WebKit-https.git
2 years agoAdd MediaCapabilities as an Experimental Feature
jer.noble@apple.com [Wed, 5 Sep 2018 17:26:54 +0000 (17:26 +0000)]
Add MediaCapabilities as an Experimental Feature
https://bugs.webkit.org/show_bug.cgi?id=189209

Reviewed by Eric Carlson.

Source/WebCore:

Move mediaCapabilitiesEnabled out of RuntimeEnabledFeatures. It already (also) exists in Settings.

* Modules/mediacapabilities/MediaCapabilities.idl:
* Modules/mediacapabilities/MediaCapabilitiesInfo.idl:
* Modules/mediacapabilities/NavigatorMediaCapabilities.idl:
* Modules/mediacapabilities/ScreenLuminance.idl:
* page/RuntimeEnabledFeatures.h:
(WebCore::RuntimeEnabledFeatures::setMediaCapabilitiesEnabled): Deleted.
(WebCore::RuntimeEnabledFeatures::mediaCapabilitiesEnabled const): Deleted.

Source/WebKit:

* Shared/WebPreferences.yaml:

Source/WebKitLegacy/mac:

RuntimeEnabledFeature -> Setting.

* WebView/WebView.mm:
(-[WebView _preferencesChanged:]):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@235675 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoCancelled fullscreen exit gesture leaves bad state for next exit request.
commit-queue@webkit.org [Wed, 5 Sep 2018 17:11:25 +0000 (17:11 +0000)]
Cancelled fullscreen exit gesture leaves bad state for next exit request.
https://bugs.webkit.org/show_bug.cgi?id=189278

Patch by Jeremy Jones <jeremyj@apple.com> on 2018-09-05
Reviewed by Jer Noble.

* UIProcess/ios/fullscreen/WKFullScreenWindowControllerIOS.mm:
(-[WKFullScreenWindowController _dismissFullscreenViewController]):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@235674 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoAdd functionality to encode and decode a uint64_t in KeyedCoding
commit-queue@webkit.org [Wed, 5 Sep 2018 17:03:02 +0000 (17:03 +0000)]
Add functionality to encode and decode a uint64_t in KeyedCoding
https://bugs.webkit.org/show_bug.cgi?id=189216

Patch by Woodrow Wang <woodrow_wang@apple.com> on 2018-09-05
Reviewed by Daniel Bates.

I've added this functionality in order to be able to encode and decode the raw uint64_t
representation of an OptionSet for my patch here <https://bugs.webkit.org/show_bug.cgi?id=187773>.

The changes in the KeyedEncoder/KeyedDecoder for Glib were made because they are derived classes
of KeyedCoding which contains pure virtual functions that need to be implemented.

* platform/KeyedCoding.h:
* platform/cf/KeyedDecoderCF.cpp:
(WebCore::KeyedDecoderCF::decodeUInt64):
* platform/cf/KeyedDecoderCF.h:
* platform/cf/KeyedEncoderCF.cpp:
(WebCore::KeyedEncoderCF::encodeUInt64):
* platform/cf/KeyedEncoderCF.h:
* platform/glib/KeyedDecoderGlib.cpp:
(WebCore::KeyedDecoderGlib::decodeUInt64):
* platform/glib/KeyedDecoderGlib.h:
* platform/glib/KeyedEncoderGlib.cpp:
(WebCore::KeyedEncoderGlib::encodeUInt64):
* platform/glib/KeyedEncoderGlib.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@235673 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoREGRESSION (r209470): EditingHistory folder is missing in EditingHistory Xcode project
ddkilzer@apple.com [Wed, 5 Sep 2018 16:49:38 +0000 (16:49 +0000)]
REGRESSION (r209470): EditingHistory folder is missing in EditingHistory Xcode project

Found using `tidy-Xcode-project-file --missing` (see Bug
188754).  Fixes were made manually.

* EditingHistory/EditingHistory.xcodeproj/project.pbxproj:
(EditingHistory): Change the path of the folder to
"EditingHistory", then remove the now-redundant name.
(EditingHistory/Resources/DumpEditingHistory.js): Remove
redundant file reference since another file reference exists at
the top of the project file list, and that file reference is the
one used in the "Copy Bundle Resources" build phase.
(EditingHistory/Resources/EditingHistoryUtil.js): Ditto.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@235672 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years ago[GTK] ASSERT(!m_nicosia.imageBacking) when starting any YouTube video
zandobersek@gmail.com [Wed, 5 Sep 2018 16:20:01 +0000 (16:20 +0000)]
[GTK] ASSERT(!m_nicosia.imageBacking) when starting any YouTube video
https://bugs.webkit.org/show_bug.cgi?id=189215

Reviewed by Carlos Garcia Campos.

* platform/graphics/texmap/coordinated/CoordinatedGraphicsLayer.cpp:
(WebCore::CoordinatedGraphicsLayer::purgeBackingStores):
Manually clear out any layer-side state of any image backing that is
associated with this layer, much like we do for backing stores.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@235671 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years ago[MediaStream] Simplify logic when changing RealtimeMediaSource settings
eric.carlson@apple.com [Wed, 5 Sep 2018 15:56:44 +0000 (15:56 +0000)]
[MediaStream] Simplify logic when changing RealtimeMediaSource settings
https://bugs.webkit.org/show_bug.cgi?id=189284
<rdar://problem/44117948>

Reviewed by Youenn Fablet.

Source/WebCore:

Remove all "apply<setting>" methods from RealtimeMediaSource and derived classes, and
add a bitfield to settingsDidChange so classes can do setting-specific setup and
configuration by overriding settingsDidChange.

No new tests, no functionality changed.

* Modules/mediastream/CanvasCaptureMediaStreamTrack.cpp:
(WebCore::CanvasCaptureMediaStreamTrack::Source::canvasResized):
* Modules/mediastream/CanvasCaptureMediaStreamTrack.h:
* platform/mediastream/MediaConstraints.h:
(WebCore::NumericConstraint::fitnessDistance const):
(WebCore::NumericConstraint::valueForDiscreteCapabilityValues const):
* platform/mediastream/RealtimeIncomingVideoSource.h:
(): Deleted.
* platform/mediastream/RealtimeMediaSource.cpp:
(WebCore::RealtimeMediaSource::RealtimeMediaSource):
(WebCore::RealtimeMediaSource::settingsDidChange):
(WebCore::RealtimeMediaSource::fitnessDistance):
(WebCore::applyNumericConstraint):
(WebCore::RealtimeMediaSource::applyConstraint):
(WebCore::RealtimeMediaSource::applyConstraints):
(WebCore::RealtimeMediaSource::setSize):
(WebCore::RealtimeMediaSource::setWidth):
(WebCore::RealtimeMediaSource::setHeight):
(WebCore::RealtimeMediaSource::setFrameRate):
(WebCore::RealtimeMediaSource::setAspectRatio):
(WebCore::RealtimeMediaSource::setFacingMode):
(WebCore::RealtimeMediaSource::setVolume):
(WebCore::RealtimeMediaSource::setSampleRate):
(WebCore::RealtimeMediaSource::setSampleSize):
(WebCore::RealtimeMediaSource::setEchoCancellation):
* platform/mediastream/RealtimeMediaSource.h:
* platform/mediastream/RealtimeMediaSourceSettings.cpp:
(WebCore::RealtimeMediaSourceSettings::diff const):
* platform/mediastream/RealtimeMediaSourceSettings.h:
(WebCore::RealtimeMediaSourceSettings::allFlags):
* platform/mediastream/RealtimeVideoSource.cpp:
(WebCore::RealtimeVideoSource::applySize): Deleted.
(WebCore::RealtimeVideoSource::applyFrameRate): Deleted.
* platform/mediastream/RealtimeVideoSource.h:
* platform/mediastream/gstreamer/GStreamerAudioCaptureSource.cpp:
(WebCore::GStreamerAudioCaptureSource::settingsDidChange):
(WebCore::GStreamerAudioCaptureSource::applySampleRate): Deleted.
* platform/mediastream/gstreamer/GStreamerAudioCaptureSource.h:
* platform/mediastream/gstreamer/GStreamerVideoCaptureSource.cpp:
(WebCore::GStreamerVideoCaptureSource::settingsDidChange):
(WebCore::GStreamerVideoCaptureSource::applySize): Deleted.
(WebCore::GStreamerVideoCaptureSource::applyFrameRate): Deleted.
* platform/mediastream/gstreamer/GStreamerVideoCaptureSource.h:
* platform/mediastream/mac/AVVideoCaptureSource.h:
* platform/mediastream/mac/AVVideoCaptureSource.mm:
(WebCore::AVVideoCaptureSource::settingsDidChange):
(WebCore::AVVideoCaptureSource::setPreset):
(WebCore::AVVideoCaptureSource::setFrameRate):
(WebCore::AVVideoCaptureSource::applySizeAndFrameRate):
(WebCore::AVVideoCaptureSource::processNewFrame):
(WebCore::AVVideoCaptureSource::applySize): Deleted.
(WebCore::AVVideoCaptureSource::applyFrameRate): Deleted.
* platform/mediastream/mac/CoreAudioCaptureSource.cpp:
(WebCore::CoreAudioCaptureSource::settingsDidChange):
(WebCore::CoreAudioCaptureSource::applySampleRate): Deleted.
(WebCore::CoreAudioCaptureSource::applyEchoCancellation): Deleted.
* platform/mediastream/mac/CoreAudioCaptureSource.h:
* platform/mediastream/mac/DisplayCaptureSourceCocoa.cpp:
(WebCore::DisplayCaptureSourceCocoa::settingsDidChange):
(WebCore::DisplayCaptureSourceCocoa::startProducingData):
(WebCore::DisplayCaptureSourceCocoa::applySize): Deleted.
(WebCore::DisplayCaptureSourceCocoa::applyFrameRate): Deleted.
* platform/mediastream/mac/DisplayCaptureSourceCocoa.h:
* platform/mediastream/mac/MockRealtimeAudioSourceMac.h:
* platform/mediastream/mac/MockRealtimeAudioSourceMac.mm:
(WebCore::MockRealtimeAudioSourceMac::settingsDidChange):
(WebCore::MockRealtimeAudioSourceMac::applySampleRate): Deleted.
* platform/mediastream/mac/MockRealtimeVideoSourceMac.h:
* platform/mediastream/mac/MockRealtimeVideoSourceMac.mm:
(WebCore::MockRealtimeVideoSourceMac::settingsDidChange):
(WebCore::MockRealtimeVideoSourceMac::applySize): Deleted.
* platform/mediastream/mac/RealtimeIncomingVideoSourceCocoa.mm:
(WebCore::RealtimeIncomingVideoSourceCocoa::processNewSample):
* platform/mediastream/mac/ScreenDisplayCaptureSourceMac.h:
* platform/mediastream/mac/ScreenDisplayCaptureSourceMac.mm:
(WebCore::ScreenDisplayCaptureSourceMac::settingsDidChange):
(WebCore::ScreenDisplayCaptureSourceMac::applySize): Deleted.
(WebCore::ScreenDisplayCaptureSourceMac::applyFrameRate): Deleted.
* platform/mock/MockRealtimeAudioSource.cpp:
(WebCore::MockRealtimeAudioSource::settingsDidChange):
* platform/mock/MockRealtimeAudioSource.h:
* platform/mock/MockRealtimeVideoSource.cpp:
(WebCore::MockRealtimeVideoSource::settingsDidChange):
(WebCore::MockRealtimeVideoSource::applySize): Deleted.
* platform/mock/MockRealtimeVideoSource.h:

Source/WebKit:

* WebProcess/cocoa/UserMediaCaptureManager.cpp:
(WebKit::UserMediaCaptureManager::Source::setSettings):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@235670 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoGTK-focused gardening
Ms2ger@igalia.com [Wed, 5 Sep 2018 15:22:15 +0000 (15:22 +0000)]
GTK-focused gardening
https://bugs.webkit.org/show_bug.cgi?id=189305

Unreviewed test gardening.

* TestExpectations: Remove passing test.
* fast/mediastream/RTCPeerConnection-media-setup-callbacks-single-dialog.html: Speculative fix for r235484.
* platform/gtk/http/tests/inspector/network/resource-request-headers-expected.txt: Added.
* platform/gtk/http/tests/loading/oauth-expected.txt: Added.
* platform/gtk/http/tests/websocket/connection-refusal-in-frame-resource-load-statistics-expected.txt: Added.
* platform/gtk/imported/w3c/web-platform-tests/xhr/send-network-error-sync-events.sub-expected.txt: Updated for r235354.
* platform/gtk/js/dom/dom-static-property-for-in-iteration-expected.txt: Updated for r235245.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@235669 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years ago[LFC][BFC] Move MarginCollapse class under BlockFormattingContext::Geometry
zalan@apple.com [Wed, 5 Sep 2018 14:42:45 +0000 (14:42 +0000)]
[LFC][BFC] Move MarginCollapse class under BlockFormattingContext::Geometry
https://bugs.webkit.org/show_bug.cgi?id=189296

Reviewed by Antti Koivisto.

Margin collapsing is part of geometry after all.

* layout/blockformatting/BlockFormattingContext.cpp:
(WebCore::Layout::BlockFormattingContext::computeEstimatedMarginTop const):
* layout/blockformatting/BlockFormattingContext.h:
* layout/blockformatting/BlockFormattingContextGeometry.cpp:
(WebCore::Layout::BlockFormattingContext::Geometry::estimatedMarginTop):
* layout/blockformatting/BlockMarginCollapse.cpp:
(WebCore::Layout::BlockFormattingContext::Geometry::MarginCollapse::collapsedMarginTopFromFirstChild):
(WebCore::Layout::BlockFormattingContext::Geometry::MarginCollapse::nonCollapsedMarginTop):
(WebCore::Layout::BlockFormattingContext::Geometry::MarginCollapse::computedNonCollapsedMarginTop):
(WebCore::Layout::BlockFormattingContext::Geometry::MarginCollapse::computedNonCollapsedMarginBottom):
(WebCore::Layout::BlockFormattingContext::Geometry::MarginCollapse::marginTop):
(WebCore::Layout::BlockFormattingContext::Geometry::MarginCollapse::marginBottom):
(WebCore::Layout::BlockFormattingContext::Geometry::MarginCollapse::isMarginBottomCollapsedWithParent):
(WebCore::Layout::BlockFormattingContext::Geometry::MarginCollapse::isMarginTopCollapsedWithParentMarginBottom):
(WebCore::Layout::BlockFormattingContext::Geometry::MarginCollapse::collapsedMarginBottomFromLastChild):
(WebCore::Layout::BlockFormattingContext::Geometry::MarginCollapse::nonCollapsedMarginBottom):
(WebCore::Layout::BlockFormattingContext::MarginCollapse::collapsedMarginTopFromFirstChild): Deleted.
(WebCore::Layout::BlockFormattingContext::MarginCollapse::nonCollapsedMarginTop): Deleted.
(WebCore::Layout::BlockFormattingContext::MarginCollapse::computedNonCollapsedMarginTop): Deleted.
(WebCore::Layout::BlockFormattingContext::MarginCollapse::computedNonCollapsedMarginBottom): Deleted.
(WebCore::Layout::BlockFormattingContext::MarginCollapse::marginTop): Deleted.
(WebCore::Layout::BlockFormattingContext::MarginCollapse::marginBottom): Deleted.
(WebCore::Layout::BlockFormattingContext::MarginCollapse::isMarginBottomCollapsedWithParent): Deleted.
(WebCore::Layout::BlockFormattingContext::MarginCollapse::isMarginTopCollapsedWithParentMarginBottom): Deleted.
(WebCore::Layout::BlockFormattingContext::MarginCollapse::collapsedMarginBottomFromLastChild): Deleted.
(WebCore::Layout::BlockFormattingContext::MarginCollapse::nonCollapsedMarginBottom): Deleted.
(WebCore::Layout::BlockFormattingContext::MarginCollapse::estimatedMarginTop): Deleted.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@235668 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years ago[LFC] Make BlockFormattingContext::Geometry a subclass of FormattingContext::Geometry
zalan@apple.com [Wed, 5 Sep 2018 14:36:43 +0000 (14:36 +0000)]
[LFC] Make BlockFormattingContext::Geometry a subclass of FormattingContext::Geometry
https://bugs.webkit.org/show_bug.cgi?id=189294

Reviewed by Antti Koivisto.

This simplifies some of the call sites.

* layout/blockformatting/BlockFormattingContext.cpp:
(WebCore::Layout::BlockFormattingContext::computeWidthAndMargin const):
(WebCore::Layout::BlockFormattingContext::computeHeightAndMargin const):
* layout/blockformatting/BlockFormattingContext.h:
* layout/blockformatting/BlockFormattingContextGeometry.cpp:
(WebCore::Layout::BlockFormattingContext::Geometry::inFlowNonReplacedHeightAndMargin):
(WebCore::Layout::BlockFormattingContext::Geometry::inFlowNonReplacedWidthAndMargin):
(WebCore::Layout::BlockFormattingContext::Geometry::inFlowReplacedWidthAndMargin):
(WebCore::Layout::BlockFormattingContext::Geometry::inFlowPositionedPosition):
(WebCore::Layout::BlockFormattingContext::Geometry::inFlowHeightAndMargin):
(WebCore::Layout::BlockFormattingContext::Geometry::instrinsicWidthConstraints):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@235667 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoREGRESSION (r235419): DFGCFG.h is missing from JavaScriptCore Xcode project
ddkilzer@apple.com [Wed, 5 Sep 2018 14:19:18 +0000 (14:19 +0000)]
REGRESSION (r235419): DFGCFG.h is missing from JavaScriptCore Xcode project

Found using `tidy-Xcode-project-file --missing` (see Bug
188754).  Fix was made manually.

* JavaScriptCore.xcodeproj/project.pbxproj:
(dfg/DFGCFG.h): Revert accidental change in r235419 by restoring
`name` and `path` values to file reference.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@235666 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoREGRESSION (r160736): Bundle folder is missing in MiniBrowser Xcode project
ddkilzer@apple.com [Wed, 5 Sep 2018 12:31:31 +0000 (12:31 +0000)]
REGRESSION (r160736): Bundle folder is missing in MiniBrowser Xcode project

Found using `tidy-Xcode-project-file --missing` (see Bug
188754).  Folder fix was made manually.

* MiniBrowser/MiniBrowser.xcodeproj/project.pbxproj:
(Bundle): Change the path of the folder to "mac/Bundle", and set
the name back to "Bundle".
(Bundle/Info.plist): Change file reference to be relative to its
group.  Accomplished using Xcode after fixing the folder path.
(Bundle/MiniBrowserBundle_Prefix.pch): Ditto.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@235665 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoREGRESSION (r184033): ContentExtensionTester folder is missing in ContentExtensionTes...
ddkilzer@apple.com [Wed, 5 Sep 2018 12:09:21 +0000 (12:09 +0000)]
REGRESSION (r184033): ContentExtensionTester folder is missing in ContentExtensionTester Xcode project

Found using `tidy-Xcode-project-file --missing` (see Bug
188754).  Fixes were made manually.

* ContentExtensionTester/ContentExtensionTester.xcodeproj/project.pbxproj:
(ContentExtensionTester): Change the path of the folder to map
to "." (which exists and is where main.m is located), and then
set the folder's name back to "ContentExtensionTester".
(main.m): Change file reference to be relative to its group now
that the group's path is fixed.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@235664 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoREGRESSION (r235585): WebDownloadInternal.h is missing from WebKitLegacy Xcode project
ddkilzer@apple.com [Wed, 5 Sep 2018 11:46:43 +0000 (11:46 +0000)]
REGRESSION (r235585): WebDownloadInternal.h is missing from WebKitLegacy Xcode project
<https://bugs.webkit.org/show_bug.cgi?id=189072>

Found using tidy-Xcode-project-file (see Bug 188754).

* WebKitLegacy.xcodeproj/project.pbxproj: Remove references to
WebDownloadInternal.h since the header was removed in r235585.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@235663 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoisAsyncGeneratorMethodParseMode() should check for SourceParseMode::AsyncGeneratorWra...
mark.lam@apple.com [Wed, 5 Sep 2018 08:14:42 +0000 (08:14 +0000)]
isAsyncGeneratorMethodParseMode() should check for SourceParseMode::AsyncGeneratorWrapperMethodMode.
https://bugs.webkit.org/show_bug.cgi?id=189292
<rdar://problem/38907433>

Reviewed by Saam Barati.

JSTests:

* stress/regress-189292.js: Added.

Source/JavaScriptCore:

Previously, isAsyncGeneratorMethodParseMode() was checking for AsyncGeneratorWrapperFunctionMode
instead of AsyncGeneratorWrapperMethodMode.  This patch fixes it
to check for AsyncGeneratorWrapperMethodMode (to match what is expected as indicated
in the name isAsyncGeneratorMethodParseMode).

* parser/ParserModes.h:
(JSC::isAsyncGeneratorMethodParseMode):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@235662 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoRemove unused bad_optional_access implementation.
mark.lam@apple.com [Wed, 5 Sep 2018 07:52:17 +0000 (07:52 +0000)]
Remove unused bad_optional_access implementation.
https://bugs.webkit.org/show_bug.cgi?id=189297

Reviewed by David Kilzer.

