Timeline

Feb 20, 2017:

11:12 PM Changeset in webkit [212703] by Carlos Garcia Campos

• 10 edits
  2 deletes in trunk

Remove ENABLE_THREADED_COMPOSITOR build option
​https://bugs.webkit.org/show_bug.cgi?id=168606

Reviewed by Michael Catanzaro.

.:

• Source/cmake/OptionsGTK.cmake:
• Source/cmake/WebKitFeatures.cmake:

Source/WebCore:

• platform/TextureMapper.cmake:

Source/WebKit2:

Now that we have AC mode on demand with the threaded compositor there's no reason to keep the old AC code.

• PlatformGTK.cmake:
• WebProcess/WebPage/LayerTreeHost.cpp:

(WebKit::LayerTreeHost::create):

• WebProcess/WebPage/gtk/LayerTreeHostGtk.cpp: Removed.
• WebProcess/WebPage/gtk/LayerTreeHostGtk.h: Removed.

Tools:

• Scripts/webkitperl/FeatureList.pm:

10:37 PM Changeset in webkit [212702] by commit-queue@webkit.org

• 3 edits in trunk/Source/WebInspectorUI

Web Inspector: Tooltip for "Show console tab" should read "Show Console tab"
​https://bugs.webkit.org/show_bug.cgi?id=168630
<rdar://problem/30621662>

Patch by Joseph Pecoraro <Joseph Pecoraro> on 2017-02-20
Reviewed by Brian Burg.

• Localizations/en.lproj/localizedStrings.js:
• UserInterface/Views/LogContentView.js:

(WebInspector.LogContentView):

10:37 PM Changeset in webkit [212701] by achristensen@apple.com

• 2 edits in trunk/Source/WebCore

Fix some 32-bit builds after r212699

• platform/mediastream/mac/RealtimeIncomingVideoSource.cpp:

(WebCore::RealtimeIncomingVideoSource::OnFrame):
Explicitly cast OSStatus to int for logging.

9:51 PM Changeset in webkit [212700] by commit-queue@webkit.org

• 5 edits in trunk/Source/WebCore

Remove unnecessary PerformanceEntry virtual methods
​https://bugs.webkit.org/show_bug.cgi?id=168633

Patch by Joseph Pecoraro <Joseph Pecoraro> on 2017-02-20
Reviewed by Ryosuke Niwa.

• page/PerformanceEntry.h:

(WebCore::PerformanceEntry::isResource):
(WebCore::PerformanceEntry::isMark):
(WebCore::PerformanceEntry::isMeasure):
Implement based on the Type.

• page/PerformanceMark.h:
• page/PerformanceMeasure.h:
• page/PerformanceResourceTiming.h:

Remove virtual overrides.

9:50 PM Changeset in webkit [212699] by achristensen@apple.com

• 29 edits
  11 adds in trunk

[WebRTC][Mac] Activate libwebrtc
​https://bugs.webkit.org/show_bug.cgi?id=167293
LayoutTests/imported/w3c:

<rdar://problem/30401864>

Patch by Youenn Fablet <​youennf@gmail.com> on 2017-02-20
Reviewed by Alex Christensen.

• web-platform-tests/webrtc/datachannel-emptystring-expected.txt:
• web-platform-tests/webrtc/no-media-call-expected.txt:
• web-platform-tests/webrtc/promises-call-expected.txt:

Source/ThirdParty/libwebrtc:

<rdar://problem/30401864>

Patch by Youenn Fablet <​youenn@apple.com> on 2017-02-20
Reviewed by Alex Christensen.

Doing some clean-up.

• Configurations/libwebrtc.xcconfig:
• Source/webrtc/base/checks.h:
• libwebrtc.xcodeproj/project.pbxproj:

Source/WebCore:

Patch by Youenn Fablet <​youenn@apple.com> on 2017-02-20
Reviewed by Alex Christensen.

• Configurations/WebCore.xcconfig:
• Configurations/WebCoreTestSupport.xcconfig:

Source/WebKit2:

<rdar://problem/30401864>

Patch by Youenn Fablet <​youenn@apple.com> on 2017-02-20
Reviewed by Alex Christensen.

Activating PeerConnection runtime flag based on successful loading of libwebrtc.dylib.

• Configurations/WebKit.xcconfig:
• Shared/WebPreferencesDefinitions.h:
• UIProcess/WebPreferences.cpp:

(WebKit::checkWebRTCAvailability):

Source/WTF:

Patch by Youenn Fablet <​youenn@apple.com> on 2017-02-20
Reviewed by Alex Christensen.

• wtf/Platform.h:

Tools:

<rdar://problem/30401864>

Patch by Youenn Fablet <​youenn@apple.com> on 2017-02-20
Reviewed by Alex Christensen.

Fixing paths to allow loading libwebrtc.dylib successfully.

• Scripts/build-webkit:
• Scripts/webkitdirs.pm:

(setupMacWebKitEnvironment):
(setupIOSWebKitEnvironment):

LayoutTests:

<rdar://problem/30401864>

Patch by Youenn Fablet <​youenn@apple.com> on 2017-02-20
Reviewed by Alex Christensen.

• TestExpectations:
• fast/mediastream/RTCPeerConnection-addIceCandidate-expected.txt:
• fast/mediastream/RTCPeerConnection-addTrack-reuse-sender-expected.txt:
• fast/mediastream/RTCPeerConnection-icecandidate-event-expected.txt:
• fast/mediastream/RTCPeerConnection-legacy-stream-based-api-expected.txt:
• platform/gtk/TestExpectations:
• platform/gtk/fast/mediastream/RTCPeerConnection-AddRemoveStream-expected.txt: Added.
• platform/gtk/fast/mediastream/RTCPeerConnection-addIceCandidate-expected.txt: Copied from LayoutTests/fast/mediastream/RTCPeerConnection-addIceCandidate-expected.txt.
• platform/gtk/fast/mediastream/RTCPeerConnection-addTrack-reuse-sender-expected.txt: Copied from LayoutTests/fast/mediastream/RTCPeerConnection-addTrack-reuse-sender-expected.txt.
• platform/gtk/fast/mediastream/RTCPeerConnection-closed-state-expected.txt: Added.
• platform/gtk/fast/mediastream/RTCPeerConnection-icecandidate-event-expected.txt: Copied from LayoutTests/fast/mediastream/RTCPeerConnection-icecandidate-event-expected.txt.
• platform/gtk/fast/mediastream/RTCPeerConnection-iceconnectionstatechange-event-expected.txt: Added.
• platform/gtk/fast/mediastream/RTCPeerConnection-legacy-stream-based-api-expected.txt: Copied from LayoutTests/fast/mediastream/RTCPeerConnection-legacy-stream-based-api-expected.txt.
• platform/gtk/fast/mediastream/RTCPeerConnection-onnegotiationneeded-expected.txt: Added.
• platform/gtk/fast/mediastream/RTCPeerConnection-overloaded-operations-params-expected.txt: Added.
• platform/gtk/fast/mediastream/RTCRtpSender-replaceTrack-expected.txt: Added.
• platform/gtk/fast/mediastream/no-media-call-expected.txt: Added.
• platform/ios-simulator-wk2/TestExpectations:
• platform/mac-wk1/TestExpectations:

9:02 PM Changeset in webkit [212698] by commit-queue@webkit.org

• 6 edits in trunk

ASSERTION FAILED: m_normalWorld->hasOneRef() under WorkerThread::stop
​https://bugs.webkit.org/show_bug.cgi?id=168356
<rdar://problem/30592486>

Patch by Joseph Pecoraro <Joseph Pecoraro> on 2017-02-20
Reviewed by Ryosuke Niwa.

Source/WebCore:

• page/DOMWindow.cpp:

(WebCore::DOMWindow::removeAllEventListeners):
Remove Performance object EventListeners.

• workers/WorkerGlobalScope.cpp:

(WebCore::WorkerGlobalScope::~WorkerGlobalScope):
(WebCore::WorkerGlobalScope::removeAllEventListeners):
(WebCore::WorkerGlobalScope::performance):

• workers/WorkerGlobalScope.h:

Remove Performance object EventListeners.
Also clear Performance early in destruction since its ContextDestructionObserver
destruction makes checks about the WorkerThread.

LayoutTests:

• TestExpectations:

Unskip tests now that they no longer trigger assertions.

8:44 PM Changeset in webkit [212697] by aakash_jain@apple.com

• 4 edits in trunk/Source/WebKit/mac

Stop exporting interface WebCoreScrollView on iOS
​https://bugs.webkit.org/show_bug.cgi?id=168536

Reviewed by Dan Bernstein.

• WebKit.exp: Moved symbol for WebCoreScrollView interface to mac specific file.
• WebKit.mac.exp: Ditto.
• WebView/WebHTMLView.mm: WebCoreScrollView shouldn't be defined for iOS platform.

6:35 PM Changeset in webkit [212696] by jer.noble@apple.com

• 6 edits
  2 adds in trunk

Local audio-only stream will not trigger playback to begin
​https://bugs.webkit.org/show_bug.cgi?id=168412

Reviewed by Youenn Fablet.

Source/WebCore:

Test: fast/mediastream/local-audio-playing-event.html

With video-bearing streams, the first video sample emitted from the stream causes the
readyState to move from HAVE_NOTHING to HAVE_ENOUGH_DATA. Wire up this same behavior for
audio-bearing streams. Add a MediaStreamTrackPrivate::Observer callback method
audioSamplesAvailable(), which gets passed up to MediaPlayerPrivateMediaStreamAVFObjC.
Because this callback may be (and probably will be) called on a background thread, run the
updateReadyState() task in a scheduledDeferredTask().

• platform/graphics/avfoundation/objc/MediaPlayerPrivateMediaStreamAVFObjC.h:
• platform/graphics/avfoundation/objc/MediaPlayerPrivateMediaStreamAVFObjC.mm:

(WebCore::MediaPlayerPrivateMediaStreamAVFObjC::audioSamplesAvailable):
(WebCore::MediaPlayerPrivateMediaStreamAVFObjC::updateTracks):

• platform/mediastream/MediaStreamTrackPrivate.cpp:

(WebCore::MediaStreamTrackPrivate::audioSamplesAvailable):

• platform/mediastream/MediaStreamTrackPrivate.h:

(WebCore::MediaStreamTrackPrivate::Observer::audioSamplesAvailable):

LayoutTests:

• fast/mediastream/local-audio-playing-event-expected.txt: Added.
• fast/mediastream/local-audio-playing-event.html: Added.

6:32 PM Changeset in webkit [212695] by BJ Burg

• 4 edits in trunk/Source/WebInspectorUI

Web Inspector: RTL: back/forward navigation buttons should not be flipped
​https://bugs.webkit.org/show_bug.cgi?id=168275
<rdar://problem/30506897>

Reviewed by Matt Baker.

The leading button is always "back" and the trailing button is always "forward", but
in RTL the icons are swapped so that it still looks like '< >' from left-to-right.

Rename the two arrows in the SVG to "left" and "right" arrows, and use them appropriately
in ContentBrowser for back-forward icons based on the layout direction.

• UserInterface/Images/BackForwardArrows.svg:
• UserInterface/Views/ContentBrowser.js:

(WebInspector.ContentBrowser):

• UserInterface/Views/FindBanner.css:

(.find-banner > button.segmented.left > .glyph):
(.find-banner > button.segmented.left:active:not(:disabled) > .glyph):
(.find-banner > button.segmented.right > .glyph):
(.find-banner > button.segmented.right:active:not(:disabled) > .glyph):

6:23 PM Changeset in webkit [212694] by aakash_jain@apple.com

• 2 edits in trunk/Source/WebKit/mac

Exclude various mac-specific header files from iOS build
​https://bugs.webkit.org/show_bug.cgi?id=168621

Reviewed by Dan Bernstein.

• Configurations/WebKitLegacy.xcconfig:

6:12 PM Changeset in webkit [212693] by Alan Bujtas

• 7 edits
  2 adds in trunk

Simple line layout: Implement absoluteQuadsForRange.
​https://bugs.webkit.org/show_bug.cgi?id=168613
<rdar://problem/30614618>

Reviewed by Simon Fraser.

Source/WebCore:

This patch ensures that the commonly used Range::getClientRects calls do not
throw us off of the simple line layout path.

Test: fast/dom/Range/simple-line-layout-getclientrects.html

• rendering/RenderText.cpp:

(WebCore::RenderText::absoluteQuadsForRange):

• rendering/SimpleLineLayoutFunctions.cpp:

(WebCore::SimpleLineLayout::collectAbsoluteQuadsForRange): Special case empty ranges with multiple empty runs.

• rendering/SimpleLineLayoutFunctions.h:
• rendering/SimpleLineLayoutResolver.cpp:

(WebCore::SimpleLineLayout::RunResolver::rangeForRendererWithOffsets):

• rendering/SimpleLineLayoutResolver.h:

LayoutTests:

• fast/dom/Range/simple-line-layout-getclientrects-expected.html: Added.
• fast/dom/Range/simple-line-layout-getclientrects.html: Added.

5:51 PM Changeset in webkit [212692] by mark.lam@apple.com

• 13 edits
  1 add in trunk/Source

[Re-landing] CachedCall should let GC know to keep its arguments alive.
​https://bugs.webkit.org/show_bug.cgi?id=168567
<rdar://problem/30475767>

Reviewed by Saam Barati.

Source/JavaScriptCore:

We fix this by having CachedCall use a MarkedArgumentBuffer to store its
arguments instead of a Vector.

Also declared CachedCall, MarkedArgumentBuffer, and ProtoCallFrame as
WTF_FORBID_HEAP_ALLOCATION because they rely on being stack allocated for
correctness.

