[WebKit-https.git] / Source / WebCore / ChangeLog

2015-04-30  Oliver Hunt  <oliver@apple.com>

        DOM bindings should not be using a reference type to point to a temporary object
        https://bugs.webkit.org/show_bug.cgi?id=144474

        Reviewed by Beth Dakin.

        The DOM bindings will currently try and use a local reference to point
        to a temporary object. This currently works as a by product of the compiler's
        stack layout. This patch removes the dependency on undefined behaviour
        by ensuring that we use a value rather than reference type.

        * bindings/scripts/CodeGeneratorJS.pm:
        (GenerateParametersCheck):
        (GetNativeTypeForCallbacks):

2015-04-30  Brady Eidson  <beidson@apple.com>

        Build fix after r183646 for less enlightened platforms.

        Unreviewed.

        * Modules/webdatabase/DatabaseBackendBase.cpp:
        (WebCore::fullyQualifiedInfoTableName): Windows doesn’t have stpcpy :(

2015-04-30  Brady Eidson  <beidson@apple.com>

        Javascript using WebSQL can create their own WebKit info table.
        <rdar://problem/20688792> and https://bugs.webkit.org/show_bug.cgi?id=144466

        Reviewed by Alex Christensen.

        Test: storage/websql/alter-to-info-table.html

        * Modules/webdatabase/DatabaseBackendBase.cpp:
        (WebCore::DatabaseBackendBase::databaseInfoTableName): Return the info table name.
        (WebCore::fullyQualifiedInfoTableName): Append "main." to the info table name.
        (WebCore::DatabaseBackendBase::DatabaseBackendBase): Use the fully qualified name.
        (WebCore::DatabaseBackendBase::performOpenAndVerify): Ditto.
        (WebCore::DatabaseBackendBase::getVersionFromDatabase): Ditto.
        (WebCore::DatabaseBackendBase::setVersionInDatabase): Ditto.

2015-04-30  Beth Dakin  <bdakin@apple.com>

        Should choose UIScrollView indicatorStyle based on the document background color
        https://bugs.webkit.org/show_bug.cgi?id=144473
        -and corresponding-
        rdar://problem/19897699

        Reviewed by Simon Fraser.

        Export this function.
        * platform/graphics/Color.h:

2015-04-30  Brent Fulgham  <bfulgham@apple.com>

        [Win] Allow WebKit to build without ANGLE support
        https://bugs.webkit.org/show_bug.cgi?id=144459
        <rdar://problem/20707307>

        Reviewed by Dean Jackson.

        * platform/graphics/GLContext.cpp: Use the ENABLE(GRAPHICS_CONTEXT_3D)
        macro to avoid compiling this code if not using WEBGL/3D contexts.

2015-04-29  David Hyatt  <hyatt@apple.com>

        Avoid containingBlock() calls when no writing mode flipping is needed.
        https://bugs.webkit.org/show_bug.cgi?id=144407

        Reviewed by Simon Fraser.

        Add a bool to RenderView that indicates whether or not any flipped blocks have been
        added to the view. Once tainted, the view just stays dirty forever. If no flipped
        blocks are ever seen, we can then optimize away calls to containingBlock().

        The motivation for this patch is to improve layer position updating, which makes many
        calls to topLeftLocationOffset(), one of the functions that can be optimized by this
        change.

        * rendering/RenderBox.cpp:
        (WebCore::RenderBox::layoutOverflowRectForPropagation):
        * rendering/RenderBoxModelObject.cpp:
        (WebCore::RenderBoxModelObject::updateFromStyle):
        * rendering/RenderLayer.cpp:
        (WebCore::RenderLayer::calculateClipRects):
        * rendering/RenderLineBoxList.cpp:
        (WebCore::RenderLineBoxList::rangeIntersectsRect):
        * rendering/RenderView.cpp:
        (WebCore::RenderView::RenderView):
        * rendering/RenderView.h:

2015-04-29  Jer Noble  <jer.noble@apple.com>

        Make GenericTaskQueue even more generic (and usable inside platform/)
        https://bugs.webkit.org/show_bug.cgi?id=144414

        Reviewed by Eric Carlson.

        Templatize GenericTaskQueue so that it can be used from within platform/.

        * CMakeLists.txt:
        * WebCore.vcxproj/WebCore.vcxproj:
        * WebCore.vcxproj/WebCore.vcxproj.filters:
        * WebCore.xcodeproj/project.pbxproj:
        * dom/GenericTaskQueue.cpp: Removed.
        * html/HTMLMediaElement.h:
        * platform/GenericTaskQueue.h: Renamed from Source/WebCore/dom/GenericTaskQueue.h.
        (WebCore::TaskDispatcher::TaskDispatcher): Added default templatized class which just calls the parameter's postTask().
        (WebCore::TaskDispatcher::postTask): Call context's postTask().
        (WebCore::TaskDispatcher<Timer>::TaskDispatcher): Add a timer version which does not require a context.
        (WebCore::TaskDispatcher<Timer>::postTask): Set the timer.
        (WebCore::TaskDispatcher<Timer>::timerFired): Call the task.
        (WebCore::GenericTaskQueue::GenericTaskQueue): Moved from .cpp.
        (WebCore::GenericTaskQueue::enqueueTask): Ditto.
        (WebCore::GenericTaskQueue::close): Ditto.
        (WebCore::GenericTaskQueue::cancelAllTasks): Ditto.
        (WebCore::GenericTaskQueue::hasPendingTasks): Ditto.

2015-04-29  Joseph Pecoraro  <pecoraro@apple.com>

        Unreviewed build fix. WebCore has a weak symbol.

        WebCore has a weak external symbol in it. A common cause of weak external
        symbols is when an inline function is listed in the linker export file.
        ERROR: symbol __ZNK7WebCore4Page25expectsWheelEventTriggersEv

        * page/Page.cpp:
        (WebCore::Page::expectsWheelEventTriggers):
        * page/Page.h:
        (WebCore::Page::expectsWheelEventTriggers): Deleted.

2015-04-29  Simon Fraser  <simon.fraser@apple.com>

        Crash at WebCore::Document::absoluteRegionForEventTargets 
        https://bugs.webkit.org/show_bug.cgi?id=144426
        rdar://problem/20502166

        Reviewed by Tim Horton.

        When a frame had wheel event handlers, we would register the document itself
        as a handler in its parent document. This is problematic, because there's not
        code path that removes it when the frame is destroyed.
        
        It turns out we don't need to do this at all; the non-fast scrollable region
        already takes handlers in subframes into account.

        Tests: fast/events/wheelevent-in-frame.html
               fast/events/wheelevent-in-reattached-frame.html

        * dom/Document.cpp:
        (WebCore::Document::didAddWheelEventHandler):
        (WebCore::Document::didRemoveWheelEventHandler):

2015-04-29  Eric Carlson  <eric.carlson@apple.com>

        Not all videos should automatically play to playback target
        https://bugs.webkit.org/show_bug.cgi?id=144430
        <rdar://problem/20718523>

        Reviewed by Darin Adler.

        * Modules/mediasession/WebMediaSessionManager.cpp:
        (WebCore::WebMediaSessionManager::clientStateDidChange): Consider ExternalDeviceAutoPlayCandidate.
        Minor cleanup.

        * html/HTMLMediaElement.cpp:
        (WebCore::HTMLMediaElement::mediaState): Set ExternalDeviceAutoPlayCandidate when a <video>
        has a file with an audio track that does not loop.

        * page/MediaProducer.h: Add ExternalDeviceAutoPlayCandidate.

2015-04-29  Joseph Pecoraro  <pecoraro@apple.com>

        LiveNodeList may unexpectedly return an element for empty string
        https://bugs.webkit.org/show_bug.cgi?id=144429

        Reviewed by Darin Adler.

        * dom/LiveNodeList.cpp:
        (WebCore::LiveNodeList::namedItem):
        Never return a result for an empty string.

2015-04-29  Gyuyoung Kim  <gyuyoung.kim@webkit.org>

        Remove PassRefPtr in SVGFEFooElement classes
        https://bugs.webkit.org/show_bug.cgi?id=144425

        Reviewed by Darin Adler.

        As a step to use Ref instead of PassRefPtr, this patch purges PassRefPtr
        in SVGFEFooElement.

        No new tests, no behavior changes.

        * css/RGBColor.cpp:
        (WebCore::RGBColor::create):
        * css/RGBColor.h:
        * platform/graphics/filters/PointLightSource.h:
        (WebCore::PointLightSource::create):
        * platform/graphics/filters/SpotLightSource.h:
        (WebCore::SpotLightSource::create):
        * svg/SVGColor.cpp:
        (WebCore::SVGColor::rgbColor):
        * svg/SVGColor.h:
        * svg/SVGFEDistantLightElement.cpp:
        (WebCore::SVGFEDistantLightElement::lightSource):
        * svg/SVGFEDistantLightElement.h:
        * svg/SVGFELightElement.cpp:
        (WebCore::SVGFELightElement::findLightSource):
        * svg/SVGFELightElement.h:
        * svg/SVGFEPointLightElement.cpp:
        (WebCore::SVGFEPointLightElement::lightSource):
        * svg/SVGFEPointLightElement.h:
        * svg/SVGFESpotLightElement.cpp:
        (WebCore::SVGFESpotLightElement::lightSource):
        * svg/SVGFESpotLightElement.h:

2015-04-29  Dean Jackson  <dino@apple.com>

        Create a named CSS property for system colors
        https://bugs.webkit.org/show_bug.cgi?id=144423
        <rdar://problem/20491011>

        Reviewed by Tim Horton.

        Test: fast/css/apple-system-colors.html

        Expose the following values to CSS color properties:
            -apple-system-blue
            -apple-system-brown
            -apple-system-gray
            -apple-system-green
            -apple-system-orange
            -apple-system-pink
            -apple-system-purple
            -apple-system-red
            -apple-system-yellow

        On platforms other than OS X Yosemite and iOS, the
        actual color values are undefined and become transparent
        black. (In fact, not all are defined on iOS either.)

        * WebCore.xcodeproj/project.pbxproj: Two new SPI header files.

        * css/CSSParser.cpp:
        (WebCore::CSSParser::validSystemColorValue): New helper function that
        checks if a CSSValueID is between the two system color values. This
        was being tested everywhere, which meant adding a new system color was
        at risk of being ignored. It's a static method so it can be used
        from the SVG CSS parser too.
        (WebCore::validPrimitiveValueColor): Use the new helper.
        (WebCore::CSSParser::parseValue): Ditto.
        (WebCore::CSSParser::parseBackgroundColor): Ditto.
        (WebCore::CSSParser::parseShadow): Ditto.
        (WebCore::parseDeprecatedGradientColorStop): Ditto.
        (WebCore::parseGradientColorOrKeyword): Ditto.
        * css/CSSParser.h: New static helper function.

        * css/CSSValueKeywords.in: Add the new CSS value keywords.

        * css/SVGCSSParser.cpp:
        (WebCore::validSystemControlColorValue): SVG has a restricted
        set of system colors, so use the helper from CSSParser but with
        an extra condition.
        (WebCore::CSSParser::parseSVGValue): Use the helper.

        * platform/spi/ios/UIColorSPI.h: Added.
        * platform/spi/cocoa/NSColorSPI.h: Added.

        * rendering/RenderThemeIOS.h: New systemColor override. Also add a cache for system colors.
        * rendering/RenderThemeIOS.mm:
        (WebCore::RenderThemeIOS::systemColor): Ask UIColor to provide the color values for the new CSS values.
        * rendering/RenderThemeMac.mm:
        (WebCore::RenderThemeMac::systemColor): Ditto.

2015-04-29  Commit Queue  <commit-queue@webkit.org>

        Unreviewed, rolling out r183600.
        https://bugs.webkit.org/show_bug.cgi?id=144432

        New tests time out everywhere (Requested by ap on #webkit).

        Reverted changeset:

        "Crash at WebCore::Document::absoluteRegionForEventTargets"
        https://bugs.webkit.org/show_bug.cgi?id=144426
        http://trac.webkit.org/changeset/183600

2015-04-29  Joseph Pecoraro  <pecoraro@apple.com>

        Unreviewed iOS build fix. Unused parameter no longer exists.

        * page/scrolling/ScrollingCoordinator.cpp:
        (WebCore::ScrollingCoordinator::absoluteNonFastScrollableRegionForFrame):

2015-04-29  Simon Fraser  <simon.fraser@apple.com>

        Crash at WebCore::Document::absoluteRegionForEventTargets 
        https://bugs.webkit.org/show_bug.cgi?id=144426
        rdar://problem/20502166

        Reviewed by Tim Horton.

        When a frame had wheel event handlers, we would register the document itself
        as a handler in its parent document. This is problematic, because there's not
        code path that removes it when the frame is destroyed.
        
        It turns out we don't need to do this at all; the non-fast scrollable region
        already takes handlers in subframes into account.

        Tests: fast/events/wheelevent-in-frame.html
               fast/events/wheelevent-in-reattached-frame.html

        * dom/Document.cpp:
        (WebCore::Document::didAddWheelEventHandler):
        (WebCore::Document::didRemoveWheelEventHandler):

2015-04-29  David Kilzer  <ddkilzer@apple.com>

        Attempt #2: Switch QuickLook soft-linking to use QuickLookSoftLink.{h,mm}
        <http://webkit.org/b/144362>

        Reviewed by Andy Estes.

        This patch switches soft-linking of QuickLook.framework to
        QuickLookSoftLink.{h,mm} so that we stop exporting unintended
        symbols.

        No new tests since no change in behavior.

        * WebCore.xcodeproj/project.pbxproj:
        - Add QuickLookSoftLinking.{h,mm} to the project.

        * platform/ios/QuickLookSoftLink.h: Added.
        * platform/ios/QuickLookSoftLink.mm: Added.
        * platform/mac/SoftLinking.h:
        (SOFT_LINK_CLASS_FOR_HEADER): Add macro.
        (SOFT_LINK_CLASS_FOR_SOURCE): Add macro.
        (SOFT_LINK_POINTER_FOR_HEADER): Add macro.
        (SOFT_LINK_POINTER_FOR_SOURCE): Add macro.

        * platform/network/ios/QuickLook.h:
        - Remove unused declarations.

        * platform/network/ios/QuickLook.mm:
        (WebCore::QLPreviewConverterClass): Deleted.
        (WebCore::QLTypeCopyBestMimeTypeForFileNameAndMimeType): Deleted.
        (WebCore::QLTypeCopyBestMimeTypeForURLAndMimeType): Deleted.
        (WebCore::QLTypeCopyUTIForURLAndMimeType): Deleted.
        - Remove SOFT_LINK macros and unused methods after switching to
          new QuickLookSoftLink.{h,mm}.
        (WebCore::QLPreviewGetSupportedMIMETypesSet):
        - Switch to use NeverDestroyed<>.
        (WebCore::registerQLPreviewConverterIfNeeded):
        (createQLPreviewProtocol):
        (WebCore::QLPreviewProtocol):
        (WebCore::QuickLookHandle::QuickLookHandle):
        (WebCore::QuickLookHandle::create):
        (WebCore::QuickLookHandle::shouldCreateForMIMEType):
        - Simplify code when using QuickLookSoftLink.h.

        * platform/network/ios/WebCoreURLResponseIOS.mm:
        - Include QuickLookSoftLink.h header.

2015-04-29  Simon Fraser  <simon.fraser@apple.com>

        Compute the non-fast-scrollable region in main-document coordinates
        https://bugs.webkit.org/show_bug.cgi?id=144420

        Reviewed by Tim Horton.

        Compute the non-fast-scrollable region in document coordinates, to make it easier
        to reason about. Previously, it was document coordinates offset by top content inset.

        * page/DebugPageOverlays.cpp:
        (WebCore::MouseWheelRegionOverlay::updateRegion): Traverse all frames to compute the wheel
        event handler region, mapping each to root view coords, and then mapping back into document
        coords at the end.
        (WebCore::NonFastScrollableRegionOverlay::updateRegion): No offset needed here; the
        overlay and region are both document coordinates.
        * page/FrameView.h: Make some mapping function overrides public, and expose widgetsInRenderTree().
        * page/Page.cpp:
        (WebCore::Page::nonFastScrollableRects): Remove frame argument.
        * page/Page.h:
        * page/PageOverlay.cpp:
        (WebCore::PageOverlay::bounds):
        (WebCore::PageOverlay::viewToOverlayOffset): Convenience function to map between
        view and overlay coordinates.
        * page/PageOverlay.h:
        * page/scrolling/AsyncScrollingCoordinator.cpp: New computeNonFastScrollableRegion() signature.
        (WebCore::AsyncScrollingCoordinator::updateNonFastScrollableRegion):
        (WebCore::AsyncScrollingCoordinator::frameViewLayoutUpdated):
        (WebCore::AsyncScrollingCoordinator::scrollingStateTreeAsText):
        * page/scrolling/ScrollingCoordinator.cpp:
        (WebCore::ScrollingCoordinator::absoluteNonFastScrollableRegionForFrame): This function
        recurses on frames, computing an absolute (document-relative) region per frame. This
        removes the confusing offsetting through top content inset.
        Change how we get to plugins that want wheel events; we can't get from PluginViewBase
        to renderers, so use FrameView's list of Widgets, and their RenderWidgets. This fixes
        regions for transformed plugin-ins.
        For subframes, we get a region in the subframe's document coords. Map to that sub-frame,
        then to our frame, then to our document.
        (WebCore::ScrollingCoordinator::absoluteNonFastScrollableRegion): Wrapper that hides
        the recursive function.
        (WebCore::ScrollingCoordinator::computeNonFastScrollableRegion): Deleted.
        * page/scrolling/ScrollingCoordinator.h:
        * page/scrolling/ScrollingTree.cpp:
        (WebCore::ScrollingTree::shouldHandleWheelEventSynchronously): Map the event point
        from view coordinates to document coordinates for testing against the non-fast region.
        We previously assert that the root note is a FrameScrolling node.
        * page/scrolling/ScrollingTreeFrameScrollingNode.cpp:
        (WebCore::ScrollingTreeFrameScrollingNode::viewToContentsOffset): Similar to ScrollView::viewToContents()
        for the scrolling tree.
        * page/scrolling/ScrollingTreeFrameScrollingNode.h:
        * testing/Internals.cpp:
        (WebCore::Internals::nonFastScrollableRects): No need for frame arg.

2015-04-29  Brent Fulgham  <bfulgham@apple.com>

        Expand test infrastructure to support scrolling tests
        https://bugs.webkit.org/show_bug.cgi?id=143684
        <rdar://problem/20375516>

        Reviewed by Simon Fraser.

        Tested by various fast/scrolling and platform/mac-wk2/tiled-drawing/scrolling tests.

        This series of changes adds a new singleton class, 'WheelEventTestTrigger', which encapsulates a
        function object to be fired when scroll events are finished. The object also keeps track of reasons
        why the test should not yet fire (e.g., 'rubberbanding' is active) so that tests do not incorrectly
        check rendering state in the middle of an animation.

        Switch from the original WeakPtr design to ThreadSafeRefPtr, because WeakPtr cannot be shared
        across multiple threads.

        * page/FrameView.cpp:
        (WebCore::FrameView::layout): Make sure ScrollAnimator knows about any active test trigger.
        (WebCore::FrameView::setScrollPosition): Ditto.
        (WebCore::FrameView::didAddScrollbar): Ditto.
        * page/MainFrame.cpp:
        (WebCore::MainFrame::testTrigger): Moved to Page.
        (WebCore::MainFrame::ensureTestTrigger): Ditto.
        * page/MainFrame.h:
        * page/Page.cpp:
        (WebCore::Page::testTrigger): Moved from MainFrame, and converted to use RefPtr.
        (WebCore::Page::ensureTestTrigger): Ditto.
        * page/Page.h:
        * page/WheelEventTestTrigger.cpp:
        (WebCore::WheelEventTestTrigger::WheelEventTestTrigger): Remove WeakPtr code.
        (WebCore::WheelEventTestTrigger::createWeakPtr): Deleted.
        * page/WheelEventTestTrigger.h:
        * page/mac/EventHandlerMac.mm:
        (WebCore::EventHandler::platformPrepareForWheelEvents): Make sure the scroll animator knows about
        any active test trigger object.
        * page/scrolling/AsyncScrollingCoordinator.cpp:
        (WebCore::AsyncScrollingCoordinator::updateScrollPositionAfterAsyncScroll): Notify WheelEventTestTrigger
        that the scrolling thread is synced with the main thread.
        * platform/ScrollAnimator.h: Hold a RefPtr to the WheelEventTestTrigger. 
        (WebCore::ScrollAnimator::ScrollAnimator::setWheelEventTestTrigger):
        * platform/cocoa/ScrollController.h:
        * platform/cocoa/ScrollController.mm:
        (WebCore::ScrollController::startSnapRubberbandTimer): Notify test trigger to hold tests until rubber band
        snapping is complete.
        (WebCore::ScrollController::stopSnapRubberbandTimer): Notify test trigger that rubber band snapping is done.
        (WebCore::ScrollController::startScrollSnapTimer): Notify test trigger to hold tests until scroll snapping
        is complete.
        (WebCore::ScrollController::stopScrollSnapTimer): Notify test trigger that scroll snapping is done.
        * platform/mac/ScrollAnimatorMac.mm:
        (WebCore::ScrollAnimatorMac::didBeginScrollGesture): Notify test trigger that a content scroll is in progress.
        (WebCore::ScrollAnimatorMac::didEndScrollGesture): Notify test trigger that a content scroll is finished.
        (WebCore::ScrollAnimatorMac::sendContentAreaScrolledSoon): Notify test trigger to hold tests until the content
        scrolling is complete.
        (WebCore::ScrollAnimatorMac::sendContentAreaScrolledTimerFired): Notify test trigger that content scrolling is done.
        * rendering/RenderBox.cpp:
        (WebCore::connectScrollAnimatorToTestTrigger): Helper function.
        (WebCore::RenderBox::setScrollLeft): Call 'connectScrollAnimatorToTestTrigger' to connect the ScrollAnimator
        to the WheelEventTestTrigger so that future scroll operations can notify the test infrastructure.
        (WebCore::RenderBox::setScrollTop): Ditto.
        * rendering/RenderLayer.cpp:
        (WebCore::RenderLayer::createScrollbar): Make sure the ScrollAnimator knows about any active test triggers.
        * rendering/RenderListBox.cpp:
        (WebCore::connectScrollAnimatorToTestTrigger): Helper function.
        (WebCore::RenderListBox::setScrollLeft): Call 'connectScrollAnimatorToTestTrigger' to connect the ScrollAnimator
        to the WheelEventTestTrigger so that future scroll operations can notify the test infrastructure.
        (WebCore::RenderListBox::setScrollTop): Ditto.
        (WebCore::RenderListBox::createScrollbar): Ditto.
        * testing/js/WebCoreTestSupport.cpp:
        (WebCoreTestSupport::monitorWheelEvents): Look for WheelEventTestTrigger in Page, rather than MainFrame.
        (WebCoreTestSupport::setTestCallbackAndStartNotificationTimer): Ditto.

2015-04-29  Javier Fernandez  <jfernandez@igalia.com>

        [CSS Box Alignment] Unifying alignment data in a single class
        https://bugs.webkit.org/show_bug.cgi?id=144384

        Reviewed by David Hyatt.

        The new CSS Box Alignment specification introduces more complex
        values and syntax for defining alignment properties. Most of the
        alignment values were just keyword identifiers, but the new syntax
        allows different combinations of identifiers to determine the
        alignment behavior (eg. overflow-alignment keyword).

        This patch wll help to implement later overflow handling and
        specially Content Dstribution alignment, the most complex case by
        far. It will be more consistent against style changes and repaint,
        snce we will have just one field to monitoring for value
        modfications.

        No new tests, because no new functionality was added; it's just a
        code refactoring which is just expected to pass current tests.

