1. 18 Apr, 2013 8 commits
    • andersca@apple.com's avatar
      Change storage factory functions to take a PageGroup and Page respectively · e44d33f5
      andersca@apple.com authored
      https://bugs.webkit.org/show_bug.cgi?id=114776
      
      Reviewed by Beth Dakin.
      
      Source/WebCore:
      
      Change StorageNamespace::localStorageNamespace to take a PageGroup since WebKit2 needs
      to know which local storage namespace belongs to which page group. Also remove the quota parameter from
      the sessionStorageNamespace function since that's trivial to get from the Page.
      
      * WebCore.exp.in:
      * page/Page.cpp:
      (WebCore::Page::sessionStorage):
      * page/PageGroup.cpp:
      (WebCore::PageGroup::localStorage):
      * storage/StorageNamespace.cpp:
      (WebCore::StorageNamespace::localStorageNamespace):
      (WebCore::StorageNamespace::sessionStorageNamespace):
      * storage/StorageNamespace.h:
      * storage/StorageNamespaceImpl.cpp:
      (WebCore::StorageNamespaceImpl::localStorageNamespace):
      (WebCore::StorageNamespaceImpl::sessionStorageNamespace):
      * storage/StorageNamespaceImpl.h:
      * storage/StorageStrategy.cpp:
      (WebCore::StorageStrategy::localStorageNamespace):
      (WebCore::StorageStrategy::sessionStorageNamespace):
      * storage/StorageStrategy.h:
      
      Source/WebKit2:
      
      Update for WebCore changes.
      
      * WebProcess/WebCoreSupport/WebPlatformStrategies.cpp:
      (WebKit::WebPlatformStrategies::localStorageNamespace):
      (WebKit::WebPlatformStrategies::sessionStorageNamespace):
      * WebProcess/WebCoreSupport/WebPlatformStrategies.h:
      (WebPlatformStrategies):
      
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@148683 268f45cc-cd09-0410-ab3c-d52691b4dbfc
      e44d33f5
    • mrobinson@webkit.org's avatar
      [GTK] fast/canvas/DrawImageSinglePixelStretch.html fails · b312a845
      mrobinson@webkit.org authored
      https://bugs.webkit.org/show_bug.cgi?id=58309
      
      Reviewed by Alejandro G. Castro.
      
      Source/WebCore:
      
      No new tests. This patch unskips a test.
      
      Prevent sampling outside source boundaries, by creating subsurfaces from source surfaces.
      This also requires careful handling of negative and floating source rectangles.
      
      * platform/graphics/cairo/PlatformContextCairo.cpp:
      (WebCore::PlatformContextCairo::drawSurfaceToContext): Use a subsurface to prevent sampling
      outside rectangle boundaries.
      
      LayoutTests:
      
      * platform/gtk/TestExpectations: Unskip a test which is now passing.
      
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@148681 268f45cc-cd09-0410-ab3c-d52691b4dbfc
      b312a845
    • commit-queue@webkit.org's avatar
      [BlackBerry] Improper initialization of ANGLEResources (resubmission) · 34165f2b
      commit-queue@webkit.org authored
      https://bugs.webkit.org/show_bug.cgi?id=114814
      
      Patch by Jonathan Feldstein <jfeldstein@blackberry.com> on 2013-04-18
      Reviewed by Yong Li, Rob Buis
      Internally reviewed by Maxim Mogilnitsky
      
      MaxDrawBuffers, OES_standard_derivatives, OES_EGL_image_external and ARB_texture_rectangle
      are initialized through the ShBuiltInResources function so these fields do not need to be
      set again in GraphicsContext3DBlackBerry.cpp. In addition, the extension flags should not
      be set to true without getExtension being called (Khronos WebGL specs, section 5.14.14.).
      In fact, as a direct result of these extensions being enabled prior to calling
      getExtension, a WebGL conformance suite test for GL_OES_standard_derivatives was failing.
      
      * platform/graphics/blackberry/GraphicsContext3DBlackBerry.cpp:
      (WebCore::GraphicsContext3D::GraphicsContext3D):
      
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@148680 268f45cc-cd09-0410-ab3c-d52691b4dbfc
      34165f2b
    • commit-queue@webkit.org's avatar
      remove build warning(unused parameter) · e81c988b
      commit-queue@webkit.org authored
      https://bugs.webkit.org/show_bug.cgi?id=114670
      
      Patch by Xuefei Ren <xren@blackberry.com> on 2013-04-18
      Reviewed by Rob Buis.
      
      Source/JavaScriptCore:
      
      remove warning in Source/JavaScriptCore/runtime/GCActivityCallbackBlackBerry.cpp
      
      * runtime/GCActivityCallbackBlackBerry.cpp:
      (JSC::DefaultGCActivityCallback::didAllocate):
      
      Source/WebCore:
      
      remove warning(unused parameter) in
      Source/WebCore/html/shadow/MediaControlsBlackBerry.cpp
      Source/WebCore/loader/blackberry/CookieJarBlackBerry.cpp
      Source/WebCore/platform/blackberry/AuthenticationChallengeManager.cpp
      Source/WebCore/platform/blackberry/RenderThemeBlackBerry.cpp
      Source/WebCore/platform/network/blackberry/ResourceResponseBlackBerry.cpp
      Source/WebCore/platform/network/blackberry/SocketStreamHandleBlackBerry.cpp
      Source/WebCore/plugins/blackberry/PluginViewBlackBerry.cpp
      
      * html/shadow/MediaControlsBlackBerry.cpp:
      (WebCore::MediaControlFullscreenFullscreenButtonElement::setIsFullscreen):
      * loader/blackberry/CookieJarBlackBerry.cpp:
      (WebCore::getRawCookies):
      (WebCore::deleteCookie):
      * platform/blackberry/AuthenticationChallengeManager.cpp:
      (WebCore::AuthenticationChallengeManager::notifyChallengeResult):
      * platform/blackberry/RenderThemeBlackBerry.cpp:
      (WebCore::RenderTheme::themeForPage):
      * platform/network/blackberry/ResourceResponseBlackBerry.cpp:
      (WebCore::ResourceResponse::doPlatformAdopt):
      * platform/network/blackberry/SocketStreamHandleBlackBerry.cpp:
      (WebCore::SocketStreamHandle::notifyStatusReceived):
      * plugins/blackberry/PluginViewBlackBerry.cpp:
      (WebCore::PluginView::invalidateRegion):
      
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@148678 268f45cc-cd09-0410-ab3c-d52691b4dbfc
      e81c988b
    • commit-queue@webkit.org's avatar
      remove build warning(unused parameter) · cbe6e046
      commit-queue@webkit.org authored
      https://bugs.webkit.org/show_bug.cgi?id=114670
      
      Patch by Xuefei Ren <xren@blackberry.com> on 2013-04-18
      Reviewed by Rob Buis.
      
