WebKit-https.git
15 months ago[GLIB] jsc_context_evaluate_in_object() should receive an instance when a JSCClass...
carlosgc@webkit.org [Fri, 20 Jul 2018 06:11:41 +0000 (06:11 +0000)]
[GLIB] jsc_context_evaluate_in_object() should receive an instance when a JSCClass is given
https://bugs.webkit.org/show_bug.cgi?id=187798

Reviewed by Michael Catanzaro.

Source/JavaScriptCore:

Because a JSCClass is pretty much useless without an instance in this case. It should be similar to
jsc_value_new_object() because indeed we are creating a new object. This makes destroy function and vtable
functions to work. We can't use JSAPIWrapperObject to wrap this object, because it's a global object, so this
patch adds JSAPIWrapperGlobalObject or that.

* API/glib/JSAPIWrapperGlobalObject.cpp: Added.
(jsAPIWrapperGlobalObjectHandleOwner):
(JSAPIWrapperGlobalObjectHandleOwner::finalize):
(JSC::JSCallbackObject<JSAPIWrapperGlobalObject>::createStructure):
(JSC::JSCallbackObject<JSAPIWrapperGlobalObject>::create):
(JSC::JSAPIWrapperGlobalObject::JSAPIWrapperGlobalObject):
(JSC::JSAPIWrapperGlobalObject::finishCreation):
(JSC::JSAPIWrapperGlobalObject::visitChildren):
* API/glib/JSAPIWrapperGlobalObject.h: Added.
(JSC::JSAPIWrapperGlobalObject::wrappedObject const):
(JSC::JSAPIWrapperGlobalObject::setWrappedObject):
* API/glib/JSCClass.cpp:
(isWrappedObject): Helper to check if the given object is a JSAPIWrapperObject or JSAPIWrapperGlobalObject.
(wrappedObjectClass): Return the class of a wrapped object.
(jscContextForObject): Get the execution context of an object. If the object is a JSAPIWrapperGlobalObject, the
scope extension global object is used instead.
(getProperty): Use isWrappedObject, wrappedObjectClass and jscContextForObject.
(setProperty): Ditto.
(hasProperty): Ditto.
(deleteProperty): Ditto.
(getPropertyNames): Ditto.
(jscClassCreateContextWithJSWrapper): Call jscContextCreateContextWithJSWrapper().
* API/glib/JSCClassPrivate.h:
* API/glib/JSCContext.cpp:
(jscContextCreateContextWithJSWrapper): Call WrapperMap::createContextWithJSWrappper().
(jsc_context_evaluate_in_object): Use jscClassCreateContextWithJSWrapper() when a JSCClass is given.
* API/glib/JSCContext.h:
* API/glib/JSCContextPrivate.h:
* API/glib/JSCWrapperMap.cpp:
(JSC::WrapperMap::createContextWithJSWrappper): Create the new context for jsc_context_evaluate_in_object() here
when a JSCClass is used to create the JSAPIWrapperGlobalObject.
(JSC::WrapperMap::wrappedObject const): Return the wrapped object also in case of JSAPIWrapperGlobalObject.
* API/glib/JSCWrapperMap.h:
* GLib.cmake:

Tools:

Update test cases to the new API and use a destroy function and vtable in the test case of calling
jsc_context_evaluate_in_object() with a JSCClass.

* TestWebKitAPI/Tests/JavaScriptCore/glib/TestJSC.cpp:
(testJSCEvaluateInObject):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@234025 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 months agoTestWTF.WTF.StringConcatenate_Unsigned fails for unsigned short on Windows
ross.kirsling@sony.com [Fri, 20 Jul 2018 06:10:34 +0000 (06:10 +0000)]
TestWTF.WTF.StringConcatenate_Unsigned fails for unsigned short on Windows
https://bugs.webkit.org/show_bug.cgi?id=187712

Reviewed by Fujii Hironori.

* TestWebKitAPI/Tests/WTF/StringConcatenate.cpp:
Mark result of unsigned short test case as platform-specific,
since Windows' behavior is actually *less* surprising here.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@234024 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 months agoNull pointer dereference under WebPage::autofillLoginCredentials()
cdumez@apple.com [Fri, 20 Jul 2018 04:53:59 +0000 (04:53 +0000)]
Null pointer dereference under WebPage::autofillLoginCredentials()
https://bugs.webkit.org/show_bug.cgi?id=187823
<rdar://problem/37152195>

Reviewed by David Kilzer.

Deal with m_assistedNode being null under WebPage::autofillLoginCredentials().

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

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@234023 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 months agoConservatively make Object.assign's fast path do a two phase protocol of loading...
sbarati@apple.com [Fri, 20 Jul 2018 04:47:51 +0000 (04:47 +0000)]
Conservatively make Object.assign's fast path do a two phase protocol of loading everything then storing everything to try to prevent a crash
https://bugs.webkit.org/show_bug.cgi?id=187836
<rdar://problem/42409527>

Reviewed by Mark Lam.

We have crash reports that we're crashing on source->getDirect in Object.assign's
fast path. Mark investigated this and determined we end up with a nullptr for
butterfly. This is curious, because source's Structure indicated that it has
out of line properties. My leading hypothesis for this at the moment is a bit
handwavy, but it's essentially:
- We end up firing a watchpoint when assigning to the target (this can happen
if a watchpoint was set up for storing to that particular field)
- When we fire that watchpoint, we end up doing some kind work on the source,
perhaps causing it to flattenDictionaryStructure. Therefore, we end up
mutating source.

I'm not super convinced this is what we're running into, but just by reading
the code, I think it needs to be something similar to this. Seeing if this change
fixes the crasher will give us good data to determine if something like this is
happening or if the bug is something else entirely.

* runtime/ObjectConstructor.cpp:
(JSC::objectConstructorAssign):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@234022 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 months agoWeb Inspector: Layers visualization shouldn't select on mousedown
ross.kirsling@sony.com [Fri, 20 Jul 2018 04:29:21 +0000 (04:29 +0000)]
Web Inspector: Layers visualization shouldn't select on mousedown
https://bugs.webkit.org/show_bug.cgi?id=187488

Reviewed by Matt Baker.

* UserInterface/Views/Layers3DContentView.js:
(WI.Layers3DContentView):
(WI.Layers3DContentView.prototype.initialLayout):
(WI.Layers3DContentView.prototype._canvasMouseDown):
(WI.Layers3DContentView.prototype._canvasMouseUp):
Don't update selection on mousedown, update on mouseup!
Specifically, only update when mousedown & mouseup targets are the same and mousemove hasn't been triggered.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@234021 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 months ago[ITP] Crash under ResourceLoadStatisticsMemoryStore::removeDataRecords()
cdumez@apple.com [Fri, 20 Jul 2018 03:44:58 +0000 (03:44 +0000)]
[ITP] Crash under ResourceLoadStatisticsMemoryStore::removeDataRecords()
https://bugs.webkit.org/show_bug.cgi?id=187821
<rdar://problem/42112693>

Reviewed by David Kilzer.

In two cases, ResourceLoadStatisticsMemoryStore (which lives on a background queue) needs to call WebPageProxy
operations on the main thread and then dispatch back on the background queue when the operation completes.
However, it is possible for the ResourceLoadStatisticsMemoryStore to get destroyed on the background queue
during this time and we would then crash when trying to use m_workQueue to re-dispatch. To address the issue,
I now ref the work queue in the lambda so that we're guaranteed to be able to re-dispatch to the background
queue. When we're back on the background queue, we'll realize that weakThis in gone and we'll call the callback
and return early.

Note that I am not checking weakThis on the main thread as this would not be safe. weakThis should only be
used on the background queue.

* UIProcess/ResourceLoadStatisticsMemoryStore.cpp:
(WebKit::ResourceLoadStatisticsMemoryStore::removeDataRecords):
(WebKit::ResourceLoadStatisticsMemoryStore::grandfatherExistingWebsiteData):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@234020 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 months agoMove WebFrameListenerProxy to WebFramePolicyListenerProxy, its only subclass
Hironori.Fujii@sony.com [Fri, 20 Jul 2018 02:24:20 +0000 (02:24 +0000)]
Move WebFrameListenerProxy to WebFramePolicyListenerProxy, its only subclass
https://bugs.webkit.org/show_bug.cgi?id=187825
<rdar://problem/42405081>

Unreviewed build fix for Windows port.

RefPtr.h(45): error C2027: use of undefined type 'API::Navigation'

* UIProcess/win/WebInspectorProxyWin.cpp: Include "APINavigation.h".

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@234019 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 months agoFlaky crash in AnimationTimeline::cancelOrRemoveDeclarativeAnimation
graouts@webkit.org [Fri, 20 Jul 2018 00:06:10 +0000 (00:06 +0000)]
Flaky crash in AnimationTimeline::cancelOrRemoveDeclarativeAnimation
https://bugs.webkit.org/show_bug.cgi?id=187530
<rdar://problem/42095186>

Reviewed by Dean Jackson.

LayoutTests/imported/mozilla:

Mark a WPT progression now that we correctly ignore animation names that have no matching @keyframes rule.

* css-animations/test_element-get-animations-expected.txt:

Source/WebCore:

We would crash in cancelOrRemoveDeclarativeAnimation() because updateCSSAnimationsForElement() would pass
nullptr values due to the return value of cssAnimationsByName.take(nameOfAnimationToRemove). This is because
we would create animations for animation names that may be empty or not match an existing @keyframes rule.
Not only was that wasteful, but it was also non-compliant, and as a result of fixing this we're actually
seeing a progression in the CSS Animations WPT tests.

* animation/AnimationTimeline.cpp:
(WebCore::shouldConsiderAnimation): New function that performs all required steps to see if a provided animation
is valid and has a name that is not "none", not the empty string and matches the name of a @keyframes rule.
(WebCore::AnimationTimeline::updateCSSAnimationsForElement):
* animation/KeyframeEffectReadOnly.cpp:
(WebCore::KeyframeEffectReadOnly::computeCSSAnimationBlendingKeyframes): We no longer need to check whether we have
an empty animation name since we're no longer creating CSSAnimation objects in that circumstance.
* css/StyleResolver.cpp:
(WebCore::StyleResolver::isAnimationNameValid): Add a new method that checks whether the provided animation name
a known @keyframes rule.
* css/StyleResolver.h:

LayoutTests:

Adjust an existing test which assumes an animation might be running when it's not really, so we test the animation is
not running using an alternate method.

* animations/keyframes-dynamic-expected.txt:
* animations/keyframes-dynamic.html:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@234017 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 months agoTeach the AWS Lambda to use the [minified]-platforms database
lforschler@apple.com [Thu, 19 Jul 2018 23:37:49 +0000 (23:37 +0000)]
Teach the AWS Lambda to use the [minified]-platforms database

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@234016 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 months agoMark storage/indexeddb/modern/opendatabase-after-storage-crash.html as flaky.
ryanhaddad@apple.com [Thu, 19 Jul 2018 23:30:11 +0000 (23:30 +0000)]
Mark storage/indexeddb/modern/opendatabase-after-storage-crash.html as flaky.
https://bugs.webkit.org/show_bug.cgi?id=187648

Unreviewed test gardening.

* platform/wk2/TestExpectations:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@234015 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 months agoFix the test.
simon.fraser@apple.com [Thu, 19 Jul 2018 23:20:49 +0000 (23:20 +0000)]
Fix the test.

* editing/mac/attributed-string/attributed-string-for-typing-with-color-filter.html:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@234014 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 months agoCrash under WebCore::DocumentWriter::addData()
cdumez@apple.com [Thu, 19 Jul 2018 23:19:31 +0000 (23:19 +0000)]
Crash under WebCore::DocumentWriter::addData()
https://bugs.webkit.org/show_bug.cgi?id=187819
<rdar://problem/41328743>

Reviewed by Brady Eidson.

When AppCache is used a DocumentLoader may start a NetworkLoad even though it has substitute data.
In DocumentLoader::continueAfterContentPolicy(), if we have substitute data we commit this data
and call finishLoad(). However, if the case where there was a NetworkLoad started, we'll send the
ContinueDidReceiveResponse IPC back to the network process and it will start sending us data for
the load. This could lead to crashes such as <rdar://problem/41328743> since the DocumentLoader
has already committed data and finished loading when it gets the data from the network process.

To address the issue, we now call clearMainResource() in continueAfterContentPolicy(), after we've
decided to commit the substitute data. This effectively removes the DocumentLoader as a client of
the CachedResource so that its will not be notified of following load progress. We do not cancel
the load as other CachedResourceClients may be interested in the load (ApplicationCacheResourceLoader
in particular, in order to update its cached data).

* loader/DocumentLoader.cpp:
(WebCore::DocumentLoader::continueAfterContentPolicy):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@234013 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 months agoProvide an lldb type summary for WebCore::Color
dino@apple.com [Thu, 19 Jul 2018 23:17:45 +0000 (23:17 +0000)]
Provide an lldb type summary for WebCore::Color
https://bugs.webkit.org/show_bug.cgi?id=187776

Dan Bates helped me with this attempted fix. Fingers crossed!

* Scripts/build-lldbwebkittester:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@234012 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 months agoCrashTracer: com.apple.WebKit.WebContent.Development at com.apple.WebCore: std::optio...
dino@apple.com [Thu, 19 Jul 2018 23:17:41 +0000 (23:17 +0000)]
CrashTracer: com.apple.WebKit.WebContent.Development at com.apple.WebCore: std::optional<WTF::Vector<WebCore::PluginInfo, 0ul, WTF::CrashOnOverflow, 16ul> >::operator* & + 73
https://bugs.webkit.org/show_bug.cgi?id=187820
<rdar://problem/42017759>

Reviewed by Antoine Quint.

Speculative fix for this crash, which is accessing an optional without checking
if it exists. The crash logs didn't point to a reproducible test case.

* plugins/PluginData.cpp:
(WebCore::PluginData::supportsWebVisibleMimeTypeForURL const): Return false if
the optional doesn't exist.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@234011 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 months ago[WinCairo] Support DEVELOPER_MODE for allowing inspection of web inspector
stephan.szabo@sony.com [Thu, 19 Jul 2018 23:11:24 +0000 (23:11 +0000)]
[WinCairo] Support DEVELOPER_MODE for allowing inspection of web inspector
https://bugs.webkit.org/show_bug.cgi?id=187786

Reviewed by Fujii Hironori.

.:

* Source/cmake/OptionsWinCairo.cmake: Add ENABLE_DEVELOPER_MODE
to build when DEVELOPER_MODE is turned on at cmake time.

Tools:

* Scripts/webkitdirs.pm:
(generateBuildSystemFromCMakeProject): Turn on DEVELOPER_MODE
for WinCairo builds.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@234010 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 months agoMove WebFrameListenerProxy to WebFramePolicyListenerProxy, its only subclass
achristensen@apple.com [Thu, 19 Jul 2018 23:11:11 +0000 (23:11 +0000)]
Move WebFrameListenerProxy to WebFramePolicyListenerProxy, its only subclass
https://bugs.webkit.org/show_bug.cgi?id=187825

Reviewed by Brady Eidson.

* CMakeLists.txt:
* UIProcess/API/C/WKPage.cpp:
* UIProcess/API/Cocoa/WKBrowsingContextController.mm:
* UIProcess/Automation/WebAutomationSession.cpp:
* UIProcess/Cocoa/WebViewImpl.mm:
* UIProcess/RemoteWebInspectorProxy.cpp:
* UIProcess/WebFormSubmissionListenerProxy.h:
(WebKit::WebFormSubmissionListenerProxy::create):
(WebKit::WebFormSubmissionListenerProxy::WebFormSubmissionListenerProxy):
* UIProcess/WebFrameListenerProxy.cpp: Removed.
* UIProcess/WebFrameListenerProxy.h: Removed.
* UIProcess/WebFramePolicyListenerProxy.cpp:
(WebKit::WebFramePolicyListenerProxy::WebFramePolicyListenerProxy):
(WebKit::WebFramePolicyListenerProxy::receivedPolicyDecision):
(WebKit::WebFramePolicyListenerProxy::changeWebsiteDataStore):
(WebKit::WebFramePolicyListenerProxy::invalidate):
(WebKit::WebFramePolicyListenerProxy::isMainFrame const):
(WebKit::WebFramePolicyListenerProxy::setNavigation):
* UIProcess/WebFramePolicyListenerProxy.h:
(WebKit::WebFramePolicyListenerProxy::listenerID const):
(WebKit::WebFramePolicyListenerProxy::operator new): Deleted.
* UIProcess/WebFrameProxy.cpp:
(WebKit::WebFrameProxy::activePolicyListenerProxy):
* UIProcess/WebFrameProxy.h:
* UIProcess/WebInspectorProxy.cpp:
* UIProcess/WebProcessPool.cpp:
* UIProcess/ios/ViewGestureControllerIOS.mm:
* UIProcess/mac/ViewGestureControllerMac.mm:
* UIProcess/mac/WKInspectorViewController.mm:
* WebKit.xcodeproj/project.pbxproj:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@234009 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 months agoPlatformThread::Run does not need to log the fact that it is running
cdumez@apple.com [Thu, 19 Jul 2018 22:56:39 +0000 (22:56 +0000)]
PlatformThread::Run does not need to log the fact that it is running
https://bugs.webkit.org/show_bug.cgi?id=187801i
<rdar://problem/40331421>

Patch by Youenn Fablet <youenn@apple.com> on 2018-07-19
Reviewed by Chris Dumez.

* Source/webrtc/rtc_base/platform_thread.cc:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@234008 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 months agoEnsure DocumentTimeline is kept alive until the VM::whenIdle callback is called
graouts@webkit.org [Thu, 19 Jul 2018 21:55:41 +0000 (21:55 +0000)]
Ensure DocumentTimeline is kept alive until the VM::whenIdle callback is called
https://bugs.webkit.org/show_bug.cgi?id=187692

Reviewed by Ryosuke Niwa.

Ensure we keep the DocumentTimeline alive until the VM::whenIdle callback is called.

* animation/DocumentTimeline.cpp:
(WebCore::DocumentTimeline::currentTime):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@234007 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 months agoAdd benchmark for WebKit process launch times
commit-queue@webkit.org [Thu, 19 Jul 2018 21:54:48 +0000 (21:54 +0000)]
Add benchmark for WebKit process launch times
https://bugs.webkit.org/show_bug.cgi?id=186414

Patch by Ben Richards <benton_richards@apple.com> on 2018-07-19
Reviewed by Ryosuke Niwa.

Added two benchmarks, one for measuring browser new tab launch time and one for browser startup time.

