Timeline

Jan 25, 2020:

10:33 PM Changeset in webkit [255128] by Darin Adler

• 3 edits in trunk/Source/WebCore

Tighten up some of the drag state machine logic
​https://bugs.webkit.org/show_bug.cgi?id=206798

Reviewed by Wenson Hsieh.

• page/EventHandler.h: Added shouldDispatchEventsToDragSourceElement function and renamed

the existing dispatchDragSrcEvent function to dispatchEventToDragSourceElement.

• page/EventHandler.cpp:

(WebCore::EventHandler::updateDragAndDrop): Call the renamed dispatchEventToDragSourceElement
unconditionally, since it now has the smarts to only dispatch an event when we are supposed to.
(WebCore::EventHandler::cancelDragAndDrop): Ditto.
(WebCore::EventHandler::dragSourceEndedAt): Call the new shouldDispatchEventsToDragSourceElement
function because of the null check it does for dataTransfer, before calling setDestinationOperation
on the dataTransfer, since there isn't an obvious ironclad guarantee we might be here without an
actual drag fully in process and a dataTransfer object allocated. Also call the renamed
dispatchEventToDragSourceElement by its new name.
(WebCore::EventHandler::shouldDispatchDragSourceEvents): Added. Checks thre three conditions that
affect whether we should dispatch events to the drag source. First that there is a drag source.
Second that there is a dataTransfer object, indicating that we got far enough in the logic to
actually start a drag. Third that shouldDispatchEvents is true, indicating this is the type of
drag that should be visible to the website content and so events should be dispatched.
(WebCore::EventHandler::dispatchEventToDragSourceElement): Call shouldDispatchDragSourceEvents
before dispatching the event, so that callers don't all have to do that check.

9:43 PM Changeset in webkit [255127] by beidson@apple.com

• 20 edits in trunk/Source

Make ContentWorlds be identified by an ObjectIdentifier instead of a uint64_t
​https://bugs.webkit.org/show_bug.cgi?id=206784

Reviewed by Alex Christensen.

Source/WebKit:

Refactor: No behavior change.

• Scripts/webkit/messages.py:

• Shared/AuxiliaryProcess.cpp:

(WebKit::AuxiliaryProcess::initialize):

• Shared/ContentWorldShared.h:

(WebKit::pageContentWorldIdentifier):

• Shared/WebPageCreationParameters.cpp:

(WebKit::WebPageCreationParameters::decode):

• Shared/WebPageCreationParameters.h:

• Shared/WebUserContentControllerDataTypes.cpp:

(WebKit::WebUserScriptData::decode):
(WebKit::WebUserStyleSheetData::decode):
(WebKit::WebScriptMessageHandlerData::decode):

• Shared/WebUserContentControllerDataTypes.h:

• UIProcess/API/APIContentWorld.cpp:

(API::ContentWorld::pageContentWorld):
(API::ContentWorld::defaultClientWorld):
(API::ContentWorld::ContentWorld):
(API::ContentWorldBase::generateIdentifier): Deleted.

• UIProcess/API/APIContentWorld.h:

(API::ContentWorldBase::identifier const):
(API::ContentWorldBase::worldData const):
(API::ContentWorldBase::ContentWorldBase):

• UIProcess/API/APIUserContentWorld.cpp:

(API::UserContentWorld::UserContentWorld):

• UIProcess/UserContent/WebUserContentControllerProxy.cpp:

(WebKit::WebUserContentControllerProxy::removeUserContentWorldUses):
(WebKit::WebUserContentControllerProxy::removeAllUserScripts):
(WebKit::WebUserContentControllerProxy::removeAllUserStyleSheets):

• WebProcess/UserContent/WebUserContentController.cpp:

(WebKit::worldMap):
(WebKit::WebUserContentController::worldForIdentifier):
(WebKit::WebUserContentController::addUserContentWorld):
(WebKit::WebUserContentController::addUserContentWorlds):
(WebKit::WebUserContentController::removeUserContentWorlds):
(WebKit::WebUserContentController::addUserScripts):
(WebKit::WebUserContentController::removeUserScript):
(WebKit::WebUserContentController::removeAllUserScripts):
(WebKit::WebUserContentController::addUserStyleSheets):
(WebKit::WebUserContentController::removeUserStyleSheet):
(WebKit::WebUserContentController::removeAllUserStyleSheets):
(WebKit::WebUserContentController::addUserScriptMessageHandlers):
(WebKit::WebUserContentController::removeUserScriptMessageHandler):
(WebKit::WebUserContentController::removeAllUserScriptMessageHandlers):

• WebProcess/UserContent/WebUserContentController.h:
• WebProcess/UserContent/WebUserContentController.messages.in:

• WebProcess/WebPage/WebPage.cpp:

(WebKit::WebPage::runJavaScript):
(WebKit::WebPage::runJavaScriptInMainFrameScriptWorld):
(WebKit::WebPage::runJavaScriptInFrame):

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

Source/WTF:

• wtf/ObjectIdentifier.h:

(WTF::ObjectIdentifier::generate):
(WTF::ObjectIdentifier::generateThreadSafe):
(WTF::ObjectIdentifier::enableGenerationProtection): To allow restricting generating an identifier to the UIProcess.

2:34 PM Changeset in webkit [255126] by mark.lam@apple.com

• 62 edits in trunk/Source

Introduce a getVTablePointer() utility function.
​https://bugs.webkit.org/show_bug.cgi?id=206804
<rdar://problem/58872290>

Reviewed by Yusuke Suzuki and Oliver Hunt.

Source/WebCore:

Updated CodeGeneratorJS to use getVTablePointer() and rebased test results.

• bindings/scripts/CodeGeneratorJS.pm:

(GenerateImplementation):

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

(WebCore::toJSNewlyCreated):

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

(WebCore::toJSNewlyCreated):

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

(WebCore::toJSNewlyCreated):

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

(WebCore::toJSNewlyCreated):

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

(WebCore::toJSNewlyCreated):

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

(WebCore::toJSNewlyCreated):

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

(WebCore::toJSNewlyCreated):

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

(WebCore::toJSNewlyCreated):

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

(WebCore::toJSNewlyCreated):

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

(WebCore::toJSNewlyCreated):

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

(WebCore::toJSNewlyCreated):

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

(WebCore::toJSNewlyCreated):

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

(WebCore::toJSNewlyCreated):

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

(WebCore::toJSNewlyCreated):

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

(WebCore::toJSNewlyCreated):

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

(WebCore::toJSNewlyCreated):

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

(WebCore::toJSNewlyCreated):

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

(WebCore::toJSNewlyCreated):

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

(WebCore::toJSNewlyCreated):

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

(WebCore::toJSNewlyCreated):

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

(WebCore::toJSNewlyCreated):

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

(WebCore::toJSNewlyCreated):

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

(WebCore::toJSNewlyCreated):

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

(WebCore::toJSNewlyCreated):

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

(WebCore::toJSNewlyCreated):

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

(WebCore::toJSNewlyCreated):

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

(WebCore::toJSNewlyCreated):

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

(WebCore::toJSNewlyCreated):

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

(WebCore::toJSNewlyCreated):

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

(WebCore::toJSNewlyCreated):

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

(WebCore::toJSNewlyCreated):

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

(WebCore::toJSNewlyCreated):

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

(WebCore::toJSNewlyCreated):

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

(WebCore::toJSNewlyCreated):

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

(WebCore::toJSNewlyCreated):

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

(WebCore::toJSNewlyCreated):

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

(WebCore::toJSNewlyCreated):

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

(WebCore::toJSNewlyCreated):

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

(WebCore::toJSNewlyCreated):

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

(WebCore::toJSNewlyCreated):

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

(WebCore::toJSNewlyCreated):

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

(WebCore::toJSNewlyCreated):

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

(WebCore::toJSNewlyCreated):

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

(WebCore::toJSNewlyCreated):

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

(WebCore::toJSNewlyCreated):

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

(WebCore::toJSNewlyCreated):

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

(WebCore::toJSNewlyCreated):

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

(WebCore::toJSNewlyCreated):

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

(WebCore::toJSNewlyCreated):

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

(WebCore::toJSNewlyCreated):

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

(WebCore::toJSNewlyCreated):

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

(WebCore::toJSNewlyCreated):

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

(WebCore::toJSNewlyCreated):

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

(WebCore::toJSNewlyCreated):

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

(WebCore::toJSNewlyCreated):

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

(WebCore::toJSNewlyCreated):

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

(WebCore::toJSNewlyCreated):

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

(WebCore::toJSNewlyCreated):

Source/WTF:

With getVTablePointer(), we can abstract away how we get a vtable function pointer
without assuming the way it is signed for ARM64E. With this, we can remove the
WTF_PREPARE_VTBL_POINTER_FOR_INSPECTION macro which assumes how a vtable function
pointer is signed.

• wtf/PointerPreparations.h:

12:13 PM Changeset in webkit [255125] by mark.lam@apple.com

• 4 edits in trunk

Add some tests for dynamically allocated StaticStringImpls.
​https://bugs.webkit.org/show_bug.cgi?id=206802

Reviewed by Darin Adler.

Source/WTF:

Removed some unnecessary explicit specialization of the charactersAreAllASCII()
template function.

