiresCompositingForScrollableFrame):
* rendering/RenderLayerCompositor.h:

LayoutTests:

* platform/chromium/compositing/force-compositing-mode/no-overflow-iframe-layer-expected.txt: Added.
* platform/chromium/compositing/force-compositing-mode/no-overflow-iframe-layer.html: Added.
* platform/chromium/compositing/force-compositing-mode/overflow-hidden-iframe-layer-expected.txt: Added.
* platform/chromium/compositing/force-compositing-mode/overflow-hidden-iframe-layer.html: Added.
* platform/chromium/compositing/force-compositing-mode/overflow-iframe-enter-compositing-expected.txt: Added.
* platform/chromium/compositing/force-compositing-mode/overflow-iframe-enter-compositing.html: Added.
* platform/chromium/compositing/force-compositing-mode/overflow-iframe-layer-expected.txt: Added.
* platform/chromium/compositing/force-compositing-mode/overflow-iframe-layer.html: Added.
* platform/chromium/compositing/force-compositing-mode/overflow-iframe-leave-compositing-expected.txt: Added.
* platform/chromium/compositing/force-compositing-mode/overflow-iframe-leave-compositing.html: Added.
* platform/chromium/compositing/force-compositing-mode/resources/subframe-big.html: Added.
* platform/chromium/compositing/force-compositing-mode/resources/subframe-small.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@92873 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Remove the temporary workaround added by trac.webkit.org/changeset/51338
https://bugs.webkit.org/show_bug.cgi?id=65908

Patch by Marja Hölttä <marja@chromium.org> on 2011-08-11
Reviewed by Adam Barth.

SecurityOrigin for a file URL returns "file://", and
SecurityOrigin::createFromString("file://") creates a unique (null)
security origin (because "file://" is first canonicalized to "file:///"
and that is a directory). This means the conversion SecurityOrigin ->
string -> SecurityOrigin doens't give back the same SecurityOrigin in
case of file URLs.

WebStorageNamespaceImpl::createStorageArea contains a workaround which
changes the string "file://" into "file:///a" before giving it to
SecurityOrigin::createFromString.

If SecurityOrigin::m_enforceFilePathSeparation is true, then
SecurityOrigin::toString() returns "null" for file URLs, and this
problem does not exist, and the workaround code is not ran.

This patch removes the workaround.

* src/WebStorageNamespaceImpl.cpp:
(WebKit::WebStorageNamespaceImpl::createStorageArea):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@92872 268f45cc-cd09-0410-ab3c-d52691b4dbfc

https://bugs.webkit.org/show_bug.cgi?id=65944
Replace QuotesData::operator==() with QuotesData::equal(const QuotesData*
, const QuotesData*) because QuotesData::operator==() were using non-standard
C++ code that is not supported by all compilers.

Patch by Yong Li <yoli@rim.com> on 2011-08-11
Reviewed by Alexey Proskuryakov.

No new tests because some existing test cases can show the issue,
for example, fast/css/content/content-quotes-01.html.

* rendering/RenderQuote.cpp:
(WebCore::RenderQuote::styleDidChange):
* rendering/style/QuotesData.cpp:
(WebCore::QuotesData::equal):
* rendering/style/QuotesData.h:
* rendering/style/RenderStyle.cpp:
(WebCore::RenderStyle::setQuotes):
* rendering/style/StyleRareInheritedData.cpp:
(WebCore::StyleRareInheritedData::operator==):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@92871 268f45cc-cd09-0410-ab3c-d52691b4dbfc

https://bugs.webkit.org/show_bug.cgi?id=66075

Source/WebCore:

Turn on regions and exclusions by default. Fix one stray ifdef that I missed.

Reviewed by James Robinson.

* css/CSSComputedStyleDeclaration.cpp:
(WebCore::contentToCSSValue):

LayoutTests:

Fix the broken test result for layer dumping, since at the moment flow threads don't collect
layers (this will be fixed eventually).

Unskip the regions and exclusions tests now that they are turned on by default.

Reviewed by James Robinson.

* fast/regions/webkit-flow-renderer-layer-expected.txt:
* platform/mac/Skipped:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@92870 268f45cc-cd09-0410-ab3c-d52691b4dbfc

[Qt] fast/dom/HTMLDocument/hasFocus.html fails
https://bugs.webkit.org/show_bug.cgi?id=66076

Unreviewed gardening.

* platform/qt/Skipped:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@92869 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Removed unused variable in StackBounds::initialize() to resolve
compiler warning when building on QNX.
https://bugs.webkit.org/show_bug.cgi?id=66072

Patch by Daniel Bates <dbates@rim.com> on 2011-08-11
Reviewed by Antonio Gomes.

* wtf/StackBounds.cpp:
(WTF::StackBounds::initialize):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@92868 268f45cc-cd09-0410-ab3c-d52691b4dbfc

[Chromium] Temporarily disable layer anti-aliasing on ChromeOS.
https://bugs.webkit.org/show_bug.cgi?id=65922

Patch by David Reveman <reveman@chromium.org> on 2011-08-11
Reviewed by James Robinson.

No new tests.

* platform/graphics/chromium/LayerTilerChromium.cpp:
(WebCore::LayerTilerChromium::draw):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@92867 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Merge expectations to mute flakiness.

* platform/chromium/test_expectations.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@92866 268f45cc-cd09-0410-ab3c-d52691b4dbfc

New expectation entries should be placed next to existing ones.
https://bugs.webkit.org/show_bug.cgi?id=66066

Reviewed by Adam Barth.

* Scripts/webkitpy/layout_tests/models/test_expectations.py: Added the notion of insertion point,
    which is computed as the last found reference to the same test.
* Scripts/webkitpy/layout_tests/models/test_expectations_unittest.py: Added and adjusted unit tests.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@92861 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Implementation of monotonically increasing clock on GTK
https://bugs.webkit.org/show_bug.cgi?id=62175

Patch by Devdatta Deshpande <pwjd73@motorola.com> on 2011-08-11
Reviewed by Martin Robinson.

* wtf/CurrentTime.cpp:
(WTF::monotonicallyIncreasingTime):
The default implementation of monotonicallyIncreasingTime only
guarantees the result to be non-decreasing.
If the system time is changed to past then default implementation will
still fail and WebCore timers will not fire.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@92859 268f45cc-cd09-0410-ab3c-d52691b4dbfc

[Qt] fast/dom/HTMLDocument/hasFocus.html failed
https://bugs.webkit.org/show_bug.cgi?id=62126

This test case passed under the latest revision, remove it from skip list.