        * css/CSSComputedStyleDeclaration.cpp:
        (WebCore::resolveSelfAlignmentAuto):
        (WebCore::ComputedStyleExtractor::propertyValue):
        * css/CSSPropertyNames.in:
        * css/StyleBuilderConverter.h:
        (WebCore::StyleBuilderConverter::convertSelfOrDefaultAlignmentData):
        (WebCore::StyleBuilderConverter::convertSVGColor): Deleted.
        * css/StyleBuilderCustom.h:
        (WebCore::StyleBuilderCustom::applyInitialWebkitMaskImage): Deleted.
        (WebCore::StyleBuilderCustom::applyInheritWebkitMaskImage): Deleted.
        (WebCore::ApplyPropertyBorderImageModifier::applyInheritValue): Deleted.
        (WebCore::ApplyPropertyBorderImageModifier::applyInitialValue): Deleted.
        (WebCore::ApplyPropertyBorderImageModifier::applyValue): Deleted.
        (WebCore::ApplyPropertyBorderImageModifier::getValue): Deleted.
        (WebCore::ApplyPropertyBorderImageModifier::setValue): Deleted.
        (WebCore::StyleBuilderCustom::applyInheritLineHeight): Deleted.
        * css/StyleResolver.cpp:
        (WebCore::StyleResolver::adjustRenderStyle):
        * mathml/MathMLTextElement.cpp:
        (WebCore::MathMLTextElement::createElementRenderer):
        * rendering/RenderFlexibleBox.cpp:
        (WebCore::RenderFlexibleBox::styleDidChange):
        * rendering/RenderFullScreen.cpp:
        (WebCore::createFullScreenStyle):
        * rendering/RenderMenuList.cpp:
        (WebCore::RenderMenuList::adjustInnerStyle):
        * rendering/mathml/RenderMathMLRoot.cpp:
        (WebCore::RenderMathMLRoot::updateStyle):
        * rendering/mathml/RenderMathMLScripts.cpp:
        (WebCore::RenderMathMLScripts::fixAnonymousStyleForSubSupPair):
        (WebCore::RenderMathMLScripts::fixAnonymousStyles):
        * rendering/style/RenderStyle.cpp:
        (WebCore::RenderStyle::resolveAlignment):
        (WebCore::RenderStyle::resolveJustification):
        * rendering/style/RenderStyle.h:
        * rendering/style/StyleRareNonInheritedData.cpp:
        (WebCore::StyleRareNonInheritedData::StyleRareNonInheritedData):
        (WebCore::StyleRareNonInheritedData::operator==):
        (WebCore::StyleRareNonInheritedData::contentDataEquivalent): Deleted.
        * rendering/style/StyleRareNonInheritedData.h:
        * rendering/style/StyleSelfAlignmentData.h: Added.
        (WebCore::StyleSelfAlignmentData::StyleSelfAlignmentData):
        (WebCore::StyleSelfAlignmentData::setPosition):
        (WebCore::StyleSelfAlignmentData::setPositionType):
        (WebCore::StyleSelfAlignmentData::setOverflow):
        (WebCore::StyleSelfAlignmentData::position):
        (WebCore::StyleSelfAlignmentData::positionType):
        (WebCore::StyleSelfAlignmentData::overflow):
        (WebCore::StyleSelfAlignmentData::operator==):
        (WebCore::StyleSelfAlignmentData::operator!=):

2015-04-29  Joseph Pecoraro  <pecoraro@apple.com>

        NodeList has issues with Symbol and empty string
        https://bugs.webkit.org/show_bug.cgi?id=144310

        Reviewed by Darin Adler.

        Tests: fast/dom/StyleSheet/stylesheet-symbol-names.html
               fast/dom/dataset-name-getter-symbols.html
               fast/dom/named-items-with-empty-name.html
               fast/dom/named-items-with-symbol-name.html
               storage/domstorage/localstorage/named-items.html

        Test different bindings objects with custom named setter
        and getter handlers. Ensure that they handle Symbol properties
        gracefully. Often times avoiding the string path.

        * dom/StaticNodeList.cpp:
        (WebCore::StaticNodeList::namedItem):
        (WebCore::StaticElementList::namedItem):
        Better handle the empty string. It should not match an item.

        * bindings/js/JSDOMBinding.h:
        (WebCore::propertyNameToString):
        In cases where we would use this, we should have handled
        Symbol properties.

        (WebCore::propertyNameToAtomicString):
        For Symbols, use the unique string. This should result
        in no matches for a Symbol property.

        * bindings/js/JSDOMStringMapCustom.cpp:
        (WebCore::JSDOMStringMap::getOwnPropertySlotDelegate):
        (WebCore::JSDOMStringMap::deleteProperty):
        (WebCore::JSDOMStringMap::putDelegate):
        * bindings/js/JSStorageCustom.cpp:
        (WebCore::JSStorage::canGetItemsForName):
        (WebCore::JSStorage::nameGetter):
        (WebCore::JSStorage::deleteProperty):
        (WebCore::JSStorage::putDelegate):
        * bindings/js/JSStyleSheetListCustom.cpp:
        (WebCore::JSStyleSheetList::canGetItemsForName):
        Treat Symbol properties as private properties. They just
        go directly through to the Object, and avoid the string
        getter/setter property path.

2015-04-29  Eric Carlson  <eric.carlson@apple.com>

        [Mac] Register with device picker whenever a page has <video>
        https://bugs.webkit.org/show_bug.cgi?id=144408

        Reviewed by Jer Noble.

        * Modules/mediasession/WebMediaSessionManager.cpp:
        (WebCore::WebMediaSessionManager::setPlaybackTarget): Drive-by fix: don't tell a client to play
        to the target when it has not routes.

        * html/HTMLMediaElement.cpp:
        (WebCore::HTMLMediaElement::mediaState): Set RequiresPlaybackTargetMonitoring whenever the 
        element is <video> with a video track that does not block wireless playback.

2015-04-29  Martin Robinson  <mrobinson@igalia.com>

        [GTK] Add support for automatic hyphenation
        https://bugs.webkit.org/show_bug.cgi?id=44478

        Reviewed by Carlos Garcia Campos.

        No new tests. This patch unskips and updates results for existing hyphenation tests.

        * PlatformGTK.cmake: Add the libhypen implementation to the source list.
        * platform/gtk/GtkUtilities.cpp:
        (WebCore::topLevelPath): Add this helper function taken from the test harness.
        (WebCore::getWebKitBuildDirectory): Ditto.
        * platform/gtk/GtkUtilities.h: Add function declarations for the helper functions.
        * platform/text/Hyphenation.cpp: Surround this implementation with !USE(LIBHYPHEN),
          so that it can still be shared.
        * platform/text/gtk/HyphenationLibHyphen.cpp: Added.
        (WebCore::extractLocaleFromDictionaryFilePath): Take in a dictionary filename and
        determine the locale that it covers.
        (WebCore::scanDirectoryForDicionaries): Look for all installed dictionaries as well
        as ones in the JHBuild root for testing.
        (WebCore::scanTestDictionariesDirectoryIfNecessary): Try to scan the dictionaries installed
        in the JHBuild root.
        (WebCore::availableLocales): Getter for global hash of installed dictionaries.
        (WebCore::canHyphenate): Added libhyphen implementation.
        (WebCore::HyphenationDictionary): Helper class that properly manages the memory of
        an open libhyphen dictionary. This is useful so that they can be stored in an MRU
        cache. This is a similar approach to the CoreFoundation implementation.
        (WebCore::AtomicStringKeyedMRUCache<RefPtr<HyphenationDictionary>>::createValueForNullKey): MRU cache
        helper.
        (WebCore::AtomicStringKeyedMRUCache<RefPtr<HyphenationDictionary>>::createValueForKey): Ditto.
        (WebCore::hyphenDictionaryCache): A cache for opened hyphenation dictionaries.
        (WebCore::countLeadingSpaces): Count leading spaces, since WebCore often passes words with them.
        (WebCore::lastHyphenLocation): Added libhyphen implementation.

2015-04-29  Myles C. Maxfield  <mmaxfield@apple.com>

        Unreviewed build fix.

        * platform/graphics/mac/FontCacheMac.mm:
        (WebCore::fontWithFamily):

2015-04-29  Eric Carlson  <eric.carlson@apple.com>

        Some media tests assert after r183096
        https://bugs.webkit.org/show_bug.cgi?id=144098

        Reviewed by Brent Fulgham.

        Never assert when a MediaPlaybackTargetClient client calls one of the methods on Document
        after it has been unregistered. A media element unregisters itself when it is removed from the 
        document, but scripts can still run and modify state that results in a call to
        playbackTargetPickerClientStateDidChange. Remove the asserts instead of adding checks to
        the various call sites.

        * dom/Document.cpp:
        (WebCore::Document::showPlaybackTargetPicker): Don't assert if the client has already
        been unregistered.
        (WebCore::Document::playbackTargetPickerClientStateDidChange): Ditto.

2015-04-29  Zalan Bujtas  <zalan@apple.com>

        Simple line layout: Web process spins endlessly below layoutSimpleLines.
        https://bugs.webkit.org/show_bug.cgi?id=144403
        rdar://problem/20742783

        Reviewed by Antti Koivisto.

        When a text fragment overlaps multiple renderes and it does not fit the current line,
        we revert the text fragment iterator position so that the overlapping content
        gets processed again for the next line.
        However, TextFragmentIterator::revertToFragment() was reverting too much and
        we started processing old content all over again -> infinite loop.

        This patch ensures that text fragment iterator is reverted to the right position.

        Test: fast/text/simple-line-layout-wrapping-multiple-renderers-hang.html

        * rendering/SimpleLineLayout.cpp:
        (WebCore::SimpleLineLayout::createLineRuns):
        * rendering/SimpleLineLayoutTextFragmentIterator.cpp:
        (WebCore::SimpleLineLayout::TextFragmentIterator::revertToEndOfFragment):
        (WebCore::SimpleLineLayout::TextFragmentIterator::revertToFragment): Deleted.
        * rendering/SimpleLineLayoutTextFragmentIterator.h:

2015-04-29  Filip Pizlo  <fpizlo@apple.com>

        JSTypeInfo should have an inline type flag to indicate of getCallData() has been overridden
        https://bugs.webkit.org/show_bug.cgi?id=144397

        Reviewed by Andreas Kling.

        If you override getCallData() and you want to be called a "function", then you need to use the
        new TypeOfShouldCallGetCallData flag.

        * bindings/scripts/CodeGeneratorJS.pm:
        (GenerateHeader):
        * bridge/objc/objc_runtime.h:
        * bridge/runtime_method.h:
        * bridge/runtime_object.h:

2015-04-29  Commit Queue  <commit-queue@webkit.org>

        Unreviewed, rolling out r183553 and r183561.
        https://bugs.webkit.org/show_bug.cgi?id=144406

        broke the iOS build (Requested by thorton__ on #webkit).

        Reverted changesets:

        "Switch QuickLook soft-linking to use
        QuickLookSoftLink.{h,mm}"
        https://bugs.webkit.org/show_bug.cgi?id=144362
        http://trac.webkit.org/changeset/183553

        "Unreviewed iOS build fix after r183553: fix declaration of
        QLPreviewScheme"
        http://trac.webkit.org/changeset/183561

2015-04-29  Antti Koivisto  <antti@apple.com>

        ResourceLoadPriority should be enum class
        https://bugs.webkit.org/show_bug.cgi?id=144326

        Reviewed by Darin Adler.

        * html/HTMLLinkElement.cpp:
        (WebCore::HTMLLinkElement::process):
        * loader/LinkLoader.cpp:
        (WebCore::LinkLoader::loadLink):
        * loader/ResourceLoadScheduler.cpp:
        (WebCore::ResourceLoadScheduler::scheduleLoad):
        (WebCore::ResourceLoadScheduler::servePendingRequests):
        (WebCore::ResourceLoadScheduler::HostInformation::~HostInformation):
        (WebCore::ResourceLoadScheduler::HostInformation::priorityToIndex):
        (WebCore::ResourceLoadScheduler::HostInformation::schedule):
        (WebCore::ResourceLoadScheduler::HostInformation::remove):
        (WebCore::ResourceLoadScheduler::HostInformation::hasRequests):
        (WebCore::ResourceLoadScheduler::HostInformation::limitRequests):
        * loader/ResourceLoadScheduler.h:
        (WebCore::ResourceLoadScheduler::HostInformation::requestsPending):

            Modernize ResourceLoadScheduler code a bit while switching to enum class.

        * loader/cache/CachedResource.cpp:
        (WebCore::defaultPriorityForResourceType):
        * loader/cache/CachedResourceLoader.cpp:
        (WebCore::CachedResourceLoader::requestResource):
        * loader/icon/IconLoader.cpp:
        (WebCore::IconLoader::startLoading):
        * platform/graphics/avfoundation/objc/WebCoreAVFResourceLoader.mm:
        (WebCore::WebCoreAVFResourceLoader::startLoading):
        * platform/network/ResourceLoadPriority.h:
        (WebCore::operator++):
        (WebCore::operator--):
        * platform/network/ResourceRequestBase.cpp:
        (WebCore::ResourceRequestBase::adopt):
        (WebCore::ResourceRequestBase::copyData):
        (WebCore::ResourceRequestBase::cachePolicy):
        (WebCore::ResourceRequestBase::priority):
        (WebCore::ResourceRequestBase::setPriority):
        * platform/network/ResourceRequestBase.h:

            Remove bitfields. ResourceRequests are not present in large enough numbers to require this kind of optimization.
            Use modern initialization syntax.

        (WebCore::ResourceRequestBase::ResourceRequestBase):
        (WebCore::CrossThreadResourceRequestDataBase::CrossThreadResourceRequestDataBase): Deleted.
        * platform/network/cf/ResourceRequestCFNet.cpp:
        (WebCore::initializeMaximumHTTPConnectionCountPerHost):
        (WebCore::initializeHTTPConnectionSettingsOnStartup):
        * platform/network/cf/ResourceRequestCFNet.h:
        (WebCore::toResourceLoadPriority):
        (WebCore::toPlatformRequestPriority):
        * platform/network/soup/ResourceRequest.h:
        (WebCore::toSoupMessagePriority):
        * testing/Internals.cpp:
        (WebCore::stringToResourceLoadPriority):

2015-04-29  Myles C. Maxfield  <mmaxfield@apple.com>

        [OS X] Use CTFontCreateForCSS instead of doing font search ourselves
        https://bugs.webkit.org/show_bug.cgi?id=132159

        Reviewed by Darin Adler.

        On platforms that support it, delegate font selection logic to the platform. Currently, this is
        only supported on Mac, using CTFontCreateForCSS().

        This also changes the mechanism that enforces our font whitelist in our tests. We used to
        swizzle the implementations of NSFontManager methods. This patch migrates to using a whitelist of
        font family names instead.

        Note that this patch is a work in progress, because it makes the following tests fail:
        fast/css/font-weight-1.html
        fast/forms/validation-message-appearance.html
        fast/forms/select/optgroup-rendering.html

        No new tests, because there is no behavior change.

        * platform/graphics/FontCache.h: Add a function to set the whitlist.
        * platform/graphics/mac/FontCacheMac.mm:
        (WebCore::fontWhitelist):
        (WebCore::FontCache::setFontWhitelist):
        (WebCore::toAppKitFontWeight):
        (WebCore::toCoreTextFontWeight):
        (WebCore::fontWithFamily): If ENABLE(PLATFORM_FONT_LOOKUP), use CTFontCreateForCSS().
        (WebCore::FontCache::createFontPlatformData):
        * platform/spi/cocoa/CoreTextSPI.h: Add signature for CTFontCreateForCSS().

2015-04-29  Jer Noble  <jer.noble@apple.com>

        Unreviewed iOS build fix after r183553: fix declaration of QLPreviewScheme

        * platform/ios/QuickLookSoftLink.mm:

2015-04-29  Eric Carlson  <eric.carlson@apple.com>

        [Mac] Use new device picker menu API
        https://bugs.webkit.org/show_bug.cgi?id=144392

        Reviewed by Jer Noble.

        * Modules/mediasession/WebMediaSessionManager.cpp:
        (WebCore::WebMediaSessionManager::showPlaybackTargetPicker): Pass the route state to the picker.
        * platform/graphics/MediaPlaybackTargetPicker.h:

        * platform/graphics/avfoundation/objc/MediaPlaybackTargetPickerMac.h:
        * platform/graphics/avfoundation/objc/MediaPlaybackTargetPickerMac.mm:
        (WebCore::MediaPlaybackTargetPickerMac::MediaPlaybackTargetPickerMac):  m_deviceChangeTimer -> m_pendingActionTimer.
        (WebCore::MediaPlaybackTargetPickerMac::~MediaPlaybackTargetPickerMac): Ditto.
        (WebCore::MediaPlaybackTargetPickerMac::pendingActionTimerFired): Renamed from 
        outputeDeviceAvailabilityChangedTimerFired.
        (WebCore::MediaPlaybackTargetPickerMac::availableDevicesDidChange): Use addPendingAction.
        (WebCore::MediaPlaybackTargetPickerMac::showPlaybackTargetPicker): Call new picker API if it is
        available. New API returns true if a target is chosen, so call currentDeviceDidChange in that case.
        (WebCore::MediaPlaybackTargetPickerMac::addPendingAction): New.
        (WebCore::MediaPlaybackTargetPickerMac::currentDeviceDidChange): Make asynchronous because it 
        can now be called from showPlaybackTargetPicker.
        (WebCore::MediaPlaybackTargetPickerMac::outputeDeviceAvailabilityChangedTimerFired): Renamed
        pendingActionTimerFired.

        * platform/spi/cocoa/AVKitSPI.h: Add new SPI.

2015-04-29  Antti Koivisto  <antti@apple.com>

        Main resource loaded via 304 response becomes empty if reloaded by user
        https://bugs.webkit.org/show_bug.cgi?id=144386

        Reviewed by Darin Adler.

        Memory cache layer may make a main resource request conditional (add If-modified-since/If-none-match header).
        DocumentLoader stores the final ResourceRequest associated with document. If user triggered reload for
        conditionally loaded document DocumentLoader would include the same conditionals to the new request as well.
        Since these were not added by the memory cache it would pass any 304 response back to the DocumentLoader.
        However DocumentLoader has no code to handle 304 so we would end up with an empty document.

        Test: http/tests/cache/main-resource-304-reload.html

        * loader/DocumentLoader.cpp:
        (WebCore::DocumentLoader::startLoadingMainResource):

            Ensure DocumentLoader doesn't issue conditional requests.

2015-04-28  Jer Noble  <jer.noble@apple.com>

        Replace HTMLMediaElement's seek timer with a task queue.
        https://bugs.webkit.org/show_bug.cgi?id=144353

        Reviewed by Eric Carlson.

        No change in functionality, so no new tests added.

        Replace a zero-length timer with an explicit task queue, backed by Document's postTask(). To
        make enqueued tasks cancellable, add a new GenericTaskQueue class modelled on
        GenericEventQueue which can cancel enqueued but not-yet-executed tasks.

        * CMakeLists.txt:
        * WebCore.vcxproj/WebCore.vcxproj:
        * WebCore.vcxproj/WebCore.vcxproj.filters:
        * WebCore.xcodeproj/project.pbxproj:
        * dom/GenericTaskQueue.cpp: Added.
        (WebCore::GenericTaskQueue::GenericTaskQueue): Initialize ivars.
        (WebCore::GenericTaskQueue::enqueueTask): Wrap the task and pass it to the ScriptExecutionContext.
        (WebCore::GenericTaskQueue::close): Assert that the task queue will not accept any additional tasks.
        (WebCore::GenericTaskQueue::cancelAllTasks): Revoke outstanding weak pointers, thereby cancelling tasks.
        * dom/GenericTaskQueue.h: Added.
        (WebCore::GenericTaskQueue::hasPendingTasks):
        * html/HTMLMediaElement.cpp:
        (WebCore::HTMLMediaElement::HTMLMediaElement): Initialize the task queue.
        (WebCore::HTMLMediaElement::~HTMLMediaElement): Close the task queue.
        (WebCore::HTMLMediaElement::seekWithTolerance): Post a task, rather than start a timer.
        (WebCore::HTMLMediaElement::seekTask): Renamed from seekTimerFired().
        (WebCore::HTMLMediaElement::seekTimerFired): Deleted.
        * html/HTMLMediaElement.h:

2015-04-28  David Kilzer  <ddkilzer@apple.com>

        Switch QuickLook soft-linking to use QuickLookSoftLink.{h,mm}
        <http://webkit.org/b/144362>

        Reviewed by Andy Estes.

        This patch switches soft-linking of QuickLook.framework to
        QuickLookSoftLink.{h,mm} so that we stop exporting unintended
        symbols.

        No new tests since no change in behavior.

        * WebCore.xcodeproj/project.pbxproj:
        - Add QuickLookSoftLinking.{h,mm} to the project.

        * platform/ios/QuickLookSoftLink.h: Added.
        * platform/ios/QuickLookSoftLink.mm: Added.
        * platform/mac/SoftLinking.h:
        (SOFT_LINK_CLASS_FOR_HEADER): Add macro.
        (SOFT_LINK_CLASS_FOR_SOURCE): Add macro.

        * platform/network/ios/QuickLook.h:
        - Remove unused declarations.

        * platform/network/ios/QuickLook.mm:
        (WebCore::QLPreviewConverterClass): Deleted.
        (WebCore::QLTypeCopyBestMimeTypeForFileNameAndMimeType): Deleted.
        (WebCore::QLTypeCopyBestMimeTypeForURLAndMimeType): Deleted.
        (WebCore::QLTypeCopyUTIForURLAndMimeType): Deleted.
        - Remove SOFT_LINK macros and unused methods after switching to
          new QuickLookSoftLink.{h,mm}.
        (WebCore::QLPreviewGetSupportedMIMETypesSet):
        - Switch to use NeverDestroyed<>.
        (WebCore::registerQLPreviewConverterIfNeeded):
        (createQLPreviewProtocol):
        (WebCore::QLPreviewProtocol):
        (WebCore::QuickLookHandle::QuickLookHandle):
        (WebCore::QuickLookHandle::create):
        (WebCore::QuickLookHandle::shouldCreateForMIMEType):
        - Simplify code when using QuickLookSoftLink.h.

        * platform/network/ios/WebCoreURLResponseIOS.mm:
        - Include QuickLookSoftLink.h header.

2015-04-29  Darin Adler  <darin@apple.com>

        [ES6] Implement Unicode code point escapes
        https://bugs.webkit.org/show_bug.cgi?id=144377

        Reviewed by Antti Koivisto.

        Test: js/unicode-escape-sequences.html

        * css/CSSParser.cpp:
        (WebCore::CSSParser::parseEscape): Use ICU's UCHAR_MAX_VALUE instead of writing
        out 0x10FFFF; clearer this way. Also use our replacementCharacter instead of
        writing out 0xFFFD.

        * html/parser/HTMLEntityParser.cpp:
        (WebCore::isAlphaNumeric): Deleted.
        (WebCore::HTMLEntityParser::legalEntityFor): Use ICU's UCHAR_MAX_VALUE and
        U_IS_SURROGATE instead of writing the code out. Didn't use U_IS_UNICODE_CHAR
        because that also includes U_IS_UNICODE_NONCHAR and thus would change behavior,
        but maye it's something we want to do in the future.
        (WebCore::HTMLEntityParser::consumeNamedEntity): Use isASCIIAlphanumeric instead
        of a the function in this file that does the same thing less efficiently.

        * html/parser/InputStreamPreprocessor.h:
        (WebCore::InputStreamPreprocessor::processNextInputCharacter): Use
        replacementCharacter from CharacterNames.h instead of writing out 0xFFFd.

        * xml/parser/CharacterReferenceParserInlines.h:
        (WebCore::consumeCharacterReference): Use ICU's UCHAR_MAX_VALUE instead of
        defining our own local highestValidCharacter constant.

2015-04-29  Martin Robinson  <mrobinson@igalia.com>

        [CMake] [GTK] Organize and clean up unused CMake variables
        https://bugs.webkit.org/show_bug.cgi?id=144364

        Reviewed by Gyuyoung Kim.

        * PlatformGTK.cmake: Add variables specific to this project.

2015-04-29  Carlos Garcia Campos  <cgarcia@igalia.com>

        REGRESSION(r182573): [GTK] The default context menu contains an empty item since r182573
        https://bugs.webkit.org/show_bug.cgi?id=144388

        Reviewed by Brady Eidson.

        There used to be a method in ContextMenuItem to check if share
        menu item was supported or not, but since r182573, there's a
        method to get the share menu item. If the returned menu item is
        null, it's not added to the menu, but we are not returning a null
        ContextMenu item even though we don't support share menu item.

        * platform/gtk/ContextMenuItemGtk.cpp:
        (WebCore::ContextMenuItem::shareMenuItem): Return a null ContextMenuItem.

2015-04-29  Zan Dobersek  <zdobersek@igalia.com>

        Switch to std::function<>, std::bind() in MediaPlayerPrivateAVFoundationObjC
        https://bugs.webkit.org/show_bug.cgi?id=144232

        Reviewed by Darin Adler.

        * platform/graphics/avfoundation/objc/MediaPlayerPrivateAVFoundationObjC.mm:
        (-[WebCoreAVFMovieObserver observeValueForKeyPath:ofObject:change:context:]): 
        Replace uses of WTF::Function<> and WTF::bind() with the STL alternatives.

2015-04-29  Hyungwook Lee  <hyungwook.lee@navercorp.com>

        Fix crash in WebCore::LogicalSelectionOffsetCaches::ContainingBlockInfo::setBlock().
        https://bugs.webkit.org/show_bug.cgi?id=140261

        Reviewed by Darin Adler.

        We need to check whether RenderObject is valid in RenderView::fooSubtreeSelection functions
        because invalid object has caused a crash. This patch adds isValidObjectForNewSelection(), and use it.

