Timeline

Jul 23, 2020:

10:31 PM Changeset in webkit [264812] by Wenson Hsieh

• 11 edits in trunk

Tapping QuickType suggestions for a misspelled word does nothing in Mail compose
​https://bugs.webkit.org/show_bug.cgi?id=214705
<rdar://problem/58320785>

Reviewed by Tim Horton.

Source/WebKit:

In WebKit2-based editing contexts on iOS (such as Mail compose), tapping on a misspelled word will result in
QuickType suggestions in the input accessory row either showing up blank (i.e. with no suggestions), or will
show suggestions that are contextual to the text selection prior to selecting the misspelled word. Subsequently,
attempting to choose of these suggestions results in no text being inserted. This happens for a couple of
reasons:

1. UIKit doesn't attempt to re-request the autocorrection context after tapping on a misspelled word. UIKit

just uses -selectWordForReplacement to extend the selection over the misspelled word, which doesn't
currently fire any of the selection change callbacks on UIInputDelegate. As a result, UIKit will instead
believe that the selection is collapsed to the start or end of the misspelled word.

2. Upon attempting to compute text suggestions for QuickType, -[UIKeyboardImpl replacementsFromSelectedText]

uses -textInRange: to grab the currently selected text, which (in WKContentView) is currently hard-coded
to always return nil.

We address (1) by bookending -selectWordForReplacement with calls to -beginSelectionChange and
-endSelectionChange, and address (2) by teaching -textInRange: to return the current text selection, only in
the case where the given UITextRange is equal to the current ranged selection.

Test: SelectionTests.SelectedTextAfterSelectingWordForReplacement

• UIProcess/WebPageProxy.h:

(WebKit::WebPageProxy::extendSelection):

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

(-[WKContentView cleanUpInteraction]):
(-[WKContentView textInRange:]):

Special-case the scenario where the input text range is exactly the same as self.selectedTextRange by
returning the currently selected text that is cached on EditorState. In all other cases, continue to return
nil.

(-[WKContentView selectedTextRange]):

Cache the current selected text range on WKContentView, such that repeated calls to selectedTextRange when the
selection has not changed does not require a new WKTextRange to be constructed every time.

(-[WKContentView _selectionChanged]):

Invalidate the cached selected text range whenever the selection has changed.

(-[WKContentView selectWordForReplacement]):

Use the completion callback added below to bookend this method with -(begin|end)SelectionChange.

(-[WKContentView _updateChangedSelection:]):

Invalidate the cached selected text range whenever we're about to tell UIKit to repaint the selection.

• UIProcess/ios/WebPageProxyIOS.mm:

(WebKit::WebPageProxy::extendSelection):

Add an optional completion callback to extendSelection.

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

(WebKit::WebPage::extendSelection):

Tools:

• TestWebKitAPI/Tests/ios/SelectionByWord.mm:

Add a unit test to exercise the changes in behavior when calling UITextInput API.

(-[SelectionChangeListener setSelectionWillChangeHandler:]):
(-[SelectionChangeListener setSelectionDidChangeHandler:]):
(-[SelectionChangeListener selectionWillChangeHandler]):
(-[SelectionChangeListener selectionDidChangeHandler]):
(-[SelectionChangeListener selectionWillChange:]):
(-[SelectionChangeListener selectionDidChange:]):
(-[SelectionChangeListener textWillChange:]):
(-[SelectionChangeListener textDidChange:]):

• TestWebKitAPI/ios/UIKitSPI.h:

10:04 PM Changeset in webkit [264811] by commit-queue@webkit.org

• 27 edits
  1 add in trunk/Source

Split Windows parts from CFNetworkSPI.h to CFNetworkSPIWin.h
​https://bugs.webkit.org/show_bug.cgi?id=214719

Patch by Alex Christensen <​achristensen@webkit.org> on 2020-07-23
Reviewed by Tim Horton.

Source/WebCore:

No change in behavior.
Windows CFNetwork work is not actively developed, but Cocoa is.
This makes it easier to find where to put new things.

• platform/network/NetworkStorageSession.h:
• platform/network/ResourceHandleClient.h:
• platform/network/ResourceHandleInternal.h:
• platform/network/cf/CookieStorageCFNet.cpp:
• platform/network/cf/CredentialStorageCFNet.cpp:
• platform/network/cf/NetworkStorageSessionCFNetWin.cpp:
• platform/network/cf/ResourceErrorCF.cpp:
• platform/network/cf/ResourceHandleCFNet.cpp:
• platform/network/cf/ResourceHandleCFURLConnectionDelegate.cpp:
• platform/network/cf/ResourceHandleCFURLConnectionDelegateWithOperationQueue.cpp:
• platform/network/cf/ResourceRequestCFNet.cpp:
• platform/network/cf/ResourceRequestCFNet.h:
• platform/network/cf/ResourceResponse.h:
• platform/network/cf/ResourceResponseCFNet.cpp:
• platform/network/cf/SocketStreamHandleImplCFNet.cpp:

Source/WebCore/PAL:

• pal/PlatformAppleWin.cmake:
• pal/spi/cf/CFNetworkSPI.h:
• pal/spi/win/CFNetworkSPIWin.h: Added.

Source/WebKitLegacy/win:

• WebCache.cpp:
• WebCoreSupport/WebFrameNetworkingContext.cpp:
• WebCoreSupport/WebPlatformStrategies.cpp:
• WebError.cpp:
• WebURLResponse.cpp:
• WebView.cpp:

9:48 PM Changeset in webkit [264810] by Peng Liu

• 2 edits in trunk/Source/WebKit

Safari Related Leaked Assertions after Playing YouTube Video
​https://bugs.webkit.org/show_bug.cgi?id=214706

Reviewed by Jer Noble.

A 10 second grace period before removing the activity token was introduced in r252692.
However, some websites, e.g., youtube.com, requests the video element to play the
next video 10 seconds after completing the current one (when the Safari is in the background).
That can lead to a race condition between the web process and the UI process. They may send
IPC messages to each other at the same time:
1) The web process sends a WebPageProxy::IsPlayingMediaDidChange message to the UI process
to notify that video/audio is playing, so the UI process will take a MediaPlayback assertion.
2) The UI process sends a WebProcess::PrepareToSuspend message to the web process, which
will suspend the web process.

Eventually, the web process will suspend, which means no video/audio playback.
But the UI process believes that there is a web process playing video/audio, so it will
take a MediaPlayback assertion.

This patch fixes the race condition by ignoring the WebPageProxy::IsPlayingMediaDidChange
message in the UIProcess if the process throttler believes that the process should not be
running, so that the MediaPlayback assertion will not be created.

• UIProcess/WebPageProxy.cpp:

(WebKit::WebPageProxy::isPlayingMediaDidChange):

8:53 PM Changeset in webkit [264809] by ysuzuki@apple.com

• 7 edits
  1 add in trunk

[JSC] Arrow function |this| resolution should not be trapped by with-scope
​https://bugs.webkit.org/show_bug.cgi?id=214716
<rdar://problem/65980639>

Reviewed by Darin Adler.

JSTests:

• stress/proxy-trap-this.js: Added.

(shouldNotThrow):
(const.handler.has):
(test):

Source/JavaScriptCore:

We were using usual "this" named variable in lexical-environment to load and store arrow-function's |this|.
But since this looks normal variable, it can be trapped by "with" scope's object while it should not be.
We use thisPrivateName instead to avoid this behavior since Proxy does not trap private names.

• builtins/BuiltinNames.h:
• bytecompiler/BytecodeGenerator.cpp:

(JSC::BytecodeGenerator::initializeArrowFunctionContextScopeIfNeeded):
(JSC::BytecodeGenerator::variable):
(JSC::BytecodeGenerator::createVariable):
(JSC::BytecodeGenerator::emitLoadThisFromArrowFunctionLexicalEnvironment):
(JSC::BytecodeGenerator::emitPutThisToArrowFunctionContextScope):

• bytecompiler/NodesCodegen.cpp:

(JSC::HasOwnPropertyFunctionCallDotNode::emitBytecode):
(JSC::ForInNode::emitBytecode):

• runtime/CommonIdentifiers.cpp:

(JSC::CommonIdentifiers::CommonIdentifiers):

• runtime/CommonIdentifiers.h:

7:28 PM Changeset in webkit [264808] by commit-queue@webkit.org

• 2 edits in trunk/Source/WebCore

Remove dead code in ResourceHandleCFNet.cpp
​https://bugs.webkit.org/show_bug.cgi?id=214717

Patch by Alex Christensen <​achristensen@webkit.org> on 2020-07-23
Reviewed by Darin Adler.

ResourceHandleCFNet.cpp is not compiled on Cocoa platforms any more.
Remove Cocoa-specific code in that file, which was used on iOS before adopting NSURLSession.

• platform/network/cf/ResourceHandleCFNet.cpp:

(WebCore::ResourceHandle::createCFURLConnection):
(WebCore::ResourceHandle::start):
(WebCore::ResourceHandle::tryHandlePasswordBasedAuthentication):
(WebCore::ResourceHandle::receivedCredential):
(WebCore::ResourceHandle::schedule): Deleted.
(WebCore::ResourceHandle::unschedule): Deleted.

5:50 PM Changeset in webkit [264807] by commit-queue@webkit.org

• 10 edits in trunk

[WebGL2] Query Objects
​https://bugs.webkit.org/show_bug.cgi?id=126940

Patch by James Darpinian <James Darpinian> on 2020-07-23
Reviewed by Dean Jackson.

Source/WebCore:

Passes conformance tests webgl/2.0.0/conformance2/query/*

• html/canvas/WebGL2RenderingContext.cpp:

(WebCore::WebGL2RenderingContext::deleteQuery):
(WebCore::WebGL2RenderingContext::isQuery):
(WebCore::WebGL2RenderingContext::validateQueryTarget):
(WebCore::WebGL2RenderingContext::beginQuery):
(WebCore::WebGL2RenderingContext::endQuery):
(WebCore::WebGL2RenderingContext::getQuery):
(WebCore::WebGL2RenderingContext::getQueryParameter):

• html/canvas/WebGL2RenderingContext.h:
• html/canvas/WebGLQuery.cpp:

(WebCore::WebGLQuery::deleteObjectImpl):

• html/canvas/WebGLQuery.h:
• html/canvas/WebGLRenderingContextBase.cpp:

(WebCore::WebGLRenderingContextBase::validateWebGLObject):

• platform/graphics/angle/GraphicsContextGLANGLE.cpp:

(WebCore::GraphicsContextGLOpenGL::deleteQuery):
(WebCore::GraphicsContextGLOpenGL::isQuery):
(WebCore::GraphicsContextGLOpenGL::getQuery):

LayoutTests:

• webgl/2.0.0/conformance2/query/query-expected.txt:
• webgl/2.0.0/conformance2/state/gl-object-get-calls-expected.txt:

4:54 PM Changeset in webkit [264806] by pvollan@apple.com

• 4 edits in trunk/Source/WebKit

Crash in block in WebKit::LaunchServicesDatabaseObserver::LaunchServicesDatabaseObserver
​https://bugs.webkit.org/show_bug.cgi?id=214702

Reviewed by Chris Dumez.

Calling xpc_connection_send_message in the block in WebKit::LaunchServicesDatabaseObserver::LaunchServicesDatabaseObserver will
sometimes crash because the XPC connection object is invalid. The block is called on a secondary thread, and the m_connections
member variable needs to be protected by a lock. Also, make sure the XPC connections objects in m_connections are valid by
retaining them when captured in a block. This is done by redeclaring a local XPC connection variable as an OSObjectPtr instead
of a raw XPC connection object.

No new tests, since I have not been able to reproduce this issue.

• NetworkProcess/cocoa/LaunchServicesDatabaseObserver.h:
• NetworkProcess/cocoa/LaunchServicesDatabaseObserver.mm:

(WebKit::LaunchServicesDatabaseObserver::LaunchServicesDatabaseObserver):
(WebKit::LaunchServicesDatabaseObserver::startObserving):
(WebKit::LaunchServicesDatabaseObserver::handleEvent):

• Shared/Cocoa/XPCEndpoint.mm:

(WebKit::XPCEndpoint::XPCEndpoint):

4:42 PM Changeset in webkit [264805] by commit-queue@webkit.org

• 36 edits
  5 adds in trunk/Source/WebCore

Infrastructure Work for Integrating CoreImage for Accelerated CSS/SVG Filter Rendering
​https://bugs.webkit.org/show_bug.cgi?id=213672

Patch by Guowei Yang <​guowei_yang@apple.com> on 2020-07-23
Reviewed by Simon Fraser.

This patch is the infrastructure work to enable filter rendering using
CoreImage. The design of this code is as follows:

0) changing WebCore configuration for using CoreImage framework from

iOS/mac only to unconditional, so watchOS and tvOS could also use
CoreImage

1) adding a new API to obtain the IOSurface from an AcceleratedImageBuffer

since CoreImage needs to render to an IOSurface

2) Instead of applying the filters inside FilterEffect, a new class,

FilterEffectRendererCoreImage is designed to handle the rendering. We hide
the implementation details by subclassing from a generic FilterEffectRenderer class.
This also has the benefit of organizing CoreImage obj-c calls in one place. This class
owns a FilterEffect node, that is the last FilterEffect of a given filter graph.
The filter graph will be traversed backwards and filter parameters is gathered
along the way. Then CIFilters will be created and connected to a larger CIFilter
graph. The output of the CIFilter graph will be rendered to the IOSurface of
a newly created AcceleratedImageBuffer. This image buffer will be returned
to RenderLayerFilters and rendered to the screen

3) Inside CSSFilter::build, we first try to create a FilterEffectRenderer by passing in

a renderer setting. A valid pointer will be returned only when the following two conditions
are met:

. CoreImage rendering is enabled in settings
. All the filters inside the filter graph is implemented using CIFilters

No new tests are required as this is the
infrastructure code for integrating CoreImage
This patch should pass all the existing css3/filters tests

• Configurations/WebCore.xcconfig: modified flag definition

so that CoreImage is used on all Apple Platforms, not only iPhone.

• Sources.txt: Added the generic FilterEffectRenderer class file
• SourcesCocoa.txt: added FilterEffectRendererCoreImage.mm
• WebCore.xcodeproj/project.pbxproj: Reflecting newly added files,

with some ordering changes

• platform/graphics/ConcreteImageBuffer.h: overrides isAccelerated()

function that returns true only if the backend is IOSurface

• platform/graphics/ImageBuffer.h: defined virtual function isAccelerated()
• platform/graphics/ImageBufferBackend.h: defined virtual function isAccelerated() that returns false in the base class

(WebCore::ImageBufferBackend::isAccelerated const): returns false for the base class

• platform/graphics/PlatformImageBuffer.h: added SPECIALIZE_TYPE_TRAITS to check what type of ImageBufferBackend is
• platform/graphics/PlatformImageBuffer.h: Added API to obtain IOSurface from IOSurfaceBackend; added SPECIALIZE_TYPE_TRAITS

(WebCore::AcceleratedImageBuffer::surface): Obtain the IOSurface pointer

from ImageBufferIOSurfaceBackend

(isType): check if a given ImageBuffer is type AcceleratedImageBuffer

• platform/graphics/cg/ImageBufferIOSurfaceBackend.cpp:

(WebCore::ImageBufferIOSurfaceBackend::surface): added an API to get the

IOSurface pointer

(WebCore::ImageBufferIOSurfaceBackend::isAccelerated const): returns true

because it is an accelerated image buffer

• platform/graphics/cg/ImageBufferIOSurfaceBackend.h: added function surface()

and isAccelerated()

• platform/graphics/coreimage/FilterEffectRendererCoreImage.h: Added.
• platform/graphics/coreimage/FilterEffectRendererCoreImage.mm: Added.

(WebCore::FilterEffectRendererCoreImage::tryCreate): creates a FilterEffectRendererCoreImage

unique pointer. If the filters inside the graph are not all implemented,
we return nullptr

(WebCore::FilterEffectRendererCoreImage::supportsCoreImageRendering):

a function that takes in FilterEffect& and returns true if its a filter effect
type that is implemented in CoreImage, false otherwise

(WebCore::FilterEffectRendererCoreImage::applyEffects): takes in a FilterEffect pointer,

calls FilterEffectRendererCoreImage::connectCIFilters to create the final output CIImage

(WebCore::FilterEffectRendererCoreImage::connectCIFilters): recursive

helper function that traverses the FilterEffect graph backwards,
creates CIFilter graph based on the subclass type of current processing
FilterEffect node. It will cache the CIFilter in a hashmap, whose key
is the corresponding FilterEffect pointer.

