Timeline

Apr 22, 2022:

11:44 PM Changeset in webkit [293288] by cathiechen

• 8 edits
  4 adds in trunk

The layout is not updated when style.contain is changed from "size"/"inline-size" to empty string
​https://bugs.webkit.org/show_bug.cgi?id=239489

Reviewed by Rob Buis.

LayoutTests/imported/w3c:

• web-platform-tests/css/css-contain/contain-inline-size-removed-expected.html: Added.
• web-platform-tests/css/css-contain/contain-inline-size-removed.html: Added.
• web-platform-tests/css/css-contain/contain-size-removed-expected.html: Added.
• web-platform-tests/css/css-contain/contain-size-removed.html: Added.
• web-platform-tests/css/css-sizing/contain-intrinsic-size/auto-004-expected.txt:
• web-platform-tests/css/css-sizing/contain-intrinsic-size/auto-005-expected.txt:

Source/WebCore:

Unlike "layout", "paint" and "style" containment, being a "size" and "inline-size" containment or not affects the layout size of the element.
So it should require layout when a normal element becomes a "size"/"inline-size" containment, vice versa.

Tests: imported/w3c/web-platform-tests/css/css-contain/contain-inline-size-removed.html

imported/w3c/web-platform-tests/css/css-contain/contain-size-removed.html

• rendering/style/RenderStyle.cpp:

(WebCore::rareNonInheritedDataChangeRequiresLayout):
(WebCore::RenderStyle::effectiveContainment const): Deleted. Moved to StyleRareNonInheritedData.

• rendering/style/RenderStyle.h:

(WebCore::RenderStyle::effectiveContainment const):

• rendering/style/StyleRareNonInheritedData.cpp:

(WebCore::StyleRareNonInheritedData::effectiveContainment const):

• rendering/style/StyleRareNonInheritedData.h:

11:27 PM Changeset in webkit [293287] by commit-queue@webkit.org

• 3 edits
  2 adds in trunk

Should never be reached failure in WebCore::RenderElement::visibleInViewportStateChanged
​https://bugs.webkit.org/show_bug.cgi?id=191824

Patch by Rob Buis <​rbuis@igalia.com> on 2022-04-22
Reviewed by Eric Carlson.

Source/WebCore:

Provide an empty visibleInViewportStateChanged implementation
for RenderMedia to prevent hitting the ASSERT.

Test: media/audio-viewport-state-change-crash.html

• rendering/RenderMedia.h:

LayoutTests:

• media/audio-viewport-state-change-crash-expected.txt: Added.
• media/audio-viewport-state-change-crash.html: Added.

9:40 PM Changeset in webkit [293286] by Simon Fraser

• 2 edits in trunk/Source/WebCore

Crash under AsyncScrollingCoordinator::scrollingThreadAddedPendingUpdate()
​https://bugs.webkit.org/show_bug.cgi?id=239683
<rdar://92198272>

Reviewed by Alan Bujtas.

Crash data suggest that m_page can be null in
AsyncScrollingCoordinator::scheduleRenderingUpdate(), which does seem possible because
scheduleRenderingUpdate() is a dispatch from the scrolling thread, and
ScrollingCoordinator::pageDestroyed() may have run before it runs.

• page/scrolling/AsyncScrollingCoordinator.cpp:

(WebCore::AsyncScrollingCoordinator::scheduleRenderingUpdate):

9:21 PM Changeset in webkit [293285] by Chris Dumez

• 88 edits
  2 copies in trunk

Move global AtomStrings to a common header to promote reuse
​https://bugs.webkit.org/show_bug.cgi?id=239585

Reviewed by Geoff Garen and Darin Adler.

Source/WebCore:

• Headers.cmake:
• Modules/async-clipboard/Clipboard.cpp:

(WebCore::Clipboard::getType):

• Modules/mediastream/MediaStreamTrack.cpp:

(WebCore::MediaStreamTrack::contentHint const):
(WebCore::MediaStreamTrack::setContentHint):

• Sources.txt:
• WebCore.xcodeproj/project.pbxproj:
• animation/AnimationEffect.cpp:

(WebCore::AnimationEffect::getTiming const):
(WebCore::AnimationEffect::updateTiming):

• animation/EffectTiming.h:
• contentextensions/ContentExtensionParser.cpp:

(WebCore::ContentExtensions::isValidCSSSelector):

• css/CSSStyleSheet.h:
• css/FontFace.cpp:

(WebCore::FontFace::display const):

• css/MediaFeatureNames.cpp:

(WebCore::MediaFeatureNames::init):

• dom/DataTransfer.cpp:

(WebCore::normalizeType):
(WebCore::DataTransfer::setDataFromItemList):
(WebCore::DataTransfer::createForInputEvent):
(WebCore::DataTransfer::dropEffect const):

• dom/Document.cpp:

(WebCore::Document::designMode const):

• dom/InlineStyleSheetOwner.cpp:

(WebCore::isValidCSSContentType):

• dom/ProcessingInstruction.cpp:

(WebCore::ProcessingInstruction::checkStyleSheet):

• dom/mac/ImageControlsMac.cpp:
• editing/FontShadow.cpp:

(WebCore::serializationForCSS):

• editing/ReplaceSelectionCommand.cpp:

(WebCore::ReplaceSelectionCommand::removeRedundantStylesAndKeepStyleSpanInline):

• editing/cocoa/DataDetection.mm:
• editing/cocoa/HTMLConverter.mm:

(HTMLConverterCaches::propertyValueForNode):
(HTMLConverter::computedAttributesForElement):
(HTMLConverter::_enterElement):
(HTMLConverter::_processElement):

• editing/markup.cpp:

(WebCore::createFragmentForTransformToFragment):

• html/Autocapitalize.cpp:

(WebCore::stringForAutocapitalizeType):

• html/Autofill.cpp:

(WebCore::AutofillData::createFromHTMLFormControlElement):

• html/BaseCheckableInputType.cpp:

(WebCore::BaseCheckableInputType::saveFormControlState const):
(WebCore::BaseCheckableInputType::restoreFormControlState):
(WebCore::BaseCheckableInputType::fallbackValue const):

• html/EnterKeyHint.cpp:

(WebCore::attributeValueForEnterKeyHint):

• html/HTMLAudioElement.cpp:

(WebCore::HTMLAudioElement::createForLegacyFactoryFunction):

• html/HTMLButtonElement.cpp:

(WebCore::HTMLButtonElement::formControlType const):

• html/HTMLElement.cpp:

(WebCore::toValidDirValue):
(WebCore::HTMLElement::contentEditable const):
(WebCore::HTMLElement::setContentEditable):
(WebCore::HTMLElement::setAutocorrect):
(WebCore::plaintextOnlyName): Deleted.

• html/HTMLFormElement.cpp:

(WebCore::HTMLFormElement::autocomplete const):

• html/HTMLIFrameElement.cpp:

(WebCore::HTMLIFrameElement::loadingForBindings const):

• html/HTMLImageElement.cpp:

(WebCore::HTMLImageElement::decoding const):
(WebCore::HTMLImageElement::loadingForBindings const):

• html/HTMLMediaElement.cpp:

(WebCore::HTMLMediaElement::preload const):

• html/HTMLTextFormControlElement.cpp:

(WebCore::directionString):

• html/InputMode.cpp:

(WebCore::InputModeNames::none):
(WebCore::InputModeNames::text):
(WebCore::InputModeNames::tel):
(WebCore::InputModeNames::url):
(WebCore::InputModeNames::email):
(WebCore::InputModeNames::search):

• html/InputTypeNames.cpp:

(WebCore::InputTypeNames::email):
(WebCore::InputTypeNames::reset):
(WebCore::InputTypeNames::search):
(WebCore::InputTypeNames::submit):
(WebCore::InputTypeNames::telephone):
(WebCore::InputTypeNames::text):
(WebCore::InputTypeNames::url):

• html/shadow/TextControlInnerElements.cpp:

(WebCore::TextControlInnerTextElement::updateInnerTextElementEditabilityImpl):

• html/track/AudioTrack.cpp:

(WebCore::AudioTrack::isValidKind const):
(WebCore::AudioTrack::updateKindFromPrivate):
(WebCore::AudioTrack::alternativeKeyword): Deleted.
(WebCore::AudioTrack::mainKeyword): Deleted.
(WebCore::AudioTrack::commentaryKeyword): Deleted.

• html/track/AudioTrack.h:
• html/track/TextTrack.cpp:

(WebCore::TextTrack::TextTrack):
(WebCore::TextTrack::isValidKindKeyword):
(WebCore::TextTrack::kindKeyword const):
(WebCore::TextTrack::subtitlesKeyword): Deleted.
(WebCore::captionsKeyword): Deleted.

• html/track/TextTrack.h:
• html/track/VTTCue.cpp:

(WebCore::VTTCue::positionAlign const):
(WebCore::VTTCue::setPositionAlign):
(WebCore::VTTCue::toJSON const):
(WebCore::autoKeyword): Deleted.

• html/track/VideoTrack.cpp:

(WebCore::VideoTrack::isValidKind const):
(WebCore::VideoTrack::updateKindFromPrivate):
(WebCore::VideoTrack::alternativeKeyword): Deleted.
(WebCore::VideoTrack::captionsKeyword): Deleted.
(WebCore::VideoTrack::mainKeyword): Deleted.
(WebCore::VideoTrack::subtitlesKeyword): Deleted.
(WebCore::VideoTrack::commentaryKeyword): Deleted.

• html/track/VideoTrack.h:
• html/track/WebVTTParser.cpp:

(WebCore::WebVTTParser::WebVTTParser):

• inspector/agents/InspectorCSSAgent.cpp:

(WebCore::InspectorCSSAgent::createInspectorStyleSheetForDocument):

• loader/FormSubmission.cpp:

(WebCore::FormSubmission::Attributes::parseEncodingType):

• loader/NavigationScheduler.cpp:
• loader/TextResourceDecoder.cpp:

(WebCore::TextResourceDecoder::textFromUTF8):

• loader/archive/mhtml/MHTMLParser.cpp:

(WebCore::MHTMLParser::addResourceToArchive):

• loader/cache/CachedCSSStyleSheet.cpp:

(WebCore::CachedCSSStyleSheet::CachedCSSStyleSheet):

• mathml/MathMLPresentationElement.cpp:
• page/CaptionUserPreferencesMediaAF.cpp:

(WebCore::addAudioTrackKindDisplayNameIfNeeded):

• page/EventSource.cpp:

(WebCore::EventSource::EventSource):

• page/Page.cpp:

(WebCore::Page::userStyleSheet const):

• page/PageSerializer.cpp:

(WebCore::PageSerializer::serializeCSSStyleSheet):

• page/PrintContext.cpp:

(WebCore::PrintContext::pageProperty):

• page/ProcessWarming.cpp:

(WebCore::ProcessWarming::initializeNames):

• platform/CommonAtomStrings.cpp: Added.

(WebCore::initializeCommonAtomStrings):

• platform/CommonAtomStrings.h: Added.

(WebCore::alternativeAtom):
(WebCore::autoAtom):
(WebCore::captionsAtom):
(WebCore::commentaryAtom):
(WebCore::cssContentTypeAtom):
(WebCore::eagerAtom):
(WebCore::emailAtom):
(WebCore::falseAtom):
(WebCore::lazyAtom):
(WebCore::mainAtom):
(WebCore::noneAtom):
(WebCore::offAtom):
(WebCore::onAtom):
(WebCore::plaintextOnlyAtom):
(WebCore::resetAtom):
(WebCore::searchAtom):
(WebCore::submitAtom):
(WebCore::subtitlesAtom):
(WebCore::telAtom):
(WebCore::textAtom):
(WebCore::textPlainContentTypeAtom):
(WebCore::trueAtom):
(WebCore::urlAtom):

• platform/Pasteboard.cpp:

(WebCore::Pasteboard::isSafeTypeForDOMToReadAndWrite):

• platform/StaticPasteboard.cpp:

(WebCore::StaticPasteboard::writePlainText):

• platform/animation/Animation.cpp:

(WebCore::Animation::initialName const):

• platform/graphics/MediaPlayer.cpp:

(WebCore::MediaPlayer::load):
(WebCore::textPlain): Deleted.

• platform/ios/wak/WebCoreThread.mm:

(StartWebThread):

• platform/network/HTTPParsers.cpp:
• platform/network/ios/WebCoreURLResponseIOS.h:

(WebCore::shouldUseQuickLookForMIMEType):

• style/Styleable.cpp:

(WebCore::Styleable::animationListContainsNewlyValidAnimation const):
(WebCore::Styleable::updateCSSAnimations const):

• svg/SVGAnimateMotionElement.cpp:

(WebCore::SVGAnimateMotionElement::rotateMode const):

• svg/SVGAnimationElement.cpp:

(WebCore::sumAtom):
(WebCore::SVGAnimationElement::isAdditive const):
(WebCore::SVGAnimationElement::isAccumulated const):

• svg/SVGFEConvolveMatrixElement.cpp:
• svg/SVGFEConvolveMatrixElement.h:

(WebCore::SVGPropertyTraits<EdgeModeType>::toString):
(WebCore::SVGPropertyTraits<EdgeModeType>::fromString):

• svg/SVGMarkerTypes.h:

(WebCore::SVGPropertyTraits<SVGMarkerOrientType>::fromString):
(WebCore::SVGPropertyTraits<SVGMarkerOrientType>::toString):
(WebCore::SVGPropertyTraits<SVGMarkerOrientType>::autoString): Deleted.

• svg/SVGStyleElement.cpp:

(WebCore::SVGStyleElement::type const):

• svg/SVGTextPathElement.h:

(WebCore::SVGPropertyTraits<SVGTextPathSpacingType>::toString):
(WebCore::SVGPropertyTraits<SVGTextPathSpacingType>::fromString):

• svg/animation/SVGSMILElement.cpp:

(WebCore::indefiniteAtom):
(WebCore::SVGSMILElement::parseClockValue):
(WebCore::SVGSMILElement::repeatCount const):

• svg/properties/SVGPropertyTraits.h:
• workers/service/context/ServiceWorkerThread.cpp:

(WebCore::m_notificationClient):

• xml/XPathValue.cpp:

Source/WebKit:

• GPUProcess/GPUProcess.cpp:

(WebKit::GPUProcess::initializeGPUProcess):

• NetworkProcess/NetworkProcess.cpp:

(WebKit::NetworkProcess::initializeNetworkProcess):

• Shared/Cocoa/WebKit2InitializeCocoa.mm:

(WebKit::runInitializationCode):

• Shared/WebKit2Initialize.cpp:

(WebKit::InitializeWebKit2):

• UIProcess/API/APIContentRuleListStore.cpp:

(API::ContentRuleListStore::compileContentRuleList):

Source/WebKitLegacy/win:

• WebKitGraphics.cpp:

(makeFont):

Source/WTF:

• wtf/text/AtomString.cpp:

(WTF::AtomString::init):

• wtf/text/AtomString.h:

(WTF::xmlnsAtom):
(WTF::trueAtom): Deleted.
(WTF::falseAtom): Deleted.

Tools:

• TestWebKitAPI/Tests/WebCore/ContentExtensions.cpp:

(TestWebKitAPI::makeBackend):

8:28 PM Changeset in webkit [293284] by Russell Epstein

• 1 edit in branches/safari-613.2.7.1-branch/Source/WebCore/page/Page.cpp

Unreviewed build fix. rdar://80544133

./page/Page.cpp:389:9: error: incomplete type 'WebCore::FontCache' named in nested name specifier

8:27 PM Changeset in webkit [293283] by Russell Epstein

• 1 edit in branches/safari-613.2.7.0-branch/Source/WebCore/page/Page.cpp

Unreviewed build fix. rdar://80544133

./page/Page.cpp:389:9: error: incomplete type 'WebCore::FontCache' named in nested name specifier

7:58 PM Changeset in webkit [293282] by Russell Epstein

• 2 edits in branches/safari-613.2.7.1-branch/Source

Apply patch. rdar://problem/90968687

7:54 PM Changeset in webkit [293281] by Russell Epstein

• 18 edits in branches/safari-613.2.7.1-branch

Revert "Apply patch. rdar://problem/88904160"

This reverts commit r293269.

7:42 PM Changeset in webkit [293280] by Russell Epstein

• 9 edits
  2 copies in branches/safari-614.1.10-branch/Source

Cherry-pick r293106. rdar://problem/91740931

[WebGPU] WebKit and WebKitLegacy shouldn't need to know about WebGPU.framework
​https://bugs.webkit.org/show_bug.cgi?id=239528
<rdar://problem/91740931>

Unreviewed build fix.

Source/WebCore/PAL:

WebKit and WebKitLegacy create a WebGPU implementation from pal/graphics/WebGPU/Impl. This implementation
knows about WebGPU.framework. However, the create() function doesn't need to know about WebGPU.framework,
and its header doesn't need to include <WebGPU/WebGPU.h>. Only the .cpp files need to know about
<WebGPU/WebGPU.h>, and those .cpp files are invisible from WebKit and WebKitLegacy's points of view. So,
this patch splits the PAL header into:

1. A header which includes a creation function, but doesn't know about WebGPU.framework, and
2. A header which knows about WebGPU.framework, but doesn't hold the creation function used by WebKit or

WebKitLegacy.

This fixes one of Apple's internal builds.

• PAL.xcodeproj/project.pbxproj:
• pal/graphics/WebGPU/Impl/WebGPUCreateImpl.cpp: Copied from Source/WebCore/PAL/pal/graphics/WebGPU/Impl/WebGPUImpl.cpp. (PAL::WebGPU::create):
• pal/graphics/WebGPU/Impl/WebGPUCreateImpl.h: Copied from Source/WebCore/PAL/pal/graphics/WebGPU/Impl/WebGPUImpl.h.
• pal/graphics/WebGPU/Impl/WebGPUImpl.cpp: (PAL::WebGPU::GPUImpl::create): Deleted.
• pal/graphics/WebGPU/Impl/WebGPUImpl.h:

Source/WebKit:

• GPUProcess/graphics/WebGPU/RemoteGPU.cpp: (WebKit::RemoteGPU::workQueueInitialize):
• WebProcess/WebCoreSupport/WebChromeClient.cpp: (WebKit::WebChromeClient::createGPUForWebGPU const):

Source/WebKitLegacy/mac:

• WebCoreSupport/WebChromeClient.mm: (WebChromeClient::createGPUForWebGPU const):

git-svn-id: ​https://svn.webkit.org/repository/webkit/trunk@293106 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7:23 PM Changeset in webkit [293279] by Diego Pino Garcia

• 6 edits in trunk/Source

Unreviewed, non-unified build fixes after r293203

Source/JavaScriptCore:

• jit/ScratchRegisterAllocator.h:

Source/WebCore:

• page/PageConfiguration.h:

Source/WTF:

• wtf/RobinHoodHashTable.h:

7:15 PM Changeset in webkit [293278] by commit-queue@webkit.org

• 4 edits in trunk/Source/WebCore

Reduce sizeof(CSSNumericType)
​https://bugs.webkit.org/show_bug.cgi?id=239680

Patch by Alex Christensen <​achristensen@webkit.org> on 2022-04-22
Reviewed by Yusuke Suzuki.

