Timeline

Nov 16, 2018:

10:46 PM Changeset in webkit [238343] by Chris Dumez

• 2 edits in trunk/Source/WebKit

Regression(ProcessPrewarming) Assertion hit in NetworkProcessProxy destructor
​https://bugs.webkit.org/show_bug.cgi?id=191800

Reviewed by Ryosuke Niwa.

Make sure the NetworkProcessProxy destructor calls its remaining m_pendingConnectionReplies
completion handlers to avoid hitting an assertion in the CompletionHandler destructor.

Prewarmed WebProcess do not keep the ProcessPool alive and thus not the NetworkProcessProxy
either. As a result, it is possible for NetworkProcessProxy to have pending connection
requests from those prewarmed processes when it is destroyed. Similar story for the
suspended page processes when they are destroyed.

• UIProcess/Network/NetworkProcessProxy.cpp:

(WebKit::NetworkProcessProxy::~NetworkProcessProxy):

10:03 PM Changeset in webkit [238342] by achristensen@apple.com

• 9 edits in trunk

Tweak _showSafeBrowsingWarningWithTitle SPI
​https://bugs.webkit.org/show_bug.cgi?id=191799

Reviewed by Wenson Hsieh.

Source/WebCore:

It turns out I needed to expose both sentinel values used in safe browsing for my application of this SPI in Mac Safari.
Allowing the caller to make its own sentinel values is insufficient because the malware confirmation needs to be over the warning.
The completion handler parameter should just be a bool indicating whether the user has chosen to continue after all warnings.

Covered by updated API tests.

• en.lproj/Localizable.strings:
• platform/LocalizedStrings.cpp:

(WebCore::formControlCancelButtonTitle):
Make "Cancel"'s description a little more generic.

Source/WebKit:

• UIProcess/API/Cocoa/WKWebView.mm:

(-[WKWebView _showSafeBrowsingWarningWithTitle:warning:details:completionHandler:]):
(+[WKWebView _visitUnsafeWebsiteSentinel]):

• UIProcess/API/Cocoa/WKWebViewPrivate.h:
• UIProcess/Cocoa/WKSafeBrowsingWarning.mm:

(-[WKSafeBrowsingWarning clickedOnLink:]):
I need to localize "Continue" and "Cancel" so they aren't always in English.

Tools:

• TestWebKitAPI/Tests/WebKitCocoa/SafeBrowsing.mm:

(TEST):

8:23 PM Changeset in webkit [238341] by Chris Dumez

• 2 edits in trunk/Source/WebKit

WKWebViewMacEditingTests.DoNotCrashWhenInterpretingKeyEventWhileDeallocatingView API test is crashing in debug
​https://bugs.webkit.org/show_bug.cgi?id=191797

Reviewed by Ryosuke Niwa.

We were hitting the assertion because the error handler was first getting called. It would
deallocate the send right, cancel/null out the XPC connection and call
didFinishLaunchingProcess(0, IPC::Connection::Identifier()).

Then the xpc_connection_send_message_with_reply() block would get called later and
would try to deallocate the send right again, and call
didFinishLaunchingProcess(0, IPC::Connection::Identifier()) again because the xpc
connection is null.

Handle this case by having the xpc_connection_send_message_with_reply() block check
that we're still launching (i.e. didFinishLaunchingProcess() has not been called
yet).

• UIProcess/Launcher/mac/ProcessLauncherMac.mm:

(WebKit::ProcessLauncher::launchProcess):

8:23 PM Changeset in webkit [238340] by rniwa@webkit.org

• 2 edits in trunk/Tools

REGRESSION (r238294): TestWebKitAPI.WKNavigation.ProcessCrashDuringCallback failing on iOS
​https://bugs.webkit.org/show_bug.cgi?id=191783

Reviewed by Chris Dumez.

Allow either WKErrorWebContentProcessTerminated or WKErrorWebViewInvalidated since the test
is closing WebView, thereby invalidating the view, which is racing with the process termination.

• TestWebKitAPI/Tests/WebKitCocoa/WebContentProcessDidTerminate.mm:

(TEST):

8:01 PM Changeset in webkit [238339] by Chris Dumez

• 2 edits in trunk/Tools

ProcessSwap.DisableForInspector API test is failing
​https://bugs.webkit.org/show_bug.cgi?id=191798

Reviewed by Joseph Pecoraro.

We support process swapping when Web Inspector is shown now so update the
API test accordingly.

• TestWebKitAPI/Tests/WebKitCocoa/ProcessSwapOnNavigation.mm:

7:06 PM Changeset in webkit [238338] by Joseph Pecoraro

• 2 edits in trunk/Source/WebKit

Unreviewed follow-up to r238330.

• UIProcess/WebPageProxy.cpp:

(WebKit::WebPageProxy::close):
Don't clear the pointer after invalidating since the pointer is
used later one. We don't mind it being used after invalidation
it will do the right thing, we'd rather have the WebInspectorProxy
lifetime match the WebPageProxy.

7:06 PM Changeset in webkit [238337] by Simon Fraser

• 3 edits in trunk/Source/WebCore

Optimize composited iframe layer hookup
​https://bugs.webkit.org/show_bug.cgi?id=191778

Reviewed by Zalan Bujtas.

The change made in r238229 can be more targeted; we only need to hook up iframe content
layers when the layer is already composited (otherwise the updateBacking() above would have triggered
the work), and when it's a RenderWidget layer.

Tested by existing tests.

• rendering/RenderLayer.cpp:

(WebCore::outputPaintOrderTreeRecursive):

• rendering/RenderLayerCompositor.cpp:

(WebCore::RenderLayerCompositor::layerStyleChanged):

6:46 PM Changeset in webkit [238336] by Ross Kirsling

• 5 edits
  1 delete in trunk/Source/WebCore

Provide default implementation of Widget
​https://bugs.webkit.org/show_bug.cgi?id=191784

Reviewed by Michael Catanzaro.

• SourcesWPE.txt:
• platform/Widget.cpp:

(WebCore::Widget::Widget):
(WebCore::Widget::frameRect const):
(WebCore::Widget::~Widget):
(WebCore::Widget::setFrameRect):
(WebCore::Widget::paint):
(WebCore::Widget::setFocus):
(WebCore::Widget::setCursor):
(WebCore::Widget::show):
(WebCore::Widget::hide):
(WebCore::Widget::setIsSelected):

• platform/gtk/WidgetGtk.cpp:

(WebCore::Widget::Widget): Deleted.
(WebCore::Widget::frameRect const): Deleted.

• platform/win/WidgetWin.cpp:

(WebCore::Widget::Widget): Deleted.
(WebCore::Widget::frameRect const): Deleted.

• platform/wpe/WidgetWPE.cpp: Removed.

6:26 PM Changeset in webkit [238335] by Chris Dumez

• 4 edits in trunk/Source/WebKit

ProcessSwap.NavigationWithLockedHistoryWithoutPSON API test is a flaky crash in debug
​https://bugs.webkit.org/show_bug.cgi?id=191795

Reviewed by Alex Christensen.

If the WebProcessPool is getting destroyed while there is a prewarmed WebContent process
that is awaiting its connnection to the network process, then we'll hit the
process->hasOneRef() assertion in the WebProcessPool destructor. This is because
NetworkProcessProxy::m_pendingConnectionReplies keeps a RefPtr to the WebProcessProxy
that is awaiting the connection. The NetworkProcessProxy does not really need to keep
the WebProcessProxy alive in this case so this patch switches to using a WeakPtr instead.

• UIProcess/Network/NetworkProcessProxy.cpp:

(WebKit::NetworkProcessProxy::getNetworkProcessConnection):
(WebKit::NetworkProcessProxy::networkProcessCrashed):

• UIProcess/Network/NetworkProcessProxy.h:
• UIProcess/WebProcessProxy.h:

6:14 PM Changeset in webkit [238334] by Devin Rousso

• 6 edits in trunk/Source/WebInspectorUI

Web Inspector: Audit: add default tests
​https://bugs.webkit.org/show_bug.cgi?id=191758

Reviewed by Matt Baker.

• UserInterface/Controllers/AuditManager.js:

(WI.AuditManager.prototype.loadStoredTests):
(WI.AuditManager.prototype.addDefaultTestsIfNeeded): Added.
If there are no previously saved tests in the database, automatically add the default set.

• UserInterface/Views/AuditNavigationSidebarPanel.js:

(WI.AuditNavigationSidebarPanel.prototype._addTest):
(WI.AuditNavigationSidebarPanel.prototype._addResult):
(WI.AuditNavigationSidebarPanel.prototype._handleAuditTestRemoved):

• UserInterface/Views/AuditNavigationSidebarPanel.css:

(.sidebar > .panel.navigation.audit.has-results:not(.has-tests) > .content > .message-text-view): Added.
(.sidebar > .panel.navigation.audit.has-results:not(.has-tests) > .content > .message-text-view > .message): Added.
(.sidebar > .panel.navigation.audit.has-results:not(.has-tests) > .content > .message-text-view > button): Added.
When the last test is removed, show a placeholder message that allows the user to re-add the
default set of tests.

• UserInterface/Views/NavigationSidebarPanel.js:

(WI.NavigationSidebarPanel.prototype.showEmptyContentPlaceholder):
(WI.NavigationSidebarPanel.prototype._createEmptyContentPlaceholderIfNeeded):

• Localizations/en.lproj/localizedStrings.js:

6:10 PM Changeset in webkit [238333] by Devin Rousso

• 10 edits in trunk

Web Inspector: Audit: minor style improvements
​https://bugs.webkit.org/show_bug.cgi?id=191727

Reviewed by Joseph Pecoraro.

Source/WebInspectorUI:

• UserInterface/Controllers/AuditManager.js:

(WI.AuditManager):
(WI.AuditManager.prototype._handleFrameMainResourceDidChange): Added.
Clear test results when the main frame navigates, but don't clear the saved results.

• UserInterface/Models/AuditTestCase.js:

(WI.AuditTestCase.prototype.async run):

• UserInterface/Models/AuditTestCaseResult.js:

(WI.AuditTestCaseResult):
(WI.AuditTestCaseResult.async fromPayload):
(WI.AuditTestCaseResult.prototype.get resolvedDOMNodes): Added.
(WI.AuditTestCaseResult.prototype.toJSON):
Capture the full CSS path for any returned DOM nodes when the test finishes running, so that
if the node moves around the path isn't modified as well.

• UserInterface/Views/AuditTestCaseContentView.js:

(WI.AuditTestCaseContentView.prototype.layout):
Always show the URL for the test when viewing a saved result.

• UserInterface/Views/AuditTestContentView.css:

(.content-view-container > .content-view.audit-test > header p):
(.content-view.audit-test > header):

• UserInterface/Views/AuditTestCaseContentView.css:

(.content-view.audit-test-case > header): Added.
(.content-view.audit-test-case > header h1 > img):
(.content-view.audit-test-case > header > .metadata > .source): Added.
(.content-view.audit-test-case > header > .metadata > .duration):
(.content-view.audit-test-case > section table > tr > td:first-child):

• UserInterface/Views/AuditTestGroupContentView.css:

(.content-view.audit-test-group > header):
Horizontally align items in the header for visual consistency.

LayoutTests:

• inspector/audit/resources/audit-utilities.js:

5:34 PM Changeset in webkit [238332] by Chris Dumez

• 9 edits in trunk/Source

[macOS] Label "prewarmed" WebContent processes in Activity Monitor
​https://bugs.webkit.org/show_bug.cgi?id=191765
<rdar://problem/45953463>

Reviewed by Geoffrey Garen.

Source/WebCore:

• en.lproj/Localizable.strings:

Source/WebKit:

Label "prewarmed" WebContent processes in Activity Monitor to facilitate debugging.

• Shared/EntryPointUtilities/mac/XPCService/XPCServiceEntryPoint.mm:

(WebKit::XPCServiceInitializerDelegate::getExtraInitializationData):

• UIProcess/WebProcessProxy.cpp:

(WebKit::WebProcessProxy::getLaunchOptions):
(WebKit::WebProcessProxy::markIsNoLongerInPrewarmedPool):

• WebProcess/WebProcess.cpp:

(WebKit::WebProcess::markIsNoLongerPrewarmed):

• WebProcess/WebProcess.h:
• WebProcess/WebProcess.messages.in:
• WebProcess/cocoa/WebProcessCocoa.mm:

(WebKit::WebProcess::initializeProcessName):
(WebKit::WebProcess::updateProcessName):
(WebKit::WebProcess::platformInitializeProcess):

5:28 PM Changeset in webkit [238331] by Joseph Pecoraro

• 4 edits in trunk/Source/WebInspectorUI

Web Inspector: Include default filtering of InspectorBackend.dumpInspectorProtocolMessages with multi target backend
​https://bugs.webkit.org/show_bug.cgi?id=191785

Reviewed by Matt Baker.

• UserInterface/Debug/Bootstrap.js:

(dumpMessagesCurrentState):
(applyDumpMessagesState):
(WI.runBootstrapOperations):
Include tri-state for the logs button.

• UserInterface/Protocol/InspectorBackend.js:

(InspectorBackendClass.prototype.set filterMultiplexingBackendInspectorProtocolMessages):
(InspectorBackendClass.prototype.get filterMultiplexingBackendInspectorProtocolMessages):
(InspectorBackendClass.prototype._startOrStopAutomaticTracing):
Add new filtering state (the Setting already got added in the past).

• UserInterface/Protocol/LoggingProtocolTracer.js:

(WI.LoggingProtocolTracer):
(WI.LoggingProtocolTracer.prototype.set filterMultiplexingBackend):
(WI.LoggingProtocolTracer.prototype.get filterMultiplexingBackend):
(WI.LoggingProtocolTracer.prototype._processEntry):
Perform new filtering.

5:28 PM Changeset in webkit [238330] by Joseph Pecoraro

• 34 edits in trunk

Web Inspector: Keep Web Inspector window alive across process swaps (PSON) (Local Inspector)
​https://bugs.webkit.org/show_bug.cgi?id=191740
<rdar://problem/45470897>

Reviewed by Timothy Hatcher.

Source/JavaScriptCore:

• inspector/InspectorFrontendChannel.h:

Expose EnumTraits for ConnectionType for WebKit IPC messages.

Source/WebInspectorUI:

• UserInterface/Protocol/Connection.js:

(InspectorBackend.Connection.prototype._dispatchResponse):
Hide the TargetAgent does not exist message. This is necessary so that WebKitLegacy
tests don't include this line in output each test.

• UserInterface/Test.html:
• UserInterface/Test/Test.js:

(WI.loaded):
(WI.resetMainExecutionContext):
(WI.performOneTimeFrontendInitializationsUsingTarget):
Behave more like Main.js and initialize a MultiplexingBackendTarget when the
TargetAgent is available.

• UserInterface/Protocol/InspectorBackend.js:

(InspectorBackendClass.prototype.runAfterPendingDispatches):
Use the main target, a.k.a. the page connection. The backend target's
messages don't have any responses, so this was effectively not waiting
at all.

Source/WebKit:

When a web page asks to open a local Web Inspector, that inspector
will now connect in the UIProcess (WebPageProxy / WebPageInspectorController)
instead of the WebContentProcess (WebKit::WebInspector / InspectorController).

Previously a WebInspectorProxy was re-created every time the WebPageProxy's
WebPage / WebProcess was changed, effectively closing the Web Inspector
frontend when the WebPage was swapped or crashed.

This change keeps the WebInspectorProxy alive as long as the WebPageProxy
is alive. During process swaps, or process crashes, the WebInspectorProxy
is reset when the page is reset and updated when the page's WebProcess
changes. Since WebInspectorProxy owns the Web Inspector view / window
the Web Inspector window can be kept alive across (and attached state)
across WebPage / WebProcess changes.

Previously the WebContentProcess's WebKit::WebInspector was the Local
FrontendChannel for the WebCore::InspectorController. This can't
outlive the single WebContentProcess.

This change makes the UIProcesses' WebInspectorProxy the Local FrontendChannel
for the WebKit::WebPageInspectorController. Given the WebInspectorProxy
now stays alive alongside the WebPageProxy this will live across process changes.
This means that the WebInspectorUI process must send its backend messages to
the WebInspectorProxy -> WebPageInspectorController now instead of the
old path WebInspector -> WebCore::InspectorController.

A direct IPC connection is still maintained between the WebContentProcess's
WebKit::WebInspector and the InspectorProcess's WebInspectorUI. Previously
this connection was established by WebKit::WebInspector vending an
IPC::Attachment to the WebInspectorUI process. This patch inverts that
relationship, because the WebInspectorUI process now lives across multiple
WebContentProcess changes. The WebInspectorUI now vends the IPC::Attachment
to the WebContentProcess each time the process changes. This way they can both
still communicate through normal Messages::WebInspector/WebInspectorUI
messages and everything behaves as previously expected.

• Scripts/webkit/messages.py:

Header for Inspector::FrontendChannel::ConnectionType.