(WebCore::FilterEffectRendererCoreImage::canRenderUsingCIFilters): unction

that takes in the last FilterEffect node and determines whether
all of the filters inside the filter graph is already implemented using
CIFilters. It returns false as long as there is one filter that hasn't
been implemented. The rendering procedure will fallback to software mode
if this function returns false

(WebCore::FilterEffectRendererCoreImage::output const): returns the ImageBuffer

that contains the data of the final output image

(WebCore::FilterEffectRendererCoreImage::renderToImageBuffer): r

enders CIImage* to an ImageBuffer

(WebCore::FilterEffectRendererCoreImage::destRect const): calculates the destination

bounding box

(WebCore::FilterEffectRendererCoreImage::clearResult): clears all the saved results
(WebCore::FilterEffectRendererCoreImage::FilterEffectRendererCoreImage): constructor

• platform/graphics/filters/FEBlend.cpp:

(WebCore::FEBlend::FEBlend): Pass an additional FilterEffect::Type argument

to base class constructor

• platform/graphics/filters/FEColorMatrix.cpp:

(WebCore::FEColorMatrix::FEColorMatrix): ditto

• platform/graphics/filters/FEComponentTransfer.cpp:

(WebCore::FEComponentTransfer::FEComponentTransfer): ditto

• platform/graphics/filters/FEComposite.cpp:

(WebCore::FEComposite::FEComposite): ditto

• platform/graphics/filters/FEConvolveMatrix.cpp:

(WebCore::FEConvolveMatrix::FEConvolveMatrix): ditto

• platform/graphics/filters/FEDiffuseLighting.cpp:

(WebCore::FEDiffuseLighting::FEDiffuseLighting):ditto

• platform/graphics/filters/FEDisplacementMap.cpp:

(WebCore::FEDisplacementMap::FEDisplacementMap):ditto

• platform/graphics/filters/FEDisplacementMap.h:
• platform/graphics/filters/FEDropShadow.cpp:

(WebCore::FEDropShadow::FEDropShadow):ditto

• platform/graphics/filters/FEFlood.cpp:

(WebCore::FEFlood::FEFlood):ditto

• platform/graphics/filters/FEGaussianBlur.cpp:

(WebCore::FEGaussianBlur::FEGaussianBlur):ditto

• platform/graphics/filters/FELighting.cpp:

(WebCore::FELighting::FELighting):ditto

• platform/graphics/filters/FELighting.h:
• platform/graphics/filters/FEMerge.cpp:

(WebCore::FEMerge::FEMerge): ditto

• platform/graphics/filters/FEMorphology.cpp:

(WebCore::FEMorphology::FEMorphology): ditto

• platform/graphics/filters/FEOffset.cpp:

(WebCore::FEOffset::FEOffset): ditto

• platform/graphics/filters/FESpecularLighting.cpp:

(WebCore::FESpecularLighting::FESpecularLighting): ditto

• platform/graphics/filters/FETile.cpp:

(WebCore::FETile::FETile): ditto

• platform/graphics/filters/FETurbulence.cpp:

(WebCore::FETurbulence::FETurbulence): ditto

• platform/graphics/filters/FilterEffect.cpp:

(WebCore::FilterEffect::FilterEffect): ditto

• platform/graphics/filters/FilterEffect.h: added a const member, m_filterEffectType

(WebCore::FilterEffect::filterEffectClassType): returns the FilterEffect

class type

• platform/graphics/filters/FilterEffectRenderer.cpp: Added.

(WebCore::FilterEffectRenderer::tryCreate): tries creates a new FilterEffectRenderer

unique pointer. Here, we check whether the renderer setting has
CoreImage accelerated filter rendering enabled; we also check if all the
filters in the given FilterEffect graph is already implemented using
CoreImage. If both conditions are satisfied, create a FilterEffectRendererCoreImage
otherwise returns nullptr.

• platform/graphics/filters/FilterEffectRenderer.h: Added.
• platform/graphics/filters/SourceAlpha.cpp:

(WebCore::SourceAlpha::SourceAlpha): Pass an additional FilterEffect::Type argument

to base class constructor

• platform/graphics/filters/SourceAlpha.h:
• platform/graphics/filters/SourceGraphic.h:

(WebCore::SourceGraphic::SourceGraphic): ditto

• rendering/CSSFilter.cpp:

(WebCore::CSSFilter::build): we try to create a FilterEffectRenderer here by passing

in the state of the CoreImage Accelerated Rendering feature switch

(WebCore::CSSFilter::apply): If we have a valid m_filterRenderer pointer, that means

we can render the result image using CoreImage path, thus we don't call effect.apply()
fall back to software mode if hasResult() returns false.

(WebCore::CSSFilter::output const): if we have a valid m_filterRenderer, then we will

obtain the ImageBuffer result from it, not FilterEffect

(WebCore::CSSFilter::outputRect const): if we have a valid m_filterRenderer, then we will

obtain the ImageBuffer result from it, not FilterEffect

• rendering/CSSFilter.h:
• svg/graphics/filters/SVGFEImage.cpp:

(WebCore::FEImage::FEImage): Pass an additional FilterEffect::Type argument

to base class constructor

4:08 PM Changeset in webkit [264804] by ysuzuki@apple.com

• 3 edits in trunk/Source/JavaScriptCore

[JSC] FTL OSR entry should store boxed |this|
​https://bugs.webkit.org/show_bug.cgi?id=214675
<rdar://problem/65474072>

Reviewed by Michael Saboff and Mark Lam.

In this patch, after ensuring that we will go to FTL OSR entry, we store boxed |this| instead of the unboxed value
to agree to the FTL assumption that all arguments should be boxed.

• dfg/DFGOperations.cpp:
• ftl/FTLOSREntry.cpp:

(JSC::FTL::prepareOSREntry):

3:48 PM Changeset in webkit [264803] by BJ Burg

• 2 edits in trunk/Source/WebKit

Web Inspector: developerExtrasEnabled should be respected when opening local Web Inspector (part 2)
​https://bugs.webkit.org/show_bug.cgi?id=214669

Reviewed by Devin Rousso and Joseph Pecoraro.

• UIProcess/Inspector/WebInspectorProxy.cpp:

(WebKit::WebInspectorProxy::save):
(WebKit::WebInspectorProxy::append): Add missing check for developerExtrasEnabled.

3:35 PM Changeset in webkit [264802] by ysuzuki@apple.com

• 3 edits
  2 adds in trunk

We should have exception check after promise operation
​https://bugs.webkit.org/show_bug.cgi?id=214610
<rdar://problem/65881794>

Reviewed by Youenn Fablet and Mark Lam.

Source/WebCore:

Test: js/dom/promise-should-have-exception-check-on-operation.html

Add exception check after promise operations.

• bindings/js/JSDOMPromiseDeferred.cpp:

(WebCore::DeferredPromise::reject):

LayoutTests:

• js/dom/promise-should-have-exception-check-on-operation-expected.txt: Added.
• js/dom/promise-should-have-exception-check-on-operation.html: Added.

3:33 PM Changeset in webkit [264801] by keith_miller@apple.com

• 5 edits
  2 deletes in trunk

iOS simulator does not support WebAssembly
​https://bugs.webkit.org/show_bug.cgi?id=191064

Reviewed by Yusuke Suzuki.

Source/WTF:

Now that it seems we can JIT in the Simulator we should enable
WebAssembly there. This just means enabling FTL/B3 so WASM gets
enabled.

• wtf/PlatformEnable.h:
• wtf/PlatformEnableCocoa.h:

LayoutTests:

Remove simulator only expectations/failures now that WebAssembly is available in the simulator.

• platform/ios-simulator/TestExpectations:
• platform/ios-simulator/imported/w3c/web-platform-tests/xhr/send-data-sharedarraybuffer.any-expected.txt:
• platform/ios-simulator/imported/w3c/web-platform-tests/xhr/send-data-sharedarraybuffer.any.worker-expected.txt:
• platform/ios-wk2/imported/w3c/web-platform-tests/webaudio/the-audio-api/the-audiobuffer-interface/audiobuffer-copy-channel-expected.txt:

3:28 PM Changeset in webkit [264800] by Alan Coon

• 1 copy in tags/Safari-610.1.23.1.1

Tag Safari-610.1.23.1.1.

3:28 PM Changeset in webkit [264799] by Russell Epstein

• 1 delete in tags/Safari-610.1.23.1.1

Delete tag.

3:26 PM Changeset in webkit [264798] by Ryan Haddad

• 3 edits in trunk/LayoutTests

[ iOS wk2 ] imported/w3c/web-platform-tests/web-share/share-without-user-gesture.https.html is a constant timeout
​https://bugs.webkit.org/show_bug.cgi?id=214694

Unreviewd test gardening.

• platform/ios-simulator-wk2/TestExpectations:
• platform/ios/TestExpectations: Skip the test since it is a consistent timeout.

3:25 PM Changeset in webkit [264797] by Russell Epstein

• 2 edits in branches/safari-610.1.23.1-branch/Source/WebCore

Cherry-pick r264796. rdar://problem/66016644

[Cocoa|WebM] Seeking can leave parser in hung state
​https://bugs.webkit.org/show_bug.cgi?id=214704
<rdar://problem/66009420>

Reviewed by Eric Carlson.

When the parser is explicitly reset, make sure to clear out all the tracks and initialization datas. Otherwise, a half-filled
CMSampleBuffer can be left waiting around for more data, which may never arrive, and no further samples will ever be parsed.

• platform/graphics/cocoa/SourceBufferParserWebM.cpp: (WebCore::SourceBufferParserWebM::resetParserState):

git-svn-id: ​https://svn.webkit.org/repository/webkit/trunk@264796 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3:21 PM Changeset in webkit [264796] by jer.noble@apple.com

• 2 edits in trunk/Source/WebCore

[Cocoa|WebM] Seeking can leave parser in hung state
​https://bugs.webkit.org/show_bug.cgi?id=214704
<rdar://problem/66009420>

Reviewed by Eric Carlson.

When the parser is explicitly reset, make sure to clear out all the tracks and initialization datas. Otherwise, a half-filled
CMSampleBuffer can be left waiting around for more data, which may never arrive, and no further samples will ever be parsed.

• platform/graphics/cocoa/SourceBufferParserWebM.cpp:

(WebCore::SourceBufferParserWebM::resetParserState):

3:20 PM Changeset in webkit [264795] by aakash_jain@apple.com

• 4 edits in trunk/Tools

[ews] Notify bot watchers about pre-existing and flaky test failures found on ews bots
​https://bugs.webkit.org/show_bug.cgi?id=207475

Reviewed by Jonathan Bedard.

• BuildSlaveSupport/ews-build/steps.py:

(ReRunWebKitTests.evaluateCommand):
(ReRunWebKitTests.send_email_for_flaky_failure): Method to send email for flaky failure.
(AnalyzeLayoutTestsResults.report_pre_existing_failures):
(AnalyzeLayoutTestsResults.send_email_for_flaky_failure): Method to send email for flaky failure.
(AnalyzeLayoutTestsResults.send_email_for_pre_existing_failure): Method to send email for pre-existing failure.

• BuildSlaveSupport/ews-build/steps.py: Updated unit-tests accordingly.
• BuildSlaveSupport/ews-build/send_email.py: Added error handling.

3:17 PM Changeset in webkit [264794] by Truitt Savell

• 2 edits in trunk/LayoutTests

[ macOS Release wk1 ] imported/w3c/web-platform-tests/css/css-overflow/overflow-recalc-001.html is a flaky failure
​https://bugs.webkit.org/show_bug.cgi?id=214703

Unreviewed test gardening.

Patch by Hector Lopez <Hector Lopez> on 2020-07-23

• platform/mac-wk1/TestExpectations:

3:15 PM Changeset in webkit [264793] by Russell Epstein

• 1 copy in tags/Safari-610.1.23.1.1

Tag Safari-610.1.23.1.1.

2:57 PM Changeset in webkit [264792] by Russell Epstein

• 2 edits in branches/safari-610.1.23.1-branch/Source/WebCore

Cherry-pick r264791. rdar://problem/66014800

[Cocoa] RELEASE_ASSERT when calling registerSupplementalVP9Decoder() when VideoToolbox.framework is not present
​https://bugs.webkit.org/show_bug.cgi?id=214697
<rdar://problem/66010949>

Reviewed by Eric Carlson.

• platform/graphics/cocoa/VP9UtilitiesCocoa.mm: (WebCore::registerSupplementalVP9Decoder): (WebCore::isVP9DecoderAvailable):

git-svn-id: ​https://svn.webkit.org/repository/webkit/trunk@264791 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2:41 PM Changeset in webkit [264791] by jer.noble@apple.com

• 2 edits in trunk/Source/WebCore

[Cocoa] RELEASE_ASSERT when calling registerSupplementalVP9Decoder() when VideoToolbox.framework is not present
​https://bugs.webkit.org/show_bug.cgi?id=214697
<rdar://problem/66010949>

Reviewed by Eric Carlson.

• platform/graphics/cocoa/VP9UtilitiesCocoa.mm:

(WebCore::registerSupplementalVP9Decoder):
(WebCore::isVP9DecoderAvailable):

2:21 PM Changeset in webkit [264790] by commit-queue@webkit.org

• 16 edits in trunk

Allow IndexedDB in third-party frames
​https://bugs.webkit.org/show_bug.cgi?id=214583
<rdar://problem/61278487>

Patch by Sihui Liu <​sihui_liu@appe.com> on 2020-07-23
Reviewed by Geoffrey Garen.

Source/WebCore:

Match IndexedDB's behavior with localStorage. Third-party iframes now have access to IndexedDB with partitioned
in-memory storage. The partitioning key is ClientOrigin, which consists of opening origin and top origin.

API Test: IndexedDB.IndexedDBThirdPartyFrameHasAccess

IndexedDB.IndexedDBThirdPartyDataRemoval
IndexedDB.IndexedDBThirdPartyWorkerHasAccess

• Modules/indexeddb/IDBDatabaseIdentifier.cpp:

(WebCore::IDBDatabaseIdentifier::IDBDatabaseIdentifier):
(WebCore::m_isTransient):
(WebCore::IDBDatabaseIdentifier::isolatedCopy const):
(WebCore::IDBDatabaseIdentifier::loggingString const):

• Modules/indexeddb/IDBDatabaseIdentifier.h:

(WebCore::IDBDatabaseIdentifier::hash const):
(WebCore::IDBDatabaseIdentifier::operator== const):
(WebCore::IDBDatabaseIdentifier::isTransient const):
(WebCore::IDBDatabaseIdentifier::encode const):
(WebCore::IDBDatabaseIdentifier::decode):

• Modules/indexeddb/IDBFactory.cpp:

(WebCore::shouldThrowSecurityException):
(WebCore::IDBFactory::openInternal):
(WebCore::IDBFactory::deleteDatabase):

• Modules/indexeddb/server/IDBServer.cpp:

(WebCore::IDBServer::IDBServer::createBackingStore):

• Modules/webdatabase/DOMWindowWebDatabase.cpp:

(WebCore::DOMWindowWebDatabase::openDatabase):

• page/SecurityOrigin.h:

(WebCore::SecurityOrigin::canAccessDatabase const):

Source/WebKit:

Do no append extra space to displayName if host is empty.

• UIProcess/WebsiteData/WebsiteDataStore.cpp:

(WebKit::WebsiteDataStore::fetchDataAndApply):

Tools:

• TestWebKitAPI/Tests/WebKitCocoa/IndexedDBPersistence.mm:

(postResult):
(catch):

LayoutTests:

Update test expectation for new behavior.

• http/tests/security/cross-origin-indexeddb-expected.txt:
• http/tests/security/cross-origin-indexeddb.html:
• http/tests/security/cross-origin-worker-indexeddb-expected.txt:
• http/tests/security/resources/document-for-cross-origin-worker-indexeddb.html:

2:05 PM Changeset in webkit [264789] by Chris Dumez

• 15 edits in trunk

Add support for SVGAElement's rel / relList attributes
​https://bugs.webkit.org/show_bug.cgi?id=214690

Reviewed by Darin Adler.

LayoutTests/imported/w3c:

Rebaseline WPT tests that are now passing.

• web-platform-tests/dom/lists/DOMTokenList-coverage-for-attributes-expected.txt:
• web-platform-tests/svg/idlharness.window-expected.txt:
• web-platform-tests/svg/linking/scripted/a.rel-getter-01-expected.txt:
• web-platform-tests/svg/linking/scripted/rellist-feature-detection-expected.txt:

Source/WebCore:

Add support for SVGAElement's rel / relList attributes:
​https://www.w3.org/TR/SVG/linking.html#InterfaceSVGAElement

No new tests, rebaselined existing tests.

• svg/SVGAElement.cpp:

(WebCore::SVGAElement::parseAttribute):
(WebCore::SVGAElement::relList const):

• svg/SVGAElement.h:
• svg/SVGAElement.idl:
• svg/svgattrs.in:

1:37 PM Changeset in webkit [264788] by ysuzuki@apple.com

• 5 edits
  2 adds in trunk

Add exception check for WebCore createRejectedPromiseWithTypeError
​https://bugs.webkit.org/show_bug.cgi?id=214680
<rdar://problem/65925490>

Reviewed by Mark Lam.

Source/WebCore:

Test: js/dom/rejected-promise-creation-should-check-exception.html

Add missing exception checks in createRejectedPromiseWithTypeError.

• bindings/js/JSDOMAttribute.h:

(WebCore::IDLAttribute::get):

• bindings/js/JSDOMOperationReturningPromise.h:

(WebCore::IDLOperationReturningPromise::call):
(WebCore::IDLOperationReturningPromise::callReturningOwnPromise):

• bindings/js/JSDOMPromiseDeferred.cpp:

(WebCore::createRejectedPromiseWithTypeError):

LayoutTests:

• js/dom/rejected-promise-creation-should-check-exception-expected.txt: Added.
• js/dom/rejected-promise-creation-should-check-exception.html: Added.

1:34 PM Changeset in webkit [264787] by Alan Coon

• 8 edits in branches/safari-610.1.23.0-branch/Source

Versioning.

WebKit-610.1.23.0.1

1:27 PM Changeset in webkit [264786] by Fujii Hironori

• 3 edits in trunk/Source/WebCore

[CMake][Win] Build StructuredExceptionHandlerSuppressor.cpp and makesafeseh.asm only for AppleWin
​https://bugs.webkit.org/show_bug.cgi?id=214671

Reviewed by Per Arne Vollan.

StructuredExceptionHandlerSuppressor is used only by AppleWin port
in platform/graphics/ca/win/LayerChangesFlusher.cpp.
makesafeseh.asm causes a building problem for 32bit WinCairo
because it is using OBJECT library type for WebCore.

• PlatformAppleWin.cmake: Added build rules for them.
• PlatformWin.cmake: Removed build rules for them.

1:16 PM Changeset in webkit [264785] by david_quesada@apple.com

• 2 edits in trunk/Source/WebKit

Unable to present file upload picker after interactively dismissing a previously opened picker
​https://bugs.webkit.org/show_bug.cgi?id=209613
rdar://problem/60931009

Reviewed by Daniel Bates.

-[<UIDocumentPickerViewController> documentPickerWasCancelled:] doesn't get called if you
dismiss the view controller using the interactive swipe dismissal gesture added in iOS 13.
As a result, the web view will continue to think it's showing a file picker after dismissing
it this way, meaning you can't present any other file pickers from the same web view. Fix
this by implementing UIAdaptivePresentationControllerDelegate in WKFileUploadPanel to cancel
and tear down the file picker operation when the view controller is interactively dismised.