This is a speculative fix for resolving build errors arising from conflicting
definitions of bad_optional_access when building WebKit with the latest clang.

* wtf/Optional.h:
(std::bad_optional_access::bad_optional_access): Deleted.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@235661 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years ago[EME] Add a layoutTest for ClearKey WebM video playback
commit-queue@webkit.org [Wed, 5 Sep 2018 06:50:13 +0000 (06:50 +0000)]
[EME] Add a layoutTest for ClearKey WebM video playback
https://bugs.webkit.org/show_bug.cgi?id=189200

Patch by Yacine Bandou <yacine.bandou_ext@softathome.com> on 2018-09-04
Reviewed by Xabier Rodriguez-Calvar.

Add a new layoutTest in order to test the playback of ClearKey encrypted WebM.

* media/encrypted-media/clearKey/clearKey-webm-video-playback-mse-expected.txt: Added.
* media/encrypted-media/clearKey/clearKey-webm-video-playback-mse.html: Added.
* platform/gtk/TestExpectations:
* platform/mac/TestExpectations:
* platform/wpe/TestExpectations:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@235660 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoAdd basic support for ScrollIntoViewOptions
commit-queue@webkit.org [Wed, 5 Sep 2018 05:19:58 +0000 (05:19 +0000)]
Add basic support for ScrollIntoViewOptions
https://bugs.webkit.org/show_bug.cgi?id=189258

Patch by Frederic Wang <fwang@igalia.com> on 2018-09-04
Reviewed by Simon Fraser.

LayoutTests/imported/w3c:

Update test expectations for WPT scrollIntoView tests.

* web-platform-tests/css/cssom-view/scrollIntoView-scrollMargin-expected.txt: This is not
supported yet but update the error message.
* web-platform-tests/css/cssom-view/scrollIntoView-scrollPadding-expected.txt: Ditto.
* web-platform-tests/css/cssom-view/scrollIntoView-shadow-expected.txt: Update expectation
now that this test passes.
* web-platform-tests/css/cssom-view/scrollIntoView-smooth-expected.txt: Ditto.
* web-platform-tests/css/cssom-view/scrollintoview-expected.txt: Ditto.

Source/WebCore:

This patch introduces a new ScrollIntoViewOptions parameter that can be passed into
Element.scrollIntoView instead of a boolean. A basic support for scroll alignments is
implemented, so that it is closer to the behavior of the CSSOMView spec while still remaining
compatible with the current boolean-parameter implementation. Full implementation that
takes into account orientation/direction will be handled in bug 161611. This patch is also a
preliminary step to support the ScrollBehavior (bug 188043) for ScrollIntoView.

Tests: web-platform-tests/css/cssom-view/scrollintoview-html
       web-platform-tests/css/cssom-view/scrollIntoView-smooth.html

* CMakeLists.txt: Add new IDL files.
* DerivedSources.make: Ditto.
* Sources.txt: Add new generated cpp JS bindings.
* WebCore.xcodeproj/project.pbxproj: Add files to build to XCode.
* dom/Element.cpp:
(WebCore::toScrollAlignment): Convert ScrollLogicalPosition to scroll alignment. Orientation
and direction are not implemented yet.
(WebCore::Element::scrollIntoView): Implement new scrollIntoView version accepting
ScrollIntoViewOptions parameter.
* dom/Element.h: Declare new scrollIntoView.
* dom/Element.idl: Make scrollIntoView accept a ScrollIntoViewOptions parameter.
* page/ScrollIntoViewOptions.h: Added.
* page/ScrollIntoViewOptions.idl: Added.
* page/ScrollLogicalPosition.h: Added.
* page/ScrollLogicalPosition.idl: Added.

Source/WebKit:

* DOM/DOMElement.mm: Add ScrollIntoViewOptions
header so that this file can build despite the new scrollIntoView function.

Source/WebKitLegacy/mac:

* WebProcess/InjectedBundle/API/gtk/DOM/WebKitDOMElementGtk.cpp: Add ScrollIntoViewOptions
header so that this file can build despite the new scrollIntoView function.

Source/WebKitLegacy/win:

* DOMCoreClasses.cpp: Add ScrollIntoViewOptions header so that this file
can build despite the new scrollIntoView function.

LayoutTests:

Update test expectations for WPT scrollIntoView tests.

* platform/ios-wk2/imported/w3c/web-platform-tests/css/cssom-view/scrollIntoView-smooth-expected.txt: Removed.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@235659 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoCheck important flags when serializing shorthand with "initial" values
commit-queue@webkit.org [Wed, 5 Sep 2018 05:02:53 +0000 (05:02 +0000)]
Check important flags when serializing shorthand with "initial" values
https://bugs.webkit.org/show_bug.cgi?id=188984

Patch by Oriol Brufau <obrufau@igalia.com> on 2018-09-04
Reviewed by Antti Koivisto.

LayoutTests/imported/w3c:

The test still has some failures due to https://bugs.webkit.org/show_bug.cgi?id=185953
but without this patch it would fail earlier.

* web-platform-tests/css/cssom/shorthand-serialization-expected.txt:
* web-platform-tests/css/cssom/shorthand-serialization.html:

Source/WebCore:

Test: imported/w3c/web-platform-tests/css/cssom/shorthand-serialization.html

The test still has some failures due to https://bugs.webkit.org/show_bug.cgi?id=185953
but without this patch it would fail earlier.

* css/StyleProperties.cpp:
(WebCore::StyleProperties::get4Values const):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@235658 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoRemove pointless RenderSVGResourceMode::ApplyToDefault enum value
antti@apple.com [Wed, 5 Sep 2018 04:32:59 +0000 (04:32 +0000)]
Remove pointless RenderSVGResourceMode::ApplyToDefault enum value
https://bugs.webkit.org/show_bug.cgi?id=189260

Reviewed by Sam Weinig.

Default does not need a bit, it is better represented with an empty OptionSet.

* rendering/svg/RenderSVGResource.cpp:
(WebCore::requestPaintingResource):
* rendering/svg/RenderSVGResource.h:
* rendering/svg/RenderSVGResourceClipper.cpp:
(WebCore::RenderSVGResourceClipper::applyResource):
* rendering/svg/RenderSVGResourceFilter.cpp:
(WebCore::RenderSVGResourceFilter::applyResource):
(WebCore::RenderSVGResourceFilter::postApplyResource):
* rendering/svg/RenderSVGResourceGradient.cpp:
(WebCore::RenderSVGResourceGradient::applyResource):
* rendering/svg/RenderSVGResourceMasker.cpp:
(WebCore::RenderSVGResourceMasker::applyResource):
* rendering/svg/RenderSVGResourcePattern.cpp:
(WebCore::RenderSVGResourcePattern::applyResource):
(WebCore::RenderSVGResourcePattern::postApplyResource):
* rendering/svg/RenderSVGResourceSolidColor.cpp:
(WebCore::RenderSVGResourceSolidColor::applyResource):
(WebCore::RenderSVGResourceSolidColor::postApplyResource):
* rendering/svg/SVGInlineTextBox.cpp:
(WebCore::SVGInlineTextBox::SVGInlineTextBox):
(WebCore::SVGInlineTextBox::paintSelectionBackground):
(WebCore::SVGInlineTextBox::paint):
(WebCore::SVGInlineTextBox::acquirePaintingResource):
(WebCore::SVGInlineTextBox::paintDecorationWithStyle):
* rendering/svg/SVGRenderingContext.cpp:
(WebCore::SVGRenderingContext::~SVGRenderingContext):
(WebCore::SVGRenderingContext::prepareToRenderSVGContent):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@235657 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoRename experimental feature for process swap on navigation
cdumez@apple.com [Wed, 5 Sep 2018 02:36:54 +0000 (02:36 +0000)]
Rename experimental feature for process swap on navigation
https://bugs.webkit.org/show_bug.cgi?id=189280

Reviewed by Antti Koivisto.

The "process swap on navigation" experimental feature on macOS Mojave does not work due to missing
browser changes, resulting in frequent load hangs. When enabling the experimental feature in WebKit
ToT or Safari Technology Preview for testing, it enables it as well for Mojave's system Safari which
is annoying.

This patch renames the experimental feature so that the feature stays disabled in Mojave's system
Safari even when enabled in WebKit ToT / STP.

* Shared/WebPreferences.yaml:
* UIProcess/API/C/WKPreferences.cpp:
(WKPreferencesSetProcessSwapOnNavigationEnabled):
(WKPreferencesGetProcessSwapOnNavigationEnabled):
* UIProcess/WebPreferences.cpp:
(WebKit::WebPreferences::updateBoolValueForExperimentalFeatureKey):
* UIProcess/WebProcessPool.cpp:
(WebKit::WebProcessPool::createWebPage):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@235656 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoYARR: Add new RegExp back reference tests to specifically test back reference JIT...
msaboff@apple.com [Wed, 5 Sep 2018 02:08:34 +0000 (02:08 +0000)]
YARR: Add new RegExp back reference tests to specifically test back reference JIT code
https://bugs.webkit.org/show_bug.cgi?id=189291

Reviewed by Saam Barati.

New tests.

* fast/regex/backreferences-expected.txt: Added.
* fast/regex/backreferences.html: Added.
* fast/regex/script-tests/backreferences.js: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@235655 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoAdd generic entrypoint and run loop in TestWebKitAPI
don.olmstead@sony.com [Wed, 5 Sep 2018 01:57:00 +0000 (01:57 +0000)]
Add generic entrypoint and run loop in TestWebKitAPI
https://bugs.webkit.org/show_bug.cgi?id=189287

Reviewed by Michael Catanzaro.

The implementations in jsconly are platform agnostic.

* TestWebKitAPI/PlatformJSCOnly.cmake:
Use the glib implementation of Utilities when using that as a run loop
otherwise use the generic implementation.

* TestWebKitAPI/PlatformUtilities.h:
Remove repeated entries from Utilities.h.

* TestWebKitAPI/PlatformWPE.cmake:
Use the generic main.cpp.

* TestWebKitAPI/generic/UtilitiesGeneric.cpp: Renamed from Tools/TestWebKitAPI/jsconly/PlatformUtilitiesJSCOnly.cpp.
(TestWebKitAPI::Util::run):
(TestWebKitAPI::Util::spinRunLoop):
Moved to a generic implementation and added spinRunLoop.

* TestWebKitAPI/generic/main.cpp: Renamed from Tools/TestWebKitAPI/jsconly/main.cpp.
Share between JSCOnly and WPE

* TestWebKitAPI/wpe/main.cpp: Removed.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@235654 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years ago[CMake] Make TestWebKitAPI dependencies explicit
don.olmstead@sony.com [Wed, 5 Sep 2018 01:52:00 +0000 (01:52 +0000)]
[CMake] Make TestWebKitAPI dependencies explicit
https://bugs.webkit.org/show_bug.cgi?id=189282

Reviewed by Fujii Hironori.

The config.h file includes files from JSC, WebCore and WebKit which
places a dependency on those headers even when just building TestWTF.
This makes the depedencies explicit for all CMake ports and shoud
remove any race conditions that can be encountered when building tests.

* TestWebKitAPI/CMakeLists.txt:
* TestWebKitAPI/PlatformGTK.cmake:
* TestWebKitAPI/PlatformWPE.cmake:
* TestWebKitAPI/PlatformWin.cmake:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@235653 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoRemove PseudoElementUserAgentCustom.
emilio@crisal.io [Wed, 5 Sep 2018 01:22:55 +0000 (01:22 +0000)]
Remove PseudoElementUserAgentCustom.
https://bugs.webkit.org/show_bug.cgi?id=189089

Reviewed by Ryosuke Niwa.

Tests at https://github.com/web-platform-tests/wpt/pull/12743.

* css/CSSSelector.cpp:
(WebCore::CSSSelector::pseudoId):
(WebCore::CSSSelector::parsePseudoElementType):
* css/CSSSelector.h:
(WebCore::CSSSelector::isCustomPseudoElement const):
* css/RuleSet.cpp:
(WebCore::RuleSet::addRule):
* css/parser/CSSParserSelector.h:
(WebCore::CSSParserSelector::needsImplicitShadowCombinatorForMatching const):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@235652 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoPost review Weinig fix-ups
dino@apple.com [Wed, 5 Sep 2018 00:50:33 +0000 (00:50 +0000)]
Post review Weinig fix-ups
https://bugs.webkit.org/show_bug.cgi?id=189288

Reviewed by Sam Weinig.

Fix-ups from https://bugs.webkit.org/show_bug.cgi?id=189252

Source/WebCore:

* platform/MIMETypeRegistry.cpp:
(WebCore::MIMETypeRegistry::getSystemPreviewMIMETypes):
(WebCore::initializeSystemPreviewMIMETypes): Deleted.
* rendering/RenderThemeIOS.mm:
(WebCore::arKitBundle):
(WebCore::loadARKitPDFPage):
(WebCore::systemPreviewLogo):

Source/WebKit:

* UIProcess/Cocoa/SystemPreviewControllerCocoa.mm:
(-[_WKPreviewControllerDataSource previewController:previewItemAtIndex:]):
(getUTIForMIMEType): Deleted.
* UIProcess/ios/WKSystemPreviewView.mm:
(getUTIForSystemPreviewMIMEType):
(-[WKSystemPreviewView web_setContentProviderData:suggestedFilename:]):
(getUTIForMIMEType): Deleted.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@235651 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoslotchange event doesn't get fired when inserting, removing, or renaming slot elements
rniwa@webkit.org [Wed, 5 Sep 2018 00:22:43 +0000 (00:22 +0000)]
slotchange event doesn't get fired when inserting, removing, or renaming slot elements
https://bugs.webkit.org/show_bug.cgi?id=189144
<rdar://problem/43871061>

Reviewed by Antti Koivisto.

LayoutTests/imported/w3c:

* web-platform-tests/shadow-dom/slotchange-customelements-expected.txt:
* web-platform-tests/shadow-dom/slotchange-event-expected.txt:
* web-platform-tests/shadow-dom/slotchange-expected.txt:

Source/WebCore:

This patch implements `slotchange` event when a slot element is inserted, removed, or renamed in the DOM tree.
Let us consider each scenario separately.

Insertion (https://dom.spec.whatwg.org/#concept-node-insert): In this case, we must fire `slotchange` event on
slot elements whose assigned nodes have changed in the tree order. When there is at most one slot element for
each name, this can be done by simply checking whether each slot has assigned nodes or not. When there are more
than one slot element, however, the newly inserted slot element may now become the first slot of a given name,
and gain assined nodes while the previously first element loses its assigned nodes. To see if the newly inserted
slot element is the first of its kind, we must travere the DOM tree to check the order of that and the previously
first slot element. To do this, we resolve the slot elements before start inserting nodes in
executeNodeInsertionWithScriptAssertion via SlotAssignment::resolveSlotsBeforeNodeInsertionOrRemoval. Note that
when the DOM tree has at most one slot element of its kind, resolveSlotsBeforeNodeInsertionOrRemoval is a no-op
and addSlotElementByName continues to operate in O(1). Becasue addSlotElementByName is called on each inserted
slot element in the tree order, we do the tree traversal upon finding the first slot element which has more than
one of its kind in the current tree. In this case, we resolve all other slot elements and enqueues slotchange
event as needed to avoid doing the tree traversal more than once.

Removal (https://dom.spec.whatwg.org/#concept-node-remove): In removal, we're concerned with removing the first
slot element of its kind. We must fire slotchange event on the remaining slot elements which became the first of
its kind after the removal as well as the ones which got removed from the tree if they had assigned nodes.
Furthermore, the DOM specification mandates that we first fire slotchange events in the tree from which a node
was removed and then in the removed subtree. Because we must only fire slotchange event on the first slot element
of its kind which has been removed, we must resolve the first slot elements of each kind before a node removal
in removeAllChildrenWithScriptAssertion and removeNodeWithScriptAssertion as we've done for insertion. Again,
in the case there was at most one slot element of each kind, resolveSlotsBeforeNodeInsertionOrRemoval is a no-op
and removeSlotElementByName would continue to operate in O(1). When there are multiple slot elements for a given
kind, we immediately enqueue slotchange event on the slot elements which newly became the first of its kind but
delay the enqueuing of slotchange event on the removed slot elements until removeSlotElementByName is called on
that element so that enqueuing of slotchange events on the slot elements still remaining in the in the tree would
happen before those which got removed as the DOM specification mandates.

Rename (https://dom.spec.whatwg.org/#shadow-tree-slots): In the case the slot element's name content attribute
is changed, the renamed element might have become the first of its kind or ceased to be the first of its kind.
There could be two other slot elements appearing later in the tree order which might have gained or lost assigned
nodes as a result. In this case, we invoke the algorithms for removing & inserting the slot with a key difference:
we enqueue slotchange event on the renamed slot immediately if it has assigned nodes.

To enqueue slotchange event in the tree order, this patch adds oldElement, which is a WeakPtr to a slot element,
to SlotAssignment::Slot. This WeakPtr is set to the slot element which used to have assigned nodes prior to the
node insertion, removal, or rename but no longer has after the mutation. This patch also adds a slot mutation
version number, m_slotMutationVersion, which is incremented whenever a node is about to be inserted or removed,
and slot resolution version, m_slotResolutionVersion, which is set to the current slot mutation version number
when the full slot resolution is triggered during slot mutations. They are used to avoid redundant tree traversals
in resolveSlotsAfterSlotMutation. This patch also makes m_needsToResolveSlotElements compiled in release builds
to avoid resolving slot elements when there is at most one slot element of each kind.

For insertion, oldElement is set to the slot which used to be the first of its kind before getting set to a slot
element being inserted in resolveSlotsAfterSlotMutation. We enqueue slotchange event on the newly inserted slot
element at that point (1). Since the slot element which used to be the first of its kind appears after the newly
inserted slot element by definition, we're guaranteed to see this oldElement later in the tree traversal upon
which we enqueue slotchange event. Note that if this slot element was the first of its kind, then we're simply
hitting (1), which is O(1) and does not invoke the tree traversal.

For removal, oldElement is set to the slot which used to be the first of its kind. Because this slot element is
getting removed, slotchange event must be enqueud after slotchange events have been enqueued on all slot elements
still remaining in the tree. To do this, we enqueue slotchange event immediately on the first slot element of
its kind during the tree traversal as we encounter it (2), and set oldElement to the previosuly-first-but-removed
slot element. slotchange event is enqueued on this slot element when removeSlotElementByName is later invoked via
HTMLSlotElement::removedFromAncestor which traverses each removed element in the tree order. Again, if this was
the last slot of its kind, we'd simply expedite (2) by enqueuing slotchange event during removeSlotElementByName,
which is O(1).

When the DOM invokes the concept to replace all children (https://dom.spec.whatwg.org/#concept-node-replace-all),
however, this algorithm isn't sufficient because the removal of each child happens one after another. We would
either need to resolve slots between each removal, or treat the removal of all children as a single operation.
While the DOM specification currently specifies the former behavior, this patch implements the latter behavior
to avoid useless work. See the DOM spec issue: https://github.com/w3c/webcomponents/issues/764

Test: fast/shadow-dom/slotchange-for-slot-mutation.html

* dom/ContainerNode.cpp:
(WebCore::ContainerNode::removeAllChildrenWithScriptAssertion): Call resolveSlotsBeforeNodeInsertionOrRemoval
before start removing children.
(WebCore::ContainerNode::removeNodeWithScriptAssertion): Ditto.
(WebCore::executeNodeInsertionWithScriptAssertion): Ditto before inserting children.
* dom/ShadowRoot.cpp:
(WebCore::ShadowRoot::~ShadowRoot): Set m_hasBegunDeletingDetachedChildren to true. This flag is used to supress
slotchange events during the shadow tree destruction.
(WebCore::ShadowRoot::renameSlotElement): Added.
(WebCore::ShadowRoot::removeSlotElementByName): Added oldParentOfRemovedTree as an argument.
* dom/ShadowRoot.h:
(WebCore::ShadowRoot::shouldFireSlotchangeEvent): Added.
* dom/SlotAssignment.cpp:
(WebCore::findSlotElement): Added.
(WebCore::nextSlotElementSkippingSubtree): Added.
(WebCore::SlotAssignment::hasAssignedNodes): Added. Returns true if the slot of a given name has assigned nodes.
(WebCore::SlotAssignment::renameSlotElement): Added.
(WebCore::SlotAssignment::addSlotElementByName): Call resolveSlotsAfterSlotMutation when slotchange event needs
to be dispatched for the current slot and there are more than one slot elements.
(WebCore::SlotAssignment::removeSlotElementByName): Ditto. When the slot's oldElement is set to the current slot
element, meaning that this slot element used to have assigned nodes, then enqueue slotchange event. It also has
a special case for oldParentOfRemovedTree is null when renaming a slot element. In this case, we want to enqueue
slot change event immediately on the renamed slot element and any affected elements as in a node insertion since
removeSlotElementByName would never be called on a slot element which newly become the first of its kind after
a slot rename.
(WebCore::SlotAssignment::resolveSlotsAfterSlotMutation): Added. This is the slot resolution algorithm invoked
when there are more than one slot elements for a given name. It has two modes dealing with insertion & removal.
The insertion mode is also used for renaming a slot element. The firs
(WebCore::SlotAssignment::resolveSlotsBeforeNodeInsertionOrRemoval): Added. Resolves all slot elements prior to
inserting or removing nodes. In many cases, this should be a no-op since m_needsToResolveSlotElements is set to
true only when there are more than one slot element of its kind.
(WebCore::SlotAssignment::willRemoveAllChildren): Ditto. Also sets m_willBeRemovingAllChildren to true.
(WebCore::SlotAssignment::didChangeSlot):
(WebCore::SlotAssignment::resolveAllSlotElements): Use seenFirstElement instead of element to indicate whether
we have seen a slot element of given name for consistency with resolveSlotsAfterSlotMutation.
* dom/SlotAssignment.h:
(WebCore::SlotAssignment::Slot): Added oldElement and seenFirstElement.
(WebCore::SlotAssignment): Always compile m_needsToResolveSlotElements. Added m_willBeRemovingAllChildren,
m_slotMutationVersion, and m_slotResolutionVersion.
(WebCore::ShadowRoot::resolveSlotsBeforeNodeInsertionOrRemoval): Added. Calls the one in SlotAssignment.
(WebCore::ShadowRoot::willRemoveAllChildren): Ditto.
* html/HTMLSlotElement.cpp:
(WebCore::HTMLSlotElement::removedFromAncestor):
(WebCore::HTMLSlotElement::attributeChanged): Calls ShadowRoot::renameSlotElement instead of
removeSlotElementByName and addSlotElementByName pair.