Update: the original patch has a bug in MarkedArgumentBuffer::expandCapacity()
where it was copying and calling addMarkSet() on values in m_buffer beyond m_size
(up to m_capacity). As a result, depending on the pre-existing values in
m_inlineBuffer, this may result in a computed Heap pointer that is wrong, and
subsequently, manifest as a crash. This is likely to be the cause of the PLT
regression.

I don't have a new test for this fix because the issue relies on sufficiently bad
values randomly showing up in m_inlineBuffer when we do an ensureCapacity() which
calls expandCapacity().

• interpreter/CachedCall.h:

(JSC::CachedCall::CachedCall):
(JSC::CachedCall::call):
(JSC::CachedCall::clearArguments):
(JSC::CachedCall::appendArgument):
(JSC::CachedCall::setArgument): Deleted.

• interpreter/CallFrame.h:

(JSC::ExecState::emptyList):

• interpreter/Interpreter.cpp:

(JSC::Interpreter::prepareForRepeatCall):

• interpreter/Interpreter.h:
• interpreter/ProtoCallFrame.h:
• runtime/ArgList.cpp:

(JSC::MarkedArgumentBuffer::slowEnsureCapacity):
(JSC::MarkedArgumentBuffer::expandCapacity):
(JSC::MarkedArgumentBuffer::slowAppend):

• runtime/ArgList.h:

(JSC::MarkedArgumentBuffer::append):
(JSC::MarkedArgumentBuffer::ensureCapacity):

• runtime/StringPrototype.cpp:

(JSC::replaceUsingRegExpSearch):

• runtime/VM.cpp:

(JSC::VM::VM):

• runtime/VM.h:

Source/WTF:

Added a WTF_FORBID_HEAP_ALLOCATION that will cause a compilation failure if
a class declared with it is malloced.

While this doesn't prevent that class declared WTF_FORBID_HEAP_ALLOCATION from
being embedded in another class that is heap allocated, it does at minimum
document the intent and gives the users of this class a chance to do the
right thing.

• WTF.xcodeproj/project.pbxproj:
• wtf/ForbidHeapAllocation.h: Added.

5:48 PM Changeset in webkit [212691] by Ryan Haddad

• 28 edits
  19 deletes in trunk

Unreviewed, rolling out r212685.

This change broke the 32-bit Sierra build.

Reverted changeset:

"Resource Load Statistics: Add alternate classification
method"
​https://bugs.webkit.org/show_bug.cgi?id=168347
​http://trac.webkit.org/changeset/212685

5:46 PM Changeset in webkit [212690] by matthew_hanson@apple.com

• 1 copy in tags/Safari-603.1.30

Tag Safari-603.1.30.

5:44 PM Changeset in webkit [212689] by commit-queue@webkit.org

• 9 edits in trunk/Source/WebCore

[WebIDL] Add support for constructors that have variadic parameters
​https://bugs.webkit.org/show_bug.cgi?id=168614

Patch by Sam Weinig <​sam@webkit.org> on 2017-02-20
Reviewed by Alex Christensen.

In preparation of removing some more custom bindings, move the constructor calling
code over to using GenerateParametersCheck completely. This required

• Removing special case of NamedConstructors automatically getting a Document passed to them. The three named constructors that expected this have been updated to add the required attributes.
• Add a version of the GenerateCallWith prologue for constructors.
• Remove now unnecessary argumentCount return value from GenerateParametersCheck.

• bindings/scripts/CodeGeneratorJS.pm:

(GenerateImplementation):
(GenerateCallWithUsingReferences):
(GenerateCallWithUsingPointers):
(GenerateConstructorCallWithUsingPointers):
(GenerateCallWith):
(GenerateParametersCheck):
(GenerateConstructorDefinition):

• bindings/scripts/test/JS/JSTestInterface.cpp:
• bindings/scripts/test/JS/JSTestNamedConstructor.cpp:
• bindings/scripts/test/JS/JSTestObj.cpp:
• bindings/scripts/test/JS/JSTestOverloadedConstructors.cpp:

Update results for more consistent placement of ScriptExecutionContext and Document
accessors, add UNLIKELY to context checks, and remove automatic Document passing
for NamedConstructors.

• html/HTMLAudioElement.idl:
• html/HTMLImageElement.idl:
• html/HTMLOptionElement.idl:

Add ConstructorCallWith=Document.

5:22 PM Changeset in webkit [212688] by BJ Burg

• 2 edits in trunk/Source/WebInspectorUI

Web Inspector: RTL: hierarchical path components need spacing and icon adjustments
​https://bugs.webkit.org/show_bug.cgi?id=168581

Reviewed by Matt Baker.

Flip a bunch of padding, margin, and offsets. Flip the chevron using a CSS transform.

• UserInterface/Views/HierarchicalPathComponent.css:

(.hierarchical-path-component):
(.hierarchical-path-component > .icon):
(.hierarchical-path-component > :matches(.icon, .selector-arrows)):
(body[dir=ltr] .hierarchical-path-component > :matches(.icon, .selector-arrows)):
(body[dir=rtl] .hierarchical-path-component > :matches(.icon, .selector-arrows)):
(.hierarchical-path-component > .selector-arrows):
(.hierarchical-path-component > select):
(body[dir=ltr] .hierarchical-path-component > select):
(body[dir=rtl] .hierarchical-path-component > select):
(.hierarchical-path-component > .separator):
(body[dir=ltr] .hierarchical-path-component > .separator):
(body[dir=rtl] .hierarchical-path-component > .separator):
(.hierarchical-path-component.text-only):
(body[dir=ltr] .hierarchical-path-component.text-only):
(body[dir=rtl] .hierarchical-path-component.text-only):
(body[dir=ltr] .hierarchical-path-component.text-only > select):
(body[dir=rtl] .hierarchical-path-component.text-only > select):

4:53 PM Changeset in webkit [212687] by matthew_hanson@apple.com

• 4 edits in branches/safari-603-branch/Source/WebCore

Roll out r212554 via r212601. rdar://problem/30339638

4:35 PM Changeset in webkit [212686] by Ryan Haddad

• 1 edit
  3 adds in trunk/LayoutTests

Add mac-wk1 baseline for fast/canvas/webgl/context-creation-attributes.html
​https://bugs.webkit.org/show_bug.cgi?id=168623

Unreviewed test gardening.

• platform/mac-wk1/fast/canvas/webgl/context-creation-attributes-expected.txt: Added.

4:34 PM Changeset in webkit [212685] by wilander@apple.com

• 28 edits
  4 copies
  24 adds in trunk

Resource Load Statistics: Add alternate classification method
​https://bugs.webkit.org/show_bug.cgi?id=168347
<rdar://problem/30352793>

Reviewed by Alex Christensen.

Source/WebCore:

This patch only adds test infrastructure in WebCore.

Tests: http/tests/loading/resourceLoadStatistics/classify-as-non-prevalent-based-on-mixed-statistics.html

http/tests/loading/resourceLoadStatistics/classify-as-non-prevalent-based-on-sub-frame-under-top-frame-origins.html
http/tests/loading/resourceLoadStatistics/classify-as-non-prevalent-based-on-subresource-under-top-frame-origins.html
http/tests/loading/resourceLoadStatistics/classify-as-non-prevalent-based-on-subresource-unique-redirects-to.html
http/tests/loading/resourceLoadStatistics/classify-as-prevalent-based-on-mixed-statistics.html
http/tests/loading/resourceLoadStatistics/classify-as-prevalent-based-on-sub-frame-under-top-frame-origins.html
http/tests/loading/resourceLoadStatistics/classify-as-prevalent-based-on-subresource-under-top-frame-origins.html
http/tests/loading/resourceLoadStatistics/classify-as-prevalent-based-on-subresource-unique-redirects-to.html

• loader/ResourceLoadObserver.cpp:

(WebCore::ResourceLoadObserver::setSubframeUnderTopFrameOrigin):
(WebCore::ResourceLoadObserver::setSubresourceUnderTopFrameOrigin):
(WebCore::ResourceLoadObserver::setSubresourceUniqueRedirectTo):

• loader/ResourceLoadObserver.h:

Source/WebKit2:

This patch adds a CorePrediction-based classifier to the WebResourceLoadStatisticsStore.
The CorePrediction framework is introduced as a dependency for macOS and iOS. The patch
also adds functions to support layout tests of the feature.

• Configurations/BaseTarget.xcconfig:
• Configurations/WebKit.xcconfig:

Added dependency on CorePrediction.

• Platform/classifier: Added.
• Platform/classifier/ResourceLoadStatisticsClassifier.h: Added.

Pulls in the Cocoa-specific classifier for Cocoa-based platforms.

(WebKit::ResourceLoadStatisticsClassifier::ResourceLoadStatisticsClassifier):

• Platform/classifier/ResourceLoadStatisticsClassifierBase.cpp: Added.

(WebKit::ResourceLoadStatisticsClassifierBase::hasPrevalentResourceCharacteristics):

Shared classifier logic.

(WebKit::ResourceLoadStatisticsClassifierBase::classifyWithVectorThreshold):

Fallback classifier for when we don't have CorePrediction.

• Platform/classifier/ResourceLoadStatisticsClassifierBase.h: Added.

(WebKit::ResourceLoadStatisticsClassifierBase::ResourceLoadStatisticsClassifierBase):
(WebKit::ResourceLoadStatisticsClassifierBase::~ResourceLoadStatisticsClassifierBase):

• Platform/classifier/cocoa: Added.
• Platform/classifier/cocoa/CorePredictionSPI.h: Added.

Includes CorePrediction if available, otherwise declares needed symbols.
Always redeclares the functions so we'll get a build error if
CorePrediction changes.

• Platform/classifier/cocoa/ResourceLoadStatisticsClassifierCocoa.cpp: Added.

(WebKit::ResourceLoadStatisticsClassifier::classify):
(WebKit::ResourceLoadStatisticsClassifier::storagePath):
(WebKit::ResourceLoadStatisticsClassifier::shouldUseCorePrediction):

The new classifier for Cocoa platforms.

• Platform/classifier/cocoa/ResourceLoadStatisticsClassifierCocoa.h: Added.

(WebKit::ResourceLoadStatisticsClassifier::ResourceLoadStatisticsClassifier):

• PlatformGTK.cmake:

Added Platform/classifier directory and ResourceLoadStatisticsClassifierBase.cpp.

• PlatformMac.cmake:

Added Platform/classifier and Platform/classifier/cocoa directories, and
source files ResourceLoadStatisticsClassifierBase.cpp and
ResourceLoadStatisticsClassifierCocoa.cpp

• Resources/ResourceLoadStatistics: Added.
• Resources/ResourceLoadStatistics/corePrediction_model: Added.

Model file to load into CorePrediction.

• UIProcess/API/C/WKResourceLoadStatisticsManager.cpp:

(WKResourceLoadStatisticsManagerSetSubframeUnderTopFrameOrigin):
(WKResourceLoadStatisticsManagerSetSubresourceUnderTopFrameOrigin):
(WKResourceLoadStatisticsManagerSetSubresourceUniqueRedirectTo):

Test infrastructure.

• UIProcess/API/C/WKResourceLoadStatisticsManager.h:
• UIProcess/WebResourceLoadStatisticsManager.cpp:

(WebKit::WebResourceLoadStatisticsManager::setSubframeUnderTopFrameOrigin):
(WebKit::WebResourceLoadStatisticsManager::setSubresourceUnderTopFrameOrigin):
(WebKit::WebResourceLoadStatisticsManager::setSubresourceUniqueRedirectTo):

Test infrastructure.

• UIProcess/WebResourceLoadStatisticsManager.h:
• UIProcess/WebResourceLoadStatisticsStore.cpp:

(WebKit::WebResourceLoadStatisticsStore::WebResourceLoadStatisticsStore):

Renamed m_storagePath to m_statisticsStoragePath for clarity.

(WebKit::WebResourceLoadStatisticsStore::classifyResource):

Now calls the classifier through its ResourceLoadStatisticsClassifier
member variable.

(WebKit::WebResourceLoadStatisticsStore::persistentStoragePath):

Renamed m_storagePath to m_statisticsStoragePath for clarity.

(WebKit::WebResourceLoadStatisticsStore::writeEncoderToDisk):

Renamed m_storagePath to m_statisticsStoragePath for clarity.

(WebKit::WebResourceLoadStatisticsStore::hasPrevalentResourceCharacteristics): Deleted.

• UIProcess/WebResourceLoadStatisticsStore.h:
• WebKit2.xcodeproj/project.pbxproj:

Added the new classifier source files under Platform/classifier and the
corePrediction_model file under Resources/ResourceLoadStatistics.

Source/WTF:

• wtf/Platform.h:

Added support for HAVE(CORE_PREDICTION).

Tools:

Added three testRunner functions to facilitate layout tests:

• setStatisticsSubframeUnderTopFrameOrigin()
• setStatisticsSubresourceUnderTopFrameOrigin()
• setStatisticsSubresourceUniqueRedirectTo()

• WebKitTestRunner/InjectedBundle/Bindings/TestRunner.idl:
• WebKitTestRunner/InjectedBundle/TestRunner.cpp:

(WTR::TestRunner::setStatisticsSubframeUnderTopFrameOrigin):
(WTR::TestRunner::setStatisticsSubresourceUnderTopFrameOrigin):
(WTR::TestRunner::setStatisticsSubresourceUniqueRedirectTo):

• WebKitTestRunner/InjectedBundle/TestRunner.h:
• WebKitTestRunner/TestController.cpp:

(WTR::TestController::setStatisticsSubframeUnderTopFrameOrigin):
(WTR::TestController::setStatisticsSubresourceUnderTopFrameOrigin):
(WTR::TestController::setStatisticsSubresourceUniqueRedirectTo):

• WebKitTestRunner/TestController.h:
• WebKitTestRunner/TestInvocation.cpp:

(WTR::TestInvocation::didReceiveSynchronousMessageFromInjectedBundle):

WebKitLibraries:

• WebKitPrivateFrameworkStubs/iOS/10/CorePrediction.framework: Added.
• WebKitPrivateFrameworkStubs/iOS/10/CorePrediction.framework/CorePrediction.tbd: Added.