• WebProcess/WebCoreSupport/WebInspectorClient.cpp:

(WebKit::WebInspectorClient::openLocalFrontend):

• WebProcess/WebPage/WebInspector.cpp:

(WebKit::WebInspector::openLocalInspectorFrontend):
(WebKit::WebInspector::setFrontendConnection):
(WebKit::WebInspector::closeFrontendConnection):
(WebKit::WebInspector::close):
(WebKit::WebInspector::openFrontendConnection): Deleted.
(WebKit::WebInspector::sendMessageToBackend): Deleted.
(WebKit::WebInspector::sendMessageToFrontend): Deleted.
On the WebProcess side we ask the UIProcess to open a local frontend and now
receive instead of vend an IPC connection to the WebInspectorUI process.

• UIProcess/WebPageProxy.cpp:

(WebKit::WebPageProxy::finishAttachingToWebProcess):
(WebKit::WebPageProxy::close):
(WebKit::WebPageProxy::resetState):
Keep the WebInspectorProxy alive alongside the WebPageProxy.
Update it as processes / states change.

• UIProcess/WebInspectorProxy.h:
• UIProcess/WebInspectorProxy.messages.in:
• UIProcess/WebInspectorProxy.cpp:

(WebKit::WebInspectorProxy::openLocalInspectorFrontend):
(WebKit::WebInspectorProxy::createInspectorPage): Deleted.
(WebKit::WebInspectorProxy::sendMessageToFrontend):
(WebKit::WebInspectorProxy::closeFrontendPageAndWindow):
Open and close a local frontend by being the FrontendChannel on the UIProcess side.

(WebKit::WebInspectorProxy::sendMessageToBackend):
Dispatch WebInspectorUI backend messages to the UIProcess's InspectorController
now that the InspectorController is here instead of in the WebProcess.

(WebKit::WebInspectorProxy::setFrontendConnection):
Transfer the WebProcess <-> InspectorProcess IPC connection through us
because we are the one link between them when processes change.

(WebKit::WebInspectorProxy::invalidate):
(WebKit::WebInspectorProxy::frontendLoaded):
Be safer and handle inspectedPage being potentially null.

• WebProcess/WebPage/WebPageInspectorTargetController.cpp:

(WebKit::WebPageInspectorTargetController::connectInspector):

• WebProcess/WebPage/WebPageInspectorTargetController.h:
• WebProcess/WebPage/WebPageInspectorTargetFrontendChannel.cpp:

(WebKit::WebPageInspectorTargetFrontendChannel::create):
(WebKit::WebPageInspectorTargetFrontendChannel::WebPageInspectorTargetFrontendChannel):

• WebProcess/WebPage/WebPageInspectorTargetFrontendChannel.h:
• UIProcess/InspectorTargetProxy.cpp:

(WebKit::InspectorTargetProxy::connect):
Proxy the ConnectionType received on the UIProcess side to the WebProcess side
when connecting to sub-targets.

• WebProcess/WebPage/WebPage.cpp:

(WebKit::WebPage::connectInspector):
(WebKit::WebPage::setHasLocalInspectorFrontend): Deleted.

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

(WebKit::WebPageProxy::hasLocalInspectorFrontend const): Deleted.
(WebKit::WebPageProxy::setHasLocalInspectorFrontend): Deleted.

• UIProcess/WebPageProxy.messages.in:
• UIProcess/WebProcessPool.cpp:

(WebKit::WebProcessPool::processForNavigationInternal):
Eliminate the hasLocalInspectorFrontend state that was only used
to disable PSON. PSON no longer needs to be disabled for this reason.

• WebProcess/WebPage/WebInspector.h:
• WebProcess/WebPage/WebInspector.messages.in:
• WebProcess/WebPage/WebInspectorUI.cpp:

(WebKit::WebInspectorUI::establishConnection):
(WebKit::WebInspectorUI::updateConnection):
(WebKit::WebInspectorUI::closeWindow):
(WebKit::WebInspectorUI::sendMessageToBackend):

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

Vend an IPC connection on demand to the WebProcess side.

LayoutTests:

• inspector/unit-tests/target-manager-expected.txt:
• inspector/unit-tests/target-manager.html:

Update test to pass in both a MultiplexingBackendTarget and DirectBackendTarget world.

5:14 PM Changeset in webkit [238329] by rniwa@webkit.org

• 2 edits in trunk/Websites/perf.webkit.org

Manifest file can contain a test metric which references a non-existent test
​https://bugs.webkit.org/show_bug.cgi?id=191796

Reviewed by Dewei Zhu.

The bug was caused by a race condition between the manifest file fetching the list of tests and test metrics
and new tests and test metrics being added. Because we would fetch tests before test metrics, it was possible
for new test metrics which references a test not included in the fetched tests to be present in the test metrics.

Fixed the bug by changing the order of the queries so that test metrics are fetched before tests. This guarantees
that any test referenced by a test metric always exists and thefore included in the manifest file.

Unfortunately no new tests beucase this involes a race condition.

• public/include/manifest-generator.php:

5:10 PM Changeset in webkit [238328] by Ryan Haddad

• 5 edits in trunk/Tools

Update buildbot dashboard after Windows 7/10 updates
​https://bugs.webkit.org/show_bug.cgi?id=191219

Reviewed by Alexey Proskuryakov.

• BuildSlaveSupport/build.webkit.org-config/public_html/dashboard/Scripts/BubbleQueueServer.js:

(BubbleQueueServer):

• BuildSlaveSupport/build.webkit.org-config/public_html/dashboard/Scripts/Dashboard.js:
• BuildSlaveSupport/build.webkit.org-config/public_html/dashboard/Scripts/WebKitBuildbot.js:

(WebKitBuildbot):

• BuildSlaveSupport/build.webkit.org-config/public_html/dashboard/Styles/Main.css:

(table.queue-grid tr.platform td.logo::before):
(table.queue-grid tr.platform.windows-10 td.logo::before):
(table.queue-grid tr.platform.wincairo-windows-10 td.logo::before):
(table.queue-grid tr.platform.windows-7 td.logo::before):
(table.queue-grid tr.platform.wincairo-windows-10 img.logo):

4:49 PM Changeset in webkit [238327] by commit-queue@webkit.org

• 9 edits in trunk/Source/WebCore

Enable external playback for video in element fullscreen.
​https://bugs.webkit.org/show_bug.cgi?id=190359
rdar://problem/42560085

Patch by Jeremy Jones <​jeremyj@apple.com> on 2018-11-16
Reviewed by Jer Noble.

No new tests because we don't have a good way to test external playback.

Any video that might be auto-pipped from element fullscreen should also enable external playback for video out.
PiP and external playback are mutually exclusive. Instead of preventing PiP when external playback is active,
allow PiP, but disable external playback while PiP is active.

• html/HTMLMediaElement.cpp:

(WebCore::HTMLMediaElement::setVideoFullscreenStandby):

• html/HTMLMediaElement.h:
• platform/graphics/MediaPlayer.cpp:

(WebCore::MediaPlayer::setVideoFullscreenStandby):
(WebCore::MediaPlayer::videoFullscreenStandby const):

• platform/graphics/MediaPlayer.h:

(WebCore::MediaPlayerClient::mediaPlayerVideoFullscreenStandby const):

• platform/graphics/MediaPlayerPrivate.h:

(WebCore::MediaPlayerPrivateInterface::setVideoFullscreenStandby):

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

(WebCore::MediaPlayerPrivateAVFoundationObjC::setVideoFullscreenStandby):
(WebCore::MediaPlayerPrivateAVFoundationObjC::updateDisableExternalPlayback):

4:42 PM Changeset in webkit [238326] by fpizlo@apple.com

• 6 edits
  3 adds in trunk

All users of ArrayBuffer should agree on the same max size
​https://bugs.webkit.org/show_bug.cgi?id=191771

Reviewed by Mark Lam.

JSTests:

• stress/big-wasm-memory-grow-no-max.js: Added.

(foo):
(catch):

• stress/big-wasm-memory-grow.js: Added.

(foo):
(catch):

• stress/big-wasm-memory.js: Added.

(foo):
(catch):

Source/JavaScriptCore:

Array buffers cannot be larger than 0x7fffffff, because otherwise loading typedArray.length in the DFG/FTL would produce
a uint32 or would require a signedness check, neither of which sounds reasonable. It's better to just bound their max size
instead.

• runtime/ArrayBuffer.cpp:

(JSC::ArrayBufferContents::ArrayBufferContents):
(JSC::ArrayBufferContents::tryAllocate):
(JSC::ArrayBufferContents::transferTo):
(JSC::ArrayBufferContents::copyTo):
(JSC::ArrayBufferContents::shareWith):

• runtime/ArrayBuffer.h:
• wasm/WasmMemory.cpp:

(JSC::Wasm::Memory::tryCreate):
(JSC::Wasm::Memory::grow):

• wasm/WasmPageCount.h:

4:31 PM Changeset in webkit [238325] by Kocsen Chung

• 1 copy in tags/Safari-606.4.1.2.1

Tag Safari-606.4.1.2.1.~

4:22 PM Changeset in webkit [238324] by fpizlo@apple.com

• 3 edits in trunk/JSTests

Unreviewed, make some more tests not crash my computer by only running on instance of it. These tests do not need to
run for each JSC config since they're regression tests for runtime bugs.

• stress/json-stringified-overflow-2.js:
• stress/json-stringified-overflow.js:

4:17 PM Changeset in webkit [238323] by Chris Dumez

• 2 edits in trunk/Tools

Regression(PSON) WebKit.GeolocationTransitionToHighAccuracy API test is timing out
​https://bugs.webkit.org/show_bug.cgi?id=191789

Reviewed by Alex Christensen.

Relate the 2 test pages so they use the same process since WKContextSetMaximumNumberOfProcesses()
SPI no longer works with PSON / Process prewarming.

• TestWebKitAPI/Tests/WebKit/Geolocation.cpp:

(TestWebKitAPI::TEST):

4:15 PM Changeset in webkit [238322] by jer.noble@apple.com

• 9 edits in trunk/Source

Regression(r233865): Causes synchronous IPC in the middle of layout
​https://bugs.webkit.org/show_bug.cgi?id=188307
<rdar://problem/42807306>

Reviewed by Eric Carlson.

Revert the changes added in r233865. Rather than make a syncronous call to the UIProcess to
query whether the view has been backgrounded while (e.g.) JS has been spinning, perform the
steps of the requestFullscreen() method on the next run loop, allowing messages from the
UIProcess about page visibilty to be delivered first.

• dom/Document.cpp:

(WebCore::Document::requestFullScreenForElement):

• html/HTMLMediaElement.cpp:

(WebCore::HTMLMediaElement::enterFullscreen):

• html/HTMLMediaElement.h:
• page/ChromeClient.h:

4:15 PM Changeset in webkit [238321] by Ross Kirsling

• 3 edits
  1 delete in trunk/Source/WebCore

Provide default implementation of Cursor
​https://bugs.webkit.org/show_bug.cgi?id=191772

Reviewed by Michael Catanzaro.

• SourcesWPE.txt:
• platform/Cursor.cpp:

(WebCore::Cursor::ensurePlatformCursor const):

• platform/wpe/CursorWPE.cpp: Removed.

4:03 PM Changeset in webkit [238320] by Kocsen Chung

• 1 copy in tags/Safari-606.4.1

Tag Safari-606.4.1.

4:02 PM Changeset in webkit [238319] by Jonathan Bedard

• 2 edits in trunk/Tools

webkitpy: Retry app installation on watchOS simulators
​https://bugs.webkit.org/show_bug.cgi?id=191769
<rdar://problem/46139850>

Reviewed by Aakash Jain.

• Scripts/webkitpy/xcode/simulated_device.py:

(SimulatedDevice.install_app): Retry app installations.

3:52 PM Changeset in webkit [238318] by Michael Catanzaro

• 3 edits in trunk/Source/WebKit

[GTK][WPE] Add "WebKitDeviceInfoPermissionRequest.h" into webkit2.h
​https://bugs.webkit.org/show_bug.cgi?id=191744
<rdar://problem/46125461>

Unreviewed, install the file.

• PlatformGTK.cmake:
• PlatformWPE.cmake:

3:27 PM Changeset in webkit [238317] by don.olmstead@sony.com

• 2 edits
  1 copy
  1 add
  1 delete in trunk/Source/WebCore

EditorWPE should be EditorLibWPE
​https://bugs.webkit.org/show_bug.cgi?id=191774

Reviewed by Michael Catanzaro.

The platform specific implementations of Editor are all around the
platform's Pasteboard. Since the Pasteboard implementation is now based
around USE(LIBWPE) EditorWPE should be renamed to EditorLibWPE.

• SourcesWPE.txt:
• editing/libwpe/EditorLibWPE.cpp: Renamed from Source/WebCore/editing/wpe/EditorWPE.cpp.

3:22 PM Changeset in webkit [238316] by Michael Catanzaro

• 34 edits
  1 copy
  1 move
  1 add
  1 delete in releases/WebKitGTK/webkit-2.22

Merge r235419 - Fix exception throwing code so that topCallFrame and topEntryFrame stay true to their names.
​https://bugs.webkit.org/show_bug.cgi?id=188577
<rdar://problem/42985684>

Reviewed by Saam Barati.

JSTests:

• stress/regress-188577.js: Added.

Source/JavaScriptCore:

1. Introduced CallFrame::convertToStackOverflowFrame() which converts the current (top) CallFrame (which may not have a valid callee) into a StackOverflowFrame.

The StackOverflowFrame is a sentinel frame that the low level code (exception
throwing code, stack visitor, and stack unwinding code) will know to skip
over. The StackOverflowFrame will also have a valid JSCallee so that client
code can compute the globalObject or VM from this frame.

As a result, client code that throws StackOverflowErrors no longer need to
compute the caller frame to throw from: it just converts the top frame into
a StackOverflowFrame and everything should *Just Work*.

2. NativeCallFrameTracerWithRestore is now obsolete.

Instead, client code should always call convertToStackOverflowFrame() on the
frame before instantiating a NativeCallFrameTracer with it.

This means that topCallFrame will always point to the top CallFrame (which
may be a StackOverflowFrame), and topEntryFrame will always point to the top
EntryFrame. We'll never temporarily point them to the previous EntryFrame
(which we used to do with NativeCallFrameTracerWithRestore).

3. genericUnwind() and Interpreter::unwind() will now always unwind from the top CallFrame, and will know how to handle a StackOverflowFrame if they see one.

This obsoletes the UnwindStart flag.

• CMakeLists.txt:
• JavaScriptCore.xcodeproj/project.pbxproj:
• Sources.txt:
• debugger/Debugger.cpp:

(JSC::Debugger::pauseIfNeeded):

• interpreter/CallFrame.cpp:

(JSC::CallFrame::callerFrame const):
(JSC::CallFrame::unsafeCallerFrame const):
(JSC::CallFrame::convertToStackOverflowFrame):
(JSC::CallFrame::callerFrame): Deleted.
(JSC::CallFrame::unsafeCallerFrame): Deleted.

• interpreter/CallFrame.h:

(JSC::ExecState::iterate):

• interpreter/CallFrameInlines.h: Added.

(JSC::CallFrame::isStackOverflowFrame const):
(JSC::CallFrame::isWasmFrame const):

• interpreter/EntryFrame.h: Added.

(JSC::EntryFrame::vmEntryRecordOffset):
(JSC::EntryFrame::calleeSaveRegistersBufferOffset):

• interpreter/FrameTracers.h:

(JSC::NativeCallFrameTracerWithRestore::NativeCallFrameTracerWithRestore): Deleted.
(JSC::NativeCallFrameTracerWithRestore::~NativeCallFrameTracerWithRestore): Deleted.

• interpreter/Interpreter.cpp:

(JSC::Interpreter::unwind):

• interpreter/Interpreter.h:
• interpreter/StackVisitor.cpp:

(JSC::StackVisitor::StackVisitor):

• interpreter/StackVisitor.h:

(JSC::StackVisitor::visit):
(JSC::StackVisitor::topEntryFrameIsEmpty const):

• interpreter/VMEntryRecord.h:

(JSC::VMEntryRecord::callee const):
(JSC::EntryFrame::vmEntryRecordOffset): Deleted.
(JSC::EntryFrame::calleeSaveRegistersBufferOffset): Deleted.