* LaunchTime/.gitignore: Added.
* LaunchTime/feedback_client.html: Added.
Displays benchmark progress in browser
* LaunchTime/feedback_server.py: Added.
(FeedbackServer): Sends data to feedback_client via websocket
(FeedbackServer.__init__):
(FeedbackServer._create_app):
(FeedbackServer._start_server):
(FeedbackServer._send_all_messages):
(FeedbackServer.start):
(FeedbackServer.stop):
(FeedbackServer.send_message): Send a message to the feedback_client
(FeedbackServer.wait_until_client_has_loaded): Wait until the feedback_client has opened a websocket connection to the feedback_server
(FeedbackServer.MainHandler): Handler factory to create handler that serves feedback_client.html
(FeedbackServer.MainHandler.Handler):
(FeedbackServer.MainHandler.Handler.get):
(FeedbackServer.WSHandler): Handler factory to create handler that sends data to feedback client
(FeedbackServer.WSHandler.Handler):
(FeedbackServer.WSHandler.Handler.open): On websocket connection opened
(FeedbackServer.WSHandler.Handler.on_close): On websocket connection closed
* LaunchTime/launch_time.py: Added.
(DefaultLaunchTimeHandler): Abstract HTTP request handler for launch time benchmarks
(DefaultLaunchTimeHandler.get_test_page): Default test page to be overridden by benchmarks
(DefaultLaunchTimeHandler.get_blank_page):
(DefaultLaunchTimeHandler.on_receive_stop_time):
(DefaultLaunchTimeHandler.do_HEAD):
(DefaultLaunchTimeHandler.do_GET):
(DefaultLaunchTimeHandler.do_POST):
(DefaultLaunchTimeHandler.log_message): Suppresses HTTP logs from SimpleHTTPRequestHandler
(LaunchTimeBenchmark): Abstract class which launch time benchmarks inherit from and override methods desired to customize
(LaunchTimeBenchmark.__init__):
(LaunchTimeBenchmark._parse_browser_bundle_path): Parser for bundle path option
(LaunchTimeBenchmark._parse_args):
(LaunchTimeBenchmark._run_server): Target function for main server thread
(LaunchTimeBenchmark._setup_servers):
(LaunchTimeBenchmark._clean_up):
(LaunchTimeBenchmark._exit_due_to_exception):
(LaunchTimeBenchmark._geometric_mean):
(LaunchTimeBenchmark._standard_deviation):
(LaunchTimeBenchmark._compute_results): Returns mean and std dev of list of results and pretty prints if should_print=True is specified
(LaunchTimeBenchmark._wait_times): Mimic numpy.linspace
(LaunchTimeBenchmark.open_tab): Open a browser tab with the html given by self.response_handler.get_test_page
(LaunchTimeBenchmark.launch_browser): Open a broser to either the feedback client (if option is set) or a blank page
(LaunchTimeBenchmark.quit_browser):
(LaunchTimeBenchmark.quit_browser.quit_app):
(LaunchTimeBenchmark.quit_browser.is_app_closed):
(LaunchTimeBenchmark.close_tab):
(LaunchTimeBenchmark.wait):
(LaunchTimeBenchmark.log): Print to console and send to feedback client if --feedback-in-browser flag is used
(LaunchTimeBenchmark.log_verbose): Only logs if --verbose flag is used
(LaunchTimeBenchmark.run):
(LaunchTimeBenchmark.group_init): Initialization done before each round of iterations
(LaunchTimeBenchmark.run_iteration):
(LaunchTimeBenchmark.initialize): Convenience method to be overriden by subclasses which is called at the end of __init__
(LaunchTimeBenchmark.will_parse_arguments): Called before argparse.parse_args to let subclasses add new command line arguments
(LaunchTimeBenchmark.did_parse_arguments): Called after argparse.parse_args to let subclass initialize based on command line arguments
* LaunchTime/new_tab.py: Added
(NewTabBenchmark):
(NewTabBenchmark._parse_wait_time): Parser for wait time option
(NewTabBenchmark.initialize):
(NewTabBenchmark.run_iteration):
(NewTabBenchmark.group_init):
(NewTabBenchmark.will_parse_arguments):
(NewTabBenchmark.did_parse_arguments):
(NewTabBenchmark.ResponseHandler):
(NewTabBenchmark.ResponseHandler.Handler):
(NewTabBenchmark.ResponseHandler.Handler.get_test_page):
* LaunchTime/startup.py: Added
(StartupBenchmark): This benchmark measures browser startup time and initial page load time
(StartupBenchmark.initialize):
(StartupBenchmark.run_iteration):
(StartupBenchmark.ResponseHandler):
(StartupBenchmark.ResponseHandler.Handler):
(StartupBenchmark.ResponseHandler.Handler.get_test_page):
* LaunchTime/thirdparty/__init__.py: Added.
(AutoinstallImportHook): Auto installs tornado package for feedback server
(AutoinstallImportHook.__init__):
(AutoinstallImportHook._ensure_autoinstalled_dir_is_in_sys_path):
(AutoinstallImportHook.find_module):
(AutoinstallImportHook._install_tornado):
(AutoinstallImportHook.greater_than_equal_to_version):
(AutoinstallImportHook._install):
(AutoinstallImportHook.get_latest_pypi_url):
(AutoinstallImportHook.install_binary):
(autoinstall_everything):
(get_os_info):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@234006 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 months agoSetting foreground color when editing should take color-filter into account, and...
simon.fraser@apple.com [Thu, 19 Jul 2018 21:18:27 +0000 (21:18 +0000)]
Setting foreground color when editing should take color-filter into account, and report the correct foreground color for collapsed selections
https://bugs.webkit.org/show_bug.cgi?id=187778

Reviewed by Ryosuke Niwa.
Source/WebCore:

Fix two aspects of editing with color-filter:

1. When setting foreground color, inverse-transform the color through -apple-color-filter so that the user gets the color
   they chose when in Dark Mode. Tested by editing/style/exec-command-foreColor-with-color-filter.html.

2. When retrieving the style of the collapsed selection, take color filter into account so that color picker
   reflects the color the users sees, instead of the content color. Tested by editing/mac/attributed-string/attributed-string-for-typing-with-color-filter.html

Add two additional tests that ensure that -apple-color-filter does not impact the NSAttributedString code
path, since -apple-color-filter should not affect the behavior of Copy.

Tests: editing/mac/attributed-string/attrib-string-colors-with-color-filter.html
       editing/mac/attributed-string/attrib-string-range-with-color-filter.html
       editing/mac/attributed-string/attributed-string-for-typing-with-color-filter.html
       editing/style/exec-command-foreColor-with-color-filter.html

* editing/EditingStyle.cpp:
(WebCore::StyleChange::StyleChange):
(WebCore::StyleChange::extractTextStyles):
* editing/EditingStyle.h:
* editing/cocoa/EditorCocoa.mm:
(WebCore::Editor::fontAttributesForSelectionStart const):
* platform/graphics/filters/FilterOperation.cpp:
(WebCore::InvertLightnessFilterOperation::inverseTransformColor const):
* platform/graphics/filters/FilterOperation.h:
(WebCore::FilterOperation::inverseTransformColor const):
* platform/graphics/filters/FilterOperations.cpp:
(WebCore::FilterOperations::transformColor const):
(WebCore::FilterOperations::inverseTransformColor const):
* platform/graphics/filters/FilterOperations.h:

LayoutTests:

* editing/mac/attributed-string/attrib-string-colors-with-color-filter-expected.txt: Added.
* editing/mac/attributed-string/attrib-string-colors-with-color-filter.html: Added.
* editing/mac/attributed-string/attrib-string-range-with-color-filter-expected.txt: Added.
* editing/mac/attributed-string/attrib-string-range-with-color-filter.html: Added.
* editing/mac/attributed-string/attributed-string-for-typing-with-color-filter-expected.txt: Added.
* editing/mac/attributed-string/attributed-string-for-typing-with-color-filter.html: Added.
* editing/style/exec-command-foreColor-with-color-filter-expected.txt: Added.
* editing/style/exec-command-foreColor-with-color-filter.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@234005 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 months agoProvide an lldb type summary for WebCore::Color
dino@apple.com [Thu, 19 Jul 2018 21:05:19 +0000 (21:05 +0000)]
Provide an lldb type summary for WebCore::Color
https://bugs.webkit.org/show_bug.cgi?id=187776

Another attempted build fix for Debug.

* lldb/lldbWebKitTester/Configurations/Base.xcconfig:
* lldb/lldbWebKitTester/lldbWebKitTester.xcodeproj/project.pbxproj:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@234004 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 months agoUnreviewed, rolling out r233994.
david_fenton@apple.com [Thu, 19 Jul 2018 20:51:48 +0000 (20:51 +0000)]
Unreviewed, rolling out r233994.

Caused EWS and bot failures due to assertions added

Reverted changeset:

"FetchResponse should close its stream when loading finishes"
https://bugs.webkit.org/show_bug.cgi?id=187790
https://trac.webkit.org/changeset/233994

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@234003 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 months agoREGRESSION(r233926): media/modern-media-controls/media-controller/media-controller...
jer.noble@apple.com [Thu, 19 Jul 2018 20:48:43 +0000 (20:48 +0000)]
REGRESSION(r233926): media/modern-media-controls/media-controller/media-controller-inline-to-fullscreen-to-pip-to-inline.html is a TIMEOUT failure
https://bugs.webkit.org/show_bug.cgi?id=187813

Reviewed by Jon Lee.

In r233926, we changed the behavior of entering PiP to exit fullscreen only after entering PiP completes. The
test in question will immediately request "inline" presentation mode once the PiP animation begins, and thus
it's asking to "exit fullscreen" when both in standard fullscreen and also in PiP. The fix is not to bail out
early if we're in standard (element) fullscreen, but to allow the remaining steps to complete and exit PiP as
well.

* html/HTMLMediaElement.cpp:
(WebCore::HTMLMediaElement::exitFullscreen):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@234002 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 months ago[WHLSL] The interpreter doesn't support boolean short-circuiting
commit-queue@webkit.org [Thu, 19 Jul 2018 20:42:20 +0000 (20:42 +0000)]
[WHLSL] The interpreter doesn't support boolean short-circuiting
https://bugs.webkit.org/show_bug.cgi?id=187779

Patch by Thomas Denney <tdenney@apple.com> on 2018-07-19
Reviewed by Alex Christensen.

* WebGPUShadingLanguageRI/Evaluator.js:
(Evaluator.prototype.visitLogicalExpression): RHS is only evaluated when necessary
* WebGPUShadingLanguageRI/Test.js:
(tests.booleanShortcircuiting): Adds 4 tests for the evaluation of logical expresions

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@234001 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 months ago[LFC] Introduce simple line breaker.
zalan@apple.com [Thu, 19 Jul 2018 20:36:13 +0000 (20:36 +0000)]
[LFC] Introduce simple line breaker.
https://bugs.webkit.org/show_bug.cgi?id=187688

Reviewed by Antti Koivisto.

This patch takes the simple line layout implementation and refactors it in a way it is no longer requires a RenderBlockFlow object to run on.
Also this patch decouples text run generation and line breaking (and this implementation is going to replace the current simple line layout codebase)

TextContentProvider: Acts both as the container for all the text content (including hard line breaks) and as an iterator for the generated text runs.
SimpleTextRunGenerator: TextContentProvider uses it as the text run generator for simple content (in the future we'll have a ComplexTextRunGenerator).
SimpleLineBreaker: Input -> text runs + line constraints; Output -> layout runs after line breaking.

* Sources.txt:
* WebCore.xcodeproj/project.pbxproj:
* layout/inlineformatting/textlayout/ContentProvider.cpp: Added.
(WebCore::Layout::TextContentProvider::TextItem::Style::Style):
(WebCore::Layout::TextContentProvider::ContentProvider):
(WebCore::Layout::TextContentProvider::~ContentProvider):
(WebCore::Layout::TextContentProvider::appendText):
(WebCore::Layout::TextContentProvider::appendLineBreak):
(WebCore::Layout::TextContentProvider::width const):
(WebCore::Layout::TextContentProvider::textWidth const):
(WebCore::Layout::TextContentProvider::fixedPitchWidth const):
(WebCore::Layout::TextContentProvider::toTextItemIndex const):
(WebCore::Layout::TextContentProvider::length const):
(WebCore::Layout::TextContentProvider::iterator):
(WebCore::Layout::TextContentProvider::findNextRun):
(WebCore::Layout::TextContentProvider::current const):
* layout/inlineformatting/textlayout/ContentProvider.h: Added.
(WebCore::Layout::TextContentProvider::textContent const):
(WebCore::Layout::TextContentProvider::hardLineBreaks const):
(WebCore::Layout::TextContentProvider::Iterator::current const):
(WebCore::Layout::TextContentProvider::contains const):
(WebCore::Layout::TextContentProvider::Iterator::Iterator):
(WebCore::Layout::TextContentProvider::Iterator::operator++):
* layout/inlineformatting/textlayout/Runs.h: Added.
(WebCore::Layout::TextRun::isWhitespace const):
(WebCore::Layout::TextRun::isNonWhitespace const):
(WebCore::Layout::TextRun::isLineBreak const):
(WebCore::Layout::TextRun::isSoftLineBreak const):
(WebCore::Layout::TextRun::isHardLineBreak const):
(WebCore::Layout::TextRun::isValid const):
(WebCore::Layout::TextRun::isCollapsed const):
(WebCore::Layout::TextRun::type const):
(WebCore::Layout::TextRun::setIsCollapsed):
(WebCore::Layout::TextRun::setWidth):
(WebCore::Layout::LayoutRun::start const):
(WebCore::Layout::LayoutRun::end const):
(WebCore::Layout::LayoutRun::length const):
(WebCore::Layout::LayoutRun::left const):
(WebCore::Layout::LayoutRun::right const):
(WebCore::Layout::LayoutRun::width const):
(WebCore::Layout::LayoutRun::isEndOfLine const):
(WebCore::Layout::LayoutRun::setEnd):
(WebCore::Layout::LayoutRun::setRight):
(WebCore::Layout::LayoutRun::setIsEndOfLine):
(WebCore::Layout::LayoutRun::LayoutRun):
(WebCore::Layout::TextRun::createWhitespaceRun):
(WebCore::Layout::TextRun::createNonWhitespaceRun):
(WebCore::Layout::TextRun::createSoftLineBreakRun):
(WebCore::Layout::TextRun::createHardLineBreakRun):
(WebCore::Layout::TextRun::TextRun):
(WebCore::Layout::TextRun::start const):
(WebCore::Layout::TextRun::end const):
(WebCore::Layout::TextRun::length const):
(WebCore::Layout::TextRun::width const):
* layout/inlineformatting/textlayout/simple/SimpleContentProvider.cpp: Added.
(WebCore::Layout::SimpleContentProvider::SimpleContentProvider):
(WebCore::Layout::SimpleContentProvider::current const):
(WebCore::Layout::SimpleContentProvider::reset):
(WebCore::Layout::SimpleContentProvider::findNextRun):
(WebCore::Layout::SimpleContentProvider::moveToNextBreakablePosition):
(WebCore::Layout::SimpleContentProvider::moveToNextNonWhitespacePosition):
(WebCore::Layout::SimpleContentProvider::isAtLineBreak const):
(WebCore::Layout::SimpleContentProvider::isAtSoftLineBreak const):
* layout/inlineformatting/textlayout/simple/SimpleContentProvider.h: Added.
(WebCore::Layout::SimpleContentProvider::Iterator::reset):
(WebCore::Layout::SimpleContentProvider::Position::operator== const):
(WebCore::Layout::SimpleContentProvider::Position::operator< const):
(WebCore::Layout::SimpleContentProvider::Position::operator ContentPosition const):
(WebCore::Layout::SimpleContentProvider::Position::resetItemPosition):
(WebCore::Layout::SimpleContentProvider::Position::contentPosition const):
(WebCore::Layout::SimpleContentProvider::Position::itemPosition const):
(WebCore::Layout::SimpleContentProvider::Iterator<T>::Iterator):
(WebCore::Layout::SimpleContentProvider::Iterator<T>::current const):
(WebCore::Layout::SimpleContentProvider::Iterator<T>::operator):
(WebCore::Layout::SimpleContentProvider::Position::operator++):
(WebCore::Layout::SimpleContentProvider::Position::operator+=):
* layout/inlineformatting/textlayout/simple/SimpleLineBreaker.cpp: Added.
(WebCore::Layout::SimpleLineBreaker::TextRunList::TextRunList):
(WebCore::Layout::SimpleLineBreaker::Line::Line):
(WebCore::Layout::adjustedEndPosition):
(WebCore::Layout::SimpleLineBreaker::Line::append):
(WebCore::Layout::SimpleLineBreaker::Line::collapseTrailingWhitespace):
(WebCore::Layout::SimpleLineBreaker::Line::reset):
(WebCore::Layout::SimpleLineBreaker::Style::Style):
(WebCore::Layout::SimpleLineBreaker::SimpleLineBreaker):
(WebCore::Layout::SimpleLineBreaker::runs):
(WebCore::Layout::SimpleLineBreaker::createRunsForLine):
(WebCore::Layout::SimpleLineBreaker::handleOverflownRun):
(WebCore::Layout::SimpleLineBreaker::collapseLeadingWhitespace):
(WebCore::Layout::SimpleLineBreaker::collapseTrailingWhitespace):
(WebCore::Layout::SimpleLineBreaker::splitTextRun):
(WebCore::Layout::SimpleLineBreaker::split const):
(WebCore::Layout::SimpleLineBreaker::availableWidth const):
(WebCore::Layout::SimpleLineBreaker::verticalPosition const):
* layout/inlineformatting/textlayout/simple/SimpleLineBreaker.h: Added.
(WebCore::Layout::SimpleLineBreaker::TextRunList::overrideCurrent):
(WebCore::Layout::SimpleLineBreaker::TextRunList::isCurrentOverridden const):
(WebCore::Layout::SimpleLineBreaker::Line::availableWidth const):
(WebCore::Layout::SimpleLineBreaker::Line::hasContent const):
(WebCore::Layout::SimpleLineBreaker::Line::setAvailableWidth):
(WebCore::Layout::SimpleLineBreaker::Line::hasTrailingWhitespace const):
(WebCore::Layout::SimpleLineBreaker::Line::isWhitespaceOnly const):
(WebCore::Layout::SimpleLineBreaker::wrapContentOnOverflow const):
(WebCore::Layout::SimpleLineBreaker::TextRunList::current const):
(WebCore::Layout::SimpleLineBreaker::TextRunList::operator++):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@234000 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 months agoUnreviewed, rolling out r233998.
commit-queue@webkit.org [Thu, 19 Jul 2018 19:43:00 +0000 (19:43 +0000)]
Unreviewed, rolling out r233998.
https://bugs.webkit.org/show_bug.cgi?id=187815

Not needed. (Requested by mlam|a on #webkit).

Reverted changeset:

"Temporarily mitigate a bug where a source provider is null
when it shouldn't be."
https://bugs.webkit.org/show_bug.cgi?id=187812
https://trac.webkit.org/changeset/233998

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@233999 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 months agoTemporarily mitigate a bug where a source provider is null when it shouldn't be.
mark.lam@apple.com [Thu, 19 Jul 2018 19:28:08 +0000 (19:28 +0000)]
Temporarily mitigate a bug where a source provider is null when it shouldn't be.
https://bugs.webkit.org/show_bug.cgi?id=187812
<rdar://problem/41192691>

Reviewed by Michael Saboff.

Adding a null check to temporarily mitigate https://bugs.webkit.org/show_bug.cgi?id=187811.

* runtime/Error.cpp:
(JSC::addErrorInfo):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@233998 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 months agoRemove duplicate compilation of WebKitNSImageExtras.mm
krollin@apple.com [Thu, 19 Jul 2018 19:06:36 +0000 (19:06 +0000)]
Remove duplicate compilation of WebKitNSImageExtras.mm
https://bugs.webkit.org/show_bug.cgi?id=187782

Reviewed by Alex Christensen.

WebKitNSImageExtras.mm gets compiled twice, once because it's in
WebCore.xcodeproj/project.pbxproj and once because it's in
Webcore/SourcesCocoa.txt. This can lead to duplicate definition
errors, particularly when building with LTO enabled. Fix this by
removing the entry from the Xcode project.

No new tests -- no change in WebKit functionality.

* WebCore.xcodeproj/project.pbxproj:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@233995 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 months agoFetchResponse should close its stream when loading finishes
youenn@apple.com [Thu, 19 Jul 2018 18:32:53 +0000 (18:32 +0000)]
FetchResponse should close its stream when loading finishes
https://bugs.webkit.org/show_bug.cgi?id=187790

Reviewed by Chris Dumez.

It simplifies for a FetchResponse to push all its data into its stream if already created at end of load time.
Did some refactoring in FetchBodyOwner to have a cleaner relationship with the stream source.
Did a minor refactoring to expose the error description when loading fails as part of the rejected promise.
This is consistent to errors sent back through callbacks.

Covered by existing tests.

* Modules/fetch/FetchBodyOwner.cpp:
(WebCore::FetchBodyOwner::~FetchBodyOwner):
* Modules/fetch/FetchBodyOwner.h:
* Modules/fetch/FetchBodySource.cpp:
(WebCore::FetchBodySource::FetchBodySource):
(WebCore::FetchBodySource::setActive):
(WebCore::FetchBodySource::setInactive):
(WebCore::FetchBodySource::doStart):
(WebCore::FetchBodySource::doPull):
(WebCore::FetchBodySource::doCancel):
(WebCore::FetchBodySource::cleanBodyOwner):
* Modules/fetch/FetchBodySource.h:
* Modules/fetch/FetchResponse.cpp:
(WebCore::FetchResponse::BodyLoader::didSucceed):
(WebCore::FetchResponse::BodyLoader::didFail):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@233994 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 months ago[ win-ews ] http/tests/preload/onload_event.html is flakey crash on win-ews
tsavell@apple.com [Thu, 19 Jul 2018 18:07:47 +0000 (18:07 +0000)]
[ win-ews ] http/tests/preload/onload_event.html is flakey crash on win-ews
https://bugs.webkit.org/show_bug.cgi?id=187809

Unreviewed test gardening.

* platform/win/TestExpectations:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@233993 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 months agoUpdate iOS fullscreen alert text again
jonlee@apple.com [Thu, 19 Jul 2018 17:45:14 +0000 (17:45 +0000)]
Update iOS fullscreen alert text again
https://bugs.webkit.org/show_bug.cgi?id=187797
rdar://problem/42373783

Reviewed by Jer Noble.

