Timeline



Jan 29, 2014:

11:42 PM Changeset in webkit [163073] by commit-queue@webkit.org
  • 23 edits
    2 deletes in trunk/Source/WebCore

Unreviewed, rolling out r163048.
​http://trac.webkit.org/changeset/163048
​https://bugs.webkit.org/show_bug.cgi?id=127890

Caused many crashes, detected by EWS prior to landing
(Requested by ap on #webkit).

  • Modules/mediacontrols/MediaControlsHost.cpp:

(WebCore::MediaControlsHost::sortedTrackListForMenu):
(WebCore::MediaControlsHost::displayNameForTrack):
(WebCore::MediaControlsHost::captionDisplayMode):

  • WebCore.exp.in:
  • WebCore.vcxproj/WebCore.vcxproj:
  • WebCore.vcxproj/WebCore.vcxproj.filters:
  • WebCore.xcodeproj/project.pbxproj:
  • css/DocumentRuleSets.cpp:

(WebCore::DocumentRuleSets::initUserStyle):

  • css/InspectorCSSOMWrappers.cpp:

(WebCore::InspectorCSSOMWrappers::collectFromDocumentStyleSheetCollection):

  • dom/Document.cpp:

(WebCore::Document::setCompatibilityMode):
(WebCore::Document::registerForCaptionPreferencesChangedCallbacks):
(WebCore::Document::captionPreferencesChanged):

  • dom/DocumentStyleSheetCollection.cpp:
  • dom/DocumentStyleSheetCollection.h:
  • html/HTMLMediaElement.cpp:

(WebCore::HTMLMediaElement::HTMLMediaElement):
(WebCore::HTMLMediaElement::parseAttribute):

  • html/shadow/MediaControlElements.cpp:

(WebCore::MediaControlClosedCaptionsTrackListElement::updateDisplay):
(WebCore::MediaControlClosedCaptionsTrackListElement::rebuildTrackListMenu):
(WebCore::MediaControlTextTrackContainerElement::updateTimerFired):

  • page/CaptionStyleSheetMediaAF.cpp: Removed.
  • page/CaptionStyleSheetMediaAF.h: Removed.
  • page/CaptionUserPreferences.cpp:

(WebCore::CaptionUserPreferences::CaptionUserPreferences):
(WebCore::CaptionUserPreferences::notify):
(WebCore::CaptionUserPreferences::setCaptionDisplayMode):
(WebCore::CaptionUserPreferences::userPrefersCaptions):
(WebCore::CaptionUserPreferences::setUserPrefersCaptions):
(WebCore::CaptionUserPreferences::userPrefersSubtitles):
(WebCore::CaptionUserPreferences::setUserPrefersSubtitles):
(WebCore::CaptionUserPreferences::userPrefersTextDescriptions):
(WebCore::CaptionUserPreferences::setUserPrefersTextDescriptions):
(WebCore::CaptionUserPreferences::captionPreferencesChanged):
(WebCore::CaptionUserPreferences::textTrackSelectionScore):
(WebCore::CaptionUserPreferences::setCaptionsStyleSheetOverride):
(WebCore::CaptionUserPreferences::updateCaptionStyleSheetOveride):

  • page/CaptionUserPreferences.h:

(WebCore::CaptionUserPreferences::captionsStyleSheetOverride):
(WebCore::CaptionUserPreferences::setInterestedInCaptionPreferenceChanges):
(WebCore::CaptionUserPreferences::testingMode):
(WebCore::CaptionUserPreferences::setTestingMode):
(WebCore::CaptionUserPreferences::pageGroup):

  • page/CaptionUserPreferencesMediaAF.cpp:

(WebCore::userCaptionPreferencesChangedNotificationCallback):
(WebCore::CaptionUserPreferencesMediaAF::CaptionUserPreferencesMediaAF):
(WebCore::CaptionUserPreferencesMediaAF::userPrefersCaptions):
(WebCore::CaptionUserPreferencesMediaAF::userPrefersSubtitles):
(WebCore::CaptionUserPreferencesMediaAF::setInterestedInCaptionPreferenceChanges):
(WebCore::CaptionUserPreferencesMediaAF::captionPreferencesChanged):
(WebCore::CaptionUserPreferencesMediaAF::captionsWindowCSS):
(WebCore::CaptionUserPreferencesMediaAF::captionsBackgroundCSS):
(WebCore::CaptionUserPreferencesMediaAF::captionsTextColor):
(WebCore::CaptionUserPreferencesMediaAF::captionsTextColorCSS):
(WebCore::CaptionUserPreferencesMediaAF::windowRoundedCornerRadiusCSS):
(WebCore::CaptionUserPreferencesMediaAF::captionsEdgeColorForTextColor):
(WebCore::CaptionUserPreferencesMediaAF::cssPropertyWithTextEdgeColor):
(WebCore::CaptionUserPreferencesMediaAF::colorPropertyCSS):
(WebCore::CaptionUserPreferencesMediaAF::captionsTextEdgeCSS):
(WebCore::CaptionUserPreferencesMediaAF::captionsDefaultFontCSS):
(WebCore::CaptionUserPreferencesMediaAF::captionsStyleSheetOverride):
(WebCore::CaptionUserPreferencesMediaAF::textTrackSelectionScore):
(WebCore::CaptionUserPreferencesMediaAF::sortedTrackListForMenu):

  • page/CaptionUserPreferencesMediaAF.h:
  • page/Page.cpp:
  • page/Page.h:
  • page/PageGroup.cpp:

(WebCore::PageGroup::captionPreferencesChanged):
(WebCore::PageGroup::captionPreferences):

  • page/PageGroup.h:
  • testing/InternalSettings.cpp:

(WebCore::InternalSettings::setShouldDisplayTrackKind):
(WebCore::InternalSettings::shouldDisplayTrackKind):

  • testing/Internals.cpp:

(WebCore::Internals::resetToConsistentState):
(WebCore::Internals::Internals):
(WebCore::Internals::captionsStyleSheetOverride):
(WebCore::Internals::setCaptionsStyleSheetOverride):
(WebCore::Internals::setPrimaryAudioTrackLanguageOverride):
(WebCore::Internals::setCaptionDisplayMode):

11:41 PM Changeset in webkit [163072] by Csaba Osztrogonác
  • 16 edits in trunk

Remove ENABLE(JAVASCRIPT_DEBUGGER) leftovers
​https://bugs.webkit.org/show_bug.cgi?id=127845

Reviewed by Joseph Pecoraro.

.:

  • Source/autotools/SetupWebKitFeatures.m4:

Source/JavaScriptCore:

  • Configurations/FeatureDefines.xcconfig:

Source/WebCore:

  • Configurations/FeatureDefines.xcconfig:
  • bindings/js/JSDOMWindowBase.cpp:

(WebCore::JSDOMWindowBase::supportsProfiling):

Source/WebKit/mac:

  • Configurations/FeatureDefines.xcconfig:

Source/WebKit2:

  • Configurations/FeatureDefines.xcconfig:

Tools:

  • Scripts/webkitperl/FeatureList.pm:

WebKitLibraries:

  • win/tools/vsprops/FeatureDefines.props:
  • win/tools/vsprops/FeatureDefinesCairo.props:
11:36 PM Changeset in webkit [163071] by Joseph Pecoraro
  • 14 edits in trunk

Web Inspector: Play Breakpoint Sound in Frontend
​https://bugs.webkit.org/show_bug.cgi?id=127885

Reviewed by Timothy Hatcher.

Source/JavaScriptCore:

  • inspector/ScriptDebugListener.h:
  • inspector/ScriptDebugServer.cpp:

(Inspector::ScriptDebugServer::evaluateBreakpointAction):
(Inspector::ScriptDebugServer::dispatchBreakpointActionSound):

  • inspector/ScriptDebugServer.h:

Pass the breakpoint action identifier through when the
sound breakpoint action is triggered.

  • inspector/protocol/Debugger.json:

New "playBreakpointActionSound" event when a "sound" breakpoint action triggers.

  • inspector/agents/InspectorDebuggerAgent.h:
  • inspector/agents/InspectorDebuggerAgent.cpp:

(Inspector::InspectorDebuggerAgent::breakpointActionSound):
Send the new event so the frontend can handle it.

Source/WebInspectorUI:

Beep in the frontend when a "sound" breakpoint action is triggered.

  • UserInterface/DebuggerManager.js:

(WebInspector.DebuggerManager.prototype.playBreakpointActionSound):

  • UserInterface/DebuggerObserver.js:

(WebInspector.DebuggerObserver.prototype.resumed):
(WebInspector.DebuggerObserver.prototype.playBreakpointActionSound):

  • UserInterface/InspectorJSBackendCommands.js:

LayoutTests:

Extend the breakpoint actions test for the "playBreakpointActionSound" event.

  • inspector-protocol/debugger/setBreakpoint-actions-expected.txt:
  • inspector-protocol/debugger/setBreakpoint-actions.html:
11:30 PM Changeset in webkit [163070] by fpizlo@apple.com
  • 6 edits
    7 adds in trunk/Source/JavaScriptCore

Merge final changesets from the jsCStack branch (r162969, r162975, r162992, r163004, r163069).

2014-01-29 Filip Pizlo <​fpizlo@apple.com>


DFG ArrayPop double array mishandles the NaN hole installation
​https://bugs.webkit.org/show_bug.cgi?id=127813


Reviewed by Mark Rowe.


Our object model for arrays inferred double dictates that we use quiet NaN (QNaN) to
mark holes. Holes, in this context, are any entries in the allocated array buffer
(i.e. from index 0 up to the vectorLength) that don't currently hold a value. Popping
creates a hole, since it deletes the value at publicLength - 1.


But, because of some sloppy copy-and-paste, we were storing (int64_t)0 when creating
the hole, instead of storing QNaN. That's likely because for other kinds of arrays,
64-bit zero is the hole marker, instead of QNaN.


The attached test case illustrates the problem. In the LLInt and Baseline JIT, the
result returned from foo() is "1.5,2.54.5", since array.pop() removes 3.5 and
replaces it with a hole and then the assignment "array[3] = 4.5" creates an element
just beyond that hole. But, once we tier-up to the DFG, the result previously became
"1.5,2.5,0,4.5", which is wrong. The 0 appeared because the IEEE double
interpretation of 64-bit zero is simply zero.


This patch fixes that problem. Now the DFG agrees with the other engines.


This patch also fixes style. For some reason that copy-pasted code wasn't even
indented correctly.


  • dfg/DFGSpeculativeJIT64.cpp: (JSC::DFG::SpeculativeJIT::compile):
  • tests/stress/array-pop-double-hole.js: Added. (foo):


2014-01-28 Filip Pizlo <​fpizlo@apple.com>


FTL should support ArrayPush
​https://bugs.webkit.org/show_bug.cgi?id=127748


Not reviewed, remove some debug code.


  • ftl/FTLLowerDFGToLLVM.cpp: (JSC::FTL::LowerDFGToLLVM::compileArrayPush):


2014-01-27 Filip Pizlo <​fpizlo@apple.com>


FTL should support ArrayPush
​https://bugs.webkit.org/show_bug.cgi?id=127748


Reviewed by Oliver Hunt.


  • ftl/FTLAbstractHeapRepository.h: (JSC::FTL::AbstractHeapRepository::forArrayType):
  • ftl/FTLCapabilities.cpp: (JSC::FTL::canCompile):
  • ftl/FTLIntrinsicRepository.h:
  • ftl/FTLLowerDFGToLLVM.cpp: (JSC::FTL::LowerDFGToLLVM::compileNode): (JSC::FTL::LowerDFGToLLVM::compileArrayPush):
  • tests/stress/array-push-contiguous.js: Added. (foo):
  • tests/stress/array-push-double.js: Added. (foo):


2014-01-28 Filip Pizlo <​fpizlo@apple.com>


FTL should support ArrayPop
​https://bugs.webkit.org/show_bug.cgi?id=127749


Reviewed by Geoffrey Garen.


  • ftl/FTLCapabilities.cpp: (JSC::FTL::canCompile):
  • ftl/FTLIntrinsicRepository.h:
  • ftl/FTLLowerDFGToLLVM.cpp: (JSC::FTL::LowerDFGToLLVM::compileNode): (JSC::FTL::LowerDFGToLLVM::compileArrayPush): (JSC::FTL::LowerDFGToLLVM::compileArrayPop):
  • tests/stress/array-pop-contiguous.js: Added. (foo):
  • tests/stress/array-pop-double.js: Added. (foo):
  • tests/stress/array-pop-int32.js: Added. (foo):
9:44 PM Changeset in webkit [163069] by fpizlo@apple.com
  • 2 edits in trunk/Source/JavaScriptCore

DFG::ByteCodeParser::m_dfgCodeBlock is sometimes uninitialized
<rdar://problem/15939032>

Reviewed by Dan Bernstein.

  • dfg/DFGByteCodeParser.cpp:

(JSC::DFG::ByteCodeParser::parse):

9:33 PM Changeset in webkit [163068] by bshafiei@apple.com
  • 4 edits in tags/Safari-538.13.1/Source/JavaScriptCore

Merged r162968.

9:25 PM Changeset in webkit [163067] by mitz@apple.com
  • 3 edits in trunk/Source/WebKit2

Moved linking against common frameworks and a library from the .xcconfig file into the target.

Reviewed by Tim Horton.

  • Configurations/WebKit2.xcconfig:
  • WebKit2.xcodeproj/project.pbxproj:
9:23 PM Changeset in webkit [163066] by bshafiei@apple.com
  • 5 edits in tags/Safari-538.13.1/Source

Versioning.

9:19 PM Changeset in webkit [163065] by bshafiei@apple.com
  • 1 copy in tags/Safari-538.13.1

New tag.

7:20 PM Changeset in webkit [163064] by andersca@apple.com
  • 3 edits in trunk/Source/WebKit2

Fix 32-bit build.

  • UIProcess/Cocoa/NavigationState.h:
  • UIProcess/Cocoa/NavigationState.mm:
6:50 PM Changeset in webkit [163063] by mitz@apple.com
  • 4 edits in trunk/Source/WebKit

Moved linking against common frameworks and a library from the .xcconfig file into the target.

Reviewed by Tim Horton.

../WebKit:

  • WebKit.xcodeproj/project.pbxproj:

../WebKit/mac:

  • Configurations/WebKit.xcconfig:
6:49 PM Changeset in webkit [163062] by andersca@apple.com
  • 7 edits
    2 copies
    1 add in trunk/Source/WebKit2

Add a NavigationState object to keep track of the navigations for a given page
​https://bugs.webkit.org/show_bug.cgi?id=127884

Reviewed by Sam Weinig.

  • UIProcess/API/Cocoa/WKNavigation.mm:

(-[WKNavigation request]):
(-[WKNavigation setRequest:]):

  • UIProcess/API/Cocoa/WKNavigationInternal.h:
  • UIProcess/API/Cocoa/WKWebView.mm:

(-[WKWebView initWithFrame:configuration:]):
(-[WKWebView loadRequest:]):

  • UIProcess/Cocoa/NavigationState.h: Copied from Source/WebKit2/UIProcess/API/Cocoa/WKNavigation.mm.
  • UIProcess/Cocoa/NavigationState.mm: Copied from Source/WebKit2/UIProcess/API/Cocoa/WKNavigation.mm.

(WebKit::NavigationState::NavigationState):
(WebKit::NavigationState::~NavigationState):
(WebKit::NavigationState::createLoadRequestNavigation):

  • UIProcess/WebPageProxy.cpp:

(WebKit::WebPageProxy::loadRequest):

  • UIProcess/WebPageProxy.h:
  • WebKit2.xcodeproj/project.pbxproj:
6:10 PM Changeset in webkit [163061] by barraclough@apple.com
  • 9 edits in trunk/Source

Add IsVisibleOrOccluded to ViewState
​https://bugs.webkit.org/show_bug.cgi?id=127875

Reviewed by Anders Carlsson.

../WebCore:

  • page/ViewState.h:
    • added IsVisibleOrOccluded

../WebKit2:

When the WKView associated with a plugin becomes occluded
we tell the plugin it is no longer visible. This is a problem
if the plugin is being occluded by its own fullscreen window.

Instead, consider plugins to be visible any time they are in
a visible window – i.e. when they are visible or occluded.

  • UIProcess/API/mac/PageClientImpl.h:
  • UIProcess/API/mac/PageClientImpl.mm:

(WebKit::PageClientImpl::isViewVisibleOrOccluded):

  • UIProcess/PageClient.h:

(WebKit::PageClient::isViewVisibleOrOccluded):

  • added isViewVisibleOrOccluded.
  • UIProcess/WebPageProxy.cpp:

(WebKit::WebPageProxy::updateViewState):

  • update IsVisibleOrOccluded.

(WebKit::WebPageProxy::viewStateDidChange):

  • when updating IsVisible also update IsVisibleOrOccluded.
  • WebProcess/Plugins/PluginView.cpp:

(WebKit::PluginView::platformViewStateDidChange):

  • use isVisibleOrOccluded to determine plugin visibility.
  • WebProcess/WebPage/WebPage.h:

(WebKit::WebPage::isVisibleOrOccluded):

  • accessor.
6:07 PM Changeset in webkit [163060] by andersca@apple.com
  • 6 edits in trunk/Source/WebKit2

Pass navigation IDs along to LoadRequest and Reload
​https://bugs.webkit.org/show_bug.cgi?id=127883

Reviewed by Sam Weinig.

  • UIProcess/WebPageProxy.cpp:

(WebKit::WebPageProxy::WebPageProxy):
(WebKit::WebPageProxy::loadRequest):
(WebKit::WebPageProxy::loadFile):
(WebKit::WebPageProxy::reload):
(WebKit::WebPageProxy::generateNavigationID):

  • UIProcess/WebPageProxy.h:
  • WebProcess/WebPage/WebDocumentLoader.cpp:

(WebKit::WebDocumentLoader::setNavigationID):

  • WebProcess/WebPage/WebPage.cpp:

(WebKit::WebPage::WebPage):
(WebKit::WebPage::loadRequest):
(WebKit::WebPage::reload):
(WebKit::WebPage::createDocumentLoader):

  • WebProcess/WebPage/WebPage.h:
  • WebProcess/WebPage/WebPage.messages.in:
6:05 PM Changeset in webkit [163059] by rniwa@webkit.org
  • 2 edits in trunk/Source/WebKit/mac

checkTextOfParagraph releases NSString twice
​https://bugs.webkit.org/show_bug.cgi?id=127881
<rdar://problem/15937446>

Reviewed by Anders Carlsson.

textString is released once by an explicit call to release and again by it going out of scope.
Don't do the explicit release.

The bug was spotted by Sam Weinig.

  • WebCoreSupport/WebEditorClient.mm:

(WebEditorClient::checkTextOfParagraph):

6:03 PM Changeset in webkit [163058] by mhahnenberg@apple.com
  • 2 edits in trunk/LayoutTests

js/function-apply-aliased.html is timing out
​https://bugs.webkit.org/show_bug.cgi?id=127860

Reviewed by Michael Saboff.

6:02 PM Changeset in webkit [163057] by ggaren@apple.com
  • 2 edits in trunk/Source/JavaScriptCore

50% time on Dromaeo Selector * benchmark spent allocating oversized backing stores (but not in Chrome)
​https://bugs.webkit.org/show_bug.cgi?id=127879

Reviewed by Gavin Barraclough.

Let's not dynamically resize an array whose size is statically known,
mmmkay?

  • runtime/ArrayPrototype.cpp:

(JSC::arrayProtoFuncConcat): Use nullptr to disambiguate vs the numeric
argument.

(JSC::arrayProtoFuncSlice): The fix.

