WebKit.git
3 years agoUnreviewed clean-up after r219323
bfulgham@apple.com [Tue, 11 Jul 2017 16:51:27 +0000 (16:51 +0000)]
Unreviewed clean-up after r219323

Simplify algorithm slightly based on a suggestion Chris Dumez made in the review.

* UIProcess/WebResourceLoadStatisticsStore.cpp:
(WebKit::pruneResources): Don't create a second variable to track the pruning operation.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@219337 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoSharedBuffer::size should return a size_t
commit-queue@webkit.org [Tue, 11 Jul 2017 16:48:02 +0000 (16:48 +0000)]
SharedBuffer::size should return a size_t
https://bugs.webkit.org/show_bug.cgi?id=174328

Patch by Alex Christensen <achristensen@webkit.org> on 2017-07-11
Reviewed by Andreas Kling.

Source/WebCore:

No change in behaviour.

* html/FTPDirectoryDocument.cpp:
(WebCore::createTemplateDocumentData):
* loader/ContentFilter.cpp:
(WebCore::ContentFilter::handleProvisionalLoadFailure):
* loader/ResourceLoader.cpp:
(WebCore::ResourceLoader::loadDataURL):
* loader/ResourceLoader.h:
* loader/appcache/ApplicationCacheStorage.cpp:
(WebCore::ApplicationCacheStorage::store):
* loader/cache/CachedScript.cpp:
(WebCore::CachedScript::script):
* platform/SharedBuffer.cpp:
(WebCore::SharedBuffer::tryCreateArrayBuffer):
* platform/SharedBuffer.h:

Source/WebKit2:

* UIProcess/WebResourceLoadStatisticsStore.cpp:
(WebKit::WebResourceLoadStatisticsStore::writeEncoderToDisk):

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@219336 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years ago[Win] Build error when building WebCore from WebCore.proj project file.
pvollan@apple.com [Tue, 11 Jul 2017 16:43:04 +0000 (16:43 +0000)]
[Win] Build error when building WebCore from WebCore.proj project file.
https://bugs.webkit.org/show_bug.cgi?id=174330

Reviewed by Brent Fulgham.

Source/WebCore:

The CMake variable PAL_DIR should be set in the project file.

* WebCore.vcxproj/WebCore.proj:

Source/WebCore/PAL:

CMAKE_BINARY_DIR is needed in the include list in order to find 'cmakeconfig.h'.

* pal/CMakeLists.txt:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@219335 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years ago[SVG] Leak in SVGAnimatedListPropertyTearOff
svillar@igalia.com [Tue, 11 Jul 2017 16:30:43 +0000 (16:30 +0000)]
[SVG] Leak in SVGAnimatedListPropertyTearOff
https://bugs.webkit.org/show_bug.cgi?id=172545

Source/WebCore:

Reviewed by Said Abou-Hallawa.

SVGAnimatedListPropertyTearOff maintains a vector m_wrappers with references to
SVGPropertyTraits<PropertyType>::ListItemTearOff. Apart from that SVGPropertyTearOff has a
reference to SVGAnimatedProperty.

When SVGListProperty::getItemValuesAndWrappers() is called, it creates a
SVGPropertyTraits<PropertyType>::ListItemTearOff pointing to the same SVGAnimatedProperty (a
SVGAnimatedListPropertyTearOff) which stores the m_wrappers vector where the ListItemTearOff
is going to be added to. This effectively creates a reference cycle between the
SVGAnimatedListPropertyTearOff and all the ListItemTearOff it stores in m_wrappers.

We should detach those wrappers in propertyWillBeDeleted() in order to break the cycle.

* svg/properties/SVGAnimatedListPropertyTearOff.h:

LayoutTests:

Reviewed by Darin Adler.

* svg/animations/animation-leak-list-property-instances-expected.txt: Added.
* svg/animations/animation-leak-list-property-instances.html: Added.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@219334 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoWeb Inspector: Node Sidebar - Improve wording for Potential Custom Element lacking...
commit-queue@webkit.org [Tue, 11 Jul 2017 16:15:11 +0000 (16:15 +0000)]
Web Inspector: Node Sidebar - Improve wording for Potential Custom Element lacking a Custom definition
https://bugs.webkit.org/show_bug.cgi?id=174346

Patch by Joseph Pecoraro <pecoraro@apple.com> on 2017-07-11
Reviewed by Matt Baker.

* Localizations/en.lproj/localizedStrings.js:
* UserInterface/Views/DOMNodeDetailsSidebarPanel.js:
(WebInspector.DOMNodeDetailsSidebarPanel.prototype._customElementState):
The text "Waiting to be upgraded" is poor. Any non-standard element with dashes
in the name is considered custom, but does not need to be upgraded. Lets use
different wording to indicate that it is detected as custom, and lacks a
specific Custom Element definition.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@219333 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years ago[GTK] Spin buttons on input type number appear over the value itself for small widths
clopez@igalia.com [Tue, 11 Jul 2017 15:34:41 +0000 (15:34 +0000)]
[GTK] Spin buttons on input type number appear over the value itself for small widths
https://bugs.webkit.org/show_bug.cgi?id=173572

Reviewed by Carlos Garcia Campos.

Source/WebCore:

When drawing the spin buttons, override the width of the input
element to increment it with the width of the spin button.
This ensures that we don't end up covering the input values with
the spin buttons.

Do this also for user controlled styles, because most web authors
won't test how their site renders on WebKitGTK+, and they will
assume spin buttons in the order of 13 pixels wide (that is what
most browsers use), but the GTK+ spin button is much wider (66 pixels).

Test: platform/gtk/fast/forms/number/number-size-spinbutton-nocover.html

* rendering/RenderTheme.cpp:
(WebCore::RenderTheme::adjustStyle):
* rendering/RenderThemeGtk.cpp:
(WebCore::RenderThemeGtk::adjustTextFieldStyle): Call the theme's adjustTextFieldStyle() also for user controlled styles.
(WebCore::RenderThemeGtk::adjustInnerSpinButtonStyle):

LayoutTests:

* fast/forms/number/number-size-expected.txt:
* fast/forms/number/number-size.html: Obtain the size of the spinbutton by inspecting the shadow dom (is more reliable)
  and make the test for user controlled style special for the case of GTK+.
* platform/gtk/fast/forms/number/number-size-spinbutton-nocover-expected.png: Added.
* platform/gtk/fast/forms/number/number-size-spinbutton-nocover-expected.txt: Added.
* platform/gtk/fast/forms/number/number-size-spinbutton-nocover.html: Added.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@219332 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoWe should do ICE candidate filtering at the Document level
commit-queue@webkit.org [Tue, 11 Jul 2017 15:32:38 +0000 (15:32 +0000)]
We should do ICE candidate filtering at the Document level
https://bugs.webkit.org/show_bug.cgi?id=173861
<rdar://problem/33122058>

Patch by Youenn Fablet <youenn@apple.com> on 2017-07-11
Reviewed by Eric Carlson.

Source/WebCore:

Tests: http/tests/webrtc/filtering-ice-candidate-cross-origin-frame.html
       http/tests/webrtc/filtering-ice-candidate-same-origin-frame.html
       http/tests/webrtc/filtering-ice-candidate-same-origin-frame2.html
       webrtc/filtering-ice-candidate-after-reload.html

Making UserMediaRequest disable the ICE candidate filtering for the page RTCController.
All RTCPeerConnection of the page that are created on a document that are same-origin as the top document
are now registered to the RTCController.
This allows disabling filtering to only these RTCPeerConnection.

The page keeps the default ICE candidate filtering policy.
This policy allows disabling ICE candidate filtering for all RTCPeerConnection.

When the top document is changing, the RTCController filtering policy is reset
and its list of RTCPeerConnection is emptied.

Internals no longer disables ICE candidate filtering by default.
This allows finer grained testing.
ICE candidate filtering is disabled for tests including testharnessreport.js
to enable web-platform-tests to run without modifications.

* Modules/mediastream/RTCController.cpp:
(WebCore::RTCController::reset):
* Modules/mediastream/RTCController.h:
* Modules/mediastream/UserMediaRequest.cpp:
(WebCore::UserMediaRequest::allow):
* page/Frame.cpp:
(WebCore::Frame::setDocument):
* page/Page.cpp:
(WebCore::Page::disableICECandidateFiltering):
* page/Page.h:
(WebCore::Page::shouldEnableICECandidateFilteringByDefault):
(WebCore::Page::disableICECandidateFiltering): Deleted.
(WebCore::Page::enableICECandidateFiltering): Deleted.
(WebCore::Page::isICECandidateFilteringEnabled): Deleted.
* testing/Internals.cpp:
(WebCore::Internals::Internals):
(WebCore::Internals::setICECandidateFiltering):
(WebCore::Internals::setEnumeratingAllNetworkInterfacesEnabled):
(WebCore::Internals::isICECandidateFilteringEnabled): Deleted.
* testing/Internals.h:
* testing/Internals.idl:

Source/WebKit2:

Removing ICE candidate filtering handling at UIProcess level.
ICE candidate filtering is now disabled at UserMediaRequest level.
WebPage forwards the ICE candidate filtering option to the page so as to set
the default option correctly for every document of the page.

* UIProcess/UserMediaProcessManager.cpp:
(WebKit::UserMediaProcessManager::willCreateMediaStream):
(WebKit::UserMediaProcessManager::endedCaptureSession):
* WebProcess/WebPage/WebPage.cpp:
(WebKit::WebPage::disableICECandidateFiltering):
(WebKit::WebPage::enableICECandidateFiltering):

LayoutTests:

* http/tests/webrtc/filtering-ice-candidate-cross-origin-frame-expected.txt: Added.
* http/tests/webrtc/filtering-ice-candidate-cross-origin-frame.html: Added.
* http/tests/webrtc/filtering-ice-candidate-same-origin-frame-expected.txt: Added.
* http/tests/webrtc/filtering-ice-candidate-same-origin-frame.html: Added.
* http/tests/webrtc/filtering-ice-candidate-same-origin-frame2-expected.txt: Added.
* http/tests/webrtc/filtering-ice-candidate-same-origin-frame2.html: Added.
* http/tests/webrtc/resources/check-ice-candidate-filtering.html: Added.
* http/tests/webrtc/resources/do-get-user-media.html: Added.
* platform/mac-wk1/TestExpectations: Disable http/tests/webrtc tests.
* resources/testharnessreport.js: Disabled ICE candidate filtering by default.
* webrtc/filtering-ice-candidate-after-reload-expected.txt: Added.
* webrtc/filtering-ice-candidate-after-reload.html: Added.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@219331 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoChanging the web page muted state for playing audio should not disable other tabs...
commit-queue@webkit.org [Tue, 11 Jul 2017 15:14:36 +0000 (15:14 +0000)]
Changing the web page muted state for playing audio should not disable other tabs capture
https://bugs.webkit.org/show_bug.cgi?id=174349
rdar://problem/33223988

Patch by Youenn Fablet <youenn@apple.com> on 2017-07-11
Reviewed by Eric Carlson.

Make sure to mute capturing streams of any other tab if the current tab will start capturing or will unmute existing capturing streams.
Manually tested by having a tab doing capture and another tab playing a video.
Muting or unmuting the tab playing video should not change the capture tab.

* UIProcess/WebPageProxy.cpp:
(WebKit::WebPageProxy::activateMediaStreamCaptureInPage):
(WebKit::WebPageProxy::setMuted):
* UIProcess/WebPageProxy.h:
(WebKit::WebPageProxy::activateMediaStreamCaptureInPage):

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@219330 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years ago[WPE] Bump GStreamer packages in jhbuild.modules to 1.10.5
zandobersek@gmail.com [Tue, 11 Jul 2017 14:59:25 +0000 (14:59 +0000)]
[WPE] Bump GStreamer packages in jhbuild.modules to 1.10.5
https://bugs.webkit.org/show_bug.cgi?id=174350

Reviewed by Xabier Rodriguez-Calvar.

Bump the various GStreamer dependencies listed in WPE's Jhbuild modules file
to the 1.10.5 version. One gstreamer patch is added. The two gst-plugins-bad
patches are removed. The patches for gst-plugins-good are updated, removing
the stale ones and adding the additional changes to qtdemux, souphttpsrc and
rtpbin elements that will help with the EMEv3 development.

* wpe/jhbuild.modules:
* wpe/patches/gst-plugins-bad-0001-dtls-port-to-OpenSSL-1.1.0.patch: Removed.
* wpe/patches/gst-plugins-bad-0002-dtlscertificate-Fix-error-checking-in-RSA_generate_k.patch: Removed.
* wpe/patches/gst-plugins-good-0003-rtpbin-receive-bundle-support.patch: Added.
* wpe/patches/gst-plugins-good-0005-souphttpsrc-cookie-jar-and-context-query-support.patch: Added.
* wpe/patches/gst-plugins-good-0006-qtdemux-add-context-for-a-preferred-protection.patch: Added.
* wpe/patches/gst-plugins-good-0008-qtdemux-also-push-buffers-without-encryption-info-in.patch: Added.
* wpe/patches/gst-plugins-good-Revert-qtdemux-expose-streams-with-first-moof-for-fr.patch: Removed.
* wpe/patches/gst-plugins-good-use-the-tfdt-decode-time.patch: Removed.
* wpe/patches/gstreamer-0001-protection-added-function-to-filter-system-ids.patch: Added.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@219329 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoNetworkProcess should close listening WebRTC sockets when being suspended
commit-queue@webkit.org [Tue, 11 Jul 2017 14:53:18 +0000 (14:53 +0000)]
NetworkProcess should close listening WebRTC sockets when being suspended
https://bugs.webkit.org/show_bug.cgi?id=174270
rdar://problem/33139844

Patch by Youenn Fablet <youenn@apple.com> on 2017-07-11
Reviewed by Chris Dumez.

To prevent potential spinning of the NetworkProcess, NetworkProcess will now close listening sockets when being notified that it will be suspended.
When the network process is being suspended, it will stop creating listening sockets, until it resumes.
Future additional efforts might be to improve select/cancel so that we can stop listening sockets at resume time,
or to reimplement part of the stack using CFStream.

Tested through manual testing by going to a website doing WebRTC, homing out so that the network process is suspended and reopening Safari.

* NetworkProcess/NetworkConnectionToWebProcess.cpp:
(WebKit::NetworkConnectionToWebProcess::cleanupForSuspension):
Proxying call to clean for suspension to rtc provider so that it will
close listening sockets.
(WebKit::NetworkConnectionToWebProcess::resumeFromSuspension): Authorizing back listening sockets.
* NetworkProcess/NetworkConnectionToWebProcess.h:
(WebKit::NetworkConnectionToWebProcess::cleanupForSuspension):
Clean-up is done asynchronously as it can happen in background threads.
Hence why passing a callback as parameter.
* NetworkProcess/NetworkProcess.cpp:
(WebKit::NetworkProcess::notifyProcessReadyToSuspend): Helper routine
to notify UI process that network process clean up is done.
(WebKit::TaskCounter::TaskCounter): Helper class to call notifyProcessReadyToSuspend when sded.
(WebKit::TaskCounter::~TaskCounter):
(WebKit::NetworkProcess::actualPrepareToSuspend): Doing the clean-up for each connection that needs it.
Making sure to notify UI process of clean-up being completed once all connections are cleaned.
(WebKit::NetworkProcess::processWillSuspendImminently):
(WebKit::NetworkProcess::prepareToSuspend):
(WebKit::NetworkProcess::processDidResume): Reenable listening sockets.
* NetworkProcess/NetworkProcess.h:
* NetworkProcess/webrtc/LibWebRTCSocketClient.cpp:
(WebKit::LibWebRTCSocketClient::LibWebRTCSocketClient):
* NetworkProcess/webrtc/LibWebRTCSocketClient.h: Adding type getter and making close public.
Used by NetworkRTCProvider to identifiy listening sockets and close them.
* NetworkProcess/webrtc/NetworkRTCProvider.cpp:
(WebKit::NetworkRTCProvider::closeListeningSockets): We close the webrtc socket
and we also notify the Web Process that the socket is closed so that it can take actions to recreate some if needed.
(WebKit::NetworkRTCProvider::finishClosingListeningSockets):
* NetworkProcess/webrtc/NetworkRTCProvider.h:
(WebKit::NetworkRTCProvider::authorizeListeningSockets): Authorize creation of listening sockets.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@219328 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoUnreviewed, rolling out r219325.
svillar@igalia.com [Tue, 11 Jul 2017 13:29:40 +0000 (13:29 +0000)]
Unreviewed, rolling out r219325.

The test is still flaky

Reverted changeset:

"[SVG] Leak in SVGAnimatedListPropertyTearOff"
https://bugs.webkit.org/show_bug.cgi?id=172545
http://trac.webkit.org/changeset/219325

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@219327 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years ago[GTK] media/video-restricted-invisible-autoplay-not-allowed.html is flaky
commit-queue@webkit.org [Tue, 11 Jul 2017 13:15:33 +0000 (13:15 +0000)]
[GTK] media/video-restricted-invisible-autoplay-not-allowed.html is flaky
https://bugs.webkit.org/show_bug.cgi?id=174353

Unreviewed test gardening.

Patch by Charlie Turner <cturner@igalia.com> on 2017-07-11

* platform/gtk/TestExpectations:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@219326 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years ago[SVG] Leak in SVGAnimatedListPropertyTearOff
svillar@igalia.com [Tue, 11 Jul 2017 10:44:18 +0000 (10:44 +0000)]
[SVG] Leak in SVGAnimatedListPropertyTearOff
https://bugs.webkit.org/show_bug.cgi?id=172545

Reviewed by Said Abou-Hallawa.

SVGAnimatedListPropertyTearOff maintains a vector m_wrappers with references to
SVGPropertyTraits<PropertyType>::ListItemTearOff. Apart from that SVGPropertyTearOff has a
reference to SVGAnimatedProperty.

When SVGListProperty::getItemValuesAndWrappers() is called, it creates a
SVGPropertyTraits<PropertyType>::ListItemTearOff pointing to the same SVGAnimatedProperty (a
SVGAnimatedListPropertyTearOff) which stores the m_wrappers vector where the ListItemTearOff
is going to be added to. This effectively creates a reference cycle between the
SVGAnimatedListPropertyTearOff and all the ListItemTearOff it stores in m_wrappers.

We should detach those wrappers in propertyWillBeDeleted() in order to break the cycle.

* svg/properties/SVGAnimatedListPropertyTearOff.h:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@219325 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoUnreviewed WPE gardening.
zandobersek@gmail.com [Tue, 11 Jul 2017 07:49:54 +0000 (07:49 +0000)]
Unreviewed WPE gardening.

* platform/wpe/TestExpectations:
Enable EMEv3 tests under media/encrypted-media/, all of them passing.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@219324 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoResource Load Statistics: Prune statistics in orders of importance
bfulgham@apple.com [Tue, 11 Jul 2017 04:08:20 +0000 (04:08 +0000)]
Resource Load Statistics: Prune statistics in orders of importance
https://bugs.webkit.org/show_bug.cgi?id=174215
<rdar://problem/33164403>

Unreviewed test correction.

Source/WebKit2:

* UIProcess/WebResourceLoadStatisticsStore.cpp:
(WebKit::pruneResources): The revised algorithm did not update the remaining
count to be pruned, causing a test failure.

LayoutTests:

Drive-by fix after recent refactoring to correct the new function names in TestRunner.

* http/tests/loading/resourceLoadStatistics/partitioned-and-unpartitioned-cookie-deletion.html:
* http/tests/loading/resourceLoadStatistics/partitioned-cookies-with-and-without-user-interaction.html:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@219323 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoAddress Antti's review comment.
rniwa@webkit.org [Tue, 11 Jul 2017 03:46:16 +0000 (03:46 +0000)]
Address Antti's review comment.

* public/v3/models/analysis-results.js:
(AnalysisResults.prototype.containsTest):

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@219322 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoA/B testing results page show results for the top-level tests instead of the one...
rniwa@webkit.org [Tue, 11 Jul 2017 03:45:00 +0000 (03:45 +0000)]
A/B testing results page show results for the top-level tests instead of the one being analyzed
https://bugs.webkit.org/show_bug.cgi?id=174304

Reviewed by Antti Koivisto.

When a specific subtest is analyzed (e.g. Images subtest of MotionMark), then TestGroupResultsViewer
should expand and highlight that specific subtest instead of simply showing the top-level test's score.
This is especially misleading since AnalysisResultsViewer (stacking bars for each test group) uses
the score of the specific subtest being analyzed.

Fixed the bug by passing in the metric associated with the analysis task from AnalysisTaskPage to
TestGroupResultsViewer via AnalysisTaskTestGroupPane. Also made TestGroupResultsViewer.setAnalysisResults
auto-expand the tests that are ancestors of the specified metric. Without that, the test won't be shown
to the user until the ancestor tests are expanded by the user.

Also fixed the bug that we were always listing sub-tests regardless of whether they have results or not.
Since tests tend to change over time, we shouldn't show a test if it doesn't have any results associated.

* public/v3/components/test-group-results-viewer.js:
(TestGroupResultsViewer.prototype.setAnalysisResults): Expand the ancestor tests of the metric.
(TestGroupResultsViewer.prototype._buildRowsForTest): Exit early if this test doesn't have any results.
* public/v3/models/analysis-results.js:
(AnalysisResults.prototype.containsTest): Added.
* public/v3/pages/analysis-task-page.js:
(AnalysisTaskTestGroupPane.prototype.setAnalysisResults): Takes a metric to pass it to the results viewer.
(AnalysisTaskPage.prototype._assignTestResultsIfPossible):

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@219321 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years ago[WK2 iOS] REGRESSION (r216803) During momentum scroll, getBoundingClientRect returns...
simon.fraser@apple.com [Tue, 11 Jul 2017 03:40:59 +0000 (03:40 +0000)]
[WK2 iOS] REGRESSION (r216803) During momentum scroll, getBoundingClientRect returns wrong coordinates (missing images on pinterest, elle.com and many other sites)
https://bugs.webkit.org/show_bug.cgi?id=174286
rdar://problem/32864180

Reviewed by Dean Jackson.