Source/WebCore:

* English.lproj/Localizable.strings:

Source/WebKit:

* UIProcess/ios/fullscreen/WKFullScreenViewController.mm:
(-[WKFullScreenViewController _showPhishingAlert]):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@233992 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 months agoAdjust WEBCORE_EXPORT annotations for LTO
krollin@apple.com [Thu, 19 Jul 2018 17:43:35 +0000 (17:43 +0000)]
Adjust WEBCORE_EXPORT annotations for LTO
https://bugs.webkit.org/show_bug.cgi?id=187781
<rdar://problem/42351124>

Reviewed by Alex Christensen.

Continuation of Bug 186944. This bug addresses issues not caught
during the first pass of adjustments. The initial work focussed on
macOS; this one addresses issues found when building for iOS. From
186944:

Adjust a number of places that result in WebKit's
'check-for-weak-vtables-and-externals' script reporting weak external
symbols:

    ERROR: WebCore has a weak external symbol in it (/Volumes/Data/dev/webkit/OpenSource/WebKitBuild/Release/WebCore.framework/Versions/A/WebCore)
    ERROR: A weak external symbol is generated when a symbol is defined in multiple compilation units and is also marked as being exported from the library.
    ERROR: A common cause of weak external symbols is when an inline function is listed in the linker export file.
    ...

These cases are caused by inline methods being marked with WTF_EXPORT
(or related macro) or with an inline function being in a class marked
as such, and when enabling LTO builds.

For the most part, address these by removing the WEBCORE_EXPORT
annotation from inline methods. In some cases, move the implementation
out-of-line because it's the class that has the WEBCORE_EXPORT on it
and removing the annotation from the class would be too disruptive.
Finally, in other cases, move the implementation out-of-line because
check-for-weak-vtables-and-externals still complains when keeping the
implementation inline and removing the annotation; this seems to
typically (but not always) happen with destructors.

Source/JavaScriptCore:

* inspector/remote/RemoteAutomationTarget.cpp:
(Inspector::RemoteAutomationTarget::~RemoteAutomationTarget):
* inspector/remote/RemoteAutomationTarget.h:
* inspector/remote/RemoteInspector.cpp:
(Inspector::RemoteInspector::Client::~Client):
* inspector/remote/RemoteInspector.h:

Source/WebCore:

No new tests. There is no changed functionality. Only the annotation
and treatment of inline methods are altered.

* platform/graphics/FourCC.h:
(WebCore::FourCC::FourCC):
* platform/graphics/IntPoint.h:
(WebCore::IntPoint::IntPoint):
* platform/mediastream/RealtimeMediaSource.cpp:
(WebCore::RealtimeMediaSource::Observer::~Observer):
(WebCore::RealtimeMediaSource::AudioCaptureFactory::~AudioCaptureFactory):
(WebCore::RealtimeMediaSource::VideoCaptureFactory::~VideoCaptureFactory):
* platform/mediastream/RealtimeMediaSource.h:
* workers/service/ServiceWorkerProvider.cpp:
(WebCore::ServiceWorkerProvider::~ServiceWorkerProvider):
* workers/service/ServiceWorkerProvider.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@233990 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 months agoWeb Inspector: Dark Mode: poor contrast for Search Tab's "the page's content has...
nvasilyev@apple.com [Thu, 19 Jul 2018 17:37:17 +0000 (17:37 +0000)]
Web Inspector: Dark Mode: poor contrast for Search Tab's "the page's content has changed" message
https://bugs.webkit.org/show_bug.cgi?id=187792

Reviewed by Brian Burg.

* UserInterface/Views/DarkMode.css:
(@media (prefers-dark-interface)):
(:root):
Make the background color slightly more yellow and less red. This slightly dicreases the contrast
but makes the text look less like an error.

* UserInterface/Views/SearchSidebarPanel.css:
(@media (prefers-dark-interface)):
(.sidebar > .panel.navigation.search.changed > .banner):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@233989 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 months agoProvide an lldb type summary for WebCore::Color
dino@apple.com [Thu, 19 Jul 2018 17:07:57 +0000 (17:07 +0000)]
Provide an lldb type summary for WebCore::Color
https://bugs.webkit.org/show_bug.cgi?id=187776

Attempted build fix.

* lldb/lldbWebKitTester/Configurations/Base.xcconfig:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@233988 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 months agoUnreviewed, check scope after performing getPropertySlot in JSON.stringify
utatane.tea@gmail.com [Thu, 19 Jul 2018 17:06:38 +0000 (17:06 +0000)]
Unreviewed, check scope after performing getPropertySlot in JSON.stringify
https://bugs.webkit.org/show_bug.cgi?id=187807

Properly putting EXCEPTION_ASSERT to tell our exception checker mechanism
that we know that exception occurrence and handle it well.

* runtime/JSONObject.cpp:
(JSC::Stringifier::Holder::appendNextProperty):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@233987 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 months agoAdd an SPI policy action to allow clients to explicitly ask for a new process on...
beidson@apple.com [Thu, 19 Jul 2018 16:56:58 +0000 (16:56 +0000)]
Add an SPI policy action to allow clients to explicitly ask for a new process on a navigation.
https://bugs.webkit.org/show_bug.cgi?id=187789

Reviewed by Andy Estes.

Source/WebKit:

At navigation policy time, when a client says "use/allow", they can now say "use/allow in a new process if possible"

* UIProcess/API/C/WKFramePolicyListener.cpp:
(WKFramePolicyListenerUseInNewProcess):
(WKFramePolicyListenerUseInNewProcessWithPolicies):
* UIProcess/API/C/WKFramePolicyListener.h:

* UIProcess/API/Cocoa/WKNavigationDelegatePrivate.h:

* UIProcess/Cocoa/NavigationState.mm:
(WebKit::NavigationState::NavigationClient::decidePolicyForNavigationAction):

* UIProcess/WebFrameListenerProxy.h:
(WebKit::WebFrameListenerProxy::setApplyPolicyInNewProcessIfPossible):
(WebKit::WebFrameListenerProxy::applyPolicyInNewProcessIfPossible const):

* UIProcess/WebPageProxy.cpp:
(WebKit::WebPageProxy::receivedPolicyDecision):

* UIProcess/WebProcessPool.cpp:
(WebKit::WebProcessPool::processForNavigation):
(WebKit::WebProcessPool::processForNavigationInternal):
* UIProcess/WebProcessPool.h:

Tools:

* TestWebKitAPI/Tests/WebKitCocoa/ProcessSwapOnNavigation.mm:
(-[PSONNavigationDelegate init]):
(-[PSONNavigationDelegate webView:decidePolicyForNavigationAction:decisionHandler:]):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@233986 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 months ago[GStreamer] Return a valid time values in unprerolled states
cturner@igalia.com [Thu, 19 Jul 2018 14:51:15 +0000 (14:51 +0000)]
[GStreamer] Return a valid time values in unprerolled states
https://bugs.webkit.org/show_bug.cgi?id=187111

Reviewed by Xabier Rodriguez-Calvar.

After r230584 in bug 180253, asserts were added in
PlatformTimeRanges::add to check that both ends of the range were
valid times. In the non-MSE GStreamer player, this assert was
firing on https://www.w3.org/2010/05/video/mediaevents.html due to
seekable being called in nonprerolled states. In this case
MediaPlayerPrivateInterface::seekable was calling GStreamer's
maxTimeSeekable, which calls in durationMediaTime. The guard
against calling gst_element_query_duration when not prerolled was
returning a different time value than when the duration query
itself failed. Hence the assert firing.

* platform/graphics/gstreamer/MediaPlayerPrivateGStreamer.cpp:
(WebCore::MediaPlayerPrivateGStreamer::durationMediaTime const):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@233984 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 months agoEnsure experimentalPlugInSandboxProfilesEnabled is set on PluginProcess
youenn@apple.com [Thu, 19 Jul 2018 14:39:50 +0000 (14:39 +0000)]
Ensure experimentalPlugInSandboxProfilesEnabled is set on PluginProcess
https://bugs.webkit.org/show_bug.cgi?id=187729

Reviewed by Ryosuke Niwa.

experimentalPlugInSandboxProfilesEnabled flag is used at initialization of the plugin process sandbox.
This flag value should be set according to the value of this flag in the UIProcess.
We set this value in the plugin process manager.
At launch of the plugin process, this flag will also be passed to it so that it is set properly.

* PluginProcess/EntryPoint/mac/XPCService/PluginServiceEntryPoint.mm:
(WebKit::PluginServiceInitializerDelegate::getExtraInitializationData):
* PluginProcess/mac/PluginProcessMac.mm:
(WebKit::PluginProcess::platformInitializeProcess):
* Shared/EntryPointUtilities/mac/XPCService/XPCServiceEntryPoint.h:
* UIProcess/API/C/WKPreferences.cpp:
(WKPreferencesSetPluginSandboxProfilesEnabledForAllPlugins):
(WKPreferencesGetPluginSandboxProfilesEnabledForAllPlugins):
* UIProcess/API/C/WKPreferencesRefPrivate.h:
* UIProcess/API/Cocoa/WKPreferences.mm:
(-[WKPreferences _setExperimentalPlugInSandboxProfilesEnabled:]):
(-[WKPreferences _experimentalPlugInSandboxProfilesEnabled]):
* UIProcess/API/Cocoa/WKPreferencesPrivate.h:
* UIProcess/Plugins/PluginProcessManager.h:
(WebKit::PluginProcessManager::experimentalPlugInSandboxProfilesEnabled const):
* UIProcess/Plugins/mac/PluginProcessManagerMac.mm:
(WebKit::PluginProcessManager::setExperimentalPlugInSandboxProfilesEnabled):
* UIProcess/Plugins/mac/PluginProcessProxyMac.mm:
(WebKit::PluginProcessProxy::platformGetLaunchOptions):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@233983 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 months agoHitting RELEASE_ASSERT(!m_activeDOMObjectAdditionForbidden) under HTMLMediaElement...
commit-queue@webkit.org [Thu, 19 Jul 2018 13:38:16 +0000 (13:38 +0000)]
Hitting RELEASE_ASSERT(!m_activeDOMObjectAdditionForbidden) under HTMLMediaElement::resume()
https://bugs.webkit.org/show_bug.cgi?id=187793
<rdar://problem/42308469>

Patch by Antoine Quint <graouts@apple.com> on 2018-07-19
Reviewed by Chris Dumez.

Ensure we do not call JS under resume(), which would happen as a result of calling configureMediaControls() in prepareForLoad().

* html/HTMLMediaElement.cpp:
(WebCore::HTMLMediaElement::~HTMLMediaElement):
(WebCore::HTMLMediaElement::contextDestroyed):
(WebCore::HTMLMediaElement::stop):
(WebCore::HTMLMediaElement::suspend):
(WebCore::HTMLMediaElement::resume):
* html/HTMLMediaElement.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@233982 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 months ago[GStreamer][MSE] imported/w3c/web-platform-tests/media-source/mediasource-is-type...
philn@webkit.org [Thu, 19 Jul 2018 12:53:29 +0000 (12:53 +0000)]
[GStreamer][MSE] imported/w3c/web-platform-tests/media-source/mediasource-is-type-supported.html crashes
https://bugs.webkit.org/show_bug.cgi?id=187469

Reviewed by Žan Doberšek.

* platform/graphics/gstreamer/mse/WebKitMediaSourceGStreamer.cpp:
(webKitMediaSrcFreeStream): Fix critical warning. The appsrc
element is created only when a valid sourcebuffer is in use.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@233981 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 months agoFix message of NotSupportedError exception thrown during custom element creation
commit-queue@webkit.org [Thu, 19 Jul 2018 10:19:29 +0000 (10:19 +0000)]
Fix message of NotSupportedError exception thrown during custom element creation
https://bugs.webkit.org/show_bug.cgi?id=187757

Patch by Frederic Wang <fwang@igalia.com> on 2018-07-19
Reviewed by Yusuke Suzuki.

Source/WebCore:

In bug 161528, some new exceptions were introduced for custom element creation [1] but the
actual text has some issues. This patch fixes one typo and one wrong message.

[1] https://dom.spec.whatwg.org/#concept-create-element

Test: fast/custom-elements/exceptions-for-synchronous-custom-element-creation.html

* bindings/js/JSCustomElementInterface.cpp:
(WebCore::constructCustomElementSynchronously):

LayoutTests:

* fast/custom-elements/exceptions-for-synchronous-custom-element-creation-expected.txt: Added.
* fast/custom-elements/exceptions-for-synchronous-custom-element-creation.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@233980 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 months ago[Flatpak] Let flatpak process handle SIGINT
commit-queue@webkit.org [Thu, 19 Jul 2018 08:24:27 +0000 (08:24 +0000)]
[Flatpak] Let flatpak process handle SIGINT
https://bugs.webkit.org/show_bug.cgi?id=187521

Patch by Thibault Saunier <tsaunier@igalia.com> on 2018-07-19
Reviewed by Philippe Normand.

Ensuring that flatpak process ends properly and that the sandbox is teard down.
It also avoids showing usless backtrace to the end user and makes gdb
much more usable.

Also make flatpakutils not verbose by default

* flatpak/flatpakutils.py:
(disable_signals):
(WebkitFlatpak.run_in_sandbox):
(WebkitFlatpak.run_gdb):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@233979 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 months agoLet clients override _WKThumbnailView's background color
rmondello@apple.com [Thu, 19 Jul 2018 01:19:40 +0000 (01:19 +0000)]
Let clients override _WKThumbnailView's background color

https://bugs.webkit.org/show_bug.cgi?id=187788

Reviewed by Tim Horton.

* UIProcess/API/Cocoa/_WKThumbnailView.h: Declare a property.
* UIProcess/API/Cocoa/_WKThumbnailView.mm: Define an ivar.
(-[_WKThumbnailView updateLayer]): Consult the background color.
(-[_WKThumbnailView setOverrideBackgroundColor:]): Notably, call -updateLayer.
(-[_WKThumbnailView overrideBackgroundColor]): Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@233945 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 months agoUnreviewed. Fix contributors entry.
stephan.szabo@sony.com [Thu, 19 Jul 2018 01:01:25 +0000 (01:01 +0000)]
Unreviewed. Fix contributors entry.

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

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@233944 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 months agoProvide an lldb type summary for WebCore::Color
dino@apple.com [Thu, 19 Jul 2018 00:47:57 +0000 (00:47 +0000)]
Provide an lldb type summary for WebCore::Color
https://bugs.webkit.org/show_bug.cgi?id=187776

Reviewed by Dan Bates.

* lldb/lldbWebKitTester/lldbWebKitTester.xcodeproj/project.pbxproj: Link against WebKit
    to get to WebCore.
* lldb/lldbWebKitTester/main.cpp:
(testSummaryProviders): Create some Color objects for testing.

* lldb/lldb_webkit.py: Add a Color summary provider.
(__lldb_init_module):
(WebCoreColor_SummaryProvider):
(WebCoreColorProvider):
(WebCoreColorProvider.__init__):
(WebCoreColorProvider._is_extended):
(WebCoreColorProvider._is_valid):
(WebCoreColorProvider._is_semantic):
(WebCoreColorProvider._to_string_extended):
(WebCoreColorProvider.to_string):

* lldb/lldb_webkit_unittest.py: Tests.
(TestSummaryProviders.serial_test_WTFVectorProvider_vector_size_and_capacity):
(TestSummaryProviders):
(TestSummaryProviders.serial_test_WebCoreColorProvider_invalid_color):
(TestSummaryProviders.serial_test_WebCoreColorProvider_extended_color):
(TestSummaryProviders.serial_test_WebCoreColorProvider_rgb_color):
(TestSummaryProviders.serial_test_WebCoreColorProvider_rgba_color):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@233943 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 months agoUnreviewed, rolling out r233934.
tsavell@apple.com [Thu, 19 Jul 2018 00:38:27 +0000 (00:38 +0000)]
Unreviewed, rolling out r233934.

Revision caused 5 webkitpy failures on Mac and broke webkitpy
testing on iOS simulator

Reverted changeset:

"Provide an lldb type summary for WebCore::Color"
https://bugs.webkit.org/show_bug.cgi?id=187776
https://trac.webkit.org/changeset/233934

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@233942 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 months agoREGRESSION (r233897): CrashTracer: com.apple.WebKit.WebContent.Development at com...
cdumez@apple.com [Wed, 18 Jul 2018 23:33:22 +0000 (23:33 +0000)]
REGRESSION (r233897): CrashTracer: com.apple.WebKit.WebContent.Development at com.apple.WebKit: WKBooleanGetValue + 9
https://bugs.webkit.org/show_bug.cgi?id=187784
<rdar://problem/42329230>

Reviewed by Brady Eidson.

When process swapping on navigation, WebPageProxy::reattachToWebProcess() unregisters the page as a MessageReceiver
from the old WebProcessProxy and registers itself as a MessageReceiver for the new WebProcessProxy instead. This
means that after this point, IPC sent by the previous WebProcess to its WebPageProxy will fail.

When we process swap, we also navigate the page in the old WebProcess to about:blank, when the navigation to
about:blank would complete, the WebKitTestRunner's injected bundle would try and send IPC to the WebPageProxy
which would fail and would cause WKBundlePagePostSynchronousMessageForTesting() to return a null result. WKTR
would crash when dereferencing this null result. This patch addresses this by dealing with the potential null
result.

* WebKitTestRunner/InjectedBundle/InjectedBundle.cpp:
(WTR::InjectedBundle::shouldProcessWorkQueue const):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@233941 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 months agoCRASH at WebKit: WebKit::WebFullScreenManagerProxy::saveScrollPosition
jer.noble@apple.com [Wed, 18 Jul 2018 23:29:57 +0000 (23:29 +0000)]
CRASH at WebKit: WebKit::WebFullScreenManagerProxy::saveScrollPosition
https://bugs.webkit.org/show_bug.cgi?id=187769
<rdar://problem/42160666>

Reviewed by Tim Horton.

Null-check all uses of _page and _manager in WKFullScreenWindowControllerIOS.

* UIProcess/ios/fullscreen/WKFullScreenWindowControllerIOS.mm:
(WebKit::WKWebViewState::applyTo):
(WebKit::WKWebViewState::store):
(-[WKFullScreenWindowController enterFullScreen]):
(-[WKFullScreenWindowController beganExitFullScreenWithInitialFrame:finalFrame:]):
(-[WKFullScreenWindowController _completedExitFullScreen]):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@233940 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 months agoWebContent crash in WebProcess::ensureNetworkProcessConnection
cdumez@apple.com [Wed, 18 Jul 2018 23:13:36 +0000 (23:13 +0000)]
WebContent crash in WebProcess::ensureNetworkProcessConnection
https://bugs.webkit.org/show_bug.cgi?id=187791
<rdar://problem/41995022>

Reviewed by Ryosuke Niwa.

If the WebProcessProxy::GetNetworkProcessConnection synchronous IPC between the WebProcess
and the UIProcess succeeded but we received an invalid connection identifier, then try
once more. This may indicate the network process has crashed, in which case it will be
relaunched.

* WebProcess/WebProcess.cpp:
(WebKit::getNetworkProcessConnection):
(WebKit::WebProcess::ensureNetworkProcessConnection):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@233939 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 months agoRebaseline fast/css/apple-system-colors.html.
ryanhaddad@apple.com [Wed, 18 Jul 2018 22:41:48 +0000 (22:41 +0000)]
Rebaseline fast/css/apple-system-colors.html.

Unreviewed test gardening.

* platform/mac/fast/css/apple-system-colors-expected.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@233938 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 months ago[JSC] Reduce size of AST nodes
utatane.tea@gmail.com [Wed, 18 Jul 2018 22:36:34 +0000 (22:36 +0000)]
[JSC] Reduce size of AST nodes
https://bugs.webkit.org/show_bug.cgi?id=187689

Reviewed by Mark Lam.

We clean up AST nodes to reduce size. By doing so, we can reduce the memory consumption
of ParserArena at peak state.

1. Annotate `final` to AST nodes to make them solid. And it allows the compiler to
devirtualize a call to the function which are implemented in a final class.

2. Use default member initializers more.

3. And use `nullptr` instead of `0`.

4. Arrange the layout of AST nodes to reduce the size. It includes changing the order
of classes in multiple inheritance. In particular, StatementNode is decreased from 48
to 40. This decreases the sizes of all the derived Statement nodes.