• UIProcess/ios/forms/WKFileUploadPanel.mm:

(-[WKFileUploadPanel showFilePickerMenu]):
(-[WKFileUploadPanel presentationControllerDidDismiss:]):

1:00 PM Changeset in webkit [264784] by Andres Gonzalez

• 3 edits in trunk/Source/WebCore

Update the isolated tree even when there is no client request.
​https://bugs.webkit.org/show_bug.cgi?id=214692

Reviewed by Chris Fleizach.

AXObjectCache::updateIsolatedTree was checking for isIsolatedTreeEnabled,
which in turn checks for whether the clientSupportsIsolatedTree. Often
at the time updateIsolatedTree is called, there is no client request
since client requests and updates occur in separate threads,
independently from each other. Thus the isolated tree wasn't being
updated. This patch removes the check for isIsolatedTreeEnabled and rely
on the presence of an isolated tree for the given PageID to perform the
update.
In addition, added an update of the isolated tree for AXLoadingEvent::AXLoadingFinished.
This fixes several instance of dynamic content page changes that were
not being reflected on the isolated tree.

• accessibility/AXObjectCache.cpp:

(WebCore::AXObjectCache::clientSupportsIsolatedTree): Some logging.
(WebCore::AXObjectCache::isIsolatedTreeEnabled): Some logging.
(WebCore::AXObjectCache::frameLoadingEventNotification): Calls updateIsolatedTree.
(WebCore::AXObjectCache::updateIsolatedTree): Handles AXLoadingEvents.

• accessibility/AXObjectCache.h:

12:46 PM Changeset in webkit [264783] by Aditya Keerthi

• 4 edits in trunk/Source

[macOS] Datalist dropdown scrollbar position does not match visible region
​https://bugs.webkit.org/show_bug.cgi?id=214656

Reviewed by Wenson Hsieh.

Source/WebKit:

Currently, the content insets for the dropdown are set on an NSScrollView's contentView,
rather than the NSScrollView itself. On the latest macOS, this results in accurate state
within the NSScrollView, leading to an inaccurate scrollbar position. To fix, set the
content insets directly on the NSScrollView.

However, on older macOS, setting content insets on the NSScrollView causes the view to start
out slightly scrolled (by a value equal to the top inset). This issue is solved by preserving
the existing behavior (setting insets on the contentView) on older macOS.

Also, set the style of the embedded NSTableView to NSTableViewStyleFullWidth, to ensure the
view respects the insets we specify.

No new tests, as the change is purely visual and affects native UI.

• UIProcess/mac/WebDataListSuggestionsDropdownMac.mm:

(-[WKDataListSuggestionTableView initWithElementRect:]):
(-[WKDataListSuggestionsController initWithInformation:inView:]):

Source/WTF:

• wtf/PlatformHave.h: Define HAVE_NSTABLEVIEWSTYLE.

12:22 PM Changeset in webkit [264782] by commit-queue@webkit.org

• 18 edits
  4 adds in trunk

Added PeriodicWave constructor according to spec
​https://bugs.webkit.org/show_bug.cgi?id=214615

Patch by Clark Wang <​clark_wang@apple.com> on 2020-07-23
Reviewed by Chris Dumez.

LayoutTests/imported/w3c:

Re-baselined existing tests that now pass, or fail further along.

• web-platform-tests/webaudio/idlharness.https.window-expected.txt:
• web-platform-tests/webaudio/the-audio-api/the-oscillatornode-interface/osc-basic-waveform-expected.txt:

Source/WebCore:

Re-baselined existing tests that now pass.

Added PeriodicWave constructor according to spec: ​https://www.w3.org/TR/webaudio/#periodicwave.
Added PeriodicWaveConstraints, PeriodicWaveOptions files. Updated PeriodicWave::createBandLimitedTables
to normalize waveform based on spec: ​https://www.w3.org/TR/webaudio/#waveform-normalization.
Added updated createPeriodicWave method in BaseAudioContext, and moved old createPeriodicWave method
Into WebKitAudioContext.

• CMakeLists.txt:
• DerivedSources-input.xcfilelist:
• DerivedSources-output.xcfilelist:
• DerivedSources.make:
• Modules/webaudio/BaseAudioContext.cpp:

(WebCore::BaseAudioContext::createPeriodicWave):

• Modules/webaudio/BaseAudioContext.h:

(WebCore::BaseAudioContext::createPeriodicWave):

• Modules/webaudio/BaseAudioContext.idl:
• Modules/webaudio/PeriodicWave.cpp:

(WebCore::PeriodicWave::create):
(WebCore::PeriodicWave::createBandLimitedTables):

• Modules/webaudio/PeriodicWave.h:
• Modules/webaudio/PeriodicWave.idl:
• Modules/webaudio/PeriodicWaveConstraints.h: Added.
• Modules/webaudio/PeriodicWaveConstraints.idl: Added.
• Modules/webaudio/PeriodicWaveOptions.h: Added.
• Modules/webaudio/PeriodicWaveOptions.idl: Added.
• Modules/webaudio/WebKitAudioContext.cpp:

(WebCore::WebKitAudioContext::createPeriodicWave):

• Modules/webaudio/WebKitAudioContext.h:
• Sources.txt:
• WebCore.xcodeproj/project.pbxproj:

11:51 AM Changeset in webkit [264781] by ysuzuki@apple.com

• 4 edits
  1 add in trunk

[JSC] BigInt can be false in boolean context in DFG AI
​https://bugs.webkit.org/show_bug.cgi?id=214678
<rdar://problem/65894751>

Reviewed by Mark Lam.

JSTests:

• stress/bigint-can-be-false-in-boolean-context.js: Added.

(foo.bar):
(foo):

Source/JavaScriptCore:

DFG::AbstractInterpreter::booleanResult returns wrong result if finite structure includes HeapBigInt structure
since HeapBigInt 0n can be evaluated false in boolean context while this function does not care it. This patch
fixes it and cleans up code by using WTF/TriState.

• dfg/DFGAbstractInterpreter.h:

(): Deleted.

• dfg/DFGAbstractInterpreterInlines.h:

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

11:45 AM Changeset in webkit [264780] by ddkilzer@apple.com

• 5 edits in trunk/Source

[IPC hardening] WebKit::ArgumentCoder<BlobPart>::decode() and encode() should use enum BlobPart::Type
<​https://webkit.org/b/214665>
<rdar://problem/65777948>

Reviewed by Darin Adler.

Source/WebCore:

• platform/network/BlobPart.h:

(WebCore::BlobPart::Type):

• Convert to an enum class of size bool.

(WebCore::BlobPart::BlobPart):
(WebCore::BlobPart::data const):
(WebCore::BlobPart::moveData):
(WebCore::BlobPart::url const):

• Change BlobPart::Type enums to use fully qualified name.
• platform/network/BlobRegistryImpl.cpp:

(WebCore::BlobRegistryImpl::registerBlobURL):

• Change BlobPart::Type enums to use fully qualified name.

Source/WebKit:

• Shared/WebCoreArgumentCoders.cpp:

(IPC::ArgumentCoder<BlobPart>::encode):

• Change BlobPart::Type enums to use fully qualified name.
• Encode using WebCore::BlobPart::Type value.
• Change break statements to early return statements.
• Add ASSERT_NOT_REACHED() to catch bugs.

(IPC::ArgumentCoder<BlobPart>::decode):

• Change BlobPart::Type enums to use fully qualified name.
• Decode using Optional<WebCore::BlobPart::Type> variable.
• Change break statements to early return statements.
• Remove default: label, move return WTF::nullopt to the end of the method, and add ASSERT_NOT_REACHED() to catch bugs.

11:26 AM Changeset in webkit [264779] by Ryan Haddad

• 2 edits in trunk/Tools

[ews] Update bot watcher email address
​https://bugs.webkit.org/show_bug.cgi?id=214693

Reviewed by Aakash Jain.

• BuildSlaveSupport/ews-build/send_email.py:

11:25 AM Changeset in webkit [264778] by Simon Fraser

• 2 edits in trunk/Source/WebCore

updateRendering trace point should exclude SVG
​https://bugs.webkit.org/show_bug.cgi?id=214649

Reviewed by Geoffrey Garen.

Don't issue the trace point for SVGImage Pages.

• page/Page.cpp:

(WebCore::Page::updateRendering):

11:19 AM Changeset in webkit [264777] by Karl Rackler

• 2 edits in trunk/LayoutTests

[ iOS wk2 ] imported/w3c/web-platform-tests/web-share/share-without-user-gesture.https.html is a constant timeout
​https://bugs.webkit.org/show_bug.cgi?id=214694

Unreviewed test gardening.

• platform/ios-simulator-wk2/TestExpectations:

11:00 AM Changeset in webkit [264776] by Jonathan Bedard

• 5 edits in trunk/Tools

[run-webkit-tests] Override suite name for results database reporting
​https://bugs.webkit.org/show_bug.cgi?id=214686
<rdar://problem/65997324>

Reviewed by Aakash Jain.

• Scripts/webkitpy/api_tests/manager.py:

(Manager.run): Allow options to override suite name.

• Scripts/webkitpy/layout_tests/controllers/manager.py:

(Manager.run): Allow options to override suite name.

• Scripts/webkitpy/results/options.py:

(upload_options): Add --suite flag to override suite name.

• Scripts/webkitpy/test/main.py:

(Tester._run_tests): Allow options to override suite name.

10:24 AM Changeset in webkit [264775] by Alan Bujtas

• 7 edits in trunk

Unreviewed, reverting r262124.

See REGRESSION(r262124) Overlapping text on Youtube's video upload page
​https://bugs.webkit.org/show_bug.cgi?id=214655

LayoutTests/imported/w3c:

• web-platform-tests/css/css-flexbox/percentage-heights-003-expected.txt:

Source/WebCore:

• rendering/RenderFlexibleBox.cpp:

(WebCore::RenderFlexibleBox::mainSizeForPercentageResolution):

LayoutTests:

• css3/flexbox/definite-main-size-expected.txt:
• css3/flexbox/definite-main-size.html:

10:22 AM Changeset in webkit [264774] by Russell Epstein

• 7 edits in branches/safari-610.1.23.1-branch/Source

Cherry-pick r264747. rdar://problem/65995978

[Cocoa] Add feature flag for WebM & VP9
​https://bugs.webkit.org/show_bug.cgi?id=214672
<rdar://problem/65959506>

Reviewed by Beth Dakin.

Source/WebKit:

• FeatureFlags/WebKit.plist:
• Shared/WebPreferences.yaml:
• Shared/WebPreferencesDefaultValues.cpp: (WebKit::defaultVP9DecoderEnabled): (WebKit::defaultWebMParserEnabled):
• Shared/WebPreferencesDefaultValues.h:

Source/WTF:

• wtf/PlatformEnableCocoa.h:

git-svn-id: ​https://svn.webkit.org/repository/webkit/trunk@264747 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10:22 AM Changeset in webkit [264773] by Russell Epstein

• 2 edits in branches/safari-610.1.23.1-branch/Source/WebCore

Cherry-pick r264742. rdar://problem/65995985

[Cocoa|WebM] Decode error when seeking
​https://bugs.webkit.org/show_bug.cgi?id=214666

Reviewed by Eric Carlson.

A decoder error is hit when seeking because we think all samples coming out of the SourceBufferParserWebM are sync-samples.
This is because the attachment added with CMSetAttachment() is ignored; the correct way to set sample attachments is by
getting the sample attachments array with CMSampleBufferGetSampleAttachmentsArray(), and then iterating over the attachments
and adding keys to the underlying mutable CFDictionary.

• platform/graphics/cocoa/SourceBufferParserWebM.cpp: (WebCore::SourceBufferParserWebM::OnFrame):

git-svn-id: ​https://svn.webkit.org/repository/webkit/trunk@264742 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10:15 AM Changeset in webkit [264772] by Alan Coon

• 8 edits in branches/safari-610.1.23.2-branch/Source

Versioning.

WebKit-610.1.23.2.1

10:11 AM Changeset in webkit [264771] by ddkilzer@apple.com

• 2 edits in trunk/Source/WebKit

[IPC hardening] FilterOperation decode/encode should use early returns
<​https://webkit.org/b/214667>
<rdar://problem/65946400>

Reviewed by Darin Adler.

• Shared/WebCoreArgumentCoders.cpp:

(IPC::ArgumentCoder<FilterOperation>::encode):

• Change break statements to early return statements.
• Add ASSERT_NOT_REACHED() to catch bugs.

(IPC::decodeFilterOperation):

• Change break statements to early return statements.
• Remove call to decoder.markInvalid() since it is redundant.
• Return false from FilterOperation::APPLE_INVERT_LIGHTNESS label to match FilterOperation::NONE and FilterOperation::REFERENCE.
• Add ASSERT_NOT_REACHED() to catch bugs.
• Change final return statement from true to false since valid decoding returns true earlier.

10:09 AM Changeset in webkit [264770] by Russell Epstein

• 8 edits in branches/safari-610.1.23.1-branch/Source

Versioning.

WebKit-610.1.23.1.1

9:36 AM Changeset in webkit [264769] by beidson@apple.com

• 15 edits
  10 adds in trunk

Add Gamepad tests that exercise the native frameworks
<rdar://problem/65343674> and ​https://bugs.webkit.org/show_bug.cgi?id=214188

Reviewed by Tim Horton.

Source/WebCore:

• platform/gamepad/mac/MultiGamepadProvider.h:

(WebCore::MultiGamepadProvider::setUsesOnlyHIDGamepadProvider):

• platform/gamepad/mac/MultiGamepadProvider.mm:

(WebCore::MultiGamepadProvider::startMonitoringGamepads):
(WebCore::MultiGamepadProvider::stopMonitoringGamepads):

Source/WebKit:

Add some testing-only SPI for TestWebKitAPI.

• UIProcess/API/Cocoa/WKProcessPool.mm:

(-[WKProcessPool _numberOfConnectedGamepadsForTesting]):
(-[WKProcessPool _setUsesOnlyHIDGamepadProviderForTesting:]):

• UIProcess/API/Cocoa/WKProcessPoolPrivate.h:

• UIProcess/Gamepad/UIGamepadProvider.h:

(WebKit::UIGamepadProvider::numberOfConnectedGamepads const):

• UIProcess/WebProcessPool.cpp:

(WebKit::WebProcessPool::numberOfConnectedGamepadsForTesting):
(WebKit::WebProcessPool::setUsesOnlyHIDGamepadProviderForTesting):

• UIProcess/WebProcessPool.h:

Source/WTF:

• wtf/PlatformHave.h:

Tools:

For Internal builds, we can create HID devices that exercise the native gamepad providers.
E.g. Actually exercise the IOHID gamepad provider codepath.

Give the Internal TestWebKitAPI build some entitlements:

• TestWebKitAPI/Configurations/TestWebKitAPI-macOS-internal.entitlements: Added.
• TestWebKitAPI/Configurations/TestWebKitAPI.xcconfig:
• TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj:

Add a very basic test that connects a device, presses a button, and verifies the web page sees it:

• TestWebKitAPI/Tests/mac/HIDGamepads.mm: Added.

(-[GamepadMessageHandler userContentController:didReceiveScriptMessage:]):
(TestWebKitAPI::handleGamepadConnect):

Add the beginnings of a VirtualGamepad device that reads in a HID descriptor (just like real gamepads):

• TestWebKitAPI/mac/VirtualGamepad.h: Added.
• TestWebKitAPI/mac/VirtualGamepad.mm: Added.

(TestWebKitAPI::VirtualGamepad::makeVirtualNimbus):
(TestWebKitAPI::VirtualGamepad::VirtualGamepad):
(TestWebKitAPI::VirtualGamepad::~VirtualGamepad):
(TestWebKitAPI::VirtualGamepad::setButtonValue):
(TestWebKitAPI::VirtualGamepad::setAxisValue):
(TestWebKitAPI::VirtualGamepad::publishReport):

WebKitLibraries:

• WebKitPrivateFrameworkStubs/Mac/101500/HID.framework/HID.tbd: Added.
• WebKitPrivateFrameworkStubs/Mac/101600/HID.framework/HID.tbd: Added.
• WebKitPrivateFrameworkStubs/Mac/110000/HID.framework/HID.tbd: Added.

9:34 AM Changeset in webkit [264768] by Alan Coon

• 1 copy in branches/safari-610.1.23.2-branch

New branch.

9:34 AM Changeset in webkit [264767] by Alan Coon

• 1 copy in branches/safari-610.1.23.1-branch

New branch.

9:34 AM Changeset in webkit [264766] by Alan Coon

• 1 copy in branches/safari-610.1.23.0-branch

New branch.

8:50 AM Changeset in webkit [264765] by Karl Rackler

• 2 edits in trunk/LayoutTests

[ iOS wk2 ] editing/selection/shift-click-includes-existing-selection.html is a flaky failure
​https://bugs.webkit.org/show_bug.cgi?id=214688

Unreviewed test gardening.

• platform/ios-simulator-wk2/TestExpectations:

8:32 AM WikiStart edited by Simon Fraser

(diff)

8:27 AM Changeset in webkit [264764] by calvaris@igalia.com

• 2 edits in trunk/Source/WebCore

Unreviewed, build fix after r264762.

• platform/graphics/gstreamer/GStreamerCommon.cpp: Added include

SharedBuffer.h.

8:14 AM Changeset in webkit [264763] by Karl Rackler

• 2 edits in trunk/LayoutTests

[ iOS ] imported/w3c/web-platform-tests/webaudio/the-audio-api/the-audiobuffer-interface/audiobuffer-copy-channel.html and imported/w3c/web-platform-tests/xhr/send-data-sharedarraybuffer.any.html and imported/w3c/web-platform-tests/xhr/send-data-sharedarraybuffer.any.worker.html do not support WebAssembly and marking Skip.
​https://bugs.webkit.org/show_bug.cgi?id=214681

Unreviewed test gardening.