        * rendering/RenderView.cpp:
        (WebCore::isValidObjectForNewSelection):
        (WebCore::RenderView::clearSubtreeSelection):
        (WebCore::RenderView::applySubtreeSelection):

2015-04-29  Youenn Fablet  <youenn.fablet@crf.canon.fr>

        Synchronous XMLHttpRequest should get access to AppCache resources stored as flat files
        https://bugs.webkit.org/show_bug.cgi?id=143711

        Reviewed by Darin Adler.

        This patch checks whether a substitute resource data is stored in memory or in file for synchronous loads.
        If data is stored in file, it reads the data through SharedBuffer::createWithContentsOfFile.
        This patch refactors some routines to replace Vector<char> by SharedBuffer to transmit response data.

        Test: http/tests/appcache/simple-video-sync.html

        * html/HTMLMediaElement.cpp:
        (WebCore::HTMLMediaElement::parseAttribute):
        * loader/DocumentThreadableLoader.cpp:
        (WebCore::DocumentThreadableLoader::loadRequest):
        * loader/FrameLoader.cpp:
        (WebCore::FrameLoader::loadResourceSynchronously):
        * loader/FrameLoader.h:
        * loader/appcache/ApplicationCacheHost.cpp:
        (WebCore::ApplicationCacheHost::maybeLoadResource):
        (WebCore::ApplicationCacheHost::createFileURL):
        (WebCore::ApplicationCacheHost::maybeLoadSynchronously):
        (WebCore::ApplicationCacheHost::maybeLoadFallbackSynchronously):
        * loader/appcache/ApplicationCacheHost.h:
        * xml/XSLTProcessorLibxslt.cpp:
        (WebCore::docLoaderFunc):
        * xml/parser/XMLDocumentParserLibxml2.cpp:
        (WebCore::openFunc):

2015-04-29  Gyuyoung Kim  <gyuyoung.kim@webkit.org>

        Purge PassRefPtr from createSVGPathSegFoo factory functions
        https://bugs.webkit.org/show_bug.cgi?id=144374

        Reviewed by Darin Adler.

        Use Ref instead of PassRefPtr in createSVGPathSegFoo functions because
        those factory functions can't return null. Additionally let's remove unnecessary
        #include<PassRefPtr.h> there.

        No new tests, no behavior changes.

        * svg/SVGColor.cpp:
        (WebCore::SVGColor::cloneForCSSOM):
        * svg/SVGColor.h:
        * svg/SVGGlyphMap.h:
        (WebCore::GlyphMapNode::create):
        * svg/SVGPaint.cpp:
        (WebCore::SVGPaint::cloneForCSSOM):
        * svg/SVGPaint.h:
        * svg/SVGPathElement.cpp:
        (WebCore::SVGPathElement::createSVGPathSegClosePath):
        (WebCore::SVGPathElement::createSVGPathSegMovetoAbs):
        (WebCore::SVGPathElement::createSVGPathSegMovetoRel):
        (WebCore::SVGPathElement::createSVGPathSegLinetoAbs):
        (WebCore::SVGPathElement::createSVGPathSegLinetoRel):
        (WebCore::SVGPathElement::createSVGPathSegCurvetoCubicAbs):
        (WebCore::SVGPathElement::createSVGPathSegCurvetoCubicRel):
        (WebCore::SVGPathElement::createSVGPathSegCurvetoQuadraticAbs):
        (WebCore::SVGPathElement::createSVGPathSegCurvetoQuadraticRel):
        (WebCore::SVGPathElement::createSVGPathSegArcAbs):
        (WebCore::SVGPathElement::createSVGPathSegArcRel):
        (WebCore::SVGPathElement::createSVGPathSegLinetoHorizontalAbs):
        (WebCore::SVGPathElement::createSVGPathSegLinetoHorizontalRel):
        (WebCore::SVGPathElement::createSVGPathSegLinetoVerticalAbs):
        (WebCore::SVGPathElement::createSVGPathSegLinetoVerticalRel):
        (WebCore::SVGPathElement::createSVGPathSegCurvetoCubicSmoothAbs):
        (WebCore::SVGPathElement::createSVGPathSegCurvetoCubicSmoothRel):
        (WebCore::SVGPathElement::createSVGPathSegCurvetoQuadraticSmoothAbs):
        (WebCore::SVGPathElement::createSVGPathSegCurvetoQuadraticSmoothRel):
        * svg/SVGPathElement.h:
        * svg/SVGPathSegArcAbs.h:
        (WebCore::SVGPathSegArcAbs::create):
        * svg/SVGPathSegArcRel.h:
        (WebCore::SVGPathSegArcRel::create):
        * svg/SVGPathSegClosePath.h:
        (WebCore::SVGPathSegClosePath::create):
        * svg/SVGPathSegCurvetoCubicAbs.h:
        (WebCore::SVGPathSegCurvetoCubicAbs::create):
        * svg/SVGPathSegCurvetoCubicRel.h:
        (WebCore::SVGPathSegCurvetoCubicRel::create):
        * svg/SVGPathSegCurvetoCubicSmoothAbs.h:
        (WebCore::SVGPathSegCurvetoCubicSmoothAbs::create):
        * svg/SVGPathSegCurvetoCubicSmoothRel.h:
        (WebCore::SVGPathSegCurvetoCubicSmoothRel::create):
        * svg/SVGPathSegCurvetoQuadraticAbs.h:
        (WebCore::SVGPathSegCurvetoQuadraticAbs::create):
        * svg/SVGPathSegCurvetoQuadraticRel.h:
        (WebCore::SVGPathSegCurvetoQuadraticRel::create):
        * svg/SVGPathSegCurvetoQuadraticSmoothAbs.h:
        (WebCore::SVGPathSegCurvetoQuadraticSmoothAbs::create):
        * svg/SVGPathSegCurvetoQuadraticSmoothRel.h:
        (WebCore::SVGPathSegCurvetoQuadraticSmoothRel::create):
        * svg/SVGPathSegLinetoAbs.h:
        (WebCore::SVGPathSegLinetoAbs::create):
        * svg/SVGPathSegLinetoHorizontalAbs.h:
        (WebCore::SVGPathSegLinetoHorizontalAbs::create):
        * svg/SVGPathSegLinetoHorizontalRel.h:
        (WebCore::SVGPathSegLinetoHorizontalRel::create):
        * svg/SVGPathSegLinetoRel.h:
        (WebCore::SVGPathSegLinetoRel::create):
        * svg/SVGPathSegLinetoVerticalAbs.h:
        (WebCore::SVGPathSegLinetoVerticalAbs::create):
        * svg/SVGPathSegLinetoVerticalRel.h:
        (WebCore::SVGPathSegLinetoVerticalRel::create):
        * svg/SVGPathSegMovetoAbs.h:
        (WebCore::SVGPathSegMovetoAbs::create):
        * svg/SVGPathSegMovetoRel.h:
        (WebCore::SVGPathSegMovetoRel::create):
        * svg/SVGViewSpec.h:
        (WebCore::SVGViewSpec::create):
        * svg/animation/SMILTimeContainer.h:
        (WebCore::SMILTimeContainer::create):
        * svg/animation/SVGSMILElement.cpp:
        (WebCore::ConditionEventListener::create):
        * svg/graphics/SVGImage.h:
        * svg/graphics/SVGImageForContainer.h:
        * svg/graphics/filters/SVGFilter.cpp:
        (WebCore::SVGFilter::create):
        * svg/graphics/filters/SVGFilter.h:

2015-04-28  Simon Fraser  <simon.fraser@apple.com>

        Make a non-static version of FrameView::yPositionForRootContentLayer()
        https://bugs.webkit.org/show_bug.cgi?id=144375

        Reviewed by Andy Estes.

        There were two calls to the static FrameView::yPositionForRootContentLayer()
        which passed in all the arguments for the same FrameView. Make a member
        function for convenience.

        * page/FrameView.cpp:
        (WebCore::FrameView::yPositionForRootContentLayer):
        * page/FrameView.h:
        * rendering/RenderLayerCompositor.cpp:
        (WebCore::RenderLayerCompositor::updateRootLayerPosition):

2015-04-28  Ryuan Choi  <ryuan.choi@navercorp.com>

        [CoordinatedGraphics] Merge TILED_BACKING_STORE guard with COORDINATED_GRAPHICS
        https://bugs.webkit.org/show_bug.cgi?id=143001

        Reviewed by Gyuyoung Kim.

        TiledBackingStore has only been used by Coordinated Graphics since Qt and WebKit1/Efl were dropped.
        So, this patch replaces USE(TILED_BACKING_STORE) with USE(COORDINATED_GRAPHICS) to merge the features.

        In addition, this moves TiledBackingStore and related files from platform to platform/texmap/coordinated
        where other coordinated graphics files are located.

        * CMakeLists.txt: Move TiledBackingStore.cpp because this is not common file.
        * PlatformEfl.cmake:
        * PlatformGTK.cmake:
        * WebCore.vcxproj/WebCore.vcxproj: Follow new location of TiledBackingStore* files.
        * WebCore.vcxproj/WebCore.vcxproj.filters:
        * loader/EmptyClients.h:
        * page/Chrome.cpp:
        * page/Chrome.h:
        * page/ChromeClient.h:
        * page/Frame.cpp:
        (WebCore::Frame::createView):
        * page/FrameView.cpp:
        (WebCore::FrameView::requestScrollPositionUpdate):
        * page/FrameView.h:
        * page/Page.cpp:
        (WebCore::Page::setPageScaleFactor):
        * platform/HostWindow.h:
        * platform/ScrollView.cpp:
        (WebCore::ScrollView::unscaledVisibleContentSizeIncludingObscuredArea):
        (WebCore::ScrollView::unscaledUnobscuredVisibleContentSize):
        (WebCore::ScrollView::visibleContentRectInternal):
        (WebCore::ScrollView::scrollTo):
        * platform/ScrollView.h:
        * platform/graphics/cairo/TileCairo.h: Removed because it is dead code since r169328
        * platform/graphics/texmap/coordinated/Tile.h: Renamed from Source/WebCore/platform/graphics/Tile.h.
        * platform/graphics/texmap/coordinated/TiledBackingStore.cpp: Renamed from Source/WebCore/platform/graphics/TiledBackingStore.cpp.
        * platform/graphics/texmap/coordinated/TiledBackingStore.h: Renamed from Source/WebCore/platform/graphics/TiledBackingStore.h.
        * platform/graphics/texmap/coordinated/TiledBackingStoreBackend.h: Renamed from Source/WebCore/platform/graphics/TiledBackingStoreBackend.h.
        * platform/graphics/texmap/coordinated/TiledBackingStoreClient.h: Renamed from Source/WebCore/platform/graphics/TiledBackingStoreClient.h.

2015-04-28  Brent Fulgham  <bfulgham@apple.com>

        REGRESSION(180076): [Mac, iOS] Correct possible null dereference in printing code
        https://bugs.webkit.org/show_bug.cgi?id=144366
        <rdar://problem/20533513>

        Reviewed by Dean Jackson.

        * rendering/RenderBlockFlow.cpp:
        (WebCore::needsAppleMailPaginationQuirk): Check if the document settings is a nullptr
        before attempting to dereference it. 

2015-04-28  Andreas Kling  <akling@apple.com>

        Simplify DOM wrapper destruction, don't deref() in finalizers.
        <https://webkit.org/b/144183>

        Reviewed by Darin Adler.

        DOM JS bindings had two mechanisms to call deref() on the WebCore object,
        once through a weak finalizer, and once through the JSCell's regular destructor.

        That was once believed to be an optimization, but these days the finalizer will
        run just moments before the destructor anyway, all in the same call stack.
        And more importantly, the finalizer is not guaranteed to run, for instance in the
        case where a Weak is assigned to after going dead, but before the WeakBlock
        has been swept by the incremental sweeper.

        Simplify this by just removing the deref() from the generated finalizers.
        This makes it easier to reason about DOM wrapper destruction, and eliminates
        the awkward time window  where a DOM wrapper could have a null impl().

        We could spend more time on figuring out a way to have finalizers manage the
        destruction of these wrappers, but that would require fundamental changes to
        our implementation of JSC::Weak pointers. It would allow us to make JSDOMWrapper
        destructor-less, and shrink each wrapper object by 1 pointer (the ClassInfo*.)
        However the risk:reward ratio does not seem justified at this point in time.

        * bindings/scripts/CodeGeneratorJS.pm:
        (GenerateHeader):
        (GenerateImplementation):
        * bindings/js/JSCSSValueCustom.cpp:
        * bindings/scripts/test/JS/JSTestActiveDOMObject.cpp:
        (WebCore::JSTestActiveDOMObject::~JSTestActiveDOMObject):
        (WebCore::JSTestActiveDOMObjectOwner::finalize):
        * bindings/scripts/test/JS/JSTestActiveDOMObject.h:
        * bindings/scripts/test/JS/JSTestCustomNamedGetter.cpp:
        (WebCore::JSTestCustomNamedGetter::~JSTestCustomNamedGetter):
        (WebCore::JSTestCustomNamedGetterOwner::finalize):
        * bindings/scripts/test/JS/JSTestCustomNamedGetter.h:
        * bindings/scripts/test/JS/JSTestEventConstructor.cpp:
        (WebCore::JSTestEventConstructor::~JSTestEventConstructor):
        (WebCore::JSTestEventConstructorOwner::finalize):
        * bindings/scripts/test/JS/JSTestEventConstructor.h:
        * bindings/scripts/test/JS/JSTestEventTarget.cpp:
        (WebCore::JSTestEventTarget::~JSTestEventTarget):
        (WebCore::JSTestEventTargetOwner::finalize):
        * bindings/scripts/test/JS/JSTestEventTarget.h:
        * bindings/scripts/test/JS/JSTestException.cpp:
        (WebCore::JSTestException::~JSTestException):
        (WebCore::JSTestExceptionOwner::finalize):
        * bindings/scripts/test/JS/JSTestException.h:
        * bindings/scripts/test/JS/JSTestGenerateIsReachable.cpp:
        (WebCore::JSTestGenerateIsReachable::~JSTestGenerateIsReachable):
        (WebCore::JSTestGenerateIsReachableOwner::finalize):
        * bindings/scripts/test/JS/JSTestGenerateIsReachable.h:
        * bindings/scripts/test/JS/JSTestInterface.cpp:
        (WebCore::JSTestInterface::~JSTestInterface):
        (WebCore::JSTestInterfaceOwner::finalize):
        * bindings/scripts/test/JS/JSTestInterface.h:
        * bindings/scripts/test/JS/JSTestMediaQueryListListener.cpp:
        (WebCore::JSTestMediaQueryListListener::~JSTestMediaQueryListListener):
        (WebCore::JSTestMediaQueryListListenerOwner::finalize):
        * bindings/scripts/test/JS/JSTestMediaQueryListListener.h:
        * bindings/scripts/test/JS/JSTestNamedConstructor.cpp:
        (WebCore::JSTestNamedConstructor::~JSTestNamedConstructor):
        (WebCore::JSTestNamedConstructorOwner::finalize):
        * bindings/scripts/test/JS/JSTestNamedConstructor.h:
        * bindings/scripts/test/JS/JSTestNondeterministic.cpp:
        (WebCore::JSTestNondeterministic::~JSTestNondeterministic):
        (WebCore::JSTestNondeterministicOwner::finalize):
        * bindings/scripts/test/JS/JSTestNondeterministic.h:
        * bindings/scripts/test/JS/JSTestObj.cpp:
        (WebCore::JSTestObj::~JSTestObj):
        (WebCore::JSTestObjOwner::finalize):
        * bindings/scripts/test/JS/JSTestObj.h:
        * bindings/scripts/test/JS/JSTestOverloadedConstructors.cpp:
        (WebCore::JSTestOverloadedConstructors::~JSTestOverloadedConstructors):
        (WebCore::JSTestOverloadedConstructorsOwner::finalize):
        * bindings/scripts/test/JS/JSTestOverloadedConstructors.h:
        * bindings/scripts/test/JS/JSTestSerializedScriptValueInterface.cpp:
        (WebCore::JSTestSerializedScriptValueInterface::~JSTestSerializedScriptValueInterface):
        (WebCore::JSTestSerializedScriptValueInterfaceOwner::finalize):
        * bindings/scripts/test/JS/JSTestSerializedScriptValueInterface.h:
        * bindings/scripts/test/JS/JSTestTypedefs.cpp:
        (WebCore::JSTestTypedefs::~JSTestTypedefs):
        (WebCore::JSTestTypedefsOwner::finalize):
        * bindings/scripts/test/JS/JSTestTypedefs.h:
        * bindings/scripts/test/JS/JSattribute.cpp:
        (WebCore::JSattribute::~JSattribute):
        (WebCore::JSattributeOwner::finalize):
        * bindings/scripts/test/JS/JSattribute.h:
        * bindings/scripts/test/JS/JSreadonly.cpp:
        (WebCore::JSreadonly::~JSreadonly):
        (WebCore::JSreadonlyOwner::finalize):
        * bindings/scripts/test/JS/JSreadonly.h:

2015-04-28  Alex Christensen  <achristensen@webkit.org>

        Build WinCairo without cygwin.
        https://bugs.webkit.org/show_bug.cgi?id=144365

        Reviewed by Myles Maxfield.

        * WebCore.vcxproj/WebCoreGeneratedWinCairo.make:
        Pass the CC executable to build-generated-files.pl like r182164.
        * WebCore.vcxproj/build-generated-files.pl:
        Default to 8 CPUs unless otherwise specified.

2015-04-28  Sam Weinig  <sam@webkit.org>

        [Content Extensions] Process NFAs individually to avoid having all NFAs live at the same time
        https://bugs.webkit.org/show_bug.cgi?id=144363

        Reviewed by Alex Christensen.

        This brings dirty memory use when compiling our test content extension down from ~300MB to ~100MB.

        * contentextensions/CombinedURLFilters.cpp:
        (WebCore::ContentExtensions::CombinedURLFilters::processNFAs):
        (WebCore::ContentExtensions::CombinedURLFilters::createNFAs): Deleted.
        * contentextensions/CombinedURLFilters.h:
        Replace function that creates a Vector of all the NFAs with one that allows incremental processing
        as they are created.

        * contentextensions/ContentExtensionCompiler.cpp:
        (WebCore::ContentExtensions::addUniversalActionsToDFA):
        Extract code to add universal actions into a helper, since we need to call it in two places now.

        (WebCore::ContentExtensions::compileRuleList):
        Adopt CombinedURLFilters::processNFAs. Now that we don't have a Vector of NFAs, we need to keep track
        of whether or not any NFAs were processed and if we are currently processing the first NFA so we can
        ensure that we have some bytecode generated event for empty rule sets, and that universal actions are
        placed on the first DFA.

2015-04-28  Timothy Horton  <timothy_horton@apple.com>

        [TextIndicator] Yellow highlight takes too long to fade out on scroll
        https://bugs.webkit.org/show_bug.cgi?id=144358
        <rdar://problem/19451011>

        Reviewed by Beth Dakin.

        * page/TextIndicator.h:
        Add Lifetime and DismissalAnimation enums.

        (WebCore::TextIndicator::contentImageWithHighlight):
        (WebCore::TextIndicator::contentImage):
        Fix style.

        * page/mac/TextIndicatorWindow.h:
        * page/mac/TextIndicatorWindow.mm:
        (WebCore::TextIndicatorWindow::TextIndicatorWindow):
        (WebCore::TextIndicatorWindow::~TextIndicatorWindow):
        (WebCore::TextIndicatorWindow::clearTextIndicator):
        (WebCore::TextIndicatorWindow::setTextIndicator):
        (WebCore::TextIndicatorWindow::closeWindow):
        (WebCore::TextIndicatorWindow::startFadeOut):
        Rename m_startFadeOutTimer to m_temporaryTextIndicatorTimer (and related).
        This is just about temporary-lifetime TextIndicators, like the ones
        you get when the find-in-page UI is hidden but you hit Cmd-G.

        Add clearTextIndicator, which takes a DismissalAnimation, providing
        clients an opportunity to avoid the normal fade-out animation, if it
        was going to happen.

2015-04-28  Michael Catanzaro  <mcatanzaro@igalia.com>

        Fully replace ENABLE_LLINT_C_LOOP with ENABLE_JIT
        https://bugs.webkit.org/show_bug.cgi?id=144304

        Reviewed by Geoffrey Garen.

        Define ENABLE_JIT, enabled by default, instead of ENABLE_LLINT_C_LOOP, disabled by default.

        * Configurations/FeatureDefines.xcconfig:

2015-04-28  Commit Queue  <commit-queue@webkit.org>

        Unreviewed, rolling out r183514.
        https://bugs.webkit.org/show_bug.cgi?id=144359

        It broke cloop test bots (Requested by mcatanzaro on #webkit).

        Reverted changeset:

        "Fully replace ENABLE_LLINT_C_LOOP with ENABLE_JIT"
        https://bugs.webkit.org/show_bug.cgi?id=144304
        http://trac.webkit.org/changeset/183514

2015-04-28  Michael Catanzaro  <mcatanzaro@igalia.com>

        Fully replace ENABLE_LLINT_C_LOOP with ENABLE_JIT
        https://bugs.webkit.org/show_bug.cgi?id=144304

        Reviewed by Geoffrey Garen.

        Define ENABLE_JIT, enabled by default, instead of ENABLE_LLINT_C_LOOP, disabled by default.

        * Configurations/FeatureDefines.xcconfig:

2015-04-28  Zalan Bujtas  <zalan@apple.com>

        Checkboxes on bugs.webkit.org are painted with stripes at some zoom levels.
        https://bugs.webkit.org/show_bug.cgi?id=144351

        Reviewed by Simon Fraser.

        This patch ensures that CG context is properly restored after painting dashed/dotted lines.

        Test: fast/forms/checkbox-painting-with-hr.html

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

2015-04-28  Simon Fraser  <simon.fraser@apple.com>

        Provide contentsToView() and viewToContents() functions on ScrollView, and use them
        https://bugs.webkit.org/show_bug.cgi?id=144357

        Reviewed by Tim Horton.

        Too much code was consulting topContentInset() and headerHeight() directly. Replace
        with calls to new contentsToView() and viewToContents() functions, which wrap the
        exisiting documentScrollOffsetRelativeToViewOrigin().
        
        Use the new functions in FrameView and ScrollView coordinate mapping functions.
        
        No behavior change.

        * page/FrameView.cpp:
        (WebCore::FrameView::convertFromRendererToContainingView):
        (WebCore::FrameView::convertFromContainingViewToRenderer):
        * platform/ScrollView.cpp:
        (WebCore::ScrollView::viewToContents):
        (WebCore::ScrollView::contentsToView):
        (WebCore::ScrollView::rootViewToContents):
        (WebCore::ScrollView::contentsToRootView):
        (WebCore::ScrollView::rootViewToTotalContents):
        (WebCore::ScrollView::windowToContents):
        (WebCore::ScrollView::contentsToWindow):
        * platform/ScrollView.h:

2015-04-28  Eric Carlson  <eric.carlson@apple.com>

        [Mac] Simplify code to support media engines which do not support target playback
        https://bugs.webkit.org/show_bug.cgi?id=144332

        Reviewed by Jer Noble.

        * Modules/mediasession/WebMediaSessionManager.cpp:
        (WebCore::WebMediaSessionManager::externalOutputDeviceAvailableDidChange): Always make client
        callback, let them decide if it is significant or not.

        * html/HTMLMediaElement.cpp:
        (WebCore::HTMLMediaElement::HTMLMediaElement): m_loadTimer -> m_pendingActionTimer.
        (WebCore::HTMLMediaElement::scheduleDelayedAction): Handle CheckPlaybackTargetCompatablity.
        (WebCore::HTMLMediaElement::scheduleNextSourceChild): m_loadTimer -> m_pendingActionTimer.
        (WebCore::HTMLMediaElement::loadTimerFired): Renamed pendingActionTimerFired.
        (WebCore::HTMLMediaElement::prepareForLoad): m_loadTimer -> m_pendingActionTimer.
        (WebCore::HTMLMediaElement::setDefaultPlaybackRate): Add logging.
        (WebCore::HTMLMediaElement::clearMediaPlayer): m_loadTimer -> m_pendingActionTimer.
        (WebCore::HTMLMediaElement::webkitCurrentPlaybackTargetIsSupported): Removed.
        (WebCore::HTMLMediaElement::dispatchEvent): If a 'webkitcurrentplaybacktargetiswirelesschanged'
        event is dispatched when the current target is wireless but the media engine does not support
        wireless playback, tell the media engine not to play to the target.
        * html/HTMLMediaElement.h:
        * html/HTMLMediaElement.idl:

        * html/HTMLMediaSession.cpp:
        (WebCore::HTMLMediaSession::showPlaybackTargetPicker): Drive-by fix to disallow audio-only files.
        (WebCore::HTMLMediaSession::currentPlaybackTargetIsSupported): Deleted.
        * html/HTMLMediaSession.h:

        * platform/graphics/MediaPlayer.cpp:
        (WebCore::MediaPlayer::isCurrentPlaybackTargetSupported): Deleted.
        * platform/graphics/MediaPlayer.h:
        * platform/graphics/MediaPlayerPrivate.h:

        * platform/graphics/avfoundation/objc/MediaPlayerPrivateAVFoundationObjC.mm:
        (WebCore::MediaPlayerPrivateAVFoundationObjC::setShouldPlayToPlaybackTarget): Use a RetainPtr
        to explicitly manage the lifetime of the temporary object.
        (WebCore::MediaPlayerPrivateAVFoundationObjC::isPlayingToWirelessPlaybackTarget): Ditto.
        * platform/graphics/avfoundation/objc/MediaPlayerPrivateMediaSourceAVFObjC.h:

        * platform/graphics/avfoundation/objc/MediaPlayerPrivateMediaSourceAVFObjC.mm:
        (WebCore::MediaPlayerPrivateMediaSourceAVFObjC::setShouldPlayToPlaybackTarget):
        (WebCore::MediaPlayerPrivateMediaSourceAVFObjC::isCurrentPlaybackTargetWireless):
        (WebCore::MediaPlayerPrivateMediaSourceAVFObjC::isCurrentPlaybackTargetSupported): Deleted.