* parser/NodeConstructors.h:
(JSC::Node::Node):
(JSC::StatementNode::StatementNode):
(JSC::ElementNode::ElementNode):
(JSC::ArrayNode::ArrayNode):
(JSC::PropertyListNode::PropertyListNode):
(JSC::ObjectLiteralNode::ObjectLiteralNode):
(JSC::ArgumentListNode::ArgumentListNode):
(JSC::ArgumentsNode::ArgumentsNode):
(JSC::NewExprNode::NewExprNode):
(JSC::BytecodeIntrinsicNode::BytecodeIntrinsicNode):
(JSC::BinaryOpNode::BinaryOpNode):
(JSC::LogicalOpNode::LogicalOpNode):
(JSC::CommaNode::CommaNode):
(JSC::SourceElements::SourceElements):
(JSC::ClauseListNode::ClauseListNode):
* parser/Nodes.cpp:
(JSC::FunctionMetadataNode::FunctionMetadataNode):
(JSC::FunctionMetadataNode::operator== const):
(JSC::FunctionMetadataNode::dump const):
* parser/Nodes.h:
(JSC::BooleanNode::value): Deleted.
(JSC::StringNode::value): Deleted.
(JSC::TemplateExpressionListNode::value): Deleted.
(JSC::TemplateExpressionListNode::next): Deleted.
(JSC::TemplateStringNode::cooked): Deleted.
(JSC::TemplateStringNode::raw): Deleted.
(JSC::TemplateStringListNode::value): Deleted.
(JSC::TemplateStringListNode::next): Deleted.
(JSC::TemplateLiteralNode::templateStrings const): Deleted.
(JSC::TemplateLiteralNode::templateExpressions const): Deleted.
(JSC::TaggedTemplateNode::templateLiteral const): Deleted.
(JSC::ResolveNode::identifier const): Deleted.
(JSC::ElementNode::elision const): Deleted.
(JSC::ElementNode::value): Deleted.
(JSC::ElementNode::next): Deleted.
(JSC::ArrayNode::elements const): Deleted.
(JSC::PropertyNode::expressionName const): Deleted.
(JSC::PropertyNode::name const): Deleted.
(JSC::PropertyNode::type const): Deleted.
(JSC::PropertyNode::needsSuperBinding const): Deleted.
(JSC::PropertyNode::isClassProperty const): Deleted.
(JSC::PropertyNode::isStaticClassProperty const): Deleted.
(JSC::PropertyNode::isInstanceClassProperty const): Deleted.
(JSC::PropertyNode::isOverriddenByDuplicate const): Deleted.
(JSC::PropertyNode::setIsOverriddenByDuplicate): Deleted.
(JSC::PropertyNode::putType const): Deleted.
(JSC::BracketAccessorNode::base const): Deleted.
(JSC::BracketAccessorNode::subscript const): Deleted.
(JSC::BracketAccessorNode::subscriptHasAssignments const): Deleted.
(JSC::DotAccessorNode::base const): Deleted.
(JSC::DotAccessorNode::identifier const): Deleted.
(JSC::SpreadExpressionNode::expression const): Deleted.
(JSC::ObjectSpreadExpressionNode::expression const): Deleted.
(JSC::BytecodeIntrinsicNode::type const): Deleted.
(JSC::BytecodeIntrinsicNode::emitter const): Deleted.
(JSC::BytecodeIntrinsicNode::identifier const): Deleted.
(JSC::TypeOfResolveNode::identifier const): Deleted.
(JSC::BitwiseNotNode::expr): Deleted.
(JSC::BitwiseNotNode::expr const): Deleted.
(JSC::AssignResolveNode::identifier const): Deleted.
(JSC::ExprStatementNode::expr const): Deleted.
(JSC::ForOfNode::isForAwait const): Deleted.
(JSC::ReturnNode::value): Deleted.
(JSC::ProgramNode::startColumn const): Deleted.
(JSC::ProgramNode::endColumn const): Deleted.
(JSC::EvalNode::startColumn const): Deleted.
(JSC::EvalNode::endColumn const): Deleted.
(JSC::ModuleProgramNode::startColumn const): Deleted.
(JSC::ModuleProgramNode::endColumn const): Deleted.
(JSC::ModuleProgramNode::moduleScopeData): Deleted.
(JSC::ModuleNameNode::moduleName): Deleted.
(JSC::ImportSpecifierNode::importedName): Deleted.
(JSC::ImportSpecifierNode::localName): Deleted.
(JSC::ImportSpecifierListNode::specifiers const): Deleted.
(JSC::ImportSpecifierListNode::append): Deleted.
(JSC::ImportDeclarationNode::specifierList const): Deleted.
(JSC::ImportDeclarationNode::moduleName const): Deleted.
(JSC::ExportAllDeclarationNode::moduleName const): Deleted.
(JSC::ExportDefaultDeclarationNode::declaration const): Deleted.
(JSC::ExportDefaultDeclarationNode::localName const): Deleted.
(JSC::ExportLocalDeclarationNode::declaration const): Deleted.
(JSC::ExportSpecifierNode::exportedName): Deleted.
(JSC::ExportSpecifierNode::localName): Deleted.
(JSC::ExportSpecifierListNode::specifiers const): Deleted.
(JSC::ExportSpecifierListNode::append): Deleted.
(JSC::ExportNamedDeclarationNode::specifierList const): Deleted.
(JSC::ExportNamedDeclarationNode::moduleName const): Deleted.
(JSC::ArrayPatternNode::appendIndex): Deleted.
(JSC::ObjectPatternNode::appendEntry): Deleted.
(JSC::ObjectPatternNode::setContainsRestElement): Deleted.
(JSC::ObjectPatternNode::setContainsComputedProperty): Deleted.
(JSC::DestructuringAssignmentNode::bindings): Deleted.
(JSC::FunctionParameters::size const): Deleted.
(JSC::FunctionParameters::append): Deleted.
(JSC::FunctionParameters::isSimpleParameterList const): Deleted.
(JSC::FuncDeclNode::metadata): Deleted.
(JSC::CaseClauseNode::expr const): Deleted.
(JSC::CaseClauseNode::setStartOffset): Deleted.
(JSC::ClauseListNode::getClause const): Deleted.
(JSC::ClauseListNode::getNext const): Deleted.
* runtime/ExceptionHelpers.cpp:
* runtime/JSObject.cpp:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@233937 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 months agoUnreviewed API Test fix; restored a line inadventantly removed in r233926.
jer.noble@apple.com [Wed, 18 Jul 2018 22:29:04 +0000 (22:29 +0000)]
Unreviewed API Test fix; restored a line inadventantly removed in r233926.

* platform/mac/VideoFullscreenInterfaceMac.mm:
(-[WebVideoFullscreenInterfaceMacObjC pipDidClose:]):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@233936 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 months ago[Web Animations] Interpolation between font-styles with a keyword value should be...
graouts@webkit.org [Wed, 18 Jul 2018 22:23:25 +0000 (22:23 +0000)]
[Web Animations] Interpolation between font-styles with a keyword value should be discrete
https://bugs.webkit.org/show_bug.cgi?id=187722

Reviewed by Myles Maxfield.

LayoutTests/imported/w3c:

Mark some WPT progressions.

* web-platform-tests/web-animations/animation-model/animation-types/discrete-expected.txt:

Source/WebCore:

Animating between "font-style: normal" or "font-style: oblique" and any another value should yield a discrete
interpolation where the from-value is used from 0 and up to (but excluding) 0.5, and the to-value from 0.5 to 1.

In order to be able to detect the "normal" value, we make the "slope" of a FontSelectionRequest an optional type
where the std::nullopt value indicates "normal" and other values an "oblique" value. Since we also need to
distinguish the "italic" value from an "oblique" value, we implement a custom PropertyWrapper for the "font-style"
property where we ensure the fontStyleAxis property of the font description matches the value we're blending to.
Indeed, in the case where we may animate from "normal" to "italic", the fontStyleAxis on the blended style would
remain "slnt" since it is the base value for "normal".

* accessibility/AccessibilityRenderObject.cpp:
(WebCore::AccessibilityRenderObject::hasPlainText const):
* css/CSSComputedStyleDeclaration.cpp:
(WebCore::ComputedStyleExtractor::fontStyleFromStyleValue):
* css/CSSComputedStyleDeclaration.h:
* css/CSSFontFace.cpp:
(WebCore::calculateItalicRange):
* css/CSSFontFaceSet.cpp:
(WebCore::computeFontSelectionRequest):
* css/FontSelectionValueInlines.h:
(WebCore::fontStyleKeyword):
(WebCore::fontStyleValue): Deleted.
* css/StyleBuilderConverter.h:
(WebCore::StyleBuilderConverter::convertFontStyleFromValue):
* page/animation/CSSPropertyAnimation.cpp:
(WebCore::blendFunc):
(WebCore::PropertyWrapperFontStyle::PropertyWrapperFontStyle):
(WebCore::CSSPropertyAnimationWrapperMap::CSSPropertyAnimationWrapperMap):
* platform/graphics/FontCache.h:
(WebCore::FontDescriptionKey::computeHash const):
* platform/graphics/FontCascade.h:
(WebCore::FontCascade::italic const):
* platform/graphics/FontDescription.h:
(WebCore::FontDescription::italic const):
(WebCore::FontDescription::setItalic):
(WebCore::FontDescription::setIsItalic):
(WebCore::FontCascadeDescription::initialItalic):
* platform/graphics/FontSelectionAlgorithm.cpp:
(WebCore::FontSelectionAlgorithm::styleDistance const):
* platform/graphics/FontSelectionAlgorithm.h:
(WebCore::isItalic):
(WebCore::FontSelectionRequest::tied const):
(WebCore::operator<<): Implement the required stream operator.
(WebCore::operator==): Mark this function as inline instead of constexpr since tied() is no longer constexpr
due to taking an std::optional<>.
(WebCore::operator!=):
* platform/graphics/cocoa/FontCacheCoreText.cpp:
(WebCore::preparePlatformFont):
* platform/graphics/win/FontCacheWin.cpp:
(WebCore::FontCache::createFontPlatformData):
* rendering/style/RenderStyle.cpp:
(WebCore::RenderStyle::setFontItalic):
* rendering/style/RenderStyle.h:
(WebCore::RenderStyle::fontItalic const):

Source/WebKitLegacy/win:

Use isItalic() since that function knows how to handle an std::optional<FontSelectionValue>.

* DOMCoreClasses.cpp:
(DOMElement::font):

LayoutTests:

Mark some WPT progressions.

* platform/mac/imported/w3c/web-platform-tests/web-animations/animation-model/animation-types/accumulation-per-property-expected.txt:
* platform/mac/imported/w3c/web-platform-tests/web-animations/animation-model/animation-types/addition-per-property-expected.txt:
* platform/mac/imported/w3c/web-platform-tests/web-animations/animation-model/animation-types/interpolation-per-property-expected.txt:
* platform/mac-sierra/imported/w3c/web-platform-tests/web-animations/animation-model/animation-types/accumulation-per-property-expected.txt:
* platform/mac-sierra/imported/w3c/web-platform-tests/web-animations/animation-model/animation-types/addition-per-property-expected.txt:
* platform/mac-sierra/imported/w3c/web-platform-tests/web-animations/animation-model/animation-types/interpolation-per-property-expected.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@233935 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 months agoProvide an lldb type summary for WebCore::Color
dino@apple.com [Wed, 18 Jul 2018 22:19:46 +0000 (22:19 +0000)]
Provide an lldb type summary for WebCore::Color
https://bugs.webkit.org/show_bug.cgi?id=187776

Reviewed by Dan Bates.

* lldb/lldbWebKitTester/lldbWebKitTester.xcodeproj/project.pbxproj: Link against WebKit
    to get to WebCore.
* lldb/lldbWebKitTester/main.cpp:
(testSummaryProviders): Create some Color objects for testing.

* lldb/lldb_webkit.py: Add a Color summary provider.
(__lldb_init_module):
(WebCoreColor_SummaryProvider):
(WebCoreColorProvider):
(WebCoreColorProvider.__init__):
(WebCoreColorProvider._is_extended):
(WebCoreColorProvider._is_valid):
(WebCoreColorProvider._is_semantic):
(WebCoreColorProvider._to_string_extended):
(WebCoreColorProvider.to_string):

* lldb/lldb_webkit_unittest.py: Tests.
(TestSummaryProviders.serial_test_WTFVectorProvider_vector_size_and_capacity):
(TestSummaryProviders):
(TestSummaryProviders.serial_test_WebCoreColorProvider_invalid_color):
(TestSummaryProviders.serial_test_WebCoreColorProvider_extended_color):
(TestSummaryProviders.serial_test_WebCoreColorProvider_rgb_color):
(TestSummaryProviders.serial_test_WebCoreColorProvider_rgba_color):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@233934 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 months agoThe WebContent process does not suspend when MiniBrowser is minimized.
pvollan@apple.com [Wed, 18 Jul 2018 21:50:00 +0000 (21:50 +0000)]
The WebContent process does not suspend when MiniBrowser is minimized.
https://bugs.webkit.org/show_bug.cgi?id=187708

Reviewed by Chris Dumez.

Source/WebCore/PAL:

Add function for enabling App nap.

* pal/spi/cf/CFUtilitiesSPI.h:

Source/WebKit:

Using the NSRunLoop runloop type prevents the WebContent process from suspending. Instead, use the new
_WebKit runloop type. Also do not leak a boost to the WebContent process, since this also prevents the
WebContent process from going to sleep. Calling SetApplicationIsDaemon prevents the WebContent process
from being assigned the application process priority level. To block WindowServer connections, call
CGSSetDenyWindowServerConnections(true) instead. Finally, App nap must be manually enabled, since the
WebContent process is no longer a NSApplication.

* Configurations/WebContentService.xcconfig:
* UIProcess/Launcher/mac/ProcessLauncherMac.mm:
(WebKit::shouldLeakBoost):
* WebProcess/WebPage/WebPage.cpp:
(WebKit::WebPage::updateThrottleState):
* WebProcess/cocoa/WebProcessCocoa.mm:
(WebKit::WebProcess::platformInitializeProcess):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@233932 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 months agoAdding myself to contributors.json
stephan.szabo@sony.com [Wed, 18 Jul 2018 21:21:07 +0000 (21:21 +0000)]
Adding myself to contributors.json

Unreviewed contributors.json change.

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

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@233931 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 months agoUnreviewed build fix after r233926; BOOL !== bool.
jer.noble@apple.com [Wed, 18 Jul 2018 21:17:07 +0000 (21:17 +0000)]
Unreviewed build fix after r233926; BOOL !== bool.

* platform/ios/VideoFullscreenInterfaceAVKit.mm:
(VideoFullscreenInterfaceAVKit::pictureInPictureWasStartedWhenEnteringBackground const):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@233930 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 months agoCrash under WebKit::SuspendedPageProxy::webProcessDidClose(WebKit::WebProcessProxy&)
cdumez@apple.com [Wed, 18 Jul 2018 21:15:41 +0000 (21:15 +0000)]
Crash under WebKit::SuspendedPageProxy::webProcessDidClose(WebKit::WebProcessProxy&)
https://bugs.webkit.org/show_bug.cgi?id=187780

Reviewed by Brady Eidson.

Protect |this| in SuspendedPageProxy::webProcessDidClose() since the call to
WebPageProxy::suspendedPageClosed() may destroy us before the method is done
executing.

* UIProcess/SuspendedPageProxy.cpp:
(WebKit::SuspendedPageProxy::webProcessDidClose):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@233929 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 months ago-_beginAnimatedResizeWithUpdates: can leave view in bad state if called during an...
jer.noble@apple.com [Wed, 18 Jul 2018 19:51:54 +0000 (19:51 +0000)]
-_beginAnimatedResizeWithUpdates: can leave view in bad state if called during an existing animation
https://bugs.webkit.org/show_bug.cgi?id=187739
Source/WebKit:

Reviewed by Tim Horton.

It's not enough to reset _dynamicViewportUpdateMode to NotResizing; other parts of the code
check whether _resizeAnimationView is non-nil, the contentView may be hidden, etc. Add a new
internal method _cancelAnimatedResize that cleans up state when a call to
_beginAnimatedResizeWithUpdates: fails.

* UIProcess/API/Cocoa/WKWebView.mm:
(-[WKWebView _processDidExit]):
(-[WKWebView _cancelAnimatedResize]):
(-[WKWebView _didCompleteAnimatedResize]):
(-[WKWebView _beginAnimatedResizeWithUpdates:]):

Tools:

<rdar://problem/42304518>

Reviewed by Tim Horton.

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

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@233927 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 months agoPiP from Element Fullscreen should match AVKit's behavior
jer.noble@apple.com [Wed, 18 Jul 2018 19:50:04 +0000 (19:50 +0000)]
PiP from Element Fullscreen should match AVKit's behavior
https://bugs.webkit.org/show_bug.cgi?id=187623
Source/WebCore:

Reviewed by Jon Lee.

PiP behavior should be defined at the WebKit2 level, and not in HTMLMediaElement:

* html/HTMLMediaElement.cpp:
(WebCore::HTMLMediaElement::enterFullscreen):

Add an accessor for pictureInPictureWasStartedWhenEnteringBackground():

* platform/cocoa/VideoFullscreenModelVideoElement.mm:
(VideoFullscreenInterfaceAVKit::pictureInPictureWasStartedWhenEnteringBackground const):

Add VideoFullscreenModelClient virutal methods for PiP change notifications:

* platform/cocoa/VideoFullscreenModel.h:
(WebCore::VideoFullscreenModelClient::hasVideoChanged):
(WebCore::VideoFullscreenModelClient::videoDimensionsChanged):
(WebCore::VideoFullscreenModelClient::willEnterPictureInPicture):
(WebCore::VideoFullscreenModelClient::didEnterPictureInPicture):
(WebCore::VideoFullscreenModelClient::failedToEnterPictureInPicture):
(WebCore::VideoFullscreenModelClient::willExitPictureInPicture):
(WebCore::VideoFullscreenModelClient::didExitPictureInPicture):
(WebCore::VideoFullscreenModelClient::failedToExitPictureInPicture):
* platform/cocoa/VideoFullscreenModelVideoElement.h:
* platform/cocoa/VideoFullscreenModelVideoElement.mm:
(VideoFullscreenModelVideoElement::willEnterPictureInPicture):
(VideoFullscreenModelVideoElement::didEnterPictureInPicture):
(VideoFullscreenModelVideoElement::failedToEnterPictureInPicture):
(VideoFullscreenModelVideoElement::willExitPictureInPicture):
(VideoFullscreenModelVideoElement::didExitPictureInPicture):
(VideoFullscreenModelVideoElement::failedToExitPictureInPicture):
* platform/ios/VideoFullscreenInterfaceAVKit.h:
* platform/ios/VideoFullscreenInterfaceAVKit.mm:
(-[WebAVPlayerLayer layoutSublayers]):
(-[WebAVPlayerLayer resolveBounds]):
(-[WebAVPlayerLayer setVideoGravity:]):
(VideoFullscreenInterfaceAVKit::setupFullscreen):
(VideoFullscreenInterfaceAVKit::presentingViewController):
(VideoFullscreenInterfaceAVKit::willStartPictureInPicture):
(VideoFullscreenInterfaceAVKit::didStartPictureInPicture):
(VideoFullscreenInterfaceAVKit::failedToStartPictureInPicture):
(VideoFullscreenInterfaceAVKit::willStopPictureInPicture):
(VideoFullscreenInterfaceAVKit::didStopPictureInPicture):
(VideoFullscreenInterfaceAVKit::shouldExitFullscreenWithReason):
(VideoFullscreenInterfaceAVKit::doSetup):
* platform/ios/WebVideoFullscreenControllerAVKit.mm:
(VideoFullscreenControllerContext::willEnterPictureInPicture):
(VideoFullscreenControllerContext::didEnterPictureInPicture):
(VideoFullscreenControllerContext::failedToEnterPictureInPicture):
(VideoFullscreenControllerContext::willExitPictureInPicture):
(VideoFullscreenControllerContext::didExitPictureInPicture):
(VideoFullscreenControllerContext::failedToExitPictureInPicture):
* platform/mac/VideoFullscreenInterfaceMac.h:
(WebCore::VideoFullscreenInterfaceMac::requestHideAndExitFullscreen): Deleted.
* platform/mac/VideoFullscreenInterfaceMac.mm:
(-[WebVideoFullscreenInterfaceMacObjC invalidateFullscreenState]):
(-[WebVideoFullscreenInterfaceMacObjC exitPIP]):
(-[WebVideoFullscreenInterfaceMacObjC exitPIPAnimatingToRect:inWindow:]):
(-[WebVideoFullscreenInterfaceMacObjC pipShouldClose:]):
(-[WebVideoFullscreenInterfaceMacObjC pipDidClose:]):
(WebCore::VideoFullscreenInterfaceMac::enterFullscreen):
(WebCore::VideoFullscreenInterfaceMac::exitFullscreen):
(WebCore::VideoFullscreenInterfaceMac::exitFullscreenWithoutAnimationToMode):
(WebCore::VideoFullscreenInterfaceMac::requestHideAndExitFullscreen):

Source/WebCore/PAL:

Reviewed by Jon Lee.