      Source/WebCore:
      
      remove builde warning (unused parameter) in
      Source/WebCore/platform/blackberry/CursorBlackBerry.cpp
      
      * platform/blackberry/CursorBlackBerry.cpp:
      (WebCore::Cursor::Cursor):
      
      Source/WebKit/blackberry:
      
      remove builde warning (unused parameter) in
      Source/WebKit/blackberry
      
      * WebKitSupport/InputHandler.cpp:
      (BlackBerry::WebKit::InputHandler::spellCheckingRequestCancelled):
      (BlackBerry::WebKit::InputHandler::spellCheckingRequestProcessed):
      (BlackBerry::WebKit::InputHandler::spannableTextInRange):
      * WebKitSupport/RenderQueue.cpp:
      (BlackBerry::WebKit::RenderQueue::backingStoreRectChanging):
      (BlackBerry::WebKit::RenderQueue::scrollZoomJobsCompleted):
      * WebKitSupport/SurfacePool.cpp:
      (BlackBerry::WebKit::SurfacePool::destroyPlatformGraphicsContext):
      (BlackBerry::WebKit::SurfacePool::waitForBuffer):
      (BlackBerry::WebKit::SurfacePool::notifyBuffersComposited):
      (BlackBerry::WebKit::SurfacePool::destroyPlatformSync):
      
      Tools:
      
      remove builde warning (unused parameter) in
      Tools/DumpRenderTree/blackberry
      
      * DumpRenderTree/blackberry/AccessibilityUIElementBlackBerry.cpp:
      (AccessibilityUIElement::scrollToMakeVisibleWithSubFocus):
      (AccessibilityUIElement::scrollToGlobalPoint):
      * DumpRenderTree/blackberry/DumpRenderTree.cpp:
      (BlackBerry::WebKit::DumpRenderTree::addMessageToConsole):
      (BlackBerry::WebKit::DumpRenderTree::didReceiveResponseForFrame):
      * DumpRenderTree/blackberry/EventSender.cpp:
      (getDragModeCallback):
      (setDragModeCallback):
      (mouseWheelToCallback):
      (contextClickCallback):
      (mouseDownCallback):
      (mouseUpCallback):
      (mouseMoveToCallback):
      (beginDragWithFilesCallback):
      (leapForwardCallback):
      (keyDownCallback):
      (textZoomInCallback):
      (textZoomOutCallback):
      (zoomPageInCallback):
      (zoomPageOutCallback):
      (addTouchPointCallback):
      (updateTouchPointCallback):
      (setTouchModifierCallback):
      (touchStartCallback):
      (touchCancelCallback):
      (touchMoveCallback):
      (touchEndCallback):
      (clearTouchPointsCallback):
      (cancelTouchPointCallback):
      (releaseTouchPointCallback):
      (scalePageByCallback):
      * DumpRenderTree/blackberry/TestRunnerBlackBerry.cpp:
      (TestRunner::authenticateSession):
      (TestRunner::setSpatialNavigationEnabled):
      (TestRunner::apiTestNewWindowDataLoadBaseURL):
      (TestRunner::setPluginsEnabled):
      (TestRunner::setApplicationCacheOriginQuota):
      (TestRunner::addMockSpeechInputResult):
      (TestRunner::setViewModeMediaFeature):
      (TestRunner::deleteLocalStorageForOrigin):
      (TestRunner::applicationCacheDiskUsageForOrigin):
      (TestRunner::grantWebNotificationPermission):
      (TestRunner::denyWebNotificationPermission):
      (TestRunner::simulateWebNotificationClick):
      (TestRunner::simulateLegacyWebNotificationClick):
      
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@148677 268f45cc-cd09-0410-ab3c-d52691b4dbfc
      cbe6e046
    • akling@apple.com's avatar
      EditingStyle: Avoid some unnecessary CSSStyleDeclaration wrappers. · 66eaa226
      akling@apple.com authored
      <http://webkit.org/b/114763>
      
      Reviewed by Antti Koivisto.
      
      removeEquivalentProperties(CSSStyleDeclaration) and removeEquivalentProperties(StylePropertySet)
      only have different behavior if the CSSStyleDeclaration is computed style (they differ in handling
      of the 'font-size' property.)
      
      Avoid creating a CSSStyleDeclaration for StylePropertySets where we can just pass them in directly.
      
      * editing/EditingStyle.cpp:
      (WebCore::EditingStyle::removeStyleAddedByNode):
      (WebCore::EditingStyle::removeStyleConflictingWithStyleOfNode):
      
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@148675 268f45cc-cd09-0410-ab3c-d52691b4dbfc
      66eaa226
    • allan.jensen@digia.com's avatar
      Unset :hover in inner documents · 9ba864fa
      allan.jensen@digia.com authored
      https://bugs.webkit.org/show_bug.cgi?id=114446
      
      Reviewed by Antonio Gomes.
      
      Source/WebCore:
      
      Fixes a regression from r145126 where hover nodes in inner documents was sometimes
      not unset. Additionally it uses the new api from r145126 to avoid an unnecessary
      hit test in touch-event handling.
      
      Test: fast/events/touch/frame-hover-update.html
      
      * dom/Document.cpp:
      (WebCore::Document::updateHoverActiveState):
      * page/EventHandler.cpp:
      (WebCore::shouldGesturesTriggerActive):
      (WebCore::EventHandler::handleTouchEvent):
      
      LayoutTests:
      
      Test expected effect of touch events on hover state.
      
      * fast/events/touch/frame-hover-update-expected.txt: Added.
      * fast/events/touch/frame-hover-update.html: Added.
      
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@148672 268f45cc-cd09-0410-ab3c-d52691b4dbfc
      9ba864fa
    • commit-queue@webkit.org's avatar
      Web Inspector: Toolbar icons are displayed incorrectly · c1c1b324
      commit-queue@webkit.org authored
      https://bugs.webkit.org/show_bug.cgi?id=114792
      
      Patch by Seokju Kwon <seokju.kwon@gmail.com> on 2013-04-18
      Reviewed by Timothy Hatcher.
      
      Inspector uses small toolbar icons when docking the inspector.
      So icons should have different x position.
      
      No tests because no behavior change is expected.
      
      * inspector/front-end/inspector.css: Add missing 'background-position-x' values.
      (body.dock-to-bottom .toolbar-item.profiles .toolbar-icon):
      (body.dock-to-bottom .toolbar-item.cpu-profiler .toolbar-icon):
      (body.dock-to-bottom .toolbar-item.css-profiler .toolbar-icon):
      (body.dock-to-bottom .toolbar-item.heap-profiler .toolbar-icon):
      (body.dock-to-bottom .toolbar-item.canvas-profiler .toolbar-icon):
      (body.dock-to-bottom .toolbar-item.memory-chart-profiler .toolbar-icon):
      (body.dock-to-bottom .toolbar-item.memory-snapshot-profiler .toolbar-icon):
      
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@148668 268f45cc-cd09-0410-ab3c-d52691b4dbfc
      c1c1b324
  2. 17 Apr, 2013 27 commits
    • commit-queue@webkit.org's avatar
      [css3-text] Rendering -webkit-hanging value for text-indent from css3-text · 1eb07872
      commit-queue@webkit.org authored
      https://bugs.webkit.org/show_bug.cgi?id=114663
      
      Patch by Jaehun Lim <ljaehun.lim@samsung.com> on 2013-04-17
      Reviewed by Beth Dakin.
      
      This patch is the rendering part to support hanging value for text-indent.
      "hanging" means "Inverts which lines are affected."
      It's prefixed and guarded by CSS3_TEXT flag.
      
      Spec: http://dev.w3.org/csswg/css-text/#text-indent
      
      Source/WebCore:
      
      Test: fast/css3-text/css3-text-indent/text-indent-each-line-hanging.html
      
      * rendering/RenderBlockLineLayout.cpp:
      (WebCore::requiresIndent): Inverted the return value when "-webkit-hanging" is applied.
      
      LayoutTests:
      
      Renamed and updated the existing testcases.
      
