2011-02-21 Pratik Solanki Reviewed by Darin Adler. Remove global initializer in CookieStorageCFNet.cpp https://bugs.webkit.org/show_bug.cgi?id=54905 * platform/network/cf/CookieStorageCFNet.cpp: (WebCore::currentCookieStorage): (WebCore::setCurrentCookieStorage): (WebCore::setCookieStoragePrivateBrowsingEnabled): 2011-02-18 Jer Noble Reviewed by Sam Weinig. WebKit2: Media document videos play only sound, no video (affects trailers.apple.com) https://bugs.webkit.org/show_bug.cgi?id=54771 Now that video is accelerated, we no longer need to special case video playing within a media document. * platform/graphics/mac/MediaPlayerPrivateQTKit.mm: (WebCore::MediaPlayerPrivateQTKit::preferredRenderingMode): (WebCore::MediaPlayerPrivateQTKit::supportsAcceleratedRendering): 2011-02-21 Alexey Proskuryakov Reviewed by Adam Roben. REGRESSION (WebKit2): HTTP requests time out after 60 seconds https://bugs.webkit.org/show_bug.cgi?id=54755 No new tests - it won't be great to have a test that runs for a minute. It's now possible to set a default timeout to be used at ResourceRequest creation. If one hasn't been set, ResourceRequest will behave as before (use NSURLRequest default on Mac, or use INT_MAX on other platforms). * WebCore.exp.in: * platform/network/ResourceRequestBase.cpp: (WebCore::ResourceRequestBase::defaultTimeoutInterval): Get the ResourceRequest notion of default timeout interval (may be 0 if using NSURLRequest default). (WebCore::ResourceRequestBase::setDefaultTimeoutInterval): Set the static member variable. (WebCore::ResourceRequestBase::updatePlatformRequest): Added an assertion that resource request is updated. Plaform code often calls updateResourceRequest() indirectly from this function, and that must obviously be a no-op. (WebCore::ResourceRequestBase::updateResourceRequest): Added an assertion in the opposite direction. * platform/network/ResourceRequestBase.h: Changed "unspecifiedTimeoutInterval" to "defaultTimeoutInterval". It has been used as default by most platforms anyway. (WebCore::ResourceRequestBase::ResourceRequestBase): * platform/network/mac/ResourceRequestMac.mm: (WebCore::ResourceRequest::doUpdatePlatformRequest): Now zero is the magic value, not INT_MAX. We'll use NSURLRequest default if neither setTimeoutInterval() nor setDefaultTimeoutInterval() has been called. 2011-02-21 Martin Robinson Reviewed by Xan Lopez. [GTK] editing/pasteboard/dataTransfer-setData-getData.html fails https://bugs.webkit.org/show_bug.cgi?id=54895 Correct the GTK+ clipboard implementation to know that "text" is an alias for "text/html" data. This corrects the failing test. * platform/gtk/ClipboardGtk.cpp: (WebCore::dataObjectTypeFromHTMLClipboardType): Accept "text" as well as "Text". (WebCore::ClipboardGtk::types): Advertise "text" as well as "Text". 2011-02-21 Vsevolod Vlasov Reviewed by Pavel Feldman. ProcessingInstruction should provide a way to know if it is of CSS type. https://bugs.webkit.org/show_bug.cgi?id=54868 * dom/ProcessingInstruction.cpp: (WebCore::ProcessingInstruction::ProcessingInstruction): (WebCore::ProcessingInstruction::checkStyleSheet): (WebCore::ProcessingInstruction::setCSSStyleSheet): * dom/ProcessingInstruction.h: (WebCore::ProcessingInstruction::isCSS): 2011-02-21 Simon Fraser Put JSDOMImplementationCustom.cpp where it belongs in the project. * WebCore.xcodeproj/project.pbxproj: 2011-02-21 Gavin Barraclough Reviewed by Sam Weinig. Bug 54894 - Make inheritance structure described by ClassInfo match C++ class hierarchy. The ClassInfo objects describe an inheritance hierarchy, with each ClassInfo instance containing a pointer to its parent class. These links should reflect the inheritance hierarchy of C++ classes below JSObject. For the large part it does, but in some cases entries in the C++ hierarchy are skipped over. This presently likely doesn't matter, since intervening C++ classes may not have ClassInfo - but would be a potential bug were ClassInfo were to be added. * bindings/js/JSAudioConstructor.cpp: * bindings/js/JSDOMGlobalObject.cpp: * bindings/js/JSImageConstructor.cpp: * bindings/js/JSOptionConstructor.cpp: * bindings/scripts/CodeGeneratorJS.pm: * bridge/objc/objc_runtime.mm: * bridge/runtime_object.cpp: 2011-02-21 Pavel Feldman Reviewed by Yury Semikhatsky. Web Inspector: navigation does not work after reopening frontend. https://bugs.webkit.org/show_bug.cgi?id=54879 Test: http/tests/inspector-enabled/open-close-open.html * inspector/InspectorAgent.cpp: (WebCore::InspectorAgent::setFrontend): * inspector/InspectorState.cpp: (WebCore::InspectorState::unmute): * inspector/InspectorState.h: 2011-02-21 Adam Roben Protect the PluginView when evaluating javascript: URLs Fixes plugins/get-url-with-javascript-destroying-plugin.html crashing on Windows since it was added Reviewed by Sam Weinig. * plugins/PluginView.cpp: (WebCore::PluginView::performRequest): Protect the PluginView, not just its parent frame, when evaluating javascript: URLs. 2011-02-21 Martin Robinson Fix GTK+ build after r79223. * GNUmakefile.am: Add file missing from the source list. 2011-02-14 Alexander Pavlov Reviewed by Yury Semikhatsky. Web Inspector: audits should not warn about gzip compression for 304s https://bugs.webkit.org/show_bug.cgi?id=54343 Do not run compression-related audits on 304 resources. Drive-by: handle multi-encoding resources (Content-Encoding: sdhc,gzip) correctly. * inspector/front-end/AuditRules.js: (WebInspector.AuditRules.GzipRule.prototype.doRun): (WebInspector.AuditRules.GzipRule.prototype._isCompressed): 2011-02-08 Anton Muhin Reviewed by Adam Barth and Alexey Proskuryakov. Propagate security origin of parent document into HTML documents created with DOMImplementation https://bugs.webkit.org/show_bug.cgi?id=53611 This restores invariant that JS wrappers residing in the same JS context should come from the same security origin. Absence of regressions is covered by the current tests. Different security origin of DOMImplementation is difficult to check with layout tests as DOMImplementation resides in the same JS context as parent document and therefore there are no security origin checks. This is observable however in C++. * Android.jscbindings.mk: * CMakeLists.txt: * WebCore.gypi: * WebCore.pro: * WebCore.vcproj/WebCore.vcproj: * WebCore.xcodeproj/project.pbxproj: * bindings/v8/V8GCController.cpp: (WebCore::NodeGrouperVisitor::visitDOMWrapper): * dom/DOMImplementation.cpp: (WebCore::DOMImplementation::DOMImplementation): (WebCore::DOMImplementation::createDocument): * dom/DOMImplementation.h: (WebCore::DOMImplementation::create): (WebCore::DOMImplementation::documentDestroyed): (WebCore::DOMImplementation::parentDocument): * dom/DOMImplementation.idl: * dom/Document.cpp: (WebCore::Document::~Document): (WebCore::Document::implementation): * dom/Document.h: 2011-02-21 Andrey Adaikin Reviewed by Pavel Feldman. Web Inspector: [Text editor] Optimize editing updates in gutter panel https://bugs.webkit.org/show_bug.cgi?id=54866 * inspector/front-end/TextViewer.js: (WebInspector.TextViewer.prototype.set startEditingListener): (WebInspector.TextViewer.prototype.set endEditingListener): (WebInspector.TextViewer.prototype.endUpdates): (WebInspector.TextViewer.prototype._enterInternalTextChangeMode): (WebInspector.TextViewer.prototype._exitInternalTextChangeMode): (WebInspector.TextEditorChunkedPanel.prototype._chunkNumberForLine): (WebInspector.TextEditorGutterPanel.prototype._expandChunks): (WebInspector.TextEditorGutterPanel.prototype.textChanged): (WebInspector.TextEditorMainPanel.prototype._updateChunksForRanges): 2011-02-21 Alexander Pavlov Reviewed by Yury Semikhatsky. Web Inspector: [Audits] Image dimensions in inline style not checked https://bugs.webkit.org/show_bug.cgi?id=54738 * inspector/front-end/AuditRules.js: (WebInspector.AuditRules.ImageDimensionsRule.prototype.doRun): 2011-02-21 Yury Semikhatsky Reviewed by Pavel Feldman. Web Inspector: create and destroy DOM agent along with InspectorAgent https://bugs.webkit.org/show_bug.cgi?id=54875 * GNUmakefile.am: * WebCore.gypi: * WebCore.pro: * WebCore.vcproj/WebCore.vcproj: * WebCore.xcodeproj/project.pbxproj: * inspector/InstrumentingAgents.h: Added. Agents that want to instrument WebCore should register themselve on this object. (WebCore::InstrumentingAgents::InstrumentingAgents): (WebCore::InstrumentingAgents::~InstrumentingAgents): (WebCore::InstrumentingAgents::inspectorBrowserDebuggerAgent): (WebCore::InstrumentingAgents::setInspectorBrowserDebuggerAgent): (WebCore::InstrumentingAgents::inspectorConsoleAgent): (WebCore::InstrumentingAgents::setInspectorConsoleAgent): (WebCore::InstrumentingAgents::inspectorDOMAgent): (WebCore::InstrumentingAgents::setInspectorDOMAgent): (WebCore::InstrumentingAgents::inspectorDOMStorageAgent): (WebCore::InstrumentingAgents::setInspectorDOMStorageAgent): (WebCore::InstrumentingAgents::inspectorDatabaseAgent): (WebCore::InstrumentingAgents::setInspectorDatabaseAgent): (WebCore::InstrumentingAgents::inspectorDebuggerAgent): (WebCore::InstrumentingAgents::setInspectorDebuggerAgent): (WebCore::InstrumentingAgents::inspectorProfilerAgent): (WebCore::InstrumentingAgents::setInspectorProfilerAgent): (WebCore::InstrumentingAgents::inspectorResourceAgent): (WebCore::InstrumentingAgents::setInspectorResourceAgent): (WebCore::InstrumentingAgents::inspectorRuntimeAgent): (WebCore::InstrumentingAgents::setInspectorRuntimeAgent): (WebCore::InstrumentingAgents::inspectorStorageAgent): (WebCore::InstrumentingAgents::setInspectorStorageAgent): (WebCore::InstrumentingAgents::inspectorTimelineAgent): (WebCore::InstrumentingAgents::setInspectorTimelineAgent): * inspector/InspectorAgent.cpp: (WebCore::InspectorAgent::InspectorAgent): (WebCore::InspectorAgent::setFrontend): (WebCore::InspectorAgent::disconnectFrontend): (WebCore::InspectorAgent::createFrontendLifetimeAgents): (WebCore::InspectorAgent::releaseFrontendLifetimeAgents): (WebCore::InspectorAgent::didCommitLoad): (WebCore::InspectorAgent::domContentLoadedEventFired): (WebCore::InspectorAgent::loadEventFired): * inspector/InspectorAgent.h: * inspector/InspectorCSSAgent.cpp: (WebCore::InspectorCSSAgent::InspectorCSSAgent): (WebCore::InspectorCSSAgent::~InspectorCSSAgent): * inspector/InspectorCSSAgent.h: * inspector/InspectorConsoleAgent.cpp: (WebCore::InspectorConsoleAgent::InspectorConsoleAgent): (WebCore::InspectorConsoleAgent::~InspectorConsoleAgent): (WebCore::InspectorConsoleAgent::clearConsoleMessages): (WebCore::InspectorConsoleAgent::clearFrontend): (WebCore::InspectorConsoleAgent::resourceRetrievedByXMLHttpRequest): (WebCore::InspectorConsoleAgent::setMonitoringXHREnabled): (WebCore::InspectorConsoleAgent::setConsoleMessagesEnabled): (WebCore::InspectorConsoleAgent::addConsoleMessage): * inspector/InspectorConsoleAgent.h: * inspector/InspectorDOMAgent.cpp: DOM agent now lives even when the front-end is not attached. (WebCore::InspectorDOMAgent::InspectorDOMAgent): (WebCore::InspectorDOMAgent::setFrontend): Add DOM agent to the set of active agents when the front-end is attached. (WebCore::InspectorDOMAgent::clearFrontend): * inspector/InspectorDOMAgent.h: (WebCore::InspectorDOMAgent::create): 2011-02-21 Csaba Osztrogonác Reviewed by Andreas Kling. [Qt][V8]REGRESSION(r79157): Fix build https://bugs.webkit.org/show_bug.cgi?id=54871 * bridge/npruntime_internal.h: Add one more undef because of evil X11 macro. 2011-02-16 Vitaly Repeshko Reviewed by Mihai Parparita. [V8] SerializedScriptValue: fix JS exception handling. https://bugs.webkit.org/show_bug.cgi?id=54555 Added checks for exceptions and empty handles: * bindings/v8/SerializedScriptValue.cpp: (WebCore::Serializer::Serializer): (WebCore::Serializer::serialize): (WebCore::Serializer::checkException): (WebCore::Serializer::reportFailure): (WebCore::Serializer::ArrayState::advance): (WebCore::Serializer::AbstractObjectState::AbstractObjectState): (WebCore::Serializer::AbstractObjectState::advance): (WebCore::Serializer::push): (WebCore::Serializer::handleError): (WebCore::Serializer::newObjectState): (WebCore::Serializer::doSerialize): (WebCore::SerializedScriptValue::SerializedScriptValue): 2011-02-21 Ryosuke Niwa Reviewed by Kent Tamura. Add selectionchange event https://bugs.webkit.org/show_bug.cgi?id=45712 Added the support for asynchronous selectionchange event, which is fired whenever selection is changed. This event is not cancelable and does not bubble. An event listener can be attached to a document by body element's onselectionchange attribute or via document's onselectionchange property. Note that WebKit's implementation fires the event asynchronously whereas Internet Explorer's implementation fires it synchronously. This implies that a script that modify selection (e.g. via selection's addRange) will not observe the event before the control returns to JavaScript. See also: http://msdn.microsoft.com/en-us/library/ms536968(VS.85).aspx Tests: fast/events/selectionchange-iframe.html fast/events/selectionchange-user-initiated.html * dom/Document.cpp: (WebCore::Document::enqueueDocumentEvent): Added. * dom/Document.h: Added selectionchange event listener. * dom/Document.idl: Added onselectionchagne attribute. * dom/EventNames.h: Added selectionchange * editing/SelectionController.cpp: (WebCore::SelectionController::setSelection): Fires selectionchange event. * html/HTMLAttributeNames.in: Added onselectionchange. * html/HTMLBodyElement.cpp: (WebCore::HTMLBodyElement::parseMappedAttribute): Handles onselectionchange attribute. 