        * platform/graphics/mac/MediaPlayerPrivateQTKit.h:
        * platform/graphics/mac/MediaPlayerPrivateQTKit.mm:
        (WebCore::MediaPlayerPrivateQTKit::setShouldPlayToPlaybackTarget):
        (WebCore::MediaPlayerPrivateQTKit::isCurrentPlaybackTargetWireless):
        (WebCore::MediaPlayerPrivateQTKit::isCurrentPlaybackTargetSupported): Deleted.

2015-04-28  Alex Christensen  <achristensen@webkit.org>

        [Content Extensions] Use less memory for CombinedURLFilters.
        https://bugs.webkit.org/show_bug.cgi?id=144290

        Reviewed by Andreas Kling.

        * contentextensions/CombinedURLFilters.cpp:
        (WebCore::ContentExtensions::recursiveMemoryUsed):
        (WebCore::ContentExtensions::CombinedURLFilters::addPattern):
        (WebCore::ContentExtensions::generateNFAForSubtree):
        (WebCore::ContentExtensions::CombinedURLFilters::createNFAs):
        * contentextensions/NFA.cpp:
        (WebCore::ContentExtensions::NFA::memoryUsed):
        (WebCore::ContentExtensions::NFA::setActions):
        * contentextensions/NFA.h:
        * contentextensions/NFANode.h:
        * contentextensions/Term.h:
        (WebCore::ContentExtensions::Term::Term::generateGraph):
        (WebCore::ContentExtensions::Term::generateSubgraphForAtom):
        Use Vectors instead of HashTables in PrefixTreeVertex because the sets stay small and need to be more memory efficient.

2015-04-28  Brady Eidson  <beidson@apple.com>

        Consolidate most "frame load" arguments into FrameLoadRequest.
        https://bugs.webkit.org/show_bug.cgi?id=144276

        Reviewed by Alexey Proskuryakov.

        No new tests (No change in behavior).
        
        This patch starts the long overdue process of wrangling the insane load methods in FrameLoader.
        
        Humble beginnings of this long process:
        - Put most of the various "frame load" arguments that are passed around within FrameLoader on FrameLoadRequest.
        - Get rid of the "easy" constructors of FrameLoadRequest, forcing users to actually think about what they're doing.
        - Change a private FrameLoader load method (urlSelected) to take a FrameLoadRequest instead of argument gobbledygook.
        - Change a public FrameLoader load methods (changeLocation) to take a FrameLoadRequest instead of gobbledygook.
        
        These cover the straightforward changes where it was easy to figure out what the values of the FrameLoadRequest
        arguments were from these various call sites.
        
        Further refactoring can be done in much smaller patches, handling fewer cases at a time. Little by little we will be
        able to reduce the number of "load" methods on FrameLoader and make sure that they mainly only take a FrameLoadRequest
        as an argument.

        * inspector/InspectorFrontendClientLocal.cpp:
        (WebCore::InspectorFrontendClientLocal::openInNewTab):
        
        * inspector/InspectorPageAgent.cpp:
        (WebCore::InspectorPageAgent::navigate):
        
        * loader/FrameLoadRequest.cpp:
        (WebCore::FrameLoadRequest::FrameLoadRequest):
        * loader/FrameLoadRequest.h:
        (WebCore::FrameLoadRequest::FrameLoadRequest):
        (WebCore::FrameLoadRequest::lockHistory):
        (WebCore::FrameLoadRequest::lockBackForwardList):
        (WebCore::FrameLoadRequest::shouldSendReferrer):
        (WebCore::FrameLoadRequest::allowNavigationToInvalidURL):
        (WebCore::FrameLoadRequest::newFrameOpenerPolicy):
        (WebCore::FrameLoadRequest::shouldReplaceDocumentIfJavaScriptURL):
        
        * loader/FrameLoader.cpp:
        (WebCore::FrameLoader::changeLocation):
        (WebCore::FrameLoader::urlSelected):
        (WebCore::FrameLoader::loadFrameRequest):
        * loader/FrameLoader.h:
        
        * loader/NavigationScheduler.cpp:
        (WebCore::NavigationScheduler::scheduleLocationChange):
        
        * page/ContextMenuController.cpp:
        (WebCore::openNewWindow):
        (WebCore::ContextMenuController::contextMenuItemSelected):
        
        * page/DOMWindow.cpp:
        (WebCore::DOMWindow::createWindow):

2015-04-27  Myles C. Maxfield  <mmaxfield@apple.com>

        [Mac] [iOS] Implement font-synthesis CSS property
        https://bugs.webkit.org/show_bug.cgi?id=144305

        Reviewed by Andreas Kling.

        This patch is fairly straightforward, though there are some particular pieces to the patch listed below.

        The implementation of this font-synthesis CSS property lives inside FontCacheIOS and FontCacheMac. We already
        determine if we should synthesize bold and italics, this CSS property simply guards that decision.

        Because FontCache interacts with FontDescriptions and not RenderStyles, this patch moves the source of truth
        regarding font-synthesis into FontDescription (from RenderStyle).

        Test: fast/css3-text/font-synthesis.html

        * css/CSSPropertyNames.in: Mark font-synthesis as a FontProperty, which means that the relevant state is kept
        inside FontDescription. This also means that the property must be marked as a "high priority" property, which
        is enforced by it existing at the top of CSSPropertyNames.in. The move to FontDescription must be done so that
        FontCacheIOS and FontCacheMac have access to the saved state.
        * css/CSSComputedStyleDeclaration.cpp:
        (WebCore::fontSynthesisFromStyle): Update to look for state in FontDescription instead of RenderStyle.
        * css/CSSSegmentedFontFace.cpp:
        (WebCore::CSSSegmentedFontFace::fontRanges): Guard the synthetic bold and synthic italic decision on the new
        FontSynthesis state inside FontDescription.
        * css/StyleBuilderConverter.h:
        (WebCore::StyleBuilderConverter::convertFontSynthesis): Moved from
        StyleBuilderCustom::applyValueFontSynthesis(). This is because of the updated options inside of
        CSSPropertyNames.in
        * css/StyleBuilderCustom.h:
        (WebCore::StyleBuilderCustom::applyValueFontSynthesis): Deleted.
        * platform/graphics/FontDescription.h:
        (WebCore::FontDescription::FontDescription): Moved from RenderStyle.
        (WebCore::FontDescription::fontSynthesis): Ditto.
        (WebCore::FontDescription::setFontSynthesis): Ditto.
        (WebCore::FontDescription::initialFontSynthesis): Ditto.
        (WebCore::FontDescription::operator==): Update to include new state.
        * platform/graphics/FontCache.h:
        (WebCore::FontDescriptionFontDataCacheKey::makeFlagKey): Make FontDescription hashes sensitive to the new
        state inside FontDescription.
        * platform/graphics/ios/FontCacheIOS.mm:
        (WebCore::FontCache::createFontPlatformData): Guard the synthetic bold and synthetic italic decision on the
        new FontSynthesis state inside FontDescription.
        * platform/graphics/mac/FontCacheMac.mm:
        (WebCore::FontCache::createFontPlatformData): Ditto.
        * platform/text/TextFlags.h: Move FontSynthesis type from RenderStyleConstants.
        * rendering/style/RenderStyle.h:
        (WebCore::RenderStyle::fontSynthesis): Moved to FontDescription.
        (WebCore::RenderStyle::setFontSynthesis): Ditto.
        (WebCore::RenderStyle::initialFontSynthesis): Ditto.
        * rendering/style/RenderStyleConstants.h: Moved FontSynthesis type to TextFlags.
        * rendering/style/StyleRareInheritedData.h: Remove state regarding font synthesis (moved to FontDescription)

2015-04-28  Youenn Fablet  <youenn.fablet@crf.canon.fr>

        SharedBuffer::copy is not computing the buffer size correctly when having m_dataArray
        https://bugs.webkit.org/show_bug.cgi?id=144321

        Reviewed by Darin Adler.

        Patch correctness covered by existing tests.

        * platform/SharedBuffer.cpp:
        (WebCore::SharedBuffer::copy): Direct appending to m_dataArray and not using append method as this method updates the SharedBuffer size.

2015-04-28  Per Arne Vollan  <peavo@outlook.com>

        [Curl] Memory leak.
        https://bugs.webkit.org/show_bug.cgi?id=144327

        Reviewed by Brent Fulgham.

        When a data url is loaded, the ResourceHandle is not released.

        * platform/network/curl/ResourceHandleManager.cpp:
        (WebCore::ResourceHandleManager::startJob):

2015-04-28  Andy Estes  <aestes@apple.com>

        Fix the Mavericks Debug build after r183467.

        * platform/network/ResourceRequestBase.h:
        (WebCore::ResourceRequestBase::requester): Removed WEBCORE_EXPORT.
        (WebCore::ResourceRequestBase::setRequester): Ditto.

2015-04-28  Csaba Osztrogonác  <ossy@webkit.org>

        Remove make-file-arrays.py
        https://bugs.webkit.org/show_bug.cgi?id=144324

        Reviewed by Gyuyoung Kim.

        * make-file-arrays.py: Removed.

2015-04-28  Antti Koivisto  <antti@apple.com>

        Build fix.

        * platform/network/ios/ResourceRequestIOS.mm:
        (WebCore::ResourceRequest::ResourceRequest):

2015-04-28  Antti Koivisto  <antti@apple.com>

        Network Cache: Disk cache getting filled by YouTube video data
        https://bugs.webkit.org/show_bug.cgi?id=144259

        Reviewed by Darin Adler.

        MSE media is loaded via XHR and tends to eventually fill the cache.

        YouTube serves the media chunks cacheable, however they are rarely (if ever) reused.
        We can reduce disk writes and keep more useful resources around by not caching them

        Test: http/tests/cache/disk-cache/disk-cache-media.html

        * loader/DocumentLoader.cpp:
        (WebCore::DocumentLoader::startLoadingMainResource):

            Set the requester.

        * loader/cache/CachedRawResource.cpp:
        (WebCore::CachedRawResource::CachedRawResource):
        * page/DiagnosticLoggingKeys.cpp:
        (WebCore::DiagnosticLoggingKeys::streamingMedia):
        * page/DiagnosticLoggingKeys.h:
        * platform/network/ResourceRequestBase.cpp:
        (WebCore::ResourceRequestBase::adopt):
        (WebCore::ResourceRequestBase::copyData):
        (WebCore::equalIgnoringHeaderFields):
        * platform/network/ResourceRequestBase.h:
        (WebCore::ResourceRequestBase::requester):
        (WebCore::ResourceRequestBase::setRequester):

            Add requester type to the request object. Currently this is main resource, xhr or unspecified.

        (WebCore::ResourceRequestBase::encodeWithoutPlatformData):
        (WebCore::ResourceRequestBase::decodeWithoutPlatformData):
        * platform/network/cf/ResourceRequest.h:
        (WebCore::ResourceRequest::deprecatedSetMainResourceRequest): Deleted.
        (WebCore::ResourceRequest::deprecatedIsMainResourceRequest): Deleted.

            Replace this iOS only field with shared mechanism.

        * platform/network/ios/QuickLook.mm:
        (WebCore::QuickLookHandle::create):
        * xml/XMLHttpRequest.cpp:
        (WebCore::XMLHttpRequest::createRequest):

            Set the requester.

2015-04-28  Namhoon Kim  <nakim@ea.com>

        Fix windows build error in WebCore related to bulk build.
        https://bugs.webkit.org/show_bug.cgi?id=144313

        Reviewed by Csaba Osztrogonác.

        No new tests because there is no behavior change.

        * css/CSSAllInOne.cpp:
        * dom/DOMAllInOne.cpp:

2015-04-27  Simon Fraser  <simon.fraser@apple.com>

        Eliminate styleDidChange with StyleDifferenceEqual when updates are actually necessary
        https://bugs.webkit.org/show_bug.cgi?id=144198

        Followup: fix assertions seen in fullscreen and pseudo-element tests. These code
        paths set the style to the existing pointer, but with a SyntheticStyleChange.
        We have to avoid an early return in this case.

        * rendering/RenderElement.cpp:
        (WebCore::RenderElement::setStyle):

2015-04-25  Simon Fraser  <simon.fraser@apple.com>

        Eliminate styleDidChange with StyleDifferenceEqual when updates are actually necessary
        https://bugs.webkit.org/show_bug.cgi?id=144198

        Reviewed by Darin Adler, Antti Koivisto.
        
        SyntheticStyleChange style recalcs are triggered for cases where behavior depends
        on state which is outside of RenderStyle; this includes triggering compositing for
        animations, for video and canvas, and for iframes with composited content.
        
        In these cases, we'd run through RenderElement::setStyle() and its fan-out, but
        with diff == StyleDifferenceEqual, and so be unable to determine if there
        is actual work to be done.
        
        This patch enforces the contract that the diff is never StyleDifferenceEqual if
        compositing or other work has to happen from setStyle(). This is achieved by
        passing in a 'hasSideEffects' flag, which causes the diff to become at least
        StyleDifferenceRecompositeLayer.
        
        RenderLayerCompositor::layerStyleChanged() can now safely early return
        if the diff is equal. Future patches will reduce redundant work even more.

        Test: compositing/animation/no-style-recalc-during-accelerated-animation.html

        * page/animation/AnimationBase.h:
        (WebCore::AnimationBase::animate): Returns a bool now if the state changed.
        (WebCore::AnimationBase::state):
        * page/animation/AnimationController.cpp:
        (WebCore::AnimationController::updateAnimations): bool out param which indicates
        whether any animations changed state.
        * page/animation/AnimationController.h:
        * page/animation/CompositeAnimation.cpp:
        (WebCore::CompositeAnimation::animate): If any transitions or animations changed
        state, set the animationStateChanged out param to true.
        * page/animation/CompositeAnimation.h:
        * page/animation/ImplicitAnimation.cpp:
        (WebCore::ImplicitAnimation::animate): Return true if the state changed.
        * page/animation/ImplicitAnimation.h:
        * page/animation/KeyframeAnimation.cpp:
        (WebCore::KeyframeAnimation::animate): Return true if the state changed.
        * page/animation/KeyframeAnimation.h:
        * rendering/RenderElement.cpp:
        (WebCore::RenderElement::adjustStyleDifference): We may enter here now with diff
        != StyleDifferenceEqual, but still need to do the check to see if layers changed.
        (WebCore::RenderElement::initializeStyle): When setting style for the first time,
        don't use StyleDifferenceEqual.
        (WebCore::RenderElement::setStyle): Additional flag to indicate whether this style
        change involves side effects. If the diff is equal but the flag is set, change
        the diff to StyleDifferenceRecompositeLayer (the "lowest" non-zero diff).
        * rendering/RenderElement.h:
        (WebCore::RenderElement::setAnimatableStyle): Pass true to setStyle() if hasSideEffects
        is true, or if animation state changed.
        * rendering/RenderLayer.cpp:
        (WebCore::RenderLayer::styleChanged): Pass the diff down.
        * rendering/RenderLayerCompositor.cpp:
        (WebCore::RenderLayerCompositor::layerStyleChanged): Return if the diff is equal.
        * rendering/RenderLayerCompositor.h:
        * rendering/style/RenderStyleConstants.h: StyleDifferenceNewStyle is used when
        setting style for the first time.
        * style/StyleResolveTree.cpp:
        (WebCore::Style::createRendererIfNeeded): Provide animationsChanged bool (which is unused).
        (WebCore::Style::resolveLocal): If the style change is synthetic, set the flag that
        says there are side-effects.

2015-04-27  Michael Catanzaro  <mcatanzaro@igalia.com>

        [GTK] Add one single option to control all OpenGL-related options
        https://bugs.webkit.org/show_bug.cgi?id=144105

        Reviewed by Martin Robinson.

        Test WTF_USE_TEXTURE_MAPPER since USE_TEXTURE_MAPPER has been removed.

        * PlatformGTK.cmake:

2015-04-27  Myles C. Maxfield  <mmaxfield@apple.com>

        Consolidate one-line flag-related header files into TextFlags.h
        https://bugs.webkit.org/show_bug.cgi?id=144295

        Reviewed by Tim Horton.

        There were a collection of single-line header files throughout platform/ which contain
        single-line type declaractions of flags related to text layout & rendering. This patch
        consolidates all these single-line headers into TextFlags.h

        No new tests because there is no behavior change.

        * WebCore.vcxproj/WebCore.vcxproj:
        * WebCore.vcxproj/WebCore.vcxproj.filters:
        * WebCore.xcodeproj/project.pbxproj:
        * css/CSSFontFace.h:
        * css/CSSPrimitiveValueMappings.h:
        * loader/cache/CachedFont.h:
        * loader/cache/CachedTextTrack.h:
        * page/Settings.h:
        * platform/DragImage.h:
        * platform/graphics/FontDescription.h:
        * platform/graphics/FontOrientation.h: Removed.
        * platform/graphics/FontPlatformData.h:
        * platform/graphics/FontRenderingMode.h: Removed.
        * platform/graphics/FontSmoothingMode.h: Removed.
        * platform/graphics/FontTraitsMask.h: Removed.
        * platform/graphics/FontWidthVariant.h: Removed.
        * platform/graphics/TextRenderingMode.h: Removed.
        * platform/graphics/cairo/FontCustomPlatformData.h:
        * platform/graphics/freetype/FontPlatformData.h:
        * platform/graphics/mac/FontCustomPlatformData.h:
        * platform/graphics/win/FontCustomPlatformData.h:
        * platform/text/NonCJKGlyphOrientation.h: Removed.
        * platform/text/TextFlags.h:
        * rendering/TextPainter.h:
        * style/StyleFontSizeFunctions.h:

2015-04-27  Daniel Bates  <dabates@apple.com>

        Form control may be associated with the wrong HTML Form element after form id change
        https://bugs.webkit.org/show_bug.cgi?id=133456
        <rdar://problem/17095055>

        Reviewed by Andy Estes.

        Fixes an issue where a form control may be associated with the wrong HTML Form element
        after the id of the HTML Form element associated with the form control is changed when
        there is more than one HTML Form element with the same id in the document. Specifically,
        a form control that has an HTML form attribute value X will always be associated with
        some HTML Form element f where f.id = X regardless of whether f.id is subsequently
        changed.

        Tests: fast/forms/change-form-id-to-be-unique-then-submit-form.html
               fast/forms/change-form-id-to-be-unique.html

        * dom/Element.cpp:
        (WebCore::Element::attributeChanged): Notify observers when the id of an element changed.
        (WebCore::Element::updateId): Added parameter NotifyObservers (defaults to NotifyObservers::Yes),
        as to whether we should notify observers of the id change.
        (WebCore::Element::updateIdForTreeScope): Ditto.
        (WebCore::Element::willModifyAttribute): Do not notify observers of the id change immediately. As
        indicated by the name of this method, we plan to modify the DOM attribute id of the element, but
        we have not actually modified it when this method is called. Instead we will notify observers
        in Element::attributeChanged(), which is called after the DOM attribute id is modified.
        (WebCore::Element::cloneAttributesFromElement): Ditto.
        * dom/Element.h: Defined enum class NotifyObservers.
        * dom/TreeScope.cpp:
        (WebCore::TreeScope::addElementById): Added boolean parameter notifyObservers (defaults to true)
        as to whether we should dispatch a notification to all observers.
        (WebCore::TreeScope::removeElementById): Ditto.
        * dom/TreeScope.h:

2015-04-27  Alex Christensen  <achristensen@webkit.org>

        Reduce allocations and memory usage when compiling content extensions.
        https://bugs.webkit.org/show_bug.cgi?id=144277

        Reviewed by Benjamin Poulain.

        Covered by existing tests.

        * contentextensions/Term.h:
        (WebCore::ContentExtensions::Term::CharacterSet::set):
        (WebCore::ContentExtensions::Term::CharacterSet::get):
        (WebCore::ContentExtensions::Term::CharacterSet::invert):
        (WebCore::ContentExtensions::Term::CharacterSet::inverted):
        (WebCore::ContentExtensions::Term::CharacterSet::bitCount):
        (WebCore::ContentExtensions::Term::CharacterSet::operator==):
        (WebCore::ContentExtensions::Term::CharacterSet::hash):
        (WebCore::ContentExtensions::Term::Term):
        (WebCore::ContentExtensions::Term::addCharacter):
        (WebCore::ContentExtensions::Term::isEndOfLineAssertion):
        (WebCore::ContentExtensions::Term::isUniversalTransition):
        (WebCore::ContentExtensions::Term::generateSubgraphForAtom):
        Use two uint64_t's instead of a BitVector with a capacity of 128 bits.

2015-04-27  Michael Catanzaro  <mcatanzaro@igalia.com>

        Rename WTF_USE_3D_GRAPHICS to ENABLE_GRAPHICS_CONTEXT_3D
        https://bugs.webkit.org/show_bug.cgi?id=144193

        Reviewed by Darin Adler.

        * CMakeLists.txt:
        * platform/graphics/ANGLEWebKitBridge.cpp:
        * platform/graphics/FormatConverter.cpp:
        * platform/graphics/FormatConverter.h:
        * platform/graphics/GLContext.h:
        * platform/graphics/GraphicsContext3D.cpp:
        * platform/graphics/GraphicsContext3DPrivate.cpp:
        * platform/graphics/OpenGLShims.cpp:
        * platform/graphics/cairo/GraphicsContext3DCairo.cpp:
        * platform/graphics/cg/GraphicsContext3DCG.cpp:
        * platform/graphics/egl/GLContextEGL.cpp:
        * platform/graphics/egl/GLContextEGL.h:
        * platform/graphics/glx/GLContextGLX.cpp:
        * platform/graphics/glx/GLContextGLX.h:
        * platform/graphics/mac/GraphicsContext3DMac.mm:
        * platform/graphics/opengl/Extensions3DOpenGL.cpp:
        * platform/graphics/opengl/Extensions3DOpenGLCommon.cpp:
        * platform/graphics/opengl/Extensions3DOpenGLES.cpp:
        * platform/graphics/opengl/GLPlatformContext.cpp:
        * platform/graphics/opengl/GLPlatformSurface.cpp:
        * platform/graphics/opengl/GraphicsContext3DOpenGL.cpp:
        * platform/graphics/opengl/GraphicsContext3DOpenGLCommon.cpp:
        * platform/graphics/opengl/GraphicsContext3DOpenGLES.cpp:
        * platform/graphics/opengl/TemporaryOpenGLSetting.cpp:
        * platform/graphics/win/GraphicsContext3DWin.cpp:

2015-04-27  Jer Noble  <jer.noble@apple.com>

        [iOS] Video not centered in element on retina devices
        https://bugs.webkit.org/show_bug.cgi?id=144274

        Reviewed by Simon Fraser.

        In r173702, a transform was added to the video layer (and a matching, inverse transform
        in the UIProcess), but this transform affects the position property of the video layer
        used to position the video content within the element bounds when their aspect ratios
        do not match.

        To work around this problem, pre-apply the transform during -setPosition:.

        * platform/graphics/avfoundation/objc/MediaPlayerPrivateAVFoundationObjC.mm:
        (-[WebVideoContainerLayer setPosition:]):

2015-04-27  Jer Noble  <jer.noble@apple.com>

        Add a setting & restriction which prevents non-interactivte playback of audible media elements.
        https://bugs.webkit.org/show_bug.cgi?id=143486

        Reviewed by Eric Carlson.

        Tests: media/audio-playback-restriction-autoplay.html
               media/audio-playback-restriction-play.html

        To allow clients who want to allow non-user-interactive video-only playback, but still
        restrict playback of audible media elements, add a new setting and matching restriction
        which disallows playback of media elements containing audible characteristics.