• bindings/IDLTypes.h:

(WebCore::IDLType::nullValue):
(WebCore::IDLType::isNullType):
(WebCore::IDLType::extractValueFromNullable):

• css/typedom/numeric/CSSMathProduct.cpp:

(WebCore::multiplyTypes):

• css/typedom/numeric/CSSNumericType.h:

(WebCore::CSSNumericType::valueForType):
(WebCore::CSSNumericType::valueForType const):

7:05 PM Changeset in webkit [293277] by Russell Epstein

• 1 edit in branches/safari-613.2.7.1-branch/Source/WebCore/page/SecurityOrigin.h

Unreviewed build fix. rdar://90511155

WebCore/page/SecurityOrigin.h:309:13: note: candidate function not viable: requires 2 arguments, but 4 were provided inline void add(Hasher& hasher, const SecurityOrigin& origin)

7:02 PM Changeset in webkit [293276] by Russell Epstein

• 1 edit in branches/safari-613.2.7.0-branch/Source/WebCore/page/SecurityOrigin.h

Unreviewed build fix. rdar://90511155

WebCore/page/SecurityOrigin.h:309:13: note: candidate function not viable: requires 2 arguments, but 4 were provided inline void add(Hasher& hasher, const SecurityOrigin& origin)

5:44 PM Changeset in webkit [293275] by Russell Epstein

• 18 edits in branches/safari-613.2.7.0-branch

Revert "Apply patch. rdar://problem/88904160"

This reverts commit r293258.

5:44 PM Changeset in webkit [293274] by Russell Epstein

• 5 edits in branches/safari-613.2.7.0-branch/Source/WebKit

Revert "Apply patch. rdar://problem/88904160"

This reverts commit r293259.

5:37 PM Changeset in webkit [293273] by Russell Epstein

• 5 edits in branches/safari-613.2.7.1-branch/Source/WebKit

Revert "Apply patch. rdar://problem/88904160"

This reverts commit r293270.

5:37 PM Changeset in webkit [293272] by Russell Epstein

• 2 edits in branches/safari-613.2.7.1-branch/Source

Revert "Apply patch. rdar://problem/90968687"

This reverts commit r293271.

5:06 PM Changeset in webkit [293271] by Russell Epstein

• 2 edits in branches/safari-613.2.7.1-branch/Source

Apply patch. rdar://problem/90968687

5:04 PM Changeset in webkit [293270] by Russell Epstein

• 5 edits in branches/safari-613.2.7.1-branch/Source/WebKit

Apply patch. rdar://problem/88904160

5:04 PM Changeset in webkit [293269] by Russell Epstein

• 18 edits in branches/safari-613.2.7.1-branch

Apply patch. rdar://problem/88904160

4:59 PM Changeset in webkit [293268] by Russell Epstein

• 10 edits in branches/safari-613.2.7.1-branch

Cherry-pick r292274. rdar://problem/80544133

[Cocoa] Automatically relayout the page when new fonts are installed
​https://bugs.webkit.org/show_bug.cgi?id=238483
<rdar://problem/80544133>

Reviewed by Chris Dumez.

Source/WebCore:

This patch simply calls setNeedsRecalcStyleInAllFrames on every Page when we receive a
kCTFontManagerRegisteredFontsChangedNotification.

FontCache::invalidateAllFontCaches() can't do this directly because it's in platform/ and
therefore isn't allowed to know what Pages are. Instead, this patch takes a process-global
callback and calls that instead. This callback is set at initialization time.

Test: fast/text/install-font-style-recalc.html

• page/Page.cpp: (WebCore::m_contentSecurityPolicyModeForExtension): (WebCore::Page::firstTimeInitialization):
• page/Page.h:
• platform/graphics/FontCache.cpp: (WebCore::Function<void): (WebCore::FontCache::registerFontCacheInvalidationCallback): (WebCore::FontCache::invalidateAllFontCaches):
• platform/graphics/FontCache.h:

Tools:

Make TestRunner::dumpResourceLoadStatistics() clear any currently-recorded statistics.
This avoids the problem where spurious layouts during the time when the page has been created but
before the test has begun record irrelevant statistics.

• WebKitTestRunner/InjectedBundle/InjectedBundle.cpp: (WTR::InjectedBundle::beginTesting): (WTR::InjectedBundle::clearResourceLoadStatistics):
• WebKitTestRunner/InjectedBundle/InjectedBundle.h:
• WebKitTestRunner/InjectedBundle/TestRunner.cpp: (WTR::TestRunner::dumpResourceLoadStatistics):

LayoutTests:

• TestExpectations:
• fast/text/install-font-style-recalc-expected.txt: Added.
• fast/text/install-font-style-recalc.html: Added.
• platform/ios-wk2/TestExpectations:
• platform/mac-wk2/TestExpectations:

git-svn-id: ​https://svn.webkit.org/repository/webkit/trunk@292274 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4:59 PM Changeset in webkit [293267] by Russell Epstein

• 8 edits
  2 adds in branches/safari-613.2.7.1-branch

Cherry-pick r292072. rdar://problem/80544133

[Cocoa] Automatically relayout the page when new fonts are installed
​https://bugs.webkit.org/show_bug.cgi?id=238483
<rdar://problem/80544133>

Reviewed by Chris Dumez.

Source/WebCore:

This patch simply calls setNeedsRecalcStyleInAllFrames on every Page when we receive a
kCTFontManagerRegisteredFontsChangedNotification.

FontCache::invalidateAllFontCaches() can't do this directly because it's in platform/ and
therefore isn't allowed to know what Pages are. Instead, this patch takes a process-global
callback and calls that instead. This callback is set at initialization time.

Test: fast/text/install-font-style-recalc.html

• page/Page.cpp: (WebCore::m_attachmentElementClient):
• platform/graphics/FontCache.cpp: (WebCore::Function<void): (WebCore::FontCache::registerFontCacheInvalidationCallback): (WebCore::FontCache::invalidateAllFontCaches):
• platform/graphics/FontCache.h:

LayoutTests:

• fast/text/install-font-style-recalc-expected.txt: Added.
• fast/text/install-font-style-recalc.html: Added.

git-svn-id: ​https://svn.webkit.org/repository/webkit/trunk@292072 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4:39 PM Changeset in webkit [293266] by commit-queue@webkit.org

• 37 edits
  5 copies
  24 adds in trunk

Revert "[WebAuthn] Clean up WebAuthenticationModern and WebAuthnProcess"

This reverts commit 6ced3b519689536e6e4d36b9e46e78be9e6f4706.

Unreviewed test gardening.

Canonical link: ​https://commits.webkit.org/249907@main

4:21 PM Changeset in webkit [293265] by dbezhetskov

• 4 edits in trunk/Source/JavaScriptCore

[Refactoring] Reduce number of const and reinterpret casts
​https://bugs.webkit.org/show_bug.cgi?id=239648

Reviewed by Yusuke Suzuki.

The statement
"reinterpret_cast<TypeDefinition*>(const_cast<char*>(functionType.m_payload)

• sizeof(TypeDefinition))"

is dangerous and it produces warning on some platfroms.
There is way to avoid it and this patch is doint that.

• wasm/WasmSlowPaths.cpp:

(JSC::LLInt::doWasmCallIndirect):
(JSC::LLInt::doWasmCallRef):

• wasm/WasmTypeDefinition.h:

(JSC::Wasm::FunctionSignature::operator== const):

• wasm/WasmTypeDefinitionInlines.h:

4:19 PM Changeset in webkit [293264] by Patrick Angle

• 9 edits in trunk

Web Inspector: Regression(r287684) Resources from the memory cache show empty content in Network, Sources, and Search tabs
​https://bugs.webkit.org/show_bug.cgi?id=239667

Reviewed by Devin Rousso.

Source/WebCore:

Updated tests:

• http/tests/inspector/network/resource-response-source-disk-cache.html
• http/tests/inspector/network/resource-response-source-memory-cache.html

r287684 introduced a subtle bug when calling InspectorInstrumentation::didReceiveData. We rely
on there being a difference between real, but empty, data buffer and not having a data buffer,
but after r287684 an empty SharedBuffer would be passed in the later case. Instead, we should
continue to pass nullptr if there is no buffer so that InspectorNetworkAgent::didReceiveData
can distiguish between the two.

The bug is the result of having a non-nullptr data in InspectorNetworkAgent::didReceiveData,
which causes us to call maybeAddResourceData, which means by the time
InspectorNetworkAgent::getResponseBody is called, the ResourceData for the request will have
an empty, not non-existant, content(), which means we will return the empty content instead,
since we believe the response had actual content.

• inspector/InspectorInstrumentation.cpp:

(WebCore::InspectorInstrumentation::didReceiveDataImpl):

• inspector/InspectorInstrumentation.h:

(WebCore::InspectorInstrumentation::didReceiveData):

• loader/ResourceLoadNotifier.cpp:

(WebCore::ResourceLoadNotifier::dispatchDidReceiveData):

LayoutTests:

Add test steps to ensure that the resource has content, and that the base64Encoded value matches our
expectations. While only the memory cache was affected by this regression, add the test steps to both
the memory and disk caches to defend this functionality.

• http/tests/inspector/network/resource-response-source-memory-cache-expected.txt:
• http/tests/inspector/network/resource-response-source-memory-cache.html:
• http/tests/inspector/network/resource-response-source-disk-cache-expected.txt:
• http/tests/inspector/network/resource-response-source-disk-cache.html:

4:06 PM Changeset in webkit [293263] by Russell Epstein

• 10 edits in branches/safari-613.2.7.0-branch

Cherry-pick r292274. rdar://problem/80544133

[Cocoa] Automatically relayout the page when new fonts are installed
​https://bugs.webkit.org/show_bug.cgi?id=238483
<rdar://problem/80544133>

Reviewed by Chris Dumez.

Source/WebCore:

This patch simply calls setNeedsRecalcStyleInAllFrames on every Page when we receive a
kCTFontManagerRegisteredFontsChangedNotification.

FontCache::invalidateAllFontCaches() can't do this directly because it's in platform/ and
therefore isn't allowed to know what Pages are. Instead, this patch takes a process-global
callback and calls that instead. This callback is set at initialization time.

Test: fast/text/install-font-style-recalc.html

• page/Page.cpp: (WebCore::m_contentSecurityPolicyModeForExtension): (WebCore::Page::firstTimeInitialization):
• page/Page.h:
• platform/graphics/FontCache.cpp: (WebCore::Function<void): (WebCore::FontCache::registerFontCacheInvalidationCallback): (WebCore::FontCache::invalidateAllFontCaches):
• platform/graphics/FontCache.h:

Tools:

Make TestRunner::dumpResourceLoadStatistics() clear any currently-recorded statistics.
This avoids the problem where spurious layouts during the time when the page has been created but
before the test has begun record irrelevant statistics.

• WebKitTestRunner/InjectedBundle/InjectedBundle.cpp: (WTR::InjectedBundle::beginTesting): (WTR::InjectedBundle::clearResourceLoadStatistics):
• WebKitTestRunner/InjectedBundle/InjectedBundle.h:
• WebKitTestRunner/InjectedBundle/TestRunner.cpp: (WTR::TestRunner::dumpResourceLoadStatistics):

LayoutTests:

• TestExpectations:
• fast/text/install-font-style-recalc-expected.txt: Added.
• fast/text/install-font-style-recalc.html: Added.
• platform/ios-wk2/TestExpectations:
• platform/mac-wk2/TestExpectations:

git-svn-id: ​https://svn.webkit.org/repository/webkit/trunk@292274 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4:06 PM Changeset in webkit [293262] by Russell Epstein

• 8 edits
  2 adds in branches/safari-613.2.7.0-branch

Cherry-pick r292072. rdar://problem/80544133

[Cocoa] Automatically relayout the page when new fonts are installed
​https://bugs.webkit.org/show_bug.cgi?id=238483
<rdar://problem/80544133>

Reviewed by Chris Dumez.

Source/WebCore:

This patch simply calls setNeedsRecalcStyleInAllFrames on every Page when we receive a
kCTFontManagerRegisteredFontsChangedNotification.

FontCache::invalidateAllFontCaches() can't do this directly because it's in platform/ and
therefore isn't allowed to know what Pages are. Instead, this patch takes a process-global
callback and calls that instead. This callback is set at initialization time.

Test: fast/text/install-font-style-recalc.html

• page/Page.cpp: (WebCore::m_attachmentElementClient):
• platform/graphics/FontCache.cpp: (WebCore::Function<void): (WebCore::FontCache::registerFontCacheInvalidationCallback): (WebCore::FontCache::invalidateAllFontCaches):
• platform/graphics/FontCache.h:

LayoutTests:

• fast/text/install-font-style-recalc-expected.txt: Added.
• fast/text/install-font-style-recalc.html: Added.

git-svn-id: ​https://svn.webkit.org/repository/webkit/trunk@292072 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4:02 PM Changeset in webkit [293261] by Russell Epstein

• 20 edits
  4 adds in branches/safari-613.2.7.1-branch

Cherry-pick r291589. rdar://problem/90511155

LayoutTests/imported/w3c:
BroadcastChannel instances in distinct opaque origins can communicate
​https://bugs.webkit.org/show_bug.cgi?id=238090
<rdar://problem/90511155>

Reviewed by Alex Christensen.

Import web-platform-tests test coverage.

• web-platform-tests/webmessaging/broadcastchannel/opaque-origin-expected.txt: Added.
• web-platform-tests/webmessaging/broadcastchannel/opaque-origin.html: Added.
• web-platform-tests/webmessaging/broadcastchannel/w3c-import.log:

Source/WebCore:
BroadcastChannel instances in distinct opaque origins can communicate
​https://bugs.webkit.org/show_bug.cgi?id=238090
<rdar://problem/90511155>

Reviewed by Alex Christensen.

The issue is that we would use a ClientOrigin to distinguish origins for BroadcastChannel,
which relies on SecurityOriginData internally. A unique/opaque SecurityOrigin becomes an empty
SecurityOriginData upon conversion. As a result, when comparing ClientOrigin objects from
unique SecurityOrigins, they would compare as equal.

To address the issue, I introduced a new PartitionedSecurityOrigin type which is similar
to ClientOrigin but stores SecurityOrigin objects internally, instead of SecurityOriginData
objects. PartitionedSecurityOrigin's operator==() is such that different SecurityOrigins
would not be equal but the same unique SecurityOrigin would be. I then used this new
PartitionedSecurityOrigin type as key in our HashMap on the WebProcess side instead of
ClientOrigin. This allows communication between several BroadcastChannels from the same
unique origin, while preventing communication between distinct opaque origins.

When the PartitionedSecurityOrigin contains an opaque security origin, we don't involve
the Network Process at all since the destination can only be in the same WebProcess.

Test: imported/w3c/web-platform-tests/webmessaging/broadcastchannel/opaque-origin.html

• Headers.cmake:
• WebCore.xcodeproj/project.pbxproj:
• dom/BroadcastChannel.cpp: (WebCore::shouldPartitionOrigin): (WebCore::BroadcastChannel::MainThreadBridge::registerChannel): (WebCore::BroadcastChannel::MainThreadBridge::unregisterChannel): (WebCore::BroadcastChannel::MainThreadBridge::postMessage):
• dom/BroadcastChannelRegistry.h:
• loader/EmptyClients.cpp:
• page/PartitionedSecurityOrigin.h: Added. (WebCore::PartitionedSecurityOrigin::PartitionedSecurityOrigin): (WebCore::PartitionedSecurityOrigin::isHashTableDeletedValue const): (WebCore::PartitionedSecurityOrigin::isHashTableEmptyValue const): (WebCore::operator==): (WTF::add): (WTF::PartitionedSecurityOriginHash::hash): (WTF::PartitionedSecurityOriginHash::equal): (WTF::HashTraits<WebCore::PartitionedSecurityOrigin>::emptyValue): (WTF::HashTraits<WebCore::PartitionedSecurityOrigin>::constructEmptyValue): (WTF::HashTraits<WebCore::PartitionedSecurityOrigin>::isEmptyValue): (WTF::HashTraits<WebCore::PartitionedSecurityOrigin>::peek): (WTF::HashTraits<WebCore::PartitionedSecurityOrigin>::take):

Source/WebKit:
BroadcastChannel instances in distinct opaque origins can communicate
​https://bugs.webkit.org/show_bug.cgi?id=238090
<rdar://problem/90511155>

Reviewed by Alex Christensen.

The issue is that we would use a ClientOrigin to distinguish origins for BroadcastChannel,
which relies on SecurityOriginData internally. A unique/opaque SecurityOrigin becomes an empty
SecurityOriginData upon conversion. As a result, when comparing ClientOrigin objects from
unique SecurityOrigins, they would compare as equal.

To address the issue, I introduced a new PartitionedSecurityOrigin type which is similar
to ClientOrigin but stores SecurityOrigin objects internally, instead of SecurityOriginData
objects. PartitionedSecurityOrigin's operator==() is such that different SecurityOrigins
would not be equal but the same unique SecurityOrigin would be. I then used this new
PartitionedSecurityOrigin type as key in our HashMap on the WebProcess side instead of
ClientOrigin. This allows communication between several BroadcastChannels from the same
unique origin, while preventing communication between distinct opaque origins.

When the PartitionedSecurityOrigin contains an opaque security origin, we don't involve
the Network Process at all since the destination can only be in the same WebProcess.

• WebProcess/WebCoreSupport/WebBroadcastChannelRegistry.cpp: (WebKit::toClientOrigin): (WebKit::WebBroadcastChannelRegistry::registerChannel): (WebKit::WebBroadcastChannelRegistry::unregisterChannel): (WebKit::WebBroadcastChannelRegistry::postMessage): (WebKit::WebBroadcastChannelRegistry::postMessageLocally): (WebKit::WebBroadcastChannelRegistry::postMessageToRemote): (WebKit::WebBroadcastChannelRegistry::networkProcessCrashed):
• WebProcess/WebCoreSupport/WebBroadcastChannelRegistry.h:

Source/WebKitLegacy:
Dust off Mac CMake build
​https://bugs.webkit.org/show_bug.cgi?id=238121

Reviewed by Yusuke Suzuki.