• platform/ios-simulator/TestExpectations:

7:55 AM Changeset in webkit [264762] by calvaris@igalia.com

• 18 edits in trunk

[GStreamer] gst_buffer_unmap: assertion 'GST_IS_BUFFER (buffer)' failed
​https://bugs.webkit.org/show_bug.cgi?id=213364

Reviewed by Philippe Normand.

Source/WebCore:

In some cases as in passing init datas around, the InitData that
contains a SharedBuffer that contains a non-reffed mapped buffer
can outlive the buffer itself, creating this problem.

In this patch we make a new class called GstMappedOwnedBuffer that
is to map read-only buffers and that is used for the InitData
SharedBuffer. GstMappedBuffer does not support SharedBuffer
anymore and is not ref counted anymore as it is intended for short
term and quick usage.

Test: imported/w3c/web-platform-tests/encrypted-media/clearkey-mp4-setmediakeys-to-multiple-video-elements.https.html

• platform/SharedBuffer.cpp:

(WebCore::SharedBuffer::create):
(WebCore::SharedBuffer::SharedBuffer):
(WebCore::SharedBuffer::DataSegment::data const):
(WebCore::SharedBuffer::DataSegment::size const):

• platform/SharedBuffer.h:
• platform/audio/gstreamer/WebKitWebAudioSourceGStreamer.cpp:

(webKitWebAudioSrcAllocateBuffersAndRenderAudio):

• platform/graphics/gstreamer/GStreamerCommon.cpp:

(WebCore::GstMappedBuffer::createVector const):
(WebCore::GstMappedOwnedBuffer::createSharedBuffer):

• platform/graphics/gstreamer/GStreamerCommon.h:

(WebCore::GstMappedBuffer::GstMappedBuffer):
(WebCore::GstMappedBuffer::~GstMappedBuffer):
(WebCore::GstMappedBuffer::isValid const):
(WebCore::GstMappedBuffer::data):
(WebCore::GstMappedBuffer::data const):
(WebCore::GstMappedBuffer::size const):
(WebCore::GstMappedBuffer::operator bool const):
(WebCore::GstMappedBuffer::operator! const):
(WebCore::GstMappedOwnedBuffer::create):
(WebCore::GstMappedOwnedBuffer::GstMappedOwnedBuffer):
(WebCore::operator==):

• platform/graphics/gstreamer/InbandTextTrackPrivateGStreamer.cpp:

(WebCore::InbandTextTrackPrivateGStreamer::notifyTrackOfSample):

• platform/graphics/gstreamer/eme/CDMProxyThunder.cpp:

(WebCore::CDMProxyThunder::getDecryptionSession const):

• platform/graphics/gstreamer/eme/GStreamerEMEUtilities.h:

(WebCore::InitData::InitData):

• platform/graphics/gstreamer/eme/WebKitClearKeyDecryptorGStreamer.cpp:

(decrypt):

• platform/mediastream/gstreamer/MockRealtimeAudioSourceGStreamer.cpp:

(WebCore::MockRealtimeAudioSourceGStreamer::render):

• platform/mediastream/gstreamer/RealtimeOutgoingAudioSourceLibWebRTC.cpp:

(WebCore::RealtimeOutgoingAudioSourceLibWebRTC::pullAudioData):

• platform/mediastream/gstreamer/RealtimeOutgoingVideoSourceLibWebRTC.cpp:

(WebCore::RealtimeOutgoingVideoSourceLibWebRTC::createBlackFrame):

• platform/mediastream/libwebrtc/GStreamerVideoEncoderFactory.cpp:

(WebCore::GStreamerVideoEncoder::Fragmentize):

Tools:

Reworked tests for GstMappedBuffer and added tests for
GstMappedOwnedBuffer.

• TestWebKitAPI/Tests/WebCore/gstreamer/GstMappedBuffer.cpp:

(TestWebKitAPI::TEST_F):

LayoutTests:

• platform/glib/TestExpectations: flagged

imported/w3c/web-platform-tests/encrypted-media/clearkey-mp4-setmediakeys-to-multiple-video-elements.https.html
as Pass.

7:33 AM Changeset in webkit [264761] by Karl Rackler

• 2 edits in trunk/LayoutTests

[ iOS wk2 debug ] imported/w3c/web-platform-tests/html/semantics/scripting-1/the-script-element/moving-between-documents/before-prepare-createHTMLDocument-fetch-error-external-classic.html is a flaky failure
​https://bugs.webkit.org/show_bug.cgi?id=214685

Unreviewed test gardening.

• platform/ios-simulator-wk2/TestExpectations:

7:16 AM Changeset in webkit [264760] by Karl Rackler

• 2 edits in trunk/LayoutTests

[ iOS wk2 ] imported/w3c/web-platform-tests/webaudio/the-audio-api/the-audiocontext-interface/suspend-after-construct.html is a flaky failure
​https://bugs.webkit.org/show_bug.cgi?id=214422

Unreviewed test gardening.

• platform/ios-simulator-wk2/TestExpectations:

6:08 AM Changeset in webkit [264759] by Karl Rackler

• 3 edits in trunk/LayoutTests

[ macOS ] media/media-source/media-source-seek-redundant-append.html is a flaky timeout
​https://bugs.webkit.org/show_bug.cgi?id=207146

Unreviewed test gardening.

• platform/mac-wk1/TestExpectations:
• platform/mac/TestExpectations:

5:51 AM Changeset in webkit [264758] by Caio Lima

• 3 edits in trunk/Source/JavaScriptCore

[32-bits] Fixing the return of doVMEntry into LowLevelInterpreter32_64.asm
​https://bugs.webkit.org/show_bug.cgi?id=214641

Reviewed by Mark Lam.

Adjusting the return of doVMEntry for 32-bits LLInt to proper set
EncodedJSValue return in little-endian architectures. It is expected
that tag is stored in r1 and payload is stored in r0.

• llint/LowLevelInterpreter.asm:
• llint/LowLevelInterpreter32_64.asm:

5:37 AM Changeset in webkit [264757] by Karl Rackler

• 2 edits in trunk/LayoutTests

[ Mac wk2 ] tiled-drawing/scrolling/fixed/four-bars-zoomed.html is flaky failing..
​https://bugs.webkit.org/show_bug.cgi?id=209624

Unreviewed test gardening.

• platform/mac-wk2/TestExpectations:

5:21 AM Changeset in webkit [264756] by Karl Rackler

• 2 edits in trunk/LayoutTests

[ macOS wk2 debug ] imported/w3c/web-platform-tests/media-source/mediasource-changetype-play-implicit.html is a flaky failure
​https://bugs.webkit.org/show_bug.cgi?id=214683

Unreviewed test gardening.

• platform/mac-wk2/TestExpectations:

4:47 AM Changeset in webkit [264755] by Karl Rackler

• 2 edits in trunk/LayoutTests

[ macOS WK2 ] imported/w3c/web-platform-tests/css/cssom/stylesheet-same-origin.sub.html is a flaky failure
​https://bugs.webkit.org/show_bug.cgi?id=214682

Unreviewed test gardening.

• platform/mac-wk2/TestExpectations:

4:23 AM Changeset in webkit [264754] by commit-queue@webkit.org

• 3 edits in trunk/JSTests

Fix logic error in r264703
​https://bugs.webkit.org/show_bug.cgi?id=214674

Unreviewed Gardening.

Patch by Angelos Oikonomopoulos <Angelos Oikonomopoulos> on 2020-07-23

• stress/intl-displaynames.js:
• stress/intl-parse-unicode-subtags.js:

3:59 AM Changeset in webkit [264753] by commit-queue@webkit.org

• 5 edits in trunk

Unreviewed, reverting r264745.
​https://bugs.webkit.org/show_bug.cgi?id=214681

Broke 3 web-platform-tests on iOS

Reverted changeset:

"iOS simulator does not support WebAssembly"
​https://bugs.webkit.org/show_bug.cgi?id=191064
​https://trac.webkit.org/changeset/264745

3:44 AM Changeset in webkit [264752] by aakash_jain@apple.com

• 1 edit
  1 add in trunk/Tools

[ews] Add support to send emails in ews
​https://bugs.webkit.org/show_bug.cgi?id=214670

Reviewed by Darin Adler.

• BuildSlaveSupport/ews-build/send_email.py: Added script to send emails.

(send_email):
(send_email_to_bot_watchers):

1:49 AM Changeset in webkit [264751] by mark.lam@apple.com

• 7 edits
  2 adds in trunk

Fix missing exception checks below RTCPeerConnection.generateCertificate().
​https://bugs.webkit.org/show_bug.cgi?id=214668
<rdar://problem/65929975>

Reviewed by Yusuke Suzuki.

Source/WebCore:

Test: webrtc/missing-exception-checks-RTCPeerConnection-generateCertificate.html

• bindings/js/JSCustomElementRegistryCustom.cpp:

(WebCore::JSCustomElementRegistry::whenDefined):

• bindings/js/JSDOMConvertUnion.h:
• bindings/js/JSDOMOperationReturningPromise.h:

(WebCore::IDLOperationReturningPromise::call):
(WebCore::IDLOperationReturningPromise::callReturningOwnPromise):
(WebCore::IDLOperationReturningPromise::callStatic):
(WebCore::IDLOperationReturningPromise::callStaticReturningOwnPromise):

• bindings/js/JSDOMPromiseDeferred.cpp:

(WebCore::DeferredPromise::reject):
(WebCore::rejectPromiseWithExceptionIfAny):

• bindings/js/JSDOMPromiseDeferred.h:

(WebCore::callPromiseFunction):

LayoutTests:

• webrtc/missing-exception-checks-RTCPeerConnection-generateCertificate-expected.txt: Added.
• webrtc/missing-exception-checks-RTCPeerConnection-generateCertificate.html: Added.

1:25 AM Changeset in webkit [264750] by Alexey Shvayka

• 7 edits
  2 adds in trunk

Remove ArrayNode::m_optional
​https://bugs.webkit.org/show_bug.cgi?id=214294

Reviewed by Darin Adler.

JSTests:

• microbenchmarks/destructuring-array-literal.js: Added.
• microbenchmarks/function-dot-apply-array-literal.js: Added.
• stress/apply-second-argument-must-be-array-like.js:
• stress/destructuring-assignment-syntax.js:

Source/JavaScriptCore:

m_optional, which dates back to KJS era, means "is this an array with optional trailing comma,
with elision, or an empty array". It was used by ArrayNode::streamTo() to preserve a trailing
comma when converting array node to source string, as well as in few other places,
before ECMA-262 clarified trailing comma in array literals [1].

Currently, m_optional is used only by ArrayNode::isSimpleArray(), along with m_elision.
Checking m_elision is enough since trailing comma doesn't add extra undefined element.

This patch completely removes m_optional, speeding up destructuring and function.apply()
with empty arrays and arrays with trailing commas by ~55% and a factor of 11 respectively.
Reflect.apply() optimization (​https://webkit.org/b/190668) will also benefit from this change.

Also, this change converts isSpreadExpression() check to an ASSERT (was enabled by r196323).

[1]: ​https://tc39.es/ecma262/#sec-array-initializer

• bytecompiler/NodesCodegen.cpp:

(JSC::ArrayNode::isSimpleArray const):
(JSC::ArrayNode::toArgumentList const):
(JSC::ArrayNode::emitDirectBinding):

• parser/NodeConstructors.h:

(JSC::ArrayNode::ArrayNode):

• parser/Nodes.h:

12:50 AM Changeset in webkit [264749] by Diego Pino Garcia

• 3 edits in trunk/LayoutTests

[GTK] Unreviewed test gardening. Update GTK baseline after r264646.

• platform/gtk/mathml/opentype/opentype-stretchy-expected.png:
• platform/gtk/mathml/opentype/opentype-stretchy-expected.txt:

12:28 AM Changeset in webkit [264748] by Alexey Shvayka

• 2 edits in trunk/Source/JavaScriptCore

Remove emitIsUndefined() from ClassExprNode::emitBytecode()
​https://bugs.webkit.org/show_bug.cgi?id=214645

Reviewed by Darin Adler.

This change removes superclass === undefined check because it's missing from
the spec [1] and isn't a common case. No behavior change: values except null are
passed to OpIsConstructor, resulting in the same error being thrown for undefined.

Test: LayoutTests/js/class-syntax-extends.html

[1]: ​https://tc39.es/ecma262/#sec-runtime-semantics-classdefinitionevaluation (step 5.e)

• bytecompiler/NodesCodegen.cpp:

(JSC::ClassExprNode::emitBytecode):

Jul 22, 2020:

9:52 PM Changeset in webkit [264747] by jer.noble@apple.com

• 7 edits in trunk/Source

[Cocoa] Add feature flag for WebM & VP9
​https://bugs.webkit.org/show_bug.cgi?id=214672
<rdar://problem/65959506>

Reviewed by Beth Dakin.

Source/WebKit:

• FeatureFlags/WebKit.plist:
• Shared/WebPreferences.yaml:
• Shared/WebPreferencesDefaultValues.cpp:

(WebKit::defaultVP9DecoderEnabled):
(WebKit::defaultWebMParserEnabled):

• Shared/WebPreferencesDefaultValues.h:

Source/WTF:

• wtf/PlatformEnableCocoa.h:

9:49 PM Changeset in webkit [264746] by Conrad Shultz

• 71 edits in trunk

Update macOS Version macros
​https://bugs.webkit.org/show_bug.cgi?id=214653

Reviewed by Tim Horton.

PerformanceTests:

• MediaTime/Configurations/Base.xcconfig:
• MediaTime/Configurations/DebugRelease.xcconfig:

Source/bmalloc:

• Configurations/Base.xcconfig:
• Configurations/DebugRelease.xcconfig:

Source/JavaScriptCore:

• Configurations/Base.xcconfig:
• Configurations/DebugRelease.xcconfig:
• Configurations/Version.xcconfig:
• Configurations/WebKitTargetConditionals.xcconfig:

Source/ThirdParty:

• gtest/xcode/Config/DebugProject.xcconfig:
• gtest/xcode/Config/ReleaseProject.xcconfig:

Source/ThirdParty/ANGLE:

• Configurations/Base.xcconfig:
• Configurations/DebugRelease.xcconfig:
• Configurations/Version.xcconfig:
• Configurations/WebKitTargetConditionals.xcconfig:

Source/ThirdParty/libwebrtc:

• Configurations/Base.xcconfig:
• Configurations/DebugRelease.xcconfig:
• Configurations/Version.xcconfig:
• Configurations/WebKitTargetConditionals.xcconfig:

Source/WebCore:

• Configurations/Base.xcconfig:
• Configurations/DebugRelease.xcconfig:
• Configurations/Version.xcconfig:
• Configurations/WebKitTargetConditionals.xcconfig:

Source/WebCore/PAL:

• Configurations/Base.xcconfig:
• Configurations/DebugRelease.xcconfig:
• Configurations/Version.xcconfig:
• Configurations/WebKitTargetConditionals.xcconfig:

Source/WebInspectorUI:

• Configurations/Base.xcconfig:
• Configurations/DebugRelease.xcconfig:
• Configurations/Version.xcconfig:
• Configurations/WebKitTargetConditionals.xcconfig:

Source/WebKit:

• Configurations/Base.xcconfig:
• Configurations/DebugRelease.xcconfig:
• Configurations/Version.xcconfig:
• Configurations/WebKitTargetConditionals.xcconfig:

Source/WebKitLegacy/mac:

• Configurations/Base.xcconfig:
• Configurations/DebugRelease.xcconfig:
• Configurations/Version.xcconfig:
• Configurations/WebKitTargetConditionals.xcconfig:

Source/WTF:

• Configurations/Base.xcconfig:
• Configurations/DebugRelease.xcconfig:
• wtf/PlatformHave.h:

Tools:

• ContentExtensionTester/Configurations/Base.xcconfig:
• ContentExtensionTester/Configurations/DebugRelease.xcconfig:
• DumpRenderTree/mac/Configurations/Base.xcconfig:
• DumpRenderTree/mac/Configurations/DebugRelease.xcconfig:
• ImageDiff/cg/Configurations/Base.xcconfig:
• ImageDiff/cg/Configurations/DebugRelease.xcconfig:
• MiniBrowser/Configurations/Base.xcconfig:
• MiniBrowser/Configurations/DebugRelease.xcconfig:
• TestWebKitAPI/Configurations/Base.xcconfig:
• TestWebKitAPI/Configurations/DebugRelease.xcconfig:
• TestWebKitAPI/Configurations/WebKitTargetConditionals.xcconfig:
• WebEditingTester/Configurations/Base.xcconfig:
• WebEditingTester/Configurations/DebugRelease.xcconfig:
• WebKitTestRunner/Configurations/Base.xcconfig:
• WebKitTestRunner/Configurations/DebugRelease.xcconfig:
• lldb/lldbWebKitTester/Configurations/Base.xcconfig:
• lldb/lldbWebKitTester/Configurations/DebugRelease.xcconfig:

9:20 PM Changeset in webkit [264745] by keith_miller@apple.com

• 4 edits in trunk

iOS simulator does not support WebAssembly
​https://bugs.webkit.org/show_bug.cgi?id=191064

Reviewed by Yusuke Suzuki.

Now that it seems we can JIT in the Simulator we should enable
WebAssembly there. This just means enabling FTL/B3 so WASM gets
enabled.

• wtf/PlatformEnable.h:
• wtf/PlatformEnableCocoa.h:

9:15 PM Changeset in webkit [264744] by Ryan Haddad

• 2 edits in trunk/LayoutTests

[Win10] js/weakref-finalizationregistry.html is frequently timing out
​https://bugs.webkit.org/show_bug.cgi?id=214673

Unreviewed test gardening.

• platform/win/TestExpectations: Mark test as flaky.

8:33 PM Changeset in webkit [264743] by ggaren@apple.com

• 5 edits in trunk/Source

WTF::Function adoption should be explicit instead of implicit
​https://bugs.webkit.org/show_bug.cgi?id=214654

Reviewed by Darin Adler.