      * fast/css3-text/css3-text-indent/text-indent-each-line-hanging-expected.html: Renamed from LayoutTests/fast/css3-text/css3-text-indent/text-indent-each-line-expected.html
      * fast/css3-text/css3-text-indent/text-indent-each-line-hanging.html: Renamed from LayoutTests/fast/css3-text/css3-text-indent/text-indent-each-line.html.
      
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@148658 268f45cc-cd09-0410-ab3c-d52691b4dbfc
      1eb07872
    • commit-queue@webkit.org's avatar
      AX: aria-level does not override implicit level on h1, h2, etc · dafa281e
      commit-queue@webkit.org authored
      https://bugs.webkit.org/show_bug.cgi?id=114692
      
      Patch by James Craig <james@cookiecrook.com> on 2013-04-17
      Reviewed by Chris Fleizach.
      
      aria-level now works on headings without an explicit role="heading" defined. Updated existing test coverage.
      
      Source/WebCore:
      
      * accessibility/AccessibilityNodeObject.cpp:
      (WebCore::AccessibilityNodeObject::headingLevel):
      
      LayoutTests:
      
      * accessibility/heading-level-expected.txt:
      * accessibility/heading-level.html:
      
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@148654 268f45cc-cd09-0410-ab3c-d52691b4dbfc
      dafa281e
    • cfleizach@apple.com's avatar
      AX: VoiceOver says everything that isn't a link is a "clickable" in Safari reader? · b99fff1c
      cfleizach@apple.com authored
      https://bugs.webkit.org/show_bug.cgi?id=114687
      
      Reviewed by Tim Horton.
      
      Source/WebCore: 
      
      VoiceOver is saying all text is clickable, because AXPress is exposed as an action on static text.
      That is happening, because there's a click handler on the body element in this case.
      
      I think the best plan to keep existing functionality, but fix this case is not to expose
      the press action for static text when the handler is on the body element.
      
      Test: platform/mac/accessibility/press-action-not-exposed-when-body-is-click-handler.html
      
      * accessibility/AccessibilityNodeObject.cpp:
      (WebCore::AccessibilityNodeObject::mouseButtonListener):
         Change from checking getAttributeEventListener to hasEventListeners. The former only
         checks if "onclick" is installed on the element and does not work with addEventListener!
      
      * accessibility/AccessibilityObject.cpp:
      (WebCore::AccessibilityObject::isAccessibilityObjectSearchMatchAtIndex):
      * accessibility/AccessibilityObject.h:
      (WebCore::AccessibilityObject::isStaticText):
      
      LayoutTests: 
      
      * platform/mac/accessibility/press-action-not-exposed-when-body-is-click-handler-expected.txt: Added.
      * platform/mac/accessibility/press-action-not-exposed-when-body-is-click-handler.html: Added.
      
      
      
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@148652 268f45cc-cd09-0410-ab3c-d52691b4dbfc
      b99fff1c
    • simon.fraser@apple.com's avatar
      Fix GraphicsLayerCA::recursiveVisibleRectChangeRequiresFlush() to do... · 87213d57
      simon.fraser@apple.com authored
      Fix GraphicsLayerCA::recursiveVisibleRectChangeRequiresFlush() to do predictive visible rect expansion
      https://bugs.webkit.org/show_bug.cgi?id=114775
      
      Reviewed by Tim Horton.
      
      GraphicsLayerCA::recursiveVisibleRectChangeRequiresFlush() is intended to answer the question
      "if your visible rect is changed to X, would any tiles be created or destroyed?".
      
      However, for compositing layer tiled layers, we do some predictive visible rect expansion based on how
      the visible rect is changing when we actually commit visible rect changes. recursiveVisibleRectChangeRequiresFlush()
      was not doing that, causing it to give confusing answers, so fix it to do so.
      
      Both now call adjustTiledLayerVisibleRect(), and it's cleaner to make this a static function.
      
      A somewhat unrelated change is to take the layer bounds origin into account
      in GraphicsLayerCA::computeVisibleRect(). Desktop WebKit never sets this, but it's used
      on other platforms for composited scrolling, so needs to be taken into account
      when computing visible rects.
      
      * platform/graphics/ca/GraphicsLayerCA.cpp:
      (WebCore::GraphicsLayerCA::recursiveVisibleRectChangeRequiresFlush):
      (WebCore::GraphicsLayerCA::computeVisibleRect):
      (WebCore::GraphicsLayerCA::adjustTiledLayerVisibleRect):
      (WebCore::GraphicsLayerCA::updateVisibleRect):
      * platform/graphics/ca/GraphicsLayerCA.h:
      (GraphicsLayerCA):
      
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@148651 268f45cc-cd09-0410-ab3c-d52691b4dbfc
      87213d57
    • oliver@apple.com's avatar
      Automate generation of toJS function for classes that need to report extra memory usage · abe8f4be
      oliver@apple.com authored
      https://bugs.webkit.org/show_bug.cgi?id=114768
      
      Reviewed by Geoff Garen.
      
      Only really used by AudioBuffer for now.  The other classes that need it can be
      trivially refactored at a later date.
      
      * Modules/webaudio/AudioBuffer.idl:
      * bindings/js/JSAudioBufferCustom.cpp:
      * bindings/js/JSDOMBinding.h:
      (WebCore):
      (HasMemoryCost):
      (NoType):
      (BaseMixin):
      * bindings/scripts/CodeGeneratorJS.pm:
      (GenerateImplementation):
      
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@148648 268f45cc-cd09-0410-ab3c-d52691b4dbfc
      abe8f4be
    • krit@webkit.org's avatar
      BasicShapeFunctions should use RenderStyle instead of StyleResolver · 4e04d157
      krit@webkit.org authored
      https://bugs.webkit.org/show_bug.cgi?id=114743
      
      Reviewed by Antti Koivisto.
      
      BasicShapeFunctions does include RenderStyle instead of StyleResolver now.
      This is a simple refactoring patch, no new tests.
      
      * css/BasicShapeFunctions.cpp:
      (WebCore::convertToLength): Use style and rootElementStyle directly.
      (WebCore::basicShapeForValue): Ditto.
      * css/BasicShapeFunctions.h:
      (WebCore):
      * css/DeprecatedStyleBuilder.cpp:
      (WebCore::ApplyPropertyClipPath::applyValue):
      (WebCore::ApplyPropertyExclusionShape::applyValue):
      
      
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@148645 268f45cc-cd09-0410-ab3c-d52691b4dbfc
      4e04d157
    • bdakin@apple.com's avatar
      Content inside frames and scrollbars in overflow areas hit-tests incorrectly when · fd0dfc86
      bdakin@apple.com authored
      the WKView has a header
      https://bugs.webkit.org/show_bug.cgi?id=114769
      
      Reviewed by Simon Fraser.
      
      Source/WebCore: 
      
      convertToRenderer() and convertFromRenderer() need to factor in the headerHeight, 
      much like all of the conversion functions on ScrollView. 
      * page/FrameView.cpp:
      (WebCore::FrameView::convertFromRenderer):
      (WebCore::FrameView::convertToRenderer):
      
      The scrollPosition equivalent of the existing scrollOffsetRelativeToDocument() 
      function. 
      * platform/ScrollView.cpp:
      (WebCore::ScrollView::scrollPositionRelativeToDocument):
      * platform/ScrollView.h:
      (ScrollView):
      
      LayoutTests: 
      
      * platform/mac-wk2/tiled-drawing/header-and-footer-hit-testing-in-frame-expected.txt: Added.
      * platform/mac-wk2/tiled-drawing/header-and-footer-hit-testing-in-frame.html: Added.
      * platform/mac-wk2/tiled-drawing/resources/iframe-to-hit-test.html: Added.
      