• wtf/text/StringImpl.cpp:

(WTF::StringImpl::createFromLiteral):
(WTF::StringImpl::createStaticStringImpl):

Tools:

• TestWebKitAPI/Tests/WTF/StringImpl.cpp:

(TestWebKitAPI::doStaticStringImplTests):
(TestWebKitAPI::TEST):

12:07 PM Changeset in webkit [255124] by Darin Adler

• 2 edits in trunk/LayoutTests

REGRESSION: [ Mac WK2 ] animations/suspend-resume-animation-events.html is a flaky failure
​https://bugs.webkit.org/show_bug.cgi?id=206615

Reviewed by Alexey Proskuryakov.

• animations/suspend-resume-animation-events.html: Make test slightly less timing

dependent.

12:02 PM Changeset in webkit [255123] by aakash_jain@apple.com

• 3 edits in trunk/Tools

webkit-patch upload should not submit automatically to old EWS
​https://bugs.webkit.org/show_bug.cgi?id=206801

Reviewed by Alexey Proskuryakov.

• Scripts/webkitpy/tool/commands/upload_unittest.py:
• Scripts/webkitpy/tool/steps/submittoews.py:

11:14 AM Changeset in webkit [255122] by commit-queue@webkit.org

• 2 edits in trunk/LayoutTests

Regression (r255103): media/modern-media-controls/placard-support/placard-support-airplay-fullscreen-no-controls.html
​https://bugs.webkit.org/show_bug.cgi?id=206800

Unreviewed test gardening; the test which this was copied from, placard-support-airplay-fullscreen.html, is skipped on
iOS test bots due to a lack of AirPlay test support.

Patch by Jer Noble <​jer.noble@apple.com> on 2020-01-25

• platform/ios/TestExpectations:

9:12 AM Changeset in webkit [255121] by Diego Pino Garcia

• 5 edits in trunk/LayoutTests

[GTK] Gardening of media related tests
​https://bugs.webkit.org/show_bug.cgi?id=206796

Unreviewed gardening.

• platform/gtk/TestExpectations:
• platform/gtk/media/audio-repaint-expected.txt:
• platform/gtk/media/controls-styling-strict-expected.txt:
• platform/gtk/media/media-controls-play-button-updates-expected.txt:

9:03 AM Changeset in webkit [255120] by mark.lam@apple.com

• 15 edits in trunk/Source

Move singleton Intl string locales out of JSGlobalObject.
​https://bugs.webkit.org/show_bug.cgi?id=206791
<rdar://problem/58889037>

Source/JavaScriptCore:

Reviewed by Yusuke Suzuki and Andy Wagoner.

We were creating an instance of these for each JSGlobalObject when they can be a
global singleton since they are always initialized with the same intl data
(barring a mid-flight change in intl settings, which we don't support even in the
existing code).

It turns out that intlPluralRulesAvailableLocales() wasn't called anywhere.
IntlPluralRules code currently just uses intlNumberFormatAvailableLocales().
To document that this is intentional, we do the following:

1. have IntlPluralRules code call intlPluralRulesAvailableLocales(), and
2. have intlPluralRulesAvailableLocales() call intlNumberFormatAvailableLocales() for its implementation.

See ​https://bugs.webkit.org/show_bug.cgi?id=206791#c7 and
​https://bugs.webkit.org/show_bug.cgi?id=206791#c8.

In addMissingScriptLocales(), I'm deliberately naming the string with underscores
because it's much easier to read pa_PK_String and see that it refers to "pa-PK"
as opposed to paPKString. Ditto for zh_CN_String, zh_HK_String, zh_SG_String,
and zh_TW_String.

• runtime/IntlCollator.cpp:

(JSC::IntlCollator::initializeCollator):

• runtime/IntlCollatorConstructor.cpp:

(JSC::IntlCollatorConstructorFuncSupportedLocalesOf):

• runtime/IntlDateTimeFormat.cpp:

(JSC::IntlDateTimeFormat::initializeDateTimeFormat):

• runtime/IntlDateTimeFormatConstructor.cpp:

(JSC::IntlDateTimeFormatConstructorFuncSupportedLocalesOf):

• runtime/IntlNumberFormat.cpp:

(JSC::IntlNumberFormat::initializeNumberFormat):

• runtime/IntlNumberFormatConstructor.cpp:

(JSC::IntlNumberFormatConstructorFuncSupportedLocalesOf):

• runtime/IntlObject.cpp:

(JSC::convertICULocaleToBCP47LanguageTag):
(JSC::addMissingScriptLocales):
(JSC::intlCollatorAvailableLocales):
(JSC::intlDateTimeFormatAvailableLocales):
(JSC::intlNumberFormatAvailableLocales):
(JSC::defaultLocale):

• runtime/IntlObject.h:
• runtime/IntlPluralRules.cpp:

(JSC::IntlPluralRules::initializePluralRules):

• runtime/IntlPluralRulesConstructor.cpp:

(JSC::IntlPluralRulesConstructorFuncSupportedLocalesOf):

• runtime/JSGlobalObject.cpp:

(JSC::addMissingScriptLocales): Deleted.
(JSC::JSGlobalObject::intlCollatorAvailableLocales): Deleted.
(JSC::JSGlobalObject::intlDateTimeFormatAvailableLocales): Deleted.
(JSC::JSGlobalObject::intlNumberFormatAvailableLocales): Deleted.
(JSC::JSGlobalObject::intlPluralRulesAvailableLocales): Deleted.

• runtime/JSGlobalObject.h:

Source/WTF:

Reviewed by Yusuke Suzuki.

Fix a bug in StringImpl::createStaticStringImpl(): I forgot to set its hash value
when I introduced it. StaticStringImpls require that its hash code be set ahead
of time, and cannot be mutated at runtime. See the comment in the definition of
StaticStringImpl in StringImpl.h.

• wtf/text/StringImpl.cpp:

(WTF::StringImpl::createStaticStringImpl):

8:51 AM Changeset in webkit [255119] by pvollan@apple.com

• 13 edits
  1 add in trunk

[Cocoa] Media mime types map should be created in the UI process
​https://bugs.webkit.org/show_bug.cgi?id=206478

Reviewed by Darin Adler.

Source/WebCore:

Creating this map in the WebContent process will access the launch services daemon, which will be blocked.
This patch creates the map in the UI process and sends it to the WebContent process as part of the WebProcess
creation parameters.

API test: WebKit.MimeTypes

• platform/MIMETypeRegistry.cpp:

(WebCore::overriddenMimeTypesMap):
(WebCore::commonMediaTypes):
(WebCore::commonMimeTypesMap):
(WebCore::typesForCommonExtension):

• platform/MIMETypeRegistry.h:
• testing/Internals.cpp:

(WebCore::Internals::mediaMIMETypeForExtension):

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

Source/WebKit:

Send the mime type map from the UI process to the WebContent process as part of the
WebProcess creation parameters.

• Shared/WebProcessCreationParameters.cpp:

(WebKit::WebProcessCreationParameters::encode const):
(WebKit::WebProcessCreationParameters::decode):

• Shared/WebProcessCreationParameters.h:
• UIProcess/Cocoa/WebProcessPoolCocoa.mm:

(WebKit::WebProcessPool::platformInitializeWebProcess):

• WebProcess/cocoa/WebProcessCocoa.mm:

(WebKit::WebProcess::platformInitializeWebProcess):

Tools:

• TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj:
• TestWebKitAPI/Tests/WebKit/MimeTypes.mm: Added.

(TEST):

7:57 AM Changeset in webkit [255118] by Alan Bujtas

• 4 edits in trunk/Source/WebCore

[LFC][IFC] Layout logic should be driven by the type of the inline box
​https://bugs.webkit.org/show_bug.cgi?id=206792
<rdar://problem/58889080>

Reviewed by Antti Koivisto.

Use the type of the inline box to decide what layout functions to call and not whether the
box has children or it establishes a formatting context.

• layout/inlineformatting/InlineFormattingContext.cpp:

(WebCore::Layout::nextInPreOrder):
(WebCore::Layout::InlineFormattingContext::computedIntrinsicWidthConstraints):
(WebCore::Layout::InlineFormattingContext::computeIntrinsicWidthForFormattingRoot):

• layout/inlineformatting/InlineFormattingContext.h:
• layout/inlineformatting/InlineLineBuilder.cpp:

(WebCore::Layout::LineBuilder::isVisuallyNonEmpty const):

6:16 AM Changeset in webkit [255117] by Antti Koivisto

• 16 edits in trunk

[LFC][Integration] Re-enable line layout integration
​https://bugs.webkit.org/show_bug.cgi?id=206795
<rdar://problem/58853907>

Reviewed by Zalan Bujtas.

Source/WebCore:

• page/RuntimeEnabledFeatures.h:

Source/WebKit:

• Shared/WebPreferences.yaml:

Source/WebKitLegacy/mac:

• WebView/WebPreferences.mm:

(+[WebPreferences initialize]):

Tools:

• DumpRenderTree/TestOptions.h:

LayoutTests:

• fast/dom/Range/getBoundingClientRect-expected.txt:
• platform/ios/fast/text/hyphenate-limit-lines-expected.txt:
• platform/ios/fast/text/whitespace/pre-break-word-expected.txt:
• platform/ios/fast/text/word-break-expected.txt:
• platform/mac/fast/text/hyphenate-limit-lines-expected.txt:
• platform/mac/fast/text/whitespace/pre-break-word-expected.txt:
• platform/mac/fast/text/word-break-expected.txt:

6:14 AM BuildingCairoOnWindows edited by Fujii Hironori

(diff)

4:27 AM Changeset in webkit [255116] by youenn@apple.com

• 12 edits in trunk

HTMLMediaElement should not remove the media session at DOM suspension time
​https://bugs.webkit.org/show_bug.cgi?id=206661
<rdar://problem/58800787>

Source/WebCore:

Reviewed by Eric Carlson.