2011-02-21 Ilya Tikhonovsky Reviewed by Pavel Feldman. Web Inspector: protocol error messages are dumping incorrectly in Layout tests. https://bugs.webkit.org/show_bug.cgi?id=54859 * inspector/CodeGeneratorInspector.pm: 2011-02-21 Pavel Feldman Reviewed by Yury Semikhatsky. Web Inspector: generate protocol documentation based on IDL. https://bugs.webkit.org/show_bug.cgi?id=54822 * inspector/CodeGeneratorInspector.pm: 2011-02-21 Andoni Morales Alastruey Reviewed by Martin Robinson. [GStreamer] Add URI queries support in webkitwebsrc https://bugs.webkit.org/show_bug.cgi?id=54627 This allow replying to URI queries from downstream elements with the uri currently set in the source element. No new tests, this feature is dedicated to internal GStreamer use, such as the upcoming HTTP Live Streaming element. * platform/graphics/gstreamer/WebKitWebSourceGStreamer.cpp: (webKitWebSrcQuery): 2011-02-21 Andoni Morales Alastruey Reviewed by Martin Robinson. [GStreamer] Add 'location' property in webkitwebsrc https://bugs.webkit.org/show_bug.cgi?id=54628 No new tests, this feature is dedicated to internal GStreamer use, such as the upcoming HTTP Live Streaming element. * platform/graphics/gstreamer/WebKitWebSourceGStreamer.cpp: (webkit_web_src_class_init): (webKitWebSrcSetProperty): (webKitWebSrcGetProperty): 2011-02-17 Ryosuke Niwa Reviewed by Kent Tamura. Rename Position::node() to Position::deprecatedNode() https://bugs.webkit.org/show_bug.cgi?id=54622 Done the rename. All calls to node() are replaced by calls to deprecatedNode() except when calls were of the form node()->document() and node()->inDocument() in which case they were replaced by anchorNode()->document() and anchorNode()->inDocument() respectively. * WebCore.exp.in: Added Position::containerNode. The rest abbreviated for simplicity. Please see svn log. 2011-02-20 Gavin Barraclough Build fix (remove includes). * bindings/js/JSDOMBinding.cpp: * bindings/js/JSDOMWindowCustom.cpp: * bindings/js/JSHistoryCustom.cpp: * bindings/js/JSLocationCustom.cpp: 2011-02-20 Gavin Barraclough Reviewed by Oliver Hunt. https://bugs.webkit.org/show_bug.cgi?id=54839 Remove PrototypeFunction, NativeFunctionWrapper, and GlobalEvalFunction. Historically, Native functions used to be represented by PrototypeFunctions, however since introducing call optimizations to the JIT this has used JSFunctions for host calls too. At the point this change was made, the interpreter continued to use PrototypeFunctions, however since fallback from the JIT to interpreter was introduced the interpreter has had to be able to run using host functions represented using JSFunctions - leading to an unnecessary and redundant divergence in behaviour between interpreter only builds, and situations where the JIT has fallen back to interpreting. NativeFunctionWrapper only existed to select between PrototypeFunction and JSFunction for wrappers for host functions, and as such can also be removed. GlobalEvalFunction is a redundant wrapper that happens to be derived from PrototypeFunction. It existed to hold a reference to the global object - but since all functions how derive from JSObjectWithGlobalObject, this no longer requires an additional class to provide this functionality. * bindings/js/JSDOMBinding.cpp: * bindings/js/JSDOMWindowCustom.cpp: * bindings/js/JSHistoryCustom.cpp: * bindings/js/JSLocationCustom.cpp: Removed use of redundant classes. 2011-02-20 Sheriff Bot Unreviewed, rolling out r79169. http://trac.webkit.org/changeset/79169 https://bugs.webkit.org/show_bug.cgi?id=54846 Made unexpected results for tests without CJK characters (Requested by tkent on #webkit). * platform/graphics/chromium/FontCacheChromiumWin.cpp: (WebCore::FontCache::createFontPlatformData): * platform/graphics/chromium/FontPlatformDataChromiumWin.cpp: (WebCore::FontPlatformData::FontPlatformData): (WebCore::FontPlatformData::operator=): * platform/graphics/chromium/FontPlatformDataChromiumWin.h: (WebCore::FontPlatformData::orientation): * platform/graphics/skia/FontCustomPlatformData.cpp: (WebCore::FontCustomPlatformData::fontPlatformData): 2011-02-20 Chun-Lung Huang Reviewed by Kent Tamura. On Chromium Windows, glyphs in vertical text tests are rotated 90 degrees clockwise. https://bugs.webkit.org/show_bug.cgi?id=51450 This platform dependent patch makes Chromium Windows show the vertical writing text correctly. Job was done by adding a prefix '@' in front of the font family name (Windows Only). No new tests added. * platform/graphics/chromium/FontCacheChromiumWin.cpp: (WebCore::FontCache::createFontPlatformData): * platform/graphics/chromium/FontPlatformDataChromiumWin.cpp: (WebCore::FontPlatformData::FontPlatformData): (WebCore::FontPlatformData::operator=): * platform/graphics/chromium/FontPlatformDataChromiumWin.h: (WebCore::FontPlatformData::orientation): * platform/graphics/skia/FontCustomPlatformData.cpp: (WebCore::FontCustomPlatformData::fontPlatformData): 2011-02-20 Simon Fraser Reviewed by Dirk Schulze. REGRESSION (r73369-r73405): transform animation interpolates incorrectly https://bugs.webkit.org/show_bug.cgi?id=54793 After the refactoring in r73380, the ending state for some types of accelerated animations was computed incorrectly, because the TransformationMatrix wasn't reset to identity before the ending value transformations were applied. Fix by cleaning up the code to use explicit, separate values for starting and ending values. Only matrix-type animations had this issue, but cleaned up other clauses similarly. Test: animations/3d/matrix-transform-type-animation.html * platform/graphics/ca/GraphicsLayerCA.cpp: (WebCore::GraphicsLayerCA::setTransformAnimationEndpoints): 2011-02-20 Dan Bernstein Reviewed by Maciej Stachowiak. REGRESSION (r75897): Scaling applied twice to an iframe with a transformed ancestor Test: fast/frames/iframe-scale-applied-twice.html * page/FrameView.cpp: (WebCore::FrameView::create): Set the initial bounds of the view to match the frame size. * platform/ScrollView.cpp: (WebCore::ScrollView::visibleContentRect): Based on bounds, not frame size. (WebCore::ScrollView::updateScrollbars): Ditto. (WebCore::ScrollView::setFrameRect): Moved code that really handles bounds size change to setBoundsSize(). (WebCore::ScrollView::setBoundsSize): Added. (WebCore::ScrollView::setInitialBoundsSize): Added. Sets the bounds size but does not update anything. (WebCore::ScrollView::frameRectsChanged): Based on bounds, not frame size. (WebCore::ScrollView::scrollbarCornerPresent): Ditto. * platform/ScrollView.h: (WebCore::ScrollView::boundsSize): Added this getter. * platform/Widget.h: (WebCore::Widget::resize): Set the bounds size to the frame size. * platform/mac/ScrollbarThemeMac.mm: (WebCore::ScrollbarThemeMac::paint): Fixed an error in the indirect drawing code path where the buffer rect was resized to capture only the damaged part, but was still drawn in the original location. 2011-02-20 Alexey Proskuryakov Reviewed by Eric Seidel. Tighten up access permissions by using libxslt API https://bugs.webkit.org/show_bug.cgi?id=52688 * xml/XSLTProcessorLibxslt.cpp: (WebCore::XSLTProcessor::transformToString): We are only interested in a string result, so let libxslt know about that. 2011-02-20 Sheriff Bot Unreviewed, rolling out r79104. http://trac.webkit.org/changeset/79104 https://bugs.webkit.org/show_bug.cgi?id=54835 Might have caused flaky canaries (Requested by tonyg-cr on #webkit). * html/parser/HTMLDocumentParser.cpp: (WebCore::HTMLDocumentParser::canTakeNextToken): (WebCore::HTMLDocumentParser::pumpTokenizer): * html/parser/HTMLParserScheduler.cpp: (WebCore::isLayoutTimerActive): * html/parser/HTMLParserScheduler.h: (WebCore::HTMLParserScheduler::checkForYieldBeforeToken): * page/FrameView.h: 2011-02-20 Dirk Schulze Reviewed by Nikolas Zimmermann. SVG animation - analyze attribute type for animation https://bugs.webkit.org/show_bug.cgi?id=54542 Analyze animated attribute type to determine the kind of animation. Removed enum PropertyType and replace it with AnimatedAttributeType instead. More cleanup of the animation code. Added missing fillAttributeToPropertyTypeMap() for SVGGElement. Use QualifiedName instead of AtomicStrings for attribute names. This makes it easier to handle animated attributes from a different namespace like xlink:href. The rest of the SVG code is using QualifiedNames as well. No change of functionality on the animation code. The changes for the -element affect animateTransform. Sadly animateTransform is not supported by our SVG animation test script at the moment. Can still be tested manually with the W3C test suite in trunk. The change on SVGUseElement gets covered by animate-elem-39-t.svg Test: svg/animations/animate-dynamic-update-attributeName.html * svg/SVGAnimateElement.cpp: (WebCore::SVGAnimateElement::SVGAnimateElement): (WebCore::SVGAnimateElement::hasValidAttributeType): (WebCore::SVGAnimateElement::determineAnimatedAttributeType): (WebCore::SVGAnimateElement::calculateAnimatedValue): (WebCore::inheritsFromProperty): (WebCore::SVGAnimateElement::calculateFromAndToValues): (WebCore::SVGAnimateElement::calculateFromAndByValues): (WebCore::SVGAnimateElement::resetToBaseValue): (WebCore::SVGAnimateElement::applyResultsToTarget): (WebCore::SVGAnimateElement::calculateDistance): * svg/SVGAnimateElement.h: * svg/SVGAnimateMotionElement.cpp: (WebCore::SVGAnimateMotionElement::hasValidAttributeType): (WebCore::parsePoint): (WebCore::SVGAnimateMotionElement::resetToBaseValue): (WebCore::SVGAnimateMotionElement::calculateAnimatedValue): (WebCore::SVGAnimateMotionElement::calculateDistance): * svg/SVGAnimateMotionElement.h: * svg/SVGAnimateTransformElement.cpp: (WebCore::SVGAnimateTransformElement::hasValidAttributeType): (WebCore::SVGAnimateTransformElement::determineAnimatedAttributeType): (WebCore::SVGAnimateTransformElement::resetToBaseValue): (WebCore::SVGAnimateTransformElement::calculateAnimatedValue): (WebCore::SVGAnimateTransformElement::calculateFromAndByValues): (WebCore::SVGAnimateTransformElement::applyResultsToTarget): (WebCore::SVGAnimateTransformElement::calculateDistance): * svg/SVGAnimateTransformElement.h: * svg/SVGAnimationElement.cpp: (WebCore::parseKeyTimes): (WebCore::parseKeySplines): (WebCore::SVGAnimationElement::isTargetAttributeCSSProperty): (WebCore::SVGAnimationElement::setTargetAttributeAnimatedValue): (WebCore::SVGAnimationElement::calculateKeyTimesForCalcModePaced): (WebCore::solveEpsilon): (WebCore::SVGAnimationElement::calculatePercentFromKeyPoints): (WebCore::SVGAnimationElement::currentValuesFromKeyPoints): (WebCore::SVGAnimationElement::currentValuesForValuesAnimation): (WebCore::SVGAnimationElement::startedActiveInterval): * svg/SVGAnimationElement.h: * svg/SVGFilterElement.cpp: (WebCore::SVGFilterElement::fillAttributeToPropertyTypeMap): * svg/SVGGElement.cpp: (WebCore::SVGGElement::attributeToPropertyTypeMap): (WebCore::SVGGElement::fillAttributeToPropertyTypeMap): Added to fill the animated attribute map for this element. * svg/SVGGElement.h: * svg/SVGUseElement.cpp: (WebCore::SVGUseElement::fillAttributeToPropertyTypeMap): Changed attribute type to AnimatedString for xlink:href. * svg/animation/SMILTimeContainer.cpp: (WebCore::SMILTimeContainer::baseValueFor): (WebCore::SMILTimeContainer::updateAnimations): * svg/animation/SMILTimeContainer.h: * svg/animation/SVGSMILElement.cpp: (WebCore::SVGSMILElement::SVGSMILElement): (WebCore::constructQualifiedName): Create a QualifiedName from a attribute name. (WebCore::SVGSMILElement::insertedIntoDocument): Update local varaible of attribute name. (WebCore::SVGSMILElement::removedFromDocument): Ditto. (WebCore::SVGSMILElement::attributeChanged): Ditto. * svg/animation/SVGSMILElement.h: (WebCore::SVGSMILElement::attributeName): 2011-02-20 David Kilzer BUILD FIX: Add missing UNUSED_PARAM() when !ENABLE(FULLSCREEN_API) Not reviewed. * rendering/RenderLayerCompositor.cpp: (WebCore::RenderLayerCompositor::requiresCompositingForFullScreen): 2011-02-20 Carlos Garcia Campos Reviewed by Martin Robinson. [GTK] Implement PlatformKeyboardEvent::getCurrentModifierState() https://bugs.webkit.org/show_bug.cgi?id=52880 * platform/gtk/KeyEventGtk.cpp: (WebCore::PlatformKeyboardEvent::getCurrentModifierState): 2011-02-19 Andreas Kling Reviewed by Oliver Hunt. DragController should use Color::serialized() when passing colors to CSSStyleDeclaration https://bugs.webkit.org/show_bug.cgi?id=54545 * page/DragController.cpp: (WebCore::DragController::concludeEditDrag): 2011-02-19 Sam Weinig Reviewed by Anders Carlsson. Add phase in addition to momentumPhase to platform wheel events Part of Rename existing phase to momentumPhase. * platform/PlatformWheelEvent.h: (WebCore::PlatformWheelEvent::PlatformWheelEvent): (WebCore::PlatformWheelEvent::momentumPhase): * platform/mac/ScrollAnimatorMac.mm: (WebCore::ScrollAnimatorMac::handleWheelEvent): (WebCore::ScrollAnimatorMac::smoothScrollWithEvent): * platform/mac/WheelEventMac.mm: (WebCore::momentumPhaseForEvent): (WebCore::phaseForEvent): (WebCore::PlatformWheelEvent::PlatformWheelEvent): 2011-02-19 Gavin Barraclough Qt build fix. * bridge/runtime_method.h: (JSC::RuntimeMethod::createStructure): 2011-02-19 Gavin Barraclough Qt build fix. * bridge/qt/qt_instance.cpp: * bridge/qt/qt_pixmapruntime.cpp: 2011-02-19 Gavin Barraclough Qt build fix. * bridge/qt/qt_instance.cpp: (JSC::Bindings::QtInstance::getMethod): * bridge/qt/qt_pixmapruntime.cpp: (JSC::Bindings::QtPixmapInstance::getMethod): 2011-02-18 Gavin Barraclough Reviewed by Sam Weinig. Bug 54786 - Devirtualize JSCell::classInfo() Instead of making a virtual function call, add a pointer to the ClassInfo onto Structure. This removes a virtual function call, and paves the way towards removing all the createStructure methods, and StructureFlags/AnonymousSlotCount properties (these should be able to move onto ClassInfo). Calls to Structure::create must now pass a pointer to the ClassInfo for the structure. All objects now have a ClassInfo pointer, non-object cell types still do not. Changes are most mechanical, involving three steps: * Remove virtual classInfo() methods. * Add &s_info parameter to calls to Structure::create. * Rename ClassInfo static members on classes from 'info' to 's_info', for consistency. * WebCore.exp.in: * bindings/js/JSAudioConstructor.cpp: * bindings/js/JSAudioConstructor.h: * bindings/js/JSDOMBinding.cpp: * bindings/js/JSDOMBinding.h: * bindings/js/JSDOMGlobalObject.cpp: * bindings/js/JSDOMGlobalObject.h: * bindings/js/JSDOMWindowBase.cpp: * bindings/js/JSDOMWindowBase.h: * bindings/js/JSDOMWindowShell.cpp: * bindings/js/JSDOMWindowShell.h: * bindings/js/JSGeolocationCustom.cpp: * bindings/js/JSImageConstructor.cpp: * bindings/js/JSImageConstructor.h: * bindings/js/JSImageDataCustom.cpp: * bindings/js/JSOptionConstructor.cpp: * bindings/js/JSOptionConstructor.h: * bindings/js/JSWorkerContextBase.cpp: * bindings/js/JSWorkerContextBase.h: * bindings/js/SerializedScriptValue.cpp: * bindings/scripts/CodeGeneratorJS.pm: * bridge/c/CRuntimeObject.cpp: * bridge/c/CRuntimeObject.h: * bridge/c/c_instance.cpp: * bridge/jni/jsc/JNIUtilityPrivate.cpp: * bridge/jni/jsc/JavaInstanceJSC.cpp: * bridge/jni/jsc/JavaRuntimeObject.cpp: * bridge/jni/jsc/JavaRuntimeObject.h: * bridge/jsc/BridgeJSC.cpp: * bridge/objc/ObjCRuntimeObject.h: * bridge/objc/ObjCRuntimeObject.mm: * bridge/objc/objc_instance.mm: * bridge/objc/objc_runtime.h: * bridge/objc/objc_runtime.mm: * bridge/runtime_array.cpp: * bridge/runtime_array.h: * bridge/runtime_method.cpp: * bridge/runtime_method.h: * bridge/runtime_object.cpp: * bridge/runtime_object.h: 2011-02-19 Zan Dobersek Reviewed by Martin Robinson. [cairo][canvas] Drawing from/into float rectangles with width or height in range 0 to 1 fails https://bugs.webkit.org/show_bug.cgi?id=54491 When width or height in float rectangle are in range (0, 0.5) or (-0.5, 0) and would round to 0, alter the behaviour to ensure that width or height are at least 1 pixel in size in these cases. * platform/graphics/cairo/GraphicsContextCairo.cpp: (WebCore::GraphicsContext::roundToDevicePixels): 2011-02-19 Dan Bernstein LLVM Compiler build fix. * platform/graphics/ShadowBlur.cpp: (WebCore::ShadowBlur::blurLayerImage): Eliminated a file static that generated a global initializer since the compiler does not evaluate sqrtf at compile time. 2011-02-19 Brian Ryner Reviewed by Adam Barth. Replace the #include of DocumentLoader.h in Document.h with a forward declaration, and add a missing #include now that this transitive include is gone. https://bugs.webkit.org/show_bug.cgi?id=50489 No new tests required. * bindings/ScriptControllerBase.cpp: * dom/Document.h: 2011-02-19 Patrick Gansterer Reviewed by Antonio Gomes. [EFL] Remove WebCore::currentTime() https://bugs.webkit.org/show_bug.cgi?id=53886 There is no declaration for it and we use WTF::currentTime() everywhere. * platform/efl/SystemTimeEfl.cpp: 2011-02-19 Patrick Gansterer Reviewed by Andreas Kling. Move KeygenWinCE from wince into win directory https://bugs.webkit.org/show_bug.cgi?id=54804 Move this file into the win directory, since it can be used on all windows platforms. * CMakeListsWinCE.txt: * platform/win/SSLKeyGeneratorWin.cpp: Renamed from platform/wince/KeygenWinCE.cpp. 2011-02-19 Patrick Gansterer Reviewed by Andreas Kling. Make FileChooserWin.cpp more portable and use it on WinCE https://bugs.webkit.org/show_bug.cgi?id=54803 * CMakeListsWinCE.txt: * platform/win/FileChooserWin.cpp: (WebCore::FileChooser::basenameForWidth): * platform/wince/FileChooserWinCE.cpp: Removed. 2011-02-19 James Simonsen Reviewed by Adam Barth. Make ScriptElement match the HTML5 spec https://bugs.webkit.org/show_bug.cgi?id=54676 This implements the "prepare a script" section of the HTML5 spec in ScriptElement::prepareScript(). http://www.whatwg.org/specs/web-apps/current-work/multipage/scripting-1.html#prepare-a-script There are a couple of things missing from the spec that would be new functionality. These will be added later. - Support for async=false - Empty src attribute should dispatch an error. There are a couple of slight behavioral changes to match the spec. - An XHTML script that is loaded then copied will not fire load on the copy. - If onbeforeload moves the script to a different document, then it will not try to execute again because wasAlreadyStarted is true. The parsers were updated to use the new API, but not rewritten to look like the spec. That will be done separately. Test: All existing tests. * dom/ScriptElement.cpp: Rewritten to match HTML5 spec. (WebCore::ScriptElement::ScriptElement): (WebCore::ScriptElement::insertedIntoDocument): Logic moved to prepareScript. (WebCore::ScriptElement::childrenChanged): Logic moved to prepareScript. (WebCore::ScriptElement::handleSourceAttribute): Logic moved to prepareScript. (WebCore::isLegacySupportedJavaScriptLanguage): Added to support old script types in layout tests. (WebCore::ScriptElement::isScriptTypeSupported): Derived from old shouldExecuteAsJavaScript(). (WebCore::ScriptElement::prepareScript): START HERE. Main change. Should read exactly like HTML5's "prepare a script." Legacy type support needed for layout tests using XML parser. (WebCore::ScriptElement::requestScript): Most logic moved to prepareScript. Check security settings here. (WebCore::ScriptElement::executeScript): Combined evaluateScript() and executeScript() from old code. Logic moved to prepareScript. (WebCore::ScriptElement::stopLoadRequest): Ignore parser executed scripts. (WebCore::ScriptElement::execute): Renamed executeScript. (WebCore::ScriptElement::notifyFinished): We should only listen for non-parser executed scripts. (WebCore::ScriptElement::ignoresLoadRequest): New variable names. (WebCore::ScriptElement::childrenAreCommentsOrEmptyText): Added for HTML5 compliance. (WebCore::ScriptElement::scriptCharset): Use HTML5 variables. * dom/ScriptElement.h: (WebCore::ScriptElement::willBeParserExecuted): Added. (WebCore::ScriptElement::readyToBeParserExecuted): Added. (WebCore::ScriptElement::willExecuteWhenDocumentFinishedParsing): Added. (WebCore::ScriptElement::cachedScript): prepareScript() is the only place that should load scripts. This accessor lets the parsers listen for when loads finish. (WebCore::ScriptElement::isParserInserted): Added. * dom/XMLDocumentParserLibxml2.cpp: (WebCore::XMLDocumentParser::endElementNs): Should behave the same. Offloads much of its work to prepareScript(). * dom/XMLDocumentParserQt.cpp: (WebCore::XMLDocumentParser::parseEndElement): Identical to libxml2 changes. * html/HTMLScriptElement.cpp: (WebCore::HTMLScriptElement::insertedIntoDocument): No longer needs url. (WebCore::HTMLScriptElement::hasSourceAttribute): Added. * html/HTMLScriptElement.h: * html/parser/HTMLScriptRunner.cpp: (WebCore::HTMLScriptRunner::requestPendingScript): Requesting scripts offloaded to ScriptElement. (WebCore::HTMLScriptRunner::runScript): Should behave the same. Offloads much of its work to prepareScript(). * svg/SVGScriptElement.cpp: (WebCore::SVGScriptElement::svgAttributeChanged): New ScriptElement function names. (WebCore::SVGScriptElement::insertedIntoDocument): No longer needs url. (WebCore::SVGScriptElement::finishParsingChildren): ScriptElement::finishParsingChildren is gone. (WebCore::SVGScriptElement::hasSourceAttribute): Added. (WebCore::SVGScriptElement::dispatchLoadEvent): New ScriptElement function names. * svg/SVGScriptElement.h: 2011-02-19 Marc-Antoine Ruel Reviewed by James Robinson. Split webcore_rendering off webcore_remaining to reduce its size for WPO builds https://bugs.webkit.org/show_bug.cgi?id=54789 * WebCore.gyp/WebCore.gyp: 2011-02-19 Bill Budge Reviewed by David Levin. ThreadableLoaderClient needs willSendRequest method https://bugs.webkit.org/show_bug.cgi?id=54688 No new tests. Exposes no new functionality * WebCore.gypi: * loader/DocumentThreadableLoader.cpp: (WebCore::DocumentThreadableLoader::willSendRequest): (WebCore::DocumentThreadableLoader::didReceiveData): (WebCore::DocumentThreadableLoader::didReceiveCachedMetadata): * loader/DocumentThreadableLoaderClient.h: Added. (WebCore::DocumentThreadableLoaderClient::isDocumentThreadableLoaderClient): (WebCore::DocumentThreadableLoaderClient::willSendRequest): * loader/ThreadableLoaderClient.h: (WebCore::ThreadableLoaderClient::isDocumentThreadableLoaderClient): 2011-02-19 Charlie Reis Reviewed by Mihai Parparita. Ensure loading has stopped in HistoryController::goToItem https://bugs.webkit.org/show_bug.cgi?id=54517 Avoid stopping all loaders in goToItem for same document navigations or pseudo-back-forward URLs. Make HistoryController::goToItem private to force callers to go through Page::goToItem. Also add a callback to FrameLoaderClient to let clients decide whether to stop loading first. Test: http/tests/navigation/forward-to-fragment-fires-onload.html * loader/EmptyClients.h: * loader/FrameLoader.h: * loader/FrameLoaderClient.h: * loader/HistoryController.cpp: * loader/HistoryController.h: * page/Page.cpp: 2011-02-19 Adam Barth Reviewed by Daniel Bates. Fix xssAuditor/iframe-injection.html https://bugs.webkit.org/show_bug.cgi?id=54591 We should block the iframe src attribute. Although this technically can't be used to run script, it's a pretty easy vector for stealing passwords. * html/parser/XSSFilter.cpp: (WebCore::XSSFilter::filterTokenInitial): (WebCore::XSSFilter::filterIframeToken): * html/parser/XSSFilter.h: 2011-02-18 Tony Gentilcore Reviewed by Eric Seidel. Let the parser yield for layout before running scripts https://bugs.webkit.org/show_bug.cgi?id=54355 Prior to this patch, the parser would yield to perform a layout/paint before running a script only if the script or a stylesheet blocking the script is not loaded yet. Since we don't preload scan into the body while parsing the head, typically we'll block on a script early in the body that causes us to yield to do the first paint within a reasonable time. However, I'm planning to change the PreloadScanner to scan into the body from the head. That significantly improves overall load time, but would hurt first paint time because fewer scripts would be blocked during parsing and thus wouldn't yield. This change causes us to yield before running scripts if we haven't painted yet (regardless of whether or not the script is loaded). In addition to allowing the above mentioned PreloadScanner change to be implemented without regressing first paint time, this also improves first paint time by itself. I tested Alexa's top 45 websites using Web Page Replay to control the content and simulate bandwidth. This patch improved average first paint time by 1% over an unlimited connection, 6% over a 1Mbps connection and 11% over a 5Mbps connection. There was no statistically signifcant change in page load time. Within the pages tested, 33 had no statistically significant change in time to first paint, 12 improved, and none regressed. Of the improved, some of the standouts from the 1Mbps set are: 20% on youtube, 37% on wiki, 27% on ebay, 13% on cnn, 16% on espn, 74% on sohu. * html/parser/HTMLDocumentParser.cpp: (WebCore::HTMLDocumentParser::canTakeNextToken): This is the new yield point. (WebCore::HTMLDocumentParser::pumpTokenizer): Remove ASSERT that we are not paused. isPaused means that we are waiting for a script. Bug 54574 changed pumpTokenizer() so that it does the right thing whether we are just before a token or waiting for a script. Now that we may yield before a token or before a script, this may be called while paused. * html/parser/HTMLParserScheduler.cpp: (WebCore::isLayoutTimerActive): Added a FIXME because r52919 changed minimumLayoutDelay() to return m_extraLayoutDelay instead of 0 as a minimum. So checking !minimumLayoutDelay() no longer works. The fix is to change it to check minimumLayoutDelay() == m_extraLayoutDelay. But this is all the more reason to move this method onto Document. I'll do this in a follow up. (WebCore::HTMLParserScheduler::checkForYieldBeforeScript): Added. * page/FrameView.h: (WebCore::FrameView::hasEverPainted): Added. 2011-02-18 Dawit Alemayehu Reviewed by Andreas Kling. [Qt] Button Element is rendered w/ text off-center. https://bugs.webkit.org/show_bug.cgi?id=53373 Test: LayoutTests/fast/forms/button-white-space.html * platform/qt/RenderThemeQt.cpp: (WebCore::RenderThemeQt::adjustButtonStyle): 2011-02-18 Jonathan Backer Reviewed by Eric Seidel. [chromium] Fix leak of texture IDs in compositor. https://bugs.webkit.org/show_bug.cgi?id=54750 No new tests. It is extremely unlikely that this leak would have any user visible impact because only a few bytes of space are wasted (we're leaking texture IDs, not actual textures) and the space of texture IDs is large (32 bits). * platform/graphics/chromium/TextureManager.cpp: (WebCore::TextureManager::requestTexture): 2011-02-18 Yi Shen Reviewed by Tor Arne Vestbø. Always display the media controls when requiresFullscreenForVideoPlayback() is true https://bugs.webkit.org/show_bug.cgi?id=54308 For video element, it should have controls when Chrome::requiresFullscreenForVideoPlayback() is true. * html/HTMLMediaElement.cpp: (WebCore::HTMLMediaElement::controls): 2011-02-18 Adrienne Walker Reviewed by Kenneth Russell. [chromium] Use nearest-neighbor filtering for root layer. https://bugs.webkit.org/show_bug.cgi?id=54409 https://bugs.webkit.org/show_bug.cgi?id=54509 This setting creates more consistent images for LayoutTests and prevents small floating point errors in texture coordinates from creating off-by-one pixel color differences. * platform/graphics/chromium/LayerTilerChromium.cpp: (WebCore::LayerTilerChromium::update): 2011-02-18 Gyuyoung Kim Reviewed by Kent Tamura. [EFL] Fix coding style errors in RenderThemeEfl.h https://bugs.webkit.org/show_bug.cgi?id=54693 Fix style errors in RenderThemeEfl.h * platform/efl/RenderThemeEfl.h: 2011-02-18 Ademar de Souza Reis Jr Reviewed by Andreas Kling. [Qt] The localized vendor name for Qt SIS packages should be "Nokia" https://bugs.webkit.org/show_bug.cgi?id=54742 This change was applied in the Qt repository (qt/src/3rdparty/webkit), so we should do the same here in QtWebKit. Patch by Eckhart Koppen a8a84f1667966acfa093c4be0b7d4b0900ddd3d9: The previously used name "Nokia, Qt" was not usable for Nokia Content Signing, which only allows "Nokia" as the visible vendor name. The unique vendor ID remains as "Nokia, Qt" * WebCore.pro: 2011-02-18 Alexis Menard Reviewed by Andreas Kling. [Qt] REGRESSION(r67516) : on www.gmail.com a strange rendering issue appears on the menu bar due to flash. https://bugs.webkit.org/show_bug.cgi?id=54741 Only show plugins with a valid size. We then don't involve X11 if there is nothing to see anyway. * plugins/qt/PluginViewQt.cpp: (WebCore::PluginView::updatePluginWidget): (WebCore::PluginView::platformStart): 2011-02-18 Ryuan Choi