WebKit-https.git
6 months agoNeed a mechanism to override navigator.userAgent
rniwa@webkit.org [Sat, 26 Jan 2019 06:50:58 +0000 (06:50 +0000)]
Need a mechanism to override navigator.userAgent
https://bugs.webkit.org/show_bug.cgi?id=193762
<rdar://problem/47504939>

Reviewed by Brent Fulgham.

Source/WebCore:

Added the ability to specify user agent string just for navigator.userAgent via DocumentLoader.

* loader/DocumentLoader.h:
(WebCore::DocumentLoader::setCustomJavaScriptUserAgent):
(WebCore::DocumentLoader::customJavaScriptUserAgent const):
* loader/FrameLoader.cpp:
(WebCore::FrameLoader::userAgentForJavaScript const):
* loader/FrameLoader.h:
* page/Navigator.cpp:
(WebCore::Navigator::userAgent const):

Source/WebKit:

This patch adds a new _WKWebsitePolicies SPI to specify the user agent string returned by
navigator.userAgent without affecting the user agent string used to send network requests.

Tests: WebKit.WebsitePoliciesCustomJavaScriptUserAgent
       WebKit.WebsitePoliciesCustomUserAgents

* Shared/WebsitePoliciesData.cpp:
(WebKit::WebsitePoliciesData::encode const):
(WebKit::WebsitePoliciesData::decode):
(WebKit::WebsitePoliciesData::applyToDocumentLoader):
* Shared/WebsitePoliciesData.h:
* UIProcess/API/APIWebsitePolicies.cpp:
(API::WebsitePolicies::data):
* UIProcess/API/APIWebsitePolicies.h:
* UIProcess/API/Cocoa/_WKWebsitePolicies.h:
* UIProcess/API/Cocoa/_WKWebsitePolicies.mm:
(-[_WKWebsitePolicies setCustomJavaScriptUserAgent:]):
(-[_WKWebsitePolicies customJavaScriptUserAgent]):

Tools:

Added test cases for _WKWebsitePolicies.customJavaScriptUserAgent.

* TestWebKitAPI/Tests/WebKitCocoa/WebsitePolicies.mm:
(-[CustomJavaScriptUserAgentDelegate _webView:decidePolicyForNavigationAction:userInfo:decisionHandler:]):
(-[CustomJavaScriptUserAgentDelegate webView:didFinishNavigation:]):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@240541 268f45cc-cd09-0410-ab3c-d52691b4dbfc

6 months agoWeb Inspector: provide a way to edit page settings on a remote target
drousso@apple.com [Sat, 26 Jan 2019 03:57:17 +0000 (03:57 +0000)]
Web Inspector: provide a way to edit page settings on a remote target
https://bugs.webkit.org/show_bug.cgi?id=193813
<rdar://problem/47359510>

Reviewed by Joseph Pecoraro.

Source/JavaScriptCore:

* inspector/protocol/Page.json:
Add `overrideSetting` command with supporting `Setting` enum type.

Source/WebCore:

Test: inspector/page/overrideSetting.html

* page/Settings.yaml:
* Scripts/GenerateSettings.rb:
* Scripts/SettingsTemplates/Settings.cpp.erb:
* Scripts/SettingsTemplates/Settings.h.erb:
Add support for an `inspectorOverride` boolean value for each setting that will take
precedence over the actual `Setting`'s value when set.

* inspector/agents/InspectorPageAgent.h:
* inspector/agents/InspectorPageAgent.cpp:
(WebCore::InspectorPageAgent::disable):
(WebCore::InspectorPageAgent::overrideSetting): Added.

* inspector/InspectorFrontendHost.idl:
* inspector/InspectorFrontendHost.h:
* inspector/InspectorFrontendHost.cpp:
(WebCore::InspectorFrontendHost::isRemote const): Added.
* inspector/InspectorFrontendClient.h:
(WebCore::InspectorFrontendClient::isRemote const): Added.
* inspector/InspectorFrontendClientLocal.h:
(WebCore::InspectorFrontendClientLocal::isRemote const): Added.

Source/WebInspectorUI:

Add toolbar button that shows a popover with the target's (page's) settings when clicked.

* UserInterface/Base/Main.js:
(WI.loaded):
(WI.contentLoaded):
(WI.initializeTarget): Added.
(WI._handleDeviceSettingsToolbarButtonClicked): Added.
(WI.didDismissPopover): Added.
* UserInterface/Views/Main.css:
(.device-settings-content): Added.
(.device-settings-content .columns): Added.
(.device-settings-content .columns > .column): Added.
(.device-settings-content .columns > .column + .column): Added.
(body[dir=ltr] .device-settings-content label > input): Added.
(body[dir=rtl] .device-settings-content label > input): Added.

* UserInterface/Views/Popover.js:
(WI.Popover.prototype._update.area):
(WI.Popover.prototype._update):
(WI.Popover.prototype._drawBackground):
(WI.Popover.prototype._bestMetricsForEdge):
(WI.Popover.prototype._drawFrame):
If the best area is negative, treat it as the worst area.
Allow areas to be clamped so long as the clamped edge is not the preferred edge.

* UserInterface/Base/Test.js:
(WI.initializeTarget): Added.

* UserInterface/Images/Device.svg: Added.
* Localizations/en.lproj/localizedStrings.js:

Source/WebKit:

* WebProcess/WebPage/WebInspectorUI.h:
(WebKit::WebInspectorUI::isRemote() const): Added.
* WebProcess/WebPage/RemoteWebInspectorUI.h:
(WebKit::RemoteWebInspectorUI::isRemote() const): Added.

LayoutTests:

* inspector/page/overrideSetting.html: Added.
* inspector/page/overrideSetting-expected.txt: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@240540 268f45cc-cd09-0410-ab3c-d52691b4dbfc

6 months agoDocument::updateMainArticleElementAfterLayout() should be a no-op when no client...
wenson_hsieh@apple.com [Sat, 26 Jan 2019 03:38:57 +0000 (03:38 +0000)]
Document::updateMainArticleElementAfterLayout() should be a no-op when no client depends on knowing the main article element
https://bugs.webkit.org/show_bug.cgi?id=193843

Reviewed by Zalan Bujtas.

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

This function currently does a bit of wasted work after every layout, on clients that don't listen to the
"significant rendered text" layout milestone and therefore don't need to guess the main article element. Simply
don't bother keeping the main article element up to date in this scenario by bailing from
FrameView::updateHasReachedSignificantRenderedTextThreshold if the client doesn't care about the significant
rendered text milestone.

* page/FrameView.cpp:
(WebCore::FrameView::updateHasReachedSignificantRenderedTextThreshold):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@240539 268f45cc-cd09-0410-ab3c-d52691b4dbfc

6 months agoWeb Inspector: Audit: unable to import audits
drousso@apple.com [Sat, 26 Jan 2019 03:24:03 +0000 (03:24 +0000)]
Web Inspector: Audit: unable to import audits
https://bugs.webkit.org/show_bug.cgi?id=193861

Reviewed by Joseph Pecoraro.

* UserInterface/Controllers/AuditManager.js:
(WI.AuditManager.prototype.async processJSON):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@240538 268f45cc-cd09-0410-ab3c-d52691b4dbfc

6 months ago<video> elements not in the DOM should be allowed to AirPlay
jer.noble@apple.com [Sat, 26 Jan 2019 01:46:50 +0000 (01:46 +0000)]
<video> elements not in the DOM should be allowed to AirPlay
https://bugs.webkit.org/show_bug.cgi?id=193837
Source/WebCore:

<rdar://42559491>

Reviewed by Eric Carlson.

Test: media/airplay-allows-buffering.html

Some websites will switch between <video> elements backed by MSE to one backed by
a media file in order to implement an AirPlay control. But when a <video> element is
removed from the DOM and paused, further buffering is blocked. For some ports (namely
Cocoa ones), this keeps AirPlay from engaging. Relax this buffering restriction for
elements who have been asked to play wirelessly, but whose wireless playback has not
started yet.

* html/MediaElementSession.cpp:
(WebCore::MediaElementSession::dataBufferingPermitted const):
(WebCore::MediaElementSession::setShouldPlayToPlaybackTarget):

LayoutTests:

Reviewed by Eric Carlson.

* media/airplay-allows-buffering-expected.txt: Added.
* media/airplay-allows-buffering.html: Added.
* platform/ios/TestExpectations:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@240537 268f45cc-cd09-0410-ab3c-d52691b4dbfc

6 months agoFix failing scrollingcoordinator/scrolling-tree/remove-scrolling-role.html
simon.fraser@apple.com [Sat, 26 Jan 2019 01:40:24 +0000 (01:40 +0000)]
Fix failing scrollingcoordinator/scrolling-tree/remove-scrolling-role.html

Test needed different macOS and iOS results. Added window.internals.settings.setAsyncOverflowScrollingEnabled(true)
so the test uses async overflow scrolling in the macOS result.

* platform/ios-wk2/scrollingcoordinator/scrolling-tree/remove-scrolling-role-expected.txt: Copied from LayoutTests/scrollingcoordinator/scrolling-tree/remove-scrolling-role-expected.txt.
* scrollingcoordinator/scrolling-tree/remove-scrolling-role-expected.txt:
* scrollingcoordinator/scrolling-tree/remove-scrolling-role.html:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@240536 268f45cc-cd09-0410-ab3c-d52691b4dbfc

6 months agoAdd a scrolling tree test that toggles scrollability on an intermediate iframe
simon.fraser@apple.com [Sat, 26 Jan 2019 01:40:20 +0000 (01:40 +0000)]
Add a scrolling tree test that toggles scrollability on an intermediate iframe
https://bugs.webkit.org/show_bug.cgi?id=193849

Reviewed by Dean Jackson.

* platform/ios-wk2/scrollingcoordinator/scrolling-tree/toggle-coordinated-frame-scrolling-expected.txt: Added.
* scrollingcoordinator/scrolling-tree/resources/scroll-toggling-frame.html: Added.
* scrollingcoordinator/scrolling-tree/toggle-coordinated-frame-scrolling-expected.txt: Added.
* scrollingcoordinator/scrolling-tree/toggle-coordinated-frame-scrolling.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@240535 268f45cc-cd09-0410-ab3c-d52691b4dbfc

6 months agoWeb Inspector: Uncaught Exception: No node with given id found
drousso@apple.com [Sat, 26 Jan 2019 01:06:51 +0000 (01:06 +0000)]
Web Inspector: Uncaught Exception: No node with given id found
https://bugs.webkit.org/show_bug.cgi?id=193833

Reviewed by Joseph Pecoraro.

* UserInterface/Views/CanvasContentView.js:
(WI.CanvasContentView):
(WI.CanvasContentView.prototype.initialLayout):
(WI.CanvasContentView.prototype.layout):
(WI.CanvasContentView.prototype._refreshPixelSize):
Drive-by: show the refresh button when viewing a specific canvas.
* UserInterface/Views/CanvasTabContentView.js:
(WI.CanvasTabContentView.prototype._removeCanvas):
Reset to the overview if the canvas is removed.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@240534 268f45cc-cd09-0410-ab3c-d52691b4dbfc

6 months agoREGRESSION: Some USDz from 3rd party websites don't go directly to AR QL
dino@apple.com [Sat, 26 Jan 2019 01:00:54 +0000 (01:00 +0000)]
REGRESSION: Some USDz from 3rd party websites don't go directly to AR QL
https://bugs.webkit.org/show_bug.cgi?id=193831
<rdar://problem/47399263>

Reviewed by Chris Dumez.

Source/WebKit:

A System Preview (<a rel="ar">) displays in a modal and doesn't trigger
a navigation. If the link was cross origin, it was causing a process swap,
which meant that the response defaulted back to a navigation.

The fix is to not cause a PSON when the navigation is a system preview.

* UIProcess/API/APINavigation.h:
(API::Navigation::shouldForceDownload const): This is now just tracking
the "download" attribute, and not including System Preview.
(API::Navigation::isSystemPreview const): New method to check for a
navigation triggered as a System Preview.
* UIProcess/WebPageProxy.cpp: Move the code from receivedPolicyDecision to
receivedNavigationPolicyDecision, so that downloads and System Previews are
detected before we decide to change process.
(WebKit::WebPageProxy::receivedNavigationPolicyDecision):
(WebKit::WebPageProxy::receivedPolicyDecision):

Tools:

Two new tests that exercise cross-origin and same-origin System
Previews.

* TestWebKitAPI/Tests/WebKitCocoa/ProcessSwapOnNavigation.mm:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@240533 268f45cc-cd09-0410-ab3c-d52691b4dbfc

6 months agoRemove expectation for inspector/css/createStyleSheet.html as test appears to be...
tsavell@apple.com [Sat, 26 Jan 2019 00:54:08 +0000 (00:54 +0000)]
Remove expectation for inspector/css/createStyleSheet.html as test appears to be passing.
https://bugs.webkit.org/show_bug.cgi?id=148636

Unreviewed test gardening.

* platform/mac/TestExpectations:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@240532 268f45cc-cd09-0410-ab3c-d52691b4dbfc

6 months agoFind-in-page on nyt.com scrolls around without touching the screen when find holes...
timothy_horton@apple.com [Sat, 26 Jan 2019 00:44:42 +0000 (00:44 +0000)]
Find-in-page on nyt.com scrolls around without touching the screen when find holes are visible
https://bugs.webkit.org/show_bug.cgi?id=193853

Reviewed by Simon Fraser.

* WebProcess/WebPage/ios/FindControllerIOS.mm:
(WebKit::FindController::updateFindIndicator):
There is no reason to scroll/zoom to the find indicator just because the
holes are up, we should only do it on initial indicator presentation.
This was a mistake in r178755 that was previously never exercised because
isShowingOverlay was always false.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@240531 268f45cc-cd09-0410-ab3c-d52691b4dbfc

6 months ago[iOS] Add support for the download attribute
cdumez@apple.com [Sat, 26 Jan 2019 00:44:15 +0000 (00:44 +0000)]
[iOS] Add support for the download attribute
https://bugs.webkit.org/show_bug.cgi?id=167341
<rdar://problem/30296281>

Reviewed by Geoffrey Garen.

Source/WTF:

Turn on HTML download attribute support on iOS.

* wtf/FeatureDefines.h:

LayoutTests:

Make download attribute layout tests iOS-friendly and unskip them now that they pass.

* fast/dom/HTMLAnchorElement/anchor-download-unset.html:
* fast/dom/HTMLAnchorElement/anchor-download-user-triggered-synthetic-click.html:
* fast/dom/HTMLAnchorElement/anchor-file-blob-convert-to-download-async-delegate.html:
* fast/dom/HTMLAnchorElement/anchor-file-blob-convert-to-download.html:
* fast/dom/HTMLAnchorElement/anchor-file-blob-download-blank-base-target-popup-not-allowed.html:
* fast/dom/HTMLAnchorElement/anchor-file-blob-download-blank-target-popup-not-allowed.html:
* fast/dom/HTMLAnchorElement/anchor-file-blob-download-blank-target.html:
* fast/dom/HTMLAnchorElement/anchor-file-blob-download-includes-backslash.html:
* fast/dom/HTMLAnchorElement/anchor-file-blob-download-includes-doublequote.html:
* fast/dom/HTMLAnchorElement/anchor-file-blob-download-includes-slashes.html:
* fast/dom/HTMLAnchorElement/anchor-file-blob-download-includes-unicode.html:
* fast/dom/HTMLAnchorElement/anchor-file-blob-download-no-extension.html:
* fast/dom/HTMLAnchorElement/anchor-file-blob-download.html:
* fast/dom/HTMLAnchorElement/anchor-nodownload-set.html:
* fast/dom/HTMLAnchorElement/anchor-nodownload.html:
* http/tests/download/anchor-download-attribute-content-disposition.html:
* http/tests/download/anchor-download-no-extension.html:
* http/tests/download/anchor-download-no-value.html:
* http/tests/download/anchor-download-redirect.html:
* http/tests/download/area-download.html:
* http/tests/download/convert-cached-load-to-download.html:
* http/tests/security/anchor-download-allow-blob.html:
* http/tests/security/anchor-download-allow-data.html:
* http/tests/security/anchor-download-allow-sameorigin.html:
* http/tests/security/anchor-download-block-crossorigin-expected.txt:
* http/tests/security/anchor-download-block-crossorigin.html:
* platform/ios-wk2/TestExpectations:
* platform/ios-wk2/http/tests/security/anchor-download-block-crossorigin-expected.txt: Copied from LayoutTests/http/tests/security/anchor-download-block-crossorigin-expected.txt.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@240530 268f45cc-cd09-0410-ab3c-d52691b4dbfc

6 months agoUpdate Xcode projects with "Check .xcfilelists" build phase
krollin@apple.com [Fri, 25 Jan 2019 23:52:51 +0000 (23:52 +0000)]
Update Xcode projects with "Check .xcfilelists" build phase
https://bugs.webkit.org/show_bug.cgi?id=193790
<rdar://problem/47201374>

Reviewed by Alex Christensen.

Support for XCBuild includes specifying inputs and outputs to various
Run Script build phases. These inputs and outputs are specified as
.xcfilelist files. Once created, these .xcfilelist files need to be
kept up-to-date. In order to check that they are up-to-date or not,
add an Xcode build step that invokes an external script that performs
the checking. If the .xcfilelists are found to be out-of-date, update
them, halt the build, and instruct the developer to restart the build
with up-to-date files.

At this time, the checking and regenerating is performed only if the
WK_ENABLE_CHECK_XCFILELISTS environment variable is set to 1. People
who want to use this facility can set this variable and test out the
checking/regenerating. Once it seems like there are no egregious
issues that upset a developer's workflow, we'll unconditionally enable
this facility.

Source/JavaScriptCore:

* JavaScriptCore.xcodeproj/project.pbxproj:
* Scripts/check-xcfilelists.sh: Added.

Source/WebCore:

No new tests since there should be no observable behavior difference.

* Scripts/check-xcfilelists.sh: Added.
* WebCore.xcodeproj/project.pbxproj:

Source/WebKit:

* Scripts/check-xcfilelists.sh: Added.
* WebKit.xcodeproj/project.pbxproj:

Tools:

* DumpRenderTree/DumpRenderTree.xcodeproj/project.pbxproj:
* DumpRenderTree/Scripts/check-xcfilelists.sh: Added.
* WebKitTestRunner/Scripts/check-xcfilelists.sh: Added.
* WebKitTestRunner/WebKitTestRunner.xcodeproj/project.pbxproj:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@240525 268f45cc-cd09-0410-ab3c-d52691b4dbfc

6 months agoWeb Inspector: Exclude Debugger Threads from CPU Usage values in Web Inspector
commit-queue@webkit.org [Fri, 25 Jan 2019 23:45:04 +0000 (23:45 +0000)]
Web Inspector: Exclude Debugger Threads from CPU Usage values in Web Inspector
https://bugs.webkit.org/show_bug.cgi?id=193796
<rdar://problem/47532910>

Patch by Joseph Pecoraro <pecoraro@apple.com> on 2019-01-25
Reviewed by Devin Rousso.

Source/JavaScriptCore:

* runtime/SamplingProfiler.cpp:
(JSC::SamplingProfiler::machThread):
* runtime/SamplingProfiler.h:
Expose the mach_port_t of the SamplingProfiler thread
so it can be tested against later.

Source/WebCore:

* page/ResourceUsageData.h:
* inspector/agents/InspectorCPUProfilerAgent.cpp:
(WebCore::InspectorCPUProfilerAgent::collectSample):
Show the CPU usage without debugger threads in the Web Inspector's timeline.

* page/ResourceUsageThread.h:
* page/cocoa/ResourceUsageThreadCocoa.mm:
(WebCore::ResourceUsageThread::platformSaveStateBeforeStarting):
For OS(DARWIN) ports, when starting to observe resource usage,
we grab the mach_port_t of SamplingProfiler on the main thread
in a thread safe way. For our purposes (Web Inspector timelines),
this will be good enough to identify the SamplingProfiler thread
during timeline recording. The SamplingProfiler thread won't change
during a timeline recording and recording start/stops will never
miss the SamplingProfiler changing.

(WebCore::filterThreads):
(WebCore::threadSendRights):
(WebCore::threadSendRightsExcludingDebuggerThreads):
(WebCore::cpuUsage):
(WebCore::ResourceUsageThread::platformCollectCPUData):
Calculate CPU usage twice, the second time excluding some threads.

* page/linux/ResourceUsageThreadLinux.cpp:
(WebCore::ResourceUsageThread::platformSaveStateBeforeStarting):
(WebCore::ResourceUsageThread::platformCollectCPUData):
Stubs for linux ports.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@240522 268f45cc-cd09-0410-ab3c-d52691b4dbfc

6 months agoFix Windows build after r240511
achristensen@apple.com [Fri, 25 Jan 2019 23:44:27 +0000 (23:44 +0000)]
Fix Windows build after r240511

* bytecode/UnlinkedFunctionExecutable.cpp:
(JSC::UnlinkedFunctionExecutable::unlinkedCodeBlockFor):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@240521 268f45cc-cd09-0410-ab3c-d52691b4dbfc

6 months agoMove logic from ChildProcess::setApplicationIsDaemon to NetworkProcess::initializeProcess
achristensen@apple.com [Fri, 25 Jan 2019 23:41:30 +0000 (23:41 +0000)]
Move logic from ChildProcess::setApplicationIsDaemon to NetworkProcess::initializeProcess
https://bugs.webkit.org/show_bug.cgi?id=193844

Reviewed by Andy Estes.

Source/WebCore/PAL:

* pal/spi/mac/HIServicesSPI.h:

Source/WebKit:

This simplifies ChildProcess and removes NetworkProcess-specific code from this generic abstraction.

* NetworkProcess/mac/NetworkProcessMac.mm:
(WebKit::NetworkProcess::initializeProcess):
* Shared/ios/ChildProcessIOS.mm:
(WebKit::ChildProcess::setApplicationIsDaemon): Deleted.
* Shared/mac/ChildProcessMac.mm:
(WebKit::ChildProcess::setApplicationIsDaemon): Deleted.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@240520 268f45cc-cd09-0410-ab3c-d52691b4dbfc

6 months agoRemove FrameView::m_significantRenderedTextMilestonePending
zalan@apple.com [Fri, 25 Jan 2019 23:40:23 +0000 (23:40 +0000)]
Remove FrameView::m_significantRenderedTextMilestonePending
https://bugs.webkit.org/show_bug.cgi?id=193842

Reviewed by Wenson Hsieh.