​https://trac.webkit.org/changeset/233560 made it so that, on HTMLMediaElement suspension,
its media session is stopped.
This was done to ensure updateNowPlayingInfo is not called synchronously but asynchronously.
The issue is that, once the media session is stopped, it is removed from the media session vector.
On updating the ready state after suspension, and playing, we try to look into the media session vector and do not find the session.
This triggers the ASSERT.

Partially revert the behavior by calling the same code as clientWillPausePlayback
but make sure updateNowPlayingInfo is calling asynchronously when suspending the media element.
Introduce clientWillBeDOMSuspended for that purpose.

Update mediaPlayerReadyStateChanged to enqueue a task to do the update if the media element is suspended.

Covered by test no longer crashing in debug.

• html/HTMLMediaElement.cpp:

(WebCore::HTMLMediaElement::mediaPlayerReadyStateChanged):
(WebCore::HTMLMediaElement::stopWithoutDestroyingMediaPlayer):

• platform/audio/PlatformMediaSession.cpp:

(WebCore::PlatformMediaSession::processClientWillPausePlayback):
(WebCore::PlatformMediaSession::clientWillPausePlayback):
(WebCore::PlatformMediaSession::clientWillBeDOMSuspended):

• platform/audio/PlatformMediaSession.h:
• platform/audio/PlatformMediaSessionManager.cpp:

(WebCore::PlatformMediaSessionManager::sessionWillEndPlayback):

• platform/audio/PlatformMediaSessionManager.h:
• platform/audio/cocoa/MediaSessionManagerCocoa.h:
• platform/audio/cocoa/MediaSessionManagerCocoa.mm:

(MediaSessionManagerCocoa::sessionWillEndPlayback):

• platform/audio/ios/MediaSessionManagerIOS.h:
• platform/audio/ios/MediaSessionManagerIOS.mm:

(WebCore::MediaSessionManageriOS::sessionWillEndPlayback):

Tools:

Reviewed by Eric Carlson.

• TestWebKitAPI/Tests/WebKitLegacy/ios/ScrollingDoesNotPauseMedia.mm:

(TestWebKitAPI::TEST):
Suspend/resume Active DOM Objects from time to time as would do scrolling.
This allows pending tasks to be executed asynchronously when not scrolling.

Jan 24, 2020:

8:24 PM Changeset in webkit [255115] by rniwa@webkit.org

• 2 edits in trunk/LayoutTests

[iOS] REGRESSION(r254699) : fast/forms/interactive-validation-remove-node-in-handler.html is timing out
​https://bugs.webkit.org/show_bug.cgi?id=206669

Reviewed by Wenson Hsieh.

The timeout was caused by the invalidation bubble in interactive-validation-prevented.html never getting closed.
Address the issue by explicitly dismissing the invalidation bubble by setting a value to the input element.

• fast/forms/interactive-validation-prevented.html:

6:09 PM Changeset in webkit [255114] by timothy_horton@apple.com

• 2 edits in trunk/Source/WebKit

macCatalyst: Don't get any custom cursors in subframes
​https://bugs.webkit.org/show_bug.cgi?id=206781
<rdar://problem/58698374>

Reviewed by Simon Fraser.

• WebProcess/WebPage/ios/WebPageIOS.mm:

(WebKit::WebPage::positionInformation):
Adjust the hit test to allow child frame content, and reorganize the
rest of the code to use the hit frame instead of the main frame.

5:37 PM Changeset in webkit [255113] by commit-queue@webkit.org

• 3 edits in trunk/Source/WebCore

Null Ptr Deref READ @ WebCore::RenderMultiColumnFlow::lastMultiColumnSet const
​https://bugs.webkit.org/show_bug.cgi?id=206106

Patch by Jack Lee <Jack Lee> on 2020-01-24
Reviewed by Ryosuke Niwa.

Could not write a reproducible fast test case for this.

• rendering/RenderMultiColumnFlow.cpp:

(WebCore::RenderMultiColumnFlow::lastMultiColumnSet const):

• rendering/updating/RenderTreeBuilderMultiColumn.cpp:

(WebCore::RenderTreeBuilder::MultiColumn::processPossibleSpannerDescendant):

5:35 PM Changeset in webkit [255112] by mark.lam@apple.com

• 6 edits
  1 add in trunk

IntlObject's cached strings should be immortal and safe for concurrent access.
​https://bugs.webkit.org/show_bug.cgi?id=206779
<rdar://problem/58831763>

Reviewed by Yusuke Suzuki.

JSTests:

• stress/numberingSystemsForLocale-cached-strings-should-be-immortal-and-safe-for-concurrent-access.js: Added.

Source/JavaScriptCore:

In IntlObject's numberingSystemsForLocale(), we have a never destroyed
cachedNumberingSystems which is a singleton vector of Strings which are shared
multiple VMs. Hence, the strings in this vector should be a StaticStringImpl
so that it will be immortal, and can be access concurrently from multiple VMs
on different threads without any ref/deref'ing race issues.

• runtime/IntlObject.cpp:

(JSC::numberingSystemsForLocale):

Source/WTF:

Add a factory for creating a dynamically allocated StaticStringImpl.

Note: StaticStringImpl is guaranteed to have the same shape as StringImpl.
The only difference is that s_refCountFlagIsStaticString is set on the refCount
for StaticStringImpl. Since the client will use the StaticStringImpl as a
StringImpl, we implement the factory by using StringImpl::createInternal() for
simplicity, and set the s_refCountFlagIsStaticString flag thereafter.

• wtf/text/StringImpl.cpp:

(WTF::StringImpl::createStaticStringImpl):

• wtf/text/StringImpl.h:

5:28 PM Changeset in webkit [255111] by jiewen_tan@apple.com

• 2 edits in trunk/Source/WebKit

Unreviewed, a speculative test fix after r255041

• UIProcess/WebAuthentication/Mock/MockLocalConnection.mm:

(WebKit::MockLocalConnection::getAttestation const):

5:22 PM Changeset in webkit [255110] by ysuzuki@apple.com

• 2 edits in trunk/JSTests

REGRESSION (r254964-r254970?): Catalina Debug JSC bot timing out while running tests
​https://bugs.webkit.org/show_bug.cgi?id=206785

Unreviewed, skipping the test if $buildType is "debug".

• stress/operand-should-fit-in-abstract-heap-encoded-payload-format.js:

5:12 PM Changeset in webkit [255109] by Alan Coon

• 1 copy in tags/Safari-609.1.15.2

Tag Safari-609.1.15.2.

5:08 PM Changeset in webkit [255108] by Wenson Hsieh

• 9 edits
  2 adds in trunk

[iOS] Long pressing text inside a selection should update the selection
​https://bugs.webkit.org/show_bug.cgi?id=206769
<rdar://problem/58704316>

Reviewed by Tim Horton.

Source/WebCore:

Remove hasSelectionAtPosition, which is no longer needed.

• page/Page.cpp:

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

• page/Page.h:

Source/WebKit:

Makes it possible to update the selection using text interaction gestures inside an existing text selection. To
do this, we remove code that was previously required in order to disambiguate text selection and drag gestures.
However, since adopting asynchronous drag interaction SPI, this check has not been necessary.

The only other purpose of this check was to avoid triggering text selection gestures after canceling a drag
that did not begin (i.e. the touch location did not move). Instead of bailing in -textInteractionGestures:
shouldBeginAtPoint: and -hasSelectablePositionAtPoint:, we can achieve the same effect by simply resetting the
text interaction gestures when the drag lift begins (at which point we know the drag interaction is starting).
This also ensures that on iPad, in the case where the drag was cancelled by the page, long pressing may still
trigger text selection.

Test: editing/selection/ios/select-text-in-existing-selection.html

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

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

Remove the hasSelectionAtPosition flag from position information, since we no longer need it.

• UIProcess/ios/WKContentViewInteraction.mm:

(-[UIGestureRecognizer _wk_cancel]):

Drive-by fix: just ignore the gesture if it is already disabled. This prevents us from causing a gesture
recognizer that was previously disabled to become enabled as a result of calling -_wk_cancel.

(-[WKContentView hasSelectablePositionAtPoint:]):
(-[WKContentView textInteractionGesture:shouldBeginAtPoint:]):
(-[WKContentView cancelActiveTextInteractionGestures]):

Instead of making the text interaction gestures return NO from -gestureRecognizerShouldBegin:, explicitly cancel
the text interaction gestures that may select text.

(-[WKContentView dragInteraction:willAnimateLiftWithAnimator:session:]):

• WebProcess/WebPage/ios/WebPageIOS.mm:

(WebKit::WebPage::positionInformation):

Remove the hasSelectionAtPosition flag from position information, since we no longer need it.

LayoutTests:

Add a layout test to verify that it's possible to select text by long pressing inside an existing text
selection. The test is runnable on both iPad and iPhone, since it programmatically prevents dragstart.

• editing/selection/ios/select-text-in-existing-selection-expected.txt: Added.
• editing/selection/ios/select-text-in-existing-selection.html: Added.

5:07 PM Changeset in webkit [255107] by Alan Bujtas

• 2 edits in trunk/Source/WebCore

[LFC][IFC] (Regression) LayoutTests/fast/text/whitespace/029.html fails
​https://bugs.webkit.org/show_bug.cgi?id=206773
<rdar://problem/58883388>

Reviewed by Simon Fraser.