      
      
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@148643 268f45cc-cd09-0410-ab3c-d52691b4dbfc
      fd0dfc86
    • akling@apple.com's avatar
      Remove FragmentScriptingPermission.h include from Element.h. · d6e5fff6
      akling@apple.com authored
      <http://webkit.org/b/114757>
      
      Rubber-stamped by Anders Carlsson.
      
      * dom/Element.h:
      
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@148642 268f45cc-cd09-0410-ab3c-d52691b4dbfc
      d6e5fff6
    • mark.lam@apple.com's avatar
      Source/JavaScriptCore: Add LLINT and baseline JIT support for timing out scripts. · dff6b22e
      mark.lam@apple.com authored
      https://bugs.webkit.org/show_bug.cgi?id=114577.
      
      Reviewed by Geoffrey Garen.
      
      Introduces the new Watchdog class which is used to track script
      execution time, and initiate script termination if needed.
      
      * API/JSContextRef.cpp:
      (internalScriptTimeoutCallback):
      (JSContextGroupSetExecutionTimeLimit):
      (JSContextGroupClearExecutionTimeLimit):
      * API/JSContextRefPrivate.h:
      - Added new script execution time limit APIs.
      * API/tests/testapi.c:
      (currentCPUTime):
      (shouldTerminateCallback):
      (cancelTerminateCallback):
      (extendTerminateCallback):
      (main):
      - Added new API tests for script execution time limit.
      * CMakeLists.txt:
      * GNUmakefile.list.am:
      * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.vcproj:
      * JavaScriptCore.vcxproj/JavaScriptCore.vcxproj:
      * JavaScriptCore.vcxproj/JavaScriptCore.vcxproj.filters:
      * JavaScriptCore.xcodeproj/project.pbxproj:
      * Target.pri:
      * bytecompiler/BytecodeGenerator.cpp:
      (JSC::BytecodeGenerator::emitLoopHint):
      - loop hints are needed for the llint as well. Hence, it will be
        emitted unconditionally.
      * interpreter/Interpreter.cpp:
      (JSC::Interpreter::addStackTraceIfNecessary):
      (JSC::Interpreter::throwException):
      (JSC::Interpreter::execute):
      (JSC::Interpreter::executeCall):
      (JSC::Interpreter::executeConstruct):
      - Added checks for script termination before entering script code.
      * jit/JIT.cpp:
      (JSC::JIT::emitWatchdogTimerCheck):
      * jit/JIT.h:
      (JSC::JIT::emit_op_loop_hint):
      * jit/JITStubs.cpp:
      (JSC::DEFINE_STUB_FUNCTION(void, handle_watchdog_timer)):
      * jit/JITStubs.h:
      * llint/LLIntExceptions.cpp:
      (JSC::LLInt::doThrow):
      - Factored out some common code from returnToThrow() and callToThrow().
      (JSC::LLInt::returnToThrow):
      (JSC::LLInt::callToThrow):
      * llint/LLIntSlowPaths.cpp:
      (JSC::LLInt::LLINT_SLOW_PATH_DECL(slow_path_handle_watchdog_timer)):
      * llint/LLIntSlowPaths.h:
      * llint/LowLevelInterpreter.asm:
      * llint/LowLevelInterpreter32_64.asm:
      * llint/LowLevelInterpreter64.asm:
      * runtime/ExceptionHelpers.cpp:
      (JSC::throwTerminatedExecutionException):
      - Also removed the now unused InterruptedExecutionException.
      * runtime/ExceptionHelpers.h:
      * runtime/JSGlobalData.cpp:
      (JSC::JSGlobalData::JSGlobalData):
      * runtime/JSGlobalData.h:
      - Added watchdog, and removed the now obsolete Terminator.
      * runtime/Terminator.h: Removed.
      * runtime/Watchdog.cpp: Added.
      (JSC::Watchdog::Watchdog):
      (JSC::Watchdog::~Watchdog):
      (JSC::Watchdog::setTimeLimit):
      (JSC::Watchdog::didFire):
      (JSC::Watchdog::isEnabled):
      (JSC::Watchdog::fire):
      (JSC::Watchdog::arm):
      (JSC::Watchdog::disarm):
      (JSC::Watchdog::startCountdownIfNeeded):
      (JSC::Watchdog::startCountdown):
      (JSC::Watchdog::stopCountdown):
      (JSC::Watchdog::Scope::Scope):
      (JSC::Watchdog::Scope::~Scope):
      * runtime/Watchdog.h: Added.
      (Watchdog):
      (JSC::Watchdog::didFire):
      (JSC::Watchdog::timerDidFireAddress):
      (JSC::Watchdog::isArmed):
      (Watchdog::Scope):
      * runtime/WatchdogMac.cpp: Added.
      (JSC::Watchdog::initTimer):
      (JSC::Watchdog::destroyTimer):
      (JSC::Watchdog::startTimer):
      (JSC::Watchdog::stopTimer):
      * runtime/WatchdogNone.cpp: Added.
      (JSC::Watchdog::initTimer):
      (JSC::Watchdog::destroyTimer):
      (JSC::Watchdog::startTimer):
      (JSC::Watchdog::stopTimer):
      
      Source/WebCore: Add LLINT and baseline JIT support for timing out scripts.
      https://bugs.webkit.org/show_bug.cgi?id=114577.
      
      Reviewed by Geoffrey Garen.
      
      Replaced use of the obsolete JSGlobalData.terminator methods with the
      JSGlobalData.watchdog equivalents.
      
      * bindings/js/JSEventListener.cpp:
      (WebCore::JSEventListener::handleEvent):
      * bindings/js/SerializedScriptValue.cpp:
      (WebCore::SerializedScriptValue::maybeThrowExceptionIfSerializationFailed):
      * bindings/js/WorkerScriptController.cpp:
      (WebCore::WorkerScriptController::evaluate):
      (WebCore::WorkerScriptController::scheduleExecutionTermination):
      (WebCore::WorkerScriptController::isExecutionTerminating):
      
      Source/WTF: Added currentCPUTime() and currentCPUTimeMS().
      https://bugs.webkit.org/show_bug.cgi?id=114577.
      
      Reviewed by Geoffrey Garen.
      
      The currentCPUTime() implementation came from the old TimeoutChecker.cpp.
      
      * wtf/CurrentTime.cpp:
      (WTF::currentCPUTime):
      (WTF::currentCPUTimeMS):
      * wtf/CurrentTime.h:
      
      
      
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@148639 268f45cc-cd09-0410-ab3c-d52691b4dbfc
      dff6b22e
    • jer.noble@apple.com's avatar
      Crash in WebCore::HTMLMediaElement::~HTMLMediaElement. · 899e1053
      jer.noble@apple.com authored
      https://bugs.webkit.org/show_bug.cgi?id=113531
      
      Reviewed by Eric Carlson.
      
      No new tests, though this is intermittently reproducible with
      http/tests/misc/delete-frame-during-readystatechange.html under ASAN.
      
      * html/HTMLMediaElement.cpp:
      (WebCore::HTMLMediaElement::~HTMLMediaElement): Clear the media player manually
          before the destructor exits. Clearing the media player may cancel a resource load,
          which can trigger a readystatechange event. It's possible for the HTMLMediaElement
          to attempt to fire an abort event within the readystatechange event, even though it is
          now in an inconsistent state. Clearling the media player before finishing the destructor
          ensures that the HTMLMediaElement will at least still be alive if this case is triggered.
          Set m_completelyLoaded to true to ensure that if userCancelledLoad() is called, it doesn't
          attempt to fire events while destructing.
      