• PlatformMac.cmake:

git-svn-id: ​https://svn.webkit.org/repository/webkit/trunk@291589 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3:52 PM Changeset in webkit [293260] by Simon Fraser

• 7 edits
  2 adds in trunk

Focusing scroll container before scrolling breaks smooth scrolling
​https://bugs.webkit.org/show_bug.cgi?id=239605

Reviewed by Wenson Hsieh.

Source/WebCore:

If script calls focus() and then scrollTo({ behavior: smooth }) in the same runloop, the
scrollToFocusedElementInternal() that happens on a zero-delay timer in FrameView will stop
the animated scroll.

Fix by having programmatic scrolls on Element and DOMWindow cancel any pending focus-related
scroll.

Test: fast/scrolling/programmatic-smooth-scroll-after-focus.html

• dom/Element.cpp:

(WebCore::Element::scrollTo):

• page/DOMWindow.cpp:

(WebCore::DOMWindow::scrollTo const):

• page/FrameView.cpp:

(WebCore::FrameView::maintainScrollPositionAtAnchor):
(WebCore::FrameView::setScrollPosition):
(WebCore::FrameView::cancelScheduledScrollToFocusedElement):
(WebCore::FrameView::scrollToAnchor):
(WebCore::FrameView::setWasScrolledByUser):

• page/FrameView.h:

LayoutTests:

• fast/scrolling/programmatic-smooth-scroll-after-focus-expected.txt: Added.
• fast/scrolling/programmatic-smooth-scroll-after-focus.html: Added.
• platform/win/TestExpectations:

3:45 PM Changeset in webkit [293259] by Russell Epstein

• 5 edits in branches/safari-613.2.7.0-branch/Source/WebKit

Apply patch. rdar://problem/88904160

3:45 PM Changeset in webkit [293258] by Russell Epstein

• 18 edits in branches/safari-613.2.7.0-branch

Apply patch. rdar://problem/88904160

3:05 PM Changeset in webkit [293257] by Russell Epstein

• 4 edits
  2 adds in branches/safari-613.2.7.0-branch

Cherry-pick r291724. rdar://problem/91975589

Expose the AX tree of PDFs loaded via iframes on the Mac
​https://bugs.webkit.org/show_bug.cgi?id=238168

Reviewed by Chris Fleizach.

Source/WebKit:

Given this markup:

<iframe src="/path/to/paystub.pdf">

An accessibility tree is built for paystub.pdf, but we don't expose it
to AX clients, making the PDF entirely inaccessible.

This happens because we were setting the AX parent to be the webpage
for full-frame PDF plugins (which an iframe with this markup is).
This behavior is correct only for main-frame (i.e. not iframe)
full-frame plugins, so this patch adds an extra condition to that logic.

Test: accessibility/mac/iframe-pdf.html

• WebProcess/Plugins/PDF/PDFPlugin.mm: (WebKit::PDFPlugin::PDFPlugin):

LayoutTests:

• accessibility/mac/iframe-pdf-expected.txt: Added.
• accessibility/mac/iframe-pdf.html: Added.
• platform/mac-wk1/TestExpectations: Skip new test as it always times out in WK1 (similar to accessibility/mac/basic-embed-pdf-accessibility.html)

git-svn-id: ​https://svn.webkit.org/repository/webkit/trunk@291724 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2:56 PM Changeset in webkit [293256] by Russell Epstein

• 2 edits in branches/safari-613.2.7.0-branch/Source

Apply patch. rdar://problem/90968687

2:53 PM Changeset in webkit [293255] by Russell Epstein

• 20 edits
  3 adds in branches/safari-613.2.7.0-branch

Cherry-pick r291589. rdar://problem/90511155

LayoutTests/imported/w3c:
BroadcastChannel instances in distinct opaque origins can communicate
​https://bugs.webkit.org/show_bug.cgi?id=238090
<rdar://problem/90511155>

Reviewed by Alex Christensen.

Import web-platform-tests test coverage.

• web-platform-tests/webmessaging/broadcastchannel/opaque-origin-expected.txt: Added.
• web-platform-tests/webmessaging/broadcastchannel/opaque-origin.html: Added.
• web-platform-tests/webmessaging/broadcastchannel/w3c-import.log:

Source/WebCore:
BroadcastChannel instances in distinct opaque origins can communicate
​https://bugs.webkit.org/show_bug.cgi?id=238090
<rdar://problem/90511155>

Reviewed by Alex Christensen.

The issue is that we would use a ClientOrigin to distinguish origins for BroadcastChannel,
which relies on SecurityOriginData internally. A unique/opaque SecurityOrigin becomes an empty
SecurityOriginData upon conversion. As a result, when comparing ClientOrigin objects from
unique SecurityOrigins, they would compare as equal.

To address the issue, I introduced a new PartitionedSecurityOrigin type which is similar
to ClientOrigin but stores SecurityOrigin objects internally, instead of SecurityOriginData
objects. PartitionedSecurityOrigin's operator==() is such that different SecurityOrigins
would not be equal but the same unique SecurityOrigin would be. I then used this new
PartitionedSecurityOrigin type as key in our HashMap on the WebProcess side instead of
ClientOrigin. This allows communication between several BroadcastChannels from the same
unique origin, while preventing communication between distinct opaque origins.

When the PartitionedSecurityOrigin contains an opaque security origin, we don't involve
the Network Process at all since the destination can only be in the same WebProcess.

Test: imported/w3c/web-platform-tests/webmessaging/broadcastchannel/opaque-origin.html

• Headers.cmake:
• WebCore.xcodeproj/project.pbxproj:
• dom/BroadcastChannel.cpp: (WebCore::shouldPartitionOrigin): (WebCore::BroadcastChannel::MainThreadBridge::registerChannel): (WebCore::BroadcastChannel::MainThreadBridge::unregisterChannel): (WebCore::BroadcastChannel::MainThreadBridge::postMessage):
• dom/BroadcastChannelRegistry.h:
• loader/EmptyClients.cpp:
• page/PartitionedSecurityOrigin.h: Added. (WebCore::PartitionedSecurityOrigin::PartitionedSecurityOrigin): (WebCore::PartitionedSecurityOrigin::isHashTableDeletedValue const): (WebCore::PartitionedSecurityOrigin::isHashTableEmptyValue const): (WebCore::operator==): (WTF::add): (WTF::PartitionedSecurityOriginHash::hash): (WTF::PartitionedSecurityOriginHash::equal): (WTF::HashTraits<WebCore::PartitionedSecurityOrigin>::emptyValue): (WTF::HashTraits<WebCore::PartitionedSecurityOrigin>::constructEmptyValue): (WTF::HashTraits<WebCore::PartitionedSecurityOrigin>::isEmptyValue): (WTF::HashTraits<WebCore::PartitionedSecurityOrigin>::peek): (WTF::HashTraits<WebCore::PartitionedSecurityOrigin>::take):

Source/WebKit:
BroadcastChannel instances in distinct opaque origins can communicate
​https://bugs.webkit.org/show_bug.cgi?id=238090
<rdar://problem/90511155>

Reviewed by Alex Christensen.

The issue is that we would use a ClientOrigin to distinguish origins for BroadcastChannel,
which relies on SecurityOriginData internally. A unique/opaque SecurityOrigin becomes an empty
SecurityOriginData upon conversion. As a result, when comparing ClientOrigin objects from
unique SecurityOrigins, they would compare as equal.

To address the issue, I introduced a new PartitionedSecurityOrigin type which is similar
to ClientOrigin but stores SecurityOrigin objects internally, instead of SecurityOriginData
objects. PartitionedSecurityOrigin's operator==() is such that different SecurityOrigins
would not be equal but the same unique SecurityOrigin would be. I then used this new
PartitionedSecurityOrigin type as key in our HashMap on the WebProcess side instead of
ClientOrigin. This allows communication between several BroadcastChannels from the same
unique origin, while preventing communication between distinct opaque origins.

When the PartitionedSecurityOrigin contains an opaque security origin, we don't involve
the Network Process at all since the destination can only be in the same WebProcess.

• WebProcess/WebCoreSupport/WebBroadcastChannelRegistry.cpp: (WebKit::toClientOrigin): (WebKit::WebBroadcastChannelRegistry::registerChannel): (WebKit::WebBroadcastChannelRegistry::unregisterChannel): (WebKit::WebBroadcastChannelRegistry::postMessage): (WebKit::WebBroadcastChannelRegistry::postMessageLocally): (WebKit::WebBroadcastChannelRegistry::postMessageToRemote): (WebKit::WebBroadcastChannelRegistry::networkProcessCrashed):
• WebProcess/WebCoreSupport/WebBroadcastChannelRegistry.h:

Source/WebKitLegacy:
Dust off Mac CMake build
​https://bugs.webkit.org/show_bug.cgi?id=238121

Reviewed by Yusuke Suzuki.

• PlatformMac.cmake:

git-svn-id: ​https://svn.webkit.org/repository/webkit/trunk@291589 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2:24 PM Changeset in webkit [293254] by Matteo Flores

• 2 edits
  1 copy in trunk/LayoutTests

REBASLINE: [ Monterey ] fast/text/khmer-lao-font.html is a constant text failure

​https://bugs.webkit.org/show_bug.cgi?id=238917

Unreviewed test gardening.

• platform/mac-bigsur/fast/text/khmer-lao-font-expected.txt: Copied from LayoutTests/platform/mac/fast/text/khmer-lao-font-expected.txt.
• platform/mac/fast/text/khmer-lao-font-expected.txt:

Canonical link: ​https://commits.webkit.org/249903@main

2:08 PM Changeset in webkit [293253] by commit-queue@webkit.org

• 2 edits in trunk

Add my github account name
​https://bugs.webkit.org/show_bug.cgi?id=239671

Patch by Alex Christensen <​achristensen@webkit.org> on 2022-04-22
Reviewed by Chris Dumez.

• metadata/contributors.json:

1:58 PM Changeset in webkit [293252] by commit-queue@webkit.org

• 3 edits in trunk/Source/JavaScriptCore

[JSC]Throw consistent exceptions for memory.init and memory.copy
​https://bugs.webkit.org/show_bug.cgi?id=239592

Patch by Geza Lore <Geza Lore> on 2022-04-22
Reviewed by Yusuke Suzuki.

For a trapping Wasm memory.init and memory.copy instruction, the LLInt
used to throw OutOfBoundsMemoryAccess, but BBQ/OMG used to throw
OutOfBoundsTableAccess.

Changed BBQ/OMG to throw OutOfBoundsMemoryAccess as well.

• wasm/WasmAirIRGenerator.cpp:

(JSC::Wasm::AirIRGenerator::addMemoryCopy):
(JSC::Wasm::AirIRGenerator::addMemoryInit):

• wasm/WasmB3IRGenerator.cpp:

(JSC::Wasm::B3IRGenerator::addMemoryInit):
(JSC::Wasm::B3IRGenerator::addMemoryCopy):

1:48 PM Changeset in webkit [293251] by Alan Bujtas

• 5 edits
  3 adds in trunk/Source/WebCore

[LFC][Integration] Add blank LayoutIntegration::FlexLayout
​https://bugs.webkit.org/show_bug.cgi?id=239663

Reviewed by Antti Koivisto.

This is in preparation for adding flex formatting context integration.

• CMakeLists.txt:
• Headers.cmake:
• Sources.txt:
• WebCore.xcodeproj/project.pbxproj:
• layout/integration/flex/LayoutIntegrationFlexLayout.cpp: Added.

(WebCore::LayoutIntegration::FlexLayout::FlexLayout):
(WebCore::LayoutIntegration::FlexLayout::updateFormattingRootGeometryAndInvalidate):
(WebCore::LayoutIntegration::FlexLayout::updateFlexItemDimensions):
(WebCore::LayoutIntegration::FlexLayout::updateStyle):
(WebCore::LayoutIntegration::FlexLayout::computeIntrinsicWidthConstraints):
(WebCore::LayoutIntegration::FlexLayout::layout):
(WebCore::LayoutIntegration::FlexLayout::paint):
(WebCore::LayoutIntegration::FlexLayout::hitTest):
(WebCore::LayoutIntegration::FlexLayout::collectOverflow):
(WebCore::LayoutIntegration::FlexLayout::contentLogicalHeight const):

• layout/integration/flex/LayoutIntegrationFlexLayout.h: Added.

1:43 PM Changeset in webkit [293250] by Russell Epstein

• 2 edits in branches/safari-613.2.7.1-branch/Source/WebCore

Cherry-pick r292878. rdar://problem/90248865

ScriptDisallowedScope::isEventAllowedInMainThread assert failure when activating AudioSession
​https://bugs.webkit.org/show_bug.cgi?id=239343

Reviewed by Eric Carlson.

As part of HTMLMediaElement::clearMediaPlayer we call PlatformMediaSession::canProduceAudioChanged
which can result in activating the AudioSession when the page is capturing audio.

This sends a synchronous IPC message to the GPU process. As part of IPC::Connection::waitForSyncReply,
we also end up dispatching enqueued messages, including a WebPage_EndPrinting message that
fires an event listener on the main thread.

This patch will instead queue the PlatformMediaSession::canProduceAudioChanged as a task to avoid
firing the event listener on the main thread, which results in assertion failure.

• html/HTMLMediaElement.cpp: (WebCore::HTMLMediaElement::clearMediaPlayer):

git-svn-id: ​https://svn.webkit.org/repository/webkit/trunk@292878 268f45cc-cd09-0410-ab3c-d52691b4dbfc

1:43 PM Changeset in webkit [293249] by Russell Epstein

• 2 edits in branches/safari-613.2.7.0-branch/Source/WebCore

Cherry-pick r292878. rdar://problem/90248865

ScriptDisallowedScope::isEventAllowedInMainThread assert failure when activating AudioSession
​https://bugs.webkit.org/show_bug.cgi?id=239343

Reviewed by Eric Carlson.

As part of HTMLMediaElement::clearMediaPlayer we call PlatformMediaSession::canProduceAudioChanged
which can result in activating the AudioSession when the page is capturing audio.

This sends a synchronous IPC message to the GPU process. As part of IPC::Connection::waitForSyncReply,
we also end up dispatching enqueued messages, including a WebPage_EndPrinting message that
fires an event listener on the main thread.

This patch will instead queue the PlatformMediaSession::canProduceAudioChanged as a task to avoid
firing the event listener on the main thread, which results in assertion failure.

• html/HTMLMediaElement.cpp: (WebCore::HTMLMediaElement::clearMediaPlayer):

git-svn-id: ​https://svn.webkit.org/repository/webkit/trunk@292878 268f45cc-cd09-0410-ab3c-d52691b4dbfc

1:39 PM Changeset in webkit [293248] by Russell Epstein

• 6 edits in branches/safari-613.2.7.1-branch

Apply patch. rdar://problem/91446317

1:39 PM Changeset in webkit [293247] by Russell Epstein

• 6 edits in branches/safari-613.2.7.1-branch

Revert "Cherry-pick r291606. rdar://problem/91446317"

This reverts commit r293072.

1:39 PM Changeset in webkit [293246] by Russell Epstein

• 1 edit in branches/safari-613.2.7.1-branch/Source/WebKit/UIProcess/WebPageProxy.cpp

Revert "Unreviewed build fix. rdar://91446317"

This reverts commit r293144.

1:35 PM Changeset in webkit [293245] by Russell Epstein

• 6 edits in branches/safari-613.2.7.0-branch

Apply patch. rdar://problem/91446317

1:35 PM Changeset in webkit [293244] by Russell Epstein

• 6 edits in branches/safari-613.2.7.0-branch

Revert "Cherry-pick r291606. rdar://problem/91446317"

This reverts commit r293072.

1:35 PM Changeset in webkit [293243] by Russell Epstein

• 1 edit in branches/safari-613.2.7.0-branch/Source/WebKit/UIProcess/WebPageProxy.cpp

Revert "Unreviewed build fix. rdar://91446317"

This reverts commit r293144.

1:18 PM Changeset in webkit [293242] by Chris Dumez

• 2 edits in trunk

Unreviewed, add Geoff's github username to contributors.json so he can review patches

• metadata/contributors.json:

1:14 PM Changeset in webkit [293241] by Russell Epstein

• 10 edits in branches/safari-613.2.7.1-branch

Apply patch. rdar://problem/91288849

1:14 PM Changeset in webkit [293240] by Russell Epstein

• 10 edits in branches/safari-613.2.7.0-branch

Apply patch. rdar://problem/91288849

1:08 PM Changeset in webkit [293239] by Russell Epstein

• 2 edits in branches/safari-613.2.7.1-branch

Apply patch. rdar://problem/91590827

1:08 PM Changeset in webkit [293238] by Russell Epstein

• 2 edits in branches/safari-613.2.7.0-branch

Apply patch. rdar://problem/91590827

12:58 PM Changeset in webkit [293237] by Chris Dumez

• 5 edits in trunk/Source

Modernize Text::replaceWholeText() ​https://bugs.webkit.org/show_bug.cgi?id=239637

Reviewed by Ryosuke Niwa.

Modernize Text::replaceWholeText():

• Drop unnecessary return value.
• Ref the Text node at call sites instead of using an internal protectedThis variable.
• Use node instead of n for variable name.
• Use modern template deduction for Ref / RefPtr.

• Source/WebKitLegacy/mac/DOM/DOMText.mm:

(-[DOMText replaceWholeText:]):

• Source/WebCore/dom/Text.cpp:

(WebCore::Text::replaceWholeText):

• Source/WebCore/dom/Text.h:
• Source/WebCore/html/shadow/DateTimeFieldElement.cpp:

(WebCore::DateTimeFieldElement::updateVisibleValue):

Canonical link: ​https://commits.webkit.org/249898@main

12:02 PM Changeset in webkit [293236] by Russell Epstein

• 4 edits
  2 adds in branches/safari-613.2.7.1-branch

Cherry-pick r291724. rdar://problem/91975589

Expose the AX tree of PDFs loaded via iframes on the Mac
​https://bugs.webkit.org/show_bug.cgi?id=238168

Reviewed by Chris Fleizach.

Source/WebKit:

Given this markup:

<iframe src="/path/to/paystub.pdf">

An accessibility tree is built for paystub.pdf, but we don't expose it
to AX clients, making the PDF entirely inaccessible.

This happens because we were setting the AX parent to be the webpage
for full-frame PDF plugins (which an iframe with this markup is).
This behavior is correct only for main-frame (i.e. not iframe)
full-frame plugins, so this patch adds an extra condition to that logic.

Test: accessibility/mac/iframe-pdf.html

• WebProcess/Plugins/PDF/PDFPlugin.mm: (WebKit::PDFPlugin::PDFPlugin):

LayoutTests:

• accessibility/mac/iframe-pdf-expected.txt: Added.
• accessibility/mac/iframe-pdf.html: Added.
• platform/mac-wk1/TestExpectations: Skip new test as it always times out in WK1 (similar to accessibility/mac/basic-embed-pdf-accessibility.html)

git-svn-id: ​https://svn.webkit.org/repository/webkit/trunk@291724 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11:46 AM Changeset in webkit [293235] by Russell Epstein

• 9 edits in branches/safari-613.2.7.1-branch/Source

Versioning.