        * html/HTMLMediaElement.cpp:
        (WebCore::HTMLMediaElement::HTMLMediaElement): Set the RequireUserGestureForAudioRateChange
            restriction if the audioPlaybackRequiresUserGesture() setting is set.
        (WebCore::HTMLMediaElement::parseAttribute): Drive-by fix. Move the opening brace
            out of the #if, as this confuses the heck out of diff and makes all subsequent
            changes appear to be within HTMLMediaElement::parseAttribute.
        (WebCore::HTMLMediaElement::autoplay): Remove the restriction check from within autoplay().
            It is checked again immediately after every autoplay() call site.
        (WebCore::HTMLMediaElement::pauseInternal): Remove the iOS-only #if.
        (WebCore::HTMLMediaElement::mediaPlayerDidAddAudioTrack): Check whether playback
            is permitted, and if not, pause.
        (WebCore::HTMLMediaElement::mediaPlayerCharacteristicChanged): Ditto.
        (WebCore::HTMLMediaElement::removeBehaviorsRestrictionsAfterFirstUserGesture): Ditto.
        * html/HTMLMediaSession.cpp:
        (WebCore::restrictionName): Since BehaviorRestrictions is a bitfield, check each bit
            individually.
        (WebCore::HTMLMediaSession::removeBehaviorRestriction): Handle RequireUserGestureForAudioRateChange.
        (WebCore::HTMLMediaSession::playbackPermitted): Check whether the element has audio and 
            audio playback is restricted, and return false if so.
        * html/HTMLMediaSession.h:
        * page/Settings.cpp:
        * page/Settings.in:
        * testing/Internals.cpp:
        (WebCore::Internals::setMediaSessionRestrictions): Added. 
        (WebCore::Internals::setMediaElementRestrictions): Added.
        * testing/Internals.h:
        * testing/Internals.idl:

2015-04-27  Jer Noble  <jer.noble@apple.com>

        [WebAudio] AudioContext does not remove user-gesture restriction during resume()
        https://bugs.webkit.org/show_bug.cgi?id=144211

        Reviewed by Eric Carlson.

        Tests: webaudio/audiocontext-restriction-audiobuffersourcenode-start.html
               webaudio/audiocontext-restriction.html

        Before the introduction of resume(), suspend(), and stop(), AudioContexts which required
        a user-gesture would start normally, but would effectively mute their outputs. Now that
        the AudioContext's state property is exposed to JavaScript, the AudioContext should stay
        in the "suspended" state until the user-gesture restriction is lifted.

        Add a new method, willBeginPlayback() which checks and potentially clears the context's
        behavior restrictions before checking with the MediaSession. Call this new willBeginPlayback()
        method when the state would transition to "running".

        Because they may be called before any nodes are created, make sure to call lazyInitialize()
        from within the JS-exposed resumePlayback(), suspendPlayback(), and stopPlayback() methods.

        Instead of clearing the behavior restrictions directly, scheduled AudioNodes should instead
        call a new method nodeWillBeginPlayback(). Because existing sites will call AudioNode.start()
        inside a user-gesture handler to clear the user-gesture restriction, call startRendering()
        from nodeWillBeginPlayback(). But because we don't want AudioNode.start() to resume playback
        unconditionally, only do so when the user-gesture restriction is set.

        Now that an AudioContext will not transition to "running" state without a user-gesture (if
        that restriction is set), there's no reason to check for that restriction from inside
        AudioDestinationNode.

        Add some internal methods to set and clear AudioContext BehaviorRestrictions for testing purposes.

        * Modules/webaudio/AudioBufferSourceNode.cpp:
        (WebCore::AudioBufferSourceNode::startPlaying):
        * Modules/webaudio/AudioContext.cpp:
        (WebCore::AudioContext::nodeWillBeginPlayback):
        (WebCore::AudioContext::willBeginPlayback):
        (WebCore::AudioContext::willPausePlayback):
        (WebCore::AudioContext::startRendering):
        (WebCore::AudioContext::suspendContext):
        (WebCore::AudioContext::resumeContext):
        (WebCore::AudioContext::closeContext):
        (WebCore::AudioContext::suspendPlayback):
        (WebCore::AudioContext::mayResumePlayback):
        * Modules/webaudio/AudioContext.h:
        (WebCore::AudioContext::behaviorRestrictions):
        (WebCore::AudioContext::userGestureRequiredForAudioStart):
        (WebCore::AudioContext::pageConsentRequiredForAudioStart):
        * Modules/webaudio/AudioDestinationNode.cpp:
        (WebCore::AudioDestinationNode::render):
        * Modules/webaudio/AudioScheduledSourceNode.cpp:
        (WebCore::AudioScheduledSourceNode::start):
        * testing/Internals.cpp:
        (WebCore::Internals::setAudioContextRestrictions):
        * testing/Internals.h:
        * testing/Internals.idl:

2015-04-27  Alexey Proskuryakov  <ap@apple.com>

        Build fix.

        * accessibility/mac/AXObjectCacheMac.mm: (WebCore::AXObjectCache::postTextReplacementPlatformNotification):
        Use -setObject:forKey:, not array subscript.

2015-04-27  Eric Carlson  <eric.carlson@apple.com>

        WirelessTargetPicker should not be visible unless a file is playable
        https://bugs.webkit.org/show_bug.cgi?id=144271
        <rdar://problem/20712003>

        Reviewed by Jer Noble.

        * Modules/mediacontrols/mediaControlsApple.js:
        (Controller.prototype.isPlayable): New.
        (Controller.prototype.setStatusHidden): Call updateWirelessTargetAvailable.
        (Controller.prototype.updateWirelessTargetAvailable): Don't show the button until the
            status message has been hidden.

2015-04-22  Martin Robinson  <mrobinson@igalia.com>

        [CMake] Autogenerate cmakeconfig.h.cmake
        https://bugs.webkit.org/show_bug.cgi?id=143997

        Reviewed by Csaba Osztrogonác.

        * PlatformGTK.cmake: Use the WTF_USE style variables, because those are the same
        as the ones exposed to the build.

2015-04-27  Zalan Bujtas  <zalan@apple.com>

        Simple line layout: Wrong text offsetting when range does not start from the first renderer.
        https://bugs.webkit.org/show_bug.cgi?id=144167
        rdar://problem/20639857

        Reviewed by Simon Fraser.

        This patch ensures that TextIterator returns the right text when the input range starts
        from a sibling node.

        TextIterator::m_previousTextLengthInFlow keeps track of the current node offset from the parent.
        It is required to map simple line layout runs to RenderText positions.
        This patch sets the offset value when the iteration start with a sibling node.

        Test: fast/text/range-text-with-simple-line-layout.html

        * editing/TextIterator.cpp:
        (WebCore::TextIterator::TextIterator):
        (WebCore::TextIterator::handleTextNode):

2015-04-27  Commit Queue  <commit-queue@webkit.org>

        Unreviewed, rolling out r183393.
        https://bugs.webkit.org/show_bug.cgi?id=144272

        Caused memory corruption detected by GuardMalloc (Requested by
        ap on #webkit).

        Reverted changeset:

        "Synchronous XMLHttpRequest should get access to AppCache
        resources stored as flat files"
        https://bugs.webkit.org/show_bug.cgi?id=143711
        http://trac.webkit.org/changeset/183393

2015-04-27  Per Arne Vollan  <peavo@outlook.com>

        [Curl] Favicons loaded from disc cache are ignored.
        https://bugs.webkit.org/show_bug.cgi?id=143953

        Reviewed by Alex Christensen.

        When a favicon is loaded from the Curl disc cache, the icon data is thrown away.
        This happens because we give a 304 response, which makes the icon loader ignore
        the response. We can solve this by responding with 200 OK.

        * platform/network/ResourceHandleInternal.h:
        (WebCore::ResourceHandleInternal::ResourceHandleInternal):
        * platform/network/curl/CurlCacheManager.cpp:
        (WebCore::CurlCacheManager::didReceiveResponse):
        * platform/network/curl/ResourceHandleManager.cpp:
        (WebCore::headerCallback):
        (WebCore::ResourceHandleManager::initializeHandle):

2015-04-27  Brady Eidson  <beidson@apple.com>

        Make 'enum NavigationType' be an enum class
        https://bugs.webkit.org/show_bug.cgi?id=144270

        Reviewed by Alex Christensen.

        No new tests (No change in behavior).

        I’ll be adding a new NavigationType for bug 144269 so it makes sense to make this
        an enum class first.

        * loader/FrameLoader.cpp:
        (WebCore::FrameLoader::load):
        (WebCore::FrameLoader::reload):
        (WebCore::FrameLoader::loadDifferentDocumentItem):
        * loader/FrameLoaderTypes.h:

        * loader/NavigationAction.cpp:
        (WebCore::navigationType):
        (WebCore::NavigationAction::NavigationAction):

        * loader/PolicyChecker.cpp:
        (WebCore::PolicyChecker::checkNavigationPolicy):

        * page/PerformanceNavigation.cpp:
        (WebCore::PerformanceNavigation::type):

2015-04-27  Yoav Weiss  <yoav@yoav.ws>

        Fix viewport units in Media Queries
        https://bugs.webkit.org/show_bug.cgi?id=144260

        Reviewed by Darin Adler.

        This patch makes sure that viewport units are considered "length units"
        in the context of Media Queries, by having MediaQueryExp use the unit logic
        that is in CSSPrimitiveValue.
        It does that by turning the relevant methods in CSSPrimitiveValue into static.

        It also makes sure that the logic for "resolution units" is not maintained separately 
        in MediaQueryExp, to avoid similiar issues in the future with resolution units.

        Test: fast/media/mq-viewport-units.html

        * css/CSSPrimitiveValue.h:
        (WebCore::CSSPrimitiveValue::isViewportPercentageLength): Added a static variant.
        (WebCore::CSSPrimitiveValue::isLength): Added a static variant.
        (WebCore::CSSPrimitiveValue::isResolution): Added a static variant.
        * css/MediaQueryExp.cpp:
        (WebCore::featureWithValidPositiveLenghtOrNumber): Call CSSPrimitiveValue's length unit logic.
        (WebCore::featureWithValidDensity): Call CSSPrimitiveValue's resolution unit logic.

2015-04-27  Javier Fernandez  <jfernandez@igalia.com>

        [CSS Grid Layout] Support for the justify-self and justify-items in grid layout
        https://bugs.webkit.org/show_bug.cgi?id=133280

        Reviewed by David Hyatt.

        Implementation of justify-self and justify-items properties for grid. It supports
        different writing-modes and direction. Margins, borders and paddings are also
        considered when computing the final position and stretched size.

        This patch applies also a quite important refactoring of the alignment logic in
        order to share code between the two alignment dimensions, row-axis and column-axis.

        Overflow handling is still missing and will be added later in a follow-up bug.

        Tests: fast/css-grid-layout/grid-align-justify-margin-border-padding-vertical-lr.html
               fast/css-grid-layout/grid-align-justify-margin-border-padding-vertical-rl.html
               fast/css-grid-layout/grid-align-justify-margin-border-padding.html
               fast/css-grid-layout/justify-self-cell.html

        * rendering/RenderGrid.cpp:
        (WebCore::RenderGrid::layoutGridItems):
        (WebCore::RenderGrid::columnAxisPositionForChild):
        (WebCore::RenderGrid::rowAxisPositionForChild):
        (WebCore::RenderGrid::rowPositionForChild):
        (WebCore::RenderGrid::columnPositionForChild):
        (WebCore::RenderGrid::findChildLogicalPosition):
        (WebCore::RenderGrid::allowedToStretchLogicalHeightForChild): Deleted.
        (WebCore::RenderGrid::needToStretchChildLogicalHeight): Deleted.
        (WebCore::RenderGrid::marginLogicalHeightForChild): Deleted.
        (WebCore::RenderGrid::availableAlignmentSpaceForChildBeforeStretching): Deleted.
        (WebCore::RenderGrid::applyStretchAlignmentToChildIfNeeded): Deleted.
        * rendering/RenderGrid.h:
        * rendering/style/RenderStyle.cpp:
        (WebCore::RenderStyle::resolveJustification):
        * rendering/style/RenderStyle.h:

2015-04-27  Darin Adler  <darin@apple.com>

        Crashes under IDBDatabase::closeConnection
        https://bugs.webkit.org/show_bug.cgi?id=141745

        Reviewed by Alexey Proskuryakov.

        * Modules/indexeddb/IDBDatabase.cpp:
        (WebCore::IDBDatabase::~IDBDatabase): Do the work of close/closeConnection without
        actually calling those functions.
        (WebCore::IDBDatabase::closeConnection): Protect the database so it's not destroyed
        in the middle of this function's execution.

2015-04-27  Xabier Rodriguez Calvar  <calvaris@igalia.com> and Youenn Fablet  <youenn.fablet@crf.canon.fr>

        [Streams API] ReadableStream constructor start function should be able to close the stream
        https://bugs.webkit.org/show_bug.cgi?id=143363

        Reviewed by Benjamin Poulain.

        Implements https://streams.spec.whatwg.org/#close-readable-stream.
        When the "close" JS function is called, the stream is getting closed.
        The stream state is changed to close and if it has a reader, the reader gets closed as well:
        The reader resolves the closed promise and releases the stream.

        Enabled the possibility to resolve a promise with any JS value.
        This is used to resolve closed promise with jsUndefined and will be used for read promises in
        the future as well, though of course it is not restricted to Streams.

        Covered by reference tests that are now passing.

        * Modules/streams/ReadableStream.h:
        * Modules/streams/ReadableStream.cpp:
        (WebCore::ReadableStream::changeStateToClosed): Called by the JS function 'close'.
        * Modules/streams/ReadableStreamReader.cpp:
        (WebCore::ReadableStreamReader::ReadableStreamReader): Initialized stream.
        (WebCore::ReadableStreamReader::initialize): Added to handle state change at constructor time
        in particular closed/errored state.
        (WebCore::ReadableStreamReader::releaseStream): Releases the stream.
        (WebCore::ReadableStreamReader::closed): Stores the promise callbacks and invokes success
        immediately if the stream is already closed.
        (WebCore::ReadableStreamReader::changeStateToClosed): Changes the internal state to closed,
        resolves the promise and releases the stream.
        * Modules/streams/ReadableStreamReader.h:
        (WebCore::ReadableStreamReader::State): Added.
        * bindings/js/JSDOMPromise.h:
        (WebCore::DeferredWrapper::resolve<JSC::JSValue>): Adds the ability to resolve a promise with
        a custom JS value.
        * bindings/js/JSReadableStreamControllerCustom.cpp:
        (WebCore::JSReadableStreamController::close): Not "notImplemented" anymore. Now it closes the
        stream.
        * bindings/js/JSReadableStreamReaderCustom.cpp:
        (WebCore::JSReadableStreamReader::closed): Resolves the promise with undefined.

2015-04-27  Csaba Osztrogonác  <ossy@webkit.org>

        Fix the !ENABLE(CSS_GRID_LAYOUT) build after r183370
        https://bugs.webkit.org/show_bug.cgi?id=144255

        Reviewed by Carlos Garcia Campos.

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

2015-04-27  Youenn Fablet  <youenn.fablet@crf.canon.fr>

        Synchronous XMLHttpRequest should get access to AppCache resources stored as flat files
        https://bugs.webkit.org/show_bug.cgi?id=143711

        Reviewed by Darin Adler.

        This patch checks whether a substitute resource data is stored in memory or in file for synchronous loads.
        If data is stored in file, it reads the data through SharedBuffer::createWithContentsOfFile.
        This patch refactors some routines to replace Vector<char> by SharedBuffer to transmit response data.

        Test: http/tests/appcache/simple-video-sync.html

        * html/HTMLMediaElement.cpp:
        (WebCore::HTMLMediaElement::parseAttribute):
        * loader/DocumentThreadableLoader.cpp:
        (WebCore::DocumentThreadableLoader::loadRequest):
        * loader/FrameLoader.cpp:
        (WebCore::FrameLoader::loadResourceSynchronously):
        * loader/FrameLoader.h:
        * loader/appcache/ApplicationCacheHost.cpp:
        (WebCore::ApplicationCacheHost::maybeLoadResource):
        (WebCore::ApplicationCacheHost::createFileURL):
        (WebCore::ApplicationCacheHost::maybeLoadSynchronously):
        (WebCore::ApplicationCacheHost::maybeLoadFallbackSynchronously):
        * loader/appcache/ApplicationCacheHost.h:
        * xml/XSLTProcessorLibxslt.cpp:
        (WebCore::docLoaderFunc):
        * xml/parser/XMLDocumentParserLibxml2.cpp:
        (WebCore::openFunc):

2015-04-24  Philippe Normand  <pnormand@igalia.com>

        [JHBuild] Move to upstream OpenWebRTC
        https://bugs.webkit.org/show_bug.cgi?id=144145

        Reviewed by Carlos Garcia Campos.

        * platform/mediastream/openwebrtc/OpenWebRTCUtilities.cpp:
        (WebCore::initializeOpenWebRTC): Update with new owr_init API.

2015-04-17  Carlos Garcia Campos  <cgarcia@igalia.com>

        [SOUP] Add initial implementation of NetworkProcess disk cache
        https://bugs.webkit.org/show_bug.cgi?id=143872

        Reviewed by Martin Robinson.

        Allow to create a GRefPtr for SoupBuffer. Even though SoupBuffer
        is not a GObject and has copy/free functions instead of ref/unref,
        it's internally refcounted, so we could use copy/free as
        ref/unref.

        * PlatformGTK.cmake:
        * platform/network/soup/GRefPtrSoup.cpp: Added.
        (WTF::refGPtr):
        (WTF::derefGPtr):
        * platform/network/soup/GRefPtrSoup.h: Added.

2015-04-26  Manuel Rego Casasnovas  <rego@igalia.com>

        [CSS Grid Layout] LayoutBox::hasDefiniteLogicalHeight() should consider abspos boxes as definite
        https://bugs.webkit.org/show_bug.cgi?id=144144

        Reviewed by Darin Adler.

        Change order in hasDefiniteLogicalHeight() ifs to consider definite
        absolutely positioned elements.

        Test: fast/css-grid-layout/percent-track-breadths-regarding-container-size.html

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

2015-04-26  Said Abou-Hallawa  <sabouhallawa@apple.com>

        SVGFilterBuilder should drive the builtin sourceAlpha from the passed sourceGraphic
        https://bugs.webkit.org/show_bug.cgi?id=144137

        Reviewed by Darin Adler.

        The SourceAlpha should be driven from the last applied FilterEffect. This
        will require the SourceAlpha FilterEffect to wrap a FilterEffect rather
        than just getting the SourceGraphic reference from FilterEffectRenderer.

        This is based on the Blink patch:
        https://src.chromium.org/viewvc/blink?revision=194115&view=revision.

        Test: svg/filters/sourceAlpha-input-filter-effect.html
        
        * platform/graphics/filters/SourceAlpha.cpp:
        (WebCore::SourceAlpha::create): Make SourceAlpha wraps a FilterEffect
        rather than a Filter.
        
        (WebCore::SourceAlpha::determineAbsolutePaintRect): Determine the paint
        rect from the wrapped FilterEffect.
        
        (WebCore::SourceAlpha::platformApplySoftware): Extract the alpha channel
        from the wrapped FilterEffect.
        
        (WebCore::SourceAlpha::SourceAlpha): 
        * platform/graphics/filters/SourceAlpha.h:
        (WebCore::SourceAlpha::SourceAlpha): Deleted.
        Reimplement the constructor of the SourceAlpha such that it wraps the
        sourceEffect in its inputEffects.
    
        * rendering/FilterEffectRenderer.cpp:
        (WebCore::FilterEffectRenderer::buildReferenceFilter): Delete the FIXME
        comment and do exactly what it said. This matches the implementation of
        Gecko and Blink.
        
        * rendering/svg/RenderSVGResourceFilter.cpp:
        (WebCore::RenderSVGResourceFilter::buildPrimitives):
        Delete the extra argument.

        * svg/graphics/filters/SVGFilterBuilder.cpp:
        (WebCore::SVGFilterBuilder::SVGFilterBuilder):
        * svg/graphics/filters/SVGFilterBuilder.h: Drive the SourceAlpha from the
        passed source FilterEffect.

        * platform/graphics/filters/FEBlend.cpp:
        * platform/graphics/filters/FEBlend.h:
        * platform/graphics/filters/FEColorMatrix.cpp:
        * platform/graphics/filters/FEColorMatrix.h:
        * platform/graphics/filters/FEComponentTransfer.cpp:
        * platform/graphics/filters/FEComponentTransfer.h:
        * platform/graphics/filters/FEComposite.cpp:
        * platform/graphics/filters/FEComposite.h:
        * platform/graphics/filters/FEConvolveMatrix.cpp:
        * platform/graphics/filters/FEConvolveMatrix.h:
        * platform/graphics/filters/FEDiffuseLighting.cpp:
        * platform/graphics/filters/FEDiffuseLighting.h:
        * platform/graphics/filters/FEDisplacementMap.cpp:
        * platform/graphics/filters/FEDisplacementMap.h:
        * platform/graphics/filters/FEDropShadow.cpp:
        * platform/graphics/filters/FEDropShadow.h:
        * platform/graphics/filters/FEFlood.cpp:
        * platform/graphics/filters/FEFlood.h:
        * platform/graphics/filters/FEGaussianBlur.cpp:
        * platform/graphics/filters/FEGaussianBlur.h:
        * platform/graphics/filters/FELighting.cpp:
        * platform/graphics/filters/FELighting.h:
        * platform/graphics/filters/FEMerge.cpp:
        * platform/graphics/filters/FEMerge.h:
        * platform/graphics/filters/FEMorphology.cpp:
        * platform/graphics/filters/FEMorphology.h:
        * platform/graphics/filters/FEOffset.cpp:
        * platform/graphics/filters/FEOffset.h:
        * platform/graphics/filters/FESpecularLighting.cpp:
        * platform/graphics/filters/FESpecularLighting.h:
        * platform/graphics/filters/FETile.cpp:
        * platform/graphics/filters/FETile.h:
        * platform/graphics/filters/FETurbulence.cpp:
        * platform/graphics/filters/FETurbulence.h:
        * platform/graphics/filters/Filter.h:
        * platform/graphics/filters/FilterEffect.cpp:
        * platform/graphics/filters/FilterEffect.h:
        * platform/graphics/filters/SourceGraphic.cpp:
        * platform/graphics/filters/SourceGraphic.h:
        * rendering/svg/RenderSVGResourceFilter.h:
        * rendering/svg/SVGRenderTreeAsText.cpp:
        Code clean up: Pointer to reference conversion. It starts by converting
        FilterEffect::m_filter from pointer to reference and it kept expanding.
        All the creations of FilterEffect were happening by passing "this" so
        passing "*this" did not require to check whether the pointer is nil.

        * svg/SVGFEBlendElement.cpp:
        * svg/SVGFEBlendElement.h:
        * svg/SVGFEColorMatrixElement.cpp:
        * svg/SVGFEColorMatrixElement.h:
        * svg/SVGFEComponentTransferElement.cpp:
        * svg/SVGFEComponentTransferElement.h:
        * svg/SVGFECompositeElement.cpp:
        * svg/SVGFECompositeElement.h:
        * svg/SVGFEConvolveMatrixElement.cpp:
        * svg/SVGFEConvolveMatrixElement.h:
        * svg/SVGFEDiffuseLightingElement.cpp:
        * svg/SVGFEDiffuseLightingElement.h:
        * svg/SVGFEDisplacementMapElement.cpp:
        * svg/SVGFEDisplacementMapElement.h:
        * svg/SVGFEDropShadowElement.cpp:
        * svg/SVGFEDropShadowElement.h:
        * svg/SVGFEFloodElement.cpp:
        * svg/SVGFEFloodElement.h:
        * svg/SVGFEGaussianBlurElement.cpp:
        * svg/SVGFEGaussianBlurElement.h:
        * svg/SVGFEImageElement.cpp:
        * svg/SVGFEImageElement.h:
        * svg/SVGFEMergeElement.cpp:
        * svg/SVGFEMergeElement.h:
        * svg/SVGFEMorphologyElement.cpp:
        * svg/SVGFEMorphologyElement.h:
        * svg/SVGFEOffsetElement.cpp:
        * svg/SVGFEOffsetElement.h:
        * svg/SVGFESpecularLightingElement.cpp:
        * svg/SVGFESpecularLightingElement.h:
        * svg/SVGFETileElement.cpp:
        * svg/SVGFETileElement.h:
        * svg/SVGFETurbulenceElement.cpp:
        * svg/SVGFETurbulenceElement.h:
        * svg/SVGFilterPrimitiveStandardAttributes.h:
        * svg/graphics/filters/SVGFEImage.cpp:
        * svg/graphics/filters/SVGFEImage.h:
        Code clean up: Pointer to reference and PassRefPtr<> to RefPtr<> or Ref<>
        conversion.

2015-04-26  Darin Adler  <darin@apple.com>

        REGRESSION (r174283): disabled default buttons use impossible-to-read white text on Mac
        https://bugs.webkit.org/show_bug.cgi?id=144237

        Reviewed by Simon Fraser.

