Unreviewed, rolling out r235500.
[WebKit-https.git] / Tools / ChangeLog
index 61682d0..209239d 100644 (file)
-2018-09-05  Ross Kirsling  <ross.kirsling@sony.com>
+2018-09-19  Michael Catanzaro  <mcatanzaro@igalia.com>
 
-        Add WinCairo to bot watcher's dashboard.
-        https://bugs.webkit.org/show_bug.cgi?id=189273
+        Unreviewed, rolling out r235500.
 
-        Reviewed by Aakash Jain.
-
-        * BuildSlaveSupport/build.webkit.org-config/public_html/dashboard/Images/Windows10.png: Added.
-        * BuildSlaveSupport/build.webkit.org-config/public_html/dashboard/Images/Windows10@2x.png: Added.
-
-        * BuildSlaveSupport/build.webkit.org-config/public_html/dashboard/Scripts/Dashboard.js:
-        * BuildSlaveSupport/build.webkit.org-config/public_html/dashboard/Styles/Main.css:
-        (table.queue-grid tr.platform.windows-10 img.logo): Added.
-        (table.queue-grid tr.platform.windows-8 img.logo): Deleted.
-        (table.queue-grid tr.platform.windows-xp img.logo): Deleted.
-        Add Win10 section to dashboard. Remove unused WinXP and Win8 sections.
-
-        * BuildSlaveSupport/build.webkit.org-config/public_html/dashboard/Scripts/BubbleQueueServer.js:
-        (BubbleQueueServer):
-        * BuildSlaveSupport/build.webkit.org-config/public_html/dashboard/Scripts/WebKitBuildbot.js:
-        (WebKitBuildbot):
-        Add WinCairo build, test, and EWS bots to Win10 section of dashboard.
-
-2018-09-05  David Kilzer  <ddkilzer@apple.com>
-
-        REGRESSION (r209470): EditingHistory folder is missing in EditingHistory Xcode project
-
-        Found using `tidy-Xcode-project-file --missing` (see Bug
-        188754).  Fixes were made manually.
-
-        * EditingHistory/EditingHistory.xcodeproj/project.pbxproj:
-        (EditingHistory): Change the path of the folder to
-        "EditingHistory", then remove the now-redundant name.
-        (EditingHistory/Resources/DumpEditingHistory.js): Remove
-        redundant file reference since another file reference exists at
-        the top of the project file list, and that file reference is the
-        one used in the "Copy Bundle Resources" build phase.
-        (EditingHistory/Resources/EditingHistoryUtil.js): Ditto.
-
-2018-09-05  David Kilzer  <ddkilzer@apple.com>
-
-        REGRESSION (r160736): Bundle folder is missing in MiniBrowser Xcode project
-
-        Found using `tidy-Xcode-project-file --missing` (see Bug
-        188754).  Folder fix was made manually.
-
-        * MiniBrowser/MiniBrowser.xcodeproj/project.pbxproj:
-        (Bundle): Change the path of the folder to "mac/Bundle", and set
-        the name back to "Bundle".
-        (Bundle/Info.plist): Change file reference to be relative to its
-        group.  Accomplished using Xcode after fixing the folder path.
-        (Bundle/MiniBrowserBundle_Prefix.pch): Ditto.
-
-2018-09-05  David Kilzer  <ddkilzer@apple.com>
-
-        REGRESSION (r184033): ContentExtensionTester folder is missing in ContentExtensionTester Xcode project
-
-        Found using `tidy-Xcode-project-file --missing` (see Bug
-        188754).  Fixes were made manually.
-
-        * ContentExtensionTester/ContentExtensionTester.xcodeproj/project.pbxproj:
-        (ContentExtensionTester): Change the path of the folder to map
-        to "." (which exists and is where main.m is located), and then
-        set the folder's name back to "ContentExtensionTester".
-        (main.m): Change file reference to be relative to its group now
-        that the group's path is fixed.
+        Time to switch back to master runtime
 
-2018-09-04  Don Olmstead  <don.olmstead@sony.com>
+        Reverted changeset:
 
-        Add generic entrypoint and run loop in TestWebKitAPI
-        https://bugs.webkit.org/show_bug.cgi?id=189287
+        "Unreviewed, rolling out r235114."
+        https://bugs.webkit.org/show_bug.cgi?id=188731
+        https://trac.webkit.org/changeset/235500
 
-        Reviewed by Michael Catanzaro.
+2018-09-19  Adrian Perez de Castro  <aperez@igalia.com>
 
-        The implementations in jsconly are platform agnostic.
+        Unreviewed. Add Pablo Saavedra to the list of contributors.
 
-        * TestWebKitAPI/PlatformJSCOnly.cmake:
-        Use the glib implementation of Utilities when using that as a run loop
-        otherwise use the generic implementation.
+        * Scripts/webkitpy/common/config/contributors.json:
 
-        * TestWebKitAPI/PlatformUtilities.h:
-        Remove repeated entries from Utilities.h.
+2018-09-19  Philippe Normand  <pnormand@igalia.com>
 
-        * TestWebKitAPI/PlatformWPE.cmake:
-        Use the generic main.cpp.
-        
-        * TestWebKitAPI/generic/UtilitiesGeneric.cpp: Renamed from Tools/TestWebKitAPI/jsconly/PlatformUtilitiesJSCOnly.cpp.
-        (TestWebKitAPI::Util::run):
-        (TestWebKitAPI::Util::spinRunLoop):
-        Moved to a generic implementation and added spinRunLoop.
+        [GStreamer] Add support for AV1 decoding
+        https://bugs.webkit.org/show_bug.cgi?id=189647
 
-        * TestWebKitAPI/generic/main.cpp: Renamed from Tools/TestWebKitAPI/jsconly/main.cpp.
-        Share between JSCOnly and WPE
+        Reviewed by Žan Doberšek.
 
-        * TestWebKitAPI/wpe/main.cpp: Removed.
+        Add patches required for AV1 decoding support. They're all
+        upstream already and will be shipped in GStreamer 1.16. The aom
+        GStreamer plugin depends on the aom library for which there's no
+        official release yet.
 
-2018-09-04  Don Olmstead  <don.olmstead@sony.com>
+        * gstreamer/jhbuild.modules:
+        * gstreamer/patches/gst-plugins-bad-0001-aomenc-Add-support-for-10-12bit-decoding.patch: Added.
+        * gstreamer/patches/gst-plugins-bad-0002-aomenc-Handle-8-bit_depth-images-with-AOM_IMG_FMT_HI.patch: Added.
+        * gstreamer/patches/gst-plugins-good-0001-qtdemux-Detect-and-expose-CEA-608-708-Closed-Caption.patch: Added. This one is unrelated but an implicit dependency of the next one.
+        * gstreamer/patches/gst-plugins-good-0004-qtdemux-Add-initial-support-for-AV1-demuxing.patch: Added.
+        * gstreamer/patches/gst-plugins-good-0005-qtdemux-Extract-AV1-codec_data-and-put-it-in-the-cap.patch: Added.
+        * gstreamer/patches/gst-plugins-good-0006-qtdemux-Recognize-more-AV1-atoms.patch: Added.
 
-        [CMake] Make TestWebKitAPI dependencies explicit
-        https://bugs.webkit.org/show_bug.cgi?id=189282
+2018-09-18  Jonathan Bedard  <jbedard@apple.com>
 
-        Reviewed by Fujii Hironori.
+        webkitpy: Clobbering and building occurs multiple times for iOS Simulator ports
+        https://bugs.webkit.org/show_bug.cgi?id=189702
+        <rdar://problem/44541704>
 
-        The config.h file includes files from JSC, WebCore and WebKit which
-        places a dependency on those headers even when just building TestWTF.
-        This makes the depedencies explicit for all CMake ports and shoud
-        remove any race conditions that can be encountered when building tests.
+        Reviewed by Aakash Jain.
 
-        * TestWebKitAPI/CMakeLists.txt:
-        * TestWebKitAPI/PlatformGTK.cmake:
-        * TestWebKitAPI/PlatformWPE.cmake:
-        * TestWebKitAPI/PlatformWin.cmake:
+        * Scripts/webkitpy/layout_tests/controllers/manager.py:
+        (Manager._set_up_run): Move build check and clobbering to run, since set up is
+        run multiple times for iOS simulator.
+        (Manager.run):
 
-2018-09-04  Wenson Hsieh  <wenson_hsieh@apple.com>
+2018-09-18  Chris Dumez  <cdumez@apple.com>
 
-        Populate "text/uri-list" with multiple URLs when the pasteboard contains multiple URLs
-        https://bugs.webkit.org/show_bug.cgi?id=188890
-        <rdar://problem/43648605>
+        "DidFirstVisuallyNonEmptyLayout" callback does not get called when restoring a page from PageCache
+        https://bugs.webkit.org/show_bug.cgi?id=189681
+        <rdar://problem/44526171>
 
-        Reviewed by Tim Horton.
+        Reviewed by Alex Christensen and Zalan Bujtas.
 
-        * DumpRenderTree/mac/DumpRenderTreePasteboard.mm:
-        (-[LocalPasteboard pasteboardItems]):
+        Add API test coverage.
 
-        Implement this method to avoid crashing when running layout tests that access the pasteboard.
+        * TestWebKitAPI/Tests/WebKit/LayoutMilestonesWithAllContentInFrame.cpp:
+        (TestWebKitAPI::didFinishNavigation):
+        (TestWebKitAPI::TEST):
 
-        * TestWebKitAPI/Tests/WebKitCocoa/DragAndDropTests.mm:
+2018-09-18  Claudio Saavedra  <csaavedra@igalia.com>
 
-        Add a test to verify that on macOS and iOS, multiple URLs dropped onto the page are accessible via
-        "text/uri-list".
+        [WPE] Implement mouse event modifiers
+        https://bugs.webkit.org/show_bug.cgi?id=189697
 
-        * TestWebKitAPI/Tests/WebKitCocoa/PasteMixedContent.mm:
+        Reviewed by Carlos Garcia Campos.
 
-        Add a test that exercises 5 different ways to write one or more URLs to the pasteboard on macOS; in all cases,
-        the URLs written to the pasteboard should be exposed to the page via "text/uri-list". In all of these different
-        cases, the results of using `-[NSPasteboardItem stringForType:]`, `-[NSURL URLFromPasteboard:]` and
-        `-[NSPasteboard stringForType:]` will yield different results, so the purpose of this API test is to ensure that
-        our logic for grabbing a list of URLs from the pasteboard on macOS is robust enough to handle all of these
-        different behaviors.
+        * WebKitTestRunner/wpe/EventSenderProxyWPE.cpp:
+        (WTR::wkEventModifiersToWPE):
+        (WTR::EventSenderProxy::mouseDown):
+        (WTR::EventSenderProxy::mouseUp):
 
-        * TestWebKitAPI/Tests/ios/UIPasteboardTests.mm:
+2018-09-17  Yusuke Suzuki  <utatane.tea@gmail.com>
 
-        Add a test to verify that on iOS, using `-[UIPasteboard setURLs:]` to write to multiple URLs to the pasteboard
-        and then pasting results in "text/uri-list" exposing a list of all the URLs written to the pasteboard.
+        [WTF] Use Semaphore and BinarySemaphore instead of dispatch_semaphore_t
+        https://bugs.webkit.org/show_bug.cgi?id=185339
 
-        * WebKitTestRunner/mac/WebKitTestRunnerPasteboard.mm:
-        (-[LocalPasteboard pasteboardItems]):
+        Reviewed by Mark Lam.
 
-        Implement this method to avoid crashing when running layout tests that access the pasteboard.
+        * TestWebKitAPI/Tests/WTF/SynchronizedFixedQueue.cpp:
+        (TestWebKitAPI::ToUpperConverter::stopProducing):
+        (TestWebKitAPI::ToUpperConverter::stopConsuming):
 
-2018-09-04  Simon Fraser  <simon.fraser@apple.com>
+2018-09-17  Simon Fraser  <simon.fraser@apple.com>
 
-        REGRESSION(r235408): GTK bots exiting early
-        https://bugs.webkit.org/show_bug.cgi?id=189063
+        Many modern media control tests leak documents in testing
+        https://bugs.webkit.org/show_bug.cgi?id=189437
 
-        Reviewed by Michael Catanzaro.
+        Reviewed by Darin Adler.
         
-        WebKitTestRunner was running the world leak checks even when run without --world-leaks,
-        causing GTK bot timeouts. So guard updateLiveDocumentsAfterTest(), checkForWorldLeaks()
-        and findAndDumpWorldLeaks() with m_checkForWorldLeaks checks, and in
-        TestController::handleControlCommand() print a message if the control script sends the
-        "#CHECK FOR WORLD LEAKS" command if WTR was not run with --world-leaks.
+        In order to accurately detect leaks in media controls tests which use lots of
+        SVGImages, we have to:
+        - Fire a zero-delay timer after the postTask, in order for ImagesLoader's m_derefElementTimer
+          to clear references to elements.
+        - Have releaseCriticalMemory() call CachedResourceLoader's garbageCollectDocumentResources()
+          to drop the last handle to the CachedResource for an SVGImage.
+        - Call WKBundleReleaseMemory() after the GC and timer, since we need garbageCollectDocumentResources()
+          to run again after that timer has fired.
         
-        I tested that running with --world-leaks still works.
-
-        * WebKitTestRunner/TestController.cpp:
-        (WTR::TestController::resetStateToConsistentValues):
-        (WTR::TestController::updateLiveDocumentsAfterTest):
-        (WTR::TestController::checkForWorldLeaks):
-        (WTR::TestController::findAndDumpWorldLeaks):
-        (WTR::TestController::handleControlCommand):
-        (WTR::TestController::run):
-
-2018-09-04  Myles C. Maxfield  <mmaxfield@apple.com>
-
-        Unreviewed follow-up to r235635
-        https://bugs.webkit.org/show_bug.cgi?id=188940
-
-        Use "let" variables instead of "var" variables.
-
-        * WebGPUShadingLanguageRI/StandardLibrary.js:
-        (let.standardLibrary):
-
-2018-09-04  Myles C. Maxfield  <mmaxfield@apple.com>
-
-        [WHLSL] Implement texture types
-        https://bugs.webkit.org/show_bug.cgi?id=188940
-
-        Rubber-stamped by Dean Jackson.
-
-        Implement the texture types and operations. This includes Sample(), Load(), Store(), Gather(), and GetDimensions().
-        These functions were implemented according to section 15 of the Vulkan 1.1.83 spec, which lists each operation
-        and how to compute its results.
+        This should fix most of the spurious leak reports involving SVGImage documents.
 
-        * WebGPUShadingLanguageRI/All.js:
-        * WebGPUShadingLanguageRI/Casts.js: Moved from Intrinsics. The texture operations need to cast too, so these are
-        moved into a common location.
-        (cast):
-        (bitwiseCast):
-        (castToUchar):
-        (castToUshort):
-        (castToUint):
-        (castToChar):
-        (castToShort):
-        (castToInt):
-        (castToHalf):
-        (castToFloat):
-        * WebGPUShadingLanguageRI/Intrinsics.js: Use Casts.js and call into TextureOperations.js.
-        (Intrinsics.):
-        (Intrinsics.checkFalse):
-        (Intrinsics.boxVector):
-        (Intrinsics.unboxVector):
-        (Intrinsics):
-        (Intrinsics.cast): Deleted.
-        (Intrinsics.bitwiseCast): Deleted.
-        (Intrinsics.castToHalf): Deleted.
-        * WebGPUShadingLanguageRI/SPIRV.html:
-        * WebGPUShadingLanguageRI/Sampler.js: Added. Represent the Sampler type. 
-        (Sampler):
-        (Sampler.prototype.get rAddressMode):
-        (Sampler.prototype.get sAddressMode):
-        (Sampler.prototype.get tAddressMode):
-        (Sampler.prototype.get minFilter):
-        (Sampler.prototype.get magFilter):
-        (Sampler.prototype.get mipmapFilter):
-        (Sampler.prototype.get lodMinClamp):
-        (Sampler.prototype.get lodMaxClamp):
-        (Sampler.prototype.get maxAnisotropy):
-        (Sampler.prototype.get compareFunction):
-        (Sampler.prototype.get borderColor):
-        (Sampler.prototype.calculateBorderColor.computeValues):
-        (Sampler.prototype.calculateBorderColor):
-        * WebGPUShadingLanguageRI/StandardLibrary.js: Update the standard library. Some of the functions were
-        either missing or wrong.
-        (let.standardLibrary):
-        * WebGPUShadingLanguageRI/Test.html:
-        * WebGPUShadingLanguageRI/Test.js: Test many, many combinations of the texture operations.
-        (makeSampler):
-        (make1DTexture):
-        (make1DTextureArray):
-        (make2DTexture):
-        (make2DTextureArray):
-        (make3DTexture):
-        (makeTextureCube):
-        (makeRW1DTexture):
-        (makeRW1DTextureArray):
-        (makeRW2DTexture):
-        (makeRW2DTextureArray):
-        (makeRW3DTexture):
-        (make2DDepthTexture):
-        (make2DDepthTextureArray):
-        (makeDepthTextureCube):
-        (makeRW2DDepthTexture):
-        (makeRW2DDepthTextureArray):
-        (checkFloat4):
-        * WebGPUShadingLanguageRI/Texture.js: Added. Represents each of the texture types.
-        (Texture):
-        (Texture.prototype.get dimension):
-        (Texture.prototype.get width):
-        (Texture.prototype.get height):
-        (Texture.prototype.get depth):
-        (Texture.prototype.get levelCount):
-        (Texture.prototype.get layerCount):
-        (Texture.prototype.get innerType):
-        (Texture.prototype.get data):
-        (Texture.prototype.elementChecked):
-        (Texture.prototype.setElementChecked):
-        (Texture1D):
-        (Texture1D.prototype.widthAtLevel):
-        (Texture1D.prototype.heightAtLevel):
-        (Texture1D.prototype.depthAtLevel):
-        (Texture1D.prototype.element):
-        (Texture1D.prototype.setElement):
-        (Texture1DArray):
-        (Texture1DArray.prototype.widthAtLevel):
-        (Texture1DArray.prototype.heightAtLevel):
-        (Texture1DArray.prototype.depthAtLevel):
-        (Texture1DArray.prototype.element):
-        (Texture1DArray.prototype.setElement):
-        (Texture2D):
-        (Texture2D.prototype.widthAtLevel):
-        (Texture2D.prototype.heightAtLevel):
-        (Texture2D.prototype.depthAtLevel):
-        (Texture2D.prototype.element):
-        (Texture2D.prototype.setElement):
-        (Texture2DArray):
-        (Texture2DArray.prototype.widthAtLevel):
-        (Texture2DArray.prototype.heightAtLevel):
-        (Texture2DArray.prototype.depthAtLevel):
-        (Texture2DArray.prototype.element):
-        (Texture2DArray.prototype.setElement):
-        (Texture3D):
-        (Texture3D.prototype.widthAtLevel):
-        (Texture3D.prototype.heightAtLevel):
-        (Texture3D.prototype.depthAtLevel):
-        (Texture3D.prototype.element):
-        (Texture3D.prototype.setElement):
-        (TextureCube):
-        (TextureCube.prototype.widthAtLevel):
-        (TextureCube.prototype.heightAtLevel):
-        (TextureCube.prototype.depthAtLevel):
-        (TextureCube.prototype.element):
-        (TextureCube.prototype.setElement):
-        (Texture1DRW):
-        (Texture1DRW.prototype.widthAtLevel):
-        (Texture1DRW.prototype.heightAtLevel):
-        (Texture1DRW.prototype.depthAtLevel):
-        (Texture1DRW.prototype.element):
-        (Texture1DRW.prototype.setElement):
-        (Texture1DArrayRW):
-        (Texture1DArrayRW.prototype.widthAtLevel):
-        (Texture1DArrayRW.prototype.heightAtLevel):
-        (Texture1DArrayRW.prototype.depthAtLevel):
-        (Texture1DArrayRW.prototype.element):
-        (Texture1DArrayRW.prototype.setElement):
-        (Texture2DRW):
-        (Texture2DRW.prototype.widthAtLevel):
-        (Texture2DRW.prototype.heightAtLevel):
-        (Texture2DRW.prototype.depthAtLevel):
-        (Texture2DRW.prototype.element):
-        (Texture2DRW.prototype.setElement):
-        (Texture2DArrayRW):
-        (Texture2DArrayRW.prototype.widthAtLevel):
-        (Texture2DArrayRW.prototype.heightAtLevel):
-        (Texture2DArrayRW.prototype.depthAtLevel):
-        (Texture2DArrayRW.prototype.element):
-        (Texture2DArrayRW.prototype.setElement):
-        (let.TextureDepth2DArrayRW.Texture2DArrayRW.Texture3DRW):
-        (let.TextureDepth2DArrayRW.Texture2DArrayRW.Texture3DRW.prototype.widthAtLevel):
-        (let.TextureDepth2DArrayRW.Texture2DArrayRW.Texture3DRW.prototype.heightAtLevel):
-        (let.TextureDepth2DArrayRW.Texture2DArrayRW.Texture3DRW.prototype.depthAtLevel):
-        (let.TextureDepth2DArrayRW.Texture2DArrayRW.Texture3DRW.prototype.element):
-        (let.TextureDepth2DArrayRW.Texture2DArrayRW.Texture3DRW.prototype.setElement):
-        * WebGPUShadingLanguageRI/TextureOperations.js: Added. This implements the texture operations
-        as described in section 15 of the Vulkan 1.1.83 spec.
-        (depthCompareOperation):
-        (conversionToRGBA):
-        (projectionOperation):
-        (cubeMapFaceSelection):
-        (cubeMapCoordinateTransformation):
-        (cubeMapDerivativeTransformation):
-        (scaleFactorOperation):
-        (levelOfDetailOperation):
-        (nearest):
-        (imageLevelSelection):
-        (strqaToUVWATransformation):
-        (rne):
-        (uvwaToIJKLNTransformationAndArrayLayerSelection):
-        (integerTexelCoordinateOperations):
-        (mirror):
-        (transform):
-        (wrappingOperation):
-        (calculateLambda):
-        (shouldBeBorder):
-        (accessColor):
-        (computeTau):
-        (reduce):
-        (texelGathering):
-        (computeColorFromLevel):
-        (texelFiltering):
-        (castToInnerTypeForGather):
-        (gatherTexture):
-        (castToInnerType):
-        (sampleTexture):
-        * WebGPUShadingLanguageRI/index.html:
-
-2018-09-04  Alexey Proskuryakov  <ap@apple.com>
-
-        Removing an accidentally committed debugging print.
-
-        * Scripts/run-leaks: (runLeaks)
-
-2018-09-04  Alexey Proskuryakov  <ap@apple.com>
-
-        run-leaks should run leaks with --list (on Mojave)
-        https://bugs.webkit.org/show_bug.cgi?id=187716
-        <rdar://problem/42261676>
-
-        Reviewed by Lucas Forschler.
-
-        Also enabled dumping memgraphs. We'll be pruning these aggressively, as they take
-        significant space.
-
-        * Scripts/run-leaks:
-        (main): Added an option to store memgraphs.
-        (runLeaks): As there is no way to test whether the new format is supported in advance,
-        we have to try with --list first, and retry if that fails. Also, made leaks operate
-        on a memgraph file if we are saving it anyway.
-
-        * Scripts/webkitpy/port/leakdetector.py:
-        (LeakDetector._leaks_args): Pass --memgraph-file to run-leaks.
-        (LeakDetector.leaks_file_name): Removed an incorrect comment.
-        (LeakDetector.memgraph_file_name): Added.
-        (LeakDetector.check_for_leaks): Changed how arguments are passed to _leaks_args.
-        It is a bit ugly that leaks path ends up being computed twice, but this is the least
-        ugly approach that I could find.
-
-        * Scripts/webkitpy/port/leakdetector_unittest.py: Updated for _leaks_args changes.
+        * WebKitTestRunner/InjectedBundle/InjectedBundle.cpp:
+        (WTR::InjectedBundle::reportLiveDocuments):
+        (WTR::InjectedBundle::didReceiveMessageToPage):
 
-2018-09-04  Chris Dumez  <cdumez@apple.com>
+2018-09-17  Chris Dumez  <cdumez@apple.com>
 
-        Add process pool configuration flag to turn on automatic process pre-warming
-        https://bugs.webkit.org/show_bug.cgi?id=189263
-        <rdar://problem/44101941>
+        PSON: window.open() with 'noopener' should only process-swap cross-site, not cross-origin
+        https://bugs.webkit.org/show_bug.cgi?id=189602
+        <rdar://problem/44430549>
 
-        Reviewed by Antti Koivisto.
+        Reviewed by Geoff Garen.
 
         Add API test coverage.
 
-        * TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj:
-        * TestWebKitAPI/Tests/WebKitCocoa/ProcessPreWarming.mm: Renamed from Tools/TestWebKitAPI/Tests/WebKitCocoa/InitialWarmedProcessUsed.mm.
-        (TEST):
         * TestWebKitAPI/Tests/WebKitCocoa/ProcessSwapOnNavigation.mm:
-        * TestWebKitAPI/Tests/WebKitCocoa/SetMaximumPrewarmedProcessCount.mm: Removed.
-
-2018-09-04  Keith Miller  <keith_miller@apple.com>
 
-        Perl doesn't like calling keys on a reference
-        https://bugs.webkit.org/show_bug.cgi?id=189261
+2018-09-17  Philippe Normand  <pnormand@igalia.com>
 
-        Reviewed by Michael Saboff.
+        [JHBuild] Update to GStreamer 1.14.3
+        https://bugs.webkit.org/show_bug.cgi?id=189450
 
-        * Scripts/test262/Runner.pm:
-        (UpdateResults):
-
-2018-09-04  Jer Noble  <jer.noble@apple.com>
-
-        REGRESSION (r234081): TestWebKitAPI.VideoControlsManager.VideoControlsManagerAudioElementFollowingUserInteraction is a flaky timeout
-        https://bugs.webkit.org/show_bug.cgi?id=187972
-        <rdar://problem/42667737>
+        Reviewed by Xabier Rodriguez-Calvar.
 
-        Reviewed by Darin Adler.
+        * gstreamer/jhbuild.modules: Bump GStreamer modules from 1.14.1 to 1.14.3.
 
-        Ensure that listeners are added in time to catch messages broadcast at them.
+2018-09-17  Philippe Normand  <pnormand@igalia.com>
 
-        * TestWebKitAPI/Tests/WebKitCocoa/VideoControlsManager.mm:
-        (TestWebKitAPI::TEST):
+        [JHBuild] Doesn't check sha256 sums
+        https://bugs.webkit.org/show_bug.cgi?id=189646
 
-2018-09-04  Ross Kirsling  <ross.kirsling@sony.com>
+        Reviewed by Michael Catanzaro.
 
-        Update googletest
-        https://bugs.webkit.org/show_bug.cgi?id=175722
+        * gstreamer/jhbuild.modules: Properly set the sha256: prefix on hashes.
 
-        Reviewed by Brent Fulgham.
+2018-09-17  Sihui Liu  <sihui_liu@apple.com>
 
-        * TestWebKitAPI/config.h:
-        Instead of modifying the gtest source, add the necessary WTF includes here.
+        Move IndexedDB to Network Process
+        https://bugs.webkit.org/show_bug.cgi?id=189415
+        <rdar://problem/44396973>
 
-        * TestWebKitAPI/Configurations/Base.xcconfig
-        Remove obsoleted definitions GTEST_HAS_TR1_TUPLE=0 and GTEST_HAS_RTTI=0.
+        Reviewed by Chris Dumez.
 