(JSC::arrayProtoFuncSort):
(JSC::arrayProtoFuncSplice):
(JSC::arrayProtoFuncFilter):
(JSC::arrayProtoFuncMap): Use nullptr.

5:53 PM Changeset in webkit [163056] by rniwa@webkit.org
  • 4 edits in trunk/Source/WebCore

EventHandler::handleMouseReleaseEvent shouldn't call updateSelectionCachesIfSelectionIsInsideTextFormControl
and selectFrameElementInParentIfFullySelected
​https://bugs.webkit.org/show_bug.cgi?id=127834

Reviewed by Alexey Proskuryakov.

Removed the calls and made setNonDirectionalSelectionIfNeeded pass in UserTriggered option.

In addition, removed the rather error-prone function override of setSelection since TextGranularity,
which is an enum, could be implicitly coerced into SetSelectionOptions which is a typedefed unsigned int.

  • editing/FrameSelection.cpp:

(WebCore::FrameSelection::setSelectionByMouseIfDifferent): Renamed from setNonDirectionalSelectionIfNeeded.
Pass in DoNotRevealSelection to avoid revealing the selection to preserve the existing behavior.
There are two layout tests that fail without this.
(WebCore::FrameSelection::setSelection): Check the newly addeed DoNotRevealSelection option.
(WebCore::FrameSelection::wordSelectionContainingCaretSelection): Call

  • editing/FrameSelection.h: Made updateSelectionCachesIfSelectionIsInsideTextFormControl and

selectFrameElementInParentIfFullySelected private as they are no longer called outside of FrameSelection.

  • page/EventHandler.cpp:

(WebCore::EventHandler::updateSelectionForMouseDownDispatchingSelectStart):
(WebCore::EventHandler::updateSelectionForMouseDrag):
(WebCore::EventHandler::handleMouseReleaseEvent): Removed calls to the functions.

5:41 PM Changeset in webkit [163055] by jer.noble@apple.com
  • 2 edits in trunk/Source/WebCore

Unreviewed iOS build fix after 163050.

Import CALayer.h explicitly as (on iOS) it is not included by other headers.

  • platform/graphics/avfoundation/objc/SourceBufferPrivateAVFObjC.mm:
5:32 PM Changeset in webkit [163054] by commit-queue@webkit.org
  • 2 edits in trunk/Source/JavaScriptCore

Web Inspector: Run JSC Inspector EventLoop in a custom run loop mode to prevent default observers from running
​https://bugs.webkit.org/show_bug.cgi?id=127865

Patch by Joseph Pecoraro <Joseph Pecoraro> on 2014-01-29
Reviewed by Geoffrey Garen.

When hitting a breakpoint in a JSContext Inspector we want to entirely
pause the process and all access to the JSContext and only move forward
based on debugger commands. Having the nested run loop run in a default
mode allowed NSTimers scheduled on the thread to regularly run and
evaluate code in the JSContext. Using a custom run loop mode gets us
a bit closer to locking down the context. This doesn't handle scenarios
where background threads also access the JSContext, but it handles the
most common scenario.

  • inspector/EventLoop.cpp:

(Inspector::EventLoop::cycle):

5:17 PM Changeset in webkit [163053] by andersca@apple.com
  • 8 edits
    2 copies
    1 add in trunk

-[WKWebView loadRequest:] should return a WKNavigation
​https://bugs.webkit.org/show_bug.cgi?id=127873

Reviewed by Dan Bernstein.

Source/WebKit2:

For now this just returns a dummy navigation.

  • UIProcess/API/Cocoa/WKNavigation.h:
  • UIProcess/API/Cocoa/WKNavigation.mm:

(-[WKNavigation initWithRequest:]):
(-[WKNavigation request]):

  • UIProcess/API/Cocoa/WKNavigationInternal.h: Added.
  • UIProcess/API/Cocoa/WKWebView.h:
  • UIProcess/API/Cocoa/WKWebView.mm:

(-[WKWebView loadRequest:]):

  • WebKit2.xcodeproj/project.pbxproj:

Tools:

  • TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj:
  • TestWebKitAPI/Tests/WebKit2Cocoa/Navigation.mm: Added.

(TEST):

5:14 PM Changeset in webkit [163052] by weinig@apple.com
  • 2 edits in trunk/Source/WebKit

Fix windows.

  • WebKit.vcxproj/WebKitExportGenerator/WebKitExports.def.in:
5:01 PM Changeset in webkit [163051] by commit-queue@webkit.org
  • 2 edits in trunk/Source/JavaScriptCore

Web Inspector: Deadlock hitting breakpoint while inspecting JSContext
​https://bugs.webkit.org/show_bug.cgi?id=127864

Patch by Joseph Pecoraro <Joseph Pecoraro> on 2014-01-29
Reviewed by Geoffrey Garen.

Temporarily drop the lock while we run the nested runloop.

  • inspector/JSGlobalObjectScriptDebugServer.cpp:

(Inspector::JSGlobalObjectScriptDebugServer::runEventLoopWhilePaused):

4:53 PM Changeset in webkit [163050] by jer.noble@apple.com
  • 2 edits in trunk/Source/WebCore

Unreviewed Mac Build fix after r163046.

Forward define AVSampleLayerDisplayLayer and its methods.

  • platform/graphics/avfoundation/objc/SourceBufferPrivateAVFObjC.mm:
4:39 PM Changeset in webkit [163049] by andersca@apple.com
  • 2 edits in trunk/Source/WebKit2
  • UIProcess/API/C/WKPage.cpp:

(WKPageLoadURL):
Don't use the resource request constructor that takes a String.

4:35 PM Changeset in webkit [163048] by weinig@apple.com
  • 23 edits
    2 adds in trunk/Source/WebCore

CaptionUserPreferences should not be on the PageGroup if they are not really per-PageGroup (Part 1)
​https://bugs.webkit.org/show_bug.cgi?id=127666

Reviewed by Eric Carlson.

This patch:

  • Makes CaptionUserPreferences a singleton (temporary) that is accessed via Page.
  • Simplifies overriding system preferences by requiring that a Document be be passed so the correct Settings object can be obtained (and not just a random one).
  • Stops using UserStyleSheets for captions style injection, and instead adds a new style sheet to the DocumentStyleSheetCollection.
  • Move caption style sheet creation into its own file - CaptionStyleSheetMediaAF.h/cpp
  • WebCore.vcxproj/WebCore.vcxproj:
  • WebCore.vcxproj/WebCore.vcxproj.filters:
  • WebCore.xcodeproj/project.pbxproj:

Add new files.

  • WebCore.exp.in:

Update exports.

  • css/DocumentRuleSets.cpp:
  • css/InspectorCSSOMWrappers.cpp:
  • dom/Document.cpp:
  • dom/DocumentStyleSheetCollection.cpp:
  • dom/DocumentStyleSheetCollection.h:

Switch from using the user style sheet mechanism, which is meant for, you guessed it, users,
to an explicit caption style sheet.

  • Modules/mediacontrols/MediaControlsHost.cpp:
  • html/HTMLMediaElement.cpp:
  • html/shadow/MediaControlElements.cpp:

Get the CaptionUserPreferences via the Page.

  • page/CaptionStyleSheetMediaAF.cpp: Added.
  • page/CaptionStyleSheetMediaAF.h: Added.

Move caption style sheet creation here. If the global style changes, the sheet is invalidated
and this is called again.

  • page/CaptionUserPreferences.cpp:
  • page/CaptionUserPreferences.h:
  • Removes unused m_havePreferences member.
  • Change userPrefersFoo() functions to take a Document&. Use it to check its Settings.
  • Remove setUserPrefersFoo() functions and just set the Settings directly.
  • Move all the testing only pieces together.


  • page/CaptionUserPreferencesMediaAF.cpp:
  • page/CaptionUserPreferencesMediaAF.h:
  • Extract caption style sheet creation into CaptionStyleSheetMediaAF.h/cpp
  • Stop waiting for an interested party, and just register for appearance change notifications right away. Media elements are common enough that this should not make a difference.
  • page/Page.cpp:

(WebCore::Page::updateStyleForAllPagesForCaptionPreferencesChanged):
Add helper to call captionPreferencesChanged() on all the Pages.

(WebCore::Page::captionPreferences):
Add accessor for the CaptionUserPreferences. Currently this returns a singleton, but I plan
to extract the singleton aspects of it into another class, and leave the per-Page parts here.

  • page/Page.h:
  • page/PageGroup.cpp:
  • page/PageGroup.h:

Remove CaptionUserPreferences from here.

  • testing/InternalSettings.cpp:

(WebCore::InternalSettings::setShouldDisplayTrackKind):
(WebCore::InternalSettings::shouldDisplayTrackKind):
Override the settings for the page directly.

  • testing/Internals.cpp:

(WebCore::Internals::resetToConsistentState):
(WebCore::Internals::Internals):
(WebCore::Internals::captionsStyleSheetOverride):
(WebCore::Internals::setCaptionsStyleSheetOverride):
(WebCore::Internals::setPrimaryAudioTrackLanguageOverride):
(WebCore::Internals::setCaptionDisplayMode):
Get the CaptionUserPreferences via the Page.

4:31 PM Changeset in webkit [163047] by Bem Jones-Bey
  • 5 edits
    2 copies
    4 adds in trunk/LayoutTests

Update clip-path tests to use new shape syntax
​https://bugs.webkit.org/show_bug.cgi?id=127862

Reviewed by Dirk Schulze.

Update the tests to reflect the new shape syntax. The inset tests are
copied from the rectangle tests that will be removed when the old
syntax is removed.

  • css3/masking/clip-path-inset-expected.txt: Added.
  • css3/masking/clip-path-inset.html: Added.
  • svg/clip-path/clip-path-shape-circle-1.svg:
  • svg/clip-path/clip-path-shape-circle-2.svg:
  • svg/clip-path/clip-path-shape-ellipse-1.svg:
  • svg/clip-path/clip-path-shape-ellipse-2.svg:
  • svg/clip-path/clip-path-shape-rounded-inset-1-expected.svg: Added.
  • svg/clip-path/clip-path-shape-rounded-inset-1.svg: Copied from LayoutTests/svg/clip-path/clip-path-shape-circle-1.svg.
  • svg/clip-path/clip-path-shape-rounded-inset-2-expected.svg: Added.
  • svg/clip-path/clip-path-shape-rounded-inset-2.svg: Copied from LayoutTests/svg/clip-path/clip-path-shape-circle-1.svg.
4:02 PM Changeset in webkit [163046] by jer.noble@apple.com
  • 2 edits in trunk/Source/WebCore

[MSE][Mac] In SourceBufferPrivateAVFObjC, only include those headers actually necessary to compile.
​https://bugs.webkit.org/show_bug.cgi?id=127846

Reviewed by Darin Adler.

To work around a compile issue, only include those paths containing classes and typedefs
which are used within SourceBufferPrivateAVFObjC, rather than all of AVFoundation.

  • platform/graphics/avfoundation/objc/SourceBufferPrivateAVFObjC.mm:
4:00 PM Changeset in webkit [163045] by beidson@apple.com
  • 9 edits in trunk/Source

IDB: Fully implement IDBKeyData encoding/decoding
​https://bugs.webkit.org/show_bug.cgi?id=127863

Rubberstamped by Alexey Proskuryakov.

Source/WebCore:

  • platform/KeyedCoding.h:
  • Modules/indexeddb/IDBKeyData.cpp:

(WebCore::IDBKeyData::encode):
(WebCore::IDBKeyData::decode):

  • WebCore.exp.in:

Source/WebKit2:

Handle invalid buffers, and pass IDBKeyData decoding along:

  • DatabaseProcess/IndexedDB/IDBSerialization.cpp:

(WebKit::deserializeIDBKeyPath):
(WebKit::serializeIDBKeyData):
(WebKit::deserializeIDBKeyData):

  • DatabaseProcess/IndexedDB/IDBSerialization.h:

Implement decodeBool and decodeDouble:

  • Shared/cf/KeyedDecoder.cpp:

(WebKit::KeyedDecoder::decodeBool):
(WebKit::KeyedDecoder::decodeDouble):

  • Shared/cf/KeyedDecoder.h:
3:30 PM Changeset in webkit [163044] by Bem Jones-Bey
  • 3 edits
    2 adds in trunk

clip-path swaps bottom radii for the inset shape
​https://bugs.webkit.org/show_bug.cgi?id=127859

Reviewed by Dirk Schulze.

Source/WebCore:

The bottom right and bottom left radii were passed in the wrong order,
causing the wrong corner to be clipped in the resultant rendering.

Test: css3/masking/clip-path-inset-corners.html

  • rendering/style/BasicShapes.cpp:

(WebCore::BasicShapeInset::path): Change the parameter order to be
correct.

LayoutTests:

Test that all the inset radii are applied to the correct corners.

  • css3/masking/clip-path-inset-corners-expected.html: Added.
  • css3/masking/clip-path-inset-corners.html: Added.
3:06 PM Changeset in webkit [163043] by timothy_horton@apple.com
  • 7 edits in trunk

WebKit2 View Gestures (Swipe): Update rubberBandsAt{Left,Right} when WebKit swipe is enabled
​https://bugs.webkit.org/show_bug.cgi?id=127855
<rdar://problem/15933878>

Reviewed by Anders Carlsson.

  • UIProcess/API/mac/WKView.mm:

(-[WKView setAllowsBackForwardNavigationGestures:]):
Enable implicit control of rubberbanding when using WebKit2's swipe implementation.

  • UIProcess/WebPageProxy.cpp:

(WebKit::WebPageProxy::WebPageProxy):
(WebKit::WebPageProxy::sendWheelEvent):

  • UIProcess/WebPageProxy.h:

(WebKit::WebPageProxy::setShouldUseImplicitRubberBandControl):
(WebKit::WebPageProxy::shouldUseImplicitRubberBandControl):
Rename away from "legacy" name, since we're using it on purpose now.

  • UIProcess/mac/WebPageProxyMac.mm:

(WebKit::expectsLegacyImplicitRubberBandControl):
(WebKit::WebPageProxy::platformInitialize):
Since NSVersionOfLinkTimeLibrary doesn't work for applications that
don't link directly against the library (like Safari), let's explicitly
check if Safari is expecting the legacy behavior, otherwise using
the linked version of WebKit2, but using the new behavior if we can't
determine the linked version of WebKit2 and are not Safari.

  • MiniBrowser/mac/WK2BrowserWindowController.m:

(-[WK2BrowserWindowController awakeFromNib]):
Enable swipe in MiniBrowser for testing purposes.

2:46 PM Changeset in webkit [163042] by andersca@apple.com
  • 4 edits in trunk/Source/WebKit2

Make WKWebView limp along on OS X
​https://bugs.webkit.org/show_bug.cgi?id=127857

Reviewed by Tim Horton.

On Mac, create a WKView and add it as a subview of the WKWebVew so we'll actually get something
on screen when creating a WKWebView.

  • UIProcess/API/Cocoa/WKWebView.mm:

(-[WKWebView initWithFrame:configuration:]):
Create a WKView.

(-[WKWebView resizeSubviewsWithOldSize:]):
Resize the WKView.

  • UIProcess/API/mac/WKView.mm:

(-[WKView initWithFrame:configuration:]):
Add new initializer.

  • UIProcess/API/mac/WKViewInternal.h:

Use a class extension for internal methods and add the new initializer.

2:14 PM Changeset in webkit [163041] by jer.noble@apple.com
  • 2 edits in trunk/Source/WebCore

[MediaControls][iOS] Hook up the fullscreen button
​https://bugs.webkit.org/show_bug.cgi?id=127850

Reviewed by Eric Carlson.

For the fullscreen button, do the same thing we do for the play button: handle touchstart,
touchend, and touchcancel, and perform the action (i.e., 'click') on touchend.

  • Modules/mediacontrols/mediaControlsiOS.js:

(ControllerIOS.prototype.createControls): Add listeners for the fullscreenButton.
(ControllerIOS.prototype.isFullScreen): Override.
(ControllerIOS.prototype.handleFullscreenButtonClicked): Override.
(ControllerIOS.prototype.handleFullscreenTouchStart): Added.
(ControllerIOS.prototype.handleFullscreenTouchEnd): Added.
(ControllerIOS.prototype.handleFullscreenTouchCancel): Added.

1:21 PM Changeset in webkit [163040] by dbates@webkit.org
  • 2 edits in trunk/Tools

Fix JSONP requests made by the WebKit Bot Watcher's Dashboard following <​http://trac.webkit.org/changeset/163034>
(​https://bugs.webkit.org/show_bug.cgi?id=127784)

Use dot-notation instead of array index notation to access the dictionary value for key jsonpCallbackName
in the specified options dictionary.

  • BuildSlaveSupport/build.webkit.org-config/public_html/dashboard/Scripts/Utilities.js:

(JSON.load):

1:17 PM Changeset in webkit [163039] by andersca@apple.com
  • 2 edits in trunk/Source/WebKit/mac

Fix build.

  • WebView/WebPDFView.mm:

(_applicationInfoForMIMEType):

1:15 PM Changeset in webkit [163038] by beidson@apple.com
  • 6 edits in trunk/Source

IDB: Opening an existing database is broken
​https://bugs.webkit.org/show_bug.cgi?id=127851

Reviewed by Tim Hatcher.

Source/WebCore:

  • Modules/indexeddb/IDBDatabaseBackend.cpp:

(WebCore::IDBDatabaseBackend::IDBDatabaseBackend):
(WebCore::IDBDatabaseBackend::didOpenInternalAsync): Remember if this method has completed.
(WebCore::IDBDatabaseBackend::processPendingCalls): If didOpenInternalAsync hasn’t

been called yet, then it is incorrect to process any pending calls right now.

  • Modules/indexeddb/IDBDatabaseBackend.h:

Source/WebKit2:

  • DatabaseProcess/IndexedDB/UniqueIDBDatabase.cpp:

(WebKit::UniqueIDBDatabase::getOrEstablishIDBDatabaseMetadata): Save the request to get metadata

to the set before posting the task to the thread.

  • DatabaseProcess/IndexedDB/sqlite/UniqueIDBDatabaseBackingStoreSQLite.cpp:

(WebKit::UniqueIDBDatabaseBackingStoreSQLite::extractExistingMetadata): Actually save the retrieved

object store metadata to the database metadata.

12:54 PM Changeset in webkit [163037] by andersca@apple.com
  • 11 edits in trunk/Source/WebKit2

Consolidate page loading functions
​https://bugs.webkit.org/show_bug.cgi?id=127847

Reviewed by Tim Horton.

Get rid of the loadURL functions (and the LoadURL message), rename
loadURLRequest to loadRequest and change it to take a ResourceRequest instead
of an API::URLRequest.

  • UIProcess/API/C/WKPage.cpp:

(WKPageLoadURL):
(WKPageLoadURLWithUserData):
(WKPageLoadURLRequest):
(WKPageLoadURLRequestWithUserData):

  • UIProcess/API/Cocoa/WKBrowsingContextController.mm:

(-[WKBrowsingContextController loadRequest:userData:]):

  • UIProcess/API/Cocoa/WKWebView.mm:

(-[WKWebView loadRequest:]):

  • UIProcess/WebInspectorProxy.cpp:

(WebKit::decidePolicyForNavigationAction):
(WebKit::WebInspectorProxy::createInspectorPage):

  • UIProcess/WebPageProxy.cpp:

(WebKit::WebPageProxy::loadRequest):
(WebKit::WebPageProxy::loadFile):

  • UIProcess/WebPageProxy.h:
  • UIProcess/cf/WebPageProxyCF.cpp:

(WebKit::WebPageProxy::restoreFromSessionStateData):

  • WebProcess/WebPage/WebPage.cpp:

(WebKit::WebPage::loadRequest):

  • WebProcess/WebPage/WebPage.h:
  • WebProcess/WebPage/WebPage.messages.in:
12:38 PM Changeset in webkit [163036] by oliver@apple.com
  • 2 edits in trunk/Source/WebCore

2014-01-29 Oliver Hunt <​oliver@apple.com>

This dereference should not actually be necessary, and
is producing deprecation warnings with newer perls

  • bindings/scripts/CodeGeneratorJS.pm: (GenerateAttributesHashTable):
12:33 PM Changeset in webkit [163035] by oliver@apple.com
  • 29 edits in trunk

Make DOM attributes appear to be faux accessor properties
​https://bugs.webkit.org/show_bug.cgi?id=127797

Reviewed by Michael Saboff.

