WebKit-https.git
2 years agoUnreviewed revert Fujii's revert in r237214 with new WinCairo build fix.
keith_miller@apple.com [Wed, 17 Oct 2018 14:10:44 +0000 (14:10 +0000)]
Unreviewed revert Fujii's revert in r237214 with new WinCairo build fix.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@237219 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoFlaky IntersectionObserver web platform tests involving style updates
ajuma@chromium.org [Wed, 17 Oct 2018 13:52:32 +0000 (13:52 +0000)]
Flaky IntersectionObserver web platform tests involving style updates
https://bugs.webkit.org/show_bug.cgi?id=189091

Reviewed by Simon Fraser.

Source/WebCore:

Update intersection observations when flushing layers from the WebProcess
to the UIProcess to make the timing of these updates more predictable, and
more consistent with the IntersectionObserver spec, since the spec expects
these updates to happen as part of the "Update the rendering" step in the
HTML EventLoop.

Getting a similar approach to work with WK1 seems to add more complexity than it's
worth, since flushes don't happen for scrolls handled by platform widgets, and
flushes for other invalidations only happen when in compositing mode.

The only remaining timer-driven intersection observation update is for handling
the initial observation on a newly added target, which needs to happen even if
there are no changes triggering a flush.

Tested by the following tests no longer being flaky:
    imported/w3c/web-platform-tests/intersection-observer/bounding-box.html
    imported/w3c/web-platform-tests/intersection-observer/display-none.html
    imported/w3c/web-platform-tests/intersection-observer/containing-block.html

* dom/Document.cpp:
(WebCore::Document::resolveStyle):
(WebCore::Document::updateIntersectionObservations):
(WebCore::Document::scheduleForcedIntersectionObservationUpdate):
(WebCore::Document::scheduleIntersectionObservationUpdate): Deleted.
* dom/Document.h:
* page/FrameView.cpp:
(WebCore::FrameView::flushCompositingStateForThisFrame):
(WebCore::FrameView::viewportContentsChanged):
* page/IntersectionObserver.cpp:
(WebCore::IntersectionObserver::observe):
* page/Page.cpp:
(WebCore::Page::Page):
(WebCore::Page::willDisplayPage):
(WebCore::Page::addDocumentNeedingIntersectionObservationUpdate):
(WebCore::Page::updateIntersectionObservations):
(WebCore::Page::scheduleForcedIntersectionObservationUpdate):
* page/Page.h:

Source/WebKit:

Add a WebPage::willDisplayPage bottleneck that gets called when flushing layers
or, in the non-composited case, when rendering the page into a bitmap.

* WebProcess/WebPage/AcceleratedDrawingArea.cpp:
(WebKit::AcceleratedDrawingArea::updateBackingStoreState):
* WebProcess/WebPage/CoordinatedGraphics/CoordinatedLayerTreeHost.cpp:
(WebKit::CoordinatedLayerTreeHost::layerFlushTimerFired):
* WebProcess/WebPage/DrawingAreaImpl.cpp:
(WebKit::DrawingAreaImpl::display):
* WebProcess/WebPage/RemoteLayerTree/RemoteLayerTreeDrawingArea.mm:
(WebKit::RemoteLayerTreeDrawingArea::flushLayers):
* WebProcess/WebPage/WebPage.cpp:
(WebKit::WebPage::willDisplayPage):
* WebProcess/WebPage/WebPage.h:
* WebProcess/WebPage/mac/TiledCoreAnimationDrawingArea.mm:
(WebKit::TiledCoreAnimationDrawingArea::flushLayers):

LayoutTests:

Remove expectation for tests that are no longer flaky.

Skip IntersectionObserver tests on WK1.

* TestExpectations:
* platform/mac-wk1/TestExpectations:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@237218 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years ago[EME] Sanity check key ID length in the keyids init data format
cturner@igalia.com [Wed, 17 Oct 2018 11:07:59 +0000 (11:07 +0000)]
[EME] Sanity check key ID length in the keyids init data format
https://bugs.webkit.org/show_bug.cgi?id=190629

Reviewed by Xabier Rodriguez-Calvar.

LayoutTests/imported/w3c:

* web-platform-tests/encrypted-media/clearkey-generate-request-disallowed-input.https-expected.txt:
Added new baseline.

Source/WebCore:

Covered by web-platform-tests/encrypted-media/clearkey-generate-request-disallowed-input.https.html

* Modules/encryptedmedia/InitDataRegistry.cpp:
(WebCore::extractKeyIDsKeyids): Ensure the decoded key id length
is at least 1 byte and no more than 512 bytes.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@237217 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years ago[EME] media/encrypted-media/mock-MediaKeySystemAccess.html crashes in CDM::createInstance
cturner@igalia.com [Wed, 17 Oct 2018 11:00:13 +0000 (11:00 +0000)]
[EME] media/encrypted-media/mock-MediaKeySystemAccess.html crashes in CDM::createInstance
https://bugs.webkit.org/show_bug.cgi?id=190578

Reviewed by Xabier Rodriguez-Calvar.

* platform/gtk/TestExpectations:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@237216 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoGetIndexedPropertyStorage can GC.
mark.lam@apple.com [Wed, 17 Oct 2018 04:16:24 +0000 (04:16 +0000)]
GetIndexedPropertyStorage can GC.
https://bugs.webkit.org/show_bug.cgi?id=190625
<rdar://problem/45309366>

Reviewed by Saam Barati.

This is because if the ArrayMode type is String, the DFG and FTL will be emitting
a call to operationResolveRope, and operationResolveRope can GC.  This patch
updates doesGC() to reflect this.

* dfg/DFGDoesGC.cpp:
(JSC::DFG::doesGC):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@237215 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoUnreviewed, rolling out r237188, r237189, and r237197.
Hironori.Fujii@sony.com [Wed, 17 Oct 2018 03:01:23 +0000 (03:01 +0000)]
Unreviewed, rolling out r237188, r237189, and r237197.

It breaks WinCairo Debug builds and Release LayoutTests

Reverted changesets:

https://bugs.webkit.org/show_bug.cgi?id=189708
https://trac.webkit.org/changeset/237188

"Unreviewed, forgot to add untracked files."
https://trac.webkit.org/changeset/237189

"isASTErroneous in offlineasm should de-macroify before
looking for Errors"
https://bugs.webkit.org/show_bug.cgi?id=190634
https://trac.webkit.org/changeset/237197

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@237214 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years ago[GTK][WPE] Fix xdg-desktop-portal permissions from a sandbox
commit-queue@webkit.org [Wed, 17 Oct 2018 02:19:11 +0000 (02:19 +0000)]
[GTK][WPE] Fix xdg-desktop-portal permissions from a sandbox
https://bugs.webkit.org/show_bug.cgi?id=190619

Patch by Patrick Griffis <pgriffis@igalia.com> on 2018-10-16
Reviewed by Michael Catanzaro.

We have to generate a /.flatpak-info file for xdg-desktop-portal
to properly treat us as a sandboxed application.

* UIProcess/Launcher/glib/BubblewrapLauncher.cpp:
(WebKit::createSealedMemFdWithData):
(WebKit::argsToFd):
(WebKit::XDGDBusProxyLauncher::setPermissions):
(WebKit::createFlatpakInfo):
(WebKit::bubblewrapSpawn):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@237213 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoWeb Inspector: ⌃G to jump to line in CSS file wipes the line
commit-queue@webkit.org [Wed, 17 Oct 2018 01:53:30 +0000 (01:53 +0000)]
Web Inspector: ⌃G to jump to line in CSS file wipes the line
https://bugs.webkit.org/show_bug.cgi?id=190645

Patch by Joseph Pecoraro <pecoraro@apple.com> on 2018-10-16
Reviewed by Devin Rousso.

* UserInterface/Base/Main.js:
* UserInterface/Views/Dialog.js:
(WI.Dialog.prototype.dismiss):
* UserInterface/Views/SourceCodeTextEditor.js:
(WI.SourceCodeTextEditor.prototype.dialogWasDismissed): Deleted.
Rename the handler when dismissing a dialog with a value so that
it is clear this is only on a successful value and not an invalid value.
Dialog now only calls this delegate with a valid value. This addresses
exceptions seen when the GoToLineDialog was dimissed via Escape and
an unexpected `null` value was being handled in the delegate.

* UserInterface/Views/GoToLineDialog.js:
(WI.GoToLineDialog.prototype._handleKeydownEvent):
* UserInterface/Views/OpenResourceDialog.js:
(WI.OpenResourceDialog.prototype._handleKeydownEvent):
Prevent default on the event whenever we dismiss the dialog.
Without this, the key event was also being sent to the editor.
In this case, that meant an "Enter" key was replacing the new
selection in the TextEditor with a newline!

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@237212 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoRegression(r236795) Check boxes are sometimes checked when they should not be
cdumez@apple.com [Wed, 17 Oct 2018 00:27:54 +0000 (00:27 +0000)]
Regression(r236795) Check boxes are sometimes checked when they should not be
https://bugs.webkit.org/show_bug.cgi?id=190651
<rdar://problem/45319934>

Reviewed by Ryosuke Niwa.

Source/WebCore:

r236795 mistakenly dropped the statement resetting m_reflectsCheckedAttribute to true
in HTMLInputElement::parseAttribute() because it looked like a no-op given that we
made sure it was true a couple of line above. However, I overlooked that calling
HTMLInputElement::setChecked() sets m_reflectsCheckedAttribute to false.

This patch thus re-introduces the statement. It also renames m_reflectsCheckedAttribute
to m_dirtyCheckednessFlag and reverses its value in order to match the specification
more closely:
- https://html.spec.whatwg.org/#concept-input-checked-dirty-flag

Test: fast/dom/HTMLInputElement/checkbox-dirty-checkedness-flag.html

* html/HTMLInputElement.cpp:
(WebCore::HTMLInputElement::HTMLInputElement):
(WebCore::HTMLInputElement::parseAttribute):
(WebCore::HTMLInputElement::finishParsingChildren):
(WebCore::HTMLInputElement::reset):
(WebCore::HTMLInputElement::setChecked):
(WebCore::HTMLInputElement::copyNonAttributePropertiesFromElement):
* html/HTMLInputElement.h:

LayoutTests:

Add layout test coverage.

* fast/dom/HTMLInputElement/checkbox-dirty-checkedness-flag-expected.txt: Added.
* fast/dom/HTMLInputElement/checkbox-dirty-checkedness-flag.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@237211 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoAdd a switch for Web SQL
sihui_liu@apple.com [Tue, 16 Oct 2018 23:59:38 +0000 (23:59 +0000)]
Add a switch for Web SQL
https://bugs.webkit.org/show_bug.cgi?id=190271

Reviewed by Ryosuke Niwa.

Source/WebCore:

Web SQL is still enabled by default.

* Modules/webdatabase/DOMWindowWebDatabase.idl:
* Modules/webdatabase/Database.idl:
* Modules/webdatabase/SQLError.idl:
* Modules/webdatabase/SQLResultSet.idl:
* Modules/webdatabase/SQLResultSetRowList.idl:
* bindings/js/WebCoreBuiltinNames.h:
* page/RuntimeEnabledFeatures.h:
(WebCore::RuntimeEnabledFeatures::setWebSQLDisabled):
(WebCore::RuntimeEnabledFeatures::webSQLEnabled const):

Source/WebKit:

* Shared/WebPreferences.yaml:
* UIProcess/API/C/WKPreferences.cpp:
(WKPreferencesSetWebSQLDisabled):
(WKPreferencesGetWebSQLDisabled):
* UIProcess/API/C/WKPreferencesRef.h:

Tools:

* WebKitTestRunner/TestController.cpp:
(WTR::TestController::resetPreferencesToConsistentValues):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@237210 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agowindow.performance should not become null after the window loses its browsing context
cdumez@apple.com [Tue, 16 Oct 2018 23:43:10 +0000 (23:43 +0000)]
window.performance should not become null after the window loses its browsing context
https://bugs.webkit.org/show_bug.cgi?id=190636

Reviewed by Ryosuke Niwa.

Source/WebCore:

window.performance should not become null after the window loses its browsing context. This
WebKit behavior does not match the HTML specification nor the behavior of other browsers.

Also note that the attribute is not nullable in the specification:
- https://www.w3.org/TR/navigation-timing/#sec-window.performance-attribute

No new tests, updated existing test.

* page/DOMWindow.cpp:
(WebCore::DOMWindow::performance const):
* page/Performance.cpp:
(WebCore::Performance::Performance):
* page/Performance.h:
* platform/GenericTaskQueue.h:
(WebCore::TaskDispatcher::TaskDispatcher):
(WebCore::TaskDispatcher::postTask):
(WebCore::GenericTaskQueue::GenericTaskQueue):
(WebCore::GenericTaskQueue::isClosed const):
* workers/WorkerGlobalScope.cpp:
(WebCore::WorkerGlobalScope::WorkerGlobalScope):

LayoutTests:

Extend layout test coverage.

* http/tests/dom/cross-origin-detached-window-properties-expected.txt:
* http/tests/dom/cross-origin-detached-window-properties.html:
* http/tests/dom/same-origin-detached-window-properties-expected.txt:
* http/tests/dom/same-origin-detached-window-properties.html:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@237209 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years ago[Media] Use nanoseconds as MaximumTimeScale
aboya@igalia.com [Tue, 16 Oct 2018 23:18:07 +0000 (23:18 +0000)]
[Media] Use nanoseconds as MaximumTimeScale
https://bugs.webkit.org/show_bug.cgi?id=190631

1e9 is a much more useful timescale than the previous one 2^31-1.
Unlike 2^31-1, which is a prime number, nanosecond scale is pretty
common among some formats like WebM and frameworks like GStreamer
where base 10 timescale is common... and it's those big timescales the
ones that are usually scaled up to MaximumTimeScale.

Reviewed by Jer Noble.

* wtf/MediaTime.cpp:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@237208 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoRefactoring: Convert HTMLMediaElement::scheduleDelayedAction() to individually schedu...
jer.noble@apple.com [Tue, 16 Oct 2018 23:12:53 +0000 (23:12 +0000)]
Refactoring: Convert HTMLMediaElement::scheduleDelayedAction() to individually schedulable & cancelable tasks
https://bugs.webkit.org/show_bug.cgi?id=188208

Reviewed by Eric Carlson.

Rather than have a single, monolithic, zero-duration-timer based dispatch for
a bunch of methods to be performed in a future run-loop, convert them all to
use a DeferrableTask, which in turn is a kind of GenericTaskQueue which can
enqueue only a single task at a time. Convert some other zero-duration-timer
and GenericTaskQueue dispatches to this new DeferrableTask.

* WebCore.xcodeproj/project.pbxproj:
* html/HTMLMediaElement.cpp:
(WebCore::HTMLMediaElement::HTMLMediaElement):
(WebCore::HTMLMediaElement::finishParsingChildren):
(WebCore::HTMLMediaElement::scheduleCheckPlaybackTargetCompatability):
(WebCore::HTMLMediaElement::checkPlaybackTargetCompatablity):
(WebCore::HTMLMediaElement::prepareForLoad):
(WebCore::HTMLMediaElement::setReadyState):
(WebCore::HTMLMediaElement::seekWithTolerance):
(WebCore::HTMLMediaElement::setMuted):
(WebCore::HTMLMediaElement::mediaPlayerDidAddTextTrack):
(WebCore::HTMLMediaElement::didAddTextTrack):
(WebCore::HTMLMediaElement::scheduleConfigureTextTracks):
(WebCore::HTMLMediaElement::mediaPlayerTimeChanged):
(WebCore::HTMLMediaElement::scheduleMediaEngineWasUpdated):
(WebCore::HTMLMediaElement::mediaEngineWasUpdated):
(WebCore::HTMLMediaElement::mediaPlayerEngineUpdated):
(WebCore::HTMLMediaElement::scheduleUpdatePlayState):
(WebCore::HTMLMediaElement::updatePlayState):
(WebCore::HTMLMediaElement::setPlaying):
(WebCore::HTMLMediaElement::setPausedInternal):
(WebCore::HTMLMediaElement::cancelPendingTasks):
(WebCore::HTMLMediaElement::userCancelledLoad):
(WebCore::HTMLMediaElement::clearMediaPlayer):
(WebCore::HTMLMediaElement::contextDestroyed):
(WebCore::HTMLMediaElement::stop):
(WebCore::HTMLMediaElement::suspend):
(WebCore::HTMLMediaElement::resume):
(WebCore::HTMLMediaElement::mediaPlayerCurrentPlaybackTargetIsWirelessChanged):
(WebCore::HTMLMediaElement::dispatchEvent):
(WebCore::HTMLMediaElement::removeEventListener):
(WebCore::HTMLMediaElement::enqueuePlaybackTargetAvailabilityChangedEvent):
(WebCore::HTMLMediaElement::didBecomeFullscreenElement):
(WebCore::HTMLMediaElement::markCaptionAndSubtitleTracksAsUnconfigured):
(WebCore::HTMLMediaElement::scheduleUpdateMediaState):
(WebCore::HTMLMediaElement::updateMediaState):
(WebCore::HTMLMediaElement::playbackControlsManagerBehaviorRestrictionsTimerFired):
(WebCore::setFlags): Deleted.
(WebCore::clearFlags): Deleted.
(WebCore::actionName): Deleted.
(WebCore::HTMLMediaElement::scheduleDelayedAction): Deleted.
(WebCore::HTMLMediaElement::pendingActionTimerFired): Deleted.
* html/HTMLMediaElement.h:
* html/HTMLMediaElementEnums.h:
* platform/DeferrableTask.h: Added.
(WebCore::DeferrableTask::DeferrableTask):
(WebCore::DeferrableTask::scheduleTask):
(WebCore::DeferrableTask::close):
(WebCore::DeferrableTask::cancelTask):
(WebCore::DeferrableTask::hasPendingTask const):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@237207 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoUnreviewed, rolling out r236996.
dewei_zhu@apple.com [Tue, 16 Oct 2018 22:58:03 +0000 (22:58 +0000)]
Unreviewed, rolling out r236996.