-2018-09-03  Pablo Saavedra  <psaavedra@igalia.com>
+        * TestWebKitAPI/Tests/WebKitCocoa/IDBDeleteRecovery.mm:
+        (TEST):
+        * TestWebKitAPI/Tests/WebKitCocoa/IDBIndexUpgradeToV2.mm:
+        (TEST):
+        * TestWebKitAPI/Tests/WebKitCocoa/IndexedDBDatabaseProcessKill.mm:
+        (TEST):
+        * TestWebKitAPI/Tests/WebKitCocoa/IndexedDBMultiProcess.mm:
+        (TEST):
+        * TestWebKitAPI/Tests/WebKitCocoa/IndexedDBPersistence.mm:
+        (TEST):
 
-        [WPE] Add more WPE bots to dashboard
-        https://bugs.webkit.org/show_bug.cgi?id=189245
+2018-09-08  Darin Adler  <darin@apple.com>
 
-        Reviewed by Daniel Bates.
+        Streamline JSRetainPtr, fix leaks of JSString and JSGlobalContext
+        https://bugs.webkit.org/show_bug.cgi?id=189455
 
-        There are two missing bots which are not listed in the dashboard for
-        a while: WPE Linux 64-bit Debug (Build) and WPE Linux 64-bit Debug (Tests)
+        Reviewed by Keith Miller.
 
-        * BuildSlaveSupport/build.webkit.org-config/public_html/dashboard/Scripts/WebKitBuildbot.js:
-        (WebKitBuildbot):
+        There is a lot of copied and pasted code for WebKit vs. Legacy WebKit
+        testing and even for macOS vs. iOS vs. Windows platform-specific code.
+        For now, this patch just makes corresponding changes to the copied code.
+        Later we might get better results by merging more code instead of having
+        all these separate copies.
+
+        * DumpRenderTree/AccessibilityController.cpp:
+        (AccessibilityController::makeWindowObject): Use the adopt function
+        instead of the special Adopt constructor of JSRetainPtr.
+
+        * DumpRenderTree/AccessibilityTextMarker.cpp: Removed unneeded include.
+
+        * DumpRenderTree/AccessibilityUIElement.cpp: Ditto.
+        (allAttributesCallback): Don't adopt at this level; changed the
+        underlying function to return a JSRetainPtr so the adopt is right next
+        to the call to the create or copy function.
+        (attributesOfLinkedUIElementsCallback): Ditto.
+        (attributesOfDocumentLinksCallback): Ditto.
+        (attributesOfChildrenCallback): Ditto.
+        (parameterizedAttributeNamesCallback): Ditto.
+        (attributesOfColumnHeadersCallback): Ditto.
+        (attributesOfRowHeadersCallback): Ditto.
+        (attributesOfColumnsCallback): Ditto.
+        (attributesOfRowsCallback): Ditto.
+        (attributesOfVisibleCellsCallback): Ditto.
+        (attributesOfHeaderCallback): Ditto.
+        (rowIndexRangeCallback): Ditto.
+        (columnIndexRangeCallback): Ditto.
+        (rangeForLineCallback): Ditto.
+        (boundsForRangeCallback): Ditto.
+        (rangeForPositionCallback): Ditto.
+        (stringForRangeCallback): Ditto.
+        (attributedStringForRangeCallback): Ditto.
+        (uiElementCountForSearchPredicateCallback): Use the free adopt
+        function instead of the adopt member function.
+        (uiElementForSearchPredicateCallback): Ditto.
+        (selectTextWithCriteriaCallback): Don't adopt at this level.
+        (attributedStringForElementCallback): Ditto.
+        (setValueCallback): Use free adopt.
+        (stringAttributeValueCallback): Don't adopt at this level.
+        (uiElementArrayAttributeValueCallback): Ditto.
+        (uiElementAttributeValueCallback): Ditto.
+        (stringForTextMarkerRangeCallback): Ditto.
+        (attributedStringForTextMarkerRangeCallback): Ditto.
+        (attributedStringForTextMarkerRangeWithOptionsCallback): Ditto.
+        (getARIADropEffectsCallback): Ditto.
+        (getClassListCallback): Ditto.
+        (getRoleCallback): Ditto.
+        (getSubroleCallback): Ditto.
+        (getRoleDescriptionCallback): Ditto.
+        (getComputedRoleStringCallback): Ditto.
+        (getTitleCallback): Ditto.
+        (getDescriptionCallback): Ditto.
+        (getStringValueCallback): Ditto.
+        (getLanguageCallback): Ditto.
+        (getHelpTextCallback): Ditto.
+        (getOrientationCallback): Ditto.
+        (getPathDescriptionCallback): Ditto.
+        (getSelectedTextRangeCallback): Ditto.
+        (speakAsCallback): Ditto.
+        (getValueDescriptionCallback): Ditto.
+        (getAccessibilityValueCallback): Ditto.
+        (getDocumentEncodingCallback): Ditto.
+        (getDocumentURICallback): Ditto.
+        (getURLCallback): Ditto.
+        (characterAtOffsetCallback): Ditto.
+        (wordAtOffsetCallback): Ditto.
+        (lineAtOffsetCallback): Ditto.
+        (sentenceAtOffsetCallback): Ditto.
+        (stringForSelectionCallback): Ditto.
+        (getIdentifierCallback): Ditto.
+        (getTraitsCallback): Ditto.
+        (supportedActionsCallback): Ditto.
+        (mathPostscriptsDescriptionCallback): Ditto.
+        (mathPrescriptsDescriptionCallback): Ditto.
+        (AccessibilityUIElement::rangeForLine): Updated to return JSRetainPtr.
+        (AccessibilityUIElement::rangeForPosition): Ditto.
+        (AccessibilityUIElement::speakAs): Ditto.
+        (AccessibilityUIElement::pathDescription const): Ditto.
+        (AccessibilityUIElement::stringForTextMarkerRange): Ditto.
+        (AccessibilityUIElement::attributedStringForTextMarkerRange): Ditto.
+        (AccessibilityUIElement::attributedStringForTextMarkerRangeWithOptions): Ditto.
 
-2018-09-01  Michael Catanzaro  <mcatanzaro@igalia.com>
+        * DumpRenderTree/AccessibilityUIElement.h: Use JSRetainPtr instead of
+        raw pointers for the results of the functions that create strings.
+        The old way was difficult to get right; the functions didn't even follow
+        the create/copy naming rule.
 
-        [WPE] 2.21.91 fails to build with ENABLE_MINIBROWSER
-        https://bugs.webkit.org/show_bug.cgi?id=189006
+        * DumpRenderTree/Bindings/CodeGeneratorDumpRenderTree.pm:
+        (_platformTypeConstructor): Use the adopt function instead of the
+        JSRetainPtr adopt constructor. Also nullptr instead of 0.
 
-        Reviewed by Darin Adler.
+        * DumpRenderTree/GCController.cpp:
+        (GCController::makeWindowObject): Use adopt function instead of constructor.
 
-        The MiniBrowser source code is missing from the tarball. Oops! Add it.
+        * DumpRenderTree/TestRunner.cpp:
+        (pathToLocalResourceCallback): Use adopt function instead of constructor.
+        (addDisallowedURLCallback): Ditto.
+        (addURLToRedirectCallback): Ditto.
+        (clearApplicationCacheForOriginCallback): Ditto.
+        (applicationCacheDiskUsageForOriginCallback): Ditto.
+        (decodeHostNameCallback): Don't adopt at this level.
+        (encodeHostNameCallback): Ditto.
+        (execCommandCallback): Use adopt instead of JSRetainPtr::adopt.
+        (findStringCallback): Use adopt function instead of constructor.
+        (isCommandEnabledCallback): Ditto.
+        (overridePreferenceCallback): Ditto.
+        (queueLoadCallback): Ditto.
+        (queueLoadHTMLStringCallback): Ditto.
+        (queueLoadingScriptCallback): Ditto.
+        (queueNonLoadingScriptCallback): Ditto.
+        (setAuthenticationPasswordCallback): Ditto.
+        (setAuthenticationUsernameCallback): Ditto.
+        (setDomainRelaxationForbiddenForURLSchemeCallback): Ditto.
+        (setMockGeolocationPositionUnavailableErrorCallback): Ditto.
+        (setPOSIXLocaleCallback): Ditto.
+        (setPersistentUserStyleSheetLocationCallback): Ditto.
+        (setUserStyleSheetLocationCallback): Ditto.
+        (setValueForUserCallback): Ditto.
+        (setWillSendRequestClearHeaderCallback): Ditto.
+        (setPageVisibilityCallback): Ditto.
+        (evaluateInWebInspectorCallback): Ditto.
+        (evaluateScriptInIsolatedWorldCallback): Ditto.
+        (evaluateScriptInIsolatedWorldAndReturnValueCallback): Ditto.
+        (addOriginAccessWhitelistEntryCallback): Ditto.
+        (removeOriginAccessWhitelistEntryCallback): Ditto.
+        (setScrollbarPolicyCallback): Ditto.
+        (addUserScriptCallback): Ditto.
+        (addUserStyleSheetCallback): Ditto.
+        (apiTestNewWindowDataLoadBaseURLCallback): Ditto.
+        (authenticateSessionCallback): Ditto.
+        (getTitleTextDirectionCallback): Ditto.
+        (getInspectorTestStubURLCallback): Ditto.
+        (simulateLegacyWebNotificationClickCallback): Ditto.
+        (setTextDirectionCallback): Ditto.
+        (grantWebNotificationPermissionCallback): Ditto.
+        (denyWebNotificationPermissionCallback): Ditto.
+        (accummulateLogsForChannel): Ditto.
+        (runUIScriptCallback): Ditto.
+        (TestRunner::makeWindowObject): Ditto.
+        (TestRunner::uiScriptDidComplete): Ditto.
+        (TestRunner::setOpenPanelFiles): Ditto.
+
+        * DumpRenderTree/TestRunner.h: Use JSRetainPtr instead of
+        raw pointers for the results of the functions that create strings.
+
+        * DumpRenderTree/ios/AccessibilityControllerIOS.mm:
+        (AccessibilityController::platformName const): Use the adopt
+        function instead of the adopt constructor.
 
-        * wpe/manifest.txt.in:
+        * DumpRenderTree/ios/AccessibilityUIElementIOS.mm:
+        (createEmptyJSString): Added. This helper function makes reduces the
+        repetitive code to create empty strings.
+        (concatenateAttributeAndValue): Changed to return a JSRetainPtr.
+        (AccessibilityUIElement::identifier): Ditto.
+        (AccessibilityUIElement::traits): Ditto.
+        (AccessibilityUIElement::url): Ditto.
+        (AccessibilityUIElement::speakAs): Ditto.
+        (AccessibilityUIElement::stringForSelection): Ditto.
+        (AccessibilityUIElement::stringForRange): Ditto.
+        (AccessibilityUIElement::attributedStringForRange): Ditto.
+        (AccessibilityUIElement::attributedStringForElement): Ditto.
+        (AccessibilityUIElement::pathDescription const): Ditto.
+        (AccessibilityUIElement::stringForTextMarkerRange): Ditto.
+        (AccessibilityUIElement::attributedStringForTextMarkerRange): Ditto.
+        (AccessibilityUIElement::attributedStringForTextMarkerRangeWithOptions): Ditto.
+        (AccessibilityUIElement::attributesOfLinkedUIElements): Ditto.
+        (AccessibilityUIElement::attributesOfDocumentLinks): Ditto.
+        (AccessibilityUIElement::attributesOfChildren): Ditto.
+        (AccessibilityUIElement::allAttributes): Ditto.
+        (AccessibilityUIElement::stringAttributeValue): Ditto.
+        (AccessibilityUIElement::parameterizedAttributeNames): Ditto.
+        (AccessibilityUIElement::role): Ditto.
+        (AccessibilityUIElement::subrole): Ditto.
+        (AccessibilityUIElement::roleDescription): Ditto.
+        (AccessibilityUIElement::computedRoleString): Ditto.
+        (AccessibilityUIElement::title): Ditto.
+        (AccessibilityUIElement::description): Ditto.
+        (AccessibilityUIElement::orientation const): Ditto.
+        (AccessibilityUIElement::stringValue): Ditto.
+        (AccessibilityUIElement::language): Ditto.
+        (AccessibilityUIElement::helpText const): Ditto.
+        (AccessibilityUIElement::valueDescription): Ditto.
+        (AccessibilityUIElement::ariaDropEffects const): Ditto.
+        (AccessibilityUIElement::boundsForRange): Ditto.
+        (AccessibilityUIElement::attributesOfColumnHeaders): Ditto.
+        (AccessibilityUIElement::attributesOfRowHeaders): Ditto.
+        (AccessibilityUIElement::attributesOfColumns): Ditto.
+        (AccessibilityUIElement::attributesOfRows): Ditto.
+        (AccessibilityUIElement::attributesOfVisibleCells): Ditto.
+        (AccessibilityUIElement::attributesOfHeader): Ditto.
+        (AccessibilityUIElement::rowIndexRange): Ditto.
+        (AccessibilityUIElement::columnIndexRange): Ditto.
+        (AccessibilityUIElement::selectedTextRange): Ditto.
+        (AccessibilityUIElement::accessibilityValue const): Ditto.
+        (AccessibilityUIElement::documentEncoding): Ditto.
+        (AccessibilityUIElement::documentURI): Ditto.
+        (AccessibilityUIElement::selectTextWithCriteria): Ditto.
+        (AccessibilityUIElement::classList const): Ditto.
+
+        * DumpRenderTree/mac/AccessibilityCommonMac.h: Changed the
+        createJSStringRef method to return a JSRetainPtr.
+        * DumpRenderTree/mac/AccessibilityCommonMac.mm:
+        (-[NSString createJSStringRef]): Ditto.
+        (searchPredicateParameterizedAttributeForSearchCriteria): Use the
+        adopt function instead of the adopt constructor.
+        * DumpRenderTree/mac/AccessibilityControllerMac.mm:
+        (AccessibilityController::platformName const): Ditto.
 
-2018-09-01  Yoshiaki Jitsukawa  <yoshiaki.jitsukawa@sony.com>
+        * DumpRenderTree/mac/AccessibilityNotificationHandler.mm:
+        (-[NSString createJSStringRef]): Return a JSRetainPtr.
+        (makeValueRefForValue): Updated for the above, no adopt needed here now.
+        (makeObjectRefForDictionary): Ditto.
+        (-[AccessibilityNotificationHandler _notificationReceived:]): Ditto.
 
-        [WinCairo] Add CryptoDigestOpenSSL
-        https://bugs.webkit.org/show_bug.cgi?id=188978
+        * DumpRenderTree/mac/AccessibilityUIElementMac.mm:
+        (createEmptyJSString): Added. This helper function makes reduces the
+        repetitive code to create empty strings.
+        (concatenateAttributeAndValue): Changed to return a JSRetainPtr.
+        (descriptionOfElements): Ditto.
+        (selectTextParameterizedAttributeForCriteria): Ditto.
+        (AccessibilityUIElement::attributesOfLinkedUIElements): Ditto.
+        (AccessibilityUIElement::attributesOfDocumentLinks): Ditto.
+        (AccessibilityUIElement::attributesOfChildren): Ditto.
+        (AccessibilityUIElement::allAttributes): Ditto.
+        (AccessibilityUIElement::stringAttributeValue): Ditto.
+        (AccessibilityUIElement::parameterizedAttributeNames): Ditto.
+        (AccessibilityUIElement::role): Ditto.
+        (AccessibilityUIElement::subrole): Ditto.
+        (AccessibilityUIElement::roleDescription): Ditto.
+        (AccessibilityUIElement::computedRoleString): Ditto.
+        (AccessibilityUIElement::title): Ditto.
+        (AccessibilityUIElement::description): Ditto.
+        (AccessibilityUIElement::orientation const): Ditto.
+        (AccessibilityUIElement::stringValue): Ditto.
+        (AccessibilityUIElement::language): Ditto.
+        (AccessibilityUIElement::helpText const): Ditto.
+        (AccessibilityUIElement::valueDescription): Ditto.
+        (AccessibilityUIElement::speakAs): Ditto.
+        (AccessibilityUIElement::classList const): Ditto.
+        (AccessibilityUIElement::ariaDropEffects const): Ditto.
+        (AccessibilityUIElement::rangeForLine): Ditto.
+        (AccessibilityUIElement::rangeForPosition): Ditto.
+        (AccessibilityUIElement::boundsForRange): Ditto.
+        (AccessibilityUIElement::stringForRange): Ditto.
+        (AccessibilityUIElement::attributedStringForRange): Ditto.
+        (AccessibilityUIElement::selectTextWithCriteria): Ditto.
+        (AccessibilityUIElement::attributesOfColumnHeaders): Ditto.
+        (AccessibilityUIElement::attributesOfRowHeaders): Ditto.
+        (AccessibilityUIElement::attributesOfColumns): Ditto.
+        (AccessibilityUIElement::attributesOfRows): Ditto.
+        (AccessibilityUIElement::attributesOfVisibleCells): Ditto.
+        (AccessibilityUIElement::attributesOfHeader): Ditto.
+        (AccessibilityUIElement::rowIndexRange): Ditto.
+        (AccessibilityUIElement::columnIndexRange): Ditto.
+        (AccessibilityUIElement::pathDescription const): Ditto.
+        (AccessibilityUIElement::selectedTextRange): Ditto.
+        (AccessibilityUIElement::accessibilityValue const): Ditto.
+        (AccessibilityUIElement::documentEncoding): Ditto.
+        (AccessibilityUIElement::documentURI): Ditto.
+        (AccessibilityUIElement::url): Ditto.
+        (AccessibilityUIElement::stringForTextMarkerRange): Ditto.
+        (createJSStringRef): Ditto.
+        (AccessibilityUIElement::attributedStringForTextMarkerRange): Ditto.
+        (AccessibilityUIElement::attributedStringForTextMarkerRangeWithOptions): Ditto.
+        (AccessibilityUIElement::supportedActions): Ditto.
+        (AccessibilityUIElement::mathPostscriptsDescription const): Ditto.
+        (AccessibilityUIElement::mathPrescriptsDescription const): Ditto.
 
-        Reviewed by Darin Adler.
+        * DumpRenderTree/mac/FrameLoadDelegate.mm:
+        (-[FrameLoadDelegate didClearWindowObjectForFrame:inIsolatedWorld:]):
+        Use the adopt function instead of the adopt constructor.
+        * DumpRenderTree/mac/TestRunnerMac.mm:
+        (originsArrayToJS): Ditto.
+        (TestRunner::copyDecodedHostName): Return a JSRetainPtr.
+        (TestRunner::copyEncodedHostName): Ditto.
+        (TestRunner::pathToLocalResource): Ditto.
+        (TestRunner::queueLoad): Use adopt function instead of constructor.
+        (TestRunner::findString): Ditto.
+        (TestRunner::inspectorTestStubURL): Return a JSRetainPtr.
 
-        Add API tests for the PAL::CryptoDigest class.
+        * DumpRenderTree/win/AccessibilityControllerWin.cpp:
+        (AccessibilityController::winNotificationReceived): Use adopt function
+        instead of adopt constructor.
+        (AccessibilityController::platformName const): Ditto.
+
+        * DumpRenderTree/win/AccessibilityUIElementWin.cpp:
+        (createEmptyJSString): Added. This helper function makes reduces the
+        repetitive code to create empty strings.
+        (AccessibilityUIElement::allAttributes): Changed to return a JSRetainPtr.
+        (AccessibilityUIElement::attributesOfLinkedUIElements): Ditto.
+        (AccessibilityUIElement::attributesOfDocumentLinks): Ditto.
+        (AccessibilityUIElement::attributesOfChildren): Ditto.
+        (AccessibilityUIElement::parameterizedAttributeNames): Ditto.
+        (AccessibilityUIElement::role): Ditto.
+        (AccessibilityUIElement::subrole): Ditto.
+        (AccessibilityUIElement::roleDescription): Ditto.
+        (AccessibilityUIElement::computedRoleString): Ditto.
+        (AccessibilityUIElement::title): Ditto.
+        (AccessibilityUIElement::description): Ditto.
+        (AccessibilityUIElement::stringValue): Ditto.
+        (AccessibilityUIElement::language): Ditto.
+        (AccessibilityUIElement::helpText const): Ditto.
+        (AccessibilityUIElement::valueDescription): Ditto.
+        (AccessibilityUIElement::ariaDropEffects const): Ditto.
+        (AccessibilityUIElement::orientation const): Ditto.
+        (AccessibilityUIElement::attributesOfColumnHeaders): Ditto.
+        (AccessibilityUIElement::attributesOfRowHeaders): Ditto.
+        (AccessibilityUIElement::attributesOfColumns): Ditto.
+        (AccessibilityUIElement::attributesOfRows): Ditto.
+        (AccessibilityUIElement::attributesOfVisibleCells): Ditto.
+        (AccessibilityUIElement::attributesOfHeader): Ditto.
+        (AccessibilityUIElement::rowIndexRange): Ditto.
+        (AccessibilityUIElement::columnIndexRange): Ditto.
+        (AccessibilityUIElement::boundsForRange): Ditto.
+        (AccessibilityUIElement::stringForRange): Ditto.
+        (AccessibilityUIElement::attributedStringForRange): Ditto.
+        (AccessibilityUIElement::selectTextWithCriteria): Ditto.
+        (AccessibilityUIElement::selectedTextRange): Ditto.
+        (AccessibilityUIElement::stringAttributeValue): Ditto.
+        (AccessibilityUIElement::accessibilityValue const): Ditto.
+        (AccessibilityUIElement::documentEncoding): Ditto.
+        (AccessibilityUIElement::documentURI): Ditto.
+        (AccessibilityUIElement::url): Ditto.
+        (AccessibilityUIElement::classList const): Ditto.
 
-        * TestWebKitAPI/PlatformWin.cmake:
-        Add CryptoDigest.cpp to TestWebCoreLib_Sources.
+        * DumpRenderTree/win/TestRunnerWin.cpp:
+        (TestRunner::copyDecodedHostName): Changed to return a JSRetainPtr.
+        (TestRunner::copyEncodedHostName): Ditto.
+        (TestRunner::pathToLocalResource): Ditto.
+        (TestRunner::queueLoad): Ditto.
+        (TestRunner::findString): Ditto.
+        (TestRunner::inspectorTestStubURL): Ditto.
 
-        Define -DSTATICALLY_LINKED_WITH_PAL=1 to avoid warnings since
-        without the definition the functions marked with PAL_EXPORT are supposed to be
-        imported from another shared library, but as it is, TestWebCoreLib
-        links PAL statically.
+        * TestRunnerShared/Bindings/JSWrapper.h:
+        (WTR::setProperty): Use adopt function instead of constructor.
+        * TestRunnerShared/UIScriptContext/UIScriptContext.cpp:
+        (UIScriptContext::UIScriptContext): Ditto.
+        (UIScriptContext::runUIScript): Ditto.
+        * TestRunnerShared/UIScriptContext/UIScriptController.cpp:
+        (WTR::toDeviceOrientation): Ditto.
 
-        * TestWebKitAPI/Tests/WebCore/CryptoDigest.cpp: Added.
-        (TestWebKitAPI::toHex):
-        (TestWebKitAPI::expect):
-        (TestWebKitAPI::expectSHA1):
-        (TestWebKitAPI::expectSHA224):
-        (TestWebKitAPI::expectSHA256):
-        (TestWebKitAPI::expectSHA384):
-        (TestWebKitAPI::expectSHA512):
-        (TestWebKitAPI::TEST):
+        * TestRunnerShared/cocoa/LayoutTestSpellChecker.mm:
+        (nsTextCheckingType): Changed this function to not take an rvalue
+        reference to a JSRetainPtr since it doesn't take ownership.
+        (-[LayoutTestSpellChecker setResultsFromJSObject:inContext:]): Updated
+        for the above change.
+
+        * TestWebKitAPI/JavaScriptTest.cpp:
+        (TestWebKitAPI::javaScriptCallback): Use adopt function instead of
+        adopt member function.
+
+        * TestWebKitAPI/Tests/WebKit/WKPageIsPlayingAudio.cpp: Removed unneeded include.
+        * TestWebKitAPI/Tests/WebKitCocoa/PictureInPictureDelegate.mm: Ditto.
+        * TestWebKitAPI/Tests/mac/DOMNodeFromJSObject.mm: Use adopt function instead of
+        adopt constructor.
+        * TestWebKitAPI/Tests/mac/JSWrapperForNodeInWebFrame.mm: Ditto.
+
+        * WebKitTestRunner/InjectedBundle/AccessibilityController.cpp:
+        Removed unneeded includes.
+        * WebKitTestRunner/InjectedBundle/AccessibilityTextMarker.cpp: Ditto.
+        * WebKitTestRunner/InjectedBundle/AccessibilityTextMarkerRange.cpp: Ditto.
+        * WebKitTestRunner/InjectedBundle/AccessibilityUIElement.cpp: Ditto.
+
+        * WebKitTestRunner/InjectedBundle/Bindings/CodeGeneratorTestRunner.pm:
+        (_platformTypeConstructor): Use the adopt function instead of the
+        JSRetainPtr adopt constructor. Also nullptr instead of 0.
+
+        * WebKitTestRunner/InjectedBundle/EventSendingController.cpp:
+        (WTR::arrayLength): Use adopt function instead of constructor.
+        (WTR::parseModifierArray): Ditto.
+        * WebKitTestRunner/InjectedBundle/InjectedBundlePage.cpp:
+        (WTR::propertyValue): Ditto.
+        (WTR::dumpPath): Ditto.
+        (WTR::toJS): Ditto.
+        * WebKitTestRunner/InjectedBundle/TestRunner.cpp:
+        (WTR::TestRunner::findString): Ditto.
+        (WTR::TestRunner::statisticsDidRunTelemetryCallback): Ditto.
+        (WTR::TestRunner::callDidReceiveAllStorageAccessEntriesCallback): Ditto.
+        (WTR::TestRunner::setOpenPanelFiles): Ditto.
+        * WebKitTestRunner/InjectedBundle/ios/AccessibilityControllerIOS.mm:
+        (WTR::AccessibilityController::platformName): Ditto.
 
