Timeline



Jun 24, 2015:

10:32 PM Changeset in webkit [185947] by Chris Dumez
  • 2 edits in trunk/LayoutTests

Unreviewed gardening, test rebaseline on Windows after r185934.

  • platform/win/fast/css/apple-system-colors-expected.txt:
10:20 PM Changeset in webkit [185946] by Chris Dumez
  • 2 edits in trunk/Source/WebCore

Unreviewed, EFL build fix after r185940.

Members were initialized in wrong order.

  • Modules/mediastream/MediaDeviceInfo.cpp:

(WebCore::MediaDeviceInfo::MediaDeviceInfo):

10:11 PM Changeset in webkit [185945] by Chris Dumez
  • 2 edits in trunk/Source/WebCore

Unreviewed, Build fix after r185940.

Move implementation inside WebCore namespace.

  • Modules/mediastream/MediaDeviceInfo.cpp:
10:02 PM Changeset in webkit [185944] by Chris Dumez
  • 2 edits in trunk/Source/WebCore

Unreviewed, speculative GTK build fix after 185940.

Add new cpp file to CMakeLists.txt.

  • CMakeLists.txt:
9:49 PM Changeset in webkit [185943] by Chris Dumez
  • 2 edits in trunk/Source/JavaScriptCore

Unreviewed, speculative build fix after r185942.

Add missing include for StrongInlines.h.

  • runtime/ArrayPrototype.cpp:
9:32 PM Changeset in webkit [185942] by Darin Adler
  • 3 edits in trunk/Source/JavaScriptCore

Optimize Array.join and Array.reverse for high speed array types
​https://bugs.webkit.org/show_bug.cgi?id=146275

Reviewed by Mark Lam.

This seems to yield another 17% speed improvement in the array
test from the Peacekeeper benchmark.

  • runtime/ArrayPrototype.cpp:

(JSC::isHole): Added. Helper to check for holes.
(JSC::containsHole): Ditto.
(JSC::arrayProtoFuncJoin): Added special cases for the various types
of arrays that could be in a butterfly.
(JSC::arrayProtoFuncReverse): Ditto.

  • runtime/JSStringJoiner.h: Made appendEmptyString public so we can

call it from the new parts of Array.join.

9:06 PM Changeset in webkit [185941] by fpizlo@apple.com
  • 4 edits in trunk/Source/JavaScriptCore

DFG::SpeculativeJIT shouldn't use filter==Contradiction when it meant isClear
​https://bugs.webkit.org/show_bug.cgi?id=146291
rdar://problem/21435366

Reviewed by Michael Saboff.

The filter() method returns Contradiction only when a value *becomes* clear. This is
necessary for supporting the convention that non-JSValue nodes have a bottom proved
type. (We should fix that convention eventually, but for now let's just be consistent
about it.)

  • dfg/DFGFiltrationResult.h: Document the issue.
  • dfg/DFGSpeculativeJIT32_64.cpp: Work around the issue.

(JSC::DFG::SpeculativeJIT::fillSpeculateInt32Internal):
(JSC::DFG::SpeculativeJIT::fillSpeculateCell):
(JSC::DFG::SpeculativeJIT::fillSpeculateBoolean):

  • dfg/DFGSpeculativeJIT64.cpp: Work around the issue.

(JSC::DFG::SpeculativeJIT::fillSpeculateInt32Internal):
(JSC::DFG::SpeculativeJIT::fillSpeculateInt52):
(JSC::DFG::SpeculativeJIT::fillSpeculateCell):
(JSC::DFG::SpeculativeJIT::fillSpeculateBoolean):

6:15 PM Changeset in webkit [185940] by commit-queue@webkit.org
  • 4 edits
    3 adds in trunk/Source/WebCore

Building MediaDeviceInfo for enumerateDevices
​https://bugs.webkit.org/show_bug.cgi?id=146257
<rdar://problem/21513462>

Patch by Matt Daiter <​mdaiter@apple.com> on 2015-06-24
Reviewed by Eric Carlson.

MediaDeviceInfo necessary for enumerateDevices

  • CMakeLists.txt: Added MediaDevicesInfo.idl
  • DerivedSources.make: Added MediaDevicesInfo.idl
  • Modules/mediastream/MediaDeviceInfo.h: Created class for MediaDevicesInfo

(WebCore::MediaDeviceInfo::create): Allocates class
(WebCore::MediaDeviceInfo::MediaDeviceInfo): Creates class for public
(WebCore::MediaDeviceInfo::~MediaDeviceInfo):
(WebCore::MediaDeviceInfo::label):
(WebCore::MediaDeviceInfo::deviceId):
(WebCore::MediaDeviceInfo::groupId):
(WebCore::MediaDeviceInfo::kind):

  • Modules/mediastream/MediaDeviceInfo.idl:
  • WebCore.xcodeproj/project.pbxproj:
5:35 PM Changeset in webkit [185939] by andersca@apple.com
  • 5 edits in trunk/Source

Detect view services by using the "com.apple.UIKit.vends-view-services" entitlement
​https://bugs.webkit.org/show_bug.cgi?id=146301

Reviewed by Tim Horton.

Source/WebCore:

  • WebCore.xcodeproj/project.pbxproj:
  • platform/spi/cocoa/SecuritySPI.h:

Source/WebKit2:

  • UIProcess/ApplicationStateTracker.mm:

(WebKit::hasEntitlement):
(WebKit::isViewService):

5:33 PM Changeset in webkit [185938] by bshafiei@apple.com
  • 5 edits in tags/Safari-601.1.35.2.4/Source

Versioning.

5:26 PM Changeset in webkit [185937] by bshafiei@apple.com
  • 1 copy in tags/Safari-601.1.35.2.4

New tag.

4:56 PM Changeset in webkit [185936] by andersca@apple.com
  • 5 edits
    1 move in trunk/Source/WebKit2

WebContent process is unreponsive/idle when WKWebView is inside a view service.
​https://bugs.webkit.org/show_bug.cgi?id=146296

Reviewed by Tim Horton.

  • Platform/spi/ios/AssertionServicesSPI.h: Renamed from Source/WebKit2/Platform/spi/ios/BKSProcessAssertionSPI.h.

Rename this and add SPI for BKSApplicationStateMonitor.

  • Platform/spi/ios/UIKitSPI.h:

Add _UIApplicationIsExtension SPI.

  • UIProcess/ApplicationStateTracker.mm:

(WebKit::isViewService):
Return whether we're an extension or the safari view controller service.

(WebKit::isBackgroundState):
Helper function that returns whether a given BKSApplicationState is a background state.

(WebKit::ApplicationStateTracker::ApplicationStateTracker):
If we're a view service, create a BKSApplicationStateMonitor and listen for state changes for our pid.

  • UIProcess/ios/ProcessAssertionIOS.mm:

Update #import.

  • WebKit2.xcodeproj/project.pbxproj:

Update for rename.

4:54 PM Changeset in webkit [185935] by commit-queue@webkit.org
  • 5 edits in trunk/Source/WebInspectorUI

Web Inspector: Pressing tab in the styles sidebar shouldn't insert a tab character
​https://bugs.webkit.org/show_bug.cgi?id=146189

Patch by Devin Rousso <Devin Rousso> on 2015-06-24
Reviewed by Timothy Hatcher.

  • UserInterface/Controllers/CodeMirrorCompletionController.js: Added variable to control whether semicolons are added to the end of autocompleted css values.

(WebInspector.CodeMirrorCompletionController):
(WebInspector.CodeMirrorCompletionController.prototype.set noEndingSemicolon):
(WebInspector.CodeMirrorCompletionController.prototype._generateCSSCompletions):

  • UserInterface/Views/CSSStyleDeclarationSection.js:

(WebInspector.CSSStyleDeclarationSection):
(WebInspector.CSSStyleDeclarationSection.prototype.cssStyleDeclarationTextEditorSwitchRule):
(WebInspector.CSSStyleDeclarationSection.prototype.focusRuleSelector):
(WebInspector.CSSStyleDeclarationSection.prototype.selectLastProperty):
(WebInspector.CSSStyleDeclarationSection.prototype.get selectorLocked):
(WebInspector.CSSStyleDeclarationSection.prototype.get locked):
(WebInspector.CSSStyleDeclarationSection.prototype._handleKeyDown):

  • UserInterface/Views/CSSStyleDeclarationTextEditor.js:

(WebInspector.CSSStyleDeclarationTextEditor): Added functions for "Tab", "Shift-Tab", and "Shift-Enter" keypresses to improve usability.
(WebInspector.CSSStyleDeclarationTextEditor.prototype.selectFirstProperty): Highlights the first property.
(WebInspector.CSSStyleDeclarationTextEditor.prototype.selectLastProperty): Highlights the last property.
(WebInspector.CSSStyleDeclarationTextEditor.prototype._insertNewlineAfterCurrentLine): Inserts a newline after the currently selected line.
(WebInspector.CSSStyleDeclarationTextEditor.prototype._handleShiftTabKey.switchRule):
(WebInspector.CSSStyleDeclarationTextEditor.prototype._handleShiftTabKey): Pressing shift-tab will move the cursor to the previous non-word character that is immediately after a word character.
(WebInspector.CSSStyleDeclarationTextEditor.prototype._handleTabKey.switchRule):
(WebInspector.CSSStyleDeclarationTextEditor.prototype._handleTabKey.highlightNextNameOrValue):
(WebInspector.CSSStyleDeclarationTextEditor.prototype._handleTabKey): Pressing tab will move the cursor to each space character until the end of the line is reached, at
which point the cursor will move to the beginning of the next line. Once the cursor is at the last line, pressing tab again will insert a newline.

  • UserInterface/Views/RulesStyleDetailsPanel.js:

(WebInspector.RulesStyleDetailsPanel.prototype.cssStyleDeclarationSectionEditorNextRule): Switches the focused rule to the next section.
(WebInspector.RulesStyleDetailsPanel.prototype.cssStyleDeclarationSectionEditorPrevRule): Switches the focused rule to the previous section.

4:48 PM Changeset in webkit [185934] by commit-queue@webkit.org
  • 13 edits in trunk

Add a new color -apple-wireless-playback-target-active
​https://bugs.webkit.org/show_bug.cgi?id=146269
<rdar://problem/21465189>

Patch by Said Abou-Hallawa <​sabouhallawa@apple.com> on 2015-06-24
Reviewed by Dean Jackson.

Source/WebCore:

At the moment we use -apple-system-blue to indicate an Airplay route is
active. This might not always be blue, so we need to add another system
color with a name that will allow us to change the underlying color.
Initially -apple-wireless-playback-target-active will be an alias to
-apple-system-blue.

Existing tests are modified to include the new color value.

  • Modules/mediacontrols/mediaControlsApple.js:

(Controller.prototype.updateWirelessTargetPickerButton):

  • Modules/mediacontrols/mediaControlsiOS.css:

(video::-webkit-media-controls-wireless-playback-picker-button.playing):
Use the new color value in the media controls.

  • css/CSSParser.cpp:

(WebCore::isAppleLegacyCssValueKeyword): Returns true if the value keyword
has an '-apple' prefix which has to be replaced by '-webkit'

(WebCore::cssValueKeywordID): Call a new function which handles all the
'-apple' legacy keywords.

  • css/CSSValueKeywords.in: Define the new color keyword.
  • rendering/RenderThemeIOS.mm:

(WebCore::RenderThemeIOS::systemColor):

  • rendering/RenderThemeMac.mm:

(WebCore::RenderThemeMac::systemColor):
Define the value of the new color keyword to be an alias of systemBlueColor.

LayoutTests:

Modify existing tests to include the new color value.

  • fast/css/apple-system-colors-expected.txt:
  • fast/css/apple-system-colors.html:
  • platform/ios-simulator/fast/css/apple-system-colors-expected.txt:
  • platform/mac-mavericks/fast/css/apple-system-colors-expected.txt:
  • platform/mac/fast/css/apple-system-colors-expected.txt:
4:44 PM Changeset in webkit [185933] by Joseph Pecoraro
  • 4 edits in trunk/Source/WebInspectorUI

Web Inspector: console.group looks poor in console
​https://bugs.webkit.org/show_bug.cgi?id=146295

Reviewed by Timothy Hatcher.

  • Localizations/en.lproj/localizedStrings.js:

String for the default group name if you use console.group() without a name parameter.

  • UserInterface/Views/LogContentView.css:

(.console-group-title::before):
Tweak the positioning and eliminate the log level ::before image.

  • UserInterface/Views/ConsoleMessageView.js:

(WebInspector.ConsoleMessageView):
Reorder a bit to reduce a couple branches.

(WebInspector.ConsoleMessageView.prototype._appendMessageTextAndArguments):
Take a very straightforward approach for group names.

3:50 PM Changeset in webkit [185932] by msaboff@apple.com
  • 3 edits in trunk/Source/JavaScriptCore

Crash on gog.com due to PolymorphicCallNode's having stale references to CallLinkInfo
​https://bugs.webkit.org/show_bug.cgi?id=146285

Reviewed by Filip Pizlo.

CallLinkInfo's contain a RefPtr to a PolymorphicCallStubRoutine, named stub, which contains
a collection of PolymorphicCallNode. Those PolymorphicCallNodes have a reference back to the
CallLinkInfo. When a CallLinkInfo replaces or clears "stub", the ref count of the
PolymorphicCallStubRoutine is decremented as expected, but since it inherits from
GCAwareJITStubRoutine, it isn't actually deleted until GC. In the mean time, the original
CallLinkInfo can go away. If PolymorphicCallNode::unlink() is called at that point,
it will try to unlink a now deleted CallLinkInfo and crash as a result.

The fix is to clear the CallLinkInfo references from any PolymorphicCallNode objects when
when we set a new stub or clear an existing stub for a CallLinkInfo. This is done by
calling PolymorphicCallNode::clearCallNodesFor() on the old stub.

The prior code would only call clearCallNodesFor() from the CallLinkInfo destructor.
This only took care of the last PolymorphicCallStubRoutine held in the CallLinkInfo.
Any prior PolymorphicCallStubRoutine would still have a, now bad, reference to the CallLinkInfo.

In the process I refactored CallLinkInfo from a struct to a class with proper accessors and
made all the data elements private.

  • bytecode/CallLinkInfo.cpp:

(JSC::CallLinkInfo::clearStub): Updated to call PolymorphicCallStubRoutine::clearCallNodesFor()
to clear the back references to this CallLinkInfo.

  • bytecode/CallLinkInfo.h:

(JSC::CallLinkInfo::~CallLinkInfo): Moved clearCallNodesFor() call to clearStub().
(JSC::CallLinkInfo::setStub): Clear any prior stub before changing to the new stub.

3:41 PM Changeset in webkit [185931] by commit-queue@webkit.org
  • 2 edits in trunk/Tools

Work towards: Add status of Safari projects on iOS Dashboard on build-safari
​https://bugs.webkit.org/show_bug.cgi?id=146126
<rdar://problem/18226291>

Patch by Jason Marcell <​jmarcell@apple.com> on 2015-06-24
Reviewed by Alexey Proskuryakov.

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

(documentReady): Using combined queues to display iOS projects

3:37 PM Changeset in webkit [185930] by msaboff@apple.com
  • 21 edits in trunk/Source/JavaScriptCore

Refactor CallLinkInfo from a struct to a class
​https://bugs.webkit.org/show_bug.cgi?id=146292

Rubber stamped by Filip Pizlo.

Refactored CallLinkInfo from a struct to a class with proper accessors and made all the
data elements private.

Done in preparation for fixing ​https://bugs.webkit.org/show_bug.cgi?id=146285.

  • bytecode/CallLinkInfo.cpp:

(JSC::CallLinkInfo::clearStub):
(JSC::CallLinkInfo::unlink):
(JSC::CallLinkInfo::visitWeak):

  • bytecode/CallLinkInfo.h:

(JSC::CallLinkInfo::callTypeFor):
(JSC::CallLinkInfo::CallLinkInfo):
(JSC::CallLinkInfo::~CallLinkInfo):
(JSC::CallLinkInfo::specializationKindFor):
(JSC::CallLinkInfo::specializationKind):
(JSC::CallLinkInfo::isLinked):
(JSC::CallLinkInfo::setUpCall):
(JSC::CallLinkInfo::setCallLocations):
(JSC::CallLinkInfo::setUpCallFromFTL):
(JSC::CallLinkInfo::callReturnLocation):
(JSC::CallLinkInfo::hotPathBegin):
(JSC::CallLinkInfo::hotPathOther):
(JSC::CallLinkInfo::setCallee):
(JSC::CallLinkInfo::clearCallee):
(JSC::CallLinkInfo::callee):
(JSC::CallLinkInfo::setLastSeenCallee):
(JSC::CallLinkInfo::clearLastSeenCallee):
(JSC::CallLinkInfo::lastSeenCallee):
(JSC::CallLinkInfo::haveLastSeenCallee):
(JSC::CallLinkInfo::setStub):
(JSC::CallLinkInfo::stub):
(JSC::CallLinkInfo::seenOnce):
(JSC::CallLinkInfo::clearSeen):
(JSC::CallLinkInfo::setSeen):
(JSC::CallLinkInfo::hasSeenClosure):
(JSC::CallLinkInfo::setHasSeenClosure):
(JSC::CallLinkInfo::clearedByGC):
(JSC::CallLinkInfo::setCallType):
(JSC::CallLinkInfo::callType):
(JSC::CallLinkInfo::addressOfMaxNumArguments):
(JSC::CallLinkInfo::maxNumArguments):
(JSC::CallLinkInfo::offsetOfSlowPathCount):
(JSC::CallLinkInfo::setCalleeGPR):
(JSC::CallLinkInfo::calleeGPR):
(JSC::CallLinkInfo::slowPathCount):
(JSC::CallLinkInfo::setCodeOrigin):
(JSC::CallLinkInfo::codeOrigin):
(JSC::getCallLinkInfoCodeOrigin):

  • bytecode/CallLinkStatus.cpp:

(JSC::CallLinkStatus::computeFor):
(JSC::CallLinkStatus::computeFromCallLinkInfo):
(JSC::CallLinkStatus::computeDFGStatuses):

  • bytecode/CallLinkStatus.h:
  • bytecode/CodeBlock.cpp:

(JSC::CodeBlock::printCallOp):
(JSC::CodeBlock::getCallLinkInfoForBytecodeIndex):

  • dfg/DFGJITCompiler.cpp:

(JSC::DFG::JITCompiler::link):

  • dfg/DFGOSRExitCompilerCommon.cpp:

(JSC::DFG::reifyInlinedCallFrames):

  • dfg/DFGSpeculativeJIT32_64.cpp:

(JSC::DFG::SpeculativeJIT::emitCall):

  • dfg/DFGSpeculativeJIT64.cpp:

(JSC::DFG::SpeculativeJIT::emitCall):

  • ftl/FTLJSCallBase.cpp:

(JSC::FTL::JSCallBase::link):

  • jit/AccessorCallJITStubRoutine.h:
  • jit/JIT.cpp:

(JSC::JIT::privateCompile):

  • jit/JIT.h:
  • jit/JITCall.cpp:

(JSC::JIT::compileSetupVarargsFrame):
(JSC::JIT::compileOpCall):

  • jit/JITCall32_64.cpp:

(JSC::JIT::compileSetupVarargsFrame):
(JSC::JIT::compileOpCall):

  • jit/JITOperations.cpp:
  • jit/PolymorphicCallStubRoutine.cpp:

(JSC::PolymorphicCallNode::unlink):
(JSC::PolymorphicCallNode::clearCallLinkInfo):

  • jit/PolymorphicCallStubRoutine.h:
  • jit/Repatch.cpp:

(JSC::generateByIdStub):
(JSC::linkSlowFor):
(JSC::linkFor):
(JSC::revertCall):
(JSC::unlinkFor):
(JSC::linkPolymorphicCall):

  • jit/ThunkGenerators.cpp:

(JSC::virtualForThunkGenerator):

3:36 PM Changeset in webkit [185929] by mrajca@apple.com
  • 11 edits in trunk/Source

MediaSession: propagate MediaSessionMetadata to WebPageProxy
​https://bugs.webkit.org/show_bug.cgi?id=146282

Reviewed by Darin Adler and Tim Horton.

WebCore:

  • Modules/mediasession/MediaSession.cpp:

(WebCore::MediaSession::setMetadata):

WebKit2:

  • Shared/WebCoreArgumentCoders.cpp: Added ArgumentEncoder support for MediaSessionMetadata so we can send it through the web page proxy.

(IPC::ArgumentCoder<MediaSessionMetadata>::encode):
(IPC::ArgumentCoder<MediaSessionMetadata>::decode):

  • Shared/WebCoreArgumentCoders.h:
  • UIProcess/WebPageProxy.cpp:

(WebKit::WebPageProxy::mediaSessionMetadataDidChange):

  • UIProcess/WebPageProxy.h: Receive mediaSessionMetadataDidChange in the web page proxy.
  • UIProcess/WebPageProxy.messages.in: Added a new MediaSessionMetadataDidChange message.
  • WebProcess/WebCoreSupport/WebChromeClient.cpp:

(WebKit::WebChromeClient::mediaSessionMetadataDidChange): Push a mediaSessionMetadataDidChange message through the web page proxy.

  • WebProcess/WebCoreSupport/WebChromeClient.h:
3:17 PM Changeset in webkit [185928] by commit-queue@webkit.org
  • 6 edits in trunk/Source/WebInspectorUI

Web Inspector: Show warning icon for invalid CSS properties and/or values
​https://bugs.webkit.org/show_bug.cgi?id=145657

Patch by Devin Rousso <Devin Rousso> on 2015-06-24
Reviewed by Timothy Hatcher.

  • UserInterface/Models/CSSCompletions.js:

(WebInspector.CSSCompletions.prototype.getClosestPropertyName): Calculates the levenshtein distance between a given property and every existing property name. Returns the property name with the smallest distance or, in the case of multiple properties having the same distance, the first property in alphabetical order.
(WebInspector.CSSCompletions.prototype.propertyRequiresWebkitPrefix): Retruns if the property name exists only with a '-webkit-' prefix.
(WebInspector.CSSCompletions):

  • UserInterface/Views/CSSStyleDeclarationTextEditor.css:

(.css-style-text-editor > .CodeMirror .CodeMirror-lines .invalid-warning-marker):
(.css-style-text-editor > .CodeMirror .CodeMirror-lines .invalid-warning-marker.clickable:hover):

  • UserInterface/Views/CSSStyleDeclarationTextEditor.js:

(WebInspector.CSSStyleDeclarationTextEditor.prototype._createTextMarkerForPropertyIfNeeded.duplicatePropertyExistsBelow): Determines if there exists a property below (visually) the given property that has the same name.
(WebInspector.CSSStyleDeclarationTextEditor.prototype._createTextMarkerForPropertyIfNeeded.generateInvalidMarker): Creates a warning icon marker at the given position with the given title.
(WebInspector.CSSStyleDeclarationTextEditor.prototype._createTextMarkerForPropertyIfNeeded.instancesOfProperty): Returns the number of properties in the rule that have the same name.
(WebInspector.CSSStyleDeclarationTextEditor.prototype._createTextMarkerForPropertyIfNeeded): A button with a warning icon is now added to the beginning of an invalid property. If this button is clicked, the property is replaced with the closest matching property and the autocomplete menu is opened.

2:43 PM Changeset in webkit [185927] by Chris Dumez
  • 2 edits in trunk/Source/WebCore

Null dereference in DocumentLoader::areAllLoadersPageCacheAcceptable()
​https://bugs.webkit.org/show_bug.cgi?id=146286
<rdar://problem/21523788>

Reviewed by Sam Weinig.

Add null check for the Page in areAllLoadersPageCacheAcceptable()
to fix this top crasher until I can investigate how this can happen.

  • loader/DocumentLoader.cpp:

(WebCore::areAllLoadersPageCacheAcceptable):

2:34 PM Changeset in webkit [185926] by andersca@apple.com
  • 11 edits
    2 moves in trunk/Source

Move PluginMainThreadScheduler to WebKit/win
​https://bugs.webkit.org/show_bug.cgi?id=146289

Reviewed by Tim Horton.