Source/WebCore:

r216803 made getBoundingClientRects relative to the layout viewport, but when scrolling we
only update that on stable viewport updates (at the end of the scroll). This meant that during
unstable updates, getBoundingClientRects() used a "frozen" viewport origin so things on-screen
would appear to be off-screen, causing sites to fail to dynamically load images etc. when
scrolling.

Fix by pushing an optional "unstable" layout viewport rect onto FrameView, which gets used by
FrameView::documentToClientOffset(). This is cleared when we do a stable update.

This is a short-term solution. Longer term, I would prefer to always call setLayoutViewportOverrideRect(),
but fix the scrolling tree logic to work correctly in this case.

Add a bit more scrolling logging.

Test: fast/visual-viewport/ios/get-bounding-client-rect-unstable.html

* page/FrameView.cpp:
(WebCore::FrameView::setUnstableLayoutViewportRect):
(WebCore::FrameView::documentToClientOffset):
* page/FrameView.h:
* page/scrolling/AsyncScrollingCoordinator.cpp:
(WebCore::AsyncScrollingCoordinator::reconcileScrollingState):
* page/scrolling/ScrollingStateFixedNode.cpp:
(WebCore::ScrollingStateFixedNode::updateConstraints):
(WebCore::ScrollingStateFixedNode::reconcileLayerPositionForViewportRect):

LayoutTests:

* fast/visual-viewport/ios/get-bounding-client-rect-unstable-expected.txt: Added.
* fast/visual-viewport/ios/get-bounding-client-rect-unstable.html: Added.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@219320 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoResource Load Statistics: Prune statistics in orders of importance
bfulgham@apple.com [Tue, 11 Jul 2017 02:52:07 +0000 (02:52 +0000)]
Resource Load Statistics: Prune statistics in orders of importance
https://bugs.webkit.org/show_bug.cgi?id=174215
<rdar://problem/33164403>

Patch by John Wilander <wilander@apple.com> on 2017-07-10
Reviewed by Chris Dumez.

Source/WebCore:

Test: http/tests/loading/resourceLoadStatistics/prune-statistics.html

* loader/ResourceLoadObserver.cpp:
(WebCore::reduceTimeResolution):
(WebCore::ResourceLoadObserver::logFrameNavigation):
(WebCore::ResourceLoadObserver::logSubresourceLoading):
(WebCore::ResourceLoadObserver::logWebSocketLoading):
(WebCore::ResourceLoadObserver::logUserInteractionWithReducedTimeResolution):
    Now all set the new statistics field lastSeen.
* loader/ResourceLoadStatistics.cpp:
(WebCore::ResourceLoadStatistics::encode):
(WebCore::ResourceLoadStatistics::decode):
(WebCore::ResourceLoadStatistics::toString):
(WebCore::ResourceLoadStatistics::merge):
    Handling of the new statistics field lastSeen.
* loader/ResourceLoadStatistics.h:

Source/WebKit2:

New functionality. Prunes statistics in this order:
1. Non-prevalent resources without user interaction.
2. Prevalent resources without user interaction.
3. Non-prevalent resources with user interaction.
4. Prevalent resources with user interaction.

* Shared/WebCoreArgumentCoders.cpp:
(IPC::ArgumentCoder<ResourceLoadStatistics>::encode):
(IPC::ArgumentCoder<ResourceLoadStatistics>::decode):
    Added timestamp field lastSeen.
* UIProcess/API/Cocoa/WKWebsiteDataStore.mm:
(-[WKWebsiteDataStore _resourceLoadStatisticsSetLastSeen:forHost:]):
(-[WKWebsiteDataStore _resourceLoadStatisticsSetMaxStatisticsEntries:]):
(-[WKWebsiteDataStore _resourceLoadStatisticsSetPruneEntriesDownTo:]):
(-[WKWebsiteDataStore _resourceLoadStatisticsResetToConsistentState]):
    Test infrastructure.
* UIProcess/API/Cocoa/WKWebsiteDataStorePrivate.h:
* UIProcess/Storage/ResourceLoadStatisticsStore.cpp:
(WebKit::ResourceLoadStatisticsStore::setMaxStatisticsEntries):
    Test infrastructure.
(WebKit::ResourceLoadStatisticsStore::setPruneEntriesDownTo):
    Test infrastructure.
(WebKit::sortAndPrune):
    Convenience function.
(WebKit::ResourceLoadStatisticsStore::pruneStatisticsIfNeeded):
    The new pruning function.
* UIProcess/Storage/ResourceLoadStatisticsStore.h:
* UIProcess/WebResourceLoadStatisticsStore.cpp:
(WebKit::WebResourceLoadStatisticsStore::processStatisticsAndDataRecords):
    Now calls ResourceLoadStatisticsStore::pruneStatisticsIfNeeded().
(WebKit::WebResourceLoadStatisticsStore::setLastSeen):
    Test infrastructure.
(WebKit::WebResourceLoadStatisticsStore::setMaxStatisticsEntries):
    Test infrastructure.
(WebKit::WebResourceLoadStatisticsStore::setPruneEntriesDownTo):
    Test infrastructure.
* UIProcess/WebResourceLoadStatisticsStore.h:

Tools:

Nest infrastructure. Adds these functions:
1. testRunner.setStatisticsLastSeen()
2. setStatisticsMaxStatisticsEntries()
3. setStatisticsPruneEntriesDownTo()

* WebKitTestRunner/InjectedBundle/Bindings/TestRunner.idl:
* WebKitTestRunner/InjectedBundle/TestRunner.cpp:
(WTR::TestRunner::setStatisticsLastSeen):
(WTR::TestRunner::setStatisticsMaxStatisticsEntries):
(WTR::TestRunner::setStatisticsPruneEntriesDownTo):
* WebKitTestRunner/InjectedBundle/TestRunner.h:
* WebKitTestRunner/TestController.cpp:
(WTR::TestController::setStatisticsLastSeen):
(WTR::TestController::setMaxStatisticsEntries):
(WTR::TestController::setPruneEntriesDownTo):
* WebKitTestRunner/TestController.h:
* WebKitTestRunner/TestInvocation.cpp:
(WTR::TestInvocation::didReceiveSynchronousMessageFromInjectedBundle):
* WebKitTestRunner/cocoa/TestControllerCocoa.mm:
(WTR::TestController::setStatisticsLastSeen):
(WTR::TestController::setStatisticsMaxStatisticsEntries):
(WTR::TestController::setStatisticsPruneEntriesDownTo):

LayoutTests:

* http/tests/loading/resourceLoadStatistics/prune-statistics-expected.txt: Added.
* http/tests/loading/resourceLoadStatistics/prune-statistics.html: Added.
* platform/wk2/TestExpectations:
    Added http/tests/loading/resourceLoadStatistics/prune-statistics.html as
    [ Pass ] since Resource Load Statistics is WK2-only.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@219319 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoconst() experimental feature should always be on by default
dino@apple.com [Tue, 11 Jul 2017 02:39:16 +0000 (02:39 +0000)]
const() experimental feature should always be on by default
https://bugs.webkit.org/show_bug.cgi?id=174341
<rdar://problem/33228603>

Reviewed by Simon Fraser.

Even though this is experimental, it should be enabled by default, and
not follow the value of DEFAULT_EXPERIMENTAL_FEATURES_ENABLED.

* Shared/WebPreferencesDefinitions.h:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@219318 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoAllocation sinking phase should consider a CheckStructure that would fail as an escape
sbarati@apple.com [Tue, 11 Jul 2017 00:29:36 +0000 (00:29 +0000)]
Allocation sinking phase should consider a CheckStructure that would fail as an escape
https://bugs.webkit.org/show_bug.cgi?id=174321
<rdar://problem/32604963>

Reviewed by Filip Pizlo.

When the allocation sinking phase was generating stores to materialize
objects in a cycle with each other, it would assume that each materialized
object had a valid, non empty, set of structures. This is an OK assumption for
the phase to make because how do you materialize an object with no structure?

The abstract interpretation part of the phase will model what's in the heap.
However, it would sometimes model that a CheckStructure would fail. The phase
did nothing special for this; it just stored the empty set of structures for
its representation of a particular allocation. However, what the phase proved
in such a scenario is that, had the CheckStructure executed, it would have exited.

This patch treats such CheckStructures and MultiGetByOffsets as escape points.
This will cause the allocation in question to be materialized just before
the CheckStructure, and then at execution time, the CheckStructure will exit.

I wasn't able to write a test case for this. However, I was able to reproduce
this crash by manually editing the IR. I've opened a separate bug to help us
create a testing framework for writing tests for hard to reproduce bugs like this:
https://bugs.webkit.org/show_bug.cgi?id=174322

* dfg/DFGObjectAllocationSinkingPhase.cpp:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@219317 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoWeb Inspector: Highlight matching CSS canvas clients when hovering contexts in the...
drousso@apple.com [Tue, 11 Jul 2017 00:01:55 +0000 (00:01 +0000)]
Web Inspector: Highlight matching CSS canvas clients when hovering contexts in the Resources tab
https://bugs.webkit.org/show_bug.cgi?id=174279

Reviewed by Matt Baker.

Source/JavaScriptCore:

* inspector/protocol/DOM.json:
Add `highlightNodeList` command that will highlight each node in the given list.

Source/WebCore:

Test: inspector/dom/highlightNodeList.html

* inspector/InspectorDOMAgent.h:
* inspector/InspectorDOMAgent.cpp:
(WebCore::InspectorDOMAgent::highlightNodeList):

Source/WebInspectorUI:

* UserInterface/Controllers/DOMTreeManager.js:
(WebInspector.DOMTreeManager.prototype.highlightDOMNodeList):
(WebInspector.DOMTreeManager.prototype.highlightSelector):
* UserInterface/Views/CanvasTreeElement.js:
(WebInspector.CanvasTreeElement.prototype._handleMouseOver):

LayoutTests:

* inspector/dom/highlightNodeList-expected.html: Added.
* inspector/dom/highlightNodeList.html: Added.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@219316 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years ago[css-align][css-flex][css-grid] 'auto' values of align-self and justify-self must...
jfernandez@igalia.com [Mon, 10 Jul 2017 23:56:21 +0000 (23:56 +0000)]
[css-align][css-flex][css-grid] 'auto' values of align-self and justify-self must not be resolved
https://bugs.webkit.org/show_bug.cgi?id=172707

Reviewed by Antti Koivisto.

LayoutTests/imported/w3c:

This change makes all the cases of the test below to pass now, hence updated expectations accordingly.

* web-platform-tests/css/css-align-3/self-alignment/place-self-shorthand-006-expected.txt:

Source/WebCore:

The CSS Box Alignment specification has been changed recently so that
now all the propeties have the specificed value as computed value. The
rationale of this change are at the associated W3C github issue [1].

This change implies that we don't need to execute the StyleAdjuter
logic we implemented specifically for supporting 'auto' values
resolution for computed style. We can live now with resolution at
layout time only.

[1] https://github.com/w3c/csswg-drafts/issues/440

No new tests, just updating the already defined tests.

* css/CSSComputedStyleDeclaration.cpp:
(WebCore::ComputedStyleExtractor::propertyValue):
* css/StyleResolver.cpp:
(WebCore::StyleResolver::adjustRenderStyle): Removed
* css/StyleResolver.h:
* html/shadow/TextControlInnerElements.cpp:
(WebCore::TextControlInnerElement::resolveCustomStyle):
* rendering/RenderBox.cpp:
(WebCore::RenderBox::columnFlexItemHasStretchAlignment):
(WebCore::RenderBox::hasStretchedLogicalWidth):
* rendering/RenderFlexibleBox.cpp:
(WebCore::RenderFlexibleBox::styleDidChange): Added
(WebCore::RenderFlexibleBox::alignmentForChild):
* rendering/RenderFlexibleBox.h:

LayoutTests:

Updated layout tests so that resolved value is as specified, even for 'auto' values.

* TestExpectations:
* css3/flexbox/css-properties-expected.txt:
* css3/flexbox/css-properties.html:
* css3/flexbox/new-alignment-values-invalid-if-grid-not-enabled-expected.txt:
* css3/flexbox/new-alignment-values-invalid-if-grid-not-enabled.html:
* css3/parse-align-self.html:
* css3/parse-alignment-of-root-elements-expected.txt:
* css3/parse-alignment-of-root-elements.html:
* css3/parse-place-items.html:
* css3/parse-place-self.html:
* fast/css/ensure-flexbox-compatibility-with-initial-values-expected.txt:
* fast/css/parse-justify-self.html:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@219315 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years ago[iOS] _didCommitLayerTree should avoid calling [scrollView setZoomScale] unnecessarily
cdumez@apple.com [Mon, 10 Jul 2017 23:28:27 +0000 (23:28 +0000)]
[iOS] _didCommitLayerTree should avoid calling [scrollView setZoomScale] unnecessarily
https://bugs.webkit.org/show_bug.cgi?id=174326
<rdar://problem/33150490>

Reviewed by Simon Fraser.

Only call [scrollView setZoomScale] if the new zoom scale is actually different to
avoid doing unnecessary work in UIKit.

* UIProcess/API/Cocoa/WKWebView.mm:
(-[WKWebView _didCommitLayerTree:]):

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@219314 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agofast/text/trak-optimizeLegibility.html is failing on macOS High Sierra
mmaxfield@apple.com [Mon, 10 Jul 2017 23:25:19 +0000 (23:25 +0000)]
fast/text/trak-optimizeLegibility.html is failing on macOS High Sierra
https://bugs.webkit.org/show_bug.cgi?id=174325
<rdar://problem/33190322>

Reviewed by Zalan Bujtas.

The name of the font changed.

* fast/text/trak-optimizeLegibility.html:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@219313 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoMerge ResourceLoadStatisticsStore into WebResourceLoadStatisticsStore
cdumez@apple.com [Mon, 10 Jul 2017 23:00:42 +0000 (23:00 +0000)]
Merge ResourceLoadStatisticsStore into WebResourceLoadStatisticsStore
https://bugs.webkit.org/show_bug.cgi?id=174203

Reviewed by Brent Fulgham.

Merge ResourceLoadStatisticsStore into WebResourceLoadStatisticsStore. The 2 classes
have a similar purpose and there is no clean separation between the 2. It makes more
sense to have a single store class for resource load statistics.

If we want to simplify the WebResourceLoadStatisticsStore class, I think it'd make
more sense to split the file system I/O code out. This code adds quite a bit of
complexity.

* CMakeLists.txt:
* UIProcess/API/Cocoa/WKWebsiteDataStore.mm:
(-[WKWebsiteDataStore _resourceLoadStatisticsUpdateCookiePartitioning]):
(-[WKWebsiteDataStore _resourceLoadStatisticsSetShouldPartitionCookies:forHost:]):
(-[WKWebsiteDataStore _resourceLoadStatisticsClearInMemoryAndPersistentStore]):
(-[WKWebsiteDataStore _resourceLoadStatisticsClearInMemoryAndPersistentStoreModifiedSinceHours:]):
(-[WKWebsiteDataStore _resourceLoadStatisticsResetToConsistentState]):
* UIProcess/Storage/ResourceLoadStatisticsStore.cpp: Removed.
* UIProcess/Storage/ResourceLoadStatisticsStore.h: Removed.
* UIProcess/WebResourceLoadStatisticsStore.cpp:
(WebKit::WebResourceLoadStatisticsStore::WebResourceLoadStatisticsStore):
(WebKit::WebResourceLoadStatisticsStore::removeDataRecords):
(WebKit::WebResourceLoadStatisticsStore::processStatisticsAndDataRecords):
(WebKit::WebResourceLoadStatisticsStore::resourceLoadStatisticsUpdated):
(WebKit::WebResourceLoadStatisticsStore::grandfatherExistingWebsiteData):
(WebKit::WebResourceLoadStatisticsStore::readDataFromDiskIfNeeded):
(WebKit::WebResourceLoadStatisticsStore::refreshFromDisk):
(WebKit::WebResourceLoadStatisticsStore::writeStoreToDisk):
(WebKit::WebResourceLoadStatisticsStore::startMonitoringStatisticsStorage):
(WebKit::WebResourceLoadStatisticsStore::performDailyTasks):
(WebKit::WebResourceLoadStatisticsStore::submitTelemetry):
(WebKit::WebResourceLoadStatisticsStore::logUserInteraction):
(WebKit::WebResourceLoadStatisticsStore::clearUserInteraction):
(WebKit::WebResourceLoadStatisticsStore::hasHadUserInteraction):
(WebKit::WebResourceLoadStatisticsStore::setPrevalentResource):
(WebKit::WebResourceLoadStatisticsStore::isPrevalentResource):
(WebKit::WebResourceLoadStatisticsStore::clearPrevalentResource):
(WebKit::WebResourceLoadStatisticsStore::setGrandfathered):
(WebKit::WebResourceLoadStatisticsStore::isGrandfathered):
(WebKit::WebResourceLoadStatisticsStore::setSubframeUnderTopFrameOrigin):
(WebKit::WebResourceLoadStatisticsStore::setSubresourceUnderTopFrameOrigin):
(WebKit::WebResourceLoadStatisticsStore::setSubresourceUniqueRedirectTo):
(WebKit::WebResourceLoadStatisticsStore::scheduleCookiePartitioningUpdate):
(WebKit::WebResourceLoadStatisticsStore::scheduleCookiePartitioningUpdateForDomains):
(WebKit::WebResourceLoadStatisticsStore::scheduleClearInMemory):
(WebKit::WebResourceLoadStatisticsStore::scheduleClearInMemoryAndPersistent):
(WebKit::WebResourceLoadStatisticsStore::setTimeToLiveUserInteraction):
(WebKit::WebResourceLoadStatisticsStore::setTimeToLiveCookiePartitionFree):
(WebKit::WebResourceLoadStatisticsStore::setGrandfatheringTime):
(WebKit::WebResourceLoadStatisticsStore::ensureResourceStatisticsForPrimaryDomain):
(WebKit::WebResourceLoadStatisticsStore::createEncoderFromData):
(WebKit::WebResourceLoadStatisticsStore::populateFromDecoder):
(WebKit::WebResourceLoadStatisticsStore::clearInMemory):
(WebKit::WebResourceLoadStatisticsStore::mergeStatistics):
(WebKit::WebResourceLoadStatisticsStore::shouldPartitionCookies):
(WebKit::WebResourceLoadStatisticsStore::updateCookiePartitioning):
(WebKit::WebResourceLoadStatisticsStore::updateCookiePartitioningForDomains):
(WebKit::WebResourceLoadStatisticsStore::processStatistics):
(WebKit::WebResourceLoadStatisticsStore::hasHadUnexpiredRecentUserInteraction):
(WebKit::WebResourceLoadStatisticsStore::topPrivatelyControlledDomainsToRemoveWebsiteDataFor):
(WebKit::WebResourceLoadStatisticsStore::includeTodayAsOperatingDateIfNecessary):
(WebKit::WebResourceLoadStatisticsStore::hasStatisticsExpired):
* UIProcess/WebResourceLoadStatisticsStore.h:
* UIProcess/WebResourceLoadStatisticsTelemetry.cpp:
(WebKit::sortedPrevalentResourceTelemetry):
(WebKit::WebResourceLoadStatisticsTelemetry::calculateAndSubmit):
* UIProcess/WebResourceLoadStatisticsTelemetry.h:
* UIProcess/WebsiteData/WebsiteDataStore.cpp:
(WebKit::WebsiteDataStore::removeData):
* WebKit2.xcodeproj/project.pbxproj:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@219312 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoUpdate webrtc expectations
commit-queue@webkit.org [Mon, 10 Jul 2017 23:00:02 +0000 (23:00 +0000)]
Update webrtc expectations
https://bugs.webkit.org/show_bug.cgi?id=174318

Unreviewed.

Updated expectation according flakiness dashboard information.

Patch by Youenn Fablet <youenn@apple.com> on 2017-07-10

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

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@219311 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years ago[WK2] Ignore touch events that interrupt platform-driven momentum scrolling
wenson_hsieh@apple.com [Mon, 10 Jul 2017 22:08:38 +0000 (22:08 +0000)]
[WK2] Ignore touch events that interrupt platform-driven momentum scrolling
https://bugs.webkit.org/show_bug.cgi?id=174300
<rdar://problem/33030639>

Reviewed by Simon Fraser.

Source/WebCore:

See Source/WebKit2/ChangeLog for more detail.

Tests: fast/events/ios/no-touch-events-when-stopping-momentum-scroll-in-mainframe.html
       fast/events/ios/no-touch-events-when-stopping-momentum-scroll-in-overflow.html
       fast/events/ios/touch-events-during-scroll-deceleration-in-overflow.html

* page/scrolling/ScrollingTree.h:
(WebCore::ScrollingTree::scrollingTreeNodeDidEndPanGesture):

Source/WebKit2:

Adds support for immediately failing WKContentView's UIWebTouchEventsGestureRecognizer if any touch interrupts
scroll view deceleration. Covered by 3 new LayoutTests (see Source/WebCore/ChangeLog).