-2018-09-01  Dan Bernstein  <mitz@apple.com>
+        * WebKitTestRunner/InjectedBundle/ios/AccessibilityUIElementIOS.mm:
+        (-[NSString createJSStringRef]): Return a JSRetainPtr.
+        This fixes storage leaks in many functions below, which were using
+        the create function and putting the value right into a JSRetainPtr
+        without adoption.
+        (WTR::createEmptyJSString): Added. This helper function reduces the
+        repetitive code to create empty strings.
+        (WTR::concatenateAttributeAndValue): Return a JSRetainPtr.
+        (WTR::AccessibilityUIElement::attributesOfLinkedUIElements):
+        Use createEmptyJSString, fixing a storage leak due to missing adoption.
+        (WTR::AccessibilityUIElement::attributesOfDocumentLinks): Ditto.
+        (WTR::AccessibilityUIElement::attributesOfChildren): Ditto.
+        (WTR::AccessibilityUIElement::allAttributes): Ditto.
+        (WTR::AccessibilityUIElement::stringAttributeValue): Ditto.
+        (WTR::AccessibilityUIElement::parameterizedAttributeNames): Ditto.
+        (WTR::AccessibilityUIElement::role): Ditto.
+        (WTR::AccessibilityUIElement::subrole): Ditto.
+        (WTR::AccessibilityUIElement::roleDescription): Ditto.
+        (WTR::AccessibilityUIElement::computedRoleString): Ditto.
+        (WTR::AccessibilityUIElement::title): Ditto.
+        (WTR::AccessibilityUIElement::orientation const): Ditto.
+        (WTR::AccessibilityUIElement::language): Ditto.
+        (WTR::AccessibilityUIElement::valueDescription): Ditto.
+        (WTR::AccessibilityUIElement::ariaDropEffects const): Ditto.
+        (WTR::AccessibilityUIElement::boundsForRange): Ditto.
+        (WTR::AccessibilityUIElement::stringForRange): Removed unneeded null check.
+        (WTR::AccessibilityUIElement::attributedStringForRange): Ditto.
+        (WTR::AccessibilityUIElement::attributesOfColumnHeaders):
+        Use createEmptyJSString, fixing a storage leak due to missing adoption.
+        (WTR::AccessibilityUIElement::attributesOfRowHeaders): Ditto.
+        (WTR::AccessibilityUIElement::attributesOfColumns): Ditto.
+        (WTR::AccessibilityUIElement::attributesOfRows): Ditto.
+        (WTR::AccessibilityUIElement::attributesOfVisibleCells): Ditto.
+        (WTR::AccessibilityUIElement::attributesOfHeader): Ditto.
+        (WTR::AccessibilityUIElement::selectedTextRange): Ditto.
+        (WTR::AccessibilityUIElement::accessibilityValue const): Ditto.
+        (WTR::AccessibilityUIElement::documentEncoding): Ditto.
+        (WTR::AccessibilityUIElement::documentURI): Ditto.
+        (WTR::AccessibilityUIElement::stringForSelection): Removed unneeded
+        null check.
+        (WTR::AccessibilityUIElement::stringForTextMarkerRange):
+        Use createEmptyJSString, fixing a storage leak due to missing adoption.
+
+        * WebKitTestRunner/InjectedBundle/mac/AccessibilityCommonMac.h:
+        Use JSRetainPtr for the return value of createJSStringRef.
+        * WebKitTestRunner/InjectedBundle/mac/AccessibilityCommonMac.mm:
+        (-[NSString createJSStringRef]): Ditto.
+        (WTR::searchPredicateParameterizedAttributeForSearchCriteria): Use the
+        adopt function instead of the adopt constructor.
+        * WebKitTestRunner/InjectedBundle/mac/AccessibilityControllerMac.mm:
+        (WTR::AccessibilityController::platformName): Ditto
 
-        [Cocoa] Crash using KVO for 'serverTrust' property of WKWebView - "this class is not key value coding-compliant for the key serverTrust"
-        https://bugs.webkit.org/show_bug.cgi?id=189222
-        <rdar://problem/33283179>
+        * WebKitTestRunner/InjectedBundle/mac/AccessibilityNotificationHandler.mm:
+        (-[NSString createJSStringRef]): Return a JSRetainPtr.
+        (makeValueRefForValue): Updated for above by removing adopt.
+        (makeObjectRefForDictionary): Ditto.
+        (-[AccessibilityNotificationHandler _notificationReceived:]): Ditto.
 
-        Reviewed by Sam Weinig.
+        * WebKitTestRunner/InjectedBundle/mac/AccessibilityUIElementMac.mm:
+        (WTR::createEmptyJSString): Added. This helper function makes reduces the
+        repetitive code to create empty strings.
+        (WTR::concatenateAttributeAndValue): Return a JSRetainPtr.
+        (WTR::descriptionOfElements): Ditto.
+        (WTR::selectTextParameterizedAttributeForCriteria): Ditto.
+        (WTR::AccessibilityUIElement::accessibilityValue const): Use createEmptyJSString,
+        fixing a storage leak due to missing adoption.
+        (WTR::AccessibilityUIElement::documentEncoding): Ditto.
+        (WTR::AccessibilityUIElement::documentURI): Ditto.
+        (WTR::createJSStringRef): Return a JSRetainPtr.
 
-        * TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj:
-        * TestWebKitAPI/Tests/WebKitCocoa/WKWebViewServerTrustKVC.mm: Added.
-        (TEST):
+        * WebKitTestRunner/StringFunctions.h:
+        (WTR::toWK): Take a const& to eliminate a little bit of reference count churn.
+        (WTR::toJS): Use the adopt function instead of the constructor.
 
-2018-08-31  Aditya Keerthi  <akeerthi@apple.com>
+2018-09-14  Dean Jackson  <dino@grorg.org>
 
-        Unreviewed, add an alternate email for Aditya Keerthi.
+        Add Justin Michaud.
 
         * Scripts/webkitpy/common/config/contributors.json:
 
-2018-08-31  Wenson Hsieh  <wenson_hsieh@apple.com>
+2018-09-14  Ryan Haddad  <ryanhaddad@apple.com>
 
-        [iOS] Consolidate the implementations of readString, stringForType, and readURL in PlatformPasteboardIOS.mm
-        https://bugs.webkit.org/show_bug.cgi?id=189054
-        <rdar://problem/43819779>
+        bots should pass --clobber-old-results to run-webkit-tests
+        https://bugs.webkit.org/show_bug.cgi?id=189623
 
-        Reviewed by Tim Horton.
+        Reviewed by Aakash Jain.
 
-        See WebCore/ChangeLog for more detail.
+        In order to ensure that bots to not produce confusing results due to transient execptions,
+        or run out of space if they continually hit an exception, old layout test results should be clobbered.
 
-        * TestWebKitAPI/Tests/WebKitCocoa/CopyURL.mm:
-        (createWebViewWithCustomPasteboardDataEnabled):
+        * BuildSlaveSupport/build.webkit.org-config/steps.py:
+        (RunWebKitTests):
 
-2018-08-31  Commit Queue  <commit-queue@webkit.org>
+2018-09-13  Ryan Haddad  <ryanhaddad@apple.com>
 
-        Unreviewed, rolling out r235565.
-        https://bugs.webkit.org/show_bug.cgi?id=189212
+        Unreviewed, rolling out r235954.
 
-        Broke the WHLSL build (Requested by litherum on #webkit).
+        Breaks the watchOS build.
 
         Reverted changeset:
 
-        "[WHLSL] Remove useless code in NameResolver"
-        https://bugs.webkit.org/show_bug.cgi?id=189176
-        https://trac.webkit.org/changeset/235565
-
-2018-08-31  Myles C. Maxfield  <mmaxfield@apple.com>
-
-        [WHLSL] Remove useless code in NameResolver
-        https://bugs.webkit.org/show_bug.cgi?id=189176
-
-        Reviewed by Dean Jackson and Thomas Denney.
-
-        The check will always pass becase the standard library contains at least one function with the correct name.
-
-        * WebGPUShadingLanguageRI/NameResolver.js:
-        (NameResolver.prototype.visitDotExpression):
-        (NameResolver.prototype._handlePropertyAccess): Deleted.
-        (NameResolver.prototype.visitIndexExpression): Deleted.
-
-2018-08-31  Simon Fraser  <simon.fraser@apple.com>
-
-        Make it possible to mark tests as leaks in TestExpectations
-        https://bugs.webkit.org/show_bug.cgi?id=189088
+        "Move IndexedDB to Network Process"
+        https://bugs.webkit.org/show_bug.cgi?id=189415
+        https://trac.webkit.org/changeset/235954
 
-        Reviewed by Jon Lee.
-        
-        Have webkitpy parse out "Leak" expectations in TestExpectations, and do the right
-        thing if the test run did not use --world-leaks. Add unit tests for leaks combined
-        with various other result combinations.
-
-        * Scripts/webkitpy/layout_tests/controllers/layout_test_runner.py:
-        (LayoutTestRunner._update_summary_with_result):
-        (LayoutTestRunner._annotate_results_with_additional_failures):
-        * Scripts/webkitpy/layout_tests/models/test_expectations.py:
-        (TestExpectationParser):
-        (TestExpectations): The 'Leak' line was duplicated here, so remove a copy.
-        (TestExpectations.remove_leak_failures):
-        (TestExpectations.matches_an_expected_result):
-        * Scripts/webkitpy/layout_tests/models/test_expectations_unittest.py:
-        (Base.get_basic_tests):
-        * Scripts/webkitpy/port/test.py:
-        (TestList.add_reftest):
-        (unit_test_list):
-
-2018-08-31  Zalan Bujtas  <zalan@apple.com>
-
-        [LFC] Add margin box verification back now that Display::Box has non-computed horizontal margin.
-        https://bugs.webkit.org/show_bug.cgi?id=189193
-
-        Reviewed by Antti Koivisto.
-
-        * LayoutReloaded/misc/LFC-passing-tests.txt:
-
-2018-08-31  Antti Koivisto  <antti@apple.com>
-
-        Replace OptionSet |= and -= operators with add() and remove() functions
-        https://bugs.webkit.org/show_bug.cgi?id=189169
-
-        Reviewed by Anders Carlsson.
-
-        * TestWebKitAPI/Tests/WTF/OptionSet.cpp:
-        (TestWebKitAPI::TEST):
-
-2018-08-31  Daniel Bates  <dabates@apple.com>
-
-        lldb-webkit: KeyError thrown for OptionSet with invalid value
-        https://bugs.webkit.org/show_bug.cgi?id=189070
-
-        Update comment to explain that we return early and do not compute the set of enumerators in the
-        OptionSet when the underlying value of the set is an invalid value.
-
-        Additionally, update the change log entry for r235482 to reflect the updated title for the bug
-        and to better describe the change that was made.
-
-        * lldb/lldb_webkit.py:
-        (WTFOptionSetProvider.update):
+2018-09-13  Daniel Bates  <dabates@apple.com>
 
-2018-08-31  David Kilzer  <ddkilzer@apple.com>
+        Add Copy WebKit Permalink plugin for Sublime Text
+        https://bugs.webkit.org/show_bug.cgi?id=189589
 
-        WebKitLauncher: Move WebKitLauncher.entitlements into Configurations directory
+        Rubber-stamped by Joseph Pecoraro.
 
-        This makes the Configurations folder in the Xcode project match
-        what's in the Configurations directory on disk.  Found by
-        tidy-Xcode-project-file (see Bug 188754).
+        Port the Copy WebKit Permalink Automator service to a Sublime Text plugin. Once installed you can
+        use the plugin to copy to the Clipboard a trac.webkit.org hyperlink to the selected line in the
+        active document with or without blame annotations.
 
-        * WebKitLauncher/Configurations/WebKitLauncher.entitlements: Rename from Tools/WebKitLauncher/WebKitLauncher.entitlements.
-        * WebKitLauncher/WebKitLauncher.xcodeproj/project.pbxproj:
-        - Update project for file move.
+        Once installed, you can Control-click (on Mac) or right-click (on Windows and Linux) on a line and
+        choose Copy WebKit Permalink or Copy WebKit Permalink to Blame from the context menu to copy to the
+        Clipboard a permanent hyperlink to the selected line without or with blame annotations, respectively.
+        On Mac you can also invoke the same functionality using the keyboard shortcuts Command + Shift + Control + C
+        and Command + Shift + Control + Option + C, respectively.
 
-2018-08-31  David Kilzer  <ddkilzer@apple.com>
+        * CopyPermalink/Sublime Text/CopyWebKitPermalink/Context.sublime-menu: Added.
+        * CopyPermalink/Sublime Text/CopyWebKitPermalink/CopyWebKitPermalink.py: Added.
+        * CopyPermalink/Sublime Text/CopyWebKitPermalink/Default (OSX).sublime-keymap: Added.
+        * CopyPermalink/Sublime Text/INSTALL: Added.
+        * CopyPermalink/Xcode/Copy WebKit Permalink.workflow/Contents/Info.plist: Renamed from Tools/CopyPermalink/Copy WebKit Permalink.workflow/Contents/Info.plist.
+        * CopyPermalink/Xcode/Copy WebKit Permalink.workflow/Contents/document.wflow: Renamed from Tools/CopyPermalink/Copy WebKit Permalink.workflow/Contents/document.wflow.
+        * CopyPermalink/Xcode/INSTALL: Renamed from Tools/CopyPermalink/README.
 
-        WebKitLauncher: Fix build failures
+2018-09-13  Carlos Garcia Campos  <cgarcia@igalia.com>
 
-        * WebKitLauncher/Configurations/Base.xcconfig:
-        (ARCHS): Use $(ARCHS_STANDARD_32_64_BIT).
-        (MACOSX_DEPLOYMENT_TARGET): Update from Snow Leopard to Sierra.
+        Unreviewed. Fix WebDriver tests after r235225.
 
-        * WebKitLauncher/WebKitNightlyEnabler.m:
-        (myApplicationWillFinishLaunching):
-        * WebKitLauncher/main.m:
-        (displayErrorAndQuit):
-        (locateSafariBundle):
-        - Use pragma to ignore deprecation warnings.
+        Use wpt serve command to run the http server.
 
-2018-08-31  David Kilzer  <ddkilzer@apple.com>
+        * Scripts/webkitpy/webdriver_tests/webdriver_w3c_web_server.py:
+        (WebDriverW3CWebServer.__init__):
+        (WebDriverW3CWebServer.start):
+        (WebDriverW3CWebServer.stop):
 
-        MobileMiniBrowser: Move test2s.mp4 into Resources directory
+2018-09-13  Carlos Garcia Campos  <cgarcia@igalia.com>
 
-        This makes the Resources folder in the Xcode project match
-        what's in the Resources directory on disk.  Found by
-        tidy-Xcode-project-file (see Bug 188754).
+        [GTK][WPE] Allow to run script dialogs asynchronously in the UI process
+        https://bugs.webkit.org/show_bug.cgi?id=189544
 
-        * MobileMiniBrowser/MobileMiniBrowser.xcodeproj/project.pbxproj:
-        - Update project for file move.
-        * MobileMiniBrowser/MobileMiniBrowserFramework/Resources/test2s.mp4: Rename from Tools/MobileMiniBrowser/MobileMiniBrowserFramework/test2s.mp4.
+        Reviewed by Michael Catanzaro.
 
-2018-08-31  Frederic Wang  <fwang@igalia.com>
+        Add test cases for the new API.
 
-        Bug 182053 - [CSSOM View] Implement standard behavior for scrollingElement
-        https://bugs.webkit.org/show_bug.cgi?id=182053
+        * TestWebKitAPI/Tests/WebKitGLib/TestUIClient.cpp:
+        (testWebViewJavaScriptDialogs):
 
-        Reviewed by Simon Fraser.
+2018-09-12  Fujii Hironori  <Hironori.Fujii@sony.com>
 
-        Add FIXMEs comments to add the new developer flag in the future and  disable it for now.
+        [Win][Clang][ImageDiff] Fix compilation error and warning of PlatformImageCairo.cpp
+        https://bugs.webkit.org/show_bug.cgi?id=189496
+        <rdar://problem/44370981>
 
-        * DumpRenderTree/mac/DumpRenderTree.mm:
-        (enableExperimentalFeatures):
-        (resetWebPreferencesToConsistentValues):
-        * DumpRenderTree/win/DumpRenderTree.cpp:
-        (enableExperimentalFeatures):
-        * WebKitTestRunner/TestController.cpp:
-        (WTR::TestController::resetPreferencesToConsistentValues):
+        Unreviewed warning fix.
 
-2018-08-30  Thomas Denney  <tdenney@apple.com>
+        r235929 has introduced a new compilation warning:
+        > Tools/ImageDiff/cairo/PlatformImageCairo.cpp:34:70: warning: unused parameter ‘imageSize’ [-Wunused-parameter]
 
-        [WHLSL] Fix array indexing behavior
-        https://bugs.webkit.org/show_bug.cgi?id=189175
+        * ImageDiff/cairo/PlatformImageCairo.cpp:
+        (ImageDiff::PlatformImage::createFromStdin): Removed the variable name.
 
-        Array, pointer, and array ref types are now parsed and then constructed
-        in reverse so that the declaration order matches the indexing order, as
-        in C/C++.
+2018-09-12  Dan Bernstein  <mitz@apple.com>
 
-        Reviewed by Myles C. Maxfield.
+        [Cocoa] Complete support for Paste as Quotation
+        https://bugs.webkit.org/show_bug.cgi?id=189504
 
-        * WebGPUShadingLanguageRI/Parse.js:
-        (parseType):
-        * WebGPUShadingLanguageRI/Test.js:
+        Reviewed by Wenson Hsieh.
 
-2018-08-30  Wenson Hsieh  <wenson_hsieh@apple.com>
+        * MiniBrowser/mac/MainMenu.xib: Added a Paste as Quotation command in the Edit menu.
 
-        Followup to [iOS] TestWebKitAPI.PasteImage tests are flaky failures
-        https://bugs.webkit.org/show_bug.cgi?id=185965
+2018-09-12  Sihui Liu  <sihui_liu@apple.com>
 
-        Reviewed by Andy Estes.
+        Move IndexedDB to Network Process
+        https://bugs.webkit.org/show_bug.cgi?id=189415
 
-        The first attempt to fix these flaky tests on iOS caused timeouts on macOS while waiting for a "loaded" message
-        from the page. To (hopefully) fix this across both platforms, make sure that we first register our script
-        message handlers in the UI process *before* evaluating script that could propagate the script message.
+        Reviewed by Geoffrey Garen.
 
-        * TestWebKitAPI/Tests/WebKitCocoa/PasteImage.mm:
-        (-[TestWKWebView waitForMessage:afterEvaluatingScript:]):
+        * TestWebKitAPI/Tests/WebKitCocoa/IDBDeleteRecovery.mm:
         (TEST):
-
-2018-08-30  Thomas Denney  <tdenney@apple.com>
-
-        [WHLSL] Implement tests to verify array indexing order matches our desires
-        https://bugs.webkit.org/show_bug.cgi?id=189099
-
-        Reviewed by Myles C. Maxfield.
-
-        * WebGPUShadingLanguageRI/Test.js: Add new test for array indexing order
-        and the option to explicitly disable tests (this test is disabled by
-        default).
-        * WebGPUShadingLanguageRI/TypeRef.js:
-        (TypeRef.wrap): Fixes  typo.
-
-2018-08-30  Sihui Liu  <sihui_liu@apple.com>
-
-        WKNavigation.ProcessCrashDuringCallback is failing on iOS
-        https://bugs.webkit.org/show_bug.cgi?id=189148
-
-        Reviewed by Chris Dumez.
-
-        We should not deference WebView or we may get WKErrorWebViewInvalidated error, because 
-        WebView could be destroyed before Web Process termination is noticed.
-
-        * TestWebKitAPI/Tests/WebKitCocoa/WebContentProcessDidTerminate.mm:
+        * TestWebKitAPI/Tests/WebKitCocoa/IDBIndexUpgradeToV2.mm:
         (TEST):
-
-2018-08-30  Zalan Bujtas  <zalan@apple.com>
-
-        [LFC][Floating] Formatting context roots avoid floats.
-        https://bugs.webkit.org/show_bug.cgi?id=189150
-
-        Reviewed by Antti Koivisto.
-
-        * LayoutReloaded/misc/LFC-passing-tests.txt:
-
-2018-08-30  Wenson Hsieh  <wenson_hsieh@apple.com>
-
-        [iOS] TestWebKitAPI.PasteImage tests are flaky failures
-        https://bugs.webkit.org/show_bug.cgi?id=185965
-
-        Reviewed by Andy Estes.
-
-        These tests pass reliably when run locally, but sometimes fail in automation when checking that the image
-        element with its "src" attribute set to a new blob URL has an empty size. It's possible that this is happening
-        because we're checking the size of the image element prior to image load after setting the blob URL; we can fix
-        this by waiting for the "load" event (which is also done in some of the other tests in this file).
-
-        * TestWebKitAPI/Tests/WebKitCocoa/PasteImage.mm:
+        * TestWebKitAPI/Tests/WebKitCocoa/IndexedDBDatabaseProcessKill.mm:
+        (TEST):
+        * TestWebKitAPI/Tests/WebKitCocoa/IndexedDBMultiProcess.mm:
+        (TEST):
+        * TestWebKitAPI/Tests/WebKitCocoa/IndexedDBPersistence.mm:
         (TEST):
 
-2018-08-30  Michael Catanzaro  <mcatanzaro@igalia.com>
-
-        Unreviewed, rolling out r235114.
-
-        ostree server deleted our commits?
-
-        Reverted changeset:
-
-        "[Flatpak] Update to GNOME master runtime"
-        https://bugs.webkit.org/show_bug.cgi?id=188731
-        https://trac.webkit.org/changeset/235114
-
-2018-08-29  Myles C. Maxfield  <mmaxfield@apple.com>
-
-        [WHLSL] Test row-majorness of matrices
-        https://bugs.webkit.org/show_bug.cgi?id=189101
-
-        The matrix multiplication functions are temporarily commented out of the standard library,
-        so I've temporarily copy/pasted them into the test. Matrix multiplication is not
-        commutative, so it requires the right indexing order.
+2018-09-12  Chris Dumez  <cdumez@apple.com>
 
-        Reviewed by Dean Jackson and Thomas Denney.
+        PSON: No process swap on back navigation after URL bar navigation
+        https://bugs.webkit.org/show_bug.cgi?id=189557
+        <rdar://problem/44353108>
 
-        * WebGPUShadingLanguageRI/Intrinsics.js:
-        * WebGPUShadingLanguageRI/StandardLibrary.js:
-        (let.standardLibrary):
-        * WebGPUShadingLanguageRI/Test.js:
+        Reviewed by Alex Christensen.
 
-2018-08-29  Jer Noble  <jer.noble@apple.com>
+        Add API test coverage.
 
-        Unreviewed test gardening; NowPlayingTest API tests require High Sierra.
+        * TestWebKitAPI/Tests/WebKitCocoa/ProcessSwapOnNavigation.mm:
 
-        * TestWebKitAPI/Tests/WebKitCocoa/NowPlaying.mm:
+2018-09-11  Dean Jackson  <dino@apple.com>
 
-2018-08-29  Youenn Fablet  <youenn@apple.com>
+        Header parsing for experimental and internal debug features
+        https://bugs.webkit.org/show_bug.cgi?id=189486
+        <rdar://problem/44320618>
 
-        Remove WebRTC legacy API implementation
-        https://bugs.webkit.org/show_bug.cgi?id=189040
+        Reviewed by Tim Horton.
 
-        Reviewed by Eric Carlson.
+        Keep track of experimental and internal debug features in TestOptions, and
+        allow webkit-test-runner headers to toggle them by name.
 
-        * DumpRenderTree/mac/DumpRenderTree.mm:
-        (enableExperimentalFeatures):
-        * WebKitTestRunner/InjectedBundle/Bindings/TestRunner.idl:
-        * WebKitTestRunner/InjectedBundle/TestRunner.cpp:
-        (WTR::TestRunner::setWebRTCLegacyAPIEnabled): Deleted.
-        * WebKitTestRunner/InjectedBundle/TestRunner.h:
+        * DumpRenderTree/TestOptions.cpp:
+        (TestOptions::TestOptions):
         * WebKitTestRunner/TestController.cpp:
         (WTR::TestController::resetPreferencesToConsistentValues):
+        (WTR::updateTestOptionsFromTestHeader):
+        * WebKitTestRunner/TestOptions.h:
+        (WTR::TestOptions::hasSameInitializationOptions const):
 
-2018-08-29  Daniel Bates  <dabates@apple.com>
-
-        lldb-webkit: KeyError thrown for OptionSet with invalid value
-        https://bugs.webkit.org/show_bug.cgi?id=189070
-
-        Reviewed by Simon Fraser.
-
-        Do not compute what enumerators are in an OptionSet with an invalid value. A local OptionSet
-        variable is only considered valid when execution passes over its assignment/constructor.
-
-        The LLDB Python API does not provide a way to determine whether a variable has a valid
-        value (SBValue.IsValid() does not seem to check if the actual bit pattern in memory for
-        a variable represents a valid value). So, we use a simple heuristic: when the value of
-        the OptionSet is greater than the value of the bitmask with all enumerators set then we
-        consider the OptionSet to be invalid. When the variable obtains a valid value then LLDB
-        will notify us to update our state.
-
-        * lldb/lldb_webkit.py:
-        (WTFOptionSetProvider.update):
-
-2018-08-29  Chris Dumez  <cdumez@apple.com>
-
-        [PSON] We should only process-swap when eTLD+1 changes on navigation
-        https://bugs.webkit.org/show_bug.cgi?id=189090
-        <rdar://problem/43799225>
-
-        Reviewed by Geoffrey Garen.
-
-        Update existing PSON API tests to use different registrable domains instead of different
-        protocols, as we now only process-swap when the eTLD+1 (aka "registrable domain") changes.
-
-        Add test coverage to make sure that we do not process swap when eTLD+1 does not change.
-
-        * TestWebKitAPI/Tests/WebKitCocoa/ProcessSwapOnNavigation.mm:
-
-2018-08-29  Youenn Fablet  <youenn@apple.com>
-
-        Add a runtime flag for WebRTC unified plan
-        https://bugs.webkit.org/show_bug.cgi?id=189068
-
-        Reviewed by Eric Carlson.
-
-        * WebKitTestRunner/InjectedBundle/Bindings/TestRunner.idl:
-        * WebKitTestRunner/InjectedBundle/TestRunner.cpp:
-        (WTR::TestRunner::setWebRTCUnifiedPlanEnabled):
-        * WebKitTestRunner/InjectedBundle/TestRunner.h:
-
-2018-08-29  Don Olmstead  <don.olmstead@sony.com>
-
-        [CMake] Split Tools/CMakeLists.txt into platform files
-        https://bugs.webkit.org/show_bug.cgi?id=189093
-
-        Reviewed by Michael Catanzaro.
-
-        * CMakeLists.txt:
-        * PlatformGTK.cmake: Added.
-        * PlatformMac.cmake: Added.
-        * PlatformWPE.cmake: Added.
-        * PlatformWin.cmake: Added.
-
-2018-08-29  Jer Noble  <jer.noble@apple.com>
-
-        Muted elements do not have their Now Playing status updated when unmuted.
-        https://bugs.webkit.org/show_bug.cgi?id=189069
-
-        Reviewed by Eric Carlson.
-
-        * TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj:
-        * TestWebKitAPI/Tests/WebKitCocoa/NowPlaying.mm: Added.
-        (userInfoHasNowPlayingApplicationPID):
-        (getNowPlayingClient):
-        (getNowPlayingClientPid):
-        (NowPlayingTest::webView):
-        (NowPlayingTest::configuration):
-        (NowPlayingTest::webViewPid):
-        (NowPlayingTest::loadPage):
-        (NowPlayingTest::runScriptWithUserGesture):
-        (NowPlayingTest::runScriptWithoutUserGesture):
-        (NowPlayingTest::executeAndWaitForPlaying):
-        (NowPlayingTest::executeAndWaitForWebViewToBecomeNowPlaying):
-        (NowPlayingTest::observers):
-        (NowPlayingTest::addObserver):
-        (NowPlayingTest::removeObserver):
-        (NowPlayingTest::notificationCallback):
-        (NowPlayingTest::receivedNotification):
-        (NowPlayingTest::performAfterReceivingNotification):
-        (TEST_F):
-        * TestWebKitAPI/Tests/WebKitCocoa/now-playing.html: Added.
-
-2018-08-29  Thomas Denney  <tdenney@apple.com>
-
-        [WHLSL] Ensure that isLValue is copied by the rewriter
-        https://bugs.webkit.org/show_bug.cgi?id=189083
-
-        The value of isLValue was not propagated when new TernaryExpressions are
-        created in Rewriter.
-
-        Reviewed by Myles C. Maxfield.
-
-        * WebGPUShadingLanguageRI/Rewriter.js:
-        (Rewriter.prototype.visitTernaryExpression):
-        * WebGPUShadingLanguageRI/Test.js: Add new test to verify the value is
-        propagated.
-
-2018-08-29  Myles C. Maxfield  <mmaxfield@apple.com>
-
-        [WHLSL] Add more functions to the standard library
-        https://bugs.webkit.org/show_bug.cgi?id=188873
-
-        Reviewed by Filip Pizlo.
-
-        This patch adds the rest of the standard library to StandardLibrary.js, and updates the compiler to be able to compile it.
-
-        There are a few major pieces:
-        1. Swizzle operators are now implemented in the language, instead of as native functions
-        2. Vector constructors are now implemented in the language, instead of as native functions
-        3. The matrix type is implemented
-        4. Vector operator&[] is illegal, and is removed from the compiler
-        5. Vector index setters & index getters are now implemented in the language, instead of as native functions
-        6. Vector and matrix equality operators are implemented in the language, instead of as native functions
-        7. Casting a scalar to a boolean is implemented in the language, instead of as native functions
-        8. Casting a vector to a boolean is not part of the language, and is removed from the compiler
-        9. Half-precision floating point types are implemented
+2018-09-12  Fujii Hironori  <Hironori.Fujii@sony.com>
 