• jit/AssemblyHelpers.h:
• jit/JITExceptions.cpp:

(JSC::genericUnwind):

• jit/JITExceptions.h:
• jit/JITOperations.cpp:
• llint/LLIntOffsetsExtractor.cpp:
• llint/LLIntSlowPaths.cpp:

(JSC::LLInt::LLINT_SLOW_PATH_DECL):

• llint/LowLevelInterpreter.asm:
• llint/LowLevelInterpreter32_64.asm:
• llint/LowLevelInterpreter64.asm:
• runtime/CallData.cpp:
• runtime/CommonSlowPaths.cpp:

(JSC::throwArityCheckStackOverflowError):
(JSC::SLOW_PATH_DECL):

• runtime/CommonSlowPathsExceptions.cpp: Removed.
• runtime/CommonSlowPathsExceptions.h: Removed.
• runtime/Completion.cpp:

(JSC::evaluateWithScopeExtension):

• runtime/JSGeneratorFunction.h:
• runtime/JSGlobalObject.cpp:

(JSC::JSGlobalObject::init):
(JSC::JSGlobalObject::visitChildren):

• runtime/JSGlobalObject.h:

(JSC::JSGlobalObject::stackOverflowFrameCallee const):

• runtime/VM.cpp:

(JSC::VM::throwException):

• runtime/VM.h:
• runtime/VMInlines.h:

(JSC::VM::topJSCallFrame const):

LayoutTests:

• http/tests/misc/large-js-program-expected.txt:

3:22 PM Changeset in webkit [238315] by Michael Catanzaro

• 3 edits in releases/WebKitGTK/webkit-2.22/Source/JavaScriptCore

Merge r235198 - Eliminate dead code operationThrowDivideError() and operationThrowOutOfBoundsAccessError()
​https://bugs.webkit.org/show_bug.cgi?id=188859

Rubber-stamped by Saam Barati.

Deleted these two functions.

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

3:22 PM Changeset in webkit [238314] by Michael Catanzaro

• 4 edits in releases/WebKitGTK/webkit-2.22/Source/WebCore

Merge r238291 - An early return from updateSchedulingInfo can leave some variables uninitialized.
​https://bugs.webkit.org/show_bug.cgi?id=191755
<rdar://problem/39910089>

Reviewed by Brent Fulgham.

• Modules/webaudio/AudioBufferSourceNode.cpp:

(WebCore::AudioBufferSourceNode::process):

• Modules/webaudio/AudioScheduledSourceNode.cpp:

(WebCore::AudioScheduledSourceNode::updateSchedulingInfo):

• Modules/webaudio/OscillatorNode.cpp:

(WebCore::OscillatorNode::process):

3:22 PM Changeset in webkit [238313] by Michael Catanzaro

• 3 edits
  2 adds in releases/WebKitGTK/webkit-2.22

Merge r238222 - Remove fonts from CSSFontFaceSet safely
​https://bugs.webkit.org/show_bug.cgi?id=191676

Reviewed by Zalan Bujtas.

Source/WebCore:

Test: fast/text/font-face-set-remove-safely.html

• css/CSSFontFaceSet.cpp:

(WebCore::CSSFontFaceSet::remove):

LayoutTests:

• fast/text/font-face-set-remove-safely-expected.txt: Added.
• fast/text/font-face-set-remove-safely.html: Added.

3:22 PM Changeset in webkit [238312] by Michael Catanzaro

• 7 edits
  2 adds in releases/WebKitGTK/webkit-2.22

Merge r238119 - Do not collapse the soon-to-be-parent anon block when we shuffle around the marker item renderer.
​https://bugs.webkit.org/show_bug.cgi?id=191554
<rdar://problem/45825265>

Reviewed by Antti Koivisto.

Source/WebCore:

While moving the marker item renderer to its correct subtree, we accidentally remove the soon-to-be parent anonymous block.
Moving a renderer is a 2 step process:

1. Detach the renderer from its current parent
2. Attach it to its new parent.

During step #1, we check if there is a chance to collapse anonymous blocks. In this case the soon-to-be-parent is a sibling anonymous block which, after detaching the marker sibling
is not needed anymore (except we use it as the new parent).

Test: fast/inline/marker-list-item-move-should-not-crash.html

• rendering/updating/RenderTreeBuilder.cpp:

(WebCore::RenderTreeBuilder::detach):

• rendering/updating/RenderTreeBuilder.h:
• rendering/updating/RenderTreeBuilderBlock.cpp:

(WebCore::RenderTreeBuilder::Block::detach):

• rendering/updating/RenderTreeBuilderBlock.h:
• rendering/updating/RenderTreeBuilderList.cpp:

(WebCore::RenderTreeBuilder::List::updateItemMarker):

LayoutTests:

• fast/inline/marker-list-item-move-should-not-crash-expected.txt: Added.
• fast/inline/marker-list-item-move-should-not-crash.html: Added.

3:19 PM Changeset in webkit [238311] by Michael Catanzaro

• 8 edits
  2 adds in releases/WebKitGTK/webkit-2.22

Merge r237837 - [MediaStream] An audio track should be muted when capture is interrupted by the OS.
Source/WebCore:

​https://bugs.webkit.org/show_bug.cgi?id= 191283

<rdar://problem/45773103>

Patch by Eric Carlson <​eric.carlson@apple.com> on 2018-11-05
Reviewed by Jon Lee.

Test: fast/mediastream/media-stream-track-interrupted.html

• platform/mediastream/RealtimeMediaSource.cpp:

(WebCore::RealtimeMediaSource::setInterruptedForTesting):

• platform/mediastream/RealtimeMediaSource.h:
• platform/mediastream/mac/CoreAudioCaptureSource.cpp:

(WebCore::CoreAudioCaptureSource::beginInterruption):
(WebCore::CoreAudioCaptureSource::endInterruption):

• testing/Internals.cpp:

(WebCore::Internals::setMediaStreamSourceInterrupted):

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

LayoutTests:

​https://bugs.webkit.org/show_bug.cgi?id=191283

<rdar://problem/45773103>

Patch by Eric Carlson <​eric.carlson@apple.com> on 2018-11-05
Reviewed by Jon Lee.

• fast/mediastream/media-stream-track-interrupted-expected.txt: Added.
• fast/mediastream/media-stream-track-interrupted.html: Added.

3:17 PM Changeset in webkit [238310] by fpizlo@apple.com

• 3 edits in trunk/JSTests

Unreviewed, make some tests not crash my computer by only running on instance of it. These tests do not need to run for each JSC
config since they're regression tests for runtime bugs.

• stress/large-unshift-splice.js:
• stress/regress-185888.js:

3:10 PM Changeset in webkit [238309] by Kocsen Chung

• 7 edits in tags/Safari-607.1.14.1/Source

Versioning.

3:08 PM Changeset in webkit [238308] by Kocsen Chung

• 1 copy in tags/Safari-607.1.14.1

New tag.

2:55 PM Changeset in webkit [238307] by commit-queue@webkit.org

• 8 edits in trunk

Include AirPlay destination name in AirPlay placard.
​https://bugs.webkit.org/show_bug.cgi?id=191574
rdar://problem/45536144

Patch by Jeremy Jones <​jeremyj@apple.com> on 2018-11-16
Reviewed by Eric Carlson.

Source/WebCore:

Updated existing tests.

Include the name of the AirPlay destination in the video element's AirPlay placard.

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

(AirplayPlacard):

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

(PlacardSupport.prototype._updatePlacard):
(PlacardSupport.prototype._updateAirPlayPlacard):
(PlacardSupport):

• en.lproj/modern-media-controls-localized-strings.js:

LayoutTests:

Updated tests for string changes.

• media/modern-media-controls/airplay-placard/airplay-placard-expected.txt:
• media/modern-media-controls/airplay-placard/airplay-placard.html:

2:54 PM Changeset in webkit [238306] by Alan Bujtas

• 14 edits
  2 copies in trunk

Add DidFirstMeaningfulPaint milestone.
​https://bugs.webkit.org/show_bug.cgi?id=191754

Reviewed by Simon Fraser.

Source/WebCore:

This milestone fires sone after the paint triggered by the first visually non-empty layout.

• page/FrameView.cpp:

(WebCore::FrameView::fireLayoutRelatedMilestonesIfNeeded):
(WebCore::FrameView::firePaintRelatedMilestonesIfNeeded):

• page/LayoutMilestone.h:

Source/WebKit:

• Shared/API/c/WKPageLoadTypes.h:
• Shared/API/c/WKSharedAPICast.h:

(WebKit::toWKLayoutMilestones):
(WebKit::toLayoutMilestones):

• UIProcess/API/C/WKPageRenderingProgressEvents.h:
• UIProcess/API/C/WKPageRenderingProgressEventsInternal.h:

(pageRenderingProgressEvents):

Tools:

• TestWebKitAPI/CMakeLists.txt:
• TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj:
• TestWebKitAPI/Tests/WebKit/FirstMeaningfulPaintMilestone.cpp: Added.

(TestWebKitAPI::didReachMilestone):
(TestWebKitAPI::setPageLoaderClient):
(TestWebKitAPI::TEST):

• TestWebKitAPI/Tests/WebKit/FirstMeaningfulPaintMilestone_Bundle.cpp: Copied from Source/WebKit/UIProcess/API/C/WKPageRenderingProgressEvents.h.

(TestWebKitAPI::FirstMeaningfulPaintMilestoneTest::FirstMeaningfulPaintMilestoneTest):
(TestWebKitAPI::FirstMeaningfulPaintMilestoneTest::didCreatePage):

2:45 PM Changeset in webkit [238305] by Matt Baker

• 2 edits in trunk/Source/WebInspectorUI

Web Inspector: TreeOutline shouldn't select first child of an expanded element on right arrow key
​https://bugs.webkit.org/show_bug.cgi?id=191766

Reviewed by Devin Rousso.

• UserInterface/Views/TreeOutline.js:

(WI.TreeOutline.prototype._treeKeyDown):

2:16 PM Changeset in webkit [238304] by don.olmstead@sony.com

• 4 edits
  1 copy
  1 delete in trunk/Source/WebCore

Provide default implementations of Image and Icon
​https://bugs.webkit.org/show_bug.cgi?id=191764

Reviewed by Michael Catanzaro.

Makes IconWPE and ImageWPE implementations the default.

• Sources.txt:
• SourcesWPE.txt:
• platform/graphics/Icon.cpp: Renamed from Source/WebCore/platform/graphics/wpe/IconWPE.cpp.
• platform/graphics/Image.cpp:

(WebCore::BitmapImage::invalidatePlatformData): Placed here for consistency with implementing ports.
(WebCore::Image::loadPlatformResource):

• platform/graphics/wpe/ImageWPE.cpp: Removed.

2:15 PM Changeset in webkit [238303] by Kocsen Chung

• 3 edits
  1 add
  1 delete in branches/safari-606.4.1.2-branch

Cherry-pick r238270. rdar://problem/46138760

2018-11-15 Mark Lam <​mark.lam@apple.com>

RegExpObject's collectMatches should not be using JSArray::push to fill in its match results.
​https://bugs.webkit.org/show_bug.cgi?id=191730
<rdar://problem/46048517>

Reviewed by Saam Barati.

JSTests:

• stress/regress-187006.js: Removed.
  • this test is invalid because its sole purpose is to test for the non-spec compliant behavior that we just fixed.

• stress/regress-191730.js: Added.

Source/JavaScriptCore:

According to the spec ​https://www.ecma-international.org/ecma-262/9.0/index.html#sec-regexp.prototype-@@match,
the RegExp match results are filled in using the spec's CreateDataProperty()
function which does not consult the prototype for setters. JSArray:push()
consults the prototype for setters. We should be using putDirectIndex() instead.

• runtime/RegExpObjectInlines.h: (JSC::collectMatches):

2:15 PM Changeset in webkit [238302] by Kocsen Chung

• 4 edits
  1 add in branches/safari-606.4.1.2-branch

Cherry-pick r238267. rdar://problem/46138752

2018-11-15 Mark Lam <​mark.lam@apple.com>

RegExp operations should not take fast patch if lastIndex is not numeric.
​https://bugs.webkit.org/show_bug.cgi?id=191731
<rdar://problem/46017305>

Reviewed by Saam Barati.

JSTests:

• stress/regress-191731.js: Added.

Source/JavaScriptCore:

This is because if lastIndex is an object with a valueOf() method, it can execute
arbitrary code which may have side effects, and side effects are not permitted by
the RegExp fast paths.

• builtins/RegExpPrototype.js: (globalPrivate.hasObservableSideEffectsForRegExpMatch): (overriddenName.string_appeared_here.search): (globalPrivate.hasObservableSideEffectsForRegExpSplit): (intrinsic.RegExpTestIntrinsic.test):
• builtins/StringPrototype.js: (globalPrivate.hasObservableSideEffectsForStringReplace):

2:11 PM Changeset in webkit [238301] by Kocsen Chung

• 7 edits in branches/safari-606.4.1.2-branch/Source

Versioning.

1:54 PM Changeset in webkit [238300] by Kocsen Chung

• 1 copy in branches/safari-606.4.1.2-branch

New branch.

1:46 PM Changeset in webkit [238299] by Ryan Haddad

• 2 edits in trunk/LayoutTests

REGRESSION (r232376): Layout Test inspector/console/webcore-logging.html is a flaky failure
​https://bugs.webkit.org/show_bug.cgi?id=186425

Unreviewed test gardening.

• platform/mac-wk2/TestExpectations: Remove flaky expectation.

1:43 PM Changeset in webkit [238298] by Brent Fulgham

• 23 edits
  1 add in trunk/Source

[Win] Reduce the use of WKSI library calls: CFNetwork
​https://bugs.webkit.org/show_bug.cgi?id=191718
<rdar://problem/46108732>

Reviewed by Alex Christensen.

Source/WebCore:

Remove custom WKSI CFNetwork calls, since the SPI is already documented in PAL. Just
make the same calls on Windows, like we did for iOS and macOS back in 2017.

Stop including WebKitSystemInterface.h for files that used to rely on it
for font-related features.

Tested by existing Windows regression tests. There should be no change in behavior.

• platform/graphics/win/FontCacheWin.cpp:
• platform/graphics/win/FontCustomPlatformData.cpp:
• platform/graphics/win/FontPlatformDataCGWin.cpp:
• platform/network/cf/CookieStorageCFNet.cpp:
• platform/network/cf/CredentialStorageCFNet.cpp:
• platform/network/cf/NetworkStorageSessionCFNet.cpp:

(WebCore::createPrivateStorageSession):
(WebCore::NetworkStorageSession::switchToNewTestingSession):
(WebCore::NetworkStorageSession::ensureSession):

• platform/network/cf/ResourceError.h:
• platform/network/cf/ResourceErrorCF.cpp:

(WebCore::getSSLPeerCertificateData):
(WebCore::setSSLPeerCertificateData):
(WebCore::ResourceError::getSSLPeerCertificateDataBytePtr):
(WebCore::ResourceError::platformLazyInit):
(WebCore::ResourceError::cfError const):

• platform/network/cf/ResourceHandleCFNet.cpp:

(WebCore::setClientCertificateInSSLProperties):
(WebCore::ResourceHandle::createCFURLConnection):

• platform/network/cf/ResourceHandleCFURLConnectionDelegate.cpp:
• platform/network/cf/ResourceRequestCFNet.cpp:
• platform/network/cf/SocketStreamHandleImplCFNet.cpp:

Source/WebCore/PAL:

Add necessary declarations for CFNetwork SPI needed by the Apple Windows build.

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

Source/WebKitLegacy/win:

Remove custom WKSI CFNetwork calls, since the SPI is already documented in PAL. Just
make the same calls on Windows, like we did for iOS and macOS back in 2017.

• WebCache.cpp:

(WebCache::cacheFolder):

• WebCoreSupport/WebFrameNetworkingContext.cpp:
• WebCoreSupport/WebPlatformStrategies.cpp:
• WebError.cpp:

(WebError::localizedDescription):
(WebError::sslPeerCertificate):

• WebURLResponse.cpp:

(WebURLResponse::sslPeerCertificate):
(WebURLResponse::certificateDictionary const):

• WebView.cpp:

(WebView::setCacheModel):

12:42 PM Changeset in webkit [238297] by sbarati@apple.com

• 3 edits
  1 add in trunk

KnownCellUse should also have SpecCellCheck as its type filter
​https://bugs.webkit.org/show_bug.cgi?id=191729
<rdar://problem/45872852>

Reviewed by Filip Pizlo.

JSTests:

• stress/known-cell-type-check-should-allow-empty-value-to-flow-through.js: Added.

(C):

Source/JavaScriptCore:

We write transformations in the compiler like this where we emit edges with
KnownCellUse if we know we're inserting code at a point where we're dominated
by a Cell check:

a: SomeValue
b: Something(Cell:@a)
c: SomethingElse(@b)
d: CheckNotEmpty(@a)

=>

a: SomeValue
b: Something(Cell:@a)
e: RandomOtherThing(KnownCellUse:@a)
c: SomethingElse(@b)
d: CheckNotEmpty(@a)

However, doing this used to lead to subtly incorrect programs since KnownCellUse
did not allow the empty value to flow through it. We used to end up incorrectly
deleting @d in the above program. We fix this, we make KnownCellUse allow the empty
value to flow through.

• dfg/DFGUseKind.h:

(JSC::DFG::typeFilterFor):

12:18 PM Changeset in webkit [238296] by commit-queue@webkit.org

• 2 edits in trunk/Tools

[WinCairo][WebKitTestRunner] Test contents window is displayed while testing.
​https://bugs.webkit.org/show_bug.cgi?id=191593

Patch by Takashi Komori <​Takashi.Komori@sony.com> on 2018-11-16
Reviewed by Alex Christensen.