* pal/spi/mac/PIPSPI.h:

Source/WebKit:

Reviewed by Jon Lee.

* UIProcess/Cocoa/PlaybackSessionManagerProxy.h:
(WebKit::PlaybackSessionManagerProxy::controlsManagerContextId const):
* UIProcess/Cocoa/VideoFullscreenManagerProxy.h:
* UIProcess/Cocoa/VideoFullscreenManagerProxy.mm:
(WebKit::VideoFullscreenModelContext::willEnterPictureInPicture):
(WebKit::VideoFullscreenModelContext::didEnterPictureInPicture):
(WebKit::VideoFullscreenModelContext::failedToEnterPictureInPicture):
(WebKit::VideoFullscreenModelContext::willExitPictureInPicture):
(WebKit::VideoFullscreenModelContext::didExitPictureInPicture):
(WebKit::VideoFullscreenModelContext::failedToExitPictureInPicture):
(WebKit::VideoFullscreenManagerProxy::controlsManagerInterface):
* UIProcess/ios/fullscreen/WKFullScreenViewController.mm:
(WKFullScreenViewControllerVideoFullscreenModelClient::setParent):
(WKFullScreenViewControllerVideoFullscreenModelClient::setInterface):
(WKFullScreenViewControllerVideoFullscreenModelClient::interface const):
(-[WKFullScreenViewController initWithWebView:]):
(-[WKFullScreenViewController dealloc]):
(-[WKFullScreenViewController videoControlsManagerDidChange]):
(-[WKFullScreenViewController ensurePiPAnimator]):
(-[WKFullScreenViewController willEnterPictureInPicture]):
(-[WKFullScreenViewController didEnterPictureInPicture]):
(-[WKFullScreenViewController failedToEnterPictureInPicture]):
(-[WKFullScreenViewController loadView]):
(-[WKFullScreenViewController viewWillAppear:]):
* UIProcess/Cocoa/WebViewImpl.mm:
(WebKit::WebViewImpl::videoControlsManagerDidChange):
* UIProcess/mac/WKFullScreenWindowController.h:
* UIProcess/mac/WKFullScreenWindowController.mm:
(WebKit::WKFullScreenWindowControllerVideoFullscreenModelClient::setParent):
(WebKit::WKFullScreenWindowControllerVideoFullscreenModelClient::setInterface):
(WebKit::WKFullScreenWindowControllerVideoFullscreenModelClient::interface const):
(-[WKFullScreenWindowController initWithWindow:webView:page:]):
(-[WKFullScreenWindowController dealloc]):
(-[WKFullScreenWindowController videoControlsManagerDidChange]):
(-[WKFullScreenWindowController willEnterPictureInPicture]):

Tools:

<rdar://problem/41212379>

Reviewed by Jon Lee.

* TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj:
* TestWebKitAPI/Tests/WebKitCocoa/ExitFullscreenOnEnterPiP.html: Added.
* TestWebKitAPI/Tests/WebKitCocoa/ExitFullscreenOnEnterPiP.mm: Added.
(-[ExitFullscreenOnEnterPiPUIDelegate _webView:hasVideoInPictureInPictureDidChange:]):
(-[ExitFullscreenOnEnterPiPUIDelegate _webViewDidEnterFullscreen:]):
(-[ExitFullscreenOnEnterPiPUIDelegate _webViewDidExitFullscreen:]):
(TestWebKitAPI::TEST):

LayoutTests:

Reviewed by Jon Lee.

* TestExpectations:
* media/fullscreen-video-going-into-pip-expected.txt: Removed.
* media/fullscreen-video-going-into-pip.html: Removed.
* media/video-contained-in-fullscreen-element-going-into-pip-expected.txt: Removed.
* media/video-contained-in-fullscreen-element-going-into-pip.html: Removed.
* platform/mac-wk2/TestExpectations:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@233926 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 months agoDissociate the VideoFullscreenInterface from its VideoFullscreenModel before removing...
jer.noble@apple.com [Wed, 18 Jul 2018 19:44:32 +0000 (19:44 +0000)]
Dissociate the VideoFullscreenInterface from its VideoFullscreenModel before removing it from the manager
https://bugs.webkit.org/show_bug.cgi?id=187775
<rdar://problem/42343229>

Reviewed by Jon Lee.

* UIProcess/Cocoa/VideoFullscreenManagerProxy.mm:
(WebKit::VideoFullscreenManagerProxy::didCleanupFullscreen):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@233925 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 months agoJSON.stringify should emit non own properties if second array argument includes
utatane.tea@gmail.com [Wed, 18 Jul 2018 19:01:34 +0000 (19:01 +0000)]
JSON.stringify should emit non own properties if second array argument includes
https://bugs.webkit.org/show_bug.cgi?id=187724

Reviewed by Mark Lam.

JSTests:

* stress/json-stringify-getter-call.js: Added.
(shouldBe):
(A.prototype.get cocoa):
(A.prototype.get cappuccino):
(A):
(shouldBe.JSON.stringify):

Source/JavaScriptCore:

According to the spec[1], JSON.stringify needs to retrieve properties by using [[Get]],
instead of [[GetOwnProperty]]. It means that we would look up a properties defined
in [[Prototype]] or upper objects in the prototype chain. While enumeration is done
by using EnumerableOwnPropertyNames typically, we can pass replacer array including
property names which does not reside in the own properties. Or we can modify the
own properties by deleting properties while JSON.stringify is calling a getter. So,
using [[Get]] instead of [[GetOwnProperty]] is user-visible.

This patch changes getOwnPropertySlot to getPropertySlot to align the behavior to the spec.
The performance of Kraken/json-stringify-tinderbox is neutral.

[1]: https://tc39.github.io/ecma262/#sec-serializejsonproperty

* runtime/JSONObject.cpp:
(JSC::Stringifier::toJSON):
(JSC::Stringifier::toJSONImpl):
(JSC::Stringifier::appendStringifiedValue):
(JSC::Stringifier::Holder::Holder):
(JSC::Stringifier::Holder::appendNextProperty):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@233924 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 months ago [ WK2 ] Layout Test http/wpt/service-workers/update-service-worker.https.html...
tsavell@apple.com [Wed, 18 Jul 2018 19:00:26 +0000 (19:00 +0000)]
   [ WK2 ] Layout Test http/wpt/service-workers/update-service-worker.https.html is a flaky Timeout
https://bugs.webkit.org/show_bug.cgi?id=187766

Unreviewed test gardening.

* platform/wk2/TestExpectations:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@233923 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 months ago[GTK] Unreviewed test gardening
aboya@igalia.com [Wed, 18 Jul 2018 18:45:42 +0000 (18:45 +0000)]
[GTK] Unreviewed test gardening
https://bugs.webkit.org/show_bug.cgi?id=187771

* TestExpectations:
* platform/gtk/TestExpectations:
* platform/wpe/TestExpectations:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@233922 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 months ago[ews-build] Add build step to run layout-test
aakash_jain@apple.com [Wed, 18 Jul 2018 18:40:20 +0000 (18:40 +0000)]
[ews-build] Add build step to run layout-test
https://bugs.webkit.org/show_bug.cgi?id=187674

Reviewed by Lucas Forschler.

* BuildSlaveSupport/ews-build/steps.py:
(RunWebKitTests): Added build step to run layout-tests. This is similar to current EWS.
* BuildSlaveSupport/ews-build/steps_unittest.py: Added unit-tests.
* BuildSlaveSupport/ews-build/factories.py: Added build-step to run layout-test in iOSSimulatorFactory.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@233921 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 months agoWeb Inspector: REGRESSION(r232591): CodeMirrorEditor should not use a RegExp lineSepa...
mattbaker@apple.com [Wed, 18 Jul 2018 18:40:14 +0000 (18:40 +0000)]
Web Inspector: REGRESSION(r232591): CodeMirrorEditor should not use a RegExp lineSeparator option
https://bugs.webkit.org/show_bug.cgi?id=187772
<rdar://problem/42331640>

Reviewed by Joseph Pecoraro.

* UserInterface/Views/CodeMirrorEditor.js:
(WI.CodeMirrorEditor.create):
(WI.CodeMirrorEditor):
CodeMirror should be left to auto-detect line separators. By default
it detects \n, \r\n, and \r. By specifying a regular expression we
merely cause problems when CodeMirror uses the supplied lineSeparator
when joining its array of lines together.

* UserInterface/Views/TextEditor.js:
(WI.TextEditor.set string.update):
(WI.TextEditor.prototype.set string):
This assertion was only true when we forced "\n" line endings everywhere.
It no longer holds for source text with "\r\n" (Windows-style) line endings.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@233920 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 months ago[Curl] Disable CURLOPT_SSL_VERIFYPEER and CURLOPT_SSL_VERIFYHOST specified by setAllo...
Basuke.Suzuki@sony.com [Wed, 18 Jul 2018 18:35:16 +0000 (18:35 +0000)]
[Curl] Disable CURLOPT_SSL_VERIFYPEER and CURLOPT_SSL_VERIFYHOST specified by setAllowsAnyHTTPSCertificate.
https://bugs.webkit.org/show_bug.cgi?id=187611

Reviewed by Fujii Hironori.

Current interface for TLS certificate validation for Curl port are as follows:

- WEBCORE_EXPORT void setHostAllowsAnyHTTPSCertificate(const String&);
- bool isAllowedHTTPSCertificateHost(const String&);
- bool canIgnoredHTTPSCertificate(const String&, const Vector<CertificateInfo::Certificate>&);

First one registers a host to be ignored for any certificate check. Once it is registered, no
further certificate validation check is executed.
Second one checks the host is registered in the list above.
Third one is weird. The method signature implies it checks the certificate for the host and detect
whether we can ignore this certificate for the host, but actually it  just check only the host and
register the certificate into the vector. Then in the next request for the host, the certificate
will be checked with the previously stored certificate.

It's hard to understand, but in short,
- We can register a host as an exception for any TLS certificate validation.
- But only certificate arrived first is ignored, not any certificates for the host
  (which is rare, but possible for mis configured web cluster).

This behavior is incomplete. To ignore any certificates of the host, these two methods are enough:

- void allowAnyHTTPSCertificatesForHost(const String&)
- bool canIgnoreAnyHTTPSCertificatesForHost(const String&)

No new tests. Covered by existing tests.

* platform/network/curl/CertificateInfo.h:
* platform/network/curl/CurlContext.cpp:
(WebCore::CurlHandle::enableSSLForHost): Ignore TLS verification for registered host.
* platform/network/curl/CurlSSLHandle.cpp:
(WebCore::CurlSSLHandle::allowAnyHTTPSCertificatesForHost): Added.
(WebCore::CurlSSLHandle::canIgnoreAnyHTTPSCertificatesForHost const): Ditto.
(WebCore::CurlSSLHandle::setClientCertificateInfo): Separate lock.
(WebCore::CurlSSLHandle::getSSLClientCertificate const): Ditto and add const.
(WebCore::CurlSSLHandle::setHostAllowsAnyHTTPSCertificate): Deleted.
(WebCore::CurlSSLHandle::isAllowedHTTPSCertificateHost): Deleted.
(WebCore::CurlSSLHandle::canIgnoredHTTPSCertificate): Deleted.
(WebCore::CurlSSLHandle::getSSLClientCertificate): Deleted.
* platform/network/curl/CurlSSLHandle.h:
* platform/network/curl/CurlSSLVerifier.cpp:
(WebCore::CurlSSLVerifier::CurlSSLVerifier):
(WebCore::CurlSSLVerifier::collectInfo): Renamed from verify.
(WebCore::CurlSSLVerifier::verifyCallback):
(WebCore::CurlSSLVerifier::verify): Renamed to collectInfo.
* platform/network/curl/CurlSSLVerifier.h:
* platform/network/curl/ResourceHandleCurl.cpp:
(WebCore::ResourceHandle::setHostAllowsAnyHTTPSCertificate): Rename calling method.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@233919 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 months ago[JSC] JSON.stringify's replacer should use `isArray` instead of JSArray checks
utatane.tea@gmail.com [Wed, 18 Jul 2018 18:31:09 +0000 (18:31 +0000)]
[JSC] JSON.stringify's replacer should use `isArray` instead of JSArray checks
https://bugs.webkit.org/show_bug.cgi?id=187755

Reviewed by Mark Lam.

JSTests:

* stress/json-stringify-gap-calculation-should-be-after-replacer-check.js: Added.
(shouldThrow):
(shouldThrow.string.toString):
* test262/expectations.yaml:

Source/JavaScriptCore:

JSON.stringify used `inherits<JSArray>(vm)` to determine whether the given replacer is an array replacer.
But this is wrong. According to the spec, we should use `isArray`[1], which accepts Proxies. This difference
makes one test262 test failed.

This patch changes the code to using `isArray()`. And we reorder the evaluations of replacer check and ident space check
to align these checks to the spec's order.

[1]: https://tc39.github.io/ecma262/#sec-json.stringify

* runtime/JSONObject.cpp:
(JSC::Stringifier::Stringifier):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@233918 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 months ago[JSC] Root wrapper object in JSON.stringify is not necessary if replacer is not callable
utatane.tea@gmail.com [Wed, 18 Jul 2018 18:27:08 +0000 (18:27 +0000)]
[JSC] Root wrapper object in JSON.stringify is not necessary if replacer is not callable
https://bugs.webkit.org/show_bug.cgi?id=187752

Reviewed by Mark Lam.

JSON.stringify has an implicit root wrapper object since we would like to call replacer
with a wrapper object and a property name. While we always create this wrapper object,
it is unnecessary if the given replacer is not callable.

This patch removes wrapper object creation when a replacer is not callable to avoid unnecessary
allocations. This change slightly improves the performance of Kraken/json-stringify-tinderbox.

                                   baseline                  patched

json-stringify-tinderbox        39.730+-0.590      ^      38.853+-0.266         ^ definitely 1.0226x faster

* runtime/JSONObject.cpp:
(JSC::Stringifier::isCallableReplacer const):
(JSC::Stringifier::Stringifier):
(JSC::Stringifier::stringify):
(JSC::Stringifier::appendStringifiedValue):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@233917 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 months agoRename WordBreak::Break to WordBreak::BreakWord
mmaxfield@apple.com [Wed, 18 Jul 2018 18:10:08 +0000 (18:10 +0000)]
Rename WordBreak::Break to WordBreak::BreakWord
https://bugs.webkit.org/show_bug.cgi?id=187767

Reviewed by Simon Fraser.

These breaking properties are very confusing. There are:

1. word-break: break-all, a standard value that allows breaking after every
character.
2. word-break: break-word, a non-standard value which allows for breaking after
every character, but only if the word is too long for the available width (otherwise
it works the same as word-break: normal). This affects the min-content-size of the
text (and makes it equal to what it would be if word-break: break-all was
specified).
3. word-wrap: break-word, which is the same as word-break: break-word, but doesn't
affect the min-content-size of the text.
4. overflow-wrap: break-word, which is the same as word-wrap: break-word.

Because this is so confusing it's valuable for our internal enums to match the names
of the official CSS properties/values.

No new tests because there is no behavior change.

* css/CSSPrimitiveValueMappings.h:
(WebCore::CSSPrimitiveValue::CSSPrimitiveValue):
(WebCore::CSSPrimitiveValue::operator WordBreak const):
* rendering/RenderText.cpp:
(WebCore::RenderText::computePreferredLogicalWidths):
* rendering/style/RenderStyle.h:
(WebCore::RenderStyle::breakWords const):
* rendering/style/RenderStyleConstants.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@233916 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 months agoAdd SPI to defer running async script until after document load
wenson_hsieh@apple.com [Wed, 18 Jul 2018 17:08:46 +0000 (17:08 +0000)]
Add SPI to defer running async script until after document load
https://bugs.webkit.org/show_bug.cgi?id=187748
<rdar://problem/42317378>

Reviewed by Ryosuke Niwa and Tim Horton.

Source/WebCore:

On watchOS, we currently observe that time-consuming async scripts can block the first paint of Reader, leaving
the user with a blank screen for tens of seconds. One way to mitigate this is to defer async script execution
until after document load (i.e. the same timing as DOMContentLoaded).

This patch introduces an SPI configuration allowing internal clients to defer execution of asynchronous script
until after document load; this, in combination with the parser yielding token introduced in r233891, allows
Safari on watchOS to avoid being blocked on slow script execution before the first paint of the Reader page on
most article-like pages. See below for more details.

Test: RunScriptAfterDocumentLoad.ExecutionOrderOfScriptsInDocument

* dom/Document.cpp:
(WebCore::Document::shouldDeferAsynchronousScriptsUntilParsingFinishes const):
(WebCore::Document::finishedParsing):

Notify ScriptRunner when the Document has finished parsing, and is about to fire DOMContentLoaded.

* dom/Document.h:
* dom/ScriptRunner.cpp:
(WebCore::ScriptRunner::documentFinishedParsing):

When the document is finished parsing, kick off the script execution timer if needed to run any async script
that has been deferred.

(WebCore::ScriptRunner::notifyFinished):
(WebCore::ScriptRunner::timerFired):

Instead of always taking from the list of async scripts to execute, check our document to see whether we should
defer this until after document load. If so, ignore `m_scriptsToExecuteSoon`.

* dom/ScriptRunner.h:
* page/Settings.yaml:

Add a WebCore setting for this behavior.

Source/WebKit:

Add plumbing for a new ShouldDeferAsynchronousScriptsUntilAfterDocumentLoad configuration that determines
whether async script execution should be deferred until document load (i.e. DOMContentLoaded). This
configuration defaults to NO on all platforms. See WebCore ChangeLog for more detail.

* Shared/WebPreferences.yaml:
* UIProcess/API/Cocoa/WKWebView.mm:
(-[WKWebView _initializeWithConfiguration:]):
* UIProcess/API/Cocoa/WKWebViewConfiguration.mm:
(-[WKWebViewConfiguration init]):
(-[WKWebViewConfiguration copyWithZone:]):
(-[WKWebViewConfiguration _shouldDeferAsynchronousScriptsUntilAfterDocumentLoad]):
(-[WKWebViewConfiguration _setShouldDeferAsynchronousScriptsUntilAfterDocumentLoad:]):
* UIProcess/API/Cocoa/WKWebViewConfigurationPrivate.h:

Tools:

Add an API test to verify that when the deferred async script configuration is set, async scripts will be
executed after the DOMContentLoaded event.

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

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@233915 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 months ago[WPE] Update WPEBackend in flatpak
cturner@igalia.com [Wed, 18 Jul 2018 16:04:33 +0000 (16:04 +0000)]
[WPE] Update WPEBackend in flatpak
https://bugs.webkit.org/show_bug.cgi?id=187753

r233763 updated WPEBackend for the jhbuild, but not the flatpak. This
caused WPE_BACKEND_CHECK_VERSION to not be defined, triggering a bug
in the version guard,

#if defined(WPE_BACKEND_CHECK_VERSION) && WPE_BACKEND_CHECK_VERSION(0, 2, 0)

This doesn't work as intended, since the C preprocessor first expands
all macro arguments in the #if expression before evaluating the
expression. When WPE_BACKEND_CHECK_VERSION is not defined, this will
lead to a syntax error and some head scratching.

A future patch should address the buggy macro check, when it is
decided whether the assume the macro is always defined and bump
the API requirements on WPEBackend, or to move the check into two
conditionals to avoid the expansion problem.

Unreviewed build fix.

* flatpak/org.webkit.WPE.yaml:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@233914 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 months agoFix the parsing of colors in attributed string tests, and make it possible to dump...
simon.fraser@apple.com [Wed, 18 Jul 2018 15:55:33 +0000 (15:55 +0000)]
Fix the parsing of colors in attributed string tests, and make it possible to dump the typing attributes
https://bugs.webkit.org/show_bug.cgi?id=187747

Reviewed by Ryosuke Niwa.

Tools:

Add textInputController.attributedStringForTyping(), which returns a one-character
attributed string whose attributes are the typing attributes, making it possible to
test -[WebView typingAttributes].

Sadly WebCore's convertObjcValueToValue() doesn't know how to convert NSDictionary,
so we can't return -typingAttributes directly.

* DumpRenderTree/mac/TextInputControllerMac.m:
(+[TextInputController isSelectorExcludedFromWebScript:]):
(-[TextInputController attributedStringForTyping]):

LayoutTests:

Fix the parsing of color properties in dump-attributed-string.js, and treat NSStrokeColor as
a color. Rebase all the affected tests. Give macOS Sierra its own expectations with the legacy NSCustomColorSpace.

Add attributed-string-for-typing.html which tests typingAttributes.