-        * WebGPUShadingLanguageRI/All.js:
-        * WebGPUShadingLanguageRI/BuiltinMatrixGetter.js: Copied from Tools/WebGPUShadingLanguageRI/BuiltinVectorIndexSetter.js.
-        (BuiltinMatrixGetter):
-        (BuiltinMatrixGetter.prototype.get height):
-        (BuiltinMatrixGetter.prototype.get width):
-        (BuiltinMatrixGetter.prototype.toString):
-        (BuiltinMatrixGetter.functions):
-        (BuiltinMatrixGetter.prototype.instantiateImplementation):
-        * WebGPUShadingLanguageRI/BuiltinMatrixSetter.js: Renamed from Tools/WebGPUShadingLanguageRI/BuiltinVectorIndexSetter.js.
-        (BuiltinMatrixSetter):
-        (BuiltinMatrixSetter.prototype.get height):
-        (BuiltinMatrixSetter.prototype.get width):
-        (BuiltinMatrixSetter.prototype.toString):
-        (BuiltinMatrixSetter.functions):
-        (BuiltinMatrixSetter.prototype.instantiateImplementation):
-        * WebGPUShadingLanguageRI/BuiltinVectorConstructors.js: Removed.
-        * WebGPUShadingLanguageRI/BuiltinVectorEqualityOperator.js: Removed.
-        * WebGPUShadingLanguageRI/BuiltinVectorGetter.js:
-        (BuiltinVectorGetter.prototype.instantiateImplementation):
-        (BuiltinVectorGetter):
-        * WebGPUShadingLanguageRI/BuiltinVectorSetter.js:
-        (BuiltinVectorSetter.functions):
-        (BuiltinVectorSetter.prototype.instantiateImplementation):
-        (BuiltinVectorSetter):
-        * WebGPUShadingLanguageRI/CallExpression.js:
-        (CallExpression.prototype.resolve):
-        * WebGPUShadingLanguageRI/CheckTypesWithArguments.js:
-        (checkTypesWithArguments.TypeWithArgumentsChecker.prototype.visitTypeRef):
-        (checkTypesWithArguments.TypeWithArgumentsChecker):
-        (checkTypesWithArguments):
-        * WebGPUShadingLanguageRI/Checker.js:
-        (Checker.prototype.visitVectorType):
-        (Checker.prototype.visitMatrixType):
-        * WebGPUShadingLanguageRI/ConstexprFolder.js:
-        (ConstexprFolder.prototype.visitCallExpression):
-        (ConstexprFolder):
-        * WebGPUShadingLanguageRI/Evaluator.js:
-        (Evaluator.prototype.visitTernaryExpression):
-        * WebGPUShadingLanguageRI/FlattenedStructOffsetGatherer.js:
-        (FlattenedStructOffsetGatherer.prototype.visitMatrixType):
-        (FlattenedStructOffsetGatherer):
-        * WebGPUShadingLanguageRI/Intrinsics.js:
-        (Intrinsics.):
-        (Intrinsics):
-        * WebGPUShadingLanguageRI/MatrixType.js: Renamed from Tools/WebGPUShadingLanguageRI/BuiltinVectorIndexGetter.js.
-        (MatrixType):
-        (MatrixType.prototype.get elementType):
-        (MatrixType.prototype.get numRows):
-        (MatrixType.prototype.get numColumns):
-        (MatrixType.prototype.get numRowsValue):
-        (MatrixType.prototype.get numColumnsValue):
-        (MatrixType.prototype.get size):
-        (MatrixType.prototype.unifyImpl):
-        (MatrixType.prototype.populateDefaultValue):
-        (MatrixType.prototype.toString):
-        * WebGPUShadingLanguageRI/NameContext.js:
-        (NameContext.prototype.add):
-        * WebGPUShadingLanguageRI/NativeType.js:
-        (NativeType.create):
-        (NativeType):
-        * WebGPUShadingLanguageRI/OperatorAnderIndexer.js: Renamed from Tools/WebGPUShadingLanguageRI/OperatorAnderIndex.js.
-        (OperatorAnderIndexer):
-        * WebGPUShadingLanguageRI/OperatorBool.js: Removed.
-        * WebGPUShadingLanguageRI/Prepare.js:
-        (let.prepare):
-        * WebGPUShadingLanguageRI/Program.js:
-        (Program.prototype.add):
-        * WebGPUShadingLanguageRI/Rewriter.js:
-        (Rewriter.prototype.visitMatrixType):
-        (Rewriter):
-        * WebGPUShadingLanguageRI/SPIRV.html:
-        * WebGPUShadingLanguageRI/StandardLibrary.js:
-        (operator.bool):
-        (bool.operator):
-        (uchar.operator):
-        (ushort.operator):
-        (char.operator):
-        (short.operator):
-        (uint.operator):
-        (int.operator):
-        (half.operator):
-        (float.operator):
-        (uchar2.operator):
-        (uchar3.operator):
-        (uchar4.operator):
-        (ushort2.operator):
-        (ushort3.operator):
-        (ushort4.operator):
-        (uint2.operator):
-        (uint3.operator):
-        (uint4.operator):
-        (char2.operator):
-        (char3.operator):
-        (char4.operator):
-        (short2.operator):
-        (short3.operator):
-        (short4.operator):
-        (int2.operator):
-        (int3.operator):
-        (int4.operator):
-        (half2.operator):
-        (half3.operator):
-        (half4.operator):
-        (float2.operator):
-        (float3.operator):
-        (float4.operator):
-        (half2x2.operator):
-        (half2x3.operator):
-        (half2x4.operator):
-        (half3x2.operator):
-        (half3x3.operator):
-        (half3x4.operator):
-        (half4x2.operator):
-        (half4x3.operator):
-        (half4x4.operator):
-        (float2x2.operator):
-        (float2x3.operator):
-        (float2x4.operator):
-        (float3x2.operator):
-        (float3x3.operator):
-        (float3x4.operator):
-        (float4x2.operator):
-        (float4x3.operator):
-        (float4x4.operator):
-        (operator.bool2):
-        (operator.bool3):
-        (operator.bool4):
-        (uint.operator.length):
-        (operator.uchar2):
-        (operator.uchar3):
-        (operator.uchar4):
-        (operator.ushort2):
-        (operator.ushort3):
-        (operator.ushort4):
-        (operator.uint2):
-        (operator.uint3):
-        (operator.uint4):
-        (operator.char2):
-        (operator.char3):
-        (operator.char4):
-        (operator.short2):
-        (operator.short3):
-        (operator.short4):
-        (operator.int2):
-        (operator.int3):
-        (operator.int4):
-        (operator.half2):
-        (operator.half3):
-        (operator.half4):
-        (operator.float2):
-        (operator.float3):
-        (operator.float4):
-        (bool2.operator):
-        (bool3.operator):
-        (bool4.operator):
-        (allVectorTypeNames):
-        * WebGPUShadingLanguageRI/SwizzleOp.js: Removed.
-        * WebGPUShadingLanguageRI/SynthesizeDefaultConstructorOperator.js:
-        (synthesizeDefaultConstructorOperator.FindAllTypes.prototype.visitMatrixType):
-        (synthesizeDefaultConstructorOperator.FindAllTypes):
-        (synthesizeDefaultConstructorOperator):
-        * WebGPUShadingLanguageRI/SynthesizeOperatorBool.js: Removed.
-        * WebGPUShadingLanguageRI/Test.html:
-        * WebGPUShadingLanguageRI/Test.js:
-        (makeHalf):
-        (checkHalf):
-        * WebGPUShadingLanguageRI/Visitor.js:
-        (Visitor.prototype.visitMatrixType):
-        (Visitor):
-        * WebGPUShadingLanguageRI/index.html:
+        [Win][Clang][DumpRenderTree] 0 and nullptr can't be implicitly converted to AccessibilityUIElement
+        https://bugs.webkit.org/show_bug.cgi?id=189492
 
-2018-08-29  Simon Fraser  <simon.fraser@apple.com>
+        Reviewed by Alex Christensen.
 
-        Teach webkitpy how to check leaks and treat leaks as test failures
-        https://bugs.webkit.org/show_bug.cgi?id=189067
+        0 and nullptr can't be implicitly converted to
+        AccessibilityUIElement because only Windows port typedefs
+        PlatformUIElement as COMPtr<IAccessible> not a pointer.
 
-        Reviewed by Darin Adler.
-        
-        Add a new "--world-leaks" argument to run-webkit-tests. When enabled, DRT/WTR are launched
-        with a --world-leaks argument (which is renamed in this patch for consistency). This enables the
-        behavior added in r235408, namely that they check for leaked documents after each test, and at
-        the end of one (if --run-singly) or a set of tests run in a single DRT/WTR instance handle the
-        "#CHECK FOR WORLD LEAKS" command to get still-live documents.
-        
-        LayoutTestRunner in webkitpy now has the notion of doing "post-tests work", called via _finished_test_group(),
-        and here it sends the "#CHECK FOR WORLD LEAKS" command to the runner and parses the resulting output block.
-        If this results block includes leaks, we convert an existing TestResult into a LEAK failure
-        in TestRunResults.change_result_to_failure(). Leaks are then added to the ouput JSON for display in results.html
+        Replaced "return 0" and "return nullptr" with "return { nullptr }".
 
-        Unit tests are updated with some leak examples.
+        * DumpRenderTree/AccessibilityUIElement.cpp:
+        (AccessibilityUIElement::horizontalScrollbar const):
+        (AccessibilityUIElement::verticalScrollbar const):
+        (AccessibilityUIElement::uiElementAttributeValue const):
+        (AccessibilityUIElement::accessibilityElementForTextMarker):
+        * DumpRenderTree/win/AccessibilityControllerWin.cpp:
+        (AccessibilityController::elementAtPoint):
+        (AccessibilityController::accessibleElementById):
+        (AccessibilityController::focusedElement):
+        (AccessibilityController::rootElement):
+        * DumpRenderTree/win/AccessibilityUIElementWin.cpp:
+        (AccessibilityUIElement::elementAtPoint):
+        (AccessibilityUIElement::linkedUIElementAtIndex):
+        (AccessibilityUIElement::getChildAtIndex):
+        (AccessibilityUIElement::titleUIElement):
+        (AccessibilityUIElement::parentElement):
+        (AccessibilityUIElement::uiElementForSearchPredicate):
+        (AccessibilityUIElement::cellForColumnAndRow):
+        (AccessibilityUIElement::disclosedRowAtIndex):
+        (AccessibilityUIElement::ariaOwnsElementAtIndex):
+        (AccessibilityUIElement::ariaFlowToElementAtIndex):
+        (AccessibilityUIElement::ariaControlsElementAtIndex):
+        (AccessibilityUIElement::selectedRowAtIndex):
+        (AccessibilityUIElement::rowAtIndex):
+        (AccessibilityUIElement::disclosedByRow):
+        (AccessibilityUIElement::selectedChildAtIndex const):
 
-        * DumpRenderTree/mac/DumpRenderTree.mm:
-        (initializeGlobalsFromCommandLineOptions):
-        * Scripts/webkitpy/common/net/resultsjsonparser_unittest.py:
-        (ParsedJSONResultsTest):
-        * Scripts/webkitpy/layout_tests/controllers/layout_test_runner.py:
-        (LayoutTestRunner._annotate_results_with_additional_failures):
-        (LayoutTestRunner._handle_finished_test_group):
-        (Worker.handle):
-        (Worker._run_test):
-        (Worker._do_post_tests_work):
-        (Worker._finished_test_group):
-        (Worker._run_test_in_another_thread):
-        * Scripts/webkitpy/layout_tests/layout_package/json_layout_results_generator.py:
-        (JSONLayoutResultsGenerator):
-        * Scripts/webkitpy/layout_tests/models/test_expectations.py:
-        (TestExpectationParser):
-        (TestExpectations):
-        * Scripts/webkitpy/layout_tests/models/test_expectations_unittest.py:
-        (Base.get_basic_tests):
-        * Scripts/webkitpy/layout_tests/models/test_failures.py:
-        (determine_result_type):
-        (FailureLeak):
-        (FailureLeak.__init__):
-        (FailureLeak.message):
-        (FailureDocumentLeak):
-        (FailureDocumentLeak.__init__):
-        (FailureDocumentLeak.message):
-        * Scripts/webkitpy/layout_tests/models/test_results.py:
-        (TestResult.convert_to_failure):
-        * Scripts/webkitpy/layout_tests/models/test_run_results.py:
-        (TestRunResults.change_result_to_failure):
-        (_interpret_test_failures):
-        (summarize_results):
-        * Scripts/webkitpy/layout_tests/models/test_run_results_unittest.py:
-        (get_result):
-        (run_results):
-        (summarized_results):
-        * Scripts/webkitpy/layout_tests/run_webkit_tests.py:
-        (parse_args):
-        * Scripts/webkitpy/layout_tests/run_webkit_tests_integrationtest.py:
-        (parse_args):
-        (RunTest.test_check_for_world_leaks):
-        * Scripts/webkitpy/port/driver.py:
-        (DriverPostTestOutput):
-        (DriverPostTestOutput.__init__):
-        (Driver.do_post_tests_work):
-        (Driver._parse_world_leaks_output):
-        (Driver.cmd_line):
-        (DriverProxy.do_post_tests_work):
-        * Scripts/webkitpy/port/test.py:
-        (unit_test_list):
-        * WebKitTestRunner/Options.cpp:
-        (WTR::OptionsHandler::OptionsHandler):
-        * WebKitTestRunner/TestController.cpp:
-        (WTR::TestController::checkForWorldLeaks):
+2018-09-11  Wenson Hsieh  <wenson_hsieh@apple.com>
 
-2018-08-29  David Kilzer  <ddkilzer@apple.com>
+        FontManagerTests.ChangeFontColorWithColorPanel fails on macOS Sierra
+        https://bugs.webkit.org/show_bug.cgi?id=189382
 
-        Remove empty directories from from svn.webkit.org repository
-        <https://webkit.org/b/189081>
+        Reviewed by Ryosuke Niwa.
 
-        * DumpRenderTree/TestNetscapePlugIn/unix: Removed.
-        * WebKitTestRunner/UIScriptContext: Removed.
+        Fix the test by specifying colors in RGB colorspace.
 
-2018-08-29  Pablo Saavedra  <psaavedra@igalia.com>
+        * TestWebKitAPI/Tests/mac/FontManagerTests.mm:
+        (TestWebKitAPI::TEST):
 
-        [GTK][WPE] Add JSCOnly Release bot for Linux ARMv7 Thumb2 SoftFP
-        https://bugs.webkit.org/show_bug.cgi?id=189077
+2018-09-11  Fujii Hironori  <Hironori.Fujii@sony.com>
 
-        Reviewed by Michael Catanzaro.
+        [Win][Clang][ImageDiff] Fix compilation error and warning of PlatformImageCairo.cpp
+        https://bugs.webkit.org/show_bug.cgi?id=189496
 
-        Deployed new bot for WPE Release in Linux ARMv7 Thumb2 SoftFP
-        This bot (jsconly-linux-igalia-bot-5) will run the 'JSCOnly
-        Linux ARMv7 Thumb2 SoftFP Release' buildjob.
+        Reviewed by Alex Christensen.
 
-        * BuildSlaveSupport/build.webkit.org-config/config.json:
-        * BuildSlaveSupport/build.webkit.org-config/steps_unittest.py:
+        * ImageDiff/cairo/PlatformImageCairo.cpp:
+        (ImageDiff::PlatformImage::createFromStdin): Removed unnecessary ReadContext struct.
+        (ImageDiff::PlatformImage::writeAsPNGToStdout): Use '%lu' format type specifier for unsigned long.
 
-2018-08-28  Daniel Bates  <dabates@apple.com>
+2018-09-07  Dean Jackson  <dino@apple.com>
 
-        lldb-webkit: KeyError thrown for OptionSet
-        https://bugs.webkit.org/show_bug.cgi?id=189062
+        Add and expose Internal features from WebKit
+        https://bugs.webkit.org/show_bug.cgi?id=189442
+        <rdar://problem/44243404>
 
         Reviewed by Simon Fraser.
 
-        For some reason lldb(1) may not retrieve the enumerator names for an enum value
-        when it calls update() on the synthetic children provider. If this occurs then
-        bail out as we cannot compute what enumerators are in the set.
-
-        * lldb/lldb_webkit.py:
-        (WTFOptionSetProvider.update):
-
-2018-08-28  Sihui Liu  <sihui_liu@apple.com>
-
-        Add error information to help debug test failure in WKNavigation.ProcessCrashDuringCallback
-        https://bugs.webkit.org/show_bug.cgi?id=189037
-
-        Reviewed by Chris Dumez.
-
-        * TestWebKitAPI/Tests/WebKitCocoa/WebContentProcessDidTerminate.mm:
-        (TEST):
+        Update MiniBrowser to expose an Internal feature menu.
 
-2018-08-28  David Kilzer  <ddkilzer@apple.com>
+        * MiniBrowser/mac/AppDelegate.m:
+        (defaultConfiguration):
+        * MiniBrowser/mac/SettingsController.m:
+        (-[SettingsController _populateMenu]):
+        (-[SettingsController validateMenuItem:]):
+        (-[SettingsController toggleExperimentalFeature:]):
+        (-[SettingsController toggleInternalDebugFeature:]):
 
-        Fix incorrect use of `sourceTree = "<group>";` for built products and frameworks in Xcode projects
+2018-09-11  Wenson Hsieh  <wenson_hsieh@apple.com>
 
-        Found by tidy-Xcode-project-file script (see Bug 188754).
+        [macOS] [WK2] Support changing foreground colors via color panel
+        https://bugs.webkit.org/show_bug.cgi?id=189382
+        <rdar://problem/44227311>
 
-        * DumpRenderTree/DumpRenderTree.xcodeproj/project.pbxproj:
-        (IOKit.framework): Use `sourceTree = "<absolute>";` to match
-        other frameworks in the project.
-        * MobileMiniBrowser/MobileMiniBrowser.xcodeproj/project.pbxproj:
-        (WebKit.framework): Use `sourceTree = BUILT_PRODUCTS_DIR;` for
-        built product instead of relative build directory path.
-        * jsc-cli/jsc-cli.xcodeproj/project.pbxproj:
-        (JavaScriptCore.framework): Ditto.
-        * lldb/lldbWebKitTester/lldbWebKitTester.xcodeproj/project.pbxproj:
-        (libWTF.a): Use `sourceTree = BUILT_PRODUCTS_DIR;` for built
-        product.
+        Reviewed by Ryosuke Niwa.
 