Source/JavaScriptCore:

Add flag so we can identify which properties should have the old
custom property semantics vs. the new faux accessors. Update the
inspector protocol accordingly.

These faux accessors produce descriptors with "get" and "set"
properties, but both values are undefined so can't be used
directly. A few custom properties actually require their
existing magical behaviour, so we now have a flag to
distinguish the expected output.

  • inspector/InjectedScriptSource.js:

(.):

  • runtime/JSObject.cpp:

(JSC::JSObject::getOwnPropertyDescriptor):

  • runtime/PropertyDescriptor.cpp:

(JSC::PropertyDescriptor::setCustomDescriptor):

  • runtime/PropertyDescriptor.h:
  • runtime/PropertySlot.h:

Source/WebCore:

Refactor the bindings generator, and make sure we emit
the CustomAccessor flag on properties that should
appear to be accessors.

  • bindings/scripts/CodeGeneratorJS.pm:

(GenerateGetOwnPropertySlotBody):
(GenerateAttributesHashTable):
(GenerateImplementation):

LayoutTests:

These properties now show up as accessors in these tests

  • fast/dom/Window/replaceable-expected.txt:
  • fast/dom/Window/replaceable.html:
  • js/dom/getOwnPropertyDescriptor-expected.txt:
  • js/resources/getOwnPropertyDescriptor.js:
12:23 PM Changeset in webkit [163034] by dbates@webkit.org
  • 5 edits in trunk/Tools

WebKit Bot Watcher's Dashboard: Add support for using HTTP authentication with Buildbot
​https://bugs.webkit.org/show_bug.cgi?id=127784

Reviewed by Alexey Proskuryakov.

Implement support for querying a Buildbot that requires HTTP authentication to access it.

  • BuildSlaveSupport/build.webkit.org-config/public_html/dashboard/Scripts/Buildbot.js:

(Buildbot): Add optional argument, options, to pass a dictionary of Buildbot options.
For now, we only support a single option, needsAuthentication, to indicate whether the
Buildbots requires HTTP authentication (defaults to false - HTTP authentication is
not required).

  • BuildSlaveSupport/build.webkit.org-config/public_html/dashboard/Scripts/BuildbotIteration.js:

(BuildbotIteration.prototype.loadLayoutTestResults): Modified to pass JSONP callback name
to JSON.load() using the option dictionary.

  • BuildSlaveSupport/build.webkit.org-config/public_html/dashboard/Scripts/BuildbotQueue.js:

(BuildbotQueue.prototype.update): Modified to pass withCredentials to JSON.load().

  • BuildSlaveSupport/build.webkit.org-config/public_html/dashboard/Scripts/Utilities.js:

(JSON.load): Removed optional argument jsonpCallbackName and added optional argument options,
which is a dictionary. The argument jsonpCallbackName is now specified as an entry in the
options dictionary. Added boolean options withCredentials to toggle querying with HTTP
authentication credentials.

12:17 PM Changeset in webkit [163033] by ryuan.choi@samsung.com
  • 10 edits
    1 delete in trunk/Source/WebKit

EFL port should stop using COMPILE_ASSERT_MATCHING_ENUM macros
​https://bugs.webkit.org/show_bug.cgi?id=127801

Reviewed by Anders Carlsson.

Source/WebKit:

  • PlatformEfl.cmake: Removed AssertMatchingEnums.cpp

Source/WebKit/efl:

Instead of COMPILE_ASSERT_MATCHING_ENUMS with static_cast, this patch
added switch and conversion functions if needed.

  • WebCoreSupport/AssertMatchingEnums.cpp: Removed.
  • WebCoreSupport/FrameLoaderClientEfl.cpp:

(WebCore::FrameLoaderClientEfl::dispatchDecidePolicyForNavigationAction):

  • ewk/ewk_contextmenu.cpp:

(toEwkContextMenuAction):

  • ewk/ewk_contextmenu.h:

Added missing enum variables. These are required not to break the build.

  • ewk/ewk_frame.cpp:

(ewk_frame_text_selection_type_get):

  • ewk/ewk_touch_event.cpp:

(toPlatformTouchPointState):
(WebKitPlatformTouchEvent::WebKitPlatformTouchEvent):

  • ewk/ewk_view.cpp:

(ewk_view_text_direction_set):
(ewk_view_navigation_policy_decision):
(toViewMode):
(toEwkViewMode):
(ewk_view_mode_set):
(ewk_view_mode_get):
(ewk_view_visibility_state_get):

  • ewk/ewk_view_private.h:
11:51 AM Changeset in webkit [163032] by Martin Robinson
  • 4 edits in trunk/Source/WebKit2

[GTK] [CMake] Fix the build after r162835
​https://bugs.webkit.org/show_bug.cgi?id=127753

Reviewed by Gustavo Noronha Silva.

  • CMakeLists.txt: Add support for using the prefix header (GCC / clang only for now)

when WebKit2_USE_PREFIX_HEADER is set to true.

  • PlatformGTK.cmake: Set WebKit2_USE_PREFIX_HEADER support and no longer process some message.in files.
  • WebKit2Prefix.h: Add support for the CMake GTK+ build.
11:40 AM Changeset in webkit [163031] by Beth Dakin
  • 2 edits in trunk/Source/JavaScriptCore

Build fix.

  • bytecode/ProfiledCodeBlockJettisoningWatchpoint.cpp:
  • llint/LowLevelInterpreter.cpp:
11:38 AM Changeset in webkit [163030] by mitz@apple.com
  • 2 edits in trunk/Source/JavaScriptCore

Build fix.

  • bytecode/ProfiledCodeBlockJettisoningWatchpoint.cpp: Added a newline at the end of the

file.

11:35 AM Changeset in webkit [163029] by ap@apple.com
  • 5 edits in trunk/Tools

build.webkit.org/dashboard links to an old failing build rather than the most recent
​https://bugs.webkit.org/show_bug.cgi?id=127765
<rdar://problem/15795228>

Reviewed by Timothy Hatcher.

This patch changes builder view to not show how many failures there were in a row,
because it is not very relevant, and complicates the UI. Instead, we just show the
latest failing build and the latest successful one.

The patch also removes commit history popovers from revisions that can not be correlated
with any change. These popovers are only helpful in the first line, when there is
another line for comparison below. In this case, it's useful to see what changed
between the revisions, but even then, only if it was an actual failure, not a buildbot
exception.

  • BuildSlaveSupport/build.webkit.org-config/public_html/dashboard/Scripts/BuildbotBuilderQueueView.js:

(BuildbotBuilderQueueView.prototype.update.appendBuilderQueueStatus):

  • BuildSlaveSupport/build.webkit.org-config/public_html/dashboard/Scripts/BuildbotQueue.js:
  • BuildSlaveSupport/build.webkit.org-config/public_html/dashboard/Scripts/BuildbotQueueView.js:

(BuildbotQueueView.prototype._presentPopoverForRevisionRange):
(BuildbotQueueView.prototype._revisionPopoverContentForIteration):
(BuildbotQueueView.prototype.revisionContentForIteration):

  • BuildSlaveSupport/build.webkit.org-config/public_html/dashboard/Scripts/BuildbotTesterQueueView.js:

(BuildbotTesterQueueView.prototype.update.appendBuilderQueueStatus):

11:20 AM Changeset in webkit [163028] by alex.christensen@flexsim.com
  • 6 edits in trunk/Source

Prepare for accelerated compositing on WinCairo.
​https://bugs.webkit.org/show_bug.cgi?id=127843

Reviewed by Anders Carlsson.

Source/WebCore:

  • platform/network/curl/ResourceHandleManager.cpp:

(WebCore::setupFormData):
Remove VS2005 warning workaround.

  • WebCore.vcxproj/WebCore.vcxproj:

Remove StyleCachedImageSet.cpp and CSSImageSetValue.cpp from the Win64 build
to avoid warnings because they are also included in StyleAllInOne.cpp and
CSSAllInOne.cpp, respectively.

Source/WebKit/win:

  • WebPreferences.cpp:

Moved CACFLayerTreeHost.h inclusion to Apple-specific includes
to avoid compile errors in WinCairo when using accelerated compositing.

11:18 AM Changeset in webkit [163027] by msaboff@apple.com
  • 219 edits
    76 adds
    4 deletes in trunk

Merge the jsCStack branch
​https://bugs.webkit.org/show_bug.cgi?id=127763

Reviewed by Mark Hahnenberg.

Source/JavaScriptCore:

Changes from ​http://svn.webkit.org/repository/webkit/branches/jsCStack
up to changeset 162958.

Source/WebCore:

Changes from ​http://svn.webkit.org/repository/webkit/branches/jsCStack
up to changeset 162958.

Source/WTF:

Changes from ​http://svn.webkit.org/repository/webkit/branches/jsCStack
up to changeset 162958.

10:52 AM Changeset in webkit [163026] by mitz@apple.com
  • 2 edits in trunk/Source/WebCore

Fixed the iOS build.

  • bindings/js/JSDOMWindowCustom.cpp: Updated for header renames.
10:51 AM Changeset in webkit [163025] by ap@apple.com
  • 1 edit
    2 adds in trunk/LayoutTests

Add a WebCrypto test to verify that BigInteger allows for leading zeroes
​https://bugs.webkit.org/show_bug.cgi?id=127804

Reviewed by Darin Adler.

This already worked, but wasn't tested for.

  • crypto/subtle/rsassa-pkcs1-v1_5-generate-key-with-leading-zeroes-in-exponent-expected.txt: Added.
  • crypto/subtle/rsassa-pkcs1-v1_5-generate-key-with-leading-zeroes-in-exponent.html: Added.
10:31 AM Changeset in webkit [163024] by Csaba Osztrogonác
  • 56 edits in trunk

Remove ENABLE(JAVASCRIPT_DEBUGGER) guards
​https://bugs.webkit.org/show_bug.cgi?id=127840

Reviewed by Mark Lam.

.:

  • Source/cmake/WebKitFeatures.cmake:
  • Source/cmakeconfig.h.cmake:

Source/JavaScriptCore:

  • inspector/scripts/CodeGeneratorInspector.py:

Source/WebCore:

  • bindings/js/JSDOMWindowBase.cpp:

(WebCore::JSDOMWindowBase::supportsProfiling):
(WebCore::JSDOMWindowBase::supportsRichSourceInfo):

  • bindings/js/PageScriptDebugServer.cpp:
  • bindings/js/PageScriptDebugServer.h:
  • bindings/js/ScriptProfile.cpp:
  • bindings/js/ScriptProfile.h:
  • bindings/js/ScriptProfileNode.h:
  • bindings/js/ScriptProfiler.cpp:
  • bindings/js/ScriptProfiler.h:
  • bindings/js/WorkerScriptDebugServer.cpp:
  • bindings/js/WorkerScriptDebugServer.h:
  • inspector/InspectorConsoleAgent.h:
  • inspector/InspectorConsoleInstrumentation.h:

(WebCore::InspectorInstrumentation::getCurrentUserInitiatedProfileName):

  • inspector/InspectorController.cpp:

(WebCore::InspectorController::InspectorController):
(WebCore::InspectorController::resume):

  • inspector/InspectorController.h:
  • inspector/InspectorDOMDebuggerAgent.cpp:
  • inspector/InspectorDOMDebuggerAgent.h:
  • inspector/InspectorHeapProfilerAgent.cpp:
  • inspector/InspectorHeapProfilerAgent.h:
  • inspector/InspectorInstrumentation.cpp:

(WebCore::InspectorInstrumentation::didClearWindowObjectInWorldImpl):
(WebCore::InspectorInstrumentation::isDebuggerPausedImpl):
(WebCore::InspectorInstrumentation::willInsertDOMNodeImpl):
(WebCore::InspectorInstrumentation::didInsertDOMNodeImpl):
(WebCore::InspectorInstrumentation::willRemoveDOMNodeImpl):
(WebCore::InspectorInstrumentation::didRemoveDOMNodeImpl):
(WebCore::InspectorInstrumentation::willModifyDOMAttrImpl):
(WebCore::InspectorInstrumentation::didInvalidateStyleAttrImpl):
(WebCore::InspectorInstrumentation::willSendXMLHttpRequestImpl):
(WebCore::InspectorInstrumentation::scriptExecutionBlockedByCSPImpl):
(WebCore::InspectorInstrumentation::didCommitLoadImpl):
(WebCore::InspectorInstrumentation::addMessageToConsoleImpl):
(WebCore::InspectorInstrumentation::profilerEnabledImpl):
(WebCore::InspectorInstrumentation::willEvaluateWorkerScript):
(WebCore::InspectorInstrumentation::pauseOnNativeEventIfNeeded):
(WebCore::InspectorInstrumentation::cancelPauseOnNativeEvent):

  • inspector/InspectorInstrumentation.h:
  • inspector/InspectorProfilerAgent.cpp:
  • inspector/InspectorProfilerAgent.h:
  • inspector/InstrumentingAgents.cpp:

(WebCore::InstrumentingAgents::InstrumentingAgents):
(WebCore::InstrumentingAgents::reset):

  • inspector/InstrumentingAgents.h:
  • inspector/PageDebuggerAgent.cpp:
  • inspector/PageDebuggerAgent.h:
  • inspector/ScriptProfile.idl:
  • inspector/ScriptProfileNode.idl:
  • inspector/WebDebuggerAgent.cpp:
  • inspector/WebDebuggerAgent.h:
  • inspector/WorkerDebuggerAgent.cpp:
  • inspector/WorkerDebuggerAgent.h:
  • inspector/WorkerInspectorController.cpp:

(WebCore::WorkerInspectorController::WorkerInspectorController):
(WebCore::WorkerInspectorController::resume):

  • inspector/WorkerInspectorController.h:
  • inspector/WorkerRuntimeAgent.cpp:

(WebCore::WorkerRuntimeAgent::pauseWorkerGlobalScope):

  • inspector/WorkerRuntimeAgent.h:
  • loader/FrameLoader.cpp:

(WebCore::FrameLoader::continueLoadAfterNavigationPolicy):

  • page/Console.cpp:
  • page/Console.h:
  • page/Console.idl:
  • testing/Internals.cpp:

(WebCore::Internals::resetToConsistentState):

  • workers/WorkerMessagingProxy.cpp:

(WebCore::connectToWorkerGlobalScopeInspectorTask):
(WebCore::WorkerMessagingProxy::connectToInspector):
(WebCore::disconnectFromWorkerGlobalScopeInspectorTask):
(WebCore::WorkerMessagingProxy::disconnectFromInspector):
(WebCore::dispatchOnInspectorBackendTask):
(WebCore::WorkerMessagingProxy::sendMessageToInspector):

Source/WebKit/gtk:

  • webkit/webkitwebinspector.cpp:

(webkit_web_inspector_set_property):
(webkit_web_inspector_get_property):

Source/WebKit2:

  • WebProcess/WebPage/WebInspector.cpp:

(WebKit::WebInspector::startJavaScriptDebugging):
(WebKit::WebInspector::stopJavaScriptDebugging):
(WebKit::WebInspector::setJavaScriptProfilingEnabled):
(WebKit::WebInspector::startJavaScriptProfiling):
(WebKit::WebInspector::stopJavaScriptProfiling):

Source/WTF:

  • wtf/FeatureDefines.h:

Tools:

  • Scripts/webkitperl/FeatureList.pm:
9:36 AM Changeset in webkit [163023] by eric.carlson@apple.com
  • 2 edits in trunk/Source/WebCore

[iOS] cleanup MediaPlayer.cpp
​https://bugs.webkit.org/show_bug.cgi?id=127821

Reviewed by Jer Noble.

Don't register MediaPlayerPrivateIOS unless the runtime setting for media player proxy is set.

  • platform/graphics/MediaPlayer.cpp: Don't define PlatformMediaEngineClassName on iOS

(WebCore::installedMediaEngines): Drive by cleanup to return early if registration has

already happened.

(WebCore::MediaPlayer::MediaPlayer): Don't force-register the first media engine unless

isVideoPluginProxyEnabled returns true.

9:03 AM Changeset in webkit [163022] by commit-queue@webkit.org
  • 9 edits
    2 adds in trunk

Source/WebCore: Have XHR.getResponseHeader() return null and XHR.getAllResponseHeader() return the empty string in initial ready states
​https://bugs.webkit.org/show_bug.cgi?id=125840

Patch by Youenn Fablet <​youennf@gmail.com> on 2014-01-29
Reviewed by Alexey Proskuryakov.

Merging ​https://chromium.googlesource.com/chromium/blink/+/d201caf874a0bd6f101f517462b3cf1d8c5fce3d
This patch makes it clear that null/empty string is returned whenever the error flag is set.
This new code path is covered by the added test.

Test: http/tests/xmlhttprequest/response-access-on-error.html

  • xml/XMLHttpRequest.cpp:

(WebCore::XMLHttpRequest::getAllResponseHeaders):
(WebCore::XMLHttpRequest::getResponseHeader):

  • xml/XMLHttpRequest.h:
  • xml/XMLHttpRequest.idl:

LayoutTests: Have XHR.getResponseHeader() return null and XHR.getAllResponseHeaders() return empty string in initial ready states
​https://bugs.webkit.org/show_bug.cgi?id=125840

Patch by Youenn Fablet <​youennf@gmail.com> on 2014-01-29
Reviewed by Alexey Proskuryakov.

Merging ​https://chromium.googlesource.com/chromium/blink/+/d201caf874a0bd6f101f517462b3cf1d8c5fce3d
The new test exercices the new code path in case xhr received HTTP headers but m_error is set

  • http/tests/xmlhttprequest/getAllResponseHeaders-expected.txt:
  • http/tests/xmlhttprequest/getAllResponseHeaders.html:
  • http/tests/xmlhttprequest/getResponseHeader-expected.txt:
  • http/tests/xmlhttprequest/getResponseHeader.html:
  • http/tests/xmlhttprequest/response-access-on-error-expected.txt: Added.
  • http/tests/xmlhttprequest/response-access-on-error.html: Added.
8:26 AM Changeset in webkit [163021] by Antti Koivisto
  • 33 edits in trunk

REGRESSION (r162947): Repaint test results are different between WK1 and WK2
​https://bugs.webkit.org/show_bug.cgi?id=127814

Source/WebCore:

Reviewed by Anders Carlsson.

  • page/FrameView.cpp:

(WebCore::FrameView::repaintContentRectangle):

Move repaint rect logging to RenderView.

  • rendering/RenderView.cpp:

(WebCore::RenderView::repaintViewRectangle):

Record raw repaint rects instead of optimized ones.

LayoutTests:

Revert test results changes. We now always log raw repaint rects which are same between platforms.