WebKit-7613.2.7.1.1

11:46 AM Changeset in webkit [293234] by Russell Epstein

• 9 edits in branches/safari-613.2.7.0-branch/Source

Versioning.

WebKit-7613.2.7.0.1

11:25 AM Changeset in webkit [293233] by Alan Bujtas

• 7 edits in trunk/Source

[LFC][Integration] Add FlexFormattingContextIntegrationEnabled setting
​https://bugs.webkit.org/show_bug.cgi?id=239654

Reviewed by Antti Koivisto.

Source/WebCore:

This is in preparation for supporting flex formatting context through integration.

• layout/integration/LayoutIntegrationCoverage.cpp:

(WebCore::LayoutIntegration::canUseForFlexLayout):

• layout/integration/LayoutIntegrationCoverage.h:
• rendering/RenderFlexibleBox.cpp:

(WebCore::RenderFlexibleBox::layoutFlexItems):
(WebCore::RenderFlexibleBox::layoutUsingFlexFormattingContext):

• rendering/RenderFlexibleBox.h:

Source/WTF:

• Scripts/Preferences/WebPreferencesInternal.yaml:

11:02 AM Changeset in webkit [293232] by Brandon

• 5 edits in trunk/Source/WebCore

Store StyleScope during CSSStyleSheet Creation
​https://bugs.webkit.org/show_bug.cgi?id=239659

Reviewed by Antti Koivisto.

To ensure that we can always access the style scope, we shall ensure we store the style scope
at CSSStyleSheet creation time. It was possible before that a style sheet could become disconnected
and then a rule would later be mutated. This resulted in the unfortunate side effect of being unable to find
the style scope.

• css/CSSStyleSheet.cpp:

(WebCore::CSSStyleSheet::CSSStyleSheet):
(WebCore::CSSStyleSheet::styleScope):

• css/CSSStyleSheet.h:
• style/StyleScope.cpp:

(WebCore::Style::Scope::collectResolverScopes):

• style/StyleScope.h:

10:46 AM Changeset in webkit [293231] by Aditya Keerthi

• 12 edits in trunk

[iOS] Add find interaction API to WKWebView
​https://bugs.webkit.org/show_bug.cgi?id=239612
rdar://88442918

Reviewed by Alex Christensen.

Source/WebKit:

Promote the _findInteraction and _findInteractionEnabled SPI to
findInteraction and findInteractionEnabled API.

In order to make this change, new UIKit objects and protocols must be
adopted. Existing (deprecated) protocol conformance is kept around to
avoid breaking internal clients.

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

(-[WKWebView findInteractionEnabled]):
(-[WKWebView setFindInteractionEnabled:]):
(-[WKWebView findInteraction]):

• UIProcess/API/Cocoa/WKWebViewInternal.h:
• UIProcess/API/Cocoa/WKWebViewPrivate.h:

Declare conformance to three protocols: _UITextSearching, UITextSearching, UIFindInteractionDelegate.

_UITextSearching: Conformance is kept around to avoid breaking internal clients.
UITextSearching: Conformance is declared for clients to migrate off of _UITextSearching.
UIFindInteractionDelegate: Conformance is necessary to begin a UIFindSession.

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

(-[WKWebView _didCommitLoadForMainFrame]):

The "searchable object" does not need to be reset anymore, as UIKit
will do that in -[WKWebView findInteraction:sessionForView:] the next
time a find session begins.

(-[WKWebView _searchableObject]):
(-[WKWebView _setFindInteractionEnabled:]):
(-[WKWebView _findInteraction]):

It is safe to cast between the SPI and API version of this object. UIKit
has guaranteed compatibility.

(-[WKWebView offsetFromPosition:toPosition:inDocument:]):
(-[WKWebView compareFoundRange:toRange:inDocument:]):
(-[WKWebView performTextSearchWithQueryString:usingOptions:resultAggregator:]):
(-[WKWebView replaceFoundTextInRange:inDocument:withText:]):
(-[WKWebView decorateFoundTextRange:inDocument:usingStyle:]):
(-[WKWebView scrollRangeToVisible:inDocument:]):
(-[WKWebView findInteraction:sessionForView:]):
(-[WKWebView findInteraction:didBeginFindSession:]):

Custom content views do not need to be aware that a find session has
begun, as they do not draw an overlay.

(-[WKWebView findInteraction:didEndFindSession:]):

Custom content views do not need to be aware that a find session has
begun, as they do not draw an overlay.

• UIProcess/ios/WKContentViewInteraction.h:

Declare the old _UITextSearching methods explicitly, as we need to
maintain support for internal clients.

• UIProcess/ios/WKContentViewInteraction.mm:

(-[WKContentView performTextSearchWithQueryString:usingOptions:resultAggregator:]):
(-[WKContentView replaceFoundTextInRange:inDocument:withText:]):
(-[WKContentView decorateFoundTextRange:inDocument:usingStyle:]):
(-[WKContentView scrollRangeToVisible:inDocument:]):
(-[WKContentView offsetFromPosition:toPosition:inDocument:]):
(-[WKContentView compareFoundRange:toRange:inDocument:]):

• UIProcess/ios/WKPDFView.mm:

(-[WKPDFView compareFoundRange:toRange:inDocument:]):
(-[WKPDFView performTextSearchWithQueryString:usingOptions:resultAggregator:]):
(-[WKPDFView decorateFoundTextRange:inDocument:usingStyle:]):

Tools:

Augment an existing API test to use the new API, in addition to the
existing SPI.

The find-and-replace test is updated to account for the fact that a
find interaction no longer exposes a searchableObject. Instead, use
test-only SPI on WKWebView to perform the replace.

• TestWebKitAPI/Tests/WebKitCocoa/FindInPage.mm:

(TEST):

10:28 AM Changeset in webkit [293230] by Wenson Hsieh

• 9 edits in trunk

[iOS] -selectWordForReplacement should select dictation alternatives that span multiple words
​https://bugs.webkit.org/show_bug.cgi?id=239622
rdar://91416535

Reviewed by Aditya Keerthi.

Source/WebKit:

Currently, -selectWordForReplacement (which is invoked by UIKit when the tapping on spelling corrections and
dictation alternatives to reveal the callout bar with text suggestions) handles multi-word dictation ranges by
only selecting one of the words (i.e., whichever word is closest to the tapped location). This makes it
impossible to handle dictation ranges that span multiple words, since only part of the range is replaced with
the chosen alternative.

To address this, instead of using the existing extendSelection method (which is still exercised when tapping
the "Select" action in the callout bar with a caret selection), we add extendSelectionForReplacement, which
extends the selection to include any dictation alternative range that encompasses the selection (and otherwise
falls back to selecting the current word).

Test: SelectionTests.SelectWordForReplacementWithDictationAlternative

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

(-[WKContentView selectWordForReplacement]):

• UIProcess/ios/WebPageProxyIOS.mm:

(WebKit::WebPageProxy::extendSelectionForReplacement):

Add a new IPC message, WebPage::ExtendSelectionForReplacement, and use it to implement `-[WKContentView
selectWordForReplacement]`.

• WebProcess/WebPage/WebPage.h:
• WebProcess/WebPage/WebPage.messages.in:
• WebProcess/WebPage/ios/WebPageIOS.mm:

(WebKit::WebPage::extendSelectionForReplacement):

Add logic to select the enclosing dictation alternative marker range (if it exists), or the current word at the
start of the selection if there is no enclosing marker range. Note that we first grab a Vector of all the marker
ranges instead of creating them as we iterate through the list of RenderedDocumentMarker*, since creating
visible position ranges from each marker range triggers layout, and each RenderedDocumentMarker* is not
ref-counted, so it's possible that a reentrant call to update the document marker controller during layout could
destroy the next RenderedDocumentMarker* during iteration.

(WebKit::WebPage::extendSelection):
(WebKit::WebPage::setSelectedRangeDispatchingSyntheticMouseEventsIfNeeded):

Pull out logic for dispatching synthetic mouse events when selecting text into a separate helper method, which
is now called from both extendSelection and extendSelectionForReplacement.

Tools:

Add an API test to exercise the change.

• TestWebKitAPI/Tests/ios/SelectionByWord.mm:

(TEST):

9:44 AM Changeset in webkit [293229] by Russell Epstein

• 9 edits in branches/safari-614.1.10-branch/Source

Versioning.

WebKit-7614.1.10.6

9:36 AM Changeset in webkit [293228] by commit-queue@webkit.org

• 3 edits in trunk/Source/WebCore/PAL

Adjust SPI headers to fix build with non-internal SDK
​https://bugs.webkit.org/show_bug.cgi?id=239626

Patch by Alex Christensen <​achristensen@webkit.org> on 2022-04-22
Reviewed by Chris Dumez.

• pal/spi/cocoa/PassKitSPI.h:
• pal/spi/mac/ScreenCaptureKitSPI.h:

9:33 AM Changeset in webkit [293227] by Alan Coon

• 1 copy in branches/safari-613.2.7.3-branch

New branch.

9:33 AM Changeset in webkit [293226] by Alan Coon

• 1 copy in branches/safari-613.2.7.2-branch

New branch.

9:33 AM Changeset in webkit [293225] by Alan Coon

• 1 copy in branches/safari-613.2.7.1-branch

New branch.

9:33 AM Changeset in webkit [293224] by Alan Coon

• 1 copy in branches/safari-613.2.7.0-branch

New branch.

9:11 AM Changeset in webkit [293223] by commit-queue@webkit.org

• 2 edits in trunk/LayoutTests

[ macOS wk2 arm64 ] imported/w3c/web-platform-tests/infrastructure/reftest-wait.html is a flaky image failure
​https://bugs.webkit.org/show_bug.cgi?id=230488

Unreviewed test gardening.

Updated text expectations in include Monterey wk2 arm64

• LayoutTests/platform/mac-wk2/TestExpectations:

Canonical link: ​https://commits.webkit.org/249892@main

8:57 AM Changeset in webkit [293222] by Alan Bujtas

• 4 edits in trunk/Source/WebCore

[LFC][Integration] BoxTree should be able to build non-inline content tree
​https://bugs.webkit.org/show_bug.cgi?id=239653

Reviewed by Antti Koivisto.

This is in preparation for supporting non-IFC content (e.g. flex).

• layout/integration/LayoutIntegrationBoxTree.cpp:

(WebCore::LayoutIntegration::rootBoxFirstLineStyle):
(WebCore::LayoutIntegration::BoxTree::BoxTree):
(WebCore::LayoutIntegration::BoxTree::buildTreeForInlineContent):
(WebCore::LayoutIntegration::BoxTree::layoutBoxForRenderer):
(WebCore::LayoutIntegration::BoxTree::rendererForLayoutBox):
(WebCore::LayoutIntegration::BoxTree::buildTree): Deleted.

• layout/integration/LayoutIntegrationBoxTree.h:

(WebCore::LayoutIntegration::BoxTree::rootRenderer const):
(WebCore::LayoutIntegration::BoxTree::rootRenderer):
(WebCore::LayoutIntegration::BoxTree::flow const): Deleted.
(WebCore::LayoutIntegration::BoxTree::flow): Deleted.

• layout/integration/inline/LayoutIntegrationLineLayout.h:

(WebCore::LayoutIntegration::LineLayout::flow const):
(WebCore::LayoutIntegration::LineLayout::flow):

8:13 AM Changeset in webkit [293221] by commit-queue@webkit.org

• 2 edits in trunk/LayoutTests

[ macOS Release arm64 ] imported/w3c/web-platform-tests/resource-timing/resource_timing.worker.html is a flakey text failure
​https://bugs.webkit.org/show_bug.cgi?id=223472

Unreviewed test gardening.

Updated test expectation [ Release arm64 ]

• LayoutTests/platform/mac/TestExpectations:

Canonical link: ​https://commits.webkit.org/249890@main

7:03 AM Changeset in webkit [293220] by mark.lam@apple.com

• 7 edits in trunk/Source

Apply purifyNaN in more places.
​https://bugs.webkit.org/show_bug.cgi?id=239619
<rdar://problem/91924480>

Reviewed by Yusuke Suzuki.

Source/JavaScriptCore:

• wasm/js/JSWebAssemblyHelpers.h:

(JSC::toJSValue):

Source/WebCore:

• bindings/js/IDBBindingUtilities.cpp:

(WebCore::toJS):

• bindings/js/JSDOMConvertWebGL.cpp:

(WebCore::convertToJSValue):

• bindings/js/SerializedScriptValue.cpp:

(WebCore::CloneDeserializer::readTerminal):

• bridge/objc/objc_utility.mm:

(JSC::Bindings::convertObjcValueToValue):

6:29 AM Changeset in webkit [293219] by Andres Gonzalez

• 9 edits in trunk

Fix for accessibility/aria-grid-with-aria-owns-rows.html in isolated tree mode.
​https://bugs.webkit.org/show_bug.cgi?id=239498
<rdar://problem/91961398>

Reviewed by Chris Fleizach and Tyler Wilcock.

Source/WebCore:

Test: accessibility/aria-grid-with-aria-owns-rows.html.

When creating AXIsolatedObjects, we were passing the object ID of the
parent to the constructor in a parent to child traversal, and storing
that parent ID in the object. This assumed that all AX objects have a
parent-child reflexive relationship, i.e., if A is a child of B, B is
the parent of A. But that is not true in several cases in the
accessibility hierarchy.
One example where the parent-child relationship is not reflexive is for table cells, where a cell can be the child of an object with role column, but its parent is an object with role row.
Another case is when aria-owns is used.
This patch fixes the problem by letting the live object determine its
own parent, instead of having the traversal algorithm pass the parent ID
to the AXIsolatedObject constructor.
A previous attempt to fix this problem was made in
​https://bugs.webkit.org/show_bug.cgi?id=236156.

• accessibility/isolatedtree/AXIsolatedObject.cpp:

(WebCore::AXIsolatedObject::AXIsolatedObject):
(WebCore::AXIsolatedObject::create):
(WebCore::AXIsolatedObject::initializeProperties):
(WebCore::AXIsolatedObject::initializeAttributeData): Renamed.

• accessibility/isolatedtree/AXIsolatedObject.h:
• accessibility/isolatedtree/AXIsolatedTree.cpp:

(WebCore::AXIsolatedTree::create):
(WebCore::AXIsolatedTree::generateSubtree):
(WebCore::AXIsolatedTree::nodeChangeForObject):
(WebCore::AXIsolatedTree::queueRemovalsAndUnresolvedChanges):
(WebCore::AXIsolatedTree::collectNodeChangesForSubtree):
(WebCore::AXIsolatedTree::updateNode):
(WebCore::AXIsolatedTree::updateChildren):
(WebCore::AXIsolatedTree::parentIDForObject): Deleted.

• accessibility/isolatedtree/AXIsolatedTree.h:
• accessibility/isolatedtree/mac/AXIsolatedObjectMac.mm:

(WebCore::AXIsolatedObject::initializePlatformProperties):

LayoutTests:

• accessibility/aria-grid-with-aria-owns-rows-expected.txt:
• accessibility/aria-grid-with-aria-owns-rows.html:

6:27 AM Changeset in webkit [293218] by Alan Bujtas

• 5 edits
  2 adds in trunk

Book content is clipped at page boundary when displaying list with bullet points
​https://bugs.webkit.org/show_bug.cgi?id=239638
<rdar://89257768>

Reviewed by Antti Koivisto.

Source/WebCore:

Table cell content needs special handling when does not fit the page.

When a table cell's first line does not fit the page, instead of the normal, "let's add a flow level pagination strut"
we have to handle it as a line level strut (not sure why it is implemted this way but this is the expected behavior).
-also implment LineBoxIteratorModernPath::isFirstAfterPageBreak.

Test: fast/multicol/pagination/table-cell-one-line-pagination.html

• layout/integration/inline/InlineIteratorLineBoxModernPath.h:

(WebCore::InlineIterator::LineBoxIteratorModernPath::isFirstAfterPageBreak const):

• layout/integration/inline/LayoutIntegrationLine.h:

(WebCore::LayoutIntegration::Line::Line):
(WebCore::LayoutIntegration::Line::isFirstAfterPageBreak const):

• layout/integration/inline/LayoutIntegrationPagination.cpp:

(WebCore::LayoutIntegration::setPageBreakForLine): According to RenderBlockFlow::adjustLinePositionForPagination, we
have to call setPageBreak, even when the first line does not fit.
(WebCore::LayoutIntegration::makeAdjustedContent):
(WebCore::LayoutIntegration::adjustLinePositionsForPagination):

LayoutTests:

• fast/multicol/pagination/table-cell-one-line-pagination-expected.html: Added.
• fast/multicol/pagination/table-cell-one-line-pagination.html: Added.

5:02 AM Changeset in webkit [293217] by Antti Koivisto

• 4 edits in trunk/Source/WebCore

Bail out from text decoration computation if none is requested
​https://bugs.webkit.org/show_bug.cgi?id=239641

Reviewed by Tim Nguyen.

• rendering/TextDecorationPainter.cpp:

(WebCore::collectStylesForRenderer):

Also avoid resolving colors unnecessarily.

(WebCore::TextDecorationPainter::stylesForRenderer):

• rendering/style/RenderStyleConstants.cpp:

(WebCore::operator<<):

• rendering/style/RenderStyleConstants.h:

Remove unused 'None' value.

4:40 AM Changeset in webkit [293216] by eocanha@igalia.com

• 5 edits in trunk/Source/WebCore

[MSE] Improve SourceBuffer memory cost reporting
​https://bugs.webkit.org/show_bug.cgi?id=238948

Reviewed by Xabier Rodriguez-Calvar.

Properly implementing the memoryCost() API in SourceBuffer would improve memory
awareness when garbage collecting.

This patch extends the SourceBuffer interface with the ReportExtraMemoryCost
attribute, implements memoryCost() method for the interface and fixes
calculations for removed buffers. It distinguishes the amount of memory
increases reported (extraMemoryCostDelta, what was reported until now, can
only grow) from the total amount of memory reported (can grow and shrink).

This patch is co-authored by Andrzej Surdej (​https://github.com/asurdej-comcast)
See: ​https://github.com/WebPlatformForEmbedded/WPEWebKit/pull/827

• Modules/mediasource/SourceBuffer.cpp: Implement memoryCost() interface and hold the absolute memory consumed as m_extraMemoryCost, keeping the original m_reportedExtraMemoryCost behaviour.
• Modules/mediasource/SourceBuffer.h: Added memoryCost() and m_extraMemoryCost.
• Modules/mediasource/SourceBuffer.idl: Require the ReportExtraMemoryCost feature.
• platform/graphics/SourceBufferPrivate.cpp: Also report the memory change to SourceBuffer when evicting frames (memory reduction).