-2018-08-28  Alex Christensen  <achristensen@webkit.org>
-
-        Fix 32-bit Mac build.
-
-        * TestWebKitAPI/Tests/WebKitCocoa/RestoreSessionStateWithoutNavigation.mm:
-
-2018-08-28  Alex Christensen  <achristensen@webkit.org>
-
-        REGRESSION (r235391): [iOS] TestWebKitAPI.WebKit.RestoreSessionStateWithoutNavigation is timing out
-        https://bugs.webkit.org/show_bug.cgi?id=189050
-
-        Reviewed by Tim Horton.
-
-        * TestWebKitAPI/Tests/WebKitCocoa/RestoreSessionStateWithoutNavigation.mm:
-        This test wasn't run on iOS before.  Let's re-introduce the WK_HAVE_C_SPI guard.
-
-2018-08-28  Wenson Hsieh  <wenson_hsieh@apple.com>
-
-        [iOS] WKAttachmentTests.DropFolderAsAttachmentAndMoveByDragging encounters an exception on iOS simulator
-        https://bugs.webkit.org/show_bug.cgi?id=189049
-
-        Reviewed by Tim Horton.
-
-        Wrap an NSURL in a RetainPtr to prevent it from being freed prematurely. It seems this broke when I changed the
-        load handler from an Objective-C block that captured `folderURL` to a C++ lambda that didn't retain `folderURL`.
-
-        * TestWebKitAPI/Tests/WebKitCocoa/WKAttachmentTests.mm:
-        (simulateFolderDragWithURL):
-
-2018-08-27  Wenson Hsieh  <wenson_hsieh@apple.com>
-
-        [Attachment Support] Remove WebCore::AttachmentDisplayOptions and friends
-        https://bugs.webkit.org/show_bug.cgi?id=189004
-
-        Reviewed by Dan Bernstein.
-
-        Move off of deprecated attachment insertion SPI.
-
-        * TestWebKitAPI/Tests/WebKitCocoa/WKAttachmentTests.mm:
-        (-[TestWKWebView synchronouslyInsertAttachmentWithFileWrapper:contentType:]):
-        (-[TestWKWebView synchronouslyInsertAttachmentWithFilename:contentType:data:]):
-        (-[_WKAttachment synchronouslySetDisplayOptions:error:]): Deleted.
-
-2018-08-27  Simon Fraser  <simon.fraser@apple.com>
-
-        Teach WebKitTestRunner and DumpRenderTree about detecting world leaks
-        https://bugs.webkit.org/show_bug.cgi?id=188994
-
-        Reviewed by Tim Horton.
-        
-        This patch adds the notion of a "control command" in the protocol between webkitpy and 
-        WebKitTestRunner/DumpRenderTree. A command is simply an input string starting with a #
-        that is checked for before trying to parse the input as test URL. For now, just one
-        commmand is supported, which is "#CHECK FOR WORLD LEAKS".
-        
-        In response to the command, the tool dumps an output block in the usual pseudo-MIME-style,
-        with a trailing "#EOF". Future patches will add support to webkitpy to parse this output.
-        
-        DumpRenderTree stubs out the command, returning an empty block.
-        
-        WebKitTestRunner responds to the command by dumping the list of live documents, if it was
-        run with the --check-for-world-leaks option.
-        
-        When run with --check-for-world-leaks, WebKitTestRunner gets the list of live documents via
-        WKBundleGetLiveDocumentURLs() after every test (this allows it to detect the first test
-        that leaked a document), and keeps them in a map of document identifier to test and live document URL.
-        Then when it receives the "#CHECK FOR WORLD LEAKS" command, it calls into the bundle to
-        clear the page and memory caches, runs a GC, then posts a task (in the Document::postTaks() sense)
-        after which it requests the list of live documents for a final time, excluding any that are loaded
-        in live Frames (thus omitting the about:blank that will be loaded at this point). Documents in this
-        list are therefore leaked (or abandoned).
-        
-        Future patches will hook up webkitpy reporting for leaked documents.
-
-        * DumpRenderTree/mac/DumpRenderTree.mm:
-        (initializeGlobalsFromCommandLineOptions):
-        (handleControlCommand):
-        (runTestingServerLoop):
-        * DumpRenderTree/win/DumpRenderTree.cpp:
-        (handleControlCommand):
-        (main):
-        * WebKitTestRunner/InjectedBundle/InjectedBundle.cpp:
-        (WTR::postGCTask):
-        (WTR::InjectedBundle::reportLiveDocuments):
-        (WTR::InjectedBundle::didReceiveMessageToPage):
-        * WebKitTestRunner/InjectedBundle/InjectedBundle.h:
-        * WebKitTestRunner/Options.cpp:
-        (WTR::handleOptionCheckForWorldLeaks):
-        (WTR::OptionsHandler::OptionsHandler):
-        * WebKitTestRunner/Options.h:
-        * WebKitTestRunner/TestController.cpp:
-        (WTR::AsyncTask::run):
-        (WTR::AsyncTask::currentTask):
-        (WTR::TestController::initialize):
-        (WTR::TestController::ensureViewSupportsOptionsForTest):
-        (WTR::TestController::resetStateToConsistentValues):
-        (WTR::TestController::updateLiveDocumentsAfterTest):
-        (WTR::TestController::checkForWorldLeaks):
-        (WTR::TestController::findAndDumpWorldLeaks):
-        (WTR::TestController::willDestroyWebView):
-        (WTR::parseInputLine):
-        (WTR::TestController::waitForCompletion):
-        (WTR::TestController::handleControlCommand):
-        (WTR::TestController::runTestingServerLoop):
-        (WTR::TestController::run):
-        (WTR::TestController::didReceiveLiveDocumentsList):
-        (WTR::TestController::didReceiveMessageFromInjectedBundle):
-        * WebKitTestRunner/TestController.h:
-        (WTR::AsyncTask::AsyncTask):
-        (WTR::AsyncTask::taskComplete):
-        (WTR::TestController::AbandonedDocumentInfo::AbandonedDocumentInfo):
-        * WebKitTestRunner/TestInvocation.cpp:
-        (WTR::TestInvocation::invoke):
-        * WebKitTestRunner/TestOptions.h:
-        (WTR::TestOptions::hasSameInitializationOptions const):
-
-2018-08-27  Alex Christensen  <achristensen@webkit.org>
-
-        Fix API test after r235398
-        https://bugs.webkit.org/show_bug.cgi?id=188997
-
-        * TestWebKitAPI/Tests/WebKit/ShouldKeepCurrentBackForwardListItemInList.cpp:
-        (TestWebKitAPI::setPageLoaderClient):
-        (TestWebKitAPI::willGoToBackForwardListItem): Deleted.
-        willGoToBackForwardListItem is unused and unsupported.  Removing its check.
-
-2018-08-27  Daniel Bates  <dabates@apple.com>
-
-        Partial revert of r235376
-        https://bugs.webkit.org/show_bug.cgi?id=189011
-
-        For now revert the unit tests added in r235376 as the following tests are failing on Apple Sierra
-        Debug and Apple High Sierra Debug bots:
-
-            lldb_webkit_unittest.TestSummaryProviders.serial_test_WTFOptionSetProvider_simple
-            lldb_webkit_unittest.TestSummaryProviders.serial_test_WTFOptionSet_SummaryProvider_simple
-
-        Will investigate offline.
-
-        * lldb/lldbWebKitTester/main.cpp:
-        (testSummaryProviders):
-        * lldb/lldb_webkit_unittest.py:
-        (TestSummaryProviders.serial_test_WTFHashSet_tablesize_and_size):
-        (TestSummaryProviders.serial_test_WTFOptionSet_SummaryProvider_empty): Deleted.
-        (TestSummaryProviders.serial_test_WTFOptionSet_SummaryProvider_simple): Deleted.
-        (TestSummaryProviders.serial_test_WTFOptionSetProvider_empty): Deleted.
-        (TestSummaryProviders.serial_test_WTFOptionSetProvider_simple): Deleted.
-
-2018-08-27  Aditya Keerthi  <akeerthi@apple.com>
-
-        Consolidate ENABLE_INPUT_TYPE_COLOR and ENABLE_INPUT_TYPE_COLOR_POPOVER
-        https://bugs.webkit.org/show_bug.cgi?id=188931
-
-        Reviewed by Wenson Hsieh.
-
-        * TestWebKitAPI/Configurations/FeatureDefines.xcconfig: Removed ENABLE_INPUT_TYPE_COLOR_POPOVER.
-
-2018-08-27  Justin Fan  <justin_fan@apple.com>
-
-        Add Justin Fan to list of WebKit contributors
-        https://bugs.webkit.org/show_bug.cgi?id=184431
-
-        * Scripts/webkitpy/common/config/contributors.json:
-
-2018-08-27  Simon Fraser  <simon.fraser@apple.com>
-
-        Convert timeout values in WebKitTestRunner to WTF::Seconds
-        https://bugs.webkit.org/show_bug.cgi?id=188987
-
-        Reviewed by Ryosuke Niwa.
-        
-        Replace various 'int' timeout values with WTF::Seconds. The timeout argument
-        comes in as milliseconds, so convert on input. When sending messages to the InjectedBundle
-        using integers, convert to and from milliseconds.
-        
-        Also do some #pragma once, and initializer cleanup.
-
-        * WebKitTestRunner/InjectedBundle/InjectedBundle.cpp:
-        (WTR::InjectedBundle::didReceiveMessageToPage):
-        (WTR::InjectedBundle::beginTesting):
-        (WTR::InjectedBundle::InjectedBundle): Deleted.
-        * WebKitTestRunner/InjectedBundle/InjectedBundle.h:
-        * WebKitTestRunner/InjectedBundle/TestRunner.h:
-        (WTR::TestRunner::timeout):
-        (WTR::TestRunner::setCustomTimeout):
-        * WebKitTestRunner/InjectedBundle/gtk/TestRunnerGtk.cpp:
-        (WTR::TestRunner::initializeWaitToDumpWatchdogTimerIfNeeded):
-        * WebKitTestRunner/InjectedBundle/mac/TestRunnerMac.mm:
-        (WTR::TestRunner::invalidateWaitToDumpWatchdogTimer):
-        (WTR::TestRunner::initializeWaitToDumpWatchdogTimerIfNeeded):
-        * WebKitTestRunner/InjectedBundle/wpe/TestRunnerWPE.cpp:
-        (WTR::TestRunner::initializeWaitToDumpWatchdogTimerIfNeeded):
-        * WebKitTestRunner/TestController.cpp:
-        (WTR::TestController::TestController):
-        (WTR::parseInputLine):
-        (WTR::TestController::runTest):
-        (WTR::TestController::runUntil):
-        (WTR::TestController::didReceiveMessageFromInjectedBundle):
-        * WebKitTestRunner/TestController.h:
-        * WebKitTestRunner/TestInvocation.cpp:
-        (WTR::TestInvocation::shortTimeout const):
-        (WTR::TestInvocation::createTestSettingsDictionary):
-        * WebKitTestRunner/TestInvocation.h:
-        * WebKitTestRunner/TestOptions.h:
-        * WebKitTestRunner/cocoa/TestControllerCocoa.mm:
-        (WTR::TestController::platformRunUntil):
-        * WebKitTestRunner/gtk/TestControllerGtk.cpp:
-        (WTR::TestController::platformRunUntil):
-        * WebKitTestRunner/wpe/TestControllerWPE.cpp:
-        (WTR::TestController::platformRunUntil):
-
-2018-08-27  Simon Fraser  <simon.fraser@apple.com>
-
-        Convert timeout values in WebKitTestRunner to WTF::Seconds
-        https://bugs.webkit.org/show_bug.cgi?id=188987
-
-        Reviewed by Ryosuke Niwa.
-        
-        Replace various 'int' timeout values with WTF::Seconds. The timeout argument
-        comes in as milliseconds, so convert on input. When sending messages to the InjectedBundle
-        using integers, convert to and from milliseconds.
-        
-        Also do some #pragma once, and initializer cleanup.
-
-        * WebKitTestRunner/InjectedBundle/InjectedBundle.cpp:
-        (WTR::InjectedBundle::didReceiveMessageToPage):
-        (WTR::InjectedBundle::beginTesting):
-        (WTR::InjectedBundle::InjectedBundle): Deleted.
-        * WebKitTestRunner/InjectedBundle/InjectedBundle.h:
-        * WebKitTestRunner/InjectedBundle/TestRunner.h:
-        (WTR::TestRunner::timeout):
-        (WTR::TestRunner::setCustomTimeout):
-        * WebKitTestRunner/InjectedBundle/gtk/TestRunnerGtk.cpp:
-        (WTR::TestRunner::initializeWaitToDumpWatchdogTimerIfNeeded):
-        * WebKitTestRunner/InjectedBundle/mac/TestRunnerMac.mm:
-        (WTR::TestRunner::invalidateWaitToDumpWatchdogTimer):
-        (WTR::TestRunner::initializeWaitToDumpWatchdogTimerIfNeeded):
-        * WebKitTestRunner/InjectedBundle/wpe/TestRunnerWPE.cpp:
-        (WTR::TestRunner::initializeWaitToDumpWatchdogTimerIfNeeded):
-        * WebKitTestRunner/TestController.cpp:
-        (WTR::TestController::TestController):
-        (WTR::parseInputLine):
-        (WTR::TestController::runTest):
-        (WTR::TestController::runUntil):
-        (WTR::TestController::didReceiveMessageFromInjectedBundle):
-        * WebKitTestRunner/TestController.h:
-        * WebKitTestRunner/TestInvocation.cpp:
-        (WTR::TestInvocation::shortTimeout const):
-        (WTR::TestInvocation::createTestSettingsDictionary):
-        * WebKitTestRunner/TestInvocation.h:
-        * WebKitTestRunner/TestOptions.h:
-        * WebKitTestRunner/cocoa/TestControllerCocoa.mm:
-        (WTR::TestController::platformRunUntil):
-        * WebKitTestRunner/gtk/TestControllerGtk.cpp:
-        (WTR::TestController::platformRunUntil):
-        * WebKitTestRunner/wpe/TestControllerWPE.cpp:
-        (WTR::TestController::platformRunUntil):
-
-2018-08-27  Alex Christensen  <achristensen@webkit.org>
-
-        REGRESSION(r234985/r234989) WKPageLoadHTMLString with a 16-bit String has the wrong encoding
-        https://bugs.webkit.org/show_bug.cgi?id=189002
-
-        Reviewed by Tim Horton.
-
-        * TestWebKitAPI/Tests/WebKit/WillLoad.cpp:
-        (TestWebKitAPI::TEST_F):
-
-2018-08-27  Wenson Hsieh  <wenson_hsieh@apple.com>
-
-        [Cocoa] Exception (fileType 'dyn.agq8u' is not a valid UTI) raised when dragging an attachment whose file wrapper is a directory
-        https://bugs.webkit.org/show_bug.cgi?id=188903
-        <rdar://problem/43702993>
-
-        Reviewed by Tim Horton.
-
-        Add two API tests and adjust existing WKAttachment API tests. The new tests exercise the following scenarios, in
-        both iOS and macOS:
-        •       Dropping a folder as an attachment element, and then moving that attachment element in the document by
-                dragging and dropping.
-        •       Using WKWebView SPI to insert a folder and a file with an unknown extension, and then using more
-                _WKAttachment SPI to swap the attachments' backing file wrappers.
-
-        * TestWebKitAPI/Tests/WebKitCocoa/WKAttachmentTests.mm:
-        (runTestWithTemporaryFolder):
-
-        Add a helper function to run a test with a new folder path, created in the temporary directory and populated
-        with some sample content. This folder is deleted after running the test.
-
-        (simulateFolderDragWithURL):
-
-        Add a helper function to prepare a given DragAndDropSimulator for simulating a dragged folder from a source
-        external to the web view.
-
-        (platformCopyRichTextWithMultipleAttachments):
-        (platformCopyRichTextWithImage):
-        (platformCopyPNG):
-        (TestWebKitAPI::TEST):
-
-        Add new API tests, and adjust existing tests to reflect new -setFileWrapper:…: behavior. Specifically,
-        ChangeAttachmentDataAndFileInformation previously required that changing a _WKAttachment's NSFileWrapper would
-        preserve the previous NSFileWrapper's preferred name if the new file wrapper does not have a preferred name, but
-        this quirk is no longer supported.
-
-        Also add a few bridging casts for the eventual transition of TestWebKitAPI to ARC.
-
-        * TestWebKitAPI/cocoa/DragAndDropSimulator.h:
-
-        Add a new hook to clear any external drag information on an existing DragAndDropSimulator. This is convenient
-        when using the same DragAndDropSimulator to perform multiple drags in a single test.
-
-        * TestWebKitAPI/ios/DragAndDropSimulatorIOS.mm:
-        (-[DragAndDropSimulator clearExternalDragInformation]):
-        * TestWebKitAPI/mac/DragAndDropSimulatorMac.mm:
-        (-[DragAndDropSimulator clearExternalDragInformation]):
-
-2018-08-27  Alex Christensen  <achristensen@webkit.org>
-
-        Translate 4 tests using WKPageLoaderClient to ObjC
-        https://bugs.webkit.org/show_bug.cgi?id=188827
-
-        Reviewed by Tim Horton.
-
-        They use processDidBecomeUnresponsive, didChangeBackForwardList, or willGoToBackForwardListItem.
-        Rather than introduce these to WKPageNavigationClient, I just translated the tests to use WKNavigationDelegate, which already have equivalent callbacks.
-        willGoToBackForwardListItem had userData from the InjectedBundle, but nobody was using it so I did not add that to the ObjC SPI, so I don't test that unused
-        bundle functionality any more.
-
-        * TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj:
-        * TestWebKitAPI/Tests/WebKit/ResponsivenessTimer.cpp: Removed.
-        * TestWebKitAPI/Tests/WebKit/ResponsivenessTimerDoesntFireEarly.cpp: Removed.
-        * TestWebKitAPI/Tests/WebKit/RestoreSessionStateWithoutNavigation.cpp: Removed.
-        * TestWebKitAPI/Tests/WebKit/ShouldGoToBackForwardListItem.cpp: Removed.
-        * TestWebKitAPI/Tests/WebKit/ShouldGoToBackForwardListItem_Bundle.cpp: Removed.
-        * TestWebKitAPI/Tests/WebKitCocoa/ResponsivenessTimer.mm: Copied from Tools/TestWebKitAPI/Tests/WebKit/ResponsivenessTimer.cpp.
-        (-[ResponsivenessTimerDelegate webView:didFinishNavigation:]):
-        (-[ResponsivenessTimerDelegate _webViewWebProcessDidBecomeUnresponsive:]):
-        (TestWebKitAPI::TEST):
-        (): Deleted.
-        (TestWebKitAPI::didFinishLoadForFrame): Deleted.
-        (TestWebKitAPI::processDidBecomeUnresponsive): Deleted.
-        (TestWebKitAPI::setPageLoaderClient): Deleted.
-        * TestWebKitAPI/Tests/WebKitCocoa/ResponsivenessTimerDoesntFireEarly.mm: Copied from Tools/TestWebKitAPI/Tests/WebKit/ResponsivenessTimerDoesntFireEarly.cpp.
-        (-[ResponsivenessDelegate webView:didFinishNavigation:]):
-        (-[ResponsivenessDelegate _webViewWebProcessDidBecomeUnresponsive:]):
-        (TestWebKitAPI::TEST):
-        (TestWebKitAPI::didReceiveMessageFromInjectedBundle): Deleted.
-        (TestWebKitAPI::didFinishLoadForFrame): Deleted.
-        (TestWebKitAPI::processDidBecomeUnresponsive): Deleted.
-        (TestWebKitAPI::setInjectedBundleClient): Deleted.
-        (TestWebKitAPI::setPageLoaderClient): Deleted.
-        * TestWebKitAPI/Tests/WebKitCocoa/RestoreSessionStateWithoutNavigation.mm: Copied from Tools/TestWebKitAPI/Tests/WebKit/RestoreSessionStateWithoutNavigation.cpp.
-        (-[SessionStateDelegate webView:didFinishNavigation:]):
-        (-[SessionStateDelegate _webView:backForwardListItemAdded:removed:]):
-        (TestWebKitAPI::createSessionStateData):
-        (TestWebKitAPI::TEST):
-        (TestWebKitAPI::didFinishLoadForFrame): Deleted.
-        (TestWebKitAPI::didChangeBackForwardListForPage): Deleted.
-        (TestWebKitAPI::setPageLoaderClient): Deleted.
-        * TestWebKitAPI/Tests/WebKitCocoa/ShouldGoToBackForwardListItem.mm: Copied from Tools/TestWebKitAPI/Tests/WebKit/ShouldGoToBackForwardListItem.cpp.
-        (-[BackForwardClient webView:didFinishNavigation:]):
-        (-[BackForwardClient _webView:willGoToBackForwardListItem:inPageCache:]):
-        (TestWebKitAPI::TEST):
-        (TestWebKitAPI::didFinishLoadForFrame): Deleted.
-        (TestWebKitAPI::willGoToBackForwardListItem): Deleted.
-        (TestWebKitAPI::setPageLoaderClient): Deleted.
-
-2018-08-27  Wenson Hsieh  <wenson_hsieh@apple.com>
-
-        [Cocoa] "video.html" appears at the top level of the TestWebKitAPI Xcode project
-        https://bugs.webkit.org/show_bug.cgi?id=188989
-
-        Reviewed by Andy Estes.
-
-        Move this into the Tests/WebKit/Resources group in the project.
-
-        * TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj:
-
-2018-08-27  Thomas Denney  <tdenney@apple.com>
-
-        Avoid an exception in the interactive interpreter
-        https://bugs.webkit.org/show_bug.cgi?id=188991
-
-        Reviewed by Myles C. Maxfield.
-
-        * WebGPUShadingLanguageRI/index.html: Corrects a typo in the name of a
-        local variable
-
-2018-08-27  Thomas Denney  <tdenney@apple.com>
-
-        Allow new vector types to work with the interactive interpreter
-        https://bugs.webkit.org/show_bug.cgi?id=188988
-
-        Reviewed by Myles C. Maxfield.
-
-        * WebGPUShadingLanguageRI/FlattenedStructOffsetGatherer.js:
-        (FlattenedStructOffsetGatherer.prototype.visitTypeRef): Do not
-        unncessarily visit the type arguments of a TypeRef, as by this point
-        there are none that are relevant.
-        * WebGPUShadingLanguageRI/Intrinsics.js:
-        (Intrinsics): Treat VectorType as a primitive type.
-
-2018-08-27  Keith Rollin  <krollin@apple.com>
-
-        Build system support for LTO
-        https://bugs.webkit.org/show_bug.cgi?id=187785
-        <rdar://problem/42353132>
-
-        Reviewed by Dan Bernstein.
-
-        Add tools/scripts support for controlling LTO builds.
-
-        * Scripts/build-webkit: Add --lto-mode={none,thin,full}.
-        * Scripts/set-webkit-configuration: Add support for saving LTO
-        configuration to WebKitBuild/LTO.
-        * Scripts/webkitdirs.pm: Add support for reading configuration from
-        WebKitBuild/LTO and providing it to xcodebuild.
-        (determineLTOMode):
-        (ltoMode):
-        (XcodeOptions):
-
-2018-08-27  Daniel Bates  <dabates@apple.com>
-
-        lldb-webkit: Pretty-print OptionSet
-        https://bugs.webkit.org/show_bug.cgi?id=188936
-
-        Reviewed by Simon Fraser.
-
-        Add LLDB formatters to pretty-print an OptionSet.
-
-        * lldb/lldbWebKitTester/main.cpp:
-        (testSummaryProviders):
-        * lldb/lldb_webkit.py:
-        (__lldb_init_module):
-        (__lldb_init_module.lldb_webkit):
-        (WTFOptionSet_SummaryProvider):
-        (WTFOptionSetProvider):
-        (WTFOptionSetProvider.__init__):
-        (WTFOptionSetProvider.has_children):
-        (WTFOptionSetProvider.num_children):
-        (WTFOptionSetProvider.get_child_index):
-        (WTFOptionSetProvider.get_child_at_index):
-        (WTFOptionSetProvider.update):
-        * lldb/lldb_webkit_unittest.py:
-        (TestSummaryProviders.serial_test_WTFHashSet_tablesize_and_size):
-        (TestSummaryProviders):
-        (TestSummaryProviders.serial_test_WTFOptionSet_SummaryProvider_empty):
-        (TestSummaryProviders.serial_test_WTFOptionSet_SummaryProvider_simple):
-        (TestSummaryProviders.serial_test_WTFOptionSetProvider_empty):
-        (TestSummaryProviders.serial_test_WTFOptionSetProvider_simple):
-
-2018-08-27  Wenson Hsieh  <wenson_hsieh@apple.com>
-
-        [Attachment Support] [WK2] Images copied from Mail message view paste with the wrong file name in compose
-        https://bugs.webkit.org/show_bug.cgi?id=188957
-        <rdar://problem/43737715>
-
-        Reviewed by Darin Adler.
-
-        Add a new API test to verify that pasting a web archive containing several image elements with alt attributes
-        generates _WKAttachments whose names reflect those alt attributes.
-
-        * TestWebKitAPI/Tests/WebKitCocoa/WKAttachmentTests.mm:
-        (testGIFFileURL):
-        (testGIFData):
-        (TestWebKitAPI::TEST):
-
-2018-08-27  Alex Christensen  <achristensen@webkit.org>
-
-        Unreviewed, rolling out r235367.
-
-        Broke iOS build.
-
-        Reverted changeset:
-
-        "Translate 4 tests using WKPageLoaderClient to ObjC"
-        https://bugs.webkit.org/show_bug.cgi?id=188827
-        https://trac.webkit.org/changeset/235367
-
-2018-08-27  Alex Christensen  <achristensen@webkit.org>
-
-        Fix GTK build.
-
-        * TestWebKitAPI/CMakeLists.txt:
-
-2018-08-27  Alex Christensen  <achristensen@webkit.org>
-
-        Fix iOS build.
-
-        * TestWebKitAPI/PlatformWebView.h:
-
-2018-08-27  Alex Christensen  <achristensen@webkit.org>
-
-        Translate 4 tests using WKPageLoaderClient to ObjC
-        https://bugs.webkit.org/show_bug.cgi?id=188827
-
-        Reviewed by Tim Horton.
-
-        They use processDidBecomeUnresponsive, didChangeBackForwardList, or willGoToBackForwardListItem.
-        Rather than introduce these to WKPageNavigationClient, I just translated the tests to use WKNavigationDelegate, which already have equivalent callbacks.
-        willGoToBackForwardListItem had userData from the InjectedBundle, but nobody was using it so I did not add that to the ObjC SPI, so I don't test that unused
-        bundle functionality any more.
-
-        * TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj:
-        * TestWebKitAPI/Tests/WebKit/ResponsivenessTimer.cpp: Removed.
-        * TestWebKitAPI/Tests/WebKit/ResponsivenessTimerDoesntFireEarly.cpp: Removed.
-        * TestWebKitAPI/Tests/WebKit/RestoreSessionStateWithoutNavigation.cpp: Removed.
-        * TestWebKitAPI/Tests/WebKit/ShouldGoToBackForwardListItem.cpp: Removed.
-        * TestWebKitAPI/Tests/WebKit/ShouldGoToBackForwardListItem_Bundle.cpp: Removed.
-        * TestWebKitAPI/Tests/WebKitCocoa/ResponsivenessTimer.mm: Copied from Tools/TestWebKitAPI/Tests/WebKit/ResponsivenessTimer.cpp.
-        (-[ResponsivenessTimerDelegate webView:didFinishNavigation:]):
-        (-[ResponsivenessTimerDelegate _webViewWebProcessDidBecomeUnresponsive:]):
-        (TestWebKitAPI::TEST):
-        (): Deleted.
-        (TestWebKitAPI::didFinishLoadForFrame): Deleted.
-        (TestWebKitAPI::processDidBecomeUnresponsive): Deleted.
-        (TestWebKitAPI::setPageLoaderClient): Deleted.
-        * TestWebKitAPI/Tests/WebKitCocoa/ResponsivenessTimerDoesntFireEarly.mm: Copied from Tools/TestWebKitAPI/Tests/WebKit/ResponsivenessTimerDoesntFireEarly.cpp.
-        (-[ResponsivenessDelegate webView:didFinishNavigation:]):
-        (-[ResponsivenessDelegate _webViewWebProcessDidBecomeUnresponsive:]):
-        (TestWebKitAPI::TEST):
-        (TestWebKitAPI::didReceiveMessageFromInjectedBundle): Deleted.
-        (TestWebKitAPI::didFinishLoadForFrame): Deleted.
-        (TestWebKitAPI::processDidBecomeUnresponsive): Deleted.
-        (TestWebKitAPI::setInjectedBundleClient): Deleted.
-        (TestWebKitAPI::setPageLoaderClient): Deleted.
-        * TestWebKitAPI/Tests/WebKitCocoa/RestoreSessionStateWithoutNavigation.mm: Copied from Tools/TestWebKitAPI/Tests/WebKit/RestoreSessionStateWithoutNavigation.cpp.
-        (-[SessionStateDelegate webView:didFinishNavigation:]):
-        (-[SessionStateDelegate _webView:backForwardListItemAdded:removed:]):
-        (TestWebKitAPI::createSessionStateData):
-        (TestWebKitAPI::TEST):
-        (TestWebKitAPI::didFinishLoadForFrame): Deleted.
-        (TestWebKitAPI::didChangeBackForwardListForPage): Deleted.
-        (TestWebKitAPI::setPageLoaderClient): Deleted.
-        * TestWebKitAPI/Tests/WebKitCocoa/ShouldGoToBackForwardListItem.mm: Copied from Tools/TestWebKitAPI/Tests/WebKit/ShouldGoToBackForwardListItem.cpp.
-        (-[BackForwardClient webView:didFinishNavigation:]):
-        (-[BackForwardClient _webView:willGoToBackForwardListItem:inPageCache:]):