Reviewed by Kent Tamura. [EFL] Remove GDK dependency. https://bugs.webkit.org/show_bug.cgi?id=53978 Remove GLIB_SUPPORT macro in GDK related code. Regardless of GLIB_SUPPORT, getDefaultFontOptions() will works same way. * CMakeListsEfl.txt: * platform/graphics/freetype/FontPlatformDataFreeType.cpp: (WebCore::getDefaultFontOptions): 2011-02-18 Noel Gordon Reviewed by James Robinson. [Chromium] Add elliptical gradient support to GradientSkia https://bugs.webkit.org/show_bug.cgi?id=51841 Covered by existing tests, these need new rebaselines once this patch lands for chrome linux, win32 fast/gradients/css3-color-stop-units.html fast/gradients/css3-color-stops.html fast/gradients/css3-linear-angle-gradients.html fast/gradients/css3-radial-gradients.html fast/gradients/css3-radial-gradients2.html fast/gradients/css3-radial-gradients3.html fast/gradients/css3-repeating-radial-gradients.html * platform/graphics/skia/GradientSkia.cpp: (WebCore::Gradient::platformGradient): 2011-02-18 James Robinson Fix typo in boundary test in ASSERT() - test is for an inclusive range, not exclusive. * platform/graphics/chromium/ContentLayerChromium.cpp: (WebCore::ContentLayerChromium::updateTextureIfNeeded): 2011-02-18 Viatcheslav Ostapenko Reviewed by Kenneth Rohde Christiansen. Tiled backing store area is too big. Error in area calculcation causes size of backing store up to 6 times bigger than viewport with default multipliers. https://bugs.webkit.org/show_bug.cgi?id=54587 * platform/graphics/TiledBackingStore.cpp: (WebCore::TiledBackingStore::createTiles): 2011-02-18 Beth Dakin Reviewed by Sam Weinig. Fix for Horizontal scroller doesn't appear when loading a page with a Horizontal scrollbar from the back/forward cache. This patch adds a new bool member variable to FrameView to keep track of whether we are loading a page from the back/ forward cache. If we are, don't suppress scrollbars on first layout. * history/CachedFrame.cpp: (WebCore::CachedFrameBase::restore): * page/FrameView.cpp: (WebCore::FrameView::FrameView): (WebCore::FrameView::reset): (WebCore::FrameView::layout): * page/FrameView.h: (WebCore::FrameView::setIsRestoringFromBackForward): (WebCore::FrameView::isRestoringFromBackForward): 2011-02-18 Patrick Gansterer Unreviewed WinCE build fix for r78846. * platform/graphics/wince/FontWinCE.cpp: (WebCore::TextRunComponent::TextRunComponent): * platform/graphics/wince/GraphicsContextWinCE.cpp: (WebCore::GraphicsContext::drawLineForText): (WebCore::GraphicsContext::drawLineForTextChecking): (WebCore::GraphicsContext::drawText): * platform/wince/FileChooserWinCE.cpp: (WebCore::FileChooser::basenameForWidth): 2011-02-18 Emil A Eklund Reviewed by Darin Adler. Crash in EventHandler::sendContextMenuEventForKey https://bugs.webkit.org/show_bug.cgi?id=54495 Test: fast/events/menu-keydown-on-hidden-element.html * page/EventHandler.cpp: (WebCore::EventHandler::sendContextMenuEventForKey): Add null check. 2011-02-15 Adrienne Walker Reviewed by James Robinson. [chromium] Clean up shader code from LayerChromium classes https://bugs.webkit.org/show_bug.cgi?id=54484 This is a refactoring and there should be no change in functionality. All shader code is pulled out into classes in ShaderChromium. The SharedValues classes are now turned into ProgramBinding, one per shader program. These contain shader classes that know about what variables they can bind. * WebCore.gypi: * platform/graphics/chromium/CanvasLayerChromium.cpp: (WebCore::CanvasLayerChromium::draw): * platform/graphics/chromium/CanvasLayerChromium.h: * platform/graphics/chromium/ContentLayerChromium.cpp: (WebCore::ContentLayerChromium::draw): * platform/graphics/chromium/ContentLayerChromium.h: * platform/graphics/chromium/GeometryBinding.cpp: Added. (WebCore::GeometryBinding::GeometryBinding): (WebCore::GeometryBinding::~GeometryBinding): (WebCore::GeometryBinding::prepareForDraw): * platform/graphics/chromium/GeometryBinding.h: Copied from Source/WebCore/platform/graphics/chromium/PluginLayerChromium.h. (WebCore::GeometryBinding::initialized): (WebCore::GeometryBinding::context): (WebCore::GeometryBinding::quadVerticesVbo): (WebCore::GeometryBinding::quadElementsVbo): (WebCore::GeometryBinding::positionAttribLocation): (WebCore::GeometryBinding::texCoordAttribLocation): * platform/graphics/chromium/LayerChromium.cpp: (WebCore::LayerChromium::drawDebugBorder): * platform/graphics/chromium/LayerChromium.h: * platform/graphics/chromium/LayerRendererChromium.cpp: (WebCore::LayerRendererChromium::drawLayers): (WebCore::LayerRendererChromium::initializeSharedObjects): (WebCore::LayerRendererChromium::cleanupSharedObjects): * platform/graphics/chromium/LayerRendererChromium.h: (WebCore::LayerRendererChromium::sharedGeometry): (WebCore::LayerRendererChromium::borderProgram): (WebCore::LayerRendererChromium::contentLayerProgram): (WebCore::LayerRendererChromium::canvasLayerProgram): (WebCore::LayerRendererChromium::videoLayerRGBAProgram): (WebCore::LayerRendererChromium::videoLayerYUVProgram): (WebCore::LayerRendererChromium::pluginLayerProgram): (WebCore::LayerRendererChromium::renderSurfaceProgram): (WebCore::LayerRendererChromium::renderSurfaceMaskProgram): (WebCore::LayerRendererChromium::tilerProgram): * platform/graphics/chromium/LayerTilerChromium.cpp: (WebCore::LayerTilerChromium::draw): (WebCore::LayerTilerChromium::drawTexturedQuad): * platform/graphics/chromium/LayerTilerChromium.h: * platform/graphics/chromium/PluginLayerChromium.cpp: (WebCore::PluginLayerChromium::draw): * platform/graphics/chromium/PluginLayerChromium.h: * platform/graphics/chromium/ProgramBinding.cpp: Added. (WebCore::ProgramBindingBase::ProgramBindingBase): (WebCore::ProgramBindingBase::~ProgramBindingBase): (WebCore::ProgramBindingBase::init): (WebCore::ProgramBindingBase::loadShader): (WebCore::ProgramBindingBase::createShaderProgram): * platform/graphics/chromium/ProgramBinding.h: Added. (WebCore::ProgramBindingBase::program): (WebCore::ProgramBindingBase::initialized): (WebCore::ProgramBinding::ProgramBinding): (WebCore::ProgramBinding::vertexShader): (WebCore::ProgramBinding::fragmentShader): * platform/graphics/chromium/RenderSurfaceChromium.cpp: (WebCore::RenderSurfaceChromium::drawSurface): * platform/graphics/chromium/RenderSurfaceChromium.h: * platform/graphics/chromium/ShaderChromium.cpp: Added. (WebCore::VertexShaderPosTex::VertexShaderPosTex): (WebCore::VertexShaderPosTex::init): (WebCore::VertexShaderPosTex::getShaderString): (WebCore::VertexShaderPosTexYUVStretch::VertexShaderPosTexYUVStretch): (WebCore::VertexShaderPosTexYUVStretch::init): (WebCore::VertexShaderPosTexYUVStretch::getShaderString): (WebCore::VertexShaderPos::VertexShaderPos): (WebCore::VertexShaderPos::init): (WebCore::VertexShaderPos::getShaderString): (WebCore::VertexShaderPosTexTransform::VertexShaderPosTexTransform): (WebCore::VertexShaderPosTexTransform::init): (WebCore::VertexShaderPosTexTransform::getShaderString): (WebCore::FragmentTexAlphaBinding::FragmentTexAlphaBinding): (WebCore::FragmentTexAlphaBinding::init): (WebCore::FragmentShaderRGBATexFlipAlpha::getShaderString): (WebCore::FragmentShaderRGBATexAlpha::getShaderString): (WebCore::FragmentShaderBGRATexAlpha::getShaderString): (WebCore::FragmentShaderRGBATexAlphaMask::FragmentShaderRGBATexAlphaMask): (WebCore::FragmentShaderRGBATexAlphaMask::init): (WebCore::FragmentShaderRGBATexAlphaMask::getShaderString): (WebCore::FragmentShaderYUVVideo::FragmentShaderYUVVideo): (WebCore::FragmentShaderYUVVideo::init): (WebCore::FragmentShaderYUVVideo::getShaderString): (WebCore::FragmentShaderColor::FragmentShaderColor): (WebCore::FragmentShaderColor::init): (WebCore::FragmentShaderColor::getShaderString): * platform/graphics/chromium/ShaderChromium.h: Added. (WebCore::VertexShaderPosTex::matrixLocation): (WebCore::VertexShaderPosTexYUVStretch::matrixLocation): (WebCore::VertexShaderPosTexYUVStretch::yWidthScaleFactorLocation): (WebCore::VertexShaderPosTexYUVStretch::uvWidthScaleFactorLocation): (WebCore::VertexShaderPos::matrixLocation): (WebCore::VertexShaderPosTexTransform::matrixLocation): (WebCore::VertexShaderPosTexTransform::texTransformLocation): (WebCore::FragmentTexAlphaBinding::alphaLocation): (WebCore::FragmentTexAlphaBinding::samplerLocation): (WebCore::FragmentShaderRGBATexAlphaMask::alphaLocation): (WebCore::FragmentShaderRGBATexAlphaMask::samplerLocation): (WebCore::FragmentShaderRGBATexAlphaMask::maskSamplerLocation): (WebCore::FragmentShaderYUVVideo::yTextureLocation): (WebCore::FragmentShaderYUVVideo::uTextureLocation): (WebCore::FragmentShaderYUVVideo::vTextureLocation): (WebCore::FragmentShaderYUVVideo::alphaLocation): (WebCore::FragmentShaderYUVVideo::ccMatrixLocation): (WebCore::FragmentShaderYUVVideo::signAdjLocation): (WebCore::FragmentShaderColor::colorLocation): * platform/graphics/chromium/VideoLayerChromium.cpp: (WebCore::VideoLayerChromium::draw): (WebCore::VideoLayerChromium::drawYUV): (WebCore::VideoLayerChromium::drawRGBA): * platform/graphics/chromium/VideoLayerChromium.h: 2011-02-18 James Robinson Reviewed by Kenneth Russell. [chromium] Update texture for ContentLayerChromiums in draw() call instead of updateContents..() call https://bugs.webkit.org/show_bug.cgi?id=54315 This defers all operations on the compositor's GL context until the draw() call which is a prerequisite for moving the draw() off-thread. Also cleans up the update cycle a bit - there were some unused local variables and whatnot. One consequence of this change is that the upload buffer is retained across updates now instead of allocated by each paint. This is necessary so that the full layer contents can be uploaded if the texture manager evicts the layer's backing texture. This costs more persistent memory but avoids lots of allocator churn on updates. Another nonobvious detail is that I have to update the texture for ContentLayerChromiums in bindContentsTexture() because mask layers never draw(), they are instead bound to the secondary texture unit. * platform/graphics/chromium/ContentLayerChromium.cpp: (WebCore::ContentLayerChromium::requiresClippedUpdateRect): (WebCore::ContentLayerChromium::updateContentsIfDirty): (WebCore::ContentLayerChromium::resizeUploadBufferForImage): (WebCore::ContentLayerChromium::resizeUploadBuffer): (WebCore::SkBitmapConditionalAutoLockerPixels::SkBitmapConditionalAutoLockerPixels): (WebCore::SkBitmapConditionalAutoLockerPixels::~SkBitmapConditionalAutoLockerPixels): (WebCore::SkBitmapConditionalAutoLockerPixels::lockPixels): (WebCore::ContentLayerChromium::updateTextureIfNeeded): (WebCore::ContentLayerChromium::draw): (WebCore::ContentLayerChromium::unreserveContentsTexture): (WebCore::ContentLayerChromium::bindContentsTexture): * platform/graphics/chromium/ContentLayerChromium.h: * platform/graphics/chromium/ImageLayerChromium.cpp: (WebCore::ImageLayerChromium::updateContentsIfDirty): * platform/graphics/chromium/LayerChromium.cpp: (WebCore::LayerChromium::setBounds): * platform/graphics/chromium/LayerChromium.h: 2011-02-18 Kenneth Russell Unreviewed, Chromium build fix on certain Linux platforms. * platform/graphics/gpu/LoopBlinnSolidFillShader.cpp: 2011-02-18 Mahesh Kulkarni Reviewed by Kenneth Rohde Christiansen. [Qt] Implement client based geolocation for qtport https://bugs.webkit.org/show_bug.cgi?id=42629 Implements client based geolocation for qtwebkit. Removed old code related to non-client based geolocation No tests as yet. This will be raised as different bug as new mock client implementation need to be done. * WebCore.pro: * features.pri: * platform/qt/GeolocationServiceQt.cpp: Removed. * platform/qt/GeolocationServiceQt.h: Removed. 2011-02-18 Yael Aharon Reviewed by Dave Hyatt. Add support for dir=auto https://bugs.webkit.org/show_bug.cgi?id=50916 When an element has dir attribute with value "auto", call defaultWritingMode to find its directionality. Added a flag SelfOrAncestorHasDirAutoFlag, and added hooks in the DOM to set and check this flag. This flag is set on every node between an element with dir=auto attribute and its first text node. Changes in the DOM between those elements will trigger re-evaluating the directionality, but changes not between those element do not need to be concerned. The DOM hooks were added to childrenChanged, and to parseMappedAttribute. The directionality is evaluated when children are added, and cleared when they are removed. Directionality flag is also cleared on a child that is no longer determining the directionality due to a sibling being added before that child. Added 2 static CSSMutableStyleDeclarations to be used for elements with dir=auto. We cannot used the mapped declaration, because it can take only one value. Tests: fast/dom/HTMLElement/attr-dir-auto-change-before-text-node.html fast/dom/HTMLElement/attr-dir-auto-change-child-node.html fast/dom/HTMLElement/attr-dir-auto-change-text.html fast/dom/HTMLElement/attr-dir-auto-children.html fast/dom/HTMLElement/attr-dir-auto-remove-add-children.html fast/dom/HTMLElement/attr-dir-auto.html fast/dom/HTMLElement/attr-dir-value-change.html * css/CSSStyleSelector.cpp: (WebCore::leftToRightDeclaration): (WebCore::rightToLeftDeclaration): (WebCore::CSSStyleSelector::canShareStyleWithElement): (WebCore::CSSStyleSelector::styleForElement): * dom/Node.h: (WebCore::Node::selfOrAncestorHasDirAutoAttribute): (WebCore::Node::setSelfOrAncestorHasDirAutoAttribute): * html/HTMLElement.cpp: (WebCore::HTMLElement::mapToEntry): (WebCore::HTMLElement::parseMappedAttribute): (WebCore::setHasDirAutoFlagRecursively): (WebCore::HTMLElement::childrenChanged): (WebCore::HTMLElement::directionalityIfhasDirAutoAttribute): (WebCore::HTMLElement::directionality): (WebCore::HTMLElement::dirAttributeChanged): (WebCore::HTMLElement::adjustDirectionalityIfNeededAfterChildAttributeChanged): (WebCore::HTMLElement::calculateAndAdjustDirectionality): (WebCore::HTMLElement::adjustDirectionalityIfNeededAfterChildrenChanged): * html/HTMLElement.h: 2011-02-18 Yael Aharon Reviewed by Antonio Gomes. Navigating downwards / upwards does not focus on the links spread across more than one line. https://bugs.webkit.org/show_bug.cgi?id=54639 