4:08 AM Changeset in webkit [293215] by zan@falconsigh.net

• 2 edits in trunk/Source/WebKit

[GLib] Make WebKitSettings XSS auditor functions no-op
​https://bugs.webkit.org/show_bug.cgi?id=239651

Reviewed by Adrian Perez de Castro.

With the XSS auditor feature being deprecated and removed, and the
preference entries removed in r293199, the GLib API for this feature
has to be adjusted as well. In both getter and setter, a warning is
printed out when invoked.

• UIProcess/API/glib/WebKitSettings.cpp:

(webkit_settings_get_enable_xss_auditor):
(webkit_settings_set_enable_xss_auditor):

12:25 AM Changeset in webkit [293214] by Adrian Perez de Castro

• 1 copy in releases/WPE WebKit/webkit-2.36.1

WPE WebKit 2.36.1

12:22 AM Changeset in webkit [293213] by Adrian Perez de Castro

• 4 edits in releases/WebKitGTK/webkit-2.36

Unreviewed. Update OptionsWPE.cmake and NEWS for the 2.36.1 release

.:

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

Source/WebKit:

• wpe/NEWS: Add release notes for 2.36.1.

12:02 AM Changeset in webkit [293212] by Tyler Wilcock

• 5 edits in trunk

AX: The isolated tree is not updated after role changes in AccessibilityRenderObject::updateRoleAfterChildrenCreation()
​https://bugs.webkit.org/show_bug.cgi?id=239549

Reviewed by Chris Fleizach and Andres Gonzalez.

Source/WebCore:

In AccessibilityRenderObject::updateRoleAfterChildrenCreation(), if the objects role changes,
we need to inform the cache via AXObjectCache::handleRoleChange which in turn will
update the isolated tree.

Fixes accessibility/mac/invalid-menu-role-does-not-send-notification.html in isolated tree mode.

• accessibility/AccessibilityRenderObject.cpp:

(WebCore::AccessibilityRenderObject::updateRoleAfterChildrenCreation):

LayoutTests:

• accessibility/mac/invalid-menu-role-does-not-send-notification-expected.txt:
• accessibility/mac/invalid-menu-role-does-not-send-notification.html:

Update test to use async-await to actually wait for changes to happen,
which is required to make the test pass in isolated tree mode.

Apr 21, 2022:

11:08 PM Changeset in webkit [293211] by commit-queue@webkit.org

• 21 edits
  1 add in trunk

Use GCGLSpanTuple to pass buffer parameters to multidraw calls.
​https://bugs.webkit.org/show_bug.cgi?id=239154

Patch by John Cunningham <​johncunningham@apple.com> on 2022-04-21
Reviewed by Kimmo Kinnunen.

Source/WebCore:

• html/canvas/WebGLMultiDraw.cpp:

(WebCore::WebGLMultiDraw::multiDrawArraysWEBGL):
(WebCore::WebGLMultiDraw::multiDrawArraysInstancedWEBGL):
(WebCore::WebGLMultiDraw::multiDrawElementsWEBGL):
(WebCore::WebGLMultiDraw::multiDrawElementsInstancedWEBGL):
(WebCore::makeSpanWithOffset): Deleted.

• platform/graphics/GraphicsContextGL.h:
• platform/graphics/GraphicsTypesGL.h:
• platform/graphics/angle/GraphicsContextGLANGLE.cpp:

(WebCore::GraphicsContextGLANGLE::multiDrawArraysANGLE):
(WebCore::GraphicsContextGLANGLE::multiDrawArraysInstancedANGLE):
(WebCore::GraphicsContextGLANGLE::multiDrawElementsANGLE):
(WebCore::GraphicsContextGLANGLE::multiDrawElementsInstancedANGLE):

• platform/graphics/angle/GraphicsContextGLANGLE.h:
• platform/graphics/opengl/GraphicsContextGLOpenGL.h:

Source/WebKit:

• GPUProcess/graphics/RemoteGraphicsContextGL.cpp:

(WebKit::WebCore::vectorCopyCast):
(WebKit::RemoteGraphicsContextGL::multiDrawArraysANGLE):
(WebKit::RemoteGraphicsContextGL::multiDrawArraysInstancedANGLE):
(WebKit::RemoteGraphicsContextGL::multiDrawElementsANGLE):
(WebKit::RemoteGraphicsContextGL::multiDrawElementsInstancedANGLE):

• GPUProcess/graphics/RemoteGraphicsContextGL.h:
• GPUProcess/graphics/RemoteGraphicsContextGL.messages.in:
• GPUProcess/graphics/RemoteGraphicsContextGLFunctionsGenerated.h:

(getInternalformativ):
(multiDrawArraysANGLE): Deleted.
(multiDrawArraysInstancedANGLE): Deleted.
(multiDrawElementsANGLE): Deleted.
(multiDrawElementsInstancedANGLE): Deleted.

• Platform/IPC/ArgumentCoders.h:
• Platform/IPC/ArrayReferenceTuple.h: Added.
• Scripts/webkit/messages.py:

(class_template_headers):

• WebKit.xcodeproj/project.pbxproj:
• WebProcess/GPU/graphics/RemoteGraphicsContextGLProxy.cpp:

(WebKit::RemoteGraphicsContextGLProxy::multiDrawArraysANGLE):
(WebKit::RemoteGraphicsContextGLProxy::multiDrawArraysInstancedANGLE):
(WebKit::RemoteGraphicsContextGLProxy::multiDrawElementsANGLE):
(WebKit::RemoteGraphicsContextGLProxy::multiDrawElementsInstancedANGLE):

• WebProcess/GPU/graphics/RemoteGraphicsContextGLProxy.h:
• WebProcess/GPU/graphics/RemoteGraphicsContextGLProxyFunctionsGenerated.cpp:

(WebKit::RemoteGraphicsContextGLProxy::multiDrawArraysANGLE): Deleted.
(WebKit::RemoteGraphicsContextGLProxy::multiDrawArraysInstancedANGLE): Deleted.
(WebKit::RemoteGraphicsContextGLProxy::multiDrawElementsANGLE): Deleted.
(WebKit::RemoteGraphicsContextGLProxy::multiDrawElementsInstancedANGLE): Deleted.

Tools:

• Scripts/generate-gpup-webgl:

10:41 PM Changeset in webkit [293210] by ysuzuki@apple.com

• 27 edits
  2 adds in trunk

[JSC] PropertyTable should have compact mode
​https://bugs.webkit.org/show_bug.cgi?id=239451

Reviewed by Saam Barati.

This patch introduces "compact" version of PropertyTable. PropertyTable needs to hold two kind of integers: index and property offset.
But for most of objects, both are pretty small and they can fit in uint8_t. If we can use uint8_t for both, we can significantly reduce
size of allocated memory for PropertyTable (only 40% of memory is required!). This is good for memory, but also good for performance.
Now each CompactPropertyTableEntry is 8bytes while original PropertyMapEntry was 16bytes, so CompactPropertyTableEntry can fit in CPU cache well.
Also, not allocating large amount of memory can reduce memory allocation / deallocation cost. One of costly destruction of GC-managed objects is
PropertyTable (and CodeBlock), and we can reduce that cost by not allocating much memory.

The approach is following.

1. For index vector, we use uint8_t if index can fit within uint8_t.
2. For proprety offset, we use uint8_t and CompactPropertyTableEntry if it is suitable.
3. Once the table gets non-compact, we keep it non-compact since we could have deleted index which has larger than uint8_t. We could improve this strategy when deleted indexes are cleared, but for now, we are taking simple approach.
4. We store isCompactFlag 1 bit in the pointer to the table.
5. We encapsulate functions modifying property table entry in PropertyTable itself, so we do not leak internal compact / non-compact mode to the user of PropertyTable. We remove begin() / end() iterators and instead use forEachproperty, which can implement iteration for each mode more efficiently.

We have a further opportunity to improve this further: we can deploy 75% load factor only for compact table. Then we can increase threshold of
compact table further and keep more and more tables compact mode. Plus, for small sized tables, small backing memory is better in terms of
CPU cache hit (and that's measured in WTF::HashTable, and that's why WTF::Hashtable deploys 75% load factor only for small tables). This is left
for the subsequent change.

This change is neutral in JetStream2, 0.3% improvement in Speedometer2 with 80% confidence, and 0.41% improvement in RAMification with 95% confidence.

• JSTests/stress/change-attribute-structure-transition.js:

(shouldBe.JSON.stringify.sd):

• Source/JavaScriptCore/bytecode/ObjectAllocationProfileInlines.h:

(JSC::ObjectAllocationProfileBase<Derived>::initializeProfile):

• Source/JavaScriptCore/dfg/DFGAbstractInterpreterInlines.h:

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

• Source/JavaScriptCore/dfg/DFGConstantFoldingPhase.cpp:

(JSC::DFG::ConstantFoldingPhase::foldConstants):

• Source/JavaScriptCore/dfg/DFGOperations.cpp:

(JSC::DFG::JSC_DEFINE_JIT_OPERATION):

• Source/JavaScriptCore/dfg/DFGSpeculativeJIT.cpp:
• Source/JavaScriptCore/ftl/FTLLowerDFGToB3.cpp:

(JSC::FTL::DFG::LowerDFGToB3::compileCompareStrictEq):

• Source/JavaScriptCore/ftl/FTLOperations.cpp:

(JSC::FTL::JSC_DEFINE_JIT_OPERATION):

• Source/JavaScriptCore/runtime/ClonedArguments.h:
• Source/JavaScriptCore/runtime/IteratorOperations.cpp:

(JSC::createIteratorResultObjectStructure):

• Source/JavaScriptCore/runtime/JSGlobalObject.cpp:

(JSC::JSGlobalObject::init):

• Source/JavaScriptCore/runtime/JSGlobalObjectFunctions.cpp:

(JSC::JSC_DEFINE_HOST_FUNCTION):

• Source/JavaScriptCore/runtime/JSONObject.cpp:

(JSC::Stringifier::Holder::appendNextProperty):

• Source/JavaScriptCore/runtime/JSObject.cpp:

(JSC::JSObject::analyzeHeap):

• Source/JavaScriptCore/runtime/JSObject.h:
• Source/JavaScriptCore/runtime/ObjectConstructor.h:

(JSC::constructEmptyObject):
(JSC::createDataPropertyDescriptorObjectStructure):
(JSC::createAccessorPropertyDescriptorObjectStructure):

• Source/JavaScriptCore/runtime/ObjectConstructorInlines.h:

(JSC::objectAssignFast):

• Source/JavaScriptCore/runtime/PropertyOffset.h:
• Source/JavaScriptCore/runtime/PropertySlot.h:
• Source/JavaScriptCore/runtime/PropertyTable.cpp:

(JSC::PropertyTable::PropertyTable):
(JSC::PropertyTable::finishCreation):
(JSC::PropertyTable::visitChildrenImpl):
(JSC::PropertyTable::~PropertyTable):
(JSC::PropertyTable::seal):
(JSC::PropertyTable::freeze):
(JSC::PropertyTable::isSealed const):
(JSC::PropertyTable::isFrozen const):
(JSC::PropertyTable::renumberPropertyOffsets):

• Source/JavaScriptCore/runtime/PropertyTable.h:

(JSC::isPowerOf2):
(JSC::nextPowerOf2):
(JSC::PropertyTable::findImpl):
(JSC::PropertyTable::find):
(JSC::PropertyTable::get):
(JSC::PropertyTable::add):
(JSC::PropertyTable::remove):
(JSC::PropertyTable::take):
(JSC::PropertyTable::updateAttributeIfExists):
(JSC::PropertyTable::sizeInMemory):
(JSC::PropertyTable::reinsert):
(JSC::PropertyTable::rehash):
(JSC::PropertyTable::skipDeletedEntries):
(JSC::PropertyTable::dataSize):
(JSC::PropertyTable::canInsert):
(JSC::PropertyTable::forEachProperty const):
(JSC::PropertyTable::forEachPropertyMutable):
(JSC::PropertyTable::begin): Deleted.
(JSC::PropertyTable::end): Deleted.
(JSC::PropertyTable::begin const): Deleted.
(JSC::PropertyTable::end const): Deleted.
(JSC::PropertyTable::table): Deleted.
(JSC::PropertyTable::table const): Deleted.

• Source/JavaScriptCore/runtime/RegExpMatchesArray.h:
• Source/JavaScriptCore/runtime/Structure.cpp:

(JSC::Structure::dumpStatistics):
(JSC::Structure::Structure):
(JSC::Structure::materializePropertyTable):
(JSC::Structure::nonPropertyTransitionSlow):
(JSC::Structure::isSealed):
(JSC::Structure::isFrozen):
(JSC::Structure::flattenDictionaryStructure):
(JSC::PropertyTableStatisticsExitLogger::~PropertyTableStatisticsExitLogger):
(JSC::Structure::getConcurrently):
(JSC::Structure::getPropertiesConcurrently):
(JSC::Structure::getPropertyNamesFromStructure):
(JSC::Structure::toStructureShape):
(JSC::Structure::dump const):

• Source/JavaScriptCore/runtime/Structure.h:

(JSC::CompactPropertyTableEntry::CompactPropertyTableEntry):
(JSC::CompactPropertyTableEntry::key const):
(JSC::CompactPropertyTableEntry::setKey):
(JSC::CompactPropertyTableEntry::offset const):
(JSC::CompactPropertyTableEntry::setOffset):
(JSC::CompactPropertyTableEntry::attributes const):
(JSC::CompactPropertyTableEntry::setAttributes):
(JSC::PropertyTableEntry::PropertyTableEntry):
(JSC::PropertyTableEntry::key const):
(JSC::PropertyTableEntry::setKey):
(JSC::PropertyTableEntry::offset const):
(JSC::PropertyTableEntry::setOffset):
(JSC::PropertyTableEntry::attributes const):
(JSC::PropertyTableEntry::setAttributes):
(JSC::PropertyMapEntry::PropertyMapEntry): Deleted.

• Source/JavaScriptCore/runtime/StructureInlines.h:

(JSC::Structure::get):
(JSC::Structure::forEachPropertyConcurrently):
(JSC::Structure::forEachProperty):
(JSC::Structure::add):
(JSC::Structure::remove):
(JSC::Structure::attributeChange):

Canonical link: ​https://commits.webkit.org/249881@main

10:35 PM Changeset in webkit [293209] by commit-queue@webkit.org

• 4 edits
  4 adds in trunk

contain: layout on the html element should change position:fixed behavior
​https://bugs.webkit.org/show_bug.cgi?id=238560

Patch by Rob Buis <​rbuis@igalia.com> on 2022-04-21
Reviewed by Simon Fraser.

LayoutTests/imported/w3c:

• web-platform-tests/css/css-contain/contain-layout-021-expected.html: Added.
• web-platform-tests/css/css-contain/contain-layout-021.html: Added.
• web-platform-tests/css/css-contain/reference/contain-layout-021-ref.html: Added.

Source/WebCore:

The mapLocalToContainer/mapAbsoluteToLocalPoint methods need to not only consider boxes
having transforms but need to include all checks in canContainFixedPositionObjects to
determine if the box acts as a fixed position containing block.

Tests: imported/w3c/web-platform-tests/css/css-contain/contain-layout-021.html

imported/w3c/web-platform-tests/css/css-contain/reference/contain-layout-021-ref.html

• rendering/RenderBox.cpp:

(WebCore::RenderBox::mapLocalToContainer const):
(WebCore::RenderBox::mapAbsoluteToLocalPoint const):

• rendering/svg/RenderSVGRoot.cpp:

(WebCore::RenderSVGRoot::mapLocalToContainer const):

10:21 PM Changeset in webkit [293208] by Chris Dumez

• 5 edits
  2 adds in trunk

No-op, instead of throwing, on dataTransfer.items.remove()
​https://bugs.webkit.org/show_bug.cgi?id=239618

Reviewed by Wenson Hsieh.

LayoutTests/imported/w3c:

Import test coverage from upstream WPT.

• web-platform-tests/html/editing/dnd/datastore/datatransferitemlist-remove-expected.txt: Added.
• web-platform-tests/html/editing/dnd/datastore/datatransferitemlist-remove.html: Added.

Source/WebCore:

Calling dataTransfer.items.remove() with an out-of-range index should be a no-op instead of
throwing as per the latest HTML specification:

• ​https://github.com/whatwg/html/pull/7844

We were the only browser behaving this way so this was a compatibility risk.

Test: imported/w3c/web-platform-tests/html/editing/dnd/datastore/datatransferitemlist-remove.html

• dom/DataTransferItemList.cpp:

(WebCore::DataTransferItemList::remove):

8:56 PM Changeset in webkit [293207] by commit-queue@webkit.org

• 2 edits in trunk/Tools

[PGO] Fix perl warnings cause by PGO for real this time.
​https://bugs.webkit.org/show_bug.cgi?id=239633

Reviewed by Yusuke Suzuki.

Add a missing 'defined' check before reading the PGO environment variable.

• Tools/Scripts/check-for-weak-vtables-and-externals:

Canonical link: ​https://commits.webkit.org/249878@main

8:27 PM Changeset in webkit [293206] by Andres Gonzalez

• 9 edits in trunk/LayoutTests

AX ITM: Fix for multiple flaky tests in isolated tree mode.
​https://bugs.webkit.org/show_bug.cgi?id=239583
<rdar://problem/92064483>

Reviewed by Chris Fleizach.

Rewrote these flaky tests so that now pass in isolated tree mode:

• Use expect() instead of shouldBe*() and log the output at the end of

the test.

• Avoid setting focus to an element just to get the accessible element.

• accessibility/legend-children-are-visible-expected.txt:
• accessibility/legend-children-are-visible.html:
• accessibility/mac/aria-label-overrides-visible-text-expected.txt:
• accessibility/mac/aria-label-overrides-visible-text.html:
• accessibility/mac/aria-popup-buttons-on-native-elements-expected.txt:
• accessibility/mac/aria-popup-buttons-on-native-elements.html:
• accessibility/mac/link-with-title-expected.txt:
• accessibility/mac/link-with-title.html:

7:48 PM Changeset in webkit [293205] by Brent Fulgham

• 4 edits in trunk/Source/WebKitLegacy/win

Unreviewed build fix after r293199.
​https://bugs.webkit.org/show_bug.cgi?id=239628

Remove unused references to XSS Auditor.