Temporarily rollout to fix a manifest generation bug.
"lastModified" field for certain platform & metric configurations may be missing.

Reverted changeset:

"ManifestGenerator shouldn't need more than 1GB of memory or
run for 30 seconds"
https://bugs.webkit.org/show_bug.cgi?id=190393
https://trac.webkit.org/changeset/236996

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@237206 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoRemove InjectedBundleBackForwardList
achristensen@apple.com [Tue, 16 Oct 2018 22:31:00 +0000 (22:31 +0000)]
Remove InjectedBundleBackForwardList
https://bugs.webkit.org/show_bug.cgi?id=190640

Reviewed by Chris Dumez.

Source/WebKit:

These objects are only used in testing now, so I made a replacement to be used for testing,
WKBundlePageDumpHistoryForTesting and WKBundleClearHistoryForTesting.
The existence of these objects is problematic with process swapping and the true back/forward list being in the UIProcess.

* Sources.txt:
* WebProcess/InjectedBundle/API/c/WKBundleBackForwardList.cpp:
(WKBundleBackForwardListGetTypeID):
(WKBundleBackForwardListCopyItemAtIndex):
(WKBundleBackForwardListGetBackListCount):
(WKBundleBackForwardListGetForwardListCount):
(WKBundleBackForwardListClear):
* WebProcess/InjectedBundle/API/c/WKBundleBackForwardList.h:
* WebProcess/InjectedBundle/API/c/WKBundleBackForwardListItem.cpp:
(WKBundleBackForwardListItemGetTypeID):
(WKBundleBackForwardListItemIsSame):
(WKBundleBackForwardListItemCopyURL):
(WKBundleBackForwardListItemCopyTarget):
(WKBundleBackForwardListItemIsTargetItem):
(WKBundleBackForwardListItemCopyChildren):
* WebProcess/InjectedBundle/API/c/WKBundleBackForwardListItem.h:
* WebProcess/InjectedBundle/API/c/WKBundlePage.cpp:
(WKBundlePageDumpHistoryForTesting):
(WKBundleClearHistoryForTesting):
(WKBundlePageGetBackForwardList):
* WebProcess/InjectedBundle/API/c/WKBundlePage.h:
* WebProcess/InjectedBundle/InjectedBundleBackForwardList.cpp: Removed.
* WebProcess/InjectedBundle/InjectedBundleBackForwardList.h: Removed.
* WebProcess/InjectedBundle/InjectedBundleBackForwardListItem.cpp: Removed.
* WebProcess/InjectedBundle/InjectedBundleBackForwardListItem.h: Removed.
* WebProcess/InjectedBundle/InjectedBundlePageLoaderClient.cpp:
* WebProcess/WebCoreSupport/WebFrameLoaderClient.cpp:
* WebProcess/WebPage/WebPage.cpp:
(WebKit::WebPage::~WebPage):
(WebKit::dumpHistoryItem):
(WebKit::WebPage::dumpHistoryForTesting):
(WebKit::WebPage::clearHistory):
(WebKit::WebPage::didReceiveSyncMessage):
(WebKit::WebPage::backForwardList): Deleted.
* WebProcess/WebPage/WebPage.h:

Tools:

* WebKitTestRunner/InjectedBundle/InjectedBundle.cpp:
(WTR::InjectedBundle::dumpBackForwardListsForAllPages):
* WebKitTestRunner/InjectedBundle/InjectedBundlePage.cpp:
(WTR::InjectedBundlePage::prepare):
(WTR::InjectedBundlePage::dumpHistory):
(WTR::hasPrefix): Deleted.
(WTR::compareByTargetName): Deleted.
(WTR::dumpBackForwardListItem): Deleted.
(WTR::InjectedBundlePage::dumpBackForwardList): Deleted.
* WebKitTestRunner/InjectedBundle/InjectedBundlePage.h:
* WebKitTestRunner/InjectedBundle/TestRunner.cpp:
(WTR::TestRunner::clearBackForwardList):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@237205 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoAdd <meta name="supported-color-schemes"> to control what color schemes the page...
timothy@apple.com [Tue, 16 Oct 2018 21:53:55 +0000 (21:53 +0000)]
Add <meta name="supported-color-schemes"> to control what color schemes the page supports
https://bugs.webkit.org/show_bug.cgi?id=190526
rdar://problem/45230140

Reviewed by Dean Jackson.

Source/WebCore:

Test: css-dark-mode/supported-color-schemes.html

* dom/Document.cpp:
(WebCore::isColorSchemeSeparator): Added.
(WebCore::processColorSchemes): Added.
(WebCore::Document::processSupportedColorSchemes): Added.
(WebCore::Document::useDarkAppearance const): Take system appearance and document's
suppoerted color shcemes into account.
* dom/Document.h:
* editing/cocoa/WebContentReaderCocoa.mm:
(WebCore::createFragment): Update use of LocalDefaultSystemAppearance.
* html/HTMLMetaElement.cpp:
(WebCore::HTMLMetaElement::process): Added supported-color-schemes behind runtime feature check.
* inspector/InspectorOverlay.cpp:
(WebCore::InspectorOverlay::paint): Update use of LocalDefaultSystemAppearance.
* platform/mac/DragImageMac.mm:
(WebCore::createDragImageForLink): Update use of LocalDefaultSystemAppearance.
* platform/mac/LocalDefaultSystemAppearance.h:
* platform/mac/LocalDefaultSystemAppearance.mm:
(WebCore::LocalDefaultSystemAppearance::LocalDefaultSystemAppearance):
Drop the need for useSystemAppearance. This is accounted for in Document::useDarkAppearance.
* platform/mac/ThemeMac.mm:
(WebCore::ThemeMac::paint): Update use of LocalDefaultSystemAppearance.
* rendering/RenderLayerBacking.cpp:
(WebCore::RenderLayerBacking::paintContents): Update use of LocalDefaultSystemAppearance.
* rendering/RenderLayerCompositor.cpp:
(WebCore::RenderLayerCompositor::paintContents): Update use of LocalDefaultSystemAppearance.
* rendering/RenderTheme.cpp:
(WebCore::RenderTheme::paint): Use document instead of page.
* rendering/RenderThemeCocoa.mm:
(WebCore::RenderThemeCocoa::drawLineForDocumentMarker): Update use of LocalDefaultSystemAppearance.
* rendering/RenderThemeMac.mm:
(WebCore::RenderThemeMac::documentViewFor const): Update use of LocalDefaultSystemAppearance.
(WebCore::RenderThemeMac::platformActiveSelectionBackgroundColor const): Ditto.
(WebCore::RenderThemeMac::platformInactiveSelectionBackgroundColor const): Ditto.
(WebCore::RenderThemeMac::transformSelectionBackgroundColor const): Ditto.
(WebCore::RenderThemeMac::supportsSelectionForegroundColors const): Ditto.
(WebCore::RenderThemeMac::platformActiveSelectionForegroundColor const): Ditto.
(WebCore::RenderThemeMac::platformInactiveSelectionForegroundColor const): Ditto.
(WebCore::RenderThemeMac::platformActiveListBoxSelectionBackgroundColor const): Ditto.
(WebCore::RenderThemeMac::platformInactiveListBoxSelectionBackgroundColor const): Ditto.
(WebCore::RenderThemeMac::platformActiveListBoxSelectionForegroundColor const): Ditto.
(WebCore::RenderThemeMac::platformInactiveListBoxSelectionForegroundColor const): Ditto.
(WebCore::RenderThemeMac::platformActiveTextSearchHighlightColor const): Ditto.
(WebCore::RenderThemeMac::colorCache const): Ditto.
(WebCore::RenderThemeMac::systemColor const): Ditto.
(WebCore::RenderThemeMac::paintCellAndSetFocusedElementNeedsRepaintIfNecessary): Ditto.
(WebCore::RenderThemeMac::paintSliderThumb): Ditto.
(WebCore::RenderThemeMac::usingDarkAppearance const): No need to limit to macOS here.
* svg/graphics/SVGImage.cpp:
(WebCore::SVGImage::draw): Update use of LocalDefaultSystemAppearance.

Source/WebKit:

* UIProcess/mac/WKPrintingView.mm:
(-[WKPrintingView drawRect:]): Update use of LocalDefaultSystemAppearance.
* WebProcess/InjectedBundle/DOM/InjectedBundleRangeHandle.cpp:
(WebKit::InjectedBundleRangeHandle::renderedImage): Ditto.
* WebProcess/Plugins/PDF/PDFPlugin.mm:
(WebKit::PDFPlugin::paintControlForLayerInContext): Ditto.
* WebProcess/WebPage/WebPage.cpp:
(WebKit::WebPage::drawRect): Ditto.

Source/WebKitLegacy/mac:

* WebCoreSupport/WebEditorClient.mm:
(_WebCreateFragment): Update use of LocalDefaultSystemAppearance.

LayoutTests:

* css-dark-mode/supported-color-schemes-expected.txt: Added.
* css-dark-mode/supported-color-schemes.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@237204 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoRemove superfluous VIDEO build guard
commit-queue@webkit.org [Tue, 16 Oct 2018 21:33:25 +0000 (21:33 +0000)]
Remove superfluous VIDEO build guard
https://bugs.webkit.org/show_bug.cgi?id=190624

Patch by Rob Buis <rbuis@igalia.com> on 2018-10-16
Reviewed by Michael Catanzaro.

Remove some instances where the VIDEO build guard is included within another VIDEO build guard.

No new tests. No change in behavior.

* platform/graphics/MediaPlayer.cpp:
(WebCore::buildMediaEnginesVector):
* platform/graphics/MediaPlayer.h:
* platform/graphics/MediaPlayerPrivate.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@237203 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoSupport RTCConfiguration.certificates
youenn@apple.com [Tue, 16 Oct 2018 21:01:15 +0000 (21:01 +0000)]
Support RTCConfiguration.certificates
https://bugs.webkit.org/show_bug.cgi?id=190603

Reviewed by Eric Carlson.

LayoutTests/imported/w3c:

* web-platform-tests/webrtc/RTCCertificate-expected.txt:
* web-platform-tests/webrtc/RTCPeerConnection-constructor-expected.txt:
* web-platform-tests/webrtc/rtcpeerconnection/rtcpeerconnection-constructor-expected.txt:

Source/ThirdParty/libwebrtc:

* Configurations/libwebrtc.iOS.exp:
* Configurations/libwebrtc.iOSsim.exp:
* Configurations/libwebrtc.mac.exp:

Source/WebCore:

Update RTCConfiguration to have a certificates member.
Add the conversion from RTCCertificate to libwebrtc certificates.
Add check to ensure that certificates are not expired.
Add check to ensure that certificates passed to setConfiguration were
the same as the ones passed to RTCPeerConnection constructor.
Once these checks are done, we reuse the certificates created at
RCPeerConnection creation time when setting the configuration again.

Covered by rebased WPT test.

* Modules/mediastream/RTCConfiguration.h:
* Modules/mediastream/RTCConfiguration.idl:
* Modules/mediastream/RTCPeerConnection.cpp:
(WebCore::certificatesFromConfiguration):
(WebCore::RTCPeerConnection::initializeConfiguration):
(WebCore::RTCPeerConnection::setConfiguration):
* Modules/mediastream/RTCPeerConnection.h:
* Modules/mediastream/libwebrtc/LibWebRTCMediaEndpoint.cpp:
(WebCore::LibWebRTCMediaEndpoint::setConfiguration):
* Modules/mediastream/libwebrtc/LibWebRTCPeerConnectionBackend.cpp:
(WebCore::configurationFromMediaEndpointConfiguration):
* platform/mediastream/MediaEndpointConfiguration.cpp:
(WebCore::MediaEndpointConfiguration::MediaEndpointConfiguration):
* platform/mediastream/MediaEndpointConfiguration.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@237202 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoUnreviewed, rolling out r237163.
ryanhaddad@apple.com [Tue, 16 Oct 2018 20:18:06 +0000 (20:18 +0000)]
Unreviewed, rolling out r237163.

Introduced layout test failures on iOS Simulator.

Reverted changeset:

"[CG] Adopt CG SPI for non-even cornered rounded rects"
https://bugs.webkit.org/show_bug.cgi?id=190155
https://trac.webkit.org/changeset/237163

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@237201 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years ago[JSC] stress/array-prototype-concat-of-long-spliced-arrays2.js times out on arm and...
dinfuehr@igalia.com [Tue, 16 Oct 2018 20:09:03 +0000 (20:09 +0000)]
[JSC] stress/array-prototype-concat-of-long-spliced-arrays2.js times out on arm and mips
https://bugs.webkit.org/show_bug.cgi?id=190611

Reviewed by Saam Barati.

Reduce array length just like in array-prototype-concat-of-long-spliced-arrays.js
to improve test runtime. On ARM/MIPS this test even timed out when running all
tests.

* stress/array-prototype-concat-of-long-spliced-arrays2.js:
(test):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@237200 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoUnreviewed, added myself to contributors.json.
dinfuehr@igalia.com [Tue, 16 Oct 2018 20:07:23 +0000 (20:07 +0000)]
Unreviewed, added myself to contributors.json.

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

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@237199 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoWeb Inspector: Canvas: capture previously saved states and add them to the recording...
drousso@apple.com [Tue, 16 Oct 2018 19:23:21 +0000 (19:23 +0000)]
Web Inspector: Canvas: capture previously saved states and add them to the recording payload
https://bugs.webkit.org/show_bug.cgi?id=190473

Reviewed by Joseph Pecoraro.

Source/JavaScriptCore:

* inspector/protocol/Recording.json:
Add `states` key to `InitialState` object.

Source/WebCore:

Updated existing tests: inspector/canvas/recording-2d.html
                        inspector/model/recording.html

Instead of sending a single object of the current state of the context, send an array of
objects, one for each restore point.

* html/canvas/CanvasRenderingContext2DBase.h:
* html/canvas/CanvasRenderingContext2DBase.cpp:
(WebCore::CanvasRenderingContext2DBase::stateStack): Added.

* inspector/InspectorCanvas.h:
* inspector/InspectorCanvas.cpp:
(WebCore::InspectorCanvas::stringIndexForKey): Added.
(WebCore::InspectorCanvas::buildInitialState):

Source/WebInspectorUI:

Instead of sending a single object of the current state of the context, send an array of
objects, one for each restore point. When replaying, recreate each restore point before
applying the selected action(s).

* UserInterface/Models/Recording.js:
(WI.Recording):
(WI.Recording.fromPayload):
(WI.Recording.prototype.toJSON):
(WI.Recording.prototype.async._process):
(WI.Recording.prototype.async._swizzleState): Added.
* UserInterface/Models/RecordingAction.js:
(WI.RecordingAction):
(WI.RecordingAction.deriveCurrentState): Added.
(WI.RecordingAction.prototype.get states): Added.
(WI.RecordingAction.prototype.process):
(WI.RecordingAction.prototype.get state): Deleted.
Drive-by: when `process`ing, also check to see if any values in the current state changed
outside of those expected in `_stateModifiers` (e.g. `restore` may modify some state values).

* UserInterface/Views/RecordingContentView.js:
(WI.RecordingContentView.prototype._generateContentCanvas2D):

* UserInterface/Views/RecordingStateDetailsSidebarPanel.js:
(WI.RecordingStateDetailsSidebarPanel.prototype._generateDetailsCanvas2D):
Default to showing the most recent (current) state.

* UserInterface/Views/CanvasTabContentView.js:
(WI.CanvasTabContentView.prototype.initialLayout): Added.
(WI.CanvasTabContentView.prototype._addCanvas):
(WI.CanvasTabContentView.prototype._removeCanvas):
(WI.CanvasTabContentView.prototype._addRecording): Added.
(WI.CanvasTabContentView.prototype._recordingImportedOrStopped):
(WI.CanvasTabContentView.prototype._recordingAdded): Deleted.
* UserInterface/Controllers/CanvasManager.js:
(WI.CanvasManager):
(WI.CanvasManager.prototype.get importedRecordings): Added.
(WI.CanvasManager.prototype.importRecording):
Drive-by: store imported recordings on `WI.CanvasManager` so that if the Canvas tab is
closed we can still show the list of imported recordings.

LayoutTests:

* inspector/canvas/recording-2d-expected.txt:
* inspector/canvas/recording-2d.html:
* inspector/canvas/resources/recording-utilities.js:
(TestPage.registerInitializer):
(TestPage.registerInitializer.async.logRecording): Added.
(TestPage.registerInitializer.logRecording): Deleted.
* inspector/model/recording-expected.txt:
* inspector/model/recording.html:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@237198 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoisASTErroneous in offlineasm should de-macroify before looking for Errors
keith_miller@apple.com [Tue, 16 Oct 2018 19:16:24 +0000 (19:16 +0000)]
isASTErroneous in offlineasm should de-macroify before looking for Errors
https://bugs.webkit.org/show_bug.cgi?id=190634

Reviewed by Mark Lam.