  • compositing/repaint/resize-repaint-expected.txt:
  • fast/box-shadow/shadow-repaint-expected.txt:
  • fast/css-grid-layout/grid-element-change-columns-repaint-expected.txt:
  • fast/css-grid-layout/grid-element-change-rows-repaint-expected.txt:
  • fast/css-grid-layout/grid-item-change-column-repaint-expected.txt:
  • fast/css-grid-layout/grid-item-change-row-repaint-expected.txt:
  • fast/regions/repaint/element-in-named-flow-absolute-from-fixed-expected.txt:
  • fast/regions/repaint/element-in-named-flow-fixed-from-absolute-expected.txt:
  • fast/regions/repaint/element-inflow-fixed-from-outflow-static-expected.txt:
  • fast/regions/repaint/element-outflow-static-from-inflow-fixed-expected.txt:
  • fast/regions/repaint/invalid-region-repaint-crash-expected.txt:
  • fast/regions/repaint/repaint-element-inside-relative-region-expected.txt:
  • fast/regions/repaint/repaint-regions-overflow-expected.txt:
  • fast/repaint/background-shorthand-with-gradient-and-height-changes-expected.txt:
  • fast/repaint/background-shorthand-with-gradient-and-height-changes.html:
  • fast/repaint/horizontal-bt-overflow-child-expected.txt:
  • fast/repaint/horizontal-bt-overflow-parent-expected.txt:
  • fast/repaint/horizontal-bt-overflow-same-expected.txt:
  • fast/repaint/negative-text-indent-with-overflow-hidden-expected.txt:
  • fast/repaint/overflow-flipped-writing-mode-table-expected.txt:
  • fast/repaint/reflection-table-layout-expected.txt:
  • fast/repaint/transform-table-layout-expected.txt:
  • fast/repaint/vertical-overflow-child-expected.txt:
  • fast/repaint/vertical-overflow-parent-expected.txt:
  • fast/repaint/vertical-overflow-same-expected.txt:
  • platform/mac-wk2/TestExpectations:
  • platform/mac/compositing/repaint/invalidations-on-composited-layers-expected.txt:
  • platform/mac/fast/images/repaint-subrect-grid-expected.txt:
  • svg/repaint/repaint-webkit-svg-shadow-container-expected.txt:
8:17 AM Changeset in webkit [163020] by beidson@apple.com
  • 10 edits in trunk/Source

IDB: Serialize IDBKeyDatas to disk, not IDBKeys
​https://bugs.webkit.org/show_bug.cgi?id=127829

Reviewed by Tim Horton.

Source/WebCore:

Move encode/decode from IDBKey to IDBKeyData.

  • Modules/indexeddb/IDBKey.cpp:
  • Modules/indexeddb/IDBKey.h:
  • Modules/indexeddb/IDBKeyData.cpp:

(WebCore::IDBKeyData::IDBKeyData):
(WebCore::IDBKeyData::encode):
(WebCore::IDBKeyData::decode):

  • Modules/indexeddb/IDBKeyData.h:
  • WebCore.exp.in:

Source/WebKit2:

Serialize IDBKeyDatas, not IDBKeys.

  • DatabaseProcess/IndexedDB/IDBSerialization.cpp:

(WebKit::deserializeIDBKeyPath):
(WebKit::serializeIDBKeyData):

  • DatabaseProcess/IndexedDB/IDBSerialization.h:
  • DatabaseProcess/IndexedDB/sqlite/UniqueIDBDatabaseBackingStoreSQLite.cpp:

(WebKit::UniqueIDBDatabaseBackingStoreSQLite::extractExistingMetadata):
(WebKit::UniqueIDBDatabaseBackingStoreSQLite::putRecord):
(WebKit::UniqueIDBDatabaseBackingStoreSQLite::getKeyRecordFromObjectStore):
(WebKit::UniqueIDBDatabaseBackingStoreSQLite::getKeyRangeRecordFromObjectStore):

8:11 AM Changeset in webkit [163019] by Antti Koivisto
  • 5 edits in trunk

REGRESSION (r162947): css3/flexbox/multiline-justify-content.html and css3/flexbox/position-absolute-child.html are timing out
​https://bugs.webkit.org/show_bug.cgi?id=127809

Reviewed by Anders Carlsson.

Source/WebCore:

These tests generate very large number of small repaint rectangles that overwhelm the region code.

  • page/FrameView.cpp:

(WebCore::FrameView::repaintContentRectangle):

  • platform/graphics/Region.h:

(WebCore::Region::gridSize):
(WebCore::Region::Shape::gridSize):

Add accessor for getting the current region grid complexity.

  • rendering/RenderView.cpp:

(WebCore::RenderView::repaintViewRectangle):

If the region gets very complex merge the repaint rects into a single big rectangle.

LayoutTests:

  • platform/mac/TestExpectations:
8:11 AM WebKitGTK/2.2.x edited by Carlos Garcia Campos
(diff)
7:20 AM Changeset in webkit [163018] by stavila@adobe.com
  • 7 edits
    4 adds in trunk

[CSSRegions] Unable to scroll a scrollable container for regions using mouse wheel
​https://bugs.webkit.org/show_bug.cgi?id=123886

Source/WebCore:

When an element flowed into a scrollable region is scrolled using the mouse wheel, the event
needs to be propagated to the region containing that element, on top of which the cursor
is located.

Reviewed by Antti Koivisto.

Tests: fast/regions/wheel-scroll-abspos.html

fast/regions/wheel-scroll.html

  • page/EventHandler.cpp:

(WebCore::scrollNode):
(WebCore::EventHandler::defaultWheelEventHandler):

  • rendering/RenderBox.cpp:

(WebCore::RenderBox::scroll):
(WebCore::RenderBox::scrollWithWheelEventLocation):

  • rendering/RenderBox.h:
  • rendering/RenderFlowThread.cpp:

(WebCore::RenderFlowThread::regionFromAbsolutePointAndBox):

  • rendering/RenderFlowThread.h:

LayoutTests:

Added tests for scrolling elements flowed into scrollable regions.

Reviewed by Antti Koivisto.

  • fast/regions/wheel-scroll-abspos-expected.html: Added.
  • fast/regions/wheel-scroll-abspos.html: Added.
  • fast/regions/wheel-scroll-expected.html: Added.
  • fast/regions/wheel-scroll.html: Added.
6:52 AM Changeset in webkit [163017] by Carlos Garcia Campos
  • 3 edits in trunk/Source/WebCore

REGRESSION(r162922): [SOUP] Several tests are failing in EFL and GTK+ after r162922
​https://bugs.webkit.org/show_bug.cgi?id=127836

Reviewed by Gustavo Noronha Silva.

The problem is that I assumed that several things done in
ResourceRequest::soupURI() were always desired, so I moved them to
URL. This patch restores the previous behaviour, so that those
tings are only done right before making a request.

  • platform/network/soup/ResourceRequestSoup.cpp:

(WebCore::ResourceRequest::createSoupURI):

  • platform/soup/URLSoup.cpp:

(WebCore::URL::createSoupURI):

6:24 AM Changeset in webkit [163016] by k.czech@samsung.com
  • 2 edits in trunk/LayoutTests

[EFL][WK2] REGRESSION (r162586-r162589): accessibility/range-alter-by-percent.html failing
​https://bugs.webkit.org/show_bug.cgi?id=127724

Unreviewed EFL gardening. Passes after r162932.

  • platform/efl/TestExpectations: Test is no longer failing.
5:57 AM Changeset in webkit [163015] by commit-queue@webkit.org
  • 2 edits in trunk/Source/WebInspectorUI

Web Inspector: [REGRESSION(r162931)] Tab navigation broken in DataGrid
​https://bugs.webkit.org/show_bug.cgi?id=127816

Patch by Diego Pino Garcia <Diego Pino Garcia> on 2014-01-29
Reviewed by Timothy Hatcher.

  • UserInterface/DataGrid.js:

(WebInspector.DataGrid.prototype.moveToNextIfNeeded):
(WebInspector.DataGrid.prototype._editingCommitted):
Change columnIdentifier values 0 and 1 to strings "0" and "1".

5:16 AM Changeset in webkit [163014] by k.czech@samsung.com
  • 16 edits in trunk

[ATK] Expose aria-describedby with ATK_RELATION_DESCRIBED_BY
​https://bugs.webkit.org/show_bug.cgi?id=121684

Reviewed by Mario Sanchez Prada.

Source/WebCore:

Exposed aria-describedby by ATK_RELATION_DESCRIBED_BY.
Moved elementsFromAttribut to AccessibilityObject to have common interface
for AccessibilityNodeObject and AccessibilityRenderObject. Implemented
supportsARIADescribedBy and ariaDescribedByElements to better deal with aria-describedby attribute.

No new tests. Covered by existed one.

  • accessibility/AccessibilityNodeObject.cpp:
  • accessibility/AccessibilityNodeObject.h:
  • accessibility/AccessibilityObject.cpp:

(WebCore::AccessibilityObject::elementsFromAttribute):

  • accessibility/AccessibilityObject.h:

(WebCore::AccessibilityObject::supportsARIADescribedBy):
(WebCore::AccessibilityObject::ariaDescribedByElements):

  • accessibility/AccessibilityRenderObject.cpp:

(WebCore::AccessibilityRenderObject::supportsARIADescribedBy):
(WebCore::AccessibilityRenderObject::ariaDescribedByElements):

  • accessibility/AccessibilityRenderObject.h:
  • accessibility/atk/WebKitAccessibleWrapperAtk.cpp:

(setAtkRelationSetFromCoreObject):

Tools:

Implementing helpText by using relation ATK_RELATION_DESCRIBED_BY.

  • DumpRenderTree/atk/AccessibilityUIElementAtk.cpp:

(AccessibilityUIElement::helpText):

  • WebKitTestRunner/InjectedBundle/atk/AccessibilityUIElementAtk.cpp:

(WTR::AccessibilityUIElement::helpText):

LayoutTests:

Slightly extended test so that it could test aria-describedby with multiple id references.

  • accessibility/aria-describedby-on-input-expected.txt:
  • accessibility/aria-describedby-on-input.html:
  • platform/efl/TestExpectations:
  • platform/gtk/TestExpectations:
5:00 AM WebKitGtkLayoutTests edited by clopez@igalia.com
Add information about running the tests when the system uses … (diff)
2:56 AM Changeset in webkit [163013] by svillar@igalia.com
  • 11 edits in trunk

[CSS Grid Layout] minmax() should be a CSSFunction instead of a CSSValueList
​https://bugs.webkit.org/show_bug.cgi?id=127057

Source/WebCore:

Reviewed by Antti Koivisto.

We were incorrectly dumping minmax(x,y) as "x y" instead of
"minmax(x,y)". That's because we were treating minmax() internally
as a CSSValueList instead of a CSSFunction. Replaced that and also
added some extra information to our tests in order to check that
we don't regress. From now on the CSS grid layout testing helper
functions print not only the computed style but also the contents
of element.style.webkitGridDefinition{Columns|Rows}.

  • css/CSSFunctionValue.h:

(WebCore::CSSFunctionValue::arguments):

  • css/CSSParser.cpp:

(WebCore::CSSParser::parseGridTrackList):
(WebCore::CSSParser::parseGridTrackRepeatFunction):
(WebCore::CSSParser::parseGridTrackSize):

  • css/CSSParser.h:
  • css/StyleResolver.cpp:

(WebCore::createGridTrackSize):

LayoutTests:

Include the output of element.style.webkitGridDefinition{Columns|Rows}
as part of the expected output.

  • fast/css-grid-layout/grid-columns-rows-get-set-expected.txt:
  • fast/css-grid-layout/grid-columns-rows-get-set-multiple-expected.txt:
  • fast/css-grid-layout/named-grid-line-get-set-expected.txt:
  • fast/css-grid-layout/named-grid-line-get-set.html:
  • fast/css-grid-layout/resources/grid-definitions-parsing-utils.js:

(testGridDefinitionsSetJSValues):
(testNonGridDefinitionsSetJSValues):
(checkGridDefinitionsSetJSValues):

1:32 AM Changeset in webkit [163012] by rniwa@webkit.org
  • 6 edits in trunk/Source/WebCore

Rename notifyRendererOfSelectionChange
​https://bugs.webkit.org/show_bug.cgi?id=127831

Reviewed by Antti Koivisto.

Renamed notifyRendererOfSelectionChange to updateSelectionCachesIfSelectionIsInsideTextFormControl.

  • WebCore.order: Removed the exported symbol name as it's not used in WebKit or WebKit2 code.
  • editing/FrameSelection.cpp:

(WebCore::FrameSelection::setSelection):
(WebCore::FrameSelection::selectAll):
(WebCore::FrameSelection::updateSelectionCachesIfSelectionIsInsideTextFormControl): Don't update the style since the existence
of text form control's shadow DOM no longer depends on renderer.

  • editing/FrameSelection.h:
  • html/HTMLTextFormControlElement.cpp:

(WebCore::HTMLTextFormControlElement::selectionChanged): Don't check renderer() for the same reason.

  • page/EventHandler.cpp:

(WebCore::EventHandler::handleMouseReleaseEvent):

12:31 AM Changeset in webkit [163011] by Carlos Garcia Campos
  • 2 edits in trunk/Source/WebKit2

Unreviewed. Fix GTK+ build after r162981.

  • GNUmakefile.list.am: Add WebProcess/WebPage/WebDocumentLoader to

the compilation.

12:23 AM Changeset in webkit [163010] by Carlos Garcia Campos
  • 2 edits in trunk/Source/WebKit2

Unreviewed. Actually fix the GTK+ build after r162967.

  • UIProcess/API/gtk/WebKitWebView.cpp:

(webkit_web_view_can_go_back):
(webkit_web_view_can_go_forward):

Jan 28, 2014:

11:59 PM Changeset in webkit [163009] by jinwoo7.song@samsung.com
  • 2 edits in trunk/LayoutTests

Unreviewed EFL gardening after r162972.

  • platform/efl/fast/table/quote-text-around-iframe-expected.txt:
11:58 PM Changeset in webkit [163008] by zandobersek@gmail.com
  • 2 edits in trunk/Source/WebKit2

Unreviewed followup to r163006. That change incorrectly modified
webkit_web_view_go_forward when webkit_web_view_can_go_forward needed fixing.

  • UIProcess/API/gtk/WebKitWebView.cpp:

(webkit_web_view_go_forward):
(webkit_web_view_can_go_forward):

11:52 PM Changeset in webkit [163007] by mmaxfield@apple.com
  • 2 edits in trunk/LayoutTests

Rebaseline quote-text-around-iframe.html after r162972
​https://bugs.webkit.org/show_bug.cgi?id=127822

Reviewed by Alexey Proskuryakov.

Forgot to rebaseline fast/table/quote-text-around-iframe.html

  • platform/mac/fast/table/quote-text-around-iframe-expected.txt:
11:21 PM Changeset in webkit [163006] by zandobersek@gmail.com
  • 2 edits in trunk/Source/WebKit2

Unreviewed GTK build fix after r162967.

  • UIProcess/API/gtk/WebKitWebView.cpp:

(webkit_web_view_can_go_back):
(webkit_web_view_go_forward):

10:42 PM Changeset in webkit [163005] by jinwoo7.song@samsung.com
  • 5 edits in trunk/LayoutTests

Unreviewed EFL gardening after r162972.

  • platform/efl/fast/css-generated-content/005-expected.txt:
  • platform/efl/fast/css-generated-content/beforeAfter-interdocument-expected.txt:
  • platform/efl/fast/css-generated-content/nested-tables-with-before-after-content-crash-expected.txt:
  • platform/efl/fast/css-generated-content/no-openclose-quote-expected.txt:
10:29 PM Changeset in webkit [163004] by fpizlo@apple.com
  • 4 edits
    3 adds in branches/jsCStack/Source/JavaScriptCore

FTL should support ArrayPop
​https://bugs.webkit.org/show_bug.cgi?id=127749

Reviewed by Geoffrey Garen.

  • ftl/FTLCapabilities.cpp:

(JSC::FTL::canCompile):

  • ftl/FTLIntrinsicRepository.h:
  • ftl/FTLLowerDFGToLLVM.cpp:

(JSC::FTL::LowerDFGToLLVM::compileNode):
(JSC::FTL::LowerDFGToLLVM::compileArrayPush):
(JSC::FTL::LowerDFGToLLVM::compileArrayPop):

  • tests/stress/array-pop-contiguous.js: Added.

(foo):

  • tests/stress/array-pop-double.js: Added.

(foo):

  • tests/stress/array-pop-int32.js: Added.

(foo):

10:12 PM Changeset in webkit [163003] by timothy_horton@apple.com
  • 2 edits in trunk/Source/WebKit2

Crashes in ViewGestureController::wheelEventWasNotHandledByWebCore
​https://bugs.webkit.org/show_bug.cgi?id=127828
<rdar://problem/15932902>

Reviewed by Dan Bernstein.

  • UIProcess/API/mac/WKView.mm:

(-[WKView _wheelEventWasNotHandledByWebCore:]):
Null check. ViewGestureController only needs unhandled wheel events after it has been created.

10:03 PM Changeset in webkit [163002] by timothy_horton@apple.com
  • 2 edits in trunk/Source/WebKit2

Fix the Mountain Lion build if you have private headers.

  • UIProcess/mac/ViewGestureController.mm:

(WebKit::ViewGestureController::beginSwipeGesture):

9:22 PM Changeset in webkit [163001] by commit-queue@webkit.org
  • 2 edits in trunk/Source/WebKit2

[iOS] UI process crashes whenever web process crashes
​https://bugs.webkit.org/show_bug.cgi?id=127826

Patch by Ian Henderson <​ianh@apple.com> on 2014-01-28
Reviewed by Tim Horton.

  • UIProcess/API/ios/WKContentView.mm:

(-[WKContentView _updateFixedPositionRect]): Null-check the drawing
area before calling size() on it.

9:14 PM Changeset in webkit [163000] by commit-queue@webkit.org
  • 2 edits in trunk/Source/WebKit2

WKRemoteObjectCoder should support more argument types
​https://bugs.webkit.org/show_bug.cgi?id=127824

Patch by Ian Henderson <​ianh@apple.com> on 2014-01-28
Reviewed by Anders Carlsson.

  • Shared/API/Cocoa/WKRemoteObjectCoder.mm:

(encodeInvocation): Support encoding char values.
(decodeInvocationArguments): Support decoding char values.

8:44 PM Changeset in webkit [162999] by jer.noble@apple.com
  • 2 edits in trunk/Source/WebCore

[Mac] Handle NSURLAuthenticationMethodServerTrust challenges from -resourceLoader:shouldWaitForResponseToAuthenticationChallenge:
​https://bugs.webkit.org/show_bug.cgi?id=127806

Reviewed by Eric Carlson.

In addition to the normal "user/password" NSURLAuthenticationChallenges, the
-resourceLoader:shouldWaitForResponseToAuthenticationChallenge: delegate method will
ask us to confirm server certificate chains. Rather than pop up an authentication
dialog (which won't work anyway) we will now just drop such requests and continue
on normally.

  • platform/graphics/avfoundation/objc/MediaPlayerPrivateAVFoundationObjC.mm:

(-[WebCoreAVFLoaderDelegate resourceLoader:shouldWaitForResponseToAuthenticationChallenge:]):

8:18 PM Changeset in webkit [162998] by commit-queue@webkit.org
  • 29 edits in trunk

Unreviewed, rolling out r162987.
​http://trac.webkit.org/changeset/162987
​https://bugs.webkit.org/show_bug.cgi?id=127825

Broke Mountain Lion build (Requested by andersca on #webkit).