Currently we keep processing the incoming text content until after the "SignificantRenderedTextMilestone" has been reached.
We can actually stop doing it right when the text content is above the threshold (regardless of whether all the conditions are met for the milestone).
This patch also ensures that we don't update Document::m_mainArticleElement once the threshold is reached.

* page/FrameView.cpp:
(WebCore::FrameView::resetLayoutMilestones):
(WebCore::FrameView::incrementVisuallyNonEmptyCharacterCount):
(WebCore::FrameView::hasReachedSignificantRenderedTextThreashold):
(WebCore::FrameView::qualifiesAsSignificantRenderedText const):
(WebCore::FrameView::fireLayoutRelatedMilestonesIfNeeded):
(WebCore::FrameView::updateSignificantRenderedTextMilestoneIfNeeded): Deleted.
* page/FrameView.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@240519 268f45cc-cd09-0410-ab3c-d52691b4dbfc

6 months agoWeb Inspector: Improve Dark Mode appearance within Memory timeline
commit-queue@webkit.org [Fri, 25 Jan 2019 23:31:17 +0000 (23:31 +0000)]
Web Inspector: Improve Dark Mode appearance within Memory timeline
https://bugs.webkit.org/show_bug.cgi?id=193804

Patch by Joseph Pecoraro <pecoraro@apple.com> on 2019-01-25
Reviewed by Matt Baker.

* UserInterface/Views/TimelineOverview.css:
(.timeline-overview.edit-instruments > .tree-outline.timelines .item.selected):
(.timeline-overview > .tree-outline.timelines .item.selected + .item,):
(@media (prefers-color-scheme: dark)):
(.timeline-overview > .tree-outline.timelines .item.selected + .item): Deleted.
(body.window-inactive .timeline-overview > .tree-outline.timelines .item.selected + .item): Deleted.
Fix some colors for Timelines edit mode.

* UserInterface/Views/CPUTimelineOverviewGraph.css:
(.timeline-overview-graph.cpu:nth-child(even) > .legend):
(@media (prefers-color-scheme: dark)):
(.timeline-overview-graph.cpu > .legend):
(.timeline-overview-graph:nth-child(even) > .legend): Deleted.
* UserInterface/Views/MemoryTimelineOverviewGraph.css:
(.timeline-overview-graph.memory:nth-child(even) > .legend):
(@media (prefers-color-scheme: dark)):
(.timeline-overview-graph.memory > .legend):
(.timeline-overview-graph:nth-child(even) > .legend): Deleted.
Improved colors in CPU / Memory overview graph legends.

* UserInterface/Views/CPUUsageView.css:
(.cpu-usage-view > .details):
* UserInterface/Views/CircleChart.css:
(.circle-chart > svg > path.background):
* UserInterface/Views/MemoryCategoryView.css:
(.memory-category-view > .details):
(.memory-category-view > .details > .name):
* UserInterface/Views/MemoryTimelineView.css:
(.timeline-view.memory > .content > .overview .total-usage,):
(.timeline-view.memory .legend > .row > .size):
Improved colors in CPU / Memory detail views.

* UserInterface/Views/Variables.css:
(:root):
Add a default --text-secondary-color which will end up slightly
lighter in dark mode where it was already implemented.

(@media (prefers-color-scheme: dark)):
Improved max-comparison colors.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@240518 268f45cc-cd09-0410-ab3c-d52691b4dbfc

6 months agoUpdate Xcode projects with "Apply Configuration to XCFileLists" build target
krollin@apple.com [Fri, 25 Jan 2019 23:19:25 +0000 (23:19 +0000)]
Update Xcode projects with "Apply Configuration to XCFileLists" build target
https://bugs.webkit.org/show_bug.cgi?id=193781
<rdar://problem/47201153>

Reviewed by Alex Christensen.

Part of generating the .xcfilelists used as part of adopting XCBuild
includes running `make DerivedSources.make` from a standalone script.
It’s important for this invocation to have the same environment as
when the actual build invokes `make DerivedSources.make`. If the
environments are different, then the two invocations will provide
different results. In order to get the same environment in the
standalone script, have the script launch xcodebuild targeting the
"Apply Configuration to XCFileLists" build target, which will then
re-invoke our standalone script. The script is now running again, this
time in an environment with all workspace, project, target, xcconfig
and other environment variables established.

The "Apply Configuration to XCFileLists" build target accomplishes
this task via a small embedded shell script that consists only of:

    eval "${WK_SUBLAUNCH_SCRIPT_PARAMETERS[@]}"

The process that invokes "Apply Configuration to XCFileLists" first
sets WK_SUBLAUNCH_SCRIPT_PARAMETERS to an array of commands to be
evaluated and exports it into the shell environment. When xcodebuild
is invoked, it inherits the value of this variable and can `eval` the
contents of that variable. Our external standalone script can then set
WK_SUBLAUNCH_SCRIPT_PARAMETERS to the path to itself, along with a set
of command-line parameters needed to restart itself in the appropriate
state.

Source/JavaScriptCore:

* JavaScriptCore.xcodeproj/project.pbxproj:

Source/WebCore:

No new tests since there should be no observable behavior difference.

* WebCore.xcodeproj/project.pbxproj:

Source/WebKit:

* WebKit.xcodeproj/project.pbxproj:

Tools:

* DumpRenderTree/DumpRenderTree.xcodeproj/project.pbxproj:
* WebKitTestRunner/WebKitTestRunner.xcodeproj/project.pbxproj:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@240517 268f45cc-cd09-0410-ab3c-d52691b4dbfc

6 months agoAdd SPI to inform an application that "Go Back" has been clicked from a safe browsing...
achristensen@apple.com [Fri, 25 Jan 2019 23:16:46 +0000 (23:16 +0000)]
Add SPI to inform an application that "Go Back" has been clicked from a safe browsing warning
https://bugs.webkit.org/show_bug.cgi?id=193802
<rdar://problem/46010580>

Reviewed by Andy Estes.

Source/WebKit:

* UIProcess/API/APIUIClient.h:
(API::UIClient::didClickGoBackFromSafeBrowsingWarning):
* UIProcess/API/Cocoa/WKUIDelegatePrivate.h:
* UIProcess/Cocoa/UIDelegate.h:
* UIProcess/Cocoa/UIDelegate.mm:
(WebKit::UIDelegate::setDelegate):
(WebKit::UIDelegate::UIClient::didClickGoBackFromSafeBrowsingWarning):
* UIProcess/WebPageProxy.cpp:
(WebKit::WebPageProxy::decidePolicyForNavigationAction):

Tools:

* TestWebKitAPI/Tests/WebKitCocoa/SafeBrowsing.mm:
(-[SafeBrowsingNavigationDelegate _webViewDidClickGoBackFromSafeBrowsingWarning:]):
(TEST):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@240516 268f45cc-cd09-0410-ab3c-d52691b4dbfc

6 months ago[ iOS Simulator Release ] Layout Test fast/animation/request-animation-frame.html...
commit-queue@webkit.org [Fri, 25 Jan 2019 23:16:27 +0000 (23:16 +0000)]
[ iOS Simulator Release ] Layout Test fast/animation/request-animation-frame.html is a flaky failure.
https://bugs.webkit.org/show_bug.cgi?id=190888

Unreviewed test gardening.

Patch by Shawn Roberts <sroberts@apple.com> on 2019-01-25

* platform/ios-wk2/TestExpectations:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@240515 268f45cc-cd09-0410-ab3c-d52691b4dbfc

6 months ago[iOS] Pressing Tab key doesn't move to next cell with Google Sheets
dbates@webkit.org [Fri, 25 Jan 2019 23:12:43 +0000 (23:12 +0000)]
[iOS] Pressing Tab key doesn't move to next cell with Google Sheets
https://bugs.webkit.org/show_bug.cgi?id=193048
<rdar://problem/46433836>

Reviewed by Brent Fulgham.

WebCore knows how interpret the Tab and Shift + Tab keys. When building with UIKit keyboard additions
we do not need to register key commands for handling them from the UIProcess code. Instead UIKit will
sends key events for these commands to us. This will make the behavior of these key commands when executed
in an editable field match the behavior of these command when executed in a non-editable field since
both code paths will now go through WebCore.

* UIProcess/ios/WKContentViewInteraction.mm:
(-[WKContentView keyCommands]): Only compile when building with !USE(UIKIT_KEYBOARD_ADDITIONS).

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@240514 268f45cc-cd09-0410-ab3c-d52691b4dbfc

6 months agoFix Linux build.
achristensen@apple.com [Fri, 25 Jan 2019 22:57:42 +0000 (22:57 +0000)]
Fix Linux build.

* NetworkProcess/NetworkConnectionToWebProcess.cpp:
(WebKit::NetworkConnectionToWebProcess::requestResourceLoadStatisticsUpdate):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@240513 268f45cc-cd09-0410-ab3c-d52691b4dbfc

6 months ago[iOS] uiController.keyDown() does not translate Control modified key
dbates@webkit.org [Fri, 25 Jan 2019 22:35:10 +0000 (22:35 +0000)]
[iOS] uiController.keyDown() does not translate Control modified key
https://bugs.webkit.org/show_bug.cgi?id=192898

Reviewed by Wenson Hsieh.

Tools:

Implement uiController.keyDown() in terms of HIDEventGenerator so that UIKit performs
key translation, as applicable. Among other cases, key translation occurs when performing
a Control key modified key sequence. Writing uiController.keyDown() in terms of HIDEventGenerator
also makes the behavior of uiController.keyDown() more closely match the behavior of a
human pressing down and releasing key(s) using a hardware keyboard.

* WebKitTestRunner/ios/HIDEventGenerator.h:
* WebKitTestRunner/ios/HIDEventGenerator.mm:
(-[HIDEventGenerator sendMarkerHIDEventWithCompletionBlock:]): Renamed; formerly named _sendMarkerHIDEventWithCompletionBlock.

(-[HIDEventGenerator touchDown:touchCount:completionBlock:]):
(-[HIDEventGenerator liftUp:touchCount:completionBlock:]):
(-[HIDEventGenerator stylusDownAtPoint:azimuthAngle:altitudeAngle:pressure:completionBlock:]):
(-[HIDEventGenerator stylusMoveToPoint:azimuthAngle:altitudeAngle:pressure:completionBlock:]):
(-[HIDEventGenerator stylusUpAtPoint:completionBlock:]):
(-[HIDEventGenerator stylusTapAtPoint:azimuthAngle:altitudeAngle:pressure:completionBlock:]):
(-[HIDEventGenerator sendTaps:location:withNumberOfTouches:completionBlock:]):
(-[HIDEventGenerator longPress:completionBlock:]):
(-[HIDEventGenerator dragWithStartPoint:endPoint:duration:completionBlock:]):
Update code for above renaming.

(hidUsageCodeForCharacter): Translate \b to a backspace and teach this function to recognize the
DumpRenderTree EventSender-compatible modifier key names (e.g. "metaKey").

(-[HIDEventGenerator keyDown:]): Added.
(-[HIDEventGenerator keyUp:]): Added.
(-[HIDEventGenerator keyPress:completionBlock:]):
(-[HIDEventGenerator eventDispatchThreadEntry:]):
(-[HIDEventGenerator _sendMarkerHIDEventWithCompletionBlock:]): Deleted; renamed to sendMarkerHIDEventWithCompletionBlock.
* WebKitTestRunner/ios/UIScriptControllerIOS.mm:
(WTR::parseModifierArray): Modified to return a vector of strings that represent the modifier keys to press.
(WTR::UIScriptController::keyDown): Write in terms of HIDEventGenerator.
(WTR::parseModifier): Deleted.

LayoutTests:

Make tests more robusts.

* fast/events/ios/key-command-italic-dispatches-keydown.html: Remove the need to special-case the test runner and
expect a single key event that represents Command + I. Now that uiController.keyDown() makes use of HIDEventGenerator
we emit separate key events for each key pressed in the sequence: keydown Command, keydown i, keyup i, keyup Command.
This matches how a human being would execute such a key sequence.
* fast/forms/auto-fill-button/caps-lock-indicator-should-not-be-visible-when-auto-fill-strong-password-button-is-visible-expected.html:
Make the test less flaky by hiding the tap highlight. Depending on timing, the tap highlight may be snapshotted and thus
cause the reference test to mismatch the test. Mitigate this by hiding the tap highlight.
* fast/text/scroll-text-overflow-ellipsis.html: Wait for a keyup event for the down arrow before declaring the
test complete.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@240512 268f45cc-cd09-0410-ab3c-d52691b4dbfc

6 months agoAdd API to generate and consume cached bytecode
tzagallo@apple.com [Fri, 25 Jan 2019 22:31:12 +0000 (22:31 +0000)]
Add API to generate and consume cached bytecode
https://bugs.webkit.org/show_bug.cgi?id=193401
<rdar://problem/47514099>

Reviewed by Keith Miller.

Add the `generateBytecode` and `generateModuleBytecode` functions to
generate serialized bytecode for a given `SourceCode`. These functions
will eagerly generate code for all the nested functions.

Additionally, update the API methods in JSScript to generate and use the
bytecode when the bytecodeCache path is provided.

* API/JSAPIGlobalObject.mm:
(JSC::JSAPIGlobalObject::moduleLoaderFetch):
* API/JSContext.mm:
(-[JSContext wrapperMap]):
* API/JSContextInternal.h:
* API/JSScript.mm:
(+[JSScript scriptWithSource:inVirtualMachine:]):
(+[JSScript scriptFromASCIIFile:inVirtualMachine:withCodeSigning:andBytecodeCache:]):
(-[JSScript dealloc]):
(-[JSScript readCache]):
(-[JSScript writeCache]):
(-[JSScript hash]):
(-[JSScript source]):
(-[JSScript cachedBytecode]):
(-[JSScript jsSourceCode:]):
* API/JSScriptInternal.h:
* API/JSScriptSourceProvider.h: Copied from Source/JavaScriptCore/API/JSScriptInternal.h.
(JSScriptSourceProvider::create):
(JSScriptSourceProvider::JSScriptSourceProvider):
* API/JSScriptSourceProvider.mm: Copied from Source/JavaScriptCore/API/JSScriptInternal.h.
(JSScriptSourceProvider::hash const):
(JSScriptSourceProvider::source const):
(JSScriptSourceProvider::cachedBytecode const):
* API/JSVirtualMachine.mm:
(-[JSVirtualMachine vm]):
* API/JSVirtualMachineInternal.h:
* API/tests/testapi.mm:
(testBytecodeCache):
(-[JSContextFileLoaderDelegate context:fetchModuleForIdentifier:withResolveHandler:andRejectHandler:]):
(testObjectiveCAPI):
* JavaScriptCore.xcodeproj/project.pbxproj:
* SourcesCocoa.txt:
* bytecode/UnlinkedFunctionExecutable.cpp:
(JSC::UnlinkedFunctionExecutable::unlinkedCodeBlockFor):
* bytecode/UnlinkedFunctionExecutable.h:
* parser/SourceCodeKey.h:
(JSC::SourceCodeKey::source const):
* parser/SourceProvider.h:
(JSC::CachedBytecode::CachedBytecode):
(JSC::CachedBytecode::operator=):
(JSC::CachedBytecode::data const):
(JSC::CachedBytecode::size const):
(JSC::CachedBytecode::owned const):
(JSC::CachedBytecode::~CachedBytecode):
(JSC::CachedBytecode::freeDataIfOwned):
(JSC::SourceProvider::cachedBytecode const):
* parser/UnlinkedSourceCode.h:
(JSC::UnlinkedSourceCode::provider const):
* runtime/CodeCache.cpp:
(JSC::generateUnlinkedCodeBlockForFunctions):
(JSC::writeCodeBlock):
(JSC::serializeBytecode):
* runtime/CodeCache.h:
(JSC::CodeCacheMap::fetchFromDiskImpl):
(JSC::CodeCacheMap::findCacheAndUpdateAge):
(JSC::generateUnlinkedCodeBlockImpl):
(JSC::generateUnlinkedCodeBlock):
* runtime/Completion.cpp:
(JSC::generateBytecode):
(JSC::generateModuleBytecode):
* runtime/Completion.h:
* runtime/Options.cpp:
(JSC::recomputeDependentOptions):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@240511 268f45cc-cd09-0410-ab3c-d52691b4dbfc

6 months agoComplete the implementation of generate-xcfilelists
krollin@apple.com [Fri, 25 Jan 2019 22:23:54 +0000 (22:23 +0000)]
Complete the implementation of generate-xcfilelists
https://bugs.webkit.org/show_bug.cgi?id=193782

Reviewed by Alex Christensen.

A preliminary version of generate-xcfilelists was added in r238008.
This patch updates that script in order to:

- Support checking of .xcfilelists
- Perform better execution of `make DerivedSources.make` by
  relaunching generate-xcfilelists under xcodebuild
- Support incremental generation and checking of .xcfilelists
- Support supplementary .xcfilelists files from WebKitAdditions
- Support being wrapped by parallel version of generate-xcfilelists in
  Internal that adds support for (re)generating WebKit projects in
  that repo.
- Support builds that have been performed in WebKitBuild as well as
  ~/Library/Developer/code/DerivedData
- Increase robustness and debugging facilities.

* Scripts/generate-xcfilelists:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@240510 268f45cc-cd09-0410-ab3c-d52691b4dbfc

6 months agoActivate the WebResourceLoadStatisticsStore in the NetworkProcess and deactivate...
bfulgham@apple.com [Fri, 25 Jan 2019 22:21:38 +0000 (22:21 +0000)]
Activate the WebResourceLoadStatisticsStore in the NetworkProcess and deactivate it in the UIProcess.
https://bugs.webkit.org/show_bug.cgi?id=193297
<rdar://problem/47158841>

Unreviewed test case fix.

These test cases require ResourceLoadStatistics to be active, so the feature must be turned on.
Previously, the default data store turned the feature on automatically, allowing these tests to pass.

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

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@240509 268f45cc-cd09-0410-ab3c-d52691b4dbfc

6 months agoUpdate WebKitAdditions.xcconfig with correct order of variable definitions
krollin@apple.com [Fri, 25 Jan 2019 22:19:25 +0000 (22:19 +0000)]
Update WebKitAdditions.xcconfig with correct order of variable definitions
https://bugs.webkit.org/show_bug.cgi?id=193793
<rdar://problem/47532439>

Reviewed by Alex Christensen.

XCBuild changes the way xcconfig variables are evaluated. In short,
all config file assignments are now considered in part of the
evaluation. When using the new build system and an .xcconfig file
contains multiple assignments of the same build setting:

- Later assignments using $(inherited) will inherit from earlier
  assignments in the xcconfig file.
- Later assignments not using $(inherited) will take precedence over
  earlier assignments. An assignment to a more general setting will
  mask an earlier assignment to a less general setting. For example,
  an assignment without a condition ('FOO = bar') will completely mask
  an earlier assignment with a condition ('FOO[sdk=macos*] = quux').

This affects some of our .xcconfig files, in that sometimes platform-
or sdk-specific definitions appear before the general definitions.
Under the new evaluations rules, the general definitions alway take
effect because they always overwrite the more-specific definitions. The
solution is to swap the order, so that the general definitions are
established first, and then conditionally overwritten by the
more-specific definitions.

Source/JavaScriptCore:

* Configurations/Version.xcconfig:

Source/ThirdParty/libwebrtc:

* Configurations/Version.xcconfig:

Source/WebCore:

No new tests since there should be no observable behavior difference.

* Configurations/Base.xcconfig:
* Configurations/Version.xcconfig:

Source/WebKit:

* Configurations/BaseTarget.xcconfig:
* Configurations/Version.xcconfig:
* Configurations/WebKit.xcconfig:

Source/WebKitLegacy/mac:

* Configurations/Version.xcconfig:
* Configurations/WebKitLegacy.xcconfig:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@240508 268f45cc-cd09-0410-ab3c-d52691b4dbfc

6 months agoUpdate existing .xcfilelists
krollin@apple.com [Fri, 25 Jan 2019 22:16:57 +0000 (22:16 +0000)]
Update existing .xcfilelists
https://bugs.webkit.org/show_bug.cgi?id=193791
<rdar://problem/47201706>

Reviewed by Alex Christensen.

Many .xcfilelist files were added in r238824 in order to support
XCBuild. Update these with recent changes to the set of build files
and with the current generate-xcfilelist script.

Source/JavaScriptCore:

* DerivedSources-input.xcfilelist:
* DerivedSources-output.xcfilelist:
* UnifiedSources-input.xcfilelist:
* UnifiedSources-output.xcfilelist:

Source/WebCore:

No new tests since there should be no observable behavior difference.

* DerivedSources-input.xcfilelist:
* DerivedSources-output.xcfilelist:
* UnifiedSources-input.xcfilelist:
* UnifiedSources-output.xcfilelist:

Source/WebKit:

* DerivedSources-input.xcfilelist:
* DerivedSources-output.xcfilelist:
* UnifiedSources-input.xcfilelist:
* UnifiedSources-output.xcfilelist:

Tools:

* DumpRenderTree/DerivedSources-input.xcfilelist:
* DumpRenderTree/DerivedSources-output.xcfilelist:
* WebKitTestRunner/DerivedSources-input.xcfilelist:
* WebKitTestRunner/DerivedSources-output.xcfilelist:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@240507 268f45cc-cd09-0410-ab3c-d52691b4dbfc

6 months agoWeb Inspector: Timelines: DOMContentLoaded and load event lines need to be more obvious
drousso@apple.com [Fri, 25 Jan 2019 22:05:09 +0000 (22:05 +0000)]
Web Inspector: Timelines: DOMContentLoaded and load event lines need to be more obvious
https://bugs.webkit.org/show_bug.cgi?id=193186
<rdar://problem/45100694>

Reviewed by Joseph Pecoraro.

Shift around z-index values to make Timelines markers appear behind record bars. Widen the
marker hit region to make it easier to see the tooltip.