LayoutTests:

Added a W3C style testharness.js test for inserting, removing, and renaming slot elements.

It has 62 distinct test cases for closed/open shadow roots in connected and disconnected trees
for the total of 248 test cases.

This test presumes the resolution of https://github.com/w3c/webcomponents/issues/764 in our favor.

Chrome fails 48 test cases because it doesn't follow the tree order when dispatching slotchange event
on the previously first slot element, and Firefox fails 84 test cases because it fails to fire slotchange
in the tree order when a node is inserted.

* fast/shadow-dom/slotchange-for-slot-mutation-expected.txt: Added.
* fast/shadow-dom/slotchange-for-slot-mutation.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@235650 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoMake LibWebRTCRtpSenderBackend own its libwebrtc RTP sender backend
youenn@apple.com [Wed, 5 Sep 2018 00:04:01 +0000 (00:04 +0000)]
Make LibWebRTCRtpSenderBackend own its libwebrtc RTP sender backend
https://bugs.webkit.org/show_bug.cgi?id=189270

Reviewed by Eric Carlson.

Previously, the libwebrtc senders were owned in LibWebRTCMediaEndpoint.
This patch makes them be owned by LibWebRTCRtpSenderBackend.
This simplifies the implementation of RTCRtpSender::getParameters.
This will help simplify the implementation of RTCRtpSender::replaceTrack.

removeTrack is also refactored so that LibWebRTCMediaEndpoint directly uses libwebrtc constructs and not WebCore libwebrtc-agnostic ones.

No change of behavior.

* Modules/mediastream/PeerConnectionBackend.h:
(WebCore::PeerConnectionBackend::removeTrack):
(WebCore::PeerConnectionBackend::notifyRemovedTrack): Deleted.
* Modules/mediastream/RTCPeerConnection.cpp:
(WebCore::RTCPeerConnection::removeTrack):
* Modules/mediastream/RTCRtpSender.cpp:
(WebCore::RTCRtpSender::getParameters):
* Modules/mediastream/RTCRtpSender.h:
(WebCore::RTCRtpSender::backend):
* Modules/mediastream/RTCRtpSenderBackend.h:
* Modules/mediastream/libwebrtc/LibWebRTCMediaEndpoint.cpp:
(WebCore::LibWebRTCMediaEndpoint::addTrack):
(WebCore::LibWebRTCMediaEndpoint::removeTrack):
(WebCore::LibWebRTCMediaEndpoint::doCreateOffer):
(WebCore::LibWebRTCMediaEndpoint::stop):
(WebCore::LibWebRTCMediaEndpoint::shouldOfferAllowToReceiveAudio const): Deleted.
(WebCore::LibWebRTCMediaEndpoint::shouldOfferAllowToReceiveVideo const): Deleted.
(WebCore::LibWebRTCMediaEndpoint::getRTCRtpSenderParameters): Deleted.
* Modules/mediastream/libwebrtc/LibWebRTCMediaEndpoint.h:
* Modules/mediastream/libwebrtc/LibWebRTCPeerConnectionBackend.cpp:
(WebCore::backendFromRTPSender):
(WebCore::LibWebRTCPeerConnectionBackend::addTrack):
(WebCore::LibWebRTCPeerConnectionBackend::addTransceiver):
(WebCore::LibWebRTCPeerConnectionBackend::removeTrack):
(WebCore::LibWebRTCPeerConnectionBackend::shouldOfferAllowToReceive const):
(WebCore::LibWebRTCPeerConnectionBackend::notifyRemovedTrack): Deleted.
(WebCore::LibWebRTCPeerConnectionBackend::getParameters const): Deleted.
* Modules/mediastream/libwebrtc/LibWebRTCPeerConnectionBackend.h:
* Modules/mediastream/libwebrtc/LibWebRTCRtpSenderBackend.cpp:
(WebCore::LibWebRTCRtpSenderBackend::getParameters const):
* Modules/mediastream/libwebrtc/LibWebRTCRtpSenderBackend.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@235649 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoUnreviewed indentations change.
msaboff@apple.com [Tue, 4 Sep 2018 23:51:12 +0000 (23:51 +0000)]
Unreviewed indentations change.

* yarr/YarrJIT.cpp:
(JSC::Yarr::YarrGenerator::matchBackreference):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@235648 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoPopulate "text/uri-list" with multiple URLs when the pasteboard contains multiple...
wenson_hsieh@apple.com [Tue, 4 Sep 2018 23:47:55 +0000 (23:47 +0000)]
Populate "text/uri-list" with multiple URLs when the pasteboard contains multiple URLs
https://bugs.webkit.org/show_bug.cgi?id=188890
<rdar://problem/43648605>

Reviewed by Tim Horton.

Source/WebCore:

Adds support for exposing a newline separated list of URLs via DataTransfer's "text/uri-list" type when pasting
or dropping multiple items on the pasteboard that can be represented as URLs. Currently on iOS, only the URL of
the first item (if present) is exposed, and on macOS, only the first out of all the URLs in the pasteboard is
exposed.

To fix this, we introduce `Pasteboard::readAllStrings`, which reads a list of pasteboard strings collected from
all available items on the platform pasteboard. Currently, this is only used to provide a list of URL strings
when fetching data for the "text/uri-list" type when calling `DataTransfer.getData()` and
`DataTransferItem.getAsString()`.

Tests:  DragAndDropTests.ExposeMultipleURLsInDataTransfer
        UIPasteboardTests.DataTransferURIListContainsMultipleURLs
        PasteMixedContent.PasteOneOrMoreURLs

* dom/DataTransfer.cpp:
(WebCore::readURLsFromPasteboardAsString):

Add a helper method that reads all URL strings from the pasteboard (for the MIME type "text/uri-list", which
corresponds to NSURLPboardType and "public.url" on macOS and iOS, respectively) and returns a single string
containing all non-empty URLs joined by newline characters. Also takes a filtering block which may be used to
reject URLs from being included in "text/uri-list" output.

(WebCore::DataTransfer::getDataForItem const):
(WebCore::DataTransfer::readStringFromPasteboard const):

Insteading of reading a single string from the pasteboard for "text/uri-list", call the above helper function to
read all URL strings in the pasteboard. If there are files present in the pasteboard, we also filter out URLs
whose schemes are not in the set of schemes that are safe to expose to the page (i.e. http(s), blob, and data).

* platform/Pasteboard.cpp:
(WebCore::Pasteboard::readAllStrings):

Add a default non-Cocoa implementation of readAllStrings() that returns a vector, which may contain the result
of readString().

* platform/Pasteboard.h:
* platform/PasteboardStrategy.h:
* platform/PlatformPasteboard.h:

Add plumbing to grab a list of strings from the pasteboard for a given type.

* platform/cocoa/PasteboardCocoa.mm:
(WebCore::Pasteboard::readAllStrings):
(WebCore::Pasteboard::readString):

Implement these two methods in terms of `readPlatformValuesAsStrings`. `readAllStrings` returns the full list of
results, while `readString` only returns the first result.

* platform/ios/PasteboardIOS.mm:
(WebCore::Pasteboard::readPlatformValuesAsStrings):
(WebCore::Pasteboard::readPlatformValueAsString): Deleted.

Refactor this Cocoa helper function to return a list of pasteboard string values for the given type, rather than
a single string.

* platform/ios/PlatformPasteboardIOS.mm:
(WebCore::PlatformPasteboard::allStringsForType const):

Grab a string for each item (represented by an NSItemProvider) in the pasteboard that has data for the given
type identifier.

(WebCore::PlatformPasteboard::readString const):

Return the absolute string of the NSURL, instead of WebCore::URL::string(). This is needed to handle the case
where the NSURL is constructed from absolute and relative parts using a Plist. While -absoluteString gets us the
full URL string, URL::string() only returns the relative portion.

* platform/mac/PasteboardMac.mm:
(WebCore::Pasteboard::readPlatformValuesAsStrings):
(WebCore::Pasteboard::readPlatformValueAsString): Deleted.

Also refactor this to retrieve a list of pasteboard strings, rather than a single result.

* platform/mac/PlatformPasteboardMac.mm:
(WebCore::typeIdentifierForPasteboardType):
(WebCore::PlatformPasteboard::allStringsForType const):

Add an implementation for `allStringsForType` on macOS. Unlike iOS, it's much trickier to get this right since
we need to maintain compatibility with legacy "NS*Pboard" types, and `NSPasteboardItem` can only provide data
for `NSPasteboardType`s (i.e. UTIs), so there's no way to just iterate through each pasteboard item and ask it
for data that matches the given type, if the types are not UTIs. However, in the case where we have multiple
items, the client must have used NSPasteboardWriting-conformant objects and/or NSPasteboardItem itself to write
data to the pasteboard. Since NSPasteboardWriting-conformant objects register modern pasteboard types when
writing to the pasteboard, we can simply iterate over these pasteboard items and ask for property lists using
type identifiers instead of having to worry about legacy pasteboard types. Furthermore, in the case where there
is only a single item in the pasteboard and we do need to handle legacy pasteboard types, using `-[NSPasteboard
stringForType:]` in the same way we do currently should yield the correct result.

As such, in the case where there is a single pasteboard item, we use `-[NSPasteboard stringForType:]` with the
original legacy type, and in the case where there are multiple items on the pasteboard, we iterate over each of
the pasteboard items and call `-[NSPasteboardItem propertyListForType:]` with the modern pasteboard type
corresponding to the given legacy pasteboard type.

The different corner cases in this logic are tested by the new API test, PasteMixedContent.PasteOneOrMoreURLs,
which exercises several different ways of writing one or more URLs to the pasteboard on macOS, which each result
in different legacy and modern pasteboard types being written to the pasteboard; our implementation of
`PlatformPasteboard::allStringsForType` on macOS handles all cases correctly.

Source/WebKit:

Add some plumbing through pasteboard classes to support `Pasteboard::readAllStrings`. See WebCore ChangeLog for
more detail.

* UIProcess/Cocoa/WebPasteboardProxyCocoa.mm:
(WebKit::WebPasteboardProxy::getPasteboardStringsForType):
* UIProcess/WebPasteboardProxy.h:
* UIProcess/WebPasteboardProxy.messages.in:
* WebProcess/WebCoreSupport/WebPlatformStrategies.cpp:
(WebKit::WebPlatformStrategies::allStringsForType):
* WebProcess/WebCoreSupport/WebPlatformStrategies.h:

Source/WebKitLegacy/mac:

Add some plumbing through pasteboard classes to support `Pasteboard::readAllStrings`. See WebCore ChangeLog for
more detail.

* WebCoreSupport/WebPlatformStrategies.h:
* WebCoreSupport/WebPlatformStrategies.mm:
(WebPlatformStrategies::allStringsForType):

Tools:

* DumpRenderTree/mac/DumpRenderTreePasteboard.mm:
(-[LocalPasteboard pasteboardItems]):

Implement this method to avoid crashing when running layout tests that access the pasteboard.

* TestWebKitAPI/Tests/WebKitCocoa/DragAndDropTests.mm:

Add a test to verify that on macOS and iOS, multiple URLs dropped onto the page are accessible via
"text/uri-list".

* TestWebKitAPI/Tests/WebKitCocoa/PasteMixedContent.mm:

Add a test that exercises 5 different ways to write one or more URLs to the pasteboard on macOS; in all cases,
the URLs written to the pasteboard should be exposed to the page via "text/uri-list". In all of these different
cases, the results of using `-[NSPasteboardItem stringForType:]`, `-[NSURL URLFromPasteboard:]` and
`-[NSPasteboard stringForType:]` will yield different results, so the purpose of this API test is to ensure that
our logic for grabbing a list of URLs from the pasteboard on macOS is robust enough to handle all of these
different behaviors.

* TestWebKitAPI/Tests/ios/UIPasteboardTests.mm:

Add a test to verify that on iOS, using `-[UIPasteboard setURLs:]` to write to multiple URLs to the pasteboard
and then pasting results in "text/uri-list" exposing a list of all the URLs written to the pasteboard.

* WebKitTestRunner/mac/WebKitTestRunnerPasteboard.mm:
(-[LocalPasteboard pasteboardItems]):

Implement this method to avoid crashing when running layout tests that access the pasteboard.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@235647 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoREGRESSION(r235408): GTK bots exiting early
simon.fraser@apple.com [Tue, 4 Sep 2018 23:45:06 +0000 (23:45 +0000)]
REGRESSION(r235408): GTK bots exiting early
https://bugs.webkit.org/show_bug.cgi?id=189063

Reviewed by Michael Catanzaro.

WebKitTestRunner was running the world leak checks even when run without --world-leaks,
causing GTK bot timeouts. So guard updateLiveDocumentsAfterTest(), checkForWorldLeaks()
and findAndDumpWorldLeaks() with m_checkForWorldLeaks checks, and in
TestController::handleControlCommand() print a message if the control script sends the
"#CHECK FOR WORLD LEAKS" command if WTR was not run with --world-leaks.

I tested that running with --world-leaks still works.

* WebKitTestRunner/TestController.cpp:
(WTR::TestController::resetStateToConsistentValues):
(WTR::TestController::updateLiveDocumentsAfterTest):
(WTR::TestController::checkForWorldLeaks):
(WTR::TestController::findAndDumpWorldLeaks):
(WTR::TestController::handleControlCommand):
(WTR::TestController::run):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@235646 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoJSC Build error when changing CPU type: offlineasm: No magic values found. Skipping...
msaboff@apple.com [Tue, 4 Sep 2018 23:13:55 +0000 (23:13 +0000)]
JSC Build error when changing CPU type: offlineasm: No magic values found. Skipping assembly file generation
https://bugs.webkit.org/show_bug.cgi?id=189274

Reviewed by Saam Barati.

Put the derived file LLIntDesiredOffsets.h in an architecture specific subdirectory to make them unique.

Some I got this change mixed up with the change for r235636.  The changes to JavaScriptCore.xcodeproj/project.pbxproj
where landed there.

* JavaScriptCore.xcodeproj/project.pbxproj:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@235645 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoCSS reference filter that references a tiled feTurbulence is blank
simon.fraser@apple.com [Tue, 4 Sep 2018 22:51:47 +0000 (22:51 +0000)]
CSS reference filter that references a tiled feTurbulence is blank
https://bugs.webkit.org/show_bug.cgi?id=188950

Reviewed by Dean Jackson.
Source/WebCore:

We need to run the code that was in RenderSVGResourceFilterPrimitive::determineFilterPrimitiveSubregion()
for CSS reference filters, to set up the various rects in the filter effects.

Do this by moving the code to FilterEffect::determineFilterPrimitiveSubregion(), which makes sense
because it recurses on the FilterEffect input chain. To make it CSS/SVGFilter agnostic, we move filterRegionInUserSpace()
to the Filter base class (for CSSFilter, it just returns m_filterRegion).

Test: css3/filters/reference-filter-set-filter-regions.html

* platform/graphics/filters/Filter.h:
* platform/graphics/filters/FilterEffect.cpp:
(WebCore::FilterEffect::determineFilterPrimitiveSubregion):
* platform/graphics/filters/FilterEffect.h:
* rendering/CSSFilter.cpp:
(WebCore::CSSFilter::determineFilterPrimitiveSubregion):
* rendering/CSSFilter.h:
* rendering/RenderLayerFilters.cpp:
(WebCore::RenderLayerFilters::beginFilterEffect):
* rendering/svg/RenderSVGResourceFilter.cpp:
(WebCore::RenderSVGResourceFilter::applyResource):
* rendering/svg/RenderSVGResourceFilterPrimitive.cpp:
(WebCore::RenderSVGResourceFilterPrimitive::determineFilterPrimitiveSubregion): Deleted.
* rendering/svg/RenderSVGResourceFilterPrimitive.h:
* svg/graphics/filters/SVGFilter.h:

LayoutTests:

* css3/filters/reference-filter-set-filter-regions-expected.html: Added.
* css3/filters/reference-filter-set-filter-regions.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@235644 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years ago[LFC] Rename LayoutPair to BoxPair
zalan@apple.com [Tue, 4 Sep 2018 22:16:44 +0000 (22:16 +0000)]
[LFC] Rename LayoutPair to BoxPair
https://bugs.webkit.org/show_bug.cgi?id=189276

Reviewed by Antti Koivisto.

* layout/FormattingContext.h:
* layout/blockformatting/BlockFormattingContext.cpp:
(WebCore::Layout::BlockFormattingContext::layout const):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@235643 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoUnreviewed follow-up to r235635
mmaxfield@apple.com [Tue, 4 Sep 2018 21:41:54 +0000 (21:41 +0000)]
Unreviewed follow-up to r235635
https://bugs.webkit.org/show_bug.cgi?id=188940

Use "let" variables instead of "var" variables.

* WebGPUShadingLanguageRI/StandardLibrary.js:
(let.standardLibrary):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@235642 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoDisable WebRTC unified plan runtime flag by default
youenn@apple.com [Tue, 4 Sep 2018 21:41:19 +0000 (21:41 +0000)]
Disable WebRTC unified plan runtime flag by default
https://bugs.webkit.org/show_bug.cgi?id=189264

Reviewed by Jer Noble.

Source/WebCore:

Covered by existing tests.

* testing/Internals.cpp:
(WebCore::Internals::resetToConsistentState):
Activate unified plan for testing.

Source/WebKit:

* Shared/WebPreferences.yaml:
Disable unified plan flag by default.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@235641 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years ago[LFC][BFC] Merge computeInFlowWidth(Height)AndMargin and computeFloatingWidth(Height...
zalan@apple.com [Tue, 4 Sep 2018 21:25:50 +0000 (21:25 +0000)]
[LFC][BFC] Merge computeInFlowWidth(Height)AndMargin and computeFloatingWidth(Height)AndMargin.
https://bugs.webkit.org/show_bug.cgi?id=189271

Reviewed by Antti Koivisto.

This is in preparation for adding min/max-widht(height).

* layout/FormattingContext.cpp:
(WebCore::Layout::FormattingContext::computeFloatingHeightAndMargin const): Deleted.
(WebCore::Layout::FormattingContext::computeFloatingWidthAndMargin const): Deleted.
* layout/FormattingContext.h:
* layout/blockformatting/BlockFormattingContext.cpp:
(WebCore::Layout::BlockFormattingContext::computeWidthAndMargin const):
(WebCore::Layout::BlockFormattingContext::computeHeightAndMargin const):
(WebCore::Layout::BlockFormattingContext::computeInFlowHeightAndMargin const): Deleted.
(WebCore::Layout::BlockFormattingContext::computeInFlowWidthAndMargin const): Deleted.
* layout/blockformatting/BlockFormattingContext.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@235639 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoYARR: JIT RegExps with back references
msaboff@apple.com [Tue, 4 Sep 2018 21:18:58 +0000 (21:18 +0000)]
YARR: JIT RegExps with back references
https://bugs.webkit.org/show_bug.cgi?id=180874

Reviewed by Filip Pizlo.

Source/JavaScriptCore:

Implemented JIT'ed back references for all counted types.  The only type of back references
not handled in the JIT are 16bit matches that ignore case.  Such support would require the
canonicalization that is currently handled in the Yarr interpreter via a C funtion call.
The back reference processing for surrogate pairs is implemented by individually comparing
each surrogate ala memcmp.

Added a generated canonicalization table for the LChar (8bit) domain to process case
ignored back references.