While line break is an inline box, it does not have inline content so do not try to cast it to Container.

• layout/inlineformatting/InlineFormattingContext.cpp:

(WebCore::Layout::InlineFormattingContext::computedIntrinsicWidthConstraints):
(WebCore::Layout::InlineFormattingContext::collectInlineContentIfNeeded):

4:49 PM Changeset in webkit [255106] by achristensen@apple.com

• 3 edits in trunk/Source/WebKit

Update EditorState from WebProcess when requesting context menu
​https://bugs.webkit.org/show_bug.cgi?id=206775
<rdar://problem/56305573>

Reviewed by Wenson Hsieh.

When generating a context menu, AppKit calls -validRequestorForSendType:returnType:
which queries the editor state in WebViewImpl::validRequestorForSendAndReturnTypes.
Without updating the editor state, the first time you right click on a link,
the Services menu item did not appear because we were looking at a stale editor state.
Verified manually this fixes the bug.

• WebProcess/WebPage/WebContextMenu.cpp:

(WebKit::WebContextMenu::show):

• WebProcess/WebPage/mac/WebPageMac.mm:

(WebKit::WebPage::handleSelectionServiceClick):

4:45 PM Changeset in webkit [255105] by timothy_horton@apple.com

• 2 edits in trunk/Source/WebKit

macCatalyst: Right clicking on links follows the link, but shouldn't
​https://bugs.webkit.org/show_bug.cgi?id=206777
<rdar://problem/56586280>

Reviewed by Wenson Hsieh.

• UIProcess/ios/WKMouseGestureRecognizer.mm:

(-[WKMouseGestureRecognizer createMouseEventWithType:forEvent:]):
(-[WKMouseGestureRecognizer touchesEnded:withEvent:]):
I got 'button' and 'buttons' backwards.

'button' indicates which button the event is about, so in MouseUp,
it should still be 2.

'buttons' indicate which buttons are still down, so in MouseUp,
it should not include 2. Since we don't currently track mouse button
chording here, we'll just say "none". Leave a FIXME about that.

This makes WebCore's behavior correct, and now it doesn't follow the link.

4:34 PM Changeset in webkit [255104] by rniwa@webkit.org

• 2 edits in trunk/LayoutTests

imported/w3c/web-platform-tests/requestidlecallback/callback-xhr-sync.html is flaky failure
​https://bugs.webkit.org/show_bug.cgi?id=206676
<rdar://problem/58839631>

Add a flaky test expectation.

• TestExpectations:

4:23 PM Changeset in webkit [255103] by commit-queue@webkit.org

• 4 edits
  2 adds in trunk

AirPlay placard not visible when AirPlay is entered in fullscreen mode.
​https://bugs.webkit.org/show_bug.cgi?id=206772
<rdar://problem/57098851>

Patch by Jer Noble <​jer.noble@apple.com> on 2020-01-24
Reviewed by Eric Carlson.

Test: media/modern-media-controls/placard-support/placard-support-airplay-fullscreen-no-controls.html

The AirPlay and Picture-in-picture placards should always become visible, regardless of
whether the <video> element uses the native media controls. So "disabling" the PlacardSupport
object should not remove the listeners for presentation mode and route change events. Instead
merely remember whether the object is disabled, so that the "error placard" will only be shown
if controls are enabled, and PiP and AirPlay placards will always be shown.

• Modules/modern-media-controls/media/placard-support.js:

(PlacardSupport.prototype.enable):
(PlacardSupport.prototype.disable):
(PlacardSupport.prototype._updatePlacard):

3:40 PM Changeset in webkit [255102] by Russell Epstein

• 4 edits in branches/safari-609-branch/Source/WebKit

Cherry-pick r255039. rdar://problem/58855996

Background thread with ITP Database should lock when the network process is suspended
​https://bugs.webkit.org/show_bug.cgi?id=206593
rdar://problem/58713379

Reviewed by Chris Dumez.

This patch adds a function to hang the background queue when the
NetworkProcess is preparing to suspend.

• NetworkProcess/Classifier/WebResourceLoadStatisticsStore.cpp: (WebKit::WebResourceLoadStatisticsStore::suspend): (WebKit::WebResourceLoadStatisticsStore::resume):
• NetworkProcess/Classifier/WebResourceLoadStatisticsStore.h:
• NetworkProcess/NetworkProcess.cpp: (WebKit::NetworkProcess::prepareToSuspend): (WebKit::NetworkProcess::resume):

git-svn-id: ​https://svn.webkit.org/repository/webkit/trunk@255039 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3:40 PM Changeset in webkit [255101] by Russell Epstein

• 2 edits in branches/safari-609-branch/Source/JavaScriptCore

Cherry-pick r254996. rdar://problem/58856010

OptimizeAssociativeExpressionTrees should reset value owners before running
​https://bugs.webkit.org/show_bug.cgi?id=206670
<rdar://problem/58535628>

Reviewed by Robin Morisset.

We have a crash inside OptimizeAssociativeExpressionTrees and we don't know
how to reproduce it. Also, based on Mark's auditing of the crash site's
assembly, Mark thinks we're crashing on a "currupt" basic block.

After I audited the code, I saw that we rely on value owners in this phase.
However, we don't actually reset them before running the phase. This patch
adds that as a speculative fix for the crash we're seeing.

• b3/B3OptimizeAssociativeExpressionTrees.cpp: (JSC::B3::OptimizeAssociativeExpressionTrees::run):

git-svn-id: ​https://svn.webkit.org/repository/webkit/trunk@254996 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3:40 PM Changeset in webkit [255100] by Russell Epstein

• 6 edits in branches/safari-609-branch/Source/WebCore

Cherry-pick r254991. rdar://problem/58856021

[Web Animations] Make AnimationList ref-counted
​https://bugs.webkit.org/show_bug.cgi?id=206664

Reviewed by Antti Koivisto.

• platform/animation/AnimationList.cpp:
• platform/animation/AnimationList.h: (WebCore::AnimationList::create): (WebCore::AnimationList::copy): (WebCore::AnimationList::AnimationList): Deleted.
• rendering/style/RenderStyle.cpp: (WebCore::RenderStyle::ensureAnimations): (WebCore::RenderStyle::ensureTransitions):
• rendering/style/StyleRareNonInheritedData.cpp: (WebCore::StyleRareNonInheritedData::StyleRareNonInheritedData):
• rendering/style/StyleRareNonInheritedData.h:

git-svn-id: ​https://svn.webkit.org/repository/webkit/trunk@254991 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3:40 PM Changeset in webkit [255099] by Russell Epstein

• 3 edits in branches/safari-609-branch/Source/JavaScriptCore

Cherry-pick r254962. rdar://problem/58856037

Restore nullification of DFG::Plan::m_vm when the plan is cancelled.
​https://bugs.webkit.org/show_bug.cgi?id=206633
<rdar://problem/58811967>

Reviewed by Robin Morisset.

In r253243, I replaced the nullification of Plan::m_vm in Plan::cancel() with
code to decorate the m_vm pointer with a nuke bit. The thinking is that keeping
the VM pointer in nuked form allows us to do certain assertions, as well as
implementing code in support of keeping Box<Identifier>s alive. It is only
correct to use the nuked VM pointer if and only if the VM is guaranteed to
outlive the Plan. r253243 guarantees this condition.

In r254464, I replaced the use of Box<Identifier> with CacheableIdentifier.
This obviated all the support code added above, and rolled out most of it.
However, I opted to keep the nuked VM pointer in the DFG::Plan to as a debugging
aid (it's nice to be able to know which VM the Plan came from).

However, r254464 also undid the guarantee that the VM will outlive the Plan.
As a result, a nuked VM pointer is no longer guaranteed to point to a valid VM.
Some worker layout tests, run on an ASAN build, detected that the pointer is
pointing to an already freed VM and failed with a crash.

This patch fixes this issue by completely reverting the nuked VM pointer code,
and restores nullification of the m_vm pointer in Plan::cancel().

• dfg/DFGPlan.cpp: (JSC::DFG::Plan::computeCompileTimes const): (JSC::DFG::Plan::cancel):
• dfg/DFGPlan.h: (JSC::DFG::Plan::vm const): (JSC::DFG::Plan::unnukedVM const): Deleted. (JSC::DFG::Plan::nuke): Deleted. (JSC::DFG::Plan::unnuke): Deleted.

git-svn-id: ​https://svn.webkit.org/repository/webkit/trunk@254962 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3:40 PM Changeset in webkit [255098] by Russell Epstein

• 3 edits in branches/safari-609-branch/Source/WebCore

Cherry-pick r254951. rdar://problem/58856000

Make a Ref to WindowEventLoop when the timer to run tasks fires
​https://bugs.webkit.org/show_bug.cgi?id=206568

Reviewed by Antti Koivisto.

Keep WindowEventLoop alive explicitly while invoking EventLoop::run.

• dom/WindowEventLoop.cpp: (WebCore::WindowEventLoop::WindowEventLoop): (WebCore::WindowEventLoop::didReachTimeToRun):
• dom/WindowEventLoop.h:

git-svn-id: ​https://svn.webkit.org/repository/webkit/trunk@254951 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3:40 PM Changeset in webkit [255097] by Russell Epstein

• 3 edits in branches/safari-609-branch/Source/WebKit

Cherry-pick r254938. rdar://problem/58856028

Fix SDK availability macros
​https://bugs.webkit.org/show_bug.cgi?id=206463
<rdar://problem/58581906>

Reviewed by Timothy Hatcher.