* Platform/spi/ios/UIKitSPI.h:
* UIProcess/API/Cocoa/WKWebView.mm:
(-[WKWebView scrollViewDidEndDragging:willDecelerate:]):
(-[WKWebView scrollViewDidEndZooming:withView:atScale:]):
* UIProcess/PageClient.h:
* UIProcess/Scrolling/RemoteScrollingCoordinatorProxy.h:
* UIProcess/Scrolling/RemoteScrollingTree.cpp:
(WebKit::RemoteScrollingTree::scrollingTreeNodeDidEndPanGesture):
* UIProcess/Scrolling/RemoteScrollingTree.h:
* UIProcess/Scrolling/ios/ScrollingTreeOverflowScrollingNodeIOS.h:
* UIProcess/Scrolling/ios/ScrollingTreeOverflowScrollingNodeIOS.mm:
(-[WKOverflowScrollViewDelegate scrollViewDidEndDragging:willDecelerate:]):
(WebKit::ScrollingTreeOverflowScrollingNodeIOS::overflowScrollViewDidEndPanGesture):

Adds plumbing in the UI process to alert the WKContentView when scroll view dragging has ended in an overflow
momentum scrolling region (tracked by an overflow scrolling tree node). This parallels UI process plumbing for
ScrollingTreeOverflowScrollingNodeIOS::overflowScrollViewWillStartPanGesture.

* UIProcess/WebPageProxy.h:
* UIProcess/ios/PageClientImplIOS.h:
* UIProcess/ios/PageClientImplIOS.mm:
(WebKit::PageClientImpl::overflowScrollViewDidEndPanGesture):
* UIProcess/ios/RemoteScrollingCoordinatorProxyIOS.mm:
(WebKit::RemoteScrollingCoordinatorProxy::scrollingTreeNodeDidEndPanGesture):
* UIProcess/ios/WKContentViewInteraction.h:
* UIProcess/ios/WKContentViewInteraction.mm:
(-[WKContentView scrollViewDidEndPanOrPinchGesture]):

When the dragging phase of a scroll gesture ends, ensure that _canSendTouchEventsAsynchronously is reset to NO.
When touches are not being prevented, this flag is set to YES in -scrollViewWillStartPanOrPinchGesture: and
normally reset to NO in -_webTouchEventsRecognized: after all touch points have been released. However, in the
case where we've prematurely failed the web touch events gesture recognizer after -touchesBegan:,
-_webTouchEventsRecognized: will not be called again in a state where all touches have been released. Thus, we
also need to bookend the place where we begin allowing async touch event dispatch in
-scrollViewWillStartPanOrPinchGesture by disallowing async touch event dispatch in
-scrollViewDidEndPanOrPinchGesture.

This has no impact in the normal case where the user scrolls without momentum, since
-scrollViewDidEndPanOrPinchGesture will fire after async touch events have already been disallowed in
-_webTouchEventsRecognized:, and in the case where the page prevents default on touches, scrolling doesn't
happen anyways, so neither scrollViewWillStartPanOrPinchGesture nor scrollViewDidEndPanOrPinchGesture will be
invoked.

(-[WKContentView gestureRecognizer:shouldIgnoreWebTouchWithEvent:]):

Implements a new SPI hook in UIKit. This serves an identical purpose as -shouldIgnoreWebTouch, but additionally
passes the triggering UIEvent, as well as the gesture recognizer itself (following standard convention for
Objective C delegate methods). Here, we return YES in the case where one or more of the touches recognized by
the gesture is currently interrupting scroll deceleration.

* UIProcess/ios/WebPageProxyIOS.mm:
(WebKit::WebPageProxy::overflowScrollViewDidEndPanGesture):

LayoutTests:

Adds 3 new LayoutTests covering cases of touch event handling during scroll deceleration in mainframe and
overflow scrolling containers. See Source/WebKit2/ChangeLog for more detail.

* fast/events/ios/no-touch-events-when-stopping-momentum-scroll-in-mainframe-expected.txt: Added.
* fast/events/ios/no-touch-events-when-stopping-momentum-scroll-in-mainframe.html: Added.
* fast/events/ios/no-touch-events-when-stopping-momentum-scroll-in-overflow-expected.txt: Added.
* fast/events/ios/no-touch-events-when-stopping-momentum-scroll-in-overflow.html: Added.
* fast/events/ios/touch-events-during-scroll-deceleration-in-overflow-expected.txt: Added.
* fast/events/ios/touch-events-during-scroll-deceleration-in-overflow.html: Added.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@219310 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoMarked animations/resume-after-page-cache.html as flaky.
jlewis3@apple.com [Mon, 10 Jul 2017 21:31:50 +0000 (21:31 +0000)]
Marked animations/resume-after-page-cache.html as flaky.
https://bugs.webkit.org/show_bug.cgi?id=174183

Unreviewed test gardening.

* platform/ios-simulator-wk2/TestExpectations:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@219309 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years ago[WK2][macOS] Allow access to com.apple.cfnetwork.cfnetworkagent in the Network Process
bfulgham@apple.com [Mon, 10 Jul 2017 21:16:40 +0000 (21:16 +0000)]
[WK2][macOS] Allow access to com.apple.cfnetwork.cfnetworkagent in the Network Process
https://bugs.webkit.org/show_bug.cgi?id=174320
<rdar://problem/33191856>

Reviewed by Geoffrey Garen.

* NetworkProcess/mac/com.apple.WebKit.NetworkProcess.sb.in:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@219308 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoCaptions and subtitles not showing up in picture-in-picture for MSE content
commit-queue@webkit.org [Mon, 10 Jul 2017 20:58:32 +0000 (20:58 +0000)]
Captions and subtitles not showing up in picture-in-picture for MSE content
https://bugs.webkit.org/show_bug.cgi?id=174317
rdar://problem/33188591

Patch by Jeremy Jones <jeremyj@apple.com> on 2017-07-10
Reviewed by Eric Carlson.

Reverts a regression created by r218403.

* html/shadow/MediaControlElements.cpp:
(WebCore::MediaControlTextTrackContainerElement::updateTextTrackRepresentation):

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@219307 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years ago[Win] Link error when building WTF from WTF.proj project file.
pvollan@apple.com [Mon, 10 Jul 2017 20:09:04 +0000 (20:09 +0000)]
[Win] Link error when building WTF from WTF.proj project file.
https://bugs.webkit.org/show_bug.cgi?id=174316
<rdar://problem/33178200>

Reviewed by Brent Fulgham.

WTF_CPU_X86 cmake variable needs to be set for link libraries directories to be correct.

Source/WebCore:

* WebCore.vcxproj/WebCore.proj:

Source/WebKit:

* WebKit.vcxproj/WebKit.proj:

Source/WTF:

* WTF.vcxproj/WTF.proj:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@219306 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agomedia element handle adding source immediately before src.
commit-queue@webkit.org [Mon, 10 Jul 2017 19:49:29 +0000 (19:49 +0000)]
media element handle adding source immediately before src.
https://bugs.webkit.org/show_bug.cgi?id=174284
rdar://problem/33115439

Patch by Jeremy Jones <jeremyj@apple.com> on 2017-07-10
Reviewed by David Kilzer.

Source/WebCore:

Test: media/video-source-before-src.html

Adding a source causes a selectMediaResource block to be enqueued.
If dataLoadingPermitted prevents creating the m_player but sets the srcAttr, then
the enqueued selectMediaResource will be in a bad state, with a srcAttr but no m_player.

This fix prevents selectMediaResource from being called, if data loading is not permitted
when adding a source element, to match how it prevents player creation when setting srcAttr.

This fix also adds a debug assert to catch the problem earlier and adds an early return to
prevent the crash in release builds.

* html/HTMLMediaElement.cpp:
(WebCore::HTMLMediaElement::selectMediaResource):
(WebCore::HTMLMediaElement::sourceWasAdded):

LayoutTests:

* media/video-source-before-src.html: Added.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@219305 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoAdd location to NavigationActionData
megan_gardner@apple.com [Mon, 10 Jul 2017 19:46:01 +0000 (19:46 +0000)]
Add location to NavigationActionData
https://bugs.webkit.org/show_bug.cgi?id=174233
<rdar://problem/29165518>

Reviewed by Simon Fraser.

Add the root view location of a tap to a NavigationAction to vend to Safari.

Source/WebCore:

Test: small enough change to not be tested alone.

* dom/MouseRelatedEvent.cpp:
(WebCore::MouseRelatedEvent::absoluteLocationConvertedToRootView):
* dom/MouseRelatedEvent.h:

Source/WebKit2:

* Shared/NavigationActionData.cpp:
(WebKit::NavigationActionData::encode):
(WebKit::NavigationActionData::decode):
* Shared/NavigationActionData.h:
* UIProcess/API/APINavigationAction.h:
* UIProcess/API/Cocoa/WKNavigationAction.mm:
(-[WKNavigationAction description]):
(-[WKNavigationAction _clickLocationInRootViewCoordinates]):
* UIProcess/API/Cocoa/WKNavigationActionPrivate.h:
* WebProcess/InjectedBundle/InjectedBundleNavigationAction.cpp:
(WebKit::clickLocationInRootViewCoordinatesForMouseEvent):
(WebKit::InjectedBundleNavigationAction::clickLocationInRootViewCoordinatesForNavigationAction):
(WebKit::InjectedBundleNavigationAction::InjectedBundleNavigationAction):
* WebProcess/InjectedBundle/InjectedBundleNavigationAction.h:
(WebKit::InjectedBundleNavigationAction::clickLocationInRootViewCoordinates):
* WebProcess/WebCoreSupport/WebChromeClient.cpp:
(WebKit::WebChromeClient::createWindow):
* WebProcess/WebCoreSupport/WebFrameLoaderClient.cpp:
(WebKit::WebFrameLoaderClient::dispatchDecidePolicyForNewWindowAction):
(WebKit::WebFrameLoaderClient::dispatchDecidePolicyForNavigationAction):

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@219304 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoFixed test expectations for http/tests/loading/resourceLoadStatistics/user-interactio...
jlewis3@apple.com [Mon, 10 Jul 2017 19:04:10 +0000 (19:04 +0000)]
Fixed test expectations for http/tests/loading/resourceLoadStatistics/user-interaction-in-cross-origin-sub-frame.html.
https://bugs.webkit.org/show_bug.cgi?id=174120

Unreviewed test gardening.

* platform/ios-wk2/TestExpectations:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@219303 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years ago[WebIDL] Move plugin object customization into the generator
commit-queue@webkit.org [Mon, 10 Jul 2017 18:27:26 +0000 (18:27 +0000)]
[WebIDL] Move plugin object customization into the generator
https://bugs.webkit.org/show_bug.cgi?id=174238

Patch by Sam Weinig <sam@webkit.org> on 2017-07-10
Reviewed by Chris Dumez.

- Added [Plugin] extended attribute to forward the necessary hooks
  for get/set/delete to the plugin code.
- Removed [CustomNamedSetter] and replaced it's remaining uses
  [CustomPut] (formally called [CustomPutFunction]).
- Renamed [CustomNamedGetterOnPrototype] to [CustomPutOnPrototype]
  because that is actually what it does.
- Removed [CustomGetOwnPropertySlotByIndex] and made
  [CustomGetOwnPropertySlot] imply it, as the other custom hooks
  do.
- Renamed [CustomEnumerateProperty] to [CustomGetOwnPropertyNames]
  to conform with other attribute names.
- Renamed [CustomCall] to [CustomGetCallData] to conform with other
  attribute names.

* CMakeLists.txt:
* WebCore.xcodeproj/project.pbxproj:
* bindings/js/JSBindingsAllInOne.cpp:
* bindings/js/JSHTMLAppletElementCustom.cpp: Removed.
* bindings/js/JSHTMLEmbedElementCustom.cpp: Removed.
* bindings/js/JSHTMLObjectElementCustom.cpp: Removed.
Remove custom bindings.

* bindings/js/JSCSSStyleDeclarationCustom.cpp:
(WebCore::putCommon):
(WebCore::JSCSSStyleDeclaration::put):
(WebCore::JSCSSStyleDeclaration::putByIndex):
(WebCore::JSCSSStyleDeclaration::putDelegate): Deleted.
Use [CustomPut] rather than [CustomNamedSetter] to allow us
to get rid of [CustomNamedSetter]. Reuse put delegate as
common code to share between put and putByIndex.

* bindings/js/JSLocationCustom.cpp:
(WebCore::getOwnPropertySlotCommon):
(WebCore::JSLocation::getOwnPropertySlot):
(WebCore::JSLocation::getOwnPropertySlotByIndex):
Replace [CustomGetOwnPropertySlotAndDescriptor] with [CustomGetOwnPropertySlot]
which is more clear and reduces the number of variants of this hook override
we need.

(WebCore::putCommon):
(WebCore::JSLocation::put):
(WebCore::JSLocation::putByIndex):
Use [CustomPut] rather than [CustomNamedSetter] to allow us
to get rid of [CustomNamedSetter]. Reuse put delegate as
common code to share between put and putByIndex.

(WebCore::JSLocationPrototype::put):
[CustomPutOnPrototype] (which weirdly used incorrectly be called
[CustomNamedGetterOnPrototype]) now works like [CustomPut] meaning
you need to call Base.

* bindings/js/JSPluginElementFunctions.cpp:
(WebCore::pluginElementPropertyGetter):
(WebCore::pluginElementCustomGetOwnPropertySlot):
(WebCore::pluginElementCustomPut):
* bindings/js/JSPluginElementFunctions.h:
(WebCore::pluginElementCustomGetOwnPropertySlot): Deleted.
Remove templatized pluginElementCustomGetOwnPropertySlot, which was
completely unnecessary and merge its functionality into the out of
line overload. Remove pluginElementPropertyGetter from the header,
since it is only used in implementation, and unify the naming and
argument position (JSHTMLElement* comes first) of the hooks.

* bindings/scripts/CodeGeneratorJS.pm:
(GenerateIndexedGetter):
(GenerateNamedGetter):
(GenerateGetOwnPropertySlot):
(GenerateGetOwnPropertySlotByIndex):
(GenerateGetOwnPropertyNames):
(GeneratePut):
(GeneratePutByIndex):
(GenerateDeleteProperty):
(GenerateDeletePropertyByIndex):
(GenerateNamedDeleterDefinition):
(InstanceOverridesGetOwnPropertySlot):
(InstanceOverridesGetOwnPropertyNames):
(InstanceOverridesPut):
(InstanceOverridesDeleteProperty):
(GenerateHeader):
(GenerateImplementation):
(GenerateGetCallData):
(GeneratePluginCall):
(GenerateLegacyCallerDefinitions):
(GenerateLegacyCallerDefinition):
(GeneratePrototypeDeclaration):
(InstanceOverridesGetCallData):
(HeaderNeedsPrototypeDeclaration):
- Add support for [Plugin]
- Remove support for [CustomNamedSetter]
- Replace [CustomGetOwnPropertySlotByIndex] with [CustomGetOwnPropertySlot]
- Replace [CustomEnumerateProperty] with [CustomGetOwnPropertyNames]
- Replace [CustomPutFunction] with [CustomPut].
- Make subroutine names more consistent (remove a few Definition suffixes)

* bindings/scripts/IDLAttributes.json:
Update for new / removed attributes.

* bindings/scripts/test/JS/JSTestInterface.cpp:
* bindings/scripts/test/JS/JSTestInterface.h:
* bindings/scripts/test/JS/JSTestPluginInterface.cpp: Added.
* bindings/scripts/test/JS/JSTestPluginInterface.h: Added.
* bindings/scripts/test/TestInterface.idl:
* bindings/scripts/test/TestPluginInterface.idl: Added.
Update / add tests.

* css/CSSStyleDeclaration.idl:
* html/HTMLAppletElement.idl:
* html/HTMLEmbedElement.idl:
* html/HTMLObjectElement.idl:
* page/DOMWindow.idl:
* page/Location.idl:
* storage/Storage.idl:
Update for new / renamed attributes.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@219302 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoWeb Replay: remove some unused code
bburg@apple.com [Mon, 10 Jul 2017 18:24:26 +0000 (18:24 +0000)]
Web Replay: remove some unused code
https://bugs.webkit.org/show_bug.cgi?id=173903

Rubber-stamped by Joseph Pecoraro.

.:

* ManualTests/inspector/replay-keyboard-events.html: Removed.
* ManualTests/inspector/replay-mouse-events.html: Removed.
* ManualTests/inspector/replay-wheel-events.html: Removed.
* ManualTests/inspector/replay-window-navigator-basic.html: Removed.
* ManualTests/inspector/replay-window-screen.html: Removed.
* Source/cmake/OptionsMac.cmake:
* Source/cmake/WebKitFeatures.cmake:

Source/JavaScriptCore:

* CMakeLists.txt:
* Configurations/FeatureDefines.xcconfig:
* DerivedSources.make:
* JavaScriptCore.xcodeproj/project.pbxproj:
* inspector/protocol/Replay.json: Removed.
* replay/EmptyInputCursor.h: Removed.
* replay/EncodedValue.cpp: Removed.
* replay/EncodedValue.h: Removed.
* replay/InputCursor.h: Removed.
* replay/JSInputs.json: Removed.
* replay/NondeterministicInput.h: Removed.
* replay/scripts/CodeGeneratorReplayInputs.py: Removed.
* replay/scripts/CodeGeneratorReplayInputsTemplates.py: Removed.
* replay/scripts/tests/expected/fail-on-c-style-enum-no-storage.json-error: Removed.
* replay/scripts/tests/expected/fail-on-duplicate-enum-type.json-error: Removed.
* replay/scripts/tests/expected/fail-on-duplicate-input-names.json-error: Removed.
* replay/scripts/tests/expected/fail-on-duplicate-type-names.json-error: Removed.
* replay/scripts/tests/expected/fail-on-enum-type-missing-values.json-error: Removed.
* replay/scripts/tests/expected/fail-on-missing-input-member-name.json-error: Removed.
* replay/scripts/tests/expected/fail-on-missing-input-name.json-error: Removed.
* replay/scripts/tests/expected/fail-on-missing-input-queue.json-error: Removed.
* replay/scripts/tests/expected/fail-on-missing-type-mode.json-error: Removed.
* replay/scripts/tests/expected/fail-on-missing-type-name.json-error: Removed.
* replay/scripts/tests/expected/fail-on-unknown-input-queue.json-error: Removed.
* replay/scripts/tests/expected/fail-on-unknown-member-type.json-error: Removed.
* replay/scripts/tests/expected/fail-on-unknown-type-mode.json-error: Removed.
* replay/scripts/tests/expected/generate-enum-encoding-helpers-with-guarded-values.json-TestReplayInputs.cpp: Removed.
* replay/scripts/tests/expected/generate-enum-encoding-helpers-with-guarded-values.json-TestReplayInputs.h: Removed.
* replay/scripts/tests/expected/generate-enum-encoding-helpers.json-TestReplayInputs.cpp: Removed.
* replay/scripts/tests/expected/generate-enum-encoding-helpers.json-TestReplayInputs.h: Removed.
* replay/scripts/tests/expected/generate-enum-with-guard.json-TestReplayInputs.cpp: Removed.
* replay/scripts/tests/expected/generate-enum-with-guard.json-TestReplayInputs.h: Removed.
* replay/scripts/tests/expected/generate-enums-with-same-base-name.json-TestReplayInputs.cpp: Removed.
* replay/scripts/tests/expected/generate-enums-with-same-base-name.json-TestReplayInputs.h: Removed.
* replay/scripts/tests/expected/generate-event-loop-shape-types.json-error: Removed.
* replay/scripts/tests/expected/generate-input-with-guard.json-TestReplayInputs.cpp: Removed.
* replay/scripts/tests/expected/generate-input-with-guard.json-TestReplayInputs.h: Removed.
* replay/scripts/tests/expected/generate-input-with-vector-members.json-TestReplayInputs.cpp: Removed.
* replay/scripts/tests/expected/generate-input-with-vector-members.json-TestReplayInputs.h: Removed.
* replay/scripts/tests/expected/generate-inputs-with-flags.json-TestReplayInputs.cpp: Removed.
* replay/scripts/tests/expected/generate-inputs-with-flags.json-TestReplayInputs.h: Removed.
* replay/scripts/tests/expected/generate-memoized-type-modes.json-TestReplayInputs.cpp: Removed.
* replay/scripts/tests/expected/generate-memoized-type-modes.json-TestReplayInputs.h: Removed.
* replay/scripts/tests/fail-on-c-style-enum-no-storage.json: Removed.
* replay/scripts/tests/fail-on-duplicate-enum-type.json: Removed.
* replay/scripts/tests/fail-on-duplicate-input-names.json: Removed.
* replay/scripts/tests/fail-on-duplicate-type-names.json: Removed.
* replay/scripts/tests/fail-on-enum-type-missing-values.json: Removed.
* replay/scripts/tests/fail-on-missing-input-member-name.json: Removed.
* replay/scripts/tests/fail-on-missing-input-name.json: Removed.
* replay/scripts/tests/fail-on-missing-input-queue.json: Removed.
* replay/scripts/tests/fail-on-missing-type-mode.json: Removed.
* replay/scripts/tests/fail-on-missing-type-name.json: Removed.
* replay/scripts/tests/fail-on-unknown-input-queue.json: Removed.
* replay/scripts/tests/fail-on-unknown-member-type.json: Removed.
* replay/scripts/tests/fail-on-unknown-type-mode.json: Removed.
* replay/scripts/tests/generate-enum-encoding-helpers-with-guarded-values.json: Removed.
* replay/scripts/tests/generate-enum-encoding-helpers.json: Removed.
* replay/scripts/tests/generate-enum-with-guard.json: Removed.
* replay/scripts/tests/generate-enums-with-same-base-name.json: Removed.
* replay/scripts/tests/generate-event-loop-shape-types.json: Removed.
* replay/scripts/tests/generate-input-with-guard.json: Removed.
* replay/scripts/tests/generate-input-with-vector-members.json: Removed.
* replay/scripts/tests/generate-inputs-with-flags.json: Removed.
* replay/scripts/tests/generate-memoized-type-modes.json: Removed.
* runtime/DateConstructor.cpp:
(JSC::constructDate):
(JSC::dateNow):
(JSC::deterministicCurrentTime): Deleted.
* runtime/JSGlobalObject.cpp:
(JSC::JSGlobalObject::JSGlobalObject):
(JSC::JSGlobalObject::setInputCursor): Deleted.
* runtime/JSGlobalObject.h:
(JSC::JSGlobalObject::inputCursor): Deleted.