Source/WebCore:

  • CMakeLists.txt:
  • Configurations/WebCore.xcconfig:
  • WebCore.vcxproj/WebCore.vcxproj:
  • WebCore.vcxproj/WebCore.vcxproj.filters:
  • WebCore.xcodeproj/project.pbxproj:
  • plugins/PluginMainThreadScheduler.cpp:

(WebCore::PluginMainThreadScheduler::scheduler): Deleted.
(WebCore::PluginMainThreadScheduler::PluginMainThreadScheduler): Deleted.
(WebCore::PluginMainThreadScheduler::scheduleCall): Deleted.
(WebCore::PluginMainThreadScheduler::registerPlugin): Deleted.
(WebCore::PluginMainThreadScheduler::unregisterPlugin): Deleted.
(WebCore::PluginMainThreadScheduler::dispatchCallsForPlugin): Deleted.
(WebCore::PluginMainThreadScheduler::dispatchCalls): Deleted.
(WebCore::PluginMainThreadScheduler::mainThreadCallback): Deleted.

  • plugins/PluginMainThreadScheduler.h:

(WebCore::PluginMainThreadScheduler::Call::Call): Deleted.
(WebCore::PluginMainThreadScheduler::Call::performCall): Deleted.

Source/WebKit:

  • WebKit.vcxproj/WebKit/WebKit.vcxproj:
  • WebKit.vcxproj/WebKit/WebKit.vcxproj.filters:

Source/WebKit/win:

  • Plugins/PluginMainThreadScheduler.cpp: Renamed from Source/WebCore/plugins/PluginMainThreadScheduler.cpp.

(WebCore::PluginMainThreadScheduler::scheduler):
(WebCore::PluginMainThreadScheduler::PluginMainThreadScheduler):
(WebCore::PluginMainThreadScheduler::scheduleCall):
(WebCore::PluginMainThreadScheduler::registerPlugin):
(WebCore::PluginMainThreadScheduler::unregisterPlugin):
(WebCore::PluginMainThreadScheduler::dispatchCallsForPlugin):
(WebCore::PluginMainThreadScheduler::dispatchCalls):
(WebCore::PluginMainThreadScheduler::mainThreadCallback):

  • Plugins/PluginMainThreadScheduler.h: Renamed from Source/WebCore/plugins/PluginMainThreadScheduler.h.

(WebCore::PluginMainThreadScheduler::Call::Call):
(WebCore::PluginMainThreadScheduler::Call::performCall):

  • Plugins/PluginView.cpp:
1:54 PM Changeset in webkit [185925] by dino@apple.com
  • 7 edits in trunk/Source/WebCore

Better fix for Element with blur backdrop-filter shows edge duplication and dark edges
​https://bugs.webkit.org/show_bug.cgi?id=146287
<rdar://problem/21530437>

Reviewed by Simon Fraser.

There is no need to carry around a bit in FilterOperations indicating what
type of source property the object came from. Whether or not to clip to the
bounds of the filter region is purely a property of the type of CALayer
we are applying the filter to.

This basically reverts r185846.

Covered by the test for the previous commit.

  • css/CSSPropertyNames.in:
  • css/StyleBuilderConverter.h:

(WebCore::StyleBuilderConverter::convertBackdropFilterOperations): Deleted.

  • page/animation/CSSPropertyAnimation.cpp:

(WebCore::blendFilterOperations): Deleted.

  • platform/graphics/ca/cocoa/PlatformCAFiltersCocoa.mm: Set the normalized

edges property if we are a CABackdropLayer.

  • platform/graphics/filters/FilterOperations.cpp:

(WebCore::FilterOperations::operator==):
(WebCore::FilterOperations::operationsMatch):
(WebCore::FilterOperations::operator=): Deleted.

  • platform/graphics/filters/FilterOperations.h:

(WebCore::FilterOperations::isUsedForBackdropFilters): Deleted.
(WebCore::FilterOperations::setUsedForBackdropFilters): Deleted.

1:51 PM Changeset in webkit [185924] by commit-queue@webkit.org
  • 13 edits
    2 adds in trunk

Source/JavaScriptCore:
Bug 146177 - AX: AXObjectCache should try to use an unignored accessibilityObject
when posting a selection notification when on the border between two accessibilityObjects
​https://bugs.webkit.org/show_bug.cgi?id=146177

Add an adopt() function to simplify JSRetainPtr<JSStringRef> { Adopt, string } to adopt(string).

Patch by Doug Russell <​d_russell@apple.com> on 2015-06-24
Reviewed by Darin Adler.

  • API/JSRetainPtr.h:

(adopt):

Source/WebCore:
Bug 146177 - AX: AXObjectCache should try to use an unignored accessibilityObject
when posting a selection notification when on the border between two accessibilityObjects
​https://bugs.webkit.org/show_bug.cgi?id=146177

Patch by Doug Russell <​d_russell@apple.com> on 2015-06-24
Reviewed by Darin Adler.

Add support for position to be passed for selection changes to allow checking.
for boundaries in the case of ignored elements.
Add support for searching for unignored siblings of AccessibilityObjects.
Include AccessibilityObject wrappers in notifications for tests.

Test: platform/mac/accessibility/selection-element-tabbing-to-link.html

  • accessibility/AXObjectCache.cpp:

(WebCore::AXObjectCache::postTextStateChangeNotification):

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

(WebCore::AccessibilityObject::previousSiblingUnignored):
(WebCore::AccessibilityObject::nextSiblingUnignored):

  • accessibility/AccessibilityObject.h:
  • accessibility/mac/WebAccessibilityObjectWrapperBase.mm:

(isValueTypeSupported):
(arrayRemovingNonSupportedTypes):
(dictionaryRemovingNonSupportedTypes):
(-[WebAccessibilityObjectWrapperBase accessibilityPostedNotification:userInfo:]):

  • editing/mac/FrameSelectionMac.mm:

(WebCore::FrameSelection::notifyAccessibilityForSelectionChange):

Tools:
Bug 146177 - AX: AXObjectCache should try to use an unignored accessibilityObject
when posting a selection notification when on the border between two accessibilityObjects
​https://bugs.webkit.org/show_bug.cgi?id=146177

Patch by Doug Russell <​d_russell@apple.com> on 2015-06-24
Reviewed by Darin Adler.

Add support for mapping WebCore Accessibility types into JSValueRef types
so they can be used in tests.

  • DumpRenderTree/mac/AccessibilityNotificationHandler.mm:

(webAccessibilityObjectWrapperClass):
(-[AccessibilityNotificationHandler startObserving]):
(makeValueRefForValue):
(makeArrayRefForArray):
(makeObjectRefForDictionary):
(-[AccessibilityNotificationHandler _notificationReceived:]):

  • WebKitTestRunner/InjectedBundle/mac/AccessibilityNotificationHandler.mm:

(webAccessibilityObjectWrapperClass):
(-[AccessibilityNotificationHandler startObserving]):
(makeValueRefForValue):
(makeArrayRefForArray):
(makeObjectRefForDictionary):
(-[AccessibilityNotificationHandler _notificationReceived:]):

LayoutTests:
Bug 146177 - AX: AXObjectCache should try to use an unignored accessibilityObject
when posting a selection notification when on the border between two accessibilityObjects
​https://bugs.webkit.org/show_bug.cgi?id=146177

Add test for forward and backward tabbing between links and corresponding notifications
change element

Patch by Doug Russell <​d_russell@apple.com> on 2015-06-24
Reviewed by Darin Adler.

  • platform/mac/accessibility/selection-element-tabbing-to-link-expected.txt: Added.
  • platform/mac/accessibility/selection-element-tabbing-to-link.html: Added.
1:48 PM Changeset in webkit [185923] by andersca@apple.com
  • 19 edits in trunk/Source/WebCore

Enable -Winconsistent-missing-override when building with Xcode
​https://bugs.webkit.org/show_bug.cgi?id=146288

Reviewed by Sam Weinig.

  • Configurations/Base.xcconfig:
  • Modules/encryptedmedia/MediaKeySession.h:
  • Modules/webdatabase/DatabaseTask.h:
  • html/DateTimeLocalInputType.h:
  • page/scrolling/ios/ScrollingCoordinatorIOS.h:
  • page/scrolling/mac/ScrollingCoordinatorMac.h:
  • platform/graphics/avfoundation/MediaPlayerPrivateAVFoundation.h:

(WebCore::MediaPlayerPrivateAVFoundation::platformLayer): Deleted.
(WebCore::MediaPlayerPrivateAVFoundation::supportsScanning): Deleted.
(WebCore::MediaPlayerPrivateAVFoundation::fileSize): Deleted.
(WebCore::MediaPlayerPrivateAVFoundation::engineDescription): Deleted.
(WebCore::MediaPlayerPrivateAVFoundation::platformErrorCode): Deleted.
(WebCore::MediaPlayerPrivateAVFoundation::notifyTrackModeChanged): Deleted.

  • platform/graphics/avfoundation/VideoTrackPrivateAVF.h:

(WebCore::VideoTrackPrivateAVF::trackIndex): Deleted.

  • platform/graphics/avfoundation/objc/CDMSessionAVFoundationObjC.h:

(WebCore::CDMSessionAVFoundationObjC::type): Deleted.

  • platform/graphics/avfoundation/objc/CDMSessionMediaSourceAVFObjC.h:

(WebCore::CDMSessionMediaSourceAVFObjC::type): Deleted.

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

(WebCore::MediaPlayerPrivateAVFoundationObjC::supportsAcceleratedRendering): Deleted.
(WebCore::MediaPlayerPrivateAVFoundationObjC::canPlayToWirelessPlaybackTarget): Deleted.

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

(WebCore::WebCoreNSURLAuthenticationChallengeClient::receivedCredential): Deleted.
(WebCore::WebCoreNSURLAuthenticationChallengeClient::receivedRequestToContinueWithoutCredential): Deleted.
(WebCore::WebCoreNSURLAuthenticationChallengeClient::receivedCancellation): Deleted.
(WebCore::WebCoreNSURLAuthenticationChallengeClient::receivedRequestToPerformDefaultHandling): Deleted.
(WebCore::WebCoreNSURLAuthenticationChallengeClient::receivedChallengeRejection): Deleted.

  • platform/graphics/avfoundation/objc/MediaPlayerPrivateMediaSourceAVFObjC.h:

(WebCore::MediaPlayerPrivateMediaSourceAVFObjC::supportsFullscreen): Deleted.

  • platform/graphics/mac/MediaPlayerPrivateQTKit.h:

(WebCore::MediaPlayerPrivateQTKit::supportsScanning): Deleted.
(WebCore::MediaPlayerPrivateQTKit::networkState): Deleted.
(WebCore::MediaPlayerPrivateQTKit::readyState): Deleted.
(WebCore::MediaPlayerPrivateQTKit::maximumDurationToCacheMediaTime): Deleted.
(WebCore::MediaPlayerPrivateQTKit::engineDescription): Deleted.

  • platform/mac/ScrollAnimatorMac.h:
  • platform/mac/SerializedPlatformRepresentationMac.h:

(WebCore::SerializedPlatformRepresentationMac::platformType): Deleted.

  • platform/mac/ThemeMac.h:

(WebCore::ThemeMac::controlRequiresPreWhiteSpace): Deleted.

  • testing/MockCDM.cpp:

(WebCore::MockCDMSession::setClient): Deleted.

1:32 PM Changeset in webkit [185922] by commit-queue@webkit.org
  • 2 edits in trunk/Source/WebKit2

[Mac] A checkmark is briefly visible next to the webpage context menu items when they're clicked
​https://bugs.webkit.org/show_bug.cgi?id=146284
<rdar://problem/20297182>

Patch by Said Abou-Hallawa <​sabouhallawa@apple.com> on 2015-06-24
Reviewed by Darin Adler.

The default value of [NSPopUpButtonCell altersStateOfSelectedItem] is YES.
This makes the popup menu items checkable by default. We need to change the
value of this attribute to NO when creating the popup menu, so the items
are uncheckable. The value of this property needs to change back to YES
if context menu is about selecting an option and not executing a command.

  • UIProcess/mac/WebContextMenuProxyMac.mm:

(WebKit::WebContextMenuProxyMac::populate):

1:24 PM Changeset in webkit [185921] by Joseph Pecoraro
  • 1 edit
    12 adds in trunk/Websites/webkit.org

Add some images for a blog post.

  • blog-files/console-improvements/array-api.png: Added.
  • blog-files/console-improvements/collections.png: Added.
  • blog-files/console-improvements/console-messages.png: Added.
  • blog-files/console-improvements/interactivity.m4v: Added.
  • blog-files/console-improvements/iterators.png: Added.
  • blog-files/console-improvements/localStorage-api.png: Added.
  • blog-files/console-improvements/object-preview.png: Added.
  • blog-files/console-improvements/object-tree.png: Added.
  • blog-files/console-improvements/promises.png: Added.
  • blog-files/console-improvements/tooltip-1.png: Added.
  • blog-files/console-improvements/tooltip-2.png: Added.
12:13 PM Changeset in webkit [185920] by commit-queue@webkit.org
  • 10 edits
    4 adds in trunk

Strict Equality on objects should only check that one of the two sides is an object.
​https://bugs.webkit.org/show_bug.cgi?id=145992

Source/JavaScriptCore:

This patch adds a new optimization for checking strict equality on objects.
If we speculate that a strict equality comparison has an object on one side
we only need to type check that side. Equality is then determined by a pointer
comparison between the two values (although in the 32-bit case we must also check
that the other side is a cell). Once LICM hoists type checks out of a loop we
can be cleverer about how we choose the operand we type check if both are
speculated to be objects.

For testing I added the addressOf function, which returns the address
of a Cell to the runtime.

Patch by Keith Miller <​keith_miller@apple.com> on 2015-06-24
Reviewed by Mark Lam.

  • dfg/DFGFixupPhase.cpp:

(JSC::DFG::FixupPhase::fixupNode):

  • dfg/DFGSpeculativeJIT.cpp:

(JSC::DFG::SpeculativeJIT::compileStrictEq):

  • dfg/DFGSpeculativeJIT.h:
  • dfg/DFGSpeculativeJIT32_64.cpp:

(JSC::DFG::SpeculativeJIT::compileObjectStrictEquality):
(JSC::DFG::SpeculativeJIT::compilePeepHoleObjectStrictEquality):

  • dfg/DFGSpeculativeJIT64.cpp:

(JSC::DFG::SpeculativeJIT::compileObjectStrictEquality):
(JSC::DFG::SpeculativeJIT::compilePeepHoleObjectStrictEquality):

  • ftl/FTLCapabilities.cpp:

(JSC::FTL::canCompile):

  • ftl/FTLLowerDFGToLLVM.cpp:

(JSC::FTL::DFG::LowerDFGToLLVM::compileCompareStrictEq):

  • jsc.cpp:

(GlobalObject::finishCreation):
(functionAddressOf):

  • tests/stress/equality-type-checking.js: Added.

(Foo):
(checkStrictEq):
(checkStrictEqOther):

LayoutTests:

Patch by Keith Miller <​keith_miller@apple.com> on 2015-06-24
Reviewed by Mark Lam.

Adds a test that checks if strict equality checks with objects properly exit out of DFG code when
dealing with document.all, which is an object that masquerades as undefined.

  • js/dom/document-all-strict-eq-expected.txt: Added.
  • js/dom/document-all-strict-eq.html: Added.
  • js/dom/script-tests/document-all-strict-eq.js: Added.

(f):

12:08 PM Changeset in webkit [185919] by youenn.fablet@crf.canon.fr
  • 2 edits in trunk/Source/WebCore

Improve JSDOMPromise callPromiseFunction naming
​https://bugs.webkit.org/show_bug.cgi?id=146280

Unreviewed.

No change in behavior.

  • bindings/js/JSDOMPromise.h:

(WebCore::callPromiseFunction):

11:41 AM Changeset in webkit [185918] by mmaxfield@apple.com
  • 1 edit
    4 adds in trunk/LayoutTests

[Win] Test gardening for fast/text/font-weight{,-zh}.html
​https://bugs.webkit.org/show_bug.cgi?id=146266
<rdar://problem/21391628>
<rdar://problem/21391673>

Unreviewed.

  • platform/win/fast/text/font-weights-expected.png: Added.
  • platform/win/fast/text/font-weights-expected.txt: Added.
  • platform/win/fast/text/font-weights-zh-expected.png: Added.
  • platform/win/fast/text/font-weights-zh-expected.txt: Added.
10:57 AM Changeset in webkit [185917] by timothy_horton@apple.com
  • 2 edits in trunk/Source/WebCore

REGRESSION (r185804): Crash selecting time in the form on ​https://backup.brighthorizons.com/BackupCareRequest.aspx
<rdar://problem/21525376>

Reviewed by Sam Weinig.

  • page/TextIndicator.cpp:

(WebCore::TextIndicator::createWithRange):
Null check.

10:53 AM Changeset in webkit [185916] by Alan Bujtas
  • 4 edits in trunk

Subpixel rendering: roundToDevicePixel() snaps to wrong value.
​https://bugs.webkit.org/show_bug.cgi?id=146273
rdar://problem/18509840

Reviewed by Simon Fraser.

Due to the floating point approximate representation, we can't always produce
the correct snap value. This patch addresses the issue by removing redundant kFixedPointDenominator multiplication
and by changing the rounding in roundToDevicePixel() from float to double.

Source/WebCore:

API test is added.

  • platform/LayoutUnit.h:

(WebCore::roundToDevicePixel):

Tools:

  • TestWebKitAPI/Tests/WebCore/LayoutUnit.cpp:

(TestWebKitAPI::TEST):

10:21 AM WebKitGTK/2.8.x edited by berto@igalia.com
(diff)
9:57 AM Changeset in webkit [185915] by beidson@apple.com
  • 11 edits
    1 copy in trunk

Update JavaScript dialog delegates to include a WKSecurityOriginRef argument.
<rdar://problem/21269187> and ​https://bugs.webkit.org/show_bug.cgi?id=146249

Reviewed by Alex Christensen.

Source/WebKit2:

  • UIProcess/API/C/WKPage.cpp:

(WKPageSetPageUIClient): Call the new signature if the client has it set.

Otherwise fall back to the old signature.

  • UIProcess/API/C/WKPageUIClient.h: Deprecate the old method signature, and add the new one to the newest V5 client structure.

Tools:

  • Update WKTR to the new client structure.
  • Update existing TestWebKitAPI tests to either use the new client structure or assign the old function signature to the updated variable name.
  • Include a new test.
  • TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj:
  • TestWebKitAPI/Tests/WebKit2/CloseFromWithinCreatePage.cpp:

(TestWebKitAPI::runJavaScriptAlert):

  • TestWebKitAPI/Tests/WebKit2/DocumentStartUserScriptAlertCrash.cpp:

(TestWebKitAPI::TEST):

  • TestWebKitAPI/Tests/WebKit2/ModalAlertsSPI.cpp:

(TestWebKitAPI::analyzeDialogArguments):
(TestWebKitAPI::runJavaScriptAlert):
(TestWebKitAPI::runJavaScriptConfirm):
(TestWebKitAPI::runJavaScriptPrompt):
(TestWebKitAPI::createNewPage):
(TestWebKitAPI::TEST):

  • TestWebKitAPI/Tests/mac/FullscreenZoomInitialFrame.mm:

(TestWebKitAPI::FullscreenZoomInitialFrame::initializeView):

  • TestWebKitAPI/Tests/mac/PageVisibilityStateWithWindowChanges.mm:

(TestWebKitAPI::PageVisibilityStateWithWindowChanges::initializeView):

  • WebKitTestRunner/TestController.cpp:

(WTR::TestController::createOtherPage):
(WTR::TestController::createWebViewWithOptions):

9:38 AM Changeset in webkit [185914] by mrajca@apple.com
  • 5 edits
    1 copy in trunk/Source/WebCore

MediaSession: Support setting of metadata
​https://bugs.webkit.org/show_bug.cgi?id=146253

Reviewed by Eric Carlson.

  • Modules/mediasession/MediaSession.cpp:

(WebCore::MediaSession::MediaSession): Store a reference to the document so we can use it in setMetadata.
(WebCore::MediaSession::setMetadata): Implemented as described in the Media Session spec.

  • Modules/mediasession/MediaSession.h:
  • Modules/mediasession/MediaSession.idl: Added the setMetadata method.
  • Modules/mediasession/MediaSessionMetadata.h:

(WebCore::MediaSessionMetadata::MediaSessionMetadata): Added to keep track of media session metadata.
(WebCore::MediaSessionMetadata::title):
(WebCore::MediaSessionMetadata::artist):
(WebCore::MediaSessionMetadata::album):
(WebCore::MediaSessionMetadata::artworkURL):

  • WebCore.xcodeproj/project.pbxproj: Added MediaSessionMetadata.h.
9:36 AM Changeset in webkit [185913] by commit-queue@webkit.org
  • 2 edits in trunk/Source/WebInspectorUI

Web Inspector: Background of Computed Styles is missing
​https://bugs.webkit.org/show_bug.cgi?id=146209

Patch by Devin Rousso <Devin Rousso> on 2015-06-24
Reviewed by Timothy Hatcher.

  • UserInterface/Views/ComputedStyleDetailsPanel.css:

(.computed-style-properties.details-section): Added white background.

8:54 AM Changeset in webkit [185912] by dbates@webkit.org
  • 5 edits in trunk

Content Security Policy error message when frame load is blocked does not read well
​https://bugs.webkit.org/show_bug.cgi?id=146279

Reviewed by Csaba Osztrogonác.

Source/WebCore:

Substitute "Refused to load frame" for "Refused to frame" as the prefix of the error message
emitted when the Content Security Policy of the page blocks a frame load.

  • page/ContentSecurityPolicy.cpp:

(WebCore::CSPDirectiveList::checkSourceAndReportViolation):

LayoutTests:

Update expected results.

  • http/tests/security/contentSecurityPolicy/frame-src-blocked-expected.txt:
  • http/tests/security/contentSecurityPolicy/frame-src-redirect-blocked-expected.txt:
8:32 AM Changeset in webkit [185911] by commit-queue@webkit.org
  • 3 edits in trunk/Source/WebInspectorUI

Web Inspector: Right/Left arrow no longer works in console to expand/collapse ObjectTrees
​https://bugs.webkit.org/show_bug.cgi?id=141949

Patch by Joseph Pecoraro <Joseph Pecoraro> on 2015-06-24
Reviewed by Timothy Hatcher.

  • UserInterface/Views/ConsoleMessageView.js:

(WebInspector.ConsoleMessageView):
(WebInspector.ConsoleMessageView.prototype.get expandable):
(WebInspector.ConsoleMessageView.prototype.set expandable): Deleted.
(WebInspector.ConsoleMessageView.prototype._appendExtraParameters):
(WebInspector.ConsoleMessageView.prototype._appendStackTrace):
(WebInspector.ConsoleMessageView.prototype._makeExpandable):
Clarify what makes a ConsoleMessageView expandable or not.

(WebInspector.ConsoleMessageView.prototype.expand):
(WebInspector.ConsoleMessageView.prototype.collapse):
Handle the expand/collapse cases which could apply to the
ConsoleMessageView itself or an inner ObjectTreeView if that is all we contain.

(WebInspector.ConsoleMessageView.prototype._formatParameterAsObject):
Delete out of date comment. A console.log("message", obj) will want the
object tree member variable to handle auto expanding/collapsing of the
single object in the message.

  • UserInterface/Views/LogContentView.js:

(WebInspector.LogContentView.prototype._leftArrowWasPressed):
(WebInspector.LogContentView.prototype._rightArrowWasPressed):
Use the higher level ConsoleMessageView objects accessible from the element
to call through to expand/collapse.

7:47 AM Changeset in webkit [185910] by dbates@webkit.org
  • 2 edits in trunk/LayoutTests

LayoutTests/http/tests/security/contentSecurityPolicy/media-src-track-block.html always passes
​https://bugs.webkit.org/show_bug.cgi?id=146272

Reviewed by Darin Adler.

Fixes an issue where the test LayoutTests/http/tests/security/contentSecurityPolicy/media-src-track-block.html
will always pass regardless of whether the Content Security Policy <meta> tag is present because
the external text track file is not valid WebVTT-formatted file.

The test verifies that the external text track file LayoutTests/http/tests/security/contentSecurityPolicy/resources/track.vtt
fails to load when the Content Security Policy of the page disallows loading of all media (media-src: "none")
by listening for a DOM error event to be dispatched to the <track>. And the text track file is not a valid
WebVTT-formatted file. So, it will always fail to load; => a DOM error event is always dispatched to the
<track>; => the test always passes.