If a macro isn't usable in a configuration it might still cause us to
think the ast is invalid. This change runs the de-macroifier before
looking for errors.

Also, it adds a missing include to Printer.h.

* assembler/Printer.h:
* offlineasm/settings.rb:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@237197 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoWeb Inspector: Canvas recording sidebar scroll position lost after switching tabs
drousso@apple.com [Tue, 16 Oct 2018 19:13:15 +0000 (19:13 +0000)]
Web Inspector: Canvas recording sidebar scroll position lost after switching tabs
https://bugs.webkit.org/show_bug.cgi?id=190482

Reviewed by Joseph Pecoraro.

* UserInterface/Views/Sidebar.js:
(WI.Sidebar.prototype.removeSidebarPanel):
(WI.Sidebar.prototype.set selectedSidebarPanel):
Call `hidden` before setting `selected` to `false`, as that sets `display: none;` on the
element, which sets the `scrollTop` to `0`.

* UserInterface/Views/SidebarPanel.js:
(WI.SidebarPanel.prototype.shown):
(WI.SidebarPanel.prototype.hidden):
(WI.SidebarPanel.prototype.get scrollElement): Added.
Allow subclasses to specify the scrolling element for saving/restoring the scroll position.

* UserInterface/Views/CanvasSidebarPanel.js:
(WI.CanvasSidebarPanel.prototype.get scrollElement): Added.
* UserInterface/Views/RecordingStateDetailsSidebarPanel.js:
(WI.RecordingStateDetailsSidebarPanel.prototype.get scrollElement): Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@237196 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoWeb Inspector: Should be a way to go directly from an event in the overview view...
drousso@apple.com [Tue, 16 Oct 2018 19:09:29 +0000 (19:09 +0000)]
Web Inspector: Should be a way to go directly from an event in the overview view to the specialized timeline for that event
https://bugs.webkit.org/show_bug.cgi?id=135307
<rdar://problem/17273966>

Reviewed by Joseph Pecoraro.

When a `WI.TimelineRecordBar` is clicked, call up the delegate chain to the overview and
adjust the currently selected `WI.TimelineRecordBar` among the `WI.TimelineOverviewGraph`s.
Similarly, selecting a `WI.DataGridNode` in any `WI.TimelineView` subclass will use the same
logic to select the corresponding `WI.TimelineRecordBar`.

* UserInterface/Views/TimelineOverview.js:
(WI.TimelineOverview):
(WI.TimelineOverview.prototype.reset):
(WI.TimelineOverview.prototype._recordSelected):

* UserInterface/Views/TimelineOverviewGraph.js:
(WI.TimelineOverviewGraph):
(WI.TimelineOverviewGraph.prototype.set selectedRecord):
(WI.TimelineOverviewGraph.prototype.get selectedRecordBar): Added.
(WI.TimelineOverviewGraph.prototype.set selectedRecordBar): Added.
(WI.TimelineOverviewGraph.prototype.timelineRecordBarClicked): Added.
(WI.TimelineOverviewGraph.prototype._needsSelectedRecordLayout):

* UserInterface/Views/TimelineRecordBar.js:
(WI.TimelineRecordBar):
(WI.TimelineRecordBar.prototype.get selected): Added.
(WI.TimelineRecordBar.prototype.set selected): Added.
(WI.TimelineRecordBar.prototype._handleClick): Added.
* UserInterface/Views/TimelineRecordBar.css:
(.timeline-record-bar.selected > .segment): Added.
Add a `delegate` that is notified whenever the element is clicked.

* UserInterface/Views/TimelineDataGridNode.js:
(WI.TimelineDataGridNode.prototype.refreshGraph.createBar):
* UserInterface/Views/LayoutTimelineOverviewGraph.js:
(WI.LayoutTimelineOverviewGraph.prototype.updateSelectedRecord):
(WI.LayoutTimelineOverviewGraph.prototype._updateRowLayout.createBar):
* UserInterface/Views/NetworkTimelineOverviewGraph.js:
(WI.NetworkTimelineOverviewGraph.prototype.layout.createBar):
* UserInterface/Views/ScriptTimelineOverviewGraph.js:
(WI.ScriptTimelineOverviewGraph.prototype.layout.createBar):
(WI.ScriptTimelineOverviewGraph.prototype.updateSelectedRecord):

* UserInterface/Views/TimelineRecordingContentView.js:
(WI.TimelineRecordingContentView.prototype.showTimelineViewForTimeline):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@237195 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoUnreviewed attempt to fix the Windows build after r237112.
ryanhaddad@apple.com [Tue, 16 Oct 2018 18:57:07 +0000 (18:57 +0000)]
Unreviewed attempt to fix the Windows build after r237112.

* Plugins/PluginView.cpp:
(WebCore::PluginView::performRequest):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@237194 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoImplement feature flag and bindings for CSS Painting API
commit-queue@webkit.org [Tue, 16 Oct 2018 18:04:43 +0000 (18:04 +0000)]
Implement feature flag and bindings for CSS Painting API
https://bugs.webkit.org/show_bug.cgi?id=190237

Patch by Justin Michaud <justin_michaud@apple.com> on 2018-10-16
Reviewed by Ryosuke Niwa.

Source/JavaScriptCore:

* Configurations/FeatureDefines.xcconfig:

Source/WebCore:

Add feature flag and bindings for CSS Painting API. This adds a new property,
CSS.paintWorkletGlobalScope, which will be a temporary way to access the paint
worklet global scope untill CSS.paintWorklet is implemented.

There are a few small changes, mostly adding headers and "using" declarations,
that were required to get this to build. This is probably related to unified sources.

Tests: fast/css-custom-paint/basic.html
       fast/css-custom-paint/registerPaintBindings.html

* CMakeLists.txt:
* Configurations/FeatureDefines.xcconfig:
* DerivedSources.make:
* Sources.txt:
* WebCore.xcodeproj/project.pbxproj:
* bindings/js/JSCSSPaintWorkletGlobalScopeCustom.cpp: Added.
(WebCore::throwInvalidModificationError):
(WebCore::JSCSSPaintWorkletGlobalScope::registerPaint):
* bindings/js/JSEventListener.cpp:
* bindings/js/JSRemoteDOMWindowCustom.cpp:
* bindings/js/JSWebMetalRenderPassAttachmentDescriptorCustom.cpp:
* bindings/js/WebCoreBuiltinNames.h:
* css/CSSPaintCallback.h: Copied from Source/WebCore/css/MediaQueryParserContext.cpp.
(WebCore::CSSPaintCallback::~CSSPaintCallback):
* css/CSSPaintCallback.idl: Added.
* css/CSSPaintWorkletGlobalScope.cpp: Copied from Source/WebCore/css/MediaQueryParserContext.cpp.
(WebCore::CSSPaintWorkletGlobalScope::create):
(WebCore::CSSPaintWorkletGlobalScope::CSSPaintWorkletGlobalScope):
(WebCore::CSSPaintWorkletGlobalScope::devicePixelRatio):
(WebCore::CSSPaintWorkletGlobalScope::addRegisteredPaint):
* css/CSSPaintWorkletGlobalScope.h: Copied from Source/WebCore/css/MediaQueryParserContext.cpp.
(WebCore::CSSPaintWorkletGlobalScope::paintDefinitionMap):
* css/CSSPaintWorkletGlobalScope.idl: Added.
* css/DOMCSSPaintWorklet.cpp: Copied from Source/WebCore/css/MediaQueryParserContext.cpp.
(WebCore::DOMCSSPaintWorklet::ensurePaintWorkletGlobalScope):
(WebCore::DOMCSSPaintWorklet::from):
(WebCore::DOMCSSPaintWorklet::supplementName):
* css/DOMCSSPaintWorklet.h: Copied from Source/WebCore/css/MediaQueryParserContext.cpp.
* css/DOMCSSPaintWorklet.idl: Added.
* css/MediaQueryParserContext.cpp:
* css/StyleBuilder.h:
* dom/Document.cpp:
(WebCore::Document::ensureCSSPaintWorkletGlobalScope):
* dom/Document.h:
* features.json:
* page/RuntimeEnabledFeatures.h:
(WebCore::RuntimeEnabledFeatures::setCSSPaintingAPIEnabled):
(WebCore::RuntimeEnabledFeatures::cssPaintingAPIEnabled const):

Source/WebCore/PAL:

* Configurations/FeatureDefines.xcconfig:

Source/WebKit:

* Configurations/FeatureDefines.xcconfig:
* Shared/WebPreferences.yaml:
* WebProcess/InjectedBundle/InjectedBundle.cpp:
(WebKit::InjectedBundle::overrideBoolPreferenceForTestRunner):

Source/WebKitLegacy/mac:

* Configurations/FeatureDefines.xcconfig:

Tools:

* TestWebKitAPI/Configurations/FeatureDefines.xcconfig:

LayoutTests:

Add basic tests for the CSS Painting API registerPaint function,
and test the custom bindings.

* fast/css-custom-paint/basic-expected.txt: Added.
* fast/css-custom-paint/basic.html: Added.
* fast/css-custom-paint/registerPaintBindings-expected.txt: Added.
* fast/css-custom-paint/registerPaintBindings.html: Added.
* platform/ios-wk1/TestExpectations:
* platform/mac-wk1/TestExpectations:
* platform/win/TestExpectations:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@237192 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years ago[GLib] Build error with RemoteInspector disabled
philn@webkit.org [Tue, 16 Oct 2018 17:54:09 +0000 (17:54 +0000)]
[GLib] Build error with RemoteInspector disabled
https://bugs.webkit.org/show_bug.cgi?id=190623

Reviewed by Michael Catanzaro.

Compilation guards added to ensure the build still works when the
Remote WebInspector has been disabled.

* UIProcess/API/glib/WebKitAutomationSession.cpp:
* UIProcess/API/glib/WebKitAutomationSessionPrivate.h:
* UIProcess/RemoteWebInspectorProxy.cpp:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@237191 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years ago[GTK][WPE] Fix sandbox on distros with suid bubblewrap
commit-queue@webkit.org [Tue, 16 Oct 2018 17:52:24 +0000 (17:52 +0000)]
[GTK][WPE] Fix sandbox on distros with suid bubblewrap
https://bugs.webkit.org/show_bug.cgi?id=190616

Patch by Patrick Griffis <pgriffis@igalia.com> on 2018-10-16
Reviewed by Michael Catanzaro.

* UIProcess/Launcher/glib/BubblewrapLauncher.cpp:
(WebKit::bubblewrapSpawn):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@237190 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoUnreviewed, forgot to add untracked files.
keith_miller@apple.com [Tue, 16 Oct 2018 17:49:00 +0000 (17:49 +0000)]
Unreviewed, forgot to add untracked files.

* llint/LLIntSettingsExtractor.cpp: Added.
(main):
* offlineasm/generate_settings_extractor.rb: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@237189 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoUnreviewed, reland https://bugs.webkit.org/show_bug.cgi?id=189708 with build fix.
keith_miller@apple.com [Tue, 16 Oct 2018 17:44:22 +0000 (17:44 +0000)]
Unreviewed, reland https://bugs.webkit.org/show_bug.cgi?id=189708 with build fix.

* CMakeLists.txt:
* JavaScriptCore.xcodeproj/project.pbxproj:
* llint/LLIntOffsetsExtractor.cpp:
(JSC::LLIntOffsetsExtractor::dummy):
* offlineasm/generate_offset_extractor.rb:
* offlineasm/offsets.rb:
* offlineasm/settings.rb:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@237188 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years ago[WebVTT] Support inline WebVTT styles
pvollan@apple.com [Tue, 16 Oct 2018 17:29:34 +0000 (17:29 +0000)]
[WebVTT] Support inline WebVTT styles
https://bugs.webkit.org/show_bug.cgi?id=190369

Reviewed by Eric Carlson.

Source/WebCore:

Add support for inline WebVTT styles, see https://w3c.github.io/webvtt/#styling. The parsed
style strings from the VTT file are added to an optional vector of style strings in the
TextTrack class. These styles are then added as HTMLStyleElement children to the parent of
the cue span element.

Test: media/track/track-cue-css.html

* html/track/InbandGenericTextTrack.cpp:
(WebCore::InbandGenericTextTrack::newStyleSheetsParsed):
* html/track/InbandGenericTextTrack.h:
* html/track/InbandWebVTTTextTrack.cpp:
(WebCore::InbandWebVTTTextTrack::newStyleSheetsParsed):
* html/track/InbandWebVTTTextTrack.h:
* html/track/LoadableTextTrack.cpp:
(WebCore::LoadableTextTrack::newStyleSheetsAvailable):
* html/track/LoadableTextTrack.h:
* html/track/TextTrack.h:
(WebCore::TextTrack::styleSheets const):
* html/track/VTTCue.cpp:
(WebCore::VTTCue::getDisplayTree):
* html/track/WebVTTParser.cpp:
(WebCore::WebVTTParser::getNewCues):
(WebCore::WebVTTParser::getStyleSheets):
(WebCore::WebVTTParser::parse):
(WebCore::WebVTTParser::collectWebVTTBlock):
(WebCore::WebVTTParser::collectStyleSheet):
(WebCore::WebVTTParser::checkStyleSheet):
(WebCore::WebVTTParser::checkAndStoreStyleSheet):
* html/track/WebVTTParser.h:
* loader/TextTrackLoader.cpp:
(WebCore::TextTrackLoader::newStyleSheetsParsed):
(WebCore::TextTrackLoader::getNewStyleSheets):
* loader/TextTrackLoader.h:

LayoutTests:

* media/track/captions-webvtt/css-styling.vtt: Added.
* media/track/captions-webvtt/no-css-styling.vtt: Added.
* media/track/track-cue-css-expected.html: Added.
* media/track/track-cue-css.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@237187 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoWeb Inspector: sequences of spaces longer than 16 don't show a dot
drousso@apple.com [Tue, 16 Oct 2018 16:47:07 +0000 (16:47 +0000)]
Web Inspector: sequences of spaces longer than 16 don't show a dot
https://bugs.webkit.org/show_bug.cgi?id=190528

Reviewed by Matt Baker.

* UserInterface/Views/CodeMirrorAdditions.js:
Create a new CSS rule and add it to a unique <style> whenever a whitespace sequence with
a length is seen for the first time.

* UserInterface/Views/CodeMirrorOverrides.css:
(.show-whitespace-characters .CodeMirror .cm-whitespace-1::before): Deleted.
(.show-whitespace-characters .CodeMirror .cm-whitespace-2::before): Deleted.
(.show-whitespace-characters .CodeMirror .cm-whitespace-3::before): Deleted.
(.show-whitespace-characters .CodeMirror .cm-whitespace-4::before): Deleted.
(.show-whitespace-characters .CodeMirror .cm-whitespace-5::before): Deleted.
(.show-whitespace-characters .CodeMirror .cm-whitespace-6::before): Deleted.
(.show-whitespace-characters .CodeMirror .cm-whitespace-7::before): Deleted.
(.show-whitespace-characters .CodeMirror .cm-whitespace-8::before): Deleted.
(.show-whitespace-characters .CodeMirror .cm-whitespace-9::before): Deleted.
(.show-whitespace-characters .CodeMirror .cm-whitespace-10::before): Deleted.
(.show-whitespace-characters .CodeMirror .cm-whitespace-11::before): Deleted.
(.show-whitespace-characters .CodeMirror .cm-whitespace-12::before): Deleted.
(.show-whitespace-characters .CodeMirror .cm-whitespace-13::before): Deleted.
(.show-whitespace-characters .CodeMirror .cm-whitespace-14::before): Deleted.
(.show-whitespace-characters .CodeMirror .cm-whitespace-15::before): Deleted.
(.show-whitespace-characters .CodeMirror .cm-whitespace-16::before): Deleted.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@237186 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agowindow.navigator should not become null after the window loses its browsing context
cdumez@apple.com [Tue, 16 Oct 2018 16:24:12 +0000 (16:24 +0000)]
window.navigator should not become null after the window loses its browsing context
https://bugs.webkit.org/show_bug.cgi?id=190595

Reviewed by Ryosuke Niwa.

LayoutTests/imported/w3c:

Rebaseline test which is not failing differently. The last check of this test is checking that
navigator.serviceWorker returns null after the frame has been detached. The test has been written
this way because this is how Chromium behaves. However, Firefox keeps returning the
ServiceWorkerContainer, as we do. Also, the specification indicates the the attribute cannot
return null (since the attribute is not nullable):
- https://w3c.github.io/ServiceWorker/#navigator-serviceworker

* web-platform-tests/service-workers/service-worker/detached-context.https-expected.txt:

Source/WebCore:

window.navigator should not become null after the window loses its browsing context.
This does not match the HTML specification or the behavior of other browsers.

No new tests, updated existing tests.