Reviewed by Chang Shu.

* platform/qt/Skipped:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@92858 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Unreviewed, rolling out r92847.
http://trac.webkit.org/changeset/92847
https://bugs.webkit.org/show_bug.cgi?id=65986

Broke check-webkit-style

* Scripts/webkitpy/style/checker.py:
* Scripts/webkitpy/style/checker_unittest.py:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@92857 268f45cc-cd09-0410-ab3c-d52691b4dbfc

[Qt] PageSerializer is specific to MHTML, we should not compile it
https://bugs.webkit.org/show_bug.cgi?id=66054

Patch by Benjamin Poulain <benjamin@webkit.org> on 2011-08-11
Reviewed by Noam Rosenthal.

PageSerializer is specific to MHTML. Do not build PageSerializer unless MHTML is enabled.

* WebCore.pro:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@92856 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Web Inspector: [refactoring] make tabbed pane a view
https://bugs.webkit.org/show_bug.cgi?id=66060

Reviewed by Pavel Feldman.

* inspector/front-end/NetworkItemView.js:
(WebInspector.NetworkItemView):
(WebInspector.NetworkItemView.prototype.show):
(WebInspector.NetworkItemView.prototype._selectTab.get if):
(WebInspector.NetworkItemView.prototype._selectTab):
* inspector/front-end/TabbedPane.js:
(WebInspector.TabbedPane):
* inspector/front-end/inspector.html:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@92855 268f45cc-cd09-0410-ab3c-d52691b4dbfc

[GTK] create pot files in builddir, not (read only) srcdir
https://bugs.webkit.org/show_bug.cgi?id=66059

Reviewed by Gustavo Noronha Silva.

* GNUmakefile.am: write and read pot files from the build dir.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@92854 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Add the javascript core dynamic library for gtk-doc's introspection helper.

Patch by Gustavo Noronha Silva  <gns@gnome.org> on 2011-08-11
Reviewed by Xan Lopez.

* docs/GNUmakefile.am:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@92853 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2011-08-11  Xan Lopez  <xlopez@igalia.com>

        [GTK] Update NEWS and configure.ac for 1.5.2 release

        Reviewed by Gustavo Noronha.

        * configure.ac: update for release.

Source/WebKit/gtk:

2011-08-11  Xan Lopez  <xlopez@igalia.com>

        [GTK] Update NEWS and configure.ac for 1.5.2 release

        Reviewed by Gustavo Noronha.

        * NEWS: update for release.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@92852 268f45cc-cd09-0410-ab3c-d52691b4dbfc

[Qt/WK2] Add initial support for viewport meta tag
https://bugs.webkit.org/show_bug.cgi?id=65975

Reviewed by Benjamin Poulain.

Redirect the viewport arguments through to the view.

* UIProcess/API/qt/qdesktopwebview.cpp:
(QDesktopWebViewPrivate::didReceiveViewportArguments):
* UIProcess/API/qt/qdesktopwebview_p.h:
* UIProcess/API/qt/qtouchwebview.cpp:
(QTouchWebView::geometryChanged):
* UIProcess/qt/QtWebPageProxy.cpp:
(QtWebPageProxy::setViewportArguments):
* UIProcess/qt/QtWebPageProxy.h:
* UIProcess/qt/TouchViewInterface.cpp:
(WebKit::TouchViewInterface::updateViewportState):
(WebKit::TouchViewInterface::didReceiveViewportArguments):
* UIProcess/qt/TouchViewInterface.h:
(WebKit::TouchViewInterface::ViewportState::ViewportState):
* UIProcess/qt/ViewInterface.h:
* UIProcess/qt/qtouchwebpageproxy.cpp:
(QTouchWebPageProxy::QTouchWebPageProxy):
* UIProcess/qt/qtouchwebpageproxy.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@92851 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Web Inspector: source frame popover simetimes pops up after user switches from Scripts panel
https://bugs.webkit.org/show_bug.cgi?id=66057

Reviewed by Pavel Feldman.

* inspector/front-end/SourceFrame.js: remove popover timer when removing popover.
(WebInspector.SourceFrame.prototype._mouseDown):
(WebInspector.SourceFrame.prototype._hidePopup):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@92850 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Web Inspector: Scripts panel: display the current search match index in the toolbar.
https://bugs.webkit.org/show_bug.cgi?id=66048

Reviewed by Pavel Feldman.

* English.lproj/localizedStrings.js:
* inspector/front-end/ScriptsPanel.js:
(WebInspector.ScriptsPanel.prototype.performSearch.finishedCallback):
(WebInspector.ScriptsPanel.prototype.performSearch):
(WebInspector.ScriptsPanel.prototype.jumpToNextSearchResult):
(WebInspector.ScriptsPanel.prototype.jumpToPreviousSearchResult):
* inspector/front-end/SearchController.js:
(WebInspector.SearchController.prototype.updateSearchMatchesCount):
(WebInspector.SearchController.prototype.updateCurrentMatchIndex):
(WebInspector.SearchController.prototype.activePanelChanged.performPanelSearch):
(WebInspector.SearchController.prototype.activePanelChanged):
(WebInspector.SearchController.prototype._updateSearchMatchesCountAndCurrentMatchIndex):
(WebInspector.SearchController.prototype._performSearch):
* inspector/front-end/SourceFrame.js:
(WebInspector.SourceFrame.prototype.get currentSearchResultIndex):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@92849 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2011-08-11  Xan Lopez  <xlopez@igalia.com>

        [GTK] Add another missing file to GNUmakefile

        Reviewed by Gustavo Noronha.

        * GNUmakefile.list.am:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@92848 268f45cc-cd09-0410-ab3c-d52691b4dbfc

[Qt] autotests shouldn't require config.h in the style check
https://bugs.webkit.org/show_bug.cgi?id=65986

Remove the include rules from the style-check for Qt tests.
Since for the most part it's just a matter of using the
API provided by QtWebKit, enforcing the WebCore set of rules
only seems to make sense in the case of MIMESniffing.