Source/JavaScriptCore:

• heap/Heap.cpp:

(JSC::Heap::LambdaFinalizerOwner::finalize): Use new adopt function.

Source/WTF:

• wtf/Function.h:

(WTF::Function<Out):
(WTF::adoptImpl): Added an adopt function for Functions and made the
adopting constructor private.

• wtf/cf/RunLoopCF.cpp:

(WTF::RunLoop::dispatch): Use new adopt function.

8:01 PM Changeset in webkit [264742] by jer.noble@apple.com

• 2 edits in trunk/Source/WebCore

[Cocoa|WebM] Decode error when seeking
​https://bugs.webkit.org/show_bug.cgi?id=214666

Reviewed by Eric Carlson.

A decoder error is hit when seeking because we think all samples coming out of the SourceBufferParserWebM are sync-samples.
This is because the attachment added with CMSetAttachment() is ignored; the correct way to set sample attachments is by
getting the sample attachments array with CMSampleBufferGetSampleAttachmentsArray(), and then iterating over the attachments
and adding keys to the underlying mutable CFDictionary.

• platform/graphics/cocoa/SourceBufferParserWebM.cpp:

(WebCore::SourceBufferParserWebM::OnFrame):

6:49 PM Changeset in webkit [264741] by Fujii Hironori

• 2 edits in trunk/Source/WebCore

[WinCairo][32bit] JPEG2000ImageDecoder.cpp: error C2664: 'std::unique_ptr<...>::unique_ptr(...)': cannot convert argument 2 from 'void (stdcall *)(opj_codec_t *)' to 'const _Dx &'
​https://bugs.webkit.org/show_bug.cgi?id=214657

Reviewed by Darin Adler.

cdecl and stdcall calling conventions don't match.
<​https://docs.microsoft.com/en-us/cpp/cpp/argument-passing-and-naming-conventions>

• platform/image-decoders/jpeg2000/JPEG2000ImageDecoder.cpp:

(WebCore::JPEG2000ImageDecoder::decode): Use decltype for the second template argument of unique_ptr.

6:44 PM Changeset in webkit [264740] by Fujii Hironori

• 2 edits in trunk/Source/WebKit

[WinCairo][32bit] ConnectionWin.cpp(327): error C2664: 'BOOL RegisterWaitForSingleObject(...)': cannot convert argument 3 from 'void (cdecl *)(...)' to 'WAITORTIMERCALLBACK'
​https://bugs.webkit.org/show_bug.cgi?id=214658

Reviewed by Darin Adler.

cdecl and stdcall calling conventions don't match.
<​https://docs.microsoft.com/en-us/cpp/cpp/argument-passing-and-naming-conventions>

• Platform/IPC/Connection.h: Added WINAPI to Connection::EventListener::callback method.

6:41 PM Changeset in webkit [264739] by Jon Davis

• 2 edits in trunk/Source/WebCore

Add feature status for CSS Containment Module Level 1
​https://bugs.webkit.org/show_bug.cgi?id=214663

Reviewed by Darin Adler.

• features.json:

6:37 PM Changeset in webkit [264738] by Jon Davis

• 2 edits in trunk/Source/WebCore

Add feature status for the focus-visible pseudo-selector
​https://bugs.webkit.org/show_bug.cgi?id=214662

Reviewed by Darin Adler.

• features.json:

5:21 PM Changeset in webkit [264737] by Wenson Hsieh

• 2 edits in trunk/Source/WebCore

Unreviewed, update feature status for async clipboard API

• features.json: Move from "In Development" to "Supported".

5:08 PM Changeset in webkit [264736] by mark.lam@apple.com

• 34 edits in trunk/Source

Disallow VM entry when doing a VMInquiry.
​https://bugs.webkit.org/show_bug.cgi?id=214624
<rdar://problem/65915314>

Reviewed by Saam Barati.

Source/JavaScriptCore:

1. In PropertySlot's constructor, automatically install a DisallowVMEntry scope if the passed in internal method type is VMInquiry. This ensures that we won't be able to enter the VM to call JS code while doing the inquiry. As a result, the PropertySlot constructor will now take an optional VM pointer, which is must be passed in in when the internal method type is VMInquiry.

Note that the handling of attempts to enter the VM depends on
Options::crashOnDisallowedVMEntry().

On Debug build (due to ASSERT_ENABLED), Options::crashOnDisallowedVMEntry()
defaults to true and the VM will crash on disallowed entry.
On Release build, Options::crashOnDisallowedVMEntry() defaults to false and
disallow entry attempts into the VM will be treated like calling an empty
function that returns undefined. This is not new behavior in this patch, but
I just want to have a reminder here of how DisallowVMEntry will be enforcing
no entry into the VM while doing a VMInquiry.

2. After VMInquiry gets, sometimes the client code wants to do other work that do entails entering the VM. In such cases, we need to reset the PropertySlot's disallowVMEntry scope. Fixed up a few places in client code to do this reset.

3. Make the DisableVMEntry scope copyable. At least one place wants to copy PropertySlot, and as a result, will need to copy its embedded DisableVMEntry scope as well if installed.

For DisableVMEntry, we'll handle copying semantics as follows: copying a
DisableVMEntry will ref the VM::disallowVMEntryCount. The count will be
decremented when both instances are destructed. As a result, VM entry will
be disallowed as long as one of the copies are still alive.

4. For the setObjectToStringValue() method of Structure and StructureRareData, we were previously passing a PropertySlot by copy. We don't really need to do this. Ultimately, only StructureRareData::setObjectToStringValue() needs to access a few of the PropertySlot query methods. So, we changed these methods to pass a const PropertySlot& instead to void the needless copying.

• API/JSCallbackObjectFunctions.h:

(JSC::JSCallbackObject<Parent>::put):
(JSC::JSCallbackObject<Parent>::staticFunctionGetter):

• heap/HeapSnapshotBuilder.cpp:

(JSC::HeapSnapshotBuilder::json):

• inspector/JSInjectedScriptHost.cpp:

(Inspector::JSInjectedScriptHost::queryInstances):

• interpreter/Interpreter.cpp:

(JSC::Interpreter::execute):

• jit/JITOperations.cpp:
• llint/LLIntSlowPaths.cpp:

(JSC::LLInt::LLINT_SLOW_PATH_DECL):

• runtime/DisallowVMEntry.h:

(JSC::DisallowVMEntryImpl::DisallowVMEntryImpl):

• runtime/ErrorInstance.cpp:

(JSC::ErrorInstance::sanitizedToString):

• runtime/JSFunction.cpp:

(JSC::JSFunction::getOwnNonIndexPropertyNames):
(JSC::JSFunction::put):
(JSC::JSFunction::defineOwnProperty):

• runtime/JSGenericTypedArrayViewConstructorInlines.h:

(JSC::constructGenericTypedArrayViewWithArguments):

• runtime/JSGlobalObject.cpp:

(JSC::getGetterById):
(JSC::JSGlobalObject::defineOwnProperty):
(JSC::JSGlobalObject::tryInstallArraySpeciesWatchpoint):

• runtime/JSObject.cpp:

(JSC::JSObject::calculatedClassName):

• runtime/JSObjectInlines.h:

(JSC::JSObject::getPrivateFieldSlot):

• runtime/JSScope.cpp:

(JSC::abstractAccess):

• runtime/PropertySlot.h:

(JSC::PropertySlot::PropertySlot):

• runtime/SamplingProfiler.cpp:

(JSC::SamplingProfiler::StackFrame::nameFromCallee):

• runtime/Structure.h:
• runtime/StructureInlines.h:

(JSC::Structure::setObjectToStringValue):

• runtime/StructureRareData.cpp:

(JSC::StructureRareData::setObjectToStringValue):

• runtime/StructureRareData.h:
• tools/JSDollarVM.cpp:

(JSC::functionGetGetterSetter):

Source/WebCore:

1. Change binding to reset the DisallowVMEntry scope in PropertySlot after doing a VMInquiry.
2. Rebase bindings test results to match.

• bindings/js/JSDOMAbstractOperations.h:

(WebCore::isVisibleNamedProperty):
(WebCore::accessVisibleNamedProperty):

• bindings/scripts/CodeGeneratorJS.pm:

(GeneratePut):
(GeneratePutByIndex):
(GenerateDefineOwnProperty):

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

(WebCore::JSTestNamedAndIndexedSetterNoIdentifier::put):
(WebCore::JSTestNamedAndIndexedSetterNoIdentifier::putByIndex):
(WebCore::JSTestNamedAndIndexedSetterNoIdentifier::defineOwnProperty):

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

(WebCore::JSTestNamedAndIndexedSetterThrowingException::put):
(WebCore::JSTestNamedAndIndexedSetterThrowingException::putByIndex):
(WebCore::JSTestNamedAndIndexedSetterThrowingException::defineOwnProperty):

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

(WebCore::JSTestNamedAndIndexedSetterWithIdentifier::put):
(WebCore::JSTestNamedAndIndexedSetterWithIdentifier::putByIndex):
(WebCore::JSTestNamedAndIndexedSetterWithIdentifier::defineOwnProperty):

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

(WebCore::JSTestNamedSetterNoIdentifier::put):
(WebCore::JSTestNamedSetterNoIdentifier::putByIndex):
(WebCore::JSTestNamedSetterNoIdentifier::defineOwnProperty):

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

(WebCore::JSTestNamedSetterThrowingException::put):
(WebCore::JSTestNamedSetterThrowingException::putByIndex):
(WebCore::JSTestNamedSetterThrowingException::defineOwnProperty):

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

(WebCore::JSTestNamedSetterWithIdentifier::put):
(WebCore::JSTestNamedSetterWithIdentifier::putByIndex):
(WebCore::JSTestNamedSetterWithIdentifier::defineOwnProperty):

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

(WebCore::JSTestNamedSetterWithIndexedGetter::put):
(WebCore::JSTestNamedSetterWithIndexedGetter::putByIndex):
(WebCore::JSTestNamedSetterWithIndexedGetter::defineOwnProperty):

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

(WebCore::JSTestNamedSetterWithIndexedGetterAndSetter::put):
(WebCore::JSTestNamedSetterWithIndexedGetterAndSetter::putByIndex):
(WebCore::JSTestNamedSetterWithIndexedGetterAndSetter::defineOwnProperty):

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

(WebCore::JSTestNamedSetterWithUnforgableProperties::put):
(WebCore::JSTestNamedSetterWithUnforgableProperties::putByIndex):
(WebCore::JSTestNamedSetterWithUnforgableProperties::defineOwnProperty):

4:57 PM Changeset in webkit [264735] by Ryan Haddad

• 1 edit
  2 deletes in trunk/LayoutTests

Unreviewed test gardening, remove old test expectation directories for iOS 12 since they are unused.

• platform/ios-12/TestExpectations: Removed.
• platform/ios-12/css2.1/t0805-c5519-brdr-r-01-e-expected.txt: Removed.
• platform/ios-12/css2.1/t0905-c5525-fltblck-00-d-ag-expected.txt: Removed.
• platform/ios-12/css2.1/t1202-counter-09-b-expected.txt: Removed.
• platform/ios-12/css2.1/t1202-counters-09-b-expected.txt: Removed.
• platform/ios-12/editing/deleting/delete-emoji-1-expected.txt: Removed.
• platform/ios-12/editing/deleting/delete-emoji-9-expected.txt: Removed.
• platform/ios-12/editing/deleting/delete-emoji-expected.txt: Removed.
• platform/ios-12/fast/css-generated-content/014-expected.txt: Removed.
• platform/ios-12/fast/css/apple-system-colors-expected.txt: Removed.
• platform/ios-12/fast/css/beforeSelectorOnCodeElement-expected.txt: Removed.
• platform/ios-12/fast/css/line-height-font-order-expected.txt: Removed.
• platform/ios-12/fast/events/touch/ios/pointer-events-with-modifiers-expected.txt: Removed.
• platform/ios-12/fast/events/updateLayoutForHitTest-expected.txt: Removed.
• platform/ios-12/fast/overflow/007-expected.txt: Removed.
• platform/ios-12/fast/table/col-and-colgroup-offsets-expected.txt: Removed.
• platform/ios-12/fast/text/hyphenate-avoid-orphaned-word-expected.txt: Removed.
• platform/ios-12/fast/text/hyphenate-character-expected.txt: Removed.
• platform/ios-12/fast/text/hyphens-expected.txt: Removed.
• platform/ios-12/imported/w3c/web-platform-tests/cors/access-control-expose-headers-parsing.window-expected.txt: Removed.
• platform/ios-12/imported/w3c/web-platform-tests/fetch/api/basic/header-value-combining.any-expected.txt: Removed.
• platform/ios-12/imported/w3c/web-platform-tests/fetch/api/basic/header-value-combining.any.worker-expected.txt: Removed.
• platform/ios-12/imported/w3c/web-platform-tests/fetch/api/basic/request-headers.any-expected.txt: Removed.
• platform/ios-12/imported/w3c/web-platform-tests/fetch/api/basic/request-headers.any.worker-expected.txt: Removed.
• platform/ios-12/imported/w3c/web-platform-tests/fetch/content-type/script.window-expected.txt: Removed.
• platform/ios-12/imported/w3c/web-platform-tests/fetch/nosniff/parsing-nosniff.window-expected.txt: Removed.
• platform/ios-12/imported/w3c/web-platform-tests/xhr/getallresponseheaders-expected.txt: Removed.
• platform/ios-12/imported/w3c/web-platform-tests/xhr/getresponseheader.any-expected.txt: Removed.
• platform/ios-12/imported/w3c/web-platform-tests/xhr/getresponseheader.any.worker-expected.txt: Removed.
• platform/ios-12/platform/ios/ios/fast/text/opticalFont-expected.txt: Removed.
• platform/ios-12/platform/ios/ios/fast/text/opticalFontWithTextStyle-expected.txt: Removed.
• platform/ios-12/platform/ios/ios/fast/text/opticalFontWithWeight-expected.txt: Removed.
• platform/ios-12/quicklook/excel-expected.txt: Removed.
• platform/ios-12/quicklook/excel-legacy-expected.txt: Removed.
• platform/ios-12/quicklook/excel-macro-enabled-expected.txt: Removed.
• platform/ios-12/quicklook/keynote-09-expected.txt: Removed.
• platform/ios-12/quicklook/numbers-09-expected.txt: Removed.
• platform/ios-12/quicklook/pages-09-expected.txt: Removed.
• platform/ios-12/quicklook/powerpoint-expected.txt: Removed.
• platform/ios-12/quicklook/powerpoint-legacy-expected.txt: Removed.
• platform/ios-12/svg/custom/object-sizing-expected.txt: Removed.
• platform/ios-12/tables/mozilla/bugs/bug10269-2-expected.txt: Removed.
• platform/ios-12/tables/mozilla/bugs/bug10296-1-expected.txt: Removed.
• platform/ios-12/tables/mozilla/bugs/bug139524-2-expected.txt: Removed.
• platform/ios-12/tables/mozilla_expected_failures/bugs/bug2479-5-expected.txt: Removed.
• platform/ipad-12/TestExpectations: Removed.
• platform/ipad-12/fast/scrolling/ios/overflow-scroll-inherited-expected.txt: Removed.
• platform/ipad-12/fast/scrolling/ios/overflow-scrolling-ancestor-clip-expected.txt: Removed.
• platform/ipad-12/fast/scrolling/ios/overflow-scrolling-ancestor-clip-size-expected.txt: Removed.
• platform/ipad-12/fast/scrolling/ios/subpixel-overflow-scrolling-with-ancestor-expected.txt: Removed.
• platform/ipad-12/fast/viewport/ios/width-is-device-width-overflowing-body-overflow-hidden-tall-expected.txt: Removed.
• platform/ipad-12/platform/ios/ios/fast/text/opticalFontWithTextStyle-expected.txt: Removed.
• platform/ipad-12/scrollingcoordinator/ios/ui-scrolling-tree-expected.txt: Removed.

4:52 PM Changeset in webkit [264734] by eric.carlson@apple.com

• 7 edits
  1 copy
  7 adds
  2 deletes in trunk

[Cocoa] canPlayType with 'codecs' parameter should never return "maybe"
​https://bugs.webkit.org/show_bug.cgi?id=214621
<rdar://problem/65146785>

Reviewed by Jer Noble.

Source/WebCore:

No new tests, updated existing tests and results.

• platform/graphics/MIMETypeCache.cpp:

(WebCore::MIMETypeCache::canDecodeType):

LayoutTests:

• media/media-can-play-av1-expected.txt:
• media/media-can-play-av1.html:
• media/vp9.html:
• platform/ios/imported/w3c/web-platform-tests/html/semantics/embedded-content/media-elements/mime-types/canPlayType-expected.txt: Added.
• platform/ios/media/media-can-play-av1-expected.txt: Removed.
• platform/ios/media/media-can-play-wav-audio-expected.txt: Copied from LayoutTests/platform/mac/media/media-can-play-wav-audio-expected.txt.
• platform/mac/imported/w3c/web-platform-tests/html/semantics/embedded-content/media-elements/mime-types/canPlayType-expected.txt: Added.
• platform/mac/media/media-can-play-av1-expected.txt: Removed.
• platform/mac/media/media-can-play-wav-audio-expected.txt:

4:38 PM Changeset in webkit [264733] by commit-queue@webkit.org

• 16 edits in trunk

[WebGL2] Implement multiple render target entry points
​https://bugs.webkit.org/show_bug.cgi?id=211156

Patch by Kenneth Russell <​kbr@chromium.org> on 2020-07-22
Reviewed by Dean Jackson.

Source/WebCore:

Implement the drawBuffers and clearBuffer entry points.

Integrate the clearBuffer APIs with preserveDrawingBuffer:false's
auto-clearing. Move some auto-clearing state from
WebGLRenderingContextBase to GraphicsContextGL and
GraphicsContextGLOpenGL.

Rename setPreserveDrawingBuffer to enablePreserveDrawingBuffer to
make it clear that arbitrary changes of preserveDrawingBuffer are
not supported. setPreserveDrawingBuffer was previously added for
capture of preserveDrawingBuffer:false WebGL canvases.