        No test because DumpRenderTree and WebKitTestRunner don't have a good method
        for testing behaviors that are specific to active windows. That's the same reason
        that the change that caused this bug didn't come with a regression test.

        * rendering/RenderTheme.cpp:
        (WebCore::RenderTheme::adjustStyle): Renamed "e" to "element". Added a comment to
        clarify the meaning of the setInsideDefaultButton function, and added an
        isDisabledFormControl check so it won't trigger on a disabled button.
        (WebCore::RenderTheme::systemColor): Moved CSSValueActivebuttontext into
        alphabetical order to match the rest of the switch statement ordering.

        * rendering/RenderThemeMac.mm:
        (WebCore::RenderThemeMac::systemColor): Moved CSSValueActivebuttontext into
        alphabetical order to match the rest of the switch statement ordering. Also
        fixed its behavior on older versions of OS X; it should just return the same
        thing that CSSValueButtontext would return in that context.

2015-04-26  Yusuke Suzuki  <utatane.tea@gmail.com>

        [ES6] Implement ES6 template literals
        https://bugs.webkit.org/show_bug.cgi?id=142691

        Reviewed by Darin Adler.

        Add ENABLE_ES6_TEMPLATE_LITERAL_SYNTAX compile time flag.

        * Configurations/FeatureDefines.xcconfig:

2015-04-26  Javier Fernandez  <jfernandez@igalia.com>

        [CSS Grid Layout] Support for align-self and align-items in grid layout
        https://bugs.webkit.org/show_bug.cgi?id=133225

        Reviewed by Darin Adler.

        Implementation of align-self and align-items properties for grid. It supports
        different writing-modes and direction. Margins, borders and paddings are also
        considered when computing the final position and stretched size.

        Overflow handling is still missing and will be added when implemented row-axis
        alignment support.

        Test: fast/css-grid-layout/grid-align.html

        * rendering/RenderBox.cpp:
        (WebCore::RenderBox::computeLogicalHeight):
        * rendering/RenderGrid.cpp:
        (WebCore::RenderGrid::logicalContentHeightForChild):
        (WebCore::RenderGrid::layoutGridItems):
        (WebCore::computeOverflowAlignmentOffset):
        (WebCore::RenderGrid::endOfRowForChild):
        (WebCore::RenderGrid::startOfRowForChild):
        (WebCore::RenderGrid::centeredRowPositionForChild):
        (WebCore::marginWidthForChild):
        (WebCore::marginHeightForChild):
        (WebCore::RenderGrid::allowedToStretchLogicalHeightForChild):
        (WebCore::RenderGrid::needToStretchChildLogicalHeight):
        (WebCore::RenderGrid::marginLogicalHeightForChild):
        (WebCore::RenderGrid::availableAlignmentSpaceForChildBeforeStretching):
        (WebCore::RenderGrid::applyStretchAlignmentToChildIfNeeded):
        (WebCore::RenderGrid::rowPositionForChild):
        (WebCore::RenderGrid::findChildLogicalPosition):
        (WebCore::RenderGrid::populateGridPositions): Deleted.
        * rendering/RenderGrid.h:

2015-04-26  Doug Russell  <d_russell@apple.com>

        AX: richer text change notifications (142719)
        https://bugs.webkit.org/show_bug.cgi?id=142719

        Reviewed by Darin Adler.

        Richer accessibility value change notifications. Introduce AXTextEditType,
        postTextStateChangeNotification and postTextReplacementNotification to give assistive
        tech apps more reliable context for responding to changes in web content. Also implement
        a mechanism to post value changes in password form fields in coalesced ticks to thwart
        analyzing the cadence of changes.

        Richer accessibility selection change notifications. Introduce AXTextStateChangeIntent,
        and an overload of postTextReplacementNotification to give assistive tech apps more
        reliable context for responding to changes in web content selection. Also block posting
        selection changes on password fields.

        Tests: platform/mac/accessibility/input-replacevalue-userinfo.html
               platform/mac/accessibility/selection-change-userinfo.html
               platform/mac/accessibility/value-change-userinfo.html

        * CMakeLists.txt:
        * WebCore.vcxproj/WebCore.vcxproj:
        * WebCore.vcxproj/WebCore.vcxproj.filters:
        * WebCore.xcodeproj/project.pbxproj:
        * accessibility/AXObjectCache.cpp:
        (WebCore::AXObjectCache::AXObjectCache):
        (WebCore::AXObjectCache::notificationPostTimerFired):
        (WebCore::AXObjectCache::passwordNotificationPostTimerFired):
        (WebCore::AXObjectCache::showIntent):
        (WebCore::AXObjectCache::setTextSelectionIntent):
        (WebCore::isPasswordFieldOrContainedByPasswordField):
        (WebCore::AXObjectCache::postTextStateChangeNotification):
        (WebCore::AXObjectCache::postTextReplacementNotification):
        (WebCore::AXObjectCache::enqueuePasswordValueChangeNotification):
        (WebCore::AXObjectCache::rootWebArea):
        (WebCore::AXObjectCache::textChangeForEditType):
        (WebCore::AXObjectCache::selectedChildrenChanged): Deleted.
        (WebCore::AXObjectCache::frameLoadingEventNotification): Deleted.
        * accessibility/AXObjectCache.h:
        (WebCore::AXObjectCache::postTextStateChangeNotification):
        (WebCore::AXObjectCache::postTextReplacementNotification):
        (WebCore::AXObjectCache::postTextStateChangePlatformNotification):
        (WebCore::AXObjectCache::postTextReplacementPlatformNotification):
        (WebCore::AXObjectCache::textChangeForEditType):
        (WebCore::AXObjectCache::nodeTextChangePlatformNotification):
        (WebCore::AXObjectCache::computedObjectAttributeCache): Deleted.
        (WebCore::AXObjectCache::getOrCreate): Deleted.
        (WebCore::AXObjectCache::attachWrapper): Deleted.
        * accessibility/AXTextStateChangeIntent.h: Added.
        (WebCore::AXTextStateChangeIntent::AXTextStateChangeIntent):
        * accessibility/AccessibilityNodeObject.cpp:
        (WebCore::AccessibilityNodeObject::passwordFieldOrContainingPasswordField):
        * accessibility/AccessibilityNodeObject.h:
        * accessibility/AccessibilityObject.cpp:
        (WebCore::AccessibilityObject::isContainedByPasswordField):
        * accessibility/AccessibilityObject.h:
        (WebCore::AccessibilityObject::passwordFieldOrContainingPasswordField):
        (WebCore::AccessibilityObject::isPasswordField): Deleted.
        * accessibility/AccessibilityRenderObject.cpp:
        (WebCore::AccessibilityRenderObject::setSelectedTextRange):
        (WebCore::AccessibilityRenderObject::setSelectedVisiblePositionRange):
        * accessibility/AccessibilityScrollView.h:
        * accessibility/atk/AXObjectCacheAtk.cpp:
        (WebCore::AXObjectCache::nodeTextChangePlatformNotification):
        * accessibility/ios/AXObjectCacheIOS.mm:
        (WebCore::AXObjectCache::postTextStateChangePlatformNotification):
        (WebCore::AXObjectCache::postTextReplacementPlatformNotification):
        * accessibility/mac/AXObjectCacheMac.mm:
        (WebCore::AXObjectCache::setShouldRepostNotificationsForTests):
        (WebCore::AXPostNotificationWithUserInfo):
        (WebCore::AXObjectCache::postPlatformNotification):
        (WebCore::AXObjectCache::postTextStateChangePlatformNotification):
        (WebCore::textReplacementChangeDictionary):
        (WebCore::AXObjectCache::postTextReplacementPlatformNotification):
        * accessibility/mac/WebAccessibilityObjectWrapperBase.h:
        * accessibility/mac/WebAccessibilityObjectWrapperBase.mm:
        (+[WebAccessibilityObjectWrapperBase accessibilitySetShouldRepostNotifications:]):
        (-[WebAccessibilityObjectWrapperBase accessibilityPostedNotification:]):
        (arrayRemovingNonJSONTypes):
        (dictionaryRemovingNonJSONTypes):
        (-[WebAccessibilityObjectWrapperBase accessibilityPostedNotification:userInfo:]):
        * accessibility/mac/WebAccessibilityObjectWrapperMac.h:
        * accessibility/mac/WebAccessibilityObjectWrapperMac.mm:
        (textMarkerRangeFromVisiblePositions):
        (-[WebAccessibilityObjectWrapper textMarkerRangeFromVisiblePositions:endPosition:]):
        * editing/AppendNodeCommand.cpp:
        (WebCore::AppendNodeCommand::AppendNodeCommand):
        (WebCore::sendAXTextChangedIgnoringLineBreaks):
        (WebCore::AppendNodeCommand::doApply):
        (WebCore::AppendNodeCommand::doUnapply):
        * editing/AppendNodeCommand.h:
        (WebCore::AppendNodeCommand::create):
        * editing/ApplyStyleCommand.cpp:
        (WebCore::ApplyStyleCommand::ApplyStyleCommand):
        (WebCore::ApplyStyleCommand::applyBlockStyle): Deleted.
        * editing/ApplyStyleCommand.h:
        * editing/CompositeEditCommand.cpp:
        (WebCore::EditCommandComposition::unapplyEditType):
        (WebCore::CompositeEditCommand::CompositeEditCommand):
        (WebCore::CompositeEditCommand::apply):
        (WebCore::CompositeEditCommand::insertParagraphSeparator):
        (WebCore::CompositeEditCommand::insertNodeBefore):
        (WebCore::CompositeEditCommand::appendNode):
        (WebCore::CompositeEditCommand::removeNodePreservingChildren):
        (WebCore::CompositeEditCommand::insertTextIntoNode):
        (WebCore::CompositeEditCommand::deleteTextFromNode):
        (WebCore::CompositeEditCommand::replaceTextInNode):
        (WebCore::CompositeEditCommand::moveParagraphs):
        (WebCore::EditCommandComposition::getNodesInCommand): Deleted.
        (WebCore::CompositeEditCommand::applyStyle): Deleted.
        (WebCore::CompositeEditCommand::insertLineBreak): Deleted.
        (WebCore::CompositeEditCommand::insertNodeAt): Deleted.
        (WebCore::CompositeEditCommand::removeChildrenInRange): Deleted.
        (WebCore::CompositeEditCommand::inputText): Deleted.
        * editing/CompositeEditCommand.h:
        * editing/DeleteFromTextNodeCommand.cpp:
        (WebCore::DeleteFromTextNodeCommand::DeleteFromTextNodeCommand):
        (WebCore::DeleteFromTextNodeCommand::doApply):
        (WebCore::DeleteFromTextNodeCommand::doUnapply):
        * editing/DeleteFromTextNodeCommand.h:
        (WebCore::DeleteFromTextNodeCommand::create):
        (WebCore::DeleteFromTextNodeCommand::deletedText):
        * editing/DeleteSelectionCommand.cpp:
        (WebCore::DeleteSelectionCommand::DeleteSelectionCommand):
        (WebCore::DeleteSelectionCommand::preservesTypingStyle): Deleted.
        * editing/DeleteSelectionCommand.h:
        (WebCore::DeleteSelectionCommand::create):
        * editing/DictationCommand.cpp:
        (WebCore::DictationCommand::insertTextRunWithoutNewlines):
        (WebCore::DictationCommand::insertParagraphSeparator):
        * editing/EditAction.h:
        * editing/EditCommand.cpp:
        (WebCore::EditCommand::EditCommand):
        (WebCore::EditCommand::editingAction):
        (WebCore::EditCommand::applyEditType):
        (WebCore::EditCommand::unapplyEditType):
        (WebCore::SimpleEditCommand::SimpleEditCommand):
        (WebCore::SimpleEditCommand::notifyAccessibilityForTextChange):
        (WebCore::EditCommand::setParent): Deleted.
        * editing/EditCommand.h:
        * editing/EditingAllInOne.cpp:
        * editing/Editor.cpp:
        (WebCore::Editor::handleTextEvent):
        (WebCore::Editor::deleteSelectionWithSmartDelete):
        (WebCore::Editor::replaceSelectionWithFragment):
        (WebCore::Editor::replaceSelectionWithText):
        (WebCore::Editor::appliedEditing):
        (WebCore::Editor::unappliedEditing):
        (WebCore::Editor::performCutOrCopy):
        (WebCore::Editor::markMisspellingsAfterTypingToWord):
        (WebCore::Editor::changeBackToReplacedString):
        (WebCore::Editor::transpose):
        (WebCore::Editor::changeSelectionAfterCommand):
        * editing/Editor.h:
        * editing/EditorCommand.cpp:
        (WebCore::executeInsertFragment):
        * editing/FrameSelection.cpp:
        (WebCore::FrameSelection::moveTo):
        (WebCore::FrameSelection::moveWithoutValidationTo):
        (WebCore::FrameSelection::setSelectionByMouseIfDifferent):
        (WebCore::FrameSelection::setSelectionWithoutUpdatingAppearance):
        (WebCore::FrameSelection::setSelection):
        (WebCore::FrameSelection::updateAndRevealSelection):
        (WebCore::isBoundary):
        (WebCore::FrameSelection::textSelectionIntent):
        (WebCore::FrameSelection::modify):
        (WebCore::FrameSelection::selectAll):
        (WebCore::FrameSelection::wordSelectionContainingCaretSelection):
        (WebCore::FrameSelection::modifyMovingBackward): Deleted.
        (WebCore::FrameSelection::selectFrameElementInParentIfFullySelected): Deleted.
        (WebCore::FrameSelection::selectionAtWordStart): Deleted.
        * editing/FrameSelection.h:
        (WebCore::FrameSelection::notifyAccessibilityForSelectionChange):
        (WebCore::FrameSelection::selection): Deleted.
        * editing/InsertIntoTextNodeCommand.cpp:
        (WebCore::InsertIntoTextNodeCommand::InsertIntoTextNodeCommand):
        (WebCore::InsertIntoTextNodeCommand::doApply):
        (WebCore::InsertIntoTextNodeCommand::doUnapply):
        * editing/InsertIntoTextNodeCommand.h:
        (WebCore::InsertIntoTextNodeCommand::create):
        (WebCore::InsertIntoTextNodeCommand::insertedText):
        * editing/InsertNodeBeforeCommand.cpp:
        (WebCore::InsertNodeBeforeCommand::InsertNodeBeforeCommand):
        (WebCore::InsertNodeBeforeCommand::doApply):
        (WebCore::InsertNodeBeforeCommand::doUnapply):
        * editing/InsertNodeBeforeCommand.h:
        (WebCore::InsertNodeBeforeCommand::create):
        * editing/InsertParagraphSeparatorCommand.cpp:
        (WebCore::InsertParagraphSeparatorCommand::InsertParagraphSeparatorCommand):
        * editing/InsertParagraphSeparatorCommand.h:
        (WebCore::InsertParagraphSeparatorCommand::create):
        * editing/InsertTextCommand.cpp:
        (WebCore::InsertTextCommand::InsertTextCommand):
        * editing/InsertTextCommand.h:
        (WebCore::InsertTextCommand::create):
        (WebCore::InsertTextCommand::createWithMarkerSupplier):
        * editing/MoveSelectionCommand.cpp:
        (WebCore::MoveSelectionCommand::doApply):
        * editing/RemoveNodePreservingChildrenCommand.cpp:
        (WebCore::RemoveNodePreservingChildrenCommand::RemoveNodePreservingChildrenCommand):
        * editing/RemoveNodePreservingChildrenCommand.h:
        (WebCore::RemoveNodePreservingChildrenCommand::create):
        * editing/ReplaceDeleteFromTextNodeCommand.cpp: Copied from Source/WebCore/editing/AppendNodeCommand.h.
        (WebCore::ReplaceDeleteFromTextNodeCommand::ReplaceDeleteFromTextNodeCommand):
        (WebCore::ReplaceDeleteFromTextNodeCommand::notifyAccessibilityForTextChange):
        * editing/ReplaceDeleteFromTextNodeCommand.h: Copied from Source/WebCore/editing/AppendNodeCommand.h.
        * editing/ReplaceInsertIntoTextNodeCommand.cpp: Added.
        (WebCore::ReplaceInsertIntoTextNodeCommand::ReplaceInsertIntoTextNodeCommand):
        (WebCore::ReplaceInsertIntoTextNodeCommand::notifyAccessibilityForTextChange):
        * editing/ReplaceInsertIntoTextNodeCommand.h: Copied from Source/WebCore/editing/RemoveNodePreservingChildrenCommand.h.
        * editing/ReplaceSelectionCommand.cpp:
        (WebCore::ReplaceSelectionCommand::ReplaceSelectionCommand):
        (WebCore::ReplaceSelectionCommand::InsertedNodes::didReplaceNode): Deleted.
        (WebCore::ReplaceSelectionCommand::insertAsListItems): Deleted.
        * editing/ReplaceSelectionCommand.h:
        (WebCore::ReplaceSelectionCommand::create):
        * editing/TypingCommand.cpp:
        (WebCore::TypingCommand::insertTextRunWithoutNewlines):
        (WebCore::TypingCommand::insertParagraphSeparator):
        * editing/atk/FrameSelectionAtk.cpp:
        (WebCore::FrameSelection::notifyAccessibilityForSelectionChange):
        * editing/ios/DictationCommandIOS.cpp:
        (WebCore::DictationCommandIOS::DictationCommandIOS):
        * editing/ios/DictationCommandIOS.h:
        * editing/mac/FrameSelectionMac.mm:
        (WebCore::FrameSelection::notifyAccessibilityForSelectionChange):
        * html/HTMLTextFormControlElement.cpp:
        (WebCore::HTMLTextFormControlElement::setInnerTextValue):
        * page/DragController.cpp:
        (WebCore::DragController::concludeEditDrag):
        * page/EventHandler.cpp:
        (WebCore::setInitialKeyboardSelection):
        * page/FocusController.cpp:
        (WebCore::FocusController::advanceFocusInDocumentOrder):

2015-04-26  Darin Adler  <darin@apple.com>

        REGRESSION (r176751): line-height ignored in <button> elements
        https://bugs.webkit.org/show_bug.cgi?id=144234

        Reviewed by Antti Koivisto.

        Test: fast/forms/button-line-height.html

        The theme code was overriding line heights on all themed form controls.
        Instead it should have only overridden line height when the theme overrides the font.
        For <input type=button> on Mac, the theme always overrides the font, but for
        other elements like <button> it does not.

        * platform/Theme.h: Changed controlFont to return an Optional so we can tell
        when the theme is overriding the font. Otherwise if the font from the user-agent
        style sheet and the font from the theme are the same, we will think we are not
        overriding the font when we actually are.

        * platform/mac/ThemeMac.h: Updated controlFont to return Optional.
        * platform/mac/ThemeMac.mm:
        (WebCore::ThemeMac::controlFont): Ditto.

        * rendering/RenderTheme.cpp:
        (WebCore::RenderTheme::adjustStyle): Set line height only if the font is
        overriden by the theme, all the time for PushButtonPart on Mac, and not at all
        for other parts. Also tightened up the logic a little since RenderStyle's
        setFontDescription already does an "==" comparison; we don't have to do
        that twice.

2015-04-26  Darin Adler  <darin@apple.com>

        REGRESSION (r173801): Use after free in WebCore::NotificationCenter::~NotificationCenter
        https://bugs.webkit.org/show_bug.cgi?id=137163

        Reviewed by Andy Estes.

        Test: fast/notifications/request-notification-permission-while-reloading.html

        The test doesn't crash under WebKit2, but that's still OK for our purposes.

        * Modules/notifications/NotificationCenter.cpp:
        (WebCore::NotificationCenter::NotificationCenter): Initialize m_timer.
        (WebCore::NotificationCenter::createNotification): Moved here from the header.
        (WebCore::NotificationCenter::requestPermission): Start the timer and ref the notification
        center when we need to defer a callback. Also use a lambda for the callback and changed
        the argument to match what the bindings actually pass. Before we said PassRefPtr, but the
        bindings were not transferring ownership of the VoidCallback. The new type is a little
        strange but it's consistent with how the bindings work right now.
        (WebCore::NotificationCenter::timerFired): Added. Calls all the callbacks, then does a deref
        to match the ref we did above.
        (WebCore::NotificationCenter::requestTimedOut): Deleted.
        (WebCore::NotificationCenter::NotificationRequestCallback::createAndStartTimer): Deleted.
        (WebCore::NotificationCenter::NotificationRequestCallback::NotificationRequestCallback): Deleted.
        (WebCore::NotificationCenter::NotificationRequestCallback::startTimer): Deleted.
        (WebCore::NotificationCenter::NotificationRequestCallback::timerFired): Deleted.

        * Modules/notifications/NotificationCenter.h: Reorganized the header to be a bit more tidy.
        Changed the argument type for requestPermission to match the reality of what's passed by the
        bindings. Removed NotificationRequestCallback and replaced the m_callbacks HashSet with a
        vector of std::function.

2015-04-26  Simon Fraser  <simon.fraser@apple.com>

        Modernize animations code
        https://bugs.webkit.org/show_bug.cgi?id=144196

        Reviewed by Darin Adler.
        
        General cleanup in animation-related code.
        
        Use C++ references where possible. Use modern loops. Make classes fast-allocated.
        Use C++11 member initialization. Order member variables to optimize packing.
        
        This removes some null checks in CompositeAnimation, but the values in m_transitions
        and m_keyframeAnimations can never be null.
        
        No behavior change.

        * page/animation/AnimationBase.cpp:
        (WebCore::AnimationBase::AnimationBase):
        (WebCore::AnimationBase::animationsMatch):
        (WebCore::AnimationBase::updateStateMachine):
        (WebCore::AnimationBase::fireAnimationEventsIfNeeded):
        (WebCore::AnimationBase::beginAnimationUpdateTime):
        (WebCore::AnimationBase::getElapsedTime):
        * page/animation/AnimationBase.h:
        (WebCore::AnimationBase::animation):
        (WebCore::AnimationBase::setAnimation):
        * page/animation/AnimationController.cpp:
        (WebCore::AnimationControllerPrivate::ensureCompositeAnimation):
        (WebCore::AnimationControllerPrivate::addToAnimationsWaitingForStartTimeResponse):
        * page/animation/CompositeAnimation.cpp:
        (WebCore::CompositeAnimation::CompositeAnimation):
        (WebCore::CompositeAnimation::clearRenderer):
        (WebCore::CompositeAnimation::updateTransitions):
        (WebCore::CompositeAnimation::updateKeyframeAnimations):
        (WebCore::CompositeAnimation::animate):
        (WebCore::CompositeAnimation::getAnimatedStyle):
        (WebCore::CompositeAnimation::timeToNextService):
        (WebCore::CompositeAnimation::getAnimationForProperty):
        (WebCore::CompositeAnimation::computeExtentOfTransformAnimation):
        (WebCore::CompositeAnimation::suspendAnimations):
        (WebCore::CompositeAnimation::resumeAnimations):
        (WebCore::CompositeAnimation::overrideImplicitAnimations):
        (WebCore::CompositeAnimation::resumeOverriddenImplicitAnimations):
        (WebCore::CompositeAnimation::isAnimatingProperty):
        (WebCore::CompositeAnimation::pauseTransitionAtTime):
        (WebCore::CompositeAnimation::numberOfActiveAnimations):
        * page/animation/CompositeAnimation.h:
        (WebCore::CompositeAnimation::create):
        (WebCore::CompositeAnimation::animationController):
        * page/animation/ImplicitAnimation.cpp:
        (WebCore::ImplicitAnimation::ImplicitAnimation):
        (WebCore::ImplicitAnimation::sendTransitionEvent):
        * page/animation/ImplicitAnimation.h:
        (WebCore::ImplicitAnimation::create):
        * page/animation/KeyframeAnimation.cpp:
        (WebCore::KeyframeAnimation::KeyframeAnimation):
        (WebCore::KeyframeAnimation::animate):
        (WebCore::KeyframeAnimation::getAnimatedStyle):
        (WebCore::KeyframeAnimation::computeExtentOfTransformAnimation):
        (WebCore::KeyframeAnimation::startAnimation):
        (WebCore::KeyframeAnimation::sendAnimationEvent):
        (WebCore::KeyframeAnimation::overrideAnimations):
        (WebCore::KeyframeAnimation::resumeOverriddenAnimations):
        (WebCore::KeyframeAnimation::timeToNextService):
        * page/animation/KeyframeAnimation.h:
        * platform/animation/Animation.cpp:
        (WebCore::Animation::animationsMatch):
        * platform/animation/Animation.h:
        * rendering/RenderLayer.cpp:
        (WebCore::RenderLayer::paint):
        (WebCore::performOverlapTests):
        * rendering/style/KeyframeList.h:
        (WebCore::KeyframeList::properties):
        (WebCore::KeyframeList::keyframes):
        (WebCore::KeyframeList::beginProperties): Deleted.
        (WebCore::KeyframeList::endProperties): Deleted.
        (WebCore::KeyframeList::operator[]): Deleted.
        * rendering/style/RenderStyle.h:

2015-04-25  Simon Fraser  <simon.fraser@apple.com>

        Expand compositing coverage rect when scrolling and animating
        https://bugs.webkit.org/show_bug.cgi?id=144214
        rdar://problem/20695926

        Reviewed by Darin Adler.
        
        r183300 added a secondary rect that gets mapped through graphics layers during
        layer flushing, which represents a rect, in layer coordinates, that represents
        the portion of this layer that's visible in the viewport.
        