Source/JavaScriptCore:

  • inspector/InjectedScriptSource.js:

(.):

  • runtime/JSObject.cpp:

(JSC::JSObject::getOwnPropertyDescriptor):

  • runtime/PropertyDescriptor.cpp:
  • runtime/PropertyDescriptor.h:
  • runtime/PropertySlot.h:

Source/WebCore:

  • bindings/scripts/CodeGeneratorJS.pm:

(GenerateGetOwnPropertySlotBody):
(GenerateAttributesHashTable):
(GenerateImplementation):

  • bindings/scripts/test/JS/JSTestActiveDOMObject.cpp:

(WebCore::jsTestActiveDOMObjectConstructor):

  • bindings/scripts/test/JS/JSTestCustomNamedGetter.cpp:

(WebCore::jsTestCustomNamedGetterConstructor):

  • bindings/scripts/test/JS/JSTestEventConstructor.cpp:

(WebCore::jsTestEventConstructorConstructor):

  • bindings/scripts/test/JS/JSTestEventTarget.cpp:

(WebCore::jsTestEventTargetConstructor):

  • bindings/scripts/test/JS/JSTestException.cpp:

(WebCore::jsTestExceptionConstructor):

  • bindings/scripts/test/JS/JSTestGenerateIsReachable.cpp:

(WebCore::jsTestGenerateIsReachableConstructor):

  • bindings/scripts/test/JS/JSTestInterface.cpp:

(WebCore::jsTestInterfaceConstructor):

  • bindings/scripts/test/JS/JSTestMediaQueryListListener.cpp:

(WebCore::jsTestMediaQueryListListenerConstructor):

  • bindings/scripts/test/JS/JSTestNamedConstructor.cpp:

(WebCore::jsTestNamedConstructorConstructor):

  • bindings/scripts/test/JS/JSTestNode.cpp:

(WebCore::jsTestNodeConstructor):

  • bindings/scripts/test/JS/JSTestObj.cpp:

(WebCore::jsTestObjConstructor):

  • bindings/scripts/test/JS/JSTestOverloadedConstructors.cpp:

(WebCore::jsTestOverloadedConstructorsConstructor):

  • bindings/scripts/test/JS/JSTestSerializedScriptValueInterface.cpp:

(WebCore::jsTestSerializedScriptValueInterfaceConstructor):

  • bindings/scripts/test/JS/JSTestTypedefs.cpp:

(WebCore::jsTestTypedefsConstructor):

  • bindings/scripts/test/JS/JSattribute.cpp:

(WebCore::jsattributeConstructor):

  • bindings/scripts/test/JS/JSreadonly.cpp:

(WebCore::jsreadonlyConstructor):

LayoutTests:

  • fast/dom/Window/replaceable-expected.txt:
  • fast/dom/Window/replaceable.html:
  • js/dom/getOwnPropertyDescriptor-expected.txt:
  • js/resources/getOwnPropertyDescriptor.js:
8:09 PM Changeset in webkit [162997] by enrica@apple.com
  • 8 edits in trunk/Source/WebCore

Add support for ActionSheets in WK2 for iOS.
​https://bugs.webkit.org/show_bug.cgi?id=127586
<rdar://problem/15283667>

Reviewed by Benjamin Poulain.

This patch contains some WebCore changes required to
implement support for ActionSheets.
It adds copyURL implementation for iOS in the Editor
class and fixes some bugs in the iOS specific pasteboard
writing code.
The changes to Localizable.strings are the result of
running update-webkit-localizedstrings script.

  • English.lproj/Localizable.strings:
  • WebCore.exp.in:
  • editing/Editor.cpp:
  • editing/Editor.h:
  • editing/ios/EditorIOS.mm:

(WebCore::Editor::writeImageToPasteboard):

  • platform/ios/PasteboardIOS.mm:

(WebCore::Pasteboard::write):

  • platform/ios/PlatformPasteboardIOS.mm:

(WebCore::PlatformPasteboard::write):

7:24 PM Changeset in webkit [162996] by ryuan.choi@samsung.com
  • 2 edits in trunk/Source/WebKit2

Unreviewed build fix on the cmake based port after r162981

One more build fix.

  • CMakeLists.txt: Added WebDocumentLoader.cpp
7:12 PM Changeset in webkit [162995] by matthew_hanson@apple.com
  • 2 edits in tags/Safari-538.14.1/Source/WebCore

Merge r162994: <rdar://problem/15920823>

7:06 PM Changeset in webkit [162994] by aestes@apple.com
  • 2 edits in trunk/Source/WebCore

[iOS] REGRESSION (r162663): Webpages have strange artifacts or corruption around tile boundaries
​https://bugs.webkit.org/show_bug.cgi?id=127823

Reviewed by Daniel Bates.

Original patch by Simon Fraser.

  • platform/ScrollView.cpp: Specify LegacyIOSDocumentViewRect when calling visibleContentRect().
6:58 PM Changeset in webkit [162993] by ryuan.choi@samsung.com
  • 2 edits in trunk/Source/WebKit2

Unreviewed build fix on the EFL build after r162984

r162984 added wheelEventWasNotHandledByWebCore which is guarded to PLATFORM(MAC).

  • UIProcess/WebPageProxy.cpp:

(WebKit::WebPageProxy::didReceiveEvent):

6:42 PM Changeset in webkit [162992] by fpizlo@apple.com
  • 2 edits
    1 add in branches/jsCStack/Source/JavaScriptCore

DFG ArrayPop double array mishandles the NaN hole installation
​https://bugs.webkit.org/show_bug.cgi?id=127813

Reviewed by Mark Rowe.

Our object model for arrays inferred double dictates that we use quiet NaN (QNaN) to
mark holes. Holes, in this context, are any entries in the allocated array buffer
(i.e. from index 0 up to the vectorLength) that don't currently hold a value. Popping
creates a hole, since it deletes the value at publicLength - 1.

But, because of some sloppy copy-and-paste, we were storing (int64_t)0 when creating
the hole, instead of storing QNaN. That's likely because for other kinds of arrays,
64-bit zero is the hole marker, instead of QNaN.

The attached test case illustrates the problem. In the LLInt and Baseline JIT, the
result returned from foo() is "1.5,2.54.5", since array.pop() removes 3.5 and
replaces it with a hole and then the assignment "array[3] = 4.5" creates an element
just beyond that hole. But, once we tier-up to the DFG, the result previously became
"1.5,2.5,0,4.5", which is wrong. The 0 appeared because the IEEE double
interpretation of 64-bit zero is simply zero.

This patch fixes that problem. Now the DFG agrees with the other engines.

This patch also fixes style. For some reason that copy-pasted code wasn't even
indented correctly.

  • dfg/DFGSpeculativeJIT64.cpp:

(JSC::DFG::SpeculativeJIT::compile):

  • tests/stress/array-pop-double-hole.js: Added.

(foo):

5:44 PM Changeset in webkit [162991] by andersca@apple.com
  • 2 edits in trunk/Source/WebKit2

Fix a thinko.

  • UIProcess/WebPageProxy.cpp:

(WebKit::WebPageProxy::goBack):
Remove early return if isValid() is true.

5:39 PM Changeset in webkit [162990] by andersca@apple.com
  • 4 edits in trunk/Source/WebKit2

WebPage should be in charge of creating document loaders
​https://bugs.webkit.org/show_bug.cgi?id=127818

Reviewed by Dan Bernstein.

  • WebProcess/WebCoreSupport/WebFrameLoaderClient.cpp:

(WebKit::WebFrameLoaderClient::createDocumentLoader):

  • WebProcess/WebPage/WebPage.cpp:

(WebKit::WebPage::createDocumentLoader):

  • WebProcess/WebPage/WebPage.h:
5:22 PM Changeset in webkit [162989] by zoltan@webkit.org
  • 7 edits in trunk

[CSS Shapes] Adjust inset sizing syntax to the latest specification
​https://bugs.webkit.org/show_bug.cgi?id=127785

Reviewed by Bem Jones-Bey.

Source/WebCore:

According to the latest CSS Shapes specification [1], the width arguments of inset should
follow the syntax of the margin shorthand, which let us set all four insets with one, two
or four values. This patch updates the behavior and updates the affected tests.

[1] ​http://dev.w3.org/csswg/css-shapes/#funcdef-inset

Existing tests have been updated.

  • css/CSSParser.cpp:

(WebCore::CSSParser::parseInsetRoundedCorners):
(WebCore::CSSParser::parseBasicShapeInset):

LayoutTests:

  • fast/shapes/parsing/parsing-shape-inside-expected.txt:
  • fast/shapes/parsing/parsing-shape-outside-expected.txt:
  • fast/shapes/parsing/parsing-test-utils.js:
  • fast/shapes/shape-outside-floats/shape-outside-floats-inset.html:
5:19 PM Changeset in webkit [162988] by commit-queue@webkit.org
  • 2 edits in trunk/Source/WebKit2

Unreviewed GTK build fix after r162951.
​https://bugs.webkit.org/show_bug.cgi?id=127815

Patch by Carlos Alberto Lopez Perez <​clopez@igalia.com> on 2014-01-28

  • GNUmakefile.list.am: Adding missing build targets.
5:13 PM Changeset in webkit [162987] by oliver@apple.com
  • 29 edits in trunk

Make DOM attributes appear to be faux accessor properties
​https://bugs.webkit.org/show_bug.cgi?id=127797

Reviewed by Michael Saboff.

Source/JavaScriptCore:

Add flag so we can identify which properties should have the old
custom property semantics vs. the new faux accessors. Update the
inspector protocol accordingly.

These faux accessors produce descriptors with "get" and "set"
properties, but both values are undefined so can't be used
directly. A few custom properties actually require their
existing magical behaviour, so we now have a flag to
distinguish the expected output.

  • inspector/InjectedScriptSource.js:

(.):

  • runtime/JSObject.cpp:

(JSC::JSObject::getOwnPropertyDescriptor):

  • runtime/PropertyDescriptor.cpp:

(JSC::PropertyDescriptor::setCustomDescriptor):

  • runtime/PropertyDescriptor.h:
  • runtime/PropertySlot.h:

Source/WebCore:

Refactor the bindings generator, and make sure we emit
the CustomAccessor flag on properties that should
appear to be accessors.

  • bindings/scripts/CodeGeneratorJS.pm:

(GenerateGetOwnPropertySlotBody):
(GenerateAttributesHashTable):
(GenerateImplementation):

LayoutTests:

These properties now show up as accessors in these tests

  • fast/dom/Window/replaceable-expected.txt:
  • fast/dom/Window/replaceable.html:
  • js/dom/getOwnPropertyDescriptor-expected.txt:
  • js/resources/getOwnPropertyDescriptor.js:
5:05 PM Changeset in webkit [162986] by Chris Fleizach
  • 26 edits
    2 adds in trunk

AX: Support @scope in HTML tables
​https://bugs.webkit.org/show_bug.cgi?id=127688

Reviewed by Mario Sanchez Prada.

Source/WebCore:

Add support for scope attribute so that row headers and column headers
are now exposed at the table cell level where appropriate.

Test: platform/mac/accessibility/table-scope.html

  • accessibility/AccessibilityARIAGridCell.cpp:

(WebCore::AccessibilityARIAGridCell::parentTable):

  • accessibility/AccessibilityARIAGridCell.h:
  • accessibility/AccessibilityARIAGridRow.cpp:

(WebCore::AccessibilityARIAGridRow::parentTable):

  • accessibility/AccessibilityARIAGridRow.h:
  • accessibility/AccessibilityTable.cpp:

(WebCore::AccessibilityTable::columns):
(WebCore::AccessibilityTable::rows):

  • accessibility/AccessibilityTable.h:
  • accessibility/AccessibilityTableCell.cpp:

(WebCore::AccessibilityTableCell::parentTable):
(WebCore::AccessibilityTableCell::isTableHeaderCell):
(WebCore::AccessibilityTableCell::isTableCellInSameRowGroup):
(WebCore::AccessibilityTableCell::isTableCellInSameColGroup):
(WebCore::AccessibilityTableCell::columnHeaders):
(WebCore::AccessibilityTableCell::rowHeaders):

  • accessibility/AccessibilityTableCell.h:
  • accessibility/AccessibilityTableRow.cpp:

(WebCore::AccessibilityTableRow::parentTable):

  • accessibility/AccessibilityTableRow.h:
  • accessibility/mac/WebAccessibilityObjectWrapperMac.mm:

(-[WebAccessibilityObjectWrapper accessibilityAttributeNames]):
(-[WebAccessibilityObjectWrapper accessibilityAttributeValue:]):

Tools:

Add support for rowHeaders and columnHeaders.

  • DumpRenderTree/AccessibilityUIElement.cpp:

(convertElementsToObjectArray):
(columnHeadersCallback):
(rowHeadersCallback):
(uiElementArrayAttributeValueCallback):
(AccessibilityUIElement::columnHeaders):
(AccessibilityUIElement::rowHeaders):
(AccessibilityUIElement::getJSClass):

  • DumpRenderTree/AccessibilityUIElement.h:
  • DumpRenderTree/mac/AccessibilityUIElementMac.mm:

(AccessibilityUIElement::rowHeaders):
(AccessibilityUIElement::columnHeaders):

  • WebKitTestRunner/InjectedBundle/AccessibilityUIElement.h:
  • WebKitTestRunner/InjectedBundle/Bindings/AccessibilityUIElement.idl:
  • WebKitTestRunner/InjectedBundle/atk/AccessibilityUIElementAtk.cpp:

(WTR::AccessibilityUIElement::uiElementArrayAttributeValue):
(WTR::AccessibilityUIElement::rowHeaders):
(WTR::AccessibilityUIElement::columnHeaders):

  • WebKitTestRunner/InjectedBundle/mac/AccessibilityUIElementMac.mm:

(WTR::convertElementsToObjectArray):
(WTR::AccessibilityUIElement::rowHeaders):
(WTR::AccessibilityUIElement::columnHeaders):
(WTR::AccessibilityUIElement::uiElementArrayAttributeValue):

LayoutTests:

  • platform/mac/accessibility/aria-columnrowheaders-expected.txt:
  • platform/mac/accessibility/table-attributes-expected.txt:
  • platform/mac/accessibility/table-cell-spans-expected.txt:
  • platform/mac/accessibility/table-cells-expected.txt:
  • platform/mac/accessibility/table-scope-expected.txt: Added.
  • platform/mac/accessibility/table-scope.html: Added.
  • platform/mac/accessibility/table-sections-expected.txt:
4:59 PM Changeset in webkit [162985] by Brent Fulgham
  • 5 edits in trunk/Source/WebCore

Improve latching behavior for wheel events
​https://bugs.webkit.org/show_bug.cgi?id=127386
<rdar://problem/12176858>

Reviewed by Simon Fraser.

  • page/scrolling/ScrollingTree.cpp:

(WebCore::ScrollingTree::clearLatchedNode): Added
(WebCore::ScrollingTree::latchedNode): Added
(WebCore::ScrollingTree::removeDestroyedNodes): Clear latched node if it's being removed.
(WebCore::ScrollingTree::ScrollingTree): Initialize new value used for tracking
scroll latching state.
(WebCore::ScrollingTree::setLatchedNode): Added
(WebCore::ScrollingTree::setOrClearLatchedNode): Added
Set latched node when beginning a swipe event, or clear latched node when scroll/momentum ends.
(WebCore::ScrollingTree::shouldHandleWheelEventSynchronously): Check for an existing
latched node and stay in fast scrolling mode if possible. If the current event should
start a swipe event, clear the current latched node so we can correctly find and assign
the new latch node.

  • page/scrolling/ScrollingTree.h:

(WebCore::ScrollingTree::hasLatchedNode): Added

  • page/scrolling/mac/ScrollingTreeScrollingNodeMac.mm:

(WebCore::ScrollingTreeScrollingNodeMac::handleWheelEvent): Determine latching state
based on wheel event state and position of mouse pointer in the document.

  • platform/ScrollAnimator.cpp:

(ScrollAnimator::handleWheelEvent): Always treat PlatformWheelEventPhaseMayBegin
as successfully handled so that it does not "bubble back up" to the root of
the scrolling tree.

4:50 PM Changeset in webkit [162984] by timothy_horton@apple.com
  • 11 edits in trunk/Source/WebKit2

WebKit2 View Gestures (Swipe): Give WebCore the first crack at scroll events if it needs it
​https://bugs.webkit.org/show_bug.cgi?id=127396

Reviewed by Anders Carlsson.

  • UIProcess/API/ios/PageClientImplIOS.h:
  • UIProcess/API/ios/PageClientImplIOS.mm:

(WebKit::PageClientImpl::wheelEventWasNotHandledByWebCore):

  • UIProcess/API/mac/PageClientImpl.h:
  • UIProcess/API/mac/PageClientImpl.mm:

(WebKit::PageClientImpl::wheelEventWasNotHandledByWebCore):

  • UIProcess/API/mac/WKView.mm:

(-[WKView _wheelEventWasNotHandledByWebCore:]):

  • UIProcess/API/mac/WKViewInternal.h:
  • UIProcess/PageClient.h:
  • UIProcess/WebPageProxy.cpp:

(WebKit::WebPageProxy::didReceiveEvent):
In addition to informing the page client about un-handled wheel events,
inform the ViewGestureController, in a roundabout way.

  • UIProcess/mac/ViewGestureController.h:

Add and use SwipeDirection enum instead of 'bool willSwipeLeft'.

  • UIProcess/mac/ViewGestureController.mm:

(WebKit::ViewGestureController::ViewGestureController):
(WebKit::ViewGestureController::handleScrollWheelEvent):
If we get a Begin scroll wheel event, any previously deferred
swipes are invalid.

If the page has subframes or wheel handlers, defer starting the
swipe until we get a wheelEventWasNotHandledByWebCore back from the WebProcess.

(WebKit::ViewGestureController::wheelEventWasNotHandledByWebCore):
Start the swipe if the WebProcess didn't eat the event.

(WebKit::ViewGestureController::trackSwipeGesture):
(WebKit::ViewGestureController::beginSwipeGesture):
(WebKit::ViewGestureController::handleSwipeGesture):
Use SwipeDirection.

4:48 PM Changeset in webkit [162983] by matthew_hanson@apple.com
  • 2 edits in tags/Safari-538.14.1/Source/WebKit2

Merge r162977: <rdar://problem/15930116>

4:42 PM Changeset in webkit [162982] by andersca@apple.com
  • 2 edits in trunk/Source/WebKit2

Fix build.

  • WebProcess/WebPage/WebFrame.cpp:
4:39 PM Changeset in webkit [162981] by andersca@apple.com
  • 3 edits
    2 adds in trunk/Source/WebKit2

Add WebDocumentLoader, a WebCore::DocumentLoader subclass
​https://bugs.webkit.org/show_bug.cgi?id=127817

Reviewed by Tim Horton.

We need a place to store navigation IDs, and since they are per document loader, putting them
in a DocumentLoader subclass seems like a good idea.

  • WebKit2.xcodeproj/project.pbxproj:
  • WebProcess/WebCoreSupport/WebFrameLoaderClient.cpp:

(WebKit::WebFrameLoaderClient::createDocumentLoader):

  • WebProcess/WebPage/WebDocumentLoader.cpp: Added.

(WebKit::WebDocumentLoader::WebDocumentLoader):
(WebKit::WebDocumentLoader::setNavigationID):

  • WebProcess/WebPage/WebDocumentLoader.h: Added.

(WebKit::WebDocumentLoader::create):
(WebKit::WebDocumentLoader::navigationID):

4:16 PM Changeset in webkit [162980] by andersca@apple.com
  • 2 edits in trunk/Source/WebKit2

Simplify WebFrame::isMainFrame
​https://bugs.webkit.org/show_bug.cgi?id=127812

Reviewed by Sam Weinig.

  • WebProcess/WebPage/WebFrame.cpp:

(WebKit::WebFrame::isMainFrame):

4:11 PM Changeset in webkit [162979] by matthew_hanson@apple.com
  • 5 edits in tags/Safari-538.14.1/Source

Versioning.

4:07 PM Changeset in webkit [162978] by ap@apple.com
  • 2 edits in trunk/LayoutTests

REGRESSION (r162947): Repaint test results are different between WK1 and WK2
​https://bugs.webkit.org/show_bug.cgi?id=127814

  • platform/mac-wk2/TestExpectations: Marking them as failing for now.
3:55 PM Changeset in webkit [162977] by aestes@apple.com
  • 2 edits in trunk/Source/WebKit2

[iOS] Set INSTALL_PATH_ACTUAL when building WebKit2 for the iOS Simulator
​https://bugs.webkit.org/show_bug.cgi?id=127811

Reviewed by Dan Bernstein.