-        (TestWebKitAPI::TEST):
-        (TestWebKitAPI::didFinishLoadForFrame): Deleted.
-        (TestWebKitAPI::willGoToBackForwardListItem): Deleted.
-        (TestWebKitAPI::setPageLoaderClient): Deleted.
-
-2018-08-27  Keith Miller  <keith_miller@apple.com>
-
-        test262-runner -s --test-only should replace test results
-        https://bugs.webkit.org/show_bug.cgi?id=188450
-
-        Reviewed by Michael Saboff.
-
-        * Scripts/test262/Runner.pm:
-        (main):
-        (SetFailureForTest):
-        (UpdateResults):
-
-2018-08-26  Sam Weinig  <sam@webkit.org>
-
-        Using _WKRemoteObjectInterface with a protocol that inherits from a non-NSObject protocol crashes
-        https://bugs.webkit.org/show_bug.cgi?id=188958
-
-        Reviewed by Anders Carlsson.
-
-        * TestWebKitAPI/Tests/WebKitCocoa/RemoteObjectRegistry.h:
-        Update test protocol to have inheritance.
-
-2018-08-26  Wenson Hsieh  <wenson_hsieh@apple.com>
-
-        [Attachment Support] Dropping and pasting images should insert inline image elements with _WKAttachments
-        https://bugs.webkit.org/show_bug.cgi?id=188933
-        <rdar://problem/43699724>
-
-        Reviewed by Darin Adler.
-
-        Rebaseline existing API tests that involve dropping or pasting image files, and additionally write some new
-        tests. These new tests exercise the following cases:
-        •       Inserting and removing newlines before an inline image with an attachment element does not cause new
-                _WKAttachments to be created and destroyed.
-        •       Pasting an image, cutting it, and then pasting it again propagates an attachment update to the UI
-                process with the original _WKAttachment.
-        •       A pasted attachment in the document can be moved around by dragging, and doing so does not cause us to
-                lose a _WKAttachment.
-
-        * TestWebKitAPI/Tests/WebKitCocoa/WKAttachmentTests.mm:
-        (-[TestWKWebView expectElementCount:tagName:]):
-        (TestWebKitAPI::TEST):
-
-        Add the new tests described above, and also adjust existing tests to check that images are dropped or pasted
-        as image elements, but still have associated attachment elements whose attachment identifiers (observed via
-        script) match that of the corresponding _WKAttachment's uniqueIdentifier in the UI process.
-
-        * TestWebKitAPI/mac/DragAndDropSimulatorMac.mm:
-        (-[DragAndDropSimulator runFrom:to:]):
-        (-[DragAndDropSimulator continueDragSession]):
-        (-[DragAndDropSimulator performDragInWebView:atLocation:withImage:pasteboard:source:]):
-
-        Teach DragAndDropSimulator on macOS to wait until the drop has been handled by the web process before returning
-        execution to the caller. This ensures that tests which involve dropping promised files as attachments aren't
-        flaky, due to how the promised data is retrieved asynchronously when performing the drop.
-
-        (-[DragAndDropSimulator _webView:didPerformDragOperation:]):
-
-2018-08-26  Lucas Forschler  <lforschler@apple.com>
-
-        Open svn.webkit.org for commits.
-
-2018-08-24  Myles C. Maxfield  <mmaxfield@apple.com>
-
-        Unreviewed test fix after r235249
-        https://bugs.webkit.org/show_bug.cgi?id=178981
-
-        * WebGPUShadingLanguageRI/Test.js:
-
-2018-08-24  Alex Christensen  <achristensen@webkit.org>
-
-        Introduce _WKInspector
-        https://bugs.webkit.org/show_bug.cgi?id=188923
-        <rdar://problem/34657861>
-
-        Reviewed by Brian Burg.
-
-        * MiniBrowser/mac/WK2BrowserWindowController.m:
-        (-[WK2BrowserWindowController validateMenuItem:]):
-        (-[WK2BrowserWindowController showHideWebInspector:]):
-
-2018-08-24  Jonathan Bedard  <jbedard@apple.com>
-
-        Fix handling of iOS minor versions in default_baseline_search_path
-        https://bugs.webkit.org/show_bug.cgi?id=188902
-
-        Reviewed by Aakash Jain.
-
-        * Scripts/webkitpy/port/ios.py:
-        (IOSPort.default_baseline_search_path): When the major version matches the major version of the CURRENT_VERSION, treat
-        it as a the CURRENT_VERSION.
-
-2018-08-24  Antti Koivisto  <antti@apple.com>
-
-        Allow creating WeakPtrs to const objects
-        https://bugs.webkit.org/show_bug.cgi?id=188785
-
-        Reviewed by Geoff Garen.
-
-        * TestWebKitAPI/Tests/WTF/WeakPtr.cpp:
-        (TestWebKitAPI::Base::weakPtrFactory const):
-        (TestWebKitAPI::TEST):
-        (TestWebKitAPI::Base::weakPtrFactory): Deleted.
-
-2018-08-23  Carlos Garcia Campos  <cgarcia@igalia.com>
-
-        [GTK][WPE] Add API to inject/register user content in isolated worlds
-        https://bugs.webkit.org/show_bug.cgi?id=188883
-
-        Reviewed by Michael Catanzaro.
-
-        Add test cases for the new API.
-
-        * TestWebKitAPI/Tests/WebKitGLib/TestWebKitUserContentManager.cpp:
-        (isStyleSheetInjectedForURLAtPath):
-        (isScriptInjectedForURLAtPath):
-        (testUserContentManagerInjectedStyleSheet):
-        (testUserContentManagerInjectedScript):
-        (UserScriptMessageTest::registerHandler):
-        (UserScriptMessageTest::unregisterHandler):
-        (UserScriptMessageTest::postMessageAndWaitUntilReceived):
-        (testUserContentManagerScriptMessageInWorldReceived):
-        (beforeAll):
-
-2018-08-23  Simon Fraser  <simon.fraser@apple.com>
-
-        Add support for dumping GC heap snapshots, and a viewer
-        https://bugs.webkit.org/show_bug.cgi?id=186416
-
-        Reviewed by Joseph Pecoraro.
-
-        Add a viewer for GC heap snapshots. A snapshot JSON file can be dragged into this
-        page for inspection (or set via the 'filename' URL parameter).
-
-        For now, this page shows all objects, all roots, and the shortest path from a root
-        to all HTMLDocuments and Windows.
-
-        * GCHeapInspector/gc-heap-inspector.html: Added.
-        * GCHeapInspector/heap-analysis/HeapSnapshot.js: Copied from Source/WebInspectorUI/UserInterface/Workers/HeapSnapshot/HeapSnapshot.js.
-        * GCHeapInspector/script/interface.js: Added.
-
-2018-08-23  Alex Christensen  <achristensen@webkit.org>
-
-        Add new _webViewRequestPointerLock SPI with a completionHandler
-        https://bugs.webkit.org/show_bug.cgi?id=188907
-        <rdar://problem/35871109>
-
-        Reviewed by Andy Estes.
-
-        * TestWebKitAPI/Tests/WebKitCocoa/UIDelegate.mm:
-        (-[PointerLockDelegate _webViewRequestPointerLock:completionHandler:]):
-        (-[PointerLockDelegate webView:runJavaScriptAlertPanelWithMessage:initiatedByFrame:completionHandler:]):
-        (TEST):
-
-2018-08-23  David Fenton  <david_fenton@apple.com>
-
-        Unreviewed, rolling out r235129.
-
-        broke internal builds
-
-        Reverted changeset:
-
-        "Allow creating WeakPtrs to const objects"
-        https://bugs.webkit.org/show_bug.cgi?id=188785
-        https://trac.webkit.org/changeset/235129
-
-2018-08-23  Jonathan Bedard  <jbedard@apple.com>
-
-        API tests should output json results
-        https://bugs.webkit.org/show_bug.cgi?id=188869
-        <rdar://problem/43615652>
-
-        Reviewed by Aakash Jain.
-
-        JSON output for API tests is of the form:
-            {
-                "Failed": [{"name": <test name>, "output": <test log>}],
-                "Timedout": [...],
-                "Skipped": [...],
-                "Crashed": [...]
-            }
-        Tests which are successful are not displayed in the json output.
-
-        * Scripts/webkitpy/api_tests/manager.py:
-        (Manager.run): Print test results to provided file as a json dictionary.
-        * Scripts/webkitpy/api_tests/run_api_tests.py:
-        (run): Pass json option.
-        (parse_args): Add --json-output flag.
-
-2018-08-23  Andy Estes  <aestes@apple.com>
-
-        [Apple Pay] Introduce Apple Pay JS v4 on iOS 12 and macOS Mojave
-        https://bugs.webkit.org/show_bug.cgi?id=188829
-
-        Reviewed by Tim Horton.
-
-        * TestWebKitAPI/Configurations/FeatureDefines.xcconfig:
-
-2018-08-23  Myles C. Maxfield  <mmaxfield@apple.com>
-
-        [WSL] Ternary expressions appear to be unimplemented
-        https://bugs.webkit.org/show_bug.cgi?id=178981
-
-        Reviewed by Saam Barati.
-
-        Implement ternary statements. These can be both lvalues and rvalues. (a ? b : c ? d : e)
-        is parsed as (a ? b : (c ? d : e)).
-
-        * WebGPUShadingLanguageRI/All.js:
-        * WebGPUShadingLanguageRI/Checker.js:
-        (Checker.prototype.visitTernaryExpression):
-        * WebGPUShadingLanguageRI/Evaluator.js:
-        (Evaluator.prototype.visitTernaryExpression):
-        * WebGPUShadingLanguageRI/NormalUsePropertyResolver.js:
-        (NormalUsePropertyResolver.prototype.visitTernaryExpression):
-        (NormalUsePropertyResolver):
-        * WebGPUShadingLanguageRI/Parse.js:
-        (parsePossibleTernaryConditional):
-        * WebGPUShadingLanguageRI/PropertyResolver.js:
-        (PropertyResolver.prototype._visitRValuesWithinLValue.RValueFinder.prototype.visitTernaryExpression):
-        (PropertyResolver.prototype._visitRValuesWithinLValue.RValueFinder):
-        (PropertyResolver.prototype._visitRValuesWithinLValue):
-        * WebGPUShadingLanguageRI/Rewriter.js:
-        (Rewriter.prototype.visitTernaryExpression):
-        * WebGPUShadingLanguageRI/SPIRV.html:
-        * WebGPUShadingLanguageRI/Test.html:
-        * WebGPUShadingLanguageRI/Test.js:
-        * WebGPUShadingLanguageRI/Visitor.js:
-        (Visitor.prototype.visitProtocolDecl):
-        * WebGPUShadingLanguageRI/index.html:
-
-2018-08-22  Ryosuke Niwa  <rniwa@webkit.org>
-
-        Assert in NetworkBlobRegistry::unregisterBlobURL after network process had terminated
-        https://bugs.webkit.org/show_bug.cgi?id=188880
-
-        Reviewed by Saam Barati.
-
-        Fixed the bug that testRunner's terminateNetworkProcess, terminateServiceWorkerProcess, and terminateStorageProcess
-        were asynchronously terminating respective processes. Do so synchronously so that we can deterministically
-        test WebKit's behavior in layout tests.
-
-        * WebKitTestRunner/InjectedBundle/TestRunner.cpp:
-        (WTR::TestRunner::terminateNetworkProcess):
-        (WTR::TestRunner::terminateServiceWorkerProcess):
-        (WTR::TestRunner::terminateStorageProcess):
-        * WebKitTestRunner/TestInvocation.cpp:
-        (WTR::TestInvocation::didReceiveMessageFromInjectedBundle):
-        (WTR::TestInvocation::didReceiveSynchronousMessageFromInjectedBundle):
-
-2018-08-23  Myles C. Maxfield  <mmaxfield@apple.com>
-
-        [WHLSL] Allow native types to have type arguments (like "vector<float, 4>")
-        https://bugs.webkit.org/show_bug.cgi?id=188773
-
-        Reviewed by Filip Pizlo.
-
-        Before this patch, it was impossible to represent "native typedef vector<float, 4>" because NativeTypes couldn't have
-        typeArguments.
-
-        Previously, the way to identify a type was strictly by name, which was represented by a string. Therefore, when something like
-        "vector<int, 3>" was parsed, it would produce a TypeRef with the name "vector" and typeArguments [TypeRef, IntLiteral]. Then,
-        there was a pass to convert the TypeRef to have the name "int3" and no typeArguments. After this transformation, each type could
-        be uniquely identified by name. That name was then matched to the string-only NativeType name.
-
-        This is okay for vectors and matrices, but it is unfortunate for textures (e.g. Texture2D<float4>) because they don't have any
-        natural string-only name. In addition, the canonicalization would have to be made aware of the fact that Texture2D<float4> is
-        the same as Texture2D<vector<float, 4>>. Similarly, an author may wish to typedef float4 to a different name.
-
-        It would be possible to mangle the names of the texture types to something unique, but then we lose information about the inner
-        type. For example, if we did this, Visitor wouldn't recurse into the float4 when encountering Texture2D<float4> because that
-        information would be lost. This could potentially make operations like programWithUnnecessaryThingsRemoved() more difficult to
-        implement in the future.
-
-        So, it would be better to have each type uniquely identified by (name, typeArguments). TypeRef will therefore also have
-        typeArguments which are used to determine which type it is referencing. After this analysis is done to determine what each
-        TypeRef is referencing, subsequent passes shouldn't care about the typeArguments and should only care about the .type field
-        which had been set - this was true even before this patch.
-
-        This means that NameContext has to aggregate types that accept typeArguments into arrays, where each array holds all the Types
-        that have the same name but different typeArguments. Then, when we need to match a TypeRef with a Type, we can ask the
-        NameContext for the appropriate array. This is the same way that function resolution works.
-
-        We can use Node.unify() to determine whether a TypeRef matches a NativeType. Eventually, this will go away, but for now, this is
-        an okay start. This works just about the same way that function overload resolution works. 
-
-        * WebGPUShadingLanguageRI/All.js:
-        * WebGPUShadingLanguageRI/CallExpression.js:
-        (CallExpression.prototype.resolve):
-        * WebGPUShadingLanguageRI/CheckTypesWithArguments.js: Copied from Tools/WebGPUShadingLanguageRI/ResolveTypeDefs.js. After types
-        have been resolved, there should be no TypeRefs with name "vector" that don't have type arguments. This is just a sanity check.
-        (checkTypesWithArguments.TypeWithArgumentsChecker.prototype.visitTypeRef):
-        (checkTypesWithArguments.TypeWithArgumentsChecker):
-        (checkTypesWithArguments):
-        * WebGPUShadingLanguageRI/Checker.js:
-        (Checker.prototype.visitProgram): Program.types mirrors NameContext: it's a Map that maps strings to types. Because types with
-        typeArguments share names, this has to be updated to map strings to arrays for these types.
-        (Checker.prototype.visitTypeRef):
-        * WebGPUShadingLanguageRI/InferTypesForCall.js:
-        (inferTypesForCall): Don't know why this was here.
-        (inferTypesForTypeArguments): Same as inferTypesForCall, but this one is for matching type arguments.
-        * WebGPUShadingLanguageRI/Intrinsics.js: Adding the types. This patch also adds some scalar types like half, char, etc, but they
-        don't have any functions which accept them. Those will be tested in my next patch which adds math functions for these types. This
-        moves in the direction of matching the standard library in the spec.
-        (Intrinsics.cast):
-        (Intrinsics.bitwiseCast):
-        (Intrinsics.castToHalf):
-        (Intrinsics.):
-        (Intrinsics):
-        * WebGPUShadingLanguageRI/NameContext.js: Aggregate types with typeArguments into arrays.
-        (NameContext.prototype.add):
-        (NameContext.prototype.get let):
-        (NameContext.underlyingThings.prototype.else):
-        (NameContext.prototype.Symbol.iterator):
-        (NameContext):
-        * WebGPUShadingLanguageRI/NameResolver.js:
-        (NameResolver.prototype.visitTypeRef): Call TypeRef.resolve().
-        (NameResolver.prototype.visitCallExpression):
-        (NameResolver):
-        (NameResolver.prototype.visitVectorType): Deleted.
-        * WebGPUShadingLanguageRI/NativeType.js: NativeTypes can have type arguments now.
-        (NativeType):
-        (NativeType.prototype.get typeArguments):
-        (NativeType.prototype.toString):
-        (NativeType.create):
-        * WebGPUShadingLanguageRI/Prepare.js:
-        (let.prepare):
-        * WebGPUShadingLanguageRI/Program.js: Update to work with types aggregated into arrays.
-        (Program.prototype.add):
-        (Program.prototype.toString):
-        (Program):
-        * WebGPUShadingLanguageRI/RemoveTypeArguments.js: Removed.
-        * WebGPUShadingLanguageRI/ResolveNames.js: Update to work with types aggregated into arrays.
-        (resolveNamesInTypes):
-        * WebGPUShadingLanguageRI/ResolveOverloadImpl.js: Resolve the type overload for types with typeArguments.
-        * WebGPUShadingLanguageRI/ResolveTypeDefs.js: Update to work with types aggregated into arrays.
-        (resolveTypeDefsInTypes):
-        * WebGPUShadingLanguageRI/Rewriter.js: TypeRefs and Native/Vector types can have typeArguments.
-        (Rewriter.prototype.visitTypeRef):
-        (Rewriter.prototype.visitVectorType):
-        (Rewriter):
-        * WebGPUShadingLanguageRI/SPIRV.html:
-        * WebGPUShadingLanguageRI/StandardLibrary.js: Matches Intrinsics.
-        (bool.operator):
-        * WebGPUShadingLanguageRI/StatementCloner.js: Native types can have typeArguments.
-        (StatementCloner.prototype.visitNativeType):
-        * WebGPUShadingLanguageRI/SynthesizeDefaultConstructorOperator.js: Vector types need constructors too.
-        (synthesizeDefaultConstructorOperator.FindAllTypes.prototype.visitVectorType):
-        (synthesizeDefaultConstructorOperator.FindAllTypes):
-        (synthesizeDefaultConstructorOperator):
-        * WebGPUShadingLanguageRI/SynthesizeStructAccessors.js: No reason to distinguish between wrapping and instantiating a TypeRef.
-        (synthesizeStructAccessors.createTypeRef):
-        * WebGPUShadingLanguageRI/Test.html:
-        * WebGPUShadingLanguageRI/Test.js:
-        * WebGPUShadingLanguageRI/TypeOverloadResolutionFailure.js: Copied from Tools/WebGPUShadingLanguageRI/ResolveTypeDefs.js.
-        (TypeOverloadResolutionFailure):
-        (TypeOverloadResolutionFailure.prototype.get type):
-        (TypeOverloadResolutionFailure.prototype.get reason):
-        (TypeOverloadResolutionFailure.prototype.toString):
-        * WebGPUShadingLanguageRI/TypeRef.js:
-        (TypeRef.wrap):
-        (TypeRef.prototype.resolve): Figure out which item in the possibleOverloads array matches this.
-        (TypeRef.prototype.toString):
-        (TypeRef):
-        (TypeRef.instantiate): Deleted.
-        * WebGPUShadingLanguageRI/UnificationContext.js: We need to give literals a chance to assume their preferred type. This
-        adds this facility back into the compiler (it was previously deleted).
-        (UnificationContext.prototype.verify):
-        * WebGPUShadingLanguageRI/VectorType.js: Vector types have type arguments.
-        (VectorType):
-        (VectorType.prototype.get elementType):
-        (VectorType.prototype.get numElements):
-        (VectorType.prototype.get numElementsValue):
-        (VectorType.prototype.toString):
-        * WebGPUShadingLanguageRI/Visitor.js: Iterate over the typeArguments.
-        (Visitor.prototype.visitTypeRef):
-        (Visitor.prototype.visitNativeType):
-        (Visitor.prototype.visitVectorType):
-        (Visitor):
-        * WebGPUShadingLanguageRI/index.html:
-
-2018-08-23  Wenson Hsieh  <wenson_hsieh@apple.com>
-
-        [Attachment Support] Attachment elements don't appear in drag images on macOS
-        https://bugs.webkit.org/show_bug.cgi?id=188823
-        <rdar://problem/43616378>
-
-        Reviewed by Tim Horton.
-
-        Adjusts a couple of existing tests to additionally verify that the drag image generated when dragging an
-        attachment element in macOS is not completely transparent.
-
-        * TestWebKitAPI/Tests/WebKitCocoa/WKAttachmentTests.mm:
-        (isCompletelyTransparent):
-        (TestWebKitAPI::TEST):
-
-2018-08-23  Jonathan Bedard  <jbedard@apple.com>
-
-        Explain test name matching in run-api-tests help
-        https://bugs.webkit.org/show_bug.cgi?id=188280
-
-        Reviewed by Ryosuke Niwa.
-
-        Improve the run-api-tests help message to explain how test
-        name matching works.
-
-        * Scripts/webkitpy/api_tests/run_api_tests.py:
-        (parse_args):
-        * Scripts/webkitpy/port/base.py:
-        (Port):
-        (Port.path_to_api_test_binaries): Make binary names a globally accessible array.
-        * Scripts/webkitpy/port/win.py:
-        (WinPort):
-        (WinPort.path_to_api_test_binaries): Ditto.
-
-2018-08-23  Youenn Fablet  <youenn@apple.com>
-
-        Use "wpt serve" to launch WPT server
-        https://bugs.webkit.org/show_bug.cgi?id=188848
-
-        Reviewed by Ryosuke Niwa.
-
-        Use "wpt serve" provided by WPT instead of using our custom launcher.
-        This simplifies things and will avoid future breakage.
-        Further simplify web_platform_test_server.py by removing no longer needed actions.
-        - Do not copy files but use alias
-        - Do not kill main pid, which leaves subprocesses alive, use interrupt instead.
-        - Stop enumerating subprocess pids.
-
-        * Scripts/webkitpy/common/system/executive_mock.py:
-        (MockExecutive.interrupt):
-        * Scripts/webkitpy/layout_tests/servers/web_platform_test_launcher.py: Removed.
-        * Scripts/webkitpy/layout_tests/servers/web_platform_test_server.py:
-        (WebPlatformTestServer.__init__):
-        (WebPlatformTestServer.ports_to_forward):
-        (WebPlatformTestServer._prepare_config):
-        (WebPlatformTestServer._spawn_process):
-        (WebPlatformTestServer._stop_running_server):
-
-2018-08-23  Jonathan Bedard  <jbedard@apple.com>
-
-        run-api-tests: Add --webkit-only, --webcore-only and --webkit-legacy-only options to run WebKit, WebCore and WebKitLegacy tests 
-        https://bugs.webkit.org/show_bug.cgi?id=188262
-
-        Reviewed by Ryosuke Niwa.
-
-        * Scripts/webkitpy/api_tests/run_api_tests.py:
-        (parse_args):
-
-2018-08-22  Wenson Hsieh  <wenson_hsieh@apple.com>
-
-        [Attachment Support] Support dragging attachment elements out as files on macOS
-        https://bugs.webkit.org/show_bug.cgi?id=181294
-        <rdar://problem/36298801>
-
-        Reviewed by Tim Horton.
-
-        Add DragAndDropSimulator support for intercepting calls to -beginDraggingSessionWithitems:event:source:. This
-        enables us to write API tests for macOS that exercise the attachment SPI in combination with dragging attachment
-        elements.
-
-        * TestWebKitAPI/Tests/WebKitCocoa/WKAttachmentTests.mm:
-        (-[TestWKWebView attachmentElementMidPoint]):
-
-        Add a helper method local to this test suite that grabs the midpoint (in client coordinates) or the first
-        attachment element in the document.
-
-        (TestWebKitAPI::TEST):
-
-        Add a new API test to verify that dragging an attachment element on macOS produces file providers which may be
-        used to write attachment data to a path on disk. Additionally, refactor an existing API test,
-        MoveAttachmentElementAsIconByDragging, so that it runs on both iOS and macOS, to test the ability to move
-        attachment elements around in a document by using drag and drop.
-
-        * TestWebKitAPI/cocoa/DragAndDropSimulator.h:
-        * TestWebKitAPI/mac/DragAndDropSimulatorMac.mm:
-        (-[DragAndDropTestWKWebView beginDraggingSessionWithItems:event:source:]):
-        (-[DragAndDropSimulator initWithWebViewFrame:configuration:]):
-        (-[DragAndDropSimulator dealloc]):
-        (-[DragAndDropSimulator runFrom:to:]):
-        (-[DragAndDropSimulator beginDraggingSessionInWebView:withItems:source:]):
-
-        Begin a drag session and kick off the -continueDragSession loop. Unlike -performDragInWebView:…, which spins
-        the main runloop until dragging ends, this version returns execution to the web view and schedules dragging
-        updates asynchronously. This matches AppKit behavior.
-
-        (-[DragAndDropSimulator continueDragSession]):
-
-        Increment the dragging progress amount, send a drag update to the web view, and continue scheduling calls to
-        itself until the progress reaches 1.
-
-        (-[DragAndDropSimulator performDragInWebView:atLocation:withImage:pasteboard:source:]):
-        (-[DragAndDropSimulator initializeDraggingInfo:dragImage:source:]):
-
-        Pull out common logic for creating a new TestDraggingInfo after starting a drag.
-
-        (-[DragAndDropSimulator insertedAttachments]):
-        (-[DragAndDropSimulator removedAttachments]):
-        (-[DragAndDropSimulator draggingSession]):
-        (-[DragAndDropSimulator receivePromisedFiles]):
-
-        Helper method to save promised files written to the pasteboard after a drag to the temporary directory. These
-        files are tracked by DragAndDropSimulator and automatically cleaned up after the test finishes.
-
-        (-[DragAndDropSimulator endDataTransfer]):
-
-        Add a method stub on macOS, so that MoveAttachmentElementAsIconByDragging can be made cross-platform.
-
-        (-[DragAndDropSimulator _webView:didInsertAttachment:withSource:]):
-        (-[DragAndDropSimulator _webView:didRemoveAttachment:]):
-
-        Implement method stubs to keep track of inserted or removed attachments while simulating a drag.
-
-2018-08-22  Aditya Keerthi  <akeerthi@apple.com>
-
-        [iOS] Add support for the inputmode attribute
-        https://bugs.webkit.org/show_bug.cgi?id=183621
-
-        Reviewed by Tim Horton.
-
-        Added an API test to verify that the correct UIKeyboardType is set depending on
-        the type, inputmode, and pattern specified for an input field.
-
-        * TestWebKitAPI/Tests/ios/KeyboardInputTestsIOS.mm:
-        (TestWebKitAPI::TEST):
-        * TestWebKitAPI/ios/UIKitSPI.h:
-
-2018-08-22  Alex Christensen  <achristensen@webkit.org>
-
-        Fix API test on Sierra after r235139
-        https://bugs.webkit.org/show_bug.cgi?id=188810
-
-        * TestWebKitAPI/Tests/mac/LimitTitleSize.mm:
-        (TEST):
-
-2018-08-22  David Kilzer  <ddkilzer@apple.com>
-
-        Move files in WebCore project to match Xcode folder structure
-        <https://webkit.org/b/188851>
-
-        Reviewed by Tim Horton.
-
-        * Scripts/extract-localizable-strings:
-        * Scripts/update-webkit-localizable-strings:
-        - Update to match new location of extract-localizable-strings.pl.
-
-2018-08-21  Alex Christensen  <achristensen@webkit.org>
-
-        Roll out r235139 and r235146