* UserInterface/Views/Variables.css:
(:root):

* UserInterface/Views/TimelineRuler.css:
(.timeline-ruler):
(.timeline-ruler > .markers):
(.timeline-ruler > .markers > .marker):
(body[dir=ltr] .timeline-ruler > .markers > .marker):
(body[dir=rtl] .timeline-ruler > .markers > .marker):
(.timeline-ruler > .markers > .marker::before):
(body[dir=ltr] .timeline-ruler > .markers > .marker::before):
(body[dir=rtl] .timeline-ruler > .markers > .marker::before):
(.timeline-ruler > .markers > .marker::after): Added.
(body[dir=ltr] .timeline-ruler > .markers > .marker::after): Added.
(body[dir=rtl] .timeline-ruler > .markers > .marker::after): Added.
(.timeline-ruler > .markers > .marker.current-time):
(.timeline-ruler > .markers > .marker.current-time::after): Added.
(.timeline-ruler > .markers > .marker.load-event):
(.timeline-ruler > .markers > .marker.dom-content-event):
(.timeline-ruler > .markers > .marker.timestamp):
(.timeline-ruler > .selection-handle):
(.timeline-ruler.both-handles-clamped > .selection-handle):
(.timeline-ruler > .shaded-area):
(.timeline-ruler > .markers > .marker.current-time::before): Deleted.

* UserInterface/Views/TimelineRecordBar.css:
(.timeline-record-bar):
(.timeline-record-bar > .segment):

* UserInterface/Views/CPUTimelineOverviewGraph.css:
(.timeline-overview-graph.cpu > .legend):
* UserInterface/Views/MemoryTimelineOverviewGraph.css:
(.timeline-overview-graph.memory > .legend):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@240506 268f45cc-cd09-0410-ab3c-d52691b4dbfc

6 months agoMove more scrolling tree tests into scrollingcoordinator/scrolling-tree
simon.fraser@apple.com [Fri, 25 Jan 2019 21:56:20 +0000 (21:56 +0000)]
Move more scrolling tree tests into scrollingcoordinator/scrolling-tree
https://bugs.webkit.org/show_bug.cgi?id=193845

Reviewed by Tim Horton.

Gather more tests of the scrolling tree into scrollingcoordinator/scrolling-tree.

* platform/gtk/TestExpectations:
* platform/ios-wk1/fast/scrolling/scrolling-tree-includes-frame-expected.txt: Removed.
* platform/ios-wk2/scrollingcoordinator/scrolling-tree/coordinated-frame-expected.txt: Added.
* platform/ios-wk2/scrollingcoordinator/scrolling-tree/coordinated-frame-gain-scrolling-ancestor-expected.txt: Copied from LayoutTests/tiled-drawing/scrolling/frames/coordinated-frame-gain-scrolling-ancestor-expected.txt.
* platform/ios-wk2/scrollingcoordinator/scrolling-tree/coordinated-frame-in-fixed-expected.txt: Added.
* platform/ios-wk2/scrollingcoordinator/scrolling-tree/coordinated-frame-lose-scrolling-ancestor-expected.txt: Added.
* platform/ios-wk2/scrollingcoordinator/scrolling-tree/remove-coordinated-frame-expected.txt: Renamed from LayoutTests/platform/ios/fast/scrolling/scrolling-tree-includes-frame-expected.txt.
* platform/ios-wk2/scrollingcoordinator/scrolling-tree/scrolling-tree-includes-frame-expected.txt: Renamed from LayoutTests/platform/ios-wk2/fast/scrolling/scrolling-tree-includes-frame-expected.txt.
* platform/mac-wk1/fast/scrolling/scrolling-tree-includes-frame-expected.txt: Removed.
* platform/win/TestExpectations:
* scrollingcoordinator/scrolling-tree/coordinated-frame-expected.txt: Renamed from LayoutTests/tiled-drawing/scrolling/frames/coordinated-frame-expected.txt.
* scrollingcoordinator/scrolling-tree/coordinated-frame-gain-scrolling-ancestor-expected.txt: Renamed from LayoutTests/tiled-drawing/scrolling/frames/coordinated-frame-gain-scrolling-ancestor-expected.txt.
* scrollingcoordinator/scrolling-tree/coordinated-frame-gain-scrolling-ancestor.html: Renamed from LayoutTests/tiled-drawing/scrolling/frames/coordinated-frame-gain-scrolling-ancestor.html.
* scrollingcoordinator/scrolling-tree/coordinated-frame-in-fixed-expected.txt: Renamed from LayoutTests/tiled-drawing/scrolling/frames/coordinated-frame-in-fixed-expected.txt.
* scrollingcoordinator/scrolling-tree/coordinated-frame-in-fixed.html: Renamed from LayoutTests/tiled-drawing/scrolling/frames/coordinated-frame-in-fixed.html.
* scrollingcoordinator/scrolling-tree/coordinated-frame-lose-scrolling-ancestor-expected.txt: Renamed from LayoutTests/tiled-drawing/scrolling/frames/coordinated-frame-lose-scrolling-ancestor-expected.txt.
* scrollingcoordinator/scrolling-tree/coordinated-frame-lose-scrolling-ancestor.html: Renamed from LayoutTests/tiled-drawing/scrolling/frames/coordinated-frame-lose-scrolling-ancestor.html.
* scrollingcoordinator/scrolling-tree/coordinated-frame.html: Renamed from LayoutTests/tiled-drawing/scrolling/frames/coordinated-frame.html.
* scrollingcoordinator/scrolling-tree/remove-coordinated-frame-expected.txt: Renamed from LayoutTests/tiled-drawing/scrolling/frames/remove-coordinated-frame-expected.txt.
* scrollingcoordinator/scrolling-tree/remove-coordinated-frame.html: Renamed from LayoutTests/tiled-drawing/scrolling/frames/remove-coordinated-frame.html.
* scrollingcoordinator/scrolling-tree/remove-scrolling-role-expected.txt: Renamed from LayoutTests/fast/scrolling/ios/remove-scrolling-role-expected.txt.
* scrollingcoordinator/scrolling-tree/remove-scrolling-role.html: Renamed from LayoutTests/fast/scrolling/ios/remove-scrolling-role.html.
* scrollingcoordinator/scrolling-tree/resources/doc-with-sticky.html: Renamed from LayoutTests/tiled-drawing/scrolling/frames/resources/doc-with-sticky.html.
* scrollingcoordinator/scrolling-tree/scrolling-tree-includes-frame-expected.txt: Renamed from LayoutTests/fast/scrolling/scrolling-tree-includes-frame-expected.txt.
* scrollingcoordinator/scrolling-tree/scrolling-tree-includes-frame.html: Renamed from LayoutTests/fast/scrolling/scrolling-tree-includes-frame.html.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@240505 268f45cc-cd09-0410-ab3c-d52691b4dbfc

6 months agoMove tests for scrolling tree mutations into their own directory
simon.fraser@apple.com [Fri, 25 Jan 2019 21:56:05 +0000 (21:56 +0000)]
Move tests for scrolling tree mutations into their own directory
https://bugs.webkit.org/show_bug.cgi?id=193836

Reviewed by Tim Horton.

Add a scrollingcoordinator/scrolling-tree directory, which will contain tests for scrolling
tree mutations.

* platform/ios-wk2/scrollingcoordinator/scrolling-tree/gain-scrolling-node-parent-expected.txt: Renamed from LayoutTests/platform/ios/scrollingcoordinator/gain-scrolling-node-parent-expected.txt.
* platform/ios-wk2/scrollingcoordinator/scrolling-tree/lose-scrolling-node-parent-expected.txt: Renamed from LayoutTests/platform/ios/scrollingcoordinator/lose-scrolling-node-parent-expected.txt.
* platform/ios-wk2/scrollingcoordinator/scrolling-tree/reparent-across-compositing-layers-expected.txt: Renamed from LayoutTests/platform/ios-wk2/scrollingcoordinator/reparent-across-compositing-layers-expected.txt.
* platform/ios-wk2/scrollingcoordinator/scrolling-tree/reparent-with-layer-removal-expected.txt: Renamed from LayoutTests/platform/ios-wk2/scrollingcoordinator/reparent-with-layer-removal-expected.txt.
* scrollingcoordinator/scrolling-tree/gain-scrolling-node-parent-expected.txt: Renamed from LayoutTests/scrollingcoordinator/gain-scrolling-node-parent-expected.txt.
* scrollingcoordinator/scrolling-tree/gain-scrolling-node-parent.html: Renamed from LayoutTests/scrollingcoordinator/gain-scrolling-node-parent.html.
* scrollingcoordinator/scrolling-tree/lose-scrolling-node-parent-expected.txt: Renamed from LayoutTests/scrollingcoordinator/lose-scrolling-node-parent-expected.txt.
* scrollingcoordinator/scrolling-tree/lose-scrolling-node-parent.html: Renamed from LayoutTests/scrollingcoordinator/lose-scrolling-node-parent.html.
* scrollingcoordinator/scrolling-tree/reparent-across-compositing-layers-expected.txt: Renamed from LayoutTests/scrollingcoordinator/reparent-across-compositing-layers-expected.txt.
* scrollingcoordinator/scrolling-tree/reparent-across-compositing-layers.html: Renamed from LayoutTests/scrollingcoordinator/reparent-across-compositing-layers.html.
* scrollingcoordinator/scrolling-tree/reparent-with-layer-removal-expected.txt: Renamed from LayoutTests/scrollingcoordinator/reparent-with-layer-removal-expected.txt.
* scrollingcoordinator/scrolling-tree/reparent-with-layer-removal.html: Renamed from LayoutTests/scrollingcoordinator/reparent-with-layer-removal.html.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@240504 268f45cc-cd09-0410-ab3c-d52691b4dbfc

6 months agoFix missing dependencies in extract-dependencies-from-makefile
krollin@apple.com [Fri, 25 Jan 2019 21:51:03 +0000 (21:51 +0000)]
Fix missing dependencies in extract-dependencies-from-makefile
https://bugs.webkit.org/show_bug.cgi?id=193783
<rdar://problem/47201571>

Reviewed by Alex Christensen.

The extract-dependencies-from-makefile script generates .xcfilelists
for XCBuild by invoking a makefile in --debug mode, parsing the
dependency information in the output, and extracting information
regarding targets and dependents. However, the way `make` emits this
dependency information is not rigorous, and so we need to determine
what lines to look for and parse by trial and error. This approach
didn't coriginally atch all the information we needed to collect, so
update the script to look for the additional lines we now know to look
for.

* Scripts/extract-dependencies-from-makefile:
(Parser):
(Parser.addTarget):
(Parser.addPrereq):
(Parser.doParse):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@240503 268f45cc-cd09-0410-ab3c-d52691b4dbfc

6 months agoWeb Inspector: Audit: remove experimental setting
drousso@apple.com [Fri, 25 Jan 2019 21:49:03 +0000 (21:49 +0000)]
Web Inspector: Audit: remove experimental setting
https://bugs.webkit.org/show_bug.cgi?id=193743
<rdar://problem/28234022>

Reviewed by Joseph Pecoraro.

* UserInterface/Base/Setting.js:
(WI.Setting):
(WI.Setting.localStorageKey): Added.
(WI.Setting.migrateValue):
Drive-by: add the localStorage key prefix to the key passed into `WI.Setting.migrateValue`.
* UserInterface/Base/Main.js:
(WI.loaded):

* UserInterface/Views/AuditTabContentView.js:
(WI.AuditTabContentView.isTabAllowed):

* UserInterface/Views/SettingsTabContentView.js:
(WI.SettingsTabContentView.prototype._createExperimentalSettingsView):

* Localizations/en.lproj/localizedStrings.js:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@240502 268f45cc-cd09-0410-ab3c-d52691b4dbfc

6 months ago[iOS] Deny mach lookups to services not used.
pvollan@apple.com [Fri, 25 Jan 2019 21:40:11 +0000 (21:40 +0000)]
[iOS] Deny mach lookups to services not used.
https://bugs.webkit.org/show_bug.cgi?id=193828

Reviewed by Brent Fulgham.

Start denying mach lookups to iOS services, which were previously allowed with reporting.
Living-on has indicated that these services are not used.

* Resources/SandboxProfiles/ios/com.apple.WebKit.WebContent.sb:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@240500 268f45cc-cd09-0410-ab3c-d52691b4dbfc

6 months agoWeb Inspector: show uncaught exception view for unhandled promise rejections
drousso@apple.com [Fri, 25 Jan 2019 21:35:03 +0000 (21:35 +0000)]
Web Inspector: show uncaught exception view for unhandled promise rejections
https://bugs.webkit.org/show_bug.cgi?id=193832

Reviewed by Matt Baker.

* UserInterface/Debug/UncaughtExceptionReporter.js:
(handleUnhandledPromiseRejection): Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@240499 268f45cc-cd09-0410-ab3c-d52691b4dbfc

6 months agoActivate the WebResourceLoadStatisticsStore in the NetworkProcess and deactivate...
bfulgham@apple.com [Fri, 25 Jan 2019 20:52:11 +0000 (20:52 +0000)]
Activate the WebResourceLoadStatisticsStore in the NetworkProcess and deactivate it in the UIProcess.
https://bugs.webkit.org/show_bug.cgi?id=193297
<rdar://problem/47158841>

Reviewed by Alex Christensen.

Source/WebCore:

Trigger logging to the UIProcess when the ResourceLoadObserver is used in the NetworkProcess.

* Modules/websockets/WebSocket.cpp:
(WebCore::WebSocket::connect): Notify NetworkProcess a connection was made to a resource.
* loader/ResourceLoadObserver.cpp:
(WebCore::ResourceLoadObserver::setLogWebSocketLoadingNotificationCallback): Added.
(WebCore::ResourceLoadObserver::setLogSubresourceLoadingNotificationCallback): Added.
(WebCore::ResourceLoadObserver::setLogSubresourceRedirectNotificationCallback): Added.
(WebCore::ResourceLoadObserver::logSubresourceLoading): Notify NetworkProcess of the load.
(WebCore::ResourceLoadObserver::logWebSocketLoading): Ditto.
(WebCore::ResourceLoadObserver::logUserInteractionWithReducedTimeResolution): Ditto.

Source/WebKit:

This patch activates the ResourceLoadStatistics code in the NetworkProcess, and turns
it off in the UIProcess. It also updates test infrastructure to work with this change
in architecture.

* NetworkProcess/Classifier/ResourceLoadStatisticsMemoryStore.cpp:
(WebKit::ResourceLoadStatisticsMemoryStore::logSubresourceLoading): Added.
(WebKit::ResourceLoadStatisticsMemoryStore::logSubresourceRedirect): Added.
(WebKit::ResourceLoadStatisticsMemoryStore::logWebSocketLoading): Added.
* NetworkProcess/Classifier/WebResourceLoadStatisticsStore.cpp:
(WebKit::WebResourceLoadStatisticsStore::logSubresourceLoading): Added.
(WebKit::WebResourceLoadStatisticsStore::logSubresourceRedirect): Added.
(WebKit::WebResourceLoadStatisticsStore::logWebSocketLoading): Added.
(WebKit::WebResourceLoadStatisticsStore::removeAllStorageAccess):
(WebKit::WebResourceLoadStatisticsStore::setCacheMaxAgeCap):
(WebKit::WebResourceLoadStatisticsStore::setCacheMaxAgeCapForPrevalentResources): Deleted.
* NetworkProcess/Classifier/WebResourceLoadStatisticsStore.h:
* NetworkProcess/NetworkConnectionToWebProcess.cpp:
(WebKit::NetworkConnectionToWebProcess::logSubresourceLoading): Added.
(WebKit::NetworkConnectionToWebProcess::logSubresourceRedirect): Added.
(WebKit::NetworkConnectionToWebProcess::logWebSocketLoading): Added.
* NetworkProcess/NetworkConnectionToWebProcess.messages.in:
* NetworkProcess/NetworkSession.cpp:
(WebKit::NetworkSession::setResourceLoadStatisticsEnabled):
* UIProcess/WebProcessPool.cpp:
(WebKit::WebProcessPool::ensureNetworkProcess):
* UIProcess/WebsiteData/Cocoa/WebsiteDataStoreCocoa.mm:
(WebKit::WebsiteDataStore::parameters):
* UIProcess/WebsiteData/WebsiteDataStore.cpp:
(WebKit::WebsiteDataStore::removeData):
(WebKit::WebsiteDataStore::setResourceLoadStatisticsEnabled):
(WebKit::WebsiteDataStore::enableResourceLoadStatisticsAndSetTestingCallback):
* Webprocess/WebProcess.cpp:
(WebKit::WebProcess::initializeWebProcess): Register new ResourceLoadObserver callbacks.

Tools:

* TestWebKitAPI/Tests/WebKit Cocoa/ResourceLoadStatistics.mm:
    Revise tests to ensure a running NetworkProcess before attempting ResourceLoadStatistics operations.
* WebKitTestRunner/InjectedBundle/TestRunner.cpp:
(WTR::TestRunner::installStatisticsDidScanDataRecordsCallback): Simplify test configuration by
activating the message used to trigger the callback when it is set.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@240498 268f45cc-cd09-0410-ab3c-d52691b4dbfc

6 months agoiOS: inputmode="none" disables hardware keyboard's globe key
rniwa@webkit.org [Fri, 25 Jan 2019 20:45:04 +0000 (20:45 +0000)]
iOS: inputmode="none" disables hardware keyboard's globe key
https://bugs.webkit.org/show_bug.cgi?id=193811
<rdar://problem/47406553>

Reviewed by Wenson Hsieh.

Source/WebKit:

Removed the support for inputmode="none" for now since we need a new SPI from UIKit
to properly implement this feature some time in the future.

* UIProcess/ios/WKContentViewInteraction.mm:
(-[WKContentView _zoomToRevealFocusedElement]):
(-[WKContentView inputView]):
(-[WKContentView requiresAccessoryView]):
(-[WKContentView textInputTraits]):

LayoutTests:

Updated and renamed the test expecting the keyboard to update upon inputmode content attribute changed
to use inputmode="decimal" instead of inputmode="none", and updated another test to expect inputmode="none"
has no effect instead of hiding the keyboard.

* fast/forms/ios/inputmode-change-update-keyboard-expected.txt: Renamed from inputmode-none-removed-expected.txt.
* fast/forms/ios/inputmode-change-update-keyboard.html: Renamed from inputmode-none-removed.html.
* fast/forms/ios/inputmode-none-expected.txt:
* fast/forms/ios/inputmode-none.html:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@240497 268f45cc-cd09-0410-ab3c-d52691b4dbfc

6 months agoRemove FrameView::m_firstVisuallyNonEmptyLayoutCallbackPending
zalan@apple.com [Fri, 25 Jan 2019 20:38:59 +0000 (20:38 +0000)]
Remove FrameView::m_firstVisuallyNonEmptyLayoutCallbackPending
https://bugs.webkit.org/show_bug.cgi?id=193835

Reviewed by Simon Fraser.

Currently updateIsVisuallyNonEmpty() is called from fireLayoutRelatedMilestonesIfNeeded() and from the incrementVisually*() functions.
By calling it from incrementVisually*() and setting the m_isVisuallyNonEmpty flag to true early does not have any impact on when the milestone is fired.
The milestone firing, as part of the post-layout tasks is triggered by a subsequent layout.
However having multiple callers of updateIsVisuallyNonEmpty() requires an extra boolen (m_firstVisuallyNonEmptyLayoutCallbackPending) to maintain.
Also calling updateIsVisuallyNonEmpty() repeatedly could be costly (with the current threshold of 200 characters, I don't think it is though).

This patch removes m_firstVisuallyNonEmptyLayoutCallbackPending and moves the logic from updateIsVisuallyNonEmpty() to fireLayoutRelatedMilestonesIfNeeded().

* page/FrameView.cpp:
(WebCore::FrameView::resetLayoutMilestones):
(WebCore::FrameView::loadProgressingStatusChanged):
(WebCore::FrameView::incrementVisuallyNonEmptyCharacterCount):
(WebCore::FrameView::fireLayoutRelatedMilestonesIfNeeded):
(WebCore::FrameView::updateIsVisuallyNonEmpty): Deleted.
* page/FrameView.h:
(WebCore::FrameView::incrementVisuallyNonEmptyPixelCount):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@240495 268f45cc-cd09-0410-ab3c-d52691b4dbfc

6 months agoMove soft-linking of Lookup.framework out of LookupSPI.h
ddkilzer@apple.com [Fri, 25 Jan 2019 20:24:14 +0000 (20:24 +0000)]
Move soft-linking of Lookup.framework out of LookupSPI.h
<https://webkit.org/b/193815>

Reviewed by Tim Horton.

Source/WebCore:

* editing/cocoa/DictionaryLookup.mm:
- Remove unused header.

* editing/mac/DictionaryLookupLegacy.mm:
(WebCore::tokenRange):
(WebCore::showPopupOrCreateAnimationController):
(WebCore::DictionaryLookup::hidePopup):
- Move soft-linking to LookupSoftLink.{h,mm}.

* platform/ios/ValidationBubbleIOS.mm:
(WebCore::ValidationBubble::show):
- Update for changes to UIKitSoftLink.{h,mm} now that
  UIAccessibilityAnnouncementNotification is using
  SOFT_LINK_CONSTANT*().

Source/WebCore/PAL:

* PAL.xcodeproj/project.pbxproj:
* pal/PlatformMac.cmake:
- Add LookupSoftLink.{h,mm} files to the project.

* pal/ios/UIKitSoftLink.h:
* pal/ios/UIKitSoftLink.mm:
- Change SOFT_LINK_VARIABLE_FOR_HEADER() to
  SOFT_LINK_CONSTANT_FOR_HEADER() for
  UIAccessibilityAnnouncementNotification constant.

* pal/mac/LookupSoftLink.h: Copied from Source/WebCore/PAL/pal/spi/mac/LookupSPI.h.
* pal/mac/LookupSoftLink.mm: Copied from Source/WebCore/PAL/pal/spi/mac/LookupSPI.h.
- Migrate soft-linking from LookupSPI.h to here.

* pal/spi/mac/LookupSPI.h:
- Replace soft-linking of Lookup.framework with
  LookupSoftLink.h.

Source/WebKit:

* UIProcess/Cocoa/WebViewImpl.mm:
(-[WKWindowVisibilityObserver dealloc]):
(-[WKWindowVisibilityObserver startObservingLookupDismissalIfNeeded]):
- Move soft-linking to LookupSoftLink.{h,mm}.

Source/WebKitLegacy/mac:

* WebView/WebImmediateActionController.mm:
(-[WebImmediateActionController _animationControllerForText]):
* WebView/WebView.mm:
(-[WebView _prepareForDictionaryLookup]):
- Move soft-linking to LookupSoftLink.{h,mm}.