Source/WebCore:

* CMakeLists.txt:
* Configurations/FeatureDefines.xcconfig:
* DerivedSources.make:
* PlatformMac.cmake:
* WebCore.xcodeproj/project.pbxproj:
* dom/Document.cpp:
(WebCore::Document::Document):
(WebCore::Document::lastModified):
(WebCore::Document::inputCursor): Deleted.
(WebCore::Document::setInputCursor): Deleted.
* dom/Document.h:
* inspector/InspectorController.cpp:
(WebCore::InspectorController::InspectorController):
* inspector/InspectorInstrumentation.cpp:
(WebCore::InspectorInstrumentation::willDispatchEventImpl):
(WebCore::InspectorInstrumentation::willDispatchEventOnWindowImpl):
(WebCore::InspectorInstrumentation::frameDetachedFromParentImpl):
(WebCore::InspectorInstrumentation::didCommitLoadImpl):
(WebCore::InspectorInstrumentation::sessionCreatedImpl): Deleted.
(WebCore::InspectorInstrumentation::sessionLoadedImpl): Deleted.
(WebCore::InspectorInstrumentation::sessionModifiedImpl): Deleted.
(WebCore::InspectorInstrumentation::segmentCreatedImpl): Deleted.
(WebCore::InspectorInstrumentation::segmentCompletedImpl): Deleted.
(WebCore::InspectorInstrumentation::segmentLoadedImpl): Deleted.
(WebCore::InspectorInstrumentation::segmentUnloadedImpl): Deleted.
(WebCore::InspectorInstrumentation::captureStartedImpl): Deleted.
(WebCore::InspectorInstrumentation::captureStoppedImpl): Deleted.
(WebCore::InspectorInstrumentation::playbackStartedImpl): Deleted.
(WebCore::InspectorInstrumentation::playbackPausedImpl): Deleted.
(WebCore::InspectorInstrumentation::playbackHitPositionImpl): Deleted.
(WebCore::InspectorInstrumentation::playbackFinishedImpl): Deleted.
(WebCore::InspectorInstrumentation::replayAgentEnabled): Deleted.
* inspector/InspectorInstrumentation.h:
(WebCore::InspectorInstrumentation::sessionCreated): Deleted.
(WebCore::InspectorInstrumentation::sessionLoaded): Deleted.
(WebCore::InspectorInstrumentation::sessionModified): Deleted.
(WebCore::InspectorInstrumentation::segmentCreated): Deleted.
(WebCore::InspectorInstrumentation::segmentCompleted): Deleted.
(WebCore::InspectorInstrumentation::segmentLoaded): Deleted.
(WebCore::InspectorInstrumentation::segmentUnloaded): Deleted.
(WebCore::InspectorInstrumentation::captureStarted): Deleted.
(WebCore::InspectorInstrumentation::captureStopped): Deleted.
(WebCore::InspectorInstrumentation::playbackStarted): Deleted.
(WebCore::InspectorInstrumentation::playbackPaused): Deleted.
(WebCore::InspectorInstrumentation::playbackFinished): Deleted.
(WebCore::InspectorInstrumentation::playbackHitPosition): Deleted.
* inspector/InspectorReplayAgent.cpp: Removed.
* inspector/InspectorReplayAgent.h: Removed.
* inspector/InstrumentingAgents.cpp:
(WebCore::InstrumentingAgents::reset):
* inspector/InstrumentingAgents.h:
(WebCore::InstrumentingAgents::inspectorReplayAgent): Deleted.
(WebCore::InstrumentingAgents::setInspectorReplayAgent): Deleted.
* page/EventHandler.h:
* page/Page.cpp:
(WebCore::Page::Page):
* page/Page.h:
(WebCore::Page::replayController): Deleted.
* page/scrolling/ScrollingCoordinator.cpp:
(WebCore::ScrollingCoordinator::synchronousScrollingReasons):
(WebCore::ScrollingCoordinator::replaySessionStateDidChange): Deleted.
* page/scrolling/ScrollingCoordinator.h:
* platform/Logging.h:
* plugins/DOMMimeTypeArray.cpp:
(WebCore::DOMMimeTypeArray::getPluginData):
* plugins/DOMPluginArray.cpp:
(WebCore::DOMPluginArray::pluginData):
* replay/AllReplayInputs.h: Removed.
* replay/CapturingInputCursor.cpp: Removed.
* replay/CapturingInputCursor.h: Removed.
* replay/EventLoopInput.cpp: Removed.
* replay/EventLoopInput.h: Removed.
* replay/EventLoopInputDispatcher.cpp: Removed.
* replay/EventLoopInputDispatcher.h: Removed.
* replay/FunctorInputCursor.h: Removed.
* replay/MemoizedDOMResult.cpp: Removed.
* replay/MemoizedDOMResult.h: Removed.
* replay/ReplayController.cpp: Removed.
* replay/ReplayController.h: Removed.
* replay/ReplayInputCreationMethods.cpp: Removed.
* replay/ReplayInputDispatchMethods.cpp: Removed.
* replay/ReplaySession.cpp: Removed.
* replay/ReplaySession.h: Removed.
* replay/ReplaySessionSegment.cpp: Removed.
* replay/ReplaySessionSegment.h: Removed.
* replay/ReplayingInputCursor.cpp: Removed.
* replay/ReplayingInputCursor.h: Removed.
* replay/SegmentedInputStorage.cpp: Removed.
* replay/SegmentedInputStorage.h: Removed.
* replay/SerializationMethods.cpp: Removed.
* replay/SerializationMethods.h: Removed.
* replay/WebInputs.json: Removed.

Source/WebCore/PAL:

* Configurations/FeatureDefines.xcconfig:

Source/WebInspectorUI:

* .eslintrc:
* UserInterface/Base/Main.js:
(WebInspector.loaded):
* UserInterface/Controllers/DashboardManager.js:
(WebInspector.DashboardManager):
* UserInterface/Controllers/ReplayManager.js: Removed.
* UserInterface/Images/ReplayPauseButton.svg: Removed.
* UserInterface/Images/ReplayPlayButton.svg: Removed.
* UserInterface/Images/ReplayRecordingButton.svg: Removed.
* UserInterface/Images/gtk/ReplayPauseButton.svg: Removed.
* UserInterface/Images/gtk/ReplayPlayButton.svg: Removed.
* UserInterface/Images/gtk/ReplayRecordingButton.svg: Removed.
* UserInterface/Main.html:
* UserInterface/Models/ReplayDashboard.js: Removed.
* UserInterface/Models/ReplaySession.js: Removed.
* UserInterface/Models/ReplaySessionSegment.js: Removed.
* UserInterface/Protocol/ReplayObserver.js: Removed.
* UserInterface/Test.html:
* UserInterface/Test/Test.js:
(WebInspector.loaded):
* UserInterface/Views/DashboardView.js:
(WebInspector.DashboardView.create):
* UserInterface/Views/ReplayDashboardView.css: Removed.
* UserInterface/Views/ReplayDashboardView.js: Removed.
* UserInterface/Views/TimelineRecordingContentView.js:
(WebInspector.TimelineRecordingContentView.prototype._debuggerPaused):
(WebInspector.TimelineRecordingContentView.prototype._debuggerResumed):

Source/WebKit/mac:

* Configurations/FeatureDefines.xcconfig:

Source/WebKit2:

* Configurations/FeatureDefines.xcconfig:
* WebProcess/WebPage/WebPage.cpp:
(WebKit::WebPage::tryClose):
(WebKit::WebPage::loadRequest):
(WebKit::WebPage::stopLoadingFrame):
(WebKit::WebPage::stopLoading):
(WebKit::WebPage::reload):
(WebKit::WebPage::contextMenuAtPointInWindow):
(WebKit::handleContextMenuEvent):
(WebKit::handleMouseEvent):
(WebKit::handleWheelEvent):
(WebKit::handleKeyEvent):
(WebKit::WebPage::scroll):
(WebKit::WebPage::logicalScroll):

Source/WTF:

* wtf/FeatureDefines.h:

Tools:

* Scripts/run-input-generator-tests: Removed.
* Scripts/webkitpy/replay/__init__.py: Removed.
* Scripts/webkitpy/replay/main.py: Removed.

LayoutTests:

* inspector/replay/javascript-date-now-expected.txt: Removed.
* inspector/replay/javascript-date-now.html: Removed.
* inspector/replay/javascript-random-seed-expected.txt: Removed.
* inspector/replay/javascript-random-seed.html: Removed.
* inspector/replay/window-navigator-plugins-memoized-expected.txt: Removed.
* inspector/replay/window-navigator-plugins-memoized.html: Removed.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@219301 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years ago[WPE] Fix layout test baseline and TestExpectations hierarchy
zandobersek@gmail.com [Mon, 10 Jul 2017 18:07:10 +0000 (18:07 +0000)]
[WPE] Fix layout test baseline and TestExpectations hierarchy
https://bugs.webkit.org/show_bug.cgi?id=174092

Reviewed by Michael Catanzaro.

Like for the GTK+ port, override the default_baseline_search_path() and
_port_specific_expectations_files() methods in the webkitpy.port.WPEPort class.
Both methods operate on a short list of search paths that defaults to 'wpe' and
'wk2', along with any other user-specified platform directory.

The test_expectations_file_position() method is overridden for testing purposes,
and relevant unit tests are added.

* Scripts/webkitpy/port/wpe.py:
(WPEPort._search_paths):
(WPEPort):
(WPEPort.default_baseline_search_path):
(WPEPort._port_specific_expectations_files):
(WPEPort.test_expectations_file_position):
* Scripts/webkitpy/port/wpe_unittest.py:
(WPEPortTest.test_default_baseline_search_path):
(WPEPortTest):
(WPEPortTest.test_port_specific_expectations_files):

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@219300 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoSkipped http/tests/loading/resourceLoadStatistics/user-interaction-in-cross-origin...
jlewis3@apple.com [Mon, 10 Jul 2017 17:52:06 +0000 (17:52 +0000)]
Skipped http/tests/loading/resourceLoadStatistics/user-interaction-in-cross-origin-sub-frame.html.
https://bugs.webkit.org/show_bug.cgi?id=174120

Unreviewed test gardening.

* platform/ios/TestExpectations:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@219299 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoCleanup lifetime issues of UniqueIDBDatabase and IDBBackingStore.
beidson@apple.com [Mon, 10 Jul 2017 17:43:38 +0000 (17:43 +0000)]
Cleanup lifetime issues of UniqueIDBDatabase and IDBBackingStore.
<rdar://problem/32908525> and https://bugs.webkit.org/show_bug.cgi?id=174244

Reviewed by David Kilzer and Alex Christensen.

Source/WebCore:

No targeted test possible, implicitly covered by all IDB tests.

The original idea behind UniqueIDBDatabase lifetime was that they are ThreadSafeRefCounted and
we take protector Refs when any operation that needs it alive is in flight.

This added variability to their lifetime which made it difficult to enforce a few different
design invariants, namely:
    - UniqueIBDDatabase objects are always created and destroyed only on the main thread.
    - IDBBackingStore objects are always created and destroyed only on the database thread.

This patch removes the ref counting and instead ties UniqueIDBDatabase lifetime to a
std::unique_ptr that is owned by the IDBServer.

Whenever any operations on the UniqueIDBDatabase are in flight it is kept alive by virtue
of that unique_ptr in the IDBServer. Once a UniqueIDBDatabase is completely done with all of
its work, the following happens:
    - On the main thread the IDBServer removes the unique_ptr owning the UniqueIDBDatabase
      from its map.
    - It hands the unique_ptr to the UniqueIDBDatabase itself, which schedules one final
      database thread task.
    - That database thread task is to destroy the IDBBackingStore, kill its message queues,
      and then message back to the main thread for one final task.
    - That main thread task is to release the unique_ptr, resulting in destruction of the
      UniqueIDBDatabase object.

This is safe, predictable, solves the lifetime issues that r218516 originally tried to solve,
and solves the lifetime issues that r218516 introduced.

(This patch also adds many more assertions to cover various design invariants throughout the
lifecycle of a particular UniqueIDBDatabase)

ASSERT that IDBBackingStores are only ever created and destroyed on the background thread:
* Modules/indexeddb/server/IDBBackingStore.h:
(WebCore::IDBServer::IDBBackingStore::~IDBBackingStore):
(WebCore::IDBServer::IDBBackingStore::IDBBackingStore):

Transition UniqueIDBDatabase ownership from a RefPtr to a std::unique_ptr:
* Modules/indexeddb/server/IDBServer.cpp:
(WebCore::IDBServer::IDBServer::getOrCreateUniqueIDBDatabase):
(WebCore::IDBServer::IDBServer::closeAndTakeUniqueIDBDatabase):
(WebCore::IDBServer::IDBServer::closeAndDeleteDatabasesModifiedSince):
(WebCore::IDBServer::IDBServer::closeAndDeleteDatabasesForOrigins):
(WebCore::IDBServer::IDBServer::closeUniqueIDBDatabase): Deleted.
* Modules/indexeddb/server/IDBServer.h:

Make all the other changes mentioned above:
* Modules/indexeddb/server/UniqueIDBDatabase.cpp:
(WebCore::IDBServer::UniqueIDBDatabase::~UniqueIDBDatabase): Bulk up on ASSERTs
(WebCore::IDBServer::UniqueIDBDatabase::openDatabaseConnection):
(WebCore::IDBServer::UniqueIDBDatabase::performUnconditionalDeleteBackingStore):
(WebCore::IDBServer::UniqueIDBDatabase::scheduleShutdownForClose):
(WebCore::IDBServer::UniqueIDBDatabase::shutdownForClose):
(WebCore::IDBServer::UniqueIDBDatabase::didShutdownForClose):
(WebCore::IDBServer::UniqueIDBDatabase::didDeleteBackingStore):
(WebCore::IDBServer::UniqueIDBDatabase::handleCurrentOperation):
(WebCore::IDBServer::UniqueIDBDatabase::performIterateCursor):
(WebCore::IDBServer::UniqueIDBDatabase::performPrefetchCursor):
(WebCore::IDBServer::UniqueIDBDatabase::operationAndTransactionTimerFired):
(WebCore::IDBServer::UniqueIDBDatabase::activateTransactionInBackingStore):
(WebCore::IDBServer::UniqueIDBDatabase::transactionCompleted):
(WebCore::IDBServer::UniqueIDBDatabase::postDatabaseTask):
(WebCore::IDBServer::UniqueIDBDatabase::postDatabaseTaskReply):
(WebCore::IDBServer::UniqueIDBDatabase::executeNextDatabaseTask):
(WebCore::IDBServer::UniqueIDBDatabase::executeNextDatabaseTaskReply):
(WebCore::IDBServer::UniqueIDBDatabase::maybeFinishHardClose):
(WebCore::IDBServer::UniqueIDBDatabase::isDoneWithHardClose):
(WebCore::IDBServer::UniqueIDBDatabase::immediateCloseForUserDelete):
(WebCore::IDBServer::UniqueIDBDatabase::didPerformUnconditionalDeleteBackingStore): Deleted.
* Modules/indexeddb/server/UniqueIDBDatabase.h:
(WebCore::IDBServer::UniqueIDBDatabase::create): Deleted.

Source/WTF:

Add proper "kill" support to CrossThreadQueue, as well as isEmpty() support.

* wtf/CrossThreadQueue.h:
(WTF::CrossThreadQueue<DataType>::append):
(WTF::CrossThreadQueue<DataType>::kill):
(WTF::CrossThreadQueue<DataType>::isKilled):
(WTF::CrossThreadQueue<DataType>::isEmpty):
(WTF::CrossThreadQueue::isKilled): Deleted.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@219298 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoFurther WebResourceLoadStatisticsStore / ResourceLoadStatisticsStore clean up
cdumez@apple.com [Mon, 10 Jul 2017 17:09:38 +0000 (17:09 +0000)]
Further WebResourceLoadStatisticsStore / ResourceLoadStatisticsStore clean up
https://bugs.webkit.org/show_bug.cgi?id=174301

Reviewed by Brent Fulgham.

Source/WebCore:

Moved some generic file system utility functions down to platform's FileSystem.h.

* platform/FileSystem.cpp:
(WebCore::openAndLockFile):
(WebCore::unlockAndCloseFile):
* platform/FileSystem.h:

Source/WebKit2:

General clean up and various simplifications / optimizations to WebResourceLoadStatisticsStore
and ResourceLoadStatisticsStore classes.

* UIProcess/API/Cocoa/WKWebsiteDataStore.mm:
(-[WKWebsiteDataStore _resourceLoadStatisticsUpdateCookiePartitioning]):
(-[WKWebsiteDataStore _resourceLoadStatisticsSetShouldPartitionCookies:forHost:]):
* UIProcess/API/Cocoa/WKWebsiteDataStorePrivate.h:
* UIProcess/Storage/ResourceLoadStatisticsStore.cpp:
(WebKit::ResourceLoadStatisticsStore::create):
(WebKit::ResourceLoadStatisticsStore::ResourceLoadStatisticsStore):
(WebKit::ResourceLoadStatisticsStore::hasHadRecentUserInteraction):
(WebKit::ResourceLoadStatisticsStore::createEncoderFromData):
(WebKit::ResourceLoadStatisticsStore::populateFromDecoder):
(WebKit::ResourceLoadStatisticsStore::clearInMemory):
(WebKit::ResourceLoadStatisticsStore::mergeStatistics):
(WebKit::ResourceLoadStatisticsStore::updateCookiePartitioning):
(WebKit::ResourceLoadStatisticsStore::updateCookiePartitioningForDomains):
(WebKit::ResourceLoadStatisticsStore::processStatistics):
(WebKit::ResourceLoadStatisticsStore::hasHadUnexpiredRecentUserInteraction):
(WebKit::ResourceLoadStatisticsStore::topPrivatelyControlledDomainsToRemoveWebsiteDataFor):
* UIProcess/Storage/ResourceLoadStatisticsStore.h:
* UIProcess/WebResourceLoadStatisticsStore.cpp:
(WebKit::WebResourceLoadStatisticsStore::WebResourceLoadStatisticsStore):
(WebKit::WebResourceLoadStatisticsStore::~WebResourceLoadStatisticsStore):
(WebKit::WebResourceLoadStatisticsStore::removeDataRecords):
(WebKit::WebResourceLoadStatisticsStore::processStatisticsAndDataRecords):
(WebKit::WebResourceLoadStatisticsStore::resourceLoadStatisticsUpdated):
(WebKit::WebResourceLoadStatisticsStore::grandfatherExistingWebsiteData):
(WebKit::WebResourceLoadStatisticsStore::readDataFromDiskIfNeeded):
(WebKit::WebResourceLoadStatisticsStore::refreshFromDisk):
(WebKit::WebResourceLoadStatisticsStore::applicationWillTerminate):
(WebKit::WebResourceLoadStatisticsStore::resourceLogFilePath):
(WebKit::WebResourceLoadStatisticsStore::writeStoreToDisk):
(WebKit::WebResourceLoadStatisticsStore::scheduleOrWriteStoreToDisk):
(WebKit::WebResourceLoadStatisticsStore::writeEncoderToDisk):
(WebKit::WebResourceLoadStatisticsStore::startMonitoringStatisticsStorage):
(WebKit::WebResourceLoadStatisticsStore::platformExcludeFromBackup):
(WebKit::WebResourceLoadStatisticsStore::createDecoderFromDisk):
(WebKit::WebResourceLoadStatisticsStore::performDailyTasks):
(WebKit::WebResourceLoadStatisticsStore::logUserInteraction):
(WebKit::WebResourceLoadStatisticsStore::hasHadUserInteraction):
(WebKit::WebResourceLoadStatisticsStore::updateCookiePartitioning):
(WebKit::WebResourceLoadStatisticsStore::updateCookiePartitioningForDomains):
(WebKit::WebResourceLoadStatisticsStore::clearInMemory):
(WebKit::WebResourceLoadStatisticsStore::clearInMemoryAndPersistent):
(WebKit::WebResourceLoadStatisticsStore::shouldRemoveDataRecords):
(WebKit::WebResourceLoadStatisticsStore::setDataRecordsBeingRemoved):
* UIProcess/WebResourceLoadStatisticsStore.h:
* UIProcess/WebResourceLoadStatisticsTelemetry.cpp:
(WebKit::sortedPrevalentResourceTelemetry):
(WebKit::WebResourceLoadStatisticsTelemetry::calculateAndSubmit):
* UIProcess/WebsiteData/WebsiteDataStore.cpp:
(WebKit::WebsiteDataStore::updateCookiePartitioningForTopPrivatelyOwnedDomains):
(WebKit::WebsiteDataStore::setResourceLoadStatisticsEnabled):
* UIProcess/WebsiteData/WebsiteDataStore.h:

Tools:

Rename some of the testRunner API for clarity and to better match what it calls
internally.

* WebKitTestRunner/InjectedBundle/Bindings/TestRunner.idl:
* WebKitTestRunner/InjectedBundle/TestRunner.cpp:
(WTR::TestRunner::statisticsUpdateCookiePartitioning):
(WTR::TestRunner::statisticsSetShouldPartitionCookiesForHost):
* WebKitTestRunner/InjectedBundle/TestRunner.h:
* WebKitTestRunner/TestController.cpp:
(WTR::TestController::statisticsUpdateCookiePartitioning):
(WTR::TestController::statisticsSetShouldPartitionCookiesForHost):
* WebKitTestRunner/TestController.h:
* WebKitTestRunner/TestInvocation.cpp:
(WTR::TestInvocation::didReceiveSynchronousMessageFromInjectedBundle):
* WebKitTestRunner/cocoa/TestControllerCocoa.mm:
(WTR::TestController::statisticsUpdateCookiePartitioning):
(WTR::TestController::statisticsSetShouldPartitionCookiesForHost):

LayoutTests:

Update existing layout tests to reflect testRunner API changes.