Hide TestRunner window and taskbar button.

• WebKitTestRunner/win/PlatformWebViewWin.cpp:

(WTR::PlatformWebView::PlatformWebView):
(WTR::PlatformWebView::windowSnapshotImage):

12:13 PM Changeset in webkit [238295] by timothy@apple.com

• 2 edits in trunk/Source/WebCore

Add html{color:text} to the simpleUserAgentStyleSheet on macOS.
​https://bugs.webkit.org/show_bug.cgi?id=191760

Reviewed by Antti Koivisto.

• css/CSSDefaultStyleSheets.cpp: Added html{color:text} to simpleUserAgentStyleSheet inside

a PLATFORM(MAC) copy of the string.

12:11 PM Changeset in webkit [238294] by Chris Dumez

• 2 edits in trunk/Source/WebKit

Automatic process prewarming should be turned on if PSON is enabled via experimental features
​https://bugs.webkit.org/show_bug.cgi?id=191761

Reviewed by Ryosuke Niwa.

m_isAutomaticProcessWarmingEnabledByClient is a std::optional and should not be initialized to
true / false until the client calls API to toggle the feature. When m_isAutomaticProcessWarmingEnabledByClient
is std::nullopt then m_processSwapsOnNavigationFromExperimentalFeatures controls automatic
process prewarming.

• UIProcess/API/APIProcessPoolConfiguration.h:

11:59 AM Changeset in webkit [238293] by achristensen@apple.com

• 6 edits in trunk

Add API to disable safe browsing checks
​https://bugs.webkit.org/show_bug.cgi?id=191610

Reviewed by Dean Jackson.

Source/WebKit:

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

(-[WKPreferences setSafeBrowsingEnabled:]):
(-[WKPreferences isSafeBrowsingEnabled]):
(-[WKPreferences _setSafeBrowsingEnabled:]): Deleted.
(-[WKPreferences _safeBrowsingEnabled]): Deleted.

• UIProcess/API/Cocoa/WKPreferencesPrivate.h:

Tools:

• TestWebKitAPI/Tests/WebKitCocoa/SafeBrowsing.mm:

(resourceURL):
(TEST):
(safeBrowsingView):
(simpleURL): Deleted.

11:19 AM Changeset in webkit [238292] by Alan Coon

• 3 edits
  2 adds in branches/safari-606-branch

Cherry-pick r238222. rdar://problem/45997333

Remove fonts from CSSFontFaceSet safely
​https://bugs.webkit.org/show_bug.cgi?id=191676

Reviewed by Zalan Bujtas.

Source/WebCore:

Test: fast/text/font-face-set-remove-safely.html

• css/CSSFontFaceSet.cpp: (WebCore::CSSFontFaceSet::remove):

LayoutTests:

• fast/text/font-face-set-remove-safely-expected.txt: Added.
• fast/text/font-face-set-remove-safely.html: Added.

git-svn-id: ​https://svn.webkit.org/repository/webkit/trunk@238222 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11:06 AM Changeset in webkit [238291] by jer.noble@apple.com

• 4 edits in trunk/Source/WebCore

An early return from updateSchedulingInfo can leave some variables uninitialized.
​https://bugs.webkit.org/show_bug.cgi?id=191755
<rdar://problem/39910089>

Reviewed by Brent Fulgham.

• Modules/webaudio/AudioBufferSourceNode.cpp:

(WebCore::AudioBufferSourceNode::process):

• Modules/webaudio/AudioScheduledSourceNode.cpp:

(WebCore::AudioScheduledSourceNode::updateSchedulingInfo):

• Modules/webaudio/OscillatorNode.cpp:

(WebCore::OscillatorNode::process):

11:00 AM Changeset in webkit [238290] by Chris Dumez

• 2 edits in trunk/Source/WebKit

WebKit.UnavailablePlugIn API test is crashing in debug
​https://bugs.webkit.org/show_bug.cgi?id=191753

Reviewed by Alex Christensen.

Add missing early returns. Those were mistakenly dropped in r235200.

• UIProcess/Cocoa/NavigationState.mm:

(WebKit::NavigationState::NavigationClient::decidePolicyForPluginLoad):

10:52 AM Changeset in webkit [238289] by Chris Dumez

• 4 edits in trunk/Source

[Mac] Regression: WebContent process's display name is no longer set
​https://bugs.webkit.org/show_bug.cgi?id=191722
<rdar://problem/45960550>

Reviewed by Per Arne Vollan.

Source/WebCore/PAL:

• pal/spi/cocoa/LaunchServicesSPI.h:

Source/WebKit:

Make sure The WebContent process calls _RegisterApplication() during initialization and
*before* trying to set its display name. Setting the display name fails otherwise.

• WebProcess/cocoa/WebProcessCocoa.mm:

(WebKit::WebProcess::initializeProcessName):
(WebKit::WebProcess::platformInitializeProcess):

10:20 AM Changeset in webkit [238288] by rniwa@webkit.org

• 2 edits in trunk/Source/WebKit

Enable process swap on cross-site navigation by default on iOS
​https://bugs.webkit.org/show_bug.cgi?id=191735

Reviewed by Chris Dumez.

Enabled the feature.

• Shared/WebPreferencesDefaultValues.h:

10:14 AM Changeset in webkit [238287] by aboya@igalia.com

• 2 edits in trunk/Source/WebCore

[MSE][GStreamer] Ignore caps notifications when tearing down the pipeline
​https://bugs.webkit.org/show_bug.cgi?id=191578

Reviewed by Xabier Rodriguez-Calvar.

Changing the demuxer to READY state (which is done only in the main
thread) triggers the unlinking of its srcpads, which in turns emits a
caps change notification in the previously linked element since they
become unnegotiated again.

We are not interested in caps notifications in these cases, so let's
just ignore caps notifications emitted from the main thread.

This fixes an assertion failure in the debug builds.

• platform/graphics/gstreamer/mse/AppendPipeline.cpp:

(WebCore::AppendPipeline::AppendPipeline):

9:51 AM Changeset in webkit [238286] by Antti Koivisto

• 2 edits in trunk/Source/WebKit

REGRESSION(r237467) [PSON] iOS: Going back to a page sometimes doesn't restore the full page
​https://bugs.webkit.org/show_bug.cgi?id=191737

Reviewed by Simon Fraser.

• WebProcess/WebPage/WebPage.cpp:

(WebKit::WebPage::setLayerTreeStateIsFrozen):

Layer tree should stay frozen in m_shouldResetDrawingArea state. Otherwise spurious commit transactions may
mess up state on UI process side.

Specifically WKWebView._needsResetViewStateAfterCommitLoadForMainFrame would get stuck to TRUE and prevent
all viewport updates because WKWebView._firstPaintAfterCommitLoadTransactionID has value greater
than any current transaction IDs.

9:28 AM Changeset in webkit [238285] by Kocsen Chung

• 7 edits
  2 adds in branches/safari-606-branch

Cherry-pick r238119. rdar://problem/45997459

Do not collapse the soon-to-be-parent anon block when we shuffle around the marker item renderer.
​https://bugs.webkit.org/show_bug.cgi?id=191554
<rdar://problem/45825265>

Reviewed by Antti Koivisto.

Source/WebCore:

While moving the marker item renderer to its correct subtree, we accidentally remove the soon-to-be parent anonymous block.
Moving a renderer is a 2 step process:

1. Detach the renderer from its current parent
2. Attach it to its new parent. During step #1, we check if there is a chance to collapse anonymous blocks. In this case the soon-to-be-parent is a sibling anonymous block which, after detaching the marker sibling is not needed anymore (except we use it as the new parent).

Test: fast/inline/marker-list-item-move-should-not-crash.html

• rendering/updating/RenderTreeBuilder.cpp: (WebCore::RenderTreeBuilder::detach):
• rendering/updating/RenderTreeBuilder.h:
• rendering/updating/RenderTreeBuilderBlock.cpp: (WebCore::RenderTreeBuilder::Block::detach):
• rendering/updating/RenderTreeBuilderBlock.h:
• rendering/updating/RenderTreeBuilderList.cpp: (WebCore::RenderTreeBuilder::List::updateItemMarker):

LayoutTests:

• fast/inline/marker-list-item-move-should-not-crash-expected.txt: Added.
• fast/inline/marker-list-item-move-should-not-crash.html: Added.

git-svn-id: ​https://svn.webkit.org/repository/webkit/trunk@238119 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9:12 AM Changeset in webkit [238284] by Chris Dumez

• 2 edits in trunk/Source/WebKit

Unreviewed typo fix in comment added in r238250.

• UIProcess/WebProcessPool.cpp:

(WebKit::WebProcessPool::ensureNetworkProcess):

8:58 AM Changeset in webkit [238283] by sihui_liu@apple.com

• 19 edits in trunk

Storing blobs in IDB on iOS: "Error preparing blob/file"
​https://bugs.webkit.org/show_bug.cgi?id=188438
<rdar://problem/43097279>

Reviewed by Alex Christensen.

Source/WebCore:

Grant sandbox extension of temp folder to network process, because we suggested network process had access to
the folder on iOS device but it didn't.

This makes some existing tests about IDB blob pass on iOS device.

• Modules/indexeddb/server/IDBBackingStore.h:
• Modules/indexeddb/server/SQLiteIDBTransaction.cpp:

(WebCore::IDBServer::SQLiteIDBTransaction::moveBlobFilesIfNecessary):
(WebCore::IDBServer::SQLiteIDBTransaction::deleteBlobFilesIfNecessary):
(WebCore::IDBServer::SQLiteIDBTransaction::abort):

• Modules/indexeddb/shared/InProcessIDBServer.h:

Source/WebKit:

Grant sandbox extension of temp folder to network process, and clean up some code that managed sandbox extension
of IDB temporary blob files in network process.

• NetworkProcess/FileAPI/NetworkBlobRegistry.cpp:

(WebKit::NetworkBlobRegistry::registerBlobURLOptionallyFileBacked):

• NetworkProcess/IndexedDB/WebIDBConnectionToClient.cpp:

(WebKit::WebIDBConnectionToClient::handleGetResult):

• NetworkProcess/NetworkConnectionToWebProcess.cpp:

(WebKit::NetworkConnectionToWebProcess::writeBlobsToTemporaryFiles):
(WebKit::NetworkConnectionToWebProcess::preregisterSandboxExtensionsForOptionallyFileBackedBlob): Deleted.
(WebKit::NetworkConnectionToWebProcess::getBlobDataFileReferenceForPath): Deleted.

• NetworkProcess/NetworkConnectionToWebProcess.h:
• NetworkProcess/NetworkConnectionToWebProcess.messages.in:
• NetworkProcess/NetworkProcess.cpp:

(WebKit::NetworkProcess::accessToTemporaryFileComplete):
(WebKit::NetworkProcess::getSandboxExtensionsForBlobFiles):
(WebKit::NetworkProcess::prepareForAccessToTemporaryFile): Deleted.
(WebKit::NetworkProcess::updateTemporaryFileSandboxExtensions): Deleted.

• NetworkProcess/NetworkProcess.h:
• NetworkProcess/NetworkProcessCreationParameters.cpp:

(WebKit::NetworkProcessCreationParameters::encode const):
(WebKit::NetworkProcessCreationParameters::decode):

• NetworkProcess/NetworkProcessCreationParameters.h:
• NetworkProcess/cocoa/NetworkProcessCocoa.mm:

(WebKit::NetworkProcess::platformInitializeNetworkProcessCocoa):

• UIProcess/WebProcessPool.cpp:

(WebKit::WebProcessPool::ensureNetworkProcess):

• WebProcess/Databases/IndexedDB/WebIDBConnectionToServer.cpp:

(WebKit::WebIDBConnectionToServer::didGetRecord):
(WebKit::WebIDBConnectionToServer::didGetAllRecords):
(WebKit::WebIDBConnectionToServer::didOpenCursor):
(WebKit::WebIDBConnectionToServer::didIterateCursor):
(WebKit::preregisterSandboxExtensionsIfNecessary): Deleted.

LayoutTests:

Three IDB blob tests are passing on iOS device now.

• platform/ios-device/TestExpectations:

8:38 AM Changeset in webkit [238282] by don.olmstead@sony.com

• 16 edits
  6 moves
  2 adds in trunk

Add USE(LIBWPE) to WebCore
​https://bugs.webkit.org/show_bug.cgi?id=191401

Reviewed by Michael Catanzaro.

.:

Exposes USE_LIBWPE to the build.

• Source/cmake/OptionsWPE.cmake:

Source/WebCore:

No new tests. No change in behavior.

Migrates all PLATFORM(WPE) code that calls into wpe_* APIs to
USE(LIBWPE) instead.

Renames classes and files to use the suffix LibWPE.

• PlatformWPE.cmake:
• SourcesWPE.txt:
• platform/Pasteboard.h:
• platform/PasteboardStrategy.h:
• platform/PlatformKeyboardEvent.h:
• platform/PlatformPasteboard.h:
• platform/graphics/PlatformDisplay.cpp:

(WebCore::PlatformDisplay::createPlatformDisplay):

• platform/graphics/PlatformDisplay.h:
• platform/graphics/egl/GLContextEGL.h:
• platform/graphics/egl/GLContextEGLLibWPE.cpp: Renamed from Source/WebCore/platform/graphics/egl/GLContextEGLWPE.cpp.

(WebCore::GLContextEGL::createWPEContext):

• platform/graphics/gstreamer/MediaPlayerPrivateGStreamerBase.cpp:

(WebCore::MediaPlayerPrivateGStreamerBase::ensureGstGLContext):

• platform/graphics/libwpe/PlatformDisplayLibWPE.cpp: Renamed from Source/WebCore/platform/graphics/wpe/PlatformDisplayWPE.cpp.

(WebCore::PlatformDisplayLibWPE::create):
(WebCore::PlatformDisplayLibWPE::PlatformDisplayLibWPE):
(WebCore::PlatformDisplayLibWPE::~PlatformDisplayLibWPE):
(WebCore::PlatformDisplayLibWPE::initialize):

• platform/graphics/libwpe/PlatformDisplayLibWPE.h: Renamed from Source/WebCore/platform/graphics/wpe/PlatformDisplayWPE.h.
• platform/libwpe/PasteboardLibWPE.cpp: Renamed from Source/WebCore/platform/wpe/PasteboardWPE.cpp.
• platform/libwpe/PlatformKeyboardEventLibWPE.cpp: Renamed from Source/WebCore/platform/wpe/PlatformKeyboardEventWPE.cpp.
• platform/libwpe/PlatformPasteboardLibWPE.cpp: Renamed from Source/WebCore/platform/wpe/PlatformPasteboardWPE.cpp.

Source/WebKit:

Rename PlatformDisplayWPE to PlatformDisplayLibWPE.

• WebProcess/WebPage/wpe/AcceleratedSurfaceWPE.cpp:

(WebKit::AcceleratedSurfaceWPE::initialize):

• WebProcess/wpe/WebProcessMainWPE.cpp:

7:24 AM Changeset in webkit [238281] by Tadeu Zagallo

• 3 edits
  1 add in trunk

Fix assertion failure on BytecodeGenerator::recordOpcode
​https://bugs.webkit.org/show_bug.cgi?id=191724
<rdar://problem/45724395>

Reviewed by Saam Barati.

JSTests:

• stress/regress-187373-2.js: Added.

(foo):

Source/JavaScriptCore:

Since ​https://bugs.webkit.org/show_bug.cgi?id=187373, we were not
restoring m_lastInstruction after patching the bytecode when
finalizing StructureForInContexts, only m_lastOpcodeID, which led to
the assertion failure.

• bytecompiler/BytecodeGenerator.cpp:

(JSC::StructureForInContext::finalize):

7:07 AM Changeset in webkit [238280] by Alan Bujtas

• 3 edits
  2 adds in trunk

Source/WebCore:
[iOS] 2 subsequent taps are required to trigger certain tasks on the desktop version of YouTube.com (hover vs click).
​https://bugs.webkit.org/show_bug.cgi?id=191712
<rdar://problem/45612900>

Reviewed by Simon Fraser.

In handleSyntheticClick() we use WKContentObservation to figure out whether the tap should be treated as a hover or a click.
In general, if the mouse-move event triggers a visible content change, we assume we hit a hover-like drop down menu (or something similar)
and no need to dispatch a click event.
The idea here is that if the new content (result of the mouse-move event) does not respond to mouse click, it is most likely
only for tooltip-like reasons and it's ok to proceed with the click event.

Test: fast/events/touch/ios/click-instead-of-hover-simple.html

• rendering/updating/RenderTreeUpdater.cpp:

(WebCore::CheckForVisibilityChange::~CheckForVisibilityChange):

LayoutTests:
[iOS] 2 subsequent taps are required to trigger certain tasks on the desktop version of YouTube.com
​https://bugs.webkit.org/show_bug.cgi?id=191712
<rdar://problem/45612900>

Reviewed by Simon Fraser.

• fast/events/touch/ios/click-instead-of-hover-simple-expected.txt: Added.
• fast/events/touch/ios/click-instead-of-hover-simple.html: Added.