  • Configurations/WebKit2.xcconfig:
3:51 PM Changeset in webkit [162976] by jinwoo7.song@samsung.com
  • 2 edits in trunk/Source/WebKit2

Unreviewed EFL WebKit2 build fix.

  • CMakeLists.txt: Added Shared/NavigationActionData.cpp
3:49 PM Changeset in webkit [162975] by fpizlo@apple.com
  • 2 edits in branches/jsCStack/Source/JavaScriptCore

FTL should support ArrayPush
​https://bugs.webkit.org/show_bug.cgi?id=127748

Not reviewed, remove some debug code.

  • ftl/FTLLowerDFGToLLVM.cpp:

(JSC::FTL::LowerDFGToLLVM::compileArrayPush):

3:46 PM Changeset in webkit [162974] by ap@apple.com
  • 2 edits in trunk/LayoutTests

REGRESSION (r162947): css3/flexbox/multiline-justify-content.html and
css3/flexbox/position-absolute-child.html are timing out
​https://bugs.webkit.org/show_bug.cgi?id=127809

  • platform/mac/TestExpectations: Skipping for now, Antti is going to take a look tomorrow.
3:43 PM Changeset in webkit [162973] by matthew_hanson@apple.com
  • 1 copy in tags/Safari-538.14.1

New Tag.

3:36 PM Changeset in webkit [162972] by mmaxfield@apple.com
  • 14 edits
    2 adds in trunk

ASSERT_WITH_SECURITY_IMPLICATION in WebCore::InlineTextBox::paint
​https://bugs.webkit.org/show_bug.cgi?id=114586

Reviewed by Dave Hyatt.

Taken mostly from ​https://chromium.googlesource.com/chromium/blink/+/cb2297db16f2e9328cb4dd8b552093d6b22340a8

If RenderQuote is a subclass of RenderObject, it can't be split by the first-letter CSS pseudoclass.
Instead, we should make it a subclass of RenderElement, so that it can be split properly.

Source/WebCore:

Test: fast/css-generated-content/quote-first-letter.html

  • dom/PseudoElement.cpp:

(WebCore::PseudoElement::didRecalcStyle):

  • rendering/RenderQuote.cpp:

(WebCore::RenderQuote::RenderQuote):
(WebCore::RenderQuote::willBeDestroyed):
(WebCore::RenderQuote::willBeRemovedFromTree):
(WebCore::RenderQuote::styleDidChange):
(WebCore::RenderQuote::updateText):
(WebCore::RenderQuote::computeText):
(WebCore::RenderQuote::updateDepth):

  • rendering/RenderQuote.h:
  • rendering/style/ContentData.cpp:

(WebCore::QuoteContentData::createContentRenderer):

LayoutTests:

This adds a test to make sure that the splitting behavior occurs, as well as updates existing tests that
didn't use the splitting behavior.

  • fast/css-generated-content/quote-first-letter-expected.html: Added.
  • fast/css-generated-content/quote-first-letter.html: Added.
  • platform/mac/fast/css-generated-content/005-expected.txt:
  • platform/mac/fast/css-generated-content/beforeAfter-interdocument-expected.txt:
  • platform/mac/fast/css-generated-content/nested-tables-with-before-after-content-crash-expected.txt:
  • platform/mac/fast/css-generated-content/no-openclose-quote-expected.txt:
3:30 PM Changeset in webkit [162971] by dbates@webkit.org
  • 2 edits in trunk/Tools

Teach Buildbot dashboard to parse alternative revision format
​https://bugs.webkit.org/show_bug.cgi?id=127561

Reviewed by Alexey Proskuryakov.

Add support for extracting the OpenSource and Internal revision numbers when Buildbot
returns a dictionary for the value of property got_revision.

  • BuildSlaveSupport/build.webkit.org-config/public_html/dashboard/Scripts/BuildbotIteration.js:

(isMultiCodebaseGotRevisionProperty): Added.
(parseRevisionProperty): Added; parses Buildbot revision property, which can either be a
string or a dictionary.
(BuildbotIteration.prototype.update.): Modified to call parseRevisionProperty() to determine
the OpenSource and Internal revision numbers.

3:29 PM Changeset in webkit [162970] by mark.lam@apple.com
  • 21 edits in trunk/Source/JavaScriptCore

Remove some unneeded debugger code.
​https://bugs.webkit.org/show_bug.cgi?id=127805.

Reviewed by Oliver Hunt.

JSC will now always support the debugger. Hence, the #if ENABLE(JAVASCRIPT_DEBUGGER)
checks can be removed.

DebuggerCallFrame::callFrame() is also unused and will be removed.

  • debugger/Breakpoint.h:
  • debugger/Debugger.cpp:
  • debugger/DebuggerCallFrame.h:
  • inspector/InjectedScript.cpp:

(Inspector::InjectedScript::wrapCallFrames):

  • inspector/InjectedScript.h:
  • inspector/JSGlobalObjectScriptDebugServer.cpp:
  • inspector/JSGlobalObjectScriptDebugServer.h:
  • inspector/JSJavaScriptCallFrame.cpp:
  • inspector/JSJavaScriptCallFrame.h:
  • inspector/JSJavaScriptCallFramePrototype.cpp:
  • inspector/JSJavaScriptCallFramePrototype.h:
  • inspector/JavaScriptCallFrame.cpp:
  • inspector/JavaScriptCallFrame.h:
  • inspector/ScriptDebugListener.h:
  • inspector/ScriptDebugServer.cpp:
  • inspector/ScriptDebugServer.h:
  • inspector/agents/InspectorDebuggerAgent.cpp:
  • inspector/agents/InspectorDebuggerAgent.h:
  • inspector/agents/InspectorRuntimeAgent.cpp:

(Inspector::InspectorRuntimeAgent::InspectorRuntimeAgent):
(Inspector::setPauseOnExceptionsState):
(Inspector::InspectorRuntimeAgent::evaluate):
(Inspector::InspectorRuntimeAgent::callFunctionOn):
(Inspector::InspectorRuntimeAgent::getProperties):

  • inspector/agents/InspectorRuntimeAgent.h:
3:22 PM Changeset in webkit [162969] by fpizlo@apple.com
  • 5 edits
    3 adds in branches/jsCStack/Source/JavaScriptCore

FTL should support ArrayPush
​https://bugs.webkit.org/show_bug.cgi?id=127748

Reviewed by Oliver Hunt.

  • ftl/FTLAbstractHeapRepository.h:

(JSC::FTL::AbstractHeapRepository::forArrayType):

  • ftl/FTLCapabilities.cpp:

(JSC::FTL::canCompile):

  • ftl/FTLIntrinsicRepository.h:
  • ftl/FTLLowerDFGToLLVM.cpp:

(JSC::FTL::LowerDFGToLLVM::compileNode):
(JSC::FTL::LowerDFGToLLVM::compileArrayPush):

  • tests/stress/array-push-contiguous.js: Added.

(foo):

  • tests/stress/array-push-double.js: Added.

(foo):

3:12 PM Changeset in webkit [162968] by ggaren@apple.com
  • 4 edits in trunk/Source/JavaScriptCore

REGRESSION: JavascriptCore crash during OS Installation (due to
Heap::m_operationInProgress ASSERT vs DelayedReleaseScope)
​https://bugs.webkit.org/show_bug.cgi?id=127793

Reviewed by Mark Hahnenberg.

This was a mistaken ASSERT.

  • API/tests/testapi.mm:

(-[EvilAllocationObject doEvilThingsWithContext:]): Added a test to verify
that GC from a DelayedReleaseScope doesn't crash.

  • heap/DelayedReleaseScope.h:

(JSC::DelayedReleaseScope::~DelayedReleaseScope): Our contract is that
it is valid to do anything while running a DelayedReleaseScope -dealloc
method, so the Heap must be ready for new allocations and collections.

Change the Heap's operationInProgress value to NoOperation while running
-dealloc methods, so that it doesn't ASSERT in the face of new allocations
and collections.

  • heap/Heap.h: Made DelayedReleaseScope a friend because exposing a setter

for m_operationInProgress seemed like the worse of the two options for
encapsulation: we don't really want arbitrary clients to set the Heap's
m_operationInProgress.

3:04 PM Changeset in webkit [162967] by andersca@apple.com
  • 5 edits in trunk/Source/WebKit2

Remove WebPageProxy::canGoBack and WebPageProxy::canGoForward
​https://bugs.webkit.org/show_bug.cgi?id=127803

Reviewed by Andreas Kling.

  • UIProcess/API/C/WKPage.cpp:

(WKPageCanGoForward):
(WKPageCanGoBack):

  • UIProcess/API/Cocoa/WKBrowsingContextController.mm:

(-[WKBrowsingContextController canGoForward]):
(-[WKBrowsingContextController canGoBack]):

  • UIProcess/WebPageProxy.cpp:

(WebKit::WebPageProxy::goForward):
(WebKit::WebPageProxy::goBack):
(WebKit::WebPageProxy::sendWheelEvent):

  • UIProcess/WebPageProxy.h:
3:02 PM Changeset in webkit [162966] by Joseph Pecoraro
  • 2 edits in trunk/Source/WebInspectorUI

Unreviewed follow-up to r162965. Better coding style.

Timothy Hatcher pointed out a clearer style for classes
on the <body> is to specify body.class. I agree.

  • UserInterface/DashboardView.css:

(body.javascript .toolbar .dashboard):
(body.web .toolbar.collapsed .dashboard > .logs):
(body.javascript .toolbar .dashboard > .resourcesCount):

2:47 PM Changeset in webkit [162965] by Joseph Pecoraro
  • 6 edits in trunk/Source/WebInspectorUI

Web Inspector: Tweak UI when debugging a JSContext
​https://bugs.webkit.org/show_bug.cgi?id=127799

Reviewed by Timothy Hatcher.

  • UserInterface/DashboardView.css:

(.toolbar.javascript .dashboard):
(.toolbar.web.collapsed .dashboard > .logs):
(.toolbar.javascript .dashboard > .resourcesCount):
Adjust collapsed dashboard in web and javascript debugging modes.
JavaScript debugging mode is always treated as collapsed.

  • UserInterface/Main.js:

(WebInspector.contentLoaded):
Adjust what is created and shown when in javascript debugging mode.

  • UserInterface/NavigationSidebarPanel.css:

(.navigation-sidebar-panel-content-tree-outline.hide-disclosure-buttons .item.small):

  • UserInterface/ResourceSidebarPanel.js:

(WebInspector.ResourceSidebarPanel):
(WebInspector.ResourceSidebarPanel.prototype._scriptWasAdded):

  • UserInterface/TimelineSidebarPanel.css:

(.sidebar > .panel.timeline > .status-bar):
(.sidebar > .panel.timeline > .title-bar):
When javascript debugging expect to put things at the root level, so hide
disclosure buttons and adjust the style so resources look nice here.

2:46 PM Changeset in webkit [162964] by andersca@apple.com
  • 4 edits in trunk/Source/WebKit2

API::PolicyClient::decidePolicyForNavigationAction should take a NavigationActionData object
​https://bugs.webkit.org/show_bug.cgi?id=127798

Reviewed by Tim Horton.

  • UIProcess/API/APIPolicyClient.h:

(API::PolicyClient::decidePolicyForNavigationAction):

  • UIProcess/API/C/WKPage.cpp:

(WKPageSetPagePolicyClient):

  • UIProcess/WebPageProxy.cpp:

(WebKit::WebPageProxy::decidePolicyForNavigationAction):

2:44 PM Changeset in webkit [162963] by commit-queue@webkit.org
  • 4 edits in trunk/LayoutTests

Unreviewed GTK gardening.

Remove failure/timeout expectations for the following passed tests:

fast/css3-text/css3-text-decoration/text-decoration-skip/text-decoration-skip-non-intersecting-underline.html
fast/shapes/shape-inside/shape-inside-subpixel-rectangle-top.html
svg/custom/mouse-move-on-svg-container.xhtml
svg/custom/mouse-move-on-svg-root.xhtml
fast/css/style-scoped/style-scoped-with-dom-operation.html

http/tests/download failures only affect gtk-wk1 now

Report and mark flaky tests:

http/tests/w3c/webperf/approved/navigation-timing/html/test_timing_attributes_order.html
fast/spatial-navigation/snav-unit-overflow-and-scroll-in-direction.html
fast/parser/document-write-basic.html
http/tests/w3c/webperf/submission/Google/resource-timing/html/test_resource_dynamic_insertion.html
media/crash-closing-page-with-media-as-plugin-fallback.html

Add timeout to flaky test media/audio-garbage-collect.html

Report and mark failed tests:

accessibility/range-alter-by-percent.html
media/video-restricted-no-preload-auto.html

Patch by Carlos Alberto Lopez Perez <​clopez@igalia.com> on 2014-01-28

  • platform/gtk/TestExpectations:
  • platform/gtk-wk1/TestExpectations:
  • platform/gtk-wk2/TestExpectations:
2:32 PM Changeset in webkit [162962] by Antti Koivisto
  • 8 edits in trunk/Source/WebCore

Document::topDocument() should return a reference
​https://bugs.webkit.org/show_bug.cgi?id=127786

Reviewed by Darin Adler.

  • accessibility/AccessibilityObject.cpp:

(WebCore::AccessibilityObject::mainFrame):
(WebCore::AccessibilityObject::topDocument):

  • accessibility/ios/WebAccessibilityObjectWrapperIOS.mm:

(-[WebAccessibilityObjectWrapper accessibilityContainer]):

  • dom/Document.cpp:

(WebCore::Document::~Document):
(WebCore::Document::createRenderTree):
(WebCore::Document::destroyRenderTree):
(WebCore::Document::clearAXObjectCache):
(WebCore::Document::existingAXObjectCache):
(WebCore::Document::axObjectCache):
(WebCore::Document::implicitClose):
(WebCore::Document::topDocument):
(WebCore::Document::topOrigin):
(WebCore::Document::webkitCancelFullScreen):
(WebCore::Document::webkitDidExitFullScreenForElement):

  • dom/Document.h:
  • page/DOMWindow.cpp:

(WebCore::DOMWindow::incrementScrollEventListenersCount):
(WebCore::DOMWindow::decrementScrollEventListenersCount):

  • rendering/RenderEmbeddedObject.cpp:

(WebCore::RenderEmbeddedObject::isReplacementObscured):

  • rendering/RenderView.cpp:

(WebCore::RenderView::RepaintRegionAccumulator::RepaintRegionAccumulator):

2:25 PM Changeset in webkit [162961] by ap@apple.com
  • 12 edits
    1 delete in trunk/LayoutTests

js/dom/post-message-numeric-property.html is flaky
​https://bugs.webkit.org/show_bug.cgi?id=127713

Reviewed by Geoffrey Garen.

Plain testRunner and js-test-post.js approaches to being asynchronous don't work
together, you need one or the other.

  • js/dom/post-message-numeric-property-expected.txt:
  • js/dom/script-tests/post-message-numeric-property.js:

Updated to use jsTestIsAsync, as the wrapper includes js-test-post.js. The results
are now always printed in correct order, not flakily.

  • js/dom/cross-global-object-inline-global-var-expected.txt:
  • js/dom/script-tests/cross-global-object-inline-global-var.js:

Ditto.

  • js/dom/function-prototype.html:
  • js/dom/parse-error-external-script-in-eval.html:
  • js/dom/parse-error-external-script-in-new-Function.html:
  • js/dom/parse-syntax-error-in-initializer.html:

Removed waitUntilDone, these tests are not asynchronous at all.

  • js/dom/script-tests/cross-frame-bad-time.js:
  • js/dom/script-tests/cross-frame-really-bad-time-with-proto.js:
  • js/dom/script-tests/cross-frame-really-bad-time.js:

Updated to use jsTestIsAsync. These tests were not flaky because they happened to
run synchronously in practice, but document.writing a script into subframe won't
necessarily execute it synchronously in the future.

  • js/resources/fs-test-post.js: Removed. This file was unused.
2:21 PM Changeset in webkit [162960] by commit-queue@webkit.org
  • 5 edits
    4 adds in trunk

Sticky positioning is broken for table rows
​https://bugs.webkit.org/show_bug.cgi?id=112024

Patch by Viatcheslav Ostapenko <​sl.ostapenko@samsung.com> on 2014-01-28
Reviewed by Simon Fraser.

Source/WebCore:

Enable sticky positioning for table elements.
Also includes fix for bug ​wkb.ug/105654
I don't have access to this bug, but from related commit message I
assume it says about "Heap use after free problem".
Debugging showed that it wasn't heap use after free.
The real problem that RenderObject::container() returns object which is
not RenderBlock and it is used as RenderBlock in
RenderBox::containingBlockLogicalHeightForPositioned() method.
Added extra isRenderBlock check and search for containingBlock if
current block is not RenderBlock.

Tests: fast/css/sticky/sticky-table-row-top.html

fast/css/sticky/sticky-table-thead-top.html

  • css/StyleResolver.cpp:

(WebCore::StyleResolver::adjustRenderStyle):

  • rendering/RenderBox.cpp:

(WebCore::RenderBox::containingBlockLogicalHeightForPositioned):

  • rendering/RenderTableRow.h:

LayoutTests:

Add layout tests for sticky positioned table row and table thead.

  • fast/css/sticky/sticky-table-row-top-expected.html: Added.
  • fast/css/sticky/sticky-table-row-top.html: Added.
  • fast/css/sticky/sticky-table-thead-top-expected.html: Added.
  • fast/css/sticky/sticky-table-thead-top.html: Added.
2:20 PM Changeset in webkit [162959] by andersca@apple.com
  • 2 edits
    2 adds in trunk/Source/WebKit2

Add an empty WKNavigation class
​https://bugs.webkit.org/show_bug.cgi?id=127796

Reviewed by Tim Horton.

  • UIProcess/API/Cocoa/WKNavigation.h: Added.
  • UIProcess/API/Cocoa/WKNavigation.mm: Added.
  • WebKit2.xcodeproj/project.pbxproj:
2:10 PM Changeset in webkit [162958] by fpizlo@apple.com
  • 2 edits
    1 add in branches/jsCStack/Source/JavaScriptCore

Arity fixup clobbers callee-saves, causing FTL->FTL arity fixed-up calls to corrupt state
​https://bugs.webkit.org/show_bug.cgi?id=127791

Reviewed by Michael Saboff.

Make the 64-bit path - i.e. the only path where the FTL matters - use regT6 instead
of regT3. regT6 is not a callee save.

  • jit/ThunkGenerators.cpp:

(JSC::arityFixup):

  • tests/stress/ftl-to-ftl-arity-fixup.js: Added.

(foo):
(bar):

2:03 PM Changeset in webkit [162957] by mmaxfield@apple.com
  • 2 edits in trunk/Source/WebCore

Fixing several incorrect assumptions with handling isolated inlines.
​https://bugs.webkit.org/show_bug.cgi?id=127608

Reviewed by Simon Fraser.

Fixing capitalization.

  • rendering/RenderBlockLineLayout.cpp:

(WebCore::setUpResolverToResumeInIsolate):
(WebCore::constructBidiRunsForSegment):

1:53 PM Changeset in webkit [162956] by mmaxfield@apple.com
  • 5 edits
    4 adds in trunk

Fixing several incorrect assumptions with handling isolated inlines.
​https://bugs.webkit.org/show_bug.cgi?id=127608

Reviewed by Dave Hyatt.