  • http/tests/security/contentSecurityPolicy/resources/track.vtt:
7:18 AM Changeset in webkit [185909] by mark.lam@apple.com
  • 2 edits in trunk/Source/JavaScriptCore

Fixed assertion in JSStringJoiner::join() (regression from r185899).

Not reviewed.

JSStringJoiner did not account for the case where the array being joined can
have null or undefined elements. As a result, its size may be less than
its initially reserved capacity (which was estimated based on the array length).

  • runtime/JSStringJoiner.cpp:

(JSC::JSStringJoiner::join):

6:36 AM Changeset in webkit [185908] by svillar@igalia.com
  • 15 edits
    1 copy
    19 adds in trunk

intrinsic size keywords don't work for heights
​https://bugs.webkit.org/show_bug.cgi?id=113610

Reviewed by Darin Adler.

Source/WebCore:

Based on Blink's r148314 & r150355 by <​cbiesinger@chromium.org>.

Adds intrinsic values support to heigh & min/max-height. This involves the
following changes:

  • RenderBox needs to pass the content height through to computeLogicalHeight and

related functions, which needs it to resolve max-content, et. al.

  • Make the callers pass the content height to this function. Some callers pass

logicalHeight() (adjusted for border/padding) which works because if
logicalHeight is not the content height, then it is the height we ended up using,
so the constrain* functions will just constrain to that value again.

  • Parsing code needs to be adjusted to support intrinsic values for heights.

Tests: fast/css-intrinsic-dimensions/height-css-tables-collapsed.html

fast/css-intrinsic-dimensions/height-css-tables.html
fast/css-intrinsic-dimensions/height-flexbox.html
fast/css-intrinsic-dimensions/height-positioned-replaced.html
fast/css-intrinsic-dimensions/height-positioned.html
fast/css-intrinsic-dimensions/height-replaced.html
fast/css-intrinsic-dimensions/height-tables-collapsed.html
fast/css-intrinsic-dimensions/height-tables.html
fast/css-intrinsic-dimensions/height.html

  • css/CSSParser.cpp:

(WebCore::CSSParser::isValidSize): Refactored from validateWidth and
validateHeight.
(WebCore::CSSParser::parseValue):
(WebCore::CSSParser::validateWidth): Deleted.
(WebCore::CSSParser::validateHeight): Deleted.

  • css/CSSParser.h:
  • platform/Length.h:

(WebCore::Length::isFillAvailable):
(WebCore::Length::isFitContent):

  • rendering/RenderBox.cpp:

(WebCore::RenderBox::constrainLogicalHeightByMinMax):
(WebCore::RenderBox::constrainContentBoxLogicalHeightByMinMax):
(WebCore::RenderBox::computeLogicalHeight):
(WebCore::RenderBox::computeLogicalHeightUsing):
(WebCore::RenderBox::computeContentLogicalHeight):
(WebCore::RenderBox::computeIntrinsicLogicalContentHeightUsing):
(WebCore::RenderBox::computeContentAndScrollbarLogicalHeightUsing):
(WebCore::RenderBox::computePercentageLogicalHeight):
(WebCore::RenderBox::computeReplacedLogicalHeightUsing):
(WebCore::RenderBox::availableLogicalHeight):
(WebCore::RenderBox::availableLogicalHeightUsing):
(WebCore::RenderBox::computePositionedLogicalHeight):
(WebCore::RenderBox::computePositionedLogicalHeightUsing):

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

(WebCore::RenderFlexibleBox::computeMainAxisExtentForChild):
(WebCore::RenderFlexibleBox::applyStretchAlignmentToChild):

  • rendering/RenderFlowThread.cpp:

(WebCore::RenderFlowThread::addForcedRegionBreak):

  • rendering/RenderGrid.cpp:

(WebCore::RenderGrid::computeUsedBreadthOfSpecifiedLength):
(WebCore::RenderGrid::applyStretchAlignmentToChildIfNeeded):

  • rendering/RenderMultiColumnSet.cpp:

(WebCore::RenderMultiColumnSet::calculateMaxColumnHeight):

  • rendering/RenderReplaced.cpp:

(WebCore::RenderReplaced::hasReplacedLogicalHeight):

  • rendering/RenderTable.cpp:

(WebCore::RenderTable::convertStyleLogicalHeightToComputedHeight):
(WebCore::RenderTable::layout):

LayoutTests:

Based on Blink's r148314 by <​cbiesinger@chromium.org>.

  • fast/css-intrinsic-dimensions/height-css-tables-collapsed-expected.html: Added.
  • fast/css-intrinsic-dimensions/height-css-tables-collapsed.html: Added.
  • fast/css-intrinsic-dimensions/height-css-tables-expected.html: Added.
  • fast/css-intrinsic-dimensions/height-css-tables.html: Added.
  • fast/css-intrinsic-dimensions/height-expected.html: Added.
  • fast/css-intrinsic-dimensions/height-flexbox-expected.html: Added.
  • fast/css-intrinsic-dimensions/height-flexbox.html: Added.
  • fast/css-intrinsic-dimensions/height-positioned-expected.html: Added.
  • fast/css-intrinsic-dimensions/height-positioned-replaced-expected.html: Added.
  • fast/css-intrinsic-dimensions/height-positioned-replaced.html: Added.
  • fast/css-intrinsic-dimensions/height-positioned.html: Added.
  • fast/css-intrinsic-dimensions/height-property-value-expected.txt:
  • fast/css-intrinsic-dimensions/height-property-value.html:
  • fast/css-intrinsic-dimensions/height-property-value.html.orig: Copied from LayoutTests/fast/css-intrinsic-dimensions/height-property-value.html.
  • fast/css-intrinsic-dimensions/height-replaced-expected.html: Added.
  • fast/css-intrinsic-dimensions/height-replaced.html: Added.
  • fast/css-intrinsic-dimensions/height-tables-collapsed-expected.html: Added.
  • fast/css-intrinsic-dimensions/height-tables-collapsed.html: Added.
  • fast/css-intrinsic-dimensions/height-tables-expected.html: Added.
  • fast/css-intrinsic-dimensions/height-tables.html: Added.
  • fast/css-intrinsic-dimensions/height.html: Added.
  • fast/css-intrinsic-dimensions/resources/height-keyword-classes.css: Added.

(.min-content):
(.max-content):
(.fill-available):
(.fit-content):
(.max-height-min-content):
(.max-height-max-content):
(.max-height-fill-available):
(.max-height-fit-content):
(.min-height-min-content):
(.min-height-max-content):
(.min-height-fill-available):
(.min-height-fit-content):

6:09 AM Changeset in webkit [185907] by commit-queue@webkit.org
  • 5 edits in trunk/Source

Unreviewed, rolling out r185906.
​https://bugs.webkit.org/show_bug.cgi?id=146276

MSVC doesn't yet provide a const-qualified std::array<T,
N>::size(), failing to compile the static_assert (Requested by
zdobersek on #webkit).

Reverted changeset:

"Improve the source code generated by make_names.pl"
​https://bugs.webkit.org/show_bug.cgi?id=146208
​http://trac.webkit.org/changeset/185906

5:19 AM Changeset in webkit [185906] by zandobersek@gmail.com
  • 5 edits in trunk/Source

Improve the source code generated by make_names.pl
​https://bugs.webkit.org/show_bug.cgi?id=146208

Reviewed by Darin Adler.

Source/WebCore:

Clean up and optimize the output that's generated by the make_names.pl script
when generating large sets of DOM names for attributes and tags.

The GenerateStrings() function in the StaticString.pm module is split into
GenerateStringData() and GenerateASCIILiteral() so that the two new functions
can be used independently, with the original function still being called when
generating font names.

Tags and attributes have the corresponding static QualifiedName globals defined
as before. After that, two static const std::array<> objects are defined for
both types -- the first is an ordered array of addresses of the QualifiedName
objects (corresponding to the C array that was defined in getHTMLTags(),
getSVGAttrs() etc.), and the second is an ordered array of StringImpl::StaticASCIILiteral
objects that replaces separately defined StringImpl::StaticASCIILiteral objects
and the additional tables that contained pairs of QualifiedName object addresses
and the corresponding StaticASCIILiteral object references in the init() function.
This is all generated by the printStaticData() function in make_names.pl.

The printQualifiedNameCreation() function generates a static_assert() that ensures
that the corresponding std::array<QualifiedName*> and std::array<StaticASCIILiteral>
objects have the same amount of items, and then sets up a loop that walks through
the two arrays and properly constructs the QualifiedName objects from the static
literal data.

On the GTK port, this shaves off ~54kB from the final stripped shared library
on a 64-bit build, and ~21kB on a 32-bit build.

  • bindings/scripts/StaticString.pm:

(GenerateStringData):
(GenerateASCIILiteral):
(GenerateStrings):

  • dom/make_names.pl:

(printCppHead):
(printNamesCppFile):
(printStaticData):
(printQualifiedNameCreation):
(printInit): Deleted.
(printDefinitions): Deleted.

Source/WTF:

  • wtf/text/StringImpl.h:

(WTF::StringImpl::assertHashIsCorrect): Make this method const-qualified.

2:25 AM Changeset in webkit [185905] by ryuan.choi@navercorp.com
  • 6 edits in trunk/Source/WebKit2

Remove some dead code from Download
​https://bugs.webkit.org/show_bug.cgi?id=146268

Reviewed by Darin Adler.

retrieveDestinationWithSuggestedFilename() was introduced for DownloadCFNetWin.cpp since r77585.
didDecideDestination and others are also introduced for WebKit2 Windows port since r77055.
Since WebKit2 Windows port is dropped, no one uses them now.

  • Shared/Downloads/Download.cpp:

(WebKit::Download::decideDestinationWithSuggestedFilename):
(WebKit::Download::retrieveDestinationWithSuggestedFilename): Deleted.

  • Shared/Downloads/Download.h:

(WebKit::Download::destination): Deleted.

  • Shared/Downloads/ios/DownloadIOS.mm:

(WebKit::Download::didDecideDestination): Deleted.

  • Shared/Downloads/mac/DownloadMac.mm:

(WebKit::Download::didDecideDestination): Deleted.

  • Shared/Downloads/soup/DownloadSoup.cpp:

(WebKit::Download::didDecideDestination): Deleted.

1:14 AM Changeset in webkit [185904] by Darin Adler
  • 4 edits in trunk/Source/JavaScriptCore

2015-06-24 Darin Adler <Darin Adler>

Fix Array.concat with RuntimeArray (regression from my last patch)

  • runtime/ArrayPrototype.cpp: (JSC::arrayProtoFuncConcat): Use getLength instead of JSArray::length.
  • runtime/JSArray.cpp: (JSC::JSArray::defineOwnProperty): Added comment about use of JSArray::length here that is incorrect (in a really non-obvious way). (JSC::JSArray::fillArgList): Ditto. (JSC::JSArray::copyToArguments): Ditto.
  • runtime/JSArray.h: Added a comment explaining that it is not always safe to use JSArray::length.
12:42 AM Changeset in webkit [185903] by youenn.fablet@crf.canon.fr
  • 7 edits in trunk/Source/WebCore

Refactor UserMediaRequest to share more codes between MediaDevices.getUserMedia and legacy webkitGetUserMedia
​https://bugs.webkit.org/show_bug.cgi?id=146237

Reviewed by Darin Adler.

Covered by existing tests, no change in behavior.

  • Modules/mediastream/MediaConstraintsImpl.cpp:

(WebCore::MediaConstraintsImpl::create): Simplified error handling by removing exception code parameter.

  • Modules/mediastream/MediaConstraintsImpl.h: Ditto.
  • Modules/mediastream/MediaDevices.cpp:

(WebCore::MediaDevices::getUserMedia): Removed code now in UserMediaRequest::start.

  • Modules/mediastream/NavigatorUserMedia.cpp:

(WebCore::NavigatorUserMedia::webkitGetUserMedia): Ditto.

  • Modules/mediastream/UserMediaRequest.cpp:

(WebCore::parseOptions): Simplified error handling by removing exception code parameter.
(WebCore::UserMediaRequest::start): Renamed create in start and added common code.

  • Modules/mediastream/UserMediaRequest.h:

Jun 23, 2015:

10:44 PM Changeset in webkit [185902] by mark.lam@apple.com
  • 2 edits in trunk/Source/JavaScriptCore

Gardening: Fixing 2 bad asserts from r185889.
​https://bugs.webkit.org/show_bug.cgi?id=140575

Not reviewed.

  • runtime/JSBoundSlotBaseFunction.cpp:

(JSC::JSBoundSlotBaseFunction::finishCreation):

9:34 PM Changeset in webkit [185901] by andersca@apple.com
  • 2 edits in trunk/Source/WebKit2

Fix build.

  • UIProcess/ios/ProcessAssertionIOS.mm:

(-[WKProcessAssertionBackgroundTaskManager init]):

7:34 PM Changeset in webkit [185900] by mitz@apple.com
  • 2 edits in trunk/Source/JavaScriptCore

Fixed iOS production builds.

7:33 PM Changeset in webkit [185899] by Darin Adler
  • 12 edits in trunk/Source

Make Array.join work directly on substrings without reifying them
​https://bugs.webkit.org/show_bug.cgi?id=146191

Reviewed by Andreas Kling.

Source/JavaScriptCore:

Besides the Array.join change, this has other optimizations based on
profiling the Peacekeeper array benchmark.

I measured a 14% speed improvement in the Peacekeeper array benchmark.

Still a lot of low hanging fruit in that test because so many of functions
on the array prototype are not optimizing for simple cases. For example,
the reverse function does individual get and put calls even when the array
is entirely made up of integers in contiguous storage.

  • runtime/ArrayPrototype.cpp:

(JSC::getProperty): Use tryGetIndexQuickly first before getPropertySlot.
(JSC::argumentClampedIndexFromStartOrEnd): Marked inline.
(JSC::shift): Use the getProperty helper in this file instead of using
getPropertySlot. Use putByIndexInline instead of calling putByIndex directly.
In both cases this can yield a faster code path.
(JSC::unshift): Ditto.
(JSC::arrayProtoFuncToString): Updated to use the new JSStringJoiner
interface. Changed local variable name to thisArray since it's not a
JSObject*. Changed loop index to i instead of k.
(JSC::arrayProtoFuncToLocaleString): Updated to use the new JSStringJoiner
interface. Renamed thisObj to thisObject. Added a missing exception check
after the toLocaleString function is called, but before toString is called
the result of that function.
(JSC::arrayProtoFuncJoin): Updated to use the new JSStringJointer interface.
Added a missing exception check after calling toString on the separator
but before calling get to get the first element in the array-like object
being joined. Changed loop index to i instead of k. Added missing exception
check after calling toString on each string from the array before calling
get for the next element.
(JSC::arrayProtoFuncConcat): Use JSArray::length instead of using the
getLength function.
(JSC::arrayProtoFuncReverse): Ditto. Also use putByIndexInline.
(JSC::arrayProtoFuncShift): Ditto.
(JSC::arrayProtoFuncSplice): Use getIndex instead of get, which includes some
additional optimizations.
(JSC::getOrHole): Deleted. Unused function.
(JSC::arrayProtoFuncUnShift): Use putByIndexInline.

  • runtime/ExceptionHelpers.cpp:

(JSC::errorDescriptionForValue): Removed the duplicate copy of the the logic
from JSValue::toString.

  • runtime/JSCJSValue.cpp:

(JSC::JSValue::toStringSlowCase): Improved the performance when converting a
small integer to a single character string.
(JSC::JSValue::toWTFStringSlowCase): Moved the contents of the
inlineJSValueNotStringtoString function here.

  • runtime/JSCJSValue.h: Removed no longer used toWTFStringInline and fixed

a comment with a typo.

  • runtime/JSObject.h:

(JSC::JSObject::putByIndexInline): Marked ALWAYS_INLINE because this was not
getting inlined at some call sites.
(JSC::JSObject::indexingData): Deleted. Unused function.
(JSC::JSObject::currentIndexingData): Deleted. Unused function.
(JSC::JSObject::getHolyIndexQuickly): Deleted. Unused function.
(JSC::JSObject::relevantLength): Deleted. Unused function.
(JSC::JSObject::currentRelevantLength): Deleted. Unused function.

  • runtime/JSString.h: Added the StringViewWithUnderlyingString struct and

the viewWithUnderlyingString function. Removed the inlineJSValueNotStringtoString
and toWTFStringInline functions.

  • runtime/JSStringJoiner.cpp:

(JSC::appendStringToData): Changed this to be a template instead of writing
it out, since StringView::getCharactersWithUpconvert does almsot exactly what
this function was trying to do.
(JSC::joinStrings): Rewrote this to use StringView.
(JSC::JSStringJoiner::joinedLength): Added. Factored out from the join function.
(JSC::JSStringJoiner::join): Rewrote to make it a bit simpler. Added an assertion
that we entirely filled capacity, since we are now reserving capacity and using
uncheckedAppend. Use String instead of RefPtr<StringImpl> because there was no
particular value to using the impl directly.

  • runtime/JSStringJoiner.h: Changed the interface to the class to use StringView.

Also changed this class so it now has the responsibility to convert each JSValue
into a string. This let us share more code between toString and join, and also
lets us use the new viewWithUnderlyingString function, which could be confusing at
all the call sites, but is easier to understand here.

Source/WTF:

  • wtf/Vector.h: Added an overload of uncheckedAppend like the one we added

a while back, a non-template function that forwards through to the function
template. This lets us call uncheckedAppend on an argument list and have it
properly convert it to the Vector's element type.

  • wtf/text/StringView.h:

(WTF::StringView::getCharactersWithUpconvert): Changed to not use memcpy;
saw some indication the hand-written loop was faster when profiling. Also
use m_length directly when we know we are dealing with an 8-bit string,
since the masking that the index function does is not needed in that case.
(WTF::StringView::UpconvertedCharacters::UpconvertedCharacters): Ditto.
(WTF::StringView::toString): Ditto.
(WTF::StringView::toFloat): Ditto.
(WTF::StringView::toInt): Ditto.
(WTF::StringView::toStringWithoutCopying): Ditto.
(WTF::StringView::find): Ditto.

7:13 PM Changeset in webkit [185898] by commit-queue@webkit.org
  • 11 edits in trunk/Tools

Refactoring benchmark runner script to follow pep8 code style.
​https://bugs.webkit.org/show_bug.cgi?id=146264

Patch by Dewei Zhu <Dewei Zhu> on 2015-06-23
Reviewed by Ryosuke Niwa.

Refactor benchmark runner script to follow pep8 code style.

  • Scripts/webkitpy/benchmark_runner/benchmark_builder/generic_benchmark_builder.py:

(GenericBenchmarkBuilder.prepare):
(GenericBenchmarkBuilder._run_create_script):
(GenericBenchmarkBuilder._copy_benchmark_to_temp_dir):
(GenericBenchmarkBuilder._fetch_remote_archive):
(GenericBenchmarkBuilder):
(GenericBenchmarkBuilder._checkout_with_subversion):
(GenericBenchmarkBuilder._apply_patch):
(GenericBenchmarkBuilder.clean):
(GenericBenchmarkBuilder._runCreateScript): Deleted.
(GenericBenchmarkBuilder._copyBenchmarkToTempDir): Deleted.
(GenericBenchmarkBuilder._fetchRemoteArchive): Deleted.
(GenericBenchmarkBuilder._checkoutWithSubverion): Deleted.
(GenericBenchmarkBuilder._applyPatch): Deleted.

  • Scripts/webkitpy/benchmark_runner/benchmark_runner.py:

(BenchmarkRunner.init):
(BenchmarkRunner._find_plan_file):
(BenchmarkRunner.execute):
(BenchmarkRunner._cleanup):
(BenchmarkRunner._dump):
(BenchmarkRunner._wrap):
(BenchmarkRunner._merge):
(BenchmarkRunner._show_results):
(BenchmarkRunner._findPlanFile): Deleted.
(BenchmarkRunner.cleanup): Deleted.
(BenchmarkRunner.dump): Deleted.
(BenchmarkRunner.wrap): Deleted.
(BenchmarkRunner.merge): Deleted.
(BenchmarkRunner.show_results): Deleted.

  • Scripts/webkitpy/benchmark_runner/browser_driver/browser_driver.py:

(BrowserDriver.prepare_env):
(BrowserDriver.launch_url):
(BrowserDriver.close_browsers):
(BrowserDriver.prepareEnv): Deleted.
(BrowserDriver.launchUrl): Deleted.
(BrowserDriver.closeBrowser): Deleted.

  • Scripts/webkitpy/benchmark_runner/browser_driver/osx_browser_driver.py:

(OSXBrowserDriver):
(OSXBrowserDriver.prepare_env):
(OSXBrowserDriver.close_browsers):
(OSXBrowserDriver._launch_process):
(OSXBrowserDriver._terminiate_processes):
(OSXBrowserDriver._launch_process_with_caffinate):
(OSXBrowserDriver._screen_size):
(OSXBrowserDriver.prepareEnv): Deleted.
(OSXBrowserDriver.closeBrowsers): Deleted.
(OSXBrowserDriver.launchProcess): Deleted.
(OSXBrowserDriver.terminateProcesses): Deleted.
(OSXBrowserDriver.launchProcessWithCaffinate): Deleted.
(OSXBrowserDriver.screenSize): Deleted.

  • Scripts/webkitpy/benchmark_runner/browser_driver/osx_chrome_driver.py:

(OSXChromeDriver):
(OSXChromeDriver.launch_url):
(OSXChromeCanaryDriver):
(OSXChromeCanaryDriver.launch_url):
(OSXChromeDriver.launchUrl): Deleted.
(OSXChromeCanaryDriver.launchUrl): Deleted.

  • Scripts/webkitpy/benchmark_runner/browser_driver/osx_firefox_driver.py:

(OSXFirefoxDriver):
(OSXFirefoxDriver.launch_url):
(OSXFirefoxNightlyDriver):
(OSXFirefoxNightlyDriver.launch_url):
(OSXFirefoxDriver.launchUrl): Deleted.
(OSXFirefoxNightlyDriver.launchUrl): Deleted.

  • Scripts/webkitpy/benchmark_runner/browser_driver/osx_safari_driver.py:

(OSXSafariDriver):
(OSXSafariDriver.prepare_env):
(OSXSafariDriver.launch_url):
(OSXSafariDriver.close_browsers):
(OSXSafariDriver._maximize_window):
(OSXSafariDriver.prepareEnv): Deleted.
(OSXSafariDriver.launchUrl): Deleted.
(OSXSafariDriver.closeBrowsers): Deleted.
(OSXSafariDriver.maximizeWindow): Deleted.

  • Scripts/webkitpy/benchmark_runner/http_server_driver/http_server_driver.py:

(HTTPServerDriver.fetch_result):
(HTTPServerDriver.kill_server):
(HTTPServerDriver.get_return_code):
(HTTPServerDriver.set_device_id):
(HTTPServerDriver.fetchResult): Deleted.
(HTTPServerDriver.killServer): Deleted.
(HTTPServerDriver.getReturnCode): Deleted.
(HTTPServerDriver.setDeviceID): Deleted.

  • Scripts/webkitpy/benchmark_runner/http_server_driver/simple_http_server_driver.py:

(SimpleHTTPServerDriver.serve):
(SimpleHTTPServerDriver.base_url):
(SimpleHTTPServerDriver.fetch_result):
(SimpleHTTPServerDriver.kill_server):
(SimpleHTTPServerDriver.get_return_code):
(SimpleHTTPServerDriver): Deleted.
(SimpleHTTPServerDriver.baseUrl): Deleted.
(SimpleHTTPServerDriver.fetchResult): Deleted.
(SimpleHTTPServerDriver.killServer): Deleted.
(SimpleHTTPServerDriver.getReturnCode): Deleted.

  • Scripts/webkitpy/benchmark_runner/utils.py:

(get_path_from_project_root):
(force_remove):
(getPathFromProjectRoot): Deleted.
(forceRemove): Deleted.

7:09 PM Changeset in webkit [185897] by akling@apple.com
  • 3 edits in trunk/Source/WebCore

Should reduce tile coverage for the first paint after a tab switch.
<​https://webkit.org/b/146252>
<rdar://problem/19821583>

Reviewed by Darin Adler.