• WebPreferenceKeysPrivate.h:
• WebPreferences.cpp:

(WebPreferences::initializeDefaultSettings):
(WebPreferences::isXSSAuditorEnabled):
(WebPreferences::setXSSAuditorEnabled):

• WebView.cpp:

(WebView::notifyPreferencesChanged):

6:21 PM Changeset in webkit [293204] by commit-queue@webkit.org

• 2 edits in trunk/LayoutTests

[ macOS wk2 arm64 ] media/modern-media-controls/tracks-support/text-track-selected-via-media-api.html is a flaky failure
​https://bugs.webkit.org/show_bug.cgi?id=239634

Unreviewed test gardening.

• LayoutTests/platform/mac-wk2/TestExpectations:

Canonical link: ​https://commits.webkit.org/249875@main

5:51 PM Changeset in webkit [293203] by ysuzuki@apple.com

• 12 edits
  2 deletes in trunk/Source/JavaScriptCore

[JSC] Remove TempRegisterSet
​https://bugs.webkit.org/show_bug.cgi?id=239578

Reviewed by Mark Lam.

We can always use RegisterSet. TempRegisterSet can save several bytes, but we have no code using TempRegisterSet in
heap-allocated classes. So, this does not make sense anymore. Instead of TempRegisterSet, we will consistently use
RegisterSet to pass register info and ScratchRegisterAllocator to manage allocation of temp / scratch registers.

We also remove the copyCalleeSavesToEntryFrameCalleeSavesBuffer function which takes no scratch register. It was
using TempRegisterSet to allocate a scratch register, but the caller of this function was making assumptions on how
TempRegisterSet will allocate that scratch. This is very fragile and dangerous. We should explicitly pass a scratch
register instead in that case.

• CMakeLists.txt:
• JavaScriptCore.xcodeproj/project.pbxproj:
• Sources.txt:
• ftl/FTLLowerDFGToB3.cpp:

(JSC::FTL::DFG::LowerDFGToB3::lower):

• jit/AssemblyHelpers.cpp:

(JSC::AssemblyHelpers::copyLLIntBaselineCalleeSavesFromFrameOrRegisterToEntryFrameCalleeSavesBuffer):

• jit/AssemblyHelpers.h:

(JSC::AssemblyHelpers::copyLLIntBaselineCalleeSavesFromFrameOrRegisterToEntryFrameCalleeSavesBuffer): Deleted.

• jit/RegisterSet.h:
• jit/ScratchRegisterAllocator.cpp:

(JSC::ScratchRegisterAllocator::lock):
(JSC::ScratchRegisterAllocator::allocateScratch):
(JSC::ScratchRegisterAllocator::preserveReusedRegistersByPushing):
(JSC::ScratchRegisterAllocator::restoreReusedRegistersByPopping):

• jit/ScratchRegisterAllocator.h:
• jit/TempRegisterSet.cpp: Removed.
• jit/TempRegisterSet.h: Removed.
• jit/ThunkGenerators.cpp:

(JSC::handleExceptionGenerator):
(JSC::handleExceptionWithCallFrameRollbackGenerator):
(JSC::throwExceptionFromCallSlowPathGenerator):
(JSC::nativeForGenerator):
(JSC::boundFunctionCallGenerator):
(JSC::remoteFunctionCallGenerator):

• wasm/js/WasmToJS.cpp:

(JSC::Wasm::wasmToJS):

5:17 PM Changeset in webkit [293202] by ysuzuki@apple.com

• 24 edits in trunk

[WTF] Add string concatenate adapter for UUID
​https://bugs.webkit.org/show_bug.cgi?id=239590

Reviewed by Chris Dumez and Darin Adler.

This patch adds string concatenate adapter for UUID.
UUID's stringifier consists of multiple string concatenate adapters.
This patch adds WTF::handleWithAdapters so that we can define adapters
once, and we can semi-automatically define length() and writeTo method for UUID.

And we use UUID + makeString instead of createVersion4UUIDString. This is more
efficient since we do not need to allocate string twice.

• Source/WTF/wtf/UUID.cpp:

(WTF::UUID::generateWeakRandomUUIDVersion4):
(WTF::createVersion4UUIDString):
(WTF::generateWeakRandomUUIDVersion4): Deleted.
(WTF::UUID::toString const): Deleted.
(WTF::createVersion4UUIDStringWeak): Deleted.

• Source/WTF/wtf/UUID.h:

(WTF::UUID::createVersion4Weak):
(WTF::UUID::isHashTableDeletedValue const):
(WTF::UUID::data const):
(WTF::StringTypeAdapter<UUID>::StringTypeAdapter):
(WTF::StringTypeAdapter<UUID>::length const):
(WTF::StringTypeAdapter<UUID>::is8Bit const):
(WTF::StringTypeAdapter<UUID>::writeTo const):

• Source/WebCore/animation/KeyframeEffect.cpp:

(WebCore::KeyframeEffect::copyPropertiesFromSource):
(WebCore::KeyframeEffect::updateBlendingKeyframes):
(WebCore::KeyframeEffect::computeCSSTransitionBlendingKeyframes):

• Tools/TestWebKitAPI/Tests/WTF/UUID.cpp:

(TEST):

Canonical link: ​https://commits.webkit.org/249873@main

5:09 PM Changeset in webkit [293201] by Alan Coon

• 1 copy in tags/WebKit-7613.2.7

Tag WebKit-7613.2.7.

5:09 PM Changeset in webkit [293200] by Alan Coon

• 1 delete in tags/WebKit-7613.2.7

Delete tag.

5:05 PM Changeset in webkit [293199] by Brent Fulgham

• 4 edits in trunk/Source

Remove XSS Auditor: Part 4 (Settings)
​https://bugs.webkit.org/show_bug.cgi?id=239628

Reviewed by Alex Christensen.

This change is a follow-up to Bug 230661 that removes the unused
Source/WebKitLegacy/mac:

WebKitXSSAuditorEnabledPreferenceKey.

• WebView/WebPreferenceKeysPrivate.h:

Source/WTF:

WebPreferences.yaml content.

• Scripts/Preferences/WebPreferences.yaml:

5:02 PM Changeset in webkit [293198] by Alan Coon

• 2 edits in branches/safari-613-branch/Source/WebCore

Cherry-pick r292095. rdar://problem/90240658

Ensure m_layerRulesBeforeImportRules is properly copied during StyleSheetContents instantiation
​https://bugs.webkit.org/show_bug.cgi?id=238537

Reviewed by Antti Koivisto.

During the instantiation of the StyleSheetContents the m_layerRulesBeforeImportRules was not being
properly copied.

• css/StyleSheetContents.cpp: (WebCore::StyleSheetContents::StyleSheetContents):

git-svn-id: ​https://svn.webkit.org/repository/webkit/trunk@292095 268f45cc-cd09-0410-ab3c-d52691b4dbfc

5:01 PM Changeset in webkit [293197] by Chris Dumez

• 4 edits in trunk/Source/WebCore

Simplify FrameSelection::textWasReplaced() and callers
​https://bugs.webkit.org/show_bug.cgi?id=239620

Reviewed by Geoffrey Garen.

• dom/CharacterData.cpp:

(WebCore::CharacterData::setData):

• Cache document().frame()
• Call textWasReplaced() with *this instead of this
• Use modern template deduction for Ref<>.

(WebCore::CharacterData::setDataAndUpdate):

• Use WTFMove() for m_data to avoid ref-counting churn. We're about to overwrite m_data anyway.
• Use else if for the ProcessingInstruction branch instead of if to avoid this check in the common case where |this| is a Text node. A Text node cannot be a ProcessingInstruction node.
• Cache document().frame()
• Call textWasReplaced() with *this instead of this

• editing/FrameSelection.cpp:

(WebCore::updatePositionAfterAdoptingTextReplacement):

• Pass node by reference now that the caller now has a reference instead of a pointer
• Drop !position.anchorNode() check. It is not necessary as the position.anchorNode() != &node check would already be true if the anchorNode were null.

(WebCore::FrameSelection::textWasReplaced):

• Drop outdated comment about a fragment check since there is no such check anymore
• Pass node by reference instead of pointer as all call sites have a non-null pointer
• Drop null check for node that is no longer necessary

• editing/FrameSelection.h:

4:55 PM Changeset in webkit [293196] by Russell Epstein

• 1 copy in tags/WebKit-7613.2.7

Tag WebKit-7613.2.7.

4:27 PM Changeset in webkit [293195] by Chris Dumez

• 85 edits in trunk/Source

Adopt RobinHoodHashMap / RobinHoodHashSet more broadly in WebCore
​https://bugs.webkit.org/show_bug.cgi?id=239576

Reviewed by Yusuke Suzuki.

Source/WebCore:

Adopt RobinHoodHashMap / RobinHoodHashSet more broadly in WebCore to avoid wasting memory in hash tables.
RobinHoodHashMap / RobinHoodHashSet have more restrictions on what key types they work with and may result
in slightly lower performance but they have a much higher load factor that the regular HashMap / HashSet,
thus reducing memory usage. This patch adopts RobinHoodHashMap / RobinHoodHashSet on non performance
sensitive maps / sets in WebCore that have compatible keys (String / AtomString / URL because they cache
their hash).

This is perf-neutral on all our benchmarks according to A/B testing.

• Modules/encryptedmedia/InitDataRegistry.h:
• Modules/encryptedmedia/legacy/LegacyCDMSessionClearKey.h:
• Modules/indexeddb/server/SQLiteIDBTransaction.h:
• Modules/mediasource/MediaSourceRegistry.h:
• Modules/mediastream/MediaDevices.h:
• Modules/mediastream/MediaStream.h:
• Modules/mediastream/libwebrtc/LibWebRTCMediaEndpoint.h:
• Modules/web-locks/WebLockRegistry.cpp:
• Modules/webaudio/AudioWorkletGlobalScope.h:
• Modules/webaudio/AudioWorkletNode.h:
• Modules/webaudio/AudioWorkletProcessor.cpp:

(WebCore::constructFrozenKeyValueObject):
(WebCore::copyDataFromParameterMapToJSObject):
(WebCore::AudioWorkletProcessor::buildJSArguments):
(WebCore::AudioWorkletProcessor::process):

• Modules/webaudio/AudioWorkletProcessor.h:
• Modules/webaudio/BaseAudioContext.h:

(WebCore::BaseAudioContext::parameterDescriptorMap const):

• Modules/webdatabase/Database.cpp:

(WebCore::WTF_REQUIRES_LOCK):

• Modules/webdatabase/DatabaseTracker.cpp:

(WebCore::DatabaseTracker::recordDeletingDatabase):

• Modules/webdatabase/DatabaseTracker.h:
• bindings/js/JSCustomElementInterface.h:
• bindings/js/ScriptModuleLoader.h:
• dom/CustomElementRegistry.h:

(WebCore::CustomElementRegistry::promiseMap):

• dom/ExtensionStyleSheets.h:
• dom/MutationObserver.cpp:

(WebCore::MutationObserver::observe):

• dom/MutationObserverRegistration.cpp:

(WebCore::MutationObserverRegistration::MutationObserverRegistration):
(WebCore::MutationObserverRegistration::resetObservation):

• dom/MutationObserverRegistration.h:
• dom/Node.cpp:

(WebCore::Node::registerMutationObserver):

• dom/Node.h:
• dom/WindowEventLoop.cpp:
• editing/Editor.h:
• inspector/InspectorAuditResourcesObject.h:
• inspector/NetworkResourcesData.cpp:

(WebCore::NetworkResourcesData::clear):

• inspector/NetworkResourcesData.h:
• inspector/agents/InspectorAnimationAgent.h:
• inspector/agents/InspectorCanvasAgent.h:
• inspector/agents/InspectorDOMDebuggerAgent.h:
• inspector/agents/InspectorDatabaseAgent.h:
• inspector/agents/InspectorNetworkAgent.h:
• inspector/agents/InspectorPageAgent.h:
• inspector/agents/InspectorWorkerAgent.h:
• loader/DocumentLoader.cpp:

(WebCore::DocumentLoader::setActiveContentRuleListActionPatterns):

• loader/DocumentLoader.h:
• loader/cache/CachedResourceLoader.h:
• page/Page.h:
• page/PageConfiguration.h:
• page/csp/ContentSecurityPolicyMediaListDirective.h:
• page/csp/ContentSecurityPolicySourceList.h:
• platform/LegacySchemeRegistry.cpp:

(WebCore::WTF_REQUIRES_LOCK):

• platform/StaticPasteboard.h:
• platform/graphics/SourceBufferPrivate.h:
• platform/graphics/avfoundation/cf/MediaPlayerPrivateAVFoundationCF.cpp:
• platform/graphics/avfoundation/objc/MediaPlayerPrivateAVFoundationObjC.h:
• platform/graphics/avfoundation/objc/MediaPlayerPrivateMediaStreamAVFObjC.h:
• platform/graphics/avfoundation/objc/MediaPlayerPrivateMediaStreamAVFObjC.mm:

(WebCore::updateTracksOfKind):

• platform/graphics/cg/UTIRegistry.cpp:

(WebCore::additionalSupportedImageTypes):

• platform/graphics/cg/UTIRegistry.h:
• platform/graphics/cocoa/FontCacheCoreText.cpp:
• platform/graphics/cocoa/SystemFontDatabaseCoreText.cpp:

(WebCore::genericFamily):

• platform/graphics/cocoa/SystemFontDatabaseCoreText.h:
• platform/mac/ThreadCheck.mm:

(WebCoreReportThreadViolation):

• platform/mediastream/MediaStreamPrivate.h:
• platform/network/BlobRegistryImpl.h:
• platform/network/CredentialStorage.h:
• platform/network/NetworkStorageSession.h:
• rendering/ReferencedSVGResources.h:
• svg/SVGDocumentExtensions.h:
• workers/Worker.h:
• workers/WorkerGlobalScope.h:
• workers/service/ServiceWorkerContextData.h:

(WebCore::ServiceWorkerContextData::decode):

• workers/service/server/RegistrationDatabase.cpp:

(WebCore::stripScriptSources):
(WebCore::populateScriptSourcesFromDisk):
(WebCore::RegistrationDatabase::doPushChanges):
(WebCore::RegistrationDatabase::importRecords):

• workers/service/server/RegistrationStore.cpp:

(WebCore::RegistrationStore::didSaveWorkerScriptsToDisk):

• workers/service/server/RegistrationStore.h:
• workers/service/server/SWServer.cpp:

(WebCore::SWServer::didSaveWorkerScriptsToDisk):
(WebCore::SWServer::updateWorker):
(WebCore::SWServer::installContextData):

• workers/service/server/SWServer.h:
• workers/service/server/SWServerToContextConnection.h:
• workers/service/server/SWServerWorker.cpp:

(WebCore::SWServerWorker::SWServerWorker):
(WebCore::SWServerWorker::didSaveScriptsToDisk):

• workers/service/server/SWServerWorker.h:

Source/WebKit:

Adopt RobinHoodHashMap / RobinHoodHashSet more broadly in WebKit to avoid wasting memory in hash tables.
RobinHoodHashMap / RobinHoodHashSet have more restrictions on what key types they work with and may result
in slightly lower performance but they have a much higher load factor that the regular HashMap / HashSet,
thus reducing memory usage. This patch adopts RobinHoodHashMap / RobinHoodHashSet on non performance
sensitive maps / sets in WebCore that have compatible keys (String / AtomString / URL because they cache
their hash).

• NetworkProcess/ServiceWorker/WebSWServerToContextConnection.cpp:

(WebKit::WebSWServerToContextConnection::didSaveScriptsToDisk):

• NetworkProcess/ServiceWorker/WebSWServerToContextConnection.h:
• Shared/WebPageCreationParameters.cpp:

(WebKit::WebPageCreationParameters::decode):

• Shared/WebPageCreationParameters.h:
• UIProcess/API/APIPageConfiguration.h:

(API::PageConfiguration::allowedNetworkHosts const):
(API::PageConfiguration::setAllowedNetworkHosts):

• UIProcess/API/Cocoa/WKWebViewConfiguration.mm:

(-[WKWebViewConfiguration _setLoadsFromNetwork:]):
(-[WKWebViewConfiguration _setAllowedNetworkHosts:]):

Source/WTF:

Allow using RobinHoodHashMap / RobinHoodHashSet with URL keys since a URL wraps a String
and uses its internal String's cached hash as hash.

• wtf/URLHash.h:

4:11 PM Changeset in webkit [293194] by commit-queue@webkit.org

• 2 edits in trunk/LayoutTests

[ macOS wk2 arm64 ] fast/scrolling/mac/adjust-scroll-snap-during-gesture.html is a flaky failure
​https://bugs.webkit.org/show_bug.cgi?id=239627

Unreviewed test gardening.

• LayoutTests/platform/mac-wk2/TestExpectations:

Canonical link: ​https://commits.webkit.org/249869@main

3:56 PM Changeset in webkit [293193] by commit-queue@webkit.org

• 2 edits in trunk/LayoutTests

[ iOS 15.4 ] imported/w3c/web-platform-tests/css/css-color/opacity-overlapping-letters.html is failing
​https://bugs.webkit.org/show_bug.cgi?id=239625

Unreviewed test gardening.

• LayoutTests/platform/ios/TestExpectations:

Canonical link: ​https://commits.webkit.org/249868@main

3:31 PM Changeset in webkit [293192] by commit-queue@webkit.org

• 2 edits in trunk/LayoutTests

[ macOS wk2 arm64 ] REGRESSION (r242313): Layout Test scrollingcoordinator/mac/multiple-fixed.html is a flaky time out and image diff
​https://bugs.webkit.org/show_bug.cgi?id=195635

Unreviewed test gardening.

Adding new test expectation.

• LayoutTests/platform/mac-wk2/TestExpectations:

Canonical link: ​https://commits.webkit.org/249867@main

3:07 PM Changeset in webkit [293191] by Chris Dumez

• 2 edits in trunk/Source/WebKit

Build is broken on shipping macOS due to TBA annotations replacement in API headers
​https://bugs.webkit.org/show_bug.cgi?id=239623

Unreviewed, bump versions in WebKit.xcconfig too to disable TBA annotation replacement on shipping OS
so that we don't get warnings about API not yet being available.

• Configurations/WebKit.xcconfig:

3:07 PM Changeset in webkit [293190] by Russell Epstein

• 1 copy in tags/WebKit-7614.1.10.5

Tag WebKit-7614.1.10.5.

2:46 PM Changeset in webkit [293189] by Devin Rousso

• 3 edits in trunk/Source/WebInspectorUI

Web Inspector: Elements: Layout: hovering over a node should highlight it in the page
​https://bugs.webkit.org/show_bug.cgi?id=239621

Reviewed by Patrick Angle.

• UserInterface/Views/NodeOverlayListSection.js:

(WI.NodeOverlayListSection.prototype.layout):
Use the existing WI.linkifyNodeReference to create the UI for the WI.DOMNode.

• UserInterface/Base/DOMUtilities.js:

(WI.linkifyNodeReferenceElement):
Don't style the UI as a link if clicking it doesn't do anything.