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@148636 268f45cc-cd09-0410-ab3c-d52691b4dbfc
      899e1053
    • commit-queue@webkit.org's avatar
      Web Inspector: make generate-inspector-protocol-version work with python3 · bd95c851
      commit-queue@webkit.org authored
      https://bugs.webkit.org/show_bug.cgi?id=114717
      
      Revision r146765 added print() calls that made the script complain about
      invalid syntax when using python3.
      
      This commit replaces such calls with calls to sys.stdout.write(), analogous
      to the sys.stderr.write() ones already used throughout the file.
      
      Patch by Sergio Correia <sergio.correia@openbossa.org> on 2013-04-17
      Reviewed by Timothy Hatcher.
      
      No new tests. No user visible behavior changed.
      
      * inspector/generate-inspector-protocol-version:
      (main):
      
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@148630 268f45cc-cd09-0410-ab3c-d52691b4dbfc
      bd95c851
    • commit-queue@webkit.org's avatar
      Web Inspector: Fix Localized string warngs · cc348d4f
      commit-queue@webkit.org authored
      https://bugs.webkit.org/show_bug.cgi?id=114718
      
      Patch by Seokju Kwon <seokju.kwon@gmail.com> on 2013-04-17
      Reviewed by Timothy Hatcher.
      
      No tests because no behavior change is expected.
      
      * English.lproj/localizedStrings.js:
      
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@148628 268f45cc-cd09-0410-ab3c-d52691b4dbfc
      cc348d4f
    • ossy@webkit.org's avatar
      [Qt] Unreviewed Windows buildfix. · 77d5d31b
      ossy@webkit.org authored
      * platform/qt/DragDataQt.cpp: Add a missing include.
      
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@148626 268f45cc-cd09-0410-ab3c-d52691b4dbfc
      77d5d31b
    • commit-queue@webkit.org's avatar
      Breaking Float: floated block level element following inline element in... · 17761407
      commit-queue@webkit.org authored
      Breaking Float: floated block level element following inline element in floated container breaks to next line
      https://bugs.webkit.org/show_bug.cgi?id=45274
      
      Patch by ChangSeok Oh <shivamidow@gmail.com> on 2013-04-17
      Reviewed by David Hyatt.
      
      Source/WebCore:
      
      Fix position issue of floating element in floating element.
      Inner floating element has placed at next line when outer floating element has text,
      even though previous line has spaces enough to fit it.
      To solve this, the width of a space added temporarily for prohibiting duplication more than 2 empty spaces
      is subtracted when floating element is checked whether it fits on a line.
      
      Tests: css2.1/20110323/floats-001.html
             css2.1/20110323/floats-102.html
             fast/inline-block/float-both-whitespace.html
             fast/inline-block/float-leading-whitespace.html
             fast/inline-block/float-no-whitespace.html
             fast/inline-block/float-trailing-whitespace.html
             fast/inline-block/multiple-floats-with-whitespace.html
      
      * rendering/RenderBlockLineLayout.cpp:
      (WebCore::LineWidth::LineWidth):
      (WebCore::LineWidth::fitsOnLine):
      (WebCore::LineWidth::trailingWhitespaceWidth):
      (WebCore::LineWidth::setTrailingWhitespaceWidth):
      (LineWidth):
      (WebCore::RenderBlock::LineBreaker::nextSegmentBreak):
      
      LayoutTests:
      
      * css2.1/20110323/floats-001-expected.html: Added.
      * css2.1/20110323/floats-001.html: Added.
      * css2.1/20110323/floats-102-expected.html: Added.
      * css2.1/20110323/floats-102.html: Added.
      
      Tests for whitespace around floating elements.
      Following test cases are provied by Joseph Pecoraro.
      See https://bugs.webkit.org/show_bug.cgi?id=58806
      
      * fast/exclusions/shape-outside-floats/shape-outside-floats-stacked.html: Updated to match new treatment of trailing space.
      * fast/inline-block/float-both-whitespace-expected.png: Added.
      * fast/inline-block/float-both-whitespace-expected.txt: Added.
      * fast/inline-block/float-both-whitespace.html: Added.
      * fast/inline-block/float-leading-whitespace-expected.png: Added.
      * fast/inline-block/float-leading-whitespace-expected.txt: Added.
      * fast/inline-block/float-leading-whitespace.html: Added.
      * fast/inline-block/float-no-whitespace-expected.png: Added.
      * fast/inline-block/float-no-whitespace-expected.txt: Added.
      * fast/inline-block/float-no-whitespace.html: Added.
      * fast/inline-block/float-trailing-whitespace-expected.png: Added.
      * fast/inline-block/float-trailing-whitespace-expected.txt: Added.
      * fast/inline-block/float-trailing-whitespace.html: Added.
      * fast/inline-block/multiple-floats-with-whitespace-expected.png: Added.
      * fast/inline-block/multiple-floats-with-whitespace-expected.txt: Added.
      * fast/inline-block/multiple-floats-with-whitespace.html: Added.
      
      Update the following tests so that wrapping happens as it
      did before. With this change the float:left div progressed
      and could fit on an earlier line when we position after
      collapsing whitespace. This moved the float a line earlier
      and changed the results of the test. By adding a character
      to the line, the float won't fit and goes back on to the
      line that it was on before this change.
      
      * fast/multicol/float-truncation.html:
      * fast/multicol/vertical-lr/float-truncation.html:
      * fast/multicol/vertical-rl/float-truncation.html:
      
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@148622 268f45cc-cd09-0410-ab3c-d52691b4dbfc
      17761407
    • leoyang@rim.com's avatar
      Lots of unused parameter warnings in filesystem code · a30cf776
      leoyang@rim.com authored
      https://bugs.webkit.org/show_bug.cgi?id=114747
      
      Reviewed by Carlos Garcia Campos.
      
      Comment out or remove unused parameter identifiers.
      
      No functionalities changed, no new tests.
      
      * Modules/filesystem/DOMFileSystem.cpp: Comment out |snapshot| because
      it is referred in the comments inside the function.
      (WebCore):
      * Modules/filesystem/DOMFileSystemSync.cpp: Ditto.
      (WebCore):
      * Modules/filesystem/FileWriter.cpp: Remove unused |ec|.
      (WebCore::FileWriter::abort):
      * Modules/filesystem/FileWriterSync.cpp:
      (WebCore::FileWriterSync::didWrite): Remove unused |bytes|.
      * platform/AsyncFileSystemCallbacks.h:
      (WebCore::AsyncFileSystemCallbacks::didOpenFileSystem): Comment out unused parameters
      because the function is inlined.
      (WebCore::AsyncFileSystemCallbacks::didCreateSnapshotFile): Ditto.
      (WebCore::AsyncFileSystemCallbacks::didReadDirectoryEntry): Ditto.
      (WebCore::AsyncFileSystemCallbacks::didReadDirectoryEntries): Ditto.
      (WebCore::AsyncFileSystemCallbacks::didCreateFileWriter): Ditto.
      * platform/blackberry/WorkerAsyncFileSystemBlackBerry.cpp:
      (WebCore::WorkerAsyncFileSystemBlackBerry::createWriterOnMainThread): Remove unused |client|.
      
      
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@148621 268f45cc-cd09-0410-ab3c-d52691b4dbfc
      a30cf776
    • bfulgham@webkit.org's avatar
      Source/JavaScriptCore: [Windows, WinCairo] Stop individually building WTF files in JSC. · 92d7c3f2
      bfulgham@webkit.org authored
      https://bugs.webkit.org/show_bug.cgi?id=114705
      
      Reviewed by Anders Carlsson.
      
      * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCoreExports.def:
      Export additional String/fastMalloc symbols needed by JSC program.
      * JavaScriptCore.vcproj/jsc/jsc.vcproj: Don't manually build
      WTF implementation files (a second time!) in this project.
      * JavaScriptCore.vcxproj/JavaScriptCoreExportGenerator/JavaScriptCoreExports.def.in:
      Export additional String/fastMalloc symbols needed by JSC program.
      * JavaScriptCore.vcxproj/jsc/jsc.vcxproj: Don't manually
      build WTF implementation files (a second time!) in this project.
      * JavaScriptCore.vcxproj/jsc/jsc.vcxproj.filters: Ditto.
      
      Source/WebCore: [Windows, WinCairo] Stop individually building WTF files in WebCore
      https://bugs.webkit.org/show_bug.cgi?id=114705
      
      Reviewed by Anders Carlsson.
      
      * WebCore.vcproj/WebCore.vcproj: Remove references to WTF objects.
      * WebCore.vcxproj/WebCore.vcxproj: Ditto.
      * WebCore.vcxproj/WebCore.vcxproj.filters: Ditto
      
      
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@148620 268f45cc-cd09-0410-ab3c-d52691b4dbfc
      92d7c3f2
    • krit@webkit.org's avatar
      Make lengthConversion methods and arguments const · af0fcb80
      krit@webkit.org authored
      https://bugs.webkit.org/show_bug.cgi?id=114749
      
      Reviewed by Andreas Kling.
      
      Refactoring, no new tests.
      
      * css/CSSCalculationValue.cpp:
      (WebCore::CSSCalcValue::computeLengthPx):
      (WebCore::CSSCalcPrimitiveValue::toCalcValue):
      (WebCore::CSSCalcPrimitiveValue::computeLengthPx):
      (WebCore::CSSCalcBinaryOperation::toCalcValue):
      (WebCore::CSSCalcBinaryOperation::computeLengthPx):
      * css/CSSCalculationValue.h:
      (WebCore::CSSCalcValue::toCalcValue):
      (CSSCalcValue):
      * css/CSSPrimitiveValue.cpp:
      (WebCore::CSSPrimitiveValue::computeLength):
      (WebCore::CSSPrimitiveValue::computeLengthDouble):
      (WebCore::CSSPrimitiveValue::viewportPercentageLength):
      * css/CSSPrimitiveValue.h:
      (CSSPrimitiveValue):
      * css/CSSPrimitiveValueMappings.h:
      (WebCore::CSSPrimitiveValue::convertToLength):
      * css/StyleResolver.cpp:
      (WebCore::StyleResolver::convertToIntLength):
      (WebCore::StyleResolver::convertToFloatLength):
      * css/StyleResolver.h:
      (StyleResolver):
      
      
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@148619 268f45cc-cd09-0410-ab3c-d52691b4dbfc
      af0fcb80
    • bruno.abinader@basyskom.com's avatar
      [refactor] Moved ScriptedAnimationController common code to inline function · bd448dc8
      bruno.abinader@basyskom.com authored
      https://bugs.webkit.org/show_bug.cgi?id=114681
      
      Reviewed by Daniel Bates.
      
      Avoided duplicated code by moving ScriptedAnimationController clearance
      code to a common function used by both Document::dispose() and
      Document::detach().
      
      No new tests, no behavior changes.
      
      * dom/Document.cpp:
      (WebCore::Document::dispose):
      (WebCore::Document::detach):
      (WebCore::Document::clearScriptedAnimationController): Added.
      * dom/Document.h:
      (Document):
      
      
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@148618 268f45cc-cd09-0410-ab3c-d52691b4dbfc
      bd448dc8
    • cfleizach@apple.com's avatar
      Source/WebCore: AX: When img@alt is undefined, WebKit should use @title as... · 57796844
      cfleizach@apple.com authored
      Source/WebCore: AX: When img@alt is undefined, WebKit should use @title as accessibility label if available
      https://bugs.webkit.org/show_bug.cgi?id=114535
      
      Reviewed by Tim Horton.
      
      Don't hide images from Accessibility that have the title attribute on them.
      
      Test: accessibility/empty-image-with-title.html
      
      * accessibility/AccessibilityRenderObject.cpp:
      (WebCore::AccessibilityRenderObject::computeAccessibilityIsIgnored):
      
      LayoutTests: When img@alt is undefined, WebKit should use @title as accessibility label if available
      https://bugs.webkit.org/show_bug.cgi?id=114535
      
      Reviewed by Tim Horton.
      
      * accessibility/empty-image-with-title-expected.txt: Added.
      * accessibility/empty-image-with-title.html: Added.
      
      
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@148617 268f45cc-cd09-0410-ab3c-d52691b4dbfc
      57796844
    • commit-queue@webkit.org's avatar
      getAttribute does not behave correctly for mixed-case attributes on HTML elements · 7727e793
      commit-queue@webkit.org authored
      https://bugs.webkit.org/show_bug.cgi?id=105713
      
      Patch by Arpita Bahuguna <a.bah@samsung.com> on 2013-04-17
      Reviewed by Andreas Kling.
      
      Source/WebCore:
      
      getAttribute() and getAttributeNode() APIs do not convert the
      passed attribute name to lowercase before comparing against the
      existing attributes.
      The specification however states that the passed name should
      be converted to ASCII lowercase before checking for the existence
      of the given attribute. [www.w3.org/TR/domcore/#dom-element-getattribute]
      
      Test: fast/dom/Element/getAttribute-case-insensitivity.html
      
      * dom/Element.h:
      (WebCore::ElementData::getAttributeItemIndex):
      getAttributeItemIndex() accepts a bool param 'shouldIgnoreAttributeCase'
      which specifies whether or not the attribute's case should be ignored
      before comparison but we don't really convert the passed name to lowercase
      before carrying out the comparison.
      
      Thus, when called from APIs such as getAttribute() and getAttributeNode()
      which do not explicitally convert the attribute name to lowercase
      before calling on this method, it fails to carry out a case-insensitive
      search.
      
      Have thus made changes to convert the passed attribute's name to
      lowercase if 'shouldIgnoreAttributeCase' is true.
      
      LayoutTests:
      
      * fast/dom/Element/getAttribute-case-insensitivity-expected.txt: Added.
      * fast/dom/Element/getAttribute-case-insensitivity.html: Added.
      Layout test added for verifying that getAttribute() and getAttributeNode()
      APIs convert the passed attribute name to lowercase before comparing
      against the existing attributes.
      
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@148614 268f45cc-cd09-0410-ab3c-d52691b4dbfc
      7727e793
    • commit-queue@webkit.org's avatar
      [BlackBerry] Add support for filesystem: URLs to BlackBerry Media Player. · 44432562
      commit-queue@webkit.org authored
      https://bugs.webkit.org/show_bug.cgi?id=114686
      https://przilla.ott.qnx.com/bugzilla/show_bug.cgi?id=314865
      
      Patch by John Griggs <jgriggs@blackberry.com> on 2013-04-17
      Reviewed by Rob Buis.
      
      Translate filesystem: URLs to file:// URLs for use by the media player, but only after the filesystem: URL has been checked for security, etc.
      