Reduce the number of tiles we need to paint after switching tabs,
to shorten the time it takes before we can flush pixels to screen.

We accomplish this by piggybacking on the "speculative tiling enabled"
mode of FrameView, which was previously only used to throttle painting
and layer flushes during page load.

When a FrameView becomes visible, which is what happens when you
switch to its tab, we revert the speculative tiling optimization to
its initial state, and reset the "scrolled by user" flag.

In practice this means that after switching tabs, we only generate
enough tiles to fill the viewport. Then, after 500ms has passed
or the user scrolls the page, we go back to the usual speculative
tiling mode.

  • page/FrameView.cpp:

(WebCore::FrameView::show):

7:04 PM Changeset in webkit [185896] by Michael Catanzaro
  • 2 edits in trunk/Source/WebKit2

[GTK] Crash performing drag-and-drop
​https://bugs.webkit.org/show_bug.cgi?id=146267

Reviewed by Darin Adler.

Return early if gtk_get_current_event() returns null to avoid a crash. Note that this does
not fix drag-and-drop. Note also this prevents the web process from forcing the UI process
to crash by sending fake startDrag messages.

  • UIProcess/gtk/DragAndDropHandler.cpp:

(WebKit::DragAndDropHandler::startDrag):

7:00 PM Changeset in webkit [185895] by mmaxfield@apple.com
  • 7 edits
    3 copies
    1 move
    2 adds in trunk/LayoutTests

[OS X] Rebaseline fast/text/font-weight{,-zh}.html for El Capitan
​https://bugs.webkit.org/show_bug.cgi?id=146266
<rdar://problem/21391628>
<rdar://problem/21391673>

Unreviewed.

  • fast/text/font-weights.html: Renamed from LayoutTests/platform/mac/fast/text/font-weights.html.
  • platform/mac/TestExpectations:
  • platform/efl/TestExpectations:
  • platform/gtk/TestExpectations:
  • platform/win/TestExpectations:
  • platform/mac/fast/text/font-weights-expected.png: Added.
  • platform/mac/fast/text/font-weights-expected.txt:
  • platform/mac/fast/text/font-weights-zh-expected.png: Added.
  • platform/mac/fast/text/font-weights-zh-expected.txt:
  • platform/mac-mavericks/fast/text/font-weights-expected.txt: Moved from platform/mac/fast/text/font-weights-expected.txt
  • platform/mac-yosemite/fast/text/font-weights-expected.txt: Ditto.
  • platform/mac-yosemite/fast/text/font-weights-zh-expected.txt: Moved from platform/mac/fast/text/font-weights-zh-expected.txt
5:50 PM Changeset in webkit [185894] by andersca@apple.com
  • 5 edits
    2 adds in trunk/Source/WebKit2

Add a class that tracks whether we're a background or foreground application
​https://bugs.webkit.org/show_bug.cgi?id=146259
rdar://problem/19833808

Reviewed by Darin Adler.

First step towards being able to send out these notifications for view services.

  • UIProcess/ApplicationStateTracker.h: Added.

(WebKit::ApplicationStateTracker::isInBackground):

  • UIProcess/ApplicationStateTracker.mm: Added.

(WebKit::ApplicationStateTracker::singleton):
(WebKit::ApplicationStateTracker::ApplicationStateTracker):
(WebKit::ApplicationStateTracker::addListener):
(WebKit::ApplicationStateTracker::applicationDidEnterBackground):
(WebKit::ApplicationStateTracker::applicationWillEnterForeground):
(WebKit::ApplicationStateTracker::invokeListeners):
(WebKit::ApplicationStateTracker::pruneListeners):

  • UIProcess/ios/ProcessAssertionIOS.mm:

(-[WKProcessAssertionBackgroundTaskManager init]):
(-[WKProcessAssertionBackgroundTaskManager _updateBackgroundTask]):
(-[WKProcessAssertionBackgroundTaskManager _applicationWillEnterForeground]):
(-[WKProcessAssertionBackgroundTaskManager _applicationDidEnterBackground]):
(-[WKProcessAssertionBackgroundTaskManager dealloc]): Deleted.
(-[WKProcessAssertionBackgroundTaskManager _applicationWillEnterForeground:]): Deleted.
(-[WKProcessAssertionBackgroundTaskManager _applicationDidEnterBackground:]): Deleted.

  • UIProcess/ios/WKContentView.h:
  • UIProcess/ios/WKContentView.mm:

(-[WKContentView _commonInitializationWithProcessPool:configuration:]):
(-[WKContentView isBackground]):
(-[WKContentView _applicationDidEnterBackground]):
(-[WKContentView _applicationWillEnterForeground]):
(-[WKContentView _applicationDidEnterBackground:]): Deleted.
(-[WKContentView _applicationWillEnterForeground:]): Deleted.

  • WebKit2.xcodeproj/project.pbxproj:
5:43 PM Changeset in webkit [185893] by andersca@apple.com
  • 22 edits in trunk/Source

Remove windowResizerRect code, nobody is using it anymore
​https://bugs.webkit.org/show_bug.cgi?id=146265

Reviewed by Beth Dakin.

Source/WebCore:

  • loader/EmptyClients.h:
  • page/Chrome.cpp:

(WebCore::Chrome::windowResizerRect): Deleted.

  • page/Chrome.h:
  • page/ChromeClient.h:
  • page/FrameView.cpp:

(WebCore::FrameView::windowResizerRect): Deleted.

  • page/FrameView.h:
  • platform/ScrollView.cpp:

(WebCore::ScrollView::ScrollView): Deleted.
(WebCore::ScrollView::windowResizerRectChanged): Deleted.
(WebCore::ScrollView::containsScrollbarsAvoidingResizer): Deleted.
(WebCore::ScrollView::adjustScrollbarsAvoidingResizerCount): Deleted.
(WebCore::ScrollView::setParent): Deleted.

  • platform/ScrollView.h:

(WebCore::ScrollView::windowResizerRect): Deleted.

  • platform/Scrollbar.cpp:

(WebCore::Scrollbar::Scrollbar): Deleted.
(WebCore::Scrollbar::setFrameRect): Deleted.
(WebCore::Scrollbar::setParent): Deleted.

  • platform/Scrollbar.h:

Source/WebKit/mac:

  • WebCoreSupport/WebChromeClient.h:
  • WebCoreSupport/WebChromeClient.mm:

(WebChromeClient::windowResizerRect): Deleted.

Source/WebKit/win:

  • WebCoreSupport/WebChromeClient.cpp:

(WebChromeClient::windowResizerRect): Deleted.

  • WebCoreSupport/WebChromeClient.h:

Source/WebKit2:

  • WebProcess/WebCoreSupport/WebChromeClient.cpp:

(WebKit::WebChromeClient::windowResizerRect): Deleted.

  • WebProcess/WebCoreSupport/WebChromeClient.h:
  • WebProcess/WebPage/WebPage.cpp:

(WebKit::WebPage::windowResizerRect): Deleted.

  • WebProcess/WebPage/WebPage.h:
5:29 PM Changeset in webkit [185892] by commit-queue@webkit.org
  • 2 edits in trunk/Source/WebInspectorUI

Web Inspector: Reduce QuickConsole DidResize events if it did not change
​https://bugs.webkit.org/show_bug.cgi?id=146258

Patch by Joseph Pecoraro <Joseph Pecoraro> on 2015-06-23
Reviewed by Timothy Hatcher.

  • UserInterface/Views/QuickConsole.js:

(WebInspector.QuickConsole.prototype.consoleLogVisibilityChanged):
Do not trigger the event unless there was a change in visibility.

5:27 PM Changeset in webkit [185891] by commit-queue@webkit.org
  • 2 edits in trunk/Source/WebInspectorUI

Web Inspector: TextEditor scroll position not correctly restored when switching tabs
​https://bugs.webkit.org/show_bug.cgi?id=146254

Patch by Joseph Pecoraro <Joseph Pecoraro> on 2015-06-23
Reviewed by Timothy Hatcher.

  • UserInterface/Views/TextEditor.js:

(WebInspector.TextEditor.prototype.updateLayout):
Workaround a larger issue with ContentView restoration so that
TextEditors restore their scroll position as expected.

5:21 PM Changeset in webkit [185890] by mitz@apple.com
  • 3 edits
    2 adds in trunk/Source/WebKit2

Restore binary compatibility with iOS 8 Safari after r185877
​https://bugs.webkit.org/show_bug.cgi?id=146263

Reviewed by Anders Carlsson.

  • Platform/spi/Cocoa: Added.
  • Platform/spi/Cocoa/NSInvocationSPI.h: Added.
  • UIProcess/API/Cocoa/WKSecurityOrigin.mm:

(-[WKSecurityOrigin methodSignatureForSelector:]): Override to check _WKSecurityOrigin for
possible category methods with the specified selector.
(-[WKSecurityOrigin forwardInvocation:]): Invoke using the implementation of the
_WKSecurityOrigin category method.

  • WebKit2.xcodeproj/project.pbxproj:
4:36 PM Changeset in webkit [185889] by mmirman@apple.com
  • 13 edits
    5 adds in trunk

Completes native binding descriptors with native getters and potentially setters.
​https://bugs.webkit.org/show_bug.cgi?id=140575
rdar://problem/19506502

Reviewed by Mark Lam.

Source/JavaScriptCore:

  • CMakeLists.txt: Added JSBoundSlotBaseFunction.cpp
  • JavaScriptCore.vcxproj/JavaScriptCore.vcxproj:
  • JavaScriptCore.vcxproj/JavaScriptCore.vcxproj.filters:
  • JavaScriptCore.xcodeproj/project.pbxproj:
  • inspector/InjectedScriptSource.js: Added case for descriptor having a native getter.
  • runtime/JSBoundSlotBaseFunction.cpp: Added.

(JSC::boundSlotBaseFunctionCall):
(JSC::JSBoundSlotBaseFunction::JSBoundSlotBaseFunction):
Necessary wrapper for custom getters and setters as objects.
(JSC::JSBoundSlotBaseFunction::create):
(JSC::JSBoundSlotBaseFunction::visitChildren):
(JSC::JSBoundSlotBaseFunction::finishCreation):

  • runtime/JSBoundSlotBaseFunction.h: Added.

(JSC::JSBoundSlotBaseFunction::createStructure):
(JSC::JSBoundSlotBaseFunction::boundSlotBase):
(JSC::JSBoundSlotBaseFunction::customGetterSetter):
(JSC::JSBoundSlotBaseFunction::isGetter):

  • runtime/JSGlobalObject.cpp:

(JSC::JSGlobalObject::init): Added a globally initialized structure for JSBoundSlotBaseFunction
(JSC::JSGlobalObject::visitChildren): visits that structure

  • runtime/JSGlobalObject.h:

(JSC::JSGlobalObject::boundSlotBaseFunctionStructure): added a getter for that structure

  • runtime/JSObject.cpp:

(JSC::JSObject::getOwnPropertyDescriptor): extends the case for CustomGetterSetter to
actually include GetterSetter as a JSBoundSlotBaseFunction

  • runtime/VM.cpp: Added initializer for customGetterSetterFunctionMap
  • runtime/VM.h: Added cache for JSBoundSlotBaseFunction

LayoutTests:

  • inspector-protocol/runtime/getProperties-expected.txt: updated.
  • js/dom/native-bindings-descriptors-expected.txt: Added.
  • js/dom/native-bindings-descriptors.html: Added.
  • js/dom/script-tests/native-bindings-descriptors.js: Added.
3:47 PM Changeset in webkit [185888] by dino@apple.com
  • 3 edits
    2 adds in trunk/Source/WebCore

Media controls are missing the white backdrop in UIWebViews
​https://bugs.webkit.org/show_bug.cgi?id=146251
<rdar://problem/20181345>

Reviewed by Simon Fraser.

Implement two new CALayer subclasses that explicitly set
something that resembles the system appearance for
media-controls-dark-bar-background and media-controls-light-bar-background.
This way, WebKit1 clients get a visible result.

Creating the actual system recipes is tracked by:
​https://bugs.webkit.org/show_bug.cgi?id=146250

  • WebCore.xcodeproj/project.pbxproj: Add new files.
  • platform/graphics/ca/cocoa/PlatformCALayerCocoa.mm:

(PlatformCALayerCocoa::PlatformCALayerCocoa): Use the new CALayer subclasses for
the appropriate layer types.

  • platform/graphics/ca/cocoa/WebSystemBackdropLayer.h: Added.
  • platform/graphics/ca/cocoa/WebSystemBackdropLayer.mm: Added.

(-[WebLightSystemBackdropLayer init]):
(-[WebLightSystemBackdropLayer setBackgroundColor:]): Only set to a light grey.
(-[WebDarkSystemBackdropLayer init]):
(-[WebDarkSystemBackdropLayer setBackgroundColor:]): Only set to a dark grey.

3:46 PM Changeset in webkit [185887] by andersca@apple.com
  • 4 edits in trunk

Add operator! and copy/move constructors to WeakObjCPtr
​https://bugs.webkit.org/show_bug.cgi?id=146255

Reviewed by Andreas Kling.

Source/WebKit2:

  • Shared/mac/WeakObjCPtr.h:

(WebKit::WeakObjCPtr::WeakObjCPtr):
(WebKit::WeakObjCPtr::operator!):

Tools:

  • TestWebKitAPI/Tests/WebKit2/cocoa/WeakObjCPtr.mm:

(TEST):

3:09 PM Changeset in webkit [185886] by commit-queue@webkit.org
  • 2 edits in trunk/Source/WebInspectorUI

Web Inspector: Some brief previews are incorrectly treated as lossless
​https://bugs.webkit.org/show_bug.cgi?id=146247

Patch by Joseph Pecoraro <Joseph Pecoraro> on 2015-06-23
Reviewed by Timothy Hatcher.

  • UserInterface/Views/ObjectPreviewView.js:

(WebInspector.ObjectPreviewView.prototype._appendEntryPreviews):
(WebInspector.ObjectPreviewView.prototype._appendPropertyPreviews):
When we have a brief preview we may need to override lossless / overflow
if the preview view itself shows less properties than were in the preview.

3:02 PM Changeset in webkit [185885] by mrajca@apple.com
  • 3 edits in trunk/Source/WebCore

Support releasing media sessions
​https://bugs.webkit.org/show_bug.cgi?id=146132

Reviewed by Darin Adler.

  • Modules/mediasession/MediaSession.cpp: Implemented as described in the Media Session spec.

(WebCore::MediaSession::releaseSession):
(WebCore::MediaSession::releaseInternal):

  • Modules/mediasession/MediaSession.h:
2:41 PM Changeset in webkit [185884] by Chris Fleizach
  • 2 edits in trunk/Source/WebCore

AX: iOS: VoiceOver and ARIA: has popup property not announced
​https://bugs.webkit.org/show_bug.cgi?id=146188

Reviewed by Darin Adler.

Expose the existing "has popup" property to the iOS Accessibility API.

  • accessibility/ios/WebAccessibilityObjectWrapperIOS.mm:

(-[WebAccessibilityObjectWrapper accessibilityHasPopup]):

2:09 PM Changeset in webkit [185883] by mmaxfield@apple.com
  • 3 edits
    2 copies
    2 adds in trunk/LayoutTests

Unreviewed test gardening after r185842
​https://bugs.webkit.org/show_bug.cgi?id=145681
<rdar://problem/21169844>

  • platform/mac-yosemite/fast/text/arabic-times-new-roman-expected.png: Copied from LayoutTests/platform/mac/fast/text/arabic-times-new-roman-expected.png.
  • platform/mac-yosemite/fast/text/arabic-times-new-roman-expected.txt: Copied from LayoutTests/platform/mac/fast/text/arabic-times-new-roman-expected.txt.
  • platform/mac/fast/text/arabic-times-new-roman-expected.png:
  • platform/mac/fast/text/arabic-times-new-roman-expected.txt:
  • platform/win/fast/text/arabic-times-new-roman-expected.png: Added.
  • platform/win/fast/text/arabic-times-new-roman-expected.txt: Added.
1:34 PM Changeset in webkit [185882] by Antti Koivisto
  • 2 edits in trunk/Source/WebKit2

CrashTracer: com.apple.WebKit.Networking at JavaScriptCore: WTF::String::isolatedCopy const &
​https://bugs.webkit.org/show_bug.cgi?id=146236
rdar://problem/21380700

Reviewed by Andreas Kling.

  • NetworkProcess/cache/NetworkCache.cpp:

(WebKit::NetworkCache::Cache::clear):

Don't crash if this is called with network cache disabled.

11:54 AM Changeset in webkit [185881] by dbates@webkit.org
  • 2 edits in trunk/LayoutTests

AX Tests hit-test-input-{auto-fill, search}-button.html and input-search-cancel-button.html fail on Windows
​https://bugs.webkit.org/show_bug.cgi?id=146243

Mark the tests as failing for now.

  • platform/win/TestExpectations:
11:27 AM Changeset in webkit [185880] by Simon Fraser
  • 7 edits in trunk

Expose some more rendering progress events, and have MiniBrowser log for each of them
​https://bugs.webkit.org/show_bug.cgi?id=146227

Reviewed by Darin Adler.

Source/WebKit2:

Add _WKRenderingProgressEventFirstLayoutAfterSuppressedIncrementalRendering and
_WKRenderingProgressEventFirstPaintAfterSuppressedIncrementalRendering to the private
header, and convert them to/from WebCore milestones.

  • Shared/API/Cocoa/_WKRenderingProgressEvents.h:
  • Shared/API/Cocoa/_WKRenderingProgressEventsInternal.h:

(renderingProgressEvents):

  • UIProcess/API/Cocoa/WKWebView.mm:

(layoutMilestones):

Tools:

Register for layout milestones/rendering progress events in the WK1 and WK2
browser window controllers, and, if logging is turned on, log when they are
reached.

  • MiniBrowser/mac/WK1BrowserWindowController.m:

(-[WK1BrowserWindowController awakeFromNib]):
(-[WK1BrowserWindowController webView:didLayout:]):

  • MiniBrowser/mac/WK2BrowserWindowController.m:

(-[WK2BrowserWindowController awakeFromNib]):
(-[WK2BrowserWindowController _webView:renderingProgressDidChange:]):

11:04 AM Changeset in webkit [185879] by Carlos Garcia Campos
  • 4 edits in trunk/Source/WebKit2

Unreviewed. Fix GTK+ build after r185876 and r185877.

  • UIProcess/API/gtk/WebKitUIClient.cpp:
  • UIProcess/API/gtk/WebKitUserContentManager.cpp:
  • UIProcess/API/gtk/WebKitWebView.cpp:
10:37 AM Changeset in webkit [185878] by andersca@apple.com
  • 7 edits in trunk/Source/WebKit2

Get rid of the window resizer size
​https://bugs.webkit.org/show_bug.cgi?id=146240

Reviewed by Darin Adler.

This code was only used by older versions of GTK+, so get rid of it.

  • UIProcess/API/gtk/WebKitWebViewBase.cpp:

(webkitWebViewBaseNotifyResizerSize): Deleted.
(toplevelWindowResizeGripVisibilityChanged): Deleted.
(webkitWebViewBaseSetToplevelOnScreenWindow): Deleted.
(resizeWebKitWebViewBaseFromAllocation): Deleted.

  • UIProcess/WebPageProxy.cpp:

(WebKit::WebPageProxy::setWindowResizerSize): Deleted.

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

(WebKit::WebPage::windowResizerRect):
(WebKit::WebPage::setWindowResizerSize): Deleted.

  • WebProcess/WebPage/WebPage.h:
  • WebProcess/WebPage/WebPage.messages.in:
10:28 AM Changeset in webkit [185877] by beidson@apple.com
  • 39 edits
    2 copies
    1 move
    2 adds in trunk

WKFrameInfo should have an accessor for the Frame's current security origin.
​https://bugs.webkit.org/show_bug.cgi?id=146162

Reviewed by Dan Bernstein.

Source/WebKit2:

This patch:

  • Promotes _WKSecurityOrigin to API.
  • Makes the WKSecurityOrigin Cocoa API object work with API::SecurityOrigin.
  • Adds a WKSecurityOrigin accessor to WKFrameInfo.
  • Changes IPC messaging so every handler that calls out with a WKFrameInfo object gets a relevant SecurityOrigin to wrap.
  • Shared/API/APISecurityOrigin.h:

(API::SecurityOrigin::create):
(API::SecurityOrigin::SecurityOrigin):

  • Shared/API/Cocoa/WebKit.h:
  • Shared/API/c/WKSharedAPICast.h:

(WebKit::toCopiedAPI):

  • Shared/Cocoa/APIObject.mm:

(API::Object::newObject):

  • Shared/SecurityOriginData.cpp:

(WebKit::SecurityOriginData::fromFrame):

  • Shared/SecurityOriginData.h:
  • UIProcess/API/APIFrameInfo.cpp:

(API::FrameInfo::FrameInfo):

  • UIProcess/API/APIFrameInfo.h:
  • UIProcess/API/APINavigationClient.h:

(API::NavigationClient::didFailProvisionalLoadInSubframeWithError):

  • UIProcess/API/APIUIClient.h:

(API::UIClient::createNewPage):
(API::UIClient::runJavaScriptAlert):
(API::UIClient::runJavaScriptConfirm):
(API::UIClient::runJavaScriptPrompt):

  • UIProcess/API/C/WKApplicationCacheManager.cpp:

(WKApplicationCacheManagerGetApplicationCacheOrigins):

  • UIProcess/API/C/WKKeyValueStorageManager.cpp:

(WKKeyValueStorageManagerGetKeyValueStorageOrigins):

  • UIProcess/API/C/WKOriginDataManager.cpp:

(WKOriginDataManagerGetOrigins):

  • UIProcess/API/C/WKPage.cpp:

(WKPageSetPageUIClient):
(WKPageSetPageNavigationClient):

  • UIProcess/API/C/WKResourceCacheManager.cpp:

(WKResourceCacheManagerGetCacheOrigins):

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

(-[WKFrameInfo securityOrigin]):

  • UIProcess/API/Cocoa/WKSecurityOrigin.h: Copied from Source/WebKit2/UIProcess/API/Cocoa/_WKSecurityOrigin.h.
  • UIProcess/API/Cocoa/WKSecurityOrigin.mm: Copied from Source/WebKit2/UIProcess/API/Cocoa/_WKSecurityOrigin.mm.

(-[WKSecurityOrigin dealloc]):
(-[WKSecurityOrigin description]):
(-[WKSecurityOrigin protocol]):
(-[WKSecurityOrigin host]):
(-[WKSecurityOrigin port]):
(-[WKSecurityOrigin _apiObject]):

  • UIProcess/API/Cocoa/WKSecurityOriginInternal.h: Renamed from Source/WebKit2/UIProcess/API/Cocoa/_WKSecurityOriginInternal.h.

(API::wrapper):

  • UIProcess/API/Cocoa/WKUIDelegatePrivate.h:
  • UIProcess/API/Cocoa/WKUserContentController.mm:
  • UIProcess/API/Cocoa/_WKSecurityOrigin.h:
  • UIProcess/API/Cocoa/_WKSecurityOrigin.mm:

(-[_WKSecurityOrigin _initWithSecurityOrigin:]): Deleted.
(-[_WKSecurityOrigin protocol]): Deleted.
(-[_WKSecurityOrigin host]): Deleted.
(-[_WKSecurityOrigin port]): Deleted.