-        https://bugs.webkit.org/show_bug.cgi?id=188805
-
-        * TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj:
-        * TestWebKitAPI/Tests/WebKit/ShouldKeepCurrentBackForwardListItemInList.cpp:
-
-2018-08-21  Wenson Hsieh  <wenson_hsieh@apple.com>
-
-        [Attachment Support] Remove _WKAttachments and notify the UI client upon mainframe navigation
-        https://bugs.webkit.org/show_bug.cgi?id=188715
-        <rdar://problem/43541790>
-
-        Reviewed by Tim Horton.
-
-        Adds API tests to exercises cases where (1) the UI client is notified of attachment removal upon mainframe
-        navigation, and (2) the UI client is notified of attachment removal upon web content process termination.
-
-        * TestWebKitAPI/Tests/WebKitCocoa/WKAttachmentTests.mm:
-        (TestWebKitAPI::ObserveAttachmentUpdatesForScope::expectAttachmentUpdates):
-        (TestWebKitAPI::TEST):
-
-2018-08-21  Alex Christensen  <achristensen@webkit.org>
-
-        Transition ResizeReversePaginatedWebView API test from WKPageLoaderClient to WKPageNavigationClient
-        https://bugs.webkit.org/show_bug.cgi?id=188821
-
-        Reviewed by Simon Fraser.
-
-        * TestWebKitAPI/Tests/WebKit/ResizeReversePaginatedWebView.cpp:
-        (TestWebKitAPI::didLayout):
-        (TestWebKitAPI::TEST):
-
-2018-08-21  Myles C. Maxfield  <mmaxfield@apple.com>
-
-        [WHLSL] Call expressions shouldn't have type arguments
-        https://bugs.webkit.org/show_bug.cgi?id=188770
-
-        Reviewed by Filip Pizlo.
-
-        Call expressions only had type arguments for casts, becuase native types can have type arguments.
-        However, instead of putting those type arguments on the CallExpression, we should parse the casted
-        type as a real type and not as an identifier, which puts the type arguments in the TypeRef.
-
-        Test: casts
-
-        * WebGPUShadingLanguageRI/CallExpression.js:
-        (CallExpression):
-        (CallExpression.prototype.get name):
-        (CallExpression.resolve):
-        (CallExpression.prototype.get typeArguments): Deleted.
-        (CallExpression.prototype.becomeCast): Deleted.
-        * WebGPUShadingLanguageRI/NameResolver.js:
-        (NameResolver.prototype.visitCallExpression):
-        * WebGPUShadingLanguageRI/Parse.js:
-        (parseConstexpr):
-        (parseTypeDef):
-        (parseLeftOperatorCall):
-        (parseCallExpression.let.parseArguments):
-        (parsePossiblePrefix):
-        (parsePossibleRelationalEquality):
-        (parseLeftLogicalExpression):
-        (parseIfStatement):
-        (parseWhile):
-        (parseFor):
-        (parseDo):
-        * WebGPUShadingLanguageRI/RemoveTypeArguments.js:
-        * WebGPUShadingLanguageRI/Rewriter.js:
-        (Rewriter.prototype.visitCallExpression):
-
-2018-08-21  Alex Christensen  <achristensen@webkit.org>
-
-        Translate WebKit.LimitTitleSize API test into ObjC
-        https://bugs.webkit.org/show_bug.cgi?id=188810
-
-        Reviewed by Andy Estes.
-
-        It uses WKPageLoaderClient.didReceiveTitleForFrame which has been replaced by KVO.
-        I'm translating the test into ObjC rather than introducing a new callback to WKPageNavigationClient.
-
-        * TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj:
-        * TestWebKitAPI/Tests/WebKit/LimitTitleSize.cpp: Removed.
-        * TestWebKitAPI/Tests/mac/LimitTitleSize.mm:
-        (TEST):
-        (-[LimitTitleSizeTestObserver observeValueForKeyPath:ofObject:change:context:]):
-        (TestWebKitAPI::TEST): Deleted.
-
-2018-08-21  Alex Christensen  <achristensen@webkit.org>
-
-        Remove unused shouldKeepCurrentBackForwardListItemInList check
-        https://bugs.webkit.org/show_bug.cgi?id=188805
-
-        Reviewed by Andy Estes.
-
-        * TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj:
-        * TestWebKitAPI/Tests/WebKit/ShouldKeepCurrentBackForwardListItemInList.cpp: Removed.
-
-2018-08-21  Alex Christensen  <achristensen@webkit.org>
-
-        Transition more API tests from WKPageLoaderClient to WKPageNavigationClient
-        https://bugs.webkit.org/show_bug.cgi?id=188813
-
-        Reviewed by Andy Estes.
-
-        * TestWebKitAPI/Tests/WebKit/NewFirstVisuallyNonEmptyLayout.cpp:
-        (TestWebKitAPI::didLayout):
-        (TestWebKitAPI::setPageLoaderClient):
-        * TestWebKitAPI/Tests/WebKit/NewFirstVisuallyNonEmptyLayoutFails.cpp:
-        (TestWebKitAPI::didFinishNavigation):
-        (TestWebKitAPI::didLayout):
-        (TestWebKitAPI::setPageLoaderClient):
-        (TestWebKitAPI::didFinishLoadForFrame): Deleted.
-        * TestWebKitAPI/Tests/WebKit/NewFirstVisuallyNonEmptyLayoutForImages.cpp:
-        (TestWebKitAPI::didLayout):
-        (TestWebKitAPI::setPageLoaderClient):
-
-2018-08-21  Wenson Hsieh  <wenson_hsieh@apple.com>
-
-        [Attachment Support] Augment _WKAttachment SPI to handle NSFileWrappers in addition to NSData
-        https://bugs.webkit.org/show_bug.cgi?id=188496
-        <rdar://problem/43216836>
-
-        Reviewed by Tim Horton.
-
-        Adjusts existing attachment API tests. See below for more detail.
-
-        * TestWebKitAPI/Tests/WebKitCocoa/WKAttachmentTests.mm:
-        (-[TestWKWebView synchronouslyInsertAttachmentWithFileWrapper:contentType:]):
-        (-[TestWKWebView synchronouslyInsertAttachmentWithFilename:contentType:data:]):
-        (-[_WKAttachment synchronouslySetData:newContentType:newFilename:error:]):
-        (-[_WKAttachment synchronouslySetFileWrapper:newContentType:error:]):
-
-        Move off of deprecated attachment SPI, and add new helper functions to synchronously insert a new attachment or
-        update an existing attachment with a file wrapper.
-
-        (-[_WKAttachment expectRequestedDataToBe:]):
-        (TestWebKitAPI::TEST):
-
-        Add a new test to verify that file-URL-backed NSFileWrappers can be used to insert and update attachment data.
-        Also augment an existing test to check that an attachment element which has been copied and pasted within the
-        same document has a different _WKAttachment wrapper object than its duplicate, but both _WKAttachments are
-        backed by the same NSFileWrapper that was originally used to insert the attachment.
-
-        Additionally, add another macOS test to verify that dropping promised files in an attachment-element-enabled
-        editable area inserts attachment elements into the document and notifies the UI client with the inserted
-        attachment data.
-
-        (-[_WKAttachment synchronouslyRequestInfo:]): Deleted.
-        (-[_WKAttachment synchronouslyRequestData:]): Deleted.
-        * TestWebKitAPI/ios/DragAndDropSimulatorIOS.mm:
-        (-[DragAndDropSimulator _webView:didInsertAttachment:withSource:]):
-
-        Move off of -_webView:didInsertAttachment:.
-
-        (-[DragAndDropSimulator _webView:didInsertAttachment:]): Deleted.
-
-2018-08-21  David Kilzer  <ddkilzer@apple.com>
-
-        sort-Xcode-project-file: top-level files and folders are sorted unintentionally when `mainGroup` has no name
-        <https://webkit.org/b/188755>
-
-        Reviewed by Andy Estes.
-
-        * Scripts/sort-Xcode-project-file: Fix regular expression to
-        make `mainGroup` name optional.
-
-2018-08-21  Andy VanWagoner  <andy@vanwagoner.family>
-
-        Unreviewed, add myself to committers list.
-
-        * Scripts/webkitpy/common/config/contributors.json:
-
-2018-08-21  Antti Koivisto  <antti@apple.com>
-
-        Allow creating WeakPtrs to const objects
-        https://bugs.webkit.org/show_bug.cgi?id=188785
-
-        Reviewed by Geoffrey Garen.
-
-        * TestWebKitAPI/Tests/WTF/WeakPtr.cpp:
-        (TestWebKitAPI::Base::weakPtrFactory const):
-        (TestWebKitAPI::TEST):
-        (TestWebKitAPI::Base::weakPtrFactory): Deleted.
-
-2018-08-21  Per Arne Vollan  <pvollan@apple.com>
-
-        [WebGL] Contexts are not updated when display configuration changed.
-        https://bugs.webkit.org/show_bug.cgi?id=188750
-
-        Reviewed by Brent Fulgham.
-
-        * WebKitTestRunner/InjectedBundle/Bindings/TestRunner.idl:
-        * WebKitTestRunner/InjectedBundle/TestRunner.cpp:
-        (WTR::TestRunner::sendDisplayConfigurationChangedMessageForTesting):
-        * WebKitTestRunner/InjectedBundle/TestRunner.h:
-        * WebKitTestRunner/TestController.cpp:
-        (WTR::TestController::sendDisplayConfigurationChangedMessageForTesting):
-        * WebKitTestRunner/TestController.h:
-        * WebKitTestRunner/TestInvocation.cpp:
-        (WTR::TestInvocation::didReceiveSynchronousMessageFromInjectedBundle):
-
-2018-08-21  John Wilander  <wilander@apple.com>
-
-        Make ResourceLoadObserver::logWebSocketLoading() handle websockets in detached frames
-        https://bugs.webkit.org/show_bug.cgi?id=188757
-        <rdar://problem/38713390>
-
-        Reviewed by Alex Christensen.
-
-        These changes add the new WebKitTestRunner function
-        isStatisticsRegisteredAsSubresourceUnder().
-
-        * WebKitTestRunner/InjectedBundle/Bindings/TestRunner.idl:
-        * WebKitTestRunner/InjectedBundle/TestRunner.cpp:
-        (WTR::TestRunner::isStatisticsRegisteredAsSubresourceUnder):
-        * WebKitTestRunner/InjectedBundle/TestRunner.h:
-        * WebKitTestRunner/TestController.cpp:
-        (WTR::TestController::isStatisticsRegisteredAsSubresourceUnder):
-        * WebKitTestRunner/TestController.h:
-        * WebKitTestRunner/TestInvocation.cpp:
-        (WTR::TestInvocation::didReceiveSynchronousMessageFromInjectedBundle):
-
-2018-08-21  Alex Christensen  <achristensen@webkit.org>
-
-        Transition more API tests from WKPageLoaderClient to WKPageNavigationClient
-        https://bugs.webkit.org/show_bug.cgi?id=188797
-
-        Reviewed by Tim Horton.
-
-        * TestWebKitAPI/Tests/WebKit/AboutBlankLoad.cpp:
-        (TestWebKitAPI::didFinishNavigation):
-        (TestWebKitAPI::TEST):
-        (TestWebKitAPI::didFinishLoadForFrame): Deleted.
-        * TestWebKitAPI/Tests/WebKit/EphemeralSessionPushStateNoHistoryCallback.cpp:
-        (TestWebKitAPI::didSameDocumentNavigation):
-        (TestWebKitAPI::TEST):
-        (TestWebKitAPI::didSameDocumentNavigationForFrame): Deleted.
-        * TestWebKitAPI/Tests/WebKit/FailedLoad.cpp:
-        (TestWebKitAPI::didFailProvisionalNavigation):
-        (TestWebKitAPI::TEST):
-        (TestWebKitAPI::didFailProvisionalLoadWithErrorForFrame): Deleted.
-        * TestWebKitAPI/Tests/WebKit/FrameMIMETypeHTML.cpp:
-        (TestWebKitAPI::didStartProvisionalNavigation):
-        (TestWebKitAPI::didCommitNavigation):
-        (TestWebKitAPI::didFinishNavigation):
-        (TestWebKitAPI::TEST):
-        (TestWebKitAPI::didStartProvisionalLoadForFrame): Deleted.
-        (TestWebKitAPI::didCommitLoadForFrame): Deleted.
-        (TestWebKitAPI::didFinishLoadForFrame): Deleted.
-        * TestWebKitAPI/Tests/WebKit/FrameMIMETypePNG.cpp:
-        (TestWebKitAPI::didStartProvisionalNavigation):
-        (TestWebKitAPI::didCommitNavigation):
-        (TestWebKitAPI::didFinishNavigation):
-        (TestWebKitAPI::TEST):
-        (TestWebKitAPI::didStartProvisionalLoadForFrame): Deleted.
-        (TestWebKitAPI::didCommitLoadForFrame): Deleted.
-        (TestWebKitAPI::didFinishLoadForFrame): Deleted.
-        * TestWebKitAPI/Tests/WebKit/LayoutMilestonesWithAllContentInFrame.cpp:
-        (TestWebKitAPI::renderingProgressDidChange):
-        (TestWebKitAPI::TEST):
-        (TestWebKitAPI::didLayout): Deleted.
-        * TestWebKitAPI/Tests/WebKit/NewFirstVisuallyNonEmptyLayoutFrames.cpp:
-        (TestWebKitAPI::didFinishNavigation):
-        (TestWebKitAPI::renderingProgressDidChange):
-        (TestWebKitAPI::setPageLoaderClient):
-        (TestWebKitAPI::didFinishLoadForFrame): Deleted.
-        (TestWebKitAPI::didLayout): Deleted.
-        * TestWebKitAPI/Tests/WebKit/PageLoadBasic.cpp:
-        (TestWebKitAPI::State::State):
-        (TestWebKitAPI::didStartProvisionalNavigation):
-        (TestWebKitAPI::didCommitNavigation):
-        (TestWebKitAPI::didFinishNavigation):
-        (TestWebKitAPI::decidePolicyForNavigationAction):
-        (TestWebKitAPI::decidePolicyForResponse):
-        (TestWebKitAPI::TEST):
-        (TestWebKitAPI::didStartProvisionalLoadForFrame): Deleted.
-        (TestWebKitAPI::didCommitLoadForFrame): Deleted.
-        (TestWebKitAPI::didFinishLoadForFrame): Deleted.
-        (TestWebKitAPI::decidePolicyForNewWindowAction): Deleted.
-        * TestWebKitAPI/Tests/WebKit/PageLoadDidChangeLocationWithinPageForFrame.cpp:
-        (TestWebKitAPI::didSameDocumentNavigation):
-        (TestWebKitAPI::TEST):
-        (TestWebKitAPI::didFinishLoadForFrame): Deleted.
-        (TestWebKitAPI::didSameDocumentNavigationForFrame): Deleted.
-        * TestWebKitAPI/Tests/WebKit/PrivateBrowsingPushStateNoHistoryCallback.cpp:
-        (TestWebKitAPI::TEST):
-        (TestWebKitAPI::didSameDocumentNavigationForFrame): Deleted.
-        * TestWebKitAPI/Tests/WebKit/ReloadPageAfterCrash.cpp:
-        (TestWebKitAPI::didFinishLoad):
-        (TestWebKitAPI::TEST):
-        * TestWebKitAPI/Tests/WebKit/WKBundleFileHandle.cpp:
-        (TestWebKitAPI::didFinishNavigation):
-        (TestWebKitAPI::TEST):
-        (TestWebKitAPI::didFinishLoadForFrame): Deleted.
-        * TestWebKitAPI/Tests/WebKit/mac/RestoreStateAfterTermination.mm:
-        (TestWebKitAPI::didFinishLoad):
-        (TestWebKitAPI::TEST):
-        * TestWebKitAPI/Tests/WebKitCocoa/CommandBackForward.mm:
-        (WebKit2_CommandBackForwardTestWKView::SetUp):
-        (WebKit2_CommandBackForwardTestWKView::didFinishLoadForFrame): Deleted.
-
-2018-08-20  Alex Christensen  <achristensen@webkit.org>
-
-        Replace WKPageLoaderClient with WKPageNavigationClient in many API tests
-        https://bugs.webkit.org/show_bug.cgi?id=188771
-
-        Reviewed by Tim Horton.
-
-        * TestWebKitAPI/Tests/WebKit/CloseThenTerminate.cpp:
-        (TestWebKitAPI::didFinishNavigation):
-        (TestWebKitAPI::TEST):
-        (TestWebKitAPI::didFinishLoadForFrame): Deleted.
-        * TestWebKitAPI/Tests/WebKit/CookieManager.cpp:
-        (TestWebKitAPI::didFinishNavigation):
-        (TestWebKitAPI::TEST):
-        (TestWebKitAPI::didFinishLoadForFrame): Deleted.
-        * TestWebKitAPI/Tests/WebKit/DeferredViewInWindowStateChange.mm:
-        (TestWebKitAPI::didFinishNavigation):
-        (TestWebKitAPI::setPageLoaderClient):
-        (TestWebKitAPI::didFinishLoadForFrame): Deleted.
-        * TestWebKitAPI/Tests/WebKit/DidNotHandleKeyDown.cpp:
-        (TestWebKitAPI::TEST):
-        (TestWebKitAPI::didFinishLoadForFrame): Deleted.
-        * TestWebKitAPI/Tests/WebKit/DidRemoveFrameFromHiearchyInPageCache.cpp:
-        (TestWebKitAPI::didFinishNavigation):
-        (TestWebKitAPI::setPageLoaderClient):
-        (TestWebKitAPI::didFinishLoadForFrame): Deleted.
-        * TestWebKitAPI/Tests/WebKit/EventModifiers.cpp:
-        (TestWebKitAPI::didFinishNavigation):
-        (TestWebKitAPI::setClients):
-        (TestWebKitAPI::didFinishLoadForFrame): Deleted.
-        * TestWebKitAPI/Tests/WebKit/Find.cpp:
-        (TestWebKitAPI::didFinishNavigation):
-        (TestWebKitAPI::TEST):
-        (TestWebKitAPI::didFinishLoadForFrame): Deleted.
-        * TestWebKitAPI/Tests/WebKit/FindMatches.mm:
-        (TestWebKitAPI::didFinishNavigation):
-        (TestWebKitAPI::TEST):
-        (TestWebKitAPI::didFinishLoadForFrame): Deleted.
-        * TestWebKitAPI/Tests/WebKit/ForceRepaint.cpp:
-        (TestWebKitAPI::didFinishNavigation):
-        (TestWebKitAPI::TEST):
-        (TestWebKitAPI::didFinishLoadForFrame): Deleted.
-        * TestWebKitAPI/Tests/WebKit/FrameHandle.cpp:
-        (TestWebKitAPI::didFinishNavigation):
-        (TestWebKitAPI::TEST):
-        (TestWebKitAPI::didFinishLoadForFrame): Deleted.
-        * TestWebKitAPI/Tests/WebKit/Geolocation.cpp:
-        (TestWebKitAPI::didFinishNavigation):
-        (TestWebKitAPI::TEST):
-        (TestWebKitAPI::didFinishLoadForFrame): Deleted.
-        * TestWebKitAPI/Tests/WebKit/HitTestResultNodeHandle.cpp:
-        (TestWebKitAPI::didFinishNavigation):
-        (TestWebKitAPI::setPageLoaderClient):
-        (TestWebKitAPI::didFinishLoadForFrame): Deleted.
-        * TestWebKitAPI/Tests/WebKit/InjectedBundleBasic.cpp:
-        (TestWebKitAPI::didFinishNavigation):
-        (TestWebKitAPI::TEST):
-        (TestWebKitAPI::didFinishLoadForFrame): Deleted.
-        * TestWebKitAPI/Tests/WebKit/LimitTitleSize.cpp:
-        (TestWebKitAPI::didFinishLoadForFrame):
-        * TestWebKitAPI/Tests/WebKit/LoadAlternateHTMLStringWithNonDirectoryURL.cpp:
-        (TestWebKitAPI::didFinishNavigation):
-        (TestWebKitAPI::loadAlternateHTMLString):
-        (TestWebKitAPI::didFinishLoadForFrame): Deleted.
-        * TestWebKitAPI/Tests/WebKit/LoadCanceledNoServerRedirectCallback.cpp:
-        (TestWebKitAPI::didFinishNavigation):
-        (TestWebKitAPI::TEST):
-        (TestWebKitAPI::didFinishLoadForFrame): Deleted.
-        * TestWebKitAPI/Tests/WebKit/LoadPageOnCrash.cpp:
-        (TestWebKitAPI::WebKit2CrashLoader::WebKit2CrashLoader):
-        (TestWebKitAPI::didFinishLoad):
-        * TestWebKitAPI/Tests/WebKit/MenuTypesForMouseEvents.cpp:
-        (TestWebKitAPI::didFinishNavigation):
-        (TestWebKitAPI::setPageLoaderClient):
-        (TestWebKitAPI::didFinishLoadForFrame): Deleted.
-        * TestWebKitAPI/Tests/WebKit/MouseMoveAfterCrash.cpp:
-        (TestWebKitAPI::didFinishNavigation):
-        (TestWebKitAPI::setPageLoaderClient):
-        (TestWebKitAPI::didFinishLoadForFrame): Deleted.
-        * TestWebKitAPI/Tests/WebKit/NewFirstVisuallyNonEmptyLayoutFails.cpp:
-        (TestWebKitAPI::didFinishLoadForFrame):
-        * TestWebKitAPI/Tests/WebKit/PageLoadBasic.cpp:
-        (TestWebKitAPI::TEST):
-        * TestWebKitAPI/Tests/WebKit/PendingAPIRequestURL.cpp:
-        (TestWebKitAPI::TEST):
-        * TestWebKitAPI/Tests/WebKit/ResizeWindowAfterCrash.cpp:
-        (TestWebKitAPI::didFinishLoad):
-        (TestWebKitAPI::TEST):
-        * TestWebKitAPI/Tests/WebKit/RestoreSessionState.cpp:
-        (TestWebKitAPI::didFinishNavigation):
-        (TestWebKitAPI::setPageLoaderClient):
-        (TestWebKitAPI::didFinishLoadForFrame): Deleted.
-        * TestWebKitAPI/Tests/WebKit/RestoreSessionStateContainingFormData.cpp:
-        (TestWebKitAPI::didFinishNavigation):
-        (TestWebKitAPI::setPageLoaderClient):
-        (TestWebKitAPI::didFinishLoadForFrame): Deleted.
-        * TestWebKitAPI/Tests/WebKit/ScrollPinningBehaviors.cpp:
-        (TestWebKitAPI::didFinishNavigation):
-        (TestWebKitAPI::TEST):
-        (TestWebKitAPI::didFinishDocumentLoadForFrame): Deleted.
-        * TestWebKitAPI/Tests/WebKit/SpacebarScrolling.cpp:
-        (TestWebKitAPI::didFinishNavigation):
-        (TestWebKitAPI::TEST):
-        (TestWebKitAPI::didFinishLoadForFrame): Deleted.
-        * TestWebKitAPI/Tests/WebKit/StopLoadingDuringDidFailProvisionalLoad.cpp:
-        (TestWebKitAPI::didFailProvisionalNavigation):
-        (TestWebKitAPI::TEST):
-        (TestWebKitAPI::didFailProvisionalLoadWithErrorForFrame): Deleted.
-        * TestWebKitAPI/Tests/WebKit/TerminateTwice.cpp:
-        (TestWebKitAPI::didFinishNavigation):
-        (TestWebKitAPI::TEST):
-        (TestWebKitAPI::didFinishLoadForFrame): Deleted.
-        * TestWebKitAPI/Tests/WebKit/TextFieldDidBeginAndEndEditing.cpp:
-        (TestWebKitAPI::WebKit2TextFieldBeginAndEditEditingTest::didFinishNavigation):
-        (TestWebKitAPI::WebKit2TextFieldBeginAndEditEditingTest::setPageLoaderClient):
-        (TestWebKitAPI::WebKit2TextFieldBeginAndEditEditingTest::didFinishLoadForFrame): Deleted.
-        * TestWebKitAPI/Tests/WebKit/UserMessage.cpp:
-        (TestWebKitAPI::WebKit2UserMessageRoundTripTest::didFinishNavigation):
-        (TestWebKitAPI::WebKit2UserMessageRoundTripTest::setPageLoaderClient):
-        (TestWebKitAPI::WebKit2UserMessageRoundTripTest::didFinishLoadForFrame): Deleted.
-        * TestWebKitAPI/Tests/WebKit/WKPageConfiguration.cpp:
-        (TestWebKitAPI::didFinishNavigation):
-        (TestWebKitAPI::setPageLoaderClient):
-        (TestWebKitAPI::didFinishLoadForFrame): Deleted.
-        * TestWebKitAPI/Tests/WebKit/WKPageCopySessionStateWithFiltering.cpp:
-        (TestWebKitAPI::didFinishNavigation):
-        (TestWebKitAPI::setPageLoaderClient):
-        (TestWebKitAPI::didFinishLoadForFrame): Deleted.
-        * TestWebKitAPI/Tests/WebKit/WKPageGetScaleFactorNotZero.cpp:
-        (TestWebKitAPI::didFinishNavigation):
-        (TestWebKitAPI::setPageLoaderClient):
-        (TestWebKitAPI::didFinishLoadForFrame): Deleted.
-        * TestWebKitAPI/Tests/WebKit/WKPageIsPlayingAudio.cpp:
-        (TestWebKitAPI::didFinishNavigation):
-        (TestWebKitAPI::setUpClients):
-        (TestWebKitAPI::didFinishLoadForFrame): Deleted.
-        * TestWebKitAPI/Tests/WebKit/WKThumbnailView.mm:
-        (TestWebKitAPI::didFinishNavigation):
-        (TestWebKitAPI::setPageLoaderClient):
-        (TestWebKitAPI::didFinishLoadForFrame): Deleted.
-        * TestWebKitAPI/Tests/WebKit/WebArchive.cpp:
-        (TestWebKitAPI::didFinishNavigation):
-        (TestWebKitAPI::TEST):
-        (TestWebKitAPI::didFinishLoadForFrame): Deleted.
-        * TestWebKitAPI/Tests/WebKit/mac/AttributedSubstringForProposedRangeWithImage.mm:
-        (TestWebKitAPI::didFinishNavigation):
-        (TestWebKitAPI::TEST):
-        (TestWebKitAPI::didFinishLoadForFrame): Deleted.
-        * TestWebKitAPI/Tests/WebKit/mac/ContextMenuDownload.mm:
-        (TestWebKitAPI::didFinishNavigation):
-        (TestWebKitAPI::TEST):
-        (TestWebKitAPI::didFinishLoadForFrame): Deleted.
-        * TestWebKitAPI/Tests/WebKit/mac/CustomBundleParameter.mm:
-        (TestWebKitAPI::didFinishNavigation):
-        (TestWebKitAPI::TEST):
-        (TestWebKitAPI::didFinishLoadForFrame): Deleted.
-        * TestWebKitAPI/Tests/WebKit/mac/EditorCommands.mm:
-        (TestWebKitAPI::didFinishNavigation):
-        (TestWebKitAPI::TEST):
-        (TestWebKitAPI::didFinishLoadForFrame): Deleted.
-        * TestWebKitAPI/Tests/WebKit/mac/GetPIDAfterAbortedProcessLaunch.cpp:
-        (TestWebKitAPI::didFinishNavigation):
-        (TestWebKitAPI::TEST):
-        (TestWebKitAPI::didFinishLoadForFrame): Deleted.
-        * TestWebKitAPI/Tests/WebKit/mac/InjectedBundleAppleEvent.cpp:
-        (TestWebKitAPI::didFinishNavigation):
-        (TestWebKitAPI::TEST):
-        (TestWebKitAPI::didFinishLoadForFrame): Deleted.
-        * TestWebKitAPI/Tests/WebKitCocoa/FullscreenDelegate.mm:
-        (didFinishNavigation):
-        (TestWebKitAPI::TEST):
-        (didFinishLoadForFrame): Deleted.
-        * TestWebKitAPI/Tests/WebKitCocoa/PictureInPictureDelegate.mm:
-        (didFinishNavigation):
-        (TestWebKitAPI::TEST):
-        (didFinishLoadForFrame): Deleted.
-        * TestWebKitAPI/Tests/mac/FirstResponderScrollingPosition.mm:
-        (TestWebKitAPI::didFinishNavigation):
-        (TestWebKitAPI::TEST):
-        (TestWebKitAPI::didFinishLoadForFrame): Deleted.
-        * TestWebKitAPI/mac/WebKitAgnosticTest.mm:
-        (TestWebKitAPI::didFinishNavigation):
-        (TestWebKitAPI::setPageLoaderClient):
-        (TestWebKitAPI::didFinishLoadForFrame): Deleted.
-
-2018-08-21  Adrian Perez de Castro  <aperez@igalia.com>
-
-        [WPE] Update to use libwpe-1.0.0 and WPEBackend-fdo-1.0.0
-        https://bugs.webkit.org/show_bug.cgi?id=188782
-
-        Reviewed by Michael Catanzaro.
-
-        Make the build depend on wpe-0.2, and change the Flatpak and JHBuild development
-        environments to use version 1.0.0 of libwpe and WPEBackend-fdo.
-
-        * flatpak/org.webkit.WPE.yaml: Update to use libwpe and WPEBackend-fdo version 1.0.0 from
-        release tarballs, and removed the (now unneeded) Lua and LuaJIT modules.
-        * wpe/jhbuild.modules: Ditto.
-
-2018-08-21  Michael Catanzaro  <mcatanzaro@igalia.com>
-
-        [Flatpak] Update to GNOME master runtime
-        https://bugs.webkit.org/show_bug.cgi?id=188731
-
-        Reviewed by Philippe Normand.
-
-        The 3.28 runtime is pretty stale at this point. We could wait a couple weeks for the 3.30
-        runtime, but I don't think there's any need to stick with the stable runtime when we have
-        the ability to pin to a particular version of the master runtime. There are also several
-        advantages to switching to the master runtime.
-
-        * flatpak/files/httpd-autogen.sh: Removed.
-        * flatpak/flatpakutils.py:
-        (WebkitFlatpak.clean_args):
-        * flatpak/org.webkit.GTK.yaml:
-        * flatpak/org.webkit.WebKit.yaml:
-        * flatpak/patches/httpd-0001-configure-use-pkg-config-for-PCRE-detection.patch: Removed.
-        * flatpak/patches/libgcrypt-0001-Use-pkg-config-to-detect-gpg-error.patch: Removed.
-
-2018-08-20  Jonathan Bedard  <jbedard@apple.com>
-
-        WebKitTestRunner: Add watchOS entitlements
-        https://bugs.webkit.org/show_bug.cgi?id=188758
-
-        Reviewed by Tim Horton.
-
-        watchOS needs an additional entitlement to open up a socket for on-device testing.
-
-        * WebKitTestRunner/Configurations/WebKitTestRunnerApp-watchOS.entitlements: Added.
-        * WebKitTestRunner/Configurations/WebKitTestRunnerApp.xcconfig:
-        * WebKitTestRunner/WebKitTestRunner.xcodeproj/project.pbxproj:
-
-2018-08-20  Thomas Denney  <tdenney@apple.com>
-
-        [WHLSL] Remove generics from the interpreter
-        https://bugs.webkit.org/show_bug.cgi?id=187988
-
-        Reviewed by Myles C. Maxfield.
-
-        This patch completely removes generic and protocol support from WHLSL
-        whilst adding back any features that were originally implemented with
-        generics.
-
-        Many of the changes in this patch just remove occurrences of type
-        arguments and type parameters --- although there are some cases where
-        they have been deliberately left in --- as well as removing tests only
-        relevant to those features. These changes are not listed below.
-
-        * WebGPUShadingLanguageRI/All.js: Add/remove JS dependencies
-        * WebGPUShadingLanguageRI/AutoWrapper.js:
-        (AutoWrapper.prototype.visitConstexprTypeParameter): Deleted.
-        * WebGPUShadingLanguageRI/BuiltinVectorCasts.js: Added. The built-in