Stubs for private framework.

LayoutTests:

• http/tests/loading/resourceLoadStatistics/classify-as-non-prevalent-based-on-mixed-statistics-expected.txt: Added.
• http/tests/loading/resourceLoadStatistics/classify-as-non-prevalent-based-on-mixed-statistics.html: Added.
• http/tests/loading/resourceLoadStatistics/classify-as-non-prevalent-based-on-sub-frame-under-top-frame-origins-expected.txt: Added.
• http/tests/loading/resourceLoadStatistics/classify-as-non-prevalent-based-on-sub-frame-under-top-frame-origins.html: Added.
• http/tests/loading/resourceLoadStatistics/classify-as-non-prevalent-based-on-subresource-under-top-frame-origins-expected.txt: Added.
• http/tests/loading/resourceLoadStatistics/classify-as-non-prevalent-based-on-subresource-under-top-frame-origins.html: Added.
• http/tests/loading/resourceLoadStatistics/classify-as-non-prevalent-based-on-subresource-unique-redirects-to-expected.txt: Added.
• http/tests/loading/resourceLoadStatistics/classify-as-non-prevalent-based-on-subresource-unique-redirects-to.html: Added.
• http/tests/loading/resourceLoadStatistics/classify-as-prevalent-based-on-mixed-statistics-expected.txt: Added.
• http/tests/loading/resourceLoadStatistics/classify-as-prevalent-based-on-mixed-statistics.html: Added.
• http/tests/loading/resourceLoadStatistics/classify-as-prevalent-based-on-sub-frame-under-top-frame-origins-expected.txt: Added.
• http/tests/loading/resourceLoadStatistics/classify-as-prevalent-based-on-sub-frame-under-top-frame-origins.html: Added.
• http/tests/loading/resourceLoadStatistics/classify-as-prevalent-based-on-subresource-under-top-frame-origins-expected.txt: Added.
• http/tests/loading/resourceLoadStatistics/classify-as-prevalent-based-on-subresource-under-top-frame-origins.html: Added.
• http/tests/loading/resourceLoadStatistics/classify-as-prevalent-based-on-subresource-unique-redirects-to-expected.txt: Added.
• http/tests/loading/resourceLoadStatistics/classify-as-prevalent-based-on-subresource-unique-redirects-to.html: Added.
• platform/gtk/TestExpectations:

The whole http/tests/loading/resourceLoadStatistics marked as crashing based on
Carlos Garcia Campos's assessment in ​https://bugs.webkit.org/show_bug.cgi?id=168171.

• platform/wk2/TestExpectations:

The above tests are only valid for WebKit2. Marked as [ Pass ].

4:29 PM Changeset in webkit [212684] by matthew_hanson@apple.com

• 1 delete in tags/Safari-603.1.30

Remove tag.

4:25 PM Changeset in webkit [212683] by matthew_hanson@apple.com

• 10 edits
  4 deletes in branches/safari-603-branch/Source

Roll out r212660. rdar://problem/30553220

4:18 PM Changeset in webkit [212682] by matthew_hanson@apple.com

• 6 edits
  6 deletes in branches/safari-603-branch

Roll out r212621 via r212647. rdar://problem/30563318

4:08 PM Changeset in webkit [212681] by andersca@apple.com

• 10 edits in trunk

Use the new drag code path when dragging links
​https://bugs.webkit.org/show_bug.cgi?id=168612

Reviewed by Tim Horton.

Source/WebCore:

• editing/Editor.cpp:

(WebCore::Editor::copyURL):
Use userVisibleString instead of fillInUserVisibleForm.

(WebCore::Editor::pasteboardWriterURL):
New function that returns a PasteboardWriterData::URL for a given URL + title.

• editing/Editor.h:

Add new members.

• editing/mac/EditorMac.mm:

(WebCore::Editor::userVisibleString):
(WebCore::Editor::fillInUserVisibleForm): Deleted.
Add a getter instead of a function that fills in a struct member.

• page/DragController.cpp:

(WebCore::DragController::startDrag):
Add the new DragItem code path when dragging links.

• platform/PasteboardWriterData.cpp:

(WebCore::PasteboardWriterData::isEmpty):
Check for m_url as well.

(WebCore::PasteboardWriterData::setURL):
Set m_url.

• platform/PasteboardWriterData.h:

Add new members.

• platform/mac/PasteboardWriter.mm:

(WebCore::toUTI):
New helper function that returns an UTI from a pasteboard type.

(WebCore::createPasteboardWriter):
Handle converting URLs to the various pasteboard types.

Tools:

If a type is not dynamic, just use the static UTI.

• DumpRenderTree/mac/DumpRenderTreePasteboard.mm:

(-[LocalPasteboard writeObjects:]):

4:04 PM Changeset in webkit [212680] by matthew_hanson@apple.com

• 1 copy in tags/Safari-603.1.30

Tag Safari-603.1.30.

4:04 PM Changeset in webkit [212679] by matthew_hanson@apple.com

• 1 delete in tags/Safari-603.1.30

Remove tag.

2:29 PM Changeset in webkit [212678] by ap@apple.com

• 2 edits in trunk/LayoutTests

LayoutTest fast/multicol/column-span-parent-continuation-crash.html is a flaky timeout
​https://bugs.webkit.org/show_bug.cgi?id=168341

Reviewed by Daniel Bates.

• fast/multicol/column-span-parent-continuation-crash.html: Make sure that the

document is fully parsed before the test runs. I'm not sure if the zero delay
is still needed, but it doesn't hurt.

1:48 PM Changeset in webkit [212677] by matthew_hanson@apple.com

• 13 edits
  1 delete in branches/safari-603-branch/Source

Roll out r212618 via r212646. rdar://problem/30475767

1:48 PM Changeset in webkit [212676] by matthew_hanson@apple.com

• 8 edits in branches/safari-603-branch

Merge r211750. rdar://problem/30383964

1:48 PM Changeset in webkit [212675] by matthew_hanson@apple.com

• 2 edits in branches/safari-603-branch/Source/WebCore

Merge r212651. rdar://problem/30604254

1:48 PM Changeset in webkit [212674] by matthew_hanson@apple.com

• 15 edits
  2 adds in branches/safari-603-branch

Merge r212650. rdar://problem/30604254

1:48 PM Changeset in webkit [212673] by matthew_hanson@apple.com

• 9 edits in branches/safari-603-branch

Merge r212633. rdar://problem/30592266

1:48 PM Changeset in webkit [212672] by matthew_hanson@apple.com

• 8 edits in branches/safari-603-branch/Source

Merge r212331. rdar://problem/30522092

1:27 PM Changeset in webkit [212671] by Ryan Haddad

• 2 edits in trunk/LayoutTests

Unreviewed, rolling out r212653.

This rebaseline was incorrect.

Reverted changeset:

"Rebaseline fast/canvas/webgl/context-creation-attributes.html
after r212650."
​http://trac.webkit.org/changeset/212653

1:24 PM Changeset in webkit [212670] by Ryan Haddad

• 2 edits in trunk/LayoutTests

Remove debug flag from flaky test streams/pipe-to.html.
​https://bugs.webkit.org/show_bug.cgi?id=165874

Unreviewed test gardening.

• platform/mac/TestExpectations:

1:21 PM Changeset in webkit [212669] by Wenson Hsieh

• 28 edits in trunk/Source

Add plumbing to the UI process for TextIndicatorData when computing the drag image
​https://bugs.webkit.org/show_bug.cgi?id=168583
Work towards <rdar://problem/30313681>

Source/WebCore:

Reviewed by Tim Horton.

Adds some additional arguments to drag image creation functions to allow platforms to specify TextIndicatorData
for a link or selection-based DragImage. This TextIndicatorData is an optional member of the DragImage, and is
propagated to the UI process via SetDragImage in the WebDragClient.

Additionally renames hasDataInteractionAtPosition to hasSelectionAtPosition to better reflect its purpose.

No new tests, since there is no behavior change.

• page/DragController.cpp:

(WebCore::DragController::startDrag):

• page/Page.cpp:

(WebCore::Page::hasSelectionAtPosition):
(WebCore::Page::hasDataInteractionAtPosition): Deleted.

• page/Page.h:
• platform/DragImage.cpp:

(WebCore::createDragImageForSelection):
(WebCore::createDragImageForLink):
(WebCore::DragImage::operator=):

• platform/DragImage.h:
• platform/gtk/DragImageGtk.cpp:

(WebCore::createDragImageForLink):

• platform/mac/DragImageMac.mm:

(WebCore::createDragImageForLink):

• platform/win/DragImageWin.cpp:

(WebCore::createDragImageForLink):

Source/WebKit/mac:

Accounts for changes in WebCore -- see Source/WebCore/ChangeLog for more details.

Reviewed by Tim Horton.

• WebView/WebHTMLView.mm:

(-[WebHTMLView _selectionDraggingImage]):
(-[WebHTMLView selectionImageForcingBlackText:selectionImageForcingBlackText:]):

Source/WebKit2:

Reviewed by Tim Horton.

Changes the SetDragImage IPC message to send TextIndicatorData over to the UI process. Also changes the boolean
isLink parameter to be a DragSourceAction instead, leaving it up to the receiving end (in the UI process) to
special case links.

Additionally, renames hasDataInteractionAtPosition to hasSelectionAtPosition to better reflect its purpose.

These changes are only infrastructural should not alter any behaviors.

• Shared/ios/InteractionInformationAtPosition.h:
• Shared/ios/InteractionInformationAtPosition.mm:

(WebKit::InteractionInformationAtPosition::encode):
(WebKit::InteractionInformationAtPosition::decode):

• UIProcess/API/Cocoa/WKWebView.mm:

(-[WKWebView _simulateDataInteractionGestureRecognized]):
(-[WKWebView _simulateDataInteractionEntered:]):
(-[WKWebView _simulateDataInteractionUpdated:]):
(-[WKWebView _simulateDataInteractionPerformOperation:]):
(-[WKWebView _simulateDataInteractionEnded:]):
(-[WKWebView _simulateDataInteractionSessionDidEnd:withOperation:]):
(-[WKWebView _simulateFailedDataInteractionWithIndex:]):
(-[WKWebView _simulateWillBeginDataInteractionWithIndex:withSession:]):
(-[WKWebView _simulatedItemsForDataInteractionWithIndex:]):

• UIProcess/API/Cocoa/_WKTestingDelegate.h:
• UIProcess/PageClient.h:
• UIProcess/WebPageProxy.h:
• UIProcess/WebPageProxy.messages.in:
• UIProcess/ios/PageClientImplIOS.h:
• UIProcess/ios/PageClientImplIOS.mm:

(WebKit::PageClientImpl::setDragImage):
(WebKit::PageClientImpl::startDataInteractionWithImage):

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

(-[WKContentView hasSelectablePositionAtPoint:]):
(-[WKContentView pointIsInDataInteractionContent:]):
(-[WKContentView _dataInteractionGestureRecognizer]):
(-[WKContentView pointIsInAssistedNode:]):

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

(WebKit::PageClientImpl::setDragImage):

• UIProcess/mac/WebPageProxyMac.mm:

(WebKit::WebPageProxy::setDragImage):

• WebProcess/WebCoreSupport/mac/WebDragClientMac.mm:

(WebKit::WebDragClient::startDrag):

• WebProcess/WebPage/ios/WebPageIOS.mm:

(WebKit::WebPage::getPositionInformation):

1:10 PM Changeset in webkit [212668] by matthew_hanson@apple.com

• 2 edits in branches/safari-603-branch/Source/WebKit2

Merge r212666. rdar://problem/30239618

12:57 PM Changeset in webkit [212667] by Brent Fulgham

• 2 edits in trunk/Source/WebCore

Nullptr dereferences when stopping a load
​https://bugs.webkit.org/show_bug.cgi?id=168608
<rdar://problem/29852056>

Reviewed by Ryosuke Niwa.

Don't attempt to notify a detached frame's load client that the load is
stopped.

• loader/DocumentLoader.cpp:

(WebCore::DocumentLoader::stopLoading): Check for null frame loader and
bypass dereferencing it.

12:37 PM Changeset in webkit [212666] by commit-queue@webkit.org

• 2 edits in trunk/Source/WebKit2

Don’t set NSURLSessionConfiguration’s _sourceApplicationAuditTokenData if _sourceApplicationBundleIdentifier is also set.
​https://bugs.webkit.org/show_bug.cgi?id=168607

rdar://problem/30239618

Patch by David Quesada <​david_quesada@apple.com> on 2017-02-20
Reviewed by Geoffrey Garen.

• NetworkProcess/cocoa/NetworkSessionCocoa.mm:

(WebKit::NetworkSessionCocoa::NetworkSessionCocoa):
If the process pool configuration sets a source application bundle identifier, don't give the NSURLSession configuration
the audit token from the UI process, as it may be associated with an application with a conflicting bundle identifier.

12:30 PM Changeset in webkit [212665] by commit-queue@webkit.org

• 13 edits
  1 delete in trunk/Source

Unreviewed, rolling out r212618.
​https://bugs.webkit.org/show_bug.cgi?id=168609

"Appears to cause PLT regression" (Requested by mlam on
#webkit).