* editing/mac/attributed-string/anchor-element-expected.txt:
* editing/mac/attributed-string/attributed-string-for-typing-expected.txt: Added.
* editing/mac/attributed-string/attributed-string-for-typing.html: Added.
* editing/mac/attributed-string/basic-expected.txt:
* editing/mac/attributed-string/comment-cdata-section-expected.txt:
* editing/mac/attributed-string/font-size-expected.txt:
* editing/mac/attributed-string/font-style-variant-effect-expected.txt:
* editing/mac/attributed-string/font-weight-expected.txt:
* editing/mac/attributed-string/letter-spacing-expected.txt:
* editing/mac/attributed-string/resources/dump-attributed-string.js:
(dumpAttributedString):
(formatNonParagraphAttributeValue):
(parseNSColorDescription):
(window.onload):
(serializeAttributedString.log): Deleted.
(serializeAttributedString.): Deleted.
(serializeAttributedString): Deleted.
* editing/mac/attributed-string/text-decorations-expected.txt:
* editing/mac/attributed-string/vertical-align-expected.txt:
* platform/mac-sierra/editing/mac/attributed-string/anchor-element-expected.txt:
* platform/mac-sierra/editing/mac/attributed-string/attributed-string-for-typing-expected.txt: Added.
* platform/mac-sierra/editing/mac/attributed-string/basic-expected.txt:
* platform/mac-sierra/editing/mac/attributed-string/comment-cdata-section-expected.txt:
* platform/mac-sierra/editing/mac/attributed-string/font-size-expected.txt:
* platform/mac-sierra/editing/mac/attributed-string/font-style-variant-effect-expected.txt:
* platform/mac-sierra/editing/mac/attributed-string/font-weight-expected.txt:
* platform/mac-sierra/editing/mac/attributed-string/letter-spacing-expected.txt:
* platform/mac-sierra/editing/mac/attributed-string/text-decorations-expected.txt:
* platform/mac-sierra/editing/mac/attributed-string/vertical-align-expected.txt:
* platform/mac/editing/mac/attributed-string/anchor-element-expected.txt: Removed.
* platform/mac/editing/mac/attributed-string/basic-expected.txt: Removed.
* platform/mac/editing/mac/attributed-string/comment-cdata-section-expected.txt: Removed.
* platform/mac/editing/mac/attributed-string/font-size-expected.txt: Removed.
* platform/mac/editing/mac/attributed-string/font-style-variant-effect-expected.txt: Removed.
* platform/mac/editing/mac/attributed-string/font-weight-expected.txt: Removed.
* platform/mac/editing/mac/attributed-string/letter-spacing-expected.txt: Removed.
* platform/mac/editing/mac/attributed-string/text-decorations-expected.txt: Removed.
* platform/mac/editing/mac/attributed-string/vertical-align-expected.txt: Removed.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@233913 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 months agoSwitch CMake ports back to C++ 14
mcatanzaro@igalia.com [Wed, 18 Jul 2018 14:43:50 +0000 (14:43 +0000)]
Switch CMake ports back to C++ 14
https://bugs.webkit.org/show_bug.cgi?id=187744

Reviewed by Ryosuke Niwa.

.:

The XCode build is still not using C++ 17, it's been several months since CMake ports
switched, everything builds fine without changes if we switch back, and there have been some
unfixed problems. Let's go back to C++ 14 for now. We can switch back to C++ 17 whenever we
are ready to switch over XCode at the same time, to ensure we don't wind up with divergent
behavior for std::optional.

* Source/cmake/WebKitCompilerFlags.cmake:

Source/WTF:

Always use WTF's internal std::optional implementation, since std::optional is not part of
C++ 14.

* wtf/Optional.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@233912 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 months ago[GLIB] Add jsc_context_check_syntax() to GLib API
carlosgc@webkit.org [Wed, 18 Jul 2018 12:54:38 +0000 (12:54 +0000)]
[GLIB] Add jsc_context_check_syntax() to GLib API
https://bugs.webkit.org/show_bug.cgi?id=187694

Reviewed by Yusuke Suzuki.

Source/JavaScriptCore:

A new function to be able to check for syntax errors without actually evaluating the code.

* API/glib/JSCContext.cpp:
(jsc_context_check_syntax):
* API/glib/JSCContext.h:
* API/glib/docs/jsc-glib-4.0-sections.txt:

Tools:

Add a new test case.

* TestWebKitAPI/Tests/JavaScriptCore/glib/TestJSC.cpp:
(testJSCCheckSyntax):
(main):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@233911 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 months ago[Nicosia] Add debug border, repaint counter state tracking to Nicosia::CompositionLayer
zandobersek@gmail.com [Wed, 18 Jul 2018 10:22:15 +0000 (10:22 +0000)]
[Nicosia] Add debug border, repaint counter state tracking to Nicosia::CompositionLayer
https://bugs.webkit.org/show_bug.cgi?id=187749

Reviewed by Carlos Garcia Campos.

Add the RepaintCounter and DebugBorder structs to
Nicosia::CompositionLayer::LayerState, tracking visibility as well as
repaint count or debug color and width.

Instances of RepaintCounter and DebugBorder types are kept in each
CoordinatedGraphicsLayer object, updating the relevant data as it is
changed (since the GraphicsLayer object isn't tracking these values on
its own). During layer flush these values (if changed) are then copied
over into the CompositionLayer state.

* platform/graphics/nicosia/NicosiaPlatformLayer.cpp:
Also fix the year in the license header.
* platform/graphics/nicosia/NicosiaPlatformLayer.h:
Also fix the year in the license header.
* platform/graphics/texmap/coordinated/CoordinatedGraphicsLayer.cpp:
(WebCore::CoordinatedGraphicsLayer::didUpdateTileBuffers):
(WebCore::CoordinatedGraphicsLayer::setShowDebugBorder):
(WebCore::CoordinatedGraphicsLayer::setShowRepaintCounter):
(WebCore::CoordinatedGraphicsLayer::setDebugBorder):
(WebCore::CoordinatedGraphicsLayer::flushCompositingStateForThisLayerOnly):
* platform/graphics/texmap/coordinated/CoordinatedGraphicsLayer.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@233910 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 months ago[CoordGraphics] Start tracking Nicosia layers in CoordinatedGraphicsState
zandobersek@gmail.com [Wed, 18 Jul 2018 10:20:21 +0000 (10:20 +0000)]
[CoordGraphics] Start tracking Nicosia layers in CoordinatedGraphicsState
https://bugs.webkit.org/show_bug.cgi?id=187751

Reviewed by Carlos Garcia Campos.

Start including the Nicosia::CompositionLayer objects in the
CoordinatedGraphicsState struct, under a separate NicosiaState struct.
References to all the layers in a given scene are kept in a HashSet,
and a separate reference to the root layer kept in a separate member
variable.

Source/WebCore:

* platform/graphics/texmap/coordinated/CoordinatedGraphicsLayer.cpp:
(WebCore::CoordinatedGraphicsLayer::compositionLayer const):
* platform/graphics/texmap/coordinated/CoordinatedGraphicsLayer.h:
Add the getter method that returns internal Nicosia::CompositionLayer
object. This can't be defined in the class definition because of
WEBCORE_EXPORT used on the CoordinatedGraphicsLayer class.
* platform/graphics/texmap/coordinated/CoordinatedGraphicsState.h:

Source/WebKit:

CompositingCoordinator now takes care of adding or removing the
Nicosia::CompositionLayer objects to the NicosiaState's HashSet, as well
as setting the root layer object when it's being initialized. Additions
and removals of Nicosia::CompositionLayer correspond to the additions
and removals of CoordinatedGraphicsLayer objects to the coordinator's
m_registeredLayers HashMap.

Upon each state commit that's done in CoordinatedGraphicsScene, the
NicosiaState object will be copied into the member variable. Nothing is
done yet with that state object, but in the near future it will be used
to finally commit all the state details into the TextureMapper layers.

* Shared/CoordinatedGraphics/CoordinatedGraphicsScene.cpp:
(WebKit::CoordinatedGraphicsScene::commitSceneState):
* Shared/CoordinatedGraphics/CoordinatedGraphicsScene.h:
* WebProcess/WebPage/CoordinatedGraphics/CompositingCoordinator.cpp:
(WebKit::CompositingCoordinator::initializeRootCompositingLayerIfNeeded):
(WebKit::CompositingCoordinator::createGraphicsLayer):
(WebKit::CompositingCoordinator::detachLayer):
(WebKit::CompositingCoordinator::attachLayer):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@233909 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 months agoShrink CompositeAnimation and AnimationBase
simon.fraser@apple.com [Wed, 18 Jul 2018 07:02:02 +0000 (07:02 +0000)]
Shrink CompositeAnimation and AnimationBase
https://bugs.webkit.org/show_bug.cgi?id=187683

Reviewed by Daniel Bates.

Reduce the size of CompositeAnimation and AnimationBase.

* page/animation/AnimationBase.h:
* page/animation/CompositeAnimation.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@233908 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 months ago[WPE][GTK] Update xdgmime
mcatanzaro@igalia.com [Wed, 18 Jul 2018 06:08:36 +0000 (06:08 +0000)]
[WPE][GTK] Update xdgmime
https://bugs.webkit.org/show_bug.cgi?id=187727

Reviewed by Carlos Garcia Campos.

* xdgmime/README:
* xdgmime/README.webkit:
* xdgmime/src/xdgmime.c:
(_xdg_mime_mime_type_subclass):
* xdgmime/src/xdgmimecache.c:
(_xdg_mime_cache_new_from_file):
(_xdg_mime_cache_get_mime_type_for_file):
(_xdg_mime_cache_mime_type_subclass):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@233907 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 months agoRevert r233630 since it broke internal wasm benchmarks
keith_miller@apple.com [Wed, 18 Jul 2018 05:14:25 +0000 (05:14 +0000)]
Revert r233630 since it broke internal wasm benchmarks
https://bugs.webkit.org/show_bug.cgi?id=187746

Unreviewed revert.

Source/JavaScriptCore:

This patch seems to have broken internal Wasm benchmarks. This
issue is likely due to an underlying bug but let's rollout while
we investigate.

* bytecode/CodeType.h:
* bytecode/UnlinkedCodeBlock.cpp:
(JSC::UnlinkedCodeBlock::UnlinkedCodeBlock):
* bytecode/UnlinkedCodeBlock.h:
(JSC::UnlinkedCodeBlock::codeType const):
(JSC::UnlinkedCodeBlock::didOptimize const):
(JSC::UnlinkedCodeBlock::setDidOptimize):
* bytecode/VirtualRegister.h:
(JSC::VirtualRegister::VirtualRegister):
(): Deleted.

Source/WTF:

* wtf/TriState.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@233906 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 months agoREGRESSION (iOS 12): Can't scroll to the bottom of the page in WKWebView while keyboa...
timothy_horton@apple.com [Wed, 18 Jul 2018 03:50:12 +0000 (03:50 +0000)]
REGRESSION (iOS 12): Can't scroll to the bottom of the page in WKWebView while keyboard is up on pages with viewport-fit=cover
https://bugs.webkit.org/show_bug.cgi?id=187743
<rdar://problem/41651546>

Reviewed by Simon Fraser.

UIScrollView's _systemContentInset no longer includes keyboard insets
in apps linked on iOS 12+ when contentInsetAdjustmentBehavior is None.

We use contentInsetAdjustmentBehavior to control adjustment of other
sources of insets, but expect the keyboard inset to always be applied.

For now, barring a more comprehensive way to separate insets, re-add
the keyboard inset in cases where UIKit does not.

* Platform/spi/ios/UIKitSPI.h:
Move some IPI to the SPI header.

* UIProcess/API/Cocoa/WKWebView.mm:
(-[WKWebView _haveSetObscuredInsets]):
* UIProcess/API/Cocoa/WKWebViewInternal.h:
Make it possible for WKScrollView to check whether the client is overriding insets.

* UIProcess/Cocoa/VersionChecks.h:
Add a linkedOnOrAfter() version for this new UIScrollView behavior.

* UIProcess/ios/WKScrollView.mm:
(-[WKScrollView initWithFrame:]):
Force WKScrollView's scroll indicator to always respect insets. We always
want the scroll bars in a sensible place, even if the page sets
viewport-fit=cover.

(-[WKScrollView _adjustForAutomaticKeyboardInfo:animated:lastAdjustment:]):
Store the bottom inset due to the keyboard.

(-[WKScrollView _systemContentInset]):
Add the bottom inset due to the keyboard to the systemContentInset
in all cases where UIKit does not. Also avoid adding it if the client takes
full control of the insets, because the only client that does (MobileSafari)
includes insets-due-to-keyboard in their custom insets.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@233905 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 months agoRELEASE_ASSERT() under IPC::Connection::sendSync() from PluginProxy::supportsSnapshot...
cdumez@apple.com [Wed, 18 Jul 2018 03:43:37 +0000 (03:43 +0000)]
RELEASE_ASSERT() under IPC::Connection::sendSync() from PluginProxy::supportsSnapshotting()
https://bugs.webkit.org/show_bug.cgi?id=187740
<rdar://problem/41818955>

Reviewed by Youenn Fablet.

As per the crash trace, PluginProxy::supportsSnapshotting() can be called during layout but does synchronous
IPC. As a result, we need to prevent WebCore re-entrancy by using DoNotProcessIncomingMessagesWhenWaitingForSyncReply
sendOption.

* WebProcess/Plugins/PluginProxy.cpp:
(WebKit::PluginProxy::supportsSnapshotting const):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@233904 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 months agoEnsure timingFunctionForKeyframeAtIndex() can be used from setAnimatedPropertiesInSty...
graouts@webkit.org [Wed, 18 Jul 2018 01:05:29 +0000 (01:05 +0000)]
Ensure timingFunctionForKeyframeAtIndex() can be used from setAnimatedPropertiesInStyle().
https://bugs.webkit.org/show_bug.cgi?id=187637
<rdar://problem/42157915>

Reviewed by Dean Jackson.

Source/WebCore:

Test: webanimations/empty-keyframes-crash.html

Unlike what we assumed, it is possible to have a non-declarative animation without any parsed keyframes.
This can happen as a result of calling `Element.animate({}, …)`. In this case, we want to return a null
value in timingFunctionForKeyframeAtIndex() so we update the call site in setAnimatedPropertiesInStyle()
which is the only place where we didn't check for a null value and didn't know for sure that there would
be parsed keyframes to rely on in the case of a WebAnimation instance.

* animation/KeyframeEffectReadOnly.cpp:
(WebCore::KeyframeEffectReadOnly::setAnimatedPropertiesInStyle):
(WebCore::KeyframeEffectReadOnly::timingFunctionForKeyframeAtIndex):

LayoutTests:

Add a new test that would crash prior to this change.

* webanimations/empty-keyframes-crash-expected.txt: Added.
* webanimations/empty-keyframes-crash.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@233903 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 months agoRebaseline imported/w3c/web-platform-tests/WebCryptoAPI/derive_bits_keys/pbkdf2.https...
ryanhaddad@apple.com [Wed, 18 Jul 2018 00:59:33 +0000 (00:59 +0000)]
Rebaseline imported/w3c/web-platform-tests/WebCryptoAPI/derive_bits_keys/pbkdf2.https.worker.html for Sierra after r233898.

Unreviewed test gardening.

* platform/mac-sierra/imported/w3c/web-platform-tests/WebCryptoAPI/derive_bits_keys/pbkdf2.https.worker-expected.txt: Added.
* platform/mac/imported/w3c/web-platform-tests/WebCryptoAPI/derive_bits_keys/pbkdf2.https.worker-expected.txt: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@233902 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 months agoMarking test as flakey
tsavell@apple.com [Wed, 18 Jul 2018 00:39:25 +0000 (00:39 +0000)]
Marking test as flakey

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@233901 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 months agoUnreviewed, test gardening after r233898.
jiewen_tan@apple.com [Wed, 18 Jul 2018 00:24:23 +0000 (00:24 +0000)]
Unreviewed, test gardening after r233898.

* TestExpectations:
* platform/gtk/TestExpectations:
* platform/ios/TestExpectations:
* platform/mac/TestExpectations:
* platform/wpe/TestExpectations:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@233900 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 months agoREGRESSION: [macOS Sierra] TestWebKitAPI.WebKit.WebsiteDataStoreCustomPaths is a...
sihui_liu@apple.com [Tue, 17 Jul 2018 23:34:36 +0000 (23:34 +0000)]
REGRESSION: [macOS Sierra] TestWebKitAPI.WebKit.WebsiteDataStoreCustomPaths is a flaky failure
https://bugs.webkit.org/show_bug.cgi?id=187066
<rdar://problem/41609065>

Reviewed by Chris Dumez.

In r232668 we started to do reload if web process crashes by default. As we killed the web
process explicitly in this test, if web page got reloaded, messages would be sent again,
and flag set in message handler could keep the test continue to evaluate the expectation
without waiting for removeDataOfTypes to finish.

* TestWebKitAPI/Tests/WebKitCocoa/WebsiteDataStoreCustomPaths.mm:
(-[WebsiteDataStoreCustomPathsMessageHandler webViewWebContentProcessDidTerminate:]):
(TEST):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@233899 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 months ago[WebCrypto] Crypto operations should copy their parameters before hoping to another...
jiewen_tan@apple.com [Tue, 17 Jul 2018 23:06:12 +0000 (23:06 +0000)]
[WebCrypto] Crypto operations should copy their parameters before hoping to another thread
https://bugs.webkit.org/show_bug.cgi?id=187501
<rdar://problem/41438160>

Reviewed by Youenn Fablet.

Source/WebCore:

This patch aims at making all captured variables in all crypto lambdas that need to be passed
to a worker thread thread safe, which includes:
1) changing ref counted objects to thread safe ref counted object.
2) adding isolatedCopy methods to non ref counted classes, so they can be called by CrossThreadCopy().

In addition to above changes, this patch also does the following things:
1) change the name CryptoAlgorithm::dispatchOperation => CryptoAlgorithm::dispatchOperationInWorkQueue
to make it clear that lambdas will be passed to a secondary thread.
2) make CryptoAlgorithmParameters as const parameters for all methods.
3) add null checks on BufferSource.length() and .data().