Added macro assembler load16(ExtendedAddress) for indexed access to the canonicalization table.

Added a new JIT failure reason for forward references as the check to JIT expressions with
forward references we're handled synonimously those containing back references.

This change is only enabled for 64 bit platforms.

* assembler/MacroAssemblerARM64.h:
(JSC::MacroAssemblerARM64::load16):
* assembler/MacroAssemblerX86_64.h:
(JSC::MacroAssemblerX86_64::load16):
* runtime/RegExp.cpp:
(JSC::RegExp::compile):
(JSC::RegExp::compileMatchOnly):
* yarr/YarrCanonicalize.h:
* yarr/YarrCanonicalizeUCS2.cpp:
* yarr/YarrCanonicalizeUCS2.js:
(set characters.hex.set string_appeared_here):
* yarr/YarrJIT.cpp:
(JSC::Yarr::YarrGenerator::checkNotEnoughInput):
(JSC::Yarr::YarrGenerator::readCharacterDontDecodeSurrogates):
(JSC::Yarr::YarrGenerator::matchBackreference):
(JSC::Yarr::YarrGenerator::generateBackReference):
(JSC::Yarr::YarrGenerator::backtrackBackReference):
(JSC::Yarr::YarrGenerator::generateTerm):
(JSC::Yarr::YarrGenerator::backtrackTerm):
(JSC::Yarr::YarrGenerator::compile):
(JSC::Yarr::dumpCompileFailure):
* yarr/YarrJIT.h:
* yarr/YarrPattern.h:
(JSC::Yarr::BackTrackInfoBackReference::beginIndex):
(JSC::Yarr::BackTrackInfoBackReference::matchAmountIndex):

Source/WTF:

Added ENABLE_YARR_JIT_BACKREFERENCES to enable RegExp JIT'ing of back references
for 64 bit platforms only.

* wtf/Platform.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@235636 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years ago[WHLSL] Implement texture types
mmaxfield@apple.com [Tue, 4 Sep 2018 21:18:22 +0000 (21:18 +0000)]
[WHLSL] Implement texture types
https://bugs.webkit.org/show_bug.cgi?id=188940

Rubber-stamped by Dean Jackson.

Implement the texture types and operations. This includes Sample(), Load(), Store(), Gather(), and GetDimensions().
These functions were implemented according to section 15 of the Vulkan 1.1.83 spec, which lists each operation
and how to compute its results.

* WebGPUShadingLanguageRI/All.js:
* WebGPUShadingLanguageRI/Casts.js: Moved from Intrinsics. The texture operations need to cast too, so these are
moved into a common location.
(cast):
(bitwiseCast):
(castToUchar):
(castToUshort):
(castToUint):
(castToChar):
(castToShort):
(castToInt):
(castToHalf):
(castToFloat):
* WebGPUShadingLanguageRI/Intrinsics.js: Use Casts.js and call into TextureOperations.js.
(Intrinsics.):
(Intrinsics.checkFalse):
(Intrinsics.boxVector):
(Intrinsics.unboxVector):
(Intrinsics):
(Intrinsics.cast): Deleted.
(Intrinsics.bitwiseCast): Deleted.
(Intrinsics.castToHalf): Deleted.
* WebGPUShadingLanguageRI/SPIRV.html:
* WebGPUShadingLanguageRI/Sampler.js: Added. Represent the Sampler type.
(Sampler):
(Sampler.prototype.get rAddressMode):
(Sampler.prototype.get sAddressMode):
(Sampler.prototype.get tAddressMode):
(Sampler.prototype.get minFilter):
(Sampler.prototype.get magFilter):
(Sampler.prototype.get mipmapFilter):
(Sampler.prototype.get lodMinClamp):
(Sampler.prototype.get lodMaxClamp):
(Sampler.prototype.get maxAnisotropy):
(Sampler.prototype.get compareFunction):
(Sampler.prototype.get borderColor):
(Sampler.prototype.calculateBorderColor.computeValues):
(Sampler.prototype.calculateBorderColor):
* WebGPUShadingLanguageRI/StandardLibrary.js: Update the standard library. Some of the functions were
either missing or wrong.
(let.standardLibrary):
* WebGPUShadingLanguageRI/Test.html:
* WebGPUShadingLanguageRI/Test.js: Test many, many combinations of the texture operations.
(makeSampler):
(make1DTexture):
(make1DTextureArray):
(make2DTexture):
(make2DTextureArray):
(make3DTexture):
(makeTextureCube):
(makeRW1DTexture):
(makeRW1DTextureArray):
(makeRW2DTexture):
(makeRW2DTextureArray):
(makeRW3DTexture):
(make2DDepthTexture):
(make2DDepthTextureArray):
(makeDepthTextureCube):
(makeRW2DDepthTexture):
(makeRW2DDepthTextureArray):
(checkFloat4):
* WebGPUShadingLanguageRI/Texture.js: Added. Represents each of the texture types.
(Texture):
(Texture.prototype.get dimension):
(Texture.prototype.get width):
(Texture.prototype.get height):
(Texture.prototype.get depth):
(Texture.prototype.get levelCount):
(Texture.prototype.get layerCount):
(Texture.prototype.get innerType):
(Texture.prototype.get data):
(Texture.prototype.elementChecked):
(Texture.prototype.setElementChecked):
(Texture1D):
(Texture1D.prototype.widthAtLevel):
(Texture1D.prototype.heightAtLevel):
(Texture1D.prototype.depthAtLevel):
(Texture1D.prototype.element):
(Texture1D.prototype.setElement):
(Texture1DArray):
(Texture1DArray.prototype.widthAtLevel):
(Texture1DArray.prototype.heightAtLevel):
(Texture1DArray.prototype.depthAtLevel):
(Texture1DArray.prototype.element):
(Texture1DArray.prototype.setElement):
(Texture2D):
(Texture2D.prototype.widthAtLevel):
(Texture2D.prototype.heightAtLevel):
(Texture2D.prototype.depthAtLevel):
(Texture2D.prototype.element):
(Texture2D.prototype.setElement):
(Texture2DArray):
(Texture2DArray.prototype.widthAtLevel):
(Texture2DArray.prototype.heightAtLevel):
(Texture2DArray.prototype.depthAtLevel):
(Texture2DArray.prototype.element):
(Texture2DArray.prototype.setElement):
(Texture3D):
(Texture3D.prototype.widthAtLevel):
(Texture3D.prototype.heightAtLevel):
(Texture3D.prototype.depthAtLevel):
(Texture3D.prototype.element):
(Texture3D.prototype.setElement):
(TextureCube):
(TextureCube.prototype.widthAtLevel):
(TextureCube.prototype.heightAtLevel):
(TextureCube.prototype.depthAtLevel):
(TextureCube.prototype.element):
(TextureCube.prototype.setElement):
(Texture1DRW):
(Texture1DRW.prototype.widthAtLevel):
(Texture1DRW.prototype.heightAtLevel):
(Texture1DRW.prototype.depthAtLevel):
(Texture1DRW.prototype.element):
(Texture1DRW.prototype.setElement):
(Texture1DArrayRW):
(Texture1DArrayRW.prototype.widthAtLevel):
(Texture1DArrayRW.prototype.heightAtLevel):
(Texture1DArrayRW.prototype.depthAtLevel):
(Texture1DArrayRW.prototype.element):
(Texture1DArrayRW.prototype.setElement):
(Texture2DRW):
(Texture2DRW.prototype.widthAtLevel):
(Texture2DRW.prototype.heightAtLevel):
(Texture2DRW.prototype.depthAtLevel):
(Texture2DRW.prototype.element):
(Texture2DRW.prototype.setElement):
(Texture2DArrayRW):
(Texture2DArrayRW.prototype.widthAtLevel):
(Texture2DArrayRW.prototype.heightAtLevel):
(Texture2DArrayRW.prototype.depthAtLevel):
(Texture2DArrayRW.prototype.element):
(Texture2DArrayRW.prototype.setElement):
(let.TextureDepth2DArrayRW.Texture2DArrayRW.Texture3DRW):
(let.TextureDepth2DArrayRW.Texture2DArrayRW.Texture3DRW.prototype.widthAtLevel):
(let.TextureDepth2DArrayRW.Texture2DArrayRW.Texture3DRW.prototype.heightAtLevel):
(let.TextureDepth2DArrayRW.Texture2DArrayRW.Texture3DRW.prototype.depthAtLevel):
(let.TextureDepth2DArrayRW.Texture2DArrayRW.Texture3DRW.prototype.element):
(let.TextureDepth2DArrayRW.Texture2DArrayRW.Texture3DRW.prototype.setElement):
* WebGPUShadingLanguageRI/TextureOperations.js: Added. This implements the texture operations
as described in section 15 of the Vulkan 1.1.83 spec.
(depthCompareOperation):
(conversionToRGBA):
(projectionOperation):
(cubeMapFaceSelection):
(cubeMapCoordinateTransformation):
(cubeMapDerivativeTransformation):
(scaleFactorOperation):
(levelOfDetailOperation):
(nearest):
(imageLevelSelection):
(strqaToUVWATransformation):
(rne):
(uvwaToIJKLNTransformationAndArrayLayerSelection):
(integerTexelCoordinateOperations):
(mirror):
(transform):
(wrappingOperation):
(calculateLambda):
(shouldBeBorder):
(accessColor):
(computeTau):
(reduce):
(texelGathering):
(computeColorFromLevel):
(texelFiltering):
(castToInnerTypeForGather):
(gatherTexture):
(castToInnerType):
(sampleTexture):
* WebGPUShadingLanguageRI/index.html:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@235635 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoRemoving an accidentally committed debugging print.
ap@apple.com [Tue, 4 Sep 2018 21:15:43 +0000 (21:15 +0000)]
Removing an accidentally committed debugging print.

* Scripts/run-leaks: (runLeaks)

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@235634 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agorun-leaks should run leaks with --list (on Mojave)
ap@apple.com [Tue, 4 Sep 2018 21:06:31 +0000 (21:06 +0000)]
run-leaks should run leaks with --list (on Mojave)
https://bugs.webkit.org/show_bug.cgi?id=187716
<rdar://problem/42261676>

Reviewed by Lucas Forschler.

Also enabled dumping memgraphs. We'll be pruning these aggressively, as they take
significant space.

* Scripts/run-leaks:
(main): Added an option to store memgraphs.
(runLeaks): As there is no way to test whether the new format is supported in advance,
we have to try with --list first, and retry if that fails. Also, made leaks operate
on a memgraph file if we are saving it anyway.

* Scripts/webkitpy/port/leakdetector.py:
(LeakDetector._leaks_args): Pass --memgraph-file to run-leaks.
(LeakDetector.leaks_file_name): Removed an incorrect comment.
(LeakDetector.memgraph_file_name): Added.
(LeakDetector.check_for_leaks): Changed how arguments are passed to _leaks_args.
It is a bit ugly that leaks path ends up being computed twice, but this is the least
ugly approach that I could find.

* Scripts/webkitpy/port/leakdetector_unittest.py: Updated for _leaks_args changes.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@235633 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoAdd process pool configuration flag to turn on automatic process pre-warming
cdumez@apple.com [Tue, 4 Sep 2018 20:47:23 +0000 (20:47 +0000)]
Add process pool configuration flag to turn on automatic process pre-warming
https://bugs.webkit.org/show_bug.cgi?id=189263
<rdar://problem/44101941>

Reviewed by Antti Koivisto.

Source/WebKit:

Add process pool configuration flag to turn on automatic process pre-warming and disassociate
it from the warmInitialProcess() SPI.

For now, turning on "process swap on navigation" via experimental features also turns on
automatic process pre-warming.

* UIProcess/API/APIProcessPoolConfiguration.cpp:
(API::ProcessPoolConfiguration::copy):
* UIProcess/API/APIProcessPoolConfiguration.h:
* UIProcess/API/C/WKContext.cpp:
(WKContextSetPrewarmsProcessesAutomatically):
(WKContextWarmInitialProcess):
* UIProcess/API/C/WKContextPrivate.h:
* UIProcess/API/Cocoa/WKProcessPool.mm:
(-[WKProcessPool _warmInitialProcess]):
(-[WKProcessPool _hasPrewarmedWebProcess]):
(-[WKProcessPool _webProcessCountIgnoringPrewarmed]):
* UIProcess/API/Cocoa/WKProcessPoolPrivate.h:
* UIProcess/API/Cocoa/_WKProcessPoolConfiguration.h:
* UIProcess/API/Cocoa/_WKProcessPoolConfiguration.mm:
(-[_WKProcessPoolConfiguration setPrewarmsProcessesAutomatically:]):
(-[_WKProcessPoolConfiguration prewarmsProcessesAutomatically]):
* UIProcess/ServiceWorkerProcessProxy.cpp:
(WebKit::ServiceWorkerProcessProxy::ServiceWorkerProcessProxy):
* UIProcess/WebProcessPool.cpp:
(WebKit::WebProcessPool::setMaximumNumberOfProcesses):
(WebKit::WebProcessPool::createNewWebProcess):
(WebKit::WebProcessPool::tryTakePrewarmedProcess):
(WebKit::WebProcessPool::prewarmProcess):
(WebKit::WebProcessPool::disconnectProcess):
(WebKit::WebProcessPool::createWebPage):
(WebKit::WebProcessPool::didReachGoodTimeToPrewarm):
* UIProcess/WebProcessPool.h:
(WebKit::WebProcessPool::sendToOneProcess):
* UIProcess/WebProcessProxy.cpp:
(WebKit::WebProcessProxy::create):
(WebKit::WebProcessProxy::WebProcessProxy):
(WebKit::m_isInPrewarmedPool):
* UIProcess/WebProcessProxy.h:

Tools:

Add API test coverage.

* TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj:
* TestWebKitAPI/Tests/WebKitCocoa/ProcessPreWarming.mm: Renamed from Tools/TestWebKitAPI/Tests/WebKitCocoa/InitialWarmedProcessUsed.mm.
(TEST):
* TestWebKitAPI/Tests/WebKitCocoa/ProcessSwapOnNavigation.mm:
* TestWebKitAPI/Tests/WebKitCocoa/SetMaximumPrewarmedProcessCount.mm: Removed.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@235632 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoAdjust XMLHttpRequest username/password precedence rules
commit-queue@webkit.org [Tue, 4 Sep 2018 20:37:29 +0000 (20:37 +0000)]
Adjust XMLHttpRequest username/password precedence rules
https://bugs.webkit.org/show_bug.cgi?id=184910

Patch by Rob Buis <rbuis@igalia.com> on 2018-09-04
Reviewed by Chris Dumez.

LayoutTests/imported/w3c:

Update test result.

* web-platform-tests/xhr/send-authentication-competing-names-passwords-expected.txt:

Source/WebCore:

Steps 9.1 and 9.2 in the XMLHTTPRequest::open [1] algorithm
specify that non null user or non null password ought
to be set on the URL, so implement this.

Behavior matches Firefox and Chrome.

[1] https://xhr.spec.whatwg.org/#dom-xmlhttprequest-open

Test: xhr/send-authentication-competing-names-passwords.htm

* xml/XMLHttpRequest.cpp:
(WebCore::XMLHttpRequest::open):

LayoutTests:

Adjust test because now we do set password on the url in
open(), even if the username is null.

* http/tests/xmlhttprequest/basic-auth.html:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@235631 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoSimplify RenderLayer filter code
simon.fraser@apple.com [Tue, 4 Sep 2018 20:27:09 +0000 (20:27 +0000)]
Simplify RenderLayer filter code
https://bugs.webkit.org/show_bug.cgi?id=189248

Reviewed by Dean Jackson.

Painting filters in RenderLayer involved the confusingly named RenderLayer::FilterInfo,
and FilterEffectRenderer/FilterEffectRendererHelper. Reduce confusion by just collapsing
these together into RenderLayerFilters. RenderLayerFilters stores a reference to the
CSSFilter, and knows how to set it up, and vend a GraphicsContext* for painting the content
to be filtered.

It's also simpler to just give RenderLayer a member variable for RenderLayerFilters, rather
than using a side hash (we can recoup the space in other ways layer).

* Sources.txt:
* WebCore.xcodeproj/project.pbxproj:
* css/CSSFilterImageValue.cpp:
* rendering/CSSFilter.h:
* rendering/FilterEffectRenderer.cpp: Removed.
* rendering/FilterEffectRenderer.h: Removed.
* rendering/RenderLayer.cpp:
(WebCore::RenderLayer::RenderLayer):
(WebCore::RenderLayer::~RenderLayer):
(WebCore::RenderLayer::paintsWithFilters const):
(WebCore::RenderLayer::requiresFullLayerImageForFilters const):
(WebCore::RenderLayer::setFilterBackendNeedsRepaintingInRect):
(WebCore::RenderLayer::filtersForPainting const):
(WebCore::RenderLayer::setupFilters):
(WebCore::RenderLayer::applyFilters):
(WebCore::RenderLayer::paintLayerContents):
(WebCore::RenderLayer::calculateClipRects const):
(WebCore::RenderLayer::filter const): Deleted.
(WebCore::RenderLayer::filterPainter const): Deleted.
(WebCore::RenderLayer::hasFilterThatIsPainting const): Deleted.
* rendering/RenderLayer.h:
* rendering/RenderLayerFilterInfo.cpp: Removed.
* rendering/RenderLayerFilters.cpp: Added.
(WebCore::RenderLayerFilters::RenderLayerFilters):
(WebCore::RenderLayerFilters::~RenderLayerFilters):
(WebCore::RenderLayerFilters::setFilter):
(WebCore::RenderLayerFilters::hasFilterThatMovesPixels const):
(WebCore::RenderLayerFilters::hasFilterThatShouldBeRestrictedBySecurityOrigin const):
(WebCore::RenderLayerFilters::notifyFinished):
(WebCore::RenderLayerFilters::updateReferenceFilterClients):
(WebCore::RenderLayerFilters::removeReferenceFilterClients):
(WebCore::RenderLayerFilters::buildFilter):
(WebCore::RenderLayerFilters::beginFilterEffect):
(WebCore::RenderLayerFilters::applyFilterEffect):
* rendering/RenderLayerFilters.h: Renamed from Source/WebCore/rendering/RenderLayerFilterInfo.h.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@235630 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoBuild fix.
dino@apple.com [Tue, 4 Sep 2018 20:26:27 +0000 (20:26 +0000)]
Build fix.

* testing/Internals.cpp:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@235629 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoPerl doesn't like calling keys on a reference
keith_miller@apple.com [Tue, 4 Sep 2018 20:16:55 +0000 (20:16 +0000)]
Perl doesn't like calling keys on a reference
https://bugs.webkit.org/show_bug.cgi?id=189261

Reviewed by Michael Saboff.

* Scripts/test262/Runner.pm:
(UpdateResults):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@235628 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoMake the jsc shell print, printErr, and debug functions more robust.
mark.lam@apple.com [Tue, 4 Sep 2018 19:48:39 +0000 (19:48 +0000)]
Make the jsc shell print, printErr, and debug functions more robust.
https://bugs.webkit.org/show_bug.cgi?id=189268
<rdar://problem/41192690>

Reviewed by Keith Miller.

We'll now check for UTF8 conversion errors.

* jsc.cpp:
(cStringFromViewWithString):
(printInternal):
(functionDebug):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@235627 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoMove SystemPreview code from WebKitAdditions to WebKit
dino@apple.com [Tue, 4 Sep 2018 19:38:33 +0000 (19:38 +0000)]
Move SystemPreview code from WebKitAdditions to WebKit
https://bugs.webkit.org/show_bug.cgi?id=189252
<rdar://problem/44080245>

Reviewed by Wenson Hsieh.

Source/WebCore:

Move the WebKitAdditions code into WebKit/WebCore.

* html/HTMLAnchorElement.cpp:
(WebCore::HTMLAnchorElement::relList const): Look for "ar".
(WebCore::HTMLAnchorElement::isSystemPreviewLink const): Ditto.

* platform/MIMETypeRegistry.cpp: Add a couple of new static methods
for System Preview MIME types.
(WebCore::initializeSystemPreviewMIMETypes):
(WebCore::MIMETypeRegistry::getSystemPreviewMIMETypes):
(WebCore::MIMETypeRegistry::isSystemPreviewMIMEType):
* platform/MIMETypeRegistry.h:

* rendering/RenderThemeIOS.mm: Load the ARKit artwork from the framework.
(WebCore::arKitBundle):
(WebCore::loadARKitPDFPage):
(WebCore::systemPreviewLogo):
(WebCore::RenderThemeIOS::paintSystemPreviewBadge):

* testing/Internals.cpp: Remove internal helper.
(WebCore::Internals::systemPreviewRelType): Deleted.
* testing/Internals.h:
* testing/Internals.idl:

Source/WebKit:

Move the WebKitAdditions code into WebKit/WebCore.