Patch by Pierre Rossi <pierre.rossi@gmail.com> on 2011-08-11
Reviewed by Benjamin Poulain.

* Scripts/webkitpy/style/checker.py:
* Scripts/webkitpy/style/checker_unittest.py:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@92847 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2011-08-11  Xan Lopez  <xlopez@igalia.com>

        [GTK] Add missing files to GNUMakefile

        Reviewed by Gustavo Noronha.

        * GNUmakefile.am:
        * GNUmakefile.list.am:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@92846 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2011-08-11  Andrei Popescu  <andreip@google.com>

[Chromium] Rebaseline these tests on Leopard.

Unreviewed gardening.

* platform/chromium-mac-leopard/editing/pasteboard/paste-text-016-expected.png: Added
* platform/chromium-mac-leopard/editing/spelling/inline_spelling_markers-expected.png: Added

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@92845 268f45cc-cd09-0410-ab3c-d52691b4dbfc

WebKitTestRunner needs a more-complete implementation of eventSender
https://bugs.webkit.org/show_bug.cgi?id=42194

fast/encoding/mailto-always-utf-8.html needs eventSender
Move it to the common wk2 Skipped list from the qt-wk2.

Unreviewed gardening.

Patch by Zsolt Fehér <h490267@stud.u-szeged.hu> on 2011-08-11

* platform/qt-wk2/Skipped:
* platform/wk2/Skipped:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@92844 268f45cc-cd09-0410-ab3c-d52691b4dbfc

WebSocket: Use PassRefPtr<> in function arguments in WorkerThreadableWebSocketChannel.cpp
https://bugs.webkit.org/show_bug.cgi?id=66047

Reviewed by Kent Tamura.

It's safe to use PassRefPtr<> in these functions because:
- CrossThreadTask knows how to pass a RefPtr<> across threads, and
- Call sites do not pass the pointers to anywhere else.

No change in functionality, thus no new tests.

* websockets/WorkerThreadableWebSocketChannel.cpp:
(WebCore::workerContextDidSend):
(WebCore::workerContextDidGetBufferedAmount):
(WebCore::workerContextDidConnect):
(WebCore::workerContextDidReceiveMessage):
(WebCore::workerContextDidStartClosingHandshake):
(WebCore::workerContextDidClose):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@92843 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Web Inspector: [Mac] search as you type does not work with the new indent in the script selector.
https://bugs.webkit.org/show_bug.cgi?id=66039

Reviewed by Yury Semikhatsky.

Source/WebCore:

* inspector/front-end/ScriptsPanel.js:
(WebInspector.ScriptsPanel.prototype._addOptionToFilesSelect):
* inspector/front-end/inspector.css:
(select.status-bar-item):

LayoutTests:

* inspector/debugger/scripts-panel-expected.txt:
* platform/chromium/inspector/debugger/scripts-sorting-expected.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@92842 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Unreviewed build fix for Qt.

Make sure we build when Qt is namespaced.

* platform/graphics/gstreamer/PlatformVideoWindowPrivate.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@92841 268f45cc-cd09-0410-ab3c-d52691b4dbfc

[WK2] Implement waitForPolicyDelegate
https://bugs.webkit.org/show_bug.cgi?id=42330

Patch by Zsolt Fehér <h490267@stud.u-szeged.hu> on 2011-08-11
Reviewed by Csaba Osztrogonác.

Tools:

* WebKitTestRunner/InjectedBundle/Bindings/LayoutTestController.idl:
* WebKitTestRunner/InjectedBundle/InjectedBundlePage.cpp:
(WTR::navigationTypeToString):
(WTR::InjectedBundlePage::decidePolicyForNavigationAction):
* WebKitTestRunner/InjectedBundle/LayoutTestController.cpp:
(WTR::LayoutTestController::LayoutTestController):
(WTR::LayoutTestController::setCustomPolicyDelegate):
(WTR::LayoutTestController::waitForPolicyDelegate):
* WebKitTestRunner/InjectedBundle/LayoutTestController.h:
(WTR::LayoutTestController::isPolicyDelegateEnabled):
(WTR::LayoutTestController::isPolicyDelegatePermissive):

LayoutTests:

Unskip passing tests.

* platform/qt-wk2/Skipped:
* platform/wk2/Skipped:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@92839 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Web Inspector: use source-file-replaced instead of source-file-changed event.
https://bugs.webkit.org/show_bug.cgi?id=65551

Reviewed by Pavel Feldman.

Source/WebCore:

Test: inspector/debugger/source-frame-count.html

* inspector/front-end/DebuggerPresentationModel.js:
(WebInspector.DebuggerPresentationModel.prototype._addScript):
(WebInspector.DebuggerPresentationModel.prototype._uiSourceCodeReplaced):
* inspector/front-end/ScriptsPanel.js:
(WebInspector.ScriptsPanel.prototype._removeSourceFrame):
(WebInspector.ScriptsPanel.prototype._sourceFileReplaced):
* inspector/front-end/SourceFile.js:
(WebInspector.RawSourceCode):
(WebInspector.RawSourceCode.prototype.get uiSourceCode):
(WebInspector.RawSourceCode.prototype.reload):
* inspector/front-end/inspector.js:
(WebInspector._createPanels):

LayoutTests:

* inspector/debugger/scripts-panel-expected.txt:
* inspector/debugger/scripts-panel.html:
* inspector/debugger/source-file-expected.txt:
* inspector/debugger/source-file.html:
* inspector/debugger/source-frame-count-expected.txt: Added.
* inspector/debugger/source-frame-count.html: Added.
* platform/chromium/test_expectations.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@92838 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2011-08-11  Andrei Popescu  <andreip@google.com>

Master: Support CSS3 font feature properties
https://bugs.webkit.org/show_bug.cgi?id=63796

Unreviewed gardening.

css3/font-feature-settings-rendering.html is missing expectations. Looking at the
actual test output, it appears clearly wrong so we cannot rebaseline it either.
We'll therefore update the expectations file to mark this test as missing expected
results and update the bug to say it needs investigation and fixing.

* platform/chromium/test_expectations.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@92837 268f45cc-cd09-0410-ab3c-d52691b4dbfc

[Qt]Fix warnings after r92805.

Reviewed by Csaba Osztrogonác.

Control reached the end non-void functions.