      * platform/graphics/blackberry/MediaPlayerPrivateBlackBerry.cpp:
      (WebCore::MediaPlayerPrivate::load):
      (WebCore::MediaPlayerPrivate::onError):
      (WebCore::MediaPlayerPrivate::onDurationChanged):
      (WebCore::MediaPlayerPrivate::notifyChallengeResult):
      
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@148613 268f45cc-cd09-0410-ab3c-d52691b4dbfc
      44432562
    • commit-queue@webkit.org's avatar
      [GStreamer] Eclipse warnings in MediaPlayerPrivateGStreamer · c74d372a
      commit-queue@webkit.org authored
      https://bugs.webkit.org/show_bug.cgi?id=114654
      
      Patch by Brendan Long <b.long@cablelabs.com> on 2013-04-17
      Reviewed by Philippe Normand.
      
      No new tests (nothing added).
      
      * platform/graphics/gstreamer/MediaPlayerPrivateGStreamer.cpp:
      (WebCore::MediaPlayerPrivateGStreamer::MediaPlayerPrivateGStreamer):
      Initialize m_volumeAndMuteInitialized
      * platform/graphics/gstreamer/MediaPlayerPrivateGStreamerBase.cpp:
      (WebCore::MediaPlayerPrivateGStreamerBase::MediaPlayerPrivateGStreamerBase):
      Initialize signal handlers to 0.
      
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@148611 268f45cc-cd09-0410-ab3c-d52691b4dbfc
      c74d372a
    • abucur@adobe.com's avatar
      [CSS Regions] Remove m_flowThread from NodeRenderingContext · 4cdc00a5
      abucur@adobe.com authored
      https://bugs.webkit.org/show_bug.cgi?id=114732
      
      Reviewed by Antti Koivisto.
      
      Cleanup NodeRenderingContext. Remove unused member m_flowThread.
      
      Tests: None needed.
      
      * dom/NodeRenderingContext.cpp:
      (WebCore::NodeRenderingContext::moveToFlowThreadIfNeeded):
      * dom/NodeRenderingContext.h: Removed m_flowThread.
      
      
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@148605 268f45cc-cd09-0410-ab3c-d52691b4dbfc
      4cdc00a5
    • allan.jensen@digia.com's avatar
      [Qt] MediaPlayerPrivateQt.cpp doesn't build in debug · 84bb53e6
      allan.jensen@digia.com authored
      Unreviewed build fix.
      
      We need to include Logging.h to get the LOG definitions.
      
      * platform/graphics/qt/MediaPlayerPrivateQt.cpp:
      
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@148600 268f45cc-cd09-0410-ab3c-d52691b4dbfc
      84bb53e6
    • csaavedra@igalia.com's avatar
      execCommand("RemoveFormat") might remove format after the selection · b2fa9322
      csaavedra@igalia.com authored
      https://bugs.webkit.org/show_bug.cgi?id=112240
      
      Reviewed by Ryosuke Niwa.
      
      Source/WebCore:
      
      Tests: editing/execCommand/remove-format-multiple-elements-mac.html
      
      This bug is hit when ApplyStyleCommand is used to change the
      style and the current selection ends in the beginning of a new node.
      The bug is actually a two-fold thing:
      
      1. There was no check as to whether the end node is really
      selected or not, and format was always removed from it with
      pushDownInlineStyleAroundNode(). An equivalent check for the start
      node was already in place, so fix it analogously.
      
      2. Previous stage might change the dom tree, resulting in a render
      tree that is not up-to-date. Position::upstream() is later used
      and, in order to be able to find a visually equivalent position in
      a text node, this method needs the render tree to be up-to-date,
      therefore, a call to updateLayoutIgnorePendingStylesheets() is
      necessary.
      
      * editing/ApplyStyleCommand.cpp:
      (WebCore::ApplyStyleCommand::removeInlineStyle): Make sure that no
      format is removed from the end node if it's not fully selected.
      (WebCore::ApplyStyleCommand::nodeFullySelected): Call updateLayoutIgnorePendingStylesheets()
      
      LayoutTests:
      
      * editing/execCommand/remove-format-multiple-elements-mac-expected.txt: Updated.
      * editing/execCommand/script-tests/remove-format-multiple-elements-mac.js:
      (selectFirstLine): Add this method to check that RemoveFormat works when
      a whole line is selected.
      
      
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@148597 268f45cc-cd09-0410-ab3c-d52691b4dbfc
      b2fa9322
    • commit-queue@webkit.org's avatar
      [BlackBerry] GraphicsContext3D: don't initialize m_extensions in the constructor · 39354ff2
      commit-queue@webkit.org authored
      https://bugs.webkit.org/show_bug.cgi?id=114726
      
      Patch by Alberto Garcia <agarcia@igalia.com> on 2013-04-17
      Reviewed by Carlos Garcia Campos.
      
      m_extensions is now initialized on demand by
      GraphicsContext3D::getExtensions().
      
      Doing it in the constructor produces a crash, because
      Extensions3DOpenGLES() calls glGetString before the WebGL context
      is current.
      
      * platform/graphics/blackberry/GraphicsContext3DBlackBerry.cpp:
      (WebCore::GraphicsContext3D::GraphicsContext3D):
      
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@148595 268f45cc-cd09-0410-ab3c-d52691b4dbfc
      39354ff2
    • mjs@apple.com's avatar
      Replace JSC-specific IDL extended attributes with generic (JSC+V8) ones, now... · bfd58303
      mjs@apple.com authored
      Replace JSC-specific IDL extended attributes with generic (JSC+V8) ones, now that the distinction no longer matters
      https://bugs.webkit.org/show_bug.cgi?id=114712
      
      Reviewed by Dan Bernstein.
      
      No behavior change expected.
      
      * bindings/scripts/IDLAttributes.txt: Remove the JSFoo attributes
      that have bare Foo equivalents.
      * bindings/scripts/CodeGeneratorJS.pm: Remove support for JSFoo
      aliases.
      (GetGenerateIsReachable):
      (GetCustomIsReachable):
      (ShouldGenerateToJSDeclaration):
      (ShouldGenerateToJSImplementation):
      (HasCustomConstructor):
      (HasCustomGetter):
      (HasCustomSetter):
      (HasCustomMethod):
      
      Replace JSFoo attributes with equivalen Foo attributs in all files
      below:
              