Reverted changeset:

"CachedCall should let GC know to keep its arguments alive."
​https://bugs.webkit.org/show_bug.cgi?id=168567
​http://trac.webkit.org/changeset/212618

11:57 AM Changeset in webkit [212664] by matthew_hanson@apple.com

• 15 edits in branches/safari-603-branch/Source

Roll out r212518 via r212538. rdar://problem/30541748

11:32 AM Changeset in webkit [212663] by andersca@apple.com

• 15 edits
  1 copy
  1 move in trunk

Add a new drag code path and use it for dragging plain text
​https://bugs.webkit.org/show_bug.cgi?id=168534

Reviewed by Geoffrey Garen.

Source/WebCore:

The new code path is currently only used for plain text selection dragging.

• WebCore.xcodeproj/project.pbxproj:
• page/DragClient.h:

(WebCore::DragClient::useLegacyDragClient):
Add a new member function. This currently returns true everywhere except legacy WebKit on Mac.

(WebCore::DragClient::beginDrag):
Add new empty member function.

• page/DragController.cpp:

(WebCore::dragImageAnchorPointForSelectionDrag):
Helper function that returns the anchor point for a selection drag image.

(WebCore::DragController::startDrag):
If we can use the new code path, set up a drag item with an image and pasteboard data.

(WebCore::DragController::beginDrag):
New function that calls out to the client.

• page/DragController.h:

Add new members.

• platform/DragItem.h:

Add a new class - it's essentially a tuple of a drag image, an anchor point and the pasteboard data the item represents.

• platform/mac/PasteboardWriter.h:

Rename createPasteboardWriting to createPasteboardWriter.

• platform/mac/PasteboardWriter.mm:

(WebCore::createPasteboardWriter):
Use the UTI pasteboard type.

Source/WebKit/mac:

• WebCoreSupport/WebDragClient.h:

Add new members.

• WebCoreSupport/WebDragClient.mm:

(WebDragClient::useLegacyDragClient):
Return false here.

(WebDragClient::beginDrag):
New client member function. Create an NSDraggingItem and use it to begin a dragging session.

• WebView/WebHTMLView.mm:

(-[WebHTMLView draggingSession:sourceOperationMaskForDraggingContext:]):
Get the source operation mask from WebCore.

(-[WebHTMLView draggingSession:endedAtPoint:operation:]):
Send a fake mouse up event.

• WebView/WebHTMLViewInternal.h:

Declare that WebHTMLView conforms to NSDraggingSource.

Tools:

Update the testing infrastructure to handle NSDraggingSession.

• DumpRenderTree/DumpRenderTree.xcodeproj/project.pbxproj:
• DumpRenderTree/mac/DumpRenderTreePasteboard.mm: Renamed from Tools/DumpRenderTree/mac/DumpRenderTreePasteboard.m.

(-[LocalPasteboard writeObjects:]):
Implement this in terms of the legacy pasteboard API.

• DumpRenderTree/mac/EventSendingController.mm:

(drt_WebHTMLView_beginDraggingSessionWithItemsEventSource):
(+[EventSendingController initialize]):
Swizzle the drag method instead of adding a delegate method (like the old code does).

11:14 AM Changeset in webkit [212662] by BJ Burg

• 2 edits in trunk/Source/WebInspectorUI

Web Inspector: RTL: adjust positioning of scope bar items with multiple values
​https://bugs.webkit.org/show_bug.cgi?id=168575

Reviewed by Matt Baker.

Mirror the leading margin as necessary. Adjust the <select> positioning
so that the popup is really close to the scope bar item's position.

• UserInterface/Views/ScopeBar.css:

(body[dir=ltr] .scope-bar > li.multiple > select):
(body[dir=rtl] .scope-bar > li.multiple > select):
(.scope-bar > li.multiple > .arrows):
(body[dir=ltr] .scope-bar > li.multiple > .arrows):
(body[dir=rtl] .scope-bar > li.multiple > .arrows):

10:39 AM Changeset in webkit [212661] by achristensen@apple.com

• 2 edits in trunk/Source/WebCore

Fix libwebrtc build after r212644.
​https://bugs.webkit.org/show_bug.cgi?id=168596

• Modules/mediastream/libwebrtc/LibWebRTCMediaEndpoint.cpp:

(WebCore::LibWebRTCMediaEndpoint::StatsCollector::OnStatsDelivered):
m_endpoint changed from a c++ reference to a WTF::Ref, so we need to use operator -> instead of .

10:33 AM Changeset in webkit [212660] by matthew_hanson@apple.com

• 10 edits
  1 copy
  3 adds in branches/safari-603-branch/Source

rdar://problem/30553220

10:32 AM Changeset in webkit [212659] by ap@apple.com

• 6 edits in trunk/Source/WebKit

List of WebCore symbols to re-export from WebKitLegacy should be built automatically
​https://bugs.webkit.org/show_bug.cgi?id=168342
rdar://problem/26182266

Reviewed by Tim Horton.

Source/WebKit:

• WebKit.xcodeproj/project.pbxproj: Append ReexportedWebCoreSymbols_${CURRENT_ARCH}.exp

to WebKitLegacy.generated.exp if it exists. Added it to build phase inputs for
dependency tracking.

Source/WebKit/ios:

• WebKit.iOS.exp: Removed manually added symbols.

Source/WebKit/mac:

• MigrateHeaders.make: On iOS only, process re-exported headers to build a list of

symbols to re-export.

10:30 AM Changeset in webkit [212658] by matthew_hanson@apple.com

• 10 edits in branches/safari-603-branch/Source

Merge r212652. rdar://problem/30435303

10:22 AM Changeset in webkit [212657] by BJ Burg

• 2 edits in trunk/Source/WebInspectorUI

Web Inspector: RTL: sidebar expand/collapse icons need to be flipped
​https://bugs.webkit.org/show_bug.cgi?id=168276
<rdar://problem/30506999>

Reviewed by Matt Baker.

The sidebars themselves are already flipped due to the layout direction change, but the
icons have a directional design and need to be manually flipped based on the layout direction.

• UserInterface/Views/ContentBrowserTabContentView.js:

(WebInspector.ContentBrowserTabContentView):

10:18 AM Changeset in webkit [212656] by commit-queue@webkit.org

• 17 edits
  1 copy
  4 adds in trunk

[Readable Streams API] Add ReadableByteStreamController byobRequest attribute
​https://bugs.webkit.org/show_bug.cgi?id=168319

Patch by Romain Bellessort <​romain.bellessort@crf.canon.fr> on 2017-02-20
Reviewed by Youenn Fablet.

LayoutTests/imported/w3c:

Updated expectations for imported w3c tests (new type of error is now obtained).

• web-platform-tests/streams/readable-byte-streams/general-expected.txt: Updated.
• web-platform-tests/streams/readable-byte-streams/general.dedicatedworker-expected.txt: Updated.

Source/WebCore:

Added ReadableByteStreamController byobRequest attribute.

Added tests related to ReadableStreamBYOBRequest as a dedicated object
and also as an attribute of ReadableByteStreamController.

• CMakeLists.txt: Added new files.
• DerivedSources.cpp: Added new files.
• DerivedSources.make: Added new files.
• Modules/streams/ReadableByteStreamController.idl: Added byobRequest support.
• Modules/streams/ReadableByteStreamController.js: Added byobRequest support.
• Modules/streams/ReadableByteStreamInternals.js:

(privateInitializeReadableStreamBYOBRequest): Added constructor.

• Modules/streams/ReadableStreamBYOBRequest.idl: Added.
• Modules/streams/ReadableStreamBYOBRequest.js: Added.

(respond): Not implemented yet.
(respondWithNewView): Not implemented yet.
(view): Not implemented yet.

• WebCore.xcodeproj/project.pbxproj: Added new files.
• bindings/js/JSDOMGlobalObject.cpp: Added ReadableStreamBYOBRequest support.

(WebCore::JSDOMGlobalObject::addBuiltinGlobals):
(WebCore::JSDOMGlobalObject::finishCreation):
(WebCore::JSDOMGlobalObject::visitChildren):

• bindings/js/JSReadableStreamPrivateConstructors.cpp: Added ReadableStreamBYOBRequest support.

(WebCore::constructJSReadableByteStreamController):
(WebCore::constructJSReadableStreamBYOBRequest):
(WebCore::JSBuiltinReadableStreamBYOBRequestPrivateConstructor::initializeExecutable):
(WebCore::createReadableStreamBYOBRequestPrivateConstructor):

• bindings/js/JSReadableStreamPrivateConstructors.h: Added ReadableStreamBYOBRequest support.
• bindings/js/WebCoreBuiltinNames.h: Added ReadableStreamBYOBRequest support.

LayoutTests:

Added tests related to ReadableStreamBYOBRequest as a dedicated object
and also as an attribute of ReadableByteStreamController.

• streams/readable-byte-stream-controller.js: Updated with byobRequest attribute test.
• streams/readable-stream-byob-request-expected.txt: Added.
• streams/readable-stream-byob-request.html: Added.
• streams/readable-stream-byob-request.js: Added to cover tests related to ReadableStreamBYOBRequest.

10:16 AM Changeset in webkit [212655] by Ryan Haddad

• 13 edits
  2 moves
  2 adds
  1 delete in trunk/Source/WebKit2

Unreviewed, rolling out r212632.

This change appears to have caused API test failures.

Reverted changeset:

"Reduce platform ifdefs in WebKit2 custom protocols
implementation"
​https://bugs.webkit.org/show_bug.cgi?id=165028
​http://trac.webkit.org/changeset/212632

10:14 AM Changeset in webkit [212654] by aakash_jain@apple.com

• 2 edits in trunk/Source/WebKit/ios

Stop exporting WebPluginController from WebKit
​https://bugs.webkit.org/show_bug.cgi?id=168603

Reviewed by Alexey Proskuryakov.

• WebKit.iOS.exp:

10:00 AM Changeset in webkit [212653] by Ryan Haddad

• 2 edits in trunk/LayoutTests

Rebaseline fast/canvas/webgl/context-creation-attributes.html after r212650.

Unreviewed test gardening.

• fast/canvas/webgl/context-creation-attributes-expected.txt:

10:00 AM Changeset in webkit [212652] by pvollan@apple.com

• 10 edits in trunk/Source

[Win] Custom scale factor is not applied in all cases.
​https://bugs.webkit.org/show_bug.cgi?id=168117

Reviewed by Brent Fulgham.

We should not call the function deviceScaleFactorForWindow directly, since this
will return the system scale factor, and ignore the custom scale factor.

Source/WebCore:

• platform/graphics/ca/win/CACFLayerTreeHost.h:
• platform/graphics/ca/win/WKCACFViewLayerTreeHost.cpp:

(WebCore::WKCACFViewLayerTreeHost::initializeContext):
(WebCore::WKCACFViewLayerTreeHost::resize):
(WebCore::WKCACFViewLayerTreeHost::setScaleFactor):

• platform/graphics/ca/win/WKCACFViewLayerTreeHost.h:
• platform/win/GDIUtilities.h:

(WebCore::makeScaledPoint):

• platform/win/PlatformMouseEventWin.cpp:

(WebCore::positionForEvent):

• platform/win/PopupMenuWin.cpp:

(WebCore::PopupMenuWin::show):
(WebCore::PopupMenuWin::wndProc):

• platform/win/PopupMenuWin.h:

Source/WebKit/win:

• WebView.cpp:

(WebView::handleContextMenuEvent):
(WebView::handleMouseEvent):
(WebView::setAcceleratedCompositing):

9:54 AM Changeset in webkit [212651] by jonlee@apple.com

• 2 edits in trunk/Source/WebCore

Fix build error for iOS. Unreviewed.

• platform/graphics/GraphicsContext3D.cpp: Move stub function into !PLATFORM(COCOA) instead of !PLATFORM(MAC).

9:52 AM Changeset in webkit [212650] by jonlee@apple.com

• 15 edits
  2 adds in trunk

Unreviewed, rolling out r212639.

Have a fix for the linker error on iOS.

Reverted changeset:

"Unreviewed, rolling out r212637."
​https://bugs.webkit.org/show_bug.cgi?id=168595
​http://trac.webkit.org/changeset/212639

9:52 AM Changeset in webkit [212649] by Carlos Garcia Campos

• 1 copy in releases/WebKitGTK/webkit-2.15.90

WebKitGTK+ 2.15.90

9:45 AM Changeset in webkit [212648] by Carlos Garcia Campos

• 4 edits in releases/WebKitGTK/webkit-2.16

Unreviewed. Update OptionsGTK.cmake and NEWS for 2.15.90 release.

.:

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

Source/WebKit2:

• gtk/NEWS: Add release notes for 2.15.90.

9:29 AM Changeset in webkit [212647] by matthew_hanson@apple.com

• 6 edits
  6 adds in branches/safari-603-branch

Merge r212621. rdar://problem/30563318

9:29 AM Changeset in webkit [212646] by matthew_hanson@apple.com

• 13 edits
  1 add in branches/safari-603-branch/Source

Merge r212618. rdar://problem/30475767

9:29 AM Changeset in webkit [212645] by matthew_hanson@apple.com

• 5 edits
  2 adds in branches/safari-603-branch

Merge r212558. rdar://problem/30451563

9:24 AM Changeset in webkit [212644] by commit-queue@webkit.org

• 6 edits in trunk/Source

[WebRTC] Fix some memory leaks in libwebrtc binding code
​https://bugs.webkit.org/show_bug.cgi?id=168596

Patch by Youenn Fablet <​youenn@apple.com> on 2017-02-20
Reviewed by Eric Carlson.