* http/tests/loading/resourceLoadStatistics/partitioned-and-unpartitioned-cookie-deletion.html:
* http/tests/loading/resourceLoadStatistics/partitioned-and-unpartitioned-cookie-with-partitioning-timeout.html:
* http/tests/loading/resourceLoadStatistics/partitioned-cookies-with-and-without-user-interaction.html:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@219297 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoREGRESSION(r210226): Keyboard-focused element not preserved when navigating back...
akling@apple.com [Mon, 10 Jul 2017 17:03:14 +0000 (17:03 +0000)]
REGRESSION(r210226): Keyboard-focused element not preserved when navigating back through page cache, causing multiple elements to have focus
https://bugs.webkit.org/show_bug.cgi?id=174302
<rdar://problem/33204273>

Reviewed by Antti Koivisto.

Source/WebCore:

Don't clear the active/hovered/focused elements when destroying the render tree,
since we might need to reconstruct it later, and would like to remember which
elements those were.

Only the focused state actually stuck when going in and out of the page cache,
but this patch removes all the element pointer clearing for consistency.

Test: fast/history/page-cache-element-state-focused.html

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

LayoutTests:

* fast/history/page-cache-element-state-focused-expected.html: Added.
* fast/history/page-cache-element-state-focused.html: Added.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@219296 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years ago[SOUP] Remove WebKitSoupCookieJarSqlite
mcatanzaro@igalia.com [Mon, 10 Jul 2017 16:23:19 +0000 (16:23 +0000)]
[SOUP] Remove WebKitSoupCookieJarSqlite
https://bugs.webkit.org/show_bug.cgi?id=174310

Reviewed by Carlos Garcia Campos.

It seems WebKitSoupCookieJarSqlite uses exactly the same database format as the upstream
SoupCookieJarDb class. WebKitSoupCookieJarSqlite is a copy of SoupCookieJarSqlite that uses
WebCore's SQLite wrapper and WTF stuff. WebKitSoupCookieJarSqlite exists only because
SoupCookieJarDb is relatively new, and the older SoupCookieJarSqlite class existed only in
libsoup-gnome rather than libsoup proper. The advantage of continuing to use our
WebKitSoupCookieJarSqlite class, besides that we already know it works, is that it utilizes
the same SQLiteDatabase class that is used elsewhere in WebKit. But that's not really any
reason to keep the class around, when SoupCookieJarDb does exactly the same thing. It was
introduced in libsoup 2.42 and that is already our minimum required version, so we can
switch now.

This passes a quick sanity-check of Epiphany's cookies dialog and also TestCookieManager.

* PlatformGTK.cmake:
* PlatformWPE.cmake:
* WebProcess/Cookies/soup/WebCookieManagerSoup.cpp:
(WebKit::WebCookieManager::setCookiePersistentStorage):
* WebProcess/Cookies/soup/WebKitSoupCookieJarSqlite.cpp: Removed.
* WebProcess/Cookies/soup/WebKitSoupCookieJarSqlite.h: Removed.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@219295 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoREGRESSION (r218616): Cannot build WebCore for macOS 10.12 with macOS 10.13 SDK
dbates@webkit.org [Mon, 10 Jul 2017 16:21:41 +0000 (16:21 +0000)]
REGRESSION (r218616): Cannot build WebCore for macOS 10.12 with macOS 10.13 SDK
https://bugs.webkit.org/show_bug.cgi?id=173939

Reviewed by Dan Bernstein.

(The code in this change was either suggested or written by Dan Bernstein with a very
minor adjustment to get it to build).

Allow WebCore to link even though CTFontCreatePhysicalFontForCharactersWithLanguage() is
undefined when building against the macOS 10.13 SDK targeting macOS 10.12. Let the dynamic
linker resolve the undefined symbol.

For completeness the SPI CTFontCreatePhysicalFontForCharactersWithLanguage() was removed
from the macOS 10.13 SDK.

* Configurations/WebCore.xcconfig: Tell the linker that CTFontCreatePhysicalFontForCharactersWithLanguage()
can be undefined when building against macOS 10.13 or later SDK.
* platform/spi/cocoa/CoreTextSPI.h: Annotate CTFontCreatePhysicalFontForCharactersWithLanguage()
with its availability information.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@219294 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoREGRESSION(r213590): UI process updates the visible content rects more often than...
mmaxfield@apple.com [Mon, 10 Jul 2017 16:17:00 +0000 (16:17 +0000)]
REGRESSION(r213590): UI process updates the visible content rects more often than it did in iOS 10
https://bugs.webkit.org/show_bug.cgi?id=174282
<rdar://problem/33144344>

Reviewed by Simon Fraser.

In r213590, we accidentally refactored -[WKWebView _didCommitLayerTree:] to call _scheduleVisibleContentRectUpdate
more often. Instead, this should be triggered from inside the "if" statements.

No tests because there is no behavior change. Performance change is tested in MotionMark.

* UIProcess/API/Cocoa/WKWebView.mm:
(-[WKWebView _isShowingVideoPictureInPicture]):
(-[WKWebView _initializeWithConfiguration:]):
(-[WKWebView goBack]):
(-[WKWebView _didCommitLayerTree:]):
(-[WKWebView _restorePageStateToUnobscuredCenter:scale:]):
(-[WKWebView _zoomToRect:withOrigin:fitEntireRect:minimumScale:maximumScale:minimumScrollDistance:]):
(-[WKWebView scrollViewWillEndDragging:withVelocity:targetContentOffset:]):
(-[WKWebView scrollViewDidScroll:]):
(-[WKWebView _visibleContentRect]):
(-[WKWebView _frameOrBoundsChanged]):
(-[WKWebView _scheduleVisibleContentRectUpdateAfterScrollInView:]):
(-[WKWebView _updateVisibleContentRects]):
(-[WKWebView _navigationGestureDidBegin]):
(-[WKWebView _reloadWithoutContentBlockers]):
(-[WKWebView _reloadExpiredOnly]):
(-[WKWebView _setObscuredInsets:]):
(-[WKWebView _setUnobscuredSafeAreaInsets:]):
(-[WKWebView _endAnimatedResize]):
(-[WKWebView _snapshotRect:intoImageOfWidth:completionHandler:]):
(-[WKWebView _clearOverrideLayoutParameters]):
(-[WKWebView _requestActivatedElementAtPosition:completionBlock:]):
(-[WKWebView _propertiesOfLayerWithID:]):

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@219293 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years ago[GStreamer] Live twitch.tv videos do not play
commit-queue@webkit.org [Mon, 10 Jul 2017 16:10:18 +0000 (16:10 +0000)]
[GStreamer] Live twitch.tv videos do not play
https://bugs.webkit.org/show_bug.cgi?id=174222

Patch by Charlie Turner <cturner@igalia.com> on 2017-07-10
Reviewed by Michael Catanzaro.

Upgrade GStreamer to 1.10.5. It contains a change to typefind that
fixes videos not playing on twitch.tv.

* gtk/jhbuild.modules:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@219292 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoBlock of text is missing in iBooks sample books.
zalan@apple.com [Mon, 10 Jul 2017 16:03:03 +0000 (16:03 +0000)]
Block of text is missing in iBooks sample books.
https://bugs.webkit.org/show_bug.cgi?id=174295
<rdar://problem/32955620>

Reviewed by Antti Koivisto.

Source/WebCore:

In the simple line layout context, translating y coordinate to a line index is
normally just a (y / line height) operation. However in case of strut offsets (pagination)
we need to take these extra paddings into account while resolving the line index.
This patch fixes the boundary checking for a given line by using the font size only
when the font is taller than the line.

* rendering/SimpleLineLayoutResolver.cpp:
(WebCore::SimpleLineLayout::RunResolver::adjustLineIndexForStruts):

LayoutTests:

* fast/multicol/simple-line-layout-line-index-after-strut-2-expected.html: Added.
* fast/multicol/simple-line-layout-line-index-after-strut-2.html: Added.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@219291 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years ago[SOUP] SoupCookieJar is never released (resulting in sqlite temp files lying around)
carlosgc@webkit.org [Mon, 10 Jul 2017 13:21:35 +0000 (13:21 +0000)]
[SOUP] SoupCookieJar is never released (resulting in sqlite temp files lying around)
https://bugs.webkit.org/show_bug.cgi?id=166029

Reviewed by Michael Catanzaro.

Source/WebCore:

Add clearSoupNetworkSessionAndCookieStorage() to clear the SoupNetworkSession and cookie storage of the main
network session, ensuring the cookies database is properly closed.

* platform/network/NetworkStorageSession.h:
* platform/network/soup/NetworkStorageSessionSoup.cpp:
(WebCore::NetworkStorageSession::clearSoupNetworkSessionAndCookieStorage):

Source/WebKit2:

Clear the SoupNetworkSession and cookie storage after the main loop quits.

* NetworkProcess/soup/NetworkProcessMainSoup.cpp:
(WebKit::NetworkProcessMainUnix):

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@219290 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoMove make-js-file-arrays.py from WebCore to JavaScriptCore
carlosgc@webkit.org [Mon, 10 Jul 2017 13:19:55 +0000 (13:19 +0000)]
Move make-js-file-arrays.py from WebCore to JavaScriptCore
https://bugs.webkit.org/show_bug.cgi?id=174024

Reviewed by Michael Catanzaro.

.:

Make MAKE_JS_FILE_ARRAYS independent of WebCore and update it to use make-js-file-arrays.py from
JavaScriptCore. It's no longer needed to set PYTHON_PATH to find jsmin.py.

* Source/cmake/WebKitMacros.cmake:

Source/JavaScriptCore:

It's currently used only by WebCore, but it depends on other JavaScriptCore scripts and it's not WebCore
specific at all. I plan to use it to compile the JavaScript atoms used by the WebDriver implementation.
Added command line option to pass the namespace to use instead of using WebCore.

* JavaScriptCore.xcodeproj/project.pbxproj:
* Scripts/make-js-file-arrays.py: Renamed from Source/WebCore/Scripts/make-js-file-arrays.py.
(main):

Source/WebCore:

* CMakeLists.txt: Explicitly add files generated by MAKE_JS_FILE_ARRAYS to the build, since the macro no longer
does it.
* DerivedSources.make: Updated to use make-js-file-arrays.py from JavaScriptCore. It's no longer needed to set
PYTHON_PATH to find jsmin.py.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@219289 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years ago[GTK] http/tests/media/video-redirect.html is failing
commit-queue@webkit.org [Mon, 10 Jul 2017 10:13:47 +0000 (10:13 +0000)]
[GTK] http/tests/media/video-redirect.html is failing
https://bugs.webkit.org/show_bug.cgi?id=174260

Patch by Charlie Turner <cturner@igalia.com> on 2017-07-10
Reviewed by Carlos Garcia Campos.

Make sure we're testing new URLs within the same security origin.

Covered by existing tests.

* platform/graphics/gstreamer/MediaPlayerPrivateGStreamer.cpp:
(WebCore::convertToInternalProtocol): Factor out setting our
internal URL schema.
(WebCore::MediaPlayerPrivateGStreamer::setPlaybinURL): Use the
refactored helper.
(WebCore::MediaPlayerPrivateGStreamer::loadNextLocation): Use
refactored helper to ensure both URLs have the same origin.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@219288 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years ago[WK2] Action sheets for links fail to present in WebKit2 PDF view
wenson_hsieh@apple.com [Mon, 10 Jul 2017 10:04:53 +0000 (10:04 +0000)]
[WK2] Action sheets for links fail to present in WebKit2 PDF view
https://bugs.webkit.org/show_bug.cgi?id=174307
<rdar://problem/31412128>

Reviewed by Tim Horton.

Currently, presenting an action sheet for a link always uses the WKActionSheetPresentAtClosestIndicatorRect
codepath, which requires text indicator data for the link. However, when showing an action sheet for a link via
WKPDFView, a text indicator for the link is not included, so the popover rect ends up being an empty rect at the
origin, which causes us to bail from presenting the popover.

To address this, we tweak our heuristic for determining which action sheet presentation style to use, so that we
only use the closest indicator rect for a link if the text indicator data is also present (otherwise, we fall
back to using the element rect). All other behavior is the same.

* UIProcess/ios/WKActionSheetAssistant.mm:
(-[WKActionSheetAssistant showImageSheet]):
(presentationStyleForView):

Refactor _shouldPresentAtTouchLocationForElementRect into presentationStyleForView, a static function that
returns a WKActionSheetPresentationStyle.

(-[WKActionSheetAssistant showLinkSheet]):
(-[WKActionSheetAssistant _shouldPresentAtTouchLocationForElementRect:]): Deleted.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@219287 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoAdd WebGL2 configuration option to build-webkit
zandobersek@gmail.com [Mon, 10 Jul 2017 05:58:57 +0000 (05:58 +0000)]
Add WebGL2 configuration option to build-webkit
https://bugs.webkit.org/show_bug.cgi?id=174251

Reviewed by Michael Catanzaro.

* Scripts/webkitperl/FeatureList.pm: Add the 'webgl2' option, making it possible
to enable the ENABLE_WEBGL2 flag through the build-webkit script. The feature
is marked as enabled for the Apple Cocoa ports.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@219286 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years ago[JSC] Drop LineNumberAdder since we no longer treat <LF><CR> (not <CR><LF>) as one...
utatane.tea@gmail.com [Mon, 10 Jul 2017 02:34:55 +0000 (02:34 +0000)]
[JSC] Drop LineNumberAdder since we no longer treat <LF><CR> (not <CR><LF>) as one line terminator
https://bugs.webkit.org/show_bug.cgi?id=174296

Reviewed by Mark Lam.

Previously, we treat <LF><CR> as one line terminator. So we increase line number by one.
It caused a problem in scanning template literals. While template literals normalize
<LF><CR> to <LF><LF>, we still needed to increase line number by only one.
To handle it correctly, LineNumberAdder is introduced.

As of r219263, <LF><CR> is counted as two line terminators. So we do not need to have
LineNumberAdder. Let's just use shiftLineTerminator() instead.

* parser/Lexer.cpp:
(JSC::Lexer<T>::parseTemplateLiteral):
(JSC::LineNumberAdder::LineNumberAdder): Deleted.
(JSC::LineNumberAdder::clear): Deleted.
(JSC::LineNumberAdder::add): Deleted.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@219285 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoResource Load Statistics: User interaction should always go to top document
bfulgham@apple.com [Mon, 10 Jul 2017 02:18:56 +0000 (02:18 +0000)]
Resource Load Statistics: User interaction should always go to top document
https://bugs.webkit.org/show_bug.cgi?id=174120
<rdar://problem/33117899>

Patch by John Wilander <wilander@apple.com> on 2017-07-08
Reviewed by Chris Dumez.

Source/WebCore:

Test: http/tests/loading/resourceLoadStatistics/user-interaction-in-cross-origin-sub-frame.html

* dom/UserGestureIndicator.cpp:
(WebCore::UserGestureIndicator::UserGestureIndicator):
    Now logs user interaction for the top document.
* loader/ResourceLoadObserver.cpp:
(WebCore::ResourceLoadObserver::setThrottledObserverNotifications):
    Test infrastructure.
(WebCore::ResourceLoadObserver::setNotificationCallback):
    Callback now takes a ResourceLoadObserver::NotificationType.
(WebCore::ResourceLoadObserver::logFrameNavigation):
    Submits the configured ResourceLoadObserver::NotificationType.
(WebCore::ResourceLoadObserver::logSubresourceLoading):
    Submits the configured ResourceLoadObserver::NotificationType.
(WebCore::ResourceLoadObserver::logWebSocketLoading):
    Submits the configured ResourceLoadObserver::NotificationType.
(WebCore::ResourceLoadObserver::logUserInteractionWithReducedTimeResolution):
    Submits the configured ResourceLoadObserver::NotificationType.
* loader/ResourceLoadObserver.h:
* testing/Internals.cpp:
(WebCore::Internals::resetToConsistentState):
    Resets to throttled notifications.
(WebCore::Internals::setResourceLoadStatisticsThrottledObserverNotifications):
    Test infrastructure.
* testing/Internals.h:
* testing/Internals.idl:
    Added internals.setResourceLoadStatisticsThrottledObserverNotifications().

LayoutTests:

* http/tests/loading/resourceLoadStatistics/resources/dummy.html: Added.
* http/tests/loading/resourceLoadStatistics/user-interaction-in-cross-origin-sub-frame-expected.txt: Added.
* http/tests/loading/resourceLoadStatistics/user-interaction-in-cross-origin-sub-frame.html: Added.
* platform/mac-wk2/TestExpectations:
    Added as [ Pass ].

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@219284 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoRemove some obsolete WebKitVersionChecks.
beidson@apple.com [Sun, 9 Jul 2017 21:26:15 +0000 (21:26 +0000)]
Remove some obsolete WebKitVersionChecks.
https://bugs.webkit.org/show_bug.cgi?id=174294

Reviewed by Dan Bernstein.

Source/WebCore:

No new tests (No change to testable behavior)

* dom/ScriptExecutionContext.cpp:
(WebCore::ScriptExecutionContext::dispatchErrorEvent):

* page/Settings.in:

* platform/RuntimeApplicationChecks.h:
* platform/cocoa/RuntimeApplicationChecksCocoa.mm:
(WebCore::IOSApplication::isOkCupid): Deleted.
(WebCore::IOSApplication::isFacebook): Deleted.

* rendering/RenderBox.cpp:
(WebCore::RenderBox::layoutOverflowRectForPropagation):

Source/WebKit/ios:

* WebView/WebPDFViewIOS.mm:
(-[WebPDFView drawPage:]):

Source/WebKit/mac:

* Misc/WebKitVersionChecks.h:

* WebView/WebHTMLView.mm:
(-[WebHTMLView hitTest:]):

* WebView/WebView.mm:
(shouldRespectPriorityInCSSAttributeSetters):
(-[WebView _commonInitializationWithFrameName:groupName:]):
(-[WebView _preferencesChanged:]):
(shouldTransformsAffectOverflow): Deleted.
(shouldDispatchJavaScriptWindowOnErrorEvents): Deleted.

Source/WebKit2:

* WebProcess/WebPage/WebPage.cpp:
(WebKit::WebPage::updatePreferences):

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@219283 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years ago[Xcode] ICU headers aren’t treated as system headers after r219155
mitz@apple.com [Sun, 9 Jul 2017 20:19:47 +0000 (20:19 +0000)]
[Xcode] ICU headers aren’t treated as system headers after r219155
https://bugs.webkit.org/show_bug.cgi?id=174299

Reviewed by Sam Weinig.

* Configurations/JavaScriptCore.xcconfig: Pass --system-header-prefix=unicode/ to the C and
  C++ compilers.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@219282 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years ago[JSC] Use fastMalloc / fastFree for STL containers
utatane.tea@gmail.com [Sun, 9 Jul 2017 12:22:58 +0000 (12:22 +0000)]
[JSC] Use fastMalloc / fastFree for STL containers
https://bugs.webkit.org/show_bug.cgi?id=174297

Reviewed by Sam Weinig.

Source/JavaScriptCore:

In some places, we intentionally use STL containers over WTF containers.
For example, we sometimes use std::unordered_{set,map} instead of WTF::Hash{Set,Map}
because we do not have effective empty / deleted representations in the space of key's value.
But just using STL container means using libc's malloc instead of our fast malloc (bmalloc if it is enabled).

We introduce WTF::FastAllocator. This is C++ allocator implementation using fastMalloc and fastFree.
We specify this allocator to STL containers' template parameter to allocate memory from fastMalloc.

This WTF::FastAllocator gives us a chance to use STL containers if it is necessary
without compromising memory allocation throughput.

* dfg/DFGGraph.h:
* dfg/DFGIntegerCheckCombiningPhase.cpp:
* ftl/FTLLowerDFGToB3.cpp:
(JSC::FTL::DFG::LowerDFGToB3::switchStringSlow):
* runtime/FunctionHasExecutedCache.h:
* runtime/TypeLocationCache.h:

Source/WTF:

* wtf/FastMalloc.h:
(WTF::FastAllocator::FastAllocator):
(WTF::FastAllocator::allocate):
(WTF::FastAllocator::deallocate):
(WTF::FastAllocator::operator==):
(WTF::FastAllocator::operator!=):

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@219281 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoRemove some obsolete RuntimeApplicationChecks.
beidson@apple.com [Sun, 9 Jul 2017 05:05:41 +0000 (05:05 +0000)]
Remove some obsolete RuntimeApplicationChecks.
https://bugs.webkit.org/show_bug.cgi?id=174293

Reviewed by Dan Bernstein.

Source/WebCore:

No new tests (No change to testable behavior)

* html/HTMLObjectElement.cpp:
(WebCore::HTMLObjectElement::parametersForPlugin):
(WebCore::shouldNotPerformURLAdjustment): Deleted.

* platform/RuntimeApplicationChecks.h:
* platform/cocoa/RuntimeApplicationChecksCocoa.mm:
(WebCore::IOSApplication::isDaijisenDictionary): Deleted.
(WebCore::IOSApplication::isNASAHD): Deleted.
(WebCore::IOSApplication::isTheEconomistOnIphone): Deleted.

* platform/ios/wak/WebCoreThread.h:
* platform/ios/wak/WebCoreThread.mm:
(StartWebThread):
(WebThreadSetDelegateSourceRunLoopMode): Deleted.