Covered by existing WebGL conformance tests and canvas capture
layout tests.

• Modules/mediastream/CanvasCaptureMediaStreamTrack.cpp:

(WebCore::CanvasCaptureMediaStreamTrack::Source::canvasChanged):

• html/canvas/WebGL2RenderingContext.cpp:

(WebCore::WebGL2RenderingContext::drawBuffers):
(WebCore::WebGL2RenderingContext::clearBufferiv):
(WebCore::WebGL2RenderingContext::clearBufferuiv):
(WebCore::WebGL2RenderingContext::clearBufferfv):
(WebCore::WebGL2RenderingContext::clearBufferfi):
(WebCore::WebGL2RenderingContext::validateClearBuffer):
(WebCore::WebGL2RenderingContext::updateBuffersToAutoClear):

• html/canvas/WebGL2RenderingContext.h:
• html/canvas/WebGLFramebuffer.cpp:

(WebCore::WebGLFramebuffer::drawBuffersIfNecessary):

• html/canvas/WebGLRenderingContextBase.cpp:

(WebCore::WebGLRenderingContextBase::clearIfComposited):
(WebCore::WebGLRenderingContextBase::enablePreserveDrawingBuffer):

• html/canvas/WebGLRenderingContextBase.h:

(WebCore::WebGLRenderingContextBase::setPreserveDrawingBuffer): Deleted.

• platform/graphics/GraphicsContextGL.cpp:

(WebCore::GraphicsContextGL::enablePreserveDrawingBuffer):

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

(WebCore::GraphicsContextGLOpenGL::markLayerComposited):
(WebCore::GraphicsContextGLOpenGL::drawBuffers):
(WebCore::GraphicsContextGLOpenGL::clearBufferiv):
(WebCore::GraphicsContextGLOpenGL::clearBufferuiv):
(WebCore::GraphicsContextGLOpenGL::clearBufferfv):
(WebCore::GraphicsContextGLOpenGL::clearBufferfi):

• platform/graphics/opengl/GraphicsContextGLOpenGL.cpp:

(WebCore::GraphicsContextGLOpenGL::resetBuffersToAutoClear):
(WebCore::GraphicsContextGLOpenGL::setBuffersToAutoClear):
(WebCore::GraphicsContextGLOpenGL::getBuffersToAutoClear const):
(WebCore::GraphicsContextGLOpenGL::enablePreserveDrawingBuffer):

• platform/graphics/opengl/GraphicsContextGLOpenGL.h:
• platform/graphics/opengl/GraphicsContextGLOpenGLCommon.cpp:

(WebCore::GraphicsContextGLOpenGL::markLayerComposited):
(WebCore::GraphicsContextGLOpenGL::drawBuffers):

LayoutTests:

Rebaseline one layout test which is now fully passing, and another
which now properly detects errors.

• fast/canvas/webgl/webgl2/sequences-expected.txt:
• webgl/2.0.0/conformance2/reading/read-pixels-from-fbo-test-expected.txt:

4:36 PM Changeset in webkit [264732] by ddkilzer@apple.com

• 2 edits in trunk/Source/WebKit

Follow-up: [IPC hardening] WebKit::ObjCObjectGraph::decode() and encode() should use enum ObjCType type
<​https://webkit.org/b/214618>
<rdar://problem/65777899>

Unreviewed fix to get rid of extra namespace blocks.

• Shared/mac/ObjCObjectGraph.mm:

(enum class WebKit::ObjCType):

• Move enum class definition back to its original location.

(WTF::EnumTraits<WebKit::ObjCType>):

• Move to the end of the file. Noticed this worked in ColorSpaceData.mm.

3:31 PM Changeset in webkit [264731] by Karl Rackler

• 1 edit in trunk/LayoutTests/ChangeLog

[ macOS debug wk2 ] Set expectation for imported/w3c/web-platform-tests/webrtc/RTCSctpTransport-events.html is a flaky crash.
​https://bugs.webkit.org/show_bug.cgi?id=214661

Unreviewed test gardening.

• platform/mac-wk2/TestExpectations:

3:22 PM Changeset in webkit [264730] by Karl Rackler

• 2 edits in trunk/LayoutTests

[ macOS debug wk2 ] Set expectation for imported/w3c/web-platform-tests/webrtc/RTCSctpTransport-events.html is a flaky crash.
webkit.org/b/214661

Unreviewed test gardening.

• platform/mac-wk2/TestExpectations:

3:10 PM Changeset in webkit [264729] by Wenson Hsieh

• 4 edits in trunk

Article headlines are split across multiple lines after translating tagesschau.de
​https://bugs.webkit.org/show_bug.cgi?id=214652
<rdar://problem/63995571>

Reviewed by Tim Horton.

Source/WebCore:

Treat block-level heading elements (h1 through h6) as item boundaries. On this website, we currently vend
the main heading and sub-heading as separate tokens in the same item, which sometimes causes the machine
translation framework to try and rearrange and split tokens, since it treats the combination of the main heading
and sub-heading text as a single sentence to be translated.

Test: TextManipulation.StartTextManipulationExtractsHeadingElementsAsSeparateItems

• editing/TextManipulationController.cpp:

(WebCore::isEnclosingItemBoundaryElement):

Tools:

Adds a new API test (a reduced version of article headlines from the front page of tagesschau.de).

• TestWebKitAPI/Tests/WebKitCocoa/TextManipulation.mm:

2:55 PM Changeset in webkit [264728] by Alan Coon

• 1 copy in tags/Safari-610.1.23

Tag Safari-610.1.23.

2:28 PM Changeset in webkit [264727] by Chris Dumez

• 2 edits in trunk/Source/WebCore/PAL

Unreviewed attempt to fix internal build with very recent SDKs.

• pal/spi/cocoa/LaunchServicesSPI.h:

2:15 PM Changeset in webkit [264726] by Alan Coon

• 8 edits in branches/safari-610.1.22.0-branch/Source

Versioning.

WebKit-610.1.22.0.2

2:01 PM Changeset in webkit [264725] by Karl Rackler

• 2 edits in trunk/LayoutTests

Set test expectation for tiled-drawing/scrolling/fast-scroll-mainframe-zoom.html as it is flaky crash
​https://bugs.webkit.org/show_bug.cgi?id=214651

Unreviewed test gardening.

• platform/mac-wk2/TestExpectations:

1:17 PM Changeset in webkit [264724] by achristensen@apple.com

• 27 edits in trunk

Safari does not present CertificateInfo for service-worker served documents
​https://bugs.webkit.org/show_bug.cgi?id=206403
<rdar://problem/65410875>

Reviewed by Darin Adler and Youenn Fablet.

Source/WebCore:

When loading a service worker JavaScript resource, we need to tell the network process to include certificate info with the response.
We then pass that response to the ServiceWorkerContextData for storage. When the service worker generates responses to fetches, use that
certificate chain for WKWebView.serverTrust when the page is committed.

Covered by updating a test I wrote in r264687.

• platform/network/ResourceResponseBase.h:

(WebCore::ResourceResponseBase::setCertificateInfo):

• platform/network/cf/CertificateInfo.h:
• platform/network/cf/CertificateInfoCFNet.cpp:

(WebCore::CertificateInfo::isolatedCopy const):

• platform/network/soup/CertificateInfo.h:

(WebCore::CertificateInfo::isolatedCopy const):

• workers/Worker.cpp:

(WebCore::Worker::create):

• workers/WorkerScriptLoader.cpp:

(WebCore::WorkerScriptLoader::loadAsynchronously):
(WebCore::WorkerScriptLoader::didReceiveResponse):

• workers/WorkerScriptLoader.h:

(WebCore::WorkerScriptLoader::certificateInfo const):

• workers/service/ServiceWorkerContainer.cpp:

(WebCore::ServiceWorkerContainer::jobFinishedLoadingScript):

• workers/service/ServiceWorkerContainer.h:
• workers/service/ServiceWorkerContextData.cpp:

(WebCore::ServiceWorkerContextData::isolatedCopy const):

• workers/service/ServiceWorkerContextData.h:

(WebCore::ServiceWorkerContextData::encode const):
(WebCore::ServiceWorkerContextData::decode):

• workers/service/ServiceWorkerFetchResult.h:

(WebCore::ServiceWorkerFetchResult::isolatedCopy const):
(WebCore::serviceWorkerFetchError):
(WebCore::ServiceWorkerFetchResult::encode const):
(WebCore::ServiceWorkerFetchResult::decode):

• workers/service/ServiceWorkerGlobalScope.h:
• workers/service/ServiceWorkerJob.cpp:

(WebCore::ServiceWorkerJob::fetchScriptWithContext):
(WebCore::ServiceWorkerJob::notifyFinished):

• workers/service/ServiceWorkerJobClient.h:
• workers/service/context/ServiceWorkerFetch.cpp:

(WebCore::ServiceWorkerFetch::processResponse):
(WebCore::ServiceWorkerFetch::dispatchFetchEvent):

• workers/service/server/RegistrationDatabase.cpp:

(WebCore::RegistrationDatabase::importRecords):

• workers/service/server/SWServer.cpp:

(WebCore::SWServer::addRegistrationFromStore):
Also fix a possible use-after-move bug, depending on parameter evaluation order which is undefined behavior.
(WebCore::SWServer::updateWorker):
(WebCore::SWServer::installContextData):

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

(WebCore::SWServerJobQueue::scriptFetchFinished):

• workers/service/server/SWServerWorker.cpp:

(WebCore::SWServerWorker::SWServerWorker):
(WebCore::SWServerWorker::contextData const):

• workers/service/server/SWServerWorker.h:

Tools:

• TestWebKitAPI/Tests/WebKitCocoa/ServiceWorkerBasic.mm:

1:16 PM Changeset in webkit [264723] by ddkilzer@apple.com

• 3 edits in trunk/Source/WebCore

Unreviewed build fix after r264710, r264719, r264721: use more HAVE_AVPLAYER_VIDEORANGEOVERRIDE.

• platform/PlatformScreen.h:
• platform/mac/PlatformScreenMac.mm:

(WebCore::collectScreenProperties):

• Guard more code with HAVE(AVPLAYER_VIDEORANGEOVERRIDE).

1:04 PM Changeset in webkit [264722] by Fujii Hironori

• 3 edits
  3 adds in trunk

[Win][Uniscribe] Material icons containing underscore or numbers aren't shown because ScriptItemize splits them apart
​https://bugs.webkit.org/show_bug.cgi?id=201214

Reviewed by Don Olmstead.

Source/WebCore:

ScriptItemize split "3d_rotation" into "3", "d", "_" and
"rotation". It has a option fMergeNeutralItems to control it.

Test: fonts/ligature.html

• platform/graphics/win/ComplexTextControllerUniscribe.cpp:

(WebCore::ComplexTextController::collectComplexTextRunsForCharacters):
Set fMergeNeutralItems of SCRIPT_CONTROL true.

LayoutTests:

• fonts/ligature-expected.html: Added.
• fonts/ligature.html: Added.
• fonts/ligature.woff: Added. Created a font for testing ligature

containing only 0-9, underscore, 'a' and U+E000 glyphs. U+E000 has
a ligature string "0_a".

12:58 PM Changeset in webkit [264721] by jer.noble@apple.com

• 2 edits in trunk/Source/WebCore

Further unreviewed build fix after r264710; need more HAVE_AVPLAYER_VIDEORANGEOVERRIDE guards.

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

12:33 PM Changeset in webkit [264720] by ddkilzer@apple.com

• 2 edits in trunk/Source/WebKit

[IPC hardening] WebKit::ObjCObjectGraph::decode() and encode() should use enum ObjCType type
<​https://webkit.org/b/214618>
<rdar://problem/65777899>

Unreviewed fix to include the better header.

• Shared/mac/ObjCObjectGraph.mm:
• Switch #include from OptionSet.h to EnumTraits.h.

12:27 PM Changeset in webkit [264719] by jer.noble@apple.com

• 8 edits in trunk/Source

Unreviewed build fix after r264710; add a HAVE_AVPLAYER_VIDEORANGEOVERRIDE guard.

Source/WebCore:

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

(WebCore::MediaPlayerPrivateAVFoundationObjC::createAVPlayer):
(WebCore::MediaPlayerPrivateAVFoundationObjC::setPreferredDynamicRangeMode):

Source/WebCore/PAL:

• pal/spi/cocoa/AVFoundationSPI.h:

Source/WebKit:

• WebProcess/cocoa/WebProcessCocoa.mm:

(WebKit::WebProcess::updatePageScreenProperties):

Source/WTF:

• wtf/PlatformHave.h:

12:00 PM Changeset in webkit [264718] by ysuzuki@apple.com

• 2 edits in trunk/Tools

TestWTF.WTF_Packed.PackedAlignedPtr should be adjusted because ARM macOS is using 48 bits for effective address width
​https://bugs.webkit.org/show_bug.cgi?id=214633

Reviewed by Mark Lam.

While PlatformOS.h's condition is updated, the test is not changed. This patch just aligns the condition used in this test
to the one in PlatformOS.h.

• TestWebKitAPI/Tests/WTF/Packed.cpp:

(TestWebKitAPI::TEST):

11:59 AM Changeset in webkit [264717] by BJ Burg

• 2 edits in trunk/Source/WebInspectorUI

Web Inspector: Sources tab empty, sidebar nonfunctional
​https://bugs.webkit.org/show_bug.cgi?id=214611
<rdar://problem/65898051>

Reviewed by Devin Rousso.

This is caused when a subresource fails its initial load. This generates a
networking IssueMessage without a source code location, which is not adequately handled.

• UserInterface/Views/IssueTreeElement.js:

(WI.IssueTreeElement.prototype._updateTitles):
(WI.IssueTreeElement):
If there is no sourceCodeLocation for the issue, then don't try to compute line numbers.

11:47 AM Changeset in webkit [264716] by Simon Fraser

• 2 edits in trunk/LayoutTests

compositing/repaint/iframes/compositing-iframe-scroll-repaint.html fails on Mojave
​https://bugs.webkit.org/show_bug.cgi?id=214647

Unreviewed test gardening. This test fails on Mojave.

• platform/mac-wk1/TestExpectations:

11:35 AM Changeset in webkit [264715] by Jonathan Bedard

• 5 edits
  2 adds in trunk/Tools

[webkitcorepy] Add string_utils (Part 1)
​https://bugs.webkit.org/show_bug.cgi?id=214405

Reviewed by Dewei Zhu.

Centralize handling of unicode encoding/decoding along with various tools
for printing strings.

• Scripts/libraries/webkitcorepy/webkitcorepy/init.py:
• Scripts/libraries/webkitcorepy/webkitcorepy/string_utils.py: Added.

(encode): Encode a string as bytes.
(decode): Decode a bytes as a string.
(ordinal): Convert a number to 1st, 2nd, 3rd, 4th, ect.
(pluralize): Convert a string to it's plurlarized version, if provided number indicates
it should be.
(join): Join a list of elements in a human-readable form.
(out_of): Create a fraction which takes up a constant amount of characters.
(elapsed): Describe the amount of time elapsed as a human-readable string.

• Scripts/libraries/webkitcorepy/webkitcorepy/tests/string_utils_unittest.py: Added.

(StringUtils): Test string_utils.

• Scripts/webkitpy/common/unicode_compatibility.py: Replaced with webkitcorepy.string_utils.

11:28 AM Changeset in webkit [264714] by mmaxfield@apple.com

• 4 edits
  3 adds
  1 delete in trunk

REGRESSION(r205826): narrowNoBreakSpace (U+202F) has zero width, regardless of font
​https://bugs.webkit.org/show_bug.cgi?id=214626
<rdar://problem/65193790>

Reviewed by Timothy Hatcher.

Source/WebCore:

There's no reason to overwrite narrowNoBreakSpace with zeroWidthSpace.

This is becoming much more prominent now because a lot of the French unit formatting data in CLDR - and hence in e.g. NSMeasurementFormatter - uses 202F between numeric value and unit, e.g. "37 km"

It works in native content on macOS, and in all other browsers on macOS, but not in WebKit.

Test: fast/text/narrowNoBreakSpace.html

• platform/graphics/Font.cpp:

(WebCore::overrideControlCharacters):

LayoutTests:

• fast/text/narrowNoBreakSpace-expected-mismatch.html: Added.
• fast/text/narrowNoBreakSpace.html: Added.
• fast/text/narrow-non-breaking-space-expected-mismatch.html: Renamed from fast/text/narrow-non-breaking-space-expected.html.
• fast/text/narrow-non-breaking-space.html: Made the test more robust.

11:25 AM Changeset in webkit [264713] by commit-queue@webkit.org

• 4 edits in trunk/Source/WebCore

Fix validation of deleted WebGL objects
​https://bugs.webkit.org/show_bug.cgi?id=214622

Patch by James Darpinian <James Darpinian> on 2020-07-22
Reviewed by Dean Jackson.

Programs and shaders need to be validated differently than other WebGL objects.
Using a deleted WebGL object should cause an INVALID_OPERATION error, except
for programs and shaders which should be INVALID_VALUE.