Tests: crypto/subtle/aes-gcm-import-key-unwrap-ec-raw-key.html
       http/wpt/crypto/aes-cbc-crash.any.html
       http/wpt/crypto/aes-cbc-crash.any.worker.html
       http/wpt/crypto/aes-ctr-crash.any.html
       http/wpt/crypto/aes-ctr-crash.any.worker.html
       http/wpt/crypto/aes-gcm-crash.any.html
       http/wpt/crypto/aes-gcm-crash.any.worker.html
       http/wpt/crypto/derive-hmac-key-crash.any.html
       http/wpt/crypto/derive-hmac-key-crash.any.worker.html
       http/wpt/crypto/ecdsa-crash.any.html
       http/wpt/crypto/ecdsa-crash.any.worker.html
       http/wpt/crypto/hkdf-crash.any.html
       http/wpt/crypto/hkdf-crash.any.worker.html
       http/wpt/crypto/pbkdf2-crash.any.html
       http/wpt/crypto/pbkdf2-crash.any.worker.html
       http/wpt/crypto/rsa-oaep-crash.any.html
       http/wpt/crypto/rsa-oaep-crash.any.worker.html
       http/wpt/crypto/rsa-pss-crash.any.html
       http/wpt/crypto/rsa-pss-crash.any.worker.html
       http/wpt/crypto/unwrap-ec-key-crash.any.html
       http/wpt/crypto/unwrap-ec-key-crash.any.worker.html
       http/wpt/crypto/unwrap-rsa-key-crash.any.html
       http/wpt/crypto/unwrap-rsa-key-crash.any.worker.html

* bindings/js/BufferSource.h:
(WebCore::BufferSource::data const):
(WebCore::BufferSource::length const):
* crypto/CryptoAlgorithm.cpp:
(WebCore::CryptoAlgorithm::encrypt):
(WebCore::CryptoAlgorithm::decrypt):
(WebCore::CryptoAlgorithm::sign):
(WebCore::CryptoAlgorithm::verify):
(WebCore::CryptoAlgorithm::deriveBits):
(WebCore::CryptoAlgorithm::importKey):
(WebCore::dispatchAlgorithmOperation):
(WebCore::CryptoAlgorithm::dispatchOperationInWorkQueue):
(WebCore::CryptoAlgorithm::dispatchOperation): Deleted.
* crypto/CryptoAlgorithm.h:
* crypto/SubtleCrypto.cpp:
(WebCore::crossThreadCopyImportParams):
(WebCore::SubtleCrypto::encrypt):
(WebCore::SubtleCrypto::decrypt):
(WebCore::SubtleCrypto::sign):
(WebCore::SubtleCrypto::verify):
(WebCore::SubtleCrypto::deriveKey):
(WebCore::SubtleCrypto::deriveBits):
(WebCore::SubtleCrypto::importKey):
(WebCore::SubtleCrypto::wrapKey):
(WebCore::SubtleCrypto::unwrapKey):
* crypto/algorithms/CryptoAlgorithmAES_CBC.cpp:
(WebCore::CryptoAlgorithmAES_CBC::encrypt):
(WebCore::CryptoAlgorithmAES_CBC::decrypt):
(WebCore::CryptoAlgorithmAES_CBC::importKey):
* crypto/algorithms/CryptoAlgorithmAES_CBC.h:
* crypto/algorithms/CryptoAlgorithmAES_CFB.cpp:
(WebCore::CryptoAlgorithmAES_CFB::encrypt):
(WebCore::CryptoAlgorithmAES_CFB::decrypt):
(WebCore::CryptoAlgorithmAES_CFB::importKey):
* crypto/algorithms/CryptoAlgorithmAES_CFB.h:
* crypto/algorithms/CryptoAlgorithmAES_CTR.cpp:
(WebCore::parametersAreValid):
(WebCore::CryptoAlgorithmAES_CTR::encrypt):
(WebCore::CryptoAlgorithmAES_CTR::decrypt):
(WebCore::CryptoAlgorithmAES_CTR::importKey):
* crypto/algorithms/CryptoAlgorithmAES_CTR.h:
* crypto/algorithms/CryptoAlgorithmAES_GCM.cpp:
(WebCore::CryptoAlgorithmAES_GCM::encrypt):
(WebCore::CryptoAlgorithmAES_GCM::decrypt):
(WebCore::CryptoAlgorithmAES_GCM::importKey):
* crypto/algorithms/CryptoAlgorithmAES_GCM.h:
* crypto/algorithms/CryptoAlgorithmAES_KW.cpp:
(WebCore::CryptoAlgorithmAES_KW::importKey):
* crypto/algorithms/CryptoAlgorithmAES_KW.h:
* crypto/algorithms/CryptoAlgorithmECDH.cpp:
(WebCore::CryptoAlgorithmECDH::deriveBits):
(WebCore::CryptoAlgorithmECDH::importKey):
* crypto/algorithms/CryptoAlgorithmECDH.h:
* crypto/algorithms/CryptoAlgorithmECDSA.cpp:
(WebCore::CryptoAlgorithmECDSA::sign):
(WebCore::CryptoAlgorithmECDSA::verify):
(WebCore::CryptoAlgorithmECDSA::importKey):
* crypto/algorithms/CryptoAlgorithmECDSA.h:
* crypto/algorithms/CryptoAlgorithmHKDF.cpp:
(WebCore::CryptoAlgorithmHKDF::deriveBits):
(WebCore::CryptoAlgorithmHKDF::importKey):
* crypto/algorithms/CryptoAlgorithmHKDF.h:
* crypto/algorithms/CryptoAlgorithmHMAC.cpp:
(WebCore::CryptoAlgorithmHMAC::sign):
(WebCore::CryptoAlgorithmHMAC::verify):
(WebCore::CryptoAlgorithmHMAC::importKey):
* crypto/algorithms/CryptoAlgorithmHMAC.h:
* crypto/algorithms/CryptoAlgorithmPBKDF2.cpp:
(WebCore::CryptoAlgorithmPBKDF2::deriveBits):
(WebCore::CryptoAlgorithmPBKDF2::importKey):
* crypto/algorithms/CryptoAlgorithmPBKDF2.h:
* crypto/algorithms/CryptoAlgorithmRSAES_PKCS1_v1_5.cpp:
(WebCore::CryptoAlgorithmRSAES_PKCS1_v1_5::encrypt):
(WebCore::CryptoAlgorithmRSAES_PKCS1_v1_5::decrypt):
(WebCore::CryptoAlgorithmRSAES_PKCS1_v1_5::importKey):
* crypto/algorithms/CryptoAlgorithmRSAES_PKCS1_v1_5.h:
* crypto/algorithms/CryptoAlgorithmRSASSA_PKCS1_v1_5.cpp:
(WebCore::CryptoAlgorithmRSASSA_PKCS1_v1_5::sign):
(WebCore::CryptoAlgorithmRSASSA_PKCS1_v1_5::verify):
(WebCore::CryptoAlgorithmRSASSA_PKCS1_v1_5::importKey):
* crypto/algorithms/CryptoAlgorithmRSASSA_PKCS1_v1_5.h:
* crypto/algorithms/CryptoAlgorithmRSA_OAEP.cpp:
(WebCore::CryptoAlgorithmRSA_OAEP::encrypt):
(WebCore::CryptoAlgorithmRSA_OAEP::decrypt):
(WebCore::CryptoAlgorithmRSA_OAEP::importKey):
* crypto/algorithms/CryptoAlgorithmRSA_OAEP.h:
* crypto/algorithms/CryptoAlgorithmRSA_PSS.cpp:
(WebCore::CryptoAlgorithmRSA_PSS::sign):
(WebCore::CryptoAlgorithmRSA_PSS::verify):
(WebCore::CryptoAlgorithmRSA_PSS::importKey):
* crypto/algorithms/CryptoAlgorithmRSA_PSS.h:
* crypto/gcrypt/CryptoAlgorithmAES_CBCGCrypt.cpp:
(WebCore::CryptoAlgorithmAES_CBC::platformEncrypt):
(WebCore::CryptoAlgorithmAES_CBC::platformDecrypt):
* crypto/gcrypt/CryptoAlgorithmAES_CFBGCrypt.cpp:
(WebCore::CryptoAlgorithmAES_CFB::platformEncrypt):
(WebCore::CryptoAlgorithmAES_CFB::platformDecrypt):
* crypto/gcrypt/CryptoAlgorithmAES_CTRGCrypt.cpp:
(WebCore::CryptoAlgorithmAES_CTR::platformEncrypt):
(WebCore::CryptoAlgorithmAES_CTR::platformDecrypt):
* crypto/gcrypt/CryptoAlgorithmAES_GCMGCrypt.cpp:
(WebCore::CryptoAlgorithmAES_GCM::platformEncrypt):
(WebCore::CryptoAlgorithmAES_GCM::platformDecrypt):
* crypto/gcrypt/CryptoAlgorithmHKDFGCrypt.cpp:
(WebCore::CryptoAlgorithmHKDF::platformDeriveBits):
* crypto/gcrypt/CryptoAlgorithmPBKDF2GCrypt.cpp:
(WebCore::CryptoAlgorithmPBKDF2::platformDeriveBits):
* crypto/gcrypt/CryptoAlgorithmRSA_OAEPGCrypt.cpp:
(WebCore::CryptoAlgorithmRSA_OAEP::platformEncrypt):
(WebCore::CryptoAlgorithmRSA_OAEP::platformDecrypt):
* crypto/gcrypt/CryptoAlgorithmRSA_PSSGCrypt.cpp:
(WebCore::CryptoAlgorithmRSA_PSS::platformSign):
(WebCore::CryptoAlgorithmRSA_PSS::platformVerify):
* crypto/mac/CryptoAlgorithmAES_CBCMac.cpp:
(WebCore::CryptoAlgorithmAES_CBC::platformEncrypt):
(WebCore::CryptoAlgorithmAES_CBC::platformDecrypt):
* crypto/mac/CryptoAlgorithmAES_CFBMac.cpp:
(WebCore::CryptoAlgorithmAES_CFB::platformEncrypt):
(WebCore::CryptoAlgorithmAES_CFB::platformDecrypt):
* crypto/mac/CryptoAlgorithmAES_CTRMac.cpp:
(WebCore::CryptoAlgorithmAES_CTR::platformEncrypt):
(WebCore::CryptoAlgorithmAES_CTR::platformDecrypt):
* crypto/mac/CryptoAlgorithmAES_GCMMac.cpp:
(WebCore::CryptoAlgorithmAES_GCM::platformEncrypt):
(WebCore::CryptoAlgorithmAES_GCM::platformDecrypt):
* crypto/mac/CryptoAlgorithmHKDFMac.cpp:
(WebCore::CryptoAlgorithmHKDF::platformDeriveBits):
* crypto/mac/CryptoAlgorithmPBKDF2Mac.cpp:
(WebCore::CryptoAlgorithmPBKDF2::platformDeriveBits):
* crypto/mac/CryptoAlgorithmRSA_OAEPMac.cpp:
(WebCore::CryptoAlgorithmRSA_OAEP::platformEncrypt):
(WebCore::CryptoAlgorithmRSA_OAEP::platformDecrypt):
* crypto/mac/CryptoAlgorithmRSA_PSSMac.cpp:
(WebCore::CryptoAlgorithmRSA_PSS::platformSign):
(WebCore::CryptoAlgorithmRSA_PSS::platformVerify):
* crypto/parameters/CryptoAlgorithmAesCbcCfbParams.h:
* crypto/parameters/CryptoAlgorithmAesCtrParams.h:
* crypto/parameters/CryptoAlgorithmAesGcmParams.h:
* crypto/parameters/CryptoAlgorithmEcKeyParams.h:
* crypto/parameters/CryptoAlgorithmEcdsaParams.h:
* crypto/parameters/CryptoAlgorithmHkdfParams.h:
* crypto/parameters/CryptoAlgorithmHmacKeyParams.h:
* crypto/parameters/CryptoAlgorithmPbkdf2Params.h:
* crypto/parameters/CryptoAlgorithmRsaHashedImportParams.h:
* crypto/parameters/CryptoAlgorithmRsaOaepParams.h:
* crypto/parameters/CryptoAlgorithmRsaPssParams.h:

LayoutTests:

crypto/subtle/aes-gcm-import-key-unwrap-ec-raw-key.html is an exception of this series of tests as
it only aims to test the correct behavoir of suggested algorithms. This patch aslo does some test
gardening.

* TestExpectations:
* crypto/subtle/aes-gcm-import-key-unwrap-ec-raw-key-expected.txt: Added.
* crypto/subtle/aes-gcm-import-key-unwrap-ec-raw-key.html: Added.
* http/wpt/crypto/aes-cbc-crash.any-expected.txt: Added.
* http/wpt/crypto/aes-cbc-crash.any.html: Added.
* http/wpt/crypto/aes-cbc-crash.any.js: Added.
* http/wpt/crypto/aes-cbc-crash.any.worker-expected.txt: Added.
* http/wpt/crypto/aes-cbc-crash.any.worker.html: Added.
* http/wpt/crypto/aes-ctr-crash.any-expected.txt: Added.
* http/wpt/crypto/aes-ctr-crash.any.html: Added.
* http/wpt/crypto/aes-ctr-crash.any.js: Added.
* http/wpt/crypto/aes-ctr-crash.any.worker-expected.txt: Added.
* http/wpt/crypto/aes-ctr-crash.any.worker.html: Added.
* http/wpt/crypto/aes-gcm-crash.any-expected.txt: Added.
* http/wpt/crypto/aes-gcm-crash.any.html: Added.
* http/wpt/crypto/aes-gcm-crash.any.js: Added.
* http/wpt/crypto/aes-gcm-crash.any.worker-expected.txt: Added.
* http/wpt/crypto/aes-gcm-crash.any.worker.html: Added.
* http/wpt/crypto/derive-hmac-key-crash.any-expected.txt: Added.
* http/wpt/crypto/derive-hmac-key-crash.any.html: Added.
* http/wpt/crypto/derive-hmac-key-crash.any.js: Added.
* http/wpt/crypto/derive-hmac-key-crash.any.worker-expected.txt: Added.
* http/wpt/crypto/derive-hmac-key-crash.any.worker.html: Added.
* http/wpt/crypto/ecdsa-crash.any-expected.txt: Added.
* http/wpt/crypto/ecdsa-crash.any.html: Added.
* http/wpt/crypto/ecdsa-crash.any.js: Added.
* http/wpt/crypto/ecdsa-crash.any.worker-expected.txt: Added.
* http/wpt/crypto/ecdsa-crash.any.worker.html: Added.
* http/wpt/crypto/hkdf-crash.any-expected.txt: Added.
* http/wpt/crypto/hkdf-crash.any.html: Added.
* http/wpt/crypto/hkdf-crash.any.js: Added.
* http/wpt/crypto/hkdf-crash.any.worker-expected.txt: Added.
* http/wpt/crypto/hkdf-crash.any.worker.html: Added.
* http/wpt/crypto/pbkdf2-crash.any-expected.txt: Added.
* http/wpt/crypto/pbkdf2-crash.any.html: Added.
* http/wpt/crypto/pbkdf2-crash.any.js: Added.
* http/wpt/crypto/pbkdf2-crash.any.worker-expected.txt: Added.
* http/wpt/crypto/pbkdf2-crash.any.worker.html: Added.
* http/wpt/crypto/resources/common.js: Added.
* http/wpt/crypto/rsa-oaep-crash.any-expected.txt: Added.
* http/wpt/crypto/rsa-oaep-crash.any.html: Added.
* http/wpt/crypto/rsa-oaep-crash.any.js: Added.
* http/wpt/crypto/rsa-oaep-crash.any.worker-expected.txt: Added.
* http/wpt/crypto/rsa-oaep-crash.any.worker.html: Added.
* http/wpt/crypto/rsa-pss-crash.any-expected.txt: Added.
* http/wpt/crypto/rsa-pss-crash.any.html: Added.
* http/wpt/crypto/rsa-pss-crash.any.js: Added.
* http/wpt/crypto/rsa-pss-crash.any.worker-expected.txt: Added.
* http/wpt/crypto/rsa-pss-crash.any.worker.html: Added.
* http/wpt/crypto/unwrap-ec-key-crash.any-expected.txt: Added.
* http/wpt/crypto/unwrap-ec-key-crash.any.html: Added.
* http/wpt/crypto/unwrap-ec-key-crash.any.js: Added.
* http/wpt/crypto/unwrap-ec-key-crash.any.worker-expected.txt: Added.
* http/wpt/crypto/unwrap-ec-key-crash.any.worker.html: Added.
* http/wpt/crypto/unwrap-rsa-key-crash.any-expected.txt: Added.
* http/wpt/crypto/unwrap-rsa-key-crash.any.html: Added.
* http/wpt/crypto/unwrap-rsa-key-crash.any.js: Added.
* http/wpt/crypto/unwrap-rsa-key-crash.any.worker.html: Added.
* http/wpt/crypto/unwrap-rsa-key-crash.any.worker-expected.txt: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@233898 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 months agoTurn on PSON in WebKitTestRunner
cdumez@apple.com [Tue, 17 Jul 2018 21:50:09 +0000 (21:50 +0000)]
Turn on PSON in WebKitTestRunner
https://bugs.webkit.org/show_bug.cgi?id=186542

Reviewed by Brady Eidson.

Source/WebKit:

Fix leaking of pre-warmed WebContent processes which became obvious when turning
on process-swap-on-navigation by default in WebKitTestRunner. The issue was that
the WebProcessPool holds a strong reference to its WebProcessProxy objects via
m_processes data members. In turn, WebProcessProxy objects hold a strong reference
to their WebProcessPool via their m_processPool data member. This reference cycle
normally gets broken by calling WebProcessProxy::shutDown() which removes the
WebProcessProxy objects from its WebProcessPool's m_processes container.
WebProcessProxy::shutDown() normally gets called when a WebProcessProxy no longer
has any WebPageProxy objects. However, in the case of unused pre-warmed
WebProcessProxy objects, those never get any page and therefore we'd never break
the reference cycle. To address the issue, pre-warmed WebProcessProxy objects
now only hold a Weak reference to their process pool, only regular WebProcessProxy
objects hold a strong reference to their pool.

* UIProcess/WebProcessPool.cpp:
(WebKit::WebProcessPool::~WebProcessPool):
(WebKit::WebProcessPool::tryTakePrewarmedProcess):
* UIProcess/WebProcessPool.h:
* UIProcess/WebProcessProxy.cpp:
(WebKit::WebProcessProxy::WebProcessProxy):
(WebKit::WebProcessProxy::getLaunchOptions):
(WebKit::WebProcessProxy::markIsNoLongerInPrewarmedPool):
* UIProcess/WebProcessProxy.h:
(WebKit::WebProcessProxy::processPool):
(WebKit::WebProcessProxy::WeakOrStrongPtr::WeakOrStrongPtr):
(WebKit::WebProcessProxy::WeakOrStrongPtr::setIsWeak):
(WebKit::WebProcessProxy::WeakOrStrongPtr::get const):
(WebKit::WebProcessProxy::WeakOrStrongPtr::operator-> const):
(WebKit::WebProcessProxy::WeakOrStrongPtr::operator* const):
(WebKit::WebProcessProxy::WeakOrStrongPtr::operator bool const):
(WebKit::WebProcessProxy::WeakOrStrongPtr::updateStrongReference):

Tools:

Turn on PSON by default in WebKitTestRunner.

* WebKitTestRunner/TestOptions.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@233897 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 months agoUnskip LayoutTest imported/w3c/web-platform-tests/css/css-display/display-contents...
ryanhaddad@apple.com [Tue, 17 Jul 2018 20:50:52 +0000 (20:50 +0000)]
Unskip LayoutTest imported/w3c/web-platform-tests/css/css-display/display-contents-first-letter-002.html.
https://bugs.webkit.org/show_bug.cgi?id=186901

Unreviewed test gardening.

* TestExpectations:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@233895 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 months agoCodeBlock::baselineVersion() should account for executables with purged codeBlocks.
mark.lam@apple.com [Tue, 17 Jul 2018 20:20:21 +0000 (20:20 +0000)]
CodeBlock::baselineVersion() should account for executables with purged codeBlocks.
https://bugs.webkit.org/show_bug.cgi?id=187736
<rdar://problem/42114371>

Reviewed by Michael Saboff.

CodeBlock::baselineVersion() currently checks for a null replacement but does not
account for the fact that that the replacement can also be null due to the
executable having being purged of its codeBlocks due to a memory event (see
ExecutableBase::clearCode()).  This patch adds code to account for this.

* bytecode/CodeBlock.cpp:
(JSC::CodeBlock::baselineVersion):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@233893 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 months ago[Web Animations] Interpolation between lengths with an "auto" value should be discrete
commit-queue@webkit.org [Tue, 17 Jul 2018 18:54:05 +0000 (18:54 +0000)]
[Web Animations] Interpolation between lengths with an "auto" value should be discrete
https://bugs.webkit.org/show_bug.cgi?id=187721

Patch by Antoine Quint <graouts@apple.com> on 2018-07-17
Reviewed by Dean Jackson.

LayoutTests/imported/w3c:

Mark 2 new WPT progressions.

* web-platform-tests/web-animations/animation-model/animation-types/discrete-expected.txt:

Source/WebCore:

When interpolating between two Length values, if one is "auto", we should use the from-value
from 0 and up to (but excluding) 0.5, and use the to-value from 0.5 to 1.

This change caused a regression in the legacy animation engine since it would create a CSS
transition even when the underlying and target values were non-interpolable. As such, the
underlying value would be used until the transition's mid-point and the tests at
legacy-animation-engine/imported/blink/transitions/transition-not-interpolable.html and
legacy-animation-engine/fast/animation/height-auto-transition-computed-value.html would fail
expecting the target value to be used immediately. We now ensure that no transition is actually
started if two values for a given property cannot be interpolated.

* page/animation/CompositeAnimation.cpp:
(WebCore::CompositeAnimation::updateTransitions):
* platform/Length.cpp:
(WebCore::blend):

LayoutTests:

Make two more tests opt into the new animation engine since they pass and they're not in the legacy-animation-engine directory.
A third test now has some logging due to transitions not actually running, which is expected and correct.

* fast/animation/height-auto-transition-computed-value.html:
* imported/blink/transitions/transition-not-interpolable.html:
* legacy-animation-engine/transitions/transition-to-from-auto-expected.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@233892 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 months agoAdd an SPI hook to allow clients to yield document parsing and script execution
wenson_hsieh@apple.com [Tue, 17 Jul 2018 18:23:36 +0000 (18:23 +0000)]
Add an SPI hook to allow clients to yield document parsing and script execution
https://bugs.webkit.org/show_bug.cgi?id=187682
<rdar://problem/42207453>

Reviewed by Ryosuke Niwa.

Source/WebCore:

Using a single web process for both the Reader page and original web page on watchOS has multiple benefits,
including: (1) allowing the user to bail out of Reader and view the original web page without having to load it
again, and (2) improving the bringup time of the Reader page, since subresources are already cached in process
and we don't eat the additional cost of a web process launch if prewarming fails.