* Modules/geolocation/NavigatorGeolocation.cpp:
(WebCore::NavigatorGeolocation::geolocation const):
* page/DOMWindow.cpp:
(WebCore::DOMWindow::navigator):
* page/Navigator.cpp:
(WebCore::Navigator::Navigator):
* page/Navigator.h:
* page/NavigatorBase.cpp:
(WebCore::NavigatorBase::NavigatorBase):
* page/NavigatorBase.h:
* page/WorkerNavigator.cpp:
(WebCore::WorkerNavigator::WorkerNavigator):
* workers/service/ServiceWorkerContainer.cpp:
(WebCore::ServiceWorkerContainer::ServiceWorkerContainer):
* workers/service/ServiceWorkerContainer.h:

LayoutTests:

Extend layout test coverage.

* fast/frames/detached-frame-property-expected.txt:
* fast/frames/detached-frame-property.html:
* http/tests/dom/cross-origin-detached-window-properties-expected.txt:
* http/tests/dom/cross-origin-detached-window-properties.html:
* http/tests/dom/same-origin-detached-window-properties-expected.txt:
* http/tests/dom/same-origin-detached-window-properties.html:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@237185 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoReplace HistoryItem* with HistoryItem& where possible
achristensen@apple.com [Tue, 16 Oct 2018 15:58:10 +0000 (15:58 +0000)]
Replace HistoryItem* with HistoryItem& where possible
https://bugs.webkit.org/show_bug.cgi?id=190617

Reviewed by Chris Dumez.

Source/WebCore:

* history/BackForwardClient.h:
* history/BackForwardController.cpp:
(WebCore::BackForwardController::setCurrentItem):
* history/BackForwardController.h:
* history/HistoryItem.cpp:
(WebCore::defaultNotifyHistoryItemChanged):
(WebCore::HistoryItem::setAlternateTitle):
(WebCore::HistoryItem::setURLString):
(WebCore::HistoryItem::setOriginalURLString):
(WebCore::HistoryItem::setReferrer):
(WebCore::HistoryItem::setTitle):
(WebCore::HistoryItem::setTarget):
(WebCore::HistoryItem::setShouldRestoreScrollPosition):
(WebCore::HistoryItem::setStateObject):
(WebCore::HistoryItem::notifyChanged):
* history/HistoryItem.h:
* loader/EmptyClients.cpp:
* loader/FrameLoader.cpp:
(WebCore::FrameLoader::checkLoadCompleteForThisFrame):
(WebCore::FrameLoader::continueLoadAfterNavigationPolicy):
(WebCore::FrameLoader::loadSameDocumentItem):
* loader/HistoryController.cpp:
(WebCore::HistoryController::goToItem):
(WebCore::HistoryController::updateForCommit):
(WebCore::HistoryController::recursiveUpdateForCommit):
(WebCore::HistoryController::recursiveUpdateForSameDocumentNavigation):
(WebCore::HistoryController::setCurrentItem):
(WebCore::HistoryController::createItem):
(WebCore::HistoryController::itemsAreClones const):
(WebCore::HistoryController::currentFramesMatchItem const):
* loader/HistoryController.h:

Source/WebKit:

* WebProcess/WebCoreSupport/SessionStateConversion.h:
* WebProcess/WebPage/WebBackForwardListProxy.cpp:
(WebKit::WK2NotifyHistoryItemChanged):
(WebKit::WebBackForwardListProxy::goToItem):
* WebProcess/WebPage/WebBackForwardListProxy.h:

Source/WebKitLegacy/mac:

* History/BackForwardList.h:
* History/BackForwardList.mm:
(BackForwardList::goToItem):
* History/WebBackForwardList.mm:
(-[WebBackForwardList goToItem:]):
* History/WebHistoryItem.mm:
(WKNotifyHistoryItemChanged):
* History/WebHistoryItemInternal.h:

Source/WebKitLegacy/win:

* BackForwardList.cpp:
(BackForwardList::goToItem):
* BackForwardList.h:
* WebBackForwardList.cpp:
(WebBackForwardList::goToItem):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@237184 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years ago[EME] Add some ClearKey baselines for passing tests
cturner@igalia.com [Tue, 16 Oct 2018 15:49:15 +0000 (15:49 +0000)]
[EME] Add some ClearKey baselines for passing tests
https://bugs.webkit.org/show_bug.cgi?id=190579

Reviewed by Unreviewed test gardening.

* web-platform-tests/encrypted-media/clearkey-check-initdata-type.https-expected.txt: Added.
* web-platform-tests/encrypted-media/clearkey-events-session-closed-event.https-expected.txt: Added.
* web-platform-tests/encrypted-media/clearkey-events.https-expected.txt: Added.
* web-platform-tests/encrypted-media/clearkey-invalid-license.https-expected.txt: Added.
* web-platform-tests/encrypted-media/clearkey-keystatuses-multiple-sessions.https-expected.txt: Added.
* web-platform-tests/encrypted-media/clearkey-keystatuses.https-expected.txt: Added.
* web-platform-tests/encrypted-media/clearkey-mp4-setmediakeys-at-same-time.https-expected.txt: Added.
* web-platform-tests/encrypted-media/clearkey-mp4-setmediakeys-multiple-times-with-the-same-mediakeys.https-expected.txt: Added.
* web-platform-tests/encrypted-media/clearkey-mp4-setmediakeys-to-multiple-video-elements.https-expected.txt: Added.
* web-platform-tests/encrypted-media/clearkey-mp4-setmediakeys.https-expected.txt: Added.
* web-platform-tests/encrypted-media/clearkey-mp4-syntax-mediakeysession.https-expected.txt: Added.
* web-platform-tests/encrypted-media/clearkey-mp4-syntax-mediakeysystemaccess.https-expected.txt: Added.
* web-platform-tests/encrypted-media/clearkey-not-callable-after-createsession.https-expected.txt: Added.
* web-platform-tests/encrypted-media/clearkey-update-non-ascii-input.https-expected.txt: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@237183 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoRemove unused WebHistoryItem._transientPropertyForKey
achristensen@apple.com [Tue, 16 Oct 2018 15:39:42 +0000 (15:39 +0000)]
Remove unused WebHistoryItem._transientPropertyForKey
https://bugs.webkit.org/show_bug.cgi?id=190606

Reviewed by Chris Dumez.

Source/WebCore:

This is unnecessary complexity in HistoryItem.

* history/HistoryItem.h:
* history/mac/HistoryItemMac.mm:
(WebCore::HistoryItem::getTransientProperty const): Deleted.
(WebCore::HistoryItem::setTransientProperty): Deleted.

Source/WebKitLegacy/mac:

* History/WebHistoryItem.mm:
* History/WebHistoryItemPrivate.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@237182 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years ago[Curl] Separate SSL Class for platform dependent way of setup.
Basuke.Suzuki@sony.com [Tue, 16 Oct 2018 15:21:16 +0000 (15:21 +0000)]
[Curl] Separate SSL Class for platform dependent way of setup.
https://bugs.webkit.org/show_bug.cgi?id=190597

Reviewed by Alex Christensen.

SSL setup is very platform dependent. This patch separate the class to allow
platform-dependent setup at the SSL handle instantiation.

No new tests because there's no behaviro changes.

* PlatformWinCairo.cmake:
* platform/network/curl/CurlSSLHandle.cpp:
(WebCore::CurlSSLHandle::CurlSSLHandle):
(WebCore::CurlSSLHandle::getCACertPathEnv): Moved.
* platform/network/curl/CurlSSLHandle.h:
* platform/network/win/CurlSSLHandleWin.cpp:
(WebCore::getCACertPathEnv):
(WebCore::CurlSSLHandle::platformInitialize):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@237181 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years ago[EME] Multiple ClearKey tests crashing in gst_qtdemux_request_protection_context
cturner@igalia.com [Tue, 16 Oct 2018 14:54:24 +0000 (14:54 +0000)]
[EME] Multiple ClearKey tests crashing in gst_qtdemux_request_protection_context
https://bugs.webkit.org/show_bug.cgi?id=190577

Reviewed by Michael Catanzaro.

* platform/gtk/TestExpectations:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@237180 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoUnreviewed, GTK Ubuntu LTS Jhbuild prospective build fix
philn@webkit.org [Tue, 16 Oct 2018 11:01:36 +0000 (11:01 +0000)]
Unreviewed, GTK Ubuntu LTS Jhbuild prospective build fix

* gstreamer/jhbuild.modules:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@237179 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years ago[GTK][JHBuild] Disable libmount support in glib
philn@webkit.org [Tue, 16 Oct 2018 10:04:01 +0000 (10:04 +0000)]
[GTK][JHBuild] Disable libmount support in glib

Rubber-stamped by Alejandro G. Castro.

This feature doesn't seem required for our setup.

* gtk/jhbuild.modules:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@237178 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoUnreviewed, another GTK build fix after r237146
philn@webkit.org [Tue, 16 Oct 2018 09:44:08 +0000 (09:44 +0000)]
Unreviewed, another GTK build fix after r237146

* UIProcess/gtk/RemoteWebInspectorProxyGtk.cpp:
(WebKit::RemoteWebInspectorProxy::platformCreateFrontendPageAndWindow):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@237177 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoUnreviewed, GTK bots build fix
philn@webkit.org [Tue, 16 Oct 2018 08:47:39 +0000 (08:47 +0000)]
Unreviewed, GTK bots build fix

* Source/cmake/OptionsGTK.cmake: Disable Sandbox support until the
bots have all the needed dependencies installed.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@237176 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoUnreviewed, add missing include.
keith_miller@apple.com [Tue, 16 Oct 2018 07:52:11 +0000 (07:52 +0000)]
Unreviewed, add missing include.

* runtime/BasicBlockLocation.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@237175 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years ago[GTK][WPE] Make libwebrtc compile using the system opus library
commit-queue@webkit.org [Tue, 16 Oct 2018 07:39:06 +0000 (07:39 +0000)]
[GTK][WPE] Make libwebrtc compile using the system opus library
https://bugs.webkit.org/show_bug.cgi?id=190573

Patch by Alejandro G. Castro <alex@igalia.com> on 2018-10-16
Reviewed by Philippe Normand.

We found some situations where gstreamer gets confused when it
tries to use opus because it finds opus symbols compiled for
liwebrtc. We are going to try the option to use the system opus
library also for libwebrtc.

* CMakeLists.txt: Added opus dependency.
* cmake/FindOpus.cmake: Added the hints to find the opus library
in the compilation.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@237174 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoSupport arm64 CPUs with a 32-bit address space
keith_miller@apple.com [Tue, 16 Oct 2018 07:19:13 +0000 (07:19 +0000)]
Support arm64 CPUs with a 32-bit address space
https://bugs.webkit.org/show_bug.cgi?id=190273

Reviewed by Michael Saboff.

Source/JavaScriptCore:

This patch adds support for arm64_32 in the LLInt. In order to
make this work we needed to add a new type that reflects the size
of a cpu register. This type is called CPURegister or UCPURegister
for the unsigned version. Most places that used void* or intptr_t
to refer to a register have been changed to use this new type.

* JavaScriptCore.xcodeproj/project.pbxproj:
* assembler/ARM64Assembler.h:
(JSC::isInt):
(JSC::is4ByteAligned):
(JSC::PairPostIndex::PairPostIndex):
(JSC::PairPreIndex::PairPreIndex):
(JSC::ARM64Assembler::readPointer):
(JSC::ARM64Assembler::readCallTarget):
(JSC::ARM64Assembler::computeJumpType):
(JSC::ARM64Assembler::linkCompareAndBranch):
(JSC::ARM64Assembler::linkConditionalBranch):
(JSC::ARM64Assembler::linkTestAndBranch):
(JSC::ARM64Assembler::loadRegisterLiteral):
(JSC::ARM64Assembler::loadStoreRegisterPairPostIndex):
(JSC::ARM64Assembler::loadStoreRegisterPairPreIndex):
(JSC::ARM64Assembler::loadStoreRegisterPairOffset):
(JSC::ARM64Assembler::loadStoreRegisterPairNonTemporal):
(JSC::isInt7): Deleted.
(JSC::isInt11): Deleted.
* assembler/CPU.h:
(JSC::isAddress64Bit):
(JSC::isAddress32Bit):
* assembler/MacroAssembler.h:
(JSC::MacroAssembler::shouldBlind):
* assembler/MacroAssemblerARM64.cpp:
(JSC::MacroAssemblerARM64::collectCPUFeatures):
* assembler/MacroAssemblerARM64.h:
(JSC::MacroAssemblerARM64::load):
(JSC::MacroAssemblerARM64::store):
(JSC::MacroAssemblerARM64::isInIntRange): Deleted.
* assembler/Printer.h:
* assembler/ProbeContext.h:
(JSC::Probe::CPUState::gpr):
(JSC::Probe::CPUState::spr):
(JSC::Probe::Context::gpr):
(JSC::Probe::Context::spr):
* b3/B3ConstPtrValue.h:
* b3/B3StackmapSpecial.cpp:
(JSC::B3::StackmapSpecial::isArgValidForRep):
* b3/air/AirArg.h:
(JSC::B3::Air::Arg::stackSlot const):
(JSC::B3::Air::Arg::special const):
* b3/air/testair.cpp:
* b3/testb3.cpp:
(JSC::B3::testStoreConstantPtr):
(JSC::B3::testInterpreter):
(JSC::B3::testAddShl32):
(JSC::B3::testLoadBaseIndexShift32):
* bindings/ScriptFunctionCall.cpp:
(Deprecated::ScriptCallArgumentHandler::appendArgument):
* bindings/ScriptFunctionCall.h:
* bytecode/CodeBlock.cpp:
(JSC::roundCalleeSaveSpaceAsVirtualRegisters):
* dfg/DFGOSRExit.cpp:
(JSC::DFG::restoreCalleeSavesFor):
(JSC::DFG::saveCalleeSavesFor):
(JSC::DFG::restoreCalleeSavesFromVMEntryFrameCalleeSavesBuffer):
(JSC::DFG::copyCalleeSavesToVMEntryFrameCalleeSavesBuffer):
* dfg/DFGOSRExitCompilerCommon.cpp:
(JSC::DFG::reifyInlinedCallFrames):
* dfg/DFGSpeculativeJIT64.cpp:
(JSC::DFG::SpeculativeJIT::compile):
* disassembler/UDis86Disassembler.cpp:
(JSC::tryToDisassembleWithUDis86):
* ftl/FTLLowerDFGToB3.cpp:
(JSC::FTL::DFG::LowerDFGToB3::compileWeakMapGet):
* heap/MachineStackMarker.cpp:
(JSC::copyMemory):
* interpreter/CallFrame.h:
(JSC::ExecState::returnPC const):
(JSC::ExecState::hasReturnPC const):
(JSC::ExecState::clearReturnPC):
(JSC::ExecState::returnPCOffset):
(JSC::ExecState::isGlobalExec const):
(JSC::ExecState::setReturnPC):
* interpreter/CalleeBits.h:
(JSC::CalleeBits::boxWasm):
(JSC::CalleeBits::isWasm const):
(JSC::CalleeBits::asWasmCallee const):
* interpreter/Interpreter.cpp:
(JSC::UnwindFunctor::copyCalleeSavesToEntryFrameCalleeSavesBuffer const):
* interpreter/VMEntryRecord.h:
* jit/AssemblyHelpers.h:
(JSC::AssemblyHelpers::clearStackFrame):
* jit/RegisterAtOffset.h:
(JSC::RegisterAtOffset::offsetAsIndex const):
* jit/RegisterAtOffsetList.cpp:
(JSC::RegisterAtOffsetList::RegisterAtOffsetList):
* llint/LLIntData.cpp:
(JSC::LLInt::Data::performAssertions):
* llint/LLIntOfflineAsmConfig.h:
* llint/LowLevelInterpreter.asm:
* llint/LowLevelInterpreter64.asm:
* offlineasm/arm64.rb:
* offlineasm/asm.rb:
* offlineasm/ast.rb:
* offlineasm/backends.rb:
* offlineasm/parser.rb:
* offlineasm/x86.rb:
* runtime/BasicBlockLocation.cpp:
(JSC::BasicBlockLocation::dumpData const):
(JSC::BasicBlockLocation::emitExecuteCode const):
* runtime/BasicBlockLocation.h:
* runtime/HasOwnPropertyCache.h:
* runtime/JSBigInt.cpp:
(JSC::JSBigInt::inplaceMultiplyAdd):
(JSC::JSBigInt::digitDiv):
* runtime/JSBigInt.h:
* runtime/JSObject.h:
* runtime/Options.cpp:
(JSC::jitEnabledByDefault):
* runtime/Options.h:
* runtime/RegExp.cpp:
(JSC::RegExp::printTraceData):
* runtime/SamplingProfiler.cpp:
(JSC::CFrameWalker::walk):
* runtime/SlowPathReturnType.h:
(JSC::encodeResult):
(JSC::decodeResult):
* tools/SigillCrashAnalyzer.cpp:
(JSC::SigillCrashAnalyzer::dumpCodeBlock):

Source/WebCore:

Fix missing namespace annotation.

* cssjit/SelectorCompiler.cpp:
(WebCore::SelectorCompiler::SelectorCodeGenerator::generateAddStyleRelation):

Source/WTF:

Use WTF_CPU_ADDRESS64/32 to decide if the system is running on arm64_32.

* wtf/MathExtras.h:
(getLSBSet):
* wtf/Platform.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@237173 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoAdd WebGPU 2018 feature flag and experimental feature flag
justin_fan@apple.com [Tue, 16 Oct 2018 04:57:47 +0000 (04:57 +0000)]
Add WebGPU 2018 feature flag and experimental feature flag
https://bugs.webkit.org/show_bug.cgi?id=190509

Reviewed by Dean Jackson.