      * Modules/geolocation/Geolocation.idl:
      * Modules/indexeddb/IDBDatabase.idl:
      * Modules/indexeddb/IDBObjectStore.idl:
      * Modules/webaudio/DOMWindowWebAudio.idl:
      * Modules/websockets/DOMWindowWebSocket.idl:
      * Modules/websockets/WorkerContextWebSocket.idl:
      * css/CSSRule.idl:
      * css/CSSRuleList.idl:
      * css/CSSStyleDeclaration.idl:
      * css/CSSValue.idl:
      * css/MediaList.idl:
      * css/StyleMedia.idl:
      * css/StyleSheet.idl:
      * dom/MessagePort.idl:
      * dom/MutationObserver.idl:
      * dom/Node.idl:
      * fileapi/Blob.idl:
      * html/HTMLDocument.idl:
      * html/HTMLTemplateElement.idl:
      * html/canvas/ArrayBuffer.idl:
      * html/canvas/CanvasRenderingContext.idl:
      * html/canvas/DataView.idl:
      * html/canvas/EXTDrawBuffers.idl:
      * html/canvas/EXTTextureFilterAnisotropic.idl:
      * html/canvas/OESElementIndexUint.idl:
      * html/canvas/OESStandardDerivatives.idl:
      * html/canvas/OESTextureFloat.idl:
      * html/canvas/OESTextureHalfFloat.idl:
      * html/canvas/OESVertexArrayObject.idl:
      * html/canvas/WebGLCompressedTextureATC.idl:
      * html/canvas/WebGLCompressedTexturePVRTC.idl:
      * html/canvas/WebGLCompressedTextureS3TC.idl:
      * html/canvas/WebGLDebugRendererInfo.idl:
      * html/canvas/WebGLDebugShaders.idl:
      * html/canvas/WebGLDepthTexture.idl:
      * html/canvas/WebGLLoseContext.idl:
      * html/track/TextTrack.idl:
      * html/track/TextTrackCue.idl:
      * html/track/TextTrackList.idl:
      * loader/appcache/DOMApplicationCache.idl:
      * page/BarInfo.idl:
      * page/Console.idl:
      * page/DOMSelection.idl:
      * page/DOMWindow.idl:
      * page/History.idl:
      * page/Location.idl:
      * page/MemoryInfo.idl:
      * page/Navigator.idl:
      * page/Screen.idl:
      * page/WorkerNavigator.idl:
      * plugins/DOMMimeTypeArray.idl:
      * plugins/DOMPluginArray.idl:
      * storage/Storage.idl:
      * workers/AbstractWorker.idl:
      * workers/SharedWorker.idl:
      * workers/Worker.idl:
      * workers/WorkerContext.idl:
      * workers/WorkerLocation.idl:
      * xml/XMLHttpRequestUpload.idl:
      
      
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@148593 268f45cc-cd09-0410-ab3c-d52691b4dbfc
      bfd58303
  3. 16 Apr, 2013 5 commits
    • bdakin@apple.com's avatar
      Re-name Page::layoutMilestones() to Page::requestedLayoutMilestones() · 6209dbe1
      bdakin@apple.com authored
      https://bugs.webkit.org/show_bug.cgi?id=114713
      
      Reviewed by Simon Fraser.
      
      Source/WebCore: 
      
      * page/FrameView.cpp:
      (WebCore::FrameView::performPostLayoutTasks):
      * page/Page.cpp:
      (WebCore::Page::Page):
      (WebCore::Page::addLayoutMilestones):
      (WebCore::Page::removeLayoutMilestones):
      (WebCore::Page::isCountingRelevantRepaintedObjects):
      * page/Page.h:
      (WebCore::Page::requestedLayoutMilestones):
      (Page):
      * rendering/RenderLayerCompositor.cpp:
      (WebCore::RenderLayerCompositor::flushPendingLayerChanges):
      
      Source/WebKit/mac: 
      
      * WebView/WebView.mm:
      (-[WebView _layoutMilestones]):
      
      
      
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@148590 268f45cc-cd09-0410-ab3c-d52691b4dbfc
      6209dbe1
    • paroga@webkit.org's avatar
      [CMake] Fix dependecy calculation for generated inspector files · 78cb8680
      paroga@webkit.org authored
      https://bugs.webkit.org/show_bug.cgi?id=114092
      
      Reviewed by Brent Fulgham.
      
      At least in a generated Visual Studio solution the files generated via
      CodeGeneratorInspector.py are considered outdated all the time because
      the code generator only touches the file when the content changed.
      Add an --write_always parameter to the script to touch the file always.
      
      * CMakeLists.txt:
      * inspector/CodeGeneratorInspector.py:
      
      
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@148586 268f45cc-cd09-0410-ab3c-d52691b4dbfc
      78cb8680
    • andersca@apple.com's avatar
      Clone storage namespaces for window.open · 55fda687
      andersca@apple.com authored
      https://bugs.webkit.org/show_bug.cgi?id=114703
      
      Reviewed by Sam Weinig.
      
      Source/WebCore:
      
      Pass the new page to StorageNamespace::copy.
      
      * page/Chrome.cpp:
      (WebCore::Chrome::createWindow):
      * storage/StorageNamespace.h:
      (StorageNamespace):
      * storage/StorageNamespaceImpl.cpp:
      (WebCore::StorageNamespaceImpl::copy):
      * storage/StorageNamespaceImpl.h:
      (WebCore):
      (StorageNamespaceImpl):
      
      Source/WebKit2:
      
      * UIProcess/Storage/StorageManager.cpp:
      (WebKit::StorageManager::StorageArea::clone):
      New helper function for cloning a storage area.
      
      (WebKit::StorageManager::StorageArea::setItem):
      Correctly handle the copy-on-write feature of StorageMap if it has multiple storage areas pointing to it.
      
      (WebKit::StorageManager::StorageArea::removeItem):
      Ditto.
      
      (WebKit::StorageManager::SessionStorageNamespace::cloneTo):
      Add cloned storage areas.
      
      * WebProcess/Storage/StorageNamespaceImpl.cpp:
      (WebKit::StorageNamespaceImpl::copy):
      Create a new session storage namespace for the new page.
      
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@148585 268f45cc-cd09-0410-ab3c-d52691b4dbfc
      55fda687
    • weinig@apple.com's avatar
      Fix fallout after r148545. · 7fb9aaa3
      weinig@apple.com authored
      Source/WebCore: 
      
      * platform/text/TextChecking.h:
      Move Platform defines that were incorrectly in WebCore, into Platform.h
      
      Source/WTF: 
      
      * wtf/Platform.h:
      Move Platform defines that were incorrectly in WebCore, into Platform.h
      
      Tools: 
      
      * WebKitTestRunner/TestController.cpp:
      (WTR::TestController::resetStateToConsistentValues):
      Remove calls to non-existent API functions.
      
      
      
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@148584 268f45cc-cd09-0410-ab3c-d52691b4dbfc
      7fb9aaa3
    • hmuller@adobe.com's avatar
      [CSS Exclusions] polygon shape-inside layout fails · 242d1366
      hmuller@adobe.com authored
      https://bugs.webkit.org/show_bug.cgi?id=114402
      
      Source/WebCore:
      
      Reviewed by Dirk Schulze.
      
      The firstIncludedIntervalLogicalTop() method's implementation relied on optimistic
      assumptions about floating point accuracy which, in rare cases, caused it to discard
      first-fit locations based on the intersection of the minLogicalIntervalTop offset edge
      and a polygon offset edge. Now: we do not verify that first-fit locations based on the
      intersection of an offset edge and the minLogicalIntervalTop offset edge are below
      the horizontal minLogicalIntervalTop line. They're essentially below the line "by definition".
      
      Test: fast/exclusions/shape-inside/shape-inside-polygon-layout.html
      
      * rendering/ExclusionPolygon.cpp:
      (WebCore::ExclusionPolygon::firstIncludedIntervalLogicalTop): Avoid floating point problems
          when checking intersections with the offset edge based on minLogicalIntervalTop.
      * rendering/ExclusionPolygon.h:
      (WebCore::OffsetPolygonEdge::OffsetPolygonEdge): Initialize the basis field.
      (WebCore::OffsetPolygonEdge::basis): Report what the offset edge is "based on": a polygon
          edge, the top of the line, or a (reflex) vertex.
      (OffsetPolygonEdge): Added the Basis enum to enable tracking what the geometry of
          an offset edge is based on.
      
      LayoutTests:
      
      Verify that subsequent polygon shape-inside lines are vertically adjacent.
      
      Reviewed by Dirk Schulze.
      
      * fast/exclusions/shape-inside/shape-inside-polygon-layout-expected.txt: Added.
      * fast/exclusions/shape-inside/shape-inside-polygon-layout.html: Added.
      
      
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@148582 268f45cc-cd09-0410-ab3c-d52691b4dbfc
      242d1366