When 2 anchor elements span more than one line each, and one ends on the same line that the second starts on, the rects reported by their renderers are overlapping. When handling 2 overlapping nodes, check for this case, and don't assume that one of the nodes is on a higher layer. Test: fast/spatial-navigation/snav-two-elements-one-line.html * page/FocusController.cpp: (WebCore::updateFocusCandidateIfNeeded): (WebCore::FocusController::findFocusCandidateInContainer): * page/SpatialNavigation.cpp: (WebCore::areElementsOnSameLine): (WebCore::distanceDataForNode): * page/SpatialNavigation.h: 2011-02-18 Ben Vanik Reviewed by Kenneth Russell. Bug 53940: Implement the OES_vertex_array_object WebGL extension https://bugs.webkit.org/show_bug.cgi?id=53940 Initial implementation of the OES_vertex_array_object extension adding the OESVertexArrayObject extension container and WebGLVertexArrayObjectOES VAO object. The extension is plumbed through the Extensions3D interface and implemented in the Extensions3DOpenGL (WebKit/OSX) version when it is available. Two big changes touching code outside of the extension files: * Moved the typedefs at the top of GraphicsContext3D.h to GraphicsTypes3D.h (modeled after GraphicsTypes.h). They are not namespaced as they weren't before. * To make the code cleaner/clearer all vertex attribute state has been moved to the WebGLVertexArrayObjectOES type (struct VertexAttribState) except for values which are still on the WebGLRenderingContext. A default VAO is now used to store the existing attribute states for when no other VAO is used. Code in WebGLRenderingContext dealing with buffers and vertex attributes now defers to or stores values in the bound array object. Tested against the WebGL conformance suite and the new oes-vertex-array-object test: https://cvs.khronos.org/svn/repos/registry/trunk/public/webgl/sdk/tests/conformance/oes-vertex-array-object.html * CMakeLists.txt: * CodeGenerators.pri: * DerivedSources.make: * GNUmakefile.am: * WebCore.gyp: Modified property svn:ignore. * WebCore.gypi: * WebCore.pro: * WebCore.xcodeproj/project.pbxproj: * bindings/js/JSWebGLRenderingContextCustom.cpp: (WebCore::toJS): * bindings/v8/custom/V8WebGLRenderingContextCustom.cpp: (WebCore::toV8Object): * html/canvas/OESVertexArrayObject.cpp: Added. (WebCore::OESVertexArrayObject::OESVertexArrayObject): (WebCore::OESVertexArrayObject::~OESVertexArrayObject): (WebCore::OESVertexArrayObject::getName): (WebCore::OESVertexArrayObject::create): (WebCore::OESVertexArrayObject::createVertexArrayOES): (WebCore::OESVertexArrayObject::deleteVertexArrayOES): (WebCore::OESVertexArrayObject::isVertexArrayOES): (WebCore::OESVertexArrayObject::bindVertexArrayOES): * html/canvas/OESVertexArrayObject.h: Added. * html/canvas/OESVertexArrayObject.idl: Added. * html/canvas/WebGLExtension.h: * html/canvas/WebGLGetInfo.cpp: (WebCore::WebGLGetInfo::WebGLGetInfo): (WebCore::WebGLGetInfo::getWebGLVertexArrayObjectOES): * html/canvas/WebGLGetInfo.h: * html/canvas/WebGLRenderingContext.cpp: (WebCore::WebGLRenderingContext::initializeNewContext): (WebCore::WebGLRenderingContext::bindBuffer): (WebCore::WebGLRenderingContext::deleteBuffer): (WebCore::WebGLRenderingContext::disableVertexAttribArray): (WebCore::WebGLRenderingContext::validateElementArraySize): (WebCore::WebGLRenderingContext::validateIndexArrayConservative): (WebCore::WebGLRenderingContext::validateIndexArrayPrecise): (WebCore::WebGLRenderingContext::validateRenderingState): (WebCore::WebGLRenderingContext::drawElements): (WebCore::WebGLRenderingContext::enableVertexAttribArray): (WebCore::WebGLRenderingContext::getExtension): (WebCore::WebGLRenderingContext::getParameter): (WebCore::WebGLRenderingContext::getSupportedExtensions): (WebCore::WebGLRenderingContext::getVertexAttrib): (WebCore::WebGLRenderingContext::vertexAttribPointer): (WebCore::WebGLRenderingContext::validateBufferDataParameters): (WebCore::WebGLRenderingContext::vertexAttribfImpl): (WebCore::WebGLRenderingContext::vertexAttribfvImpl): (WebCore::WebGLRenderingContext::initVertexAttrib0): (WebCore::WebGLRenderingContext::simulateVertexAttrib0): (WebCore::WebGLRenderingContext::restoreStatesAfterVertexAttrib0Simulation): (WebCore::WebGLRenderingContext::getNumberOfExtensions): (WebCore::WebGLRenderingContext::getExtensionNumber): * html/canvas/WebGLRenderingContext.h: (WebCore::WebGLRenderingContext::getMaxVertexAttribs): (WebCore::WebGLRenderingContext::setBoundVertexArrayObject): (WebCore::WebGLRenderingContext::VertexAttribValue::VertexAttribValue): * html/canvas/WebGLVertexArrayObjectOES.cpp: Added. (WebCore::WebGLVertexArrayObjectOES::create): (WebCore::WebGLVertexArrayObjectOES::WebGLVertexArrayObjectOES): (WebCore::WebGLVertexArrayObjectOES::deleteObjectImpl): * html/canvas/WebGLVertexArrayObjectOES.h: Added. (WebCore::WebGLVertexArrayObjectOES::~WebGLVertexArrayObjectOES): (WebCore::WebGLVertexArrayObjectOES::VertexAttribState::VertexAttribState): (WebCore::WebGLVertexArrayObjectOES::isDefaultObject): (WebCore::WebGLVertexArrayObjectOES::hasEverBeenBound): (WebCore::WebGLVertexArrayObjectOES::setHasEverBeenBound): (WebCore::WebGLVertexArrayObjectOES::getElementArrayBuffer): (WebCore::WebGLVertexArrayObjectOES::setElementArrayBuffer): (WebCore::WebGLVertexArrayObjectOES::getVertexAttribState): (WebCore::WebGLVertexArrayObjectOES::isVertexArray): * html/canvas/WebGLVertexArrayObjectOES.idl: Added. * platform/graphics/Extensions3D.h: * platform/graphics/GraphicsContext3D.h: * platform/graphics/GraphicsTypes3D.h: Added. * platform/graphics/chromium/Extensions3DChromium.h: * platform/graphics/opengl/Extensions3DOpenGL.cpp: (WebCore::Extensions3DOpenGL::supports): (WebCore::Extensions3DOpenGL::createVertexArrayOES): (WebCore::Extensions3DOpenGL::deleteVertexArrayOES): (WebCore::Extensions3DOpenGL::isVertexArrayOES): (WebCore::Extensions3DOpenGL::bindVertexArrayOES): * platform/graphics/opengl/Extensions3DOpenGL.h: * platform/graphics/qt/Extensions3DQt.cpp: (WebCore::Extensions3DQt::createVertexArrayOES): (WebCore::Extensions3DQt::deleteVertexArrayOES): (WebCore::Extensions3DQt::isVertexArrayOES): (WebCore::Extensions3DQt::bindVertexArrayOES): * platform/graphics/qt/Extensions3DQt.h: 2011-02-17 Alexander Pavlov Reviewed by Pavel Feldman. Web Inspector: reflect changes to styles when they happen outside inspector. https://bugs.webkit.org/show_bug.cgi?id=44620 Notify InspectorInstrumentation of inline style changes from CSSMutableStyleDeclaration whenever the style change does not come from a direct "style" attribute modification. The performance regression occurs only with the Web Inspector opened, when inline styles are modified through setting style.cssText or style.: according to the Web Inspector protocol, an attribute change requires that all attributes for the node in question be pushed into the frontend. Test: inspector/styles/styles-update-from-js.html * css/CSSMutableStyleDeclaration.cpp: (WebCore::CSSMutableStyleDeclaration::setNeedsStyleRecalc): * inspector/InspectorDOMAgent.cpp: (WebCore::RevalidateStyleAttributeTask::reset): (WebCore::RevalidateStyleAttributeTask::RevalidateStyleAttributeTask): (WebCore::RevalidateStyleAttributeTask::scheduleFor): (WebCore::RevalidateStyleAttributeTask::onTimer): (WebCore::InspectorDOMAgent::reset): (WebCore::InspectorDOMAgent::didInvalidateStyleAttr): * inspector/InspectorDOMAgent.h: * inspector/InspectorInstrumentation.cpp: (WebCore::InspectorInstrumentation::didInvalidateStyleAttrImpl): * inspector/InspectorInstrumentation.h: (WebCore::InspectorInstrumentation::didInvalidateStyleAttr): * inspector/InspectorStyleSheet.cpp: (WebCore::InspectorStyleSheetForInlineStyle::didModifyElementAttribute): (WebCore::InspectorStyleSheetForInlineStyle::ensureParsedDataReady): (WebCore::InspectorStyleSheetForInlineStyle::elementStyleText): * inspector/InspectorStyleSheet.h: * inspector/front-end/ElementsPanel.js: (WebInspector.ElementsPanel.prototype._attributesUpdated): * inspector/front-end/StylesSidebarPane.js: (WebInspector.StylesSidebarPane.prototype._rebuildSectionsForStyleRules): (WebInspector.StylesSidebarPane.prototype.addBlankSection): (WebInspector.StylePropertiesSection.prototype.onpopulate): (WebInspector.StylePropertiesSection.prototype.addNewBlankProperty): (WebInspector.ComputedStylePropertiesSection.prototype.onpopulate): (WebInspector.BlankStylePropertiesSection): (WebInspector.StylePropertyTreeElement): (WebInspector.StylePropertyTreeElement.prototype): (WebInspector.StylePropertyTreeElement.prototype.element.userInput.previousContent.context.moveDirection): 2011-02-16 Tony Gentilcore Reviewed by Eric Seidel. Refactor pumpTokenizer loop https://bugs.webkit.org/show_bug.cgi?id=54574 1. This makes pumpTokenizer() safe to call when waiting for a script or when about to get the next token, although ASSERTs still enforce that we aren't waiting for a script. This enables resumeParsingAfterYield() to call pumpTokenizer with no modifications even if we yield before running a script rather than before taking a token (see bug 54355). 2. This also picks up the refCount >= 1 assert when stopped. Tested PerformanceTests/Parser to verify no regression. If anything it got faster. Before: avg 985.05 median 985.5 stdev 3.007906248539007 min 980 max 990 After: avg 980.05 median 981 stdev 3.122098653149833 min 974 max 985 No new tests because no new functionality. * html/parser/HTMLDocumentParser.cpp: (WebCore::HTMLDocumentParser::canTakeNextToken): Added. (WebCore::HTMLDocumentParser::pumpTokenizer): * html/parser/HTMLDocumentParser.h: 2011-02-18 Sheriff Bot Unreviewed, rolling out r79003. http://trac.webkit.org/changeset/79003 https://bugs.webkit.org/show_bug.cgi?id=54753 It broke tests on GTK bots (Requested by Ossy on #webkit). * accessibility/AccessibilityRenderObject.cpp: (WebCore::lastChildConsideringContinuation): * dom/Node.cpp: (WebCore::Node::removeEventListener): * html/DateComponents.cpp: (WebCore::DateComponents::parseTime): 2011-02-18 Csaba Osztrogonác Rubber-stamped by Andreas Kling. Warning fix. Use ASSERT_UNUSED() instead of ASSERT(). * accessibility/AccessibilityRenderObject.cpp: (WebCore::lastChildConsideringContinuation): * dom/Node.cpp: (WebCore::Node::removeEventListener): * html/DateComponents.cpp: (WebCore::DateComponents::parseTime): 2011-02-18 Andrey Adaikin Reviewed by Pavel Feldman. Web Inspector: [Text editor] Optimize editing updates in main panel https://bugs.webkit.org/show_bug.cgi?id=54661 * inspector/front-end/TextViewer.js: (WebInspector.TextViewer): (WebInspector.TextViewer.prototype._textChanged): (WebInspector.TextViewer.prototype._enterInternalTextChangeMode): (WebInspector.TextViewer.prototype._exitInternalTextChangeMode): (WebInspector.TextViewer.prototype._syncDecorationsForLine): (WebInspector.TextEditorChunkedPanel.prototype.textChanged): (WebInspector.TextEditorChunkedPanel.prototype._scroll): (WebInspector.TextEditorChunkedPanel.prototype.chunkForLine): (WebInspector.TextEditorChunkedPanel.prototype._totalHeight): (WebInspector.TextEditorGutterPanel.prototype._expandChunks): (WebInspector.TextEditorGutterChunk): (WebInspector.TextEditorGutterChunk.prototype.addDecoration): (WebInspector.TextEditorGutterChunk.prototype.removeDecoration): (WebInspector.TextEditorMainPanel): (WebInspector.TextEditorMainPanel.prototype._expandChunks): (WebInspector.TextEditorMainPanel.prototype._highlightDataReady): (WebInspector.TextEditorMainPanel.prototype._markSkippedPaintLines): (WebInspector.TextEditorMainPanel.prototype._paintSkippedLines): (WebInspector.TextEditorMainPanel.prototype._paintLines): (WebInspector.TextEditorMainPanel.prototype._paintLine): (WebInspector.TextEditorMainPanel.prototype._positionToSelection): (WebInspector.TextEditorMainPanel.prototype._applyDomUpdates): (WebInspector.TextEditorMainPanel.prototype._updateChunksForRanges): (WebInspector.TextEditorMainPanel.prototype._updateHighlightsForRange): (WebInspector.TextEditorMainPanel.prototype._collectLinesFromDiv): (WebInspector.TextEditorMainChunk): (WebInspector.TextEditorMainChunk.prototype.get startLine): (WebInspector.TextEditorMainChunk.prototype.set startLine): (WebInspector.TextEditorMainChunk.prototype.getExpandedLineRow): (WebInspector.TextEditorMainChunk.prototype.updateCollapsedLineRow): 2011-02-18 Steve Block Reviewed by Andreas Kling Memory allocation error in convertV8ObjectToNPVariant() for strings https://bugs.webkit.org/show_bug.cgi?id=54737 Include the termination character in the length when allocating memory and copying the string. This fixes a crashing bug on Android. This should be tested by the java/ tests on Chromium, but these tests are currently skipped. * bindings/v8/V8NPUtils.cpp: (WebCore::convertV8ObjectToNPVariant): 2011-02-18 Philippe Normand Unreviewed, rolling out r78979. http://trac.webkit.org/changeset/78979 https://bugs.webkit.org/show_bug.cgi?id=53146 causes multiple crashes on GTK * accessibility/gtk/AXObjectCacheAtk.cpp: (WebCore::notifyChildrenSelectionChange): (WebCore::AXObjectCache::postPlatformNotification): 2011-02-18 Philippe Normand Reviewed by Martin Robinson. [GTK] minimal build unrecognized options https://bugs.webkit.org/show_bug.cgi?id=50890 * GNUmakefile.am: new feature defines for optional features build. 2011-02-17 Hans Wennborg Reviewed by Jeremy Orlow. IndexedDB: Populate indexes created for object stores with data https://bugs.webkit.org/show_bug.cgi?id=54669 Make sure that indices for object stores that already hold data get populated. * storage/IDBIndexBackendImpl.h: (WebCore::IDBIndexBackendImpl::hasValidId): * storage/IDBObjectStoreBackendImpl.cpp: (WebCore::IDBObjectStoreBackendImpl::putInternal): (WebCore::populateIndex): (WebCore::IDBObjectStoreBackendImpl::createIndexInternal): 2011-02-18 Mario Sanchez Prada Reviewed by Martin Robinson. [GTK] Combo boxes should emit object:selection-changed even when collapsed https://bugs.webkit.org/show_bug.cgi?id=53146 Emit the selection-changed signals when the menu list value has changed Test: platform/gtk/accessibility/combo-box-collapsed-selection-changed.html * accessibility/gtk/AXObjectCacheAtk.cpp: (WebCore::getListObject): New, return the right list object for menu lists and list boxes. (WebCore::notifyChildrenSelectionChange): Support menu lists. (WebCore::AXObjectCache::postPlatformNotification): Call function notifyChildrenSelectionChange for AXMenuListValueChanged. 