Bug 197841 updated WebKit.xcconfig with regards to when it runs a
header post-processing script. This post-processing script grovels
over our exported headers, changing macros like WK_API_AVAILABLE to
the standard API_AVAILABLE. Because of the change in Bug 197841, the
script was no longer being run when preparing macOS 10.15 SDKs, which
was breaking those SDKs. Fix this by reverting the change, allowing
the macros to be replaced for macOS 10.15.

We also need to touch WKFoundation.h in order to support incremental
builds. If we don't, then the version of WKFoundation.h that was
exported in the previous build and that was incorrectly post-processed
would remain incorrectly post-processed. Touch the file so that it
will get re-exported and re-post-processed.

No new tests - no added or changed functionality.

• Configurations/WebKit.xcconfig:
• Shared/API/Cocoa/WKFoundation.h:

git-svn-id: ​https://svn.webkit.org/repository/webkit/trunk@254938 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3:40 PM Changeset in webkit [255096] by Russell Epstein

• 4 edits in branches/safari-609-branch/Tools

Cherry-pick r254318. rdar://problem/58856031

TestController::platformLibraryPathForTesting() returns inner pointer to autoreleased NSString
<​https://webkit.org/b/206018>
<rdar://problem/58449733>

Reviewed by Darin Adler.

• WebKitTestRunner/TestController.cpp: (WTR::TestController::initialize):
• Use WTF::AutodrainedPool to fix remaining autoreleasePool leaks.
• WebKitTestRunner/ios/TestControllerIOS.mm: (WTR::TestController::platformLibraryPathForTesting):
• WebKitTestRunner/mac/TestControllerMac.mm: (WTR::TestController::platformLibraryPathForTesting):
• Retain NSString since these methods return an inner pointer. This also only needs to be initialized once.

git-svn-id: ​https://svn.webkit.org/repository/webkit/trunk@254318 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2:52 PM Changeset in webkit [255095] by Andres Gonzalez

• 2 edits in trunk/Source/WebCore

Crash in AXIsolatedObject::tagName.
​https://bugs.webkit.org/show_bug.cgi?id=206765

Reviewed by Chris Fleizach.

Fixes crash in AXIsolatedObject::tagName.

• accessibility/isolatedtree/AXIsolatedObject.cpp:

(WebCore::AXIsolatedObject::initializeAttributeData): Was missing the isolatedCopy when caching the tagName.

2:19 PM Changeset in webkit [255094] by Keith Rollin

• 2 edits in trunk/Source/WebKitLegacy

Remove check-xcfilelists.sh from WebKitLegacy Xcode build target
​https://bugs.webkit.org/show_bug.cgi?id=206762
<rdar://problem/58825997>

Reviewed by Alexey Proskuryakov.

When check-xcfilelists.sh got added to the WebKitLegacy Xcode project,
it was also accidentally added to the WebKitLegacy build target. This
caused it to be treated as a "resource", and so was copied to the
output directory in the Resources sub-folder during the build. This
copying is not expected, needed, or desired. Remove
check-xcfilelists.sh from the target, which also removes it from the
Copy Resources build phase, which removes it from the
WebKitLegacy.framework/Resources directory.

• WebKitLegacy.xcodeproj/project.pbxproj:

1:58 PM Changeset in webkit [255093] by jiewen_tan@apple.com

• 2 edits in trunk/Source/WebKit

Unreviewed, a build fix after r255056

• UIProcess/WebAuthentication/Mock/MockLocalConnection.mm:

(WebKit::MockLocalConnection::getAttestation const):

1:18 PM Changeset in webkit [255092] by Ryan Haddad

• 2 edits in trunk/Source/WebCore

Unreviewed, rolling out r255089.

Breaks tvOS build.

Reverted changeset:

"WebCore: Remove iOS 11 macros from
WebItemProviderPasteboard.h"
​https://bugs.webkit.org/show_bug.cgi?id=206719
​https://trac.webkit.org/changeset/255089

12:57 PM Changeset in webkit [255091] by Alan Coon

• 5 edits in branches/safari-609.1.15-branch/Source/WebKit

Cherry-pick r254552. rdar://problem/58812968

REGRESSION (r253394): After swiping back during a navigation, WKWebView gets stuck with the forward content, stops repainting
​https://bugs.webkit.org/show_bug.cgi?id=206268
<rdar://problem/58536702>

Reviewed by Simon Fraser.

• UIProcess/ViewGestureController.cpp: (WebKit::ViewGestureController::endSwipeGesture):
• UIProcess/ViewGestureController.h:
• UIProcess/ios/ViewGestureControllerIOS.mm: (WebKit::ViewGestureController::beginSwipeGesture): (WebKit::ViewGestureController::endSwipeGesture): (WebKit::ViewGestureController::removeSwipeSnapshot):
• UIProcess/mac/ViewGestureControllerMac.mm: (WebKit::ViewGestureController::beginSwipeGesture): (WebKit::ViewGestureController::removeSwipeSnapshot): After r253394, we start loading the destination page in willEndSwipeGesture, when we know the gesture will complete, instead of in endSwipeGesture, when it is actually done.

This means that if we decide that we should tear down the snapshot immediately,
this can now happen in the window between willEndSwipeGesture and endSwipeGesture.

However, removeSwipeSnapshot has numerous dependencies on endSwipeGesture
(especially on iOS, where there are /also/ dependencies in the other direction -
endSwipeGesture will never be called after removeSwipeSnapshot because
of the gestureID mismatch).

Regardless, it does not make sense to remove the snapshot while the animation
is still running. So, if something causes removeSwipeSnapshot to be called
before endSwipeGesture, we just set a bit and call it inside endSwipeGesture instead.

This ends up putting the snapshot removal ordering back as it was before r253394.

git-svn-id: ​https://svn.webkit.org/repository/webkit/trunk@254552 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12:54 PM Changeset in webkit [255090] by Alan Coon

• 8 edits in branches/safari-609.1.15-branch/Source

Versioning.

12:40 PM Changeset in webkit [255089] by Jonathan Bedard

• 2 edits in trunk/Source/WebCore

WebCore: Remove iOS 11 macros from WebItemProviderPasteboard.h
​https://bugs.webkit.org/show_bug.cgi?id=206719

Reviewed by Tim Horton.

No functional changes, covered by existing tests.

• platform/ios/WebItemProviderPasteboard.h:

12:26 PM Changeset in webkit [255088] by Caio Lima

• 7 edits in trunk/Source/JavaScriptCore

[ARMv7][JIT] Implement checkpoint support
​https://bugs.webkit.org/show_bug.cgi?id=206611

Reviewed by Yusuke Suzuki.

This patch is adding code path to properly support checkpoint on ARMv7.

• dfg/DFGOSREntry.cpp:

(JSC::DFG::prepareOSREntry):

We changed the scratch buffer population to properly handle
callee-saved registers on 32-bits. For more details, check comments on
source code.

• dfg/DFGOSRExit.cpp:

(JSC::DFG::OSRExit::compileExit):

• jit/GPRInfo.h:

(JSC::GPRInfo::toRegister):
(JSC::GPRInfo::toIndex):

We are adding back regCS1 to be used as GPRReg on DFG and
scratch registers. This was necessary because some code generated by
GetByVal IC requires at least 9 registers available.
Given that, we also added such register to dfgCalleeSaveRegisters
set.

• jit/JITPropertyAccess32_64.cpp:

(JSC::JIT::emitPutByValWithCachedId):

• jit/RegisterSet.cpp:

(JSC::RegisterSet::dfgCalleeSaveRegisters):

• llint/LowLevelInterpreter.asm:

12:25 PM Changeset in webkit [255087] by Keith Rollin

• 2 edits in trunk/Source/WTF

Fix internal Apple builds after r254411
​https://bugs.webkit.org/show_bug.cgi?id=206723
<rdar://problem/58844735>

Reviewed by Maciej Stachowiak.

The default Xcode build configuration has changed between macOS 10.13
and macOS 10.15, such that references to some frameworks when bulding
for the latter don't work when building for the former. Fix this by
explicitly adding support for the desired build location rather than
assuming such support will be provided for us.

• Configurations/WTF.xcconfig:

12:00 PM Changeset in webkit [255086] by keith_miller@apple.com

• 3 edits in trunk/Source/JavaScriptCore

JSC should produce a module map.
​https://bugs.webkit.org/show_bug.cgi?id=206717

Reviewed by Mark Lam.

• Configurations/JavaScriptCore.xcconfig:
• JavaScriptCore.modulemap:

11:33 AM Changeset in webkit [255085] by youenn@apple.com

• 8 edits in trunk

Add support for MediaRecorder.requestData
​https://bugs.webkit.org/show_bug.cgi?id=206731

LayoutTests/imported/w3c:

Reviewed by Eric Carlson.

• web-platform-tests/mediacapture-record/idlharness.window-expected.txt:

Source/WebCore:

Reviewed by Eric Carlson.

Add support for requestData that fetches any available data.
Expose MediaRecorder stream.
Covered by updated tests.

• Modules/mediarecorder/MediaRecorder.cpp:

(WebCore::MediaRecorder::requestData):

• Modules/mediarecorder/MediaRecorder.h:
• Modules/mediarecorder/MediaRecorder.idl:

LayoutTests:

Reviewed by Eric Carlson.