Source/WebKit/mac:

* Misc/WebKitVersionChecks.h:
* WebView/WebView.mm:
(+[WebView enableWebThread]):
(-[WebView _needsPreHTML5ParserQuirks]):

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@219280 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years ago[Curl] Safe access and life cycle management of bare Curl handle
commit-queue@webkit.org [Sun, 9 Jul 2017 02:58:36 +0000 (02:58 +0000)]
[Curl] Safe access and life cycle management of bare Curl handle
by wrapping with C++ class
https://bugs.webkit.org/show_bug.cgi?id=174002

Patch by Basuke Suzuki <Basuke.Suzuki@sony.com> on 2017-07-08
Reviewed by Alex Christensen.

* platform/network/ResourceHandle.h:
* platform/network/ResourceHandleInternal.h:
* platform/network/curl/CookieJarCurl.cpp:
(WebCore::setCookiesFromDOM):
(WebCore::cookiesForSession):
* platform/network/curl/CurlContext.cpp:
(WebCore::CurlContext::CurlContext):
(WebCore::CurlContext::~CurlContext):
(WebCore::CurlContext::initCookieSession):
(WebCore::CurlShareHandle::CurlShareHandle):
(WebCore::CurlShareHandle::~CurlShareHandle):
(WebCore::CurlShareHandle::lockCallback):
(WebCore::CurlShareHandle::unlockCallback):
(WebCore::CurlShareHandle::mutexFor):
(WebCore::CurlMultiHandle::CurlMultiHandle):
(WebCore::CurlMultiHandle::~CurlMultiHandle):
(WebCore::CurlMultiHandle::addHandle):
(WebCore::CurlMultiHandle::removeHandle):
(WebCore::CurlMultiHandle::getFdSet):
(WebCore::CurlMultiHandle::perform):
(WebCore::CurlMultiHandle::readInfo):
(WebCore::CurlHandle::CurlHandle):
(WebCore::CurlHandle::~CurlHandle):
(WebCore::CurlHandle::perform):
(WebCore::CurlHandle::pause):
(WebCore::CurlHandle::enableShareHandle):
(WebCore::CurlHandle::setPrivateData):
(WebCore::CurlHandle::setUrl):
(WebCore::CurlHandle::clearUrl):
(WebCore::CurlHandle::clearRequestHeaders):
(WebCore::CurlHandle::appendRequestHeader):
(WebCore::CurlHandle::enableRequestHeaders):
(WebCore::CurlHandle::enableHttpGetRequest):
(WebCore::CurlHandle::enableHttpHeadRequest):
(WebCore::CurlHandle::enableHttpPostRequest):
(WebCore::CurlHandle::setPostFields):
(WebCore::CurlHandle::setPostFieldLarge):
(WebCore::CurlHandle::enableHttpPutRequest):
(WebCore::CurlHandle::setInFileSizeLarge):
(WebCore::CurlHandle::setHttpCustomRequest):
(WebCore::CurlHandle::enableAcceptEncoding):
(WebCore::CurlHandle::enableAllowedProtocols):
(WebCore::CurlHandle::enableFollowLocation):
(WebCore::CurlHandle::enableAutoReferer):
(WebCore::CurlHandle::enableHttpAuthentication):
(WebCore::CurlHandle::setHttpAuthUserPass):
(WebCore::CurlHandle::enableCAInfoIfExists):
(WebCore::CurlHandle::setSslVerifyPeer):
(WebCore::CurlHandle::setSslVerifyHost):
(WebCore::CurlHandle::setSslCert):
(WebCore::CurlHandle::setSslCertType):
(WebCore::CurlHandle::setSslKeyPassword):
(WebCore::CurlHandle::enableCookieJarIfExists):
(WebCore::CurlHandle::setCookieList):
(WebCore::CurlHandle::getCookieList):
(WebCore::CurlHandle::clearCookieList):
(WebCore::CurlHandle::enableProxyIfExists):
(WebCore::CurlHandle::enableTimeout):
(WebCore::CurlHandle::setHeaderCallbackFunction):
(WebCore::CurlHandle::setWriteCallbackFunction):
(WebCore::CurlHandle::setReadCallbackFunction):
(WebCore::CurlHandle::setSslCtxCallbackFunction):
(WebCore::CurlHandle::getEffectiveURL):
(WebCore::CurlHandle::getPrimaryPort):
(WebCore::CurlHandle::getResponseCode):
(WebCore::CurlHandle::getContentLenghtDownload):
(WebCore::CurlHandle::getHttpAuthAvail):
(WebCore::CurlHandle::getTimes):
(WebCore::CurlHandle::maxCurlOffT):
(WebCore::CurlHandle::expectedSizeOfCurlOffT):
(WebCore::CurlHandle::enableVerboseIfUsed):
(WebCore::CurlHandle::enableStdErrIfUsed):
(WebCore::CurlContext::getEffectiveURL): Deleted.
(WebCore::CurlContext::createMultiHandle): Deleted.
(WebCore::CurlContext::mutexFor): Deleted.
(WebCore::CurlContext::lock): Deleted.
(WebCore::CurlContext::unlock): Deleted.
* platform/network/curl/CurlContext.h:
(WebCore::CurlGlobal::CurlGlobal):
(WebCore::CurlGlobal::~CurlGlobal):
(WebCore::CurlShareHandle::handle):
(WebCore::CurlContext::shareHandle):
(WebCore::CurlHandle::handle):
(WebCore::CurlHandle::url):
(WebCore::CurlContext::curlShareHandle): Deleted.
* platform/network/curl/CurlDownload.cpp:
(WebCore::CurlDownload::~CurlDownload):
(WebCore::CurlDownload::init):
(WebCore::CurlDownload::start):
(WebCore::CurlDownload::cancel):
(WebCore::CurlDownload::getUrl):
(WebCore::CurlDownload::addHeaders):
(WebCore::CurlDownload::didReceiveHeader):
(WebCore::CurlDownload::writeCallback):
* platform/network/curl/CurlDownload.h:
* platform/network/curl/CurlJobManager.cpp:
(WebCore::CurlJobManager::CurlJobManager):
(WebCore::CurlJobManager::~CurlJobManager):
(WebCore::CurlJobManager::addToCurl):
(WebCore::CurlJobManager::removeFromCurl):
(WebCore::CurlJobManager::workerThread):
* platform/network/curl/CurlJobManager.h:
(WebCore::CurlJobManager::getMultiHandle): Deleted.
* platform/network/curl/ResourceHandleCurl.cpp:
(WebCore::ResourceHandleInternal::~ResourceHandleInternal):
(WebCore::ResourceHandle::platformSetDefersLoading):
(WebCore::ResourceHandle::didReceiveAuthenticationChallenge):
(WebCore::ResourceHandle::receivedCredential):
(WebCore::ResourceHandle::receivedRequestToContinueWithoutCredential):
(WebCore::calculateWebTimingInformations):
(WebCore::handleLocalReceiveResponse):
(WebCore::writeCallback):
(WebCore::getProtectionSpace):
(WebCore::headerCallback):
(WebCore::readCallback):
(WebCore::setupFormData):
(WebCore::ResourceHandle::setupPUT):
(WebCore::ResourceHandle::setupPOST):
(WebCore::ResourceHandle::dispatchSynchronousJob):
(WebCore::ResourceHandle::applyAuthentication):
(WebCore::ResourceHandle::initialize):
(WebCore::ResourceHandle::handleCurlMsg):
* platform/network/curl/ResourceHandleManager.cpp:
(WebCore::ResourceHandleManager::ResourceHandleManager):
(WebCore::ResourceHandleManager::~ResourceHandleManager):
(WebCore::ResourceHandleManager::downloadTimerCallback):
(WebCore::ResourceHandleManager::removeFromCurl):
(WebCore::ResourceHandleManager::startJob):
* platform/network/curl/ResourceHandleManager.h:
* platform/network/curl/SSLHandle.cpp:
(WebCore::setSSLClientCertificate):
(WebCore::certVerifyCallback):
(WebCore::setSSLVerifyOptions):

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@219279 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoREGRESSION: "visibility:hidden" does not hide play button for video elements
commit-queue@webkit.org [Sat, 8 Jul 2017 23:59:04 +0000 (23:59 +0000)]
REGRESSION: "visibility:hidden" does not hide play button for video elements
https://bugs.webkit.org/show_bug.cgi?id=174258
<rdar://problem/33181452>

Patch by Antoine Quint <graouts@apple.com> on 2017-07-08
Reviewed by Dean Jackson.

Source/WebCore:

In order to not have most styles from the page affect the shadow root, we set "all: initial" on the
media controls container. However, we need to still make the "visibility" property inherit from its
host such that "visibility: hidden" on the host won't be overridden by setting the property back to
its initial value, which is "visible".

Test: media/modern-media-controls/css/visibility-hidden.html

* Modules/modern-media-controls/controls/media-controls.css:
(.media-controls-container):

LayoutTests:

Add a test that checks that <video controls style="visibility:hidden"> does not render anything.

* media/modern-media-controls/css/visibility-hidden-expected.html: Added.
* media/modern-media-controls/css/visibility-hidden.html: Added.
* platform/ios-simulator/TestExpectations:
* platform/mac/TestExpectations:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@219278 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoSimplify WebResourceLoadStatisticsStore / ResourceLoadStatisticsStore
cdumez@apple.com [Sat, 8 Jul 2017 23:11:16 +0000 (23:11 +0000)]
Simplify WebResourceLoadStatisticsStore / ResourceLoadStatisticsStore
https://bugs.webkit.org/show_bug.cgi?id=174290

Reviewed by Brent Fulgham.

Source/WebKit2:

* UIProcess/API/Cocoa/WKWebsiteDataStore.mm:
(-[WKWebsiteDataStore _resourceLoadStatisticsProcessStatisticsAndDataRecords]):
(-[WKWebsiteDataStore _resourceLoadStatisticsSubmitTelemetry]):
* UIProcess/API/Cocoa/WKWebsiteDataStorePrivate.h:
Rename a couple of functions in the Cocoa SPI for clarity and to match the name
of their internal implementation.

* UIProcess/Storage/ResourceLoadStatisticsStore.cpp:
(WebKit::ResourceLoadStatisticsStore::create):
(WebKit::ResourceLoadStatisticsStore::ensureResourceStatisticsForPrimaryDomain):
Stylistic changes.

(WebKit::ResourceLoadStatisticsStore::createEncoderFromData):
Make const.

(WebKit::ResourceLoadStatisticsStore::readDataFromDecoder):
Use HashMap::add() instead of HashMap::set(). There is not supposed to have any duplicate
keys so add() is sufficient and more efficient.

(WebKit::ResourceLoadStatisticsStore::clearInMemoryAndPersistent): Removed.
clearInMemoryAndPersistent() was only called from WebResourceLoadStatisticsStore and
the m_deletePersistentStoreHandler / m_grandfatherExistingWebsiteDataHandler were both
set by the WebResourceLoadStatisticsStore as well. Therefore, we don't really need this
function, WebResourceLoadStatisticsStore now calls ResourceLoadStatisticsStore::clearInMemory()
instead and takes care of doing what the m_deletePersistentStoreHandler /
m_grandfatherExistingWebsiteDataHandler handlers were doing on its side.

(WebKit::ResourceLoadStatisticsStore::mergeStatistics):
Pass Vector as rvalue reference, as a preparation for a future optimization. Added FIXME
comment about the optimization.

(WebKit::ResourceLoadStatisticsStore::setNotificationCallback): Removed.
(WebKit::ResourceLoadStatisticsStore::setGrandfatherExistingWebsiteDataCallback): Removed.
(WebKit::ResourceLoadStatisticsStore::setDeletePersistentStoreCallback): Removed.
(WebKit::ResourceLoadStatisticsStore::setFireTelemetryCallback): Removed.
Drop those callback setters. WebResourceLoadStatisticsStore is the only client of
ResourceLoadStatisticsStore and those callbacks were always called as a result of
an operation requested by the WebResourceLoadStatisticsStore. Therefore, those are
not needed. WebResourceLoadStatisticsStore can take care of doing what those
callbacks were doing on its side.

(WebKit::ResourceLoadStatisticsStore::processStatistics):
Pass parameter by const reference since we clearly do not intend to transfer ownership of it.

(WebKit::ResourceLoadStatisticsStore::topPrivatelyControlledDomainsToRemoveWebsiteDataFor):
(WebKit::ResourceLoadStatisticsStore::updateStatisticsForRemovedDataRecords):
(WebKit::ResourceLoadStatisticsStore::handleFreshStartWithEmptyOrNoStore):
Stylistic changes.

(WebKit::ResourceLoadStatisticsStore::shouldRemoveDataRecords): Removed.
(WebKit::ResourceLoadStatisticsStore::dataRecordsBeingRemoved): Removed.
(WebKit::ResourceLoadStatisticsStore::dataRecordsWereRemoved): Removed.
Those were moved to WebResourceLoadStatisticsStore since they do not require
any access to the statistics, and are only used by WebResourceLoadStatisticsStore.

* UIProcess/Storage/ResourceLoadStatisticsStore.h:

* UIProcess/WebResourceLoadStatisticsStore.cpp:
(WebKit::WebResourceLoadStatisticsStore::WebResourceLoadStatisticsStore):
Store registering most callbacks on the coreStore. Instead, the logic of those
callbacks is now inlined in the few methods in WebResourceLoadStatisticsStore
that were causing these callbacks to be called.

(WebKit::WebResourceLoadStatisticsStore::removeDataRecords):
Access some methods on WebResourceLoadStatisticsStore instead of the coreStore
since those were moved.

(WebKit::WebResourceLoadStatisticsStore::processStatisticsAndDataRecords):
Drop unnecessary mutable for lambda.

(WebKit::WebResourceLoadStatisticsStore::resourceLoadStatisticsUpdated):
Pass parameter as rvalue reference as a preparation for a future optimization.

(WebKit::WebResourceLoadStatisticsStore::hasStatisticsFileChangedSinceLastSync):
Make const.

(WebKit::WebResourceLoadStatisticsStore::submitTelemetry):
Add assertion to make sure this is called on the main thread.

(WebKit::WebResourceLoadStatisticsStore::fireDataModificationHandler): Removed.
This was dispatching to a background queue to call fireDataModificationHandler()
on the ResourceLoadStatisticsStore. The coreStore would then dispatch back on the
main thread to call the WebResourceLoadStatisticsStore's modification handler.
This modification handler was merely calling processStatisticsAndDataRecords().
Therefore, callers of WebResourceLoadStatisticsStore::fireDataModificationHandler()
can call WebResourceLoadStatisticsStore::processStatisticsAndDataRecords() directly
instead.

(WebKit::WebResourceLoadStatisticsStore::fireTelemetryHandler): Removed.
This was calling ResourceLoadStatisticsStore::fireTelemetryHandler(), which was
calling back WebResourceLoadStatisticsStore's fireTelemetryHandler. The handler
in question was merely calling submitTelemetry(). Therefore, callers of
WebResourceLoadStatisticsStore::fireTelemetryHandler() can call
WebResourceLoadStatisticsStore::submitTelemetry() directly instead.

(WebKit::WebResourceLoadStatisticsStore::clearInMemoryAndPersistent):
Call ResourceLoadStatisticsStore::clearInMemory() instead of clearInMemoryAndPersistent(),
which was removed. ResourceLoadStatisticsStore::clearInMemoryAndPersistent() was calling
clearInMemory() and then 2 of WebResourceLoadStatisticsStore's handlers. Instead, we now
call the functions that those handlers were calling directly, thus eliminating the need
for those handlers.

(WebKit::WebResourceLoadStatisticsStore::setMinimumTimeBetweenDataRecordsRemoval):
Set m_minimumTimeBetweenDataRecordsRemoval directly as it is now store on this class
instead of the coreStore.

(WebKit::WebResourceLoadStatisticsStore::shouldRemoveDataRecords):
(WebKit::WebResourceLoadStatisticsStore::dataRecordsBeingRemoved):
(WebKit::WebResourceLoadStatisticsStore::dataRecordsWereRemoved):
Those were merely moved from the ResourceLoadStatisticsStore.

* UIProcess/WebResourceLoadStatisticsStore.h:

* UIProcess/WebResourceLoadStatisticsTelemetry.cpp:
(WebKit::WebResourceLoadStatisticsTelemetry::calculateAndSubmit):
Check for sortedPrevalentResources.isEmpty() instead of checking if the size
if less than minimumPrevalentResourcesForTelemetry.
ResourceLoadStatisticsStore::sortedPrevalentResourceTelemetry() either returns
an empty vector or a vector with more than minimumPrevalentResourcesForTelemetry
items. Therefore, there is no need to expose minimumPrevalentResourcesForTelemetry.

Tools:

Rename testRunner API to match the new internal API names.

* WebKitTestRunner/InjectedBundle/Bindings/TestRunner.idl:
* WebKitTestRunner/InjectedBundle/TestRunner.cpp:
(WTR::TestRunner::statisticsProcessStatisticsAndDataRecords):
(WTR::TestRunner::statisticsSubmitTelemetry):
* WebKitTestRunner/InjectedBundle/TestRunner.h:
* WebKitTestRunner/TestController.cpp:
(WTR::TestController::statisticsProcessStatisticsAndDataRecords):
(WTR::TestController::statisticsSubmitTelemetry):
* WebKitTestRunner/TestController.h:
* WebKitTestRunner/TestInvocation.cpp:
(WTR::TestInvocation::didReceiveSynchronousMessageFromInjectedBundle):
* WebKitTestRunner/cocoa/TestControllerCocoa.mm:
(WTR::TestController::statisticsProcessStatisticsAndDataRecords):
(WTR::TestController::statisticsSubmitTelemetry):

LayoutTests:

Update layout tests to reflect changes to the testRunner API naming.

* http/tests/loading/resourceLoadStatistics/classify-as-non-prevalent-based-on-mixed-statistics.html:
* http/tests/loading/resourceLoadStatistics/classify-as-non-prevalent-based-on-sub-frame-under-top-frame-origins.html:
* http/tests/loading/resourceLoadStatistics/classify-as-non-prevalent-based-on-subresource-under-top-frame-origins.html:
* http/tests/loading/resourceLoadStatistics/classify-as-non-prevalent-based-on-subresource-unique-redirects-to.html:
* http/tests/loading/resourceLoadStatistics/classify-as-prevalent-based-on-mixed-statistics.html:
* http/tests/loading/resourceLoadStatistics/classify-as-prevalent-based-on-sub-frame-under-top-frame-origins.html:
* http/tests/loading/resourceLoadStatistics/classify-as-prevalent-based-on-subresource-under-top-frame-origins.html:
* http/tests/loading/resourceLoadStatistics/classify-as-prevalent-based-on-subresource-unique-redirects-to.html:
* http/tests/loading/resourceLoadStatistics/clear-in-memory-and-persistent-store-one-hour.html:
* http/tests/loading/resourceLoadStatistics/clear-in-memory-and-persistent-store.html:
* http/tests/loading/resourceLoadStatistics/grandfathering.html:
* http/tests/loading/resourceLoadStatistics/non-prevalent-resource-with-user-interaction.html:
* http/tests/loading/resourceLoadStatistics/non-prevalent-resource-without-user-interaction.html:
* http/tests/loading/resourceLoadStatistics/partitioned-and-unpartitioned-cookie-deletion.html:
* http/tests/loading/resourceLoadStatistics/partitioned-and-unpartitioned-cookie-with-partitioning-timeout.html:
* http/tests/loading/resourceLoadStatistics/partitioned-cookies-with-and-without-user-interaction.html:
* http/tests/loading/resourceLoadStatistics/prevalent-resource-with-user-interaction-timeout.html:
* http/tests/loading/resourceLoadStatistics/prevalent-resource-with-user-interaction.html:
* http/tests/loading/resourceLoadStatistics/prevalent-resource-without-user-interaction.html:
* http/tests/loading/resourceLoadStatistics/telemetry-generation.html:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@219277 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoDrop NOSNIFF compile flag
utatane.tea@gmail.com [Sat, 8 Jul 2017 15:58:16 +0000 (15:58 +0000)]
Drop NOSNIFF compile flag
https://bugs.webkit.org/show_bug.cgi?id=174289

Reviewed by Michael Catanzaro.

.:

* Source/cmake/WebKitFeatures.cmake:

Source/JavaScriptCore:

* Configurations/FeatureDefines.xcconfig:

Source/WebCore:

* Configurations/FeatureDefines.xcconfig:
* css/StyleSheetContents.cpp:
(WebCore::StyleSheetContents::parseAuthorStyleSheet):
(WebCore::StyleSheetContents::notifyLoadedSheet):
* dom/LoadableClassicScript.cpp:
(WebCore::LoadableClassicScript::notifyFinished):
* loader/cache/CachedCSSStyleSheet.cpp:
(WebCore::CachedCSSStyleSheet::mimeTypeAllowedByNosniff):
(WebCore::CachedCSSStyleSheet::canUseSheet):
* loader/cache/CachedCSSStyleSheet.h:
* platform/network/HTTPParsers.cpp:
(WebCore::parseContentTypeOptionsHeader):
* platform/network/HTTPParsers.h:
* platform/network/ResourceResponseBase.cpp:
(WebCore::isScriptAllowedByNosniff):
* platform/network/ResourceResponseBase.h:
* workers/WorkerScriptLoader.cpp:
(WebCore::WorkerScriptLoader::didReceiveResponse):

Source/WebCore/PAL:

* Configurations/FeatureDefines.xcconfig:

Source/WebKit/mac:

* Configurations/FeatureDefines.xcconfig:

Source/WebKit2:

* Configurations/FeatureDefines.xcconfig:

Tools:

* Scripts/webkitperl/FeatureList.pm:
* TestWebKitAPI/Configurations/FeatureDefines.xcconfig:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@219276 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years ago[WK2] Use a rolling 30-day uptime for processing statistics
cdumez@apple.com [Sat, 8 Jul 2017 05:09:55 +0000 (05:09 +0000)]
[WK2] Use a rolling 30-day uptime for processing statistics
https://bugs.webkit.org/show_bug.cgi?id=174235
<rdar://problem/33164381>

Reviewed by Brent Fulgham.