  • UIProcess/Cocoa/NavigationState.h:
  • UIProcess/Cocoa/NavigationState.mm:

(WebKit::NavigationState::NavigationClient::didFailProvisionalLoadInSubframeWithError):

  • UIProcess/Cocoa/UIDelegate.h:
  • UIProcess/Cocoa/UIDelegate.mm:

(WebKit::UIDelegate::UIClient::createNewPage):
(WebKit::UIDelegate::UIClient::runJavaScriptAlert):
(WebKit::UIDelegate::UIClient::runJavaScriptConfirm):
(WebKit::UIDelegate::UIClient::runJavaScriptPrompt):
(WebKit::UIDelegate::UIClient::exceededDatabaseQuota):
(WebKit::UIDelegate::UIClient::reachedApplicationCacheOriginQuota):

  • UIProcess/UserContent/WebScriptMessageHandler.h:
  • UIProcess/UserContent/WebUserContentControllerProxy.cpp:

(WebKit::WebUserContentControllerProxy::didPostMessage):

  • UIProcess/UserContent/WebUserContentControllerProxy.h:
  • UIProcess/UserContent/WebUserContentControllerProxy.messages.in:
  • UIProcess/WebPageProxy.cpp:

(WebKit::WebPageProxy::didFailProvisionalLoadForFrame):
(WebKit::WebPageProxy::decidePolicyForNavigationAction):
(WebKit::WebPageProxy::decidePolicyForNewWindowAction):
(WebKit::WebPageProxy::decidePolicyForResponse):
(WebKit::WebPageProxy::decidePolicyForResponseSync):
(WebKit::WebPageProxy::createNewPage):
(WebKit::WebPageProxy::runJavaScriptAlert):
(WebKit::WebPageProxy::runJavaScriptConfirm):
(WebKit::WebPageProxy::runJavaScriptPrompt):

  • UIProcess/WebPageProxy.h:
  • UIProcess/WebPageProxy.messages.in:
  • WebKit2.xcodeproj/project.pbxproj:
  • WebProcess/UserContent/WebUserContentController.cpp:

(WebKit::WebUserMessageHandlerDescriptorProxy::didPostMessage):

  • WebProcess/WebCoreSupport/WebChromeClient.cpp:

(WebKit::WebChromeClient::createWindow):
(WebKit::WebChromeClient::runJavaScriptAlert):
(WebKit::WebChromeClient::runJavaScriptConfirm):
(WebKit::WebChromeClient::runJavaScriptPrompt):

  • WebProcess/WebCoreSupport/WebFrameLoaderClient.cpp:

(WebKit::WebFrameLoaderClient::dispatchDidFailProvisionalLoad):
(WebKit::WebFrameLoaderClient::dispatchDecidePolicyForResponse):
(WebKit::WebFrameLoaderClient::dispatchDecidePolicyForNewWindowAction):
(WebKit::WebFrameLoaderClient::dispatchDecidePolicyForNavigationAction):

Tools:

  • TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj:
  • TestWebKitAPI/Tests/WebKit2/modal-alerts-in-new-about-blank-window.html: Added.
  • TestWebKitAPI/Tests/WebKit2Cocoa/ModalAlerts.mm: Added.

(sawDialog):
(-[ModalAlertsUIDelegate webView:createWebViewWithConfiguration:forNavigationAction:windowFeatures:]):
(-[ModalAlertsUIDelegate webView:runJavaScriptAlertPanelWithMessage:initiatedByFrame:completionHandler:]):
(-[ModalAlertsUIDelegate webView:runJavaScriptConfirmPanelWithMessage:initiatedByFrame:completionHandler:]):
(-[ModalAlertsUIDelegate webView:runJavaScriptTextInputPanelWithPrompt:defaultText:initiatedByFrame:completionHandler:]):
(TEST):

9:45 AM Changeset in webkit [185876] by mitz@apple.com
  • 15 edits in trunk/Source/WebKit2

<rdar://problem/21342465> Make -[WKWebView _certificateChain] public.
​https://bugs.webkit.org/show_bug.cgi?id=145886

Reviewed by Sam Weinig.

  • Shared/WebCertificateInfo.h:

(WebKit::WebCertificateInfo::create): Changed to return a Ref rather than a PassRefPtr.

  • UIProcess/API/Cocoa/WKBrowsingContextController.mm: Added no-op overrides of new

PageLoadState::Observer functions to PageLoadStateObserver.

  • UIProcess/API/Cocoa/WKProcessGroup.mm:

(-[WKProcessGroup _setAllowsSpecificHTTPSCertificate:forHost:]): Updated for change to
WebCertificateInfo::create.

  • UIProcess/API/Cocoa/WKProcessPool.mm:

(-[WKProcessPool _setAllowsSpecificHTTPSCertificate:forHost:]): Ditto.

  • UIProcess/API/Cocoa/WKWebView.h: Declared new certificateChain property on WKWebView.
  • UIProcess/API/Cocoa/WKWebView.mm:

(-[WKWebView certificateChain]): Added. Returns the certificate chain from the page load
state, or an empty array if there is no certificate chain.

  • UIProcess/API/Cocoa/WKWebViewPrivate.h: Deprecated _certificateChain.
  • UIProcess/Cocoa/NavigationState.h: Declared overrides of new PageLoadState::Observer

functions.

  • UIProcess/Cocoa/NavigationState.mm:

(WebKit::NavigationState::willChangeCertificateInfo): Override to call
-willChangeValueForKey:.
(WebKit::NavigationState::didChangeCertificateInfo): Override to call
-didChangeValueForKey:.

  • UIProcess/PageLoadState.cpp:

(WebKit::PageLoadState::commitChanges): Call the new observer functions
willChangeCertificateInfo and didChangeCertificateInfo.
(WebKit::PageLoadState::didCommitLoad): Added new certificateInfo parameter, which is stored
in the uncommitted state.

  • UIProcess/PageLoadState.h:

(WebKit::PageLoadState::certificateInfo): Added this getter.

  • UIProcess/WebFrameProxy.cpp:

(WebKit::WebFrameProxy::didCommitLoad): Changed the parameter type from
WebCore::CertificateInfo to WebCertificateInfo.

  • UIProcess/WebFrameProxy.h:
  • UIProcess/WebPageProxy.cpp:

(WebKit::WebPageProxy::didCommitLoadForFrame): Pass the certificate info to
PageLoadState::didCommitLoad.

9:26 AM Changeset in webkit [185875] by Matt Baker
  • 5 edits in trunk/Source/WebInspectorUI

Web Inspector: Layout & Rendering timeline should show paint and layout records in separate rows
​https://bugs.webkit.org/show_bug.cgi?id=146119

Reviewed by Timothy Hatcher.

This patch makes the original Layout & Rendering timeline visually consistent with the Rendering Frames
timeline, which uses green to distinguish Paint from Layout. In order to support having record bars with
different colors in the same overview graph, the timeline has been split into two rows.

  • UserInterface/Views/LayoutTimelineOverviewGraph.css:

(.timeline-overview-graph.layout > .graph-row):
(.timeline-overview-graph.layout > .graph-row > .timeline-record-bar):
(.timeline-overview-graph.layout > .graph-row > .timeline-record-bar > .segment):
New row styles.

  • UserInterface/Views/LayoutTimelineOverviewGraph.js:

(WebInspector.LayoutTimelineOverviewGraph.prototype.reset.createRecordRow):
(WebInspector.LayoutTimelineOverviewGraph.prototype.reset):
(WebInspector.LayoutTimelineOverviewGraph.prototype.updateLayout):
(WebInspector.LayoutTimelineOverviewGraph.prototype._updateRowLayout.createBar):
(WebInspector.LayoutTimelineOverviewGraph.prototype._updateRowLayout):
(WebInspector.LayoutTimelineOverviewGraph.prototype._layoutTimelineRecordAdded):
(WebInspector.LayoutTimelineOverviewGraph): Deleted.
Added bookkeeping objects for timeline row elements and their associated records.

  • UserInterface/Views/TimelineRecordBar.css:

(.timeline-record-bar.timeline-record-type-layout.layout-timeline-record-paint > .segment):
New style for layout record event types.

  • UserInterface/Views/TimelineRecordBar.js:

(WebInspector.TimelineRecordBar.prototype.set records):
Add style class for eventType, if present.

7:18 AM Changeset in webkit [185874] by jfernandez@igalia.com
  • 2 edits in trunk/Source/WebCore

[CSS Grid Layout] Performance optimization: avoid computing overflow alignment if not needed
​https://bugs.webkit.org/show_bug.cgi?id=146231

Reviewed by Sergio Villar Senin.

We don't need to apply any overflow handling if alignment value don't have a potential
risk of data loss, as it's the case of 'start' value.

This patch avoid computing the overflow in all the cases, since it adds an unneeded
overhead which affects performance.

New code improves performance around 3%-8%, depending on the grid tests.

No new tests, no new funcitonality.

  • rendering/RenderGrid.cpp:

(WebCore::RenderGrid::rowPositionForChild):
(WebCore::RenderGrid::columnPositionForChild):

7:10 AM Changeset in webkit [185873] by youenn.fablet@crf.canon.fr
  • 13 edits
    1 delete in trunk

MediaDevices.getUserMedia should migrate from callbacks to DOMPromise
​https://bugs.webkit.org/show_bug.cgi?id=146200

Reviewed by Darin Adler.

Source/WebCore:

Introduced DOMPromiseWithCallback to resolve/reject promises while allowing promise clients to use a typed callback approach.
Migrated MediaDevices.getUserMedia from callbacks to DOMPromiseWithCallback.
Removed MediaDevices.getUserMedia custom binding.

Covered by existing tests.

  • CMakeLists.txt: Removing custom binding.
  • Modules/mediastream/MediaDevices.cpp:

(WebCore::MediaDevices::getUserMedia): Moving from callback to promise.

  • Modules/mediastream/MediaDevices.h: Ditto.
  • Modules/mediastream/MediaDevices.idl: Removing custom binding.
  • Modules/mediastream/NavigatorUserMedia.cpp:

(WebCore::NavigatorUserMedia::webkitGetUserMedia): Converting from promise callback to API callbacks.

  • Modules/mediastream/UserMediaRequest.cpp:

(WebCore::UserMediaRequest::create): Moving from callback to promise.
(WebCore::UserMediaRequest::UserMediaRequest): Ditto.
(WebCore::UserMediaRequest::didCreateStream): Ditto.
(WebCore::UserMediaRequest::failedToCreateStreamWithConstraintsError): Ditto.
(WebCore::UserMediaRequest::failedToCreateStreamWithPermissionError): Ditto.

  • Modules/mediastream/UserMediaRequest.h: Ditto.
  • bindings/js/JSDOMPromise.h: Introducing DOMPromiseWithCallback and removing crypto specific header.

(WebCore::DOMPromiseWithCallback::DOMPromiseWithCallback):
(WebCore::Error>::resolve):
(WebCore::Error>::reject):

  • bindings/js/JSMediaDevicesCustom.cpp: Removed.
  • bindings/js/JSSubtleCryptoCustom.cpp: Updating headers.

LayoutTests:

  • fast/mediastream/MediaDevices-getUserMedia-expected.txt:
4:29 AM Changeset in webkit [185872] by youenn.fablet@crf.canon.fr
  • 8 edits in trunk

[Streams API] Implement ReadableStream js source "'cancel" callback
​https://bugs.webkit.org/show_bug.cgi?id=146204

Reviewed by Darin Adler.

Source/WebCore:

Calling "cancel" JS function when web app is cancelling a JS readable stream.
Handling of promise returned value in case of async cancel.

Covered by rebased tests.

  • bindings/js/ReadableJSStream.cpp:

(WebCore::ReadableJSStream::invoke): Refactoring to pass cancel reason or controller to the JS function.
(WebCore::ReadableJSStream::doStart): Ditto.
(WebCore::startReadableStreamAsync): Renaming readableStream as protectedStream.
(WebCore::createPullResultFulfilledFunction): Ditto.
(WebCore::ReadableJSStream::doPull): Refactoring to pass cancel reason or controller to the JS function.
(WebCore::createCancelResultFulfilledFunction): Cancel promise fullfil callback.
(WebCore::createCancelResultRejectedFunction): Cancel promise reject callback.
(WebCore::ReadableJSStream::doCancel): Calling cancel JS callback and handling promise returned value.

  • bindings/js/ReadableJSStream.h: Refactoring to pass cancel reason or controller to the JS function.

LayoutTests:

  • streams/reference-implementation/bad-underlying-sources-expected.txt:
  • streams/reference-implementation/readable-stream-cancel-expected.txt:
  • streams/reference-implementation/readable-stream-expected.txt:
  • streams/reference-implementation/readable-stream-reader-expected.txt:
3:45 AM Changeset in webkit [185871] by commit-queue@webkit.org
  • 4 edits in trunk/Tools

Remove build warnings in Tools/DumpRenderTree/TestNetscapePlugIn
​https://bugs.webkit.org/show_bug.cgi?id=146007

Patch by Tanay C <​tanay.c@samsung.com> on 2015-06-23
Reviewed by Darin Adler.

  • DumpRenderTree/TestNetscapePlugIn/CMakeLists.txt:
  • DumpRenderTree/TestNetscapePlugIn/PluginObject.cpp: Add attribute to functions.

(pluginLogWithWindowObject):
(pluginLogWithArguments):
(pluginLog):

  • DumpRenderTree/TestNetscapePlugIn/PluginTest.cpp: Add attribute to functions.

(PluginTest::executeScript):
(PluginTest::log):

2:34 AM Changeset in webkit [185870] by Gyuyoung Kim
  • 2 edits in trunk/LayoutTests

[EFL] Unreviewed, gardening.

Mark fast/forms/listbox-visible-size.html to failure and skip
fast/text/arabic-times-new-roman.html because it is for iOS.

  • platform/efl/TestExpectations:
2:03 AM Changeset in webkit [185869] by rniwa@webkit.org
  • 2 edits in trunk/Websites/perf.webkit.org

Build fix. A/B testing is broken when continuous builders report revisions out of order.

  • public/v2/app.js:

(App.AnalysisTaskController.Ember.Controller.extend.):

1:56 AM Changeset in webkit [185868] by peavo@outlook.com
  • 2 edits in trunk/Source/WebKit/win

[WinCairo] WebDownload::initWithRequest is not implemented.
​https://bugs.webkit.org/show_bug.cgi?id=146203

Reviewed by Alex Christensen.

Implement method to start download from a IWebURLRequest object.

  • WebDownloadCurl.cpp:

(WebDownload::initWithRequest): Implemented.

12:10 AM Changeset in webkit [185867] by Csaba Osztrogonác
  • 2 edits in trunk/Tools

[EFL] Add libhyphen-dev as dependency after r185862
​https://bugs.webkit.org/show_bug.cgi?id=146230

Reviewed by Gyuyoung Kim.

  • efl/install-dependencies:
12:00 AM Changeset in webkit [185866] by Gyuyoung Kim
  • 4 edits in trunk/Source/WebKit2

[EFL][CustomProtocol] Do not add duplicated custom scheme
​https://bugs.webkit.org/show_bug.cgi?id=146199

Reviewed by Carlos Garcia Campos.

WebSoupCustomProtocolRequestManager::registerSchemeForCustomProtocol generates
a crash when duplicated scheme is registered on debug mode, or just registers it on release mode.
However application can register duplicate scheme by mistake or on purpose. Thus it would be good
if we don't register it instead of registering it or generating a crash when trying to regiseter
duplicated scheme.

EFL port want to allow user to change registered callback, thus EWK2ContextTest::ewk_context_url_scheme_register()
is modified to test it.

Test: ewk_context_url_scheme_register() in test_ewk2_context.cpp.

  • UIProcess/API/efl/ewk_context.h: Added a comment to replace registered callback.
  • UIProcess/API/efl/tests/test_ewk2_context.cpp:

(EWK2ContextTest::schemeRequestCallback1):
(EWK2ContextTest::schemeRequestCallback2):
(TEST_F):
(EWK2ContextTest::schemeRequestCallback): Deleted.

  • UIProcess/Network/CustomProtocols/soup/WebSoupCustomProtocolRequestManager.cpp:

(WebKit::WebSoupCustomProtocolRequestManager::registerSchemeForCustomProtocol):

Jun 22, 2015:

11:47 PM Changeset in webkit [185865] by Carlos Garcia Campos
  • 1 copy in releases/WebKitGTK/webkit-2.9.3

WebKitGTK+ 2.9.3

11:45 PM Changeset in webkit [185864] by Carlos Garcia Campos
  • 4 edits in trunk

Unreviewed. Update OptionsGTK.cmake and NEWS for 2.9.3 release.

.:

  • Source/cmake/OptionsGTK.cmake: Bump version numbers.

Source/WebKit2:

  • gtk/NEWS: Add release notes for 2.9.3.
11:33 PM Changeset in webkit [185863] by Csaba Osztrogonác
  • 2 edits in trunk/Source/WTF

[WTF] Platform.h: use _ABI64 instead of _MIPS_SIM_ABI64 to determine MIPS N64
​https://bugs.webkit.org/show_bug.cgi?id=145113

Patch by YunQiang Su <​wzssyqa@gmail.com> on 2015-06-22
Reviewed by Csaba Osztrogonác.

  • wtf/Platform.h:
10:23 PM Changeset in webkit [185862] by ryuan.choi@navercorp.com
  • 21 edits
    1 move
    3 adds in trunk

[EFL] Hyphenation is not supported
​https://bugs.webkit.org/show_bug.cgi?id=89830

Reviewed by Gyuyoung Kim.

.:

  • Source/cmake/OptionsEfl.cmake: Added an option for LibHyphen.

Source/WebCore:

Share libHyphen backend of GTK port.

Rebased fast/text/hyphenate-*.html

  • PlatformEfl.cmake:
  • PlatformGTK.cmake:
  • platform/efl/FileSystemEfl.cpp:

(WebCore::listDirectory): Deleted because of lack of functionality.
eina_file_ls returns full directory path so fnmatch fails to check dict file.
This patch reuse Posix implementation instead of EFL port specific function.

  • platform/posix/FileSystemPOSIX.cpp: Ditto.
  • platform/text/gtk/HyphenationLibHyphen.cpp: Moved to platform/text/hyphen
  • platform/text/hyphen/HyphenationLibHyphen.cpp:

Renamed from Source/WebCore/platform/text/gtk/HyphenationLibHyphen.cpp.
(WebCore::scanTestDictionariesDirectoryIfNecessary):
Added PLATFORM guard and EFL implementation for the test directory

Tools:

  • efl/jhbuild.modules: Added webkitgtk-test-dicts for layout test.

LayoutTests:

Rebaseline expected results which is related to hyphenation.

  • platform/efl/TestExpectations: Unskip hyphenate-locale.html
  • platform/efl/fast/text/hyphenate-character-expected.png:
  • platform/efl/fast/text/hyphenate-character-expected.txt:
  • platform/efl/fast/text/hyphenate-first-word-expected.png:
  • platform/efl/fast/text/hyphenate-first-word-expected.txt:
  • platform/efl/fast/text/hyphenate-limit-before-after-expected.png:
  • platform/efl/fast/text/hyphenate-limit-before-after-expected.txt:
  • platform/efl/fast/text/hyphenate-limit-lines-expected.png:
  • platform/efl/fast/text/hyphenate-limit-lines-expected.txt:
  • platform/efl/fast/text/hyphenate-locale-expected.png: Added.
  • platform/efl/fast/text/hyphenate-locale-expected.txt: Added.
  • platform/efl/fast/text/hyphens-expected.png:
  • platform/efl/fast/text/hyphens-expected.txt:
9:10 PM Changeset in webkit [185861] by mitz@apple.com
  • 2 edits in trunk/Source/WebKit2

Part 2 of [Xcode] Fold the WebKit2SandboxProfiles-IOSOverride target into the Sandbox Profiles target
​https://bugs.webkit.org/show_bug.cgi?id=146197

Rubber-stamped by Anders Carlsson.

  • WebKit2.xcodeproj/project.pbxproj: Deleted the WebKit2SandboxProfiles-IOSOverride target.
8:54 PM Changeset in webkit [185860] by Chris Dumez
  • 4 edits in trunk/Source

[WK1] WebAllowDenyPolicyListener.denyOnlyThisRequest() should not start a new permission request
​https://bugs.webkit.org/show_bug.cgi?id=146228
<rdar://problem/15179262>

Reviewed by Daniel Bates.

Source/WebCore:

Add Geolocation::resetIsAllowed() API that merely resets
m_allowGeolocation to Unknown, so that we will request the permission
again the next time a position is requested.

  • Modules/geolocation/Geolocation.h:

(WebCore::Geolocation::resetIsAllowed):

Source/WebKit/mac:

Call the new Geolocation::resetIsAllowed() API after denying the
current request, instead of calling
Geolocation::resetAllGeolocationPermission(). In addition to resetting
m_allowGeolocation to Unknown, the latter would also start a new
permission request. However, for
WebAllowDenyPolicyListener.denyOnlyThisRequest(), we really only want
to deny the current request and then reset m_allowGeolocation to
Unknown so that permission is requested again later if the app requests
a position again.

The previous implementation meant that if the client application keeps
calling WebAllowDenyPolicyListener.denyOnlyThisRequest(), we would end
up in an infinite loop (requesting for permission), even though the
application did not make any new geolocation requests.

  • WebCoreSupport/WebGeolocationClient.mm:

(-[WebGeolocationPolicyListener denyOnlyThisRequest]):

8:47 PM Changeset in webkit [185859] by commit-queue@webkit.org
  • 18 edits
    3 deletes in trunk/Tools

Get rid of factory json files in benchmark_runner
​https://bugs.webkit.org/show_bug.cgi?id=146175

Patch by Dewei Zhu <Dewei Zhu> on 2015-06-22
Reviewed by Ryosuke Niwa.

Get rid of factory json files in benchmark_runner and refactoring the code.

  • Scripts/webkitpy/benchmark_runner/benchmark_builder/init.py:

(benchmark_builder_loader):

  • Scripts/webkitpy/benchmark_runner/benchmark_builder/benchmark_builder_factory.py:

(BenchmarkBuilderFactory):

  • Scripts/webkitpy/benchmark_runner/benchmark_builder/benchmark_builders.json: Removed.
  • Scripts/webkitpy/benchmark_runner/benchmark_builder/generic_benchmark_builder.py:

(GenericBenchmarkBuilder):

  • Scripts/webkitpy/benchmark_runner/benchmark_runner.py:

(BenchmarkRunner.init):
(BenchmarkRunner.execute):

  • Scripts/webkitpy/benchmark_runner/browser_driver/init.py:

(browser_driver_loader):

  • Scripts/webkitpy/benchmark_runner/browser_driver/browser_driver.py:

(BrowserDriver):

  • Scripts/webkitpy/benchmark_runner/browser_driver/browser_driver_factory.py:

(BrowserDriverFactory):
(BrowserDriverFactory.available_platforms):
(BrowserDriverFactory.available_browsers):
(BrowserDriverFactory.add_browser_driver):
(BrowserDriverFactory.create):

  • Scripts/webkitpy/benchmark_runner/browser_driver/browser_drivers.json: Removed.
  • Scripts/webkitpy/benchmark_runner/browser_driver/osx_browser_driver.py:

(OSXBrowserDriver):
(OSXBrowserDriver.prepareEnv):
(OSXBrowserDriver.terminateProcesses):
(OSXBrowserDriver.screenSize):

  • Scripts/webkitpy/benchmark_runner/browser_driver/osx_chrome_driver.py:

(OSXChromeDriver):
(OSXChromeCanaryDriver):

  • Scripts/webkitpy/benchmark_runner/browser_driver/osx_firefox_driver.py:

(OSXFirefoxDriver):
(OSXFirefoxNightlyDriver):

  • Scripts/webkitpy/benchmark_runner/browser_driver/osx_safari_driver.py:

(OSXSafariDriver):

  • Scripts/webkitpy/benchmark_runner/generic_factory.py:

(GenericFactory.create):
(GenericFactory.add):
(GenericFactory.iterateGetItem): Deleted.

  • Scripts/webkitpy/benchmark_runner/http_server_driver/init.py:

(http_server_driver_loader):

  • Scripts/webkitpy/benchmark_runner/http_server_driver/http_server_driver.py:

(HTTPServerDriver):

  • Scripts/webkitpy/benchmark_runner/http_server_driver/http_server_driver_factory.py:

(HTTPServerDriverFactory):

  • Scripts/webkitpy/benchmark_runner/http_server_driver/http_server_drivers.json: Removed.
  • Scripts/webkitpy/benchmark_runner/http_server_driver/simple_http_server_driver.py:

(SimpleHTTPServerDriver):

  • Scripts/webkitpy/benchmark_runner/utils.py:

(is_subclass):
(load_subclasses):
(ModuleNotFoundError): Deleted.
(loadModule): Deleted.
(loadJSONFromFile): Deleted.

6:59 PM Changeset in webkit [185858] by Simon Fraser
  • 2 edits in trunk/Source/WebCore

ASSERT(!m_zOrderListsDirty) when mousing over web view with incremental rendering suppressed
​https://bugs.webkit.org/show_bug.cgi?id=146225

Reviewed by Zalan Bujtas.