6:51 AM Changeset in webkit [238279] by Alan Bujtas

• 7 edits
  2 adds in trunk

[LFC][IFC] Add support for out-of-flow positioned boxes
​https://bugs.webkit.org/show_bug.cgi?id=191726

Reviewed by Antti Koivisto.

Source/WebCore:

While laying out formatting context roots (inline-block, floats) in an inline formatting context, we need to make sure
that their out-of-flow descendants get laid out as well.

Test: fast/inline/simple-inline-with-out-of-flow-descendant.html

• layout/inlineformatting/InlineFormattingContext.cpp:

(WebCore::Layout::InlineFormattingContext::layoutFormattingContextRoot const):

• layout/layouttree/LayoutTreeBuilder.cpp: This was returning the wrong context root when the container was also a context root.

(WebCore::Layout::TreeBuilder::createSubTree):

Tools:

• LayoutReloaded/misc/LFC-passing-tests.txt:

LayoutTests:

• fast/inline/simple-inline-with-out-of-flow-descendant-expected.txt: Added.
• fast/inline/simple-inline-with-out-of-flow-descendant.html: Added.
• platform/ios/TestExpectations:

6:09 AM Changeset in webkit [238278] by commit-queue@webkit.org

• 10 edits in trunk/Source

[GStreamer][WebRTC] Add API to enable/disable device mocks
​https://bugs.webkit.org/show_bug.cgi?id=191699

This basically us to test MediaStream/WebRTC support without
requiring cameras or microphones and is quite useful.

Also fix the GStreamerAudioMock by:

• Stop setting leaky-upstream on the GStreamerCapturer queue, this was usefull when we were trying to bring the MediaStream sources inside the main pipeline, it is not the case anymore (and not doable with latest version of LibWebRTC).
• Use a 'ticks' wave on the gstreamer audiotestsrc so the test stream is similar to what Apple port does.

Patch by Thibault Saunier <​tsaunier@igalia.com> on 2018-11-16
Reviewed by Xabier Rodriguez-Calvar.

The mocks are already tested and the API is really simple.

Source/WebCore:

• platform/mediastream/gstreamer/GStreamerAudioCapturer.cpp:

(WebCore::GStreamerAudioCapturer::createSource):

• platform/mediastream/gstreamer/GStreamerAudioCapturer.h:
• platform/mediastream/gstreamer/GStreamerCapturer.cpp:

(WebCore::GStreamerCapturer::addSink):

• platform/mediastream/gstreamer/GStreamerCapturer.h:

Source/WebKit:

• UIProcess/API/glib/WebKitSettings.cpp:

(webKitSettingsSetProperty):
(webKitSettingsGetProperty):
(webkit_settings_class_init):
(webkit_settings_get_enable_mock_capture_devices):
(webkit_settings_set_enable_mock_capture_devices):

• UIProcess/API/gtk/WebKitSettings.h:
• UIProcess/API/gtk/docs/webkit2gtk-4.0-sections.txt:

6:02 AM Changeset in webkit [238277] by commit-queue@webkit.org

• 3 edits in trunk/Source/WebKit

[GTK][WPE] Add "WebKitDeviceInfoPermissionRequest.h" into webkit2.h
​https://bugs.webkit.org/show_bug.cgi?id=191744

It should always have been there.

Patch by Thibault Saunier <​tsaunier@igalia.com> on 2018-11-16
Reviewed by Carlos Garcia Campos.

• UIProcess/API/gtk/webkit2.h:

6:01 AM Changeset in webkit [238276] by commit-queue@webkit.org

• 4 edits
  2 adds in trunk

[GStreamer][MediaStream] Handle track addition and removal
​https://bugs.webkit.org/show_bug.cgi?id=191599

Patch by Thibault Saunier <​tsaunier@igalia.com> on 2018-11-16
Reviewed by Xabier Rodriguez-Calvar.

Source/WebCore:

Test: fast/mediastream/MediaStream-video-element-remove-track.html

• platform/mediastream/gstreamer/GStreamerMediaStreamSource.cpp:

(WebCore::WebKitMediaStreamObserver::~WebKitMediaStreamObserver):
(WebCore::WebKitMediaStreamObserver::WebKitMediaStreamObserver):
(WebCore::webkitMediaStreamSrcFinalize):
(WebCore::webkitMediaStreamSrcChangeState):
(WebCore::webkit_media_stream_src_init):
(WebCore::webkitMediaStreamSrcSetupSrc):
(WebCore::webkitMediaStreamSrcAddTrack):
(WebCore::webkitMediaStreamSrcRemoveTrackByType):
(WebCore::webkitMediaStreamSrcSetStream):

LayoutTests:

• fast/mediastream/MediaStream-video-element-remove-track-expected.txt: Added.
• fast/mediastream/MediaStream-video-element-remove-track.html: Added.

2:57 AM Changeset in webkit [238275] by zandobersek@gmail.com

• 2 edits in trunk/Source/WebCore

ScalableImageDecoder: don't forcefully decode image data when querying frame completeness, duration
​https://bugs.webkit.org/show_bug.cgi?id=191354

Reviewed by Michael Catanzaro.

ScalableImageDecoder::frameIsCompleteAtIndex() should only check the
index validity and, if the index is valid, check for completeness of the
corresponding frame. ScalableImageDecoder::frameDurationAtIndex() should
also only retrieve duration for already-complete frames.

Both methods avoid calling ScalableImageDecoder::frameBufferAtIndex()
as that method goes on and decodes image data to determine specific
information. The ImageSource class that's querying this information
doesn't anticipate this, and doesn't handle the increased memory
consumption of the decoded data, leaving MemoryCache in the blind about
the image resource's actual amount of consumed memory. ImageSource can
instead gracefully handle any incomplete frame by marking the decoding
status for this frame as only partial.

• platform/image-decoders/ScalableImageDecoder.cpp:

(WebCore::ScalableImageDecoder::frameIsCompleteAtIndex const):
(WebCore::ScalableImageDecoder::frameHasAlphaAtIndex const):
(WebCore::ScalableImageDecoder::frameDurationAtIndex const):

1:09 AM Changeset in webkit [238274] by graouts@webkit.org

• 8 edits
  6 adds in trunk

PointerEvents should not require touch event listeners to be registered
​https://bugs.webkit.org/show_bug.cgi?id=191333
<rdar://problem/45857523>

Reviewed by Dean Jackson.

Source/WebCore:

Tests: pointerevents/ios/pointer-events-dispatch-on-touch.html

pointerevents/ios/pointer-events-prevent-default.html

• dom/EventNames.h:

(WebCore::EventNames::isTouchEventType const):
(WebCore::EventNames::touchAndPointerEventNames const):
(WebCore::EventNames::touchEventNames const): Deleted.

• dom/Node.cpp:

(WebCore::Node::moveNodeToNewDocument):

Source/WebKit:

• UIProcess/WebPageProxy.cpp:

(WebKit::WebPageProxy::updateTouchEventTracking):
(WebKit::WebPageProxy::touchEventTrackingType const):
(WebKit::WebPageProxy::handleTouchEventSynchronously):
(WebKit::WebPageProxy::handleTouchEventAsynchronously):
(WebKit::WebPageProxy::handleTouchEvent):
(WebKit::WebPageProxy::resetState):

• UIProcess/WebPageProxy.h:

LayoutTests:

Add two iOS tests that check that we dispatch pointer events when only those events are registered
and that we correctly account for calls to preventDefault(). To support pointer events test, we add
a new utility where we will be adding user interaction functions that should allow these tests to be
submitted to the WPT repository provided browsers implement their own versions of the UIController.

• TestExpectations:
• pointerevents/ios/pointer-events-dispatch-on-touch-expected.txt: Added.
• pointerevents/ios/pointer-events-dispatch-on-touch.html: Added.
• pointerevents/ios/pointer-events-prevent-default-expected.txt: Added.
• pointerevents/ios/pointer-events-prevent-default.html: Added.
• pointerevents/utils.js: Added.

Nov 15, 2018:

9:46 PM Changeset in webkit [238273] by Alan Bujtas

• 3 edits
  2 adds in trunk

[iOS] Do not get stuck in indeterminate content observation state.
​https://bugs.webkit.org/show_bug.cgi?id=191719

Reviewed by Simon Fraser.

Source/WebCore:

Reset the _WKContentChange flag to WKContentNoChange when the last observing timer is removed and we are in
the "can't decide yet if it's a hover or click" state.
This bug prevents us from firing click event when JS installs and removes the same set of timer(s) during mouse-move dispatch.

Test: fast/events/touch/ios/stuck-with-hover-state.html

• platform/ios/wak/WKContentObservation.cpp:

(WebThreadRemoveObservedContentModifier):

LayoutTests:

• fast/events/touch/ios/stuck-with-hover-state-expected.txt: Added.
• fast/events/touch/ios/stuck-with-hover-state.html: Added.

9:42 PM Changeset in webkit [238272] by mark.lam@apple.com

• 4 edits
  1 copy
  1 delete in branches/safari-606-branch

Cherry-pick r238270. rdar://problem/46085279

2018-11-15 Mark Lam <​mark.lam@apple.com>

RegExpObject's collectMatches should not be using JSArray::push to fill in its match results.
​https://bugs.webkit.org/show_bug.cgi?id=191730
<rdar://problem/46048517>

Reviewed by Saam Barati.

JSTests:

• stress/regress-187006.js: Removed.
  • this test is invalid because its sole purpose is to test for the non-spec compliant behavior that we just fixed.

• stress/regress-191730.js: Added.

Source/JavaScriptCore:

According to the spec ​https://www.ecma-international.org/ecma-262/9.0/index.html#sec-regexp.prototype-@@match,
the RegExp match results are filled in using the spec's CreateDataProperty()
function which does not consult the prototype for setters. JSArray:push()
consults the prototype for setters. We should be using putDirectIndex() instead.

• runtime/RegExpObjectInlines.h: (JSC::collectMatches):

9:39 PM Changeset in webkit [238271] by Simon Fraser

• 1 edit
  4 adds in trunk/LayoutTests

Add layout test for iframe contents disappearing on view unparent/reparent
​https://bugs.webkit.org/show_bug.cgi?id=191725

Reviewed by Alex Christensen.

This test tests the changes made in r238229.

• compositing/iframes/remove-reinsert-webview-with-iframe-expected.txt: Added.
• compositing/iframes/remove-reinsert-webview-with-iframe.html: Added.
• platform/ios/compositing/iframes/remove-reinsert-webview-with-iframe-expected.txt: Added.
• platform/mac-wk1/compositing/iframes/remove-reinsert-webview-with-iframe-expected.txt: Added.

9:32 PM Changeset in webkit [238270] by mark.lam@apple.com

• 3 edits
  1 add
  1 delete in trunk

RegExpObject's collectMatches should not be using JSArray::push to fill in its match results.
​https://bugs.webkit.org/show_bug.cgi?id=191730
<rdar://problem/46048517>

Reviewed by Saam Barati.

JSTests:

• stress/regress-187006.js: Removed.
  • this test is invalid because its sole purpose is to test for the non-spec compliant behavior that we just fixed.

• stress/regress-191730.js: Added.

Source/JavaScriptCore:

According to the spec ​https://www.ecma-international.org/ecma-262/9.0/index.html#sec-regexp.prototype-@@match,
the RegExp match results are filled in using the spec's CreateDataProperty()
function which does not consult the prototype for setters. JSArray:push()
consults the prototype for setters. We should be using putDirectIndex() instead.

• runtime/RegExpObjectInlines.h:

(JSC::collectMatches):

9:29 PM Changeset in webkit [238269] by Simon Fraser

• 3 edits
  2 adds in trunk

REGRESSION (r238090) Composited iframes that resize from zero don't show
​https://bugs.webkit.org/show_bug.cgi?id=191733
rdar://problem/46107764

Reviewed by Zalan Bujtas.
Source/WebCore:

A zero-sized iframe whose contents are composited should not trigger compositing in the
parent document (see code in requiresCompositingForFrame()), but when the <iframe> element
was resized without a style change (e.g. because it's width: 100%, height: 100% and the
parent resizes), there was no code that triggered a compositing update.

Fix by having RenderLayer::updateLayerPosition() trigger an update when the size changes,
for a RenderWidget whose contents are composited.

Test: compositing/iframes/resize-from-zero-size.html

• rendering/RenderLayer.cpp:

(WebCore::RenderLayer::updateLayerPosition):

LayoutTests:

• compositing/iframes/resize-from-zero-size-expected.txt: Added.
• compositing/iframes/resize-from-zero-size.html: Added.

9:23 PM Changeset in webkit [238268] by mark.lam@apple.com

• 5 edits
  1 copy in branches/safari-606-branch

Cherry-pick r238267. rdar://problem/46032438

2018-11-15 Mark Lam <​mark.lam@apple.com>

RegExp operations should not take fast patch if lastIndex is not numeric.
​https://bugs.webkit.org/show_bug.cgi?id=191731
<rdar://problem/46017305>

Reviewed by Saam Barati.

JSTests:

• stress/regress-191731.js: Added.

Source/JavaScriptCore:

This is because if lastIndex is an object with a valueOf() method, it can execute
arbitrary code which may have side effects, and side effects are not permitted by
the RegExp fast paths.

• builtins/RegExpPrototype.js: (globalPrivate.hasObservableSideEffectsForRegExpMatch): (overriddenName.string_appeared_here.search): (globalPrivate.hasObservableSideEffectsForRegExpSplit): (intrinsic.RegExpTestIntrinsic.test):
• builtins/StringPrototype.js: (globalPrivate.hasObservableSideEffectsForStringReplace):

9:12 PM Changeset in webkit [238267] by mark.lam@apple.com

• 4 edits
  1 add in trunk

RegExp operations should not take fast patch if lastIndex is not numeric.
​https://bugs.webkit.org/show_bug.cgi?id=191731
<rdar://problem/46017305>

Reviewed by Saam Barati.

JSTests:

• stress/regress-191731.js: Added.

Source/JavaScriptCore:

This is because if lastIndex is an object with a valueOf() method, it can execute
arbitrary code which may have side effects, and side effects are not permitted by
the RegExp fast paths.

• builtins/RegExpPrototype.js:

(globalPrivate.hasObservableSideEffectsForRegExpMatch):
(overriddenName.string_appeared_here.search):
(globalPrivate.hasObservableSideEffectsForRegExpSplit):
(intrinsic.RegExpTestIntrinsic.test):

• builtins/StringPrototype.js:

(globalPrivate.hasObservableSideEffectsForStringReplace):

9:02 PM Changeset in webkit [238266] by Simon Fraser

• 4 edits
  4 adds in trunk

Overlay with -webkit-overflow-scrolling:touch doesn't become scrollable after added text makes it taller
​https://bugs.webkit.org/show_bug.cgi?id=158342
rdar://problem/26652811

Reviewed by Zalan Bujtas.
Source/WebCore:

Patch partly by Frédéric Wang.

This commit fixes an issue when resizing the content of a -webkit-overflow-scrolling: touch
overflow node on iOS. Indeed, the RenderLayerBacking's scrolling layer may not be properly
created and hence the UIProcess receives a null UIScrollView pointer. This triggers an
assertion in debug mode and prevents the user from scrolling the overflow node in release
mode. This was partially fixed by the refactoring of bug 90342 but this commit addresses
the remaining issues by forcing a configuration update after layout in order to ensure that
RenderLayerBacking's scrolling layer is available. For an overflow element that is not yet
composited, trigger a post-layout update that is necessary to check if we need to make it
composited when it gains scrollable overflow.

Tests: fast/scrolling/ios/change-scrollability-on-content-resize-nested.html

fast/scrolling/ios/change-scrollability-on-content-resize.html

• rendering/RenderLayer.cpp:

(WebCore::RenderLayer::updateScrollInfoAfterLayout): Force a configuration update so that
RenderLayerCompositor::updateBackingAndHierarchy will later instantiate
RenderLayerBacking::m_scrollingLayer.

LayoutTests:

Patch partly by Frédéric Wang.

Add a test to check that scrollability of composited/non-composited overflow divs is properly
updated. This test used to fail and assert before #238090. A second test with nested divs
stills exhibit the issue and is fixed by the present commit.

• fast/scrolling/ios/change-scrollability-on-content-resize-expected.txt: Added.
• fast/scrolling/ios/change-scrollability-on-content-resize-nested-expected.txt: Added.
• fast/scrolling/ios/change-scrollability-on-content-resize-nested.html: Added.
• fast/scrolling/ios/change-scrollability-on-content-resize.html: Added.

7:09 PM Changeset in webkit [238265] by Brent Fulgham

• 2 edits in trunk/Source/WebCore/PAL

Additional Windows build fix after r238253.

• pal/spi/cg/CoreGraphicsSPI.h:

5:55 PM Changeset in webkit [238264] by mmaxfield@apple.com

• 12 edits
  1 delete in trunk

WKPreferencesSetFontSmoothingLevel doesn't actually do anything
​https://bugs.webkit.org/show_bug.cgi?id=191708

Reviewed by Simon Fraser.