• html/canvas/WebGL2RenderingContext.cpp:

(WebCore::WebGL2RenderingContext::getFragDataLocation):
(WebCore::WebGL2RenderingContext::transformFeedbackVaryings):
(WebCore::WebGL2RenderingContext::getTransformFeedbackVarying):
(WebCore::WebGL2RenderingContext::getUniformIndices):
(WebCore::WebGL2RenderingContext::getActiveUniforms):
(WebCore::WebGL2RenderingContext::getUniformBlockIndex):
(WebCore::WebGL2RenderingContext::getActiveUniformBlockParameter):
(WebCore::WebGL2RenderingContext::getActiveUniformBlockName):
(WebCore::WebGL2RenderingContext::uniformBlockBinding):

• html/canvas/WebGLRenderingContextBase.cpp:

(WebCore::WebGLRenderingContextBase::attachShader):
(WebCore::WebGLRenderingContextBase::bindAttribLocation):
(WebCore::WebGLRenderingContextBase::compileShader):
(WebCore::WebGLRenderingContextBase::detachShader):
(WebCore::WebGLRenderingContextBase::validateWebGLObject):
(WebCore::WebGLRenderingContextBase::validateWebGLProgramOrShader):
(WebCore::WebGLRenderingContextBase::getActiveAttrib):
(WebCore::WebGLRenderingContextBase::getActiveUniform):
(WebCore::WebGLRenderingContextBase::getAttachedShaders):
(WebCore::WebGLRenderingContextBase::getAttribLocation):
(WebCore::WebGLRenderingContextBase::getProgramParameter):
(WebCore::WebGLRenderingContextBase::getProgramInfoLog):
(WebCore::WebGLRenderingContextBase::getShaderParameter):
(WebCore::WebGLRenderingContextBase::getShaderInfoLog):
(WebCore::WebGLRenderingContextBase::getShaderSource):
(WebCore::WebGLRenderingContextBase::getUniform):
(WebCore::WebGLRenderingContextBase::getUniformLocation):
(WebCore::WebGLRenderingContextBase::linkProgramWithoutInvalidatingAttribLocations):
(WebCore::WebGLRenderingContextBase::shaderSource):
(WebCore::WebGLRenderingContextBase::validateProgram):

• html/canvas/WebGLRenderingContextBase.h:

11:23 AM Changeset in webkit [264712] by Jon Davis

• 2 edits in trunk/Source/WebCore

Update feature status to include the inert attribute
​https://bugs.webkit.org/show_bug.cgi?id=214644

Reviewed by Darin Adler.

• features.json:

11:16 AM Changeset in webkit [264711] by aakash_jain@apple.com

• 3 edits in trunk/Tools

[ews-app] Pass api key in more secure manner
​https://bugs.webkit.org/show_bug.cgi?id=214639

Reviewed by Jonathan Bedard.

• BuildSlaveSupport/ews-app/ews/common/bugzilla.py:

(Bugzilla.get_bugzilla_api_key): Method to read api key from file instead of environment variable.
(Bugzilla.fetch_data_from_bugzilla_with_authentication): Method to fetch data from bugzilla using api key.
(Bugzilla._fetch_attachment_json): Do not append api key in the url, use new method instead.
(Bugzilla._get_bug_json): Ditto.
(Bugzilla._get_commit_queue_patches_from_bug): Ditto.

• BuildSlaveSupport/ews-app/ews/common/util.py:

(fetch_data_from_url): Log the url to help in debugging any possible issues.

11:12 AM Changeset in webkit [264710] by jer.noble@apple.com

• 23 edits in trunk

[macOS] Adopt AVPlayer.videoRangeOverride
​https://bugs.webkit.org/show_bug.cgi?id=213902
<rdar://problem/63953509>

Reviewed by Youenn Fablet.

Source/WebCore:

AVPlayer.videoRangeOverride and also +[AVPlayer preferredVideoRangeForDisplays:] are replacements for the
process-global MTShouldPlayHDR() and MTOverrideShouldPlayHDRVideo() and should be used preferentially if
present.

• html/HTMLMediaElement.cpp:

(WebCore::HTMLMediaElement::createMediaPlayer):
(WebCore::HTMLMediaElement::setPreferredDynamicRangeMode):

• html/HTMLMediaElement.h:
• page/Page.cpp:

(WebCore::Page::screenPropertiesDidChange):
(WebCore::Page::windowScreenDidChange):

• page/Page.h:
• platform/PlatformScreen.h:

(WebCore::preferredDynamicRangeMode):

• platform/ScreenProperties.h:

(WebCore::ScreenData::encode const):
(WebCore::ScreenData::decode):

• platform/graphics/MediaPlayer.cpp:

(WebCore::MediaPlayer::setPreferredDynamicRangeMode):

• platform/graphics/MediaPlayer.h:
• platform/graphics/MediaPlayerPrivate.h:

(WebCore::MediaPlayerPrivateInterface::setPreferredDynamicRangeMode):

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

(WebCore::CDMSessionAVStreamSession::update):

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

(WebCore::convertDynamicRangeModeEnumToAVVideoRange):
(WebCore::MediaPlayerPrivateAVFoundationObjC::createAVPlayer):
(WebCore::MediaPlayerPrivateAVFoundationObjC::setPreferredDynamicRangeMode):

• platform/mac/PlatformScreenMac.mm:

(WebCore::convertAVVideoRangeToEnum):
(WebCore::collectScreenProperties):
(WebCore::preferredDynamicRangeMode):

Source/WebCore/PAL:

• pal/cocoa/AVFoundationSoftLink.h:
• pal/cocoa/AVFoundationSoftLink.mm:
• pal/spi/cocoa/AVFoundationSPI.h:

Source/WebKit:

• WebProcess/WebPage/WebPage.cpp:

(WebKit::WebPage::screenPropertiesDidChange):

• WebProcess/cocoa/WebProcessCocoa.mm:

(WebKit::WebProcess::updatePageScreenProperties):

Tools:

• TestWebKitAPI/Tests/WebCore/cocoa/AVFoundationSoftLinkTest.mm:

(TestWebKitAPI::TEST):

11:00 AM Changeset in webkit [264709] by Alan Coon

• 3 edits in branches/safari-610.1.23-branch/Source/ThirdParty/libwebrtc

Cherry-pick r264693. rdar://problem/65944676

Unreviewed build fix after r264685; copy headers to the correct path during a DEPLOYMENT_LOCATION build.

• Configurations/libwebm.xcconfig:
• libwebrtc.xcodeproj/project.pbxproj:

git-svn-id: ​https://svn.webkit.org/repository/webkit/trunk@264693 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11:00 AM Changeset in webkit [264708] by Alan Coon

• 34 edits
  5 copies
  9 adds in branches/safari-610.1.23-branch

Cherry-pick r264685. rdar://problem/65944676

[Cocoa] Add experimental MSE WebM parser
​https://bugs.webkit.org/show_bug.cgi?id=214529
Source/ThirdParty/libwebrtc:

<rdar://problem/65782467>

Reviewed by Eric Carlson.

Add the vp9_header_parser from libwebrtc to the project, and add its
symbols to the list of exported symbols. Because this header wants to
be installed to a "common/vp9_header_parser.h" path, split up the default
CopyHeadersInstallPhase into two separate CopyFiles build phases that
put headers into different locations.

• Configurations/libwebm.xcconfig:
• Configurations/libwebrtc.iOS.exp:
• Configurations/libwebrtc.iOSsim.exp:
• Configurations/libwebrtc.mac.exp:
• Source/third_party/libwebm/common/vp9_header_parser.h: (vp9_parser::Vp9HeaderParser::color_range const): (vp9_parser::Vp9HeaderParser::subsampling_x const): (vp9_parser::Vp9HeaderParser::subsampling_y const):
• libwebrtc.xcodeproj/project.pbxproj:

Source/WebCore:

Reviewed by Eric Carlson.

Test: media/media-source/media-source-webm.html

Add an experimental MSE WebM parser to SourceBufferPrivateAVFObjC. This will require extensive refactoring
of SourceBufferPrivateAVFObjC, to extract out the parser into a new, virtualized class. This new parser can
take advantage of new C++>11 features such as lambdas to clean up some problematic existing code. The new
virtualized SourceBufferParser class uses passed-in Function objects, rather than pure-virtual Client classes
to notify it's parent object when the parser encounters initialization- or media-data.

Because SourceBufferPrivateAVFObjC will now have to deal with AudioTrackPrivate and VideoTrackPrivate objects
that are not necessarily Audio- and VideoTrackPrivateMediaSourceAVFObjC subclasses, move those classes notification
features up into Audio- and VideoTrackPrivate, again using a Function object rather than a client class.

Add a new SourceBufferParserWebM and associated AudioTrackPrivateWebM and VideoTrackPrivateWebM classes, which
use libwebm via libwebrtc to parse appended buffers. SourceBufferParserWebM feeds passed-in data into a
libwebm-provided parser of its own. It then builds up tracks and samples based on callbacks from that same parser,
constructing CMSampleBuffers and VideoTrackPrivate objects out of the underlying libwebm primitives.

Because WebM uses a UUID as its track identifier, the representation of a "trackID" inside WebCore must be
increased from an int to a uint64_t, or trackIDs will be truncated or zeroed when passed between classes and
when converted to and from AtomStrings.

This early implementation is not yet able to pass the WebM WPT media-source/ tests, as we only have a VP9 decoder
available, and those tests use both VP8 and Vorbis.

• SourcesCocoa.txt:
• WebCore.xcodeproj/project.pbxproj:
• page/RuntimeEnabledFeatures.h: (WebCore::RuntimeEnabledFeatures::setWebMParserEnabled): (WebCore::RuntimeEnabledFeatures::webMParserEnabled const):
• platform/graphics/AudioTrackPrivate.h: (WebCore::AudioTrackPrivate::setEnabled): (WebCore::AudioTrackPrivate::setEnabledChangedCallback):
• platform/graphics/VP9Utilities.cpp: (WebCore::isValidVPColorPrimaries): (WebCore::isValidVPTransferCharacteristics): (WebCore::isValidVPMatrixCoefficients): (WebCore::parseVPCodecParameters):
• platform/graphics/VP9Utilities.h:
• platform/graphics/VideoTrackPrivate.h: (WebCore::VideoTrackPrivate::setSelected): (WebCore::VideoTrackPrivate::setSelectedChangedCallback):
• platform/graphics/avfoundation/objc/AudioTrackPrivateMediaSourceAVFObjC.cpp: (WebCore::AudioTrackPrivateMediaSourceAVFObjC::AudioTrackPrivateMediaSourceAVFObjC): (WebCore::AudioTrackPrivateMediaSourceAVFObjC::setEnabled):
• platform/graphics/avfoundation/objc/AudioTrackPrivateMediaSourceAVFObjC.h:
• platform/graphics/avfoundation/objc/CDMSessionAVContentKeySession.mm: (WebCore::CDMSessionAVContentKeySession::update):
• platform/graphics/avfoundation/objc/CDMSessionAVStreamSession.mm: (WebCore::CDMSessionAVStreamSession::update):
• platform/graphics/avfoundation/objc/MediaPlayerPrivateMediaSourceAVFObjC.mm: (WebCore::MediaPlayerPrivateMediaSourceAVFObjC::supportsType):
• platform/graphics/avfoundation/objc/MediaSampleAVFObjC.h: (WebCore::MediaSampleAVFObjC::create): (WebCore::MediaSampleAVFObjC::MediaSampleAVFObjC):
• platform/graphics/avfoundation/objc/MediaSourcePrivateAVFObjC.mm: (WebCore::MediaSourcePrivateAVFObjC::addSourceBuffer):
• platform/graphics/avfoundation/objc/SourceBufferParserAVFObjC.h: Added. (isType):
• platform/graphics/avfoundation/objc/SourceBufferParserAVFObjC.mm: Added. (-[WebAVStreamDataParserListener initWithParser:parent:]): (-[WebAVStreamDataParserListener dealloc]): (-[WebAVStreamDataParserListener invalidate]): (-[WebAVStreamDataParserListener streamDataParser:didParseStreamDataAsAsset:]): (-[WebAVStreamDataParserListener streamDataParser:didParseStreamDataAsAsset:withDiscontinuity:]): (-[WebAVStreamDataParserListener streamDataParser:didFailToParseStreamDataWithError:]): (-[WebAVStreamDataParserListener streamDataParser:didProvideMediaData:forTrackID:mediaType:flags:]): (-[WebAVStreamDataParserListener streamDataParserWillProvideContentKeyRequestInitializationData:forTrackID:]): (-[WebAVStreamDataParserListener streamDataParser:didProvideContentKeyRequestInitializationData:forTrackID:]): (WebCore::SourceBufferParserAVFObjC::isContentTypeSupported): (WebCore::SourceBufferParserAVFObjC::SourceBufferParserAVFObjC): (WebCore::SourceBufferParserAVFObjC::~SourceBufferParserAVFObjC): (WebCore::SourceBufferParserAVFObjC::appendData): (WebCore::SourceBufferParserAVFObjC::flushPendingMediaData): (WebCore::SourceBufferParserAVFObjC::setShouldProvideMediaDataForTrackID): (WebCore::SourceBufferParserAVFObjC::shouldProvideMediadataForTrackID): (WebCore::SourceBufferParserAVFObjC::resetParserState): (WebCore::SourceBufferParserAVFObjC::didParseStreamDataAsAsset): (WebCore::SourceBufferParserAVFObjC::didFailToParseStreamDataWithError): (WebCore::SourceBufferParserAVFObjC::didProvideMediaDataForTrackID): (WebCore::SourceBufferParserAVFObjC::willProvideContentKeyRequestInitializationDataForTrackID): (WebCore::SourceBufferParserAVFObjC::didProvideContentKeyRequestInitializationDataForTrackID):
• platform/graphics/avfoundation/objc/SourceBufferPrivateAVFObjC.h:
• platform/graphics/avfoundation/objc/SourceBufferPrivateAVFObjC.mm: (WebCore::SourceBufferPrivateAVFObjC::create): (WebCore::SourceBufferPrivateAVFObjC::SourceBufferPrivateAVFObjC): (WebCore::SourceBufferPrivateAVFObjC::~SourceBufferPrivateAVFObjC): (WebCore::SourceBufferPrivateAVFObjC::didParseInitializationData): (WebCore::SourceBufferPrivateAVFObjC::didEncounterErrorDuringParsing): (WebCore::SourceBufferPrivateAVFObjC::didProvideMediaDataForTrackID): (WebCore::SourceBufferPrivateAVFObjC::willProvideContentKeyRequestInitializationDataForTrackID): (WebCore::SourceBufferPrivateAVFObjC::didProvideContentKeyRequestInitializationDataForTrackID): (WebCore::SourceBufferPrivateAVFObjC::append): (WebCore::SourceBufferPrivateAVFObjC::appendCompleted): (WebCore::SourceBufferPrivateAVFObjC::abort): (WebCore::SourceBufferPrivateAVFObjC::resetParserState): (WebCore::SourceBufferPrivateAVFObjC::destroyParser): (WebCore::SourceBufferPrivateAVFObjC::clearTracks): (WebCore::SourceBufferPrivateAVFObjC::hasSelectedVideo const): (WebCore::SourceBufferPrivateAVFObjC::trackDidChangeSelected): (WebCore::SourceBufferPrivateAVFObjC::trackDidChangeEnabled): (WebCore::SourceBufferPrivateAVFObjC::parser const): (WebCore::SourceBufferPrivateAVFObjC::attemptToDecrypt): (WebCore::SourceBufferPrivateAVFObjC::flush): (WebCore::SourceBufferPrivateAVFObjC::enqueueSample): (WebCore::SourceBufferPrivateAVFObjC::isReadyForMoreSamples): (WebCore::SourceBufferPrivateAVFObjC::didBecomeReadyForMoreSamples): (WebCore::SourceBufferPrivateAVFObjC::notifyClientWhenReadyForMoreSamples): (WebCore::SourceBufferPrivateAVFObjC::canSetMinimumUpcomingPresentationTime const):
• platform/graphics/avfoundation/objc/VideoTrackPrivateMediaSourceAVFObjC.h:
• platform/graphics/avfoundation/objc/VideoTrackPrivateMediaSourceAVFObjC.mm: (WebCore::VideoTrackPrivateMediaSourceAVFObjC::VideoTrackPrivateMediaSourceAVFObjC):
• platform/graphics/cocoa/AudioTrackPrivateWebM.cpp: Copied from Source/WebCore/platform/graphics/cocoa/VP9UtilitiesCocoa.h. (WebCore::AudioTrackPrivateWebM::create): (WebCore::AudioTrackPrivateWebM::AudioTrackPrivateWebM): (WebCore::AudioTrackPrivateWebM::id const): (WebCore::AudioTrackPrivateWebM::label const): (WebCore::AudioTrackPrivateWebM::language const): (WebCore::AudioTrackPrivateWebM::trackIndex const):
• platform/graphics/cocoa/AudioTrackPrivateWebM.h: Copied from Source/WebCore/platform/graphics/cocoa/VP9UtilitiesCocoa.h.
• platform/graphics/cocoa/SourceBufferParser.cpp: Copied from Source/WebCore/platform/graphics/cocoa/VP9UtilitiesCocoa.h. (WebCore::SourceBufferParser::isContentTypeSupported): (WebCore::SourceBufferParser::create):
• platform/graphics/cocoa/SourceBufferParser.h: Added. (WebCore::SourceBufferParser::setDidParseInitializationDataCallback): (WebCore::SourceBufferParser::setDidEncounterErrorDuringParsingCallback): (WebCore::SourceBufferParser::setDidProvideMediaDataCallback): (WebCore::SourceBufferParser::setWillProvideContentKeyRequestInitializationDataForTrackIDCallback): (WebCore::SourceBufferParser::setDidProvideContentKeyRequestInitializationDataForTrackIDCallback):
• platform/graphics/cocoa/SourceBufferParserWebM.cpp: Added. (WebCore::isWebmParserAvailable): (WebCore::SourceBufferParserWebM::isContentTypeSupported): (WebCore::SourceBufferParserWebM::SourceBufferParserWebM): (WebCore::SourceBufferParserWebM::appendData): (WebCore::SourceBufferParserWebM::flushPendingMediaData): (WebCore::SourceBufferParserWebM::setShouldProvideMediaDataForTrackID): (WebCore::SourceBufferParserWebM::shouldProvideMediadataForTrackID): (WebCore::SourceBufferParserWebM::resetParserState): (WebCore::SourceBufferParserWebM::trackDataForTrackNumber): (WebCore::SourceBufferParserWebM::OnElementBegin): (WebCore::SourceBufferParserWebM::OnEbml): (WebCore::SourceBufferParserWebM::OnSegmentBegin): (WebCore::SourceBufferParserWebM::OnInfo): (WebCore::SourceBufferParserWebM::OnClusterBegin): (WebCore::SourceBufferParserWebM::OnTrackEntry): (WebCore::SourceBufferParserWebM::OnBlockBegin): (WebCore::SourceBufferParserWebM::OnBlockEnd): (WebCore::SourceBufferParserWebM::OnSimpleBlockBegin): (WebCore::SourceBufferParserWebM::OnSimpleBlockEnd): (WebCore::SourceBufferParserWebM::OnBlockGroupBegin): (WebCore::SourceBufferParserWebM::OnBlockGroupEnd): (WebCore::convertToColorPrimaries): (WebCore::convertToTransferCharacteristics): (WebCore::convertToMatrixCoefficients): (WebCore::convertSubsamplingXYToChromaSubsampling): (WebCore::createFormatDescriptionFromVP9HeaderParser): (WebCore::SourceBufferParserWebM::OnFrame):
• platform/graphics/cocoa/SourceBufferParserWebM.h: Added. (WebCore::SourceBufferParserWebM::status const): (WebCore::SourceBufferParserWebM::type const): (isType):
• platform/graphics/cocoa/VP9UtilitiesCocoa.h:
• platform/graphics/cocoa/VP9UtilitiesCocoa.mm: (WebCore::isVP9DecoderAvailable): (WebCore::validateVPParameters):
• platform/graphics/cocoa/VideoTrackPrivateWebM.cpp: (WebCore::VideoTrackPrivateWebM::create): (WebCore::VideoTrackPrivateWebM::VideoTrackPrivateWebM): (WebCore::VideoTrackPrivateWebM::id const): (WebCore::VideoTrackPrivateWebM::label const): (WebCore::VideoTrackPrivateWebM::language const): (WebCore::VideoTrackPrivateWebM::trackIndex const):
• platform/graphics/cocoa/VideoTrackPrivateWebM.h:

Source/WebKit:

<rdar://problem/65782467>

Reviewed by Eric Carlson.