* UIProcess/Cocoa/SystemPreviewControllerCocoa.mm:
(getUTIForMIMEType): Helper to get the system UTI for USDZ files.
(-[_WKPreviewControllerDataSource previewController:previewItemAtIndex:]):

* UIProcess/Cocoa/WKWebViewContentProviderRegistry.mm: Use the new methods
in MIMETypeRegistry.
(-[WKWebViewContentProviderRegistry initWithConfiguration:]):

* UIProcess/SystemPreviewController.cpp: Ditto.
(WebKit::SystemPreviewController::canPreview const):

* UIProcess/ios/WKSystemPreviewView.mm:
(getUTIForMIMEType): Similar helper to above.
(-[WKSystemPreviewView web_setContentProviderData:suggestedFilename:]):

LayoutTests:

This test no longer needs an internal helper.

* system-preview/detection.html:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@235626 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years ago[Payment Request] PaymentResponse should have an onpayerdetailchange event handler
aestes@apple.com [Tue, 4 Sep 2018 19:14:23 +0000 (19:14 +0000)]
[Payment Request] PaymentResponse should have an onpayerdetailchange event handler
https://bugs.webkit.org/show_bug.cgi?id=189249

Reviewed by Alex Christensen.

LayoutTests/imported/w3c:

* web-platform-tests/payment-request/payment-response/onpayerdetailchange-attribute.https-expected.txt:

Source/WebCore:

Implemented the onpayerdetailchange event handler as defined in the Payment Request API W3C Editor's Draft of 30 August 2018.

The "payer detail changed" algorithm will be implemented in a follow-up patch.

Covered by existing web platform tests.

* Modules/paymentrequest/PaymentResponse.cpp:
(WebCore::PaymentResponse::scriptExecutionContext const):
* Modules/paymentrequest/PaymentResponse.h:
* Modules/paymentrequest/PaymentResponse.idl:
* dom/EventNames.h:
* dom/EventTargetFactory.in:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@235625 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoSkip Test on 32-bit platforms.
david_fenton@apple.com [Tue, 4 Sep 2018 18:53:47 +0000 (18:53 +0000)]
Skip Test on 32-bit platforms.
https://bugs.webkit.org/show_bug.cgi?id=189185.

Unreviewed test gardening.

* stress/regress-189185.js:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@235624 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoIntroduce LibWebRTC backends for sender and receiver
youenn@apple.com [Tue, 4 Sep 2018 18:48:38 +0000 (18:48 +0000)]
Introduce LibWebRTC backends for sender and receiver
https://bugs.webkit.org/show_bug.cgi?id=189171

Reviewed by Alejandro G. Castro.

Rename RTCRtpSender::Backend to RTCRtpSenderBackend, ditto for RTCRtpReceiver::Backend.
Make RTCRtpSender/RTCRtpReceiver own their backend.
This will allow the backends to own a libwebrtc sender/receiver ref counted object
and might allow us to cleanly separate code from unified plan vs plan B.

Update code so that libwebrtc code specific code is now creating senders and receivers.
This moves code from RTCPeerConnection down to LibWebRTCPeerConnectionBackend, in particular for
addTrack and addTransceiver methods.

Moved some code from LibWebRTCMediaEndpoint to LibWebRTCUtils to ease readability.

A future patch will allow to tie the libwebrtc sender/receiver to WebKit DOM sender/receiver.

Covered by existing tests.

* Modules/mediastream/PeerConnectionBackend.cpp:
(WebCore::PeerConnectionBackend::addTrack):
(WebCore::PeerConnectionBackend::addTransceiver):
* Modules/mediastream/PeerConnectionBackend.h:
* Modules/mediastream/RTCPeerConnection.cpp:
(WebCore::RTCPeerConnection::addTrack):
(WebCore::RTCPeerConnection::addTransceiver):
(WebCore::RTCPeerConnection::enqueueReplaceTrackTask):
* Modules/mediastream/RTCPeerConnection.h:
* Modules/mediastream/RTCRtpReceiver.cpp:
(WebCore::RTCRtpReceiver::RTCRtpReceiver):
* Modules/mediastream/RTCRtpReceiver.h:
(WebCore::RTCRtpReceiver::create):
(WebCore::RTCRtpReceiver::setBackend):
* Modules/mediastream/RTCRtpReceiverBackend.h: Added.
(WebCore::RTCRtpReceiverBackend::getParameters):
* Modules/mediastream/RTCRtpSender.cpp:
(WebCore::RTCRtpSender::create):
(WebCore::RTCRtpSender::RTCRtpSender):
* Modules/mediastream/RTCRtpSender.h:
* Modules/mediastream/RTCRtpSenderBackend.h: Added.
* Modules/mediastream/libwebrtc/LibWebRTCMediaEndpoint.cpp:
(WebCore::LibWebRTCMediaEndpoint::addRemoteTrack):
* Modules/mediastream/libwebrtc/LibWebRTCPeerConnectionBackend.cpp:
(WebCore::LibWebRTCPeerConnectionBackend::videoReceiver):
(WebCore::LibWebRTCPeerConnectionBackend::audioReceiver):
(WebCore::LibWebRTCPeerConnectionBackend::addTrack):
(WebCore::LibWebRTCPeerConnectionBackend::addTransceiver):
(WebCore::LibWebRTCPeerConnectionBackend::completeAddTransceiver):
* Modules/mediastream/libwebrtc/LibWebRTCPeerConnectionBackend.h:
* Modules/mediastream/libwebrtc/LibWebRTCRtpReceiverBackend.cpp: Added.
(WebCore::LibWebRTCRtpReceiverBackend::getParameters):
* Modules/mediastream/libwebrtc/LibWebRTCRtpReceiverBackend.h: Added.
* Modules/mediastream/libwebrtc/LibWebRTCRtpSenderBackend.cpp: Added.
(WebCore::LibWebRTCRtpSenderBackend::replaceTrack):
(WebCore::LibWebRTCRtpSenderBackend::getParameters const):
* Modules/mediastream/libwebrtc/LibWebRTCRtpSenderBackend.h: Added.
* Modules/mediastream/libwebrtc/LibWebRTCUtils.cpp: Added.
(WebCore::fillEncodingParameters):
(WebCore::fillHeaderExtensionParameters):
(WebCore::fillCodecParameters):
(WebCore::fillRtpParameters):
* Modules/mediastream/libwebrtc/LibWebRTCUtils.h: Added.
(WebCore::fromStdString):
* WebCore.xcodeproj/project.pbxproj:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@235623 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years ago[Modern Media Controls] Disabling both fullscreen and picture-in-picture shows an...
graouts@webkit.org [Tue, 4 Sep 2018 18:40:12 +0000 (18:40 +0000)]
[Modern Media Controls] Disabling both fullscreen and picture-in-picture shows an empty top left container
https://bugs.webkit.org/show_bug.cgi?id=189259
<rdar://problem/42026625>

Reviewed by Dean Jackson.

Source/WebCore:

Test: media/modern-media-controls/macos-inline-media-controls/macos-inline-media-controls-disabled-fullscreen-and-pip.html

Ensure the top left controls bar is hidden when none of its children are visible.

* Modules/modern-media-controls/controls/inline-media-controls.js:
(InlineMediaControls.prototype.layout):

LayoutTests:

Adding a new test that checks that the top left container is hidden when both the fullscreen and picture-in-picture buttons are disabled.

* media/modern-media-controls/macos-inline-media-controls/macos-inline-media-controls-disabled-fullscreen-and-pip-expected.txt: Added.
* media/modern-media-controls/macos-inline-media-controls/macos-inline-media-controls-disabled-fullscreen-and-pip.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@235622 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoAttempt to fix failing tests following r235615 (https://bugs.webkit.org/show_bug...
dbates@webkit.org [Tue, 4 Sep 2018 18:20:04 +0000 (18:20 +0000)]
Attempt to fix failing tests following r235615 (https://bugs.webkit.org/show_bug.cgi?id=187925)

An inline text box that does not have combined text and contains a single character
should be considered as having text content.

* rendering/InlineTextBox.cpp:
(WebCore::InlineTextBox::hasTextContent const):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@235621 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoUnreviewed, attempt to fix an internal build after r235585.
ryanhaddad@apple.com [Tue, 4 Sep 2018 18:10:52 +0000 (18:10 +0000)]
Unreviewed, attempt to fix an internal build after r235585.

* pal/spi/cf/CFNetworkSPI.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@235620 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoAdd test for fix in https://bugs.webkit.org/show_bug.cgi?id=187922
pvollan@apple.com [Tue, 4 Sep 2018 18:03:10 +0000 (18:03 +0000)]
Add test for fix in https://bugs.webkit.org/show_bug.cgi?id=187922
https://bugs.webkit.org/show_bug.cgi?id=187925

Reviewed by Simon Fraser.

This will test that the function primaryScreenDisplayID() returns a valid display ID.

Source/WebCore:

Test: fast/misc/valid-primary-screen-displayID.html

* platform/PlatformScreen.h:
* platform/mac/PlatformScreenMac.mm:
(WebCore::primaryScreenDisplayID):
* testing/Internals.cpp:
(WebCore::Internals::primaryScreenDisplayID):
* testing/Internals.h:
* testing/Internals.idl:

LayoutTests:

* TestExpectations:
* fast/misc/valid-primary-screen-displayID-expected.txt: Added.
* fast/misc/valid-primary-screen-displayID.html: Added.
* platform/mac-wk2/TestExpectations:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@235619 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoBug 189190 - REGRESSION(r235398) ASSERTION failure !m_client.didFinishDocumentLoadFor...
commit-queue@webkit.org [Tue, 4 Sep 2018 17:28:28 +0000 (17:28 +0000)]
Bug 189190 - REGRESSION(r235398) ASSERTION failure !m_client.didFinishDocumentLoadForFrame
https://bugs.webkit.org/show_bug.cgi?id=189190

Patch by Frederic Wang <fwang@igalia.com> on 2018-09-04
Reviewed by Alex Christensen.

Assertions to prevent use of the deprecated WKPageSetPageLoaderClient class are hit when
running system Safari with the current macOS/iOS releases, making impossible to launch it.
This patch conditionally skips these assertions so that they are only verified for
macOS > 10.14 or other platforms.

* UIProcess/API/C/WKPage.cpp:
(WKPageSetPageLoaderClient): Only assert on future OS releases.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@235618 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoAdd helper function to create a potential CORS request
dbates@webkit.org [Tue, 4 Sep 2018 17:11:03 +0000 (17:11 +0000)]
Add helper function to create a potential CORS request
https://bugs.webkit.org/show_bug.cgi?id=189251

Reviewed by Andy Estes.

Add a new function, createPotentialAccessControlRequest, that we will use to implement the algorithm Create a potential-
CORS request from the HTML standard: <https://html.spec.whatwg.org/multipage/urls-and-fetching.html#create-a-potential-cors-request> (31 August 2018).
This function replaces CachedResourceRequest::setAsPotentiallyCrossOrigin() and is the first step towards separating
the concepts of CORS settings states and module script credentials mode as well as implementing the aforementioned
algorithm. Rename CachedResourceRequest::setAsPotentiallyCrossOrigin() to deprecatedSetAsPotentiallyCrossOrigin()
and switch existing callers to use createPotentialAccessControlRequest(). For now, createPotentialAccessControlRequest()
is implemented in terms of deprecatedSetAsPotentiallyCrossOrigin().

No functionality changed. So, no new tests.

* bindings/js/CachedScriptFetcher.cpp:
(WebCore::CachedScriptFetcher::requestScriptWithCache const): Write in terms of WebCore::createPotentialAccessControlRequest().
* html/HTMLLinkElement.cpp:
(WebCore::HTMLLinkElement::process): Ditto.
* html/parser/HTMLResourcePreloader.cpp:
(WebCore::PreloadRequest::resourceRequest): Ditto.
* loader/CrossOriginAccessControl.cpp:
(WebCore::createPotentialAccessControlRequest): For now, implemented in terms of CachedResourceRequest::deprecatedSetAsPotentiallyCrossOrigin().
* loader/CrossOriginAccessControl.h:
* loader/ImageLoader.cpp:
(WebCore::ImageLoader::updateFromElement): Write in terms of WebCore::createPotentialAccessControlRequest().
* loader/LinkLoader.cpp:
(WebCore::LinkLoader::preloadIfNeeded): Ditto.
* loader/MediaResourceLoader.cpp:
(WebCore::MediaResourceLoader::requestResource): Ditto. Also renamed local variable cacheRequest to cachedRequest.
* loader/TextTrackLoader.cpp:
(WebCore::TextTrackLoader::load): Write in terms of WebCore::createPotentialAccessControlRequest(). Also change local variable
document from a pointer to a reference since this function asserts that the script execution context is a non-null Document.
* loader/cache/CachedResourceRequest.cpp:
(WebCore::CachedResourceRequest::deprecatedSetAsPotentiallyCrossOrigin): Renamed; formerly named setAsPotentiallyCrossOrigin.
(WebCore::CachedResourceRequest::setAsPotentiallyCrossOrigin): Deleted.
* loader/cache/CachedResourceRequest.h:
(WebCore::CachedResourceRequest::setPriority): Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@235617 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoClean up WebGL policy getting code path
commit-queue@webkit.org [Tue, 4 Sep 2018 17:06:52 +0000 (17:06 +0000)]
Clean up WebGL policy getting code path
https://bugs.webkit.org/show_bug.cgi?id=189003

Patch by Alex Christensen <achristensen@webkit.org> on 2018-09-04
Reviewed by Darin Adler.

* UIProcess/API/APINavigationClient.h:
(API::NavigationClient::webGLLoadPolicy const):
(API::NavigationClient::resolveWebGLLoadPolicy const):
* UIProcess/Cocoa/NavigationState.h:
* UIProcess/Cocoa/NavigationState.mm:
(WebKit::NavigationState::NavigationClient::webGLLoadPolicy const):
(WebKit::NavigationState::NavigationClient::resolveWebGLLoadPolicy const):
* WebProcess/WebPage/WebPage.cpp:
* WebProcess/WebPage/WebPage.h:
* WebProcess/WebPage/ios/WebPageIOS.mm:
(WebKit::WebPage::webGLPolicyForURL): Deleted.
(WebKit::WebPage::resolveWebGLPolicyForURL): Deleted.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@235616 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoRemove redundant inline text boxes for empty combined text
dbates@webkit.org [Tue, 4 Sep 2018 17:02:33 +0000 (17:02 +0000)]
Remove redundant inline text boxes for empty combined text
https://bugs.webkit.org/show_bug.cgi?id=189119

Reviewed by Zalan Bujtas.

Source/WebCore:

We should consider inline text boxes that have a combined text renderer (RenderCombineText)
whose composed string is empty as "redundant" just as we do for inline text boxes that have
a non-combined text renderer that have zero length so that we remove them. Such boxes are
visibly empty and do not take up space visually. By removing them we reduce memory and make
it easier to reason about the line box tree.

Currently RenderBlockFlow::computeBlockDirectionPositionsForLine() tests if an inline text
box is empty by checking if it has a zero length (InlineTextBox::len()). However an inline
text box associated with a RenderCombineText always has length 1 regardless of whether the
composed string it represents is the empty string. Instead we should expose a way to check
if an inline text box is visually empty and have RenderBlockFlow::computeBlockDirectionPositionsForLine()
query the inline text box for this answer.

* rendering/InlineTextBox.cpp:
(WebCore::InlineTextBox::hasTextContent const): Added. Returns whether an inline text box
has text content. We do not need to consider hypenation since hypens are an embellishment (i.e.
they are not part of the markup of the page).
(WebCore::InlineTextBox::paint): Write in terms of hasTextContent().
(WebCore::InlineTextBox::subdivideAndResolveStyle): Assert that WebCore::subdivide() always
returns a non-empty list of subdivisions. A non-empty text box should always have at least
one subdivision, say for the unmarked text. I left the existing conditonal (though marked
it as UNLIKELY()) so as to be forgiving and avoid a bad user experience should WebCore::subdivide()
return an empty vector in a non-debug build.
* rendering/InlineTextBox.h:
* rendering/RenderBlockLineLayout.cpp:
(WebCore::RenderBlockFlow::computeBlockDirectionPositionsForLine): Write in terms of InlineTextBox::hasTextContent()
so that we remove empty inline text boxes associated with combined text.
* rendering/RenderText.cpp:
(WebCore::RenderText::positionLineBox): Write in terms of InlineTextBox::hasTextContent().

LayoutTests:

Update expected result now that we do not create an inline text box associated with combined text
when we do not have any combined text to render.

* fast/text/text-combine-surroundContents-crash-expected.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@235615 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoREGRESSION (r234081): TestWebKitAPI.VideoControlsManager.VideoControlsManagerAudioEle...
jer.noble@apple.com [Tue, 4 Sep 2018 17:01:55 +0000 (17:01 +0000)]
REGRESSION (r234081): TestWebKitAPI.VideoControlsManager.VideoControlsManagerAudioElementFollowingUserInteraction is a flaky timeout
https://bugs.webkit.org/show_bug.cgi?id=187972
<rdar://problem/42667737>

Reviewed by Darin Adler.

Ensure that listeners are added in time to catch messages broadcast at them.

* TestWebKitAPI/Tests/WebKitCocoa/VideoControlsManager.mm:
(TestWebKitAPI::TEST):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@235614 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoUpdate googletest
ross.kirsling@sony.com [Tue, 4 Sep 2018 16:55:46 +0000 (16:55 +0000)]
Update googletest
https://bugs.webkit.org/show_bug.cgi?id=175722

Reviewed by Brent Fulgham.

Source/ThirdParty:

* gtest/:
Update to 1.8.1.

* gtest/CMakeLists.txt:
* gtest/xcode/:
Keep our versions of these build files, making any necessary additive updates
and removing WTF/JSC dependencies so that we stop forking third-party source.

Tools:

* TestWebKitAPI/config.h:
Instead of modifying the gtest source, add the necessary WTF includes here.

* TestWebKitAPI/Configurations/Base.xcconfig
Remove obsoleted definitions GTEST_HAS_TR1_TUPLE=0 and GTEST_HAS_RTTI=0.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@235613 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years ago[WPE][GTK] Add more unused result warnings to JSC API
mcatanzaro@igalia.com [Tue, 4 Sep 2018 14:48:55 +0000 (14:48 +0000)]
[WPE][GTK] Add more unused result warnings to JSC API
https://bugs.webkit.org/show_bug.cgi?id=189243

Reviewed by Carlos Garcia Campos.

The jsc_context_evaluate() family of functions has a (transfer full) return value, but the
caller may be tempted to not inspect it if uninterested in the return value. This would be
an error, because it must be freed.

* API/glib/JSCContext.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@235612 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoImplement support for passing ImageBitmap to texImage2D/texSubImage2D
Ms2ger@igalia.com [Tue, 4 Sep 2018 12:10:34 +0000 (12:10 +0000)]
Implement support for passing ImageBitmap to texImage2D/texSubImage2D
https://bugs.webkit.org/show_bug.cgi?id=187584

Reviewed by Dean Jackson.

Source/WebCore:

Test: fast/canvas/webgl/gl-teximage-imagebitmap.html

* html/canvas/WebGL2RenderingContext.h: Update union type definition.
* html/canvas/WebGL2RenderingContext.idl: Update union type definition.
* html/canvas/WebGLRenderingContext.idl: Update union type definition.
* html/canvas/WebGLRenderingContextBase.cpp:
(WebCore::WebGLRenderingContextBase::texSubImage2D): implement.
(WebCore::WebGLRenderingContextBase::texImage2D): implement.
* html/canvas/WebGLRenderingContextBase.h: Update union type definition; add to TexFuncValidationSourceType.
* html/canvas/WebGLRenderingContextBase.idl: Remove unused union type definition.

LayoutTests:

* fast/canvas/webgl/gl-teximage-imagebitmap-expected.txt: Added.
* fast/canvas/webgl/gl-teximage-imagebitmap.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@235611 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years ago[EME] Add the WebM initData support in ClearKey CDM
commit-queue@webkit.org [Tue, 4 Sep 2018 10:27:27 +0000 (10:27 +0000)]
[EME] Add the WebM initData support in ClearKey CDM
https://bugs.webkit.org/show_bug.cgi?id=189240

Patch by Yacine Bandou <yacine.bandou_ext@softathome.com> on 2018-09-04
Reviewed by Xabier Rodriguez-Calvar.

Add the "webm" initDataType support in ClearKey CDM.
Read the WebM initData by following the W3C spec https://www.w3.org/TR/eme-initdata-webm/#common-system,
and put it in JSON object format like is specified in https://www.w3.org/TR/encrypted-media/#clear-key-request-format.

Tests: media/encrypted-media/clearKey/clearKey-encrypted-webm-event-mse.html
       media/encrypted-media/clearKey/clearKey-webm-video-playback-mse.html

* platform/encryptedmedia/clearkey/CDMClearKey.cpp:
(WebCore::extractKeyIdFromWebMInitData):
(WebCore::CDMPrivateClearKey::supportsInitDataType const):
(WebCore::CDMPrivateClearKey::supportsInitData const):
(WebCore::CDMInstanceClearKey::requestLicense):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@235610 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years ago[EME] Add layout test for InitData and InitDataType in WebM encrypted event
commit-queue@webkit.org [Tue, 4 Sep 2018 10:27:20 +0000 (10:27 +0000)]
[EME] Add layout test for InitData and InitDataType in WebM encrypted event
https://bugs.webkit.org/show_bug.cgi?id=189196

Patch by Yacine Bandou <yacine.bandou@softathome.com> on 2018-09-04
Reviewed by Xabier Rodriguez-Calvar.