Delete the symbol because no one calls it and it doesn't do anything

Source/WebKit:

• Scripts/PreferencesTemplates/WebPreferencesStoreDefaultsMap.cpp.erb:
• Shared/FontSmoothingLevel.h: Removed.
• Shared/WebPreferences.yaml:
• UIProcess/API/C/WKAPICast.h:

(WebKit::toFontSmoothingLevel): Deleted.

• UIProcess/API/C/WKPreferences.cpp:

(WKPreferencesSetFontSmoothingLevel): Deleted.
(WKPreferencesGetFontSmoothingLevel): Deleted.

• UIProcess/API/C/WKPreferencesRefPrivate.h:
• UIProcess/WebPreferences.h:
• WebKit.xcodeproj/project.pbxproj:
• mac/WebKit2.order:

Tools:

• TestWebKitAPI/Tests/WebKit/WKPreferences.cpp:

(TestWebKitAPI::TEST):

• WebKitTestRunner/TestController.cpp:

(WTR::TestController::resetPreferencesToConsistentValues):

5:37 PM Changeset in webkit [238263] by Fujii Hironori

• 2 edits in trunk/Source/WebCore

[curl] warning: delete called on non-final 'WebCore::CurlDownload' that has virtual functions but non-virtual destructor [-Wdelete-non-virtual-dtor]
​https://bugs.webkit.org/show_bug.cgi?id=191582

Reviewed by Alex Christensen.

No new tests because there's no behaviour change.

• platform/network/curl/CurlDownload.h: Marked CurlDownload final.

5:07 PM Changeset in webkit [238262] by Truitt Savell

• 43 edits
  5 deletes in trunk

Unreviewed, rolling out r238244.

Caused High Sierra test runs to fail early with 50 crashes and
casued 25 API failures.

Reverted changeset:

"[css-logical] Implement flow-relative margin, padding and
border shorthands"
​https://bugs.webkit.org/show_bug.cgi?id=188697
​https://trac.webkit.org/changeset/238244

4:48 PM Changeset in webkit [238261] by Ryan Haddad

• 2 edits in trunk/LayoutTests

imported/w3c/web-platform-tests/encoding tests take > 30 minutes to run on debug bots
​https://bugs.webkit.org/show_bug.cgi?id=191720

Unreviewed test gardening.

• TestExpectations: Skip the directory on debug.

4:35 PM Changeset in webkit [238260] by Alan Coon

• 22 edits
  1 move
  6 deletes in tags/Safari-607.1.14

Revert r238090. rdar://problem/46111475

4:35 PM Changeset in webkit [238259] by Alan Coon

• 3 edits
  2 deletes in tags/Safari-607.1.14

Revert r238170. rdar://problem/46111475

4:34 PM Changeset in webkit [238258] by Alan Coon

• 2 edits in tags/Safari-607.1.14/Source/WebCore

Revert r238229. rdar://problem/46111475

3:51 PM Changeset in webkit [238257] by jer.noble@apple.com

• 2 edits in trunk/Source/WebCore

AVKit will set videoGravity to a nil string when building against iosmac
​https://bugs.webkit.org/show_bug.cgi?id=191573

Reviewed by Dean Jackson.

Workaround AVKit behavior by treating nil videoGravity as the default,
which is AVLayerVideoGravityResizeAspect.

• platform/ios/VideoFullscreenInterfaceAVKit.mm:

(-[WebAVPlayerLayer setVideoGravity:]):

3:18 PM Changeset in webkit [238256] by Brent Fulgham

• 2 edits in trunk/Source/WebCore/PAL

Unreviewed build fix after removing some WKSI calls on Windows.

• pal/spi/cg/CoreGraphicsSPI.h:

3:15 PM Changeset in webkit [238255] by Chris Dumez

• 3 edits in trunk/LayoutTests

REGRESSION (r238038): [iOS] Layout test fast/forms/ios/validation-bubble-dismiss-on-tap.html is a flaky failure
​https://bugs.webkit.org/show_bug.cgi?id=191710
<rdar://problem/46104093>

Reviewed by Simon Fraser.

There is a small amount of time after we ask to show the popover before the popover is actually shown on screen.
During this period, clicks are eaten and not reported to the view. For this reason, we keep tapping until the
popup is dismissed.

• fast/forms/ios/validation-bubble-dismiss-on-tap-expected.txt:
• fast/forms/ios/validation-bubble-dismiss-on-tap.html:

2:42 PM Changeset in webkit [238254] by commit-queue@webkit.org

• 5 edits in trunk/Source/WebKit

Web Inspector: Remove unused WebInspectorProxy code
​https://bugs.webkit.org/show_bug.cgi?id=191674

Patch by Joseph Pecoraro <Joseph Pecoraro> on 2018-11-15
Reviewed by Dean Jackson.

• UIProcess/WebInspectorProxy.cpp:

(WebKit::WebInspectorProxy::createInspectorPage):
(WebKit::WebInspectorProxy::closeFrontendPageAndWindow):

• UIProcess/WebInspectorProxy.h:
• UIProcess/mac/WebInspectorProxyMac.mm:

(WebKit::WebInspectorProxy::setInspectorWindowFrame): Deleted.
(WebKit::WebInspectorProxy::inspectorWindowFrame): Deleted.

• WebProcess/WebPage/WebInspector.h:

2:38 PM Changeset in webkit [238253] by Brent Fulgham

• 10 edits in trunk/Source

[Win] Reduce the use of WKSI library calls: Font Handling
​https://bugs.webkit.org/show_bug.cgi?id=191701
<rdar://problem/46104809>

Reviewed by Myles C. Maxfield.

Source/WebCore:

Move the old Windows font handling code out of WKSI to our regular
repository. We now handle SPI differently, and don't need to keep
these implementations in a separate library. This should also help
avoid the somewhat frequent build failures caused when WKSI is not
updated in sync with WebKit.

Tested by existing Windows test cases.

• platform/graphics/FontCascade.h:
• platform/graphics/win/FontCGWin.cpp:

(WebCore::FontCascade::drawGlyphs):
(WebCore::FontCascade::setFontSmoothingLevel):
(WebCore::setCGFontSmoothingStyle):
(WebCore::FontCascade::setFontSmoothingStyle):
(WebCore::FontCascade::setFontSmoothingContrast):
(WebCore::clearTypeContrast):
(WebCore::FontCascade::systemFontSmoothingChanged):
(WebCore::FontCascade::setCGContextFontRenderingStyle):
(WebCore::renderingStyleForFont):
(WebCore::FontCascade::getGlyphAdvances):

• platform/graphics/win/GlyphPageTreeNodeCGWin.cpp:

(WebCore::GlyphPage::fill):

• platform/graphics/win/GraphicsContextCGWin.cpp:

(WebCore::GraphicsContext::drawFocusRing):
(WebCore::GraphicsContext::drawDotsForDocumentMarker):

• platform/graphics/win/SimpleFontDataCGWin.cpp:

(WebCore::Font::platformWidthForGlyph const):

• rendering/RenderMediaControls.cpp:

(WebCore::RenderMediaControls::adjustMediaSliderThumbSize):

Source/WebKitLegacy/win:

Update to use the old font handling routines moved from WKSI to FontCascade.

• WebPreferences.cpp:

(WebPreferences::setFontSmoothing):
(WebPreferences::setFontSmoothingContrast):

• WebView.cpp:

(systemParameterChanged):

2:07 PM Changeset in webkit [238252] by rniwa@webkit.org

• 3 edits in trunk/Source/WebKit

Enable process swap on cross-site navigation by default on macOS
​https://bugs.webkit.org/show_bug.cgi?id=191572

Reviewed by Chris Dumez.

Enabled the feature by default on macOS.

• Shared/WebPreferences.yaml:
• Shared/WebPreferencesDefaultValues.h:

2:05 PM Changeset in webkit [238251] by aakash_jain@apple.com

• 2 edits in trunk/Tools

Increase EWS lock timeout to 3 hours
​https://bugs.webkit.org/show_bug.cgi?id=191705

Reviewed by Alexey Proskuryakov.

• QueueStatusServer/config/queues.py: Increased EWS lock timeout to 3 hours.

2:01 PM Changeset in webkit [238250] by Chris Dumez

• 5 edits in trunk

IndexedDB.IndexedDBTempFileSize API test times out with process prewarming enabled
​https://bugs.webkit.org/show_bug.cgi?id=191671
<rdar://problem/46086062>

Reviewed by Alex Christensen.

Source/WebKit:

Enabling process prewarming caused IndexedDB.IndexedDBTempFileSize API to time out and print
the following line:
"Attempted to create a NetworkLoad with a session (id=2) that does not exist."

This actually identified a pre-existing bug with our handling of non-default data store.
Whenever a page starts using a data store, we call WebProcessPool::pageBeginUsingWebsiteDataStore()
which will call NetworkProcessProxy::addSession() if the network process was already started
to let the network process know about this non-default session. There are several issues with the
current model:

1. If the network process was not created yet when pageBeginUsingWebsiteDataStore() is called, then the network process will not know about the non-default session when it actually gets started later on. This is unlikely to happen in practice, except in case of network process crash because we create the network process as soon as we initialize the first WebProcessProxy.
2. Even if we successfuly managed to register the session with the network process proxy, we get in trouble if the network process crashes or is terminated later on as we do not re-register those sessions with the new network process.

To address these 2 issues, WebProcessPool::ensureNetworkProcess() now takes care of registering
all the non-default sessions (that are associated with this process pool) with the new network
process. The WebProcessPool knows about these sessions because it adds them to m_sessionToPagesMap
whenever WebProcessPool::pageBeginUsingWebsiteDataStore() is called, even if the network process
proxy was not created yet.

The reason the IndexedDB.IndexedDBTempFileSize API test was failing was because it did:

1. A load in a WebView V1 with a non-default session
2. Process prewarming kicked in after this load and would create a new WebProcessProxy.
3. Terminate the network process
4. Another load in a WebView V2 with the same non-defaut session, which would reuse the prewarmed process. Because the network process was terminated, constructing the new page would not register the session ID with the new network process when pageBeginUsingWebsiteDataStore() is called.

-> The load would hang because the new network process would not know about the

non-default session when started later on.

The bug was previously hidden without process prewarming because step 4 would create a *new*
WebProcessProxy and WebProcessPool::initializeNewWebProcess() would call ensureNetworkProcess()
so that pageBeginUsingWebsiteDataStore() would successfuly register the session with the
network process later on.

I wrote a second API test (WebKit.DoLoadWithNonDefaultDataStoreAfterTerminatingNetworkProcess)
to demonstrate the pre-existing bug without process prewarming enabled.

• UIProcess/WebProcessPool.cpp:

(WebKit::WebProcessPool::ensureNetworkProcess):

Tools:

Add API test coverage.

• TestWebKitAPI/Tests/WebKitCocoa/ProcessSwapOnNavigation.mm:
• TestWebKitAPI/Tests/WebKitCocoa/WebsiteDataStoreCustomPaths.mm:

(TEST):

1:58 PM Changeset in webkit [238249] by commit-queue@webkit.org

• 2 edits in trunk/Source/WebKit

Create feature flag (HTTPS_UPGRADE)
​https://bugs.webkit.org/show_bug.cgi?id=191613
<rdar://problem/45851068>

Patch by Vivek Seth <​v_seth@apple.com> on 2018-11-15
Reviewed by Dean Jackson.

• config.h:

1:53 PM Changeset in webkit [238248] by Chris Dumez

• 5 edits in trunk/Source/WebKit

Regression(r238215) WKProcessPool.WarmInitialProcess API test is failing
​https://bugs.webkit.org/show_bug.cgi?id=191693

Reviewed by Alex Christensen.

Allow *explicit* process prewarming to construct the default data store if necessary or we would
not be obeying the client's request. Only automatic process prewarming should avoid constructing
the default data store unnecessarily to avoid causing memory regressions.

• UIProcess/API/C/WKContext.cpp:

(WKContextWarmInitialProcess):

• UIProcess/API/Cocoa/WKProcessPool.mm:

(-[WKProcessPool _warmInitialProcess]):

• UIProcess/WebProcessPool.cpp:

(WebKit::WebProcessPool::prewarmProcess):
(WebKit::WebProcessPool::didReachGoodTimeToPrewarm):

• UIProcess/WebProcessPool.h:

(WebKit::WebProcessPool::sendToOneProcess):

1:31 PM Changeset in webkit [238247] by jiewen_tan@apple.com

• 3 edits in trunk/Source/WebKit

Unreviewed, comment improvement over r238166.

• UIProcess/API/C/WKWebsiteDataStoreRef.cpp:

(WKWebsiteDataStoreSetWebAuthenticationMockConfiguration):

• UIProcess/WebAuthentication/fido/CtapHidAuthenticator.h:

1:24 PM Changeset in webkit [238246] by jiewen_tan@apple.com

• 15 edits in trunk

[WebAuthN] Use a real nonce for CTAPHID_INIT
​https://bugs.webkit.org/show_bug.cgi?id=191533
<rdar://problem/46103502>

Reviewed by Brent Fulgham.

Source/WebCore:

New tests are added into existing test files.

• Modules/webauthn/fido/FidoConstants.h:

Source/WebKit:

Use a real nonce for CTAPHID_INIT request according to:
​https://fidoalliance.org/specs/fido-v2.0-ps-20170927/fido-client-to-authenticator-protocol-v2.0-ps-20170927.html#ctaphid_init-0x06.
The challenge here is the new transaction needs to start in the next runloop otherwise a dead lock will form:
wrong nonce -> new transaction -> new nonce -> write init request -> read init response from last run as it
piped in the run loop -> wrong nonce of course -> ...
To break the above dead lock, we have to start the new transaction in the next run. However, that isn't
sufficient as the arrived init response will be piped in HidConnection::m_inputReports, which is designed
on purpose to store any data packets within (initialized, terminated) time interval to prevent data loss in
the case when HidConnection::registerDataReceivedCallback is called after the first data packet's arrival.
In order to break the dead lock completely, HidConnection::invalidateCache will bnnne called prior to every
send to delete any potential init response from last run. HidConnection::invalidateCache is not necessary
for other protocols though. The above scenario is more or less a design flaw in CTAP HID.

Of course, all above scenarios are covered in our mock tests.

• UIProcess/API/C/WKWebsiteDataStoreRef.cpp:

(WKWebsiteDataStoreSetWebAuthenticationMockConfiguration):

• UIProcess/WebAuthentication/Cocoa/HidConnection.h:

(WebKit::HidConnection::invalidateCache):

• UIProcess/WebAuthentication/Mock/MockHidConnection.cpp:

(WebKit::MockHidConnection::send):
(WebKit::MockHidConnection::parseRequest):
(WebKit::MockHidConnection::feedReports):

• UIProcess/WebAuthentication/Mock/MockHidConnection.h:
• UIProcess/WebAuthentication/Mock/MockWebAuthenticationConfiguration.h:
• UIProcess/WebAuthentication/fido/CtapHidDriver.cpp:

(WebKit::CtapHidDriver::Worker::transact):
(WebKit::CtapHidDriver::CtapHidDriver):
(WebKit::CtapHidDriver::transact):
(WebKit::CtapHidDriver::continueAfterChannelAllocated):
(WebKit::CtapHidDriver::returnResponse):

• UIProcess/WebAuthentication/fido/CtapHidDriver.h:

LayoutTests:

• http/wpt/webauthn/ctap-hid-failure.https-expected.txt:
• http/wpt/webauthn/ctap-hid-failure.https.html:
• http/wpt/webauthn/ctap-hid-success.https-expected.txt:
• http/wpt/webauthn/ctap-hid-success.https.html:

1:20 PM Changeset in webkit [238245] by Justin Fan

• 18 edits
  7 copies
  2 moves
  2 adds in trunk

[WebGPU] WebGPUCommandBuffer prototype
​https://bugs.webkit.org/show_bug.cgi?id=191663

Reviewed by Dean Jackson.

Begin implementation of WebGPUCommandBuffers as well as GPUQueues (MTLCommandBuffer, MTLCommandQueue).

Source/WebCore:

Test: webgpu/command-buffers.html

• CMakeLists.txt:
• DerivedSources.make:
• Modules/webgpu/GPUCommandBuffer.h: Added. Wrapper class around a MTLCommandBuffer.
• Modules/webgpu/GPUDevice.cpp:

(WebCore::GPUDevice::createCommandBuffer): Added.
(WebCore::GPUDevice::getQueue): Returns RefPtr to the device's singleton queue.

• Modules/webgpu/GPUDevice.h: Now manages the device's GPUQueue.

(WebCore::GPUDevice::platformDevice const):

• Modules/webgpu/GPUQueue.h: Added. Wrapper class around a MTLCommandQueue.

(WebCore::GPUQueue::platformQueue const):

• Modules/webgpu/GPURenderPipeline.h: Moved from Source/WebCore/Modules/webgpu/cocoa/GPURenderPipeline.h.