Source/WTF:

* wtf/cocoa/SoftLinking.h:
(SOFT_LINK_CONSTANT_FOR_HEADER):
(SOFT_LINK_CONSTANT_FOR_SOURCE_WITH_EXPORT):
(SOFT_LINK_CONSTANT_MAY_FAIL_FOR_HEADER):
- Remove local declaration of constant.  This should be handled
  by including *SPI.h or "real" headers.
(SOFT_LINK_CONSTANT_MAY_FAIL_FOR_SOURCE_WITH_EXPORT):
- Remove local declaration of constant.
- Rename from SOFT_LINK_CONSTANT_MAY_FAIL_FOR_SOURCE() and add
  `export` argument to support exporting of functions.
(SOFT_LINK_CONSTANT_MAY_FAIL_FOR_SOURCE):
- Redefine in terms of
  SOFT_LINK_CONSTANT_MAY_FAIL_FOR_SOURCE_WITH_EXPORT().

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@240494 268f45cc-cd09-0410-ab3c-d52691b4dbfc

6 months agoREGRESSION (PSON): After swipe back, first few attempts to swipe forward not recogniz...
antti@apple.com [Fri, 25 Jan 2019 20:20:17 +0000 (20:20 +0000)]
REGRESSION (PSON): After swipe back, first few attempts to swipe forward not recognized as swipe gestures
https://bugs.webkit.org/show_bug.cgi?id=193818
<rdar://problem/47456584>

Reviewed by Tim Horton.

We don't hit the RenderTreeSizeThreshold milestone to remove the snapshot and allow interaction because we request it
from a wrong process (the one we are leaving behind).

* UIProcess/Cocoa/ViewGestureController.cpp:
(WebKit::ViewGestureController::didStartProvisionalLoadForMainFrame):

Request the RenderTreeSizeThreshold notification when the load has started and we know the process to use.

(WebKit::ViewGestureController::SnapshotRemovalTracker::hasOutstandingEvent):
* UIProcess/Cocoa/ViewGestureController.h:
(WebKit::ViewGestureController::SnapshotRemovalTracker::renderTreeSizeThreshold const):
(WebKit::ViewGestureController::SnapshotRemovalTracker::setRenderTreeSizeThreshold):
* UIProcess/mac/ViewGestureControllerMac.mm:
(WebKit::ViewGestureController::endSwipeGesture):

Don't request immediately, save the threshold to SnapshotRemovalTracker.

(WebKit::ViewGestureController::requestRenderTreeSizeNotificationIfNeeded):

Do the request to the right process.

* WebProcess/WebPage/ViewGestureGeometryCollector.cpp:
(WebKit::ViewGestureGeometryCollector::setRenderTreeSizeNotificationThreshold):

Ensure we send the notification even if we reached the threshold before it was requested.

(WebKit::ViewGestureGeometryCollector::sendDidHitRenderTreeSizeThresholdIfNeeded):
(WebKit::ViewGestureGeometryCollector::mainFrameDidLayout):
* WebProcess/WebPage/ViewGestureGeometryCollector.h:
(WebKit::ViewGestureGeometryCollector::setRenderTreeSizeNotificationThreshold): Deleted.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@240493 268f45cc-cd09-0410-ab3c-d52691b4dbfc

6 months agowebkitpy: Implement device type specific expected results (Gardening)
jbedard@apple.com [Fri, 25 Jan 2019 19:56:56 +0000 (19:56 +0000)]
webkitpy: Implement device type specific expected results (Gardening)
https://bugs.webkit.org/show_bug.cgi?id=192162
<rdar://problem/46345449>

Unreviewed test gardening.

* platform/ios/TestExpectations: Remove non-existant directories.
* platform/mac/TestExpectations: Ditto.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@240492 268f45cc-cd09-0410-ab3c-d52691b4dbfc

6 months ago[iOS] Rename some WebKit-internal functions and variables that reference "data intera...
wenson_hsieh@apple.com [Fri, 25 Jan 2019 19:45:07 +0000 (19:45 +0000)]
[iOS] Rename some WebKit-internal functions and variables that reference "data interaction"
https://bugs.webkit.org/show_bug.cgi?id=193829

Reviewed by Tim Horton.

Source/WebCore:

No change in behavior.

* page/EventHandler.h:
* page/ios/EventHandlerIOS.mm:
(WebCore::EventHandler::tryToBeginDragAtPoint):
(WebCore::EventHandler::tryToBeginDataInteractionAtPoint): Deleted.
* platform/ios/WebItemProviderPasteboard.mm:
(linkTemporaryItemProviderFilesToDropStagingDirectory):

Source/WebKit:

* UIProcess/Cocoa/WebPageProxyCocoa.mm:
(WebKit::WebPageProxy::setDragCaretRect):
* UIProcess/PageClient.h:
* UIProcess/WebPageProxy.h:
* UIProcess/WebPageProxy.messages.in:
* UIProcess/ios/PageClientImplIOS.h:
* UIProcess/ios/PageClientImplIOS.mm:
(WebKit::PageClientImpl::didHandleDragStartRequest):
(WebKit::PageClientImpl::didConcludeEditDrag):
(WebKit::PageClientImpl::didChangeDragCaretRect):
(WebKit::PageClientImpl::didHandleStartDataInteractionRequest): Deleted.
(WebKit::PageClientImpl::didConcludeEditDataInteraction): Deleted.
(WebKit::PageClientImpl::didChangeDataInteractionCaretRect): Deleted.
* UIProcess/ios/WKContentViewInteraction.h:
* UIProcess/ios/WKContentViewInteraction.mm:
(-[WKContentView setupInteraction]):
(-[WKContentView cleanupInteraction]):
(-[WKContentView setupDragAndDropInteractions]):
(-[WKContentView teardownDragAndDropInteractions]):
(-[WKContentView _didHandleDragStartRequest:]):
(-[WKContentView _didConcludeEditDrag:]):
(-[WKContentView _didChangeDragCaretRect:currentRect:]):
(-[WKContentView _dragInteraction:prepareForSession:completion:]):
(-[WKContentView setupDataInteractionDelegates]): Deleted.
(-[WKContentView teardownDataInteractionDelegates]): Deleted.
(-[WKContentView _didHandleStartDataInteractionRequest:]): Deleted.
(-[WKContentView _didConcludeEditDataInteraction:]): Deleted.
(-[WKContentView _didChangeDataInteractionCaretRect:currentRect:]): Deleted.
* UIProcess/ios/WebPageProxyIOS.mm:
(WebKit::WebPageProxy::didHandleDragStartRequest):
(WebKit::WebPageProxy::requestDragStart):
(WebKit::WebPageProxy::didConcludeEditDrag):
(WebKit::WebPageProxy::didHandleStartDataInteractionRequest): Deleted.
(WebKit::WebPageProxy::requestStartDataInteraction): Deleted.
(WebKit::WebPageProxy::didConcludeEditDataInteraction): Deleted.
* WebProcess/WebCoreSupport/mac/WebDragClientMac.mm:
(WebKit::WebDragClient::didConcludeEditDrag):
* WebProcess/WebPage/WebPage.h:
* WebProcess/WebPage/WebPage.messages.in:
* WebProcess/WebPage/ios/WebPageIOS.mm:
(WebKit::WebPage::requestDragStart):
(WebKit::WebPage::requestAdditionalItemsForDragSession):
(WebKit::WebPage::didConcludeEditDrag):
(WebKit::WebPage::requestStartDataInteraction): Deleted.
(WebKit::WebPage::didConcludeEditDataInteraction): Deleted.

Source/WebKitLegacy/mac:

* WebCoreSupport/WebDragClient.mm:
(WebDragClient::didConcludeEditDrag):
* WebView/WebView.mm:
(-[WebView _requestStartDataInteraction:globalPosition:]):
(-[WebView _didConcludeEditDrag]):
(-[WebView _didConcludeEditDataInteraction]): Deleted.
* WebView/WebViewInternal.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@240491 268f45cc-cd09-0410-ab3c-d52691b4dbfc

6 months agoWKWebView.goBack should reload if there is a safe browsing warning
achristensen@apple.com [Fri, 25 Jan 2019 19:42:27 +0000 (19:42 +0000)]
WKWebView.goBack should reload if there is a safe browsing warning
https://bugs.webkit.org/show_bug.cgi?id=193805
<rdar://problem/46908216>

Reviewed by Geoff Garen.

Source/WebKit:

If a WKWebView is showing a safe browsing warning and the user clicks a back button
in the app which calls WKWebView.goBack, the WKWebView is in a state where it has not navigated yet,
so actually going back will appear to the user to go back twice.  We can't just do nothing because the
app is in a state where it is expecting a navigation to happen.  Reloading achieves what the user expects
and makes the app work like the app expects.

* UIProcess/API/C/WKPage.cpp:
(WKPageGoBack):
* UIProcess/API/Cocoa/WKWebView.mm:
(-[WKWebView goBack]):
* UIProcess/PageClient.h:
(WebKit::PageClient::hasSafeBrowsingWarning const):
* UIProcess/mac/PageClientImplMac.h:
* UIProcess/mac/PageClientImplMac.mm:
(WebKit::PageClientImpl::hasSafeBrowsingWarning const):

Tools:

* TestWebKitAPI/Tests/WebKitCocoa/SafeBrowsing.mm:
(+[Simple3LookupContext sharedLookupContext]):
(-[Simple3LookupContext lookUpURL:completionHandler:]):
(-[WKWebViewGoBackNavigationDelegate webView:didFinishNavigation:]):
(TEST):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@240490 268f45cc-cd09-0410-ab3c-d52691b4dbfc

6 months agoWeb Inspector: REGRESSION (r237808): offscreen path warning doesn't work
drousso@apple.com [Fri, 25 Jan 2019 19:39:38 +0000 (19:39 +0000)]
Web Inspector: REGRESSION (r237808): offscreen path warning doesn't work
https://bugs.webkit.org/show_bug.cgi?id=193830

Reviewed by Matt Baker.

* UserInterface/Models/RecordingAction.js:
(WI.RecordingAction.prototype.process):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@240489 268f45cc-cd09-0410-ab3c-d52691b4dbfc

6 months agoWeb Inspector: Add another Protocol Version (iOS 12.2)
commit-queue@webkit.org [Fri, 25 Jan 2019 19:37:38 +0000 (19:37 +0000)]
Web Inspector: Add another Protocol Version (iOS 12.2)
https://bugs.webkit.org/show_bug.cgi?id=193810
<rdar://problem/42981838>

Patch by Joseph Pecoraro <pecoraro@apple.com> on 2019-01-25
Reviewed by Matt Baker.

* UserInterface/Protocol/Legacy/12.2/InspectorBackendCommands.js: Added.
* Versions/Inspector-iOS-12.2.json: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@240488 268f45cc-cd09-0410-ab3c-d52691b4dbfc

6 months agoWeb Inspector: `WI.Setting.migrateValue` doesn't take into account the key prefix
drousso@apple.com [Fri, 25 Jan 2019 19:35:29 +0000 (19:35 +0000)]
Web Inspector: `WI.Setting.migrateValue` doesn't take into account the key prefix
https://bugs.webkit.org/show_bug.cgi?id=193814

Reviewed by Matt Baker.

* UserInterface/Base/Setting.js:
(WI.Setting):
(WI.Setting.migrateValue):
(WI.Setting._localStorageKey): Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@240487 268f45cc-cd09-0410-ab3c-d52691b4dbfc

6 months agoRegression(PSON) cross-site provisional page is not canceled if a new same-site one...
cdumez@apple.com [Fri, 25 Jan 2019 18:54:54 +0000 (18:54 +0000)]
Regression(PSON) cross-site provisional page is not canceled if a new same-site one is started
https://bugs.webkit.org/show_bug.cgi?id=193788
<rdar://problem/47531231>

Reviewed by Alex Christensen.

Source/WebKit:

When the page starts a new provisional load, make sure we cancel any pending one in the provisional
process, as it would have happened in the first provisional load happened in the same process.
Without this, we could have 2 parallel loads happening, one in the committed process and another
in the provisional one, leading to assertion failures in debug.

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

Tools:

Add API test coverage.

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

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@240485 268f45cc-cd09-0410-ab3c-d52691b4dbfc

6 months agoDrop WebKit::WebKitPolicyAction type as it is no longer needed
cdumez@apple.com [Fri, 25 Jan 2019 18:49:01 +0000 (18:49 +0000)]
Drop WebKit::WebKitPolicyAction type as it is no longer needed
https://bugs.webkit.org/show_bug.cgi?id=193827

Reviewed by Antti Koivisto.

Drop WebKit::WebKitPolicyAction type as it is no longer needed. It is now identical to
WebCore::PolicyAction.

* Shared/WebPolicyAction.h: Removed.
* UIProcess/WebFramePolicyListenerProxy.cpp:
(WebKit::WebFramePolicyListenerProxy::didReceiveSafeBrowsingResults):
(WebKit::WebFramePolicyListenerProxy::use):
(WebKit::WebFramePolicyListenerProxy::download):
(WebKit::WebFramePolicyListenerProxy::ignore):
* UIProcess/WebFramePolicyListenerProxy.h:
* UIProcess/WebFrameProxy.cpp:
(WebKit::WebFrameProxy::setUpPolicyListenerProxy):
* UIProcess/WebFrameProxy.h:
* UIProcess/WebPageProxy.cpp:
(WebKit::WebPageProxy::receivedNavigationPolicyDecision):
(WebKit::WebPageProxy::receivedPolicyDecision):
(WebKit::WebPageProxy::decidePolicyForNavigationAction):
(WebKit::WebPageProxy::decidePolicyForNavigationActionSync):
(WebKit::WebPageProxy::decidePolicyForNewWindowAction):
(WebKit::WebPageProxy::decidePolicyForResponseShared):
* UIProcess/WebPageProxy.h:
* WebKit.xcodeproj/project.pbxproj:
* WebProcess/WebCoreSupport/WebFrameLoaderClient.cpp:
(WebKit::WebFrameLoaderClient::dispatchDecidePolicyForResponse):
(WebKit::WebFrameLoaderClient::dispatchDecidePolicyForNavigationAction):
* WebProcess/WebPage/WebFrame.cpp:
(WebKit::WebFrame::didReceivePolicyDecision):
(WebKit::toPolicyAction): Deleted.
* WebProcess/WebPage/WebFrame.h:
* WebProcess/WebPage/WebPage.cpp:
(WebKit::WebPage::didReceivePolicyDecision):
* WebProcess/WebPage/WebPage.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@240484 268f45cc-cd09-0410-ab3c-d52691b4dbfc

6 months agowebkitpy: Missing PID in crashlog name should not be fatal (Follow-up fix)
jbedard@apple.com [Fri, 25 Jan 2019 18:48:29 +0000 (18:48 +0000)]
webkitpy: Missing PID in crashlog name should not be fatal (Follow-up fix)
https://bugs.webkit.org/show_bug.cgi?id=193771

Unreviewed infrastructure fix.

* Scripts/webkitpy/port/darwin.py:
(DarwinPort._merge_crash_logs): Correct syntax for catching multiple exception types.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@240483 268f45cc-cd09-0410-ab3c-d52691b4dbfc

6 months agoSource/WebCore:
jond@apple.com [Fri, 25 Jan 2019 18:46:49 +0000 (18:46 +0000)]
Source/WebCore:
Updated feature status for several features
https://bugs.webkit.org/show_bug.cgi?id=193794

Reviewed by Joseph Pecoraro.

Updated feature status for the following: CSS Font Display,
CSS Text Decoration Level 4, SVG in OpenType Fonts, Web SQL,
File and Directory Entries API, MediaStream Recording API,
Readable Streams, Subresource Integrity, Visual Viewport API,
and Web Audio.

* features.json:

Websites/webkit.org:
Added feature status entry for "Removed in Preview"
https://bugs.webkit.org/show_bug.cgi?id=193794

Reviewed by Joseph Pecoraro.

* wp-content/themes/webkit/status.php:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@240482 268f45cc-cd09-0410-ab3c-d52691b4dbfc

6 months agoUpdate JavaScriptCore feature status entries.
jond@apple.com [Fri, 25 Jan 2019 18:45:09 +0000 (18:45 +0000)]
Update JavaScriptCore feature status entries.
https://bugs.webkit.org/show_bug.cgi?id=193797

Reviewed by Mark Lam.

Updated feature status for Async Iteration, and Object rest/spread.

* features.json:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@240481 268f45cc-cd09-0410-ab3c-d52691b4dbfc

6 months ago* Scripts/webkitpy/common/config/contributors.json: Removing a stray comma.
ap@apple.com [Fri, 25 Jan 2019 18:42:00 +0000 (18:42 +0000)]
* Scripts/webkitpy/common/config/contributors.json: Removing a stray comma.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@240480 268f45cc-cd09-0410-ab3c-d52691b4dbfc

6 months agoAdd contributors.json entries for Ling Ho and for a new bot watchers group.
ap@apple.com [Fri, 25 Jan 2019 18:39:48 +0000 (18:39 +0000)]
Add contributors.json entries for Ling Ho and for a new bot watchers group.

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

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@240479 268f45cc-cd09-0410-ab3c-d52691b4dbfc

6 months ago[iOS] Add logging of calls
pvollan@apple.com [Fri, 25 Jan 2019 17:58:58 +0000 (17:58 +0000)]
[iOS] Add logging of calls
https://bugs.webkit.org/show_bug.cgi?id=193784

Reviewed by Brent Fulgham.

Add permissive logging of calls on iOS.

* Resources/SandboxProfiles/ios/com.apple.WebKit.WebContent.sb:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@240478 268f45cc-cd09-0410-ab3c-d52691b4dbfc

6 months agoRegression(PSON?) Crash under NavigationState::NavigationClient::decidePolicyForNavig...
cdumez@apple.com [Fri, 25 Jan 2019 17:32:53 +0000 (17:32 +0000)]
Regression(PSON?) Crash under NavigationState::NavigationClient::decidePolicyForNavigationAction()
https://bugs.webkit.org/show_bug.cgi?id=193779
<rdar://problem/46170903>

Reviewed by Antti Koivisto.

Source/WebKit:

* UIProcess/Cocoa/NavigationState.mm:
(WebKit::tryAppLink):
(WebKit::NavigationState::NavigationClient::decidePolicyForNavigationAction):
We were crashing when trying to get the URL of the main frame, which was sad because we never
ended up using the main frame URL. Therefore, this patch drops the code in question.

* UIProcess/ProvisionalPageProxy.cpp:
(WebKit::ProvisionalPageProxy::decidePolicyForNavigationActionAsync):
Add assertion to make sure that the DecidePolicyForNavigationActionAsync IPC it is getting
from the process is related to its main frame.

Tools:

Add API test that quickly navigates forward to a previous process without waiting for it to
suspend. I suspect the crash could have been happening due to receiving leftover IPC from
the process' previous page when reconnecting the it for the forward navigation.

* TestWebKitAPI/Tests/WebKitCocoa/ProcessSwapOnNavigation.mm:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@240477 268f45cc-cd09-0410-ab3c-d52691b4dbfc

6 months agoNeed a way for JavaScript (or bundle) code to participate in undo
wenson_hsieh@apple.com [Fri, 25 Jan 2019 17:23:06 +0000 (17:23 +0000)]
Need a way for JavaScript (or bundle) code to participate in undo
https://bugs.webkit.org/show_bug.cgi?id=190009
<rdar://problem/44807048>

Reviewed by Ryosuke Niwa.

Source/WebCore:

Finish hooking up `UndoManager::addItems()` to CustomUndoStep.

Tests: editing/undo-manager/undo-manager-add-item-exceptions.html
       editing/undo-manager/undo-manager-add-item.html
       editing/undo-manager/undo-manager-delete-stale-undo-items.html
       editing/undo-manager/undo-manager-item-labels.html
       editing/undo-manager/undo-manager-undo-redo-after-garbage-collection.html

* editing/CompositeEditCommand.h:
* editing/CustomUndoStep.cpp:
(WebCore::CustomUndoStep::didRemoveFromUndoManager):

Add a method to invalidate CustomUndoStep. This clears out the pointer to the undo item, and also invalidates
the UndoItem, removing it from its UndoManager.

* editing/CustomUndoStep.h:
* editing/Editor.cpp:
(WebCore::Editor::registerCustomUndoStep):

Add a helper method to register a CustomUndoStep as a platform undoable step.

* editing/Editor.h:
* editing/UndoStep.h:
* page/UndoItem.h:
(WebCore::UndoItem::undoManager const):
* page/UndoManager.cpp:
(WebCore::UndoManager::addItem):

Create a CustomUndoStep with the given UndoItem, and register it with the platform undo manager.

* page/UndoManager.h:
* page/UndoManager.idl:

Mark addItem() as capable of throwing exceptions.

Source/WebKit:

Invalidate undo steps when removing them from WebPage. Invalidation is a no-op for editing actions that come
from the UA, but for custom undo steps backed by an UndoItem, we clear out the custom undo step's pointer to its
UndoItem and additionally disconnect the UndoItem from its UndoManager.

* WebProcess/WebPage/WebPage.cpp:
(WebKit::WebPage::addWebUndoStep):
(WebKit::WebPage::removeWebEditCommand):
* WebProcess/WebPage/WebUndoStep.h:
(WebKit::WebUndoStep::invalidate):

Tools:

Add UIScriptController helpers to grab the platform undo and redo action labels. Currently only implemented for
Cocoa platforms in WebKit2. See other ChangeLogs for more detail.

* DumpRenderTree/ios/UIScriptControllerIOS.mm:
(WTR::UIScriptController::lastUndoLabel const):
(WTR::UIScriptController::firstRedoLabel const):
(WTR::UIScriptController::platformUndoManager const):
* DumpRenderTree/mac/UIScriptControllerMac.mm:
(WTR::UIScriptController::lastUndoLabel const):
(WTR::UIScriptController::firstRedoLabel const):
(WTR::UIScriptController::platformUndoManager const):
* TestRunnerShared/UIScriptContext/Bindings/UIScriptController.idl:
* TestRunnerShared/UIScriptContext/UIScriptController.cpp:
(WTR::UIScriptController::lastUndoLabel const):
(WTR::UIScriptController::firstRedoLabel const):
* TestRunnerShared/UIScriptContext/UIScriptController.h:
* WebKitTestRunner/UIScriptControllerCocoa.mm:
(WTR::UIScriptController::lastUndoLabel const):
(WTR::UIScriptController::firstRedoLabel const):
* WebKitTestRunner/ios/UIScriptControllerIOS.mm:
(WTR::UIScriptController::platformUndoManager const):
* WebKitTestRunner/mac/UIScriptControllerMac.mm:
(WTR::UIScriptController::platformUndoManager const):

LayoutTests:

Add a few new layout tests covering `UndoManager.addItem()`.