        This patch extends that rect based on scrolling and animating, with the extended
        coverage rect being mapped into descendants (where it's used to determine whether
        they should detach their backing store). This required moving the responsibility for
        extending the coverage rect from TileController (where it was used to compute
        tiled extent), to GraphicsLayerCA. Thus, GraphicsLayerCA now owns the concept
        of coverage rect, and pushes it onto TileControllers.

        Test: compositing/tiling/transform-origin-tiled.html

        * platform/graphics/TiledBacking.h: Coverage rect is now pushed onto
        TiledBacking from outside; new functions reflect that.
        * platform/graphics/ca/GraphicsLayerCA.cpp:
        (WebCore::GraphicsLayerCA::flushCompositingStateForThisLayerOnly): No need to pass
        old visible rect to commitLayerChangesBeforeSublayers.
        (WebCore::accumulatesTransform):
        (WebCore::GraphicsLayerCA::recursiveVisibleRectChangeRequiresFlush): Now calls computeVisibleAndCoverageRect(),
        then adjustCoverageRect() which adjusts for scrolling and animations.
        (WebCore::GraphicsLayerCA::computeVisibleAndCoverageRect): Fetch the coverage rect from the
        TransformState, but we can only use this if mapping it through various transforms didn't clamp
        it (which can happen with severe rotations in perspective).
        (WebCore::GraphicsLayerCA::adjustCoverageRect): If this layer represents the page tiled backing,
        ask the TileController to do coverage adjustment (code which should move into here eventually).
        For other tiled layers, do motion-based adjustment. This should really be done for all layers
        (an animating non-tiled layer may contain a tiled layer).
        (WebCore::GraphicsLayerCA::setVisibleAndCoverageRects): The VisibleRectChanged wasn't needed
        any more; layers only react to changed in coverage rect now.
        (WebCore::GraphicsLayerCA::recursiveCommitChanges): If adjustCoverageRect() changed the
        coverage rect, here we push the new rect back into the TransformState so it will affect
        descendants.
        (WebCore::GraphicsLayerCA::commitLayerChangesBeforeSublayers):
        (WebCore::GraphicsLayerCA::updateCoverage): Coverage and backing store attachment go
        hand-in-hand now.
        (WebCore::GraphicsLayerCA::changeLayerTypeTo):
        (WebCore::GraphicsLayerCA::computeVisibleRect): Deleted.
        (WebCore::GraphicsLayerCA::updateBackingStoreAttachment): Deleted.
        (WebCore::GraphicsLayerCA::updateVisibleRect): Deleted.
        * platform/graphics/ca/GraphicsLayerCA.h:
        * platform/graphics/ca/TileController.cpp:
        (WebCore::TileController::setVisibleRect): Visible rect is only used for scroll performance
        logging now.
        (WebCore::TileController::setCoverageRect):
        (WebCore::TileController::tilesWouldChangeForCoverageRect):
        (WebCore::TileController::revalidateTiles):
        (WebCore::TileController::boundsForSize): This is required because when computeTileCoverageRect()
        is called at the top of committing the owning layer, we haven't yet pushed a new size to the
        underlying platform layer (which TileController::bounds() consults).
        (WebCore::TileController::bounds):
        (WebCore::TileController::computeTileCoverageRect): Have to push the new size in, since the
        underlying layer hasn't been updated yet.
        (WebCore::TileController::didRevalidateTiles):
        (WebCore::TileController::tilesWouldChangeForVisibleRect): Deleted.
        * platform/graphics/ca/TileController.h:
        * platform/graphics/ca/TileCoverageMap.cpp:
        (WebCore::TileCoverageMap::TileCoverageMap): Add a layer to the tile map that shows 
        a dark blue outline for the coverage rect.
        (WebCore::TileCoverageMap::update): Fix some minor issues with map position when top
        content inset is non-zero.
        * platform/graphics/ca/TileCoverageMap.h:
        * platform/graphics/ca/TileGrid.cpp: Everything is in terms of coverage rect now.
        (WebCore::TileGrid::tilesWouldChangeForCoverageRect):
        (WebCore::TileGrid::revalidateTiles):
        (WebCore::TileGrid::tilesWouldChangeForVisibleRect): Deleted.
        * platform/graphics/ca/TileGrid.h:
        * platform/graphics/transforms/TransformState.cpp:
        (WebCore::TransformState::operator=):
        (WebCore::TransformState::mappedQuad): Pass direction into mapQuad().
        (WebCore::TransformState::mappedSecondaryQuad): Ditto.
        (WebCore::TransformState::setLastPlanarSecondaryQuad): This function allows a caller
        to pass in a secondary quad, and get it mapped into the coordinate space of the last
        "flattening" state (since the TransformState may have in-flight XY offset, or transform).
        This mapping is achieved by re-using mapQuad(), but in the inverse direction.
        (WebCore::TransformState::mapQuad): Make direction a parameter.
        (WebCore::TransformState::flattenWithTransform): This code omitted to map the secondary
        quad. Failure was detected by tests.
        * platform/graphics/transforms/TransformState.h:
        (WebCore::TransformState::isMappingSecondaryQuad):
        (WebCore::TransformState::direction):
        (WebCore::TransformState::inverseDirection):
        (WebCore::TransformState::lastPlanarSecondaryQuad): Deleted.

2015-04-25  Dan Bernstein  <mitz@apple.com>

        Updated expected bindings generation test results after r183343.

        * bindings/scripts/test/ObjC/DOMTestActiveDOMObject.h:
        * bindings/scripts/test/ObjC/DOMTestCallback.h:
        * bindings/scripts/test/ObjC/DOMTestCustomNamedGetter.h:
        * bindings/scripts/test/ObjC/DOMTestEventConstructor.h:
        * bindings/scripts/test/ObjC/DOMTestEventTarget.h:
        * bindings/scripts/test/ObjC/DOMTestException.h:
        * bindings/scripts/test/ObjC/DOMTestGenerateIsReachable.h:
        * bindings/scripts/test/ObjC/DOMTestInterface.h:
        * bindings/scripts/test/ObjC/DOMTestMediaQueryListListener.h:
        * bindings/scripts/test/ObjC/DOMTestNamedConstructor.h:
        * bindings/scripts/test/ObjC/DOMTestNode.h:
        * bindings/scripts/test/ObjC/DOMTestNondeterministic.h:
        * bindings/scripts/test/ObjC/DOMTestObj.h:
        * bindings/scripts/test/ObjC/DOMTestOverloadedConstructors.h:
        * bindings/scripts/test/ObjC/DOMTestSerializedScriptValueInterface.h:
        * bindings/scripts/test/ObjC/DOMTestTypedefs.h:
        * bindings/scripts/test/ObjC/DOMattribute.h:
        * bindings/scripts/test/ObjC/DOMreadonly.h:

2015-04-25  Myles C. Maxfield  <mmaxfield@apple.com>

        font-synthesis's initial value is "style weight"
        https://bugs.webkit.org/show_bug.cgi?id=144195

        Reviewed by Darin Adler.

        Updated fast/css3-text/font-synthesis-parse.html.

        * rendering/style/RenderStyle.h:

2015-04-25  Commit Queue  <commit-queue@webkit.org>

        Unreviewed, rolling out r183305.
        https://bugs.webkit.org/show_bug.cgi?id=144213

        Crashes on wikipedia (Requested by litherum on #webkit).

        Reverted changeset:

        "[Cocoa] FontPlatformData's equality check should always use
        reference URLs"
        https://bugs.webkit.org/show_bug.cgi?id=144168
        http://trac.webkit.org/changeset/183305

2015-04-25  Dan Bernstein  <mitz@apple.com>

        WebCore part of <rdar://problem/20697966> Avoid using TBD as an argument to NS_AVAILABLE
        https://bugs.webkit.org/show_bug.cgi?id=144201

        Reviewed by Darin Adler.

        Instead of TBD, use the high version number to which it corresponds in CFAvailability.h,
        9876.5.

        * bindings/objc/WebKitAvailability.h: Added a definition of __NSi_9876_5, so that we can
        use 9876_5 in availability macros. Removed the definition of
        __AVAILABILITY_INTERNAL__MAC_TBD, which will no longer be used.

        * bindings/scripts/CodeGeneratorObjC.pm:
        (ReadPublicInterfaces): Use 9876_5 instead of TBD.

2015-04-25  Timothy Hatcher  <timothy@apple.com>

        Update localized strings.

        * English.lproj/Localizable.strings: Updated.

2015-04-25  Darin Adler  <darin@apple.com>

        Crash if IDBTransaction is aborted right after it's created
        https://bugs.webkit.org/show_bug.cgi?id=144199

        Reviewed by Simon Fraser.

        Test: storage/indexeddb/transaction-abort.html

        * Modules/indexeddb/IDBTransactionBackend.cpp:
        (WebCore::IDBTransactionBackend::IDBTransactionBackend): Don't start the
        transaction if it is already complete (has been aborted).

2015-04-25  Martin Robinson  <mrobinson@igalia.com>

        Rename ENABLE_3D_RENDERING to ENABLE_3D_TRANSFORMS
        https://bugs.webkit.org/show_bug.cgi?id=144182

        Reviewed by Simon Fraser.

        * Configurations/FeatureDefines.xcconfig: Replace all instances of 3D_RENDERING with 3D_TRANSFORMS.
        * WebCore.vcxproj/WebCoreCommon.props: Replace all instances of 3D_RENDERING with 3D_TRANSFORMS.
        * WebCore.vcxproj/WebCoreTestSupportCommon.props: Replace all instances of 3D_RENDERING with 3D_TRANSFORMS.
        * css/CSSComputedStyleDeclaration.cpp: Replace all instances of 3D_RENDERING with 3D_TRANSFORMS.
        * css/MediaQueryEvaluator.cpp: Replace all instances of 3D_RENDERING with 3D_TRANSFORMS.
        * platform/graphics/GraphicsContext.h: Replace all instances of 3D_RENDERING with 3D_TRANSFORMS.
        * platform/graphics/cairo/GraphicsContextCairo.cpp: Replace all instances of 3D_RENDERING with 3D_TRANSFORMS.
        * platform/graphics/texmap/TextureMapperImageBuffer.cpp: Replace all instances of 3D_RENDERING with 3D_TRANSFORMS.
        * rendering/RenderLayer.cpp: Replace all instances of 3D_RENDERING with 3D_TRANSFORMS.
        * rendering/RenderLayerCompositor.cpp: Replace all instances of 3D_RENDERING with 3D_TRANSFORMS.
        * rendering/RenderObject.cpp: Replace all instances of 3D_RENDERING with 3D_TRANSFORMS.

2015-04-24  Myles C. Maxfield  <mmaxfield@apple.com>

        [Cocoa] FontPlatformData's equality check should always use reference URLs
        https://bugs.webkit.org/show_bug.cgi?id=144168

        Reviewed by Tim Horton.

        <rdar://problem/18985642> is not fixed, so we need to continue using the older
        objectForEqualityCheck().

        No new tests because there is no behavior change.

        * platform/graphics/cocoa/FontPlatformDataCocoa.mm:
        (WebCore::FontPlatformData::objectForEqualityCheck):

2015-04-24  Myles C. Maxfield  <mmaxfield@apple.com>

        Implement parsing support for font-synthesis CSS property
        https://bugs.webkit.org/show_bug.cgi?id=144180
        <rdar://problem/20692791>
        <rdar://problem/20625024>

        Reviewed by Tim Horton.

        See http://www.w3.org/TR/css-fonts-3/#font-synthesis-prop

        This property allows web authors to opt out of font synthesis. This patch only
        implements parsing support for the property.

        Test: fast/css3-text/font-synthesis-parse.html

        * css/CSSComputedStyleDeclaration.cpp:
        (WebCore::fontSynthesisFromStyle): Create a CSSValue for the existing style object.
        (WebCore::ComputedStyleExtractor::propertyValue): Call fontSynthesisFromStyle().
        * css/CSSParser.cpp:
        (WebCore::CSSParser::parseValue): Call parseFontSynthesis().
        (WebCore::CSSParser::parseFontSynthesis): Accept the grammar
        "none | [ weight || style ]"
        * css/CSSParser.h: Function signature.
        * css/CSSPropertyNames.in: New CSS property name.
        * css/CSSValueKeywords.in: New CSS value names.
        * css/StyleBuilderCustom.h:
        (WebCore::StyleBuilderCustom::applyValueFontSynthesis): Construct an internal
        representation of font-synthesis for a given CSSValue.
        * rendering/style/RenderStyle.h: Function signature.
        * rendering/style/RenderStyleConstants.h: Internal representation of
        font-synthesis.
        * rendering/style/StyleRareInheritedData.h: Where we actually store the bits
        related to this internal representation.

2015-04-24  Simon Fraser  <simon.fraser@apple.com>

        Have the web inspector report accurate memory use for layers with detached backing store
        https://bugs.webkit.org/show_bug.cgi?id=144179

        Reviewed by Dean Jackson.

        Only report memory use for layers whose backing store is attached.

        * platform/graphics/ca/GraphicsLayerCA.cpp:
        * platform/graphics/ca/PlatformCALayer.h: Add backingContributesToMemoryEstimate()
        which platforms can override if they allow backing store detaching.

2015-04-24  Simon Fraser  <simon.fraser@apple.com>

        Track a coverage rect through GraphicsLayerCA flushes, use it to mark backing store attached
        https://bugs.webkit.org/show_bug.cgi?id=144165

        Reviewed by Dean Jackson.
        
        Push in the clipRect as the secondary rect when doing a GraphicsLayerCA flush,
        which maps this rect through all the layers. Each layer tests intersection with
        this rect to determine if its backing store should be attached.
        
        This will cause us to drop backing store for layers that don't intersect the
        visible part of the view, which is too aggressive. A future patch will add
        some padding.

        Tests: compositing/visible-rect/coverage-clipped.html
               compositing/visible-rect/coverage-scrolling.html

        * platform/graphics/ca/GraphicsLayerCA.cpp:
        (WebCore::GraphicsLayerCA::GraphicsLayerCA): Remove initialization of things with
        C++11 initializers. Initialize the new m_intersectsCoverageRect bit.
        (WebCore::GraphicsLayerCA::flushCompositingState): Push the clipRect in as the
        secondary coverage rect. A future patch will inflate this on the way down the tree.
        (WebCore::GraphicsLayerCA::computeVisibleRect): If we're clipping, and
        we have a secondary coverage rect, that rect can be clipped to us too.
        (WebCore::GraphicsLayerCA::setVisibleAndCoverageRects):
        (WebCore::GraphicsLayerCA::recursiveCommitChanges):
        (WebCore::GraphicsLayerCA::commitLayerChangesBeforeSublayers):
        (WebCore::GraphicsLayerCA::updateBackingStoreAttachment):
        (WebCore::GraphicsLayerCA::dumpAdditionalProperties):
        * platform/graphics/ca/GraphicsLayerCA.h: New CoverageRectChanged bit.
        Use more C++11 initializers.
        (WebCore::GraphicsLayerCA::coverageRect):

2015-04-24  Dean Jackson  <dino@apple.com>

        AnimationController::scrollWasUpdated() shows up in scrolling profiles on pages that don't use scroll triggers
        https://bugs.webkit.org/show_bug.cgi?id=144173
        <rdar://problem/20526168>

        Reviewed by Simon Fraser.

        Keep a list of Animations that care about scroll updates, and only
        run the animation update if the list is not empty.

        Covered by existing tests.

        * page/animation/AnimationBase.cpp:
        (WebCore::AnimationBase::updateStateMachine): Tell the AnimationController
        if this is an animation that depends on scrolling.
        * page/animation/AnimationController.cpp:
        (WebCore::AnimationControllerPrivate::animationWillBeRemoved): Call the
        new removeFromAnimationsDependentOnScroll as this animation is deleted.
        (WebCore::AnimationControllerPrivate::addToAnimationsDependentOnScroll):
        (WebCore::AnimationControllerPrivate::removeFromAnimationsDependentOnScroll):
        (WebCore::AnimationControllerPrivate::scrollWasUpdated): Only update if
        there are animations that care.
        (WebCore::AnimationController::wantsScrollUpdates): Helper to expose this
        value to FrameView.
        * page/animation/AnimationController.h:
        * page/animation/AnimationControllerPrivate.h:
        (WebCore::AnimationControllerPrivate::wantsScrollUpdates):

2015-04-24  Tim Horton  <timothy_horton@apple.com>

        WKPDFView does not support password-protected PDFs
        https://bugs.webkit.org/show_bug.cgi?id=144162
        <rdar://problem/18411512>

        Reviewed by Andy Estes.

        * English.lproj/Localizable.strings:
        Add some localizable strings.

2015-04-24  David Kilzer  <ddkilzer@apple.com>

        Fix iOS EWS builds after updating to iOS 8.3 SDK

        * platform/spi/cocoa/CoreTextSPI.h:
        * rendering/RenderThemeIOS.mm:
        (WebCore::RenderThemeIOS::updateCachedSystemFontDescription):
        - Update __IPHONE_OS_VERSION_MIN_REQUIRED for these constants.

2015-04-24  Jer Noble  <jer.noble@apple.com>

        Unreviewed iOS build fix; add declarations to AVKitSPI.h

        * platform/spi/cocoa/AVKitSPI.h:

2015-04-24  Enrica Casucci  <enrica@apple.com>

        Make rangeExpandedAroundPositionByCharacters to all platforms.
        https://bugs.webkit.org/show_bug.cgi?id=144161

        Reviewed by Tim Horton.

        No change in functionality, simply making this available
        to all platforms.

        * editing/VisibleUnits.cpp:
        (WebCore::rangeExpandedAroundPositionByCharacters):
        * editing/VisibleUnits.h:
        * editing/mac/DataDetection.mm:
        * editing/mac/DictionaryLookup.h:
        * editing/mac/DictionaryLookup.mm:
        (WebCore::rangeExpandedAroundPositionByCharacters): Deleted.

2015-04-24  Andreas Kling  <akling@apple.com>

        Purge PassRefPtr from NamedNodeMap.
        <https://webkit.org/b/144091>

        Reviewed by Antti Koivisto.

        Use RefPtr instead of PassRefPtr in NamedNodeMap.

        * dom/NamedNodeMap.cpp:
        (WebCore::NamedNodeMap::getNamedItem):
        (WebCore::NamedNodeMap::getNamedItemNS):
        (WebCore::NamedNodeMap::removeNamedItem):
        (WebCore::NamedNodeMap::removeNamedItemNS):
        (WebCore::NamedNodeMap::setNamedItem):
        (WebCore::NamedNodeMap::setNamedItemNS):
        (WebCore::NamedNodeMap::item):
        * dom/NamedNodeMap.h:

2015-04-24  Commit Queue  <commit-queue@webkit.org>

        Unreviewed, rolling out r183266.
        https://bugs.webkit.org/show_bug.cgi?id=144164

        Broke 32-bit build and a test. (Requested by ap on #webkit).

        Reverted changeset:

        "AX: richer text change notifications (142719)"
        https://bugs.webkit.org/show_bug.cgi?id=142719
        http://trac.webkit.org/changeset/183266

2015-04-24  Brady Eidson  <beidson@apple.com>

        Origin header is preserved on cross-origin redirects.
        https://bugs.webkit.org/show_bug.cgi?id=144157.

        Reviewed by Sam Weinig.

        Tests: http/tests/security/cors-post-redirect-301.html
               http/tests/security/cors-post-redirect-302.html
               http/tests/security/cors-post-redirect-307.html
               http/tests/security/cors-post-redirect-308.html

        * platform/network/cf/ResourceHandleCFNet.cpp:
        (WebCore::ResourceHandle::willSendRequest): Always clear any origin header for cross-origin redirects.
        * platform/network/mac/ResourceHandleMac.mm:
        (WebCore::ResourceHandle::willSendRequest): Ditto.

2015-04-24  Brent Fulgham  <bfulgham@apple.com>

        Immediate action not functional for embedded PDFs
        https://bugs.webkit.org/show_bug.cgi?id=143952
        <rdar://problem/19842365>        

        Reviewed by Tim Horton.

        Add a dictionary lookup method to allow us to retrieve
        relevant information for PDFs.

        * editing/mac/DictionaryLookup.mm:
        (WebCore::dictionaryLookupForPDFSelection): Updated to properly handle cases
        where the Lookup engine selects a smaller range than the initial automatic
        selection performed on clicking.

2015-04-24  Simon Fraser  <simon.fraser@apple.com>

        Make it possible to map a secondary quad through TransformState
        https://bugs.webkit.org/show_bug.cgi?id=144156

        Reviewed by Dean Jackson.

        A future patch will need to map two quads simultaneously through TransformState,
        so add the ability to provide an optional secondary quad.
        
        This patch also firms up the setQuad() contract, fixing webkit.org/b/106680,
        requiring the state to be flattened when setting the quad (and now, the secondary quad).
        Previously, setQuad implicitly flattened but failed to update m_mapPoint when
        doing so.

        * platform/graphics/ca/GraphicsLayerCA.cpp:
        (WebCore::GraphicsLayerCA::computeVisibleRect): Now we have to explicitly flatten
        before setting the quad.
        * platform/graphics/ca/GraphicsLayerCA.h: Drive-up #include removal.
        * platform/graphics/transforms/TransformState.cpp:
        (WebCore::TransformState::operator=): Copy the secondary quad if we have one.
        (WebCore::TransformState::translateMappedCoordinates): Move the secondary quad
        if we have one.
        (WebCore::TransformState::mappedQuad): Code factored into mapQuad().
        (WebCore::TransformState::mappedSecondaryQuad): Return the secondary quad mapped
        into the state's current coordinate space.
        (WebCore::TransformState::mapQuad): Factored code.
        * platform/graphics/transforms/TransformState.h:
        (WebCore::TransformState::setQuad): Make the contract more explicit with assertions.
        (WebCore::TransformState::setSecondaryQuad): Ditto when setting the secondary quad.
        (WebCore::TransformState::lastPlanarSecondaryQuad):
        (WebCore::TransformState::lastPlanarQuad): Deleted.

2015-04-24  Myles C. Maxfield  <mmaxfield@apple.com>

        [iOS] Reimplement r182512 and r183153 in a cleaner way
        https://bugs.webkit.org/show_bug.cgi?id=144151

        Reviewed by Enrica Casucci.

        On iOS, we create FontPlatformData's ctFont() by round tripping it through a CGFontRef.
        This causes the resultant ctFont() to lose system-font-ness. Patches r182512 and r183153
        react to this fact by making users of a FontPlatformData use the original font passed in
        to the FontPlatformData instead of the FontPlatformData's ctFont(), but only if we
        detect that the FontPlatformData represents a system font (the underlying APIs have
        different behavior for system fonts and non-system-fonts).

        However, on OS X, we create a FontPlatformData's ctFont() directly from the original
        font passed in to the constructor. This preserves system-font-ness (because it no
        longer has the CGFontRef in the middle of the transformation). Therefore, OS X has the
        correct behavior regarding system fonts.

        The difference between the two ctFont() creation codepaths seems to be historical
        rather than intentional. Rather than change all the call sites of ctFont() to use a
        different platform font object if a system font is detected, a cleaner solution is to
        simply unify the two ctFont() creation codepaths to the version which preserves system-
        font-ness. Doing this will make all users of FontPlatformData automatically have the
        correct behavior with no updates.

        This patch reverts the relevant parts of r182512 and r183153 in favor of this new
        solution.

        No new tests because there is no behavior change.

        * platform/graphics/FontPlatformData.h:
        (WebCore::FontPlatformData::hashTableDeletedFontValue): Deleted.
        * platform/graphics/cocoa/FontCocoa.mm:
        (WebCore::Font::platformWidthForGlyph):
        * platform/graphics/cocoa/FontPlatformDataCocoa.mm:
        (WebCore::FontPlatformData::ctFontSize):
        (WebCore::FontPlatformData::ctFont):
        * platform/graphics/mac/SimpleFontDataCoreText.cpp:
        (WebCore::Font::getCFStringAttributes):

2015-04-24  Myles C. Maxfield  <mmaxfield@apple.com>

        Unreviewed build fix.

        * platform/graphics/cocoa/FontPlatformDataCocoa.mm:
        (WebCore::FontPlatformData::FontPlatformData):

2015-04-24  Myles C. Maxfield  <mmaxfield@apple.com>

        [Cocoa] Start cleaning up FontPlatformData
        https://bugs.webkit.org/show_bug.cgi?id=144133

        Reviewed by Enrica Casucci.