* src/compiler/ossource_posix.cpp:
(OS_AllocTLSIndex):
(OS_SetTLSValue):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@92836 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Web Inspector: [REGRESSION] Resources panel search fails when search result is found in a resource used in several frames.
https://bugs.webkit.org/show_bug.cgi?id=66007

Reviewed by Pavel Feldman.

* inspector/front-end/ResourcesPanel.js:
(WebInspector.ResourcesPanel.prototype.performSearch.callback):
(WebInspector.ResourcesPanel.prototype.performSearch):
(WebInspector.FrameTreeElement.prototype.resourceByURL):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@92835 268f45cc-cd09-0410-ab3c-d52691b4dbfc

[GTK] Fix the build after r92792.

Rubber-stamped by Csaba Osztrogonác.

No new tests needed.

* GNUmakefile.list.am: add missing
   RenderFlowThread.{cpp|h}
   RenderRegion.{cpp|h}

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@92834 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Rebaseline these tests, as requested by morrita.

* platform/chromium-mac/editing/pasteboard/paste-text-016-expected.png: Added.
* platform/chromium-mac/editing/selection/move-by-character-005-expected.png: Added.
* platform/chromium-mac/editing/spelling/inline_spelling_markers-expected.png: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@92833 268f45cc-cd09-0410-ab3c-d52691b4dbfc

rebaseline-expectations should use the release bots, not the debug
bots.  The debug bots don't all run all the tests.  Also, delay
updating test_expectations a long as possible to avoid clearing out
lines that we'll need for other ports.

* Scripts/webkitpy/layout_tests/port/builders.py:
* Scripts/webkitpy/tool/commands/rebaseline.py:
* Scripts/webkitpy/tool/commands/rebaseline_unittest.py:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@92832 268f45cc-cd09-0410-ab3c-d52691b4dbfc

[Qt] Enable WebGL by default for QtWebKit
https://bugs.webkit.org/show_bug.cgi?id=65998

disable failing tests after https://trac.webkit.org/changeset/92805

* platform/qt/Skipped: Skip fast/canvas/webgl directory

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@92831 268f45cc-cd09-0410-ab3c-d52691b4dbfc

[Qt] Fix WebGL on Windows after r92805
https://bugs.webkit.org/show_bug.cgi?id=66042

WEBGL is temporarly disabled, because it broked the bots.

* features.pri:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@92830 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Fix Snow Leopard expectation after r92518.

* platform/mac-snowleopard/fast/loader/file-url-mimetypes-expected.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@92829 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Test how SpeechInputResultList behaves with not enough arguments
https://bugs.webkit.org/show_bug.cgi?id=65812

Reviewed by Sam Weinig.

SpeechInputResultList is only exposed via the x-webkit-speech
attribute, which means there is little compatibility risk in aligning
our too-few-arguments behavior with the specification.  There's a
question about whether all these "fake array" objects should work
consistently, but we'll probably want the "fake array" objects we
create in the future to match the spec, so even if we're consistent
now, we'll end up not being consistent in the future.

* fast/speech/speech-input-result-list-not-enough-arguments-expected.txt: Added.
* fast/speech/speech-input-result-list-not-enough-arguments.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@92828 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Web Inspector: do not evaluate watch expressions on load.
https://bugs.webkit.org/show_bug.cgi?id=66002

Patch by Pavel Feldman <pfeldman@google.com> on 2011-08-11
Reviewed by Yury Semikhatsky.

* inspector/front-end/ScriptsPanel.js:
(WebInspector.ScriptsPanel.prototype.show):
(WebInspector.ScriptsPanel.prototype.hide):
(WebInspector.ScriptsPanel.prototype.reset):
* inspector/front-end/WatchExpressionsSidebarPane.js:
(WebInspector.WatchExpressionsSidebarPane):
(WebInspector.WatchExpressionsSidebarPane.prototype.hide):
(WebInspector.WatchExpressionsSidebarPane.prototype.reset):
(WebInspector.WatchExpressionsSidebarPane.prototype.refreshExpressions):
(WebInspector.WatchExpressionsSidebarPane.prototype._refreshExpressionsIfNeeded):
(WebInspector.WatchExpressionsSidebarPane.prototype._refreshButtonClicked):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@92827 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Build fix on Qt Windows 32-bit Release/Debug after r92805.

Unreviewed.

* src/compiler/ossource_posix.cpp:
(OS_FreeTLSIndex):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@92826 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Skip tests added by r92769 on WebKit2 because TestRunner doesn't support
drag and drop by eventSender yet.

* platform/wk2/Skipped:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@92825 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Unreviewed pixel expectations update. New expectations now have
bad grammar markers. (The settings has been enabled since r82159.)

* platform/mac/editing/pasteboard/paste-text-016-expected.png:
* platform/mac/editing/selection/move-by-character-005-expected.png:
* platform/mac/editing/spelling/inline_spelling_markers-expected.png:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@92824 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Copying can result in span around block elements on the clipboard
https://bugs.webkit.org/show_bug.cgi?id=34564

Reviewed by Tony Chang.

Source/WebCore:

Completely overhauled the way WebKit preserves style in copy and paste. Instead of wrapping the entire
serialized contents by a Apple style span, WebKit now adds inline style to the top level elements,
wrap top level text nodes by a style span.

* editing/EditingStyle.cpp:
(WebCore::EditingStyle::collapseTextDecorationProperties): Remove text-decoration property when the value
of -webkit-text-decorations-in-effect is none.
(WebCore::EditingStyle::removeStyleFromRulesAndContext): Since display: inline and float: none are now
added on copy, remove these properties on paste.
(WebCore::EditingStyle::removePropertiesInElementDefaultStyle): Takes Element* instead of StyledElement*.
(WebCore::EditingStyle::forceInline): Added.
(WebCore::getPropertiesNotIn): Remove properties only when the base style has them.
* editing/EditingStyle.h:
* editing/markup.cpp:
(WebCore::StyledMarkupAccumulator::shouldApplyWrappingStyle): Added.
(WebCore::StyledMarkupAccumulator::StyledMarkupAccumulator): Takes highestNodeToBeSerialized.
(WebCore::StyledMarkupAccumulator::wrapWithStyleNode): Calls appendStyleNodeOpenTag and styleNodeCloseTag.
(WebCore::StyledMarkupAccumulator::appendStyleNodeOpenTag): Extracted from wrapWithStyleNode.
(WebCore::StyledMarkupAccumulator::styleNodeCloseTag): Ditto.
(WebCore::StyledMarkupAccumulator::appendText): Wraps text node with a style span if needed.
Set display: inline and float: none so that it won't be converted to a block on paste side.
(WebCore::StyledMarkupAccumulator::appendElement): Add wrapping style if appropriate; Remove any properties
that are overridden by default style and any style that may conflict with the computed style of node to
avoid modifying the appearance of the serialized nodes.
(WebCore::StyledMarkupAccumulator::serializeNodes): Compute wrapping style; copies of this style are
modified as needed when serializing top-level elements or text nodes. We call traverseNodesForSerialization
with NodeTraversalMode set to DoNotEmitString first to compute the highest node to be serialized. The second
call to the function actually serialize the nodes.
(WebCore::StyledMarkupAccumulator::traverseNodesForSerialization): Extracted from serializeNodes.
Outputs string only if NodeTraversalMode is set to EmitString.
(WebCore::createMarkup): No longer adds wrapping spans.