Update RenderLayer's z-order lists when hit testing. There's no guarantee that they've
been updated; this happens to work most of the time because painting updates them,
but if incremental rendering is suppressed, we may not have painted yet.

Easy to hit on webkit.org in MiniBrowser, but I wasn't able to make a reduced testcase.

  • rendering/RenderLayer.cpp:

(WebCore::RenderLayer::hitTest):
(WebCore::RenderLayer::updateLayerListsIfNeeded): Flip the order of the tests, since checking
dirty bits is cheaper than calling isStackingContext().

6:59 PM Changeset in webkit [185857] by Simon Fraser
  • 6 edits in trunk/Tools

Make it possible to enable incremental rendering suppression in MiniBrowser
​https://bugs.webkit.org/show_bug.cgi?id=146223

Reviewed by Tim Horton.

Add a menu item to enable incremental rendering suppression to MiniBrowser.

For WK1, this just toggles a pref, and takes effect immediately.

For WK2, this is a WKConfiguration property, so only affects new web views.

Also remove a toolbar item connection to toggleUseMinimumViewSize: that caused
logging on launch.

  • MiniBrowser/mac/AppDelegate.m:

(defaultConfiguration):

  • MiniBrowser/mac/BrowserWindow.xib:
  • MiniBrowser/mac/SettingsController.h:
  • MiniBrowser/mac/SettingsController.m:

(-[SettingsController _populateMenu]):
(-[SettingsController validateMenuItem:]):
(-[SettingsController toggleIncrementalRenderingSuppressed:]):
(-[SettingsController incrementalRenderingSuppressed]):

  • MiniBrowser/mac/WK1BrowserWindowController.m:

(-[WK1BrowserWindowController didChangeSettings]):

6:51 PM Changeset in webkit [185856] by dino@apple.com
  • 12 edits
    8 moves
    1 add in trunk/Source

Rename PlatformCA*Mac to PlatformCA*Cocoa
​https://bugs.webkit.org/show_bug.cgi?id=146224
<rdar://problem/21497182>

Reviewed by Simon Fraser.

Rename PlatformCALayerMac and related files in platform/graphics/ca/mac
to the more accurate Cocoa suffix.

Source/WebCore:

  • WebCore.xcodeproj/project.pbxproj:
  • page/mac/ServicesOverlayController.mm:
  • platform/graphics/ca/GraphicsLayerCA.cpp:
  • platform/graphics/ca/PlatformCAAnimation.h:
  • platform/graphics/ca/PlatformCALayer.h:
  • platform/graphics/ca/cocoa/LayerFlushSchedulerMac.cpp: Renamed from Source/WebCore/platform/graphics/ca/mac/LayerFlushSchedulerMac.cpp.
  • platform/graphics/ca/cocoa/PlatformCAAnimationCocoa.h: Renamed from Source/WebCore/platform/graphics/ca/mac/PlatformCAAnimationMac.h.
  • platform/graphics/ca/cocoa/PlatformCAAnimationCocoa.mm: Renamed from Source/WebCore/platform/graphics/ca/mac/PlatformCAAnimationMac.mm.
  • platform/graphics/ca/cocoa/PlatformCAFiltersCocoa.mm: Renamed from Source/WebCore/platform/graphics/ca/mac/PlatformCAFiltersMac.mm.
  • platform/graphics/ca/cocoa/PlatformCALayerCocoa.h: Renamed from Source/WebCore/platform/graphics/ca/mac/PlatformCALayerMac.h.
  • platform/graphics/ca/cocoa/PlatformCALayerCocoa.mm: Renamed from Source/WebCore/platform/graphics/ca/mac/PlatformCALayerMac.mm.
  • platform/graphics/ca/cocoa/WebTiledBackingLayer.h: Renamed from Source/WebCore/platform/graphics/ca/mac/WebTiledBackingLayer.h.
  • platform/graphics/ca/cocoa/WebTiledBackingLayer.mm: Renamed from Source/WebCore/platform/graphics/ca/mac/WebTiledBackingLayer.mm.
  • platform/graphics/ca/mac/LayerFlushSchedulerMac.cpp:
  • platform/graphics/ca/mac/PlatformCAAnimationMac.mm:
  • platform/graphics/ca/mac/PlatformCAFiltersMac.mm:
  • platform/graphics/ca/mac/PlatformCALayerMac.mm:
  • platform/graphics/ca/mac/WebTiledBackingLayer.mm:

Source/WebKit2:

  • WebProcess/WebPage/mac/PlatformCAAnimationRemote.mm:
  • WebProcess/WebPage/mac/PlatformCALayerRemote.cpp:

(WebKit::PlatformCALayerRemote::filtersCanBeComposited):

  • WebProcess/WebPage/mac/PlatformCALayerRemoteCustom.mm:

(WebKit::PlatformCALayerRemoteCustom::create):

  • WebProcess/WebPage/mac/PlatformCALayerRemoteTiledBacking.cpp:
  • WebProcess/WebPage/mac/TiledCoreAnimationDrawingArea.mm:

(WebKit::TiledCoreAnimationDrawingArea::commitTransientZoom):

6:24 PM Changeset in webkit [185855] by andersca@apple.com
  • 2 edits in trunk/Source/WebKit/mac

Delegates should be formal protocols
​https://bugs.webkit.org/show_bug.cgi?id=146222
rdar://problem/17380856

Reviewed by Dan Bernstein.

Flip the switch on OS X and make delegates formal protocols.

  • postprocess-headers.sh:
6:09 PM Changeset in webkit [185854] by achristensen@apple.com
  • 2 edits in trunk/Source/WebCore

[cssjit] Disable compiling scrollbar pseudoclass selectors
​https://bugs.webkit.org/show_bug.cgi?id=146220

Reviewed by Benjamin Poulain.

  • cssjit/SelectorCompiler.cpp:

(WebCore::SelectorCompiler::addScrollbarPseudoClassType):
Don't compile selectors with scrollbar pseudoclasses.

5:51 PM Changeset in webkit [185853] by Yusuke Suzuki
  • 5 edits
    1 add in trunk

[ES6] Allow trailing comma in ArrayBindingPattern and ObjectBindingPattern
​https://bugs.webkit.org/show_bug.cgi?id=146192

Reviewed by Darin Adler.

Source/JavaScriptCore:

According to the ES6 spec, trailing comma in ArrayBindingPattern and ObjectBindingPattern is allowed.
And empty ArrayBindingPattern and ObjectBindingPattern is also allowed.

This patch allows trailing comma and empty binding patterns.

  • bytecompiler/NodesCodegen.cpp:

(JSC::ArrayPatternNode::bindValue):

  • parser/Parser.cpp:

(JSC::Parser<LexerType>::parseDeconstructionPattern):

  • tests/stress/trailing-comma-in-patterns.js: Added.

(shouldBe):
(iterator):

LayoutTests:

  • js/object-literal-syntax-expected.txt:
5:47 PM Changeset in webkit [185852] by rniwa@webkit.org
  • 3 edits in trunk/Websites/perf.webkit.org

A/B testing results should be shown even if they were submitted to different platforms
​https://bugs.webkit.org/show_bug.cgi?id=146219

Reviewed by Andreas Kling.

Fetch A/B testing results regardless of the platform to which results are submitted
by providing the platform ID to which the results were submitted for each test group.

  • public/api/test-groups.php:

(main): Include the platform id in the test groups.

  • public/v2/analysis.js:

(App.TestGroup._fetchTestResults): Fetch results from the platform associated with the group.

4:32 PM Changeset in webkit [185851] by Simon Fraser
  • 3 edits
    2 adds in trunk

-webkit-clip-path clips incorrectly if the element bounds go beyond the top edge of the page
​https://bugs.webkit.org/show_bug.cgi?id=146218
rdar://problem/21127840

Reviewed by Zalan Bujtas.
Source/WebCore:

The clip path is computed using the RenderLayer's bounding box, so needs to be offset
by the offsetFromRenderer when set on the mask layer.

Test: compositing/masks/compositing-clip-path-origin.html

  • rendering/RenderLayerBacking.cpp:

(WebCore::RenderLayerBacking::updateMaskingLayerGeometry):

LayoutTests:

Test clip path on layers with non-zero bounds offsetFromRenderer because of box shadow,
or vertical writing mode.

  • compositing/masks/compositing-clip-path-origin-expected.html: Added.
  • compositing/masks/compositing-clip-path-origin.html: Added.
4:02 PM Changeset in webkit [185850] by ddkilzer@apple.com
  • 2 edits in trunk/LayoutTests

fast/text/justify-ideograph-{complex,simple,vertical}.html tests are flaky on El Capitan Debug builds

Tracked by <rdar://problem/21486062>.

  • platform/mac/TestExpectations: Mark tests as flaky:
  • fast/text/justify-ideograph-complex.html
  • fast/text/justify-ideograph-simple.html
  • fast/text/justify-ideograph-vertical.html
3:36 PM Changeset in webkit [185849] by andersca@apple.com
  • 2 edits in trunk/Source/WTF

Fix build.

  • wtf/threads/BinarySemaphore.cpp:
3:22 PM Changeset in webkit [185848] by Michael Catanzaro
  • 4 edits
    6 adds in trunk

Web sockets should be treated as active mixed content
​https://bugs.webkit.org/show_bug.cgi?id=140624

Reviewed by Sam Weinig.

Source/WebCore:

Tests: http/tests/security/mixedContent/websocket/insecure-websocket-in-iframe.html

http/tests/security/mixedContent/websocket/insecure-websocket-in-main-frame.html

  • Modules/websockets/WebSocket.cpp:

(WebCore::WebSocket::connect): Block ws:// WebSocket connections from ​https:// pages, and
emit the onerror event after doing so.

  • platform/SchemeRegistry.cpp:

(WebCore::secureSchemes): Add wss:// to the list of secure schemes.

LayoutTests:

  • http/tests/security/mixedContent/resources/frame-with-insecure-websocket.html: Added.
  • http/tests/security/mixedContent/websocket/insecure-websocket-in-iframe-expected.txt: Added.
  • http/tests/security/mixedContent/websocket/insecure-websocket-in-iframe.html: Added.
  • http/tests/security/mixedContent/websocket/insecure-websocket-in-main-frame-expected.txt: Added.
  • http/tests/security/mixedContent/websocket/insecure-websocket-in-main-frame.html: Added.
3:07 PM Changeset in webkit [185847] by andersca@apple.com
  • 4 edits
    1 delete in trunk/Source/WTF

Get rid of the Windows specific BinarySemaphore implementation
​https://bugs.webkit.org/show_bug.cgi?id=146216

Reviewed by Andreas Kling.

The fact that the Windows implementation uses a HEVENT internally was only useful
to the Windows port of WebKit2; we can get rid of it now.

  • WTF.vcxproj/WTF.vcxproj:
  • WTF.vcxproj/WTF.vcxproj.filters:
  • wtf/threads/BinarySemaphore.h:

(WTF::BinarySemaphore::event): Deleted.

  • wtf/threads/win/BinarySemaphoreWin.cpp: Removed.

(WTF::BinarySemaphore::BinarySemaphore): Deleted.
(WTF::BinarySemaphore::~BinarySemaphore): Deleted.
(WTF::BinarySemaphore::signal): Deleted.
(WTF::BinarySemaphore::wait): Deleted.

3:06 PM Changeset in webkit [185846] by dino@apple.com
  • 8 edits
    4 adds in trunk

Element with blur backdrop-filter shows edge duplication and dark edges
​https://bugs.webkit.org/show_bug.cgi?id=146215
<rdar://problem/20367695>

Reviewed by Tim Horton.

Source/WebCore:

The input images to backdrop filters should duplicate their edge pixels
outwards, rather than using transparent pixels. This is a flag we
set on the Gaussian blur, but means we have to remember if the
FilterOperations list came from a regular filter or a backdrop filter.

Test: css3/filters/backdrop/blur-input-bounds.html

  • css/CSSPropertyNames.in: New custom convertor for backdrop-filter.
  • css/StyleBuilderConverter.h:

(WebCore::StyleBuilderConverter::convertBackdropFilterOperations): New convertor
that sets the backdrop flag, but is otherwise the same as a normal filter
convertor.

  • page/animation/CSSPropertyAnimation.cpp:

(WebCore::blendFilterOperations): Inherit the backdrop flag if either of our
inputs has it.

  • platform/graphics/ca/mac/PlatformCAFiltersMac.mm: Set the inputNormalizeEdges

key on the CAFilter if necessary.

  • platform/graphics/filters/FilterOperations.cpp: Add a new flag indicating if

these operations are intended for backdrops.
(WebCore::FilterOperations::operator=):
(WebCore::FilterOperations::operator==):

  • platform/graphics/filters/FilterOperations.h:

(WebCore::FilterOperations::isUsedForBackdropFilters):
(WebCore::FilterOperations::setUsedForBackdropFilters):

LayoutTests:

Add a pixel test to show that the input images to backdrop filters should duplicate their
edge pixels. Unfortunately this is not reproducible with normal filters, so it
can't be a reference test.

  • css3/filters/backdrop/blur-input-bounds.html: Added.
  • platform/mac/css3/filters/backdrop/blur-input-bounds-expected.png: Added.
  • platform/mac/css3/filters/backdrop/blur-input-bounds-expected.txt: Added.
3:02 PM Changeset in webkit [185845] by commit-queue@webkit.org
  • 3 edits in trunk/Source/WebInspectorUI

Web Inspector: gaps between sections of the styles sidebar rules tab confusing, should say "Media: all"
​https://bugs.webkit.org/show_bug.cgi?id=142918

Patch by Devin Rousso <Devin Rousso> on 2015-06-22
Reviewed by Timothy Hatcher.

  • UserInterface/Views/RulesStyleDetailsPanel.css:

(.sidebar > .panel.details.css-style > .content.filter-in-progress .label:not(.filter-section-non-matching) ~ .label):
Now properly adds padding to filtered labels.
(.sidebar > .panel.details.css-style > .content.filter-in-progress .label ~ .label): Deleted.

  • UserInterface/Views/RulesStyleDetailsPanel.js:

(WebInspector.RulesStyleDetailsPanel.prototype.refresh): If a section of CSS rules has no media or inheritance, add
a label that says "Media: all" above the section.

2:33 PM Changeset in webkit [185844] by timothy_horton@apple.com
  • 2 edits in trunk/Source/WebCore

[TextIndicator] Text shifts one pixel to the left when I force click to bring up Lookup in Mail messages
​https://bugs.webkit.org/show_bug.cgi?id=146214
<rdar://problem/20782970>

Reviewed by Dean Jackson.

  • page/mac/TextIndicatorWindow.mm:

(-[WebTextIndicatorView initWithFrame:textIndicator:margin:offset:]):
(WebCore::TextIndicatorWindow::setTextIndicator):
(-[WebTextIndicatorView initWithFrame:textIndicator:margin:]): Deleted.
When the WebView is at a nonintegral position, we can end up needing a TextIndicator
with a nonintegral position. We need to round the window position, so we need to apply
the fractional part to the indicator layers inside, not to the window.

2:11 PM Changeset in webkit [185843] by andersca@apple.com
  • 3 edits in trunk/Source/WebKit2

Simplify Connection::SyncMessageState
​https://bugs.webkit.org/show_bug.cgi?id=146213

Reviewed by Andreas Kling.

Since we no longer support Connections dispatching to multiple threads, we can make SyncMessageState
into a singleton and get rid of the RunLoop -> SyncMessageState hash map.

  • Platform/IPC/Connection.cpp:

(IPC::Connection::SyncMessageState::singleton):
(IPC::Connection::SyncMessageState::SyncMessageState):
(IPC::Connection::SyncMessageState::processIncomingMessage):
(IPC::Connection::SyncMessageState::dispatchMessages):
(IPC::Connection::waitForSyncReply):
(IPC::Connection::processIncomingSyncReply):
(IPC::Connection::processIncomingMessage):
(IPC::Connection::connectionDidClose):
(IPC::Connection::SyncMessageState::syncMessageStateMap): Deleted.
(IPC::Connection::SyncMessageState::syncMessageStateMapMutex): Deleted.
(IPC::Connection::SyncMessageState::getOrCreate): Deleted.
(IPC::Connection::SyncMessageState::~SyncMessageState): Deleted.
(IPC::Connection::Connection): Deleted.

  • Platform/IPC/Connection.h:
1:53 PM Changeset in webkit [185842] by mmaxfield@apple.com
  • 5 edits
    6 adds in trunk

[iOS] Arabic text styled with Georgia is rendered as boxes
​https://bugs.webkit.org/show_bug.cgi?id=145681
<rdar://problem/21169844>

Reviewed by Darin Adler.

Source/WebCore:

Georgia doesn't support Arabic, so we ask CoreText what font does support Arabic. It returns
TimesNewRomanPSMT. However, WebKit explicitly disallows this font on iOS. Therefore, instead
of using TimesNewRomanPSMT, we will simply just use GeezaPro.

Test: fast/text/arabic-times-new-roman.html

  • platform/graphics/ios/FontCacheIOS.mm:

(WebCore::FontCache::systemFallbackForCharacters):

  • platform/graphics/Font.h: Let FontCacheIOS call fontFamilyShouldNotBeUsedForArabic()
  • platform/graphics/cocoa/FontCocoa.mm:

(WebCore::fontFamilyShouldNotBeUsedForArabic):

LayoutTests:

  • fast/text/arabic-times-new-roman.html: Added.
  • platform/ios-simulator/fast/text/arabic-times-new-roman-expected.txt: Added.
  • platform/ios-simulator/fast/text/arabic-times-new-roman-expected.png: Added.
  • platform/mac-mavericks/fast/text/arabic-times-new-roman-expected.txt: Added.
  • platform/mac/fast/text/arabic-times-new-roman-expected.txt: Added.
  • platform/mac/fast/text/arabic-times-new-roman-expected.png: Added.
1:49 PM Changeset in webkit [185841] by achristensen@apple.com
  • 2 edits in trunk/Source/WebCore

Unreviewed non-mac debug build fix after r185840.

  • loader/ResourceLoader.cpp:

(WebCore::ResourceLoader::willSendRequest):
Added enable flag around assertion.

1:10 PM Changeset in webkit [185840] by achristensen@apple.com
  • 20 edits in trunk/Source

[Content Extensions] Add SPI to reload without content blocking.
​https://bugs.webkit.org/show_bug.cgi?id=146128
rdar://problem/20351903

Reviewed by Sam Weinig.

Source/WebCore:

  • html/HTMLMediaElement.cpp:

(WebCore::HTMLMediaElement::loadResource):

  • loader/ResourceLoader.cpp:

(WebCore::ResourceLoader::willSendRequest):

  • loader/cache/CachedResourceLoader.cpp:

(WebCore::CachedResourceLoader::requestResource):

  • page/DOMWindow.cpp:

(WebCore::DOMWindow::open):

  • page/Page.h:

(WebCore::Page::userContentController):
(WebCore::Page::userContentExtensionsEnabled):
(WebCore::Page::setUserContentExtensionsEnabled):
(WebCore::Page::group):

  • page/UserContentController.cpp:

(WebCore::UserContentController::removeAllUserContentExtensions):
(WebCore::UserContentController::processContentExtensionRulesForLoad):
(WebCore::UserContentController::actionsForResourceLoad):

  • page/UserContentController.h:

Source/WebKit2:

  • Shared/WebPageCreationParameters.cpp:

(WebKit::WebPageCreationParameters::encode):
(WebKit::WebPageCreationParameters::decode):

  • Shared/WebPageCreationParameters.h:
  • UIProcess/API/C/WKPage.cpp:

(WKPageSetCustomUserAgent):
(WKPageSetUserContentExtensionsEnabled):
(WKPageSupportsTextEncoding):

  • UIProcess/API/C/WKPage.h:
  • UIProcess/API/Cocoa/WKWebView.mm:

(-[WKWebView _customUserAgent]):
(-[WKWebView _setUserContentExtensionsEnabled:]):
(-[WKWebView _userContentExtensionsEnabled]):
(-[WKWebView _setCustomUserAgent:]):

  • UIProcess/API/Cocoa/WKWebViewPrivate.h:
  • UIProcess/WebPageProxy.cpp:

(WebKit::WebPageProxy::creationParameters):
(WebKit::WebPageProxy::setShouldScaleViewToFitDocument):
(WebKit::WebPageProxy::setUserContentExtensionsEnabled):

  • UIProcess/WebPageProxy.h:

(WebKit::WebPageProxy::userContentExtensionsEnabled):

  • WebProcess/WebPage/WebPage.cpp:

(WebKit::WebPage::WebPage):
(WebKit::WebPage::reinitializeWebPage):
(WebKit::WebPage::setShouldScaleViewToFitDocument):
(WebKit::WebPage::setUserContentExtensionsEnabled):

  • WebProcess/WebPage/WebPage.h:
  • WebProcess/WebPage/WebPage.messages.in:

Pass a boolean from the API to WebCore.

12:52 PM Changeset in webkit [185839] by andersca@apple.com
  • 15 edits in trunk/Source/WebKit2

Remove m_clientRunLoop from IPC::Connection
​https://bugs.webkit.org/show_bug.cgi?id=146212

Reviewed by Sam Weinig.

We only ever create connections whose messages are dispatched to the main run loop, so we can
vastly simplify the code by only allowing messages to be dispatched there.

  • DatabaseProcess/DatabaseToWebProcessConnection.cpp:

(WebKit::DatabaseToWebProcessConnection::DatabaseToWebProcessConnection):

  • NetworkProcess/NetworkConnectionToWebProcess.cpp:

(WebKit::NetworkConnectionToWebProcess::NetworkConnectionToWebProcess):

  • Platform/IPC/Connection.cpp:

(IPC::Connection::createServerConnection):
(IPC::Connection::createClientConnection):
(IPC::Connection::Connection):
(IPC::Connection::addWorkQueueMessageReceiver):
(IPC::Connection::removeWorkQueueMessageReceiver):
(IPC::Connection::waitForMessage):
(IPC::Connection::sendSyncMessage):
(IPC::Connection::sendSyncMessageFromSecondaryThread):
(IPC::Connection::processIncomingMessage):
(IPC::Connection::connectionDidClose):
(IPC::Connection::dispatchDidReceiveInvalidMessage):
(IPC::Connection::enqueueIncomingMessage):
(IPC::Connection::wakeUpRunLoop):

  • Platform/IPC/Connection.h:
  • Platform/IPC/mac/ConnectionMac.mm:

(IPC::Connection::receiveSourceEventHandler):

  • PluginProcess/WebProcessConnection.cpp:

(WebKit::WebProcessConnection::WebProcessConnection):

  • Shared/ChildProcess.cpp:

(WebKit::ChildProcess::initialize):

  • Shared/ChildProcessProxy.cpp:

(WebKit::ChildProcessProxy::didFinishLaunching):

  • UIProcess/Plugins/PluginProcessProxy.cpp:

(WebKit::PluginProcessProxy::didFinishLaunching):

  • WebProcess/Databases/WebToDatabaseProcessConnection.cpp:

(WebKit::WebToDatabaseProcessConnection::WebToDatabaseProcessConnection):

  • WebProcess/Network/NetworkProcessConnection.cpp:

(WebKit::NetworkProcessConnection::NetworkProcessConnection):

  • WebProcess/Plugins/PluginProcessConnection.cpp:

(WebKit::PluginProcessConnection::PluginProcessConnection):

  • WebProcess/WebPage/WebInspector.cpp:

(WebKit::WebInspector::createInspectorPage):

  • WebProcess/WebPage/WebInspectorUI.cpp:

(WebKit::WebInspectorUI::establishConnection):

12:35 PM Changeset in webkit [185838] by Alan Bujtas
  • 3 edits
    2 adds in trunk

REGRESSION(r169105) Dangling renderer pointer in SelectionSubtreeRoot::SelectionSubtreeData.
​https://bugs.webkit.org/show_bug.cgi?id=146116
rdar://problem/20959369

Reviewed by Brent Fulgham.

This patch ensures that we don't adjust the selection unless the visual selection still matches this subtree root.