* editing/undo-manager/undo-manager-add-item-exceptions-expected.txt: Added.
* editing/undo-manager/undo-manager-add-item-exceptions.html: Added.

Add a test to verify that we throw exceptions when calling addItem() in a couple of circumstances.

* editing/undo-manager/undo-manager-add-item-expected.txt: Added.
* editing/undo-manager/undo-manager-add-item.html: Added.

Add a test that exercises the new API in both the top-level context and a child frame.

* editing/undo-manager/undo-manager-delete-stale-undo-items-expected.txt: Added.
* editing/undo-manager/undo-manager-delete-stale-undo-items.html: Added.

Add a test to verify that after adding undo items, undoing, and then performing other edit actions, garbage
collection will destroy JS wrappers for the previously added UndoItems, since these undo items' handlers can no
longer be invoked.

* editing/undo-manager/undo-manager-item-labels-expected.txt: Added.
* editing/undo-manager/undo-manager-item-labels.html: Added.

Add a test verifying that the undo and redo action labels are updated correctly when undoing and redoing.

* editing/undo-manager/undo-manager-undo-redo-after-garbage-collection-expected.txt: Added.
* editing/undo-manager/undo-manager-undo-redo-after-garbage-collection.html: Added.

Add a test to verify that triggering garbage collection after adding an undo item without keeping references to
the item (or its undo/redo handlers) doesn't break the API.

* resources/ui-helper.js:
(window.UIHelper.undoAndRedoLabels):

Add a helper method to grab the platform's current undo and redo action names.

(window.UIHelper):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@240476 268f45cc-cd09-0410-ab3c-d52691b4dbfc

6 months ago[LFC][BFC][MarginCollapsing] Add "clear" to static position computation.
zalan@apple.com [Fri, 25 Jan 2019 16:50:18 +0000 (16:50 +0000)]
[LFC][BFC][MarginCollapsing] Add "clear" to static position computation.
https://bugs.webkit.org/show_bug.cgi?id=193824

Reviewed by Antti Koivisto.

Source/WebCore:

When clear property is set and floats are present, we have to estimate and set the box's vertical position during
static positioning to be able to properly layout its subtree.

<div style="float: left; width: 100px; height: 100px;"></div>
<div style="clear: left;">
  <div style="float: left; width: 100px; height: 100px;"></div>
</div>

In the above example since the second float's parent clears the first float, the second float is positioned below
the first float. If we didn't push down (clear) the box, the float child would get placed next to the first float.

* layout/blockformatting/BlockFormattingContext.cpp:
(WebCore::Layout::BlockFormattingContext::layout const):
(WebCore::Layout::BlockFormattingContext::layoutFormattingContextRoot const):
(WebCore::Layout::BlockFormattingContext::computeStaticPosition const):
(WebCore::Layout::BlockFormattingContext::computeEstimatedVerticalPosition const):
(WebCore::Layout::BlockFormattingContext::computeEstimatedVerticalPositionForFloatClear const):
(WebCore::Layout::BlockFormattingContext::computeHeightAndMargin const):
(WebCore::Layout::BlockFormattingContext::verticalPositionWithMargin const):
(WebCore::Layout::BlockFormattingContext::computeVerticalPositionForFloatClear const): Deleted.
(WebCore::Layout::BlockFormattingContext::adjustedVerticalPositionAfterMarginCollapsing const): Deleted.
* layout/blockformatting/BlockFormattingContext.h:
* layout/blockformatting/BlockMarginCollapse.cpp:
(WebCore::Layout::BlockFormattingContext::MarginCollapse::estimatedMarginBefore):
* layout/displaytree/DisplayBox.h:

Tools:

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

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@240475 268f45cc-cd09-0410-ab3c-d52691b4dbfc

6 months ago[LFC][BFC][MarginCollapsing] Move positive/negative margin value updating to a dedica...
zalan@apple.com [Fri, 25 Jan 2019 16:46:47 +0000 (16:46 +0000)]
[LFC][BFC][MarginCollapsing] Move positive/negative margin value updating to a dedicated function
https://bugs.webkit.org/show_bug.cgi?id=193812

Reviewed by Antti Koivisto.

Move update logic to BlockFormattingContext::MarginCollapse::updatePositiveNegativeMarginValues().

* layout/blockformatting/BlockFormattingContext.cpp:
(WebCore::Layout::BlockFormattingContext::computeHeightAndMargin const):
* layout/blockformatting/BlockFormattingContext.h:
* layout/blockformatting/BlockMarginCollapse.cpp:
(WebCore::Layout::BlockFormattingContext::MarginCollapse::updatePositiveNegativeMarginValues):
(WebCore::Layout::BlockFormattingContext::MarginCollapse::collapsedVerticalValues):
* page/FrameViewLayoutContext.cpp:
(WebCore::layoutUsingFormattingContext):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@240474 268f45cc-cd09-0410-ab3c-d52691b4dbfc

6 months ago[GTK][WPE] Add API to add paths to sandbox
commit-queue@webkit.org [Fri, 25 Jan 2019 15:54:54 +0000 (15:54 +0000)]
[GTK][WPE] Add API to add paths to sandbox
https://bugs.webkit.org/show_bug.cgi?id=193571

This allows applications to add paths to the web process
if required by web extensions.

Patch by Patrick Griffis <pgriffis@igalia.com> on 2019-01-25
Reviewed by Michael Catanzaro.

* UIProcess/API/glib/WebKitWebContext.cpp:
(webkit_web_context_add_path_to_sandbox):
* UIProcess/API/gtk/WebKitWebContext.h:
* UIProcess/API/gtk/docs/webkit2gtk-4.0-sections.txt:
* UIProcess/API/wpe/docs/wpe-0.1-sections.txt:
* UIProcess/Launcher/glib/BubblewrapLauncher.cpp:
(WebKit::bubblewrapSpawn):
* UIProcess/WebProcessPool.h:
* UIProcess/glib/WebProcessProxyGLib.cpp:
(WebKit::WebProcessProxy::platformGetLaunchOptions):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@240473 268f45cc-cd09-0410-ab3c-d52691b4dbfc

6 months agoUse ENABLE_POINTER_EVENTS for the touch-action property
graouts@webkit.org [Fri, 25 Jan 2019 11:36:57 +0000 (11:36 +0000)]
Use ENABLE_POINTER_EVENTS for the touch-action property
https://bugs.webkit.org/show_bug.cgi?id=193819

Reviewed by Antti Koivisto.

Since we've added an ENABLE_POINTER_EVENTS we should be using it for anything related to the implementation of the
Pointer Events specification of which the touch-action property is a part.

* css/CSSComputedStyleDeclaration.cpp:
(WebCore::ComputedStyleExtractor::valueForPropertyinStyle):
* css/CSSPrimitiveValueMappings.h:
* css/CSSProperties.json:
* css/CSSValueKeywords.in:
* css/StyleBuilderConverter.h:
* css/parser/CSSPropertyParser.cpp:
(WebCore::CSSPropertyParser::parseSingleValue):
* dom/Element.cpp:
(WebCore::Element::allowsDoubleTapGesture const):
* platform/TouchAction.h:
* rendering/style/RenderStyle.h:
* rendering/style/StyleRareNonInheritedData.cpp:
(WebCore::StyleRareNonInheritedData::StyleRareNonInheritedData):
(WebCore::StyleRareNonInheritedData::operator== const):
* rendering/style/StyleRareNonInheritedData.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@240472 268f45cc-cd09-0410-ab3c-d52691b4dbfc

6 months agoWeb Inspector: improve invalid Audit/Recording JSON error messages
drousso@apple.com [Fri, 25 Jan 2019 09:27:50 +0000 (09:27 +0000)]
Web Inspector: improve invalid Audit/Recording JSON error messages
https://bugs.webkit.org/show_bug.cgi?id=193476
<rdar://problem/47303659>

Reviewed by Joseph Pecoraro.

Source/WebInspectorUI:

* UserInterface/Models/AuditTestBase.js:
(WI.AuditTestBase):
* UserInterface/Models/AuditTestCase.js:
(WI.AuditTestCase.async fromPayload):
* UserInterface/Models/AuditTestGroup.js:
(WI.AuditTestGroup.async fromPayload):
* UserInterface/Models/AuditTestCaseResult.js:
(WI.AuditTestCaseResult.async fromPayload.checkArray):
(WI.AuditTestCaseResult.async fromPayload):
* UserInterface/Models/AuditTestGroupResult.js:
(WI.AuditTestGroupResult.async fromPayload):
* UserInterface/Controllers/AuditManager.js:
(WI.AuditManager.synthesizeWarning): Added.
(WI.AuditManager.synthesizeError):
(WI.AuditManager.prototype.async processJSON):

* UserInterface/Models/Recording.js:
(WI.Recording.fromPayload):
(WI.Recording.synthesizeWarning): Added.
(WI.Recording.synthesizeError):
* UserInterface/Models/RecordingFrame.js:
(WI.RecordingFrame.fromPayload):
* UserInterface/Models/RecordingAction.js:
(WI.RecordingAction.fromPayload):
(WI.RecordingAction.prototype.async swizzle):
(WI.RecordingAction.prototype.apply):
* UserInterface/Controllers/CanvasManager.js:
(WI.CanvasManager.prototype.processJSON):

* Localizations/en.lproj/localizedStrings.js:

LayoutTests:

* inspector/model/auditTestCase.html:
* inspector/model/auditTestCase-expected.txt:
* inspector/model/auditTestCaseResult-expected.txt:
* inspector/model/auditTestGroup.html:
* inspector/model/auditTestGroup-expected.txt:
* inspector/model/auditTestGroupResult-expected.txt:
* inspector/model/recording-expected.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@240471 268f45cc-cd09-0410-ab3c-d52691b4dbfc

6 months agoUpdate macOS JSC bot configurations
ryanhaddad@apple.com [Fri, 25 Jan 2019 07:12:00 +0000 (07:12 +0000)]
Update macOS JSC bot configurations
https://bugs.webkit.org/show_bug.cgi?id=193739

Reviewed by Alexey Proskuryakov.

* BuildSlaveSupport/build.webkit.org-config/config.json: Remove 32-bit BuildAndTest queue.
* BuildSlaveSupport/build.webkit.org-config/public_html/dashboard/Scripts/BubbleQueueServer.js:
(BubbleQueueServer): Move JSC EWS to Mojave section.
* BuildSlaveSupport/build.webkit.org-config/public_html/dashboard/Scripts/WebKitBuildbot.js:
(WebKitBuildbot): Remove 32-bit BuildAndTest queue from dashboard.
* BuildSlaveSupport/build.webkit.org-config/steps_unittest.py: Update unit test.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@240470 268f45cc-cd09-0410-ab3c-d52691b4dbfc

6 months agoWeb Inspector: Audit: add supports key to test/group for compatibility
drousso@apple.com [Fri, 25 Jan 2019 06:47:49 +0000 (06:47 +0000)]
Web Inspector: Audit: add supports key to test/group for compatibility
https://bugs.webkit.org/show_bug.cgi?id=193686
<rdar://problem/47460872>

Reviewed by Joseph Pecoraro.

Source/WebInspectorUI:

* UserInterface/Models/AuditTestBase.js:
(WI.AuditTestBase):
(WI.AuditTestBase.prototype.get supported): Added.
(WI.AuditTestBase.prototype.set supported): Added.
(WI.AuditTestBase.prototype.set disabled):
(WI.AuditTestBase.prototype.async start):
(WI.AuditTestBase.prototype.stop):
(WI.AuditTestBase.toJSON):
* UserInterface/Models/AuditTestCase.js:
(WI.AuditTestCase.async fromPayload):
* UserInterface/Models/AuditTestGroup.js:
(WI.AuditTestGroup):
(WI.AuditTestGroup.async fromPayload):
(WI.AuditTestGroup.prototype.get supported): Added.
(WI.AuditTestGroup.prototype.set supported): Added.
Add support for a "supports" key in the test JSON that prevents the test from being run if
it's value is lower than the frontend/backend version.

* UserInterface/Views/AuditTreeElement.js:
(WI.AuditTreeElement.prototype.onattach):
(WI.AuditTreeElement.prototype.ondelete):
(WI.AuditTreeElement.prototype.populateContextMenu):
(WI.AuditTreeElement.prototype._updateStatus): Added.
(WI.AuditTreeElement.prototype._showRunningSpinner):
(WI.AuditTreeElement.prototype._showRunningProgress):
(WI.AuditTreeElement.prototype._updateTestGroupDisabled):
(WI.AuditTreeElement.prototype._handleTestCaseCompleted):
(WI.AuditTreeElement.prototype._handleTestResultCleared):
(WI.AuditTreeElement.prototype._handleTestGroupCompleted):
(WI.AuditTreeElement.prototype._handleManagerEditingChanged):
(WI.AuditTreeElement.prototype.canSelectOnMouseDown): Deleted.
(WI.AuditTreeElement.prototype._updateLevel): Deleted.
* UserInterface/Views/AuditTreeElement.css:
(.tree-outline .item.audit:matches(.test-case, .test-group):not(.unsupported, .manager-active) > .status:hover > img): Added.
(.tree-outline .item.audit > .status:not(:hover) > img.show-on-hover, .tree-outline .item.audit.test-group.expanded:not(.unsupported, .editing-audits) > .status:not(:hover)): Added.
(.tree-outline .item.audit.manager-active > .status > img.show-on-hover, .tree-outline .item.audit.test-group.expanded:not(.editing-audits) > .status:hover > :not(img), .tree-outline .item.audit.test-group-result.expanded > .status, .tree-outline .item.audit.unsupported + .children .item.audit.unsupported  > .status > img): Added.
(.tree-outline .item.audit.unsupported:not(.selected) > :matches(.icon, .titles)): Added.
(.tree-outline .item.audit.unsupported > .status > img): Added.
(.tree-outline .item.audit:matches(.test-case, .test-group):not(.manager-active) > .status:hover > img): Deleted.
(.tree-outline .item.audit > .status:not(:hover) > img.show-on-hover, .tree-outline .item.audit.test-group.expanded:not(.editing-audits) > .status:not(:hover)): Deleted.
(.tree-outline .item.audit.manager-active > .status > img.show-on-hover, .tree-outline .item.audit.test-group.expanded:not(.editing-audits) > .status:hover > :not(img), .tree-outline .item.audit.test-group-result.expanded > .status): Deleted.
Hide unsupported tests unless in edit mode, where they are greyed out and cannot be enabled.
Drive-by: only allow tests to be deleted when in edit mode.
* UserInterface/Views/AuditNavigationSidebarPanel.js:
(WI.AuditNavigationSidebarPanel.prototype._treeSelectionDidChange):

* Localizations/en.lproj/localizedStrings.js:

LayoutTests:

* inspector/audit/version.html:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@240469 268f45cc-cd09-0410-ab3c-d52691b4dbfc

6 months agoRemove usage of internal macro from private header
keith_miller@apple.com [Fri, 25 Jan 2019 04:39:32 +0000 (04:39 +0000)]
Remove usage of internal macro from private header
https://bugs.webkit.org/show_bug.cgi?id=193809

Reviewed by Saam Barati.

Also, add a new file to include all of our API headers to make sure
they don't accidentally include C++ or internal values.

* API/JSScript.h:
* API/tests/testIncludes.m: Added.
* JavaScriptCore.xcodeproj/project.pbxproj:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@240468 268f45cc-cd09-0410-ab3c-d52691b4dbfc

6 months ago[LFC][BFC][MarginCollapsing] Refactor MarginCollapse::updateCollapsedMarginAfter
zalan@apple.com [Fri, 25 Jan 2019 04:14:32 +0000 (04:14 +0000)]
[LFC][BFC][MarginCollapsing] Refactor MarginCollapse::updateCollapsedMarginAfter
https://bugs.webkit.org/show_bug.cgi?id=193807

Reviewed by Simon Fraser.

Rename updateCollapsedMarginAfter to updateMarginAfterForPreviousSibling and make the margin updating logic more explicit.

* layout/blockformatting/BlockFormattingContext.cpp:
(WebCore::Layout::BlockFormattingContext::computeHeightAndMargin const):
* layout/blockformatting/BlockFormattingContext.h:
* layout/blockformatting/BlockMarginCollapse.cpp:
(WebCore::Layout::BlockFormattingContext::MarginCollapse::updateMarginAfterForPreviousSibling):
(WebCore::Layout::BlockFormattingContext::MarginCollapse::updateCollapsedMarginAfter): Deleted.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@240467 268f45cc-cd09-0410-ab3c-d52691b4dbfc

6 months agoiOS: Split keyboard should not shrink visualViewport.height
rniwa@webkit.org [Fri, 25 Jan 2019 02:53:27 +0000 (02:53 +0000)]
iOS: Split keyboard should not shrink visualViewport.height
https://bugs.webkit.org/show_bug.cgi?id=193798

Reviewed by Tim Horton.

Treat a split keyboard like a floating keyboard and don't consider its input view bounds in computing the visible content rects.

No new tests since while it's possible to transition between split and merged keyboard using some SPI,
there isn't a reliable mechanism to reset the keyboard state. e.g. we need to wait for a hard-coded amount of time.

* Platform/spi/ios/UIKitSPI.h:
* UIProcess/API/Cocoa/WKWebView.mm:
(-[WKWebView _keyboardChangedWithInfo:adjustScrollView:]):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@240466 268f45cc-cd09-0410-ab3c-d52691b4dbfc

6 months ago[JSC] ErrorConstructor should not have own IsoSubspace
ysuzuki@apple.com [Fri, 25 Jan 2019 02:47:58 +0000 (02:47 +0000)]
[JSC] ErrorConstructor should not have own IsoSubspace
https://bugs.webkit.org/show_bug.cgi?id=193800

Reviewed by Saam Barati.

Similar to r240456, sizeof(ErrorConstructor) != sizeof(InternalFunction), and that is why we have
IsoSubspace errorConstructorSpace in VM. But it is allocated only one-per-JSGlobalObject, and it is
too costly to have IsoSubspace which allocates 16KB. Since stackTraceLimit information is per
JSGlobalObject information, we should have m_stackTraceLimit in JSGlobalObject instead and put
ErrorConstructor in InternalFunction's IsoSubspace. As r230813 (moving InternalFunction and subclasses
into IsoSubspaces) described,

    "subclasses that are the same size as InternalFunction share its subspace. I did this because the subclasses
    appear to just override methods, which are called dynamically via the structure or class of the object.
    So, I don't see a type confusion risk if UAF is used to allocate one kind of InternalFunction over another."

Then, putting ErrorConstructor in InternalFunction IsoSubspace is fine since it meets the above condition.
This patch removes m_stackTraceLimit in ErrorConstructor, and drops IsoSubspace for errorConstructorSpace.
This reduces the memory usage.

* interpreter/Interpreter.h:
* runtime/Error.cpp:
(JSC::getStackTrace):
* runtime/ErrorConstructor.cpp:
(JSC::ErrorConstructor::ErrorConstructor):
(JSC::ErrorConstructor::finishCreation):
(JSC::constructErrorConstructor):
(JSC::callErrorConstructor):
(JSC::ErrorConstructor::put):
(JSC::ErrorConstructor::deleteProperty):
(JSC::Interpreter::constructWithErrorConstructor): Deleted.
(JSC::Interpreter::callErrorConstructor): Deleted.
* runtime/ErrorConstructor.h:
* runtime/JSGlobalObject.cpp:
(JSC::JSGlobalObject::JSGlobalObject):
(JSC::JSGlobalObject::init):
(JSC::JSGlobalObject::visitChildren):
* runtime/JSGlobalObject.h:
(JSC::JSGlobalObject::stackTraceLimit const):
(JSC::JSGlobalObject::setStackTraceLimit):
(JSC::JSGlobalObject::errorConstructor const): Deleted.
* runtime/VM.cpp:
(JSC::VM::VM):
* runtime/VM.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@240465 268f45cc-cd09-0410-ab3c-d52691b4dbfc

6 months ago[iOS] Silence MediaPlayer compile warnings
eric.carlson@apple.com [Fri, 25 Jan 2019 02:00:01 +0000 (02:00 +0000)]
[iOS] Silence MediaPlayer compile warnings
https://bugs.webkit.org/show_bug.cgi?id=193780
<rdar://problem/47518428>

Reviewed by Jer Noble.

Source/WebCore/PAL:

* pal/spi/ios/MediaPlayerSPI.h:

Source/WebKit:

* UIProcess/ios/forms/WKAirPlayRoutePicker.mm:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@240461 268f45cc-cd09-0410-ab3c-d52691b4dbfc

6 months ago[Win][WebKitTestRunner] Implement EventSenderProxy
Hironori.Fujii@sony.com [Fri, 25 Jan 2019 01:49:10 +0000 (01:49 +0000)]
[Win][WebKitTestRunner] Implement EventSenderProxy
https://bugs.webkit.org/show_bug.cgi?id=193755

Reviewed by Don Olmstead.

* WebKitTestRunner/EventSenderProxy.h: Removed unused members m_buttonState and m_mouseButtonsCurrentlyDown.
* WebKitTestRunner/win/EventSenderProxyWin.cpp:
(WTR::EventSenderProxy::dispatchMessage):
(WTR::EventSenderProxy::EventSenderProxy):
(WTR::EventSenderProxy::mouseDown):
(WTR::EventSenderProxy::mouseUp):
(WTR::EventSenderProxy::mouseMoveTo):
(WTR::EventSenderProxy::mouseScrollBy):
(WTR::EventSenderProxy::mouseScrollByWithWheelAndMomentumPhases):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@240460 268f45cc-cd09-0410-ab3c-d52691b4dbfc

6 months agoUnreviewed fix for API test failure after r240437.
ross.kirsling@sony.com [Fri, 25 Jan 2019 01:16:35 +0000 (01:16 +0000)]
Unreviewed fix for API test failure after r240437.

TestWTF.WTF.StringOperators apparently can't deal with another test using String::operator+. This is fine...

* TestWebKitAPI/Tests/WTF/FileSystem.cpp:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@240459 268f45cc-cd09-0410-ab3c-d52691b4dbfc

6 months agoWeb Inspector: CPU Usage Timeline
commit-queue@webkit.org [Fri, 25 Jan 2019 01:06:17 +0000 (01:06 +0000)]
Web Inspector: CPU Usage Timeline
https://bugs.webkit.org/show_bug.cgi?id=193730
<rdar://problem/46797201>

Patch by Joseph Pecoraro <pecoraro@apple.com> on 2019-01-24
Reviewed by Devin Rousso.

Source/JavaScriptCore:

* CMakeLists.txt:
* DerivedSources-input.xcfilelist:
* DerivedSources.make:
New files.