LayoutTests:

This patch overhauled the way we preserve styles when copying and pasting HTML contents. Many tests progressed
and lost wrapping spans while others are observing different markup on the clipboard as expected.

* editing/deleting/deleting-line-break-preserves-underline-color-expected.txt: span's style attribute now has
a trailing space.
* editing/execCommand/insert-list-with-noneditable-content-expected.txt: LightGray is now rgb(211, 211, 211).
* editing/pasteboard/data-transfer-items-expected.txt: Different serialization.
* editing/pasteboard/onpaste-text-html-expected.txt: Ditto.
* editing/pasteboard/paste-4039777-fix-expected.txt: No longer nests ul erroneously or aligns the inner list
to the right.
* editing/pasteboard/paste-code-in-pre-expected.txt: No longer produces erroneous wrapping span.
* editing/pasteboard/paste-pre-001-expected.txt: Ditto.
* editing/pasteboard/paste-pre-002-expected.txt: Ditto.
* editing/pasteboard/paste-text-012-expected.txt: Ditto.
* editing/pasteboard/paste-text-with-style-4-expected.txt: Ditto.
* fast/events/ondrop-text-html-expected.txt: Different serialization.
* platform/chromium-win/editing/pasteboard/paste-code-in-pre-expected.txt: Removed.
* platform/chromium-win/editing/pasteboard/paste-pre-001-expected.txt: Removed.
* platform/chromium-win/editing/pasteboard/paste-pre-002-expected.txt: Removed.
* platform/qt-mac/editing/pasteboard/paste-code-in-pre-expected.txt: Removed.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@92823 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Skip a test added by r92769 on Windows port since drag and drop won't work well on Windows port.

* platform/win/Skipped:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@92822 268f45cc-cd09-0410-ab3c-d52691b4dbfc

GTK and Qt rebaselines after r92792.

* platform/gtk/fast/css/getComputedStyle/computed-style-without-renderer-expected.txt:
* platform/qt/fast/css/getComputedStyle/computed-style-without-renderer-expected.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@92821 268f45cc-cd09-0410-ab3c-d52691b4dbfc

r92792 inadvertently removed PlatformSpecificScheduler for all Snow Leopard bots.
Revert it partially so that the remaining Snow Leopard bots can fulfill their duty.

* BuildSlaveSupport/build.webkit.org-config/config.json:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@92820 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Move <input type=date> tests to fast/forms/date/
https://bugs.webkit.org/show_bug.cgi?id=66029

Reviewed by Hajime Morita.

* fast/forms/ValidityState-rangeOverflow-expected.txt: Removed "date" part.
* fast/forms/ValidityState-rangeUnderflow-expected.txt: ditto.
* fast/forms/ValidityState-stepMismatch-expected.txt: ditto.
* fast/forms/ValidityState-typeMismatch-date.html: Removed.
* fast/forms/date: Added.
* fast/forms/date-input-visible-strings-expected.txt: Removed.
* fast/forms/date-input-visible-strings.html: Removed.
* fast/forms/date/ValidityState-rangeOverflow-date-expected.txt: Copied from LayoutTests/fast/forms/ValidityState-rangeOverflow-expected.txt.
* fast/forms/date/ValidityState-rangeOverflow-date.html: Copied from LayoutTests/fast/forms/script-tests/ValidityState-rangeOverflow.js.
* fast/forms/date/ValidityState-rangeUnderflow-date-expected.txt: Copied from LayoutTests/fast/forms/ValidityState-rangeUnderflow-expected.txt.
* fast/forms/date/ValidityState-rangeUnderflow-date.html: Copied from LayoutTests/fast/forms/script-tests/ValidityState-rangeUnderflow.js.
* fast/forms/date/ValidityState-stepMismatch-date-expected.txt: Copied from LayoutTests/fast/forms/ValidityState-stepMismatch-expected.txt.
* fast/forms/date/ValidityState-stepMismatch-date.html: Copied from LayoutTests/fast/forms/script-tests/ValidityState-stepMismatch.js.
* fast/forms/date/ValidityState-typeMismatch-date-expected.txt: Copied from LayoutTests/fast/forms/ValidityState-typeMismatch-date-expected.txt.
* fast/forms/date/ValidityState-typeMismatch-date.html: Copied from LayoutTests/fast/forms/script-tests/ValidityState-typeMismatch-date.js.
* fast/forms/date/date-input-type-expected.txt: Copied from LayoutTests/fast/forms/week/week-input-type-expected.txt.
* fast/forms/date/date-input-type.html: Copied from LayoutTests/fast/forms/week/week-input-type.html.
* fast/forms/date/date-input-visible-strings-expected.txt: Copied from LayoutTests/fast/forms/date-input-visible-strings-expected.txt.
* fast/forms/date/date-input-visible-strings.html: Copied from LayoutTests/fast/forms/date-input-visible-strings.html.
* fast/forms/date/date-stepup-stepdown-expected.txt: Copied from LayoutTests/fast/forms/input-stepup-stepdown-expected.txt.
* fast/forms/date/date-stepup-stepdown-from-renderer-expected.txt: Copied from LayoutTests/fast/forms/input-stepup-stepdown-from-renderer-expected.txt.
* fast/forms/date/date-stepup-stepdown-from-renderer.html: Copied from LayoutTests/fast/forms/script-tests/input-stepup-stepdown-from-renderer.js.
* fast/forms/date/date-stepup-stepdown.html: Copied from LayoutTests/fast/forms/script-tests/input-stepup-stepdown.js.
* fast/forms/date/input-valueasdate-date-expected.txt: Copied from LayoutTests/fast/forms/input-valueasdate-date-expected.txt.
* fast/forms/date/input-valueasdate-date.html: Copied from LayoutTests/fast/forms/script-tests/input-valueasdate-date.js.
* fast/forms/date/input-valueasnumber-date-expected.txt: Copied from LayoutTests/fast/forms/input-valueasnumber-date-expected.txt.
* fast/forms/date/input-valueasnumber-date.html: Copied from LayoutTests/fast/forms/script-tests/input-valueasnumber-date.js.
* fast/forms/input-stepup-stepdown-expected.txt: Removed "date" part.
* fast/forms/input-stepup-stepdown-from-renderer-expected.txt: ditto.
* fast/forms/input-type-change3-expected.txt: ditto.
* fast/forms/input-valueasdate-date-expected.txt: Removed.
* fast/forms/input-valueasdate-date.html: Removed.
* fast/forms/input-valueasnumber-date-expected.txt: Removed.
* fast/forms/input-valueasnumber-date.html: Removed.
* fast/forms/script-tests/ValidityState-rangeOverflow.js: Removed "date" part.
* fast/forms/script-tests/ValidityState-rangeUnderflow.js: ditto.
* fast/forms/script-tests/ValidityState-stepMismatch.js: ditto.
* fast/forms/script-tests/input-stepup-stepdown-from-renderer.js: ditto.
* fast/forms/script-tests/input-stepup-stepdown.js: ditto.
* fast/forms/script-tests/input-type-change3.js: ditto.
* fast/forms/script-tests/input-valueasdate-date.js: Removed.
* fast/forms/script-tests/input-valueasnumber-date.js: Removed.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@92819 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Test how TouchList behaves with too few arguments
https://bugs.webkit.org/show_bug.cgi?id=65815