Source/WebCore:

First, when an isolated inline spans multiple lines, we aren't guaranteed
to exit BidiResolver with a nested inline count of zero. Removing the
assert that says otherwise.

Previously in constructBidiRunsForSegment, we called bidiFirst in
an attempt to properly setup the isolatedResolver for any dom/style
that applied, but this only worked on the first line the isolated
inline appeared in. Adding a function that approaches this properly
by recursing through the parents of the starting object for the line
and post-fixing direction attributes to the resolver.

Finally, addressing an issue where the line following a removed isolated
inline (with a continuation) failed to be marked dirty.

Merged from Blink: ​https://chromium.googlesource.com/chromium/blink/+/72698f203b1c50900e535b80945563b92b7eef23

Tests: fast/text/nested-bidi-assert.html

fast/text/nested-bidi-with-continuation-crash.html

  • platform/text/BidiResolver.h:

(WebCore::Run>::~BidiResolver):

  • rendering/RenderBlockLineLayout.cpp:

(WebCore::setupResolverToResumeInIsolate):
(WebCore::constructBidiRunsForSegment):

  • rendering/RenderLineBoxList.cpp:

(WebCore::RenderLineBoxList::dirtyLinesFromChangedChild):

LayoutTests:

See file-specific info.

Merged from Blink: ​https://chromium.googlesource.com/chromium/blink/+/72698f203b1c50900e535b80945563b92b7eef23

  • fast/text/nested-bidi-assert-expected.txt: Added.
  • fast/text/nested-bidi-assert.html: Added. Tests that no crash occurs

when an isolated inline spans several lines

  • fast/text/nested-bidi-with-continuation-crash-expected.txt: Added.
  • fast/text/nested-bidi-with-continuation-crash.html: Added. Tests that

no crash occurs when additional isolates need to be set up for a
continuation

1:42 PM Changeset in webkit [162955] by fpizlo@apple.com
  • 2 edits in branches/jsCStack/Source/JavaScriptCore

FTL should really, *really* support NotCellUse
​https://bugs.webkit.org/show_bug.cgi?id=127790

Reviewed by Mark Hahnenberg.

It's easy to forget about Phantoms. Any use kind that appears in any node can
eventually appear in a Phantom. This was covered by Octane/mandreel.

  • ftl/FTLLowerDFGToLLVM.cpp:

(JSC::FTL::LowerDFGToLLVM::speculate):
(JSC::FTL::LowerDFGToLLVM::speculateNotCell):

1:37 PM Changeset in webkit [162954] by Antti Koivisto
  • 3 edits in trunk/LayoutTests

Rebase fast/repaint/background-shorthand-with-gradient-and-height-changes.html after ​https://trac.webkit.org/r162947

  • fast/repaint/background-shorthand-with-gradient-and-height-changes-expected.txt:
  • fast/repaint/background-shorthand-with-gradient-and-height-changes.html:
1:31 PM Changeset in webkit [162953] by fpizlo@apple.com
  • 2 edits
    1 add in branches/jsCStack/Source/JavaScriptCore

FTL should really support NotCellUse
​https://bugs.webkit.org/show_bug.cgi?id=127789

Reviewed by Mark Hahnenberg.

This is for ValueToInt32(NotCell:). That support was added ages ago, and it even
works as demonstrated by the new test case.

  • ftl/FTLCapabilities.cpp:

(JSC::FTL::canCompile):

  • tests/stress/not-cell-use.js: Added.

(foo):
(bar):

1:26 PM Changeset in webkit [162952] by Antti Koivisto
  • 27 edits in trunk/LayoutTests

Rebaseline some repaint test results after ​https://trac.webkit.org/r162947

There are less repaint rects.

  • compositing/repaint/resize-repaint-expected.txt:
  • fast/box-shadow/shadow-repaint-expected.txt:
  • fast/css-grid-layout/grid-element-change-columns-repaint-expected.txt:
  • fast/css-grid-layout/grid-element-change-rows-repaint-expected.txt:
  • fast/css-grid-layout/grid-item-change-column-repaint-expected.txt:
  • fast/css-grid-layout/grid-item-change-row-repaint-expected.txt:
  • fast/regions/repaint/element-in-named-flow-absolute-from-fixed-expected.txt:
  • fast/regions/repaint/element-in-named-flow-fixed-from-absolute-expected.txt:
  • fast/regions/repaint/element-inflow-fixed-from-outflow-static-expected.txt:
  • fast/regions/repaint/element-outflow-static-from-inflow-fixed-expected.txt:
  • fast/regions/repaint/invalid-region-repaint-crash-expected.txt:
  • fast/regions/repaint/repaint-element-inside-relative-region-expected.txt:
  • fast/regions/repaint/repaint-regions-overflow-expected.txt:
  • fast/repaint/horizontal-bt-overflow-child-expected.txt:
  • fast/repaint/horizontal-bt-overflow-parent-expected.txt:
  • fast/repaint/horizontal-bt-overflow-same-expected.txt:
  • fast/repaint/negative-text-indent-with-overflow-hidden-expected.txt:
  • fast/repaint/overflow-flipped-writing-mode-table-expected.txt:
  • fast/repaint/reflection-table-layout-expected.txt:
  • fast/repaint/transform-table-layout-expected.txt:
  • fast/repaint/vertical-overflow-child-expected.txt:
  • fast/repaint/vertical-overflow-parent-expected.txt:
  • fast/repaint/vertical-overflow-same-expected.txt:
  • platform/mac/compositing/repaint/invalidations-on-composited-layers-expected.txt:
  • platform/mac/fast/images/repaint-subrect-grid-expected.txt:
  • svg/repaint/repaint-webkit-svg-shadow-container-expected.txt:
1:15 PM Changeset in webkit [162951] by andersca@apple.com
  • 7 edits
    2 adds in trunk/Source/WebKit2

Add a NavigationActionData object for easy storage of navigation action data
​https://bugs.webkit.org/show_bug.cgi?id=127787

Reviewed by Tim Horton.

  • Scripts/webkit2/messages.py:

(struct_or_class):

  • Shared/NavigationActionData.cpp: Added.

(WebKit::NavigationActionData::NavigationActionData):
(WebKit::NavigationActionData::encode):
(WebKit::NavigationActionData::decode):

  • Shared/NavigationActionData.h: Added.
  • UIProcess/WebPageProxy.cpp:

(WebKit::WebPageProxy::decidePolicyForNavigationAction):
(WebKit::WebPageProxy::decidePolicyForNewWindowAction):

  • UIProcess/WebPageProxy.h:
  • UIProcess/WebPageProxy.messages.in:
  • WebKit2.xcodeproj/project.pbxproj:
  • WebProcess/WebCoreSupport/WebFrameLoaderClient.cpp:

(WebKit::WebFrameLoaderClient::dispatchDecidePolicyForNewWindowAction):
(WebKit::WebFrameLoaderClient::dispatchDecidePolicyForNavigationAction):

12:49 PM Changeset in webkit [162950] by timothy_horton@apple.com
  • 3 edits in trunk/Source/WebKit2

Fix the Mountain Lion build.

  • UIProcess/mac/ViewGestureController.mm:

(WebKit::ViewGestureController::beginSwipeGesture):
(WebKit::ViewGestureController::removeSwipeSnapshot):

  • UIProcess/mac/ViewSnapshotStore.mm:

(WebKit::ViewSnapshotStore::recordSnapshot):

12:25 PM Changeset in webkit [162949] by timothy_horton@apple.com
  • 4 edits in trunk/Source/WebKit2

Fix the iOS build.

  • UIProcess/WebPageProxy.cpp:

(WebKit::WebPageProxy::recordNavigationSnapshot):

  • UIProcess/mac/ViewSnapshotStore.h:
  • UIProcess/mac/ViewSnapshotStore.mm:
12:01 PM Changeset in webkit [162948] by beidson@apple.com
  • 3 edits in trunk/Source/WebKit2

Remove unneeded “WebCore::” prefixing in two .cpp files.

Rubberstamped by Alexey Proskuryakov.

  • DatabaseProcess/IndexedDB/UniqueIDBDatabase.cpp:
  • DatabaseProcess/IndexedDB/sqlite/UniqueIDBDatabaseBackingStoreSQLite.cpp:
11:40 AM Changeset in webkit [162947] by Antti Koivisto
  • 5 edits in trunk/Source/WebCore

REGRESSION(r162837): 5% regression on html5-full-render and 3% regression in DoYouEvenBench
​https://bugs.webkit.org/show_bug.cgi?id=127722

Reviewed by Darin Adler.

Accumulate repaint region in RendeView instead of flushing rects directly to the system.

  • dom/Document.cpp:

(WebCore::Document::recalcStyle):
(WebCore::Document::updateLayout):
(WebCore::Document::topDocument):

Make less silly.

  • page/FrameView.cpp:

(WebCore::FrameView::layout):

  • rendering/RenderView.cpp:

(WebCore::RenderView::repaintViewRectangle):
(WebCore::RenderView::flushAccumulatedRepaintRegion):
(WebCore::RenderView::RepaintRegionAccumulator::RepaintRegionAccumulator):
(WebCore::RenderView::RepaintRegionAccumulator::~RepaintRegionAccumulator):

  • rendering/RenderView.h:
11:33 AM Changeset in webkit [162946] by timothy_horton@apple.com
  • 4 edits in trunk/Source/WebKit2

WebKit2 View Gestures (Swipe): Snapshots should be purgeable
​https://bugs.webkit.org/show_bug.cgi?id=127390
<rdar://problem/15876775>

Reviewed by Anders Carlsson.

Make snapshots be purgeable, and implement a straightforward
(if perhaps expensive, for now) cache eviction strategy to limit the
number of snapshots to 20.

  • UIProcess/mac/ViewGestureController.mm:

(WebKit::ViewGestureController::beginSwipeGesture):
When beginning a gesture, attempt to make the retrieved snapshot
non-volatile. If it was purged while volatile, we won't use it, but if it
is still valid, we'll go ahead and use it as the swipe layer's contents.

(WebKit::ViewGestureController::removeSwipeSnapshot):
When removing the swipe snapshot, make it volatile once again.

  • UIProcess/mac/ViewSnapshotStore.h:

Store a creation time along with the image.
Store and return IOSurfaces instead of CGImages.
Store snapshots and render tree sizes separately, so that we can
throw away snapshots but keep the render tree sizes indefinitely.

  • UIProcess/mac/ViewSnapshotStore.mm:

(WebKit::ViewSnapshotStore::pruneSnapshots):
Cap the number of snapshots we'll ever have live at 20.
Enforce this cap by first trying to remove snapshots farthest
from the current back-forward list's current item, falling back
to removing the least recently created snapshot if there are no
snapshots owned by the current back-forward list.

(WebKit::createIOSurfaceFromImage):
Build an IOSurface from the CGImage snapshot we took, for ease of
use of its purgeability API.

(WebKit::ViewSnapshotStore::recordSnapshot):
Bail from taking the snapshot if the image is empty; this can happen
if the view is out of the window when the snapshot is taken.

Mark snapshots as purgeable as soon as they go into the cache.

(WebKit::ViewSnapshotStore::snapshotAndRenderTreeSize):
Return the target render tree size even if there is no snapshot image.
Take care not to look up an empty UUID.

11:26 AM Changeset in webkit [162945] by commit-queue@webkit.org
  • 2 edits in trunk/Source/WebCore

HTMLMediaElement::m_platformLayerBorrowed is uninitialized.
​https://bugs.webkit.org/show_bug.cgi?id=127759

Patch by Jeremy Jones <​jeremyj@apple.com> on 2014-01-28
Reviewed by Eric Carlson.

Initalize m_platformLayerBorrowed to false.

  • html/HTMLMediaElement.cpp:

(WebCore::HTMLMediaElement::HTMLMediaElement):

10:22 AM Changeset in webkit [162944] by jer.noble@apple.com
  • 3 edits in trunk/Source/WebCore

[Mac] Use explicit, rather than canonical, AudioFormatFlags and types.
​https://bugs.webkit.org/show_bug.cgi?id=127781

Reviewed by Eric Carlson.

Since every mac platform uses Float32 audio samples, use these types explicitly
rather than the implicit, canonical AudioSampleType type. Once we're using Float32
explicity, we can also use the explicit kAudioFormatFlagsNativeFloatPacked flag
rather than the canonical version.

  • platform/audio/mac/AudioDestinationMac.cpp:

(WebCore::AudioDestinationMac::configure):

  • platform/audio/mac/AudioFileReaderMac.cpp:

(WebCore::AudioFileReader::createBus):

10:18 AM Changeset in webkit [162943] by ap@apple.com
  • 18 edits in trunk/Source

Expose SQL database creation and modification times
​https://bugs.webkit.org/show_bug.cgi?id=127728

Reviewed by Brady Eidson.

Source/WebCore:

  • platform/FileSystem.h:
  • platform/posix/FileSystemPOSIX.cpp:

(WebCore::getFileCreationTime):

  • platform/win/FileSystemWin.cpp:

(WebCore::getFileCreationTimeFromFindData):
(WebCore::getFileCreationTime):
Added functions to get file creation times. Not all OSes support that, but Darwin
and Windows do, as (I think) various BSD flavors.

  • platform/gtk/FileSystemGtk.cpp: Added a stub for getFileCreationTime().
  • platform/sql/SQLiteFileSystem.h:
  • platform/sql/SQLiteFileSystem.cpp:

(WebCore::SQLiteFileSystem::databaseCreationTime):
(WebCore::SQLiteFileSystem::databaseModificationTime):
Expose it in the same strange manner other database file operations are.

  • Modules/webdatabase/DatabaseDetails.h:

(WebCore::DatabaseDetails::DatabaseDetails):
(WebCore::DatabaseDetails::creationTime):
(WebCore::DatabaseDetails::modificationTime):
Added creation and modification times to DatabaseDetails.

  • Modules/webdatabase/DatabaseManager.cpp:

(WebCore::DatabaseManager::ProposedDatabase::ProposedDatabase):
(WebCore::DatabaseManager::openDatabaseBackend):
ProposedDatabase is an unfortunate mechanism for communicating quota errors to the
client, we should really straighten it up. But for now, just leave the times uninitialized.

  • Modules/webdatabase/DatabaseBackendBase.cpp: (WebCore::DatabaseBackendBase::details):

This code path is for handling quota errors, so there is no need to initialize most
of DatabaseDetails here as well.

  • Modules/webdatabase/DatabaseTracker.cpp: (WebCore::DatabaseTracker::detailsForNameAndOrigin):
  • Modules/webdatabase/DatabaseTracker.h:

Fill in database file times. Inlined and removed usageForDatabase function to
avoid rebuilding the path multiple times.

Source/WebKit2:

  • Shared/WebCoreArgumentCoders.cpp:

(IPC::ArgumentCoder<DatabaseDetails>::encode):
(IPC::ArgumentCoder<DatabaseDetails>::decode):
Encode and decode new DatabaseDetails member variables.

  • UIProcess/API/C/WKDatabaseManager.cpp:

(WKDatabaseManagerGetDatabaseDetailsCreationTimeKey):
(WKDatabaseManagerGetDatabaseDetailsModificationTimeKey):

  • UIProcess/API/C/WKDatabaseManager.h:

Added new keys for database details.

  • UIProcess/WebDatabaseManagerProxy.cpp:

(WebKit::WebDatabaseManagerProxy::databaseDetailsCreationTimeKey):
(WebKit::WebDatabaseManagerProxy::databaseDetailsModificationTimeKey):
(WebKit::WebDatabaseManagerProxy::didGetDatabasesByOrigin):

  • UIProcess/WebDatabaseManagerProxy.h:

Set the new keys if times are valid.

9:55 AM Changeset in webkit [162942] by Michał Pakuła vel Rutka
  • 2 edits in trunk/LayoutTests

Unreviewed EFL gardening

Skip crashing http tests.

  • platform/efl-wk2/TestExpectations:
9:49 AM Changeset in webkit [162941] by jer.noble@apple.com
  • 6 edits in trunk

Setting muted attribute on <video> element is not reflected in controls
​https://bugs.webkit.org/show_bug.cgi?id=127726

Reviewed by Eric Carlson.

Source/WebCore:

If the 'muted' IDL attribute is queried after the 'muted' and 'src' content attributes are
set but before loading begins, it will return 'false' until loading begins, but with no
way to query whether that value is valid, nor firing a volumechange event when its value
changes.

The HTML spec says that the 'muted' content attribute controls whether audio output is
muted "when the media element is created", not when loading begins, but we don't
necessarily have a signal that the element is fully parsed. Additionally, this means its
impossible to make an element via script and get this behavior.

So the new behavior is that the 'muted' IDL attribute will always reflect the 'muted'
content attribute up until one of the following three conditions:

  • The 'muted' IDL attribute is set via script.
  • The element is inserted in the document.
  • The element begins loading.

After the first one of the above conditions, the 'muted' IDL attribute will no longer
change when the 'muted' content attribute changes.

  • html/HTMLMediaElement.cpp:

(WebCore::HTMLMediaElement::HTMLMediaElement):
(WebCore::HTMLMediaElement::parseAttribute):

  • html/HTMLMediaElement.h:

LayoutTests:

  • media/video-defaultmuted-expected.txt:
  • media/video-defaultmuted.html:
9:43 AM Changeset in webkit [162940] by mark.lam@apple.com
  • 15 edits
    2 adds in trunk

Jettison DFG code when neither breakpoints or the profiler are active.
<​https://webkit.org/b/127766>

Reviewed by Geoffrey Garen.

Source/JavaScriptCore:

We need to jettison the DFG CodeBlocks under the following circumstances:

  1. When adding breakpoints to a CodeBlock, jettison it if it is a DFG CodeBlock.
  2. When enabling stepping mode in a CodeBlock, jettison it if it a DFG CodeBlock.
  3. When settign the enabled profiler in the VM, we need to jettison all DFG CodeBlocks.

Instead of emitting speculation checks, the DFG code will now treat Breakpoint,
ProfileWillCall, and ProfileDidCall as no-ops similar to a Phantom node. We
still need to track these nodes so that they match the corresponding opcodes
in the baseline JIT when we jettison and OSR exit. Without them, we would OSR
exit to the wrong location in the baseline JIT code.

In DFGDriver's compileImpl() and DFGPlan's finalizeWithoutNotifyingCallback()
we fail the compilation effort with a CompilationInvalidated result. This allows
the DFG compiler to re-attampt the compilation of the function after some time
if it is hot. The CompilationInvalidated result is supposed to cause the DFG
to exercise an exponential back off before re-attempting compilation again
(see runtime/CompilationResult.h).

This patch improves the Octane score from ~2950 to ~3067.

  • bytecode/CodeBlock.cpp:

(JSC::CodeBlock::addBreakpoint):
(JSC::CodeBlock::setSteppingMode):

  • bytecode/CodeBlock.h:
  • debugger/Debugger.h:
  • dfg/DFGAbstractInterpreterInlines.h:

(JSC::DFG::AbstractInterpreter<AbstractStateType>::executeEffects):

  • dfg/DFGClobberize.h:

(JSC::DFG::clobberize):

  • dfg/DFGDriver.cpp:

(JSC::DFG::compileImpl):

  • dfg/DFGPlan.cpp:

(JSC::DFG::Plan::finalizeWithoutNotifyingCallback):

  • dfg/DFGSpeculativeJIT32_64.cpp:

(JSC::DFG::SpeculativeJIT::compile):

  • dfg/DFGSpeculativeJIT64.cpp:

(JSC::DFG::SpeculativeJIT::compile):

  • profiler/LegacyProfiler.cpp:

(JSC::LegacyProfiler::startProfiling):
(JSC::LegacyProfiler::stopProfiling):

  • runtime/VM.cpp:

(JSC::VM::VM):
(JSC::SetEnabledProfilerFunctor::operator()):
(JSC::VM::setEnabledProfiler):

  • runtime/VM.h:

(JSC::VM::enabledProfiler):

LayoutTests:

Added a test to exercise setting a breakpoint in 2 DFG compiled functions:
1 not inlined, and 1 inlined.