Source/WebCore:

Covered by manual testing.

• Modules/mediastream/libwebrtc/LibWebRTCMediaEndpoint.cpp:

(WebCore::LibWebRTCMediaEndpoint::StatsCollector::StatsCollector): Using the libwebrtc way of creating such objects.

• Modules/mediastream/libwebrtc/LibWebRTCMediaEndpoint.h:

(WebCore::LibWebRTCMediaEndpoint::StatsCollector::create):

• platform/mediastream/libwebrtc/LibWebRTCProvider.cpp:

(WebCore::PeerConnectionFactoryAndThreads::OnMessage):

Source/WebKit2:

• NetworkProcess/webrtc/NetworkRTCProvider.cpp:

(WebKit::NetworkRTCProvider::OnMessage):

9:24 AM Changeset in webkit [212643] by aakash_jain@apple.com

• 2 edits in trunk/Source/WebKit/mac

Stop exporting DOMTokenList from WebKit
​https://bugs.webkit.org/show_bug.cgi?id=168544

Reviewed by Alexey Proskuryakov.

• WebKit.exp:

9:22 AM Changeset in webkit [212642] by ap@apple.com

• 7 edits
  1 add in trunk/Source

Clean up how WebKit exports _WebCreateFragment
​https://bugs.webkit.org/show_bug.cgi?id=168579

Reviewed by Dan Bernstein.

Source/WebCore:

• editing/Editor.h: Removed the declaration of _WebCreateFragment, SOFT_LINK doesn't need it.
• editing/cocoa/EditorCocoa.mm: Moved a FIXME comment here.

Source/WebKit:

• WebKit.xcodeproj/project.pbxproj: Added WebCreateFragmentInternal.h.

Source/WebKit/mac:

• WebCoreSupport/WebCreateFragmentInternal.h: Added a private header.
• WebCoreSupport/WebEditorClient.mm: Include the new header that declares _WebCreateFragment.

9:20 AM Changeset in webkit [212641] by aakash_jain@apple.com

• 2 edits in trunk/Source/WebKit/mac

Stop exporting DOMNodeFilter
​https://bugs.webkit.org/show_bug.cgi?id=168542

Reviewed by Alexey Proskuryakov.

• WebKit.exp:

9:18 AM Changeset in webkit [212640] by mark.lam@apple.com

• 2 edits in trunk/Source/JavaScriptCore

BytecodeGenerator should not iterate its m_controlFlowScopeStack using a pointer bump.
​https://bugs.webkit.org/show_bug.cgi?id=168585

Reviewed by Yusuke Suzuki.

This is because m_controlFlowScopeStack is a SegmentedVector, and entries for
consecutive indices in the vector are not guaranteed to be consecutive in memory
layout. Instead, we should be using indexing instead.

This issue was detected by the marathon.js test from
​https://bugs.webkit.org/show_bug.cgi?id=168580.

• bytecompiler/BytecodeGenerator.cpp:

(JSC::BytecodeGenerator::emitJumpViaFinallyIfNeeded):
(JSC::BytecodeGenerator::emitReturnViaFinallyIfNeeded):

9:12 AM Changeset in webkit [212639] by Ryan Haddad

• 15 edits
  2 deletes in trunk

Unreviewed, rolling out r212637.

This change broke iOS builds.

Reverted changeset:

"Add 'webglcontextchanged' WebGLContextEvent"
​https://bugs.webkit.org/show_bug.cgi?id=168595
​http://trac.webkit.org/changeset/212637

8:52 AM Changeset in webkit [212638] by commit-queue@webkit.org

• 15 edits
  4 deletes in trunk/Source

Remove code under USE(GRAPHICS_SURFACE)
​https://bugs.webkit.org/show_bug.cgi?id=168592

Patch by Carlos Garcia Campos <​cgarcia@igalia.com> on 2017-02-20
Reviewed by Michael Catanzaro.

This was only used by EFL port.

Source/WebCore:

• platform/TextureMapper.cmake:
• platform/graphics/opengl/GLPlatformSurface.cpp: Removed.
• platform/graphics/surfaces/GLTransportSurface.cpp: Removed.
• platform/graphics/surfaces/GLTransportSurface.h: Removed.
• platform/graphics/surfaces/GraphicsSurface.cpp: Removed.
• platform/graphics/surfaces/GraphicsSurface.h: Removed.
• platform/graphics/surfaces/GraphicsSurfaceToken.h: Removed.
• platform/graphics/surfaces/egl/EGLConfigSelector.cpp: Removed.
• platform/graphics/surfaces/egl/EGLConfigSelector.h: Removed.
• platform/graphics/surfaces/egl/EGLContext.cpp: Removed.
• platform/graphics/surfaces/egl/EGLContext.h: Removed.
• platform/graphics/surfaces/egl/EGLHelper.cpp: Removed.
• platform/graphics/surfaces/egl/EGLHelper.h: Removed.
• platform/graphics/surfaces/egl/EGLSurface.cpp: Removed.
• platform/graphics/surfaces/egl/EGLSurface.h: Removed.
• platform/graphics/surfaces/egl/EGLXSurface.cpp: Removed.
• platform/graphics/surfaces/egl/EGLXSurface.h: Removed.
• platform/graphics/surfaces/glx/GLXConfigSelector.h: Removed.
• platform/graphics/surfaces/glx/GLXContext.cpp: Removed.
• platform/graphics/surfaces/glx/GLXContext.h: Removed.
• platform/graphics/surfaces/glx/GLXSurface.cpp: Removed.
• platform/graphics/surfaces/glx/GLXSurface.h: Removed.
• platform/graphics/surfaces/glx/X11Helper.cpp: Removed.
• platform/graphics/surfaces/glx/X11Helper.h: Removed.
• platform/graphics/texmap/TextureMapperBackingStore.cpp:
• platform/graphics/texmap/TextureMapperBackingStore.h:
• platform/graphics/texmap/TextureMapperPlatformLayer.h:

(WebCore::TextureMapperPlatformLayer::setClient):

• platform/graphics/texmap/TextureMapperSurfaceBackingStore.cpp: Removed.
• platform/graphics/texmap/TextureMapperSurfaceBackingStore.h: Removed.
• platform/graphics/texmap/coordinated/CoordinatedGraphicsLayer.cpp:

(WebCore::CoordinatedGraphicsLayer::CoordinatedGraphicsLayer):
(WebCore::CoordinatedGraphicsLayer::setContentsNeedsDisplay):
(WebCore::CoordinatedGraphicsLayer::setContentsToPlatformLayer):
(WebCore::CoordinatedGraphicsLayer::syncPlatformLayer):

• platform/graphics/texmap/coordinated/CoordinatedGraphicsLayer.h:
• platform/graphics/texmap/coordinated/CoordinatedGraphicsState.h:

(WebCore::CoordinatedGraphicsLayerState::CoordinatedGraphicsLayerState):

Source/WebKit2:

• Shared/CoordinatedGraphics/CoordinatedGraphicsArgumentCoders.cpp:

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

• Shared/CoordinatedGraphics/CoordinatedGraphicsArgumentCoders.h:
• Shared/CoordinatedGraphics/CoordinatedGraphicsScene.cpp:

(WebKit::CoordinatedGraphicsScene::syncPlatformLayerIfNeeded):
(WebKit::CoordinatedGraphicsScene::deleteLayer):
(WebKit::CoordinatedGraphicsScene::assignImageBackingToLayer):
(WebKit::CoordinatedGraphicsScene::purgeGLResources):

• Shared/CoordinatedGraphics/CoordinatedGraphicsScene.h:
• Shared/CoordinatedGraphics/WebCoordinatedSurface.cpp:

(WebKit::WebCoordinatedSurface::Handle::encode):
(WebKit::WebCoordinatedSurface::Handle::decode):
(WebKit::WebCoordinatedSurface::create):
(WebKit::WebCoordinatedSurface::createGraphicsContext):
(WebKit::WebCoordinatedSurface::createHandle):
(WebKit::WebCoordinatedSurface::copyToTexture):

• Shared/CoordinatedGraphics/WebCoordinatedSurface.h:

8:15 AM WebKitGTK/2.16.x created by Carlos Garcia Campos

7:47 AM Changeset in webkit [212637] by dino@apple.com

• 15 edits
  2 adds in trunk

Add 'webglcontextchanged' WebGLContextEvent
​https://bugs.webkit.org/show_bug.cgi?id=168595
<rdar://problem/30604254>

Reviewed by Antoine Quint.

Source/WebCore:

Add a new event that is dispatched when the GraphicsContext3D
notices that the active GPU has changed.

Test: fast/canvas/webgl/webglcontextchangedevent.html

• dom/EventNames.h: Add 'webglcontextchanged'.
• html/canvas/WebGLRenderingContextBase.cpp:

(WebCore::WebGLRenderingContextBase::dispatchContextChangedEvent): Dispatch the new event.
(WebCore::WebGLRenderingContextBase::simulateContextChanged): Tell the GraphicsContext3D to
pretend there was a GPU switch. This is called from Internals for testing.

• html/canvas/WebGLRenderingContextBase.h:
• html/canvas/WebGLRenderingContextBase.idl: Export the interface so Internals can see it.

• platform/graphics/GraphicsContext3D.cpp:

(WebCore::GraphicsContext3D::simulateContextChanged): Tell the GraphicsContext3DManager
that it should pretend a GPU change has happened, which causes it to notify all the
active contexts.

• platform/graphics/GraphicsContext3D.h:

• platform/graphics/mac/GraphicsContext3DMac.mm:

(WebCore::GraphicsContext3DManager::updateAllContexts): Tell the GraphicsContext3Ds
to tell their WebGLRenderingContexts to dispatch an event.
(WebCore::GraphicsContext3D::simulateContextChanged): Fake the display change.

• platform/graphics/opengl/GraphicsContext3DOpenGLCommon.cpp:

(WebCore::GraphicsContext3D::dispatchContextChangedNotification): Tell the WebGLRenderingContext
that a change has occurred, and that it should dispatch an event. This is a layering
violation which will need to be fixed (although with other communication from GC3D back
to WebGL).

• testing/Internals.cpp:

(WebCore::Internals::simulateWebGLContextChanged):

• testing/Internals.h:
• testing/Internals.idl:

LayoutTests:

Add a new webglcontextchanged event. This test only
runs on Cocoa platforms at the moment, because it
needs GraphicsContext3D to keep track of all
active contexts in order to dispatch events.

• fast/canvas/webgl/webglcontextchangedevent-expected.txt: Added.
• fast/canvas/webgl/webglcontextchangedevent.html: Added.
• platform/gtk/TestExpectations:
• platform/win/TestExpectations:

7:30 AM Changeset in webkit [212636] by Manuel Rego Casasnovas

• 1 edit
  3 adds in trunk/Websites/webkit.org

Unreviewed, add CSS Grid Layout examples for a blog post.

• demos/css-grid/index.html: Added.
• demos/css-grid/responsive-grid.html: Added.

5:46 AM Changeset in webkit [212635] by Carlos Garcia Campos

• 1 copy in releases/WebKitGTK/webkit-2.16

Brach WebKitGTK+ for 2.16

4:30 AM Changeset in webkit [212634] by Carlos Garcia Campos

• 2 edits
  1 move in trunk/LayoutTests

Unreviewed GTK+ gardening. Fix tests reporting missing results.

• media/media-session-restrictions-expected.txt: Renamed from LayoutTests/platform/mac/media/media-session-restrictions-expected.txt.
• platform/gtk/TestExpectations:

4:28 AM Changeset in webkit [212633] by dino@apple.com

• 9 edits in trunk

Activate/deactivate high performance GPU when requested
​https://bugs.webkit.org/show_bug.cgi?id=168559
<rdar://problem/30592266>

Reviewed by Jon Lee.

Source/WebCore:

Respect the high-performance powerPreference for WebGL, by managing an
object that enables the high-performance GPU. If a WebGL context wants
high-performance, and it is visible, then a manager class in GraphicsContext3D
creates and retains the object, causing all the WebGL contexts to move GPUs.
If all the high-performance contexts are not visible, such as in a background tab,
then the manager will release the object, allowing the GPU to power down.

The swapping back from the high-performance GPU happens on a timer, to make
sure we don't churn between GPUs if the user is swapping between a lot of tabs,
or windows.

Unfortunately testing this change properly requires hardware with
multiple GPUs. I plan to write an API test that fakes most of the
system interaction, such as occluding the page. An API test might
also be able to verify if the system has more than one GPU. Otherwise
I'll have to plumb everything through Internals.

• html/canvas/WebGLRenderingContextBase.cpp:

(WebCore::isHighPerformanceContext): Helper to detect if the GraphicsContext3D actually
used high-performance mode.
(WebCore::WebGLRenderingContextBase::create): Add logging if we are
actually overriding a high-performance request.
(WebCore::WebGLRenderingContextBase::WebGLRenderingContextBase): If we are high-performance,
then register for activity state changes.
(WebCore::WebGLRenderingContextBase::addActivityStateChangeObserverIfNecessary):
(WebCore::WebGLRenderingContextBase::removeActivityStateChangeObserver):
(WebCore::WebGLRenderingContextBase::destroyGraphicsContext3D): Call removeActivityStateChangeObserver
as the GC3D is destroyed.
(WebCore::WebGLRenderingContextBase::maybeRestoreContext): If the context was
restored successfully, and came back in high-performance, then we need
to listen for activity state changes as usual.
(WebCore::WebGLRenderingContextBase::activityStateDidChange): If we changed visibility,
tell the GC3D.