This commit adds a new layoutTest in order to test the reception of the encrypted WebM
event with an expected initData and initDataType fields.

* media/content/encrypted/VideoClearKeyWebM.webm: Added.
    This is a simple encrypted video.
    It has been encrypted with webm_crypt see https://github.com/webmproject/webm-tools.
    #webm_crypt  -i VideoWebM.webm -audio false -video true -video_options base_file=bear.key,content_id="0123456789012345" -o VideoClearKeyWebM.webm
* media/encrypted-media/clearKey/clearKey-encrypted-webm-event-mse-expected.txt: Added.
* media/encrypted-media/clearKey/clearKey-encrypted-webm-event-mse.html: Added.
* media/encrypted-media/medias-enc.js:
* platform/gtk/TestExpectations:
* platform/mac/TestExpectations:
* platform/wpe/TestExpectations:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@235609 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoREGRESSION(r235165): [GStreamer] Major video performance regression
zandobersek@gmail.com [Tue, 4 Sep 2018 08:38:28 +0000 (08:38 +0000)]
REGRESSION(r235165): [GStreamer] Major video performance regression
https://bugs.webkit.org/show_bug.cgi?id=189224

Reviewed by Michael Catanzaro.

* Shared/CoordinatedGraphics/CoordinatedGraphicsScene.cpp:
(WebKit::CoordinatedGraphicsScene::updateSceneState): Drop the early
return used when the scene state wasn't updated through a layer flush,
and instead always run through layer changes and backings. This way we
won't miss proxy objects that need swapping even if none of the layer
states were changed. Reducing the amount of work done here when scene
hasn't been updated might be possible, but it's not a huge bottleneck.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@235608 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years ago[Payment Request] Remove PaymentAddress.languageCode
aestes@apple.com [Tue, 4 Sep 2018 04:25:11 +0000 (04:25 +0000)]
[Payment Request] Remove PaymentAddress.languageCode
https://bugs.webkit.org/show_bug.cgi?id=189254

Reviewed by Sam Weinig.

Source/WebCore:

PaymentAddress.languageCode is marked as "at risk" in the latest Payment Request Editor's
Draft (30 August 2018). Firefox has already removed it, and Chrome is considering it.

There is some compatibility risk in removing this attribute, but considering we never
populated it with a meaningful value, the risk seems very low. If we learn about
compatibility problems, we can address them by evangelising or even reverting this change.

Updated existing tests.

* Modules/applepay/paymentrequest/ApplePayPaymentHandler.cpp:
(WebCore::convert):
* Modules/paymentrequest/PaymentAddress.cpp:
(WebCore::PaymentAddress::PaymentAddress):
* Modules/paymentrequest/PaymentAddress.h:
* Modules/paymentrequest/PaymentAddress.idl:

LayoutTests:

* http/tests/paymentrequest/payment-address-attributes-and-toJSON-method.https.html:
* http/tests/ssl/applepay/ApplePayRequestShippingContact.https-expected.txt:
* http/tests/ssl/applepay/ApplePayRequestShippingContact.https.html:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@235607 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years ago[Payment Request] Add a stub for PaymentResponse.retry
aestes@apple.com [Tue, 4 Sep 2018 03:36:22 +0000 (03:36 +0000)]
[Payment Request] Add a stub for PaymentResponse.retry
https://bugs.webkit.org/show_bug.cgi?id=189253

Reviewed by Sam Weinig.

Implemented the AddressErrors, PayerErrorFields, and PaymentValidationErrors dictionaries as
defined in the Payment Request API W3C Editor's Draft of 30 August 2018.

Also added a stub implementation of PaymentResponse.retry. This method will be implemented
with tests in a follow-up patch.

* CMakeLists.txt:
* DerivedSources.make:
* Modules/paymentrequest/AddressErrors.h: Added.
* Modules/paymentrequest/AddressErrors.idl: Added.
* Modules/paymentrequest/PayerErrorFields.h: Added.
* Modules/paymentrequest/PayerErrorFields.idl: Added.
* Modules/paymentrequest/PaymentResponse.cpp:
(WebCore::PaymentResponse::retry):
* Modules/paymentrequest/PaymentResponse.h:
* Modules/paymentrequest/PaymentResponse.idl:
* Modules/paymentrequest/PaymentValidationErrors.h: Added.
* Modules/paymentrequest/PaymentValidationErrors.idl: Added.
* Sources.txt:
* WebCore.xcodeproj/project.pbxproj:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@235606 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoThe watchdog sometimes fails to terminate a script.
mark.lam@apple.com [Tue, 4 Sep 2018 02:48:13 +0000 (02:48 +0000)]
The watchdog sometimes fails to terminate a script.
https://bugs.webkit.org/show_bug.cgi?id=189227
<rdar://problem/39932857>

Reviewed by Saam Barati.

JSTests:

* stress/regress-189227-watchdog-on-infinite-loop.js: Added.

Source/JavaScriptCore:

Consider the following scenario:

1. We have an infinite loop bytecode sequence as follows:

    [  13] loop_hint
    [  14] check_traps
    [  15] jmp               -2(->13)

2. The VM tiers up from LLInt -> BaselineJIT -> DFG -> FTL.

   Note that op_check_traps is represented as a CheckTraps node in the DFG and FTL.
   When we're not using pollingTraps (JSC_usePollingTraps is false by default),
   we emit no code for CheckTraps, but only record an InvalidationPoint there.

3. The watchdog fires, and invalidates all InvalidationPoints in the FTL CodeBlock.

   InvalidationPoints OSR exits to the next instruction by design.  In this case,
   that means the VM will resumes executing at the op_jmp, which jumps to the
   op_loop_hint opcode.  At the loop_hint, the VM discovers that the function is
   already hot, and attempts to tier up.  It immediately discovers that a replacement
   CodeBlock is available because we still haven't jettisoned the DFG CodeBlock
   nor the FTL CodeBlock that was previously compiled for this function.

   Note that jettisoning a CodeBlock necessarily means the VM will invalidate
   its InvalidationPoints (if the CodeBlock is DFG/FTL).  However, the reverse
   is not true: merely invalidating the InvalidationPoints does not necessarily
   mean that the CodeBlock is jettisoned.

   VMTraps::tryInstallTrapBreakpoints() runs from a separate thread.  Hence,
   it is only safe for it to invalidate a CodeBlock's InvalidationPoints.  It
   is not safe for the CodeBlock to be jettisoned from another thread.  Instead,
   the VMTraps mechanism relies on the script thread running to an op_check_traps
   in the baseline JIT code where it will do the necessary jettisoning of optimized
   CodeBlocks.

Since the op_check_traps never get executed, the VM will perpetually tier up in
the op_loop_hint, OSR exit to the op_jmp, jump to the op_loop_hint, and repeat.
Consequently, the watchdog fails to terminate this script.

In this patch, we fix this by making the DFG BytecodeParser emit an InvalidationPoint
node directly (when the VM is not configured to use polling traps).  This ensures
that the check traps invalidation point will OSR exit to the op_check_traps opcode
in the baseline JIT.

In this patch, we also change VMTraps::tryInstallTrapBreakpoints() to use
CallFrame::unsafeCodeBlock() instead of CallFrame::codeBlock().  This is because
we don't really know if the frame is properly set up.  We're just conservatively
probing the stack.  ASAN does not like this probing.  Using unsafeCodeBlock() here
will suppress the false positive ASAN complaint.

* dfg/DFGByteCodeParser.cpp:
(JSC::DFG::ByteCodeParser::parseBlock):
* dfg/DFGClobberize.h:
(JSC::DFG::clobberize):
* dfg/DFGFixupPhase.cpp:
(JSC::DFG::FixupPhase::fixupNode):
* dfg/DFGPredictionPropagationPhase.cpp:
* dfg/DFGSpeculativeJIT.cpp:
(JSC::DFG::SpeculativeJIT::compileCheckTraps):
* dfg/DFGSpeculativeJIT32_64.cpp:
(JSC::DFG::SpeculativeJIT::compile):
* dfg/DFGSpeculativeJIT64.cpp:
(JSC::DFG::SpeculativeJIT::compile):
* ftl/FTLLowerDFGToB3.cpp:
(JSC::FTL::DFG::LowerDFGToB3::compileNode):
* runtime/VMTraps.cpp:
(JSC::VMTraps::tryInstallTrapBreakpoints):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@235605 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years ago[Payment Request] MerchantValidationEvent should be enabled by the PaymentRequest...
aestes@apple.com [Mon, 3 Sep 2018 23:48:40 +0000 (23:48 +0000)]
[Payment Request] MerchantValidationEvent should be enabled by the PaymentRequest setting
https://bugs.webkit.org/show_bug.cgi?id=189250

Reviewed by Daniel Bates.

* Modules/paymentrequest/MerchantValidationEvent.idl:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@235604 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoCallFrame::unsafeCallee() should use an ASAN suppressed Register::asanUnsafePointer().
mark.lam@apple.com [Mon, 3 Sep 2018 23:29:34 +0000 (23:29 +0000)]
CallFrame::unsafeCallee() should use an ASAN suppressed Register::asanUnsafePointer().
https://bugs.webkit.org/show_bug.cgi?id=189247

Reviewed by Saam Barati.

* interpreter/CallFrame.h:
(JSC::ExecState::unsafeCallee const):
* interpreter/Register.h:
(JSC::Register::asanUnsafePointer const):
(JSC::Register::unsafePayload const):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@235603 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoREGRESSION: Layout Test http/tests/security/bypassing-cors-checks-for-extension-urls...
youenn@apple.com [Mon, 3 Sep 2018 23:17:22 +0000 (23:17 +0000)]
REGRESSION: Layout Test http/tests/security/bypassing-cors-checks-for-extension-urls.html is Flaky
https://bugs.webkit.org/show_bug.cgi?id=187658
<rdar://problem/42306442>

Reviewed by Alexey Proskuryakov.

Source/WebCore:

Test is flaky as a previous test was setting the isRunningUserScripts state on the Page and it was never reset.
This patch moves this state to the topDocument so that it will be reset for every navigation.
Covered by existing test being no longer flaky.

* dom/Document.h:
(WebCore::Document::setAsRunningUserScripts):
(WebCore::Document::isRunningUserScripts const):
* loader/DocumentThreadableLoader.cpp:
(WebCore::DocumentThreadableLoader::DocumentThreadableLoader):
* page/Frame.cpp:
(WebCore::Frame::injectUserScriptImmediately):
* page/Page.h:
(WebCore::Page::setAsRunningUserScripts): Deleted.
(WebCore::Page::isRunningUserScripts const): Deleted.
* testing/Internals.cpp:
(WebCore::Internals::setAsRunningUserScripts):

LayoutTests:

* platform/mac-wk2/TestExpectations:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@235602 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years ago[WPE] Add more WPE bots to dashboard
commit-queue@webkit.org [Mon, 3 Sep 2018 19:18:36 +0000 (19:18 +0000)]
[WPE] Add more WPE bots to dashboard
https://bugs.webkit.org/show_bug.cgi?id=189245

Patch by Pablo Saavedra <psaavedra@igalia.com> on 2018-09-03
Reviewed by Daniel Bates.

There are two missing bots which are not listed in the dashboard for
a while: WPE Linux 64-bit Debug (Build) and WPE Linux 64-bit Debug (Tests)

* BuildSlaveSupport/build.webkit.org-config/public_html/dashboard/Scripts/WebKitBuildbot.js:
(WebKitBuildbot):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@235601 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years ago[Payment Request] Implement the MerchantValidationEvent constructor
aestes@apple.com [Mon, 3 Sep 2018 17:16:00 +0000 (17:16 +0000)]
[Payment Request] Implement the MerchantValidationEvent constructor
https://bugs.webkit.org/show_bug.cgi?id=189235

Reviewed by Daniel Bates.

LayoutTests/imported/w3c:

Updated payment-request web platform tests to 0313d9f.

* resources/TestRepositories:
* resources/import-expectations.json:
* web-platform-tests/payment-request/MerchantValidationEvent/complete-method-manual.https.html: Added.
* web-platform-tests/payment-request/MerchantValidationEvent/constructor.http-expected.txt: Added.
* web-platform-tests/payment-request/MerchantValidationEvent/constructor.http.html: Added.
* web-platform-tests/payment-request/MerchantValidationEvent/constructor.https-expected.txt: Added.
* web-platform-tests/payment-request/MerchantValidationEvent/constructor.https.html: Added.
* web-platform-tests/payment-request/MerchantValidationEvent/w3c-import.log: Added.
* web-platform-tests/payment-request/PaymentRequestUpdateEvent/constructor.http-expected.txt: Added.
* web-platform-tests/payment-request/PaymentRequestUpdateEvent/constructor.http.html: Added.
* web-platform-tests/payment-request/PaymentRequestUpdateEvent/constructor.https-expected.txt: Added.
* web-platform-tests/payment-request/PaymentRequestUpdateEvent/constructor.https.html: Added.
* web-platform-tests/payment-request/PaymentRequestUpdateEvent/updateWith-call-immediate-manual.https.html: Added.
* web-platform-tests/payment-request/PaymentRequestUpdateEvent/updateWith-duplicate-shipping-options-manual.https.html: Added.
* web-platform-tests/payment-request/PaymentRequestUpdateEvent/updateWith-incremental-update-manual.https.html: Added.
* web-platform-tests/payment-request/PaymentRequestUpdateEvent/updateWith-method-abort-update-manual.https.html: Added.
* web-platform-tests/payment-request/PaymentRequestUpdateEvent/updateWith-state-checks-manual.https.html: Added.
* web-platform-tests/payment-request/PaymentRequestUpdateEvent/updatewith-method.https-expected.txt: Added.
* web-platform-tests/payment-request/PaymentRequestUpdateEvent/updatewith-method.https.html: Added.
* web-platform-tests/payment-request/PaymentRequestUpdateEvent/w3c-import.log: Added.
* web-platform-tests/payment-request/allowpaymentrequest/common.sub.js:
(window.onmessage.e.t.step):
(string_appeared_here.forEach):
(window.onmessage): Deleted.
* web-platform-tests/payment-request/allowpaymentrequest/echo-PaymentRequest.html:
* web-platform-tests/payment-request/onmerchantvalidation-attribute.https-expected.txt: Added.
* web-platform-tests/payment-request/onmerchantvalidation-attribute.https.html: Added.
* web-platform-tests/payment-request/w3c-import.log:

Source/WebCore:

Implemented the constructor for MerchantValidationEvent as defined in the Payment Request API W3C Editor's Draft of 30 August 2018.

Updated the payment-request web platform tests to include MerchantValidationEvent tests.

Tests: imported/w3c/web-platform-tests/payment-request/MerchantValidationEvent/constructor.http.html
       imported/w3c/web-platform-tests/payment-request/MerchantValidationEvent/constructor.https.html
       imported/w3c/web-platform-tests/payment-request/PaymentRequestUpdateEvent/constructor.http.html
       imported/w3c/web-platform-tests/payment-request/PaymentRequestUpdateEvent/constructor.https.html
       imported/w3c/web-platform-tests/payment-request/PaymentRequestUpdateEvent/updatewith-method.https.html
       imported/w3c/web-platform-tests/payment-request/onmerchantvalidation-attribute.https.html

* CMakeLists.txt:
* DerivedSources.make:
* Modules/paymentrequest/MerchantValidationEvent.cpp:
(WebCore::MerchantValidationEvent::create):
(WebCore::MerchantValidationEvent::MerchantValidationEvent):
(WebCore::MerchantValidationEvent::complete):
* Modules/paymentrequest/MerchantValidationEvent.h:
* Modules/paymentrequest/MerchantValidationEvent.idl:
* Modules/paymentrequest/MerchantValidationEventInit.h: Added.
* Modules/paymentrequest/MerchantValidationEventInit.idl: Added.
* Modules/paymentrequest/PaymentRequest.idl:
* Sources.txt:
* WebCore.xcodeproj/project.pbxproj:
* bindings/js/WebCoreBuiltinNames.h:

LayoutTests:

Skipped new manual tests.

* platform/ios-wk2/TestExpectations:
* platform/mac-wk2/TestExpectations:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@235600 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoRemove duplicate directory web-platform-tests/cssom-view
fred.wang@free.fr [Mon, 3 Sep 2018 13:41:59 +0000 (13:41 +0000)]
Remove duplicate directory web-platform-tests/cssom-view
https://bugs.webkit.org/show_bug.cgi?id=189241

LayoutTests/imported/w3c:

Patch by Frederic Wang <fwang@igalia.com> on 2018-09-03
Reviewed by Daniel Bates.

cssom-view tests were moved into web-platform-tests/css at r232903 but the old
directory and expectation were left. This patch finishes the migration.