.:

Update FeatureDefines for WebGPU -> WebMetal renaming, while preserving WebGPU
references for 2018 WebGPU Sketch implementation.

* Source/cmake/OptionsMac.cmake:
* Source/cmake/tools/vsprops/FeatureDefines.props:
* Source/cmake/tools/vsprops/FeatureDefinesCairo.props:

Source/JavaScriptCore:

Re-add ENABLE_WEBGPU, an experimental feature flag, and a RuntimeEnabledFeature
for the 2018 WebGPU prototype.

* Configurations/FeatureDefines.xcconfig:

Source/WebCore:

Re-add ENABLE_WEBGPU, an experimental feature flag, and a RuntimeEnabledFeature
for the 2018 WebGPU prototype.

* Configurations/FeatureDefines.xcconfig:
* page/RuntimeEnabledFeatures.h:
(WebCore::RuntimeEnabledFeatures::setWebGPUEnabled):
(WebCore::RuntimeEnabledFeatures::webGPUEnabled const):

Source/WebCore/PAL:

Re-add ENABLE_WEBGPU, an experimental feature flag, and a RuntimeEnabledFeature
for the 2018 WebGPU prototype.

* Configurations/FeatureDefines.xcconfig:

Source/WebKit:

Re-add ENABLE_WEBGPU, an experimental feature flag, and a RuntimeEnabledFeature
for the 2018 WebGPU prototype.

* Configurations/FeatureDefines.xcconfig:
* Shared/WebPreferences.yaml:
* WebProcess/InjectedBundle/InjectedBundle.cpp:
(WebKit::InjectedBundle::overrideBoolPreferenceForTestRunner):

Source/WebKitLegacy/mac:

Re-add ENABLE_WEBGPU, an experimental feature flag, and a RuntimeEnabledFeature
for the 2018 WebGPU prototype.

* Configurations/FeatureDefines.xcconfig:
* WebView/WebPreferenceKeysPrivate.h:
* WebView/WebPreferences.mm:
(+[WebPreferences initialize]):
(-[WebPreferences webGPUEnabled]):
(-[WebPreferences setWebGPUEnabled:]):
* WebView/WebPreferencesPrivate.h:
* WebView/WebView.mm:
(-[WebView _preferencesChanged:]):

Tools:

Re-add ENABLE_WEBGPU, an experimental feature flag, and a RuntimeEnabledFeature
for the 2018 WebGPU prototype.

* DumpRenderTree/TestOptions.cpp:
(TestOptions::TestOptions):
* DumpRenderTree/TestOptions.h:
* DumpRenderTree/mac/DumpRenderTree.mm:
(setWebPreferencesForTestOptions):
* TestWebKitAPI/Configurations/FeatureDefines.xcconfig:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@237170 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoSkip stress/array-prototype-concat-of-long-spliced-arrays2.js on arm and mips/linux
guijemont@igalia.com [Tue, 16 Oct 2018 01:30:21 +0000 (01:30 +0000)]
Skip stress/array-prototype-concat-of-long-spliced-arrays2.js on arm and mips/linux

Unreviewed gardening.

* stress/array-prototype-concat-of-long-spliced-arrays2.js:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@237169 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoUnreviewed, rolling out r237162.
commit-queue@webkit.org [Tue, 16 Oct 2018 01:15:27 +0000 (01:15 +0000)]
Unreviewed, rolling out r237162.
https://bugs.webkit.org/show_bug.cgi?id=190612

Broke Windows build (Requested by dolmstead on #webkit).

Reverted changeset:

"Add stub implementation for accessibility objects"
https://bugs.webkit.org/show_bug.cgi?id=190608
https://trac.webkit.org/changeset/237162

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@237168 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoFix API tests after r237146
achristensen@apple.com [Tue, 16 Oct 2018 01:12:59 +0000 (01:12 +0000)]
Fix API tests after r237146

* Shared/WebPageGroupData.cpp:
(WebKit::WebPageGroupData::decode):
I accidentally changed the decoder.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@237167 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoHSTS cache entries cannot be removed by origin
ryanhaddad@apple.com [Tue, 16 Oct 2018 01:04:37 +0000 (01:04 +0000)]
HSTS cache entries cannot be removed by origin
https://bugs.webkit.org/show_bug.cgi?id=190455
<rdar://problem/45181322>

Patch by Sihui Liu <sihui_liu@apple.com> on 2018-10-15
Reviewed by Chris Dumez.

The SPI _CFNetworkResetHSTS does not recognize url without protocol, so we need to add protocol manually.

* NetworkProcess/cocoa/NetworkProcessCocoa.mm:
(WebKit::NetworkProcess::deleteHSTSCacheForHostNames):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@237166 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoSource/WebKit:
achristensen@apple.com [Tue, 16 Oct 2018 01:02:54 +0000 (01:02 +0000)]
Source/WebKit:
Fix GTK build, even though bots are not working right now.

* UIProcess/gtk/WebInspectorProxyGtk.cpp:
(WebKit::WebInspectorProxy::platformCreateFrontendPage):

Source/WebKitLegacy/win:
Fix Windows build.

* WebView.cpp:
(WebView::loadBackForwardListFromOtherView):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@237165 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoFix WinCairo build.
achristensen@apple.com [Tue, 16 Oct 2018 00:56:23 +0000 (00:56 +0000)]
Fix WinCairo build.

* WebView.cpp:
(WebView::backForwardList):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@237164 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years ago[CG] Adopt CG SPI for non-even cornered rounded rects
commit-queue@webkit.org [Tue, 16 Oct 2018 00:53:13 +0000 (00:53 +0000)]
[CG] Adopt CG SPI for non-even cornered rounded rects
https://bugs.webkit.org/show_bug.cgi?id=190155

Patch by Said Abou-Hallawa <sabouhallawa@apple.com> on 2018-10-15
Reviewed by Simon Fraser.

Source/WebCore:

Instead of creating bezier curves for the non-even corners of the rounded
rects, we should use the optimized SPI provided by CG.

* platform/graphics/cg/PathCG.cpp:
(WebCore::Path::platformAddPathForRoundedRect):

Source/WebCore/PAL:

* pal/spi/cg/CoreGraphicsSPI.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@237163 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoAdd stub implementation for accessibility objects
don.olmstead@sony.com [Tue, 16 Oct 2018 00:46:37 +0000 (00:46 +0000)]
Add stub implementation for accessibility objects
https://bugs.webkit.org/show_bug.cgi?id=190608

Reviewed by Michael Catanzaro.

No new tests. No change in behavior.

Add default implementation to associated cpp file. Removes the WPE
specific stubs.

* SourcesWPE.txt:
* accessibility/AXObjectCache.cpp:
(WebCore::AXObjectCache::detachWrapper):
(WebCore::AXObjectCache::attachWrapper):
(WebCore::AXObjectCache::postPlatformNotification):
(WebCore::AXObjectCache::nodeTextChangePlatformNotification):
(WebCore::AXObjectCache::frameLoadingEventPlatformNotification):
(WebCore::AXObjectCache::platformHandleFocusedUIElementChanged):
(WebCore::AXObjectCache::handleScrolledToAnchor):
* accessibility/AccessibilityObject.cpp:
(WebCore::AccessibilityObject::accessibilityIgnoreAttachment const):
(WebCore::AccessibilityObject::accessibilityPlatformIncludesObject const):
* accessibility/AccessibilityObject.h:
* accessibility/wpe/AXObjectCacheWPE.cpp: Removed.
* accessibility/wpe/AccessibilityObjectWPE.cpp: Removed.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@237162 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoFix WinCairo build after r237146
achristensen@apple.com [Tue, 16 Oct 2018 00:38:30 +0000 (00:38 +0000)]
Fix WinCairo build after r237146

* UIProcess/win/WebInspectorProxyWin.cpp:
(WebKit::WebInspectorProxy::platformCreateFrontendPage):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@237161 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoFix iOS build after r237157
achristensen@apple.com [Tue, 16 Oct 2018 00:34:02 +0000 (00:34 +0000)]
Fix iOS build after r237157

* WebView/WebView.mm:
(-[WebView initSimpleHTMLDocumentWithStyle:frame:preferences:groupName:]):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@237160 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoREGRESSION (r236512): [ Mac WK2 ] Layout Test editing/undo/undo-smart-delete-word...
cdumez@apple.com [Tue, 16 Oct 2018 00:32:56 +0000 (00:32 +0000)]
REGRESSION (r236512): [ Mac WK2 ] Layout Test editing/undo/undo-smart-delete-word.html is flaky
https://bugs.webkit.org/show_bug.cgi?id=190375
<rdar://problem/45177807>

Reviewed by Ryosuke Niwa.

The test is doing the following:
> document.execCommand("Delete"); document.execCommand("Undo");

Both operation are synchronous and rely on synchronous IPC from the WebProcess to the UIProcess.
However, for the undo operation to succeed, the first command need to have been registered with
with the UIProcess via the WebPageProxy::RegisterEditCommandForUndo IPC to the UIProcess, which
is asynchronous.

After r236512, the UIProcess no longer processes incoming sync / async IPC in order and thus,
it is possible for the WebPageProxy::ExecuteUndoRedo synchronous IPC to get processed by the
UIProcess *before* the WebPageProxy::RegisterEditCommandForUndo asynchronous IPC for the
previous "Delete" command. As a result, canUndo would return false and we would fail to undo.

To address the issue, use SendOption::DispatchMessageEvenWhenWaitingForSyncReply flag when sending
the WebPageProxy::RegisterEditCommandForUndo IPC, so that it gets processed in order with regards
to surrounding synchronous IPC.

* WebProcess/WebCoreSupport/WebEditorClient.cpp:
(WebKit::WebEditorClient::registerUndoStep):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@237159 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoUpdate bot configuration for Sierra archive queues
ryanhaddad@apple.com [Tue, 16 Oct 2018 00:30:49 +0000 (00:30 +0000)]
Update bot configuration for Sierra archive queues
https://bugs.webkit.org/show_bug.cgi?id=190605

Reviewed by Aakash Jain.

* BuildSlaveSupport/build.webkit.org-config/config.json:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@237158 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoModernize BackForwardClient.h
achristensen@apple.com [Tue, 16 Oct 2018 00:14:41 +0000 (00:14 +0000)]
Modernize BackForwardClient.h
https://bugs.webkit.org/show_bug.cgi?id=190610

Reviewed by Chris Dumez.

Source/WebCore:

* editing/markup.cpp:
(WebCore::createPageForSanitizingWebContent):
* history/BackForwardClient.h:
* history/BackForwardController.h:
(WebCore::BackForwardController::client):
(WebCore::BackForwardController::client const):
* inspector/InspectorOverlay.cpp:
(WebCore::InspectorOverlay::overlayPage):
* loader/EmptyClients.cpp:
(WebCore::pageConfigurationWithEmptyClients):
(WebCore::createEmptyFrameNetworkingContext): Deleted.
(WebCore::fillWithEmptyClients): Deleted.
(WebCore::createEmptyEditorClient): Deleted.
* loader/EmptyClients.h:
* page/Page.cpp:
(WebCore::Page::Page):
* page/PageConfiguration.cpp:
(WebCore::PageConfiguration::PageConfiguration):
* page/PageConfiguration.h:
* svg/graphics/SVGImage.cpp:
(WebCore::SVGImage::dataChanged):

Source/WebKit:

* WebProcess/InjectedBundle/InjectedBundleBackForwardList.cpp:
(WebKit::InjectedBundleBackForwardList::clear):
* WebProcess/Storage/WebSWContextManagerConnection.cpp:
(WebKit::WebSWContextManagerConnection::installServiceWorker):
* WebProcess/WebPage/WebBackForwardListProxy.cpp:
(WebKit::WebBackForwardListProxy::WebBackForwardListProxy):
(WebKit::WebBackForwardListProxy::backListCount const):
(WebKit::WebBackForwardListProxy::forwardListCount const):
(WebKit::WebBackForwardListProxy::backListCount): Deleted.
(WebKit::WebBackForwardListProxy::forwardListCount): Deleted.
* WebProcess/WebPage/WebBackForwardListProxy.h:
(WebKit::WebBackForwardListProxy::create):
* WebProcess/WebPage/WebPage.cpp:
(WebKit::m_cpuLimit):
(WebKit::WebPage::restoreSessionInternal):

Source/WebKitLegacy/mac:

* History/BackForwardList.h:
* History/BackForwardList.mm:
(BackForwardList::backListCount const):
(BackForwardList::forwardListCount const):
(BackForwardList::backListCount): Deleted.
(BackForwardList::forwardListCount): Deleted.
* History/WebBackForwardList.mm:
(-[WebBackForwardList init]):
* WebCoreSupport/WebFrameLoaderClient.mm:
(WebFrameLoaderClient::canCachePage const):
* WebView/WebFrameView.mm:
(-[WebFrameView keyDown:keyDown:]):
* WebView/WebView.mm:
(-[WebView _commonInitializationWithFrameName:groupName:]):
(-[WebView initSimpleHTMLDocumentWithStyle:frame:preferences:groupName:]):
(-[WebView _loadBackForwardListFromOtherView:]):
(-[WebView initWithCoder:]):
(-[WebView encodeWithCoder:]):
(-[WebView backForwardList]):
(-[WebView setMaintainsBackForwardList:]):

Source/WebKitLegacy/win:

* BackForwardList.cpp:
(BackForwardList::backListCount const):
(BackForwardList::forwardListCount const):
(BackForwardList::backListCount): Deleted.
(BackForwardList::forwardListCount): Deleted.
* BackForwardList.h:
* WebView.cpp:
(WebView::initWithFrame):
(WebView::backForwardList):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@237157 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoAdd support for prefers-color-scheme media query
timothy@apple.com [Mon, 15 Oct 2018 23:43:02 +0000 (23:43 +0000)]
Add support for prefers-color-scheme media query
https://bugs.webkit.org/show_bug.cgi?id=190499
rdar://problem/45212025

Reviewed by Dean Jackson.

Source/JavaScriptCore:

* Configurations/FeatureDefines.xcconfig: Added ENABLE_DARK_MODE_CSS.

Source/WebCore:

Test: css-dark-mode/prefers-color-scheme.html

* Configurations/FeatureDefines.xcconfig: Added ENABLE_DARK_MODE_CSS.
* css/CSSValueKeywords.in: Added light and dark.
* css/MediaFeatureNames.h: Added prefers-color-scheme.
* css/MediaQueryEvaluator.cpp:
(WebCore::prefersColorSchemeEvaluate): Added.
* css/MediaQueryExpression.cpp:
(WebCore::featureWithValidIdent): Added prefers-color-scheme.
* page/RuntimeEnabledFeatures.h:
(WebCore::RuntimeEnabledFeatures::setDarkModeCSSEnabled): Added.
(WebCore::RuntimeEnabledFeatures::darkModeCSSEnabled const): Added.
* testing/InternalSettings.cpp:
(WebCore::InternalSettings::resetToConsistentState): Reset setUseDarkAppearance.
(WebCore::InternalSettings::setUseDarkAppearance): Added.
* testing/InternalSettings.h: Added setUseDarkAppearance.
* testing/InternalSettings.idl: Ditto.

Source/WebCore/PAL:

* Configurations/FeatureDefines.xcconfig: Added ENABLE_DARK_MODE_CSS.

Source/WebKit:

* Configurations/FeatureDefines.xcconfig: Added ENABLE_DARK_MODE_CSS.
* Shared/WebPreferences.yaml: Added DarkModeCSSEnabled as experimental.

Source/WebKitLegacy/mac:

* Configurations/FeatureDefines.xcconfig: Added ENABLE_DARK_MODE_CSS.

Source/WTF:

* wtf/FeatureDefines.h: Added ENABLE_DARK_MODE_CSS.

Tools:

* Scripts/webkitperl/FeatureList.pm: Added ENABLE_DARK_MODE_CSS as dark-mode-css.
* TestWebKitAPI/Configurations/FeatureDefines.xcconfig: Added ENABLE_DARK_MODE_CSS.

LayoutTests:

* css-dark-mode/prefers-color-scheme-expected.txt: Added.
* css-dark-mode/prefers-color-scheme.html: Added.
* platform/gtk/TestExpectations: Skip css-dark-mode.
* platform/ios/TestExpectations: Skip css-dark-mode.
* platform/mac-wk1/TestExpectations: Skip css-dark-mode.
* platform/win/TestExpectations: Skip css-dark-mode.
* platform/wincairo/TestExpectations: Skip css-dark-mode.
* platform/wpe/TestExpectations: Skip css-dark-mode.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@237156 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoWeb Inspector: Cleanup Table test utilities
mattbaker@apple.com [Mon, 15 Oct 2018 23:29:28 +0000 (23:29 +0000)]
Web Inspector: Cleanup Table test utilities
https://bugs.webkit.org/show_bug.cgi?id=190538
<rdar://problem/45237730>

Reviewed by Joseph Pecoraro.

Cleanup Table test utilities by exposing removing the artificial dependency
between TableDelegate and TableDataSource, and exposing a cleaner set of
utility methods to InspectorTest. This patch changes `InspectorTest.createTable`
to take no parameters, and adds `InspectorTest.createTableWithDelegate` for
tests that need to supply a custom delegate.