• html/canvas/WebGLRenderingContextBase.h: Class inherits ActivityStateChangeObserver.

• page/Settings.in: No longer force low-power everywhere.

• platform/graphics/GraphicsContext3D.cpp:

(WebCore::GraphicsContext3D::setContextVisibility): Empty implementation for non-Mac.

• platform/graphics/GraphicsContext3D.h:

(WebCore::GraphicsContext3D::powerPreferenceUsedForCreation): Tells clients what power preference
was actually used during creation (e.g. a single GPU system will use default, even if
they requested high-performance).

• platform/graphics/mac/GraphicsContext3DMac.mm:

(WebCore::GraphicsContext3DManager::GraphicsContext3DManager): Helper class to
look after all GraphicsContext3Ds.
(WebCore::GraphicsContext3DManager::hasTooManyContexts): We have a limit on the
number of contexts we can keep alive at any one time.
(WebCore::manager): Helper to return the static instance.
(WebCore::displayWasReconfigured): Send a message to all the contexts.
(WebCore::GraphicsContext3DManager::addContext):
(WebCore::GraphicsContext3DManager::removeContext):
(WebCore::GraphicsContext3DManager::addContextRequiringHighPerformance):
(WebCore::GraphicsContext3DManager::removeContextRequiringHighPerformance):
(WebCore::GraphicsContext3DManager::updateHighPerformanceState): Check if the number
of contexts requiring high-performance means we need to enable/disable that GPU.
(WebCore::GraphicsContext3DManager::disableHighPerformanceGPUTimerFired): Releases our
object that keeps the high-performance GPU on.
(WebCore::GraphicsContext3DManager::recycleContextIfNecessary): Get rid of the first (oldest)
context. This code was in GC3D proper, but it made more sense here in the helper.
(WebCore::setPixelFormat): All contexts are created muxable now.
(WebCore::GraphicsContext3D::create): Use the manager.
(WebCore::GraphicsContext3D::GraphicsContext3D): Ditto.
(WebCore::GraphicsContext3D::~GraphicsContext3D): Add logging.
(WebCore::GraphicsContext3D::checkGPUStatusIfNecessary): Better logging.
(WebCore::GraphicsContext3D::updateCGLContext):
(WebCore::GraphicsContext3D::setContextVisibility): This is the responder to the
ActivityStateChanges in the WebGLRenderingContext.
(WebCore::activeContexts): Deleted.
(WebCore::addActiveContext): Deleted.
(WebCore::removeActiveContext): Deleted.

LayoutTests:

We no longer force low-power, so the WebGL canvas creation attributes now
return the value that was passed in.

• fast/canvas/webgl/context-creation-attributes-expected.txt:

4:28 AM Changeset in webkit [212632] by commit-queue@webkit.org

• 13 edits
  3 copies
  2 deletes in trunk/Source/WebKit2

Reduce platform ifdefs in WebKit2 custom protocols implementation
​https://bugs.webkit.org/show_bug.cgi?id=165028

Patch by Carlos Garcia Campos <​cgarcia@igalia.com> on 2017-02-20
Reviewed by Darin Adler.

Make mac implementation use the APICustomProtocolManagerClient to share more code with the soup based ports.

• CMakeLists.txt: Add CustomProtocolManagerProxy.cpp.
• PlatformEfl.cmake: Remove CustomProtocolManagerProxySoup.cpp.
• PlatformGTK.cmake: Ditto.
• PlatformMac.cmake: Remove CustomProtocolManagerProxyMac.mm.
• UIProcess/API/APICustomProtocolManagerClient.h:

(API::CustomProtocolManagerClient::startLoading): Make it void, since we were always returning true and the
returned value was ignored.

• UIProcess/API/efl/ewk_url_scheme_request.cpp:

(EwkUrlSchemeRequest::finish): Update to new CustomProtocolManagerProxy API.

• UIProcess/API/gtk/WebKitCustomProtocolManagerClient.cpp:
• UIProcess/API/gtk/WebKitURISchemeRequest.cpp:

(webkitURISchemeRequestReadCallback): Ditto.

• UIProcess/Cocoa/CustomProtocolManagerClient.h: Added.
• UIProcess/Cocoa/CustomProtocolManagerClient.mm: Added.

(-[WKCustomProtocolLoader initWithCustomProtocolManagerProxy:customProtocolID:request:]):
(-[WKCustomProtocolLoader dealloc]):
(-[WKCustomProtocolLoader connection:didFailWithError:]):
(-[WKCustomProtocolLoader connection:didReceiveResponse:]):
(-[WKCustomProtocolLoader connection:didReceiveData:]):
(-[WKCustomProtocolLoader connection:willSendRequest:redirectResponse:]):
(-[WKCustomProtocolLoader connectionDidFinishLoading:]):
(WebKit::CustomProtocolManagerClient::startLoading):
(WebKit::CustomProtocolManagerClient::stopLoading):
(WebKit::CustomProtocolManagerClient::invalidate):

• UIProcess/Cocoa/WebProcessPoolCocoa.mm:

(WebKit::WebProcessPool::platformInitialize): Set the CustomProtocolManagerClient.
(WebKit::WebProcessPool::platformInitializeNetworkProcess): Remove code that is now shared.

• UIProcess/Network/CustomProtocols/CustomProtocolManagerProxy.cpp: Added.

(WebKit::CustomProtocolManagerProxy::startLoading):
(WebKit::CustomProtocolManagerProxy::processDidClose):
(WebKit::CustomProtocolManagerProxy::wasRedirectedToRequest):
(WebKit::CustomProtocolManagerProxy::didReceiveResponse):

• UIProcess/Network/CustomProtocols/CustomProtocolManagerProxy.h:
• UIProcess/WebProcessPool.cpp:

(WebKit::globalURLSchemesWithCustomProtocolHandlers): Just moved here to use it from ensureNetworkProcess().
(WebKit::WebProcessPool::WebProcessPool): Remove soup ifdefs.
(WebKit::WebProcessPool::setCustomProtocolManagerClient): Moved from the soup implementation.
(WebKit::WebProcessPool::ensureNetworkProcess): Initialize custom protocols parameters.
(WebKit::WebProcessPool::registerSchemeForCustomProtocol): Register the protocol if not already registered globally.
(WebKit::WebProcessPool::unregisterSchemeForCustomProtocol): Remove soup ifdefs.

• UIProcess/WebProcessPool.h: Remove soup ifdefs.
• UIProcess/efl/RequestManagerClientEfl.cpp:
• UIProcess/soup/WebProcessPoolSoup.cpp:

(WebKit::WebProcessPool::platformInitializeNetworkProcess): Remove code that is now shared.

• WebKit2.xcodeproj/project.pbxproj:

4:23 AM Changeset in webkit [212631] by commit-queue@webkit.org

• 2 edits in trunk/Tools

Handle EINTR during webkitpy server process select()
​https://bugs.webkit.org/show_bug.cgi?id=168361

Patch by Florian Bruhin <​git@the-compiler.org> on 2017-02-20
Reviewed by Michael Catanzaro.

Otherwise, on my machine, resizing the terminal run-webkit-tests is running in
makes it die.

• Scripts/webkitpy/port/server_process.py:

(ServerProcess._wait_for_data_and_update_buffers_using_select):

2:03 AM Changeset in webkit [212630] by Carlos Garcia Campos

• 2 edits in trunk/Tools

Unreviewed. Fix GTK+ test /webkit2/WebKitWebView/default-menu.

It can happen that the window onload event is dispatched after the load finished is received in the UI
process. Move the script to select the text to the body instead of using a window onload handler.

• TestWebKitAPI/Tests/WebKit2Gtk/TestContextMenu.cpp:

(testContextMenuDefaultMenu):

2:01 AM Changeset in webkit [212629] by Manuel Rego Casasnovas

• 76 edits in trunk

[css-grid] Remove compilation flag ENABLE_CSS_GRID_LAYOUT
​https://bugs.webkit.org/show_bug.cgi?id=167693

Reviewed by Sergio Villar Senin.

.:

CSS Grid Layout has been enabled by default in r211415,
and Safari 10.1 is shipping it.
It seems like a good moment to remove the compilation flag.

• Source/cmake/OptionsMac.cmake:
• Source/cmake/OptionsWin.cmake:
• Source/cmake/WebKitFeatures.cmake:
• Source/cmake/tools/vsprops/FeatureDefines.props:
• Source/cmake/tools/vsprops/FeatureDefinesCairo.props:

Source/JavaScriptCore:

• Configurations/FeatureDefines.xcconfig:

Source/WebCore:

• Configurations/FeatureDefines.xcconfig:
• css/CSSComputedStyleDeclaration.cpp:

(WebCore::valueForGridPosition):
(WebCore::isLayoutDependent):
(WebCore::resolveJustifySelfAuto):
(WebCore::isImplicitlyInheritedGridOrFlexProperty):
(WebCore::valueForContentPositionAndDistributionWithOverflowAlignment):
(WebCore::ComputedStyleExtractor::propertyValue):

• css/CSSGridAutoRepeatValue.cpp:
• css/CSSGridAutoRepeatValue.h:
• css/CSSGridLineNamesValue.cpp:
• css/CSSGridLineNamesValue.h:
• css/CSSGridTemplateAreasValue.cpp:
• css/CSSGridTemplateAreasValue.h:
• css/CSSPrimitiveValueMappings.h:

(WebCore::CSSPrimitiveValue::CSSPrimitiveValue):

• css/CSSProperties.json:
• css/CSSValue.cpp:

(WebCore::CSSValue::equals):
(WebCore::CSSValue::cssText):
(WebCore::CSSValue::destroy):

• css/CSSValue.h:

(WebCore::CSSValue::isContentDistributionValue):
(WebCore::CSSValue::isGridLineNamesValue):

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

(WebCore::StyleBuilderConverter::convertGridAutoFlow):
(WebCore::StyleBuilderConverter::convertContentAlignmentData):

• css/StyleBuilderCustom.h:

(WebCore::StyleBuilderCustom::applyValueGridTemplateRows):

• css/StyleProperties.cpp:

(WebCore::StyleProperties::getPropertyValue):

• css/StyleResolver.cpp:

(WebCore::equivalentBlockDisplay):

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

(WebCore::CSSParserContext::CSSParserContext):
(WebCore::operator==):

• css/parser/CSSParserFastPaths.cpp:

(WebCore::isSimpleLengthPropertyID):
(WebCore::CSSParserFastPaths::isValidKeywordPropertyAndValue):
(WebCore::CSSParserFastPaths::isKeywordPropertyID):

• css/parser/CSSParserMode.h:

(WebCore::CSSParserContextHash::hash):

• css/parser/CSSPropertyParser.cpp:

(WebCore::consumeGridAutoFlow):
(WebCore::consumeGridTemplateAreas):
(WebCore::CSSPropertyParser::parseSingleValue):
(WebCore::CSSPropertyParser::consumeGridShorthand):
(WebCore::CSSPropertyParser::parseShorthand):

• dom/Document.cpp:
• dom/Document.h:
• dom/Position.cpp:

(WebCore::Position::isCandidate):

• page/RuntimeEnabledFeatures.h:

(WebCore::RuntimeEnabledFeatures::isCSSGridLayoutEnabled):

• rendering/Grid.cpp:
• rendering/Grid.h:
• rendering/RenderBlock.cpp:

(WebCore::RenderBlock::getFirstLetter):

• rendering/RenderBox.cpp:

(WebCore::RenderBox::~RenderBox):
(WebCore::RenderBox::styleDidChange):
(WebCore::RenderBox::clearExtraInlineAndBlockOffests):
(WebCore::RenderBox::containingBlockLogicalWidthForContent):
(WebCore::RenderBox::containingBlockLogicalHeightForContent):
(WebCore::RenderBox::perpendicularContainingBlockLogicalHeight):
(WebCore::RenderBox::computeLogicalWidthInRegion):
(WebCore::RenderBox::sizesLogicalWidthToFitContent):
(WebCore::RenderBox::computeLogicalHeight):
(WebCore::RenderBox::computePercentageLogicalHeight):
(WebCore::RenderBox::containingBlockLogicalWidthForPositioned):
(WebCore::RenderBox::containingBlockLogicalHeightForPositioned):
(WebCore::RenderBox::computePositionedLogicalWidth):
(WebCore::RenderBox::computePositionedLogicalHeight):
(WebCore::RenderBox::createsNewFormattingContext):
(WebCore::logicalWidthIsResolvable):
(WebCore::RenderBox::percentageLogicalHeightIsResolvableFromBlock):
(WebCore::RenderBox::hasDefiniteLogicalHeight):

• rendering/RenderBox.h:

(WebCore::RenderBox::isGridItem):

• rendering/RenderElement.cpp:

(WebCore::RenderElement::createFor):

• rendering/RenderFlexibleBox.cpp:

(WebCore::RenderFlexibleBox::alignChildren):

• rendering/RenderGrid.cpp:
• rendering/RenderGrid.h:
• rendering/RenderObject.h:

(WebCore::RenderObject::isRenderGrid):

• rendering/style/GridArea.h:
• rendering/style/GridLength.h:
• rendering/style/GridPosition.h:
• rendering/style/GridPositionsResolver.cpp:
• rendering/style/GridPositionsResolver.h:
• rendering/style/GridTrackSize.h:
• rendering/style/RenderStyle.cpp:

(WebCore::RenderStyle::isCSSGridLayoutEnabled):
(WebCore::RenderStyle::changeRequiresLayout):

• rendering/style/RenderStyle.h:

(WebCore::RenderStyle::gridItemRowEnd):
(WebCore::RenderStyle::setGridRowGap):
(WebCore::RenderStyle::initialGridItemRowEnd):
(WebCore::RenderStyle::isDisplayReplacedType):
(WebCore::RenderStyle::isDisplayGridBox):

• rendering/style/RenderStyleConstants.h:
• rendering/style/StyleGridData.cpp:
• rendering/style/StyleGridData.h:
• rendering/style/StyleGridItemData.cpp:
• rendering/style/StyleGridItemData.h:
• rendering/style/StyleRareNonInheritedData.cpp:

(WebCore::StyleRareNonInheritedData::StyleRareNonInheritedData):
(WebCore::StyleRareNonInheritedData::operator==):

• rendering/style/StyleRareNonInheritedData.h:
• testing/InternalSettings.cpp:

(WebCore::InternalSettings::Backup::Backup):
(WebCore::InternalSettings::Backup::restoreTo):
(WebCore::InternalSettings::setCSSGridLayoutEnabled):

Source/WebCore/PAL:

• Configurations/FeatureDefines.xcconfig:

Source/WebKit/mac:

• Configurations/FeatureDefines.xcconfig:
• WebView/WebPreferences.mm:

(+[WebPreferences initialize]):

• WebView/WebView.mm:

(-[WebView _preferencesChanged:]):

Source/WebKit2:

• Configurations/FeatureDefines.xcconfig:
• WebProcess/InjectedBundle/InjectedBundle.cpp:

(WebKit::InjectedBundle::overrideBoolPreferenceForTestRunner):

• WebProcess/WebPage/WebPage.cpp:

(WebKit::WebPage::updatePreferences):

Tools:

• Scripts/webkitperl/FeatureList.pm:
• TestWebKitAPI/Configurations/FeatureDefines.xcconfig:
• TestWebKitAPI/Tests/WebCore/CSSParser.cpp:

(TestWebKitAPI::computeNumberOfTracks):
(TestWebKitAPI::TEST):

• TestWebKitAPI/Tests/WebCore/GridPosition.cpp:

(TestWebKitAPI::TEST):

1:13 AM Changeset in webkit [212628] by Carlos Garcia Campos

• 2 edits in trunk/LayoutTests

Unreviewed GTK+ gardening. Mark fast/events/message-port-postMessage-recursive.html as timing out.

• platform/gtk/TestExpectations:

1:09 AM Changeset in webkit [212627] by Carlos Garcia Campos

• 2 edits in trunk/Source/WTF

Unreviewed, rolling out r212622.

Caused several test failures

Reverted changeset:

"[GTK] Test fast/events/message-port-postMessage-
recursive.html times out"
​https://bugs.webkit.org/show_bug.cgi?id=168570
​http://trac.webkit.org/changeset/212622

12:30 AM Changeset in webkit [212626] by Carlos Garcia Campos

• 14 edits in trunk/Source

REGRESSION(r212345): [GTK] ASSERT(cookieChangeCallbackMap().contains(jar)) failed in WebCore::stopObservingCookieChanges:54
​https://bugs.webkit.org/show_bug.cgi?id=168375

Reviewed by Alex Christensen.

Source/WebCore:

Simplify the way cookie storage is handled in NetworkStorageSessionSoup. Now we always keep a reference to the
cookie jar, no matter if we have a session or not, that is always kept in sync with the one actually used by the
session. We always connect to the changed signal of the current cookie jar and notify the observer if it has been
set.

• platform/network/NetworkStorageSession.h:
• platform/network/soup/CookieStorageSoup.cpp:

(WebCore::startObservingCookieChanges): Remove this implementation and add an ASSERT to ensure we don't use this anymore.
(WebCore::stopObservingCookieChanges): Ditto.
(WebCore::cookieChangeCallbackMap): Deleted.
(WebCore::soupCookiesChanged): Deleted.

• platform/network/soup/NetworkStorageSessionSoup.cpp:

(WebCore::NetworkStorageSession::NetworkStorageSession): Initialize the cookie storage.
(WebCore::NetworkStorageSession::~NetworkStorageSession): Disconnect the changed signal.
(WebCore::NetworkStorageSession::getOrCreateSoupNetworkSession): Always pass the current cookie storage.
(WebCore::NetworkStorageSession::cookiesDidChange): Notify the observer.
(WebCore::NetworkStorageSession::cookieStorage): Return the current cookie storage.
(WebCore::NetworkStorageSession::setCookieStorage): Update the current cookie storage and update the session if
needed too. We always have a valid cookie storage so if nullptr is given we create a new one.
(WebCore::NetworkStorageSession::setCookieObserverHandler): Set a cookie observer handler.

Source/WebKit2:

Simplify the cookies change notification. The current approach of the UI process telling the network process
when to start and stop observing cookie changes is quite fragile and error prone. In the GTK+ port we always
observe cookies, so our cookie manager always calls start in the constructor and stop in the destructor.
Everything would be a lot simpler if the network process always monitored cookies.

• NetworkProcess/soup/NetworkSessionSoup.cpp:

(WebKit::NetworkSessionSoup::NetworkSessionSoup): Set cookies observer handler for the session.
(WebKit::NetworkSessionSoup::~NetworkSessionSoup): Remove the cookies observer handler.

• UIProcess/API/gtk/WebKitCookieManager.cpp:

(_WebKitCookieManagerPrivate::~_WebKitCookieManagerPrivate): Remove the cookies observer callback.
(webkit_cookie_manager_class_init): Remove dispose implementation that is no longer needed.
(webkitCookieManagerCreate): Set the cookies observer callback.
(webkit_cookie_manager_set_persistent_storage): Remove code to start/stop observing cookies when the jar
changes, the network process deals with that now.
(webkitCookieManagerDispose): Deleted.

• UIProcess/WebCookieManagerProxy.cpp:

(WebKit::WebCookieManagerProxy::startObservingCookieChanges): Remove the optional callback method in favor of an
explicit method to the set the callback that doesn't send the StartObservingCookieChanges message to the network
process.
(WebKit::WebCookieManagerProxy::stopObservingCookieChanges): Do not change the callback map here.
(WebKit::WebCookieManagerProxy::setCookieObserverCallback): Add the callback to the map or remove the session if
nullptr is given.

• UIProcess/WebCookieManagerProxy.h:
• WebProcess/Cookies/WebCookieManager.cpp:

(WebKit::WebCookieManager::notifyCookiesDidChange): Send CookiesDidChange message to the UI process.
(WebKit::WebCookieManager::startObservingCookieChanges): Use notifyCookiesDidChange().

• WebProcess/Cookies/WebCookieManager.h:
• WebProcess/Cookies/soup/WebCookieManagerSoup.cpp:

(WebKit::WebCookieManager::platformSetHTTPCookieAcceptPolicy): Set the policy on all sessions. Code moved from
WebFrameNetworkingContext since it doesn't belong there.
(WebKit::WebCookieManager::setCookiePersistentStorage): Use new method to set the cookie storage in the
NetworkStorageSession.

• WebProcess/WebCoreSupport/soup/WebFrameNetworkingContext.cpp:

(WebKit::WebFrameNetworkingContext::setCookieAcceptPolicyForAllContexts): Deleted.

• WebProcess/WebCoreSupport/soup/WebFrameNetworkingContext.h:

Feb 19, 2017:

10:36 PM Changeset in webkit [212625] by Chris Dumez

• 8 edits
  6 adds in trunk

onbeforeunload event return value coercion is not per-spec
​https://bugs.webkit.org/show_bug.cgi?id=168382

Reviewed by Darin Adler.

LayoutTests/imported/w3c:

Import test coverage from web-platform-tests. We were failing half the checks
before this patch.

• resources/resource-files.json:
• web-platform-tests/html/browsers/browsing-the-web/unloading-documents/MANIFEST:
• web-platform-tests/html/browsers/browsing-the-web/unloading-documents/beforeunload-canceling-1.html: Added.
• web-platform-tests/html/browsers/browsing-the-web/unloading-documents/beforeunload-canceling-expected.txt: Added.
• web-platform-tests/html/browsers/browsing-the-web/unloading-documents/beforeunload-canceling.html: Added.
• web-platform-tests/html/browsers/browsing-the-web/unloading-documents/w3c-import.log:

Source/WebCore:

Update handling of value returned by onbeforeunload event listeners
to match Firefox and the specification:

• ​https://html.spec.whatwg.org/#the-event-handler-processing-algorithm (step 4)

Namely, the following changes were made:

• Only set the event's returnValue attribute to the returned value if the attribute value is the empty string (so as to not override the attribute value if it has explicitly been set by JS).
• Cancel the event when the return value is not null by calling preventDefault().

Additionally, the following changes were made:

• Ask the user to confirm the navigation if the event was canceled, not just if the returnValue attribute was set to a non-empty string.

as per:

• ​https://html.spec.whatwg.org/#prompt-to-unload-a-document (step 8)

Tests: fast/events/before-unload-return-string-conversion.html

imported/w3c/web-platform-tests/html/browsers/browsing-the-web/unloading-documents/beforeunload-canceling.html

• bindings/js/JSEventListener.cpp:

(WebCore::handleBeforeUnloadEventReturnValue):
(WebCore::JSEventListener::handleEvent):

• loader/FrameLoader.cpp:

(WebCore::shouldAskForNavigationConfirmation):
(WebCore::FrameLoader::dispatchBeforeUnloadEvent):

LayoutTests:

Add test case to check that the value returned by a beforeunload event handler
is already converted to a string, even if the returnValue attribute is also
set on the BeforeUnloadEvent. The existing code did not handle this properly
and it has been fixed in this patch.

• fast/events/before-unload-return-string-conversion-expected.txt: Added.
• fast/events/before-unload-return-string-conversion.html: Added.
• fast/events/resources/before-unload-return-string-conversion-frame.html: Added.

10:34 PM Changeset in webkit [212624] by commit-queue@webkit.org

• 3 edits in trunk/Source/JavaScriptCore

Unreviewed, rolling out r212472.
​https://bugs.webkit.org/show_bug.cgi?id=168584

Broke CLoop builds when r212466 was rolled out in r212616
(Requested by rniwa on #webkit).

Reverted changeset:

"Unreviewed, fix cloop build."
​http://trac.webkit.org/changeset/212472

10:30 PM Changeset in webkit [212623] by Carlos Garcia Campos

• 2 edits in trunk/Source/WebCore

[SOUP] Call SoupNetworkSession::setShouldIgnoreTLSErrors when testRunner.setAllowsAnySSLCertificate() is called
​https://bugs.webkit.org/show_bug.cgi?id=168571

Reviewed by Michael Catanzaro.

Fixes: http/tests/ssl/upgrade-origin-usage.html

• testing/InternalSettings.cpp:

(WebCore::InternalSettings::setAllowsAnySSLCertificate):

10:29 PM Changeset in webkit [212622] by Carlos Garcia Campos

• 2 edits in trunk/Source/WTF

[GTK] Test fast/events/message-port-postMessage-recursive.html times out
​https://bugs.webkit.org/show_bug.cgi?id=168570

Reviewed by Michael Catanzaro.

This has recently been added and the patch is good. It's just revealing a problem with our timers. The test is
posting a message recursively, and also starts a timeout timer to finish the test. The timeout timer is never
fired for us, because WebCore timers have lower priority than the one used by
postMessage. ScriptExecutionContext uses Document::postTask, that uses scheduleOnMainThread, that uses
RunLoop::dispatch(). We are not setting any priority for the timer used by RunLoop::dispatch, so it's using the
default. RunLoop::dispatch is normally used to schedule tasks between threads, or just to ensure something is
run in a different run loop iteration, but in general nothing urgent as a graphics redraw or something like
that. It's quite common to use g_idle_add to schedule tasks between threads, so I think it makes sense to use
G_PRIORITY_DEFAULT_IDLE for the RunLoop timer.

• wtf/glib/RunLoopGLib.cpp:

(WTF::RunLoop::RunLoop):

10:22 PM Changeset in webkit [212621] by rniwa@webkit.org

• 6 edits
  6 adds in trunk

REGRESSION(r212218): Assertion failures in and after parserRemoveChild
​https://bugs.webkit.org/show_bug.cgi?id=168458

Reviewed by Antti Koivisto.

Source/WebCore:

The bug was caused by parserRemoveChild not preceeding to remove oldChild even when
oldChild had been inserted elsewhere during unload evnets of the disconnected frames.
Fixed the bug by checking this condition and exiting early.

Also fixed various callers of parserRemoveChild to not call parserAppendChild when
the removed node had already been inserted elsewhere by scripts.

Tests: fast/parser/adoption-agency-unload-iframe-3.html

fast/parser/adoption-agency-unload-iframe-4.html
fast/parser/xml-error-unload-iframe.html

• dom/ContainerNode.cpp:

(WebCore::ContainerNode::parserRemoveChild): Exit early when the node had been
inserted elsewhere while firing unload events. Also moved the call to
notifyRemovePendingSheetIfNeeded outside NoEventDispatchAssertion since it can
synchrnously fire a focus event.
(WebCore::ContainerNode::parserAppendChild): Moved adoptNode call to inside
NoEventDispatchAssertion since adoptNode call here should never mutate DOM.

• html/parser/HTMLConstructionSite.cpp:

(WebCore::executeReparentTask): Added an early exit when the node had already been
inserted elsewhere.
(WebCore::executeInsertAlreadyParsedChildTask): Ditto.

• xml/XMLErrors.cpp:

(WebCore::XMLErrors::insertErrorMessageBlock): Ditto.

• xml/parser/XMLDocumentParser.cpp:

(WebCore::XMLDocumentParser::end): Fixed a crash unveiled by one of the test cases.
Exit early when insertErrorMessageBlock detached the parser (by author scripts).
(WebCore::XMLDocumentParser::finish): Keep the parser alive until we exit.