* resources/import-expectations.json:
* web-platform-tests/css/cssom-view/scrollingElement-expected.txt: Removed.
* web-platform-tests/cssom-view/CaretPosition-001-expected.txt: Removed.
* web-platform-tests/cssom-view/CaretPosition-001.html: Removed.
* web-platform-tests/cssom-view/DOMRectList-expected.txt: Removed.
* web-platform-tests/cssom-view/DOMRectList.html: Removed.
* web-platform-tests/cssom-view/HTMLBody-ScrollArea_quirksmode-expected.txt: Removed.
* web-platform-tests/cssom-view/HTMLBody-ScrollArea_quirksmode.html: Removed.
* web-platform-tests/cssom-view/MediaQueryList-001-expected.txt: Removed.
* web-platform-tests/cssom-view/MediaQueryList-001.html: Removed.
* web-platform-tests/cssom-view/MediaQueryList-with-empty-string-expected.txt: Removed.
* web-platform-tests/cssom-view/MediaQueryList-with-empty-string.html: Removed.
* web-platform-tests/cssom-view/OWNERS: Removed.
* web-platform-tests/cssom-view/Screen-pixelDepth-Screen-colorDepth001-expected.txt: Removed.
* web-platform-tests/cssom-view/Screen-pixelDepth-Screen-colorDepth001.html: Removed.
* web-platform-tests/cssom-view/cssom-getBoundingClientRect-001-expected.txt: Removed.
* web-platform-tests/cssom-view/cssom-getBoundingClientRect-001.html: Removed.
* web-platform-tests/cssom-view/cssom-getBoundingClientRect-002-expected.txt: Removed.
* web-platform-tests/cssom-view/cssom-getBoundingClientRect-002.html: Removed.
* web-platform-tests/cssom-view/cssom-getClientRects-002-expected.txt: Removed.
* web-platform-tests/cssom-view/cssom-getClientRects-002.html: Removed.
* web-platform-tests/cssom-view/cssom-getClientRects-expected.txt: Removed.
* web-platform-tests/cssom-view/cssom-getClientRects.html: Removed.
* web-platform-tests/cssom-view/cssom-view-img-attributes-001-expected.txt: Removed.
* web-platform-tests/cssom-view/cssom-view-img-attributes-001.html: Removed.
* web-platform-tests/cssom-view/cssom-view-window-screen-interface-expected.txt: Removed.
* web-platform-tests/cssom-view/cssom-view-window-screen-interface.html: Removed.
* web-platform-tests/cssom-view/cssom-view/media-query-list-interface-expected.txt: Removed.
* web-platform-tests/cssom-view/cssom-view/media-query-list-interface.xht: Removed.
* web-platform-tests/cssom-view/cssom-view/w3c-import.log: Removed.
* web-platform-tests/cssom-view/cssom-view/window-interface-expected.txt: Removed.
* web-platform-tests/cssom-view/cssom-view/window-interface.xht: Removed.
* web-platform-tests/cssom-view/elementFromPoint-001-expected.txt: Removed.
* web-platform-tests/cssom-view/elementFromPoint-001.html: Removed.
* web-platform-tests/cssom-view/elementFromPoint-002-expected.txt: Removed.
* web-platform-tests/cssom-view/elementFromPoint-002.html: Removed.
* web-platform-tests/cssom-view/elementFromPoint-003-expected.txt: Removed.
* web-platform-tests/cssom-view/elementFromPoint-003.html: Removed.
* web-platform-tests/cssom-view/elementFromPoint-expected.txt: Removed.
* web-platform-tests/cssom-view/elementFromPoint.html: Removed.
* web-platform-tests/cssom-view/elementFromPosition-expected.txt: Removed.
* web-platform-tests/cssom-view/elementFromPosition.html: Removed.
* web-platform-tests/cssom-view/elementScroll-expected.txt: Removed.
* web-platform-tests/cssom-view/elementScroll.html: Removed.
* web-platform-tests/cssom-view/elementsFromPoint-expected.txt: Removed.
* web-platform-tests/cssom-view/elementsFromPoint-iframes-expected.txt: Removed.
* web-platform-tests/cssom-view/elementsFromPoint-iframes.html: Removed.
* web-platform-tests/cssom-view/elementsFromPoint-invalid-cases-expected.txt: Removed.
* web-platform-tests/cssom-view/elementsFromPoint-invalid-cases.html: Removed.
* web-platform-tests/cssom-view/elementsFromPoint-shadowroot-expected.txt: Removed.
* web-platform-tests/cssom-view/elementsFromPoint-shadowroot.html: Removed.
* web-platform-tests/cssom-view/elementsFromPoint-simple-expected.txt: Removed.
* web-platform-tests/cssom-view/elementsFromPoint-simple.html: Removed.
* web-platform-tests/cssom-view/elementsFromPoint-svg-expected.txt: Removed.
* web-platform-tests/cssom-view/elementsFromPoint-svg.html: Removed.
* web-platform-tests/cssom-view/elementsFromPoint-table-expected.txt: Removed.
* web-platform-tests/cssom-view/elementsFromPoint-table.html: Removed.
* web-platform-tests/cssom-view/elementsFromPoint.html: Removed.
* web-platform-tests/cssom-view/historical-expected.txt: Removed.
* web-platform-tests/cssom-view/historical.html: Removed.
* web-platform-tests/cssom-view/htmlelement-offset-width-001-expected.txt: Removed.
* web-platform-tests/cssom-view/htmlelement-offset-width-001.html: Removed.
* web-platform-tests/cssom-view/iframe.html: Removed.
* web-platform-tests/cssom-view/interfaces-expected.txt: Removed.
* web-platform-tests/cssom-view/interfaces.html: Removed.
* web-platform-tests/cssom-view/matchMedia-expected.txt: Removed.
* web-platform-tests/cssom-view/matchMedia.xht: Removed.
* web-platform-tests/cssom-view/matchMediaAddListener-expected.txt: Removed.
* web-platform-tests/cssom-view/matchMediaAddListener.html: Removed.
* web-platform-tests/cssom-view/mouseEvent-expected.txt: Removed.
* web-platform-tests/cssom-view/mouseEvent.html: Removed.
* web-platform-tests/cssom-view/negativeMargins-expected.txt: Removed.
* web-platform-tests/cssom-view/negativeMargins.html: Removed.
* web-platform-tests/cssom-view/offsetParent_element_test-expected.txt: Removed.
* web-platform-tests/cssom-view/offsetParent_element_test.html: Removed.
* web-platform-tests/cssom-view/offsetTopLeftInScrollableParent-expected.txt: Removed.
* web-platform-tests/cssom-view/offsetTopLeftInScrollableParent.html: Removed.
* web-platform-tests/cssom-view/resources/elementsFromPoint.js: Removed.
* web-platform-tests/cssom-view/resources/iframe1.html: Removed.
* web-platform-tests/cssom-view/resources/iframe2.html: Removed.
* web-platform-tests/cssom-view/resources/w3c-import.log: Removed.
* web-platform-tests/cssom-view/scrollIntoView-shadow-expected.txt: Removed.
* web-platform-tests/cssom-view/scrollIntoView-shadow.html: Removed.
* web-platform-tests/cssom-view/scrollIntoView-smooth-expected.txt: Removed.
* web-platform-tests/cssom-view/scrollIntoView-smooth.html: Removed.
* web-platform-tests/cssom-view/scrollWidthHeight-expected.txt: Removed.
* web-platform-tests/cssom-view/scrollWidthHeight.xht: Removed.
* web-platform-tests/cssom-view/scrollWidthHeightWhenNotScrollable-expected.txt: Removed.
* web-platform-tests/cssom-view/scrollWidthHeightWhenNotScrollable.xht: Removed.
* web-platform-tests/cssom-view/scrolling-no-browsing-context-expected.txt: Removed.
* web-platform-tests/cssom-view/scrolling-no-browsing-context.html: Removed.
* web-platform-tests/cssom-view/scrolling-quirks-vs-nonquirks-expected.txt: Removed.
* web-platform-tests/cssom-view/scrolling-quirks-vs-nonquirks.html: Removed.
* web-platform-tests/cssom-view/scrollingElement-quirks-dynamic-001-expected.html: Removed.
* web-platform-tests/cssom-view/scrollingElement-quirks-dynamic-001.html: Removed.
* web-platform-tests/cssom-view/scrollingElement-quirks-dynamic-002-expected.html: Removed.
* web-platform-tests/cssom-view/scrollingElement-quirks-dynamic-002.html: Removed.
* web-platform-tests/cssom-view/scrollingElement.html: Removed.
* web-platform-tests/cssom-view/scrollintoview-expected.txt: Removed.
* web-platform-tests/cssom-view/scrollintoview.html: Removed.
* web-platform-tests/cssom-view/support/1x1-green.png: Removed.
* web-platform-tests/cssom-view/support/1x1-lime.png: Removed.
* web-platform-tests/cssom-view/support/1x1-maroon.png: Removed.
* web-platform-tests/cssom-view/support/1x1-navy.png: Removed.
* web-platform-tests/cssom-view/support/1x1-red.png: Removed.
* web-platform-tests/cssom-view/support/1x1-white.png: Removed.
* web-platform-tests/cssom-view/support/60x60-gg-rr.png: Removed.
* web-platform-tests/cssom-view/support/60x60-green.png: Removed.
* web-platform-tests/cssom-view/support/60x60-red.png: Removed.
* web-platform-tests/cssom-view/support/README: Removed.
* web-platform-tests/cssom-view/support/a-green.css: Removed.
* web-platform-tests/cssom-view/support/b-green.css: Removed.
* web-platform-tests/cssom-view/support/c-red.css: Removed.
* web-platform-tests/cssom-view/support/cat.png: Removed.
* web-platform-tests/cssom-view/support/import-green.css: Removed.
* web-platform-tests/cssom-view/support/import-red.css: Removed.
* web-platform-tests/cssom-view/support/pattern-grg-rgr-grg.png: Removed.
* web-platform-tests/cssom-view/support/pattern-grg-rrg-rgg.png: Removed.
* web-platform-tests/cssom-view/support/pattern-rgr-grg-rgr.png: Removed.
* web-platform-tests/cssom-view/support/pattern-tr.png: Removed.
* web-platform-tests/cssom-view/support/ruler-h-50%.png: Removed.
* web-platform-tests/cssom-view/support/ruler-h-50px.png: Removed.
* web-platform-tests/cssom-view/support/ruler-v-100px.png: Removed.
* web-platform-tests/cssom-view/support/ruler-v-50px.png: Removed.
* web-platform-tests/cssom-view/support/square-purple.png: Removed.
* web-platform-tests/cssom-view/support/square-teal.png: Removed.
* web-platform-tests/cssom-view/support/square-white.png: Removed.
* web-platform-tests/cssom-view/support/support/README: Removed.
* web-platform-tests/cssom-view/support/support/swatch-green.png: Removed.
* web-platform-tests/cssom-view/support/support/swatch-red.png: Removed.
* web-platform-tests/cssom-view/support/support/w3c-import.log: Removed.
* web-platform-tests/cssom-view/support/swatch-blue.png: Removed.
* web-platform-tests/cssom-view/support/swatch-green.png: Removed.
* web-platform-tests/cssom-view/support/swatch-lime.png: Removed.
* web-platform-tests/cssom-view/support/swatch-orange.png: Removed.
* web-platform-tests/cssom-view/support/swatch-red.png: Removed.
* web-platform-tests/cssom-view/support/swatch-teal.png: Removed.
* web-platform-tests/cssom-view/support/swatch-white.png: Removed.
* web-platform-tests/cssom-view/support/swatch-yellow.png: Removed.
* web-platform-tests/cssom-view/support/test-bl.png: Removed.
* web-platform-tests/cssom-view/support/test-br.png: Removed.
* web-platform-tests/cssom-view/support/test-inner-half-size.png: Removed.
* web-platform-tests/cssom-view/support/test-outer.png: Removed.
* web-platform-tests/cssom-view/support/test-tl.png: Removed.
* web-platform-tests/cssom-view/support/test-tr.png: Removed.
* web-platform-tests/cssom-view/support/w3c-import.log: Removed.
* web-platform-tests/cssom-view/ttwf-js-cssomview-getclientrects-length-expected.txt: Removed.
* web-platform-tests/cssom-view/ttwf-js-cssomview-getclientrects-length.html: Removed.
* web-platform-tests/cssom-view/w3c-import.log: Removed.
* web-platform-tests/cssom-view/window-screen-height-expected.txt: Removed.
* web-platform-tests/cssom-view/window-screen-height-immutable-expected.txt: Removed.
* web-platform-tests/cssom-view/window-screen-height-immutable.html: Removed.
* web-platform-tests/cssom-view/window-screen-height.html: Removed.
* web-platform-tests/cssom-view/window-screen-width-expected.txt: Removed.
* web-platform-tests/cssom-view/window-screen-width-immutable-expected.txt: Removed.
* web-platform-tests/cssom-view/window-screen-width-immutable.html: Removed.
* web-platform-tests/cssom-view/window-screen-width.html: Removed.

LayoutTests:

Patch by Frederic Wang <fwang@free.fr> on 2018-09-03
Reviewed by Daniel Bates.

cssom-view tests were moved into web-platform-tests/css at r232903 but the old
directory and expectation were left. This patch finishes the migration.

* TestExpectations:
* platform/ios-simulator/imported/w3c/web-platform-tests/cssom-view/Screen-pixelDepth-Screen-colorDepth001-expected.txt: Removed.
* platform/ios-simulator/imported/w3c/web-platform-tests/cssom-view/scrolling-quirks-vs-nonquirks-expected.txt: Removed.
* platform/ios-wk2/imported/w3c/web-platform-tests/cssom-view/elementFromPoint-expected.txt: Removed.
* platform/ios-wk2/imported/w3c/web-platform-tests/cssom-view/scrollIntoView-smooth-expected.txt: Removed.
* platform/ios/TestExpectations:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@235599 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years ago[GStreamer] elements registration clean-ups
philn@webkit.org [Mon, 3 Sep 2018 11:17:07 +0000 (11:17 +0000)]
[GStreamer] elements registration clean-ups
https://bugs.webkit.org/show_bug.cgi?id=189192

Reviewed by Xabier Rodriguez-Calvar.

Register all our elements from the base player class. This
simplifies the MSE player by removing duplicated code.

* platform/graphics/gstreamer/MediaPlayerPrivateGStreamer.h:
* platform/graphics/gstreamer/MediaPlayerPrivateGStreamerBase.cpp:
(WebCore::MediaPlayerPrivateGStreamerBase::initializeGStreamerAndRegisterWebKitElements):
(WebCore::registerWebKitGStreamerElements): Deleted.
* platform/graphics/gstreamer/mse/MediaPlayerPrivateGStreamerMSE.cpp:
(WebCore::MediaPlayerPrivateGStreamerMSE::registerMediaEngine):
(WebCore::MediaPlayerPrivateGStreamerMSE::load):
(WebCore::mimeTypeCache):
(WebCore::initializeGStreamerAndRegisterWebKitMSEElement): Deleted.
(WebCore::MediaPlayerPrivateGStreamerMSE::isAvailable): Deleted.
* platform/graphics/gstreamer/mse/MediaPlayerPrivateGStreamerMSE.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@235598 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoREGRESSION (r191336): RenderFlexibleBox::adjustChildSizeForMinAndMax crashes in std...
zalan@apple.com [Mon, 3 Sep 2018 03:41:11 +0000 (03:41 +0000)]
REGRESSION (r191336): RenderFlexibleBox::adjustChildSizeForMinAndMax crashes in std::optional<>::value()
https://bugs.webkit.org/show_bug.cgi?id=189232
<rdar://problem/43886373>

Reviewed by Brent Fulgham.

Source/WebCore:

It's not guaranteed that RenderFlexibleBox::computeMainAxisExtentForChild() always returns with a valid value.

Test: fast/flexbox/crash-when-min-max-content-is-not-computed.html

* rendering/RenderFlexibleBox.cpp:
(WebCore::RenderFlexibleBox::adjustChildSizeForMinAndMax):

LayoutTests:

* fast/flexbox/crash-when-min-max-content-is-not-computed-expected.txt: Added.
* fast/flexbox/crash-when-min-max-content-is-not-computed.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@235590 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoImplement Object.fromEntries
yusukesuzuki@slowstart.org [Sun, 2 Sep 2018 16:41:45 +0000 (16:41 +0000)]
Implement Object.fromEntries
https://bugs.webkit.org/show_bug.cgi?id=188481

Reviewed by Darin Adler.

JSTests:

* stress/object-from-entries.js: Added.
(shouldBe):
(shouldThrow):
(shouldBe.JSON.stringify.Object.getOwnPropertyDescriptor):
(shouldBe.set get shouldThrow):

Source/JavaScriptCore:

Object.fromEntries becomes stage 3[1]. This patch implements it by using builtin JS.

[1]: https://tc39.github.io/proposal-object-from-entries/

* builtins/ObjectConstructor.js:
(fromEntries):
* runtime/ObjectConstructor.cpp:

LayoutTests:

* js/Object-getOwnPropertyNames-expected.txt:
* js/script-tests/Object-getOwnPropertyNames.js:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@235589 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years ago[WPE] 2.21.91 fails to build with ENABLE_MINIBROWSER
mcatanzaro@igalia.com [Sun, 2 Sep 2018 05:35:49 +0000 (05:35 +0000)]
[WPE] 2.21.91 fails to build with ENABLE_MINIBROWSER
https://bugs.webkit.org/show_bug.cgi?id=189006

Reviewed by Darin Adler.

The MiniBrowser source code is missing from the tarball. Oops! Add it.

* wpe/manifest.txt.in:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@235588 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years ago[WinCairo] Add CryptoDigestOpenSSL
yoshiaki.jitsukawa@sony.com [Sun, 2 Sep 2018 05:33:52 +0000 (05:33 +0000)]
[WinCairo] Add CryptoDigestOpenSSL
https://bugs.webkit.org/show_bug.cgi?id=188978

Reviewed by Darin Adler.

Source/WebCore/PAL:

Add a CryptoDigest implementation with OpenSSL/LibreSSL and let the
WinCairo port use it.

* pal/PlatformAppleWin.cmake: Add CryptoDigestWin.cpp to PAL_SOURCES
* pal/PlatformWin.cmake: Remove CryptoDigestWin.cpp from PAL_SOURCES
* pal/PlatformWinCairo.cmake: Add CryptoDigestOpenSSL to PAL_SOURCES
* pal/crypto/openssl/CryptoDigestOpenSSL.cpp: Added.
(PAL::toSHA1Context):
(PAL::toSHA224Context):
(PAL::toSHA256Context):
(PAL::toSHA384Context):
(PAL::toSHA512Context):
(PAL::CryptoDigest::CryptoDigest):
(PAL::CryptoDigest::~CryptoDigest):
(PAL::CryptoDigest::create):
(PAL::CryptoDigest::addBytes):
(PAL::CryptoDigest::computeHash):

Tools:

Add API tests for the PAL::CryptoDigest class.

* TestWebKitAPI/PlatformWin.cmake:
Add CryptoDigest.cpp to TestWebCoreLib_Sources.

Define -DSTATICALLY_LINKED_WITH_PAL=1 to avoid warnings since
without the definition the functions marked with PAL_EXPORT are supposed to be
imported from another shared library, but as it is, TestWebCoreLib
links PAL statically.

* TestWebKitAPI/Tests/WebCore/CryptoDigest.cpp: Added.
(TestWebKitAPI::toHex):
(TestWebKitAPI::expect):
(TestWebKitAPI::expectSHA1):
(TestWebKitAPI::expectSHA224):
(TestWebKitAPI::expectSHA256):
(TestWebKitAPI::expectSHA384):
(TestWebKitAPI::expectSHA512):
(TestWebKitAPI::TEST):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@235587 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoRename FilterEffectRenderer to CSSFilter
simon.fraser@apple.com [Sun, 2 Sep 2018 03:04:24 +0000 (03:04 +0000)]
Rename FilterEffectRenderer to CSSFilter
https://bugs.webkit.org/show_bug.cgi?id=189223

Reviewed by Sam Weinig.

Filter was subclassed by SVGFilter for SVG, and FilterEffectRenderer for CSS,
which was very confusing, especially when the code uses renderer() to get it.

Rename FilterEffectRenderer to CSSFilter, and access via filter().

Future cleanup will deal with FilterEffectRendererHelper.

* Sources.txt:
* WebCore.xcodeproj/project.pbxproj:
* css/CSSFilterImageValue.cpp:
(WebCore::CSSFilterImageValue::image):
* platform/graphics/ca/cocoa/PlatformCAFiltersCocoa.mm:
* platform/graphics/filters/Filter.h:
(WebCore::Filter::isCSSFilter const):
* rendering/CSSFilter.cpp: Copied from Source/WebCore/rendering/FilterEffectRenderer.cpp.
(WebCore::endMatrixRow):
(WebCore::lastMatrixRow):
(WebCore::CSSFilter::create):
(WebCore::CSSFilter::CSSFilter):
(WebCore::m_sourceGraphic):
(WebCore::CSSFilter::inputContext):
(WebCore::CSSFilter::buildReferenceFilter):
(WebCore::CSSFilter::build):
(WebCore::CSSFilter::updateBackingStoreRect):
(WebCore::CSSFilter::allocateBackingStoreIfNeeded):
(WebCore::CSSFilter::clearIntermediateResults):
(WebCore::CSSFilter::apply):
(WebCore::CSSFilter::computeSourceImageRectForDirtyRect):
(WebCore::CSSFilter::output const):
(WebCore::CSSFilter::setSourceImageRect):
(WebCore::CSSFilter::setMaxEffectRects):
(WebCore::CSSFilter::outputRect const):
* rendering/CSSFilter.h: Copied from Source/WebCore/rendering/FilterEffectRenderer.h.
(isType):
* rendering/FilterEffectRenderer.cpp:
(WebCore::FilterEffectRendererHelper::prepareFilterEffect):
(WebCore::FilterEffectRendererHelper::filterContext const):
(WebCore::FilterEffectRendererHelper::beginFilterEffect):
(WebCore::FilterEffectRendererHelper::applyFilterEffect):
(WebCore::endMatrixRow): Deleted.
(WebCore::lastMatrixRow): Deleted.
(WebCore::FilterEffectRenderer::FilterEffectRenderer): Deleted.
(WebCore::m_sourceGraphic): Deleted.
(WebCore::FilterEffectRenderer::create): Deleted.
(WebCore::FilterEffectRenderer::inputContext): Deleted.
(WebCore::FilterEffectRenderer::buildReferenceFilter): Deleted.
(WebCore::FilterEffectRenderer::build): Deleted.
(WebCore::FilterEffectRenderer::updateBackingStoreRect): Deleted.
(WebCore::FilterEffectRenderer::allocateBackingStoreIfNeeded): Deleted.
(WebCore::FilterEffectRenderer::clearIntermediateResults): Deleted.
(WebCore::FilterEffectRenderer::apply): Deleted.
(WebCore::FilterEffectRenderer::computeSourceImageRectForDirtyRect): Deleted.
(WebCore::FilterEffectRenderer::output const): Deleted.
(WebCore::FilterEffectRenderer::setMaxEffectRects): Deleted.
(WebCore::FilterEffectRenderer::outputRect const): Deleted.
* rendering/FilterEffectRenderer.h:
(WebCore::FilterEffectRendererHelper::FilterEffectRendererHelper):
(WebCore::FilterEffectRenderer::setSourceImageRect): Deleted.
* rendering/RenderBlockFlow.cpp:
* rendering/RenderFrame.cpp:
* rendering/RenderImage.cpp:
* rendering/RenderLayer.cpp:
(WebCore::RenderLayer::requiresFullLayerImageForFilters const):
(WebCore::RenderLayer::filter const):
(WebCore::RenderLayer::filterPainter const):
(WebCore::RenderLayer::setupFilters):
(WebCore::RenderLayer::calculateClipRects const):
(WebCore::RenderLayer::filterRenderer const): Deleted.
* rendering/RenderLayer.h:
* rendering/RenderLayerBacking.cpp:
* rendering/RenderLayerFilterInfo.cpp:
(WebCore::RenderLayer::FilterInfo::setFilter):
(WebCore::RenderLayer::FilterInfo::updateReferenceFilterClients):
(WebCore::RenderLayer::FilterInfo::removeReferenceFilterClients):
(WebCore::RenderLayer::FilterInfo::setRenderer): Deleted.
* rendering/RenderLayerFilterInfo.h:
* rendering/updating/RenderTreeBuilderBlock.cpp:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@235586 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years ago[CFNetwork] Update CFNetwork SPI use to use CFNetworkSPI.h more consistently
darin@apple.com [Sun, 2 Sep 2018 02:54:58 +0000 (02:54 +0000)]
[CFNetwork] Update CFNetwork SPI use to use CFNetworkSPI.h more consistently
https://bugs.webkit.org/show_bug.cgi?id=189072
Source/WebCore:

Reviewed by Dan Bernstein.