* inspector/table/resources/table-utilities.js:
(TestPage.registerInitializer.InspectorTest.TableDataSource):
(TestPage.registerInitializer.InspectorTest.TableDelegate):
(TestPage.registerInitializer.InspectorTest.TableDelegate.prototype.tablePopulateCell):
(TestPage.registerInitializer.createDataSource):
(TestPage.registerInitializer.createTableInternal):
(TestPage.registerInitializer.InspectorTest.createTable):
(TestPage.registerInitializer.InspectorTest.createTableWithDelegate):
(TestPage.registerInitializer):

* inspector/table/table-selection-expected.txt:
Update with new test suite name.

* inspector/table/table-selection.html:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@237155 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoRemove StorageProcess leftover
commit-queue@webkit.org [Mon, 15 Oct 2018 23:18:52 +0000 (23:18 +0000)]
Remove StorageProcess leftover
https://bugs.webkit.org/show_bug.cgi?id=190591

Patch by Patrick Griffis <pgriffis@igalia.com> on 2018-10-15
Reviewed by Michael Catanzaro.

The process was deleted.

* Shared/ChildProcess.h:
* Shared/mac/ChildProcessMac.mm:
(WebKit::processStorageClass):
(WebKit::sandboxDirectory):
* UIProcess/ChildProcessProxy.cpp:
(WebKit::ChildProcessProxy::getLaunchOptions):
* UIProcess/Launcher/ProcessLauncher.h:
* UIProcess/Launcher/mac/ProcessLauncherMac.mm:
(WebKit::serviceName):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@237154 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years ago[JHBuild] Add xdg-dbus-proxy module
commit-queue@webkit.org [Mon, 15 Oct 2018 23:14:33 +0000 (23:14 +0000)]
[JHBuild] Add xdg-dbus-proxy module
https://bugs.webkit.org/show_bug.cgi?id=190589

Patch by Patrick Griffis <pgriffis@igalia.com> on 2018-10-15
Reviewed by Michael Catanzaro.

* gtk/jhbuild.modules:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@237153 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years ago[GTK] Add new sandboxing dependencies to install-dependencies
commit-queue@webkit.org [Mon, 15 Oct 2018 23:14:00 +0000 (23:14 +0000)]
[GTK] Add new sandboxing dependencies to install-dependencies
https://bugs.webkit.org/show_bug.cgi?id=190588

Patch by Patrick Griffis <pgriffis@igalia.com> on 2018-10-15
Reviewed by Michael Catanzaro.

* gtk/install-dependencies:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@237152 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoWeb Inspector: Expose Server Timing Response Headers in Network Tab
commit-queue@webkit.org [Mon, 15 Oct 2018 23:10:20 +0000 (23:10 +0000)]
Web Inspector: Expose Server Timing Response Headers in Network Tab
https://bugs.webkit.org/show_bug.cgi?id=190440

Patch by Charles Vazac <cvazac@gmail.com> on 2018-10-15
Reviewed by Joseph Pecoraro.

Source/WebInspectorUI:

* Localizations/en.lproj/localizedStrings.js: new key "Server Timing:"
* UserInterface/Main.html: add reference to Models/ServerTimingEntry.js
* UserInterface/Models/Resource.js:
(WI.Resource.prototype.get serverTiming):
(WI.Resource.prototype.updateForResponse):
* UserInterface/Models/ServerTimingEntry.js: Added.
(WI.ServerTimingEntry):
(WI.ServerTimingEntry.parseHeaders): parse raw response headers into an array of ServerTimingEntry objects
(WI.ServerTimingEntry.parseHeaders.consumeDelimiter):
(WI.ServerTimingEntry.parseHeaders.consumeToken):
(WI.ServerTimingEntry.):
* UserInterface/Test.html: add reference to Models/ServerTimingEntry.js
* UserInterface/Views/ResourceTimingBreakdownView.js:
(WI.ResourceTimingBreakdownView.prototype._appendServerTimingRow): render a table row per ServerTimingEntry object
(WI.ResourceTimingBreakdownView.prototype.initialLayout):
(WI.ResourceTimingBreakdownView):

LayoutTests:

* inspector/unit-tests/server-timing-entry-expected.txt:
* inspector/unit-tests/server-timing-entry.html:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@237151 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoWeb Inspector: Dark Mode: pseudo elements in DOM tree are too dark
nvasilyev@apple.com [Mon, 15 Oct 2018 22:51:45 +0000 (22:51 +0000)]
Web Inspector: Dark Mode: pseudo elements in DOM tree are too dark
https://bugs.webkit.org/show_bug.cgi?id=190541
<rdar://problem/45238443>

Reviewed by Matt Baker.

* UserInterface/Views/DOMTreeOutline.css:
(@media (prefers-dark-interface)):
(.tree-outline.dom .html-pseudo-element):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@237150 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoGarden WK2 tests after r237104
achristensen@apple.com [Mon, 15 Oct 2018 22:35:35 +0000 (22:35 +0000)]
Garden WK2 tests after r237104
https://bugs.webkit.org/show_bug.cgi?id=190522

* platform/wk2/http/tests/navigation/keyboard-events-during-provisional-navigation-expected.txt: Removed.
I added  this wk2 expectation in r234658 after r234640.  r237104 effectively did the inverse of r234640,
so behavior is back to how it was before then.  This makes the expectations reflect such.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@237149 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoRemove unused WebView._globalHistoryItem
achristensen@apple.com [Mon, 15 Oct 2018 21:45:55 +0000 (21:45 +0000)]
Remove unused WebView._globalHistoryItem
https://bugs.webkit.org/show_bug.cgi?id=190601

Reviewed by Chris Dumez.

Source/WebCore:

This was a hack to give the client a more reliable current back/forward item, but it hasn't been used for years
and it's adding unnecessary complexity to the history code.

* loader/EmptyFrameLoaderClient.h:
* loader/FrameLoader.cpp:
(WebCore::FrameLoader::checkLoadCompleteForThisFrame):
(WebCore::FrameLoader::continueLoadAfterNavigationPolicy):
* loader/FrameLoaderClient.h:
* loader/HistoryController.cpp:
(WebCore::HistoryController::goToItem):
(WebCore::HistoryController::updateForStandardLoad):
(WebCore::HistoryController::updateForRedirectWithLockedBackForwardList):

Source/WebKitLegacy/mac:

* WebCoreSupport/WebFrameLoaderClient.h:
* WebCoreSupport/WebFrameLoaderClient.mm:
(WebFrameLoaderClient::updateGlobalHistoryItemForPage): Deleted.
* WebView/WebView.mm:
(-[WebView _globalHistoryItem]): Deleted.
(-[WebView _setGlobalHistoryItem:]): Deleted.
* WebView/WebViewData.h:
* WebView/WebViewInternal.h:
* WebView/WebViewPrivate.h:

Source/WebKitLegacy/win:

* WebCoreSupport/WebFrameLoaderClient.cpp:
(WebFrameLoaderClient::updateGlobalHistoryItemForPage): Deleted.
* WebCoreSupport/WebFrameLoaderClient.h:
* WebView.cpp:
(WebView::globalHistoryItem):
(WebView::setGlobalHistoryItem): Deleted.
* WebView.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@237148 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoUnreviewed, rolling out r237084, r237088, r237098, and
ryanhaddad@apple.com [Mon, 15 Oct 2018 21:40:49 +0000 (21:40 +0000)]
Unreviewed, rolling out r237084, r237088, r237098, and
r237114.
https://bugs.webkit.org/show_bug.cgi?id=190602

Breaks internal builds. (Requested by ryanhaddad on #webkit).

Reverted changesets:

"Separate configuration extraction from offset extraction"
https://bugs.webkit.org/show_bug.cgi?id=189708
https://trac.webkit.org/changeset/237084

"Gardening: Build fix after r237084."
https://bugs.webkit.org/show_bug.cgi?id=189708
https://trac.webkit.org/changeset/237088

"Gardening: Build fix after r237084."
https://bugs.webkit.org/show_bug.cgi?id=189708
https://trac.webkit.org/changeset/237098

"REGRESSION (r237084): JavaScriptCore fails to build on Linux"
https://trac.webkit.org/changeset/237114

Patch by Commit Queue <commit-queue@webkit.org> on 2018-10-15

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@237147 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoRemove unused parameters of WebPageGroupData
achristensen@apple.com [Mon, 15 Oct 2018 21:33:52 +0000 (21:33 +0000)]
Remove unused parameters of WebPageGroupData
https://bugs.webkit.org/show_bug.cgi?id=190600

Reviewed by Chris Dumez.

visibleToInjectedBundle and visibleToHistoryClient are both always true.
This removes a mysterious check in the history code.

* Shared/WebPageGroupData.cpp:
(WebKit::WebPageGroupData::encode const):
(WebKit::WebPageGroupData::decode):
* Shared/WebPageGroupData.h:
* UIProcess/WebPageGroup.cpp:
(WebKit::WebPageGroup::create):
(WebKit::pageGroupData):
(WebKit::WebPageGroup::WebPageGroup):
* UIProcess/WebPageGroup.h:
(WebKit::WebPageGroup::WebPageGroup):
(WebKit::WebPageGroup::create):
* WebProcess/WebCoreSupport/WebFrameLoaderClient.cpp:
(WebKit::WebFrameLoaderClient::updateGlobalHistory):
(WebKit::WebFrameLoaderClient::updateGlobalHistoryRedirectLinks):
(WebKit::WebFrameLoaderClient::setTitle):
* WebProcess/WebPage/WebPage.cpp:
(WebKit::WebPage::create):
(WebKit::WebPage::close):
* WebProcess/WebPage/WebPageGroupProxy.cpp:
(WebKit::WebPageGroupProxy::create):
* WebProcess/WebPage/WebPageGroupProxy.h:
(WebKit::WebPageGroupProxy::pageGroupID const):
(WebKit::WebPageGroupProxy::isVisibleToInjectedBundle const): Deleted.
(WebKit::WebPageGroupProxy::isVisibleToHistoryClient const): Deleted.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@237146 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoWeb Inspector: REGRESSION (r233824): execution highlight range missing/incorrect...
mattbaker@apple.com [Mon, 15 Oct 2018 21:21:16 +0000 (21:21 +0000)]
Web Inspector: REGRESSION (r233824): execution highlight range missing/incorrect in pretty printed code
https://bugs.webkit.org/show_bug.cgi?id=188082
<rdar://problem/42640580>

Reviewed by Joseph Pecoraro.

* UserInterface/Views/SourceCodeTextEditor.js:
(WI.SourceCodeTextEditor.prototype.textEditorExecutionHighlightRange):
TextEditor expects positions returned by the delegate to be relative to
the editor's text content, not the original source code content.

* UserInterface/Views/TextEditor.js:
(WI.TextEditor.prototype._updateExecutionRangeHighlight):
Convert positions to CodeMirror format here rather than in the delegate
method, which is a layer removed from CodeMirror.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@237145 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoExperiment: target=_blank on anchors should imply rel=noopener
cdumez@apple.com [Mon, 15 Oct 2018 21:21:07 +0000 (21:21 +0000)]
Experiment: target=_blank on anchors should imply rel=noopener
https://bugs.webkit.org/show_bug.cgi?id=190481

Reviewed by Alex Christensen.

Source/WebCore:

As an experiment, try and make it so that target=_blank on anchors implies `rel=noopener` for improved security.
WebContent can then request an opener relationship by using `rel=opener` instead.

This change was discussed at:
- https://github.com/whatwg/html/issues/4078

We want to attempt this change is STP to see if it is Web-compatible. Preliminary testing seems to indicate
that OAuth workflows still work.

* html/HTMLAnchorElement.cpp:
(WebCore::HTMLAnchorElement::parseAttribute):
(WebCore::HTMLAnchorElement::handleClick):
(WebCore::HTMLAnchorElement::effectiveTarget const):
* html/HTMLAnchorElement.h:
* page/RuntimeEnabledFeatures.h:
(WebCore::RuntimeEnabledFeatures::setBlankAnchorTargetImpliesNoOpenerEnabled):
(WebCore::RuntimeEnabledFeatures::blankAnchorTargetImpliesNoOpenerEnabled const):

Source/WebKit:

* Shared/WebPreferences.yaml:

Tools:

Add API test coverage to make sure we can now swap process when target=_blank
is specified on an anchor but rel=noopener is not.

* TestWebKitAPI/Tests/WebKitCocoa/ProcessSwapOnNavigation.mm:

LayoutTests:

Update existing tests to reflect behavior change.

* TestExpectations:
* http/tests/navigation/no-referrer-reset.html:
* http/tests/security/resources/referrer-policy-redirect-link.html:
* http/tests/security/xss-DENIED-script-inject-into-inactive-window2-pson.html:
* http/tests/security/xss-DENIED-script-inject-into-inactive-window2.html:
* http/tests/security/xssAuditor/link-opens-new-window.html:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@237144 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoWeb Inspector: Dark Mode: unreadable text when hovering CSS properties while holding...
nvasilyev@apple.com [Mon, 15 Oct 2018 21:14:52 +0000 (21:14 +0000)]
Web Inspector: Dark Mode: unreadable text when hovering CSS properties while holding Command
https://bugs.webkit.org/show_bug.cgi?id=190548
<rdar://problem/45242098>

Reviewed by Joseph Pecoraro.

* UserInterface/Views/SpreadsheetCSSStyleDeclarationEditor.css:
(.meta-key-pressed .spreadsheet-css-declaration:not(.locked) :matches(.name, .value):not(.editing):hover):
In the light mode the color remained the same.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@237143 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years ago[Apple Pay] New shipping methods are ignored when updating after the shippingaddressc...
aestes@apple.com [Mon, 15 Oct 2018 21:10:56 +0000 (21:10 +0000)]
[Apple Pay] New shipping methods are ignored when updating after the shippingaddresschange event
https://bugs.webkit.org/show_bug.cgi?id=190560
<rdar://problem/44559075>

Reviewed by Youenn Fablet.

Source/WebCore:

When a PaymentDetailsUpdate with shipping options was specified to updateWith() after the
shippingaddresschange event fires, ApplePayPaymentHandler needs to convert the shipping
options to ShippingMethods and add them to the ShippingContactUpdate so that the Apple Pay
UI renders the new options.

Added test cases to http/tests/ssl/applepay/ApplePayShippingAddressChangeEventErrors.https.html.

* Modules/applepay/ApplePayShippingMethod.idl:
* Modules/applepay/paymentrequest/ApplePayPaymentHandler.cpp:
(WebCore::ApplePayPaymentHandler::computeShippingMethods):
(WebCore::ApplePayPaymentHandler::shippingAddressUpdated):
* Modules/applepay/paymentrequest/ApplePayPaymentHandler.h:
* testing/MockPaymentCoordinator.cpp:
(WebCore::convert):
(WebCore::MockPaymentCoordinator::showPaymentUI):
(WebCore::MockPaymentCoordinator::completeShippingContactSelection):
* testing/MockPaymentCoordinator.h:
* testing/MockPaymentCoordinator.idl:

LayoutTests:

* http/tests/ssl/applepay/ApplePayShippingAddressChangeEventErrors.https-expected.txt:
* http/tests/ssl/applepay/ApplePayShippingAddressChangeEventErrors.https.html:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@237142 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoBytecodeDumper should print all switch labels
keith_miller@apple.com [Mon, 15 Oct 2018 20:58:24 +0000 (20:58 +0000)]
BytecodeDumper should print all switch labels
https://bugs.webkit.org/show_bug.cgi?id=190596

Reviewed by Saam Barati.

Right now the bytecode dumper only prints the default target not any of the
non-default targets.

* bytecode/BytecodeDumper.cpp:
(JSC::BytecodeDumper<Block>::dumpBytecode):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@237141 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoRTCPeerConnection.generateCertificate is not a function
youenn@apple.com [Mon, 15 Oct 2018 20:52:15 +0000 (20:52 +0000)]
RTCPeerConnection.generateCertificate is not a function
https://bugs.webkit.org/show_bug.cgi?id=173541
<rdar://problem/32638029>

Reviewed by Eric Carlson.

LayoutTests/imported/w3c:

* web-platform-tests/webrtc/RTCCertificate-expected.txt:
* web-platform-tests/webrtc/RTCPeerConnection-generateCertificate-expected.txt:
* web-platform-tests/webrtc/interfaces.https-expected.txt:
* web-platform-tests/webrtc/rtcpeerconnection/rtcpeerconnection-constructor-expected.txt:

Source/ThirdParty/libwebrtc:

* Configurations/libwebrtc.iOS.exp:
* Configurations/libwebrtc.iOSsim.exp:
* Configurations/libwebrtc.mac.exp:

Source/WebCore:

Add support for RTCCertificate generation through libwebrtc certificate generator.
Make generation in the webrtc network thread.
Support is as per spec (ECDSA and RSASSA_PKCS).

Partially covered by WPT tests.