Reviewed by Sam Weinig.

This API is very new and the compatibility cost of making this API
match the specification is minimal.  We should align with the
specification and (future) implementations in other browsers by making
this argument required.

* fast/events/touch/document-create-touch-list-expected.txt:
* fast/events/touch/script-tests/document-create-touch-list.js:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@92818 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Test how AudioBufferSourceNode reacts when given too few arguments
https://bugs.webkit.org/show_bug.cgi?id=65816

Reviewed by Sam Weinig.

The WebAudio API is extremely new.  It has only been available as an
experimental API in a single browser.  The compatibility cost of
tightening up these APIs is tiny and these APIs make very little sense
to call with too-few arguments.

* webaudio/audionode-expected.txt:
* webaudio/audionode.html:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@92817 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Add tests of optional arguments for Geolocation
https://bugs.webkit.org/show_bug.cgi?id=65810

Reviewed by Sam Weinig.

Source/WebCore:

Geolocation's treatment of not-enough-arguments almost matches the
spec.  getCurrentPosition and watchPosition work properly, but, because
they are custom, we should use the prettier form of the [Optional]
attribute.

Before this series of patches, clearWatch argument was optional, but
that doesn't match the spec or other browsers.  Calling this function
with zero arguments is pretty non-sensical, so there shouldn't be much
compat risk to tightening up our behavior here.  Matching other
browsers and the spec seems like the bigger win.  If we run into compat
problems, we can re-evaluate this decision.

Test: fast/dom/Geolocation/not-enough-arguments.html

* page/Geolocation.idl:

LayoutTests:

Test what happens when calling geolocation APIs with too few arguments.

* fast/dom/Geolocation/not-enough-arguments-expected.txt: Added.
* fast/dom/Geolocation/not-enough-arguments.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@92816 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Skip some more tests on Lion.

* platform/mac-lion/Skipped:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@92813 268f45cc-cd09-0410-ab3c-d52691b4dbfc

SVGAElement check for internal non-view anchors is inside ENABLE(SVG_ANIMATION)
https://bugs.webkit.org/show_bug.cgi?id=66026

Reviewed by Rob Buis.

Move only the SMIL-related code inside the ENABLE(SVG_ANIMATION) block.

No new tests, as it depends on the enabled features.

* svg/SVGAElement.cpp:
(WebCore::SVGAElement::defaultEventHandler):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@92812 268f45cc-cd09-0410-ab3c-d52691b4dbfc

[chromium] Update WebScrollbar so that it works with overlay scrollbars on Lion
https://bugs.webkit.org/show_bug.cgi?id=66008

Reviewed by James Robinson.