        This is the start of a crusade to ultimately remove one of the CTFontRefs which we have inside
        FontPlatformData. This patch starts this effort out by removing the FontPlatformData
        constructor which accepts an NSFont. This constructor simply casts the NSFont to a CTFontRef,
        and forwards to another constructor. This patch moves this cast to callers of this constructor,
        thereby enumerating the places we have left that use NSFont with FontPlatformData.

        This patch also cleans up two of the constructors in FontPlatformData.cpp to forward on to
        a common constructor.

        This patch also collects places inside FontPlatformDataCocoa.mm where we check if a pointer is
        not 0 and not -1, and puts this common code inside a new function,
        FontPlatformData::isValidCTFontRef().

        No new tests because there is no behavior change.

        * platform/graphics/FontPlatformData.cpp:
        (WebCore::FontPlatformData::FontPlatformData): Coalesce constructors.
        * platform/graphics/FontPlatformData.h:
        (WebCore::FontPlatformData::isValidCTFontRef): Check if a pointer is not 0 and not -1.
        (WebCore::FontPlatformData::hashTableDeletedFontValue): Deleted.
        * platform/graphics/cocoa/FontCocoa.mm:
        (WebCore::Font::platformCreateScaledFont): Receive cast which has been hoisted from the deleted
        FontPlatformData constructor.
        (WebCore::Font::compositeFontReferenceFont): Ditto.
        * platform/graphics/cocoa/FontPlatformDataCocoa.mm:
        (WebCore::FontPlatformData::~FontPlatformData): Use isValidCTFontRef().
        (WebCore::FontPlatformData::platformDataInit): Ditto.
        (WebCore::FontPlatformData::platformDataAssign): Ditto.
        (WebCore::FontPlatformData::platformIsEqual): Simplify preprocessor macros.
        (WebCore::FontPlatformData::allowsLigatures): Remove unnecessary NSFont use.
        (WebCore::FontPlatformData::FontPlatformData): Deleted.
        * platform/graphics/mac/ComplexTextControllerCoreText.mm:
        (WebCore::ComplexTextController::collectComplexTextRunsForCharacters): Remove unnecessary
        NSFont use.
        * platform/graphics/mac/FontCacheMac.mm:
        (WebCore::FontCache::systemFallbackForCharacters): Receive cast which has been hoisted from the
        deleted FontPlatformData constructor.
        (WebCore::FontCache::createFontPlatformData): Ditto.
        * platform/mac/DragImageMac.mm:
        (WebCore::fontFromNSFont): Ditto.
        (WebCore::widthWithFont): Ditto.
        (WebCore::drawAtPoint): Ditto.

2015-04-24  Doug Russell  <d_russell@apple.com>

        AX: richer text change notifications (142719)
        https://bugs.webkit.org/show_bug.cgi?id=142719

        Reviewed by Darin Adler.

        Richer accessibility value change notifications. Introduce AXTextEditType, postTextStateChangeNotification and postTextReplacementNotification to give assistive tech apps more reliable context for responding to changes in web content. Also implement a mechanism to post value changes in password form fields in coalesced ticks to thwart analyzing the cadence of changes.

        Richer accessibility selection change notifications. Introduce AXTextStateChangeIntent, and an overload of postTextReplacementNotification to give assistive tech apps more reliable context for responding to changes in web content selection. Also block posting selection changes on password fields.

        Tests: platform/mac/accessibility/input-replacevalue-userinfo.html
               platform/mac/accessibility/selection-change-userinfo.html
               platform/mac/accessibility/value-change-userinfo.html

        * CMakeLists.txt:
        * WebCore.vcxproj/WebCore.vcxproj:
        * WebCore.vcxproj/WebCore.vcxproj.filters:
        * WebCore.xcodeproj/project.pbxproj:
        * accessibility/AXObjectCache.cpp:
        (WebCore::AXObjectCache::AXObjectCache):
        (WebCore::AXObjectCache::notificationPostTimerFired):
        (WebCore::AXObjectCache::passwordNotificationPostTimerFired):
        (WebCore::AXObjectCache::showIntent):
        (WebCore::AXObjectCache::setTextSelectionIntent):
        (WebCore::isPasswordFieldOrContainedByPasswordField):
        (WebCore::AXObjectCache::postTextStateChangeNotification):
        (WebCore::AXObjectCache::postTextReplacementNotification):
        (WebCore::AXObjectCache::enqueuePasswordValueChangeNotification):
        (WebCore::AXObjectCache::rootWebArea):
        (WebCore::AXObjectCache::selectedChildrenChanged): Deleted.
        * accessibility/AXObjectCache.h:
        (WebCore::AXObjectCache::postTextStateChangeNotification):
        (WebCore::AXObjectCache::postTextReplacementNotification):
        (WebCore::AXObjectCache::postTextStateChangePlatformNotification):
        (WebCore::AXObjectCache::postTextReplacementPlatformNotification):
        (WebCore::AXObjectCache::textChangeForEditType):
        (WebCore::AXObjectCache::nodeTextChangePlatformNotification):
        (WebCore::AXObjectCache::computedObjectAttributeCache): Deleted.
        (WebCore::AXObjectCache::getOrCreate): Deleted.
        (WebCore::AXObjectCache::attachWrapper): Deleted.
        * accessibility/AXTextStateChangeIntent.h: Added.
        (WebCore::AXTextStateChangeIntent::AXTextStateChangeIntent):
        * accessibility/AccessibilityNodeObject.cpp:
        (WebCore::AccessibilityNodeObject::passwordFieldOrContainingPasswordField):
        * accessibility/AccessibilityNodeObject.h:
        * accessibility/AccessibilityObject.cpp:
        (WebCore::AccessibilityObject::isContainedByPasswordField):
        * accessibility/AccessibilityObject.h:
        (WebCore::AccessibilityObject::passwordFieldOrContainingPasswordField):
        (WebCore::AccessibilityObject::isPasswordField): Deleted.
        * accessibility/AccessibilityRenderObject.cpp:
        (WebCore::AccessibilityRenderObject::setSelectedTextRange):
        (WebCore::AccessibilityRenderObject::setSelectedVisiblePositionRange):
        * accessibility/AccessibilityScrollView.h:
        * accessibility/atk/AXObjectCacheAtk.cpp:
        (WebCore::AXObjectCache::textChangeForEditType):
        (WebCore::AXObjectCache::nodeTextChangePlatformNotification):
        (WebCore::AXObjectCache::postPlatformNotification): Deleted.
        * accessibility/ios/AXObjectCacheIOS.mm:
        (WebCore::AXObjectCache::postTextStateChangePlatformNotification):
        (WebCore::AXObjectCache::postTextReplacementPlatformNotification):
        * accessibility/mac/AXObjectCacheMac.mm:
        (WebCore::AXObjectCache::setShouldRepostNotificationsForTests):
        (WebCore::AXPostNotificationWithUserInfo):
        (WebCore::AXObjectCache::postPlatformNotification):
        (WebCore::AXObjectCache::postTextStateChangePlatformNotification):
        (WebCore::textReplacementChangeDictionary):
        (WebCore::AXObjectCache::postTextReplacementPlatformNotification):
        * accessibility/mac/WebAccessibilityObjectWrapperBase.h:
        * accessibility/mac/WebAccessibilityObjectWrapperBase.mm:
        (+[WebAccessibilityObjectWrapperBase accessibilitySetShouldRepostNotifications:]):
        (-[WebAccessibilityObjectWrapperBase accessibilityPostedNotification:]):
        (arrayRemovingNonJSONTypes):
        (dictionaryRemovingNonJSONTypes):
        (-[WebAccessibilityObjectWrapperBase accessibilityPostedNotification:userInfo:]):
        * accessibility/mac/WebAccessibilityObjectWrapperMac.h:
        * accessibility/mac/WebAccessibilityObjectWrapperMac.mm:
        (textMarkerRangeFromVisiblePositions):
        (-[WebAccessibilityObjectWrapper textMarkerRangeFromVisiblePositions:endPosition:]):
        * editing/AppendNodeCommand.cpp:
        (WebCore::AppendNodeCommand::AppendNodeCommand):
        (WebCore::sendAXTextChangedIgnoringLineBreaks):
        (WebCore::AppendNodeCommand::doApply):
        (WebCore::AppendNodeCommand::doUnapply):
        * editing/AppendNodeCommand.h:
        (WebCore::AppendNodeCommand::create):
        * editing/ApplyStyleCommand.cpp:
        (WebCore::ApplyStyleCommand::ApplyStyleCommand):
        (WebCore::ApplyStyleCommand::applyBlockStyle): Deleted.
        * editing/ApplyStyleCommand.h:
        * editing/CompositeEditCommand.cpp:
        (WebCore::EditCommandComposition::unapplyEditType):
        (WebCore::CompositeEditCommand::CompositeEditCommand):
        (WebCore::CompositeEditCommand::apply):
        (WebCore::CompositeEditCommand::insertParagraphSeparator):
        (WebCore::CompositeEditCommand::insertNodeBefore):
        (WebCore::CompositeEditCommand::appendNode):
        (WebCore::CompositeEditCommand::removeNodePreservingChildren):
        (WebCore::CompositeEditCommand::insertTextIntoNode):
        (WebCore::CompositeEditCommand::deleteTextFromNode):
        (WebCore::CompositeEditCommand::replaceTextInNode):
        (WebCore::CompositeEditCommand::moveParagraphs):
        (WebCore::EditCommandComposition::getNodesInCommand): Deleted.
        (WebCore::CompositeEditCommand::applyStyle): Deleted.
        (WebCore::CompositeEditCommand::insertLineBreak): Deleted.
        (WebCore::CompositeEditCommand::insertNodeAt): Deleted.
        (WebCore::CompositeEditCommand::removeChildrenInRange): Deleted.
        (WebCore::CompositeEditCommand::inputText): Deleted.
        * editing/CompositeEditCommand.h:
        * editing/DeleteFromTextNodeCommand.cpp:
        (WebCore::DeleteFromTextNodeCommand::DeleteFromTextNodeCommand):
        (WebCore::DeleteFromTextNodeCommand::doApply):
        (WebCore::DeleteFromTextNodeCommand::doUnapply):
        * editing/DeleteFromTextNodeCommand.h:
        (WebCore::DeleteFromTextNodeCommand::create):
        (WebCore::DeleteFromTextNodeCommand::deletedText):
        * editing/DeleteSelectionCommand.cpp:
        (WebCore::DeleteSelectionCommand::DeleteSelectionCommand):
        (WebCore::DeleteSelectionCommand::preservesTypingStyle): Deleted.
        * editing/DeleteSelectionCommand.h:
        (WebCore::DeleteSelectionCommand::create):
        * editing/DictationCommand.cpp:
        (WebCore::DictationCommand::insertTextRunWithoutNewlines):
        (WebCore::DictationCommand::insertParagraphSeparator):
        * editing/EditAction.h:
        * editing/EditCommand.cpp:
        (WebCore::EditCommand::EditCommand):
        (WebCore::EditCommand::editingAction):
        (WebCore::EditCommand::applyEditType):
        (WebCore::EditCommand::unapplyEditType):
        (WebCore::SimpleEditCommand::SimpleEditCommand):
        (WebCore::SimpleEditCommand::notifyAccessibilityForTextChange):
        (WebCore::EditCommand::setParent): Deleted.
        * editing/EditCommand.h:
        * editing/EditingAllInOne.cpp:
        * editing/Editor.cpp:
        (WebCore::Editor::handleTextEvent):
        (WebCore::Editor::deleteSelectionWithSmartDelete):
        (WebCore::Editor::replaceSelectionWithFragment):
        (WebCore::Editor::replaceSelectionWithText):
        (WebCore::Editor::appliedEditing):
        (WebCore::Editor::unappliedEditing):
        (WebCore::Editor::performCutOrCopy):
        (WebCore::Editor::markMisspellingsAfterTypingToWord):
        (WebCore::Editor::changeBackToReplacedString):
        (WebCore::Editor::transpose):
        (WebCore::Editor::changeSelectionAfterCommand):
        * editing/Editor.h:
        * editing/EditorCommand.cpp:
        (WebCore::executeInsertFragment):
        * editing/FrameSelection.cpp:
        (WebCore::FrameSelection::moveTo):
        (WebCore::FrameSelection::moveWithoutValidationTo):
        (WebCore::FrameSelection::setSelectionByMouseIfDifferent):
        (WebCore::FrameSelection::setSelectionWithoutUpdatingAppearance):
        (WebCore::FrameSelection::setSelection):
        (WebCore::FrameSelection::updateAndRevealSelection):
        (WebCore::isBoundary):
        (WebCore::FrameSelection::textSelectionIntent):
        (WebCore::FrameSelection::modify):
        (WebCore::FrameSelection::selectAll):
        (WebCore::FrameSelection::wordSelectionContainingCaretSelection):
        (WebCore::FrameSelection::modifyMovingBackward): Deleted.
        (WebCore::FrameSelection::selectFrameElementInParentIfFullySelected): Deleted.
        (WebCore::FrameSelection::selectionAtWordStart): Deleted.
        * editing/FrameSelection.h:
        (WebCore::FrameSelection::notifyAccessibilityForSelectionChange):
        (WebCore::FrameSelection::selection): Deleted.
        * editing/InsertIntoTextNodeCommand.cpp:
        (WebCore::InsertIntoTextNodeCommand::InsertIntoTextNodeCommand):
        (WebCore::InsertIntoTextNodeCommand::doApply):
        (WebCore::InsertIntoTextNodeCommand::doUnapply):
        * editing/InsertIntoTextNodeCommand.h:
        (WebCore::InsertIntoTextNodeCommand::create):
        (WebCore::InsertIntoTextNodeCommand::insertedText):
        * editing/InsertNodeBeforeCommand.cpp:
        (WebCore::InsertNodeBeforeCommand::InsertNodeBeforeCommand):
        (WebCore::InsertNodeBeforeCommand::doApply):
        (WebCore::InsertNodeBeforeCommand::doUnapply):
        * editing/InsertNodeBeforeCommand.h:
        (WebCore::InsertNodeBeforeCommand::create):
        * editing/InsertParagraphSeparatorCommand.cpp:
        (WebCore::InsertParagraphSeparatorCommand::InsertParagraphSeparatorCommand):
        * editing/InsertParagraphSeparatorCommand.h:
        (WebCore::InsertParagraphSeparatorCommand::create):
        * editing/InsertTextCommand.cpp:
        (WebCore::InsertTextCommand::InsertTextCommand):
        * editing/InsertTextCommand.h:
        (WebCore::InsertTextCommand::create):
        (WebCore::InsertTextCommand::createWithMarkerSupplier):
        * editing/MoveSelectionCommand.cpp:
        (WebCore::MoveSelectionCommand::doApply):
        * editing/RemoveNodePreservingChildrenCommand.cpp:
        (WebCore::RemoveNodePreservingChildrenCommand::RemoveNodePreservingChildrenCommand):
        * editing/RemoveNodePreservingChildrenCommand.h:
        (WebCore::RemoveNodePreservingChildrenCommand::create):
        * editing/ReplaceDeleteFromTextNodeCommand.cpp: Copied from Source/WebCore/editing/AppendNodeCommand.h.
        (WebCore::ReplaceDeleteFromTextNodeCommand::ReplaceDeleteFromTextNodeCommand):
        (WebCore::ReplaceDeleteFromTextNodeCommand::notifyAccessibilityForTextChange):
        * editing/ReplaceDeleteFromTextNodeCommand.h: Copied from Source/WebCore/editing/AppendNodeCommand.h.
        * editing/ReplaceInsertIntoTextNodeCommand.cpp: Added.
        (WebCore::ReplaceInsertIntoTextNodeCommand::ReplaceInsertIntoTextNodeCommand):
        (WebCore::ReplaceInsertIntoTextNodeCommand::notifyAccessibilityForTextChange):
        * editing/ReplaceInsertIntoTextNodeCommand.h: Copied from Source/WebCore/editing/RemoveNodePreservingChildrenCommand.h.
        * editing/ReplaceSelectionCommand.cpp:
        (WebCore::ReplaceSelectionCommand::ReplaceSelectionCommand):
        (WebCore::ReplaceSelectionCommand::InsertedNodes::didReplaceNode): Deleted.
        (WebCore::ReplaceSelectionCommand::insertAsListItems): Deleted.
        * editing/ReplaceSelectionCommand.h:
        (WebCore::ReplaceSelectionCommand::create):
        * editing/TypingCommand.cpp:
        (WebCore::TypingCommand::insertTextRunWithoutNewlines):
        (WebCore::TypingCommand::insertParagraphSeparator):
        * editing/atk/FrameSelectionAtk.cpp:
        (WebCore::FrameSelection::notifyAccessibilityForSelectionChange):
        * editing/ios/DictationCommandIOS.cpp:
        (WebCore::DictationCommandIOS::DictationCommandIOS):
        * editing/ios/DictationCommandIOS.h:
        * editing/mac/FrameSelectionMac.mm:
        (WebCore::FrameSelection::notifyAccessibilityForSelectionChange):
        * html/HTMLTextFormControlElement.cpp:
        (WebCore::HTMLTextFormControlElement::setInnerTextValue):
        * page/DragController.cpp:
        (WebCore::DragController::concludeEditDrag):
        * page/EventHandler.cpp:
        (WebCore::setInitialKeyboardSelection):
        * page/FocusController.cpp:
        (WebCore::FocusController::advanceFocusInDocumentOrder):

2015-04-24  Darin Adler  <darin@apple.com>

        Convert OwnPtr and PassOwnPtr uses to std::unique_ptr
        https://bugs.webkit.org/show_bug.cgi?id=128007

        Reviewed by Anders Carlsson.

        * PlatformEfl.cmake: Removed OwnPtrCairo source files.
        * PlatformGTK.cmake: Ditto.
        * PlatformWinCairo.cmake: Ditto.
        * WebCore.vcxproj/WebCore.vcxproj: Ditto.
        * WebCore.vcxproj/WebCore.vcxproj.filters: Ditto.

        * platform/CrossThreadCopier.cpp: Removed OwnPtr specialization.
        * platform/CrossThreadCopier.h: Ditto.

        * platform/ScrollAnimatorNone.cpp: Removed unneeded include.
        * platform/ThreadGlobalData.cpp: Ditto.
        * platform/ThreadGlobalData.h: Ditto.
        * platform/graphics/avfoundation/cf/CDMSessionAVFoundationCF.h: Ditto.

        * platform/graphics/avfoundation/cf/MediaPlayerPrivateAVFoundationCF.cpp:
        (WebCore::AVFWrapper::processNotification): Use unique_ptr.
        (WebCore::AVFWrapper::notificationCallback): Use make_unique.
        (WebCore::AVFWrapper::platformLayer): Ditto.

        * platform/graphics/cairo/CairoUtilities.cpp:
        (WebCore::appendPathToCairoContext): Use cairo_path_destroy directly
        instead of a smart pointer; the code path is simple enough that it is
        clear this way.

        * platform/graphics/cairo/ImageBufferCairo.cpp:
        (WebCore::ImageBuffer::ImageBuffer): Use make_unique.
        * platform/graphics/cairo/ImageBufferDataCairo.h: Use unique_ptr.

        * platform/graphics/cairo/OwnPtrCairo.cpp: Removed.
        * platform/graphics/cairo/OwnPtrCairo.h: Removed.

        * platform/graphics/cairo/PathCairo.cpp:
        (WebCore::Path::Path): Use cairo_path_destroy directly instead of a
        smart pointer; the code path is simple enough that it is clear this way.
        (WebCore::Path::operator=): Ditto.
        (WebCore::Path::addPath): Ditto.
        (WebCore::Path::apply): Ditto.
        * platform/graphics/cairo/PlatformContextCairo.cpp:
        (WebCore::PlatformContextCairo::clipForPatternFilling): Ditto.

        * platform/graphics/freetype/FontCacheFreeType.cpp: Removed unneeded include.
        * platform/graphics/glx/GLContextGLX.cpp: Ditto.

        * platform/graphics/surfaces/egl/EGLSurface.cpp:
        (WebCore::EGLOffScreenSurface::EGLOffScreenSurface): Use make_unique.

        * platform/graphics/surfaces/glx/GLXConfigSelector.h:
        (WebCore::GLXConfigSelector::findMatchingConfig): Use unique_ptr with X11Deleter.
        (WebCore::GLXConfigSelector::findMatchingConfigWithVisualId): Ditto.
        * platform/graphics/surfaces/glx/GLXSurface.cpp:
        (WebCore::GLXTransportSurface::GLXTransportSurface): Ditto.
        (WebCore::GLXOffScreenSurface::initialize): Ditto.

        * platform/graphics/surfaces/glx/OwnPtrX11.h: Removed.

        * platform/graphics/surfaces/glx/X11Helper.cpp:
        (WebCore::X11Helper::createOffScreenWindow): Use unique_ptr with X11Deleter.
        (WebCore::X11Helper::createPixmap): Ditto.

        * platform/graphics/surfaces/glx/X11Helper.h: Added X11Deleter,
        which can be used with unique_ptr to make it call XFree instead of delete.

        * platform/graphics/win/FullScreenController.cpp: Use unique_ptr.
        (FullScreenController::FullScreenController): Use make_unique.
        (FullScreenController::enterFullScreen): Ditto.
        * platform/graphics/win/FullScreenController.h: Use unique_ptr.

        * platform/graphics/win/GraphicsContext3DWin.cpp: Removed unneeded includes.

        * platform/graphics/win/WKCAImageQueue.cpp:
        (WebCore::WKCAImageQueue::WKCAImageQueue): Use make_unique.
        * platform/graphics/win/WKCAImageQueue.h: Use unique_ptr.

        * platform/network/NetworkStorageSessionStub.cpp: Removed unneeded include.

        * platform/network/ResourceHandleInternal.h: Use unique_ptr.

        * platform/network/curl/MultipartHandle.cpp:
        (WebCore::MultipartHandle::create): Deleted.
        * platform/network/curl/MultipartHandle.h: Deleted the create function.

        * platform/network/curl/ResourceHandleManager.cpp:
        (WebCore::headerCallback): Use make_unique.

2015-04-24  Per Arne Vollan  <peavo@outlook.com>

        [Win] Popup menu is not accessible.
        https://bugs.webkit.org/show_bug.cgi?id=141704

        Reviewed by Brent Fulgham.

        Get accessibility information for items in popup menus.

        * platform/win/PopupMenuWin.cpp:
        (WebCore::PopupMenuWin::onGetObject):
        (WebCore::PopupMenuWin::wndProc):
        (WebCore::AccessiblePopupMenu::AccessiblePopupMenu):
        (WebCore::AccessiblePopupMenu::~AccessiblePopupMenu):
        (WebCore::AccessiblePopupMenu::QueryInterface):
        (WebCore::AccessiblePopupMenu::AddRef):
        (WebCore::AccessiblePopupMenu::Release):
        (WebCore::AccessiblePopupMenu::GetTypeInfoCount):
        (WebCore::AccessiblePopupMenu::GetTypeInfo):
        (WebCore::AccessiblePopupMenu::GetIDsOfNames):
        (WebCore::AccessiblePopupMenu::Invoke):
        (WebCore::AccessiblePopupMenu::get_accParent):
        (WebCore::AccessiblePopupMenu::get_accChildCount):
        (WebCore::AccessiblePopupMenu::get_accChild):
        (WebCore::AccessiblePopupMenu::get_accName):
        (WebCore::AccessiblePopupMenu::get_accValue):
        (WebCore::AccessiblePopupMenu::get_accDescription):
        (WebCore::AccessiblePopupMenu::get_accRole):
        (WebCore::AccessiblePopupMenu::get_accState):
        (WebCore::AccessiblePopupMenu::get_accHelp):
        (WebCore::AccessiblePopupMenu::get_accKeyboardShortcut):
        (WebCore::AccessiblePopupMenu::get_accFocus):
        (WebCore::AccessiblePopupMenu::get_accSelection):
        (WebCore::AccessiblePopupMenu::get_accDefaultAction):
        (WebCore::AccessiblePopupMenu::accSelect):
        (WebCore::AccessiblePopupMenu::accLocation):
        (WebCore::AccessiblePopupMenu::accNavigate):
        (WebCore::AccessiblePopupMenu::accHitTest):
        (WebCore::AccessiblePopupMenu::accDoDefaultAction):
        (WebCore::AccessiblePopupMenu::put_accName):
        (WebCore::AccessiblePopupMenu::put_accValue):
        (WebCore::AccessiblePopupMenu::get_accHelpTopic):
        * platform/win/PopupMenuWin.h:
        * platform/win/ScrollbarThemeWin.h:
        (WebCore::ScrollbarThemeWin::hasButtons): Deleted.

2015-04-23  Antti Koivisto  <antti@apple.com>

        Memory cache live resources repeatedly purged during painting
        https://bugs.webkit.org/show_bug.cgi?id=144104
        <rdar://problem/20667695>

        Reviewed by Chris Dumez.

        On some PLT pa