(WebCore::GPURenderPipeline::platformRenderPipeline const):

• Modules/webgpu/GPUShaderModule.h:

(WebCore::GPUShaderModule::platformShaderModule const):

• Modules/webgpu/GPUSwapChain.h: Moved from Source/WebCore/Modules/webgpu/cocoa/GPUSwapChain.h.

(WebCore::GPUSwapChain::platformLayer const):

• Modules/webgpu/WebGPUCommandBuffer.cpp: Added. Web interface for a GPU device's command buffer.

(WebCore::WebGPUCommandBuffer::create):
(WebCore::WebGPUCommandBuffer::WebGPUCommandBuffer):

• Modules/webgpu/WebGPUCommandBuffer.h: Added.
• Modules/webgpu/WebGPUCommandBuffer.idl: Added.
• Modules/webgpu/WebGPUDevice.cpp:

(WebCore::WebGPUDevice::createCommandBuffer const): Added.

• Modules/webgpu/WebGPUDevice.h:
• Modules/webgpu/WebGPUDevice.idl:
• Modules/webgpu/cocoa/GPUCommandBufferMetal.mm: Added. MTLCommandBuffer impl for GPUCommandBuffer.

(WebCore::GPUCommandBuffer::create): Create a MTLCommandBuffer from the MTLCommandQueue.
(WebCore::GPUCommandBuffer::GPUCommandBuffer):

• Modules/webgpu/cocoa/GPUDeviceMetal.mm:

(WebCore::GPUDevice::GPUDevice):

• Modules/webgpu/cocoa/GPUQueueMetal.mm: Added. MTLCommandQueue impl for GPUQueue.

(WebCore::GPUQueue::create):
(WebCore::GPUQueue::GPUQueue):

• Modules/webgpu/cocoa/GPURenderPipelineMetal.mm:

(WebCore::GPURenderPipeline::GPURenderPipeline):

• Modules/webgpu/cocoa/GPUShaderModuleMetal.mm:

(WebCore::GPUShaderModule::create):
(WebCore::GPUShaderModule::GPUShaderModule):

• Sources.txt:
• SourcesCocoa.txt:
• WebCore.xcodeproj/project.pbxproj:
• bindings/js/WebCoreBuiltinNames.h:

LayoutTests:

New testing for WebGPUCommandBuffers, to be fleshed out when WebGPUCommandBufferDescriptors
are added to the specification.

• webgpu/command-buffers-expected.txt: Added.
• webgpu/command-buffers.html: Basic functionality, with more stress tests to be added when the spec is updated.
• webgpu/webgpu-basics.html: Updated to create the command buffer used by later rendering operations.

1:15 PM Changeset in webkit [238244] by commit-queue@webkit.org

• 43 edits
  17 adds in trunk

[css-logical] Implement flow-relative margin, padding and border shorthands
​https://bugs.webkit.org/show_bug.cgi?id=188697

Patch by Oriol Brufau <Oriol Brufau> on 2018-11-15
Reviewed by Simon Fraser and Antti Koivisto.

LayoutTests/imported/w3c:

Import tests from the WPT test suite for CSS Logical Properties and Values,
and enable the CSSLogicalEnabled runtime flag for them.

They still have some failures because sideways writing modes have not been
implemented yet (​https://bugs.webkit.org/show_bug.cgi?id=166941).

• web-platform-tests/css/css-logical/logical-box-border-color-expected.txt:
• web-platform-tests/css/css-logical/logical-box-border-color.html:
• web-platform-tests/css/css-logical/logical-box-border-shorthands-expected.txt:
• web-platform-tests/css/css-logical/logical-box-border-shorthands.html:
• web-platform-tests/css/css-logical/logical-box-border-style-expected.txt:
• web-platform-tests/css/css-logical/logical-box-border-style.html:
• web-platform-tests/css/css-logical/logical-box-border-width-expected.txt:
• web-platform-tests/css/css-logical/logical-box-border-width.html:
• web-platform-tests/css/css-logical/logical-box-inset-expected.txt:
• web-platform-tests/css/css-logical/logical-box-inset.html:
• web-platform-tests/css/css-logical/logical-box-margin-expected.txt:
• web-platform-tests/css/css-logical/logical-box-margin.html:
• web-platform-tests/css/css-logical/logical-box-padding-expected.txt:
• web-platform-tests/css/css-logical/logical-box-padding.html:
• web-platform-tests/css/css-logical/logical-box-size.html:
• web-platform-tests/css/css-logical/resources/test-box-properties.js:

(export.createBoxPropertyGroup):
(export.createSizingPropertyGroup):
(export.runTests.set test):
(export.runTests):

• web-platform-tests/css/css-logical/resources/test-shared.js: Added.

(export.testCSSValues):
(export.testComputedValues):
(export.makeDeclaration):

• web-platform-tests/css/css-logical/resources/w3c-import.log:

Source/WebCore:

Tests: imported/w3c/web-platform-tests/css/css-logical/logical-box-border-color.html

imported/w3c/web-platform-tests/css/css-logical/logical-box-border-shorthands.html
imported/w3c/web-platform-tests/css/css-logical/logical-box-border-style.html
imported/w3c/web-platform-tests/css/css-logical/logical-box-border-width.html
imported/w3c/web-platform-tests/css/css-logical/logical-box-margin.html
imported/w3c/web-platform-tests/css/css-logical/logical-box-padding.html
webexposed/css-properties-as-js-properties.html
webexposed/css-properties-behind-flags.html
webexposed/css-property-listing.html

• css/CSSComputedStyleDeclaration.cpp:

(WebCore::ComputedStyleExtractor::valueForPropertyinStyle):
Allow the new properties to serialize their computed value.

(WebCore::ComputedStyleExtractor::getCSSPropertyValuesFor2SidesShorthand):
(WebCore::ComputedStyleExtractor::getCSSPropertyValuesFor4SidesShorthand):

• css/CSSComputedStyleDeclaration.h:

Rename getCSSPropertyValuesForSidesShorthand to getCSSPropertyValuesFor4SidesShorthand,
and add analogous getCSSPropertyValuesFor2SidesShorthand for serializing 2-sided
shorthands.

• css/CSSProperties.json:

Add the new properties behind the CSSLogicalEnabled runtime flag.

• css/CSSStyleDeclaration.cpp:

(WebCore::CSSStyleDeclaration::supportedPropertyNames const):
Prevent CSS properties disabled behind a runtime flag from being exposed in
style declarations.

• css/StyleProperties.cpp:

(WebCore::StyleProperties::getPropertyValue const):
Allow the new properties to serialize their specified value.

(WebCore::StyleProperties::get2Values const):
Add get2Values, analogous to get4Values, for serializing 2-sided shorthands.

(WebCore::StyleProperties::borderPropertyValue const):
Allow borderPropertyValue to serialize arbitrary multi-sided border shorthands
corresponding to width, style and color.

(WebCore::MutableStyleProperties::setProperty):
Prevent CSS properties disabled behind a runtime flag from being set a value.

(WebCore::StyleProperties::asText const):
Allow the new properties to be serialized in cssText.
Prevent CSS shorthands disabled behind a runtime flag from appearing in cssText,
and serialize the longhands instead. Note that there could be another shorthand
available which is enabled, but a proper solution would require bug 190496.

• css/StyleProperties.h:

Update declarations of borderPropertyValue and get2Values.

• css/makeprop.pl:

(addProperty):
Add isEnabledCSSProperty function for checking that a CSS property is not
disabled behind a runtime flag.

• css/parser/CSSPropertyParser.cpp:

(WebCore::cssPropertyID):
Prevent CSS properties disabled behind a runtime flag from being exposed in
computed styles.

(WebCore::CSSPropertyParser::addProperty):
Prevent CSS properties disabled behind a runtime flag from being set a value.

(WebCore::CSSPropertyParser::consumeBorder):
Change consumeBorder to provide the caller with the parsed values instead of
setting properties. Then the caller can decide to which properties the values
should be set, and whether border-image should be reset or not.

(WebCore::CSSPropertyParser::consume2ValueShorthand):
(WebCore::CSSPropertyParser::consume4ValueShorthand):
Rename consume4Values to consume4ValueShorthand, and add analogous
consume2ValueShorthand for parsing shorthands with two longhands.

(WebCore::CSSPropertyParser::parseShorthand):
Allow the new properties to be parsed.

• css/parser/CSSPropertyParser.h:

Update declarations of consumeBorder, consume2ValueShorthand and
consume4ValueShorthand.

• inspector/agents/InspectorCSSAgent.cpp:

(WebCore::InspectorCSSAgent::getSupportedCSSProperties):
Prevent CSS properties disabled behind a runtime flag from being exposed in
the CSS inspector tool.

• page/RuntimeEnabledFeatures.h:

(WebCore::RuntimeEnabledFeatures::setCSSLogicalEnabled):
(WebCore::RuntimeEnabledFeatures::cssLogicalEnabled const):
Add the CSSLogicalEnabled runtime flag.

Source/WebKit:

Add a CSSLogicalEnabled runtime flag.

• Shared/WebPreferences.yaml:

Source/WebKitLegacy/mac:

Add CSSLogicalEnabled runtime flag.

• WebView/WebPreferenceKeysPrivate.h:
• WebView/WebPreferences.mm:

(+[WebPreferences initialize]):
(-[WebPreferences cssLogicalEnabled]):
(-[WebPreferences setCSSLogicalEnabled:]):

• WebView/WebPreferencesPrivate.h:
• WebView/WebView.mm:

(-[WebView _preferencesChanged:]):

Tools:

• DumpRenderTree/TestOptions.cpp:

(TestOptions::TestOptions):

• DumpRenderTree/TestOptions.h:
• DumpRenderTree/mac/DumpRenderTree.mm:

(setWebPreferencesForTestOptions):
Allow tests to enable the CSSLogicalEnabled flag in WK1.

• Scripts/webkitpy/style/checkers/jsonchecker.py:

(JSONCSSPropertiesChecker.check_codegen_properties):
Allow CSS property definitions to have a 'runtime-flag' parameter which
disables the property when the specified runtime flag is disabled.

LayoutTests:

Add tests checking that CSS properties disabled behind runtime flags
are not exposed.

• platform/ios/webexposed/css-properties-as-js-properties-expected.txt: Added.
• platform/ios/webexposed/css-property-listing-expected.txt: Added.
• platform/mac/webexposed/css-properties-as-js-properties-expected.txt: Added.
• platform/mac/webexposed/css-property-listing-expected.txt: Added.
• platform/win/webexposed/css-properties-as-js-properties-expected.txt: Added.
• platform/win/webexposed/css-property-listing-expected.txt: Added.
• webexposed/css-properties-as-js-properties-expected.txt: Added.
• webexposed/css-properties-as-js-properties.html: Added.
• webexposed/css-properties-behind-flags-expected.txt: Added.
• webexposed/css-properties-behind-flags.html: Added.
• webexposed/css-property-listing-expected.txt: Added.
• webexposed/css-property-listing.html: Added.

1:03 PM Changeset in webkit [238243] by jiewen_tan@apple.com

• 2 edits in trunk/LayoutTests

Unreviewed, test gardening.

• platform/mac-wk2/TestExpectations:

12:40 PM Changeset in webkit [238242] by Truitt Savell

• 4 edits
  6 deletes in trunk

Unreviewed, rolling out r238220.

Introduced failing tests to iOS and is slowing down EWS

Reverted changeset:

"[css-grid] Consider scrollbars in
populateGridPositionsForDirection()"
​https://bugs.webkit.org/show_bug.cgi?id=191656
​https://trac.webkit.org/changeset/238220

12:32 PM Changeset in webkit [238241] by timothy@apple.com

• 2 edits in trunk/Tools

REGRESSION (r238212): WebKit.BackgroundColor* API test failures.
​https://bugs.webkit.org/show_bug.cgi?id=191697
rdar://problem/46098570

Reviewed by Dean Jackson.

• TestWebKitAPI/Tests/mac/BackgroundColor.mm:

(TestWebKitAPI::TEST): Updated for new default background color.

12:28 PM Changeset in webkit [238240] by jiewen_tan@apple.com

• 4 edits in trunk/Source

[WebAuthN] PublicKeyCredentialCreationOptions::AuthenticatorSelectionCriteria::AuthenticatorAttachment should be optional
​https://bugs.webkit.org/show_bug.cgi?id=191522

Reviewed by Brent Fulgham.

Source/WebCore:

Accordign to the WebIDL, AuthenticatorSelectionCriteria::AuthenticatorAttachment should be optional.
​https://www.w3.org/TR/webauthn/#dictdef-authenticatorselectioncriteria

Covered by existing tests.

• Modules/webauthn/PublicKeyCredentialCreationOptions.h:

(WebCore::PublicKeyCredentialCreationOptions::AuthenticatorSelectionCriteria::decode):

Source/WebKit:

• UIProcess/WebAuthentication/AuthenticatorManager.cpp:

(WebKit::AuthenticatorManagerInternal::collectTransports):

12:23 PM Changeset in webkit [238239] by Ross Kirsling

• 3 edits
  1 delete in trunk/Source/WebCore

DragImage should have a complete default implementation
​https://bugs.webkit.org/show_bug.cgi?id=191666

Reviewed by Dean Jackson.

Move WPE's stub implementation down into the base implementation file.

• SourcesWPE.txt:
• platform/DragImage.cpp:

(WebCore::dragImageSize):
(WebCore::deleteDragImage):
(WebCore::scaleDragImage):
(WebCore::dissolveDragImageToFraction):
(WebCore::createDragImageFromImage):
(WebCore::createDragImageIconForCachedImageFilename):
(WebCore::createDragImageForLink):

• platform/wpe/DragImageWPE.cpp: Removed.

12:01 PM Changeset in webkit [238238] by jiewen_tan@apple.com

• 8 edits in trunk

[WebAuthN] UserHandle can be null
​https://bugs.webkit.org/show_bug.cgi?id=191521

Reviewed by Alex Christensen.

Source/WebCore:

According to the newest spec as of 7 August, 2018: ​https://www.w3.org/TR/webauthn/#conforming-authenticators-u2f.
UserHandle can now be null.

Covered by existing tests.

• Modules/webauthn/AuthenticatorAssertionResponse.h:

(WebCore::AuthenticatorAssertionResponse::create):
(WebCore::AuthenticatorAssertionResponse::userHandle const):
(WebCore::AuthenticatorAssertionResponse::AuthenticatorAssertionResponse):

• Modules/webauthn/AuthenticatorAssertionResponse.idl:
• Modules/webauthn/PublicKeyCredential.cpp:

(WebCore::PublicKeyCredential::tryCreate):

• Modules/webauthn/PublicKeyCredentialData.h:

(WebCore::PublicKeyCredentialData::encode const):
(WebCore::PublicKeyCredentialData::decode):

• Modules/webauthn/fido/DeviceResponseConverter.cpp:

(fido::readCTAPGetAssertionResponse):

LayoutTests:

• http/wpt/webauthn/public-key-credential-get-success-hid.https.html:

11:48 AM Changeset in webkit [238237] by Jonathan Bedard

• 2 edits in trunk/Tools

webkitpy: Check for specific process instead of using data migrator (Follow-up fix)
​https://bugs.webkit.org/show_bug.cgi?id=191551
<rdar://problem/45993156>

Reviewed by Aakash Jain.

• Scripts/webkitpy/xcode/simulated_device.py:

(SimulatedDevice.is_usable): Use software_variant instead of software_version for watchOS.

11:43 AM Changeset in webkit [238236] by Michael Catanzaro

• 3 edits in trunk/Source/WebKit

[SOUP] Make Accept-Languages header look more like other browsers
​https://bugs.webkit.org/show_bug.cgi?id=191614

Reviewed by Carlos Garcia Campos.

This changes our Accept-Language header to better match other browsers, from something
like this:

Accept-Language: en-us, en;q=0.90, es-es;q=0.80, es;q=0.70

to something like this:

Accept-Language: en-US,en;q=0.90,es-ES;q=0.80,es;q=0.70

There's no particular motivation for this change. I just noticed we were different for no
clear reason.

• NetworkProcess/soup/NetworkProcessSoup.cpp:

(WebKit::buildAcceptLanguages):

• UIProcess/API/glib/WebKitWebContext.cpp:

(webkit_web_context_set_preferred_languages):

11:15 AM Changeset in webkit [238235] by dbates@webkit.org

• 16 edits
  4 adds in trunk

[iOS] Shift + Tab does not focus previous field
​https://bugs.webkit.org/show_bug.cgi?id=191596
<rdar://problem/45892053>

Reviewed by Wenson Hsieh.

Source/WebKit:

Wire up the the tab and shift + tab key commands to the WKWebView/WKContentView's
action forwarding mechanism. Also rename -_prevAccessoryTab to -_previousAccessoryTab.