* WebKit.gyp:
* public/WebScrollbar.h:
* public/WebScrollbarClient.h:
* src/ScrollbarGroup.cpp: Added.
(WebKit::ScrollbarGroup::ScrollbarGroup):
(WebKit::ScrollbarGroup::~ScrollbarGroup):
(WebKit::ScrollbarGroup::scrollbarCreated):
(WebKit::ScrollbarGroup::scrollbarDestroyed):
(WebKit::ScrollbarGroup::setLastMousePosition):
(WebKit::ScrollbarGroup::scrollSize):
(WebKit::ScrollbarGroup::scrollPosition):
(WebKit::ScrollbarGroup::setScrollOffset):
(WebKit::ScrollbarGroup::invalidateScrollbarRect):
(WebKit::ScrollbarGroup::invalidateScrollCornerRect):
(WebKit::ScrollbarGroup::isActive):
(WebKit::ScrollbarGroup::enclosingScrollableArea):
(WebKit::ScrollbarGroup::isScrollCornerVisible):
(WebKit::ScrollbarGroup::getTickmarks):
(WebKit::ScrollbarGroup::convertFromContainingViewToScrollbar):
(WebKit::ScrollbarGroup::horizontalScrollbar):
(WebKit::ScrollbarGroup::verticalScrollbar):
(WebKit::ScrollbarGroup::minimumScrollPosition):
(WebKit::ScrollbarGroup::maximumScrollPosition):
(WebKit::ScrollbarGroup::visibleHeight):
(WebKit::ScrollbarGroup::visibleWidth):
(WebKit::ScrollbarGroup::contentsSize):
(WebKit::ScrollbarGroup::overhangAmount):
(WebKit::ScrollbarGroup::currentMousePosition):
(WebKit::ScrollbarGroup::shouldSuspendScrollAnimations):
(WebKit::ScrollbarGroup::scrollbarStyleChanged):
(WebKit::ScrollbarGroup::isOnActivePage):
(WebKit::ScrollbarGroup::disconnectFromPage):
* src/ScrollbarGroup.h: Added.
(WebKit::ScrollbarGroup::scrollCornerRect):
* src/WebPluginContainerImpl.cpp:
(WebKit::WebPluginContainerImpl::paint):
(WebKit::WebPluginContainerImpl::reportGeometry):
(WebKit::WebPluginContainerImpl::scrollbarGroup):
(WebKit::WebPluginContainerImpl::willStartLiveResize):
(WebKit::WebPluginContainerImpl::willEndLiveResize):
(WebKit::WebPluginContainerImpl::handleMouseEvent):
* src/WebPluginContainerImpl.h:
* src/WebScrollbarImpl.cpp:
(WebKit::WebScrollbar::createForPlugin):
(WebKit::WebScrollbarImpl::WebScrollbarImpl):
(WebKit::WebScrollbarImpl::~WebScrollbarImpl):
(WebKit::WebScrollbarImpl::setScrollOffset):
(WebKit::WebScrollbarImpl::invalidateScrollbarRect):
(WebKit::WebScrollbarImpl::getTickmarks):
(WebKit::WebScrollbarImpl::convertFromContainingViewToScrollbar):
(WebKit::WebScrollbarImpl::scrollbarStyleChanged):
(WebKit::WebScrollbarImpl::isOverlay):
(WebKit::WebScrollbarImpl::setValue):
(WebKit::WebScrollbarImpl::scroll):
(WebKit::WebScrollbarImpl::onMouseMove):
(WebKit::WebScrollbarImpl::onMouseLeave):
(WebKit::WebScrollbarImpl::onMouseWheel):
(WebKit::WebScrollbarImpl::onKeyDown):
* src/WebScrollbarImpl.h:
(WebKit::WebScrollbarImpl::scrollOffset):
(WebKit::WebScrollbarImpl::scrollbar):
* src/WebViewImpl.cpp:
(WebKit::WebViewImpl::willStartLiveResize):
(WebKit::WebViewImpl::willEndLiveResize):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@92811 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Removed some incorrect code that was dead.

Reviewed by Oliver Hunt.

clearSingleTransition() wasn't resetting m_data. Luckily,
no one cares, because its caller was unused. Removed both.

* runtime/Structure.cpp:
* runtime/StructureTransitionTable.h:
(JSC::StructureTransitionTable::~StructureTransitionTable):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@92810 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Crash when clicking an SVG <a> link to the local document
https://bugs.webkit.org/show_bug.cgi?id=66019
<rdar://problem/9933311>

Reviewed by Rob Buis.

Source/WebCore:

Check if the target element of an internal link is valid before
making use of it.

Test: svg/custom/click-internal-anchor-with-use-crash.xhtml

* svg/SVGAElement.cpp:
(WebCore::SVGAElement::defaultEventHandler):

LayoutTests:

Add a test which programmatically clicks a link referring to "#",
and should crash between r89745 and this commit.

* svg/custom/click-internal-anchor-with-use-crash-expected.txt: Added.
* svg/custom/click-internal-anchor-with-use-crash.xhtml: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@92809 268f45cc-cd09-0410-ab3c-d52691b4dbfc

[EFL] Return from _ewk_view_smart_add if smart data can not be allocated
https://bugs.webkit.org/show_bug.cgi?id=65853

Patch by Grzegorz Czajkowski <g.czajkowski@samsung.com> on 2011-08-10
Reviewed by Kenneth Rohde Christiansen.

It prevents the crash while allocating a memory for view's smart data.
Others checking are needed for support single and tiled backing store.
As a result an empty window will be displayed.

* ewk/ewk_view.cpp:
(_ewk_view_smart_add):
(_ewk_view_smart_resize):
(_ewk_view_smart_move):
(_ewk_view_smart_show):
(_ewk_view_smart_hide):
* ewk/ewk_view_single.c:
(_ewk_view_single_smart_add):
(_ewk_view_single_smart_resize):
* ewk/ewk_view_tiled.c:
(_ewk_view_tiled_smart_add):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@92808 268f45cc-cd09-0410-ab3c-d52691b4dbfc

JSC should always throw when function arg list is too long
https://bugs.webkit.org/show_bug.cgi?id=65869

Patch by Mark Hahnenberg <mhahnenberg@apple.com> on 2011-08-10
Reviewed by Oliver Hunt.

Previous change forgot to update some of the tests in fast/js/function-apply.html so we're doing it now.

* fast/js/function-apply-expected.txt:
* fast/js/script-tests/function-apply.js:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@92807 268f45cc-cd09-0410-ab3c-d52691b4dbfc

WebViewImpl::selectionRange should return false if there is no selection
https://bugs.webkit.org/show_bug.cgi?id=66012

Patch by Varun Jain <varunjain@google.com> on 2011-08-10
Reviewed by Darin Fisher.

*  Source/WebKit/chromium/src/WebViewImpl.cpp:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@92806 268f45cc-cd09-0410-ab3c-d52691b4dbfc

[Qt] Enable WebGL by default for QtWebKit
https://bugs.webkit.org/show_bug.cgi?id=65998

Reviewed by Benjamin Poulain.

Make ENABLE_WEBGL true by default when Qt is configured with OpenGL.

No new tests. Build change only.

* features.pri:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@92805 268f45cc-cd09-0410-ab3c-d52691b4dbfc

REGRESSION(r92670-r92744): WebKit crashes when opening Gmail
https://bugs.webkit.org/show_bug.cgi?id=66010

Reviewed by Oliver Hunt.

Made sure that Construct calls use() on the this argument.

* dfg/DFGJITCodeGenerator.cpp:
(JSC::DFG::JITCodeGenerator::emitCall):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@92804 268f45cc-cd09-0410-ab3c-d52691b4dbfc

new-run-webkit-test's WinPort has no fallback logic
https://bugs.webkit.org/show_bug.cgi?id=64486

Unreviewed.