LayoutTests:

Add tests to make sure parserAppendChild aren't called when a node removed by parserRemoveChild
had already been been inserted elsewhere by scripts.

• fast/parser/adoption-agency-unload-iframe-3-expected.txt: Added.
• fast/parser/adoption-agency-unload-iframe-3.html: Added.
• fast/parser/adoption-agency-unload-iframe-4-expected.txt: Added.
• fast/parser/adoption-agency-unload-iframe-4.html: Added.
• fast/parser/xml-error-unload-iframe-expected.txt: Added.
• fast/parser/xml-error-unload-iframe.html: Added.

7:59 PM Changeset in webkit [212620] by mark.lam@apple.com

• 3 edits in trunk/Source/JavaScriptCore

functionTestWasmModuleFunctions() should use a MarkedArgumentBuffer for storing args instead of a Vector.
​https://bugs.webkit.org/show_bug.cgi?id=168574

Reviewed by Filip Pizlo.

• jsc.cpp:

(callWasmFunction):
(functionTestWasmModuleFunctions):

• runtime/ArgList.h:

6:38 PM Changeset in webkit [212619] by Chris Dumez

• 11 edits
  2 adds in trunk/Source/WebKit2

[WK2] Support termination of unresponsive WebContent processes that are in the background
​https://bugs.webkit.org/show_bug.cgi?id=168479
<rdar://problem/30558745>

Reviewed by Antti Koivisto.

Support termination of unresponsive WebContent processes that are in the background.
This protects us against cases where a background tab is unresponsive and has high CPU
usage, thus draining the battery without the user knowing about it (e.g.
<rdar://problem/29808005>).

The feature works as follows:

• If a WebProcessProxy has pages but none of them are visible, then it will check regularly if its associated WebContent process is still responsive.
• Every time we find that the WebContent process is responsive, we exponentially back off the timer to avoid waking up the process too often. The timer initially uses a 20 seconds interval and can back off up to 8 hours.
• If the WebContent process is unresponsive, then we terminate it. No notification is shown to the user and the termination is not immediately observable since the tab is not visible and its title remains. The tab is not immediately reloaded so as to avoid getting back into a bad state in the background (and have a yo-yo effect).
• If the user switches to a tab that was previously killed for being unresponsive while in the background, we will then reload the tab in a fresh WebContent process. The crash banner will only be shown if the tab has ever been visible. If it wasn't ever visible (e.g. session restore case), then it will look as if the tab loads for the first time when the user switches to it.

The functionality is disabled by default and can be enabled by the client via the new
WKContextSetUnresponsiveBackgroundProcessesTerminationEnabled() private API.

• CMakeLists.txt:
• UIProcess/API/APIProcessPoolConfiguration.cpp:

(API::ProcessPoolConfiguration::copy):

• UIProcess/API/APIProcessPoolConfiguration.h:
• UIProcess/API/C/WKContext.cpp:

(WKContextSetUnresponsiveBackgroundProcessesTerminationEnabled):

• UIProcess/API/C/WKContextPrivate.h:
• UIProcess/UnresponsiveWebProcessTerminator.cpp: Added.

(WebKit::UnresponsiveWebProcessTerminator::UnresponsiveWebProcessTerminator):
(WebKit::UnresponsiveWebProcessTerminator::updateState):
(WebKit::pagesCopy):
(WebKit::UnresponsiveWebProcessTerminator::timerFired):

• UIProcess/UnresponsiveWebProcessTerminator.h: Added.

(WebKit::UnresponsiveWebProcessTerminator::shouldBeActive):

• UIProcess/WebPageProxy.cpp:

(WebKit::WebPageProxy::reattachToWebProcess):
(WebKit::WebPageProxy::reloadAfterBeingKilledInBackground):
(WebKit::WebPageProxy::dispatchActivityStateChange):
(WebKit::WebPageProxy::terminateProcess):
(WebKit::WebPageProxy::resetStateAfterProcessExited):

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

(WebKit::WebProcessProxy::WebProcessProxy):
(WebKit::m_backgroundResponsivenessTimer):
(WebKit::WebProcessProxy::createWebPage):
(WebKit::WebProcessProxy::addExistingWebPage):
(WebKit::WebProcessProxy::removeWebPage):
(WebKit::WebProcessProxy::updateBackgroundResponsivenessTimer):

• UIProcess/WebProcessProxy.h:

(WebKit::WebProcessProxy::visiblePageCount):

• WebKit2.xcodeproj/project.pbxproj:

5:46 PM Changeset in webkit [212618] by mark.lam@apple.com

• 13 edits
  1 add in trunk/Source

CachedCall should let GC know to keep its arguments alive.
​https://bugs.webkit.org/show_bug.cgi?id=168567
<rdar://problem/30475767>

Reviewed by Saam Barati.

Source/JavaScriptCore:

We fix this by having CachedCall use a MarkedArgumentBuffer to store its
arguments instead of a Vector.

Also declared CachedCall, MarkedArgumentBuffer, and ProtoCallFrame as
WTF_FORBID_HEAP_ALLOCATION because they rely on being stack allocated for
correctness.

• interpreter/CachedCall.h:

(JSC::CachedCall::CachedCall):
(JSC::CachedCall::call):
(JSC::CachedCall::clearArguments):
(JSC::CachedCall::appendArgument):
(JSC::CachedCall::setArgument): Deleted.

• interpreter/CallFrame.h:

(JSC::ExecState::emptyList):

• interpreter/Interpreter.cpp:

(JSC::Interpreter::prepareForRepeatCall):

• interpreter/Interpreter.h:
• interpreter/ProtoCallFrame.h:
• runtime/ArgList.cpp:

(JSC::MarkedArgumentBuffer::expandCapacity):

• runtime/ArgList.h:

(JSC::MarkedArgumentBuffer::ensureCapacity):

• runtime/StringPrototype.cpp:

(JSC::replaceUsingRegExpSearch):

• runtime/VM.cpp:

(JSC::VM::VM):

• runtime/VM.h:

Source/WTF:

Added a WTF_FORBID_HEAP_ALLOCATION that will cause a compilation failure if
a class declared with it is malloced.

While this doesn't prevent that class declared WTF_FORBID_HEAP_ALLOCATION from
being embedded in another class that is heap allocated, it does at minimum
document the intent and gives the users of this class a chance to do the
right thing.

• WTF.xcodeproj/project.pbxproj:
• wtf/ForbidHeapAllocation.h: Added.

4:46 PM Changeset in webkit [212617] by Chris Dumez

• 5 edits
  4 adds in trunk

<input type=color readonly> doesn't act per spec
​https://bugs.webkit.org/show_bug.cgi?id=168421
<rdar://problem/30593185>

Reviewed by Ryosuke Niwa.

Source/WebCore:

The 'readonly' attribute should not apply to <input type=color> as per the
HTML specification:

• ​https://html.spec.whatwg.org/#the-input-element:attr-input-readonly-3

Chrome / Firefox and Edge already behave as per the specification.

Tests: fast/forms/color/input-color-disabled.html

fast/forms/color/input-color-readonly.html

• html/ColorInputType.cpp:

(WebCore::ColorInputType::handleDOMActivateEvent):
(WebCore::ColorInputType::didChooseColor):

LayoutTests:

Add layout test coverage.

• fast/forms/color/input-color-disabled-expected.txt: Added.
• fast/forms/color/input-color-disabled.html: Added.
• fast/forms/color/input-color-readonly-expected.txt: Added.
• fast/forms/color/input-color-readonly.html: Added.

2:07 PM Changeset in webkit [212616] by commit-queue@webkit.org

• 45 edits
  2 moves
  1 add
  16 deletes in trunk

Unreviewed, rolling out r212466.
​https://bugs.webkit.org/show_bug.cgi?id=168577

causes crashes on AArch64 on linux, maybe it's causing crashes
on iOS too (Requested by pizlo on #webkit).

Reverted changeset:

"The collector thread should only start when the mutator
doesn't have heap access"
​https://bugs.webkit.org/show_bug.cgi?id=167737
​http://trac.webkit.org/changeset/212466

8:24 AM Changeset in webkit [212615] by Alan Bujtas

• 12 edits in trunk/Source/WebCore

Simple line layout: Implement positionForPoint.
​https://bugs.webkit.org/show_bug.cgi?id=168565
<rdar://problem/30593370>

Reviewed by Antti Koivisto.

With this patch, now we don't need to switch over to line box tree anymore when positionForPoint()
is called on a single RenderText content (which is the majority of the simple line content).
This patch also adds a new positionForPoint method which returns a Position object instead of VisiblePosition.
Some of the positionForPoint() callsites are only interested in the Position object so constructing a VisiblePosition
(which could potentially kick off a layout) is a waste. At this point the non-simple line layout codepath still constructs
a VisiblePosition and that should be addresses as part of webkit.org/b/168566.

Covered by existing tests.

• dom/Document.cpp:

(WebCore::Document::caretRangeFromPoint):

• rendering/RenderBlockFlow.cpp:

(WebCore::RenderBlockFlow::positionForPoint):

• rendering/RenderBlockFlow.h:
• rendering/RenderObject.cpp:

(WebCore::RenderObject::positionForPoint):

• rendering/RenderObject.h:
• rendering/RenderText.cpp:

(WebCore::RenderText::positionForPoint):

• rendering/RenderText.h:
• rendering/SimpleLineLayoutFunctions.cpp:

(WebCore::SimpleLineLayout::positionForPoint):

• rendering/SimpleLineLayoutFunctions.h:
• rendering/SimpleLineLayoutResolver.cpp:

(WebCore::SimpleLineLayout::RunResolver::runForPoint):

• rendering/SimpleLineLayoutResolver.h:

(WebCore::SimpleLineLayout::RunResolver::Run::logicalLeft):
(WebCore::SimpleLineLayout::RunResolver::Run::logicalRight):

4:12 AM Changeset in webkit [212614] by Antti Koivisto

• 15 edits in trunk/Source/WebCore

Execute pending scripts asynchronously after stylesheet loads complete
​https://bugs.webkit.org/show_bug.cgi?id=168367
rdar://problem/30561379

Reviewed by Andreas Kling.

The current synchronous execution is fragile and creates various problems.

• css/StyleSheetContents.cpp:

(WebCore::StyleSheetContents::checkLoaded):

• dom/ContainerNode.cpp:

(WebCore::ContainerNode::takeAllChildrenFrom):
(WebCore::ContainerNode::notifyChildInserted):
(WebCore::ContainerNode::removeChild):
(WebCore::ContainerNode::parserRemoveChild):
(WebCore::ContainerNode::removeChildren):

Remove various places where we would trigger delayed synchronous execution.

• dom/Document.cpp:

(WebCore::Document::Document):
(WebCore::Document::recalcStyle):

Trigger scroll to anchor at the end of style resolution instead of when style sheet load completes.

(WebCore::Document::didRemoveAllPendingStylesheet):

Call asynchronous script execution function.

• dom/Document.h:

(WebCore::Document::setNeedsNotifyRemoveAllPendingStylesheet): Deleted.
(WebCore::Document::notifyRemovePendingSheetIfNeeded): Deleted.

• dom/ScriptableDocumentParser.cpp:

(WebCore::ScriptableDocumentParser::ScriptableDocumentParser):
(WebCore::ScriptableDocumentParser::executeScriptsWaitingForStylesheetsSoon):
(WebCore::ScriptableDocumentParser::scriptsWaitingForStylesheetsExecutionTimerFired):

Add a timer for executing pending scripts.

(WebCore::ScriptableDocumentParser::detach):

• dom/ScriptableDocumentParser.h:

(WebCore::ScriptableDocumentParser::executeScriptsWaitingForStylesheets):

• html/HTMLLinkElement.cpp:

(WebCore::HTMLLinkElement::removedFrom):
(WebCore::HTMLLinkElement::removePendingSheet):

• html/HTMLLinkElement.h:
• html/parser/HTMLDocumentParser.cpp:

(WebCore::HTMLDocumentParser::detach):

• loader/DocumentLoader.cpp:

(WebCore::DocumentLoader::isLoadingInAPISense):

Stay in loading state if we have a pending script. This matches existing behavior.

• style/StyleScope.cpp:

(WebCore::Style::Scope::removePendingSheet):

• style/StyleScope.h:

3:54 AM Changeset in webkit [212613] by Carlos Garcia Campos

• 2 edits in trunk/LayoutTests

Unreviewed GTK+ gardening. Rebaseline fast/visual-viewport/rubberbanding-viewport-rects-extended-background.html again.

This matches what bots expect now.

• platform/gtk/fast/visual-viewport/rubberbanding-viewport-rects-extended-background-expected.txt:

3:39 AM Changeset in webkit [212612] by Carlos Garcia Campos

• 1 edit
  1 add in trunk/LayoutTests

Unreviewed GTK+ gardening. Rebaseline http/tests/loading/server-redirect-for-provisional-load-caching.html.

I think we are just setting the FirstPartyForCookies to our request a bit earlier when redirecting.

• platform/gtk/http/tests/loading/server-redirect-for-provisional-load-caching-expected.txt: Added.

3:14 AM Changeset in webkit [212611] by Carlos Garcia Campos

• 2 edits
  2 adds in trunk/LayoutTests

Unreviewed GTK+ gardening. Update expectations of a few more tests and rebaseline fast/visual-viewport/rubberbanding-viewport-rects-extended-background.html.

• platform/gtk/TestExpectations:
• platform/gtk/fast/visual-viewport/rubberbanding-viewport-rects-extended-background-expected.txt: Added.