• http/wpt/mediarecorder/MediaRecorder-dataavailable.html:

11:30 AM Changeset in webkit [255084] by Jon Davis

• 2 edits in trunk/Websites/webkit.org

Fix editing permissions for the Web Inspector Editor role
​https://bugs.webkit.org/show_bug.cgi?id=206761

Reviewed by Devin Rousso.

• wp-content/plugins/web-inspector-pages.php:

11:12 AM Changeset in webkit [255083] by commit-queue@webkit.org

• 3 edits
  2 adds in trunk

Nullptr deref in WebCore::RenderTreeBuilder::Block::attachIgnoringContinuation when an element is inserted before legend under multi-column layout.
​https://bugs.webkit.org/show_bug.cgi?id=206109

Patch by Jack Lee <Jack Lee> on 2020-01-24
Reviewed by Antti Koivisto.

Source/WebCore:

Test: fast/forms/fieldset/fieldset-crash-insert-before-legend-under-multicol.html

• rendering/updating/RenderTreeBuilderBlockFlow.cpp:

(WebCore::RenderTreeBuilder::BlockFlow::attach):

LayoutTests:

• fast/forms/fieldset/fieldset-crash-insert-before-legend-under-multicol-expected.txt: Added.
• fast/forms/fieldset/fieldset-crash-insert-before-legend-under-multicol.html: Added.

10:57 AM Changeset in webkit [255082] by Matt Lewis

• 5 edits in trunk/Tools

Remove Apple windows 7 queues.
​https://bugs.webkit.org/show_bug.cgi?id=206702

Reviewed by Jonathan Bedard.

• BuildSlaveSupport/build.webkit.org-config/config.json:
• BuildSlaveSupport/build.webkit.org-config/public_html/dashboard/Scripts/WebKitBuildbot.js:
• BuildSlaveSupport/build.webkit.org-config/public_html/dashboard/Scripts/Dashboard.js:
• BuildSlaveSupport/build.webkit.org-config/public_html/dashboard/Styles/Main.css:
• BuildSlaveSupport/build.webkit.org-config/steps_unittest.py:

Removed Windows 7 queues from config and from dashboard.

10:53 AM Changeset in webkit [255081] by pvollan@apple.com

• 13 edits
  1 delete in trunk

Unreviewed, rolling out r255050.

Introduced crashes on bots

Reverted changeset:

"[Cocoa] Media mime types map should be created in the UI
process"
​https://bugs.webkit.org/show_bug.cgi?id=206478
​https://trac.webkit.org/changeset/255050

10:40 AM Changeset in webkit [255080] by Andres Gonzalez

• 5 edits in trunk/Source/WebCore

Implementation of AXIsolatedObject::isStaticText and isLandmark methods.
​https://bugs.webkit.org/show_bug.cgi?id=206745

Reviewed by Chris Fleizach.

• isStaticText moved from the AXObject class to the base AXCoreObject so

that it is used by isolated objects as well.

• Implemented isLandmark.
• Ordered alphabetically the setProperty calls for most IsXXX

properties in initializeAttributeData for easier identification.

• accessibility/AccessibilityObject.h:
• accessibility/AccessibilityObjectInterface.h:

(WebCore::AXCoreObject::isStaticText const):

• accessibility/isolatedtree/AXIsolatedObject.cpp:

(WebCore::AXIsolatedObject::initializeAttributeData):
(WebCore::AXIsolatedObject::isLandmark const): Deleted, inline in header file.
(WebCore::AXIsolatedObject::isStaticText const): Deleted, implemented in AXCoreObject.

• accessibility/isolatedtree/AXIsolatedObject.h:

10:29 AM Changeset in webkit [255079] by commit-queue@webkit.org

• 2 edits
  5 adds in trunk/LayoutTests

[GTK][WPE] Websockets close-on-x tests are failing
​https://bugs.webkit.org/show_bug.cgi?id=206392

Unreviewed gardening.

Rebaseline after libsoup commit 35f1bac5.

Patch by Lauro Moura <Lauro Moura> on 2020-01-24

• platform/gtk/TestExpectations:
• platform/gtk/http/tests/websocket/tests/hybi/close-on-navigate-new-location-expected.txt: Added.
• platform/gtk/http/tests/websocket/tests/hybi/close-on-unload-and-force-gc-expected.txt: Added.
• platform/gtk/http/tests/websocket/tests/hybi/close-on-unload-expected.txt: Added.
• platform/gtk/http/tests/websocket/tests/hybi/close-on-unload-reference-in-parent-expected.txt: Added.
• platform/gtk/http/tests/websocket/tests/hybi/send-after-close-on-unload-expected.txt: Added.

10:13 AM Changeset in webkit [255078] by commit-queue@webkit.org

• 3 edits in trunk/LayoutTests

[GTK][WPE] Several WPT fetch tests are failing
​https://bugs.webkit.org/show_bug.cgi?id=206416

Patch by Rob Buis <​rbuis@igalia.com> on 2020-01-24
Reviewed by Frédéric Wang.

It is better to skip these as more tests can be added
and this feature is not enabled on WPE/GTK.

• platform/gtk/TestExpectations:
• platform/wpe/TestExpectations:

9:58 AM Changeset in webkit [255077] by commit-queue@webkit.org

• 3 edits in trunk/Source/JavaScriptCore

Unreviewed, rolling out r255052.
​https://bugs.webkit.org/show_bug.cgi?id=206758

"Broke production builds" (Requested by rmorisset on #webkit).

Reverted changeset:

"Break the dependency between jsc and DerivedSources"
​https://bugs.webkit.org/show_bug.cgi?id=206720
​https://trac.webkit.org/changeset/255052

9:35 AM Changeset in webkit [255076] by graouts@webkit.org

• 14 edits in trunk

[Web Animations] Support multiple CSS Animations with the same name in animation-name
​https://bugs.webkit.org/show_bug.cgi?id=206688

Reviewed by Dean Jackson.

LayoutTests/imported/w3c:

Mark some new WPT progressions.

• web-platform-tests/css/css-animations/Element-getAnimations-dynamic-changes.tentative-expected.txt:
• web-platform-tests/css/css-animations/Element-getAnimations.tentative-expected.txt:
• web-platform-tests/web-animations/timing-model/timelines/update-and-send-events-replacement-expected.txt:

Source/WebCore:

AnimationTimeline would keep track of registered CSS Animations by name for a given element in m_elementToCSSAnimationByName which would map one CSSAnimation
per String (the animation-name) for a given Element. However, within the same animation-name property, the name of a given @keyframes rules may appear more
than once, and the CSS Animations specification explains how to handle this scenario.

We now correctly handle this by replacing m_elementToCSSAnimationByName with the new m_elementToCSSAnimationsCreatedByMarkupMap which simply maps an Element
to a ListHashSet of CSSAnimation objects. Removing the string that appeared in animation-name to create this animation requires us to keep the AnimationList
used for the last style update for sorting purposes, since having multiple instances of the same string would not allow disambiguation when sorting the
KeyframeEffectStack.

So we also replace m_cssAnimationNames, a Vector<String>, with m_cssAnimationList, a RefPtr<const AnimationList>, and use this to compare Animation objects
stored in the AnimationList against the backing animation of each CSSAnimation.

Storing the AnimationList on the KeyframeEffectStack also has the benefit of allowing us to use this as the previous state when updating CSS Animations in
AnimationTimeline::updateCSSAnimationsForElement(). We used to rely on the previous RenderStyle provided to that function, but it's possible that this style
is null and we would unnecessarily create additional CSSAnimation objects for animations that actually were retained since the last time CSS Animations were
invalidated. We now use the stored AnimationList on the invalidated element's KeyframeEffectStack and create a new animation list that will replace the old
list stored in the m_elementToCSSAnimationsCreatedByMarkupMap map for that element. We can also compare the old list with the new list to find out which
animations are no longer current.

Finally, we refactor things a bit to have some new aliases AnimationCollection and CSSAnimationCollection instead of using ListHashSet<> in our types.

• animation/AnimationTimeline.cpp:

(WebCore::AnimationTimeline::animationWasAddedToElement): Use the new AnimationCollection alias.
(WebCore::AnimationTimeline::removeDeclarativeAnimationFromListsForOwningElement): We no longer need to do any work for CSSAnimation here since the
m_elementToCSSAnimationByName map is no more and the m_elementToCSSAnimationsCreatedByMarkupMap that replaces it is updated in updateCSSAnimationsForElement()
and elementWasRemoved().
(WebCore::AnimationTimeline::animationsForElement const): Since animations are correctly sorted accounting for their composite order in KeyframeEffectStack,
call KeyframeEffectStack::sortedEffects() when we're called with Ordering::Sorted.
(WebCore::AnimationTimeline::removeCSSAnimationCreatedByMarkup): New method called by elementWasRemoved() to ensure that when an element is removed, we remove
its CSSAnimation objects from the new m_elementToCSSAnimationsCreatedByMarkupMap and also update the AnimationList on the relevant KeyframeEffectStack.
(WebCore::AnimationTimeline::elementWasRemoved): Call the new removeCSSAnimationCreatedByMarkup() method before canceling a CSSAnimation.
(WebCore::AnimationTimeline::cancelDeclarativeAnimationsForElement): Call the new removeCSSAnimationCreatedByMarkup() method before canceling a CSSAnimation.
(WebCore::AnimationTimeline::updateCSSAnimationsForElement): Use the AnimationList recoreded on the relevant KeyframeEffectStack to determine which CSSAnimation
objects to create, cancel or merely update depending on the AnimationList in the current style.