* platform/mac/WebCoreNSURLExtras.mm: Include CFNetworkSPI.h and remove a
declaration of +[NSURLProtocol _protocolClassForRequest:].

* platform/network/mac/ResourceHandleMac.mm: Remove a declaration of
-[NSURLConnection _initWithRequest:delegate:usesCache:maxContentLength:startImmediately:connectionProperties:].
(WebCore::ResourceHandle::createNSURLConnection): Add ignore of
"-Wdeprecated-declarations" around use of the above method, since it
is deprecated.

Source/WebCore/PAL:

Reviewed by Dan Bernstein.

* pal/spi/cf/CFNetworkSPI.h: Added a FIXME about USE(CFNETWORK_IGNORE_HSTS).
Sorted Objective-C class interfaces alphabetically instead of having them
relatively randomly ordered in the file. Reduced the platform #if statements
since it's not critical to have these SPIs declared only where they are
present, as long as they are used only when they are present. Moved some
additional declarations into this file.

* pal/spi/cocoa/NSURLDownloadSPI.h: Added more NSURLDownload methods.

Source/WebKit:

Reviewed by Dan Bernstein.

* NetworkProcess/cocoa/NetworkDataTaskCocoa.mm: Removed NSURLSessionTask
method declarations from this file, they are in CFNetworkSPI.h.

* NetworkProcess/cocoa/NetworkSessionCocoa.mm: Removed NSURLSessionConfiguration
method declarations from this file, they are in CFNetworkSPI.h.

Source/WebKitLegacy/mac:

Reviewed by Dan Bernstein.

* Misc/WebDownload.mm: Removed forward declaration of NSURLConnectionDelegateProxy
and declarations for NSURLDownload init methods.

* Misc/WebDownloadInternal.h: Removed.

* WebCoreSupport/WebFrameLoaderClient.mm: Removed include of WebDownloadInternal.h.
(WebFrameLoaderClient::convertMainResourceLoadToDownload): Ignore
"-Wdeprecated-declarations" because the method used here is deprecated.

* WebView/WebView.mm:
(-[WebView _downloadURL:]): Ignore "-Wdeprecated-declarations" because the method
used here is deprecated.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@235585 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years ago[Cocoa] Crash using KVO for 'serverTrust' property of WKWebView - "this class is...
mitz@apple.com [Sat, 1 Sep 2018 23:14:31 +0000 (23:14 +0000)]
[Cocoa] Crash using KVO for 'serverTrust' property of WKWebView - "this class is not key value coding-compliant for the key serverTrust"
https://bugs.webkit.org/show_bug.cgi?id=189222
<rdar://problem/33283179>

Reviewed by Sam Weinig.

Source/WebKit:

Test: TestWebKitAPI/Tests/WebKitCocoa/WKWebViewServerTrustKVC.mm

* UIProcess/API/Cocoa/WKWebView.mm:
(-[WKWebView valueForUndefinedKey:]): Override to handle the "serverTrust" key, which is not
  handled automatically by the KVC machinery since the getter’s return type is not an
  Objective-C type.

Tools:

* TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj:
* TestWebKitAPI/Tests/WebKitCocoa/WKWebViewServerTrustKVC.mm: Added.
(TEST):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@235584 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoAdd some more filter logging
simon.fraser@apple.com [Sat, 1 Sep 2018 20:16:01 +0000 (20:16 +0000)]
Add some more filter logging
https://bugs.webkit.org/show_bug.cgi?id=189221

Reviewed by Anders Carlsson.

Add logging in FilterEffect::apply() that shows all the filter rects, and logging
that brackets the application of a whole filter chain, for both SVG and CSS filters.

* platform/graphics/filters/FilterEffect.cpp:
(WebCore::FilterEffect::apply):
* rendering/FilterEffectRenderer.cpp:
(WebCore::FilterEffectRendererHelper::applyFilterEffect):
* rendering/svg/RenderSVGResourceFilter.cpp:
(WebCore::RenderSVGResourceFilter::postApplyResource):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@235583 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoFunction object should convert params to string before throw a parsing error
yusukesuzuki@slowstart.org [Sat, 1 Sep 2018 08:03:43 +0000 (08:03 +0000)]
Function object should convert params to string before throw a parsing error
https://bugs.webkit.org/show_bug.cgi?id=188874

Reviewed by Darin Adler.

JSTests:

* stress/function-body-to-string-before-parameter-syntax-check.js: Added.
(shouldThrow):

Source/JavaScriptCore:

ToString operation onto the `body` of the Function constructor should be performed
before checking syntax correctness of the parameters.

* runtime/FunctionConstructor.cpp:
(JSC::constructFunctionSkippingEvalEnabledCheck):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@235582 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoAdd support for USDZ to webkit.org for a sample file
dino@apple.com [Sat, 1 Sep 2018 00:16:57 +0000 (00:16 +0000)]
Add support for USDZ to webkit.org for a sample file
https://bugs.webkit.org/show_bug.cgi?id=189217

Rubber-stamped by Anders Carlsson.

Use the accepted MIME-type for USDZ files, and provide a
simple example file.

* .htaccess: Serve USD with a MIME-type that we will accept.
* demos/ar/heart.usdz: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@235579 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoUnreviewed, add an alternate email for Aditya Keerthi.
akeerthi@apple.com [Sat, 1 Sep 2018 00:02:12 +0000 (00:02 +0000)]
Unreviewed, add an alternate email for Aditya Keerthi.

* Scripts/webkitpy/common/config/contributors.json:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@235578 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoAdded a CSS rule to DarkMode.css to fix color on sidebar toggle.
commit-queue@webkit.org [Fri, 31 Aug 2018 23:28:26 +0000 (23:28 +0000)]
Added a CSS rule to DarkMode.css to fix color on sidebar toggle.
<https://webkit.org/b/189139>

Patch by Jamal Nasser <xnasser@outlook.com> on 2018-08-31
Reviewed by Matt Baker.

* UserInterface/Views/DarkMode.css:
(@media (prefers-dark-interface)):
(.sidebar > .panel.details.css-style > .content ~ .options-container > .toggle-class-toggle):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@235577 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoMove stats gathering out of LibWebRTCMediaEndpoint
youenn@apple.com [Fri, 31 Aug 2018 23:27:20 +0000 (23:27 +0000)]
Move stats gathering out of LibWebRTCMediaEndpoint
https://bugs.webkit.org/show_bug.cgi?id=189180

Reviewed by Alejandro G. Castro.

Move stats gathering in LibWebRTCStatsCollector.
Make sure that the lambda given to the collector is always called and destroyed from the main thread.
This allows capturing the promise here instead of storing it into the peer connection backend.
No change of behavior.

* Modules/mediastream/libwebrtc/LibWebRTCMediaEndpoint.cpp:
(WebCore::LibWebRTCMediaEndpoint::getStats):
* Modules/mediastream/libwebrtc/LibWebRTCMediaEndpoint.h:
* Modules/mediastream/libwebrtc/LibWebRTCPeerConnectionBackend.cpp:
(WebCore::LibWebRTCPeerConnectionBackend::getStats):
(WebCore::LibWebRTCPeerConnectionBackend::doStop):
* Modules/mediastream/libwebrtc/LibWebRTCPeerConnectionBackend.h:
* Modules/mediastream/libwebrtc/LibWebRTCStatsCollector.cpp: Added.
(WebCore::LibWebRTCStatsCollector::LibWebRTCStatsCollector):
(WebCore::LibWebRTCStatsCollector::~LibWebRTCStatsCollector):
(WebCore::fromStdString):
(WebCore::fillRTCStats):
(WebCore::fillRTCRTPStreamStats):
(WebCore::fillInboundRTPStreamStats):
(WebCore::fillOutboundRTPStreamStats):
(WebCore::fillRTCMediaStreamTrackStats):
(WebCore::fillRTCDataChannelStats):
(WebCore::iceCandidatePairState):
(WebCore::fillRTCIceCandidatePairStats):
(WebCore::fillRTCCertificateStats):
(WebCore::LibWebRTCStatsCollector::OnStatsDelivered):
* Modules/mediastream/libwebrtc/LibWebRTCStatsCollector.h: Added.
(WebCore::LibWebRTCStatsCollector::create):
* WebCore.xcodeproj/project.pbxproj:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@235576 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoCompile error in RealtimeOutgoingVideoSource.cpp; unused parameter in libwebrtc header
jer.noble@apple.com [Fri, 31 Aug 2018 23:22:47 +0000 (23:22 +0000)]
Compile error in RealtimeOutgoingVideoSource.cpp; unused parameter in libwebrtc header
https://bugs.webkit.org/show_bug.cgi?id=189203

Reviewed by Youenn Fablet.

If RealtimeOutgoingVideoSource.cpp is compiled alone (non-unified) or if it is the first file compiled in
a unified build, there is no pragma set to ignore unused parameter warnings in place.

* platform/mediastream/RealtimeOutgoingVideoSource.cpp:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@235575 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years ago[iOS] Consolidate the implementations of readString, stringForType, and readURL in...
wenson_hsieh@apple.com [Fri, 31 Aug 2018 23:18:30 +0000 (23:18 +0000)]
[iOS] Consolidate the implementations of readString, stringForType, and readURL in PlatformPasteboardIOS.mm
https://bugs.webkit.org/show_bug.cgi?id=189054
<rdar://problem/43819779>

Reviewed by Tim Horton.

Source/WebCore:

Remove redundant and unnecessary logic for reading from the pasteboard on iOS.

* platform/PlatformPasteboard.h:
* platform/ios/PlatformPasteboardIOS.mm:
(WebCore::PlatformPasteboard::stringForType const):

This currently grabs a string corresponding to the given type from the first item in the pasteboard. Make
stringForType instead call readString with pasteboard item index 0.

(WebCore::PlatformPasteboard::count const):
(WebCore::PlatformPasteboard::readBuffer const):
(WebCore::PlatformPasteboard::readString const):

Make readString with "public.url" call readURL.

(WebCore::PlatformPasteboard::readURL const):

Remove logic for reading URLs from the pasteboard as property lists deserialized from properly lists. This was
added in r223195 due to fix a case "when UIPasteboard serializes NSURL as a plist" when grabbing pasteboard data
using -valuesForPasteboardType:inItemSet:. However, this case only arises in non-UI applications (i.e. when
UIApplicationInitialize() has not been invoked); this is currently exercised by the test CopyURL.ValidURL, but
doesn't really correspond to a real-world use case, since all UI applications where a user would be able to
paste in a web view already invoke UIApplicationInitialize().

Instead of handling the case where the pasteboard contains a property list that has not been coerced to an
NSURL, simply remove the code from PlatformPasteboard::readURL and allow UIKit to perform the coercion when
running the test.

(WebCore::PlatformPasteboard::count): Deleted.
(WebCore::PlatformPasteboard::readBuffer): Deleted.
(WebCore::PlatformPasteboard::readString): Deleted.
(WebCore::PlatformPasteboard::readURL): Deleted.

Mark these functions as `const`.

* platform/wpe/PlatformPasteboardWPE.cpp:
(WebCore::PlatformPasteboard::readString const):
(WebCore::PlatformPasteboard::readString): Deleted.

Mark this function as const.

Tools:

See WebCore/ChangeLog for more detail.

* TestWebKitAPI/Tests/WebKitCocoa/CopyURL.mm:
(createWebViewWithCustomPasteboardDataEnabled):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@235574 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoCompilation error in FormData.cpp: incomplete type 'WebCore::SharedBuffer'
jer.noble@apple.com [Fri, 31 Aug 2018 23:14:06 +0000 (23:14 +0000)]
Compilation error in FormData.cpp: incomplete type 'WebCore::SharedBuffer'
https://bugs.webkit.org/show_bug.cgi?id=189207

Reviewed by Youenn Fablet.

If FormData.cpp is compiled alone (non-unified) or if it is the first file compiled in
a unified build, SharedBuffer.h is not included.

* platform/network/FormData.cpp:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@235573 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoUnreviewed, rolling out r235565.
commit-queue@webkit.org [Fri, 31 Aug 2018 22:50:08 +0000 (22:50 +0000)]
Unreviewed, rolling out r235565.
https://bugs.webkit.org/show_bug.cgi?id=189212

Broke the WHLSL build (Requested by litherum on #webkit).

Reverted changeset:

"[WHLSL] Remove useless code in NameResolver"
https://bugs.webkit.org/show_bug.cgi?id=189176
https://trac.webkit.org/changeset/235565

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@235572 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoClean up TestExpectations so that ./Tools/Scripts/lint-test-expectations
simon.fraser@apple.com [Fri, 31 Aug 2018 22:23:22 +0000 (22:23 +0000)]
Clean up TestExpectations so that ./Tools/Scripts/lint-test-expectations
succeeds now.

* platform/ios-wk2/TestExpectations:
* platform/mac-wk1/TestExpectations:
* platform/win/TestExpectations:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@235571 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years ago[ WK2 ] http/tests/workers/service/client-*-page-cache.html LayoutTests are flaky
cdumez@apple.com [Fri, 31 Aug 2018 21:48:18 +0000 (21:48 +0000)]
[ WK2 ] http/tests/workers/service/client-*-page-cache.html LayoutTests are flaky
https://bugs.webkit.org/show_bug.cgi?id=183705
<rdar://problem/42440606>

Reviewed by Youenn Fablet.

Source/WebCore:

Add internals.serviceWorkerClientIdentifier() utility function so that a layout test can get the
service worker client identifier of a document.

* testing/Internals.cpp:
(WebCore::Internals::serviceWorkerClientIdentifier const):
* testing/Internals.h:
* testing/Internals.idl:

LayoutTests:

Update Layout tests to not rely on the total number of clients as this is flaky. Instead, check for specific client
identifiers to see if they are present or not.

* http/tests/workers/service/client-added-to-clients-when-restored-from-page-cache.html:
* http/tests/workers/service/client-removed-from-clients-while-in-page-cache.html:
* http/tests/workers/service/resources/getClientIds-worker.js: Renamed from LayoutTests/http/tests/workers/service/resources/getClientCount-worker.js.
(event.then):
* http/tests/workers/service/serviceworkerclients-matchAll-worker.js:
(async.doTestAfterMessage):
* http/tests/workers/service/serviceworkerclients-matchAll.https.html:

* platform/ios-wk2/TestExpectations:
* platform/mac-wk2/TestExpectations:
Unskip tests as they should no longer be flaky.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@235570 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoStorage Access API: Maintain access through same-site navigations
wilander@apple.com [Fri, 31 Aug 2018 21:09:36 +0000 (21:09 +0000)]
Storage Access API: Maintain access through same-site navigations
https://bugs.webkit.org/show_bug.cgi?id=188564
<rdar://problem/43445160>

Reviewed by Alex Christensen.

Source/WebCore:

Tests: http/tests/storageAccess/request-and-grant-access-then-navigate-cross-site-should-not-have-access.html
       http/tests/storageAccess/request-and-grant-access-then-navigate-same-site-should-have-access.html

* dom/Document.h:
    Removed unused member variable m_hasFrameSpecificStorageAccess.
* loader/DocumentLoader.cpp:
(WebCore::DocumentLoader::willSendRequest):
* loader/FrameLoaderClient.h:

Source/WebKit:

* WebProcess/WebCoreSupport/WebFrameLoaderClient.cpp:
(WebKit::WebFrameLoaderClient::dispatchWillChangeDocument):
    Now takes the current URL and the new URL as parameters
    and only clears out storage access if the navigation is
    cross-site, i.e. with differing eTLD+1s.
* WebProcess/WebCoreSupport/WebFrameLoaderClient.h:

LayoutTests:

* http/tests/storageAccess/request-and-grant-access-cross-origin-sandboxed-iframe-from-prevalent-domain-with-user-interaction-and-access-from-right-frame-expected.txt:
* http/tests/storageAccess/request-and-grant-access-cross-origin-sandboxed-iframe-from-prevalent-domain-with-user-interaction-and-access-from-right-frame.html:
    Added both a same-site and a cross-site navigation to the sub frame to test maintained and cleared access.
* http/tests/storageAccess/request-and-grant-access-then-navigate-cross-site-should-not-have-access-expected.txt: Copied from LayoutTests/http/tests/storageAccess/request-and-grant-access-then-navigate-should-not-have-access-expected.txt.
* http/tests/storageAccess/request-and-grant-access-then-navigate-cross-site-should-not-have-access.html: Copied from LayoutTests/http/tests/storageAccess/request-and-grant-access-then-navigate-should-not-have-access.html.
* http/tests/storageAccess/request-and-grant-access-then-navigate-same-site-should-have-access-expected.txt: Renamed from LayoutTests/http/tests/storageAccess/request-and-grant-access-then-navigate-should-not-have-access-expected.txt.
* http/tests/storageAccess/request-and-grant-access-then-navigate-same-site-should-have-access.html: Renamed from LayoutTests/http/tests/storageAccess/request-and-grant-access-then-navigate-should-not-have-access.html.
* http/tests/storageAccess/resources/get-cookies.php:
    Support for the new test scenarios.
* http/tests/storageAccess/resources/self-navigating-frame-after-granted-access.html:
    Support for the new test scenarios.
* platform/mac-wk2/TestExpectations:
    Marked the new and the renamed tests as [ Pass ].

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@235569 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoslotchange-event-bubbling.html fails in Chrome & Firefox
rniwa@webkit.org [Fri, 31 Aug 2018 20:41:57 +0000 (20:41 +0000)]
slotchange-event-bubbling.html fails in Chrome & Firefox
https://bugs.webkit.org/show_bug.cgi?id=189183

Reviewed by Chris Dumez.

slotchange event is supposed to fire when a slot element is inserted and there are assigned nodes.

Firefox and Chrome implement this behavior and fail this test since the test doesn't let microstask
check point run before start mutating the DOM.

Fixed the test by manually waiting for the current microtask queue to be emptied after constructing
the test shadow tree so that slotchange event from the initial tree construction doesn't get merged
with slotchange event fired for the actual DOM mutation being tested.

Also fixed typos in slotchange-in-fallback.html and a bug that we were removing the slot instead
of the shadow host at the end of each test case.

* fast/shadow-dom/slotchange-event-bubbling.html:
* fast/shadow-dom/slotchange-in-fallback.html:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@235568 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoUnreviewed Windows build fix after r235562.
cdumez@apple.com [Fri, 31 Aug 2018 20:15:28 +0000 (20:15 +0000)]
Unreviewed Windows build fix after r235562.

* WebCoreSupport/WebFrameLoaderClient.cpp:
(WebFrameLoaderClient::dispatchWillSubmitForm):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@235566 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years ago[WHLSL] Remove useless code in NameResolver
mmaxfield@apple.com [Fri, 31 Aug 2018 20:08:08 +0000 (20:08 +0000)]
[WHLSL] Remove useless code in NameResolver
https://bugs.webkit.org/show_bug.cgi?id=189176

Reviewed by Dean Jackson and Thomas Denney.

The check will always pass becase the standard library contains at least one function with the correct name.

* WebGPUShadingLanguageRI/NameResolver.js:
(NameResolver.prototype.visitDotExpression):
(NameResolver.prototype._handlePropertyAccess): Deleted.
(NameResolver.prototype.visitIndexExpression): Deleted.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@235565 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoMake it possible to mark tests as leaks in TestExpectations
simon.fraser@apple.com [Fri, 31 Aug 2018 20:00:22 +0000 (20:00 +0000)]
Make it possible to mark tests as leaks in TestExpectations
https://bugs.webkit.org/show_bug.cgi?id=189088

Reviewed by Jon Lee.

Have webkitpy parse out "Leak" expectations in TestExpectations, and do the right
thing if the test run did not use --world-leaks. Add unit tests for leaks combined
with various other result combinations.

* Scripts/webkitpy/layout_tests/controllers/layout_test_runner.py:
(LayoutTestRunner._update_summary_with_result):
(LayoutTestRunner._annotate_results_with_additional_failures):
* Scripts/webkitpy/layout_tests/models/test_expectations.py:
(TestExpectationParser):
(TestExpectations): The 'Leak' line was duplicated here, so remove a copy.
(TestExpectations.remove_leak_failures):
(TestExpectations.matches_an_expected_result):
* Scripts/webkitpy/layout_tests/models/test_expectations_unittest.py:
(Base.get_basic_tests):
* Scripts/webkitpy/port/test.py:
(TestList.add_reftest):
(unit_test_list):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@235564 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoWebAudioBufferList.cpp/.h is named as cocoa/WebAudioBufferList.cpp/.h
youenn@apple.com [Fri, 31 Aug 2018 19:13:04 +0000 (19:13 +0000)]
WebAudioBufferList.cpp/.h is named as cocoa/WebAudioBufferList.cpp/.h
https://bugs.webkit.org/show_bug.cgi?id=189160

Reviewed by Jer Noble.

Add WebAudioBufferList.cpp to unified build.
Update XCode project so that the file names are right.
No change of behavior.

* PlatformMac.cmake:
* SourcesCocoa.txt:
* WebCore.xcodeproj/project.pbxproj:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@235563 268f45cc-cd09-0410-ab3c-d52691b4dbfc