Turns out that we've long had a subtle bug whereby if you didn't
already have DumpRenderTree built, we might end up running unsupported tests
because the DumpRenderTree --print-supported-features check would silently fail.

When I moved the call from using popen to using Executive.run_command, this silent
failure became noisy.  (Breaking some bots.)

This re-orders our steps so that we always build DRT before collecting expectations,
as well as adds a warning message when DumpRenderTree can't be found.

* Scripts/webkitpy/layout_tests/port/webkit.py:
* Scripts/webkitpy/layout_tests/run_webkit_tests.py:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@92803 268f45cc-cd09-0410-ab3c-d52691b4dbfc

[Chromium] Decouple implementation of allowFileSystem, openFileSystem and allowDatabase from WebWorkerBase.
https://bugs.webkit.org/show_bug.cgi?id=65997.

This patch moves implementation of allowFileSystem, openFileSystem and allowDatabase from
WebWorkerBase to respectively LocalFileSystemChromium and DatabaseObserver,
parameterizing them with relevant data from WebWorker.

Reviewed by Jian Li.

* src/DatabaseObserver.cpp: Move allowDatabase from WebWorkerBase and update the caller.
(WebKit::AllowDatabaseMainThreadBridge::create):
(WebKit::AllowDatabaseMainThreadBridge::cancel):
(WebKit::AllowDatabaseMainThreadBridge::result):
(WebKit::AllowDatabaseMainThreadBridge::signalCompleted):
(WebKit::AllowDatabaseMainThreadBridge::AllowDatabaseMainThreadBridge):
(WebKit::AllowDatabaseMainThreadBridge::allowDatabaseTask):
(WebKit::AllowDatabaseMainThreadBridge::didComplete):
(WebKit::allowDatabaseForWorker):
(WebCore::DatabaseObserver::canEstablishDatabase):
* src/LocalFileSystemChromium.cpp: Move allowFileSystem and openFileSystem from WebWorkerBase and update the caller.
(WebCore::openFileSystemHelper):
* src/WebWorkerBase.cpp: Move allowFileSystem, openFileSystem and allowDatabase to LocalFileSystemChromium and DatabaseObserver resp.
* src/WebWorkerBase.h:
(WebKit::WebWorkerBase::webView):
* src/WorkerFileSystemCallbacksBridge.cpp: Generalized WorkerFileSystemCallbacksBridge to work on WorkerLoaderProxy, not on WebWorkerBase.
(WebKit::WorkerFileSystemCallbacksBridge::stop):
(WebKit::WorkerFileSystemCallbacksBridge::WorkerFileSystemCallbacksBridge):
(WebKit::WorkerFileSystemCallbacksBridge::dispatchTaskToMainThread):
(WebKit::WorkerFileSystemCallbacksBridge::mayPostTaskToWorker):
* src/WorkerFileSystemCallbacksBridge.h:
(WebKit::WorkerFileSystemCallbacksBridge::create):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@92802 268f45cc-cd09-0410-ab3c-d52691b4dbfc

[wx] Unreviewed build fix. Do not build LocaleToScriptMappingICU.cpp for now as ICU on Leopard is too old to build this.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@92801 268f45cc-cd09-0410-ab3c-d52691b4dbfc

new-run-webkit-test's WinPort has no fallback logic
https://bugs.webkit.org/show_bug.cgi?id=64486

Unreviewed.  Fix test-webkitpy under Python 2.5.

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

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@92800 268f45cc-cd09-0410-ab3c-d52691b4dbfc

[Qt][WK2] MiniBrowser is firing twice the QDesktopWebView::mousePressEvent method
https://bugs.webkit.org/show_bug.cgi?id=65875

Patch by Benjamin Poulain <benjamin@webkit.org> on 2011-08-10
Reviewed by Kenneth Rohde Christiansen.

For some reason, QWidget send fake mouse event on some platform, under certains conditions.

This causes mouse events to be sent twice in some cases. We work around the problem
by defining the first touch point to be a primary touch point (case which hit some
weird condition in QWidget, skipping the generation of fake mouse events).

* MiniBrowser/qt/MiniBrowserApplication.cpp:
(MiniBrowserApplication::notify):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@92799 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Source/WebCore: Check that we do not need layout before trying to dirty
m_originatingLine for our floats.
https://bugs.webkit.org/show_bug.cgi?id=65938

Reviewed by Dave Hyatt.

Test: fast/block/float/float-originating-line-deleted-crash.html

* rendering/RenderBlock.cpp:
(WebCore::RenderBlock::removeFloatingObject):
(WebCore::RenderBlock::clearFloats):

LayoutTests: Tests that we do not crash when trying to access a removed
m_originatingLine for a float.
https://bugs.webkit.org/show_bug.cgi?id=65938

Reviewed by Dave Hyatt.

* fast/block/float/float-originating-line-deleted-crash-expected.txt: Added.
* fast/block/float/float-originating-line-deleted-crash.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@92798 268f45cc-cd09-0410-ab3c-d52691b4dbfc

JSC should always throw when function arg list is too long
https://bugs.webkit.org/show_bug.cgi?id=65869

Patch by Mark Hahnenberg <mhahnenberg@apple.com> on 2011-08-10
Reviewed by Oliver Hunt.

Source/JavaScriptCore:

Changed the behavior of the interpreter and JIT to throw an exception
when too many arguments are passed rather than truncating the list.  Added
a new method to create a "Too many arguments." exception used by this
new functionality.

* interpreter/Interpreter.cpp:
(JSC::Interpreter::privateExecute):
* jit/JITStubs.cpp:
(JSC::DEFINE_STUB_FUNCTION):
* runtime/ExceptionHelpers.cpp:
(JSC::createTooManyParamsError):
* runtime/ExceptionHelpers.h:

LayoutTests:

Changed the tests to make sure passing too many function arguments throws an exception
rather than truncating the arguments.

* fast/js/function-apply-many-args-expected.txt:
* fast/js/script-tests/function-apply-many-args.js:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@92797 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Add missing regions/exclusions-related files to EFL.

* CMakeLists.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@92796 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Add missing regions/exclusions-related files to Chromium.

* WebCore.gypi:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@92795 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Fix Qt build bustage. Add missing regions/exclusions-related files to their project.

* WebCore.pro:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@92794 268f45cc-cd09-0410-ab3c-d52691b4dbfc