• animation/AnimationTimeline.h:
• animation/DocumentTimeline.cpp:

(WebCore::DocumentTimeline::getAnimations const): Use compareAnimationsByCompositeOrder() to correctly sort CSS Animations since they are no longer guaranteed
to be stored in the relevant map in the expected order.

• animation/KeyframeEffectStack.cpp:

(WebCore::KeyframeEffectStack::ensureEffectsAreSorted): Use the new m_cssAnimationList instead of the old m_cssAnimationNames when sorting effects.
(WebCore::KeyframeEffectStack::setCSSAnimationList):
(WebCore::KeyframeEffectStack::setCSSAnimationNames): Deleted.

• animation/KeyframeEffectStack.h:

(WebCore::KeyframeEffectStack::cssAnimationList const):
(WebCore::KeyframeEffectStack::cssAnimationNames const): Deleted.

• animation/WebAnimation.cpp:

(WebCore::WebAnimation::commitStyles): Use the new KeyframeEffectStack::cssAnimationList() instead of the old KeyframeEffectStack::cssAnimationNames().

• animation/WebAnimationUtilities.cpp:

(WebCore::compareAnimationsByCompositeOrder): Update the composite order comparison utility to use an AnimationList rather than a list of animation names.

• animation/WebAnimationUtilities.h:
• platform/animation/AnimationList.h:

(WebCore::AnimationList::copy const):

9:18 AM Changeset in webkit [255075] by Jonathan Bedard

• 2 edits in trunk/Tools

run-webkit-tests: Handle case where device socket make handle invalid
​https://bugs.webkit.org/show_bug.cgi?id=206707

Reviewed by Stephanie Lewis.

• Scripts/webkitpy/port/simulator_process.py:

(SimulatorProcess.NonBlockingFileFromSocket.close): Catch IOError when closing file.

8:42 AM Changeset in webkit [255074] by Alan Bujtas

• 2 edits in trunk/Source/WebCore

[LFC][Painting] Add Display::Run cleanup to TextPainter::clearGlyphDisplayLists
​https://bugs.webkit.org/show_bug.cgi?id=206744
<rdar://problem/58867112>

Reviewed by Antti Koivisto.

• rendering/TextPainter.cpp:

(WebCore::TextPainter::clearGlyphDisplayLists):

8:32 AM Changeset in webkit [255073] by Jonathan Bedard

• 4 edits in trunk/Source/WTF

WTF: Remove old iOS version macros
​https://bugs.webkit.org/show_bug.cgi?id=206634
<rdar://problem/58818561>

Reviewed by Darin Adler.

• wtf/PlatformEnable.h: Remove iOS 12 version macro.
• wtf/PlatformHave.h: Ditto.
• wtf/PlatformUse.h: Ditto.

7:58 AM Changeset in webkit [255072] by Brent Fulgham

• 2 edits in trunk/Source/WebKit

[iOS] Add missing sysctl-read permission (already available in macOS sandbox)
​https://bugs.webkit.org/show_bug.cgi?id=206713
<rdar://problem/57665153>

Reviewed by Per Arne Vollan.

Correct a sandbox violation encountered when attempting to log the system hostname.
This was already handled properly on macOS, but the rule was not added to the iOS sandbox.

• Resources/SandboxProfiles/ios/com.apple.WebKit.WebContent.sb:

7:26 AM Changeset in webkit [255071] by Antti Koivisto

• 5 edits in trunk/Source/WebCore

[LFC][Integration] Clear inline item caches on low memory notification
​https://bugs.webkit.org/show_bug.cgi?id=206740
<rdar://problem/58773905>

Reviewed by Zalan Bujtas.

Clear inline item caches on low memory notification.

• layout/inlineformatting/InlineFormattingState.h:
• layout/integration/LayoutIntegrationLineLayout.cpp:

(WebCore::LayoutIntegration::LineLayout::releaseCaches):
(WebCore::LayoutIntegration::LineLayout::releaseInlineItemCache):

• layout/integration/LayoutIntegrationLineLayout.h:

Remove inline capacity. It is rarely optimal and we can afford the heap allocation.

• page/MemoryRelease.cpp:

(WebCore::releaseNoncriticalMemory):

7:14 AM Changeset in webkit [255070] by Adrian Perez de Castro

• 9 edits in trunk/Source

Fix various non-unified build issues introduced since r254751
​https://bugs.webkit.org/show_bug.cgi?id=206736

Reviewed by Carlos Garcia Campos.

Source/JavaScriptCore:

• tools/CompilerTimingScope.cpp: Add missing inclusions of wtf/Vector.h and wtf/text/WTFString.h

Source/WebCore:

No new tests needed.

• Modules/cache/DOMCacheEngine.cpp: Move errorToException() function out from the file to

make it usable from DOMCacheStorage.cpp

• Modules/cache/DOMCacheEngine.h:

(WebCore::DOMCacheEngine::errorToException): Added.

• Modules/cache/DOMCacheStorage.cpp:

(WebCore::DOMCacheStorage::retrieveCaches): Adapt to use DOMCacheEngine::errorToException().

• platform/graphics/cairo/ImageBufferUtilitiesCairo.cpp: Add missing RefPtrCairo.h inclusion.

Source/WebKit:

• WebProcess/WebPage/WebURLSchemeHandlerProxy.cpp: Add missing WebFrame.h inclusion.

7:14 AM Changeset in webkit [255069] by clopez@igalia.com

• 2 edits in trunk/Tools

[GTK] API test runner fails with error: too many open files.
​https://bugs.webkit.org/show_bug.cgi?id=206743

Reviewed by Carlos Garcia Campos.

The API test runner for google tests was leaking file descriptors because
it was not closing the file descriptor of the pseudo-terminal of the child,
which was referred in the parent because of the forkpty() call.
This was causing a small but constant increase of open file descriptors,
until the point it reached the limit of maximum open files (1024 usually)
causing this error.

Fix that by closing the reference to the fd of the child's pseudo-terminal.

• glib/api_test_runner.py:

(TestRunner._run_google_test):

6:36 AM Changeset in webkit [255068] by Alan Bujtas

• 4 edits in trunk/Source/WebCore

[LFC][Integration] LineLayout::paint should take advantage of the glyph display list cache
​https://bugs.webkit.org/show_bug.cgi?id=206727
<rdar://problem/58809696>

Reviewed by Antti Koivisto.

• layout/displaytree/DisplayInlineContent.cpp:

(WebCore::Display::InlineContent::~InlineContent):

• layout/displaytree/DisplayInlineContent.h:
• layout/integration/LayoutIntegrationLineLayout.cpp:

(WebCore::LayoutIntegration::LineLayout::paint):

6:11 AM Changeset in webkit [255067] by don.olmstead@sony.com

• 16 edits
  3 copies
  3 moves
  3 deletes in trunk/Source/WebKit

Centralize AuxiliaryProcessMainBase implementations
​https://bugs.webkit.org/show_bug.cgi?id=206700

Reviewed by Carlos Garcia Campos.

Non-Cocoa ports share a common usage of AuxiliaryProcessMainBase but the implementations
don't share a lot of code. This patch centralizes the code and the naming used.

• NetworkProcess/EntryPoint/unix/NetworkProcessMain.cpp:

(main):

• NetworkProcess/EntryPoint/win/NetworkProcessMain.cpp:

(main):

• NetworkProcess/NetworkProcessMain.h: Renamed from Source/WebKit/NetworkProcess/win/NetworkProcessMainWin.h.
• NetworkProcess/curl/NetworkProcessMainCurl.cpp: Renamed from Source/WebKit/NetworkProcess/win/NetworkProcessMainWin.cpp.

(WebKit::initializeAuxiliaryProcess<NetworkProcess>):
(WebKit::NetworkProcessMain):

• NetworkProcess/soup/NetworkProcessMainSoup.cpp:

(WebKit::NetworkProcessMain):
(WebKit::NetworkProcessMainUnix): Deleted.

• NetworkProcess/unix/NetworkProcessMainUnix.h: Removed.
• PlatformFTW.cmake:
• PlatformGTK.cmake:
• PlatformWPE.cmake:
• PlatformWin.cmake:
• PluginProcess/EntryPoint/unix/PluginProcessMain.cpp:

(main):

• PluginProcess/PluginProcessMain.h: Renamed from Source/WebKit/PluginProcess/unix/PluginProcessMainUnix.h.
• PluginProcess/unix/PluginProcessMainUnix.cpp:

(WebKit::PluginProcessMain):
(WebKit::PluginProcessMainUnix): Deleted.

• Shared/AuxiliaryProcessMain.h: Renamed from Source/WebKit/Shared/unix/AuxiliaryProcessMain.h.
• WebProcess/EntryPoint/unix/WebProcessMain.cpp:

(main):

• WebProcess/EntryPoint/win/WebProcessMain.cpp:

(main):

• WebProcess/WebProcessMain.h: Renamed from Source/WebKit/WebProcess/win/WebProcessMainWin.h.
• WebProcess/gtk/WebProcessMainGtk.cpp:

(WebKit::WebProcessMain):
(WebKit::WebProcessMainUnix): Deleted.

• WebProcess/playstation/WebProcessMainPlayStation.cpp: Renamed from Source/WebKit/WebProcess/unix/WebProcessMainUnix.h.

(WebKit::WebProcessMain):

• WebProcess/win/WebProcessMainWin.cpp:

(WebKit::WebProcessMain):
(WebKit::WebProcessMainWin): Deleted.