* inspector/protocol/CPUProfiler.json: Added.
New domain that follows the pattern of Memory/ScriptProfiler.

* inspector/protocol/Timeline.json:
New enum to auto-start a CPU instrument in the backend.

Source/WebCore:

Test: inspector/cpu-profiler/tracking.html

* Sources.txt:
* UnifiedSources-input.xcfilelist:
* WebCore.xcodeproj/project.pbxproj:
New files.

* inspector/InspectorController.cpp:
(WebCore::InspectorController::createLazyAgents):
* inspector/InstrumentingAgents.cpp:
(WebCore::InstrumentingAgents::reset):
* inspector/InstrumentingAgents.h:
(WebCore::InstrumentingAgents::inspectorCPUProfilerAgent const):
(WebCore::InstrumentingAgents::setInspectorCPUProfilerAgent):
Create and track the CPUProfilerAgent.

* inspector/agents/InspectorTimelineAgent.cpp:
(WebCore::InspectorTimelineAgent::toggleInstruments):
(WebCore::InspectorTimelineAgent::toggleCPUInstrument):
Handle backend auto-start of the CPU instrument / timeline.

* inspector/agents/InspectorCPUProfilerAgent.h:
* inspector/agents/InspectorCPUProfilerAgent.cpp: Added.
(WebCore::InspectorCPUProfilerAgent::InspectorCPUProfilerAgent):
(WebCore::InspectorCPUProfilerAgent::didCreateFrontendAndBackend):
(WebCore::InspectorCPUProfilerAgent::willDestroyFrontendAndBackend):
(WebCore::InspectorCPUProfilerAgent::startTracking):
(WebCore::InspectorCPUProfilerAgent::stopTracking):
(WebCore::InspectorCPUProfilerAgent::collectSample):
CPUProfilerAgent uses the ResourceUsageThread to get CPU data.

* inspector/agents/InspectorTimelineAgent.h:
* inspector/agents/InspectorMemoryAgent.cpp:
(WebCore::InspectorMemoryAgent::startTracking):
(WebCore::InspectorMemoryAgent::collectSample):
Update the MemoryAgent to collect only Memory data and use a more accurate sample timestamp.

* page/ResourceUsageData.h:
* page/ResourceUsageThread.cpp:
(WebCore::ResourceUsageThread::addObserver):
(WebCore::ResourceUsageThread::removeObserver):
(WebCore::ResourceUsageThread::notifyObservers):
(WebCore::ResourceUsageThread::recomputeCollectionMode):
(WebCore::ResourceUsageThread::threadBody):
* page/ResourceUsageThread.h:
* page/cocoa/ResourceUsageOverlayCocoa.mm:
(WebCore::ResourceUsageOverlay::platformInitialize):
* page/cocoa/ResourceUsageThreadCocoa.mm:
(WebCore::ResourceUsageThread::platformCollectCPUData):
(WebCore::ResourceUsageThread::platformCollectMemoryData):
(WebCore::ResourceUsageThread::platformThreadBody): Deleted.
* page/linux/ResourceUsageOverlayLinux.cpp:
(WebCore::ResourceUsageOverlay::platformInitialize):
* page/linux/ResourceUsageThreadLinux.cpp:
(WebCore::ResourceUsageThread::platformCollectCPUData):
(WebCore::ResourceUsageThread::platformCollectMemoryData):
(WebCore::ResourceUsageThread::platformThreadBody):
Give each observer their own collection mode. The ResourceUsageThread
will then collect data that is the union of all of the active observers.
This allows collecting CPU and Memory data separately, reducing the cost
of each when gathered individually.

Source/WebInspectorUI:

CPU Usage is gathered in the backend twice a second, the frequency of the
ResourceUsageThread in WebCore. The frontend displays cpu usage in a few
ways in the Timeline.

We use a column chart in the timeline overview to display the frequency and
relative distance of samples. This helps show if the samples were close
together or far apart, which indicates how meaningful they will be at a
particular scale.

We use a line chart in the timeline detail view which will be easier to see
the changes over a particular time range selection.

* Localizations/en.lproj/localizedStrings.js:
New strings.

* UserInterface/Main.html:
* UserInterface/Base/Main.js:
(WI.loaded):
* UserInterface/Test.html:
* UserInterface/Test/Test.js:
(WI.loaded):
* UserInterface/Protocol/CPUProfilerObserver.js:
(WI.CPUProfilerObserver.prototype.trackingStart):
(WI.CPUProfilerObserver.prototype.trackingUpdate):
(WI.CPUProfilerObserver.prototype.trackingComplete):
(WI.CPUProfilerObserver):
New files and default registration.

* UserInterface/Protocol/Target.js:
(WI.Target.prototype.get CPUProfilerAgent):
New Agent.

* UserInterface/Controllers/TimelineManager.js:
(WI.TimelineManager.availableTimelineTypes):
(WI.TimelineManager.prototype.cpuProfilerTrackingStarted):
(WI.TimelineManager.prototype.cpuProfilerTrackingUpdated):
(WI.TimelineManager.prototype.cpuProfilerTrackingCompleted):
(WI.TimelineManager.prototype._updateAutoCaptureInstruments):
(WI.TimelineManager.prototype.memoryTrackingStart): Renamed.
(WI.TimelineManager.prototype.memoryTrackingUpdate): Renamed.
(WI.TimelineManager.prototype.memoryTrackingComplete): Renamed.
* UserInterface/Models/CPUInstrument.js:
(WI.CPUInstrument):
(WI.CPUInstrument.supported):
(WI.CPUInstrument.prototype.get timelineRecordType):
(WI.CPUInstrument.prototype.startInstrumentation):
(WI.CPUInstrument.prototype.stopInstrumentation):
* UserInterface/Models/CPUTimelineRecord.js:
(WI.CPUTimelineRecord):
(WI.CPUTimelineRecord.prototype.get timestamp):
(WI.CPUTimelineRecord.prototype.get usage):
* UserInterface/Models/Instrument.js:
(WI.Instrument.createForTimelineType):
* UserInterface/Models/TimelineRecord.js:
* UserInterface/Models/TimelineRecording.js:
(WI.TimelineRecording.prototype.addRecord):
Expose a new CPU instrument and timeline.

* UserInterface/Views/ColumnChart.js: Added.
(WI.ColumnChart):
(WI.ColumnChart.prototype.get element):
(WI.ColumnChart.prototype.get bars):
(WI.ColumnChart.prototype.get size):
(WI.ColumnChart.prototype.set size):
(WI.ColumnChart.prototype.addBar):
(WI.ColumnChart.prototype.clear):
(WI.ColumnChart.prototype.needsLayout):
(WI.ColumnChart.prototype.updateLayout):
View that will draw vertical bars with independent widths.
This is meant to be used similiar to WI.LineChart.

* UserInterface/Images/CPUInstrument.svg: Added.
* UserInterface/Views/Variables.css:
(:root):
CPU timeline colors and icon.

* UserInterface/Views/CPUTimelineOverviewGraph.css:
(body .sidebar > .panel.navigation.timeline > .timelines-content li.item.cpu,):
(.timeline-overview-graph.cpu):
(.timeline-overview-graph.cpu > .legend):
(body[dir=ltr] .timeline-overview-graph.cpu > .legend):
(body[dir=rtl] .timeline-overview-graph.cpu > .legend):
(.timeline-overview-graph:nth-child(even) > .legend):
(body[dir=rtl] .timeline-overview-graph.cpu > .bar-chart):
(.timeline-overview-graph.cpu > .bar-chart > svg > g > rect):
* UserInterface/Views/CPUTimelineOverviewGraph.js: Added.
(WI.CPUTimelineOverviewGraph):
(WI.CPUTimelineOverviewGraph.prototype.get height):
(WI.CPUTimelineOverviewGraph.prototype.reset):
(WI.CPUTimelineOverviewGraph.prototype.layout.xScale):
(WI.CPUTimelineOverviewGraph.prototype.layout.yScale):
(WI.CPUTimelineOverviewGraph.prototype.layout.yScaleForRecord):
(WI.CPUTimelineOverviewGraph.prototype.layout):
(WI.CPUTimelineOverviewGraph.prototype._updateLegend):
(WI.CPUTimelineOverviewGraph.prototype._cpuTimelineRecordAdded):
* UserInterface/Views/CPUTimelineView.css:
(.timeline-view.cpu):
(.timeline-view.cpu > .content):
(.timeline-view.cpu > .content .subtitle):
(.timeline-view.cpu > .content > .details):
(.timeline-view.cpu > .content > .details > .timeline-ruler):
(body[dir=ltr] .timeline-view.cpu > .content > .details > .timeline-ruler):
(body[dir=rtl] .timeline-view.cpu > .content > .details > .timeline-ruler):
(.timeline-view.cpu > .content > .details > .subtitle):
(.cpu-usage-view .line-chart > svg > path):
(.timeline-view.cpu .legend > .row > .swatch.current):
* UserInterface/Views/CPUTimelineView.js: Added.
(WI.CPUTimelineView):
(WI.CPUTimelineView.prototype.shown):
(WI.CPUTimelineView.prototype.hidden):
(WI.CPUTimelineView.prototype.closed):
(WI.CPUTimelineView.prototype.reset):
(WI.CPUTimelineView.prototype.get scrollableElements):
(WI.CPUTimelineView.prototype.get showsFilterBar):
(WI.CPUTimelineView.prototype.layout.layoutView):
(WI.CPUTimelineView.prototype.layout.xScale):
(WI.CPUTimelineView.prototype.layout.yScale):
(WI.CPUTimelineView.prototype.layout):
(WI.CPUTimelineView.prototype._cpuTimelineRecordAdded):
* UserInterface/Views/CPUUsageView.css:
(.cpu-usage-view):
(.cpu-usage-view > .details):
(body[dir=ltr] .cpu-usage-view > .details):
(body[dir=rtl] .cpu-usage-view > .details):
(.cpu-usage-view > .graph):
(body[dir=rtl] .cpu-usage-view > .graph):
* UserInterface/Views/CPUUsageView.js:
(WI.CPUUsageView):
(WI.CPUUsageView.prototype.get element):
(WI.CPUUsageView.prototype.clear):
(WI.CPUUsageView.prototype.layoutWithDataPoints):
(WI.CPUUsageView.prototype._updateDetails):
* UserInterface/Views/ContentView.js:
(WI.ContentView.createFromRepresentedObject):
* UserInterface/Views/TimelineIcons.css:
(.cpu-icon .icon):
* UserInterface/Views/TimelineOverviewGraph.js:
(WI.TimelineOverviewGraph.createForTimeline):
* UserInterface/Views/TimelineTabContentView.js:
(WI.TimelineTabContentView.displayNameForTimelineType):
(WI.TimelineTabContentView.iconClassNameForTimelineType):
(WI.TimelineTabContentView.genericClassNameForTimelineType):
(WI.TimelineTabContentView.iconClassNameForRecord):
(WI.TimelineTabContentView.displayNameForRecord):
Timeline views for CPU usage.

* UserInterface/Views/MemoryCategoryView.js:
(WI.MemoryCategoryView):
* UserInterface/Views/MemoryTimelineView.js:
(WI.MemoryTimelineView.createChartContainer):
(WI.MemoryTimelineView):
(WI.MemoryTimelineView.prototype._clearMaxComparisonLegend):
Minor updates to style and comments.

LayoutTests:

* inspector/cpu-profiler/tracking-expected.txt: Added.
* inspector/cpu-profiler/tracking.html: Added.
Test the CPUProfiler domain emits events.

* inspector/heap/tracking-expected.txt:
* inspector/heap/tracking.html:
* inspector/memory/tracking-expected.txt:
* inspector/memory/tracking.html:
* inspector/script-profiler/tracking-expected.txt:
* inspector/script-profiler/tracking.html:
Update test naming.

* platform/win/TestExpectations:
Skip on platforms without RESOURCE_USAGE.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@240457 268f45cc-cd09-0410-ab3c-d52691b4dbfc

6 months ago[JSC] SharedArrayBufferConstructor and ArrayBufferConstructor should not have their...
ysuzuki@apple.com [Fri, 25 Jan 2019 00:49:44 +0000 (00:49 +0000)]
[JSC] SharedArrayBufferConstructor and ArrayBufferConstructor should not have their own IsoSubspace
https://bugs.webkit.org/show_bug.cgi?id=193774

Reviewed by Mark Lam.

We put all the instances of InternalFunction and its subclasses in IsoSubspace to make safer from UAF.
But since IsoSubspace requires the memory layout of instances is the same, we created different IsoSubspace
for subclasses of InternalFunction if sizeof(subclass) != sizeof(InternalFunction). One example is
ArrayBufferConstructor and SharedArrayBufferConstructor. But it is too costly to allocate 16KB page just
for these two constructor instances. They are only two instances per JSGlobalObject.

This patch makes sizeof(ArrayBufferConstructor) == sizeof(InternalFunction) so that they can use IsoSubspace
of InternalFunction. We introduce JSGenericArrayBufferConstructor, and it takes ArrayBufferSharingMode as
its template parameter. We define JSArrayBufferConstructor as JSGenericArrayBufferConstructor<ArrayBufferSharingMode::Default>
and JSSharedArrayBufferConstructor as JSGenericArrayBufferConstructor<ArrayBufferSharingMode::Shared> so that
we do not need to hold ArrayBufferSharingMode in the field of the constructor. This change removes IsoSubspace
for ArrayBufferConstructors, and reduces the memory usage.

* runtime/JSArrayBufferConstructor.cpp:
(JSC::JSGenericArrayBufferConstructor<sharingMode>::JSGenericArrayBufferConstructor):
(JSC::JSGenericArrayBufferConstructor<sharingMode>::finishCreation):
(JSC::JSGenericArrayBufferConstructor<sharingMode>::constructArrayBuffer):
(JSC::JSGenericArrayBufferConstructor<sharingMode>::createStructure):
(JSC::JSGenericArrayBufferConstructor<sharingMode>::info):
(JSC::JSArrayBufferConstructor::JSArrayBufferConstructor): Deleted.
(JSC::JSArrayBufferConstructor::finishCreation): Deleted.
(JSC::JSArrayBufferConstructor::create): Deleted.
(JSC::JSArrayBufferConstructor::createStructure): Deleted.
(JSC::constructArrayBuffer): Deleted.
* runtime/JSArrayBufferConstructor.h:
* runtime/JSGlobalObject.cpp:
(JSC::JSGlobalObject::init):
* runtime/JSGlobalObject.h:
* runtime/VM.cpp:
(JSC::VM::VM):
* runtime/VM.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@240456 268f45cc-cd09-0410-ab3c-d52691b4dbfc

6 months ago[webkitdirs] iosSimulatorApplicationsPath may not have iOS.simruntime in its path
jbedard@apple.com [Fri, 25 Jan 2019 00:35:19 +0000 (00:35 +0000)]
[webkitdirs] iosSimulatorApplicationsPath may not have iOS.simruntime in its path
https://bugs.webkit.org/show_bug.cgi?id=193787
<rdar://problem/47497634>

Reviewed by Alexey Proskuryakov.

* Scripts/webkitdirs.pm:
(iosSimulatorApplicationsPath): Pick from available runtime directories.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@240455 268f45cc-cd09-0410-ab3c-d52691b4dbfc

6 months agoImplement PerformanceObserver.supportedEntryTypes
commit-queue@webkit.org [Fri, 25 Jan 2019 00:04:52 +0000 (00:04 +0000)]
Implement PerformanceObserver.supportedEntryTypes
https://bugs.webkit.org/show_bug.cgi?id=193428

LayoutTests/imported/w3c:

Patch by Charles Vazac <cvazac@akamai.com> on 2019-01-24
Reviewed by Joseph Pecoraro.

* web-platform-tests/resource-timing/supported_resource_type.any-expected.txt:
* web-platform-tests/resource-timing/supported_resource_type.any.worker-expected.txt:

Source/WebCore:

PerformanceObserver.supportedEntryTypes should return an array of
entryTypes that can be observed per specification
https://w3c.github.io/performance-timeline/#supportedentrytypes-attribute

Patch by Charles Vazac <cvazac@akamai.com> on 2019-01-24
Reviewed by Joseph Pecoraro.

This is covered by web-platform-tests
LayoutTests/imported/w3c/web-platform-tests/resource-timing/supported_resource_type.*.html.

* page/PerformanceObserver.cpp:
(WebCore::PerformanceObserver::supportedEntryTypes):
* page/PerformanceObserver.h:
* page/PerformanceObserver.idl:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@240454 268f45cc-cd09-0410-ab3c-d52691b4dbfc

6 months agoUnreviewed, rolling out r240446.
tsavell@apple.com [Thu, 24 Jan 2019 23:41:30 +0000 (23:41 +0000)]
Unreviewed, rolling out r240446.

Casued 5 API failures

Reverted changeset:

"Activate the WebResourceLoadStatisticsStore in the
NetworkProcess and deactivate it in the UIProcess."
https://bugs.webkit.org/show_bug.cgi?id=193297
https://trac.webkit.org/changeset/240446

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@240453 268f45cc-cd09-0410-ab3c-d52691b4dbfc

6 months ago[iOS] Unable to make a selection in jsfiddle.net using arrow keys when requesting...
wenson_hsieh@apple.com [Thu, 24 Jan 2019 23:30:37 +0000 (23:30 +0000)]
[iOS] Unable to make a selection in jsfiddle.net using arrow keys when requesting desktop site
https://bugs.webkit.org/show_bug.cgi?id=193758
<rdar://problem/43614978>

Reviewed by Tim Horton.

Source/WebCore:

CodeMirror's script adds a repeating timer that periodically normalizes the logical selection in the editor
(this is distinct from the actual DOM selection, which is inside a hidden textarea element). This script defines
a helper method to select all the text inside of a text form control, called `selectInput`, which normally
invokes `node.select()`. However, in the case of iOS, the script works around broken `select()` behavior by
setting `selectionStart` and `selectionEnd` to encompass all the content in the form control. When requesting
the desktop version of the site, CodeMirror no longer attempts to apply its iOS workaround.

This iOS-specific behavior was introduced to fix <rdar://problem/4901923>. However, the original bug no longer
reproduces even without this quirk. To fix CodeMirror, we make two adjustments:

1.  Roll out this ancient demo hack, in favor of standardized behavior.
2.  Note that `select()` is also used when focusing an input. However, when focusing an input element on iOS, we
    want to match the platform (i.e. UITextField behavior) and move focus to the end of the text field. To
    achieve this, we introduce a new helper on HTMLInputElement that is called when setting the default
    selection of a text input after focus; on iOS, this helper method moves the selection to the end of the
    input, but everywhere else, it selects all the text in the input element.

This causes 6 existing layout tests to begin passing on iOS.

* html/HTMLInputElement.cpp:
(WebCore::HTMLInputElement::updateFocusAppearance):
(WebCore::HTMLInputElement::setDefaultSelectionAfterFocus):
* html/HTMLInputElement.h:
* html/HTMLTextFormControlElement.cpp:
(WebCore::HTMLTextFormControlElement::select):

LayoutTests:

Mark some existing layout tests as passing on iOS. Additionally, remove failing expectations for another
existing layout test on iOS.

* platform/ios/TestExpectations:
* platform/ios/editing/text-iterator/hidden-textarea-selection-quirk-expected.txt: Removed.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@240452 268f45cc-cd09-0410-ab3c-d52691b4dbfc

6 months agoFix leak of AVSampleBufferRenderSynchronizer boundaryObserver object.
jer.noble@apple.com [Thu, 24 Jan 2019 23:15:56 +0000 (23:15 +0000)]
Fix leak of AVSampleBufferRenderSynchronizer boundaryObserver object.
https://bugs.webkit.org/show_bug.cgi?id=193778

Reviewed by Jon Lee.

* platform/graphics/avfoundation/objc/MediaPlayerPrivateMediaSourceAVFObjC.mm:
(WebCore::MediaPlayerPrivateMediaSourceAVFObjC::performTaskAtMediaTime):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@240451 268f45cc-cd09-0410-ab3c-d52691b4dbfc

6 months agoDidFirstVisuallyNonEmptyLayout milestone should always fire at some point.
zalan@apple.com [Thu, 24 Jan 2019 22:58:09 +0000 (22:58 +0000)]
DidFirstVisuallyNonEmptyLayout milestone should always fire at some point.
https://bugs.webkit.org/show_bug.cgi?id=193741
<rdar://problem/47135030>

Reviewed by Antti Koivisto and Simon Fraser.

Source/WebCore:

fireLayoutRelatedMilestonesIfNeeded() is part of the post-layout tasks. In certain cases when
    1. the received data is not "contentful" yet
    2. and we are expecting some more (loading is not complete yet)
    3. but no layout is initiated anymore
nothing triggers the milestone firing.

This patch ensures that we fire the DidFirstVisuallyNonEmptyLayout when the frame load is complete unless we already did.

* page/FrameView.cpp:
(WebCore::FrameView::FrameView):
(WebCore::FrameView::loadProgressingStatusChanged):

Tools:

* TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj:
* TestWebKitAPI/Tests/WebKit/LayoutMilestonesWithAllContentInFrame.cpp:
(TestWebKitAPI::TEST):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@240450 268f45cc-cd09-0410-ab3c-d52691b4dbfc

6 months agostress/const-semantics.js fails a dfg-eager / ftl-eager run with an ASAN release...
ysuzuki@apple.com [Thu, 24 Jan 2019 22:37:01 +0000 (22:37 +0000)]
stress/const-semantics.js fails a dfg-eager / ftl-eager run with an ASAN release build.
https://bugs.webkit.org/show_bug.cgi?id=190693

Reviewed by Michael Saboff.

JSTests:

* stress/regress-190693.js: Added.
(truth):
(assert):
(shouldThrowInvalidConstAssignment):
(taz):

Source/JavaScriptCore:

JITStubRoutine's fields are marked only when JITStubRoutine::m_mayBeExecuting is true.
This becomes true when we find the executable address in our conservative roots, which
means that we could be executing it right now. This means that object liveness in
JITStubRoutine depends on the information gathered in ConservativeRoots. However, our
constraints are separated, "Conservative Scan" and "JIT Stub Routines". They can even
be executed concurrently, so that "JIT Stub Routines" may miss to mark the actually
executing JITStubRoutine because "Conservative Scan" finds it later.
When finalizing the GC, we delete the dead JITStubRoutines. At that time, since
"Conservative Scan" already finishes, we do not delete some JITStubRoutines which do not
mark the depending objects. Then, in the next cycle, we find JITStubRoutines still live,
attempt to mark the depending objects, and encounter the dead objects which are collected
in the previous cycles.