When multiple selection roots are present we need to ensure that a RenderObject
only shows up in one of them.
RenderView::splitSelectionBetweenSubtrees(), as the name implies, splits the
selection and sets the selection range (start/end) on each selection root.
However, SelectionSubtreeRoot::adjustForVisibleSelection() later recomputes the range
based on visible selection and that could end up collecting renderers as selection start/end
from another selection subtree.
RenderObject's holds the last selection state (RenderObject::setSelectionState).
If we set a renderer first as "on selection border" and later "inside" using multiple selection roots,
we can't clean up selections properly when this object gets destroyed.
One of the roots ends up with a dangling RenderObject pointer.

Source/WebCore:

Test: fast/regions/crash-when-renderer-is-in-multiple-selection-subtrees.html

  • rendering/SelectionSubtreeRoot.cpp:

(WebCore::SelectionSubtreeRoot::adjustForVisibleSelection):

LayoutTests:

  • fast/regions/crash-when-renderer-is-in-multiple-selection-subtrees-expected.txt: Added.
  • fast/regions/crash-when-renderer-is-in-multiple-selection-subtrees.html: Added.
12:30 PM Changeset in webkit [185837] by andersca@apple.com
  • 2 edits in trunk/Source/WebKit2

Crashes in hit testing under WebPage::acceptsFirstMouse() while handling sync message in plug-in teardown
​https://bugs.webkit.org/show_bug.cgi?id=146211
rdar://problem/17180615

Reviewed by Sam Weinig.

If we're inside a sendSync message when we're being called, just bail.

  • WebProcess/WebPage/mac/WebPageMac.mm:

(WebKit::WebPage::acceptsFirstMouse):

10:53 AM Changeset in webkit [185836] by Nikita Vasilyev
  • 3 edits in trunk/Source/WebInspectorUI

Web Inspector: Consider making read-only style rules have a darker background
​https://bugs.webkit.org/show_bug.cgi?id=145983

Reviewed by Timothy Hatcher.

  • UserInterface/Views/CSSStyleDeclarationSection.css:

(.style-declaration-section:not(.locked)):
(.style-declaration-section): Deleted.

  • UserInterface/Views/CSSStyleDeclarationTextEditor.css:

(.css-style-text-editor.read-only > .CodeMirror):

10:47 AM Changeset in webkit [185835] by Nikita Vasilyev
  • 2 edits in trunk/Source/WebInspectorUI

Web Inspector: Unable to select text of user input messages in the console
​https://bugs.webkit.org/show_bug.cgi?id=145888

Reviewed by Timothy Hatcher.

  • UserInterface/Views/ConsoleMessageView.css:

(.console-user-command > .console-message-text):

10:41 AM Changeset in webkit [185834] by commit-queue@webkit.org
  • 4 edits in trunk/Source/WebCore

Do not exit fullscreen when starting PiP since this is done automatically.
​https://bugs.webkit.org/show_bug.cgi?id=144871

Patch by Jeremy Jones <​jeremyj@apple.com> on 2015-06-22
Reviewed by Darin Adler.

Since we don't explicitly exit fullscreen, update state in shouldExitFullscreenWithReason()

  • platform/ios/WebVideoFullscreenInterfaceAVKit.h: Declare shouldExitFullscreenWithReason().
  • platform/ios/WebVideoFullscreenInterfaceAVKit.mm:

(-[WebAVPlayerController playerViewController:shouldExitFullScreenWithReason:]): Forward to WebVideoFullscreenInterfaceAVKit.
(WebVideoFullscreenInterfaceAVKit::shouldExitFullscreenWithReason): Added.
(WebVideoFullscreenInterfaceAVKit::willStartPictureInPicture): Remove enter fullscreen code.

  • platform/spi/cocoa/AVKitSPI.h: Add missing enums.
10:11 AM Changeset in webkit [185833] by fpizlo@apple.com
  • 6 edits
    1 add in trunk

Run CDjs as part of JSC stress testing
​https://bugs.webkit.org/show_bug.cgi?id=146174

Reviewed by Geoffrey Garen.

PerformanceTests:

  • JetStream/cdjs/cdjs-tests.yaml: Added. This tells JSC stress tests what tests to run. It uses new syntax ("tests" being a list) that I add in this change.
  • JetStream/cdjs/main.js: Mark this as a slow test.
  • JetStream/create.rb: Don't copy the JSC stress tests artifacts into the JetStream bundle.

Tools:

  • Scripts/run-javascriptcore-tests:

(runJSCStressTests): Make this aware of the cdjs-tests.yaml.

  • Scripts/run-jsc-stress-tests:
  • Teach this about tests that indicate error by returning an error code while also having lots of output even when they succeed.
  • Add the ability to have "tests" be a list of tests rather than just one test. This could also be a list of directories that have tests.
  • Fix a bug with bundle copying: whether the $collection should have the basename appended depends on whether we copy into bundleDir.dirname, not on whether absoluteCollection is a directory.
10:04 AM Changeset in webkit [185832] by mitz@apple.com
  • 2 edits
    1 add in trunk/Source/WebKit2

Part 1 of [Xcode] Fold the WebKit2SandboxProfiles-IOSOverride target into the Sandbox Profiles target
​https://bugs.webkit.org/show_bug.cgi?id=146197

Reviewed by Anders Carlsson.

  • Configurations/SandboxProfiles.xcconfig: Added. Defines INSTALL_PATH based on the value

of WK_INSTALL_OVERRIDE_SANDBOX_PROFILES.

  • WebKit2.xcodeproj/project.pbxproj: Use SandboxProfiles.xcconfig for the Sandbox Profiles

target, and change the destination path in its Copy Files build phase to INSTALL_PATH.

9:35 AM Changeset in webkit [185831] by bshafiei@apple.com
  • 3 edits in tags/Safari-601.1.35.2.3/Source/WebKit2

Merged r185814. rdar://problem/21444762

9:34 AM Changeset in webkit [185830] by bshafiei@apple.com
  • 5 edits in tags/Safari-601.1.35.2.3/Source

Versioning.

9:29 AM Changeset in webkit [185829] by bshafiei@apple.com
  • 1 copy in tags/Safari-601.1.35.2.3

New tag.

9:17 AM Changeset in webkit [185828] by dbates@webkit.org
  • 5 edits
    7 adds in trunk

AX: UI Automation cannot find AutoFill or search cancel buttons
​https://bugs.webkit.org/show_bug.cgi?id=145241
<rdar://problem/21051411>

Reviewed by Chris Fleizach.

Source/WebCore:

Add support for hit testing the search cancel button and AutoFill button so that
they can be accessed by UI Automation.

Currently the accessibility hit test machinery ignores nodes in a shadow tree.
So, it neither finds the <input type="search"> cancel button nor the AutoFill button
when it performs a hit test. Therefore these buttons cannot be accessed using
UI Automation.

Tests: accessibility/hit-test-input-auto-fill-button.html

accessibility/hit-test-input-search-cancel-button.html
accessibility/input-search-cancel-button.html

  • accessibility/AccessibilityRenderObject.cpp:

(WebCore::AccessibilityRenderObject::accessibilityTextFieldDecorationHitTest): Added; returns
the AccessibilityObject for the search cancel button or AutoFill text field decoration as applicable.
(WebCore::AccessibilityRenderObject::accessibilityHitTest): Check whether the hit node
is a text field decoration.

LayoutTests:

Add tests to ensure that there exists an accessibility element for the
search cancel button and that it can be hit using a cursor position. Also
add a test to ensue that the AutoFill button can be hit using a cursor position.

  • accessibility/hit-test-input-auto-fill-button-expected.txt: Added.
  • accessibility/hit-test-input-auto-fill-button.html: Copied from LayoutTests/accessibility/input-auto-fill-button.html.
  • accessibility/hit-test-input-search-cancel-button-expected.txt: Added.
  • accessibility/hit-test-input-search-cancel-button.html: Added.
  • accessibility/input-search-cancel-button-expected.txt: Added.
  • accessibility/input-search-cancel-button.html: Copied from LayoutTests/accessibility/input-auto-fill-button.html.
  • accessibility/resources/shouldBeAccessibleByCursor.js: Added.

(shouldBeAccessibleByCursor): Tests whether an AccessibilityUIElement can be hit
using its screen position.

  • platform/wk2/TestExpectations: Mark tests hit-test-input-{auto-fill, search-cancel}-button.html

as failing due to <​https://bugs.webkit.org/show_bug.cgi?id=71298>.

7:32 AM Changeset in webkit [185827] by Antti Koivisto
  • 2 edits in trunk/Source/WebKit2

Crash replacing TabDocument in MobileSafari at WebKit: -[WKWebView(WKPrivate) _beginAnimatedResizeWithUpdates:]
​https://bugs.webkit.org/show_bug.cgi?id=146201

Reviewed by Dan Bernstein.

  • UIProcess/API/Cocoa/WKWebView.mm:

(-[WKWebView _beginAnimatedResizeWithUpdates:]):

No repro but if for some reason [_contentView bounds] width is zero we'll compute +Inf targetScale
and then NaN contentOffset.x. Verified in lldb that this gives the exact crash signature seen.

Fix by checking that [_contentView bounds] is not empty like is done with other inputs.

6:59 AM Changeset in webkit [185826] by youenn.fablet@crf.canon.fr
  • 18 edits in trunk

[Streams API] Implement ReadableStream cancel (abstract part)
​https://bugs.webkit.org/show_bug.cgi?id=146111

Reviewed by Darin Adler.

Source/WebCore:

This patch implements ReadableStream and ReadableStreamReader cancel.
The reader delegates cancellation to its stream.

This patch also ensures that controller.close() will not throw in case cancellation is on-going.

A follow-up patch will implement the calling of 'cancel' JS callback for JS sources.

Covered by rebased tests.

  • Modules/streams/ReadableStream.cpp:

(WebCore::ReadableStream::cancel): Checks whether locked or not before cancelling.
(WebCore::ReadableStream::cancelNoCheck): Cancel without lock check.
(WebCore::ReadableStream::notifyCancelSucceeded): Async cancel callback.
(WebCore::ReadableStream::notifyCancelFailed): Ditto.

  • Modules/streams/ReadableStream.h:
  • Modules/streams/ReadableStream.idl: Cleaned up IDL.
  • Modules/streams/ReadableStreamReader.cpp:

(WebCore::ReadableStreamReader::cancel):

  • Modules/streams/ReadableStreamReader.h:
  • Modules/streams/ReadableStreamReader.idl: Cleaned up IDL
  • bindings/js/JSReadableStreamControllerCustom.cpp:

(WebCore::JSReadableStreamController::close):

  • bindings/js/JSReadableStreamCustom.cpp:

(WebCore::JSReadableStream::cancel):

  • bindings/js/JSReadableStreamReaderCustom.cpp:

(WebCore::JSReadableStreamReader::cancel):

  • bindings/js/ReadableJSStream.cpp:

(WebCore::ReadableJSStream::doCancel):

  • bindings/js/ReadableJSStream.h:

LayoutTests:

Rebasing expectations.

  • streams/reference-implementation/bad-underlying-sources-expected.txt:
  • streams/reference-implementation/readable-stream-cancel-expected.txt:
  • streams/reference-implementation/readable-stream-expected.txt:
  • streams/reference-implementation/readable-stream-reader-expected.txt:
  • streams/reference-implementation/readable-stream-templated-expected.txt:
5:42 AM Changeset in webkit [185825] by Carlos Garcia Campos
  • 2 edits in trunk/Source/WebKit2

REGRESSION(r182303): [GTK] Context menu API is broken since r182303
​https://bugs.webkit.org/show_bug.cgi?id=146202

Reviewed by Žan Doberšek.

The problem is that ContextMenuclient API changed in r182303, but
we didn't notice it either, and the default handler for
getContextMenuFromProposedMenu was executed. An override keyword
would have caught this.

  • UIProcess/API/gtk/WebKitContextMenuClient.cpp: Build a Vector of

WebContextMenuItemData as expected by our API, and add add
override keyword to ensure this doesn't happen again.

5:06 AM Changeset in webkit [185824] by zandobersek@gmail.com
  • 2 edits in trunk/Source/WebKit2

[WK2] ConnectionUnix should use FastMalloc to allocate on-heap resources
​https://bugs.webkit.org/show_bug.cgi?id=146143

Reviewed by Carlos Garcia Campos.

IPC handling in Unix-specific IPC::Connection implementation should use
FastMalloc to allocate on-heap resources, instead of allocating via the
system allocator.

The AttachmentInfo class is marked as allocatable through FastMalloc.
That way it can be allocated through FastMalloc while still handled
through std::unique_ptr<>.

The char[] arrays in readBytesFromSocket() and Connection::sendOutgoingMessage()
are now handled through a MallocPtr<> object.

In Connection::sendOutgoingMessage(), both the AttachmentInfo[] and char[]
arrays are now only allocated if there are actual attachments contained
in the message. The code that's conditioned with a non-empty attachments
Vector is now also grouped together, in a single branch.

  • Platform/IPC/unix/ConnectionUnix.cpp:

(IPC::readBytesFromSocket):
(IPC::Connection::sendOutgoingMessage):

4:55 AM Changeset in webkit [185823] by zandobersek@gmail.com
  • 3 edits in trunk

[CMake] Add support for building with various sanitizer tools
​https://bugs.webkit.org/show_bug.cgi?id=131941

Reviewed by Martin Robinson.

  • Source/PlatformGTK.cmake: Don't generate any documentation

when compiling with sanitizers enabled.

  • Source/cmake/OptionsCommon.cmake: Allow linking with

undefined symbols when compiling with sanitizers enabled.

3:03 AM Changeset in webkit [185822] by Gyuyoung Kim
  • 2 edits in trunk/Source/WebKit2

[EFL] test_ewk2_application_cache_manager has been failed since r185527
​https://bugs.webkit.org/show_bug.cgi?id=146016

Reviewed by Csaba Osztrogonác.

  • UIProcess/API/efl/tests/test_ewk2_application_cache_manager.cpp:

(TEST_F): Disable this test for now. This test will be enabled again.

1:57 AM Changeset in webkit [185821] by youenn.fablet@crf.canon.fr
  • 3 edits in trunk/LayoutTests

[Streams API] Correct releaseLock test in readable-stream-templated.html
​https://bugs.webkit.org/show_bug.cgi?id=146101

Reviewed by Darin Adler.

  • streams/reference-implementation/readable-stream-templated-expected.txt: Rebased changed test.
  • streams/reference-implementation/readable-stream-templated.html: Removed testharness wrapper around function.
1:36 AM Changeset in webkit [185820] by adam.bergkvist@ericsson.com
  • 8 edits in trunk

WebRTC: Navigator.webkitGetUserMedia() requires three arguments
​https://bugs.webkit.org/show_bug.cgi?id=146022

Reviewed by Eric Carlson.

Source/WebCore:

Updated custom binding to make the third error callback argument
mandatory. Updated and unskipped three tests (for the GTK+ port).

  • bindings/js/JSNavigatorCustom.cpp:

(WebCore::JSNavigator::webkitGetUserMedia):

LayoutTests:

Updated and unskipped three existing tests.

  • fast/mediastream/argument-types-expected.txt:
  • fast/mediastream/getusermedia-expected.txt:
  • fast/mediastream/getusermedia.html:
  • fast/mediastream/script-tests/argument-types.js:
  • platform/gtk/TestExpectations:
12:57 AM Changeset in webkit [185819] by Carlos Garcia Campos
  • 2 edits in trunk/Source/WebCore

Unreviewed. Fix GTK+ build after r185818.

Actually rollout r185320.

  • platform/network/soup/DNSSoup.cpp:

(WebCore::DNSResolveQueue::platformProxyIsEnabledInSystemPreferences):
(WebCore::DNSResolveQueue::platformResolve):
(WebCore::gotProxySettingsCallback): Deleted.
(WebCore::DNSResolveQueue::platformMaybeResolveHost): Deleted.

Jun 21, 2015:

11:32 PM Changeset in webkit [185818] by barraclough@apple.com
  • 9 edits in trunk/Source

Page load performance regression due to bugs.webkit.org/show_bug.cgi?id=145542
​https://bugs.webkit.org/show_bug.cgi?id=146198

Unreviewed rollout.

Source/WebCore:

  • platform/network/DNSResolveQueue.cpp:

(WebCore::DNSResolveQueue::DNSResolveQueue):
(WebCore::DNSResolveQueue::isUsingProxy):
(WebCore::DNSResolveQueue::add):
(WebCore::DNSResolveQueue::timerFired):

  • platform/network/DNSResolveQueue.h:
  • platform/network/cf/DNSCFNet.cpp:

(WebCore::DNSResolveQueue::platformProxyIsEnabledInSystemPreferences):
(WebCore::clientCallback):
(WebCore::DNSResolveQueue::platformResolve):
(WebCore::proxyIsEnabledInSystemPreferences): Deleted.
(WebCore::isUsingProxy): Deleted.
(WebCore::DNSResolveQueue::platformMaybeResolveHost): Deleted.

  • platform/network/soup/DNSSoup.cpp:

Source/WebKit2:

  • UIProcess/API/gtk/WebKitWebContext.cpp:

(webkit_web_context_prefetch_dns):

Source/WTF:

  • wtf/glib/GUniquePtr.h:
9:48 PM Changeset in webkit [185817] by commit-queue@webkit.org
  • 2 edits in trunk/Source/WebKit2

WKApplicationCacheManagerDeleteEntriesForOrigin() has wrong WebsiteDataTypes.
​https://bugs.webkit.org/show_bug.cgi?id=146180

Patch by Hyungwook Lee <​hyungwook.lee@navercorp.com> on 2015-06-21
Reviewed by Sam Weinig.

Modify to use correct WebsiteDataTypes in WKApplicationCacheManagerDeleteEntriesForOrigin()

  • UIProcess/API/C/WKApplicationCacheManager.cpp:

(WKApplicationCacheManagerDeleteEntriesForOrigin):

9:25 PM Changeset in webkit [185816] by ap@apple.com
  • 5 edits
    1 add in trunk

REGRESSION (r172975): navigator.language unable to tell region for Traditional Chinese users
​https://bugs.webkit.org/show_bug.cgi?id=146121
rdar://problem/21395180

Reviewed by Darin Adler.

Source/WebCore:

Revert to previous behavior, which is wrong in many ways, but not as wrong as the new one.

  • platform/mac/Language.mm:

(WebCore::httpStyleLanguageCode):
(WebCore::platformUserPreferredLanguages):

  • platform/spi/cf/CFBundleSPI.h:

Tools:

  • TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj:
  • TestWebKitAPI/Tests/mac/NavigatorLanguage.mm: Added.

(-[NavigatorLanguageDelegate webView:didFinishLoadForFrame:]):
(TestWebKitAPI::overrideAppleLanguages):
(TestWebKitAPI::languageForSystemLanguage):
(TestWebKitAPI::TEST):

8:25 PM Changeset in webkit [185815] by ddkilzer@apple.com
  • 2 edits in trunk/LayoutTests

fast/canvas/webgl/tex-image-and-sub-image-2d-with-video.html is slow on Mavericks WK1 Debug builds

The Flakiness Dashboard says that this test times out, but only
on Mavericks WK1 Debug builds (and the leaks bot):

<​http://webkit-test-results.appspot.com/dashboards/flakiness_dashboard.html#showAllRuns=true&tests=fast%2Fcanvas%2Fwebgl%2Ftex-image-and-sub-image-2d-with-video.html>

  • platform/mac-wk1/TestExpectations: Mark test as slow on

Mavericks Debug builds:

  • fast/canvas/webgl/tex-image-and-sub-image-2d-with-video.html
4:44 PM Changeset in webkit [185814] by mitz@apple.com
  • 3 edits in trunk/Source/WebKit2

<rdar://problem/21444762> REGRESSION (r184215): Plug-in services crash on launch on Yosemite
​https://bugs.webkit.org/show_bug.cgi?id=146195

Reviewed by Sam Weinig.

  • Configurations/PluginService.32.xcconfig: Changed the Yosemite definition of

INFOPLIST_FILE such that it would also apply to the macosx10.10.internal SDK.

  • Configurations/PluginService.64.xcconfig: Ditto.
2:14 PM Changeset in webkit [185813] by aestes@apple.com
  • 23 edits in trunk/Source/WebCore

Give Node::didNotifySubtreeInsertions() a better name
​https://bugs.webkit.org/show_bug.cgi?id=146170

Reviewed by Darin Adler.

didNotifySubtreeInsertions() is not a good name. It sounds like we are notifying the subtree insertions, which doesn't make sense.

This function is really about notifying the Node that the subtree it's a part of has finished being inserted into the DOM
(i.e. all nodes have received their call to insertedInto()). Change the name to finishedInsertingSubtree() to better reflect this.

  • dom/ContainerNode.cpp:

(WebCore::ContainerNode::notifyChildInserted):

  • dom/ContainerNodeAlgorithms.h:

(WebCore::ChildNodeInsertionNotifier::notifyNodeInsertedIntoDocument):
(WebCore::ChildNodeInsertionNotifier::notifyNodeInsertedIntoTree):

  • dom/Element.cpp:

(WebCore::Element::addShadowRoot):

  • dom/Node.h:

(WebCore::Node::finishedInsertingSubtree): Renamed from didNotifySubtreeInsertions.
(WebCore::Node::didNotifySubtreeInsertions): Renamed to finishedInsertingSubtree.

  • dom/ScriptElement.cpp:

(WebCore::ScriptElement::finishedInsertingSubtree): Renamed from didNotifySubtreeInsertions.
(WebCore::ScriptElement::didNotifySubtreeInsertions): Renamed to finishedInsertingSubtree.

  • dom/ScriptElement.h:
  • html/HTMLFrameElementBase.cpp:

(WebCore::HTMLFrameElementBase::insertedInto):
(WebCore::HTMLFrameElementBase::finishedInsertingSubtree): Renamed from didNotifySubtreeInsertions.
(WebCore::HTMLFrameElementBase::didNotifySubtreeInsertions): Renamed to finishedInsertingSubtree.

  • html/HTMLFrameElementBase.h:
  • html/HTMLScriptElement.cpp:

(WebCore::HTMLScriptElement::insertedInto):
(WebCore::HTMLScriptElement::finishedInsertingSubtree): Renamed from didNotifySubtreeInsertions.
(WebCore::HTMLScriptElement::didNotifySubtreeInsertions): Renamed to finishedInsertingSubtree.

  • html/HTMLScriptElement.h:
  • svg/SVGFEImageElement.cpp:

(WebCore::SVGFEImageElement::insertedInto):
(WebCore::SVGFEImageElement::finishedInsertingSubtree): Renamed from didNotifySubtreeInsertions.
(WebCore::SVGFEImageElement::didNotifySubtreeInsertions): Renamed to finishedInsertingSubtree.

  • svg/SVGFEImageElement.h:
  • svg/SVGMPathElement.cpp:

(WebCore::SVGMPathElement::insertedInto):
(WebCore::SVGMPathElement::finishedInsertingSubtree): Renamed from didNotifySubtreeInsertions.
(WebCore::SVGMPathElement::didNotifySubtreeInsertions): Renamed to finishedInsertingSubtree.

  • svg/SVGMPathElement.h:
  • svg/SVGScriptElement.cpp:

(WebCore::SVGScriptElement::insertedInto):
(WebCore::SVGScriptElement::finishedInsertingSubtree): Renamed from didNotifySubtreeInsertions.
(WebCore::SVGScriptElement::didNotifySubtreeInsertions): Renamed to finishedInsertingSubtree.

  • svg/SVGScriptElement.h:
  • svg/SVGTRefElement.cpp:

(WebCore::SVGTRefElement::insertedInto):
(WebCore::SVGTRefElement::finishedInsertingSubtree): Renamed from didNotifySubtreeInsertions.
(WebCore::SVGTRefElement::didNotifySubtreeInsertions): Renamed to finishedInsertingSubtree.

  • svg/SVGTRefElement.h:
  • svg/SVGTextPathElement.cpp:

(WebCore::SVGTextPathElement::insertedInto):
(WebCore::SVGTextPathElement::finishedInsertingSubtree): Renamed from didNotifySubtreeInsertions.
(WebCore::SVGTextPathElement::didNotifySubtreeInsertions): Renamed to finishedInsertingSubtree.