However, this has some drawbacks as well, one of which is that main thread work being done on behalf of the
original page may contend with work being done to load and render the Reader page. This is especially bad when
the page is in the middle of executing heavy script after Safari has already detected that the Reader version of
the page is available, but before it has finished loading the Reader page. The result is that script on the
original page may block the first paint of the Reader page (on New York Times articles, this often leads to an
apparent page load time of 25-35 seconds before the user sees anything besides a blank screen).

To mitigate this, we introduce a way for injected bundle clients to yield parsing and async script execution on
a document. This capability is surfaced in the form of an opaque token which clients may request from a
WKDOMDocument. Construction of the token causes the document to begin yielding and defer execution of previously
scheduled scripts, only if there were no active tokens on the document already. Similarly, destruction of all
active tokens on the document causes it to stop yielding and resume execution of scripts if needed.

Tests:  ParserYieldTokenTests.PreventDocumentLoadByTakingParserYieldToken
        ParserYieldTokenTests.TakeMultipleParserYieldTokens
        ParserYieldTokenTests.DeferredScriptExecutesBeforeDocumentLoadWhenTakingParserYieldToken
        ParserYieldTokenTests.AsyncScriptRunsWhenFetched

* dom/Document.cpp:
(WebCore::Document::implicitOpen):

If the parser yield token was taken before the document's parser was created, tell the parser's scheduler to
start yielding immediately after creation.

(WebCore::DocumentParserYieldToken::DocumentParserYieldToken):
(WebCore::DocumentParserYieldToken::~DocumentParserYieldToken):
* dom/Document.h:

Introduce a parser yield count to Document; as long as this count is greater than 0, we consider the Document to
have active yield tokens. When constructing or destroying a ParserYieldToken, we increment and decrement the
parser yield count (respectively).

(WebCore::Document::createParserYieldToken):
(WebCore::Document::hasActiveParserYieldToken const):
* dom/DocumentParser.h:
(WebCore::DocumentParser::didBeginYieldingParser):
(WebCore::DocumentParser::didEndYieldingParser):

Hooks for Document to tell its parser that we've started or finished yielding. This updates a flag on the
parser's scheduler which is consulted when we determine whether to yield before a pumping token or executing
script.

* dom/ScriptRunner.cpp:
(WebCore::ScriptRunner::resume):
(WebCore::ScriptRunner::notifyFinished):
* dom/ScriptRunner.h:
(WebCore::ScriptRunner::didBeginYieldingParser):
(WebCore::ScriptRunner::didEndYieldingParser):

Hooks for Document to tell its ScriptRunner that we've started or finished yielding. These wrap calls to suspend
and resume.

* html/parser/HTMLDocumentParser.cpp:
(WebCore::HTMLDocumentParser::didBeginYieldingParser):
(WebCore::HTMLDocumentParser::didEndYieldingParser):

Plumb to didBegin/didEnd calls to the HTMLParserScheduler.

* html/parser/HTMLDocumentParser.h:
* html/parser/HTMLParserScheduler.cpp:
(WebCore::HTMLParserScheduler::shouldYieldBeforeExecutingScript):
* html/parser/HTMLParserScheduler.h:
(WebCore::HTMLParserScheduler::shouldYieldBeforeToken):

Consult a flag when determining whether to yield. This flag is set to true only while the document has an active
parser yield token.

(WebCore::HTMLParserScheduler::isScheduledForResume const):

Consider the parser scheduler to be scheduled for resume if there are active tokens. Without this change, we
incorrectly consider the document to be finished loading when we have yield tokens, since it appears that the
parser is no longer scheduled to pump its tokenizer.

(WebCore::HTMLParserScheduler::didBeginYieldingParser):
(WebCore::HTMLParserScheduler::didEndYieldingParser):

When the Document begins yielding due to the documet having active tokens or ends yielding after the document
loses all of its yield tokens, update a flag on the parser scheduler. After we finish yielding, additionally
reschedule the parser if needed to ensure that we continue parsing the document; without this additional change
to resume, we'll never get the document load or load events after relinquishing the yield token.

Source/WebKit:

Add hooks to WKDOMDocument to create and return an internal WKDOMDocumentParserYieldToken object, whose lifetime
is tied to a document parser yield token. See WebCore ChangeLog for more detail.

* WebProcess/InjectedBundle/API/mac/WKDOMDocument.h:
* WebProcess/InjectedBundle/API/mac/WKDOMDocument.mm:
(-[WKDOMDocumentParserYieldToken initWithDocument:]):
(-[WKDOMDocument parserYieldToken]):

Tools:

Add a few tests to exercise the new document yield token SPI, verifying that clients can use the SPI to defer
document load, and that doing so doesn't cause deferred `script` to execute in the wrong order (i.e. before
synchronous script, or after "DOMContentLoaded").

* TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj:
* TestWebKitAPI/Tests/WebKitCocoa/ParserYieldTokenPlugIn.mm: Added.
(-[ParserYieldTokenPlugIn takeDocumentParserTokenAfterCommittingLoad]):
(-[ParserYieldTokenPlugIn releaseDocumentParserToken]):
(-[ParserYieldTokenPlugIn webProcessPlugInBrowserContextController:didCommitLoadForFrame:]):
(-[ParserYieldTokenPlugIn webProcessPlugIn:didCreateBrowserContextController:]):
(-[ParserYieldTokenPlugIn webProcessPlugInBrowserContextController:didFinishDocumentLoadForFrame:]):
(-[ParserYieldTokenPlugIn webProcessPlugInBrowserContextController:didFinishLoadForFrame:]):

Add an injected bundle object that knows how to take and release multiple document parser yield tokens.

* TestWebKitAPI/Tests/WebKitCocoa/ParserYieldTokenTests.h: Added.
* TestWebKitAPI/Tests/WebKitCocoa/ParserYieldTokenTests.mm: Added.
(+[ParserYieldTokenTestWebView webView]):
(-[ParserYieldTokenTestWebView bundle]):
(-[ParserYieldTokenTestWebView schemeHandler]):
(-[ParserYieldTokenTestWebView didFinishDocumentLoad]):
(-[ParserYieldTokenTestWebView didFinishLoad]):
(waitForDelay):
(TEST):
* TestWebKitAPI/Tests/WebKitCocoa/TestURLSchemeHandler.h: Added.
* TestWebKitAPI/Tests/WebKitCocoa/TestURLSchemeHandler.mm: Added.
(-[TestURLSchemeHandler webView:startURLSchemeTask:]):
(-[TestURLSchemeHandler webView:stopURLSchemeTask:]):
(-[TestURLSchemeHandler setStartURLSchemeTaskHandler:]):
(-[TestURLSchemeHandler startURLSchemeTaskHandler]):
(-[TestURLSchemeHandler setStopURLSchemeTaskHandler:]):
(-[TestURLSchemeHandler stopURLSchemeTaskHandler]):

Add a new test helper class to handle custom schemes via a block-based API.

* TestWebKitAPI/Tests/WebKitCocoa/text-with-async-script.html: Added.

New test HTML page that contains a deferred script element, a synchronous script element, another deferred
script element, and then some text, images, and links.

* TestWebKitAPI/Tests/WebKitCocoa/text-with-deferred-script.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@233891 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 months agoAdding myself to Contributors.json
tsavell@apple.com [Tue, 17 Jul 2018 18:21:55 +0000 (18:21 +0000)]
Adding myself to Contributors.json

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@233890 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 months ago[macOS] TestWebKitAPI.PictureInPicture.WKUIDelegate is timing out
commit-queue@webkit.org [Tue, 17 Jul 2018 18:04:46 +0000 (18:04 +0000)]
[macOS] TestWebKitAPI.PictureInPicture.WKUIDelegate is timing out
https://bugs.webkit.org/show_bug.cgi?id=187730

Patch by Aditya Keerthi <akeerthi@apple.com> on 2018-07-17
Reviewed by Jer Noble.

This regression was introduced by r233865. PIP can now only be initiated from a
window that is on screen.

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

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@233889 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 months agoAdd completion handlers to TestRunner functions setStatisticsLastSeen(), setStatistic...
wilander@apple.com [Tue, 17 Jul 2018 17:45:41 +0000 (17:45 +0000)]
Add completion handlers to TestRunner functions setStatisticsLastSeen(), setStatisticsPrevalentResource(), setStatisticsVeryPrevalentResource(), setStatisticsHasHadUserInteraction(), and setStatisticsHasHadNonRecentUserInteraction()
https://bugs.webkit.org/show_bug.cgi?id=187710
<rdar://problem/42252757>

Reviewed by Chris Dumez.

Source/WebKit:

These changes are to back the completion handler functionality of
TestRunner functions:
- setStatisticsLastSeen(),
- setStatisticsPrevalentResource(),
- setStatisticsVeryPrevalentResource(),
- setStatisticsHasHadUserInteraction(), and
- setStatisticsHasHadNonRecentUserInteraction().

* UIProcess/API/C/WKWebsiteDataStoreRef.cpp:
(WKWebsiteDataStoreSetStatisticsLastSeen):
(WKWebsiteDataStoreSetStatisticsPrevalentResource):
(WKWebsiteDataStoreSetStatisticsVeryPrevalentResource):
(WKWebsiteDataStoreSetStatisticsHasHadUserInteraction):
(WKWebsiteDataStoreSetStatisticsHasHadNonRecentUserInteraction):
* UIProcess/API/C/WKWebsiteDataStoreRef.h:
* UIProcess/WebResourceLoadStatisticsStore.cpp:
(WebKit::WebResourceLoadStatisticsStore::logUserInteraction):
(WebKit::WebResourceLoadStatisticsStore::logNonRecentUserInteraction):
(WebKit::WebResourceLoadStatisticsStore::clearUserInteraction):
(WebKit::WebResourceLoadStatisticsStore::setLastSeen):
(WebKit::WebResourceLoadStatisticsStore::setPrevalentResource):
(WebKit::WebResourceLoadStatisticsStore::setVeryPrevalentResource):
(WebKit::WebResourceLoadStatisticsStore::clearPrevalentResource):
* UIProcess/WebResourceLoadStatisticsStore.h:

Tools:

* WebKitTestRunner/InjectedBundle/Bindings/TestRunner.idl:
* WebKitTestRunner/InjectedBundle/InjectedBundle.cpp:
(WTR::InjectedBundle::didReceiveMessageToPage):
* WebKitTestRunner/InjectedBundle/TestRunner.cpp:
(WTR::TestRunner::setStatisticsLastSeen):
(WTR::TestRunner::statisticsCallDidSetLastSeenCallback):
(WTR::TestRunner::setStatisticsPrevalentResource):
(WTR::TestRunner::statisticsCallDidSetPrevalentResourceCallback):
(WTR::TestRunner::setStatisticsVeryPrevalentResource):
(WTR::TestRunner::statisticsCallDidSetVeryPrevalentResourceCallback):
(WTR::TestRunner::setStatisticsHasHadUserInteraction):
(WTR::TestRunner::setStatisticsHasHadNonRecentUserInteraction):
(WTR::TestRunner::statisticsCallDidSetHasHadUserInteractionCallback):
* WebKitTestRunner/InjectedBundle/TestRunner.h:
* WebKitTestRunner/TestController.cpp:
(WTR::TestController::setStatisticsLastSeen):
(WTR::TestController::setStatisticsPrevalentResource):
(WTR::TestController::setStatisticsVeryPrevalentResource):
(WTR::TestController::setStatisticsHasHadUserInteraction):
(WTR::TestController::setStatisticsHasHadNonRecentUserInteraction):
* WebKitTestRunner/TestInvocation.cpp:
(WTR::TestInvocation::didSetLastSeen):
(WTR::TestInvocation::didSetPrevalentResource):
(WTR::TestInvocation::didSetVeryPrevalentResource):
(WTR::TestInvocation::didSetHasHadUserInteraction):
(WTR::TestInvocation::didSetHasHadNonRecentUserInteraction):
* WebKitTestRunner/TestInvocation.h:

LayoutTests:

These changes are to update all test cases that make use of
TestRunner functions:
- setStatisticsLastSeen(),
- setStatisticsPrevalentResource(),
- setStatisticsVeryPrevalentResource(),
- setStatisticsHasHadUserInteraction(), and
- setStatisticsHasHadNonRecentUserInteraction().

* http/tests/resourceLoadStatistics/add-blocking-to-redirect.html:
* http/tests/resourceLoadStatistics/add-partitioning-to-redirect.html:
* http/tests/resourceLoadStatistics/classify-as-non-prevalent-based-on-mixed-statistics.html:
* http/tests/resourceLoadStatistics/classify-as-non-prevalent-based-on-sub-frame-under-top-frame-origins.html:
* http/tests/resourceLoadStatistics/classify-as-non-prevalent-based-on-subresource-under-top-frame-origins.html:
* http/tests/resourceLoadStatistics/classify-as-non-prevalent-based-on-subresource-unique-redirects-to.html:
* http/tests/resourceLoadStatistics/classify-as-prevalent-based-on-mixed-statistics.html:
* http/tests/resourceLoadStatistics/classify-as-prevalent-based-on-sub-frame-under-top-frame-origins.html:
* http/tests/resourceLoadStatistics/classify-as-prevalent-based-on-subresource-redirect-collusion.html:
* http/tests/resourceLoadStatistics/classify-as-prevalent-based-on-subresource-redirect-to-prevalent.html:
* http/tests/resourceLoadStatistics/classify-as-prevalent-based-on-subresource-under-top-frame-origins.html:
* http/tests/resourceLoadStatistics/classify-as-prevalent-based-on-subresource-unique-redirects-to.html:
* http/tests/resourceLoadStatistics/classify-as-prevalent-based-on-top-frame-redirect-collusion.html:
* http/tests/resourceLoadStatistics/classify-as-prevalent-based-on-top-frame-redirect-to-prevalent.html:
* http/tests/resourceLoadStatistics/classify-as-prevalent-based-on-top-frame-unique-redirects-to.html:
* http/tests/resourceLoadStatistics/classify-as-very-prevalent-based-on-mixed-statistics.html:
* http/tests/resourceLoadStatistics/clear-in-memory-and-persistent-store-one-hour.html:
* http/tests/resourceLoadStatistics/clear-in-memory-and-persistent-store.html:
* http/tests/resourceLoadStatistics/do-not-block-top-level-navigation-redirect.html:
* http/tests/resourceLoadStatistics/grandfathering.html:
* http/tests/resourceLoadStatistics/non-prevalent-resource-with-user-interaction.html:
* http/tests/resourceLoadStatistics/non-prevalent-resource-without-user-interaction.html:
* http/tests/resourceLoadStatistics/non-prevalent-resources-can-access-cookies-in-a-third-party-context.html:
* http/tests/resourceLoadStatistics/partitioned-and-unpartitioned-cookie-deletion.html:
* http/tests/resourceLoadStatistics/partitioned-and-unpartitioned-cookie-with-partitioning-timeout-expected.txt:
* http/tests/resourceLoadStatistics/partitioned-and-unpartitioned-cookie-with-partitioning-timeout.html:
* http/tests/resourceLoadStatistics/partitioned-cookies-with-and-without-user-interaction.html:
* http/tests/resourceLoadStatistics/prevalent-resource-handled-keydown.html:
* http/tests/resourceLoadStatistics/prevalent-resource-unhandled-keydown.html:
* http/tests/resourceLoadStatistics/prevalent-resource-with-user-interaction-timeout.html:
* http/tests/resourceLoadStatistics/prevalent-resource-with-user-interaction.html:
* http/tests/resourceLoadStatistics/prevalent-resource-without-user-interaction.html:
* http/tests/resourceLoadStatistics/prune-statistics.html:
* http/tests/resourceLoadStatistics/remove-blocking-in-redirect.html:
* http/tests/resourceLoadStatistics/remove-partitioning-in-redirect.html:
* http/tests/resourceLoadStatistics/strip-referrer-to-origin-for-prevalent-subresource-redirects.html:
* http/tests/resourceLoadStatistics/strip-referrer-to-origin-for-prevalent-subresource-requests.html:
* http/tests/resourceLoadStatistics/telemetry-generation.html:
* http/tests/resourceLoadStatistics/third-party-cookie-with-and-without-user-interaction.html:
* http/tests/resourceLoadStatistics/user-interaction-only-reported-once-within-short-period-of-time.html:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@233888 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 months agoRebaseline displaylists/extent-includes-* tests for mac-wk1 after r233869.
ryanhaddad@apple.com [Tue, 17 Jul 2018 16:59:05 +0000 (16:59 +0000)]
Rebaseline displaylists/extent-includes-* tests for mac-wk1 after r233869.
https://bugs.webkit.org/show_bug.cgi?id=187574

Unreviewed test gardening.

* platform/mac-wk1/displaylists/extent-includes-shadow-expected.txt:
* platform/mac-wk1/displaylists/extent-includes-transforms-expected.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@233887 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 months ago[clip-path] Implement support for margin-box as reference box and box shape
krit@webkit.org [Tue, 17 Jul 2018 16:52:04 +0000 (16:52 +0000)]
[clip-path] Implement support for margin-box as reference box and box shape
https://bugs.webkit.org/show_bug.cgi?id=127984

Reviewed by Simon Fraser.

Compute the margin-box rectangle as needed for clip-path based on the actual
computed values for the margin-top, *-left, *-bottom, *-right properties.

Source/WebCore:

Test: css3/masking/clip-path-margin-box.html

* rendering/RenderBox.h:
(WebCore::RenderBox::marginBoxRect const):
* rendering/RenderBoxModelObject.h:
* rendering/RenderLayer.cpp:
(WebCore::computeReferenceBox):

LayoutTests:

* css3/masking/clip-path-circle-margin-box-expected.html: Added.
* css3/masking/clip-path-margin-box-expected.html: Added.
* css3/masking/clip-path-margin-box.html: Added.
* platform/mac/css3/masking/clip-path-circle-margin-box-expected.png: Removed.
* platform/mac/css3/masking/clip-path-circle-margin-box-expected.txt: Removed.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@233886 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 months agoDelete content of a single cell table should not delete the whole table
jfernandez@igalia.com [Tue, 17 Jul 2018 08:48:49 +0000 (08:48 +0000)]
Delete content of a single cell table should not delete the whole table
https://bugs.webkit.org/show_bug.cgi?id=173117

Reviewed by Ryosuke Niwa.

Source/WebCore:

We should not extend selection looking for special elements if the
delete operation has been triggered by a caret based selection.

This change is based on a recent [1] resolution of the Editing TF,
which acknowledges that behavior of single-cell tables must be the
same that multi-cell tables and even if the last character is
deleted, we should not delete the whole table structure.

A different case would be when the user actively selects the whole
content of a table; in this case, as we do in multi-cell tables,
the structure of single-cell tables should be deleted together
with the content.

[1] https://github.com/w3c/editing/issues/163

Tests: editing/deleting/backspace-delete-last-char-in-table.html
       editing/deleting/forward-delete-last-char-in-table.html
       editing/deleting/select-and-delete-last-char-in-table.html

* editing/TypingCommand.cpp:
(WebCore::TypingCommand::deleteKeyPressed):
(WebCore::TypingCommand::forwardDeleteKeyPressed):

LayoutTests:

Tests to verify that single-cell tables are not deleted when their
last character is deleted, unless it was previously selected by
the user.

Changes two expected files to adapt them to the new logic.

* LayoutTests/editing/deleting/deleting-relative-positioned-special-element-expected.txt: The paragraph is not deleted, even if it's empty. The paragraphs above are not merged, which was the goal of the test.
* editing/deleting/delete-last-char-in-table-expected.txt: The table is not removed, even if it's empty. The formatted elements are deleted, which was the goal of the test.
* editing/deleting/backspace-delete-last-char-in-table-expected.txt: Added.
* editing/deleting/backspace-delete-last-char-in-table.html: Added.
* editing/deleting/forward-delete-last-char-in-table-expected.txt: Added.
* editing/deleting/forward-delete-last-char-in-table.html: Added.
* editing/deleting/select-and-delete-last-char-in-table-expected.txt: Added.
* editing/deleting/select-and-delete-last-char-in-table.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@233885 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 months agoCustomConfigurationTestGroupForm should dispatch different arguments based on whether...
dewei_zhu@apple.com [Tue, 17 Jul 2018 05:07:34 +0000 (05:07 +0000)]
CustomConfigurationTestGroupForm should dispatch different arguments based on whether analysis task is created.
https://bugs.webkit.org/show_bug.cgi?id=187675

Reviewed by Ryosuke Niwa.

This change will fix the bug that no notification will be sent for any test groups except the
first one in any custom perf-try A/B task.

* public/v3/components/custom-configuration-test-group-form.js:
(CustomConfigurationTestGroupForm.prototype.startTesting): Conditionally includes taskName based on
whether or not analysis task is created.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@233884 268f45cc-cd09-0410-ab3c-d52691b4dbfc