Follow-up fix for r219274 because it caused this test to time out:
http/tests/loading/resourceLoadStatistics/prevalent-resource-with-user-interaction-timeout.html

The test sets TimeToLiveUserInteraction to 0 so our implementation cannot use
0 as magic value to see if it was set. Instead, use std::optional.

* UIProcess/API/Cocoa/WKWebsiteDataStore.mm:
(-[WKWebsiteDataStore _resourceLoadStatisticsResetToConsistentState]):
* UIProcess/Storage/ResourceLoadStatisticsStore.cpp:
(WebKit::ResourceLoadStatisticsStore::setTimeToLiveUserInteraction):
(WebKit::ResourceLoadStatisticsStore::hasStatisticsExpired):
* UIProcess/Storage/ResourceLoadStatisticsStore.h:
* UIProcess/WebResourceLoadStatisticsStore.cpp:
(WebKit::WebResourceLoadStatisticsStore::setTimeToLiveUserInteraction):
* UIProcess/WebResourceLoadStatisticsStore.h:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@219275 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years ago[WK2] Use a rolling 30-day uptime for processing statistics
bfulgham@apple.com [Sat, 8 Jul 2017 00:07:13 +0000 (00:07 +0000)]
[WK2] Use a rolling 30-day uptime for processing statistics
https://bugs.webkit.org/show_bug.cgi?id=174235
<rdar://problem/33164381>

Reviewed by Chris Dumez.

Source/WebCore:

Add a KeyedDecoder specialization for Deque.

* platform/KeyedCoding.h:
(WebCore::KeyedDecoder::decodeObjects):

Source/WebKit2:

* UIProcess/API/Cocoa/WKWebsiteDataStore.cpp:
(WebKit::WKWebsiteDataStore::_resourceLoadStatisticsResetToConsistentState): Initialize time-to-live to zero by default.
* UIProcess/Storage/ResourceLoadStatisticsStore.cpp:
(WebKit::ResourceLoadStatisticsStore::createEncoderFromData): Write out vector
of operating dates.
(WebKit::ResourceLoadStatisticsStore::readDataFromDecoder): Read in vector of
operating dates.
(WebKit::ResourceLoadStatisticsStore::hasHadRecentUserInteraction): Check new
convenience method.
(WebKit::ResourceLoadStatisticsStore::markTodayAsOperatingDate): Added.
(WebKit::ResourceLoadStatisticsStore::hasStatisticsExpired): Added.
* UIProcess/Storage/ResourceLoadStatisticsStore.h:
* UIProcess/WebResourceLoadStatisticsStore.cpp:
(WebKit::WebResourceLoadStatisticsStore::readDataFromDiskIfNeeded): When reading
a new data file, update the current operating date (if needed).
(WebKit::WebResourceLoadStatisticsStore::handleDailyTasks): Roll uptime dates as
needed, then handle telemetry.
* UIProcess/WebResourceLoadStatisticsStore.h:

Source/WTF:

Modify Deque to allow it to be used in a template specialization in KeyedDecoder.

* wtf/Deque.h:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@219274 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoLower the max_protection for the separated heap
commit-queue@webkit.org [Sat, 8 Jul 2017 00:03:27 +0000 (00:03 +0000)]
Lower the max_protection for the separated heap
https://bugs.webkit.org/show_bug.cgi?id=174281

Patch by AJ Ringer <aringer@apple.com> on 2017-07-07
Reviewed by Oliver Hunt.

Switch to vm_protect so we can set maximum page protection.

* jit/ExecutableAllocator.cpp:
(JSC::FixedVMPoolExecutableAllocator::initializeSeparatedWXHeaps):
(JSC::ExecutableAllocator::allocate):

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@219273 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years ago[AppCache] Ignore fallback entries whose namespace is not prefixed with manifest...
dbates@webkit.org [Fri, 7 Jul 2017 22:02:46 +0000 (22:02 +0000)]
[AppCache] Ignore fallback entries whose namespace is not prefixed with manifest path
https://bugs.webkit.org/show_bug.cgi?id=174273
<rdar://problem/33011682>

Reviewed by Brent Fulgham.

Source/WebCore:

As per <https://html.spec.whatwg.org/multipage/offline.html#parsing-cache-manifests> (07/06/2017)
we should ignore fallback entires whose fallback namespace URL is not prefixed with
the manifest path. For now we only apply this policy when the manifest is served with
a non-standard Content-Type to minimize web compatibility risk.

Test: http/tests/appcache/fallback-namespace-outside-manifest-path.html

* loader/appcache/ApplicationCacheGroup.cpp:
(WebCore::ApplicationCacheGroup::didFinishLoadingManifest): Pass the MIME type of the manifest.
* loader/appcache/ManifestParser.cpp:
(WebCore::manifestPath): Computes the manifest path from a manifest URL.
(WebCore::parseManifest): Modified to take the MIME type of the manifest. If the MIME type is
non-standard (i.e. not text/cached-manifest) then skip fallback entries whose namespace is not
prefixed with the manifest path. Otherwise, process fallback entries as we do now. Also cleaned
up the code a bit while I was here, including renaming a local variable to be more descriptive
and using a const character array for the manifest signature to avoid the need to document the
length of the manifest signature in a comment.
* loader/appcache/ManifestParser.h:

LayoutTests:

* http/tests/appcache/fallback-namespace-outside-manifest-path-expected.txt: Added.
* http/tests/appcache/fallback-namespace-outside-manifest-path.html: Added.
* http/tests/appcache/resources/fallback-namespace-outside-manifest-path.txt: Added.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@219272 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years ago[iOS DnD] For cross-app drags, 'drop' event handlers are never invoked if dataTransfe...
wenson_hsieh@apple.com [Fri, 7 Jul 2017 21:59:25 +0000 (21:59 +0000)]
[iOS DnD] For cross-app drags, 'drop' event handlers are never invoked if dataTransfer.dropEffect is not set while dragging
https://bugs.webkit.org/show_bug.cgi?id=174219
<rdar://problem/32083177>

Reviewed by Ryosuke Niwa.

Source/WebCore:

Currently, in DragController.cpp, defaultOperationForDrag maps a drag source operation mask of
DragOperationGeneric to DragOperationMove across all platforms. However, on iOS, where cross-app drag moves do
not trigger a drop, this means drop handlers won't fire unless the dropEffect is explicitly set to copy.

To fix this, we introduce DragController::platformGenericDragOperation(), which returns DragOperationCopy on iOS
and DragOperationMove (the existing behavior) elsewhere. defaultOperationForDrag then maps a drag source
operation mask of DragOperationGeneric to platformGenericDragOperation().

Tests:  DataInteractionTests.ExternalSourceHTMLToUploadArea
        DataInteractionTests.ExternalSourceImageAndHTMLToUploadArea
        DataInteractionTests.ExternalSourceMoveOperationNotAllowed

* page/DragController.cpp:
(WebCore::DragController::platformGenericDragOperation):
(WebCore::defaultOperationForDrag):
* page/DragController.h:
* page/mac/DragControllerMac.mm:
(WebCore::DragController::platformGenericDragOperation):

Source/WebKit2:

Tweak some testing SPI to return a drop operation flag instead of whether or not the drop operation was not
UIDropOperationCancel.

* UIProcess/API/Cocoa/WKWebView.mm:
(-[WKWebView _simulateDataInteractionUpdated:]):
* UIProcess/API/Cocoa/WKWebViewPrivate.h:
* UIProcess/ios/WKContentViewInteraction.h:
* UIProcess/ios/WKContentViewInteraction.mm:
(-[WKContentView _simulateDataInteractionUpdated:]):

Tools:

Add plumbing and support to mock the value of -allowsMoveOperation on the simulated UIDragDropSession objects.
Setting the DataInteractionSimulator's shouldAllowMoveOperation property to NO simulates a drag operation coming
in from another app out-of-process, for which move operations won't cause a drop to be performed in the first
place.

Also tweaks 2 existing unit tests regarding file uploads via JavaScript to simulate items coming in from a
different application, and adds a new test to check that if a drop area specifically requests a MOVE operation,
no action is taken when dropping.

* TestWebKitAPI/Tests/WebKit2Cocoa/file-uploading.html:
* TestWebKitAPI/Tests/ios/DataInteractionTests.mm:
(TestWebKitAPI::TEST):
* TestWebKitAPI/ios/DataInteractionSimulator.h:
* TestWebKitAPI/ios/DataInteractionSimulator.mm:
(-[MockDragDropSession initWithItems:location:window:allowMove:]):
(-[MockDragDropSession allowsMoveOperation]):
(-[MockDataOperationSession initWithProviders:location:window:allowMove:]):
(-[MockDataInteractionSession initWithWindow:allowMove:]):
(-[DataInteractionSimulator initWithWebView:]):
(-[DataInteractionSimulator runFrom:to:]):
(-[DataInteractionSimulator _advanceProgress]):
(-[MockDragDropSession initWithItems:location:window:]): Deleted.
(-[MockDataOperationSession initWithProviders:location:window:]): Deleted.
(-[MockDataInteractionSession initWithWindow:]): Deleted.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@219271 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoWeb Inspector: Resources are sometimes missing from the tree outline right before...
drousso@apple.com [Fri, 7 Jul 2017 21:44:18 +0000 (21:44 +0000)]
Web Inspector: Resources are sometimes missing from the tree outline right before folderizing
https://bugs.webkit.org/show_bug.cgi?id=174277

Reviewed by Joseph Pecoraro.

* UserInterface/Views/TreeElement.js:
(WebInspector.TreeElement.prototype.expand):

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@219270 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoWeb Inspector: REGRESSION(r217258): WebSocket resource tree elements don't show the...
drousso@apple.com [Fri, 7 Jul 2017 21:33:26 +0000 (21:33 +0000)]
Web Inspector: REGRESSION(r217258): WebSocket resource tree elements don't show the connection status
https://bugs.webkit.org/show_bug.cgi?id=174274

Reviewed by Brian Burg.

* UserInterface/Views/FrameTreeElement.js:
(WebInspector.FrameTreeElement):
* UserInterface/Views/ResourceTreeElement.js:
(WebInspector.ResourceTreeElement.prototype.populateContextMenu):
* UserInterface/Views/WebSocketResourceTreeElement.js:
(WebInspector.WebSocketResourceTreeElement.prototype.populateContextMenu):
Move "Log WebSocket" context menu item to WebSocketResourceTreeElement.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@219269 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoWeb Inspector: Show all elements currently using a given CSS Canvas
drousso@apple.com [Fri, 7 Jul 2017 21:30:47 +0000 (21:30 +0000)]
Web Inspector: Show all elements currently using a given CSS Canvas
https://bugs.webkit.org/show_bug.cgi?id=173965

Reviewed by Joseph Pecoraro.

Source/JavaScriptCore:

* inspector/protocol/Canvas.json:
 - Add `requestCSSCanvasClientNodes` command for getting the node IDs all nodes using this
   canvas via -webkit-canvas.
 - Add `cssCanvasClientNodesChanged` event that is dispatched whenever a node is
   added/removed from the list of -webkit-canvas clients.

Source/WebCore:

Test: inspector/canvas/css-canvas-clients.html

* css/CSSImageGeneratorValue.cpp:
(WebCore::CSSImageGeneratorValue::addClient):
(WebCore::CSSImageGeneratorValue::removeClient):
* css/CSSImageGeneratorValue.h:
(WebCore::CSSImageGeneratorValue::clients):
* html/HTMLCanvasElement.cpp:
(WebCore::HTMLCanvasElement::addObserver):
(WebCore::HTMLCanvasElement::removeObserver):
(WebCore::HTMLCanvasElement::cssCanvasClients):
Each time an observer is added/removed for a given HTMLCanvasElement, send an event to the
inspector frontend that the CSS canvas client nodes have changed. Additionally, anytime a
client/use is added/removed from one of the observing CSSCanvasValue, fire the same event.

* css/CSSCanvasValue.h:
(isType):
* html/HTMLCanvasElement.h:
(WebCore::CanvasObserver::isCSSCanvasValueObserver):
Allows type traits to distinguish CanvasObserver from CSSCanvasValue::CanvasObserverProxy.

* inspector/InspectorCanvasAgent.h:
* inspector/InspectorCanvasAgent.cpp:
(WebCore::InspectorCanvasAgent::requestCSSCanvasClientNodes):
(WebCore::InspectorCanvasAgent::didChangeCSSCanvasClientNodes):
* inspector/InspectorInstrumentation.h:
(WebCore::InspectorInstrumentation::didChangeCSSCanvasClientNodes):
* inspector/InspectorInstrumentation.cpp:
(WebCore::InspectorInstrumentation::didChangeCSSCanvasClientNodesImpl):
Notify the frontend that the list of client nodes has changed for the given canvas. Let the
frontend request the actual list of node IDs when it needs, possibly at a later time.

Source/WebInspectorUI:

* UserInterface/Controllers/CanvasManager.js:
(WebInspector.CanvasManager.prototype.cssCanvasClientNodesChanged):
* UserInterface/Models/Canvas.js:
(WebInspector.Canvas.prototype.requestCSSCanvasClientNodes):
(WebInspector.Canvas.prototype.cssCanvasClientNodesChanged):
* UserInterface/Protocol/CanvasObserver.js:
(WebInspector.CanvasObserver.prototype.cssCanvasClientNodesChanged):

* Localizations/en.lproj/localizedStrings.js:
* UserInterface/Views/CanvasDetailsSidebarPanel.js:
(WebInspector.CanvasDetailsSidebarPanel):
(WebInspector.CanvasDetailsSidebarPanel.prototype.set canvas):
(WebInspector.CanvasDetailsSidebarPanel.prototype.initialLayout):
(WebInspector.CanvasDetailsSidebarPanel.prototype.layout):
(WebInspector.CanvasDetailsSidebarPanel.prototype._refreshCSSCanvasSection):
(WebInspector.CanvasDetailsSidebarPanel.prototype._formatMemoryRow):
Add CSS section for CSS canvases. Currently displays a list of node links, each of which is
using the selected canvas via -webkit-canvas.

* UserInterface/Main.html:
* UserInterface/Views/CanvasDetailsSidebarPanel.css: Added.
(.sidebar > .panel.details.canvas .details-section > .content .row.simple > .value > .node-link):

* UserInterface/Controllers/DOMTreeManager.js:
(WebInspector.DOMTreeManager.prototype.ensureDocument):
* UserInterface/Models/Canvas.js:
(WebInspector.Canvas.prototype.requestNode):
* UserInterface/Views/SearchSidebarPanel.js:
(WebInspector.SearchSidebarPanel.prototype.performSearch):
Add convenience function that will call DOMAgent.getDocument with an empty function. Should
be used when it is necessary that the document has been sent to the frontend, but the
document node itself is not needed.

LayoutTests:

* inspector/canvas/css-canvas-clients-expected.txt: Added.
* inspector/canvas/css-canvas-clients.html: Added.
* platform/mac/TestExpectations:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@219268 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoAdjusted test expectations for webrtc/video-replace-muted-track.html.
jlewis3@apple.com [Fri, 7 Jul 2017 20:38:33 +0000 (20:38 +0000)]
Adjusted test expectations for webrtc/video-replace-muted-track.html.
https://bugs.webkit.org/show_bug.cgi?id=173608

Unreviewed test gardening.

* platform/mac-wk2/TestExpectations:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@219267 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoAVPlayer can continue to be active after released by MediaPlayerPrivateAVFoundationObjC.
jer.noble@apple.com [Fri, 7 Jul 2017 20:24:35 +0000 (20:24 +0000)]
AVPlayer can continue to be active after released by MediaPlayerPrivateAVFoundationObjC.
https://bugs.webkit.org/show_bug.cgi?id=174264

Reviewed by Eric Carlson.

If the AVPlayer is retained (by an autorelease pool, or internally by other objects in
AVFoundation), releasing the AVPlayer is not enough to cancel loading or playback. So before
releasing the AVPlayer, make sure to disassociate the current AVPlayerItem, which should
cancel all activity in the AVPlayer.

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

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@219266 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years ago[Curl] Remove data url handler for async load
commit-queue@webkit.org [Fri, 7 Jul 2017 20:20:24 +0000 (20:20 +0000)]
[Curl] Remove data url handler for async load
https://bugs.webkit.org/show_bug.cgi?id=174263

data url is handled by ResourceLoader. No need for specific handling
in platform dependent layer.

Patch by Basuke Suzuki <Basuke.Suzuki@sony.com> on 2017-07-07
Reviewed by Alex Christensen.

* platform/network/curl/ResourceHandleManager.cpp:
(WebCore::ResourceHandleManager::startJob):

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@219265 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoUnreviewed, rolling out r219257.
jlewis3@apple.com [Fri, 7 Jul 2017 20:05:27 +0000 (20:05 +0000)]
Unreviewed, rolling out r219257.

The test added in the revision was still extreamly flaky on
all testers.

Reverted changeset:

"[SVG] Leak in SVGAnimatedListPropertyTearOff"
https://bugs.webkit.org/show_bug.cgi?id=172545
http://trac.webkit.org/changeset/219257

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@219264 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years ago\n\r is not the same as \r\n.
mark.lam@apple.com [Fri, 7 Jul 2017 19:50:00 +0000 (19:50 +0000)]
\n\r is not the same as \r\n.
https://bugs.webkit.org/show_bug.cgi?id=173053

Reviewed by Keith Miller.

JSTests:

* stress/regress-173053.js: Added.
* stress/template-literal-line-terminators.js:

Source/JavaScriptCore:

* parser/Lexer.cpp:
(JSC::Lexer<T>::shiftLineTerminator):
(JSC::LineNumberAdder::add):

LayoutTests:

* js/parse-backslash-before-newline-expected.txt:
* js/script-tests/parse-backslash-before-newline.js:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@219263 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoUnreviewed, rolling out r219256.
jlewis3@apple.com [Fri, 7 Jul 2017 19:36:08 +0000 (19:36 +0000)]
Unreviewed, rolling out r219256.

Previous change to Test Expectations broke the tests

Reverted changeset:

"[WPE] Unreviewed gardening"
http://trac.webkit.org/changeset/219256

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@219262 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoUnreviewed, rolling out r219258.
jlewis3@apple.com [Fri, 7 Jul 2017 19:35:38 +0000 (19:35 +0000)]
Unreviewed, rolling out r219258.

Previous change to Test Expectations broke the tests

Reverted changeset:

"Unreviewed gardening, fix WK2 expectations after my previous
commit"
http://trac.webkit.org/changeset/219258

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@219261 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoUnreviewed, rolling out r219238, r219239, and r219241.
commit-queue@webkit.org [Fri, 7 Jul 2017 18:30:09 +0000 (18:30 +0000)]
Unreviewed, rolling out r219238, r219239, and r219241.
https://bugs.webkit.org/show_bug.cgi?id=174265

"fast/workers/dedicated-worker-lifecycle.html is flaky"
(Requested by yusukesuzuki on #webkit).

Reverted changesets:

"[WTF] Implement WTF::ThreadGroup"
https://bugs.webkit.org/show_bug.cgi?id=174081
http://trac.webkit.org/changeset/219238

"Unreviewed, build fix after r219238"
https://bugs.webkit.org/show_bug.cgi?id=174081
http://trac.webkit.org/changeset/219239

"Unreviewed, CLoop build fix after r219238"
https://bugs.webkit.org/show_bug.cgi?id=174081
http://trac.webkit.org/changeset/219241

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@219260 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoWeb Inspector: Clean up some unnecessary constructors
commit-queue@webkit.org [Fri, 7 Jul 2017 17:20:31 +0000 (17:20 +0000)]
Web Inspector: Clean up some unnecessary constructors
https://bugs.webkit.org/show_bug.cgi?id=174236

Patch by Joseph Pecoraro <pecoraro@apple.com> on 2017-07-07
Reviewed by Brian Burg.

* UserInterface/Controllers/CodeMirrorBezierEditingController.js:
* UserInterface/Controllers/CodeMirrorColorEditingController.js:
* UserInterface/Controllers/CodeMirrorGradientEditingController.js:
* UserInterface/Views/DividerNavigationItem.js:
* UserInterface/Views/FlexibleSpaceNavigationItem.js:
* UserInterface/Views/IdleTreeElement.js:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@219259 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoUnreviewed gardening, fix WK2 expectations after my previous commit
mcatanzaro@igalia.com [Fri, 7 Jul 2017 16:43:34 +0000 (16:43 +0000)]
Unreviewed gardening, fix WK2 expectations after my previous commit

* platform/wk2/TestExpectations:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@219258 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years ago[SVG] Leak in SVGAnimatedListPropertyTearOff
svillar@igalia.com [Fri, 7 Jul 2017 16:42:32 +0000 (16:42 +0000)]
[SVG] Leak in SVGAnimatedListPropertyTearOff
https://bugs.webkit.org/show_bug.cgi?id=172545

Reviewed by Said Abou-Hallawa.

SVGAnimatedListPropertyTearOff maintains a vector m_wrappers with references to
SVGPropertyTraits<PropertyType>::ListItemTearOff. Apart from that SVGPropertyTearOff has a
reference to SVGAnimatedProperty.

When SVGListProperty::getItemValuesAndWrappers() is called, it creates a
SVGPropertyTraits<PropertyType>::ListItemTearOff pointing to the same SVGAnimatedProperty (a
SVGAnimatedListPropertyTearOff) which stores the m_wrappers vector where the ListItemTearOff
is going to be added to. This effectively creates a reference cycle between the
SVGAnimatedListPropertyTearOff and all the ListItemTearOff it stores in m_wrappers.

We should detach those wrappers in propertyWillBeDeleted() in order to break the cycle.