• Platform/spi/ios/UIKitSPI.h: Add more SPI.
• UIProcess/API/Cocoa/WKWebViewInternal.h:
• UIProcess/ios/WKContentViewInteraction.h:
• UIProcess/ios/WKContentViewInteraction.mm:

(-[WKContentView canPerformActionForWebView:withSender:]):
(-[WKContentView keyCommands]):
(-[WKContentView _nextAccessoryTabForWebView:]): Added.
(-[WKContentView _previousAccessoryTabForWebView:]): Added.
(-[WKContentView _nextAccessoryTab:]): Deleted.
(-[WKContentView _prevAccessoryTab:]): Deleted.

Tools:

Add infrastructure to support testing a key down event with modifiers by creating
and dispatching a UIEvent. This infrastructure replaces the previous mechanism in
Tools/WebKitTestRunner/ios/HIDEventGenerator.mm to generate a IOHIDEvent for a
keydown as it did not support creating an event with modifier key state that would
be recognized by UIKit.

• DumpRenderTree/ios/UIScriptControllerIOS.mm:

(WTR::UIScriptController::keyDown):
(WTR::UIScriptController::keyDownUsingHardwareKeyboard): Deleted.
(WTR::UIScriptController::keyUpUsingHardwareKeyboard): Deleted.

• TestRunnerShared/UIScriptContext/Bindings/UIScriptController.idl:

Add function uiController.keyDown() that takes a character that represents a keyboard key
and an array of modifier keys. The behavior of this function is analogous to eventSender.keyDown().
Remove functions uiController.keyDownUsingHardwareKeyboard() and uiController.keyUpUsingHardwareKeyboard()
as the former is replaced by uiController.keyDown() and the latter was never used.

• TestRunnerShared/UIScriptContext/UIScriptController.cpp:

(WTR::UIScriptController::keyDown): Added.
(WTR::UIScriptController::keyUpUsingHardwareKeyboard): Deleted.
(WTR::UIScriptController::keyDownUsingHardwareKeyboard): Deleted.

• TestRunnerShared/UIScriptContext/UIScriptController.h:
• WebKitTestRunner/ios/HIDEventGenerator.h:
• WebKitTestRunner/ios/HIDEventGenerator.mm:

(createHIDKeyDownEvent): Added.
(-[HIDEventGenerator keyDown:completionBlock:]): Deleted.
(-[HIDEventGenerator keyUp:completionBlock:]): Deleted.

• WebKitTestRunner/ios/UIScriptControllerIOS.mm:

(WTR::arrayLength): Added.
(WTR::parseModifier): Added.
(WTR::parseModifierArray): Added.
(WTR::UIScriptController::keyDown):
(WTR::UIScriptController::keyDownUsingHardwareKeyboard): Deleted.
(WTR::UIScriptController::keyUpUsingHardwareKeyboard): Deleted.

LayoutTests:

Add tests to ensure that we do not regress the tab and shift + tab key commands.

• fast/events/ios/focus-tab-next-field-expected.txt: Added.
• fast/events/ios/focus-tab-next-field.html: Added.
• fast/events/ios/focus-tab-previous-field-expected.txt: Added.
• fast/events/ios/focus-tab-previous-field.html: Added.
• platform/ios-wk1/TestExpectations: Skip test focus-tab-previous-field.html in Legacy WebKit as

we do not support testing key commands with modifier keys.

• resources/ui-helper.js:

(window.UIHelper.keyDown): Updated to support taking an optional array of modifiers (defaults: [])
and use the new infrastructure to generate a key down event for the specified character.

10:26 AM Changeset in webkit [238234] by Alan Coon

• 2 edits in tags/Safari-607.1.14/Source/WebCore

Cherry-pick r238229. rdar://problem/46083440

REGRESSION(r238090): Composited iframe contents disappear after switching tabs in Safari
​https://bugs.webkit.org/show_bug.cgi?id=191673
rdar://problem/46083440

Reviewed by Antti Koivisto.

Switching tabs in Safari triggers the "setIsInWindow" code path, that detaches the layer
tree for every Frame. They get re-attached on tab show, and for subframes this involves
the triggering of a fake style recalc in the parent document via scheduleInvalidateStyleAndLayerComposition().

The style diff that's sent to RenderLayerCompositor::layerStyleChanged() as a result of that
fake style recalc is RecompositeLayer, but the code was failing to trigger the necessary
layer configuration update that gets iframe layers parented.

This stop-gap patch triggers layer config updates on every RecompositeLayer diff. A future
patch will optimize this, and add a layout test.

• rendering/RenderLayerCompositor.cpp: (WebCore::RenderLayerCompositor::layerStyleChanged):

git-svn-id: ​https://svn.webkit.org/repository/webkit/trunk@238229 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10:25 AM Changeset in webkit [238233] by Alan Coon

• 3 edits
  2 adds in tags/Safari-607.1.14

Cherry-pick r238170. rdar://problem/46083440

Transform of composited layer not updated when layer also needs repaint
​https://bugs.webkit.org/show_bug.cgi?id=191598

Reviewed by Simon Fraser.

Source/WebCore:

Trigger a compositing geometry update whenever a RenderLayer's transform changes,
even when other parts of its style have changed in a way that produces a
StyleDifference greater than RecompositeLayer.

Test: compositing/geometry/transform-and-repaint-updates-geometry.html

• rendering/RenderLayerCompositor.cpp: (WebCore::RenderLayerCompositor::layerStyleChanged):

LayoutTests:

• compositing/geometry/transform-and-repaint-updates-geometry-expected.html: Added.
• compositing/geometry/transform-and-repaint-updates-geometry.html: Added.

git-svn-id: ​https://svn.webkit.org/repository/webkit/trunk@238170 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10:25 AM Changeset in webkit [238232] by youenn@apple.com

• 3 edits in trunk/Source/WebCore

Modernize RTCPeerConnection handling of pendingActivity
​https://bugs.webkit.org/show_bug.cgi?id=191661

Reviewed by Eric Carlson.

makePendingActivity is the modern way to handle set/unset of pending activity.
No change of behavior.

• Modules/mediastream/RTCPeerConnection.cpp:

(WebCore::RTCPeerConnection::create):
(WebCore::RTCPeerConnection::doStop):

• Modules/mediastream/RTCPeerConnection.h:

10:20 AM Changeset in webkit [238231] by Keith Rollin

• 4 edits
  8 deletes in trunk

Delete old .xcfilelist files
​https://bugs.webkit.org/show_bug.cgi?id=191669
<rdar://problem/46081994>

Reviewed by Chris Dumez.

.xcfilelist files were created and added to the Xcode project files in
​https://trac.webkit.org/changeset/238008/webkit. However, they caused
build issues and they were removed from the Xcode projects in
​https://trac.webkit.org/changeset/238055/webkit. This check-in removes
the files from the repository altogether. They'll ultimately be
replaced with new files with names that indicate whether the
associated files are inputs to the Run Script phase or are files
created by the Run Script phase.

Source/JavaScriptCore:

• DerivedSources.xcfilelist: Removed.
• UnifiedSources.xcfilelist: Removed.

Source/WebCore:

No new tests -- no changed functionality.

• DerivedSources.xcfilelist: Removed.
• UnifiedSources.xcfilelist: Removed.

Source/WebKit:

• DerivedSources.xcfilelist: Removed.
• UnifiedSources.xcfilelist: Removed.

Tools:

• DumpRenderTree/DerivedSources.xcfilelist: Removed.
• WebKitTestRunner/DerivedSources.xcfilelist: Removed.

8:41 AM Changeset in webkit [238230] by youenn@apple.com

• 7 edits in trunk

Update RTCPeerConnection JS built-ins to be closer to specWe
​https://bugs.webkit.org/show_bug.cgi?id=191665

Reviewed by Eric Carlson.

LayoutTests/imported/w3c:

• web-platform-tests/webrtc/rtcpeerconnection/rtcpeerconnection-idl-expected.txt:

Source/WebCore:

Simplify JS built-ins since we no longer need to support callback versions of the API.
Make sure to have the right number of parameters in the JS built-in functions.
Make some simplification to the code.
Covered by existing tests and rebased test.

• Modules/mediastream/RTCPeerConnection.js:

(createOffer):
(createAnswer):
(setLocalDescription):
(setRemoteDescription):
(addIceCandidate):

• Modules/mediastream/RTCPeerConnectionInternals.js:

(enqueueOperation):
(callbacksAndDictionaryOverload): Deleted.

LayoutTests:

• TestExpectations:

7:57 AM Changeset in webkit [238229] by Simon Fraser

• 2 edits in trunk/Source/WebCore

REGRESSION(r238090): Composited iframe contents disappear after switching tabs in Safari
​https://bugs.webkit.org/show_bug.cgi?id=191673
rdar://problem/46083440

Reviewed by Antti Koivisto.

Switching tabs in Safari triggers the "setIsInWindow" code path, that detaches the layer
tree for every Frame. They get re-attached on tab show, and for subframes this involves
the triggering of a fake style recalc in the parent document via scheduleInvalidateStyleAndLayerComposition().

The style diff that's sent to RenderLayerCompositor::layerStyleChanged() as a result of that
fake style recalc is RecompositeLayer, but the code was failing to trigger the necessary
layer configuration update that gets iframe layers parented.

This stop-gap patch triggers layer config updates on every RecompositeLayer diff. A future
patch will optimize this, and add a layout test.

• rendering/RenderLayerCompositor.cpp:

(WebCore::RenderLayerCompositor::layerStyleChanged):

7:41 AM Changeset in webkit [238228] by magomez@igalia.com

• 2 edits in trunk/LayoutTests

Unreviewed GTK+ gardening after r238220.

• platform/gtk/TestExpectations:

7:35 AM Changeset in webkit [238227] by Alan Bujtas

• 2 edits in trunk/Source/WebCore

[LFC][IFC] Skip non-inflow boxes while splitting the inline runs.
​https://bugs.webkit.org/show_bug.cgi?id=191690

Reviewed by Antti Koivisto.

Skip all non-inflow boxes (floats, out-of-flow positioned elements). They don't participate in the inline run context.

• layout/Verification.cpp:

(WebCore::Layout::LayoutState::verifyAndOutputMismatchingLayoutTree const):

• layout/inlineformatting/InlineFormattingContext.cpp:

(WebCore::Layout::InlineFormattingContext::splitInlineRunIfNeeded const):

6:58 AM Changeset in webkit [238226] by Alan Bujtas

• 5 edits in trunk/Source/WebCore

[LFC] FormattingContext base class should not declare computeStaticPosition.
​https://bugs.webkit.org/show_bug.cgi?id=191683

Reviewed by Antti Koivisto.

Apparently only BlockFormattingContext uses it.

• layout/FormattingContext.h:
• layout/blockformatting/BlockFormattingContext.h:
• layout/inlineformatting/InlineFormattingContext.cpp:

(WebCore::Layout::InlineFormattingContext::computeStaticPosition const): Deleted.

• layout/inlineformatting/InlineFormattingContext.h:

6:42 AM Changeset in webkit [238225] by Alan Bujtas

• 10 edits
  2 adds in trunk

[LFC][IFC] Add support for in-flow positioned inline boxes.
​https://bugs.webkit.org/show_bug.cgi?id=191672

Reviewed by Antti Koivisto.

Source/WebCore:

We might offset the in-flow positioned runs differently once runs are moved over to the display tree.

Test: fast/inline/simple-inline-inflow-positioned.html

• layout/Verification.cpp:

(WebCore::Layout::outputMismatchingComplexLineInformationIfNeeded):

• layout/inlineformatting/InlineFormattingContext.cpp:

(WebCore::Layout::InlineFormattingContext::splitInlineRunsIfNeeded const):
(WebCore::Layout::InlineFormattingContext::postProcessInlineRuns const):
(WebCore::Layout::InlineFormattingContext::placeInFlowPositionedChildren const):
(WebCore::Layout::InlineFormattingContext::collectInlineContentForSubtree const):

• layout/inlineformatting/InlineFormattingContext.h:
• layout/inlineformatting/InlineRun.h:

(WebCore::Layout::InlineRun::moveVertically):

• layout/layouttree/LayoutTreeBuilder.cpp:

(WebCore::Layout::outputInlineRuns):

Tools:

• LayoutReloaded/misc/LFC-passing-tests.txt:

LayoutTests:

• fast/inline/simple-inline-inflow-positioned-expected.txt: Added.
• fast/inline/simple-inline-inflow-positioned.html: Added.
• platform/ios/TestExpectations:

6:39 AM Changeset in webkit [238224] by commit-queue@webkit.org

• 3 edits in trunk/Source/WebCore

[GStreamer][WebRTC] Add support for sending silence or silencing an incoming track
​https://bugs.webkit.org/show_bug.cgi?id=191631

Patch by Thibault Saunier <​tsaunier@igalia.com> on 2018-11-15
Reviewed by Xabier Rodriguez-Calvar.

This will be tested once webkit.org/b/186933 is implemented.

• platform/mediastream/gstreamer/RealtimeIncomingAudioSourceLibWebRTC.cpp:

(WebCore::RealtimeIncomingAudioSourceLibWebRTC::OnData):

• platform/mediastream/gstreamer/RealtimeOutgoingAudioSourceLibWebRTC.cpp:

(WebCore::RealtimeOutgoingAudioSourceLibWebRTC::pullAudioData):

5:57 AM Changeset in webkit [238223] by Antti Koivisto

• 6 edits in trunk

REGRESSION(r238178): fast/forms/access-key-mutated.html and fast/forms/access-key-case-insensitive.html are timing out
​https://bugs.webkit.org/show_bug.cgi?id=191642

Reviewed by Zalan Bujtas.

Source/WebCore:

Invalidate access key map even when thorttling style recalcs.

• dom/Document.cpp:

(WebCore::Document::scheduleStyleRecalc):

LayoutTests:

• TestExpectations: Unskip
• platform/mac-wk2/TestExpectations: Unskip
• platform/mac-wk2/accessibility/mac/selection-notification-focus-change-expected.txt:

Rebase. Focus timings depends on layout timings.

5:48 AM Changeset in webkit [238222] by Antti Koivisto

• 3 edits
  2 adds in trunk

Remove fonts from CSSFontFaceSet safely
​https://bugs.webkit.org/show_bug.cgi?id=191676

Reviewed by Zalan Bujtas.

Source/WebCore:

Test: fast/text/font-face-set-remove-safely.html

• css/CSSFontFaceSet.cpp:

(WebCore::CSSFontFaceSet::remove):

LayoutTests:

• fast/text/font-face-set-remove-safely-expected.txt: Added.
• fast/text/font-face-set-remove-safely.html: Added.

5:43 AM Changeset in webkit [238221] by fred.wang@free.fr

• 4 edits in trunk/Source/WebKit

[WebAuthN] Make AuthenticatorManager
​https://bugs.webkit.org/show_bug.cgi?id=189279

Unreviewed build fix.

Patch by Frederic Wang <​fwang@igalia.com> on 2018-11-15

• UIProcess/WebAuthentication/Cocoa/LocalAuthenticator.mm: Add missing header.
• UIProcess/WebAuthentication/Cocoa/LocalConnection.mm: Ditto.
• UIProcess/WebAuthentication/WebAuthenticatorCoordinatorProxy.cpp: Ditto.

4:19 AM Changeset in webkit [238220] by Manuel Rego Casasnovas

• 4 edits
  6 adds in trunk

[css-grid] Consider scrollbars in populateGridPositionsForDirection()
​https://bugs.webkit.org/show_bug.cgi?id=191656

Reviewed by Javier Fernandez.

LayoutTests/imported/w3c:

Imported tests from WPT.

• web-platform-tests/css/css-grid/grid-model/grid-container-scrollbar-001-expected.html: Added.
• web-platform-tests/css/css-grid/grid-model/grid-container-scrollbar-001.html: Added.
• web-platform-tests/css/css-grid/grid-model/grid-container-scrollbar-vertical-lr-001-expected.html: Added.
• web-platform-tests/css/css-grid/grid-model/grid-container-scrollbar-vertical-lr-001.html: Added.
• web-platform-tests/css/css-grid/grid-model/grid-container-scrollbar-vertical-rl-001-expected.html: Added.
• web-platform-tests/css/css-grid/grid-model/grid-container-scrollbar-vertical-rl-001.html: Added.
• web-platform-tests/css/css-grid/grid-model/w3c-import.log:

Source/WebCore:

We never care about scrollbars in RenderGrid::populateGridPositionsForDirection(),
that's fine if the scrollbars are at the end (e.g. on the right in horizontal writing mode and LTR direction)
but it causes problems when they're at the beginning (e.g. on the left in horizontal writing mode and RTL direction).

The patch modifies the method so it takes into account scrollbar size
in order to compute the position of the columns/rows depending on the direction and the writing mode.

Tests: imported/w3c/web-platform-tests/css/css-grid/grid-model/grid-container-scrollbar-001.html

imported/w3c/web-platform-tests/css/css-grid/grid-model/grid-container-scrollbar-vertical-lr-001.html
imported/w3c/web-platform-tests/css/css-grid/grid-model/grid-container-scrollbar-vertical-rl-001.html

• rendering/RenderGrid.cpp:

(WebCore::RenderGrid::populateGridPositionsForDirection):