2:03 PM Changeset in webkit [293188] by Simon Fraser

• 4 edits in trunk

<body> with overflow:hidden CSS is scrollable on iOS standalone web app
​https://bugs.webkit.org/show_bug.cgi?id=220908
<rdar://73574797>

Reviewed by Devin Rousso.

Source/WebKit:

overflow:hidden failed to make the primary scroll view unscrollable in a home screen web
app because of logic in _updateScrollViewForTransaction: that was intended to ensure that if
the current unobscured rect is equal to the max unobscured rect (i.e. testing for collapsed
UI in MobileSafari), the user could scroll to expand the UI.

This failed in Web.app because there is no collapsible UI there, i.e.
minUnobscuredSize == maxUnobscuredSize. So test for that scenario.

Tested by new API tests.

• UIProcess/API/ios/WKWebViewIOS.mm:

(-[WKWebView _updateScrollViewForTransaction:]):

Tools:

New API tests for various configurations of insets and size overrides.

• TestWebKitAPI/Tests/ios/ScrollViewInsetTests.mm:

(TestWebKitAPI::TEST):

1:57 PM Changeset in webkit [293187] by Jonathan Bedard

• 10 edits in trunk/Tools

[git-webkit] Allow caller to specify remote for PR
​https://bugs.webkit.org/show_bug.cgi?id=239452
<rdar://problem/91897384>

Reviewed by Dewei Zhu.

• Tools/Scripts/libraries/webkitscmpy/setup.py: Bump version.
• Tools/Scripts/libraries/webkitscmpy/webkitscmpy/init.py: Ditto.
• Tools/Scripts/libraries/webkitscmpy/webkitscmpy/mocks/local/git.py:

(Git.init): Allow caller to define remotes.

• Tools/Scripts/libraries/webkitscmpy/webkitscmpy/program/pull_request.py:

(PullRequest.parser): Add '--remote' option.
(PullRequest.main): Allow options to specify the remote we're making a PR against.

• Tools/Scripts/libraries/webkitscmpy/webkitscmpy/program/revert.py:

(Revert.add_comment_to_reverted_commit_bug_tracker): Derive remote from argument.

• Tools/Scripts/libraries/webkitscmpy/webkitscmpy/remote/git_hub.py:

(GitHub.PRGenerator.find): Support finding PRs if label doesn't contain username.
(GitHub.PRGenerator.create): Error 422 is a validation error, usually caused by another PR pointing to the same branch.
(GitHub.PRGenerator.update): Extract error message for more actionable errors.

• Tools/Scripts/libraries/webkitscmpy/webkitscmpy/test/clean_unittest.py:

(TestClean.test_clean_pr): Add fork remote.

• Tools/Scripts/libraries/webkitscmpy/webkitscmpy/test/pull_request_unittest.py:
• Tools/Scripts/libraries/webkitscmpy/webkitscmpy/test/revert_unittest.py:

Canonical link: ​https://commits.webkit.org/249863@main

1:51 PM Changeset in webkit [293186] by Tyler Wilcock

• 2 edits in trunk/Source/WebCore

Only AXIsolatedTree::update{Children,Node} once when processing notifications in AXObjectCache::updateIsolatedTree
​https://bugs.webkit.org/show_bug.cgi?id=239499

Reviewed by Andres Gonzalez.

Currently, we do try to prevent performing duplicate updates, but we do so only
if we have an exact match for a previous notification for any given object.
However, multiple notifications trigger AXIsolatedTree::updateNode and children,
so we'll still duplicate work if we get multiple notifications in the same group.
For example, before this patch, if the same object got both an AXValueChanged and
AXTextChanged notification, we'd call AXIsolatedTree::updateNode twice.

With this patch, we instead keep track of whether or not we've
performed a node or children update for each object to prevent
duplicate work.

No new tests because there is no behavior change.

• accessibility/AXObjectCache.cpp:

(WebCore::AXObjectCache::updateIsolatedTree):
(WebCore::appendIfNotContainsMatching): Deleted.

1:48 PM Changeset in webkit [293185] by Russell Epstein

• 9 edits in branches/safari-614.1.10-branch/Source

Versioning.

WebKit-7614.1.10.5

1:40 PM Changeset in webkit [293184] by Russell Epstein

• 1 copy in tags/WebKit-7614.1.10.4

Tag WebKit-7614.1.10.4.

1:33 PM Changeset in webkit [293183] by Matteo Flores

• 2 edits in trunk/LayoutTests

Implement ElementInternals

​https://bugs.webkit.org/show_bug.cgi?id=197960

Unreviewed test gardening.

• LayoutTests/platform/ipad/TestExpectations:

Canonical link: ​https://commits.webkit.org/249861@main

1:11 PM Changeset in webkit [293182] by Russell Epstein

• 1 edit in branches/safari-613-branch/Source/WebKit/UIProcess/ios/WebPageProxyIOS.mm

Unreviewed build fix. rdar://91584856

WebPageProxyIOS.mm:572:31: error: use of undeclared identifier 'createVersion4UUIDString'; did you mean 'createVersion4UUIDStringWeak'?

1:11 PM Changeset in webkit [293181] by Russell Epstein

• 1 edit in branches/safari-613-branch/Source/WebKit/WebProcess/WebPage/WebPage.messages.in

Unreviewed build fix. rdar://72058321

DerivedSources/WebKit/WebPageMessageReceiver.cpp:4038:99: error: indirection requires pointer operand ('UniqueRef<IPC::Encoder>' invalid)

return IPC::handleMessage<Messages::WebPage::PerformActionOnElement>(connection, decoder, *replyEncoder, this, &WebPage::performActionOnElement);

1:11 PM Changeset in webkit [293180] by Russell Epstein

• 8 edits in branches/safari-613-branch/Source/WebKit

Cherry-pick r291371. rdar://problem/72058321

1:11 PM Changeset in webkit [293179] by Russell Epstein

• 8 edits in branches/safari-613-branch/Source/WebKit

Revert r291371. rdar://problem/72058321

This reverts r293069.

10:51 AM Changeset in webkit [293178] by Devin Rousso

• 23 edits in trunk/Source/WebInspectorUI

Web Inspector: add reference page links for the Timelines Tab
​https://bugs.webkit.org/show_bug.cgi?id=239580

Reviewed by Patrick Angle.

• UserInterface/Views/TimelineView.js:

(WI.TimelineView.prototype.initialLayout): Added.

• UserInterface/Views/TimelineView.css:

(.timeline-view > .reference-page-link-container): Added.
Add a reference page link to the bottom-right of each timeline view.

• UserInterface/Views/CPUTimelineView.js:

(WI.CPUTimelineView.prototype.initialLayout):

• UserInterface/Views/HeapAllocationsTimelineView.js:
• UserInterface/Views/LayoutTimelineView.js:
• UserInterface/Views/MediaTimelineView.js:
• UserInterface/Views/MemoryTimelineView.js:
• UserInterface/Views/NetworkTimelineView.js:
• UserInterface/Views/OverviewTimelineView.js:
• UserInterface/Views/RenderingFrameTimelineView.js:
• UserInterface/Views/ScriptDetailsTimelineView.js:
• UserInterface/Views/ScriptProfileTimelineView.js:

Add WI.*.ReferencePage = WI.ReferencePage.* relationships for each timeline view.

• UserInterface/Base/ReferencePage.js:

(WI.ReferencePage): Added.
(WI.ReferencePage.prototype.get page): Added.
(WI.ReferencePage.prototype.get topic): Added.
(WI.ReferencePage.prototype.createLinkElement): Added.

• UserInterface/Base/Main.js:

(WI.createReferencePageLink): Moved to WI.ReferencePage.prototype.createLinkElement.
Create a WI.ReferencePage class that knows how to generate a <a>.

• UserInterface/Views/AuditNavigationSidebarPanel.js:

(WI.AuditNavigationSidebarPanel.prototype.showDefaultContentView):

• UserInterface/Views/AuditTestCaseContentView.js:

(WI.AuditTestCaseContentView.prototype.showRunningPlaceholder):

• UserInterface/Views/AuditTestContentView.js:

(WI.AuditTestContentView.prototype.showStoppingPlaceholder):
(WI.AuditTestContentView.prototype.showNoResultPlaceholder):
(WI.AuditTestContentView.prototype.showNoResultDataPlaceholder):
(WI.AuditTestContentView.prototype.showFilteredPlaceholder):

• UserInterface/Views/AuditTestGroupContentView.js:

(WI.AuditTestGroupContentView.prototype.showRunningPlaceholder):

• UserInterface/Views/BreakpointPopover.js:

(WI.BreakpointPopover.prototype.show):

• UserInterface/Views/CreateAuditPopover.js:

(WI.CreateAuditPopover.prototype.show):

• UserInterface/Views/Layers3DContentView.js:

(WI.Layers3DContentView.prototype.initialLayout):

• UserInterface/Views/LocalResourceOverridePopover.js:

(WI.LocalResourceOverridePopover.prototype.show):
Replace WI.createReferencePageLink with WI.ReferencePage.*.createLinkElement

10:43 AM Changeset in webkit [293177] by commit-queue@webkit.org

• 2 edits in trunk/Source/WebCore

[GBM] Debug build broken
​https://bugs.webkit.org/show_bug.cgi?id=239609

Patch by Philippe Normand <​pnormand@igalia.com> on 2022-04-21
Reviewed by Adrian Perez de Castro.

• platform/graphics/gbm/GraphicsContextGLGBM.cpp: Include Logging.h enabling use of logging facilities.

Canonical link: ​https://commits.webkit.org/249859@main

10:32 AM Changeset in webkit [293176] by commit-queue@webkit.org

• 9 edits
  2 adds in trunk/Source/JavaScriptCore

Unreviewed, reverting r293146.
​https://bugs.webkit.org/show_bug.cgi?id=239613

Broke JSC tests. Looks like EWS was blocked by label
<​https://bugs.webkit.org/show_bug.cgi?id=239411>

Reverted changeset:

"[JSC] Remove TempRegisterSet"
​https://bugs.webkit.org/show_bug.cgi?id=239578
​https://commits.webkit.org/r293146

10:16 AM Changeset in webkit [293175] by zhifei_fang@apple.com

• 6 edits
  2 adds in trunk

[git-webkit] Add revert command (237663)
​https://bugs.webkit.org/show_bug.cgi?id=237663
rdar://90039948

Reviewed by Jonathan Bedard.

• Tools/Scripts/libraries/webkitscmpy/webkitscmpy/program/pull_request.py: Break the pull request creation into serval differnt funtions so that other action can reuse it easily.

(PullRequest):
(PullRequest.check_pull_request_args):
(PullRequest.pull_request_branch_point):
(PullRequest.create_pull_request):
(PullRequest.main):

• Tools/Scripts/libraries/webkitscmpy/webkitscmpy/program/revert.py: Added.

(Revert):
(Revert.parser):
(Revert.revert_commit):
(Revert.main):

Canonical link: ​https://commits.webkit.org/249857@main

10:10 AM Changeset in webkit [293174] by Jonathan Bedard

• 3 edits in trunk/Tools

[Merge-Queue] Don't close bugs associated with test gardening
​https://bugs.webkit.org/show_bug.cgi?id=239604
<rdar://problem/92093456>

Reviewed by Ryan Haddad.

• Tools/CISupport/ews-build/steps.py:

(CloseBug.doStepIf): Skip step if we're a test gardening change.
(UpdatePullRequest):
(UpdatePullRequest.is_test_gardening): Check for "Unreviewed test gardening" in the commit message.
(UpdatePullRequest.evaluateCommand):

• Tools/CISupport/ews-build/steps_unittest.py:

Canonical link: ​https://commits.webkit.org/249856@main

9:55 AM Changeset in webkit [293173] by youenn@apple.com

• 9 edits in trunk

A suspended remote shared worker should resume when a new SharedWorker is added
​https://bugs.webkit.org/show_bug.cgi?id=239325

Reviewed by Chris Dumez.

Source/WebKit:

Check whether resuming/suspending remote shared worker in case of added/removed SharedWorker object.
Make sure to suspend the remote shared worker at launch time if needed.

Covered by updated test.

• NetworkProcess/SharedWorker/WebSharedWorker.cpp:
• NetworkProcess/SharedWorker/WebSharedWorker.h:
• NetworkProcess/SharedWorker/WebSharedWorkerServer.cpp:

LayoutTests:

• http/tests/navigation/page-cache-shared-worker-expected.txt:
• http/tests/navigation/page-cache-shared-worker.html:
• http/tests/navigation/resources/page-cache-helper-for-sharedworker.html:
• http/tests/navigation/resources/shared-worker-script.js:

9:28 AM Changeset in webkit [293172] by Jonathan Bedard

• 4 edits in trunk/Tools

[commits.webkit.org] Disable CORS
​https://bugs.webkit.org/show_bug.cgi?id=239607

Reviewed by Aakash Jain.

• Tools/Scripts/libraries/reporelaypy/reporelaypy/init.py: Bump version.
• Tools/Scripts/libraries/reporelaypy/reporelaypy/webserver.py: Disable CORS.
• Tools/Scripts/libraries/reporelaypy/setup.py: Bump version.

Canonical link: ​https://commits.webkit.org/249853@main

9:28 AM Changeset in webkit [293171] by commit-queue@webkit.org

• 2 edits in trunk/LayoutTests

[ macOS Release ] webanimations/css-transition-retargeting-during-ready-promise.html is a flaky failure
Updating test expectations to include Monterey Release
​https://bugs.webkit.org/show_bug.cgi?id=215335

Unreviewed test gardening.

• LayoutTests/platform/mac/TestExpectations:

Canonical link: ​https://commits.webkit.org/249853@main

9:21 AM Changeset in webkit [293170] by eric.carlson@apple.com

• 7 edits
  2 adds in trunk

AVSampleBufferRenderSynchronizer timeline sometimes goes backwards when playback begins
​https://bugs.webkit.org/show_bug.cgi?id=239471
rdar://91117803

Reviewed by Jer Noble.

Source/WebCore:

Test: media/media-source/media-source-current-time.html

• html/HTMLMediaElement.cpp:

(WebCore::HTMLMediaElement::mediaPlayerTimeChanged): Log current time and requested
playback rate.
(WebCore::HTMLMediaElement::setPausedInternal): Log.

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

(WebCore::MediaPlayerPrivateMediaSourceAVFObjC::MediaPlayerPrivateMediaSourceAVFObjC):
If time has gone backward, clamp to the most recent seek time.
(WebCore::MediaPlayerPrivateMediaSourceAVFObjC::playInternal): INFO_LOG -> ALWAYS_LOG
(WebCore::MediaPlayerPrivateMediaSourceAVFObjC::pauseInternal): Ditto.
(WebCore::MediaPlayerPrivateMediaSourceAVFObjC::currentMediaTime const): Use clampTimeToLastSeekTime.
(WebCore::MediaPlayerPrivateMediaSourceAVFObjC::clampTimeToLastSeekTime const): New,
clamp time value to the most recent seek time.
(WebCore::MediaPlayerPrivateMediaSourceAVFObjC::setCurrentTimeDidChangeCallback):
Use clampTimeToLastSeekTime.
(WebCore::MediaPlayerPrivateMediaSourceAVFObjC::seekWithTolerance): INFO_LOG -> ALWAYS_LOG
(WebCore::MediaPlayerPrivateMediaSourceAVFObjC::seekInternal): INFO_LOG -> ALWAYS_LOG
(WebCore::MediaPlayerPrivateMediaSourceAVFObjC::updateLastPixelBuffer): INFO_LOG -> ALWAYS_LOG
(WebCore::MediaPlayerPrivateMediaSourceAVFObjC::setHasAvailableVideoFrame): INFO_LOG -> ALWAYS_LOG
(WebCore::MediaPlayerPrivateMediaSourceAVFObjC::durationChanged): INFO_LOG -> ALWAYS_LOG

Source/WebKit:

• WebProcess/GPU/media/MediaPlayerPrivateRemote.cpp:

(WebKit::MediaPlayerPrivateRemote::currentTimeChanged): Log when time jumps backwards.

LayoutTests:

• media/media-source/media-source-current-time-expected.txt: Added.
• media/media-source/media-source-current-time.html: Added.

8:21 AM Changeset in webkit [293169] by Antti Koivisto

• 5 edits in trunk/Source/WebCore

[LFC][Integration] Add direct style accessor to inline iterator
​https://bugs.webkit.org/show_bug.cgi?id=239593

Reviewed by Alan Bujtas.

The current implementation fetches the line and the renderer and is pretty expensive.

• layout/integration/inline/InlineIteratorBox.cpp:

(WebCore::InlineIterator::Box::style const): Deleted.

• layout/integration/inline/InlineIteratorBox.h:

(WebCore::InlineIterator::Box::style const):

• layout/integration/inline/InlineIteratorBoxLegacyPath.h:

(WebCore::InlineIterator::BoxLegacyPath::style const):

• layout/integration/inline/InlineIteratorBoxModernPath.h:

(WebCore::InlineIterator::BoxModernPath::style const):

7:46 AM Changeset in webkit [293168] by Jonathan Bedard

• 3 edits in trunk/Tools

[Merge-Queue] Allow "Rubber-stamped" as review message
​https://bugs.webkit.org/show_bug.cgi?id=239601
<rdar://problem/92091819>

Reviewed by Aakash Jain.

• Tools/CISupport/ews-build/steps.py:

(ValidateCommitMessage.evaluateCommand): Allow "Rubber-stamped by."

• Tools/CISupport/ews-build/steps_unittest.py:

Canonical link: ​https://commits.webkit.org/249850@main

6:20 AM Changeset in webkit [293167] by Jonathan Bedard

• 10 edits
  2 copies in trunk/Tools

[build.webkit.org] Support identifiers on dashboard
​https://bugs.webkit.org/show_bug.cgi?id=239473
<rdar://problem/76852365>

Reviewed by Aakash Jain.

• Tools/CISupport/build-webkit-org/public_html/dashboard/Scripts/Buildbot.js:

(Buildbot.prototype.layoutTestResultsDirectoryURLForIteration): Remove leading r.

• Tools/CISupport/build-webkit-org/public_html/dashboard/Scripts/BuildbotIteration.js:

(BuildbotIteration.prototype.get buildURL): Request identifier.

• Tools/CISupport/build-webkit-org/public_html/dashboard/Scripts/Dashboard.js: OpenSource should be considered a git checkout.
• Tools/CISupport/build-webkit-org/public_html/dashboard/Scripts/Initialization.js: Add Commit interface.
• Tools/CISupport/build-webkit-org/public_html/dashboard/Scripts/Main.js: Kick off periodic commits.webkit.org refresh.
• Tools/CISupport/build-webkit-org/public_html/dashboard/Scripts/QueueView.js:

(QueueView.prototype._appendPendingRevisionCount): Use commit.webkit.org interface instead of trac.
(QueueView.prototype._formatRevisionForDisplay): Don't truncate identifiers.