* css/StyleBuilder.h: Fixing build.
* CMakeLists.txt:
* DerivedSources.make:
* Modules/webauthn/PublicKeyCredential.cpp: Fixing build.
* Modules/mediastream/PeerConnectionBackend.cpp:
(WebCore::PeerConnectionBackend::generateCertificate):
* Modules/mediastream/PeerConnectionBackend.h:
(WebCore::PeerConnectionBackend::CertificateInformation::RSASSA_PKCS1_v1_5):
(WebCore::PeerConnectionBackend::CertificateInformation::ECDSA_P256):
(WebCore::PeerConnectionBackend::CertificateInformation::CertificateInformation):
* Modules/mediastream/RTCCertificate.cpp: Added.
(WebCore::RTCCertificate::create):
(WebCore::RTCCertificate::RTCCertificate):
* Modules/mediastream/RTCCertificate.h: Added.
(WebCore::RTCCertificate::expires const):
(WebCore::RTCCertificate::getFingerprints const):
* Modules/mediastream/RTCCertificate.idl: Added.
* Modules/mediastream/RTCPeerConnection.cpp:
(WebCore::certificateTypeFromAlgorithmIdentifier):
(WebCore::RTCPeerConnection::generateCertificate):
* Modules/mediastream/RTCPeerConnection.h:
* Modules/mediastream/RTCPeerConnection.idl:
* Modules/mediastream/libwebrtc/LibWebRTCCertificateGenerator.cpp: Added.
(WebCore::LibWebRTCCertificateGenerator::RTCCertificateGeneratorCallback::RTCCertificateGeneratorCallback):
(WebCore::LibWebRTCCertificateGenerator::RTCCertificateGeneratorCallback::~RTCCertificateGeneratorCallback):
(WebCore::LibWebRTCCertificateGenerator::keyParamsFromCertificateType):
(WebCore::LibWebRTCCertificateGenerator::generateCertificate):
* Modules/mediastream/libwebrtc/LibWebRTCCertificateGenerator.h: Added.
* Sources.txt:
* WebCore.xcodeproj/project.pbxproj:
* bindings/js/WebCoreBuiltinNames.h:
* platform/mediastream/libwebrtc/LibWebRTCProvider.cpp:
(WebCore::LibWebRTCProvider::certificateGenerator):
* platform/mediastream/libwebrtc/LibWebRTCProvider.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@237140 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoAdd compact logging for the paint-order RenderLayer tree
simon.fraser@apple.com [Mon, 15 Oct 2018 20:09:15 +0000 (20:09 +0000)]
Add compact logging for the paint-order RenderLayer tree
https://bugs.webkit.org/show_bug.cgi?id=190592

Reviewed by Zalan Bujtas.

Add a way to dump the RenderLayer tree in paint order, which will be extended in future
to show more dirty bit state.

* rendering/RenderLayer.cpp:
(WebCore::outputPaintOrderTreeLegend):
(WebCore::outputIdent):
(WebCore::outputPaintOrderTreeRecursive):
(WebCore::showPaintOrderTree):
* rendering/RenderLayer.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@237139 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoRemove unused parameters from FrameLoaderClient::createFrame
achristensen@apple.com [Mon, 15 Oct 2018 19:55:39 +0000 (19:55 +0000)]
Remove unused parameters from FrameLoaderClient::createFrame
https://bugs.webkit.org/show_bug.cgi?id=190587

Reviewed by Chris Dumez.

Source/WebCore:

* loader/EmptyClients.cpp:
(WebCore::EmptyFrameLoaderClient::createFrame):
* loader/EmptyFrameLoaderClient.h:
* loader/FrameLoaderClient.h:
* loader/SubframeLoader.cpp:
(WebCore::SubframeLoader::loadSubframe):

Source/WebKit:

* WebProcess/WebCoreSupport/WebFrameLoaderClient.cpp:
(WebKit::WebFrameLoaderClient::createFrame):
* WebProcess/WebCoreSupport/WebFrameLoaderClient.h:

Source/WebKitLegacy/mac:

* WebCoreSupport/WebFrameLoaderClient.h:
* WebCoreSupport/WebFrameLoaderClient.mm:
(WebFrameLoaderClient::createFrame):

Source/WebKitLegacy/win:

* WebCoreSupport/WebFrameLoaderClient.cpp:
(WebFrameLoaderClient::createFrame):
* WebCoreSupport/WebFrameLoaderClient.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@237138 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoRemove unused WebFrame._loadType
achristensen@apple.com [Mon, 15 Oct 2018 19:53:47 +0000 (19:53 +0000)]
Remove unused WebFrame._loadType
https://bugs.webkit.org/show_bug.cgi?id=190583

Reviewed by Chris Dumez.

This is unused, and this concept needs to not be exposed from WebCore to WebKitLegacy for me to fix history bugs.

* WebView/WebFrame.mm:
(toWebFrameLoadType): Deleted.
(-[WebFrame _loadType]): Deleted.
* WebView/WebFramePrivate.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@237137 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoEmit fjcvtzs on ARM64E on Darwin
sbarati@apple.com [Mon, 15 Oct 2018 19:47:09 +0000 (19:47 +0000)]
Emit fjcvtzs on ARM64E on Darwin
https://bugs.webkit.org/show_bug.cgi?id=184023

Reviewed by Yusuke Suzuki and Filip Pizlo.

JSTests:

* stress/double-to-int32-NaN.js: Added.
(assert):
(foo):

Source/JavaScriptCore:

ARMv8.3 introduced the fjcvtzs instruction which does double->int32
conversion using the semantics defined by JavaScript:
http://infocenter.arm.com/help/index.jsp?topic=/com.arm.doc.dui0801g/hko1477562192868.html
This patch teaches JSC to use that instruction when possible.

* assembler/ARM64Assembler.h:
(JSC::ARM64Assembler::fjcvtzs):
(JSC::ARM64Assembler::fjcvtzsInsn):
* assembler/MacroAssemblerARM64.cpp:
(JSC::MacroAssemblerARM64::collectCPUFeatures):
* assembler/MacroAssemblerARM64.h:
(JSC::MacroAssemblerARM64::supportsDoubleToInt32ConversionUsingJavaScriptSemantics):
(JSC::MacroAssemblerARM64::convertDoubleToInt32UsingJavaScriptSemantics):
* dfg/DFGSpeculativeJIT.cpp:
(JSC::DFG::SpeculativeJIT::compileValueToInt32):
* disassembler/ARM64/A64DOpcode.cpp:
* disassembler/ARM64/A64DOpcode.h:
(JSC::ARM64Disassembler::A64DOpcode::appendInstructionName):
* ftl/FTLLowerDFGToB3.cpp:
(JSC::FTL::DFG::LowerDFGToB3::doubleToInt32):
* jit/JITRightShiftGenerator.cpp:
(JSC::JITRightShiftGenerator::generateFastPath):
* runtime/MathCommon.h:
(JSC::toInt32):

Source/WTF:

* wtf/Platform.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@237136 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years ago[iOS] Can't select text after dismissing the keyboard when changing focus
wenson_hsieh@apple.com [Mon, 15 Oct 2018 19:47:03 +0000 (19:47 +0000)]
[iOS] Can't select text after dismissing the keyboard when changing focus
https://bugs.webkit.org/show_bug.cgi?id=190563
<rdar://problem/44613559>

Reviewed by Tim Horton.

Source/WebKit:

In r230686, we switched from using UIWKSelectionAssistant to UIWKTextInteractionAssistant for handling selection
in non-editable content on iOS; as such, when an editable element loses focus, instead of switching from the
text interaction assistant to the web selection assistant as we've previously done, we now reset our text
interaction assistant by calling `-[UIWKTextInteractionAssistant setGestureRecognizers]`, which removes all of
the current text selection gesture recognizers from WKContentView and regenerates them by building up a tree of
`UITextInteraction`s and adding them to the assistant (see `-[UITextInteractionAssistant
addGestureRecognizersToView:]`). In particular, `_UITextSelectionForceGesture` is the gesture recognizer used to
trigger text selection when long pressing.

After dismissing the keyboard by tapping the "Done" button, the UITextInteractions and gesture recognizers on
the interaction assistant include:

    <UITextInteraction>
        …
        <UITextIndirectNonEditableInteraction>
            <_UIKeyboardBasedNonEditableTextSelectionInteraction>
              ↳ "_UIKeyboardTextSelectionGestureForcePress" → <_UITextSelectionForceGesture>

However, after the keyboard dismisses due to an editable element losing focus, the UITextInteractions on the
interaction assistant look like this:

    <UITextInteraction>
        …
        <UITextIndirectNonEditableInteraction>

Subsequently, the lack of a `_UIKeyboardBasedNonEditableTextSelectionInteraction` makes text selection by long
pressing impossible, since the `_UITextSelectionForceGesture` is never introduced to `WKContentView`. In UIKit,
`UITextIndirectNonEditableInteraction` only adds `_UIKeyboardBasedNonEditableTextSelectionInteraction` as a
child if the text input view — in our case, WKContentView — is missing an input delegate (see `-initWithView:`).
In the case where the Done button is used to dismiss the keyboard, WKContentView loses first responder, and the
input delegate of WKContentView is cleared out early on, before we call `-stopAssistingKeyboard`:

    -[WKContentView(WKInteraction) setInputDelegate:]
    -[UIKeyboardImpl setDelegate:force:]
    -[UIPeripheralHost(UIKitInternal) _reloadInputViewsForResponder:]
    -[UIResponder _finishResignFirstResponder]
    -[UIResponder resignFirstResponder]
    -[WKContentView(WKInteraction) resignFirstResponderForWebView]
    -[UIKeyboardImpl dismissKeyboard]

However, in the case where the focused element is blurred, we end up clearing out the delegate in
`-_stopAssistingNode`, *after* we've already called `-setGestureRecognizers` on the interaction assistant. This
means UIKit will skip adding `_UIKeyboardBasedNonEditableTextSelectionInteraction` to the text interaction
assistant.

    -[WKContentView(WKInteraction) setInputDelegate:]
    -[UIKeyboardImpl setDelegate:force:]
    -[UIPeripheralHost(UIKitInternal) _reloadInputViewsForResponder:]
    -[UIResponder(UIResponderInputViewAdditions) reloadInputViews]
    -[WKContentView(WKInteraction) _stopAssistingNode]

To fix this, we simply reset our `inputDelegate` earlier in `_stopAssistingKeyboard` instead of waiting until
we reload input views. This ensures that UIKit sets up the text interaction assistant's gestures when changing
focus in the same way as we would when the keyboard is dismissed via `-resignFirstResponder` (e.g. when pressing
the Done button).

Test: editing/selection/ios/select-text-after-changing-focus.html

* UIProcess/ios/WKContentViewInteraction.mm:
(-[WKContentView setupInteraction]):
(-[WKContentView setUpTextSelectionAssistant]):
(-[WKContentView _startAssistingKeyboard]):
(-[WKContentView _stopAssistingKeyboard]):
(-[WKContentView useSelectionAssistantWithGranularity:]): Deleted.

Additionally rename this to -setUpTextSelectionAssistant and remove the selection granularity argument. This was
previously used to switch between web and text interaction assistants.

Tools:

* DumpRenderTree/ios/UIScriptControllerIOS.mm:
(WTR::UIScriptController::isShowingKeyboard const):

Add a new UIScriptController method that returns whether the keyboard is shown. See `ui-helper.js` for more
details.

* TestRunnerShared/UIScriptContext/Bindings/UIScriptController.idl:
* TestRunnerShared/UIScriptContext/UIScriptController.cpp:
(WTR::UIScriptController::isShowingKeyboard const):
* TestRunnerShared/UIScriptContext/UIScriptController.h:
* WebKitTestRunner/cocoa/TestRunnerWKWebView.h:
* WebKitTestRunner/cocoa/TestRunnerWKWebView.mm:

Also rename the `isShowingKeyboard` Objective-C property to the more canonical `showingKeyboard`, with
`isShowingKeyboard` as the getter method.

(-[TestRunnerWKWebView _invokeShowKeyboardCallbackIfNecessary]):
(-[TestRunnerWKWebView _invokeHideKeyboardCallbackIfNecessary]):
* WebKitTestRunner/ios/UIScriptControllerIOS.mm:
(WTR::UIScriptController::isShowingKeyboard const):

LayoutTests:

Add a new layout test to check that the user can make a selection by long pressing after the keyboard is
dismissed due to changing the focused element.

* editing/selection/ios/select-text-after-changing-focus-expected.txt: Added.
* editing/selection/ios/select-text-after-changing-focus.html: Added.
* resources/ui-helper.js:

Also tweak the behavior of `UIHelper.waitForKeyboardToHide()`, so that it resolves immediately if the keyboard
is not shown. This allows us to ensure that tests which use `UIHelper.waitForKeyboardToHide()` are robust in the
case where they wait for another action to complete (e.g. a simulated tap) prior to registering a keyboard
hiding callback.

(window.UIHelper.waitForKeyboardToHide.return.new.Promise):
(window.UIHelper.waitForKeyboardToHide):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@237135 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years ago[Apple Pay] Payment authorization results with ApplePayErrors should never be conside...
aestes@apple.com [Mon, 15 Oct 2018 19:40:32 +0000 (19:40 +0000)]
[Apple Pay] Payment authorization results with ApplePayErrors should never be considered final
https://bugs.webkit.org/show_bug.cgi?id=190559
<rdar://problem/37250908>

Reviewed by Anders Carlsson.

Source/WebCore:

When PaymentCoordinator thinks a payment authorization result is final it releases the
active ApplePaySession. The Apple Pay UI is dismissed after a receiving a final result.

However, WebCore::isFinalStateResult had the wrong idea about what was a final state,
in some cases causing PaymentCoordinator to release the active session even when the UI is
still presented. If the user authorizes payment again, the website will not receive another
paymentauthorized event, and the Apple Pay UI will eventually time out waiting for a result.

Specifically, isFinalStateResult thought that:

(1) results with STATUS_SUCCESS were always final, even if they had errors
(2) errors with code "unknown" were final

Both of these assumptions are wrong. PassKit considers any result with errors to be
non-final, even if an error has code "unknown."

Fixed WebCore::isFinalStateResult to agree with what PassKit considers to be final results.

Test: http/tests/ssl/applepay/ApplePaySessionFinalState.https.html

* Modules/applepay/ApplePaySessionPaymentRequest.cpp:
(WebCore::isFinalStateResult):
* testing/MockPaymentCoordinator.cpp:
(WebCore::MockPaymentCoordinator::completePaymentSession):

LayoutTests:

* http/tests/ssl/applepay/ApplePaySessionFinalState.https-expected.txt: Added.
* http/tests/ssl/applepay/ApplePaySessionFinalState.https.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@237134 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years ago[Payment Request] PaymentResponse should be a ContextDestructionObserver
aestes@apple.com [Mon, 15 Oct 2018 19:37:17 +0000 (19:37 +0000)]
[Payment Request] PaymentResponse should be a ContextDestructionObserver
https://bugs.webkit.org/show_bug.cgi?id=190558

Reviewed by Alex Christensen.

Make PaymentResponse a ContextDestructionObserver so it has a ScriptExecutionContext to
return for its override of EventTarget::scriptExecutionContext(). Do this instead of
downcasting m_request to an ActiveDOMObject to ask for its context.

* Modules/paymentrequest/PaymentRequest.cpp:
(WebCore::PaymentRequest::accept):
* Modules/paymentrequest/PaymentResponse.cpp:
(WebCore::PaymentResponse::PaymentResponse):
(WebCore::PaymentResponse::scriptExecutionContext const): Deleted.
* Modules/paymentrequest/PaymentResponse.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@237133 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years ago[Payment Request] Use a PendingActivity token rather than calling (un)setPendingActivity
aestes@apple.com [Mon, 15 Oct 2018 19:34:18 +0000 (19:34 +0000)]
[Payment Request] Use a PendingActivity token rather than calling (un)setPendingActivity
https://bugs.webkit.org/show_bug.cgi?id=190557

Reviewed by Alex Christensen.

PaymentRequest has pending activity whenever there is an active payment handler, since
user-initiated events can occur whenever the payment UI is displayed.

Rather than manually track the pending activity with setPendingActivity() and
unsetPendingActivity(), use a PendingActivity token tied to the lifetime of the active
payment handler. Also, rewrite canSuspendForDocumentSuspension() in terms of
hasPendingActivity().

* Modules/paymentrequest/PaymentRequest.cpp:
(WebCore::PaymentRequest::show):
(WebCore::PaymentRequest::abortWithException):
(WebCore::PaymentRequest::canSuspendForDocumentSuspension const):
(WebCore::PaymentRequest::paymentMethodChanged):
(WebCore::PaymentRequest::completeMerchantValidation):
(WebCore::PaymentRequest::settleDetailsPromise):
(WebCore::PaymentRequest::complete):
(WebCore::PaymentRequest::cancel):
* Modules/paymentrequest/PaymentRequest.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@237132 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoWeb Inspector: RDM: Toolbar hidden in when Inspector is docked to side.
commit-queue@webkit.org [Mon, 15 Oct 2018 19:28:09 +0000 (19:28 +0000)]
Web Inspector: RDM: Toolbar hidden in when Inspector is docked to side.
https://bugs.webkit.org/show_bug.cgi?id=190545
rdar://problem/44674500

Patch by Remy Demarest <rdemarest@apple.com> on 2018-10-15
Reviewed by Brian Burg.

When the inspector is placed next to the web view it uses its _topContentInset and _totalHeightOfBanners
to lay out the inspector so it does not underlap the window toolbar, but this technique does not work when
in responsive design mode because of the different attachment view. This patch fixes the issue by using
-[NSWindow contentLayoutRect] to figure out the height of the inspector instead of relying on the content
insets of the web view. This requires observing -contentLayoutRect and ensure we only observe its changes
when the view is actually on the screen.