  • inspector-protocol/debugger/resources/breakpoint.js:

(notInlineable):
(inlineable):
(notInliningFoo):
(inliningFoo):
(dfgWithoutInline):
(dfgWithInline):

  • inspector-protocol/debugger/setBreakpoint-dfg-expected.txt: Added.
  • inspector-protocol/debugger/setBreakpoint-dfg.html: Added.
9:38 AM Changeset in webkit [162939] by andersca@apple.com
  • 3 edits in trunk/Source/WTF

Actually turn stack stats back off.

  • wtf/StackStats.cpp:
  • wtf/StackStats.h:
9:36 AM Changeset in webkit [162938] by andersca@apple.com
  • 2 edits in trunk/Source/WTF

Turn stack stats back off.

  • wtf/StackStats.cpp:
9:28 AM Changeset in webkit [162937] by andersca@apple.com
  • 8 edits
    2 adds in trunk/Source/WebCore

Add stubbed out VisitedLinkProvider class
​https://bugs.webkit.org/show_bug.cgi?id=127744

Reviewed by Andreas Kling.

This is a first step towards moving responsibility of visited links from
the page group to a separate object.

  • CMakeLists.txt:
  • GNUmakefile.list.am:
  • WebCore.vcxproj/WebCore.vcxproj:
  • WebCore.vcxproj/WebCore.vcxproj.filters:
  • WebCore.xcodeproj/project.pbxproj:
  • page/PageGroup.cpp:

(WebCore::PageGroup::PageGroup):

  • page/PageGroup.h:

(WebCore::PageGroup::visitedLinkProvider):

  • page/VisitedLinkProvider.cpp: Added.

(WebCore::VisitedLinkProvider::VisitedLinkProvider):
(WebCore::VisitedLinkProvider::~VisitedLinkProvider):

  • page/VisitedLinkProvider.h: Added.

(WebCore::VisitedLinkProvider::create):

9:19 AM Changeset in webkit [162936] by timothy_horton@apple.com
  • 2 edits in trunk/Source/WebKit2

WebKit2 View Gestures (Swipe): Use CGSCaptureWindowsContentsToRect for now
​https://bugs.webkit.org/show_bug.cgi?id=127768
<rdar://problem/15923662>

Reviewed by Simon Fraser.

  • UIProcess/API/mac/WKView.mm:

(-[WKView _takeViewSnapshot]):
Until <rdar://problem/15709646> is resolved, we'll use
CGSCaptureWindowsContentsToRect instead of CGWindowListCreateImage.

9:13 AM Changeset in webkit [162935] by andersca@apple.com
  • 5 edits in trunk/Source/WTF

Use std::mutex instead of WTF::Mutex in WTF
​https://bugs.webkit.org/show_bug.cgi?id=127783

Reviewed by Antti Koivisto.

  • wtf/CryptographicallyRandomNumber.cpp:
  • wtf/MainThread.cpp:

(WTF::mainThreadFunctionQueueMutex):
(WTF::functionQueue):
(WTF::dispatchFunctionsFromMainThread):
(WTF::callOnMainThread):
(WTF::cancelCallOnMainThread):

  • wtf/StackStats.cpp:

(WTF::StackStats::initialize):
(WTF::StackStats::CheckPoint::CheckPoint):
(WTF::StackStats::CheckPoint::~CheckPoint):
(WTF::StackStats::probe):
(WTF::StackStats::LayoutCheckPoint::LayoutCheckPoint):
(WTF::StackStats::LayoutCheckPoint::~LayoutCheckPoint):

  • wtf/StackStats.h:
9:07 AM Changeset in webkit [162934] by msaboff@apple.com
  • 3 edits in branches/jsCStack/Source/JavaScriptCore

CStack Branch: Some LLInt symbols not declared extern "C"
​https://bugs.webkit.org/show_bug.cgi?id=127782

Reviewed by Geoffrey Garen.

Added 'extern "C"' to the definitions of llint_throw_stack_overflow_error() and
llint_stack_check_at_vm_entry(). Added WTF_INTERNAL to their declarations.

(JSC::LLInt::llint_throw_stack_overflow_error):
(JSC::LLInt::llint_stack_check_at_vm_entry):

  • llint/LLIntSlowPaths.h:
8:34 AM Changeset in webkit [162933] by commit-queue@webkit.org
  • 21 edits
    2 copies
    20 adds in trunk

Add support for menclose element
​https://bugs.webkit.org/show_bug.cgi?id=85729

Patch by Gurpreet Kaur <​k.gurpreet@samsung.com> on 2014-01-28
Reviewed by Chris Fleizach.

Source/WebCore:

Added support for menclose element. MathML <menclose> element renders
its content inside an enclosing notation specified by the notation
attribute. The notation attribute can have values longdiv, box, left,
right, top, bottom , radical, madruwb, actuarial, roundedbox, circle,
updiagonalstrike, downdiagonalstrike, verticalstrike and
horizontalstrike.

Tests: mathml/presentation/menclose-add-children.html

mathml/presentation/menclose-notation-attribute-add.html
mathml/presentation/menclose-notation-attribute-change-value.html
mathml/presentation/menclose-notation-attribute-remove.html
mathml/presentation/menclose-notation-attribute-set1.html
mathml/presentation/menclose-notation-attribute-set2.html
mathml/presentation/menclose-notation-no-overlap.html
mathml/presentation/menclose-notation-radical.html
mathml/presentation/menclose-remove-children.html

  • CMakeLists.txt:
  • GNUmakefile.list.am:
  • WebCore.vcxproj/WebCore.vcxproj:
  • WebCore.vcxproj/WebCore.vcxproj.filters:
  • WebCore.xcodeproj/project.pbxproj:
  • css/mathml.css:

(mo, mrow, mfenced, mfrac, msub, msup, msubsup, mmultiscripts, mprescripts, none, munder, mover, munderover, msqrt, mroot, merror, mphantom, mstyle, menclose):
(math, mrow, mfenced, msqrt, mroot, merror, mphantom, mstyle, menclose):

  • mathml/MathMLAllInOne.cpp:
  • mathml/MathMLElement.h:
  • mathml/MathMLInlineContainerElement.cpp:
  • mathml/MathMLMencloseElement.cpp: Added.

(WebCore::MathMLMencloseElement::MathMLMencloseElement):
(WebCore::MathMLMencloseElement::create):
(WebCore::MathMLMencloseElement::createElementRenderer):
(WebCore::MathMLMencloseElement::isPresentationAttribute):
(WebCore::MathMLMencloseElement::finishParsingChildren):
(WebCore::MathMLMencloseElement::collectStyleForPresentationAttribute):
(WebCore::MathMLMencloseElement::longDivLeftPadding):

  • mathml/MathMLMencloseElement.h: Added.

(WebCore::toMathMLMencloseElement):

  • mathml/mathattrs.in:
  • mathml/mathtags.in:
  • rendering/mathml/RenderMathMLMenclose.cpp: Added.

(WebCore::RenderMathMLMenclose::RenderMathMLMenclose):
(WebCore::RenderMathMLMenclose::addChild):
(WebCore::RenderMathMLMenclose::computePreferredLogicalWidths):
(WebCore::RenderMathMLMenclose::updateLogicalHeight):
(WebCore::RenderMathMLMenclose::paint):
(WebCore::RenderMathMLMenclose::checkNotationalValuesValidity):

  • rendering/mathml/RenderMathMLMenclose.h: Added.
  • rendering/mathml/RenderMathMLRoot.cpp:

(WebCore::RenderMathMLRoot::RenderMathMLRoot):

  • rendering/mathml/RenderMathMLRoot.h:
  • rendering/mathml/RenderMathMLSquareRoot.cpp:

(WebCore::RenderMathMLSquareRoot::RenderMathMLSquareRoot):
(WebCore::RenderMathMLSquareRoot::createAnonymousWithParentRenderer):

  • rendering/mathml/RenderMathMLSquareRoot.h:

Added new file related to menclose element implementation. Menclose
element is created and while parsing its notation attribute based on
its values like top, left CSSBorder properties are applied and for
values like circle, verticalstrike, longidv its taken care in paint.
For radical value an anonymous RenderMathMLSquareRoot is created as
a child of menclose.

LayoutTests:

  • mathml/presentation/inferred-mrow-baseline-expected.txt:
  • mathml/presentation/inferred-mrow-baseline.html:
  • mathml/presentation/inferred-mrow-stretchy-expected.txt:
  • mathml/presentation/inferred-mrow-stretchy.html:

Rebaselining existing tests as per the new behavior.

  • mathml/presentation/menclose-add-children-expected.html: Added.
  • mathml/presentation/menclose-add-children.html: Added.
  • mathml/presentation/menclose-notation-attribute-add-expected.html: Added.
  • mathml/presentation/menclose-notation-attribute-add.html: Added.
  • mathml/presentation/menclose-notation-attribute-change-value-expected.html: Added.
  • mathml/presentation/menclose-notation-attribute-change-value.html: Added.
  • mathml/presentation/menclose-notation-attribute-remove-expected.html: Added.
  • mathml/presentation/menclose-notation-attribute-remove.html: Added.
  • mathml/presentation/menclose-notation-attribute-set1-expected.txt: Added.
  • mathml/presentation/menclose-notation-attribute-set1.html: Added.
  • mathml/presentation/menclose-notation-attribute-set2-expected-mismatch.html: Added.
  • mathml/presentation/menclose-notation-attribute-set2.html: Added.
  • mathml/presentation/menclose-notation-no-overlap-expected.txt: Added.
  • mathml/presentation/menclose-notation-no-overlap.html: Added.
  • mathml/presentation/menclose-notation-radical-expected.html: Added.
  • mathml/presentation/menclose-notation-radical.html: Added.
  • mathml/presentation/menclose-remove-children-expected.html: Added.
  • mathml/presentation/menclose-remove-children.html: Added.

Added new test cases for verifying the behaviour of menclose element
with different values of notation attribute and removing/changing/adding
notation attribute.

8:29 AM Changeset in webkit [162932] by k.czech@samsung.com
  • 2 edits in trunk/Source/WebCore

[ATK] accessibility/range-alter-by-percent.html is failing after r162587.
​https://bugs.webkit.org/show_bug.cgi?id=127774

Reviewed by Mario Sanchez Prada.

Implicit value of step in range type elements should be one or larger.

No new tests. Covered by exiting one.

  • accessibility/atk/WebKitAccessibleInterfaceValue.cpp:

(webkitAccessibleValueGetMinimumIncrement):

7:40 AM Changeset in webkit [162931] by commit-queue@webkit.org
  • 3 edits in trunk/Source/WebInspectorUI

Web Inspector: In a DataGrid, store value of columnIdentifier to DOM node representing a cell
​https://bugs.webkit.org/show_bug.cgi?id=127613

Patch by Diego Pino Garcia <Diego Pino Garcia> on 2014-01-28
Reviewed by Timothy Hatcher.

  • UserInterface/DOMStorageContentView.js:

(WebInspector.DOMStorageContentView.prototype._editingCallback):

  • UserInterface/DataGrid.js:

(WebInspector.DataGrid):
(WebInspector.DataGrid.prototype._keyDown):
(WebInspector.DataGrid.prototype._contextMenuInDataTable):
(WebInspector.DataGridNode.prototype.createCell):

7:08 AM Changeset in webkit [162930] by zandobersek@gmail.com
  • 2 edits in trunk/Source/WebKit/efl

Remove unnecessary forward declarations of MessagePortChannel, MessagePortChannelArray in DumpRenderTreeSupportEfl.h
​https://bugs.webkit.org/show_bug.cgi?id=127654

Reviewed by Alexey Proskuryakov.

  • WebCoreSupport/DumpRenderTreeSupportEfl.h: Remove the forward declarations of MessagePortChannel

and MessagePortChannelArray. They were most likely at one point used in DRT support methods, but are
not necessary anymore.

2:40 AM Changeset in webkit [162929] by mrowe@apple.com
  • 10 edits in trunk/Source

<​https://webkit.org/b/127767> Disable some deprecation warnings to fix the build.

Reviewed by Ryosuke Niwa.

Source/WebCore:

  • accessibility/mac/AXObjectCacheMac.mm:

(WebCore::AXObjectCache::postPlatformNotification):

  • accessibility/mac/WebAccessibilityObjectWrapperMac.mm:

(-[WebAccessibilityObjectWrapper renderWidgetChildren]):

  • platform/audio/mac/AudioDestinationMac.cpp:

(WebCore::AudioDestinationMac::configure):

  • platform/audio/mac/AudioFileReaderMac.cpp:

(WebCore::AudioFileReader::createBus):

Source/WebKit/mac:

  • WebCoreSupport/PopupMenuMac.mm:

(PopupMenuMac::populate):

Source/WebKit2:

  • UIProcess/API/mac/WKView.mm:

(-[WKView _updateWindowAndViewFrames]):

  • WebProcess/WebPage/mac/WKAccessibilityWebPageObject.mm:
2:22 AM Changeset in webkit [162928] by commit-queue@webkit.org
  • 9 edits
    1 add in trunk

[GTK] Add API to allow setting the process model in WebKitWebContext
​https://bugs.webkit.org/show_bug.cgi?id=125463

Patch by Adrian Perez de Castro <Adrian Perez de Castro> on 2014-01-28
Reviewed by Carlos Garcia Campos.

Implements accessors in WebKitWebContext for the process model.
The default behavior is unchanged, and a single web process is
used. Using WEBKIT_PROCESS_MODEL_ONE_SECONDARY_PROCESS_PER_WEB_VIEW
as the process model will make use one web process for each
WebKitWebView. This also enables the network process. Setting
the process model must be done as early as possible, before the
very first web process is spawned.

Source/WebKit2:

  • UIProcess/API/gtk/WebKitWebContext.cpp:

Add accessors in the API to get/set the process model.
(webkit_web_context_set_process_model):
(webkit_web_context_get_process_model):

  • UIProcess/API/gtk/WebKitWebContext.h:

Define the WebKitProcessModel enum.

  • UIProcess/API/gtk/docs/webkit2gtk-sections.txt:

Add new public API bits to the documentation.

Tools:

  • MiniBrowser/gtk/main.c:

(main):
Enable multiple process mode if the MINIBROWSER_MULTIPROCESS
environment variable is defined and not empty.

  • TestWebKitAPI/GNUmakefile.am:

Add new TestMultiprocess test case.

  • TestWebKitAPI/Tests/WebKit2Gtk/CMakeLists.txt:

Add new TestMultiprocess test case.

  • TestWebKitAPI/Tests/WebKit2Gtk/TestMultiprocess.cpp: Added.

(loadChanged):
(testMultipleSecondaryProcesses):
(initializeWebExtensions):
(beforeAll):
(afterAll):

  • TestWebKitAPI/Tests/WebKit2Gtk/WebExtensionTest.cpp:

(methodCallCallback):
Implement the GetProcessIdentifier D-Bus method.
(makeBusName):
Choose a different bus name when the web extension is used
from TestMultiprocess.
(webkit_web_extension_initialize_with_user_data):
Use makeBusName() to choose the bus name.

2:02 AM Changeset in webkit [162927] by bshafiei@apple.com
  • 3 edits in tags/Safari-538.14/Source/WebCore

Merged r162885.

1:56 AM Changeset in webkit [162926] by bshafiei@apple.com
  • 18 edits in tags/Safari-538.14

Merged r162881.

1:38 AM Changeset in webkit [162925] by bshafiei@apple.com
  • 2 edits in tags/Safari-538.14/Source/WebCore

Merged r162872.

1:21 AM Changeset in webkit [162924] by bshafiei@apple.com
  • 5 edits in trunk/Source

Versioning.

1:15 AM Changeset in webkit [162923] by bshafiei@apple.com
  • 1 copy in tags/Safari-538.14

New tag.

1:08 AM Changeset in webkit [162922] by Carlos Garcia Campos
  • 14 edits
    1 add
    2 deletes in trunk

[SOUP] Remove soupURIToKURL
​https://bugs.webkit.org/show_bug.cgi?id=127104

Reviewed by Martin Robinson.

Source/WebCore:

In favor of a URL constructor receiving a SoupURI. Also add a
method to URL to create a soupURI.

  • GNUmakefile.list.am: Remove SoupURIUtils and add URLSoup.
  • PlatformEfl.cmake: Ditto.
  • PlatformGTK.cmake: Ditto.
  • platform/URL.h: Add URL constructor receiving a SoupURI and

createSoupURI() to create a new soupURI for the URL.

  • platform/network/soup/CookieJarSoup.cpp:

(WebCore::setCookiesFromDOM): Use URL::createSoupURI().
(WebCore::cookiesForSession): Ditto.
(WebCore::getRawCookies): Ditto.
(WebCore::deleteCookie): Ditto.

  • platform/network/soup/ResourceHandleSoup.cpp:

(WebCore::doRedirect): Use the new URL constructor instead of
soupURIToKURL.
(WebCore::createSoupRequestAndMessageForHandle): Use URL::createSoupURI().

  • platform/network/soup/ResourceRequest.h: Rename soupURI as

createSoupURI to make it clear that the method returns a newly
created SoupURI.

  • platform/network/soup/ResourceRequestSoup.cpp:

(WebCore::ResourceRequest::updateSoupMessageMembers): Use URL::createSoupURI().
(WebCore::ResourceRequest::updateSoupMessage): Ditto.
(WebCore::ResourceRequest::updateFromSoupMessage): Use the new URL
constructor instead of soupURIToKURL.
(WebCore::ResourceRequest::createSoupURI): Use URL::createSoupURI().

  • platform/network/soup/ResourceResponseSoup.cpp:

(WebCore::ResourceResponse::updateFromSoupMessage): Use the new
URL constructor instead of soupURIToKURL.

  • platform/network/soup/SoupURIUtils.cpp: Removed.
  • platform/network/soup/SoupURIUtils.h: Removed.
  • platform/soup/URLSoup.cpp: Added.

(WebCore::URL::URL):
(WebCore::URL::createSoupURI):

Tools:

  • TestWebKitAPI/GNUmakefile.am: Add soup cppflags to WebCore unit

tests, because URL.h now includes GUniquePtrSoup.h.

  • TestWebKitAPI/CMakeLists.txt: Ditto.
  • WebKitTestRunner/PlatformEfl.cmake: Add soup cppflags to WTR,

because URL.h now includes GUniquePtrSoup.h.

12:15 AM Changeset in webkit [162921] by Carlos Garcia Campos
  • 2 edits in trunk/Source/WebCore

[GTK] Avoid unnecessary string duplications in FileSystemGtk
​https://bugs.webkit.org/show_bug.cgi?id=127469

Reviewed by Martin Robinson.

We are using fileSystemRepresentation() everywhere internally
which returns a CString that always duplicates the string.
Add unescapedFilename() internal helper function that returns a
GUniquePtr and used it everywhere instead of fileSystemRepresentation().

  • platform/gtk/FileSystemGtk.cpp:

(WebCore::unescapedFilename):
(WebCore::fileSystemRepresentation):
(WebCore::filenameForDisplay):
(WebCore::fileExists):
(WebCore::deleteFile):
(WebCore::deleteEmptyDirectory):
(WebCore::getFileStat):
(WebCore::getFileSize):
(WebCore::getFileModificationTime):
(WebCore::getFileMetadata):
(WebCore::pathByAppendingComponent):
(WebCore::makeAllDirectories):
(WebCore::pathGetFileName):
(WebCore::directoryName):
(WebCore::listDirectory):
(WebCore::openFile):

Note: See TracTimeline for information about the timeline view.