2011-02-18 Mario Sanchez Prada Reviewed by Martin Robinson. [Gtk] atk_text_get_selection/atk_text_set_selection fails for list items https://bugs.webkit.org/show_bug.cgi?id=53453 Ensure that atk_text_{get|set}_selection() work with list items. * accessibility/gtk/AccessibilityObjectWrapperAtk.cpp: (webkit_accessible_text_get_text): Properly handle list item markers when returning the text for an object for a given interval specified through the startOffset and endOffset parameters. (getSelectionOffsetsForObject): Bear in mind list item markers when returning the offsets for a selection over a list item. (webkit_accessible_text_set_selection): Adjust offsets if needed for list items with item markers. Ensure that it returns TRUE only when everything went fine setting the text selection. 2011-02-18 Antti Koivisto Reviewed by Maciej Stachowiak. https://bugs.webkit.org/show_bug.cgi?id=54728 checkSelector*Value functions used in fastCheckSelector fail to inline Wrap the functions used as template arguments to classes. * css/CSSStyleSelector.cpp: (WebCore::fastCheckSingleSelector): (WebCore::ClassCheck::checkValue): (WebCore::IdCheck::checkValue): (WebCore::TagCheck::checkValue): (WebCore::CSSStyleSelector::SelectorChecker::fastCheckSelector): 2011-02-17 Simon Fraser Reviewed by Sam Weinig. Composited iframe content is missing from snapshots in WebKit2 https://bugs.webkit.org/show_bug.cgi?id=54696 We need to propagate the 'flattening' paint behavior flag down to subviews while painting. WebKit1 does this via code in WebFrameView, but this is a more general fix that works in WebKit2 as well. Made a utility method, parentFrameView(), which I changed some other methods to use as well. * page/FrameView.cpp: (WebCore::FrameView::isEnclosedInCompositingLayer): (WebCore::FrameView::useSlowRepaints): (WebCore::FrameView::useSlowRepaintsIfNotOverlapped): (WebCore::FrameView::isOverlappedIncludingAncestors): (WebCore::FrameView::parentFrameView): (WebCore::FrameView::paintContents): * page/FrameView.h: 2011-02-17 Dan Bernstein Reviewed by Simon Fraser. Pages that use fixed positioning display poorly when scaled * html/HTMLBodyElement.cpp: (WebCore::adjustForZoom): Account for page scale. (WebCore::HTMLBodyElement::setScrollLeft): Ditto. (WebCore::HTMLBodyElement::setScrollTop): Ditto. * page/FrameView.cpp: (WebCore::FrameView::scrollXForFixedPosition): Moved from ScrollView here and changed to account for page scale: when the page is scaled, the “viewport” with respect to which fixed objects are positioned is scaled as well. Since it’s now bigger than the real viewport (that is, the frame view), we move it around in proportion to the document scroll, so that when the document is fully scrolled to the bottom-right, the bottom right of the scaled viewport is visible. (WebCore::FrameView::scrollYForFixedPosition): Ditto. (WebCore::FrameView::scrollOffsetForFixedPosition): Moved from ScrollView here. * page/FrameView.h: * platform/ScrollView.cpp: Moved functions to FrameView. * platform/ScrollView.h: * rendering/RenderLayer.cpp: (WebCore::RenderLayer::updateRepaintRectsAfterScroll): Account for the RenderView being transformed. (WebCore::RenderLayer::calculateClipRects): Used scrollOffsetForFixedPosition(). (WebCore::RenderLayer::backgroundClipRect): Ditto. 2011-02-17 Kenneth Russell Reviewed by James Robinson. Add support for GPU accelerated path rendering https://bugs.webkit.org/show_bug.cgi?id=44729 Incorporates the Loop and Blinn path rendering algorithm as an option to the GPU-accelerated canvas code, currently only compiled in to the Chromium port. Currently it's toggled by changing a hardcoded constant in SharedGraphicsContext3D::useLoopBlinnForPathRendering() and is disabled by default. This mechanism can be improved once we've gained more confidence in the implementation. There are some known bugs that need to be fixed first. No new tests; ran some 2D Canvas tests manually with the new flag both enabled and disabled. * WebCore.gypi: * platform/graphics/chromium/GLES2Canvas.cpp: (WebCore::GLES2Canvas::GLES2Canvas): (WebCore::GLES2Canvas::fillPath): * platform/graphics/chromium/GLES2Canvas.h: * platform/graphics/gpu/LoopBlinnClassifier.h: * platform/graphics/gpu/LoopBlinnLocalTriangulator.h: * platform/graphics/gpu/SharedGraphicsContext3D.cpp: (WebCore::SharedGraphicsContext3D::create): (WebCore::SharedGraphicsContext3D::SharedGraphicsContext3D): (WebCore::SharedGraphicsContext3D::createBuffer): (WebCore::SharedGraphicsContext3D::bindBuffer): (WebCore::SharedGraphicsContext3D::bufferData): (WebCore::SharedGraphicsContext3D::bufferSubData): (WebCore::SharedGraphicsContext3D::useLoopBlinnForPathRendering): (WebCore::SharedGraphicsContext3D::useLoopBlinnInteriorProgram): (WebCore::SharedGraphicsContext3D::useLoopBlinnExteriorProgram): * platform/graphics/gpu/SharedGraphicsContext3D.h: * platform/graphics/skia/GraphicsContextSkia.cpp: (WebCore::GraphicsContext::fillPath): 2011-02-16 Stephen White Reviewed by James Robinson. Layout Test fast/canvas/setWidthResetAfterForcedRender.html fails on accelerated 2D canvas w/compositor enabled. https://bugs.webkit.org/show_bug.cgi?id=54561 When resetting the CanvasRenderingContext2D, we also need to send a contentChanged() to the relevant RenderLayer. This is similar to what is done in didDraw(). Covered by fast/canvas/setWidthResetAfterForcedRender.html, but note that this test will still fail pixel tests because the compositor is not compatible with repaint tests (the green square is now white, but the half-transparent grey repaint rect does not appear). * html/canvas/CanvasRenderingContext2D.cpp: (WebCore::CanvasRenderingContext2D::reset): 2011-02-17 Sergey Glazunov Reviewed by Kenneth Russell. Null out the WEBKIT_lose_context WebGL extension's context pointer when the WebGL rendering context is removed. https://bugs.webkit.org/show_bug.cgi?id=54679 Test: fast/canvas/webgl/context-destroyed-crash.html * html/canvas/WebGLRenderingContext.cpp: (WebCore::WebGLRenderingContext::~WebGLRenderingContext): * html/canvas/WebKitLoseContext.cpp: (WebCore::WebKitLoseContext::loseContext): Add null-check for m_context. * html/canvas/WebKitLoseContext.h: (WebCore::WebKitLoseContext::contextDestroyed): 2011-02-17 Kenneth Russell Reviewed by Chris Marrin. Construction of Uint8Array from JS Array (and possibly others) incorrectly clamps values https://bugs.webkit.org/show_bug.cgi?id=52768 Removed incorrect clamping code from IntegralTypedArrayBase. Fixed code which casts from incoming double to the destination type. Changed the JSC bindings to use this code, rather than a copy of the casting code and a different constructor, in order to reuse the fix. * bindings/js/JSArrayBufferViewHelper.h: (WebCore::constructArrayBufferView): * html/canvas/Int32Array.h: (WebCore::Int32Array::set): * html/canvas/Int8Array.h: (WebCore::Int8Array::set): * html/canvas/IntegralTypedArrayBase.h: (WebCore::IntegralTypedArrayBase::set): * html/canvas/Uint16Array.h: (WebCore::Uint16Array::set): * html/canvas/Uint32Array.h: (WebCore::Uint32Array::set): * html/canvas/Uint8Array.h: (WebCore::Uint8Array::set): 2011-02-17 Sam Weinig Reviewed by Dan Bernstein. Knob proportion not quite right during rubber-band. Change knob proportion algorithm to treat overhang as making the view smaller, rather than document bigger. * platform/mac/ScrollbarThemeMac.mm: (WebCore::ScrollbarThemeMac::paint): 2011-02-16 Luiz Agostini Reviewed by Andreas Kling. Summary: HTML5
and : HTMLSummaryElement https://bugs.webkit.org/show_bug.cgi?id=54584 Adding class HTMLSummaryElement to build systems. This class will be used in the implementation of new HTML5 tag . * CMakeLists.txt: * GNUmakefile.am: * WebCore.gypi: * WebCore.pro: * WebCore.vcproj/WebCore.vcproj: * WebCore.xcodeproj/project.pbxproj: * html/HTMLElementsAllInOne.cpp: * html/HTMLSummaryElement.cpp: Added. (WebCore::HTMLSummaryElement::create): (WebCore::HTMLSummaryElement::HTMLSummaryElement): * html/HTMLSummaryElement.h: Added. * html/HTMLTagNames.in: 2011-02-17 Jeremy Orlow Reviewed by Nate Chapin. Throwing in an IndexedDB error or success event should lead to the transaction aborting https://bugs.webkit.org/show_bug.cgi?id=54249 When an exception is thrown but not handled within an IDBRequests success/error event, we should abort the transaction. Test: storage/indexeddb/exception-in-event-aborts.html * bindings/js/JSEventListener.cpp: (WebCore::JSEventListener::handleEvent): * bindings/js/JSEventTarget.cpp: (WebCore::toJS): * bindings/v8/V8AbstractEventListener.cpp: (WebCore::V8AbstractEventListener::invokeEventHandler): * dom/EventTarget.cpp: (WebCore::EventTarget::uncaughtExceptionInEventHandler): * dom/EventTarget.h: * storage/IDBRequest.cpp: (WebCore::IDBRequest::uncaughtExceptionInEventHandler): * storage/IDBRequest.h: 2011-02-17 Sam Weinig Reviewed by Maciej Stachowiak. WebKit2: Support Dictionary popup Add some necessary exports. * WebCore.exp.in: 2011-02-17 W. James MacLean Reviewed by James Robinson. [chromium] Add command-line flag to enable composite to offscreen texture. https://bugs.webkit.org/show_bug.cgi?id=52311 Add plumbing to allow command-line switch to enable offscreen compositing. Function LayerRendererChromium::copyOffscreenTextureToDisplay used for now to mimic normal renderer operation. Existing functionality not changed; offscreen compositing will be tested via GPU test framework. * platform/graphics/chromium/LayerRendererChromium.cpp: (WebCore::LayerRendererChromium::setRootLayer): (WebCore::LayerRendererChromium::setCompositeOffscreen): (WebCore::LayerRendererChromium::copyOffscreenTextureToDisplay): (WebCore::LayerRendererChromium::useRenderSurface): (WebCore::LayerRendererChromium::setScissorToRect): * platform/graphics/chromium/LayerRendererChromium.h: (WebCore::LayerRendererChromium::isCompositingOffscreen): 2011-02-17 Kevin Ollivier [wx] Build fixes after recent changes. * dom/ScriptedAnimationController.h: * platform/graphics/wx/FontWx.cpp: (WebCore::Font::drawComplexText): * platform/graphics/wx/GraphicsContextWx.cpp: (WebCore::GraphicsContext::drawLineForText): (WebCore::GraphicsContext::drawLineForTextChecking): * platform/wx/WidgetWx.cpp: (WebCore::Widget::setFrameRect): 2011-02-16 Brian Weinstein Reviewed by Brady Eidson. WebKit2: Need a way to manage the WebCore Cache https://bugs.webkit.org/show_bug.cgi?id=54501 Add a way to get a set of all of the origins that have entries in the WebCore memory cache, and a method to remove all resources from the memory cache from a given security origin. No change in behavior. * WebCore.exp.in: Add functions that need to be exported. * loader/cache/MemoryCache.cpp: (WebCore::MemoryCache::removeResourcesWithOrigin): (WebCore::MemoryCache::getOriginsWithCache): * loader/cache/MemoryCache.h: 2011-02-16 David Hyatt Reviewed by Dan Bernstein. https://bugs.webkit.org/show_bug.cgi?id=54244 Convert the line box tree to floating point and eliminate font rounding hacks. This patch removes all of the rounding hacks from the Font code and makes sure all Font APIs involving width measurement and width offsets use floats. The line box tree's x, y and logicalWidth members have all been converted to floats and all of the line box APIs have been changed as well. In terms of pixel adjustments, overflow is using an enclosing model (so it will be enclosingIntRect of a line box's x/y/width/height). Background and border painting is using a rounding model, so borders and backgrounds will round to the nearest pixel when painting. Replaced elements still snap to integer positions on lines, and they use a rounding model as well, although their underlying line boxes still have a precise floating point position. Justification will now allow subpixel positioning to occur as well. Platforms that don't support subpixel positioning should already be rounding justification spacing in their font code. Many layout test results change on Mac, since rounding hacks were used there and are now gone. * WebCore.exp.in: * html/canvas/CanvasRenderingContext2D.cpp: (WebCore::CanvasRenderingContext2D::drawTextInternal): * platform/chromium/FileChooserChromium.cpp: (WebCore::FileChooser::basenameForWidth): * platform/graphics/Font.cpp: (WebCore::Font::width): * platform/graphics/Font.h: (WebCore::Font::spaceWidth): (WebCore::Font::tabWidth): * platform/graphics/FontFastPath.cpp: (WebCore::Font::getGlyphsAndAdvancesForSimpleText): * platform/graphics/GraphicsContext.cpp: (WebCore::GraphicsContext::drawText): (WebCore::GraphicsContext::drawEmphasisMarks): (WebCore::GraphicsContext::drawBidiText): (WebCore::GraphicsContext::drawHighlightForText): * platform/graphics/GraphicsContext.h: * platform/graphics/SimpleFontData.cpp: (WebCore::SimpleFontData::SimpleFontData): (WebCore::SimpleFontData::platformGlyphInit): * platform/graphics/SimpleFontData.h: (WebCore::SimpleFontData::spaceWidth): * platform/graphics/StringTruncator.cpp: (WebCore::stringWidth): (WebCore::truncateString): (WebCore::StringTruncator::centerTruncate): (WebCore::StringTruncator::rightTruncate): (WebCore::StringTruncator::width): * platform/graphics/StringTruncator.h: * platform/graphics/TextRun.h: (WebCore::TextRun::TextRun): (WebCore::TextRun::xPos): (WebCore::TextRun::expansion): (WebCore::TextRun::directionalOverride): (WebCore::TextRun::disableSpacing): * platform/graphics/WidthIterator.cpp: (WebCore::WidthIterator::WidthIterator): (WebCore::WidthIterator::advance): * platform/graphics/WidthIterator.h: * platform/graphics/cairo/GraphicsContextCairo.cpp: (WebCore::GraphicsContext::drawLineForText): (WebCore::GraphicsContext::drawLineForTextChecking): * platform/graphics/cg/GraphicsContextCG.cpp: (WebCore::GraphicsContext::drawLineForText): * platform/graphics/mac/ComplexTextController.cpp: (WebCore::ComplexTextController::ComplexTextController): (WebCore::ComplexTextController::advance): (WebCore::ComplexTextController::adjustGlyphsAndAdvances): * platform/graphics/mac/ComplexTextController.h: * platform/graphics/mac/FontComplexTextMac.cpp: (WebCore::Font::getGlyphsAndAdvancesForComplexText): * platform/graphics/mac/GraphicsContextMac.mm: (WebCore::GraphicsContext::drawLineForTextChecking): * platform/graphics/qt/GraphicsContextQt.cpp: (WebCore::GraphicsContext::drawLineForText): (WebCore::GraphicsContext::drawLineForTextChecking): * platform/graphics/qt/SimpleFontDataQt.cpp: (WebCore::SimpleFontData::platformGlyphInit): * platform/graphics/skia/GraphicsContextSkia.cpp: (WebCore::GraphicsContext::drawLineForTextChecking): (WebCore::GraphicsContext::drawLineForText): * platform/graphics/win/GraphicsContextCGWin.cpp: (WebCore::GraphicsContext::drawLineForTextChecking): * platform/graphics/win/UniscribeController.cpp: (WebCore::UniscribeController::shapeAndPlaceItem): * platform/gtk/FileChooserGtk.cpp: (WebCore::FileChooser::basenameForWidth): * platform/mac/DragImageMac.mm: (WebCore::widthWithFont): (WebCore::drawAtPoint): * platform/mac/FileChooserMac.mm: (WebCore::FileChooser::basenameForWidth): * platform/win/DragImageWin.cpp: (WebCore::createDragImageForLink): * platform/win/FileChooserWin.cpp: (WebCore::FileChooser::basenameForWidth): * platform/win/PopupMenuWin.cpp: (WebCore::PopupMenuWin::calculatePositionAndSize): * platform/win/WebCoreTextRenderer.cpp: (WebCore::WebCoreTextFloatWidth): * rendering/HitTestResult.cpp: (WebCore::HitTestResult::addNodeToRectBasedTestResult): * rendering/HitTestResult.h: * rendering/InlineBox.cpp: (WebCore::InlineBox::adjustPosition): (WebCore::InlineBox::placeEllipsisBox): (WebCore::InlineBox::locationIncludingFlipping): (WebCore::InlineBox::flipForWritingMode): * rendering/InlineBox.h: (WebCore::InlineBox::InlineBox): (WebCore::InlineBox::adjustLineDirectionPosition): (WebCore::InlineBox::adjustBlockDirectionPosition): (WebCore::InlineBox::setX): (WebCore::InlineBox::x): (WebCore::InlineBox::setY): (WebCore::InlineBox::y): (WebCore::InlineBox::width): (WebCore::InlineBox::height): (WebCore::InlineBox::logicalLeft): (WebCore::InlineBox::logicalRight): (WebCore::InlineBox::setLogicalLeft): (WebCore::InlineBox::pixelSnappedLogicalLeft): (WebCore::InlineBox::pixelSnappedLogicalRight): (WebCore::InlineBox::setLogicalWidth): (WebCore::InlineBox::logicalWidth): (WebCore::InlineBox::verticalAlign): * rendering/InlineFlowBox.cpp: (WebCore::InlineFlowBox::roundedFrameRect): (WebCore::InlineFlowBox::adjustPosition): (WebCore::InlineFlowBox::placeBoxesInInlineDirection): (WebCore::InlineFlowBox::adjustMaxAscentAndDescent): (WebCore::verticalPositionForBox): (WebCore::InlineFlowBox::computeLogicalBoxHeights): (WebCore::InlineFlowBox::placeBoxesInBlockDirection): (WebCore::InlineFlowBox::addBoxShadowVisualOverflow): (WebCore::InlineFlowBox::addTextBoxVisualOverflow): (WebCore::InlineFlowBox::computeOverflow): (WebCore::InlineFlowBox::setLayoutOverflow): (WebCore::InlineFlowBox::setVisualOverflow): (WebCore::InlineFlowBox::nodeAtPoint): (WebCore::InlineFlowBox::paintBoxDecorations): (WebCore::InlineFlowBox::paintMask): (WebCore::InlineFlowBox::placeEllipsisBox): * rendering/InlineFlowBox.h: (WebCore::InlineFlowBox::maxYLayoutOverflow): (WebCore::InlineFlowBox::maxXLayoutOverflow): (WebCore::InlineFlowBox::layoutOverflowRect): (WebCore::InlineFlowBox::maxYVisualOverflow): (WebCore::InlineFlowBox::maxXVisualOverflow): (WebCore::InlineFlowBox::visualOverflowRect): * rendering/InlineTextBox.cpp: (WebCore::InlineTextBox::placeEllipsisBox): (WebCore::InlineTextBox::nodeAtPoint): (WebCore::paintTextWithShadows): (WebCore::InlineTextBox::paint): (WebCore::InlineTextBox::paintSelection): (WebCore::InlineTextBox::paintCompositionBackground): (WebCore::InlineTextBox::paintDecoration): (WebCore::InlineTextBox::paintSpellingOrGrammarMarker): (WebCore::InlineTextBox::paintTextMatchMarker): (WebCore::InlineTextBox::paintDocumentMarkers): (WebCore::InlineTextBox::paintCompositionUnderline): (WebCore::InlineTextBox::textPos): (WebCore::InlineTextBox::offsetForPosition): (WebCore::InlineTextBox::positionForOffset): * rendering/InlineTextBox.h: (WebCore::InlineTextBox::setExpansion): * rendering/RenderBR.h: (WebCore::RenderBR::width): * rendering/RenderBlock.cpp: (WebCore::stripTrailingSpace): (WebCore::updatePreferredWidth): (WebCore::RenderBlock::computeInlinePreferredLogicalWidths): (WebCore::RenderBlock::adjustForBorderFit): (WebCore::RenderBlock::addFocusRingRects): * rendering/RenderBlock.h: * rendering/RenderBlockLineLayout.cpp: (WebCore::RenderBlock::computeInlineDirectionPositionsForLine): (WebCore::RenderBlock::fitBelowFloats): (WebCore::textWidth): (WebCore::tryHyphenating): (WebCore::RenderBlock::findNextLineBreak): * rendering/RenderBox.cpp: (WebCore::RenderBox::positionLineBox): (WebCore::RenderBox::flipForWritingMode): * rendering/RenderBox.h: * rendering/RenderCombineText.cpp: (WebCore::RenderCombineText::width): (WebCore::RenderCombineText::adjustTextOrigin): (WebCore::RenderCombineText::combineText): * rendering/RenderCombineText.h: (WebCore::RenderCombineText::combinedTextWidth): * rendering/RenderCounter.cpp: (WebCore::RenderCounter::computePreferredLogicalWidths): * rendering/RenderCounter.h: * rendering/RenderEmbeddedObject.cpp: (WebCore::RenderEmbeddedObject::getReplacementTextGeometry): * rendering/RenderFileUploadControl.cpp: (WebCore::RenderFileUploadControl::computePreferredLogicalWidths): * rendering/RenderImage.cpp: * rendering/RenderInline.cpp: (WebCore::RenderInline::linesBoundingBox): (WebCore::RenderInline::linesVisualOverflowBoundingBox): (WebCore::RenderInline::addFocusRingRects): (WebCore::RenderInline::paintOutline): * rendering/RenderListBox.cpp: (WebCore::RenderListBox::updateFromElement): (WebCore::RenderListBox::paintItemForeground): * rendering/RenderMenuList.cpp: (WebCore::RenderMenuList::updateOptionsWidth): * rendering/RenderText.cpp: (WebCore::RenderText::localCaretRect): (WebCore::RenderText::widthFromCache): (WebCore::RenderText::trimmedPrefWidths): (WebCore::RenderText::minLogicalWidth): (WebCore::RenderText::maxLogicalWidth): (WebCore::RenderText::computePreferredLogicalWidths): (WebCore::RenderText::firstRunOrigin): (WebCore::RenderText::firstRunX): (WebCore::RenderText::firstRunY): (WebCore::RenderText::width): (WebCore::RenderText::linesBoundingBox): * rendering/RenderText.h: * rendering/RenderTextControl.cpp: (WebCore::RenderTextControl::getAvgCharWidth): (WebCore::RenderTextControl::paintPlaceholder): * rendering/RenderTreeAsText.cpp: (WebCore::writeTextRun): * rendering/RootInlineBox.cpp: (WebCore::RootInlineBox::placeEllipsis): (WebCore::RootInlineBox::placeEllipsisBox): (WebCore::RootInlineBox::adjustPosition): (WebCore::RootInlineBox::beforeAnnotationsAdjustment): (WebCore::RootInlineBox::paddedLayoutOverflowRect): * rendering/RootInlineBox.h: * rendering/VerticalPositionCache.h: * rendering/svg/SVGInlineTextBox.cpp: (WebCore::SVGInlineTextBox::offsetForPosition): (WebCore::SVGInlineTextBox::positionForOffset): (WebCore::SVGInlineTextBox::constructTextRun): * rendering/svg/SVGInlineTextBox.h: * rendering/svg/SVGRenderTreeAsText.cpp: (WebCore::writeRenderSVGTextBox): * rendering/svg/SVGTextMetrics.cpp: (WebCore::SVGTextMetrics::SVGTextMetrics): (WebCore::constructTextRun): * svg/SVGFont.cpp: (WebCore::floatWidthMissingGlyphCallback): (WebCore::Font::drawTextUsingSVGFont): 2011-02-17 Nikolas Zimmermann Reviewed by Dirk Schulze. 'ex' coordinates fail, when SVGFont doesn't provide an explicit xHeight attribute https://bugs.webkit.org/show_bug.cgi?id=54672 Measure the xHeight from the 'x' glyph of a SVGFont, if the font itself doesn't explicitely specify an x-height attribute. Fixes the modern version of SVG 1.1 2nd Edition coords-units-03-b.svg. Test: svg/W3C-SVG-1.1-SE/coords-units-03-b.svg * platform/graphics/SimpleFontData.cpp: (WebCore::SimpleFontData::SimpleFontData): 2011-02-10 Luiz Agostini Reviewed by Adam Roben. HTML5
and : localized text https://bugs.webkit.org/show_bug.cgi?id=54260 The method defaultDetailsSummaryText was added to LocalizationStrategy class and to platform/LocalizedStrings. It is used to provide the default label to be used by a
tag that has no child. * platform/LocalizationStrategy.h: * platform/LocalizedStrings.cpp: (WebCore::fileButtonNoFileSelectedLabel): (WebCore::defaultDetailsSummaryText): * platform/LocalizedStrings.h: * platform/android/LocalizedStringsAndroid.cpp: (WebCore::defaultDetailsSummaryText): * platform/brew/LocalizedStringsBrew.cpp: (WebCore::defaultDetailsSummaryText): * platform/efl/LocalizedStringsEfl.cpp: (WebCore::defaultDetailsSummaryText): * platform/gtk/LocalizedStringsGtk.cpp: (WebCore::defaultDetailsSummaryText): * platform/haiku/LocalizedStringsHaiku.cpp: (WebCore::defaultDetailsSummaryText): * platform/wx/LocalizedStringsWx.cpp: (WebCore::defaultDetailsSummaryText): 2011-02-17 Kristian Amlie Reviewed by Laszlo Gombos. Updated include paths for phonon. [Qt] WebKit patches required to work with a modularized version of Qt https://bugs.webkit.org/show_bug.cgi?id=53916 Build fix. No tests. * WebCore.pro: 2011-02-17 Hui Huang Reviewed by Laszlo Gombos. The URL of HTML5 Video Element is percent encoded at websites such as youtube. It is percent encoded again by QUrl constructor QUrl::QUrl(QString). This causes the HTTP GET request for the video to be rejected by the service provider. https://bugs.webkit.org/show_bug.cgi?id=53973. The bug is fixed by constructing QUrl from the encoded URL. New test function tst_QWebPage::loadHtml5Video() is added in Source/WebKit/qt/tests/qwebpage/tst_qwebpage.cpp * platform/graphics/qt/MediaPlayerPrivateQt.cpp: (WebCore::MediaPlayerPrivateQt::commitLoad): 2011-02-17 Andreas Kling Reviewed by Antti Koivisto. [Qt] Crash when calling QWebFrame::setUrl() while a previous load has pending requests https://bugs.webkit.org/show_bug.cgi?id=49216 CachedResourceRequest::didFail() will protect the CachedResourceLoader's document() while it runs, but if we're being called from the Document destructor, the protecting RefPtr will cause a double-delete instead. * loader/cache/CachedResourceLoader.cpp: (WebCore::CachedResourceLoader::~CachedResourceLoader): Clear the m_document pointer so CachedResourceRequest::didFail() won't try to protect it. (WebCore::CachedResourceLoader::frame): Add null-check for m_document. 2011-02-17 Andrey Adaikin Reviewed by Pavel Feldman. Web Inspector: [Text editor] Add updateHighlight method to the highlighter https://bugs.webkit.org/show_bug.cgi?id=54448 * inspector/front-end/SourceTokenizer.js: * inspector/front-end/TextEditorHighlighter.js: (WebInspector.TextEditorHighlighter.prototype.set mimeType): (WebInspector.TextEditorHighlighter.prototype.reset): (WebInspector.TextEditorHighlighter.prototype.updateHighlight): (WebInspector.TextEditorHighlighter.prototype._highlightInChunks): (WebInspector.TextEditorHighlighter.prototype._highlightLines): 2011-02-16 Pavel Podivilov Reviewed by Yury Semikhatsky. Web Inspector: show all inlined scripts from single document in the same source frame. https://bugs.webkit.org/show_bug.cgi?id=54544 Currently when debugging synchronously executed inlined scripts each script is shown in it's own source frame ("example.html:24"). We should show such scripts in the same source frame "example.html" with framing. Test: inspector/debugger/debug-inlined-scripts.html * inspector/front-end/ScriptsPanel.js: (WebInspector.ScriptsPanel.prototype._scriptSourceChanged): (WebInspector.ScriptsPanel.prototype._addScript): (WebInspector.ScriptsPanel.prototype._resourceForURL): (WebInspector.ScriptsPanel.prototype._resourceLoadingFinished): (WebInspector.ScriptsPanel.prototype.addConsoleMessage): (WebInspector.ScriptsPanel.prototype.clearConsoleMessages): (WebInspector.ScriptsPanel.prototype.reset): (WebInspector.ScriptsPanel.prototype._sourceFrameForResource): (WebInspector.ScriptsPanel.prototype._sourceFrameForScript): (WebInspector.ScriptsPanel.prototype._recreateSourceFrame): (WebInspector.ScriptsPanel.prototype._showScriptOrResource): (WebInspector.ScriptsPanel.prototype._addScriptToFilesMenu.optionCompare): (WebInspector.ScriptsPanel.prototype._addScriptToFilesMenu): (WebInspector.SourceFrameContentProviderForScript.prototype.requestContent): (WebInspector.SourceFrameContentProviderForScript.prototype._buildSource): 2011-02-17 Nikolas Zimmermann Reviewed by Dirk Schulze. svg/batik/paints/patternRegions-positioned-objects.svg fails on Windows https://bugs.webkit.org/show_bug.cgi?id=44484 Pattern of pattern defined with objectBoundingBox does not render correctly https://bugs.webkit.org/show_bug.cgi?id=53463 Fix + patternContentUnits="objectBoundingBox" support. We were incorrrectly translating the tile image transform, by the target objects bbox.x()/y(). RenderSVGResourceMask/Clipper don't have this error. Fix nesting in objectBoundingBox mode, propagate the tileImageTransform as new user-space when drawing the pattern children. + don't have the problem. Test: svg/custom/nested-pattern-boundingBoxModeContent.svg * rendering/svg/RenderSVGResourcePattern.cpp: (WebCore::RenderSVGResourcePattern::buildTileImageTransform): (WebCore::RenderSVGResourcePattern::createTileImage): 2011-02-17 Csaba Osztrogonác Unreviewed. [Qt][V8] Buildfix after r78752. * CodeGenerators.pri: Add missing IDL files. 