* UIProcess/WebInspectorProxy.h:
Declare helpers to add/remove observer on the attached inspector window.

* UIProcess/mac/WKInspectorViewController.h:
* UIProcess/mac/WKInspectorViewController.mm:
(-[WKInspectorViewController inspectorWKWebView:willMoveToWindow:]):
(-[WKInspectorViewController inspectorWKWebViewDidMoveToWindow:]):

* UIProcess/mac/WKInspectorWKWebView.h:
* UIProcess/mac/WKInspectorWKWebView.mm:
(-[WKInspectorWKWebView viewWillMoveToWindow:]):
(-[WKInspectorWKWebView viewDidMoveToWindow]):

* UIProcess/mac/WebInspectorProxyMac.mm:
(-[WKWebInspectorProxyObjCAdapter observeValueForKeyPath:ofObject:change:context:]): Update inspector layout
whenever the contentLayoutRect changes. Except when live resizing since the attachment view also sends
notifications at the same time.
(-[WKWebInspectorProxyObjCAdapter inspectorViewController:willMoveToWindow:]):
(-[WKWebInspectorProxyObjCAdapter inspectorViewControllerDidMoveToWindow:]):
(WebKit::WebInspectorProxy::attachmentWillMoveFromWindow): Remove the observer only if we set it up before.
(WebKit::WebInspectorProxy::attachmentDidMoveToWindow): Set up the observer and immediately update the frame
of the inspector since it just moved to its final destination.
(WebKit::WebInspectorProxy::inspectedViewFrameDidChange): Adjust the frame of the attached inspector based
on the contentLayoutRect of the window rather than the topContentInset of the web view.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@237131 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoUnreviewed, fix JSC tests after WebGPUShadingLanguageRI directory was removed in...
ryanhaddad@apple.com [Mon, 15 Oct 2018 19:14:50 +0000 (19:14 +0000)]
Unreviewed, fix JSC tests after WebGPUShadingLanguageRI directory was removed in r237115.

* Scripts/run-javascriptcore-tests:
(runJSCStressTests): Remove wsl-tests.yaml.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@237130 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoJSArray::shiftCountWithArrayStorage is wrong when an array has holes
sbarati@apple.com [Mon, 15 Oct 2018 19:14:42 +0000 (19:14 +0000)]
JSArray::shiftCountWithArrayStorage is wrong when an array has holes
https://bugs.webkit.org/show_bug.cgi?id=190262
<rdar://problem/44986241>

Reviewed by Mark Lam.

JSTests:

* stress/array-prototype-concat-of-long-spliced-arrays.js:
(test):
* stress/slice-array-storage-with-holes.js: Added.
(main):

Source/JavaScriptCore:

We would take the fast path for shiftCountWithArrayStorage when the array
hasHoles(). However, the code for this was wrong. It'd incorrectly update
ArrayStorage::m_numValuesInVector. Since the hasHoles() for ArrayStorage
path is never taken in JetStream 2, this patch just removes that from
the fast path. Instead, we just fallback to the slow path when hasHoles().
If we find evidence that this matters for real use cases, we can
figure out a way to make the fast path work.

* runtime/JSArray.cpp:
(JSC::JSArray::shiftCountWithArrayStorage):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@237129 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoUnreviewed, rolling out r237054.
commit-queue@webkit.org [Mon, 15 Oct 2018 18:42:44 +0000 (18:42 +0000)]
Unreviewed, rolling out r237054.
https://bugs.webkit.org/show_bug.cgi?id=190593

"this regressed JetStream 2 by 6% on iOS" (Requested by
saamyjoon on #webkit).

Reverted changeset:

"[JSC] JSC should have "parseFunction" to optimize Function
constructor"
https://bugs.webkit.org/show_bug.cgi?id=190340
https://trac.webkit.org/changeset/237054

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@237128 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoChanging view scale should zoom to initial scale if the page is already at initial...
wenson_hsieh@apple.com [Mon, 15 Oct 2018 17:36:31 +0000 (17:36 +0000)]
Changing view scale should zoom to initial scale if the page is already at initial scale
https://bugs.webkit.org/show_bug.cgi?id=190570
<rdar://problem/45261877>

Reviewed by Tim Horton.

Source/WebCore:

Add a getter for ViewportConfiguration's layout size scale factor.
See Source/WebKit/ChangeLog for more details.

* page/ViewportConfiguration.h:
(WebCore::ViewportConfiguration::layoutSizeScaleFactor const):

Source/WebKit:

r237087 added support for changing the view scale on iOS, by making it possible to let the minimum layout size
be a factor of the view size; this allows internal clients to change page zoom levels on iOS. Currently,
changing the page zoom level automatically zooms to the new initial scale only if the user has not manually
scaled the page before, even if the page is already at initial scale (e.g. after the user double taps to zoom on
a small element, and double taps again to zoom back out to initial scale).

This patch makes some minor adjustments to automatically zoom to the new initial scale after changing the view
scale, as long as the page was at initial scale when changing zoom levels.

Test: fast/viewport/ios/initial-scale-after-changing-view-scale.html

* WebProcess/WebPage/WebPage.h:
* WebProcess/WebPage/ios/WebPageIOS.mm:
(WebKit::areEssentiallyEqualAsFloat):

Move this function further up in the file, so that we can use it in `setViewportConfigurationViewLayoutSize`.

(WebKit::WebPage::setViewportConfigurationViewLayoutSize):

If the page is near initial scale and the zoom level changes, zoom to the new initial scale.

(WebKit::WebPage::viewportConfigurationChanged):

Make this take a new enum argument that determines whether we want to zoom to initial scale as a result of the
viewport configuration change (`No` by default).

LayoutTests:

* fast/viewport/ios/initial-scale-after-changing-view-scale-expected.txt: Added.
* fast/viewport/ios/initial-scale-after-changing-view-scale.html: Added.

Add a new layout test to verify that:
1.  Changing view scale before the page scale factor is changed by the user zooms to the new initial scale.
2.  Changing view scale after the user changes the page scale to something different than initial scale does not
    cause the page to zoom to the new initial scale.
3.  Changing view scale after the user changes the page scale factor back to initial scale causes the page to
    zoom to the new initial scale.

* resources/basic-gestures.js:

Make a small tweak in `doubleTapToZoomAtPoint` to ensure that both single taps as well as zooming are complete
before invoking UIScriptController completion.

(return.new.Promise):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@237127 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoFix build error with !LOG_DISABLED in Release mode
cturner@igalia.com [Mon, 15 Oct 2018 17:33:21 +0000 (17:33 +0000)]
Fix build error with !LOG_DISABLED in Release mode
https://bugs.webkit.org/show_bug.cgi?id=190465

Reviewed by Michael Catanzaro.

No tests due to no new functionality.

* workers/service/ServiceWorkerRegistrationKey.cpp:
* workers/service/ServiceWorkerRegistrationKey.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@237126 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoWeb Inspector: Dark Mode: style editor warnings should not look like errors
nvasilyev@apple.com [Mon, 15 Oct 2018 17:27:02 +0000 (17:27 +0000)]
Web Inspector: Dark Mode: style editor warnings should not look like errors
https://bugs.webkit.org/show_bug.cgi?id=190569
<rdar://problem/45261689>

Reviewed by Matt Baker.

Make warnings in the style editor look the same as warnings in the content views.
Unify warning and error background color variables.

* UserInterface/Views/LogContentView.css:
(@media (prefers-dark-interface)):
(.console-error-level):
(.console-warning-level):
* UserInterface/Views/SourceCodeTextEditor.css:
(.source-code.text-editor > .CodeMirror .warning):
(.source-code.text-editor > .CodeMirror .error):
(.source-code.text-editor > .CodeMirror .issue-widget.warning):
(.source-code.text-editor > .CodeMirror .issue-widget.inline.warning):
(.source-code.text-editor > .CodeMirror .issue-widget.error):
(.source-code.text-editor > .CodeMirror .issue-widget.inline.error):
(@media (prefers-dark-interface)):
(.source-code.text-editor > .CodeMirror .issue-widget):
* UserInterface/Views/SpreadsheetCSSStyleDeclarationEditor.css:
(.spreadsheet-style-declaration-editor .property.has-warning):
(.spreadsheet-style-declaration-editor .property.has-warning .warning):
(@media (prefers-dark-interface)):
(.spreadsheet-style-declaration-editor :matches(.name, .value).editing):
* UserInterface/Views/Variables.css:
(:root):
(@media (prefers-dark-interface)):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@237125 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoLayerListMutationDetector should take a reference
simon.fraser@apple.com [Mon, 15 Oct 2018 17:25:50 +0000 (17:25 +0000)]
LayerListMutationDetector should take a reference
https://bugs.webkit.org/show_bug.cgi?id=190586

Reviewed by Zalan Bujtas.

Change LayerListMutationDetector to take a RenderLayer&.

* rendering/RenderLayer.cpp:
(WebCore::RenderLayer::paintList):
(WebCore::RenderLayer::hitTestLayer):
(WebCore::RenderLayer::calculateClipRects const):
* rendering/RenderLayer.h:
(WebCore::LayerListMutationDetector::LayerListMutationDetector):
(WebCore::LayerListMutationDetector::~LayerListMutationDetector):
* rendering/RenderLayerBacking.cpp:
(WebCore::traverseVisibleNonCompositedDescendantLayers):
* rendering/RenderLayerCompositor.cpp:
(WebCore::RenderLayerCompositor::addToOverlapMapRecursive):
(WebCore::RenderLayerCompositor::computeCompositingRequirements):
(WebCore::RenderLayerCompositor::rebuildCompositingLayerTree):
(WebCore::RenderLayerCompositor::updateLayerTreeGeometry):
(WebCore::RenderLayerCompositor::updateCompositingDescendantGeometry):
(WebCore::RenderLayerCompositor::recursiveRepaintLayer):
(WebCore::RenderLayerCompositor::layerHas3DContent const):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@237124 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoShare some code to dirty z-order and normal flow lists when child layers are added...
simon.fraser@apple.com [Mon, 15 Oct 2018 17:25:47 +0000 (17:25 +0000)]
Share some code to dirty z-order and normal flow lists when child layers are added or removed
https://bugs.webkit.org/show_bug.cgi?id=190585

Reviewed by Zalan Bujtas.

Factor code which dirties the normal flow list and child stacking context z-order lists
when a layer is added or removed.

* rendering/RenderLayer.cpp:
(WebCore::RenderLayer::addChild):
(WebCore::RenderLayer::removeChild):
(WebCore::RenderLayer::dirtyPaintOrderListsOnChildChange):
* rendering/RenderLayer.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@237123 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoMake a helper function to check for reflection layers
simon.fraser@apple.com [Mon, 15 Oct 2018 17:25:45 +0000 (17:25 +0000)]
Make a helper function to check for reflection layers
https://bugs.webkit.org/show_bug.cgi?id=190584

Reviewed by Zalan Bujtas.

Add RenderLayer::isReflectionLayer() which returns true if the passed layer
is the layer of this layer's reflection. It's used in z-order list building.

* rendering/RenderLayer.cpp:
(WebCore::RenderLayer::updateNormalFlowList):
(WebCore::RenderLayer::rebuildZOrderLists):
(WebCore::RenderLayer::collectLayers):
(WebCore::expandClipRectForDescendantsAndReflection):
(WebCore::RenderLayer::calculateClipRects const):
* rendering/RenderLayer.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@237122 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoRenderLayer::addChild() and removeChild() should take references
simon.fraser@apple.com [Mon, 15 Oct 2018 17:25:42 +0000 (17:25 +0000)]
RenderLayer::addChild() and removeChild() should take references
https://bugs.webkit.org/show_bug.cgi?id=190582

Reviewed by Zalan Bujtas.

Pass the layer to be added or removed as a reference; it's never null.

* rendering/RenderElement.cpp:
(WebCore::addLayers):
(WebCore::RenderElement::removeLayers):
(WebCore::RenderElement::moveLayers):
* rendering/RenderLayer.cpp:
(WebCore::RenderLayer::addChild):
(WebCore::RenderLayer::removeChild):
(WebCore::RenderLayer::insertOnlyThisLayer):
(WebCore::RenderLayer::removeOnlyThisLayer):
* rendering/RenderLayer.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@237121 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoUnreviewed Windows build fix after r237112.
cdumez@apple.com [Mon, 15 Oct 2018 16:59:49 +0000 (16:59 +0000)]
Unreviewed Windows build fix after r237112.

* Plugins/PluginView.cpp:
(WebCore::PluginView::performRequest):
(WebCore::PluginView::load):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@237120 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years ago[Cairo] Incorrect rendering for 135-deg skews
yoshiaki.jitsukawa@sony.com [Mon, 15 Oct 2018 16:50:33 +0000 (16:50 +0000)]
[Cairo] Incorrect rendering for 135-deg skews
https://bugs.webkit.org/show_bug.cgi?id=190513

Source/WebCore:

Compensation value to zero the the translation components
of the transformation matrix is incorrect if the matrix
has a shear factor.

Reviewed by Žan Doberšek.

Tests: fast/transforms/skew-x-135deg-with-gradient.html
       fast/transforms/skew-y-135deg-with-gradient.html

* platform/graphics/cairo/CairoUtilities.cpp:
(WebCore::drawPatternToCairoContext):

LayoutTests:

Reviewed by Žan Doberšek.

* fast/transforms/skew-x-135deg-with-gradient-expected.html: Added.
* fast/transforms/skew-x-135deg-with-gradient.html: Added.
* fast/transforms/skew-y-135deg-with-gradient-expected.html: Added.
* fast/transforms/skew-y-135deg-with-gradient.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@237119 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years ago[Curl][WinCairo] Add Public Suffix support to WinCairo
chris.reid@sony.com [Mon, 15 Oct 2018 16:33:56 +0000 (16:33 +0000)]
[Curl][WinCairo] Add Public Suffix support to WinCairo
https://bugs.webkit.org/show_bug.cgi?id=183060

Reviewed by Alex Christensen.

.:

* Source/cmake/FindLibPSL.cmake: Added.
* Source/cmake/OptionsWin.cmake:
* Source/cmake/OptionsWinCairo.cmake:

Source/WebCore:

Add Public Suffix support to WinCairo using LibPSL.
LibPSL has been added to WinCairoRequirements in v2018.10.09.

* platform/Curl.cmake:
* platform/network/curl/CookieJarDB.cpp:
(WebCore::CookieJarDB::searchCookies): Search for cookies using the topPrivatelyControlledDomain
(WebCore::CookieJarDB::setCookie): Prevent cookies from being added under TLDs
* platform/network/curl/PublicSuffixCurl.cpp: Added.
(WebCore::isPublicSuffix):
(WebCore::topPrivatelyControlledDomainInternal):
(WebCore::topPrivatelyControlledDomain):

Tools:

Share mac's PublicSuffix tests with all ports.

* TestWebKitAPI/PlatformWin.cmake:
* TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj:
* TestWebKitAPI/Tests/WebCore/PublicSuffix.cpp:
* TestWebKitAPI/Tests/mac/PublicSuffix.mm: Removed.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@237118 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoRenderLayer tree-related cleanup
simon.fraser@apple.com [Mon, 15 Oct 2018 16:12:37 +0000 (16:12 +0000)]
RenderLayer tree-related cleanup
https://bugs.webkit.org/show_bug.cgi?id=190572

Reviewed by Zalan Bujtas.

Move code around so that functions related to parent/child and z-order tree structure
are near the top of RenderLayer.cpp, since this is one of the primary functions of layers.

Attempts to use inheritance or composition to separate out tree structure resulted in
unwieldy code and/or performance regressions.

The only behavior change is to store a bit for m_isStackingContext so that we don't have
to consult the old style to know if it changed.

* rendering/RenderLayer.cpp:
(WebCore::RenderLayer::RenderLayer):
(WebCore::RenderLayer::addChild):
(WebCore::RenderLayer::removeChild):
(WebCore::RenderLayer::insertOnlyThisLayer):
(WebCore::RenderLayer::removeOnlyThisLayer):
(WebCore::canCreateStackingContext):
(WebCore::RenderLayer::shouldBeNormalFlowOnly const):
(WebCore::RenderLayer::shouldBeStackingContext const):
(WebCore::RenderLayer::setIsNormalFlowOnly):
(WebCore::RenderLayer::setIsStackingContext):
(WebCore::RenderLayer::setParent):
(WebCore::RenderLayer::dirtyZOrderLists):
(WebCore::RenderLayer::dirtyStackingContextZOrderLists):
(WebCore::RenderLayer::dirtyNormalFlowList):
(WebCore::RenderLayer::updateNormalFlowList):
(WebCore::RenderLayer::rebuildZOrderLists):
(WebCore::RenderLayer::collectLayers):
(WebCore::RenderLayer::updateLayerListsIfNeeded):
(WebCore::RenderLayer::beginTransparencyLayers):
(WebCore::RenderLayer::willBeDestroyed):
(WebCore::RenderLayer::isDescendantOf const):
(WebCore::RenderLayer::calculateClipRects const):
(WebCore::compareZIndex): Deleted.
* rendering/RenderLayer.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@237117 268f45cc-cd09-0410-ab3c-d52691b4dbfc