• WebProcess/wpe/WebProcessMainWPE.cpp:

(WebKit::WebProcessMain):
(WebKit::WebProcessMainUnix): Deleted.

• webkitglib-symbols.map:

5:53 AM Changeset in webkit [255066] by commit-queue@webkit.org

• 2 edits in trunk/LayoutTests

[GTK] More unreviewed gardening.
​https://bugs.webkit.org/show_bug.cgi?id=206693

Unreviewed gardening.

Patch by Lauro Moura <Lauro Moura> on 2020-01-24

• platform/gtk/TestExpectations:

3:32 AM Changeset in webkit [255065] by Pablo Saavedra

• 2 edits in trunk/LayoutTests

Layout Test animations/no-style-recalc-during-accelerated-animation.html is flaky
​https://bugs.webkit.org/show_bug.cgi?id=206733

Unreviewed test gardening.

• platform/gtk/TestExpectations:

3:18 AM Changeset in webkit [255064] by commit-queue@webkit.org

• 2 edits
  4 moves
  3 adds
  2 deletes in trunk/LayoutTests

Import 2 sets of grid-align-tests from WPT
​https://bugs.webkit.org/show_bug.cgi?id=206492

Patch by Rossana Monteriso <​rmonteriso@igalia.com> on 2020-01-24
Reviewed by Manuel Rego Casasnovas.

LayoutTests/imported/w3c:

Add outdated grid-align tests, checked and adapted to the WPT project, to the css-grid/alignment folder for imported tests.
Add also blue-100.png support file inside the css/support folder.
Correct the test description inside <p>, for grid-align-stretching-replaced-items.html.

• web-platform-tests/css/css-grid/alignment/grid-align-content-vertical-lr.html:
• web-platform-tests/css/css-grid/alignment/grid-align-content-vertical-rl.html:
• web-platform-tests/css/css-grid/alignment/grid-align-stretching-replaced-items.html:
• web-platform-tests/css/css-grid/alignment/grid-align-content-vertical-lr-expected.txt:
• web-platform-tests/css/css-grid/alignment/grid-align-content-vertical-rl-expected.txt:
• web-platform-tests/css/css-grid/alignment/grid-align-stretching-replaced-items-expected.txt:
• web-platform-tests/css/support/blue-100.png:

LayoutTests:

Remove outdated grid-align tests, that are being replaced by newer tests, adapted to the WPT project and imported
to the corresponding css-grid folder inside WebKit.

• fast/css-grid-layout/grid-align-content-vertical-lr-expected.txt: Removed.
• fast/css-grid-layout/grid-align-content-vertical-rl-expected.txt: Removed.
• fast/css-grid-layout/grid-align-stretching-replaced-items-expected.txt: Removed.
• fast/css-grid-layout/grid-align-stretching-replaced-items.html: Removed.
• fast/css-grid-layout/grid-align-content-vertical-lr.html: Removed.
• fast/css-grid-layout/grid-align-content-vertical-rl.html: Removed.

3:17 AM Changeset in webkit [255063] by commit-queue@webkit.org

• 2 edits in trunk/Source/JavaScriptCore

Fix number of callee saved register count on MIPS
​https://bugs.webkit.org/show_bug.cgi?id=206732

Patch by Paulo Matos <Paulo Matos> on 2020-01-24
Reviewed by Žan Doberšek.

MIPS has been issuing illegal instruction errors due to stack corruption.
This is caused by an incorrect number of CalleeSaveRegisterCount
that should be set to 3 since r254674.

• llint/LowLevelInterpreter.asm:

2:52 AM Changeset in webkit [255062] by zandobersek@gmail.com

• 2 edits in trunk/Tools

Unreviewed WPE dependency bump.

• wpe/jhbuild.modules: Bump the libwpe checkout commit to include fixes

for configuration errors that can occur with older (but still supported)
CMake versions.

1:03 AM Changeset in webkit [255061] by Carlos Garcia Campos

• 2 edits in trunk/Tools

Unreviewed GTK gardening. Mark /webkit/WebKitWebPage/get-uri as flaky

• TestWebKitAPI/glib/TestExpectations.json:

12:47 AM Changeset in webkit [255060] by Carlos Garcia Campos

• 2 edits in trunk/Source/WebKit

[GTK] Reduce the maximum time we wait for draw events in DrawingMonitor
​https://bugs.webkit.org/show_bug.cgi?id=206662

Reviewed by Carlos Alberto Lopez Perez.

1 second is too much, we can wait up to 100_ms instead and wait for the next frame after 16_ms. We should also
ensure there's a draw event since this is called from dispatchAfterEnsuringDrawing().

• UIProcess/CoordinatedGraphics/DrawingAreaProxyCoordinatedGraphics.cpp:

(WebKit::DrawingAreaProxyCoordinatedGraphics::DrawingMonitor::start):
(WebKit::DrawingAreaProxyCoordinatedGraphics::DrawingMonitor::didDraw):

12:37 AM Changeset in webkit [255059] by Antti Koivisto

• 9 edits
  2 adds in trunk

:matches() doesn't combine correctly with pseudo elements
​https://bugs.webkit.org/show_bug.cgi?id=206654

Reviewed by Simon Fraser.

Source/WebCore:

:matches() has been renamed to :is() in the spec (​https://www.w3.org/TR/selectors-4/#matches).
Update the :matches implementation to spec behavior regarding pseudo element handling.

"Pseudo-elements cannot be represented by the matches-any pseudo-class; they are not valid within :is()."

Selectors like

::matches(foo, .parent .child)::first-letter

should work while

::matches(:first-letter)

shoudn't match anything.

Test: fast/selectors/matches-with-pseudo-element.html

• css/SelectorChecker.cpp:

(WebCore::SelectorChecker::matchRecursively const):
(WebCore::SelectorChecker::checkOne const):

Don't provide pseudoId into ::matches subselectors. Pseudo elements are not suppported inside ::matches.
If one if found, fail that subselector.

• css/SelectorChecker.h:

LayoutTests:

• TestExpectations:

Unskip fast/css/first-letter-and-float-crash.html which no longer hits assert due to this.

• fast/selectors/matches-with-pseudo-element-expected.html: Added.
• fast/selectors/matches-with-pseudo-element.html: Added.
• fast/css/currentColor-on-before-after-pseudo-elements.html:
• fast/css/matches-specificity-5.html:
• fast/selectors/pseudo-element-inside-matches-expected.html:
• fast/selectors/pseudo-element-inside-matches.html:

Update the existing tests to match the behavior.

12:23 AM Changeset in webkit [255058] by youenn@apple.com

• 13 edits in trunk

Make sure fetch tasks go to network if service worker never gets to activated
​https://bugs.webkit.org/show_bug.cgi?id=206648

Reviewed by Chris Dumez.

Source/WebCore:

In case worker context process crashes, the SWServerWorker gets set to NotRunning.
If the SWServerWorker has pending activating completion handlers, they will never be called until the worker is destroyed.
But the worker may never be destroyed until its registration is destroyed.
This may trigger service worker fetch task hangs.

To fix this, make sure to call activating completion handlers whenever the SWServerWorker state is changed to either Terminating or NotRunning.

Covered by updated test.

• workers/service/server/SWServerWorker.cpp:

(WebCore::SWServerWorker::~SWServerWorker):
(WebCore::SWServerWorker::whenActivated):
(WebCore::SWServerWorker::setState):

• workers/service/server/SWServerWorker.h:

Source/WebKit:

In case activating completion handlers are not called, the fetch task timeout should kick in and make the load go to network process.
The issue is that our code was using the context connection to do so.
If the fetch task is waiting for the worker activation, the context connection might not be set and the timeout will be a no-op.

To fix this, the fetch task will do as if its context is closed when the timeout fires.
The fetck task now has a weak pointer to the WebSWServerConnection and will use to terminate the service worker as done previously.

We no longer handle all ongoing fetch tasks of the ongoing service worker.
Each individual fetch task timeout provides the same level of protection.
The service worker will anyway get terminated which will race to finalize the service worker fetch tasks with each of their timeout.

• NetworkProcess/ServiceWorker/ServiceWorkerFetchTask.cpp:

(WebKit::ServiceWorkerFetchTask::ServiceWorkerFetchTask):
(WebKit::ServiceWorkerFetchTask::timeoutTimerFired):

• NetworkProcess/ServiceWorker/ServiceWorkerFetchTask.h:
• NetworkProcess/ServiceWorker/WebSWServerConnection.cpp:

(WebKit::WebSWServerConnection::createFetchTask):
(WebKit::WebSWServerConnection::fetchTaskTimedOut):

• NetworkProcess/ServiceWorker/WebSWServerConnection.h:
• NetworkProcess/ServiceWorker/WebSWServerToContextConnection.cpp:

(WebKit::WebSWServerToContextConnection::fetchTaskTimedOut):

LayoutTests:

• http/wpt/service-workers/service-worker-spinning-activate.https-expected.txt:
• http/wpt/service-workers/service-worker-spinning-activate.https.html:

12:19 AM Changeset in webkit [255057] by youenn@apple.com

• 2 edits in trunk/Source/WebCore

Make sure DOMCacheStorage::retrieveCaches always calls its completionHandler
​https://bugs.webkit.org/show_bug.cgi?id=206647

Reviewed by Chris Dumez.

• Modules/cache/DOMCacheStorage.cpp:

(WebCore::DOMCacheStorage::retrieveCaches):
In case of context being stopped, make sure the completion handler is still called.