  • svg/SVGTextPathElement.h:
  • svg/animation/SVGSMILElement.cpp:

(WebCore::SVGSMILElement::insertedInto):
(WebCore::SVGSMILElement::finishedInsertingSubtree): Renamed from didNotifySubtreeInsertions.
(WebCore::SVGSMILElement::didNotifySubtreeInsertions): Renamed to finishedInsertingSubtree.

  • svg/animation/SVGSMILElement.h:
9:44 AM Changeset in webkit [185812] by ddkilzer@apple.com
  • 3 edits in trunk/LayoutTests

fast/fixed-layout/fixed-layout.html times out on Apple WK2 bots

The Flakiness Dashboard says this test times out on Apple
WebKit2 platforms by not calling testRunner.notifyDone():

<​http://webkit-test-results.appspot.com/dashboards/flakiness_dashboard.html#showAllRuns=true&tests=fast%2Ffixed-layout%2Ffixed-layout.html>

  • fast/css-grid-layout/flex-content-sized-columns-resize.html:

Use the same mechanism here that was just added in r185811.

  • fast/fixed-layout/fixed-layout.html: Switch to using a

zero-delay timer instead of document.body.offsetTop.

9:16 AM Changeset in webkit [185811] by ddkilzer@apple.com
  • 2 edits in trunk/LayoutTests

REGRESSION (r185809): fast/css-grid-layout/flex-content-sized-columns-resize.html hangs on WK1

Since window.resize() occurs in the same runloop as the load on
WK1, testRunner.notifyDone() would never get called.

  • fast/css-grid-layout/flex-content-sized-columns-resize.html:

Call document.body.offsetTop to ensure layout occurs at the
initial window size before calling window.resize().

9:01 AM Changeset in webkit [185810] by mitz@apple.com
  • 2 edits in trunk/Source/WebKit2

Build fix.

  • Shared/API/Cocoa/WKFoundation.h: Don’t use generics if the SDK doesn’t support them.
8:57 AM Changeset in webkit [185809] by ddkilzer@apple.com
  • 2 edits in trunk/LayoutTests

fast/css-grid-layout/flex-content-sized-columns-resize.html is flaky

The Flakiness Dashboard says this test is flaky on Apple and EFL
WebKit2 platforms:

<​http://webkit-test-results.appspot.com/dashboards/flakiness_dashboard.html#showAllRuns=true&tests=fast%2Fcss-grid-layout%2Fflex-content-sized-columns-resize.html>

It currently doesn't wait for window.resize() to complete, which
is likely the source of flakiness.

  • fast/css-grid-layout/flex-content-sized-columns-resize.html:

Switch to using notifyDone() inside a window.onresize function,
and make the test work in a web browser without TestRunner.

6:19 AM Changeset in webkit [185808] by Michael Catanzaro
  • 2 edits in trunk/Source/WebKit2

[SOUP] Use GTask in WebKitSoupRequestInputStream
​https://bugs.webkit.org/show_bug.cgi?id=146184

Reviewed by Carlos Garcia Campos.

Replace deprecated use of GSimpleAsyncResult with GTask.

  • WebProcess/soup/WebKitSoupRequestInputStream.cpp:

(AsyncReadData::AsyncReadData):
(webkitSoupRequestInputStreamReadAsyncResultComplete):
(webkitSoupRequestInputStreamPendingReadAsyncComplete):
(webkitSoupRequestInputStreamReadAsync):
(webkitSoupRequestInputStreamReadFinish):

6:15 AM Changeset in webkit [185807] by Michael Catanzaro
  • 4 edits in trunk/Source/WebKit2

Fix format strings in NetworkCache.cpp and NetworkCacheStatistics.cpp and WebIDBServerConnection.cpp
​https://bugs.webkit.org/show_bug.cgi?id=146172

Reviewed by Darin Adler.

Use the correct %"PRIu64" and %"PRIi64" for uint64_t and int64_t, instead of %llu and %lli.

  • NetworkProcess/cache/NetworkCache.cpp:

(WebKit::NetworkCache::Cache::retrieve):

  • NetworkProcess/cache/NetworkCacheStatistics.cpp:

(WebKit::NetworkCache::Statistics::initialize):
(WebKit::NetworkCache::Statistics::recordNotUsingCacheForRequest):
(WebKit::NetworkCache::Statistics::recordRetrievalFailure):
(WebKit::NetworkCache::Statistics::recordRetrievedCachedEntry):
(WebKit::NetworkCache::Statistics::recordRevalidationSuccess):

  • WebProcess/Databases/IndexedDB/WebIDBServerConnection.cpp:

(WebKit::WebIDBServerConnection::deleteDatabase):
(WebKit::WebIDBServerConnection::didDeleteDatabase):
(WebKit::WebIDBServerConnection::getOrEstablishIDBDatabaseMetadata):
(WebKit::WebIDBServerConnection::didGetOrEstablishIDBDatabaseMetadata):
(WebKit::WebIDBServerConnection::openTransaction):
(WebKit::WebIDBServerConnection::didOpenTransaction):
(WebKit::WebIDBServerConnection::beginTransaction):
(WebKit::WebIDBServerConnection::didBeginTransaction):
(WebKit::WebIDBServerConnection::commitTransaction):
(WebKit::WebIDBServerConnection::didCommitTransaction):
(WebKit::WebIDBServerConnection::resetTransaction):
(WebKit::WebIDBServerConnection::didResetTransaction):
(WebKit::WebIDBServerConnection::rollbackTransaction):
(WebKit::WebIDBServerConnection::didRollbackTransaction):
(WebKit::WebIDBServerConnection::createObjectStore):
(WebKit::WebIDBServerConnection::didCreateObjectStore):
(WebKit::WebIDBServerConnection::createIndex):
(WebKit::WebIDBServerConnection::didCreateIndex):
(WebKit::WebIDBServerConnection::deleteIndex):
(WebKit::WebIDBServerConnection::didDeleteIndex):
(WebKit::WebIDBServerConnection::get):
(WebKit::WebIDBServerConnection::put):
(WebKit::WebIDBServerConnection::didPutRecord):
(WebKit::WebIDBServerConnection::didGetRecord):
(WebKit::WebIDBServerConnection::didOpenCursor):
(WebKit::WebIDBServerConnection::didAdvanceCursor):
(WebKit::WebIDBServerConnection::didIterateCursor):
(WebKit::WebIDBServerConnection::count):
(WebKit::WebIDBServerConnection::didCount):
(WebKit::WebIDBServerConnection::deleteRange):
(WebKit::WebIDBServerConnection::didDeleteRange):
(WebKit::WebIDBServerConnection::clearObjectStore):
(WebKit::WebIDBServerConnection::didClearObjectStore):
(WebKit::WebIDBServerConnection::deleteObjectStore):
(WebKit::WebIDBServerConnection::didDeleteObjectStore):
(WebKit::WebIDBServerConnection::changeDatabaseVersion):
(WebKit::WebIDBServerConnection::didChangeDatabaseVersion):
(WebKit::WebIDBServerConnection::openCursor):
(WebKit::WebIDBServerConnection::cursorAdvance):
(WebKit::WebIDBServerConnection::cursorIterate):

6:15 AM Changeset in webkit [185806] by Michael Catanzaro
  • 13 edits in trunk/Tools

[CMake] Ignore warnings from system headers when compiling Tools
​https://bugs.webkit.org/show_bug.cgi?id=146185

Reviewed by Martin Robinson.

Use the SYSTEM argument to the include_directories() command to silence compiler warnings
from system headers that we don't control.

  • DumpRenderTree/TestNetscapePlugIn/CMakeLists.txt:
  • ImageDiff/CMakeLists.txt: Also, remove an EFL include directory from here...
  • ImageDiff/PlatformEfl.cmake: ...and put it here where it belongs.
  • ImageDiff/PlatformGTK.cmake:
  • MiniBrowser/efl/CMakeLists.txt:
  • MiniBrowser/gtk/CMakeLists.txt:
  • TestWebKitAPI/PlatformEfl.cmake:
  • TestWebKitAPI/PlatformGTK.cmake:
  • TestWebKitAPI/Tests/WebKit2Gtk/CMakeLists.txt:
  • WebKitTestRunner/CMakeLists.txt:
  • WebKitTestRunner/PlatformEfl.cmake:
  • WebKitTestRunner/PlatformGTK.cmake:
3:06 AM Changeset in webkit [185805] by commit-queue@webkit.org
  • 5 edits in trunk

libwebkit2gtk fails to link without opengl
​https://bugs.webkit.org/show_bug.cgi?id=138332

Patch by Philip Chimento <​philip.chimento@gmail.com> on 2015-06-21
Reviewed by Carlos Garcia Campos.

.:

  • Source/cmake/OptionsGTK.cmake: USE(TEXTURE_MAPPER) must be

enabled regardless of whether OpenGL is, because certain symbols
such as WebCore::GraphicsLayer::create() need to be built.

Source/WebCore:

  • CMakeLists.txt: The third-party ANGLE directories need to be

included even if ENABLE(GRAPHICS_CONTEXT_3D) is false. They must
be included after the OpenGL headers as the comment says.

  • platform/graphics/texmap/BitmapTexturePool.cpp: Remove

reference to no longer existent header file.

Jun 20, 2015:

11:43 PM Changeset in webkit [185804] by timothy_horton@apple.com
  • 9 edits in trunk/Source

Deselection of text causes a noticeable jump on force touch machines
​https://bugs.webkit.org/show_bug.cgi?id=146173
<rdar://problem/20992842>

Reviewed by Sam Weinig.

Source/WebCore:

When we have a TextIndicator of type Crossfade, we end up putting
a layer with the blue highlight + text painted into it on top of the
content, and cross-fading that layer to the yellow-highlighted text.

This is necessary for BounceAndCrossfade TextIndicators, because the
blue highlight has to bounce, but is not necessary for Crossfade-only
ones; we can just fade in the yellow highlight on top of the
existing blue page highlight, and all is well.

So, get rid of the Crossfade TextIndicator type and use FadeIn, separately
keeping track of whether or not we can add a margin (we still can't
add a margin to TextIndicators that indicate the page's current selection,
because the blue highlight cannot have the margin applied to it, and we
want the bounds to match exactly).

  • page/TextIndicator.cpp:

(WebCore::TextIndicator::createWithRange):
If the range is the same as the selection, turn off the margin.
We were previously doing this based on the presentation transition, but now
there's no difference in presentation transition in this case.

(WebCore::TextIndicator::createWithSelectionInFrame):
(WebCore::TextIndicator::wantsBounce):
(WebCore::TextIndicator::wantsContentCrossfade):
(WebCore::TextIndicator::wantsFadeIn):
(WebCore::TextIndicator::wantsManualAnimation):

  • page/TextIndicator.h:

Get rid of TextIndicatorPresentationTransition::Crossfade.

(WebCore::TextIndicator::setWantsMargin):
(WebCore::TextIndicator::wantsMargin):
Keep track of whether we want a margin.

  • page/mac/TextIndicatorWindow.mm:

(-[WebTextIndicatorView initWithFrame:textIndicator:margin:]):
Determine if we should use a margin based on wantsMargin instead of the
presentation transition.

Source/WebKit/mac:

  • WebView/WebImmediateActionController.mm:

(-[WebImmediateActionController _animationControllerForText]):
Get rid of TextIndicatorPresentationTransition::Crossfade.

Source/WebKit2:

  • Shared/WebCoreArgumentCoders.cpp:

(IPC::ArgumentCoder<TextIndicatorData>::encode):
(IPC::ArgumentCoder<TextIndicatorData>::decode):
Encode/decode wantsMargin.

  • WebProcess/WebPage/mac/WebPageMac.mm:

(WebKit::WebPage::performImmediateActionHitTestAtLocation):
(WebKit::textIndicatorTransitionForImmediateAction): Deleted.
Get rid of TextIndicatorPresentationTransition::Crossfade.

8:27 PM Changeset in webkit [185803] by mitz@apple.com
  • 5 edits in trunk/Source/WebKit2

[Cocoa] Actually use Objective-C lightweight generics when the compiler supports them
​https://bugs.webkit.org/show_bug.cgi?id=146183

Reviewed by Sam Weinig.

  • Shared/API/Cocoa/WKFoundation.h: When the compiler supports Objective-C generics, define

WK_ARRAY and WK_SET to use them. Also added a definition of WK_DICTIONARY.

  • UIProcess/API/Cocoa/WKWebViewPrivate.h: Use WK_DICTIONARY for the options dictionary of

-_loadRequest:withOptions:.

  • UIProcess/API/Cocoa/WKWebsiteDataStore.mm:

(-[WKWebsiteDataStore fetchDataRecordsOfTypes:completionHandler:]): Made the completion
halder type match the declaration, now that NSArray and WK_ARRAY(WKWebsiteDataRecord *) are
not the same.

  • mac/postprocess-framework-headers.sh: Also rewrite WK_DICTIONARY.
4:00 PM Changeset in webkit [185802] by Michael Catanzaro
  • 4 edits in trunk

[EFL][GTK] Define GLIB_VERSION_MIN_REQUIRED and require glib 2.36 for GTK
​https://bugs.webkit.org/show_bug.cgi?id=146181

Reviewed by Martin Robinson.

.:

Bump our GLib requirement to 2.36 to reflect reality, since we are using GTask.

  • Source/cmake/OptionsGTK.cmake:

Source/WTF:

Define GLIB_VERSION_MIN_REQUIRED to avoid deprecation warnings.

  • wtf/Platform.h:
3:50 PM Changeset in webkit [185801] by Michael Catanzaro
  • 2 edits in trunk/Tools

[GTK] Define GDK_VERSION_MIN_REQUIRED when compiling MiniBrowser
​https://bugs.webkit.org/show_bug.cgi?id=146186

Reviewed by Martin Robinson.

Define GDK_VERSION_MIN_REQUIRED when compiling MiniBrowser to silence several dozen
deprecation warnings. (MiniBrowser is unique in that it does not include wtf/Platform.h.)

  • MiniBrowser/gtk/CMakeLists.txt:
3:47 PM Changeset in webkit [185800] by Nikita Vasilyev
  • 3 edits in trunk/Source/WebInspectorUI

Web Inspector: Unable to select parent element in the DOM tree path bar
​https://bugs.webkit.org/show_bug.cgi?id=145810

Reviewed by Timothy Hatcher.

  • UserInterface/Views/DOMTreeContentView.js:

(WebInspector.DOMTreeContentView.prototype.get selectionPathComponents):

  • UserInterface/Views/HierarchicalPathComponent.js:

(WebInspector.HierarchicalPathComponent.prototype.get selectedPathComponent):
(WebInspector.HierarchicalPathComponent.prototype._selectElementMouseUp):
(WebInspector.HierarchicalPathComponent.prototype._selectElementSelectionChanged):
(WebInspector.HierarchicalPathComponent):
(WebInspector.HierarchicalPathComponent.prototype._selectElementMouseDown): Deleted.

1:50 PM Changeset in webkit [185799] by Chris Dumez
  • 7 edits in trunk/Source/WebKit2

[WK2][iOS] Avoid synchronous IPC on view state change when the content is not visible
​https://bugs.webkit.org/show_bug.cgi?id=146179
<rdar://problem/20923432>

Reviewed by Tim Horton.

After r170787, viewStateChange() would cause a synchronous IPC between
the UIProcess and the WebProcess when the view becomes visible. This
was to avoid painting empty / black tiles when unsuspending the
WebProcess on tab switch, in the event volatile IOSurfaces were purged.

However, this sync IPC can have performance implications and is not
needed when the content is not actually visible yet (e.g.
hideContentUntilNextUpdate() was called, or the tab was killed).

This patch avoids the synchronous IPC when the content is hidden and
exposes a private API on WKWebView so that clients can ask for the
content to be hidden until the next update. This would allow for
clients to avoid the synchronous IPC if they don't need the content
to be displayed synchronously (e.g. the view is obscured).

  • UIProcess/API/Cocoa/WKWebView.mm:

(-[WKWebView _hideContentUntilNextUpdate]):

  • UIProcess/API/Cocoa/WKWebViewPrivate.h:
  • UIProcess/DrawingAreaProxy.h:

(WebKit::DrawingAreaProxy::hasVisibleContent):

  • UIProcess/WebPageProxy.cpp:

(WebKit::WebPageProxy::dispatchViewStateChange):

  • UIProcess/mac/RemoteLayerTreeDrawingAreaProxy.h:
  • UIProcess/mac/RemoteLayerTreeDrawingAreaProxy.mm:

(WebKit::RemoteLayerTreeDrawingAreaProxy::isContentHidden):

9:48 AM Changeset in webkit [185798] by ddkilzer@apple.com
  • 3 edits
    4 copies in trunk/LayoutTests

REGRESSION (r185779): fast/canvas/{canvas-toDataURL-crash,pattern-too-large-to-create}.html are broken

This layout test started failing after the fix for:

Extremely large canvas crashes on pre-El Capitan machines
​https://bugs.webkit.org/show_bug.cgi?id=146169
<rdar://problem/21410046>

Different limits require different test results for each
platform. This is the same fix for these tests as r185793.

  • fast/canvas/canvas-toDataURL-crash-expected.txt: Update.
  • fast/canvas/pattern-too-large-to-create-expected.txt: Update.
  • Set limit to 268435456.
  • platform/ios-simulator/fast/canvas/canvas-toDataURL-crash-expected.txt: Copied from LayoutTests/fast/canvas/canvas-toDataURL-crash-expected.txt.
  • platform/ios-simulator/fast/canvas/pattern-too-large-to-create-expected.txt: Copied from LayoutTests/fast/canvas/pattern-too-large-to-create-expected.txt.
  • Set limit to 16777216.
  • platform/mac-yosemite/fast/canvas/canvas-toDataURL-crash-expected.txt: Copied from LayoutTests/fast/canvas/canvas-toDataURL-crash-expected.txt.
  • platform/mac-yosemite/fast/canvas/pattern-too-large-to-create-expected.txt: Copied from LayoutTests/fast/canvas/pattern-too-large-to-create-expected.txt.
  • Set limit to 67108864.
9:30 AM Changeset in webkit [185797] by ddkilzer@apple.com
  • 2 edits in trunk/LayoutTests

fast/canvas/webgl/tex-image-and-uniform-binding-bugs.html is slow on Mavericks WK1 Debug builds

The Flakiness Dashboard says that this test times out, but only
on Mavericks WK1 Debug builds:

<​https://webkit-test-results.appspot.com/dashboards/flakiness_dashboard.html#showAllRuns=true&tests=fast%2Fcanvas%2Fwebgl%2Ftex-image-and-uniform-binding-bugs.html>

  • platform/mac-wk1/TestExpectations: Mark test as slow on

Mavericks Debug builds:

  • fast/canvas/webgl/tex-image-and-uniform-binding-bugs.html
9:16 AM Changeset in webkit [185796] by ddkilzer@apple.com
  • 2 edits in trunk/LayoutTests

editing/selection/leak-document-with-selection-inside.html is flaky
<​http://webkit.org/b/146182>

  • TestExpectations: Mark as flaky:
  • editing/selection/leak-document-with-selection-inside.html
9:06 AM Changeset in webkit [185795] by Michael Catanzaro
  • 2 edits in trunk/Source/WebKit2

Check for SHA1 certificates ignores subresources
​https://bugs.webkit.org/show_bug.cgi?id=146159

Reviewed by Dan Bernstein.

  • UIProcess/WebPageProxy.cpp:

(WebKit::WebPageProxy::didCommitLoadForFrame): Call
PageLoadState::didDisplayOrRunInsecureContent if a certificate chain contains a non-root
SHA1 certificate when the frame is not the main frame.

8:57 AM Changeset in webkit [185794] by mitz@apple.com
  • 2 edits in trunk/Source/WebKit2

Added availability information to a recently-added delegate method.

Suggested by Alexey Proskuryakov and Anders Carlsson.

  • UIProcess/API/Cocoa/WKUIDelegate.h: Annotated -webViewDidClose:.
8:10 AM Changeset in webkit [185793] by ddkilzer@apple.com
  • 2 edits
    3 adds in trunk/LayoutTests

REGRESSION (r185779): fast/canvas/canvas-too-large-to-draw.html is still broken

This layout test started failing after the fix for:

Extremely large canvas crashes on pre-El Capitan machines
​https://bugs.webkit.org/show_bug.cgi?id=146169
<rdar://problem/21410046>

Different limits require different test results for each
platform. Also update the line number from 35 to 36 since the
html test file was modified in r185792.

  • fast/canvas/canvas-too-large-to-draw-expected.txt:
  • Update limit to 16384x16384 for El Capitan (and other ports) and update line number.
  • platform/ios-simulator/fast/canvas/canvas-too-large-to-draw-expected.txt: Added.
  • New results with 4096x4096 limit for iOS.
  • platform/mac-yosemite/fast/canvas/canvas-too-large-to-draw-expected.txt: Added.
  • New results with 8192x8192 limit for Mac OS X 10.10.x and older.
7:26 AM Changeset in webkit [185792] by ddkilzer@apple.com
  • 2 edits in trunk/LayoutTests

REGRESSION (r185779): fast/canvas/canvas-too-large-to-draw.html broken prior to El Capitan

This layout test started failing after the fix for:

Extremely large canvas crashes on pre-El Capitan machines
​https://bugs.webkit.org/show_bug.cgi?id=146169
<rdar://problem/21410046>

  • fast/canvas/canvas-too-large-to-draw.html: Update test for new

expectations after r185779.

4:04 AM Changeset in webkit [185791] by Yusuke Suzuki
  • 12 edits
    1 add in trunk

[ES6] Destructuring assignment need to accept iterables
​https://bugs.webkit.org/show_bug.cgi?id=144111

Reviewed by Darin Adler.

Source/JavaScriptCore:

This patch makes that destructuring assignments to array binding patterns accept iterables.
Previously, it just access the indexed properties.
After this patch, it iterates the given value by using ES6 iterator protocol.

The iteration becomes different from the for-of case.

  1. Since there's no break/continue case, finally scope is not necessary.
  2. When the error is raised, the close status of the iterator becomes true. So IteratorClose is not called for that.
  3. Since the array binding patterns requires a limited count of iterations (if there is no rest(...rest) case), IteratorClose is called when the iteration does not consume the all values of the iterator.
  4. Since the array binding patterns requires a specified count of iterations, iterator's next call is skipped when iterator becomes closed.
  • bytecompiler/BytecodeGenerator.cpp:

(JSC::BytecodeGenerator::emitIteratorClose):

  • bytecompiler/BytecodeGenerator.h:
  • bytecompiler/NodesCodegen.cpp:

(JSC::ArrayPatternNode::bindValue):

  • parser/ASTBuilder.h:

(JSC::ASTBuilder::finishArrayPattern):

  • parser/Nodes.h:
  • parser/Parser.cpp:

(JSC::Parser<LexerType>::parseDeconstructionPattern):

  • parser/SyntaxChecker.h:

(JSC::SyntaxChecker::operatorStackPop):

  • tests/stress/destructuring-assignment-accepts-iterables.js: Added.

(shouldBe):
(shouldThrow):
(.set shouldThrow):

LayoutTests:

From this patch, we use iterators for destructuring assignments to array binding patterns.
So test results become different.

  • js/destructuring-assignment-expected.txt:
  • js/parser-syntax-check-expected.txt:
  • js/script-tests/destructuring-assignment.js:
2:18 AM Changeset in webkit [185790] by ryuan.choi@navercorp.com
  • 7 edits in trunk

[EFL] Do not consider test directories when DEVELOPER_MODE is OFF
​https://bugs.webkit.org/show_bug.cgi?id=146171

Reviewed by Gyuyoung Kim.

.:

  • Source/cmake/OptionsEfl.cmake:

Exposed ENABLE_DEVELOPER_MODE to Compiler and CMake definition when DEVELOPER_MODE is ON.

Source/WebCore:

  • platform/efl/EflInspectorUtilities.cpp:

(WebCore::inspectorResourcePath): Only used WEB_INSPECTOR_DIR which CMake decides properly.

Source/WebKit2:

  • PlatformEfl.cmake:

Do not define SOURCE_DIR when DEVELOPER_MODE is OFF for seccomp filter
not to allow jhbuild directory.

  • Shared/efl/ProcessExecutablePathEfl.cpp:

(WebKit::findProcessPath): Guard test code to ENABLE(DEVELOPER_MODE).

Note: See TracTimeline for information about the timeline view.