2011-02-17 Benjamin Kalman Reviewed by Ryosuke Niwa. RTL lineboundary left/right is reversed when cursor is at start of RTL container https://bugs.webkit.org/show_bug.cgi?id=54534 Test: editing/selection/extend-left-right-by-lineboundary.html Add missing cases for extending left/right by lineboundary. * editing/SelectionController.cpp: (WebCore::SelectionController::modifyExtendingRight): (WebCore::SelectionController::modifyExtendingLeft): 2011-02-16 Philippe Normand Reviewed by Martin Robinson. [GTK] libsoup critical warnings https://bugs.webkit.org/show_bug.cgi?id=54557 Avoid pausing a soup message for already downloaded resources. * platform/network/soup/ResourceHandleSoup.cpp: (WebCore::ResourceHandle::platformSetDefersLoading): 2011-02-16 Brian Ryner Reviewed by Darin Fisher. Split the socket address field into separate IP address and port fields. This will make the field less error-prone to parse, for example when dealing with IPv6 literals. https://bugs.webkit.org/show_bug.cgi?id=54607 No new tests required. * platform/network/chromium/ResourceResponse.cpp: (WebCore::ResourceResponse::doPlatformCopyData): (WebCore::ResourceResponse::doPlatformAdopt): * platform/network/chromium/ResourceResponse.h: (WebCore::ResourceResponse::ResourceResponse): (WebCore::ResourceResponse::remoteIPAddress): (WebCore::ResourceResponse::setRemoteIPAddress): (WebCore::ResourceResponse::remotePort): (WebCore::ResourceResponse::setRemotePort): 2011-02-16 Dominic Mazzoni Reviewed by Chris Fleizach. Add support for canvas fallback content. https://bugs.webkit.org/show_bug.cgi?id=50126 Test: accessibility/canvas-fallback-content.html * accessibility/AccessibilityObject.h: * accessibility/AccessibilityRenderObject.cpp: (WebCore::AccessibilityRenderObject::determineAccessibilityRole): (WebCore::AccessibilityRenderObject::canHaveChildren): * accessibility/gtk/AccessibilityObjectWrapperAtk.cpp: (atkRole): * accessibility/mac/AccessibilityObjectWrapper.mm: * html/HTMLFormControlElement.cpp: (WebCore::HTMLFormControlElement::isFocusable): * rendering/RenderHTMLCanvas.cpp: (WebCore::RenderHTMLCanvas::recursiveSetNoNeedsLayout): (WebCore::RenderHTMLCanvas::layout): (WebCore::RenderHTMLCanvas::nodeAtPoint): * rendering/RenderHTMLCanvas.h: (WebCore::RenderHTMLCanvas::children): (WebCore::RenderHTMLCanvas::canHaveChildren): (WebCore::RenderHTMLCanvas::virtualChildren): * rendering/RenderObject.cpp: (WebCore::RenderObject::repaint): * rendering/RenderTreeAsText.cpp: (WebCore::write): 2011-02-16 Matthew Delaney Reviewed by Simon Fraser. Allow acceleratesDrawing for WebKit2 https://bugs.webkit.org/show_bug.cgi?id=54511 Plumb through preference for accelerated drawing. When accelerated drawing is enabled, set a flag on new GraphicsLayers. Not testable via Layout Tests * WebCore.exp.in: * page/Settings.cpp: (WebCore::Settings::Settings): (WebCore::Settings::setAcceleratedDrawingEnabled): * page/Settings.h: (WebCore::Settings::acceleratedDrawingEnabled): * rendering/RenderLayerCompositor.cpp: (WebCore::RenderLayerCompositor::updateBacking): 2011-02-16 Victoria Kirst Reviewed by Kenneth Russell. [chromium] Fix green pixels at edge of certain GPU-accelerated videos https://bugs.webkit.org/show_bug.cgi?id=54559 Adds logic to properly resize the range of YUV textures to only select legitimate values. * platform/graphics/chromium/VideoFrameChromium.h: * platform/graphics/chromium/VideoLayerChromium.cpp: (WebCore::VideoLayerChromium::SharedValues::SharedValues): (WebCore::VideoLayerChromium::allocateTexturesIfNeeded): (WebCore::VideoLayerChromium::drawYUV): * platform/graphics/chromium/VideoLayerChromium.h: (WebCore::VideoLayerChromium::SharedValues::yWidthScaleFactorLocation): (WebCore::VideoLayerChromium::SharedValues::uvWidthScaleFactorLocation): 2011-02-16 Bill Budge Reviewed by David Levin. Need didReceiveCachedMetadata, and finishTime for didFinishLoading exposed in ThreadableLoaderClient https://bugs.webkit.org/show_bug.cgi?id=54313 No tests needed, exposes no new functionality * fileapi/FileReaderLoader.cpp: (WebCore::FileReaderLoader::didFinishLoading): * fileapi/FileReaderLoader.h: * loader/DocumentThreadableLoader.cpp: (WebCore::DocumentThreadableLoader::setDefersLoading): (WebCore::DocumentThreadableLoader::didReceiveCachedMetadata): (WebCore::DocumentThreadableLoader::didFinishLoading): (WebCore::DocumentThreadableLoader::loadRequest): * loader/DocumentThreadableLoader.h: * loader/ThreadableLoaderClient.h: (WebCore::ThreadableLoaderClient::didReceiveData): (WebCore::ThreadableLoaderClient::didReceiveCachedMetadata): (WebCore::ThreadableLoaderClient::didFinishLoading): * loader/ThreadableLoaderClientWrapper.h: (WebCore::ThreadableLoaderClientWrapper::didReceiveData): (WebCore::ThreadableLoaderClientWrapper::didReceiveCachedMetadata): (WebCore::ThreadableLoaderClientWrapper::didFinishLoading): * loader/WorkerThreadableLoader.cpp: (WebCore::workerContextDidReceiveCachedMetadata): (WebCore::WorkerThreadableLoader::MainThreadBridge::didReceiveCachedMetadata): (WebCore::workerContextDidFinishLoading): (WebCore::WorkerThreadableLoader::MainThreadBridge::didFinishLoading): * loader/WorkerThreadableLoader.h: * notifications/Notification.cpp: (WebCore::Notification::didFinishLoading): * notifications/Notification.h: * page/EventSource.cpp: (WebCore::EventSource::didFinishLoading): * page/EventSource.h: * workers/WorkerScriptLoader.cpp: (WebCore::WorkerScriptLoader::didFinishLoading): * workers/WorkerScriptLoader.h: * xml/XMLHttpRequest.cpp: (WebCore::XMLHttpRequest::didFinishLoading): * xml/XMLHttpRequest.h: 2011-02-16 Jeremy Orlow Fix uninitialized memory error. * storage/IDBDatabaseBackendImpl.cpp: (WebCore::IDBDatabaseBackendImpl::close): 2011-02-16 Adam Barth Reviewed by Eric Seidel. Fix xssAuditor/form-action.html https://bugs.webkit.org/show_bug.cgi?id=54590 We should block form actions. Although this technically can't be used to run script, it's a pretty easy vector for stealing passwords. * html/parser/XSSFilter.cpp: (WebCore::XSSFilter::filterTokenInitial): (WebCore::XSSFilter::filterFormToken): * html/parser/XSSFilter.h: 2011-02-16 Abhishek Arya Reviewed by James Robinson. Remove the early bail added in r75823 since we can run into anonymous blocks when traversing the parents chain for clearing floats. https://bugs.webkit.org/show_bug.cgi?id=54601 removeFloatingOrPositionedChildFromBlockLists tries to find the topmost parent containing "this" block and then tries to remove it from its floats list and mark all descendants blocks for layout. I added a bailout condition in r75823 because we thought that if one of the parent render block does not contain "this" float, then it is safe to assume that none of the grand parents will have it. This is a wrong assumption since anonymous blocks do not have float objects and we need to go higher in the chain to find the top most parent containing this float. Instead of breaking out of the loop, it is ok to keep traversing the chain till we find that parent. Otherwise, we will leave deleted floats in the grand parents floats list. Test: fast/block/float/floats-not-cleared-from-grand-parents.html * rendering/RenderBox.cpp: (WebCore::RenderBox::removeFloatingOrPositionedChildFromBlockLists): 2011-02-16 Andreas Kling Reviewed by Ryosuke Niwa. Editing styles should not emit #RRGGBBAA colors https://bugs.webkit.org/show_bug.cgi?id=54540 * editing/ApplyStyleCommand.cpp: (WebCore::StyleChange::extractTextStyles): Use Color::serialized() instead of Color::nameForRenderTreeAsText(). 2011-02-16 Ryosuke Niwa Reviewed by Kent Tamura. Improve showTree of Range, Position, and VisiblePosition https://bugs.webkit.org/show_bug.cgi?id=54536 Enhanced showTree of Range, Position, and VisiblePosition. * dom/Position.cpp: (WebCore::Position::showAnchorTypeAndOffset): Added; dumps "legacy" if the position is a legacy position and also dumps anchor type. (WebCore::Position::showTreeForThis): Calls showAnchorTypeAndOffset. * dom/Position.h: * dom/Range.cpp: (showTree): No longer calls deprecatedEditingOffset. * editing/VisiblePosition.cpp: (WebCore::VisiblePosition::debugPosition): No longer included in release build. Calls Position::showAnchorTypeAndOffset instead of manually calling deprecatedEditingOffset. * editing/VisiblePosition.h: * editing/VisibleSelection.cpp: (WebCore::VisibleSelection::debugPosition): Ditto. (WebCore::VisibleSelection::showTreeForThis): Ditto. * editing/VisibleSelection.h: * page/EventHandler.cpp: 2011-02-16 David Grogan Reviewed by Jeremy Orlow. fix compile error introduced in 78752 https://bugs.webkit.org/show_bug.cgi?id=54604 * storage/IDBRequest.h: 2011-02-16 Jeremy Orlow Back out IndexedDB change thats no longer necessary https://bugs.webkit.org/show_bug.cgi?id=54603 Backing out 78645 as it turns out that it's not necessary. * storage/IDBFactoryBackendImpl.cpp: (WebCore::IDBFactoryBackendImpl::open): * storage/IDBFactoryBackendImpl.h: * storage/IDBFactoryBackendInterface.h: 2011-02-16 Brian Salomon Reviewed by James Robinson. Skia's gpu backed just needs the correct context bound before drawing. It will bind the correct FBO itself and doing so externally confuses it unless resetContext is called. No new tests required. * platform/graphics/skia/PlatformContextSkia.cpp: (WebCore::PlatformContextSkia::syncSoftwareCanvas): 2011-02-16 Mike Reed Reviewed by Kenneth Russell. Use non-asserting pack function for decoding images, since webgl may want a non-premultiplied version of the image. https://bugs.webkit.org/show_bug.cgi?id=54023 No new tests. fast/canvas/webgl/gl-teximage.html fast/canvas/webgl/tex-image-with-format-and-type.html fast/canvas/webgl/texture-transparent-pixels-initialized.html * platform/image-decoders/ImageDecoder.h: (WebCore::ImageFrame::setRGBA): 2011-02-16 David Grogan Reviewed by Jeremy Orlow. indexeddb: make setVersion fire blocked event if other connections are open https://bugs.webkit.org/show_bug.cgi?id=53728 Tests: storage/indexeddb/set_version_blocked.html storage/indexeddb/set_version_queue.html * WebCore.gypi: * bindings/js/JSEventCustom.cpp: (WebCore::toJS): * bindings/v8/custom/V8EventCustom.cpp: (WebCore::toV8): * dom/Event.cpp: (WebCore::Event::isIDBVersionChangeEvent): * dom/Event.h: * dom/EventNames.h: * dom/EventTarget.cpp: (WebCore::EventTarget::toIDBVersionChangeRequest): * dom/EventTarget.h: * storage/IDBCallbacks.h: * storage/IDBDatabase.cpp: (WebCore::IDBDatabase::setVersion): (WebCore::IDBDatabase::close): * storage/IDBDatabase.h: * storage/IDBDatabase.idl: * storage/IDBDatabaseBackendImpl.cpp: (WebCore::IDBDatabaseBackendImpl::PendingSetVersionCall::create): (WebCore::IDBDatabaseBackendImpl::PendingSetVersionCall::version): (WebCore::IDBDatabaseBackendImpl::PendingSetVersionCall::callbacks): (WebCore::IDBDatabaseBackendImpl::PendingSetVersionCall::PendingSetVersionCall): (WebCore::IDBDatabaseBackendImpl::IDBDatabaseBackendImpl): (WebCore::IDBDatabaseBackendImpl::setVersion): (WebCore::IDBDatabaseBackendImpl::open): (WebCore::IDBDatabaseBackendImpl::close): * storage/IDBDatabaseBackendImpl.h: * storage/IDBDatabaseBackendInterface.h: * storage/IDBFactoryBackendImpl.cpp: (WebCore::IDBFactoryBackendImpl::open): * storage/IDBRequest.cpp: (WebCore::IDBRequest::onBlocked): (WebCore::IDBRequest::dispatchEvent): (WebCore::IDBRequest::source): * storage/IDBRequest.h: * storage/IDBVersionChangeEvent.cpp: Copied from Source/WebKit/chromium/src/WebIDBCallbacksImpl.h. (WebCore::IDBVersionChangeEvent::create): (WebCore::IDBVersionChangeEvent::IDBVersionChangeEvent): (WebCore::IDBVersionChangeEvent::~IDBVersionChangeEvent): (WebCore::IDBVersionChangeEvent::version): * storage/IDBVersionChangeEvent.h: Copied from Source/WebKit/chromium/src/WebIDBCallbacksImpl.h. (WebCore::IDBVersionChangeEvent::isIDBVersionChangeEvent): * storage/IDBVersionChangeEvent.idl: Added. * storage/IDBVersionChangeRequest.cpp: Copied from Source/WebKit/chromium/src/WebIDBCallbacksImpl.h. (WebCore::IDBVersionChangeRequest::create): (WebCore::IDBVersionChangeRequest::IDBVersionChangeRequest): (WebCore::IDBVersionChangeRequest::~IDBVersionChangeRequest): (WebCore::IDBVersionChangeRequest::onBlocked): * storage/IDBVersionChangeRequest.h: Copied from Source/WebKit/chromium/src/WebIDBCallbacksImpl.h. * storage/IDBVersionChangeRequest.idl: Copied from Source/WebKit/chromium/src/WebIDBCallbacksImpl.h. 2011-02-16 Robin Cao Reviewed by James Robinson. PlatformContextSkia::applyAntiAliasedClipPaths does not work for paths which have evenOdd property https://bugs.webkit.org/show_bug.cgi?id=54336 We need to take fill type of paths into account when drawing them. No new tests, covered by svg/W3C-SVG-1.1/masking-path-05-f.svg. * platform/graphics/skia/PlatformContextSkia.cpp: (WebCore::PlatformContextSkia::applyAntiAliasedClipPaths): 2011-02-16 Eric Seidel Reviewed by Adam Barth. REGRESSION (r61234): washingtonpost.com top bar looks wrong, doesn't animate https://bugs.webkit.org/show_bug.cgi?id=53717 Test: http/tests/local/absolute-url-strip-whitespace.html This was theoretically tested already in fast/url, however the tests were disabled due to lack of any clean way to test absolute url parsing in JavaScript. I added a test which mimics the sites behavior using our local http server. There seems to be no other way to test this at the moment. * platform/KURL.cpp: (WebCore::shouldTrimFromURL): - Any char 0-20 should be removed (matches google-url and other browsers). (WebCore::KURL::init): 2011-02-16 Abhishek Arya Reviewed by James Robinson. Traverse the next sibling tree to find the text fragment for a first letter. https://bugs.webkit.org/show_bug.cgi?id=54568 We cannot assume that the next sibling to the first letter will a text fragment since there can be intermediatary Apple-style-span inline elements wrapping the text fragment. So, we traverse the next sibling tree to find it. Test: fast/css/first-letter-text-fragment-crash.html * rendering/RenderBlock.cpp: (WebCore::RenderBlock::updateFirstLetter): 2011-02-16 Martin Robinson Reviewed by Xan Lopez. [GTK] r78718 introduced some assertion failures in some HTTP tests https://bugs.webkit.org/show_bug.cgi?id=54592 No new tests. This fix is covered by tests that are currently failing. * platform/network/soup/ResourceRequestSoup.cpp: (WebCore::ResourceRequest::updateFromSoupMessage): Instead of setting the existing headers and then selectively removing ones that do not exist in the updated soup message, just remove all headers from the map first. 2011-02-16 Jian Li Reviewed by Kenneth Russell. [V8] DataView constructor can be applied as a regular method https://bugs.webkit.org/show_bug.cgi?id=54563 Tested by adding a new test case to fast/canvas/webgl/data-view-test.html. * bindings/v8/custom/V8DataViewCustom.cpp: (WebCore::V8DataView::constructorCallback): == Rolled over to ChangeLog-2011-02-16 ==