* svg/properties/SVGAnimatedListPropertyTearOff.h:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@219257 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years ago[WPE] Unreviewed gardening
mcatanzaro@igalia.com [Fri, 7 Jul 2017 15:41:40 +0000 (15:41 +0000)]
[WPE] Unreviewed gardening

* platform/wk2/TestExpectations: Do not mark individual resourceLoadStatistics tests as
passing. Instead, mark the entire directory, so we don't have to override the expectations
for every single test individually in the WPE expectations file.
* platform/wpe/TestExpectations: Update resourceLoadStatistics expectations.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@219256 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years ago[GTK][WPE] Enable X-Content-Type-Options: nosniff
utatane.tea@gmail.com [Fri, 7 Jul 2017 12:45:18 +0000 (12:45 +0000)]
[GTK][WPE] Enable X-Content-Type-Options: nosniff
https://bugs.webkit.org/show_bug.cgi?id=174250

Reviewed by Carlos Alberto Lopez Perez.

.:

Make nosniff enabled by default in all the ports.

* Source/cmake/OptionsMac.cmake:
* Source/cmake/OptionsWin.cmake:
* Source/cmake/WebKitFeatures.cmake:

Tools:

* Scripts/webkitperl/FeatureList.pm:

LayoutTests:

* platform/gtk/TestExpectations:
* platform/gtk/imported/w3c/web-platform-tests/fetch/nosniff/stylesheet-expected.txt: Removed.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@219255 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years ago[WPE] New baseline after r218740
csaavedra@igalia.com [Fri, 7 Jul 2017 11:48:25 +0000 (11:48 +0000)]
[WPE] New baseline after r218740

Unreviewed gardening.
* platform/wpe/js/dom/global-constructors-attributes-expected.txt:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@219254 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years ago[WPE] Unreviewed gardening
csaavedra@igalia.com [Fri, 7 Jul 2017 10:44:02 +0000 (10:44 +0000)]
[WPE] Unreviewed gardening

resourceLoadingStatistics tests are timing out and failing.
* platform/wpe/TestExpectations:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@219253 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years ago[GStreamer] vid.me videos do not play
commit-queue@webkit.org [Fri, 7 Jul 2017 10:32:36 +0000 (10:32 +0000)]
[GStreamer] vid.me videos do not play
https://bugs.webkit.org/show_bug.cgi?id=172240

Patch by Charlie Turner <cturner@igalia.com> on 2017-07-07
Reviewed by Xabier Rodriguez-Calvar.

Source/WebCore:

In r142251, code to hide the WK HTTP source elements from elsewhere in
the pipeline was removed. This has the nasty side-effect of
auto-plugging the WK HTTP source into things it really should not be
used in, especially the adaptive streaming demuxers. The reasons this
is bad are documented in several places on Bugzilla, see the parent
bug report for more details. The high-level issue is that the WK HTTP
source and its use of WebCore is not thread-safe. Although work has
been recently done to improve this situation, it's still not perfect.

Another issue is the interface hlsdemux expects its HTTP source to
implement, specifically seeking in READY.

This does rely on HTTP context sharing being available in GStreamer,
upstream bug is here:
https://bugzilla.gnome.org/show_bug.cgi?id=761099. The failing case
can be demonstrated with
https://github.com/thiagoss/adaptive-test-server but manual testing on
popular video hosting sites, including vid.me, shows that this doesn't
bite us at the moment, just something else to fix in the future.

There are some QoS issues with the adaptive streaming code in
GStreamer, but it seems much better to offer a below par QoS in lieu
of crashing/livelocking when playing certain streams, and issues can be
raised upstream when they arise.

This patch does take us further away from the future goal of having all
networking operations go through the network process, but in return it
solves some nasty crashes and livelocks that have been irritating
users for some time. With the pressure off on this issue, work can be
planned to consider how to make the WK HTTP source a better citizen
inside the GStreamer pipeline when we migrate the netcode to go
through the network process.

A new test is added to check that the single file HLS playlists
(new in version 4) can be played, which was the primary cause of
this bug report.

Test: http/tests/media/hls/range-request.html

* platform/graphics/gstreamer/MediaPlayerPrivateGStreamer.cpp:
(WebCore::MediaPlayerPrivateGStreamer::setPlaybinURL): Perform
some trickery to make sure that we only ever fetch URLs handed to
us by WebCore. Any further URLs discovered inside the pipeline
will not get WKWS auto-plugged, since they'll be plain https?
schemas.
(WebCore::MediaPlayerPrivateGStreamer::load): Refactor to use the
setPlaybinURL helper method.
(WebCore::MediaPlayerPrivateGStreamer::loadNextLocation): Ditto.
* platform/graphics/gstreamer/MediaPlayerPrivateGStreamer.h: Add
the setPlaybinURL helper method.
* platform/graphics/gstreamer/WebKitWebSourceGStreamer.cpp:
(webKitWebSrcGetProtocols): Only advertise webkit+https?, this
ensures we won't get auto-plugged by pipeline elements asking for
an element to fetch https? resources (like adaptive demuxers).
(convertPlaybinURI): Undo the trick when another element asks us
for our URI.

Tools:

Build httpsoupsrc again for use in adaptive streaming pipelines, and
have the existing libsoup build against GNOME to avoid header drift
against GStreamer's linked Soup library.

* gtk/jhbuild.modules:

LayoutTests:

Add a test for single output file HLS playlists that require HTTP
range requests to playback. This failed using the WK http source
for reasons documented in the linked bug.

Generated with mp4hls --segment-duration 3 --output-single-file

* Http/tests/media/hls/range-request-expected.txt: Added.
* http/tests/media/hls/range-request.html: Added.
* http/tests/media/resources/hls/range-request-playlist.m3u8: Added.
* http/tests/media/resources/hls/range-request-playlists/iframes.m3u8: Added.
* http/tests/media/resources/hls/range-request-playlists/media.ts: Added.
* http/tests/media/resources/hls/range-request-playlists/stream.m3u8: Added.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@219252 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoSpeedometer 2.0: Fix vanilla JS example
commit-queue@webkit.org [Fri, 7 Jul 2017 10:25:00 +0000 (10:25 +0000)]
Speedometer 2.0: Fix vanilla JS example
https://bugs.webkit.org/show_bug.cgi?id=174246

Patch by Mathias Bynens <mathias@qiwi.be> on 2017-07-07
Reviewed by Ryosuke Niwa.

Previously, all TODO items were assigned an ID of `null`. This patch
fixes that bug. Manually deleting items now works again.

* Speedometer/resources/todomvc/vanilla-examples/vanillajs/index.html: Made title consistent.
* Speedometer/resources/todomvc/vanilla-examples/vanillajs/js/store.js: Fixed bug with deleting items.
* Speedometer/resources/todomvc/vanilla-examples/vanillajs/readme.md: Added docs.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@219251 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoUnreviewed, Windows build fix after r219233 part 4
utatane.tea@gmail.com [Fri, 7 Jul 2017 08:30:08 +0000 (08:30 +0000)]
Unreviewed, Windows build fix after r219233 part 4
https://bugs.webkit.org/show_bug.cgi?id=174231

* wtf/Assertions.h:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@219250 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoWKWebSiteDataStore.removeDataOfTypes should wait until disk cache files are actually...
antti@apple.com [Fri, 7 Jul 2017 07:22:12 +0000 (07:22 +0000)]
WKWebSiteDataStore.removeDataOfTypes should wait until disk cache files are actually removed before invoking completion handler
https://bugs.webkit.org/show_bug.cgi?id=174224
<rdar://problem/33067545>

Reviewed by Sam Weinig.

Currently we dispatch file deletion operations to a background queue and call the completion
handler without waiting for the I/O to complete.

* NetworkProcess/NetworkProcess.cpp:
(WebKit::clearDiskCacheEntries):

    Call a new version of NetworkCache::remove() for bulk deletion.
    Note that it is fine to call this with an empty vector.

* NetworkProcess/cache/NetworkCache.cpp:
(WebKit::NetworkCache::Cache::remove):

    Bulk deletion with a completion handler.

(WebKit::NetworkCache::Cache::deleteFiles): Added.

    Factor to a helper function.

* NetworkProcess/cache/NetworkCache.h:
* NetworkProcess/cache/NetworkCacheStorage.cpp:
(WebKit::NetworkCache::Storage::remove):

    Remove files for all the provided keys in a queue and invoke the completion handler in the main thread when done.

* NetworkProcess/cache/NetworkCacheStorage.h:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@219249 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoUnreviewed, Windows build fix after r219233 part 3
utatane.tea@gmail.com [Fri, 7 Jul 2017 06:08:02 +0000 (06:08 +0000)]
Unreviewed, Windows build fix after r219233 part 3
https://bugs.webkit.org/show_bug.cgi?id=174231

* wtf/Assertions.h:
(compilerFenceForCrash):

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@219248 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoUnreviewed, Windows build fix after r219233 part 2
utatane.tea@gmail.com [Fri, 7 Jul 2017 06:04:02 +0000 (06:04 +0000)]
Unreviewed, Windows build fix after r219233 part 2
https://bugs.webkit.org/show_bug.cgi?id=174231

* wtf/Assertions.h:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@219247 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoWrite a support script to enable buildbot to upload to S3
lforschler@apple.com [Fri, 7 Jul 2017 05:57:34 +0000 (05:57 +0000)]
Write a support script to enable buildbot to upload to S3
https://bugs.webkit.org/show_bug.cgi?id=174196

Reviewed by Stephanie Lewis.

* BuildSlaveSupport/build.webkit.org-config/transfer-archive-to-s3: Added.
(uploadToS3):
(archiveExists):

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@219246 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoUnreviewed GTK and WPE test gardening
mcatanzaro@igalia.com [Fri, 7 Jul 2017 05:53:39 +0000 (05:53 +0000)]
Unreviewed GTK and WPE test gardening

* platform/gtk/TestExpectations:
* platform/gtk/imported/w3c/web-platform-tests/WebCryptoAPI/derive_bits_keys/hkdf.worker-expected.txt:
* platform/gtk/imported/w3c/web-platform-tests/WebCryptoAPI/derive_bits_keys/test_hkdf.https-expected.txt:
* platform/wpe/TestExpectations:
* platform/wpe/imported/w3c/web-platform-tests/WebCryptoAPI/derive_bits_keys/hkdf.worker-expected.txt:
* platform/wpe/imported/w3c/web-platform-tests/WebCryptoAPI/derive_bits_keys/test_hkdf.https-expected.txt:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@219245 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoUnreviewed, Windows build fix after r219233
utatane.tea@gmail.com [Fri, 7 Jul 2017 05:33:40 +0000 (05:33 +0000)]
Unreviewed, Windows build fix after r219233
https://bugs.webkit.org/show_bug.cgi?id=174231

* wtf/Compiler.h:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@219244 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoFollow-up to "REGRESSION(r209495): materiauxlaverdure.com fails to load"
mcatanzaro@igalia.com [Fri, 7 Jul 2017 05:22:36 +0000 (05:22 +0000)]
Follow-up to "REGRESSION(r209495): materiauxlaverdure.com fails to load"
https://bugs.webkit.org/show_bug.cgi?id=173301
<rdar://problem/32624850>

Unreviewed gardening. Attempt to fix this test by updating it to expect double quotes. It
was missed because it is marked as flaky on Mac, so no way to automatically detect the
failure there.

This doesn't actually fix the test for me due to some floating point issues, but it's a step
in the right direction, and I speculate this will fix the test on the GTK and WPE bots.

* transitions/clip-path-path-transitions.html:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@219243 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoFix bad usage of static variables in ResourceLoadStatisticsStore
cdumez@apple.com [Fri, 7 Jul 2017 05:19:14 +0000 (05:19 +0000)]
Fix bad usage of static variables in ResourceLoadStatisticsStore
https://bugs.webkit.org/show_bug.cgi?id=174237

Reviewed by Brent Fulgham.

Fix bad usage of static variables in ResourceLoadStatisticsStore. Those can be set via
member functions on ResourceLoadStatisticsStore so they should be data members. The
API to set them is associated to a particular store and there can in theory be several
stores.

* UIProcess/Storage/ResourceLoadStatisticsStore.cpp:
(WebKit::ResourceLoadStatisticsStore::shouldPartitionCookies):
(WebKit::ResourceLoadStatisticsStore::setTimeToLiveUserInteraction):
(WebKit::ResourceLoadStatisticsStore::setTimeToLiveCookiePartitionFree):
(WebKit::ResourceLoadStatisticsStore::setMinimumTimeBetweenDataRecordsRemoval):
(WebKit::ResourceLoadStatisticsStore::setGrandfatheringTime):
(WebKit::ResourceLoadStatisticsStore::hasHadRecentUserInteraction):
(WebKit::ResourceLoadStatisticsStore::handleFreshStartWithEmptyOrNoStore):
(WebKit::ResourceLoadStatisticsStore::shouldRemoveDataRecords):
(WebKit::shouldPartitionCookies): Deleted.
* UIProcess/Storage/ResourceLoadStatisticsStore.h:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@219242 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoUnreviewed, CLoop build fix after r219238
utatane.tea@gmail.com [Fri, 7 Jul 2017 05:18:57 +0000 (05:18 +0000)]
Unreviewed, CLoop build fix after r219238
https://bugs.webkit.org/show_bug.cgi?id=174081

* heap/MachineStackMarker.cpp:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@219241 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoEnabling uploading archives to S3.
lforschler@apple.com [Fri, 7 Jul 2017 05:10:38 +0000 (05:10 +0000)]
Enabling uploading archives to S3.
https://bugs.webkit.org/show_bug.cgi?id=174198

Reviewed by Stephanie Lewis.

* BuildSlaveSupport/build.webkit.org-config/master.cfg:
(ArchiveMinifiedBuiltProduct): Create a minified archive from a full archive
(UploadMinifiedBuiltProduct): Upload minified archive to build master
(TransferToS3): Transfer full and minified archives to S3

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@219240 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoUnreviewed, build fix after r219238
utatane.tea@gmail.com [Fri, 7 Jul 2017 05:08:22 +0000 (05:08 +0000)]
Unreviewed, build fix after r219238
https://bugs.webkit.org/show_bug.cgi?id=174081

* wtf/Threading.cpp:
(WTF::Thread::addToThreadGroup):

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@219239 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years ago[WTF] Implement WTF::ThreadGroup
utatane.tea@gmail.com [Fri, 7 Jul 2017 04:56:23 +0000 (04:56 +0000)]
[WTF] Implement WTF::ThreadGroup
https://bugs.webkit.org/show_bug.cgi?id=174081

Reviewed by Mark Lam.

Source/JavaScriptCore:

Large part of MachineThreads are now removed and replaced with WTF::ThreadGroup.
And SamplingProfiler and others interact with WTF::Thread directly.

* API/tests/ExecutionTimeLimitTest.cpp:
* heap/MachineStackMarker.cpp:
(JSC::MachineThreads::MachineThreads):
(JSC::captureStack):
(JSC::MachineThreads::tryCopyOtherThreadStack):
(JSC::MachineThreads::tryCopyOtherThreadStacks):
(JSC::MachineThreads::gatherConservativeRoots):
(JSC::ActiveMachineThreadsManager::Locker::Locker): Deleted.
(JSC::ActiveMachineThreadsManager::add): Deleted.
(JSC::ActiveMachineThreadsManager::remove): Deleted.
(JSC::ActiveMachineThreadsManager::contains): Deleted.
(JSC::ActiveMachineThreadsManager::ActiveMachineThreadsManager): Deleted.
(JSC::activeMachineThreadsManager): Deleted.
(JSC::MachineThreads::~MachineThreads): Deleted.
(JSC::MachineThreads::addCurrentThread): Deleted.
(): Deleted.
(JSC::MachineThreads::removeThread): Deleted.
(JSC::MachineThreads::removeThreadIfFound): Deleted.
(JSC::MachineThreads::MachineThread::MachineThread): Deleted.
(JSC::MachineThreads::MachineThread::getRegisters): Deleted.
(JSC::MachineThreads::MachineThread::Registers::stackPointer): Deleted.
(JSC::MachineThreads::MachineThread::Registers::framePointer): Deleted.
(JSC::MachineThreads::MachineThread::Registers::instructionPointer): Deleted.
(JSC::MachineThreads::MachineThread::Registers::llintPC): Deleted.
(JSC::MachineThreads::MachineThread::captureStack): Deleted.
* heap/MachineStackMarker.h:
(JSC::MachineThreads::addCurrentThread):
(JSC::MachineThreads::getLock):
(JSC::MachineThreads::threads):
(JSC::MachineThreads::MachineThread::suspend): Deleted.
(JSC::MachineThreads::MachineThread::resume): Deleted.
(JSC::MachineThreads::MachineThread::threadID): Deleted.
(JSC::MachineThreads::MachineThread::stackBase): Deleted.
(JSC::MachineThreads::MachineThread::stackEnd): Deleted.
(JSC::MachineThreads::threadsListHead): Deleted.
* runtime/SamplingProfiler.cpp:
(JSC::FrameWalker::isValidFramePointer):
(JSC::SamplingProfiler::SamplingProfiler):
(JSC::SamplingProfiler::takeSample):
(JSC::SamplingProfiler::noticeCurrentThreadAsJSCExecutionThread):
* runtime/SamplingProfiler.h:
* wasm/WasmMachineThreads.cpp:
(JSC::Wasm::resetInstructionCacheOnAllThreads):

Source/WebCore:

* page/ResourceUsageThread.h:

Source/WebKit2:

* Shared/AsyncRequest.h:
* UIProcess/Storage/ResourceLoadStatisticsStore.h:

Source/WTF:

This patch implements WTF::ThreadGroup. It implements core of JSC::MachineThreads with more reliable way.
JSC::MachineThreads was complicated because of managing dead threads. Each JSC::MachineThreads has its
own TLS with a registered destructor. And everytime a thread dies, the registered TLS destructor is called.
And this destructor will remove the current dying thread from JSC::MachineThreads.

However the above implementation is tricky. And each JSC::MachineThreads requires own TLS space, which is
not considered in WTF's Windows ThreadSpecific implementation. Current design works well since we only
have small number of MachineThreads right now.

Instead, we use more reliable way. After introducing WTF::Thread, WTF::Thread has WTF::Thread::didExit,
which is called when associated TLS (with WTF::Thread) is destroyed. We leverage this mechanism to remove
WTF::Thread from MachineThreads.

This patch introduces WTF::ThreadGroup. It is tightly integrated with WTF::Thread: WTF::Thread knows
ThreadGroups which includes this thread. And WTF::ThreadGroup of course knows WTF::Threads added to it.
WTF::Thread::didExit carefully remove itself from WTF::ThreadGroups.

The most important part of this patch is locking. WTF::Thread can die. And WTF::ThreadGroup can die.
If we take a design using two fine grain locks in WTF::Thread and WTF::ThreadGroup, we easily encounter
dead lock. Consider the following case.

1. When adding WTF::Thread (TH) to WTF::ThreadGroup (THG), we first hold a lock of THG, and hold a lock of TH (locking order is THG -> TH).
2. When TH dies, TH need to hold a lock of TH to iterate THGs. And we hold a lock of THG to unregister TH from it (locking order is TH -> THG).
3. When suspending and resuming THs in THG, we first hold a lock of THG. And then, we hold a lock of TH to suspend and resume it (locking order is THG -> TH).
4. When destroying THG, we need to hold a lock of TH to unregister THG from TH. We can hold a lock of THG before that (locking order is THG -> TH).

Then, it easily causes dead lock. We cannot swap the locking order of (2) since iterating THG requires a lock of TH.
To solve this problem, we introduce one global lock ThreadGroup::destructionMutex (GL).

1. When adding WTF::Thread (TH) to WTF::ThreadGroup (THG), we first hold GL, and hold a lock of THG. Do not hold a
lock of TH. TH's thread groups information is guarded by GL instead of a lock of TH (locking order is GL -> THG).
2. When TH dies, TH need to hold GL to iterate THGs. And we hold a lock of THG to unregister TH from it (locking order is GL -> THG).
3. When suspending and resuming THs in THG, we first hold a lock of THG. And then, we hold a lock of TH to suspend and resume it (locking order is THG -> TH).
4. When destroying THG, we need to hold GL to unregister THG from TH. We can hold a lock of THG after that (locking order is GL -> THG).

We still have a lock of THG. By separating GL and THG's lock, we can hold a lock of THG while completely unrelated TH is destructed which takes a lock of GL and unrelated THG.

* WTF.xcodeproj/project.pbxproj:
* wtf/AutomaticThread.cpp:
* wtf/CMakeLists.txt:
* wtf/CrossThreadCopier.h:
* wtf/ParkingLot.h:
* wtf/ThreadGroup.cpp: Copied from Source/JavaScriptCore/wasm/WasmMachineThreads.cpp.
(WTF::ThreadGroup::destructionMutex):
(WTF::ThreadGroup::~ThreadGroup):
(WTF::ThreadGroup::add):
(WTF::ThreadGroup::addCurrentThread):
(WTF::ThreadGroup::removeCurrentThread):
* wtf/ThreadGroup.h: Copied from Source/JavaScriptCore/wasm/WasmMachineThreads.cpp.
(WTF::ThreadGroup::create):
(WTF::ThreadGroup::threads):
(WTF::ThreadGroup::getLock):
* wtf/Threading.cpp:
(WTF::Thread::didExit):
(WTF::Thread::addToThreadGroup):
(WTF::Thread::removeFromThreadGroup):
* wtf/Threading.h:
(WTF::Thread::canAddToThreadGroup):

Tools:

Add WTF::ThreadGroup tests.

* TestWebKitAPI/CMakeLists.txt:
* TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj:
* TestWebKitAPI/Tests/WTF/ThreadGroup.cpp: Added.
(TestWebKitAPI::TEST):

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@219238 268f45cc-cd09-0410-ab3c-d52691b4dbfc