• Tools/CISupport/build-webkit-org/public_html/dashboard/Scripts/tests/index.html: Import Commits.js.
• Tools/CISupport/build-webkit-org/public_html/dashboard/Scripts/tests/resources/MockCommits.js:

(MockCommits): Add mock commits.webkit.org interface.

• Tools/CISupport/build-webkit-org/public_html/dashboard/Scripts/tests/resources/tests.js:

(setup): Add commits.webkit.org interface.
(this.view._latestProductiveIteration):

• Tools/CISupport/build-webkit-org/public_html/dashboard/index.html:

Canonical link: ​https://commits.webkit.org/249849@main

5:14 AM Changeset in webkit [293166] by Carlos Garcia Campos

• 1 copy in releases/WebKitGTK/webkit-2.36.1

WebKitGTK 2.36.1

5:13 AM Changeset in webkit [293165] by Carlos Garcia Campos

• 4 edits in releases/WebKitGTK/webkit-2.36

Unreviewed. Update OptionsGTK.cmake and NEWS for 2.36.1 release

.:

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

Source/WebKit:

• gtk/NEWS: Add release notes for 2.36.1.

4:41 AM Changeset in webkit [293164] by Carlos Garcia Campos

• 2 edits in releases/WebKitGTK/webkit-2.36/Tools

Merge r293163 - [GTK][WPE] Make make-dist script python3 only
​https://bugs.webkit.org/show_bug.cgi?id=239596

Reviewed by Adrian Perez de Castro.

It fails with python2 since r291934.

• Scripts/make-dist:

4:40 AM Changeset in webkit [293163] by Carlos Garcia Campos

• 2 edits in trunk/Tools

[GTK][WPE] Make make-dist script python3 only
​https://bugs.webkit.org/show_bug.cgi?id=239596

Reviewed by Adrian Perez de Castro.

It fails with python2 since r291934.

• Scripts/make-dist:

3:53 AM Changeset in webkit [293162] by ntim@apple.com

• 2 edits
  1 add in trunk/LayoutTests

[iOS] Rebaseline fast/attachment/attachment-disabled-rendering.html expectations
​https://bugs.webkit.org/show_bug.cgi?id=238960

Unreviewed test gardening.

The page dimensions are slightly different, but the <attachment> has 0 size as expected.

• platform/ios-wk2/TestExpectations:
• platform/ios/fast/attachment/attachment-disabled-rendering-expected.txt: Added.

3:37 AM Changeset in webkit [293161] by ntim@apple.com

• 2 edits in trunk/Source/WebCore

Small cleanups to StyleBuilderCustom.cpp
​https://bugs.webkit.org/show_bug.cgi?id=239595

Reviewed by Antti Koivisto.

• Make use of DECLARE_PROPERTY_CUSTOM_HANDLERS more often
• Make code style for the containIntrinsic*Size methods consistent with

applyInitialContainIntrinsicBlockSize (use auto and better variable name)

• A bit of extra sorting

• style/StyleBuilderCustom.h:

(WebCore::Style::BuilderCustom::applyValueContainIntrinsicBlockSize):
(WebCore::Style::BuilderCustom::applyInitialContainIntrinsicInlineSize):
(WebCore::Style::BuilderCustom::applyValueContainIntrinsicInlineSize):

2:51 AM Changeset in webkit [293160] by Carlos Garcia Campos

• 3 edits in releases/WebKitGTK/webkit-2.36/Source/WebCore

Merge r292901 - [GTK] AddressSanitizer: heap-buffer-overflow in WebCore::Length::ref()
​https://bugs.webkit.org/show_bug.cgi?id=237389

Reviewed by Žan Doberšek.

• platform/graphics/nicosia/NicosiaAnimation.cpp:

(Nicosia::createThreadsafeKeyFrames): Convert Length members of transform functions to
the fixed variety before they are moved to separate threads.
(Nicosia::Animation::Animation): Use the new helper.

• platform/graphics/transforms/TranslateTransformOperation.h: Added setters.

2:51 AM Changeset in webkit [293159] by Carlos Garcia Campos

• 2 edits in releases/WebKitGTK/webkit-2.36/Source/WebKit

Merge r292628 - [WPE][2.36] Build with a11y disabled fails
​https://bugs.webkit.org/show_bug.cgi?id=239004

Patch by Philippe Normand <​pnormand@igalia.com> on 2022-04-08
Reviewed by Michael Catanzaro.

Add accessibility build guards.

• UIProcess/Launcher/glib/XDGDBusProxy.cpp:

(WebKit::XDGDBusProxy::XDGDBusProxy):
(WebKit::XDGDBusProxy::launch const):

2:51 AM WebKitGTK/2.36.x edited by Carlos Garcia Campos

(diff)

2:51 AM Changeset in webkit [293158] by Carlos Garcia Campos

• 11 edits in releases/WebKitGTK/webkit-2.36/Source

Merge r292211 - REGRESSION(r290360): [GLX] Crash on process exit
​https://bugs.webkit.org/show_bug.cgi?id=238494

Reviewed by Michael Catanzaro.

Source/WebCore:

When creating a PlatformDisplay from a shared GdkDisplay, we need to keep a reference to the display and connect
to closed signal to release our display dependent resources. Closed signal is emitted right before the native
display is actually closed by GDK. We also need to make sure that the sharing GL context is released before the
PlatformDisplay is destroyed. This is always happening for EGL because the sharing GL context is cleared before
terminating the EGL display in the atexit handler, but that's not the case of GLX. Since X11 only allows one
PlatformDisplay, which is always the shared one, we can just install an atexit handler in the constructor to
clear the sharing context before the destructor is called.

• platform/graphics/PlatformDisplay.cpp:

(WebCore::PlatformDisplay::createPlatformDisplay):
(WebCore::PlatformDisplay::PlatformDisplay):
(WebCore::PlatformDisplay::sharedDisplayDidClose):
(WebCore::PlatformDisplay::~PlatformDisplay):
(WebCore::PlatformDisplay::clearSharingGLContext):
(WebCore::PlatformDisplay::terminateEGLDisplay):

• platform/graphics/PlatformDisplay.h:
• platform/graphics/libwpe/PlatformDisplayLibWPE.cpp:

(WebCore::PlatformDisplayLibWPE::PlatformDisplayLibWPE):

• platform/graphics/wayland/PlatformDisplayWayland.cpp:

(WebCore::PlatformDisplayWayland::create):
(WebCore::PlatformDisplayWayland::PlatformDisplayWayland):
(WebCore::PlatformDisplayWayland::~PlatformDisplayWayland):
(WebCore::PlatformDisplayWayland::sharedDisplayDidClose):

• platform/graphics/wayland/PlatformDisplayWayland.h:
• platform/graphics/win/PlatformDisplayWin.h:
• platform/graphics/x11/PlatformDisplayX11.cpp:

(WebCore::PlatformDisplayX11::create):
(WebCore::clearSharingGLContextAtExit):
(WebCore::PlatformDisplayX11::PlatformDisplayX11):
(WebCore::PlatformDisplayX11::~PlatformDisplayX11):
(WebCore::PlatformDisplayX11::sharedDisplayDidClose):

• platform/graphics/x11/PlatformDisplayX11.h:

Source/WebKit:

• WebProcess/gtk/WaylandCompositorDisplay.cpp:

(WebKit::WaylandCompositorDisplay::WaylandCompositorDisplay):

2:51 AM Changeset in webkit [293157] by Carlos Garcia Campos

• 1 edit
  1 add in releases/WebKitGTK/webkit-2.36/Source/WebCore/platform/gtk/po

Merge r292212 - Croatian translation for WebKitGTK
​https://bugs.webkit.org/show_bug.cgi?id=238611

Patch by gogo <​trebelnik2@gmail.com> on 2022-04-01
Rubber-stamped by Carlos Garcia Campos.

• hr.po: Added.

2:51 AM Changeset in webkit [293156] by Carlos Garcia Campos

• 2 edits in releases/WebKitGTK/webkit-2.36/Source/WebKit

Merge r292099 - [GTK][WPE] Exit cleanly if Connection to UIProcess severed. Regression (r214307)
​https://bugs.webkit.org/show_bug.cgi?id=235224

This patch reverts the WPE and GTK specific code removed in '[WK2] Make
establishing a connection between the WebProcess and the Network process more
robust' (r12345):

GTK+ and WPE ports don't exit on send sync message failure. In those particular
cases, the network process can be terminated by the UI process while the
Web process is still initializing, so we always want to exit instead of crashing. This can
happen when the WebView is created and then destroyed quickly.
See ​https://bugs.webkit.org/show_bug.cgi?id=183348.

Reviewed by Carlos Garcia Campos.

• WebProcess/WebProcess.cpp:

(WebKit::getNetworkProcessConnection):
(WebKit::WebProcess::getGPUProcessConnection):
(WebKit::getWebAuthnProcessConnection):

2:51 AM Changeset in webkit [293155] by Carlos Garcia Campos

• 2 edits in releases/WebKitGTK/webkit-2.36/Source/WebCore

Merge r292684 - Include locale.h for LC_MESSAGES definition.
​https://bugs.webkit.org/show_bug.cgi?id=239030

Patch by Khem Raj <​raj.khem@gmail.com> on 2022-04-09
Reviewed by Michael Catanzaro.

• accessibility/atspi/AccessibilityRootAtspi.cpp:

2:51 AM Changeset in webkit [293154] by Carlos Garcia Campos

• 4 edits in releases/WebKitGTK/webkit-2.36/Source/JavaScriptCore

Merge r291815 - [JSC] JSRemoteFunction thunk should materialize code-pointer
​https://bugs.webkit.org/show_bug.cgi?id=238313

Reviewed by Mark Lam.

When invoking a JSRemoteFunction, we must first wrap the arguments passed to it.
The wrapping operation may trigger a GC, and GC can jettison JIT code. As a result,
even though we know that the target JSFunction has JIT code that we want to execute,
the JIT code may be jettisoned (while wrapping the arguments for it) before we get
to the call. This resulted in occasional crashes on the JSTests/stress/shadow-realm-evaluate.js test.

This patch fixes this by doing a null check on the JIT code just before calling it,
and if null (i.e. the JIT code has been jettisoned), re-materializing the JIT code
first before making the call.

• jit/JITOperations.cpp:

(JSC::JSC_DEFINE_JIT_OPERATION):

• jit/JITOperations.h:
• jit/ThunkGenerators.cpp:

(JSC::remoteFunctionCallGenerator):

2:51 AM Changeset in webkit [293153] by Carlos Garcia Campos

• 4 edits in releases/WebKitGTK/webkit-2.36

Merge r291785 - [JSC] Fix remoteFunctionCallGenerator on MIPS
​https://bugs.webkit.org/show_bug.cgi?id=238274

Patch by Geza Lore <Geza Lore> on 2022-03-23
Reviewed by Yusuke Suzuki.

JSTests:

• stress/shadow-realm-evaluate.js:

Source/JavaScriptCore:

Eliminate register conflict (nonArgGPR0 == regT4, which is used in
valueRegs)

• jit/ThunkGenerators.cpp:

(JSC::remoteFunctionCallGenerator):

2:51 AM Changeset in webkit [293152] by Carlos Garcia Campos

• 13 edits in releases/WebKitGTK/webkit-2.36/Source/JavaScriptCore

Merge r290788 - Add a DeferTraps scope
​https://bugs.webkit.org/show_bug.cgi?id=237306
<rdar://83494949>

Reviewed by Mark Lam.

• bytecode/CodeBlock.cpp:

(JSC::CodeBlock::CodeBlock):
(JSC::CodeBlock::jettison):
(JSC::CodeBlock::noticeIncomingCall):

• bytecode/CodeBlock.h:
• bytecode/RepatchInlines.h:

(JSC::linkFor):
(JSC::virtualForWithFunction):

• dfg/DFGOperations.cpp:

(JSC::DFG::JSC_DEFINE_JIT_OPERATION):

• interpreter/Interpreter.cpp:

(JSC::Interpreter::executeProgram):
(JSC::Interpreter::executeCall):
(JSC::Interpreter::executeConstruct):
(JSC::Interpreter::execute):
(JSC::Interpreter::executeModuleProgram):

• interpreter/InterpreterInlines.h:

(JSC::Interpreter::execute):

• llint/LLIntSlowPaths.cpp:

(JSC::LLInt::setUpCall):

• runtime/ExceptionScope.h:
• runtime/ScriptExecutable.cpp:

(JSC::ScriptExecutable::installCode):

• runtime/VMTraps.cpp:

(JSC::VMTraps::handleTraps):
(JSC::VMTraps::takeTopPriorityTrap):

• runtime/VMTraps.h:

(JSC::VMTraps::needHandling const):
(JSC::VMTraps::maybeNeedHandling const):
(JSC::VMTraps::hasTrapBit):
(JSC::VMTraps::setTrapBit):

• runtime/VMTrapsInlines.h:

(JSC::DeferTraps::DeferTraps):
(JSC::DeferTraps::~DeferTraps):

2:51 AM Changeset in webkit [293151] by Carlos Garcia Campos

• 5 edits
  4 adds in releases/WebKitGTK/webkit-2.36

Merge r291813 - Safari can't play video completely at bilibili.com
​https://bugs.webkit.org/show_bug.cgi?id=236440
rdar://88761053

Reviewed by Jer Noble.

Source/WebCore:

Video frames were incorrectly evicted during a call to appendBuffer
as the Source Buffer incorrectly assumed a discontinuity was present.

When appending data to a source buffer, the MSE specs describe a method
to detect discontinuities in the Coded Frame Processing algorithm
(​https://www.w3.org/TR/media-source/#sourcebuffer-coded-frame-processing)
step 6:
"

• If last decode timestamp for track buffer is set and decode timestamp is less than last decode timestamp:

OR

• If last decode timestamp for track buffer is set and the difference between decode timestamp and last decode timestamp is greater than 2 times last frame duration.

"
The issue being what defines the last frame duration.
Is it the frame last seen in the coded frame processing loop or the frame
whose presentation timestamp is just before the one we are currently
processing.

H264 and HEVC have a concept of b-frames: that is a frame that depends
on a future frame to be decoded.
Those frames are found in the container and can be identified by their
presentation timestamp higher than the frame following in decode order.
Those present a challenge as the frame prior the current one in
presentation order, may actually only be found several frames back in
decode order.
Bug 181891 attempted to fix a similar issue, and used the longest
"decode duration" as a workaround to detect discontinuity in the content.
It mentioned adopting the same technique as in Mozilla's MSE
implementation, but Mozilla also skip discontinuity detection within a
media segment (​https://www.w3.org/TR/media-source/#media-segment which for
fMP4 is a single moof box) an approach that can't be achieved with
CoreMedia's AVStreamDataParser.
As mentioned in bug 181891, CoreMedia ignore the decode timestamps' delta
and juggles with the sample's duration so that there's no discontinuity
in the demuxed samples' presentation time, causing false positive in the
gap detection algorithm.

Bilibili uses HEVC content, and uses an encoding that generate lots
of b-frames, with a very wide sliding window (seen up to 12 frames).
By using the longest frame duration found in either presentation or
decode duration as threshold to identify a discontinuity, we can
properly parse the content and not incorrectly evict appended frames.
(As a side note, the use of HEVC with B-Frames is peculiar as not all
hardware support it.)
It is difficult to identify here if the issue is within the bilibili's
content or CoreMedia's output, though the responsibility more than
likely lies with bilibili.

Test: media/media-source/media-mp4-hevc-bframes.html

• platform/graphics/SourceBufferPrivate.cpp:

(WebCore::SourceBufferPrivate::TrackBuffer::TrackBuffer):
(WebCore::SourceBufferPrivate::resetTrackBuffers):
(WebCore::SourceBufferPrivate::didReceiveSample):

• platform/graphics/SourceBufferPrivate.h:

LayoutTests:

• media/media-source/content/test-bframes-hevc-manifest.json: Added.
• media/media-source/content/test-bframes-hevc.mp4: Added.
• media/media-source/media-mp4-hevc-bframes-expected.txt: Added.
• media/media-source/media-mp4-hevc-bframes.html: Added.

2:50 AM Changeset in webkit [293150] by Carlos Garcia Campos

• 2 edits in releases/WebKitGTK/webkit-2.36/Source/WebKit

Merge r291621 - Ensure that proxies are invalidated before destroying them.
​https://bugs.webkit.org/show_bug.cgi?id=237187

Reviewed by Carlos Garcia Campos.

• Shared/CoordinatedGraphics/CoordinatedGraphicsScene.cpp:

(WebKit::CoordinatedGraphicsScene::updateSceneState):

2:35 AM Changeset in webkit [293149] by vjaquez@igalia.com

• 3 edits in trunk/Source/WebCore

[GTK][WPE] Update libwebrtc API usage for video encoder factory
​https://bugs.webkit.org/show_bug.cgi?id=239558

Reviewed by Philippe Normand.

Update GstreamerVideoEncoderFactory with current libwebrtc API.

No new tests required.

• platform/mediastream/libwebrtc/gstreamer/GStreamerVideoEncoderFactory.cpp:

(WebCore::GStreamerVideoEncoder::GStreamerVideoEncoder): Chain main constructor rather to
copy its initializations.
Update InitEncode method signature.
Remove deprecated attributes of EncoderInfo.
Fix code-style.
(WebCore::GStreamerH264Encoder::GStreamerH264Encoder): Remove unused class attribute and fix
a memleak.

• platform/mediastream/libwebrtc/gstreamer/GStreamerVideoEncoderFactory.h: Remove deprecated

method.

2:28 AM Changeset in webkit [293148] by Diego Pino Garcia

• 2 edits in trunk/Source/WTF

REGRESSION(r292863): std::remove_cvref_t not defined in GCC9<
​https://bugs.webkit.org/show_bug.cgi?id=239482

Reviewed by Žan Doberšek.

std::remove_cvref_t is only available since GCC9 onwards:

​https://en.cppreference.com/w/cpp/compiler_support/20

• wtf/StdLibExtras.h: add remove_cvref_t to std namespace if not defined.

1:39 AM Changeset in webkit [293147] by commit-queue@webkit.org

• 2 edits in trunk/Source/WebCore

Skip release assert during container query resolution
​https://bugs.webkit.org/show_bug.cgi?id=239537

Patch by Rob Buis <​rbuis@igalia.com> on 2022-04-21
Reviewed by Antti Koivisto.

Skip release assert during container query resolution
since resolveStyle will be a no-op as the container
query resolution was started in resolveStyle, so calling
resolveStyle the second time will bail right away.

• dom/Document.cpp:

(WebCore::Document::updateStyleIfNeeded):