This patch removes "JIT Stub Routines" and merge it to "Conservative Scan". Since
"Conservative Scan" and "JIT Stub Routines" need to be executed only when the execution
happens (ensured by GreyedByExecution and CollectionPhase check), this change is OK for
GC stop time.

* heap/ConservativeRoots.h:
(JSC::ConservativeRoots::roots const):
(JSC::ConservativeRoots::roots): Deleted.
* heap/Heap.cpp:
(JSC::Heap::addCoreConstraints):
* heap/SlotVisitor.cpp:
(JSC::SlotVisitor::append):
* heap/SlotVisitor.h:
* jit/GCAwareJITStubRoutine.cpp:
(JSC::GCAwareJITStubRoutine::GCAwareJITStubRoutine):
* jit/GCAwareJITStubRoutine.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@240449 268f45cc-cd09-0410-ab3c-d52691b4dbfc

6 months agoUpdate ARM64EHash
sbarati@apple.com [Thu, 24 Jan 2019 22:07:33 +0000 (22:07 +0000)]
Update ARM64EHash
https://bugs.webkit.org/show_bug.cgi?id=193776
<rdar://problem/47526457>

Reviewed by Mark Lam.

See radar for details.

* assembler/AssemblerBuffer.h:
(JSC::ARM64EHash::update):
(JSC::ARM64EHash::finalHash const):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@240448 268f45cc-cd09-0410-ab3c-d52691b4dbfc

6 months agoObject Allocation Sinking phase can move a node that walks the stack into a place...
sbarati@apple.com [Thu, 24 Jan 2019 21:30:55 +0000 (21:30 +0000)]
Object Allocation Sinking phase can move a node that walks the stack into a place where the InlineCallFrame is no longer valid
https://bugs.webkit.org/show_bug.cgi?id=193751
<rdar://problem/47280215>

Reviewed by Michael Saboff.

JSTests:

* stress/object-allocation-sinking-phase-must-only-move-allocations-if-stack-trace-is-still-valid.js: Added.
(let.thing):
(foo.let.hello):
(foo):

Source/JavaScriptCore:

The Object Allocation Sinking phase may move allocations around inside
of the program. However, it was not ensuring that it's still possible
to walk the stack at the point in the program that it moved the allocation to.
Certain InlineCallFrames rely on data in the stack when taking a stack trace.
All allocation sites can do a stack walk (we do a stack walk when we GC).
Conservatively, this patch says we're ok to move this allocation if we are
moving within the same InlineCallFrame. We could be more precise and do an
analysis of stack writes. However, this scenario is so rare that we just
take the conservative-and-straight-forward approach of checking that the place
we're moving to is the same InlineCallFrame as the allocation site.

In general, this issue arises anytime we do any kind of code motion.
Interestingly, LICM gets this right. It gets it right because the only
InlineCallFrames we can't move out of are the InlineCallFrames that
have metadata stored on the stack (callee for closure calls and argument
count for varargs calls). LICM doesn't have this issue because it relies
on Clobberize for doing its effects analysis. In clobberize, we model every
node within an InlineCallFrame that meets the above criteria as reading
from those stack fields. Consequently, LICM won't hoist any node in that
InlineCallFrame past the beginning of the InlineCallFrame since the IR
we generate to set up such an InlineCallFrame contains writes to that
stack location.

* dfg/DFGObjectAllocationSinkingPhase.cpp:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@240447 268f45cc-cd09-0410-ab3c-d52691b4dbfc

6 months agoActivate the WebResourceLoadStatisticsStore in the NetworkProcess and deactivate...
bfulgham@apple.com [Thu, 24 Jan 2019 21:09:37 +0000 (21:09 +0000)]
Activate the WebResourceLoadStatisticsStore in the NetworkProcess and deactivate it in the UIProcess.
https://bugs.webkit.org/show_bug.cgi?id=193297
<rdar://problem/47158841>

Reviewed by Alex Christensen.

Source/WebCore:

Trigger logging to the UIProcess when the ResourceLoadObserver is used in the NetworkProcess.

* Modules/websockets/WebSocket.cpp:
(WebCore::WebSocket::connect): Notify NetworkProcess a connection was made to a resource.
* loader/ResourceLoadObserver.cpp:
(WebCore::ResourceLoadObserver::setLogWebSocketLoadingNotificationCallback): Added.
(WebCore::ResourceLoadObserver::setLogSubresourceLoadingNotificationCallback): Added.
(WebCore::ResourceLoadObserver::setLogSubresourceRedirectNotificationCallback): Added.
(WebCore::ResourceLoadObserver::logSubresourceLoading): Notify NetworkProcess of the load.
(WebCore::ResourceLoadObserver::logWebSocketLoading): Ditto.
(WebCore::ResourceLoadObserver::logUserInteractionWithReducedTimeResolution): Ditto.

Source/WebKit:

This patch activates the ResourceLoadStatistics code in the NetworkProcess, and turns
it off in the UIProcess. It also updates test infrastructure to work with this change
in architecture.

* NetworkProcess/Classifier/ResourceLoadStatisticsMemoryStore.cpp:
(WebKit::ResourceLoadStatisticsMemoryStore::logSubresourceLoading): Added.
(WebKit::ResourceLoadStatisticsMemoryStore::logSubresourceRedirect): Added.
(WebKit::ResourceLoadStatisticsMemoryStore::logWebSocketLoading): Added.
* NetworkProcess/Classifier/WebResourceLoadStatisticsStore.cpp:
(WebKit::WebResourceLoadStatisticsStore::logSubresourceLoading): Added.
(WebKit::WebResourceLoadStatisticsStore::logSubresourceRedirect): Added.
(WebKit::WebResourceLoadStatisticsStore::logWebSocketLoading): Added.
(WebKit::WebResourceLoadStatisticsStore::removeAllStorageAccess):
(WebKit::WebResourceLoadStatisticsStore::setCacheMaxAgeCap):
(WebKit::WebResourceLoadStatisticsStore::setCacheMaxAgeCapForPrevalentResources): Deleted.
* NetworkProcess/Classifier/WebResourceLoadStatisticsStore.h:
* NetworkProcess/NetworkConnectionToWebProcess.cpp:
(WebKit::NetworkConnectionToWebProcess::logSubresourceLoading): Added.
(WebKit::NetworkConnectionToWebProcess::logSubresourceRedirect): Added.
(WebKit::NetworkConnectionToWebProcess::logWebSocketLoading): Added.
* NetworkProcess/NetworkConnectionToWebProcess.messages.in:
* NetworkProcess/NetworkSession.cpp:
(WebKit::NetworkSession::setResourceLoadStatisticsEnabled):
* UIProcess/WebProcessPool.cpp:
(WebKit::WebProcessPool::ensureNetworkProcess):
* UIProcess/WebsiteData/Cocoa/WebsiteDataStoreCocoa.mm:
(WebKit::WebsiteDataStore::parameters):
* UIProcess/WebsiteData/WebsiteDataStore.cpp:
(WebKit::WebsiteDataStore::removeData):
(WebKit::WebsiteDataStore::setResourceLoadStatisticsEnabled):
(WebKit::WebsiteDataStore::enableResourceLoadStatisticsAndSetTestingCallback):
* Webprocess/WebProcess.cpp:
(WebKit::WebProcess::initializeWebProcess): Register new ResourceLoadObserver callbacks.

Tools:

* WebKitTestRunner/InjectedBundle/TestRunner.cpp:
(WTR::TestRunner::installStatisticsDidScanDataRecordsCallback): Simplify test configuration by
activating the message used to trigger the callback when it is set.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@240446 268f45cc-cd09-0410-ab3c-d52691b4dbfc

6 months agoAdd Ad Click Attribution as an internal/experimental feature
wilander@apple.com [Thu, 24 Jan 2019 19:17:09 +0000 (19:17 +0000)]
Add Ad Click Attribution as an internal/experimental feature
https://bugs.webkit.org/show_bug.cgi?id=193685
<rdar://problem/47450399>

Reviewed by Brent Fulgham.

Source/WebCore:

Test: http/tests/adClickAttribution/anchor-tag-attributes-reflect.html

* html/HTMLAnchorElement.h:
* html/HTMLAnchorElement.idl:
* html/HTMLAttributeNames.in:
    Addeed two new experimental attributes:
    - adcampaignid: Ad campaign ID.
    - addestination: Ad link destination site.
* page/RuntimeEnabledFeatures.h:
(WebCore::RuntimeEnabledFeatures::adClickAttributionEnabled const):
(WebCore::RuntimeEnabledFeatures::setAdClickAttributionEnabled):
* page/Settings.yaml:

Source/WebKit:

* Shared/WebPreferences.yaml:
    Added AdClickAttributionEnabled as a runtime enabled feature, off by default.

Source/WebKitLegacy/mac:

* WebView/WebPreferenceKeysPrivate.h:
* WebView/WebPreferences.mm:
(+[WebPreferences initialize]):
(-[WebPreferences adClickAttributionEnabled]):
(-[WebPreferences setAdClickAttributionEnabled:]):
    Added support for experimental ad click attribution.
* WebView/WebPreferencesPrivate.h:
* WebView/WebView.mm:
(-[WebView _preferencesChanged:]):

Tools:

Added test option adClickAttributionEnabled.

* DumpRenderTree/TestOptions.cpp:
(TestOptions::TestOptions):
* DumpRenderTree/TestOptions.h:
* DumpRenderTree/mac/DumpRenderTree.mm:
(setWebPreferencesForTestOptions):

LayoutTests:

* http/tests/adClickAttribution/anchor-tag-attributes-reflect-expected.txt: Added.
* http/tests/adClickAttribution/anchor-tag-attributes-reflect.html: Added.
* platform/win/TestExpectations:
    Skipped on Windows for now.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@240444 268f45cc-cd09-0410-ab3c-d52691b4dbfc

6 months ago[PSON] Flash on back navigation on Mac
antti@apple.com [Thu, 24 Jan 2019 18:35:05 +0000 (18:35 +0000)]
[PSON] Flash on back navigation on Mac
https://bugs.webkit.org/show_bug.cgi?id=193716
<rdar://problem/47148458>

Reviewed by Chris Dumez.

Source/WebKit:

We close the page immediately if we fail to suspend. Layers disappear and we get a flash.

* UIProcess/SuspendedPageProxy.cpp:
(WebKit::SuspendedPageProxy::~SuspendedPageProxy):
(WebKit::SuspendedPageProxy::close):

Track closed state so we don't send the message twice, causing unhandled message errors in web process.

(WebKit::SuspendedPageProxy::didProcessRequestToSuspend):

Close the suspended page if the suspension fails.
Skip this if we are using web process side compositing on Mac.

* UIProcess/SuspendedPageProxy.h:
* UIProcess/WebPageProxy.cpp:
(WebKit::WebPageProxy::enterAcceleratedCompositingMode):

On Mac, close the failed SuspendedPageProxy when entering compositing mode. At this point we don't need it to keep layers alive.

* UIProcess/WebProcessPool.cpp:
(WebKit::WebProcessPool::closeFailedSuspendedPagesForPage):
* UIProcess/WebProcessPool.h:
* WebProcess/WebPage/WebPage.cpp:
(WebKit::WebPage::suspendForProcessSwap):

Don't close the page on suspension failure. This is now managed by the UI process.

Tools:

* TestWebKitAPI/Tests/WebKitCocoa/ProcessSwapOnNavigation.mm:

Closing of the previous page is delayed so waiting for didFinishNavigation is
not sufficient to guarantee we have received all the messages. Wait for them.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@240443 268f45cc-cd09-0410-ab3c-d52691b4dbfc

6 months agoRegression(PSON) Back/Forward list items' URL sometimes gets replaced with the URL...
cdumez@apple.com [Thu, 24 Jan 2019 18:30:35 +0000 (18:30 +0000)]
Regression(PSON) Back/Forward list items' URL sometimes gets replaced with the URL of a subframe
https://bugs.webkit.org/show_bug.cgi?id=193761
<rdar://problem/47456405>

Reviewed by Alex Christensen.

Source/WebKit:

When doing a history navigation cross-process, the UIProcess would first send the back/forward list items
to the destination WebProcess via WebPage::updateBackForwardListForReattach(), then ask the process to
navigate to the expected back/forward list item.

WebPage::updateBackForwardListForReattach() would call restoreSessionInternal(), which would call
toHistoryItem() on each BackForwardListItem. This may end up creating more than one HistoryItem for each
BackForwardListItem because HistoryItems are part of a tree (each frame has its own list of history items).

Note that BackForwardListItems and HistoryItem share a BackForwardItemIdentifier which is a
(processIdentifier, itemIdentifier) pair. We normally generate the HistoryItem's identifier from inside
its constructor like so:
`{ Process::identifier(), generateObjectIdentifier<BackForwardItemIdentifier::ItemIdentifierType>() }`

However, when calling updateBackForwardListForReattach() and constructing children HistoryItem,
applyFrameState() would generate the identifier by itself and passing it to the HistoryItem constructor.
Its genegates the ID the exact same way so this would in theory not be a problem. Unfortunately, both
calls to generateObjectIdentifier() get inlined and both call sites end up with their own static counter
to generate ids. As a result, we may end up with conflicts as HistoryItems for child frames (restored
by restoreSessionInternal()) can end up with the same identifier as HistoryItems for top frames.

This confusion would lead to the WebContent process sending the UIProcess bad information and the URL
of subframes could end up as the WebBackForwardListItem's mainframe URL.

* WebProcess/WebCoreSupport/SessionStateConversion.cpp:
(WebKit::applyFrameState):
Stop calling generateObjectIdentifier() explicitly and let the HistoryItem constructor take care of it.
Calling generateObjectIdentifier() for the same type from different places is not safe due to inlining.

Tools:

Add API test coverage.

* TestWebKitAPI/Tests/WebKitCocoa/ProcessSwapOnNavigation.mm:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@240442 268f45cc-cd09-0410-ab3c-d52691b4dbfc

6 months agoUse MonotonicTime in WorkerRunLoop
youenn@apple.com [Thu, 24 Jan 2019 18:30:24 +0000 (18:30 +0000)]
Use MonotonicTime in WorkerRunLoop
https://bugs.webkit.org/show_bug.cgi?id=193417

Reviewed by Saam Barati.

Source/WebCore:

Condition is based on MonotonicTime so MessageQueue should also be based on MonotonicTime.
Ditto for WorkerRunLoop.
No easy way to test the change which should not be easily observable.

* workers/WorkerRunLoop.cpp:
(WebCore::WorkerRunLoop::runInMode):

Source/WTF:

* wtf/MessageQueue.h:
(WTF::MessageQueue<DataType>::waitForMessage):
(WTF::MessageQueue<DataType>::waitForMessageFilteredWithTimeout):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@240441 268f45cc-cd09-0410-ab3c-d52691b4dbfc

6 months agowebkitpy: Missing PID in crashlog name should not be fatal
jbedard@apple.com [Thu, 24 Jan 2019 18:26:05 +0000 (18:26 +0000)]
webkitpy: Missing PID in crashlog name should not be fatal
https://bugs.webkit.org/show_bug.cgi?id=193771

If we can't determine the PID of a crashlog, just ignore the crashlog instead
of raising an exception.

Reviewed by Lucas Forschler.

* Scripts/webkitpy/port/darwin.py:
(DarwinPort._merge_crash_logs):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@240440 268f45cc-cd09-0410-ab3c-d52691b4dbfc

6 months agotest_exporter.py: webkitpy.common.system.executive.ScriptError from create_branch_wit...
mcatanzaro@igalia.com [Thu, 24 Jan 2019 17:30:10 +0000 (17:30 +0000)]
test_exporter.py: webkitpy.common.system.executive.ScriptError from create_branch_with_patch
https://bugs.webkit.org/show_bug.cgi?id=193746

Reviewed by Youenn Fablet.

It calls 'git apply --index --abort', but --abort is not a valid argument to 'git apply',
so we'd best just not call it.

* Scripts/webkitpy/w3c/test_exporter.py:
(WebPlatformTestExporter.create_branch_with_patch):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@240438 268f45cc-cd09-0410-ab3c-d52691b4dbfc

6 months agoMove FileSystem to WTF
ross.kirsling@sony.com [Thu, 24 Jan 2019 17:25:57 +0000 (17:25 +0000)]
Move FileSystem to WTF
https://bugs.webkit.org/show_bug.cgi?id=193602

Reviewed by Yusuke Suzuki.

Source/WebCore:

* Modules/encryptedmedia/CDM.cpp:
* Modules/encryptedmedia/legacy/WebKitMediaKeySession.cpp:
* Modules/entriesapi/DOMFileSystem.cpp:
* Modules/entriesapi/FileSystemEntry.cpp:
* Modules/indexeddb/IDBDatabaseIdentifier.cpp:
* Modules/indexeddb/server/SQLiteIDBBackingStore.cpp:
* Modules/indexeddb/server/SQLiteIDBTransaction.cpp:
* Modules/indexeddb/shared/InProcessIDBServer.cpp:
* Modules/webdatabase/DatabaseTracker.cpp:
* Modules/webdatabase/OriginLock.cpp:
* Modules/webdatabase/OriginLock.h:
* Modules/webdatabase/cocoa/DatabaseManagerCocoa.mm:
* PlatformMac.cmake:
* PlatformPlayStation.cmake:
* PlatformWin.cmake:
* Sources.txt:
* SourcesCocoa.txt:
* WebCore.xcodeproj/project.pbxproj:
* bindings/js/GCController.cpp:
* dom/DataTransferItem.cpp:
* editing/cocoa/WebContentReaderCocoa.mm:
* fileapi/File.cpp:
* fileapi/FileCocoa.mm:
* html/FileInputType.cpp:
* html/FileListCreator.cpp:
* loader/appcache/ApplicationCacheHost.cpp:
* loader/appcache/ApplicationCacheStorage.cpp:
* page/Page.cpp:
* page/SecurityOrigin.cpp:
* page/SecurityOriginData.cpp:
* platform/FileHandle.h:
* platform/FileStream.cpp:
* platform/FileStream.h:
* platform/SharedBuffer.h:
* platform/SourcesGLib.txt:
* platform/cocoa/FileMonitorCocoa.mm:
* platform/glib/FileMonitorGLib.cpp:
* platform/glib/SharedBufferGlib.cpp:
* platform/graphics/avfoundation/objc/CDMSessionAVContentKeySession.mm:
* platform/graphics/avfoundation/objc/CDMSessionAVStreamSession.mm:
* platform/graphics/avfoundation/objc/CDMSessionMediaSourceAVFObjC.mm:
* platform/graphics/avfoundation/objc/MediaPlayerPrivateMediaSourceAVFObjC.mm:
* platform/graphics/gstreamer/MediaPlayerPrivateGStreamer.cpp:
* platform/ios/QuickLook.mm:
* platform/ios/WebItemProviderPasteboard.mm:
* platform/mediarecorder/cocoa/MediaRecorderPrivateWriterCocoa.mm:
* platform/network/BlobDataFileReference.cpp:
* platform/network/BlobRegistryImpl.cpp:
* platform/network/BlobResourceHandle.cpp:
* platform/network/FormData.cpp:
* platform/network/cf/FormDataStreamCFNet.cpp:
* platform/network/cocoa/ResourceRequestCocoa.mm:
* platform/network/curl/CookieJarDB.cpp:
* platform/network/curl/CurlCacheEntry.h:
* platform/network/curl/CurlCacheManager.cpp:
* platform/network/curl/CurlFormDataStream.h:
* platform/network/curl/CurlRequest.h:
* platform/network/curl/NetworkStorageSessionCurl.cpp:
* platform/network/curl/ResourceHandleCurl.cpp:
* platform/network/mac/BlobDataFileReferenceMac.mm:
* platform/network/soup/ResourceHandleSoup.cpp:
* platform/network/soup/SoupNetworkSession.cpp:
* platform/posix/SharedBufferPOSIX.cpp:
* platform/sql/SQLiteFileSystem.cpp:
* platform/text/hyphen/HyphenationLibHyphen.cpp:
* platform/win/SearchPopupMenuDB.cpp:
* rendering/RenderTheme.cpp:
* rendering/RenderThemeGtk.cpp:
* rendering/RenderThemeWin.cpp:
* workers/service/server/RegistrationDatabase.cpp:

Source/WebCore/PAL:

* PAL.xcodeproj/project.pbxproj:
* pal/PlatformMac.cmake:

Source/WebKit:

* NetworkProcess/Classifier/ResourceLoadStatisticsPersistentStorage.cpp:
* NetworkProcess/NetworkDataTaskBlob.h:
* NetworkProcess/NetworkProcess.cpp:
* NetworkProcess/cache/CacheStorageEngine.cpp:
* NetworkProcess/cache/CacheStorageEngineCaches.cpp:
* NetworkProcess/cache/NetworkCache.cpp:
* NetworkProcess/cache/NetworkCacheBlobStorage.cpp:
* NetworkProcess/cache/NetworkCacheData.cpp:
* NetworkProcess/cache/NetworkCacheFileSystem.cpp:
* NetworkProcess/cache/NetworkCacheFileSystem.h:
* NetworkProcess/cache/NetworkCacheIOChannelCocoa.mm:
* NetworkProcess/cache/NetworkCacheIOChannelSoup.cpp:
* NetworkProcess/cache/NetworkCacheStatistics.cpp:
* NetworkProcess/cache/NetworkCacheStorage.cpp:
* NetworkProcess/cocoa/NetworkDataTaskCocoa.mm:
* NetworkProcess/cocoa/NetworkSessionCocoa.mm:
* NetworkProcess/mac/NetworkProcessMac.mm:
* NetworkProcess/soup/NetworkProcessSoup.cpp:
* PluginProcess/unix/PluginProcessMainUnix.cpp:
* Shared/PersistencyUtils.cpp:
* Shared/Plugins/Netscape/unix/NetscapePluginModuleUnix.cpp:
* Shared/Plugins/unix/PluginSearchPath.cpp:
* Shared/WebMemorySampler.h:
* Shared/glib/ProcessExecutablePathGLib.cpp:
* Shared/ios/ChildProcessIOS.mm:
* Shared/mac/ChildProcessMac.mm:
* Shared/mac/SandboxExtensionMac.mm:
* UIProcess/API/APIContentRuleListStore.cpp:
* UIProcess/API/Cocoa/APIWebsiteDataStoreCocoa.mm:
* UIProcess/API/glib/APIWebsiteDataStoreGLib.cpp:
* UIProcess/API/glib/IconDatabase.cpp:
* UIProcess/API/glib/WebKitFaviconDatabase.cpp:
* UIProcess/API/glib/WebKitFileChooserRequest.cpp:
* UIProcess/API/glib/WebKitWebContext.cpp:
* UIProcess/API/glib/WebKitWebsiteDataManager.cpp:
* UIProcess/API/win/APIWebsiteDataStoreWin.cpp:
* UIProcess/Automation/WebAutomationSession.cpp:
* UIProcess/Cocoa/DownloadClient.mm:
* UIProcess/Cocoa/WebProcessPoolCocoa.mm:
* UIProcess/Cocoa/WebViewImpl.mm:
* UIProcess/DeviceIdHashSaltStorage.cpp:
* UIProcess/Downloads/DownloadProxy.cpp:
* UIProcess/Launcher/glib/BubblewrapLauncher.cpp:
* UIProcess/Launcher/glib/ProcessLauncherGLib.cpp:
* UIProcess/Plugins/gtk/PluginInfoCache.cpp:
* UIProcess/Plugins/mac/PluginProcessProxyMac.mm:
* UIProcess/Plugins/unix/PluginInfoStoreUnix.cpp:
* UIProcess/Plugins/unix/PluginProcessProxyUnix.cpp:
* UIProcess/ServiceWorkerProcessProxy.cpp:
* UIProcess/WebStorage/LocalStorageDatabase.cpp:
* UIProcess/WebStorage/LocalStorageDatabaseTracker.cpp:
* UIProcess/WebsiteData/Cocoa/WebsiteDataStoreCocoa.mm:
* UIProcess/WebsiteData/WebsiteDataStore.cpp:
* UIProcess/glib/WebProcessProxyGLib.cpp:
* UIProcess/gtk/WebInspectorProxyGtk.cpp:
* UIProcess/gtk/WebProcessPoolGtk.cpp:
* UIProcess/ios/WKContentViewInteraction.mm:
* UIProcess/ios/forms/WKFileUploadPanel.mm:
* UIProcess/mac/WebPageProxyMac.mm:
* UIProcess/win/WebProcessPoolWin.cpp:
* UIProcess/win/WebView.cpp:
* UIProcess/wpe/WebProcessPoolWPE.cpp:
* WebProcess/InjectedBundle/API/glib/WebKitExtensionManager.cpp:
* WebProcess/InjectedBundle/glib/InjectedBundleGlib.cpp:
* WebProcess/MediaCache/WebMediaKeyStorageManager.cpp:
* WebProcess/Plugins/Netscape/NetscapePluginStream.h:
* WebProcess/Plugins/PluginProcessConnection.cpp:
* WebProcess/WebCoreSupport/SessionStateConversion.cpp:
* WebProcess/WebPage/gtk/WebInspectorUIGtk.cpp:
* WebProcess/WebPage/win/WebInspectorUIWin.cpp:
* WebProcess/cocoa/WebProcessCocoa.mm:

Source/WebKitLegacy:

* Storage/StorageAreaSync.cpp:
* Storage/StorageSyncManager.cpp:
* Storage/StorageTracker.cpp:

Source/WebKitLegacy/mac:

* Misc/WebNSFileManagerExtras.mm:
* Storage/WebDatabaseProvider.mm:
* WebView/WebView.mm:

Source/WebKitLegacy/win:

* Plugins/PluginDatabase.cpp:
* Plugins/PluginPackage.h:
* Plugins/PluginStream.h:
* WebApplicationCache.cpp:
* WebDatabaseManager.cpp:
* WebDownloadCurl.cpp:
* WebPreferences.cpp:
* WebView.cpp:

Source/WTF:

* WTF.xcodeproj/project.pbxproj:
* wtf/CMakeLists.txt:
* wtf/FileMetadata.h: Renamed from Source/WebCore/platform/FileMetadata.h.
* wtf/FileSystem.cpp: Renamed from Source/WebCore/platform/FileSystem.cpp.
* wtf/FileSystem.h: Renamed from Source/WebCore/platform/FileSystem.h.
* wtf/PlatformGTK.cmake:
* wtf/PlatformMac.cmake:
* wtf/PlatformPlayStation.cmake:
* wtf/PlatformWPE.cmake:
* wtf/PlatformWin.cmake:
* wtf/cf/FileSystemCF.cpp: Renamed from Source/WebCore/platform/cf/FileSystemCF.cpp.
* wtf/cocoa/FileSystemCocoa.mm: Renamed from Source/WebCore/platform/cocoa/FileSystemCocoa.mm.
* wtf/glib/FileSystemGlib.cpp: Renamed from Source/WebCore/platform/glib/FileSystemGlib.cpp.
* wtf/mac/FileSystemMac.mm: Renamed from Source/WebCore/platform/mac/FileSystemMac.mm.
* wtf/posix/FileSystemPOSIX.cpp: Renamed from Source/WebCore/platform/posix/FileSystemPOSIX.cpp.
* wtf/spi/mac/MetadataSPI.h: Renamed from Source/WebCore/PAL/pal/spi/mac/MetadataSPI.h.
* wtf/win/FileSystemWin.cpp: Renamed from Source/WebCore/platform/win/FileSystemWin.cpp.
* wtf/win/PathWalker.cpp: Renamed from Source/WebCore/platform/win/PathWalker.cpp.
* wtf/win/PathWalker.h: Renamed from Source/WebCore/platform/win/PathWalker.h.

Tools:

* DumpRenderTree/win/DumpRenderTree.cpp:
* TestWebKitAPI/CMakeLists.txt:
* TestWebKitAPI/PlatformGTK.cmake:
* TestWebKitAPI/PlatformJSCOnly.cmake:
* TestWebKitAPI/PlatformPlayStation.cmake:
* TestWebKitAPI/PlatformWPE.cmake:
* TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj:
* TestWebKitAPI/Tests/WTF/FileSystem.cpp: Renamed from Tools/TestWebKitAPI/Tests/WebCore/FileSystem.cpp.
* TestWebKitAPI/Tests/WebCore/FileMonitor.cpp:
* TestWebKitAPI/Tests/WebCore/SecurityOrigin.cpp:
* TestWebKitAPI/Tests/WebCore/SharedBufferTest.cpp:
* TestWebKitAPI/Tests/WebCore/cocoa/DatabaseTrackerTest.mm:
* TestWebKitAPI/Tests/WebCore/curl/Cookies.cpp:
* TestWebKitAPI/Tests/WebKitCocoa/Download.mm:
* TestWebKitAPI/Tests/WebKitCocoa/DownloadProgress.mm:
* TestWebKitAPI/win/PlatformUtilitiesWin.cpp:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@240437 268f45cc-cd09-0410-ab3c-d52691b4dbfc

6 months ago[LFC][BFC][MarginCollapsing] MarginCollapse::collapsedVerticalValues should not retur...
zalan@apple.com [Thu, 24 Jan 2019 17:15:46 +0000 (17:15 +0000)]
[LFC][BFC][MarginCollapsing] MarginCollapse::collapsedVerticalValues should not return computed non-collapsed values.
https://bugs.webkit.org/show_bug.cgi?id=193768

Reviewed by Antti Koivisto.

When it comes to the actual used values it does not really matter, only from correctness point of view.
(This patch also moves some checks to their correct place.)

* layout/blockformatting/BlockMarginCollapse.cpp:
(WebCore::Layout::BlockFormattingContext::MarginCollapse::marginBeforeCollapsesWithPreviousSiblingMarginAfter):
(WebCore::Layout::BlockFormattingContext::MarginCollapse::marginBeforeCollapsesWithFirstInFlowChildMarginBefore):
(WebCore::Layout::BlockFormattingContext::MarginCollapse::marginAfterCollapsesWithLastInFlowChildMarginAfter):
(WebCore::Layout::BlockFormattingContext::MarginCollapse::positiveNegativeMarginBefore):
(WebCore::Layout::BlockFormattingContext::MarginCollapse::positiveNegativeMarginAfter):
(WebCore::Layout::BlockFormattingContext::MarginCollapse::collapsedVerticalValues):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@240436 268f45cc-cd09-0410-ab3c-d52691b4dbfc

6 months agoSource/WebCore:
simon.fraser@apple.com [Thu, 24 Jan 2019 16:10:43 +0000 (16:10 +0000)]
Source/WebCore:
Add "frame hosting" nodes to the scrolling tree
https://bugs.webkit.org/show_bug.cgi?id=193753

Reviewed by Antti Koivisto.

When the scrolling tree crosses frame boundaries, mutations in the parent frame currently
require the iframe's scrolling node to get reparented in a new ancestor, which requires
a layer tree walk of the parent frame. This is error-prone, and not very future-proof.

Fix this by introducing "frame hosting" scrolling tree nodes. These are mostly inert
nodes that are owned by the RenderIFrame's layer backing in the parent frame, and exist
to provide a consistent parent node for the subframe's scrolling node.

This patch adds the node types, but does not instantiate them yet.

* Sources.txt:
* WebCore.xcodeproj/project.pbxproj:
* page/scrolling/ScrollingCoordinator.cpp:
(WebCore::operator<<):
* page/scrolling/ScrollingCoordinator.h:
* page/scrolling/ScrollingStateFrameHostingNode.cpp: Added.
(WebCore::ScrollingStateFrameHostingNode::create):
(WebCore::ScrollingStateFrameHostingNode::ScrollingStateFrameHostingNode):
(WebCore::ScrollingStateFrameHostingNode::clone):
(WebCore::ScrollingStateFrameHostingNode::dumpProperties const):
* page/scrolling/ScrollingStateFrameHostingNode.h: Added.
* page/scrolling/ScrollingStateNode.h:
(WebCore::ScrollingStateNode::isFrameHostingNode const):
* page/scrolling/ScrollingStateTree.cpp:
(WebCore::ScrollingStateTree::createNode):
* page/scrolling/ScrollingTreeFrameHostingNode.cpp: Added.
(WebCore::ScrollingTreeFrameHostingNode::create):
(WebCore::ScrollingTreeFrameHostingNode::ScrollingTreeFrameHostingNode):
(WebCore::ScrollingTreeFrameHostingNode::commitStateBeforeChildren):
(WebCore::ScrollingTreeFrameHostingNode::updateLayersAfterAncestorChange):
(WebCore::ScrollingTreeFrameHostingNode::dumpProperties const):
* page/scrolling/ScrollingTreeFrameHostingNode.h: Added.
* page/scrolling/ScrollingTreeNode.h:
(WebCore::ScrollingTreeNode::isFrameHostingNode const):
* page/scrolling/ios/ScrollingTreeIOS.cpp:
(WebCore::ScrollingTreeIOS::createScrollingTreeNode):
* page/scrolling/mac/ScrollingTreeMac.cpp:
(ScrollingTreeMac::createScrollingTreeNode):
* rendering/RenderLayerBacking.cpp:
(WebCore::RenderLayerBacking::~RenderLayerBacking):
(WebCore::RenderLayerBacking::detachFromScrollingCoordinator):
(WebCore::operator<<):
* rendering/RenderLayerBacking.h:
* rendering/RenderLayerCompositor.cpp:
(WebCore::scrollCoordinationRoleForNodeType):
(WebCore::RenderLayerCompositor::detachScrollCoordinatedLayerWithRole):
(WebCore::RenderLayerCompositor::detachScrollCoordinatedLayer):
(WebCore::RenderLayerCompositor::updateScrollCoordinatedLayer):
* rendering/RenderLayerCompositor.h:

Source/WebKit:
Create "frame hosting" nodes for the scrolling tree
https://bugs.webkit.org/show_bug.cgi?id=193753

Reviewed by Antti Koivisto.

When the scrolling tree crosses frame boundaries, mutations in the parent frame currently
require the iframe's scrolling node to get reparented in a new ancestor, which requires
a layer tree walk of the parent frame. This is error-prone, and not very future-proof.

Fix this by introducing "frame hosting" scrolling tree nodes. These are mostly inert
nodes that are owned by the RenderIFrame's layer backing in the parent frame, and exist
to provide a consistent parent node for the subframe's scrolling node.

This patch adds the node types, but does not instantiate them yet.

* Shared/RemoteLayerTree/RemoteScrollingCoordinatorTransaction.cpp:
(ArgumentCoder<ScrollingStateFrameHostingNode>::encode):
(ArgumentCoder<ScrollingStateFrameHostingNode>::decode):
(WebKit::encodeNodeAndDescendants):
(WebKit::RemoteScrollingCoordinatorTransaction::decode):
(WebKit::dump):
* UIProcess/RemoteLayerTree/RemoteScrollingCoordinatorProxy.cpp:
(WebKit::RemoteScrollingCoordinatorProxy::connectStateNodeLayers):
* UIProcess/RemoteLayerTree/RemoteScrollingTree.cpp:
(WebKit::RemoteScrollingTree::createScrollingTreeNode):
* UIProcess/RemoteLayerTree/ios/RemoteScrollingCoordinatorProxyIOS.mm:
(WebKit::RemoteScrollingCoordinatorProxy::connectStateNodeLayers):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@240435 268f45cc-cd09-0410-ab3c-d52691b4dbfc

6 months ago[iOS] Enable media element volume on iPad
eric.carlson@apple.com [Thu, 24 Jan 2019 15:31:02 +0000 (15:31 +0000)]
[iOS] Enable media element volume on iPad
https://bugs.webkit.org/show_bug.cgi?id=193745
<rdar://problem/47452297>

Reviewed by Jer Noble.

* html/HTMLMediaElement.cpp:
(WebCore::HTMLMediaElement::setVolume):
(WebCore::HTMLMediaElement::mediaPlayerVolumeChanged):
(WebCore::HTMLMediaElement::updateVolume):

* platform/graphics/avfoundation/objc/MediaPlayerPrivateAVFoundationObjC.mm:
(WebCore::MediaPlayerPrivateAVFoundationObjC::setVolume):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@240434 268f45cc-cd09-0410-ab3c-d52691b4dbfc

6 months ago[WPE] Remove hard dependency on subprocess32 for the API tests
philn@webkit.org [Thu, 24 Jan 2019 14:34:50 +0000 (14:34 +0000)]
[WPE] Remove hard dependency on subprocess32 for the API tests

Rubber-stamped by Carlos Garcia Campos.

* glib/api_test_runner.py:
(TestRunner._run_test_qt): Lazily import subprocess32 and check
its presence only when running WPEQt tests.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@240433 268f45cc-cd09-0410-ab3c-d52691b4dbfc

6 months ago[JSC] Reenable baseline JIT on mips
guijemont@igalia.com [Thu, 24 Jan 2019 12:04:36 +0000 (12:04 +0000)]
[JSC] Reenable baseline JIT on mips
https://bugs.webkit.org/show_bug.cgi?id=192983

Reviewed by Mark Lam.

.:

Use baseline JIT by default on MIPS.

* Source/cmake/WebKitFeatures.cmake:

JSTests:

Added a new test for a case that was triggering a RELEASE_ASSERT when
testing.
Disable some slow tests that were already disabled for arm and x86.

* stress/json-parse-big-object.js: Added.
* stress/new-largeish-contiguous-array-with-size.js:
* stress/op_add.js:
* stress/op_bitand.js:
* stress/op_bitor.js:
* stress/op_bitxor.js:
* stress/op_lshift-ConstVar.js:
* stress/op_lshift-VarConst.js:
* stress/op_lshift-VarVar.js:
* stress/op_mod-ConstVar.js:
* stress/op_mod-VarConst.js:
* stress/op_mod-VarVar.js:
* stress/op_mul-ConstVar.js:
* stress/op_mul-VarConst.js:
* stress/op_mul-VarVar.js:
* stress/op_rshift-ConstVar.js:
* stress/op_rshift-VarConst.js:
* stress/op_rshift-VarVar.js:
* stress/op_sub-ConstVar.js:
* stress/op_sub-VarConst.js:
* stress/op_sub-VarVar.js:
* stress/op_urshift-ConstVar.js:
* stress/op_urshift-VarConst.js:
* stress/op_urshift-VarVar.js:
* stress/sampling-profiler-richards.js:
* stress/spread-forward-call-varargs-stack-overflow.js:

Source/JavaScriptCore:

Use $s0 as metadata register and make sure it's properly saved and
restored.

* jit/GPRInfo.h:
* jit/RegisterSet.cpp:
(JSC::RegisterSet::vmCalleeSaveRegisters):
(JSC::RegisterSet::llintBaselineCalleeSaveRegisters):
* llint/LowLevelInterpreter.asm:
* offlineasm/mips.rb:

Source/WTF:

Use baseline JIT by default on MIPS.

* wtf/Platform.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@240432 268f45cc-cd09-0410-ab3c-d52691b4dbfc

6 months ago[GLIB] Expose JavaScriptCore options in GLib public API
carlosgc@webkit.org [Thu, 24 Jan 2019 11:09:54 +0000 (11:09 +0000)]
[GLIB] Expose JavaScriptCore options in GLib public API
https://bugs.webkit.org/show_bug.cgi?id=188742

Reviewed by Michael Catanzaro.

Source/JavaScriptCore:

Add new API to set, get and iterate JSC options.

* API/glib/JSCOptions.cpp: Added.
(valueFromGValue):
(valueToGValue):
(jscOptionsSetValue):
(jscOptionsGetValue):
(jsc_options_set_boolean):
(jsc_options_get_boolean):
(jsc_options_set_int):
(jsc_options_get_int):
(jsc_options_set_uint):
(jsc_options_get_uint):
(jsc_options_set_size):
(jsc_options_get_size):
(jsc_options_set_double):
(jsc_options_get_double):
(jsc_options_set_string):
(jsc_options_get_string):
(jsc_options_set_range_string):
(jsc_options_get_range_string):
(jscOptionsType):
(jsc_options_foreach):
(setOptionEntry):
(jsc_options_get_option_group):
* API/glib/JSCOptions.h: Added.
* API/glib/docs/jsc-glib-4.0-sections.txt:
* API/glib/docs/jsc-glib-docs.sgml:
* API/glib/jsc.h:
* GLib.cmake:

Source/WebCore/platform/gtk/po:

* POTFILES.in: Add JSCOptions.cpp

Tools:

Add a test for the new API.

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

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@240431 268f45cc-cd09-0410-ab3c-d52691b4dbfc

6 months agoFollow-up #2: Override the session configuration for cookieAcceptPolicy
ddkilzer@apple.com [Thu, 24 Jan 2019 10:59:47 +0000 (10:59 +0000)]
Follow-up #2: Override the session configuration for cookieAcceptPolicy
<https://bugs.webkit.org/show_bug.cgi?id=190925>
<rdar://problem/45497382>

* pal/spi/cf/CFNetworkSPI.h: Protect Objective-C code using
defined(__OBJC__) macro.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@240430 268f45cc-cd09-0410-ab3c-d52691b4dbfc

6 months agoDuplicate global variables: WTF::asciiCaseFoldTable
ddkilzer@apple.com [Thu, 24 Jan 2019 10:39:14 +0000 (10:39 +0000)]
Duplicate global variables: WTF::asciiCaseFoldTable
<https://webkit.org/b/193726>
<rdar://problem/47334622>

Reviewed by Michael Catanzaro.

* WTF.xcodeproj/project.pbxproj:
* wtf/CMakeLists.txt:
- Add ASCIICType.cpp to build systems.

* wtf/ASCIICType.cpp: Added.
(WTF::asciiCaseFoldTable): Move table to here so there is only
one copy.  Mark as exported for other frameworks to use.
* wtf/ASCIICType.h:
(WTF::asciiCaseFoldTable): Change to extern declaration, and
mark as exported for other frameworks to use.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@240429 268f45cc-cd09-0410-ab3c-d52691b4dbfc

6 months ago[GTK][WPE] Support JPEG 2000 images
carlosgc@webkit.org [Thu, 24 Jan 2019 10:02:33 +0000 (10:02 +0000)]
[GTK][WPE] Support JPEG 2000 images
https://bugs.webkit.org/show_bug.cgi?id=186272

Reviewed by Žan Doberšek.

.:

Add USE_OPENJPEG build option.

* Source/cmake/OptionsGTK.cmake:
* Source/cmake/OptionsWPE.cmake:

Source/WebCore:

Add JPEG2000ImageDecoder to support JPEG2000 images using OpenJPEG. For now only SRGB and SYCC color spaces are
supported.

* platform/ImageDecoders.cmake:
* platform/MIMETypeRegistry.cpp:
(WebCore::MIMETypeRegistry::supportedImageMIMETypes):
* platform/image-decoders/ScalableImageDecoder.cpp:
(WebCore::ScalableImageDecoder::create):
* platform/image-decoders/jpeg2000/JPEG2000ImageDecoder.cpp: Added.
(WebCore::syccToRGB):
(WebCore::sycc444ToRGB):
(WebCore::sycc422ToRGB):
(WebCore::sycc420ToRGB):
(WebCore::JPEG2000ImageDecoder::JPEG2000ImageDecoder):
(WebCore::JPEG2000ImageDecoder::frameBufferAtIndex):
(WebCore::JPEG2000ImageDecoder::decode):
* platform/image-decoders/jpeg2000/JPEG2000ImageDecoder.h: Added.

Tools:

Add OpenJPEG to jhbuild since 2.2.0 version is required and it's not available in debian stable.

* gtk/jhbuild.modules:
* wpe/jhbuild.modules:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@240428 268f45cc-cd09-0410-ab3c-d52691b4dbfc

6 months agoChange some RenderLayerCompositor functions to use references
simon.fraser@apple.com [Thu, 24 Jan 2019 06:10:54 +0000 (06:10 +0000)]
Change some RenderLayerCompositor functions to use references
https://bugs.webkit.org/show_bug.cgi?id=193760

Reviewed by Zalan Bujtas.

RenderWidget* -> RenderWidget&

* rendering/RenderLayerBacking.cpp:
(WebCore::RenderLayerBacking::updateAfterWidgetResize):
(WebCore::RenderLayerBacking::updateConfiguration):
* rendering/RenderLayerCompositor.cpp:
(WebCore::RenderLayerCompositor::updateBackingAndHierarchy):
(WebCore::RenderLayerCompositor::updateBacking):
(WebCore::RenderLayerCompositor::frameContentsCompositor):
(WebCore::RenderLayerCompositor::parentFrameContentLayers):
* rendering/RenderLayerCompositor.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@240407 268f45cc-cd09-0410-ab3c-d52691b4dbfc