Add an Experimental setting to enable the MSE WebM parser (off by default).

• Shared/WebPreferences.yaml:

LayoutTests:

<rdar://problem/65782467>

Reviewed by Eric Carlson.

• media/media-source/content/test-vp9-manifest.json: Added.
• media/media-source/content/test-vp9.webm: Added.
• media/media-source/media-source-webm-expected.txt: Added.
• media/media-source/media-source-webm.html: Added.

git-svn-id: ​https://svn.webkit.org/repository/webkit/trunk@264685 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11:00 AM Changeset in webkit [264707] by Alan Coon

• 2 edits in branches/safari-610.1.23-branch/Source/WebKit

Cherry-pick r264678. rdar://problem/65944685

[macOS] Layout tests exiting early with crashes
​https://bugs.webkit.org/show_bug.cgi?id=214612
<rdar://problem/65897905>

Reviewed by Brent Fulgham.

This is caused by the WebContent process making a syscall not allowed by the sandbox.

• WebProcess/com.apple.WebProcess.sb.in:

git-svn-id: ​https://svn.webkit.org/repository/webkit/trunk@264678 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11:00 AM Changeset in webkit [264706] by Alan Coon

• 8 edits in branches/safari-610.1.23-branch/Source

Cherry-pick r264659. rdar://problem/65944694

[Cocoa] Adopt VTRegisterSupplementalVideoDecoderIfAvailable
​https://bugs.webkit.org/show_bug.cgi?id=214585
<rdar://problem/65863651>

Reviewed by Youenn Fablet.

Source/WebCore:

• WebCore.xcodeproj/project.pbxproj:
• platform/cocoa/VideoToolboxSoftLink.cpp:
• platform/cocoa/VideoToolboxSoftLink.h:
• platform/graphics/cocoa/VP9UtilitiesCocoa.h:
• platform/graphics/cocoa/VP9UtilitiesCocoa.mm: (WebCore::registerSupplementalVP9Decoder):

Source/WebKit:

• WebProcess/WebProcess.cpp: (WebKit::WebProcess::enableVP9Decoder):

git-svn-id: ​https://svn.webkit.org/repository/webkit/trunk@264659 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10:51 AM Changeset in webkit [264705] by Jon Davis

• 2 edits in trunk/Source/WebCore

Update status for Web App Manifest
​https://bugs.webkit.org/show_bug.cgi?id=204863

Reviewed by Dean Jackson.

• features.json:

10:48 AM Changeset in webkit [264704] by Simon Fraser

• 17 edits
  4 adds in trunk

Scrolling tesla.com on iPad is not smooth
​https://bugs.webkit.org/show_bug.cgi?id=214627
Source/WebCore:

Reviewed by Tim Horton.

This is the iOS equivalent of the fix in r264203; if the user is scrolling (or we're animating for
scroll snap), layout should not trigger scroll position changes via computation of the current snap
location.

To fix this, we need to send data from the UI process to the web process about which scrolling tree
nodes the user is scrolling. That's done via RemoteScrollingUIState, which will, in the long term,
become a more general UI -> WebProcess scrolling state updating mechanism. For now, it just keeps
track of nodes undergoing scroll snap and user scrolls (only the latter is hooked up for now).

Test: fast/scrolling/ios/adjust-scroll-snap-during-gesture.html

• page/scrolling/AsyncScrollingCoordinator.h:
• page/scrolling/ScrollingTree.h:

(WebCore::ScrollingTree::scrollingTreeNodeWillStartPanGesture):
(WebCore::ScrollingTree::scrollingTreeNodeWillStartScroll):
(WebCore::ScrollingTree::scrollingTreeNodeDidEndScroll):

Source/WebKit:

<rdar://problem/65568979>

Reviewed by Tim Horton.

This is the iOS equivalent of the fix in r264203; if the user is scrolling (or we're animating for
scroll snap), layout should not trigger scroll position changes via computation of the current snap
location.

To fix this, we need to send data from the UI process to the web process about which scrolling tree
nodes the user is scrolling. That's done via RemoteScrollingUIState, which will, in the long term,
become a more general UI -> WebProcess scrolling state updating mechanism. For now, it just keeps
track of nodes undergoing scroll snap and user scrolls (only the latter is hooked up for now).

• Shared/RemoteLayerTree/RemoteScrollingUIState.cpp: Added.

(WebKit::RemoteScrollingUIState::encode const):
(WebKit::RemoteScrollingUIState::decode):
(WebKit::RemoteScrollingUIState::addNodeWithActiveScrollSnap):
(WebKit::RemoteScrollingUIState::removeNodeWithActiveScrollSnap):
(WebKit::RemoteScrollingUIState::addNodeWithActiveUserScroll):
(WebKit::RemoteScrollingUIState::removeNodeWithActiveUserScroll):

• Shared/RemoteLayerTree/RemoteScrollingUIState.h: Added.

(WebKit::RemoteScrollingUIState::changes const):
(WebKit::RemoteScrollingUIState::clearChanges):
(WebKit::RemoteScrollingUIState::nodesWithActiveScrollSnap const):
(WebKit::RemoteScrollingUIState::nodesWithActiveUserScrolls const):

• SourcesCocoa.txt:
• UIProcess/RemoteLayerTree/RemoteScrollingCoordinatorProxy.cpp:

(WebKit::RemoteScrollingCoordinatorProxy::RemoteScrollingCoordinatorProxy):
(WebKit::RemoteScrollingCoordinatorProxy::sendUIStateChangedIfNecessary):

• UIProcess/RemoteLayerTree/RemoteScrollingCoordinatorProxy.h:
• UIProcess/RemoteLayerTree/RemoteScrollingTree.cpp:

(WebKit::RemoteScrollingTree::scrollingTreeNodeWillStartPanGesture):
(WebKit::RemoteScrollingTree::scrollingTreeNodeWillStartScroll):
(WebKit::RemoteScrollingTree::scrollingTreeNodeDidEndScroll):

• UIProcess/RemoteLayerTree/RemoteScrollingTree.h:
• UIProcess/RemoteLayerTree/ios/RemoteScrollingCoordinatorProxyIOS.mm:

(WebKit::RemoteScrollingCoordinatorProxy::scrollViewForScrollingNodeID const):
(WebKit::RemoteScrollingCoordinatorProxy::scrollingTreeNodeWillStartPanGesture):
(WebKit::RemoteScrollingCoordinatorProxy::scrollingTreeNodeWillStartScroll):
(WebKit::RemoteScrollingCoordinatorProxy::scrollingTreeNodeDidEndScroll):

• UIProcess/RemoteLayerTree/ios/ScrollingTreeScrollingNodeDelegateIOS.mm:

(WebKit::ScrollingTreeScrollingNodeDelegateIOS::scrollWillStart const):
(WebKit::ScrollingTreeScrollingNodeDelegateIOS::scrollDidEnd const):
(WebKit::ScrollingTreeScrollingNodeDelegateIOS::scrollViewWillStartPanGesture const):

• WebKit.xcodeproj/project.pbxproj:
• WebProcess/WebPage/RemoteLayerTree/RemoteScrollingCoordinator.h:
• WebProcess/WebPage/RemoteLayerTree/RemoteScrollingCoordinator.messages.in:
• WebProcess/WebPage/RemoteLayerTree/RemoteScrollingCoordinator.mm:

(WebKit::RemoteScrollingCoordinator::isUserScrollInProgress const):
(WebKit::RemoteScrollingCoordinator::isScrollSnapInProgress const):
(WebKit::RemoteScrollingCoordinator::scrollingStateInUIProcessChanged):

LayoutTests:

<rdar://problem/65568979>

Reviewed by Tim Horton.

• fast/scrolling/ios/adjust-scroll-snap-during-gesture-expected.txt: Added.
• fast/scrolling/ios/adjust-scroll-snap-during-gesture.html: Added.

10:17 AM Changeset in webkit [264703] by commit-queue@webkit.org

• 3 edits in trunk/JSTests

Skip failing intl tests on ARM
​https://bugs.webkit.org/show_bug.cgi?id=214638

Unreviewed Gardening.

Patch by Angelos Oikonomopoulos <Angelos Oikonomopoulos> on 2020-07-22

• stress/intl-displaynames.js:
• stress/intl-parse-unicode-subtags.js:

9:51 AM Changeset in webkit [264702] by Truitt Savell

• 2 edits in trunk/LayoutTests

[ macOS ] imported/w3c/web-platform-tests/css/css-pseudo/marker-hit-testing.html is a flaky failure
​https://bugs.webkit.org/show_bug.cgi?id=214643

Unreviewed test gardening.

Patch by Hector Lopez <Hector Lopez> on 2020-07-22

• platform/mac/TestExpectations:

9:28 AM Changeset in webkit [264701] by ddkilzer@apple.com

• 3 edits in trunk/Source/WebKit

[IPC hardening] WebKit::ObjCObjectGraph::decode() and encode() should use enum ObjCType type
<​https://webkit.org/b/214618>
<rdar://problem/65777899>

Reviewed by Youenn Fablet.

• Shared/mac/ObjCObjectGraph.h:
• Drive-by fix to use #pragma once. This is included in some plain C++ source files.

• Shared/mac/ObjCObjectGraph.mm:

(WebKit::ObjCType):

• Move enum definition to the top of the file so that the EnumTraits<> can be defined after it, but before it's used.

(WTF::EnumTraits<WebKit::ObjCType>): Add.

• Define so that IPC::Decoder and IPC::Encoder can validate enum values.

(WebKit::ObjCObjectGraph::encode):

• Encode using WebKit::ObjCType value.
• Add ObjCType::Null label so that default: label can be removed. Change break statements to early return statements.
• Move ASSERT_NOT_REACHED() to the end of the method.

(WebKit::ObjCObjectGraph::decode):

• Decode using WebKit::ObjCType value.
• Change break statements to early return statements.
• Remove default: label, and move return false to the end of the method. Add ASSERT_NOT_REACHED().

9:21 AM Changeset in webkit [264700] by commit-queue@webkit.org

• 13 edits in trunk

Unreviewed, reverting r264691.
​https://bugs.webkit.org/show_bug.cgi?id=214642

Regressed fast/mediastream/captureStream/canvas3d.html

Reverted changeset:

"[WebGL2] Implement multiple render target entry points"
​https://bugs.webkit.org/show_bug.cgi?id=211156
​https://trac.webkit.org/changeset/264691

9:09 AM Changeset in webkit [264699] by Megan Gardner

• 2 edits in trunk/Source/WebKit

Tapped DataDetected links present sub-menus from the wrong location.
​https://bugs.webkit.org/show_bug.cgi?id=214628
<rdar://problem/64808642>

When getting the positionInformation associated with presenting DataDetectors UI
we were not indicating that we need the linkIndicator data. This caused the
resulting CGRect to not be filled out, and we were returning a zeroed out CGRect
to the DataDetectors framework for the location that it should present the subMenu from.
Setting this bit ensures that all the necessary data is calculated and returned
when a DataDetector link is tapped.

Reviewed by Wenson Hsieh.

• WebProcess/WebCoreSupport/ios/WebChromeClientIOS.mm:

(WebKit::WebChromeClient::showDataDetectorsUIForElement):

8:59 AM Changeset in webkit [264698] by youenn@apple.com

• 3 edits
  3 adds in trunk

KeepAlive fetch should not be blocked in pagehide event handlers
​https://bugs.webkit.org/show_bug.cgi?id=214630
<rdar://problem/65564772>

Reviewed by Chris Dumez.

Source/WebCore:

Test: http/wpt/fetch/fetch-in-pagehide.html

• loader/cache/CachedResource.cpp:

(WebCore::CachedResource::load):
Allow keep alive fetches to proceed similarly to ping loads.

LayoutTests:

• http/wpt/fetch/fetch-in-pagehide-expected.txt: Added.
• http/wpt/fetch/fetch-in-pagehide.html: Added.
• http/wpt/fetch/resources/fetch-in-pagehide-window.html: Added.

8:23 AM Changeset in webkit [264697] by Jonathan Bedard

• 1 edit
  1 delete in trunk/Tools

[webkitpy] Remove version.py
​https://bugs.webkit.org/show_bug.cgi?id=214580

Reviewed by Darin Adler.

• Scripts/webkitpy/common/version.py: Removed.

8:07 AM Changeset in webkit [264696] by ggaren@apple.com

• 10 edits in trunk/Source

JSRunLoopTimer should use WTF::RunLoop rather than custom CF code
​https://bugs.webkit.org/show_bug.cgi?id=214102

Unreviewed, re-landing r264242 with crash fixed.

Source/JavaScriptCore:

We needed to synchronize timer destruction with timer firing.

• runtime/DeferredWorkTimer.cpp:

(JSC::DeferredWorkTimer::doWork):
(JSC::DeferredWorkTimer::runRunLoop):

• runtime/JSRunLoopTimer.cpp:

(JSC::epochTime):
(JSC::JSRunLoopTimer::Manager::PerVMData::PerVMData):
(JSC::JSRunLoopTimer::Manager::timerDidFireCallback):
(JSC::JSRunLoopTimer::Manager::PerVMData::~PerVMData):
(JSC::JSRunLoopTimer::Manager::timerDidFire):
(JSC::JSRunLoopTimer::Manager::registerVM):
(JSC::JSRunLoopTimer::Manager::scheduleTimer):
(JSC::JSRunLoopTimer::Manager::cancelTimer):
(JSC::JSRunLoopTimer::Manager::PerVMData::setRunLoop): Deleted.
(JSC::JSRunLoopTimer::Manager::didChangeRunLoop): Deleted.

• runtime/JSRunLoopTimer.h:

(JSC::JSRunLoopTimer::Manager::PerVMData::PerVMData): Deleted.

• runtime/VM.cpp:

(JSC::VM::VM):
(JSC::VM::create):
(JSC::VM::tryCreate):
(JSC::VM::setRunLoop): Deleted.

• runtime/VM.h:

(JSC::VM::runLoop const):

Source/WebCore:

Since I was here, I applied Darin's previous review feedback that I had
missed.

• bindings/js/CommonVM.cpp:

(WebCore::commonVMSlow):

Source/WTF:

To support JSC using external locking around a RunLoop timer, we needed
to avoid modifying unsynchronized timer state while firing.

• wtf/cf/RunLoopCF.cpp:

(WTF::RunLoop::TimerBase::start):

6:30 AM BuildingGtk edited by clopez@igalia.com

(diff)

6:29 AM BuildingGtk edited by clopez@igalia.com

(diff)

5:22 AM Changeset in webkit [264695] by Diego Pino Garcia

• 5 edits
  1 add in trunk/LayoutTests

[GTK] Unreviewed test gardening. Update test expectations after r264694.

• platform/glib/TestExpectations:
• platform/glib/imported/w3c/web-platform-tests/encoding/eof-shift_jis-expected.txt: Added.
• platform/gtk/TestExpectations:
• platform/gtk/imported/w3c/web-platform-tests/css/css-sizing/percentage-height-in-flexbox-expected.txt:
• platform/wpe/TestExpectations:

4:33 AM BuildingGtk edited by clopez@igalia.com

(diff)

3:47 AM Changeset in webkit [264694] by Carlos Garcia Campos

• 5 edits in trunk

[GTK][WPE] ITP: use the database store unconditionally
​https://bugs.webkit.org/show_bug.cgi?id=214590

Reviewed by Adrian Perez de Castro.

Source/WebKit:

The memory store is now considered legacy.

• NetworkProcess/Classifier/ResourceLoadStatisticsDatabaseStore.cpp:

(WebKit::ResourceLoadStatisticsDatabaseStore::ResourceLoadStatisticsDatabaseStore): Use
FileSystem::pathByAppendingComponent() to create the database path.
(WebKit::ResourceLoadStatisticsDatabaseStore::openITPDatabase): Ensure the directory exists before trying to
open the database.

• UIProcess/soup/WebProcessPoolSoup.cpp:

(WebKit::WebProcessPool::platformInitializeNetworkProcess): Always enable the ITP databae by default.

Tools:

Update the unit test to monitor the database file instead of the old log file.

• TestWebKitAPI/Tests/WebKitGLib/TestWebsiteData.cpp:

(testWebsiteDataITP):