-        vectors have many constructors for different scalar and vector types.
-        These were originally implemented with generics.
-        * WebGPUShadingLanguageRI/BuiltinVectorEqualityOperator.js: Adds the
-        equality operator for built-in vectors, again this was removed with
-        generics.
-        * WebGPUShadingLanguageRI/BuiltinVectorGetter.js: Added, as above.
-        Built-in vectors only support the getter and setter, and not the ander.
-        * WebGPUShadingLanguageRI/BuiltinVectorIndexGetter.js: Added.
-        * WebGPUShadingLanguageRI/BuiltinVectorIndexSetter.js: Added.
-        * WebGPUShadingLanguageRI/BuiltinVectorSetter.js: Added.
-        * WebGPUShadingLanguageRI/CallExpression.js: Adds support for
-        instantiating functions which cannot be discovered earlier in
-        preparation because their types are absent (specifically support for the
-        operator &[] for array references and the length operator for array
-        references). Previously support for these was implemented with generics.
-        (CallExpression.resolve):
-        (CallExpression.prototype.resolve):
-        (CallExpression.prototype._resolveByInstantiation):
-        (CallExpression.prototype._resolveToOperatorAnderIndexer):
-        (CallExpression.prototype._resolveToOperatorLength):
-        (CallExpression.prototype.resolveToOverload):
-        (CallExpression.prototype.becomeCast):
-        (CallExpression.prototype.toString):
-        (CallExpression):
-        * WebGPUShadingLanguageRI/CallExpressionTypeArgumentResolver.js: Added.
-        We support syntax like vector<int, 2>(a, b) for constructing vectors;
-        this new preparation step resolves them to specific constructors.
-        * WebGPUShadingLanguageRI/CallFunction.js:
-        * WebGPUShadingLanguageRI/Checker.js:
-        * WebGPUShadingLanguageRI/ConstexprFolder.js:
-        * WebGPUShadingLanguageRI/CreateLiteralType.js:
-        * WebGPUShadingLanguageRI/ExpressionFinder.js:
-        * WebGPUShadingLanguageRI/FlattenProtocolExtends.js: Removed.
-        * WebGPUShadingLanguageRI/FlattenedStructOffsetGatherer.js:
-        * WebGPUShadingLanguageRI/Func.js:
-        * WebGPUShadingLanguageRI/FuncDef.js:
-        * WebGPUShadingLanguageRI/FuncInstantiator.js: Deleted.
-        * WebGPUShadingLanguageRI/InferTypesForCall.js: Removed usage of type arguments.
-        * WebGPUShadingLanguageRI/Inline.js: Remove use of FuncInstantiator.
-        * WebGPUShadingLanguageRI/Inliner.js: Ditto.
-        * WebGPUShadingLanguageRI/Intrinsics.js:
-        * WebGPUShadingLanguageRI/Lexer.js:
-        * WebGPUShadingLanguageRI/LiteralTypeChecker.js:
-        * WebGPUShadingLanguageRI/NameContext.js:
-        (NameContext.prototype.add):
-        (NameContext.prototype.get let):
-        (NameContext.underlyingThings.prototype.else):
-        (NameContext.prototype.resolveFuncOverload):
-        * WebGPUShadingLanguageRI/NameFinder.js:
-        * WebGPUShadingLanguageRI/NameResolver.js:
-        (NameResolver.prototype.visitFunc):
-        (NameResolver.prototype.visitFuncDef):
-        (NameResolver.prototype.visitTypeDef):
-        (NameResolver.prototype.visitStructType):
-        (NameResolver.prototype.visitTypeRef):
-        (NameResolver.prototype.visitCallExpression):
-        (NameResolver.prototype.visitVectorType):
-        (NameResolver):
-        (NameResolver.prototype._visitTypeParametersAndBuildNameContext): Deleted.
-        (NameResolver.prototype.visitProtocolDecl): Deleted.
-        * WebGPUShadingLanguageRI/NativeFunc.js:
-        (NativeFunc):
-        * WebGPUShadingLanguageRI/NativeFuncInstance.js: Deleted.
-        * WebGPUShadingLanguageRI/NativeType.js:
-        (NativeType):
-        (NativeType.prototype.get name):
-        (NativeType.prototype.toString):
-        (NativeType.prototype.get typeParameters): Deleted.
-        (NativeType.prototype.instantiate): Deleted.
-        * WebGPUShadingLanguageRI/Node.js:
-        (Node.prototype.commit):
-        (Node):
-        (Node.prototype.substitute): Deleted.
-        (Node.prototype.substituteToUnification): Deleted.
-        * WebGPUShadingLanguageRI/OperatorAnderIndex.js: Added, previously
-        implemented with generics.
-        * WebGPUShadingLanguageRI/OperatorArrayRefLength.js: Ditto.
-        * WebGPUShadingLanguageRI/OperatorBool.js: Ditto.
-        * WebGPUShadingLanguageRI/Parse.js: Removed everything related to
-        generics and protocols.
-        * WebGPUShadingLanguageRI/Prepare.js: Added new stages.
-        * WebGPUShadingLanguageRI/Program.js:
-        (Program):
-        (Program.prototype.get types):
-        (Program.prototype.add):
-        (Program.prototype.get protocols): Deleted.
-        * WebGPUShadingLanguageRI/ProtocolDecl.js: Removed.
-        * WebGPUShadingLanguageRI/ProtocolRef.js: Removed.
-        * WebGPUShadingLanguageRI/RecursiveTypeChecker.js:
-        * WebGPUShadingLanguageRI/ResolveCallExpressionsWithTypeArguments.js: Added.
-        * WebGPUShadingLanguageRI/ResolveNames.js:
-        * WebGPUShadingLanguageRI/ResolveOverloadImpl.js:
-        * WebGPUShadingLanguageRI/ResolveTypeDefs.js:
-        (resolveTypeDefsInProtocols): Deleted.
-        * WebGPUShadingLanguageRI/Rewriter.js: Changes for new classes.
-        * WebGPUShadingLanguageRI/SPIRV.html:
-        * WebGPUShadingLanguageRI/SPIRVCodegen.js:
-        * WebGPUShadingLanguageRI/StandardLibrary.js: Replace old generic features.
-        * WebGPUShadingLanguageRI/StatementCloner.js: Changes for new classes.
-        (StructLayoutBuilder):
-        * WebGPUShadingLanguageRI/StructType.js:
-        * WebGPUShadingLanguageRI/SwizzleOp.js:
-        (SwizzleOp):
-        (SwizzleOp.prototype.get baseTypeName):
-        (SwizzleOp.prototype.toString):
-        (SwizzleOp.functions.):
-        (SwizzleOp.functions):
-        (SwizzleOp.prototype.instantiateImplementation):
-        (SwizzleOp.allSwizzleOperators.): Deleted.
-        (SwizzleOp.allSwizzleOperators): Deleted.
-        * WebGPUShadingLanguageRI/SynthesizeArrayOperatorLength.js: Added,
-        previously implemented with generics.
-        * WebGPUShadingLanguageRI/SynthesizeCopyConstructorOperator.js: Ditto.
-        * WebGPUShadingLanguageRI/SynthesizeDefaultConstructorOperator.js: Ditto.
-        * WebGPUShadingLanguageRI/SynthesizeEnumFunctions.js: Ditto.
-        * WebGPUShadingLanguageRI/SynthesizeOperatorBool.js: Ditto.
-        * WebGPUShadingLanguageRI/SynthesizeStructAccessors.js:
-        (synthesizeStructAccessors.setupImplementationData):
-        (synthesizeStructAccessors.createFieldType):
-        (synthesizeStructAccessors.createTypeRef):
-        (synthesizeStructAccessors.setupAnder):
-        (synthesizeStructAccessors):
-        (synthesizeStructAccessors.createTypeParameters): Deleted.
-        (synthesizeStructAccessors.): Deleted.
-        * WebGPUShadingLanguageRI/Test.html:
-        * WebGPUShadingLanguageRI/Test.js: Removed redundant tests and added new
-        tests for changed language features.
-        * WebGPUShadingLanguageRI/Type.js:
-        * WebGPUShadingLanguageRI/TypeDef.js:
-        * WebGPUShadingLanguageRI/TypeDefResolver.js:
-        * WebGPUShadingLanguageRI/TypeParameterRewriter.js: Removed.
-        * WebGPUShadingLanguageRI/TypeRef.js:
-        * WebGPUShadingLanguageRI/TypeVariable.js: Removed.
-        * WebGPUShadingLanguageRI/UnificationContext.js:
-        * WebGPUShadingLanguageRI/VectorType.js: Added.
-        * WebGPUShadingLanguageRI/Visitor.js: Changes for new classes.
-        * WebGPUShadingLanguageRI/WSL.md: Removed description of
-        generics/protocols.
-        * WebGPUShadingLanguageRI/WrapChecker.js:
-        (WrapChecker):
-        (WrapChecker.visitConstexprTypeParameter): Deleted.
-        (WrapChecker.prototype.visitFuncParameter): Deleted.
-        (WrapChecker.prototype.visitTypeVariable): Deleted.
-        * WebGPUShadingLanguageRI/index.html:
-
-2018-08-20  Thomas Denney  <tdenney@apple.com>
-
-        Added Thomas Denney to contributors.json.
-        https://bugs.webkit.org/show_bug.cgi?id=188525
-
-        Reviewed by Myles C. Maxfield.
-
-        * Scripts/webkitpy/common/config/contributors.json:
-
-2018-08-20  Bernhard M. Wiedemann  <bwiedemann@suse.de>
-
-        [GTK] Sort inspector GResource manifest to ensure reproducible builds
-        https://bugs.webkit.org/show_bug.cgi?id=188738
-
-        Reviewed by Michael Catanzaro.
-
-        * glib/generate-inspector-gresource-manifest.py:
-        (get_filenames): sort list of input files
-
-2018-08-20  Wenson Hsieh  <wenson_hsieh@apple.com>
-
-        [iOS] Paste is missing from callout bar when pasteboard only contains custom data
-        https://bugs.webkit.org/show_bug.cgi?id=184271
-        <rdar://problem/39256708>
-
-        Reviewed by Andy Estes.
-
-        This API test is failing when run against older macOS builds, where custom pasteboard
-        data is not enabled by default.
-
-        * TestWebKitAPI/Tests/WebKitCocoa/PasteMixedContent.mm:
-        (TestWebKitAPI::TEST):
-
-2018-08-20  Claudio Saavedra  <csaavedra@igalia.com>
-
-        Add WPE Debug configuration to the flakiness dashboard
-        https://bugs.webkit.org/show_bug.cgi?id=188743
-
-        Reviewed by Michael Catanzaro.
-
-        * TestResultServer/static-dashboards/builders.jsonp:
-        * TestResultServer/static-dashboards/flakiness_dashboard_unittests.js:
-
-2018-08-19  Carlos Garcia Campos  <cgarcia@igalia.com>
-
-        [GLIB] Add API to throw exceptions using printf formatted strings
-        https://bugs.webkit.org/show_bug.cgi?id=188698
-
-        Reviewed by Michael Catanzaro.
-
-        Add cases to test the new API.
-
-        * TestWebKitAPI/Tests/JavaScriptCore/glib/TestJSC.cpp:
-        (createFormattedError):
-        (createCustomFormattedError):
-        (testJSCExceptions):
-
-2018-08-19  Carlos Garcia Campos  <cgarcia@igalia.com>
-
-        [GLIB] Complete the JSCException API
-        https://bugs.webkit.org/show_bug.cgi?id=188695
-
-        Reviewed by Michael Catanzaro.
-
-        Add cases to test the new API.
-
-        * TestWebKitAPI/Tests/JavaScriptCore/glib/TestJSC.cpp:
-        (testJSCCheckSyntax):
-        (createCustomError):
-        (testJSCExceptions):
-
-2018-08-18  Wenson Hsieh  <wenson_hsieh@apple.com>
-
-        [iOS] Paste is missing from callout bar when pasteboard only contains custom data
-        https://bugs.webkit.org/show_bug.cgi?id=184271
-        <rdar://problem/39256708>
-
-        Reviewed by Ryosuke Niwa.
-
-        Add a new API test to verify that on both iOS and macOS, copied custom pasteboard data can only be pasted in a
-        matching origin. Additionally verify that on iOS, the web view is capable of performing the "paste:" selector.
-
-        * TestWebKitAPI/Tests/WebKitCocoa/PasteMixedContent.mm:
-        (TestWebKitAPI::TEST):
-        (imagePath): Deleted.
-        (writeTypesAndDataToPasteboard): Deleted.
-        (setUpWebView): Deleted.
-        (markupString): Deleted.
-
-        Make this test file no longer exclusive to macOS.
-
-        * TestWebKitAPI/cocoa/TestWKWebView.h:
-        * TestWebKitAPI/cocoa/TestWKWebView.mm:
-        (-[TestWKWebView synchronouslyLoadHTMLString:baseURL:]):
-
-        Tweak this helper to also take in a `baseURL`. Defaults to using the TestWebKitAPI bundle resource URL.
-
-        (-[TestWKWebView synchronouslyLoadHTMLString:]):
-
-2018-08-17  Jonathan Bedard  <jbedard@apple.com>
-
-        Add back --wtf-only to run-api-tests
-        https://bugs.webkit.org/show_bug.cgi?id=187893
-        <rdar://problem/42483983>
-
-        Reviewed by Aakash Jain.
-
-        When doing WTF development, it is not necessary to build or run all of the API
-        tests. Generally, if a user has specified a specific binary (or binaries) that
-        they are interested in testing, it is not necessary to check all API test binaries.
-
-        * Scripts/webkitpy/api_tests/manager.py:
-        (Manager._collect_tests): Only use the binaries matching the program arguments
-        when collecting tests.
-        (Manager._binaries_for_arguments): Generate a list of binaries which match the
-        program arguments.
-        (Manager.run): Pass a list binaries to check.
-        * Scripts/webkitpy/api_tests/run_api_tests.py:
-        (parse_args):
-        * Scripts/webkitpy/port/base.py:
-        (Port.check_api_test_build): If the caller specifies which API test binaries it
-        requires, only check the ones specified.
-        (Port.path_to_api_test_binaries): Allow the caller to only build the WTF API tests.
-        (Port._build_api_tests): Allow the caller to only build the WTF API tests.
-        * Scripts/webkitpy/port/win.py:
-        (WinPort.path_to_api_test_binaries):
-
-2018-08-17  Jer Noble  <jer.noble@apple.com>
-
-        REGRESSION (234743) Timeouts in TestWebKitAPI.PreferredAudioBufferSize.AudioWithWebAudio and TestWebKitAPI.PreferredAudioBufferSize.WebAudio
-        https://bugs.webkit.org/show_bug.cgi?id=188470
-        <rdar://problem/43144969>
-
-        Reviewed by Alex Christensen.
-
-        On a loaded server, the WebContent process may send the "playing" message before the
-        UIProcess has a chance to register a listener for that message. Restructure the tests
-        so that the listener is registered before the page is loaded.
-
-        * TestWebKitAPI/Tests/WebKitCocoa/PreferredAudioBufferSize.mm:
-        (PreferredAudioBufferSize::runPlayingTestWithPageNamed):
-        (TEST_F):
-
-2018-08-16  Wenson Hsieh  <wenson_hsieh@apple.com>
-
-        [macOS] [WK2] Add infrastructure to test receiving file promises on drop
-        https://bugs.webkit.org/show_bug.cgi?id=188583
-
-        Reviewed by Andy Estes.
-
-        Enable testing file promise drop handling on the WebKit2 port of macOS by introducing a subclass of
-        NSFilePromiseReceiver and implementing `-enumerateDraggingItemsWithOptions:forView:…` on TestDraggingInfo (the
-        concrete NSDraggingInfo implementation used by DragAndDropSimulator) using the mock file receiver. Also
-        introduces 3 new macOS tests. See below for more details.
-
-        A large portion of this logic is ported over from DumpRenderTree testing infrastructure added in r229297. In a
-        future patch, we should introduce a way to write code common to both API tests, WebKitTestRunner and
-        DumpRenderTree, and make this code shared among all three testing harnesses.
-
-        Tests:  DragAndDropTests.DragImageElementIntoFileUpload
-                DragAndDropTests.DragPromisedImageFileIntoFileUpload
-                DragAndDropTests.DragImageFileIntoFileUpload
-
-        * TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj:
-        * TestWebKitAPI/Tests/WebKitCocoa/DragAndDropTests.mm:
-        (TEST):
-
-        Drive-by fix: Replace NSMakePoint with CGPointMake.
-
-        * TestWebKitAPI/Tests/WebKitCocoa/image-and-file-upload.html:
-
-        Add a new test page consisting of an image and a "file upload" area that updates exercises DataTransfer API to
-        load the dropped image.
-
-        * TestWebKitAPI/Tests/mac/DragAndDropTestsMac.mm:
-        (waitForConditionWithLogging):
-
-        Add a helper function to wait for a condition to evaluate to true. To make failures more informative,
-        additionally add a mechanism to log a warning message after a given timeout, if the condition has yet to be met.
-
-        (TEST):
-
-        Add new tests to exercise a few cases of file upload in WebKit, including: (1) dragging from an image element,
-        (2) dragging files written to the pasteboard as promises, and (3) files written to the pasteboard as file URLs.
-
-        * TestWebKitAPI/cocoa/DragAndDropSimulator.h:
-
-        Surface the new functionality provided by the file promise receiver mock via two new methods on
-        DragAndDropSimulator to (1) write files as file promises to the pasteboard, and (2) write files as file paths
-        directly to the pasteboard.
-
-        * TestWebKitAPI/mac/DragAndDropSimulatorMac.mm:
-        (-[DragAndDropSimulator performDragInWebView:atLocation:withImage:pasteboard:source:]):
-        (-[DragAndDropSimulator externalPromisedFiles]):
-        (getFilePathsAndTypeIdentifiers):
-        (-[DragAndDropSimulator writePromisedFiles:]):
-        (-[DragAndDropSimulator writeFiles:]):
-        * TestWebKitAPI/mac/TestDraggingInfo.h:
-        * TestWebKitAPI/mac/TestDraggingInfo.mm:
-        (-[TestDraggingInfo initWithDragAndDropSimulator:]):
-        (-[TestDraggingInfo filePromiseReceivers]):
-        (-[TestDraggingInfo enumerateDraggingItemsWithOptions:forView:classes:searchOptions:usingBlock:]):
-
-        Implement this by invoking the given block with a TestFilePromiseReceiver. While all other classes are
-        unhandled by this testing code, WebKit only calls into this with [NSFilePromiseReceiver class], this is
-        currently sufficient for testing purposes.
-
-        * TestWebKitAPI/mac/TestFilePromiseReceiver.h: Copied from Tools/TestWebKitAPI/mac/TestDraggingInfo.h.
-        * TestWebKitAPI/mac/TestFilePromiseReceiver.mm: Added.
-        (-[TestFilePromiseReceiver initWithPromisedTypeIdentifiers:dragAndDropSimulator:]):
-        (-[TestFilePromiseReceiver fileTypes]):
-        (-[TestFilePromiseReceiver fileNames]):
-        (-[TestFilePromiseReceiver dealloc]):
-        (-[TestFilePromiseReceiver draggingSource]):
-        (-[TestFilePromiseReceiver setDraggingSource:]):
-        (fileNameWithNumericSuffix):
-        (copyFile):
-
-        Add a helper to copy a file into a destination directory. Used to implement the main functionality of our
-        NSFilePromiseReceiver subclass, which copies the files specified by test code into the directory determined by
-        WebKit.
-
-        (-[TestFilePromiseReceiver receivePromisedFilesAtDestination:options:operationQueue:reader:]):
-
-2018-08-16  Alex Christensen  <achristensen@webkit.org>
-
-        Add entitlement to MiniBrowser to allow it to communicate with com.apple.Safari.SafeBrowsing.Service
-        https://bugs.webkit.org/show_bug.cgi?id=188677
-
-        Reviewed by Tim Horton.
-
-        * MiniBrowser/MiniBrowser.entitlements:
-        SSBLookupContext lookUpURL:completionHandler: fails to communicate with its service without this entitlement
-        on released operating systems.  See <rdar://problem/42749777>
-
-2018-08-16  Jer Noble  <jer.noble@apple.com>
-
-        Add option to run-api-tests to force running of DISABLED tests.
-        https://bugs.webkit.org/show_bug.cgi?id=188674
-
-        Reviewed by Joseph Pecoraro.
-
-        * Scripts/webkitpy/api_tests/run_api_tests.py:
-        (parse_args):
-        * Scripts/webkitpy/api_tests/runner.py:
-        (Runner.command_for_port):
-        (_Worker._run_single_test):
-
-2018-08-16  Aakash Jain  <aakash_jain@apple.com>
-
-        [ews-build] Separate queues for Builders and Testers - macOS High Sierra
-        https://bugs.webkit.org/show_bug.cgi?id=188540
-
-        Reviewed by Lucas Forschler.
-
-        * BuildSlaveSupport/ews-build/config.json: Splitted few macOS queues into builders and testers. 
-
-2018-08-16  Aakash Jain  <aakash_jain@apple.com>
-
-        [ews-build] Add build steps ArchiveTestResults, UploadTestResults and ExtractTestResults
-        https://bugs.webkit.org/show_bug.cgi?id=188666
-
-        Reviewed by Lucas Forschler.
-
-        * BuildSlaveSupport/ews-build/steps.py:
-        (ArchiveTestResults): Build step to Archive the test results.
-        (UploadTestResults): Build step to upload the archive.
-        (ExtractTestResults): Build step to unzip the archive on server and generate a link.
-        * BuildSlaveSupport/ews-build/steps_unittest.py: Added unit-tests.
-        (ExpectMasterShellCommand): Copied from other similar internal code.
-
-2018-08-16  Aakash Jain  <aakash_jain@apple.com>
-
-        [ews-build] Separate queues for Builders and Testers - macOS Sierra
-        https://bugs.webkit.org/show_bug.cgi?id=188516
-
-        Reviewed by Lucas Forschler.
-
-        * BuildSlaveSupport/ews-build/config.json: Splitted macOS Sierra Release queues into builders and testers.
-        * BuildSlaveSupport/ews-build/factories.py: Used appropriate base classes for builders and testers.
-
-2018-08-16  Basuke Suzuki  <Basuke.Suzuki@sony.com>
-
-        [Win][DumpRenderTree] Remove error message for duplicated tests.
-        https://bugs.webkit.org/show_bug.cgi?id=188662
-
-        Reviewed by Per Arne Vollan.
-
-        This was added to track down an issue where the same tests appeared
-        to be run multiple times. This turned out to not be the case.
-        See https://bugs.webkit.org/show_bug.cgi?id=166760 .
-
-        * DumpRenderTree/win/DumpRenderTree.cpp:
-        (runTest):
-
-2018-08-16  Alex Christensen  <achristensen@webkit.org>
-
-        Re-introduce assertion removed in r234890
-        https://bugs.webkit.org/show_bug.cgi?id=188611
-
-        Reviewed by Geoffrey Garen.
-
-        * TestWebKitAPI/Tests/WTF/Vector.cpp:
-        (TestWebKitAPI::TEST):
-
-2018-08-16  Jer Noble  <jer.noble@apple.com>
-
-        Add Experimental Feature support for SourceBuffer.changeType()
-        https://bugs.webkit.org/show_bug.cgi?id=188626
-        <rdar://problem/43356021>
-
-        Reviewed by Eric Carlson.
-
-        * DumpRenderTree/mac/DumpRenderTree.mm:
-        (enableExperimentalFeatures):
-        (resetWebPreferencesToConsistentValues):
-        * WebKitTestRunner/TestController.cpp:
-        (WTR::TestController::resetPreferencesToConsistentValues):
-
-2018-08-16  Aditya Keerthi  <akeerthi@apple.com>
-
-        Support drag-and-drop for input[type=color]
-        https://bugs.webkit.org/show_bug.cgi?id=188464
-
-        Reviewed by Wenson Hsieh.
-
-        Added five cross-platform API tests, to verify that dragging and dropping with
-        color inputs changes the value of the drop target if and only if both inputs are
-        enabled. Also tests that the change and input events are fired when changing the
-        value of a color input through drag and drop. 
-
-        Additionally, added a macOS-specific test to verify that dropping an item with
-        NSColorPboardType changes the value of the color input.
-
-        * TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj:
-        * TestWebKitAPI/Tests/WebKitCocoa/DragAndDropTests.mm:
-        (TestWebKitAPI::TEST):
-        * TestWebKitAPI/Tests/WebKitCocoa/color-drop.html: Added.
-        * TestWebKitAPI/Tests/mac/DragAndDropTestsMac.mm:
-        (TestWebKitAPI::TEST):
-
-2018-08-16  Antti Koivisto  <antti@apple.com>
-
-        Use OptionSet for ActivityState::Flags
-        https://bugs.webkit.org/show_bug.cgi?id=188554
-
-        Reviewed by Brent Fulgham.
-
-        * TestWebKitAPI/Tests/WTF/OptionSet.cpp:
-        (TestWebKitAPI::TEST):
-
-2018-08-16  Carlos Garcia Campos  <cgarcia@igalia.com>
-
-        [GTK] MiniBrowser: web view doesn't get the focus when new window is created
-        https://bugs.webkit.org/show_bug.cgi?id=188546
-
-        Reviewed by Michael Catanzaro.
-
-        When a new window is created, the first widget of the windo is getting the focus, the toolbar in this case. We
-        always want the WebView to grab the focus. In case of new empty tab, we want to the URL bar to be focused
-        instead. This was causing failures in WebDriver tests that create a popup window with an input element that is
-        filled with send keys command. The input element is marked as focused in the DOM, but the actual focused widget
-        is the toolbar, so key events are sent to the toolbar instead of the web view.
-
-        * MiniBrowser/gtk/BrowserWindow.c:
-        (webViewCreate):
-        (newTabCallback):
-        (openPrivateWindow):
-        (browser_window_get_or_create_web_view_for_automation):
-
-2018-08-16  Carlos Garcia Campos  <cgarcia@igalia.com>
-
-        Unreviewed. Fix WebDriver tests after r234839.
-
-        New pytest requires to autoinstall attrs too.
-
-        * Scripts/webkitpy/thirdparty/__init__.py:
-        (AutoinstallImportHook._install_pytest):
-
-2018-08-15  Ansh Shukla  <ansh_shukla@apple.com>
-
-        NSURLAuthenticationMethodOAuth challenges are surfaced to clients in -didReceiveAuthenticationChallenge as NSURLAuthenticationMethodDefault
-        https://bugs.webkit.org/show_bug.cgi?id=186870
-        <rdar://problem/41314410>
-
-        Reviewed by Alex Christensen.
-
-        * WebKitTestRunner/TestController.cpp:
-        (WTR::toString):
-        (WTR::TestController::canAuthenticateAgainstProtectionSpace): Expose type of authentication challenge so we can test OAuth.
-        (WTR::TestController::didReceiveAuthenticationChallenge):
-
-2018-08-15  Ben Richards  <benton_richards@apple.com>
-
-        We should cache the compiled sandbox profile in a data vault
-        https://bugs.webkit.org/show_bug.cgi?id=184991
-
-        Reviewed by Ryosuke Niwa.
-
-        Added trace points for sandbox initialization
-
-        * Tracing/SystemTracePoints.plist:
-
-2018-08-15  Ryan Haddad  <ryanhaddad@apple.com>
-
-        Unreviewed, rolling out r234870.
-
-        The test introduced with this change is a flaky failure.
-
-        Reverted changeset:
-
-        "NSURLAuthenticationMethodOAuth challenges are surfaced to
-        clients in -didReceiveAuthenticationChallenge as
-        NSURLAuthenticationMethodDefault"
-        https://bugs.webkit.org/show_bug.cgi?id=186870
-        https://trac.webkit.org/changeset/234870
-
-2018-08-15  Jonathan Bedard  <jbedard@apple.com>
-
-        WebKitTestRunner should support watch devices
-        https://bugs.webkit.org/show_bug.cgi?id=188570
-
-        Reviewed by Alex Christensen.
-
-        * WebKitTestRunner/Configurations/WebKitTestRunnerApp.xcconfig:
-
-2018-08-15  Wenson Hsieh  <wenson_hsieh@apple.com>
-
-        [Attachment SPI] Remove attachment display mode options
-        https://bugs.webkit.org/show_bug.cgi?id=188596
-
-        Reviewed by Dan Bernstein.
-
-        Remove API tests and API test helpers for verifying the behavior of in-place media attachment elements.
-
-        * TestWebKitAPI/Tests/WebKitCocoa/WKAttachmentTests.mm:
-        (-[TestWKWebView synchronouslyInsertAttachmentWithFilename:contentType:data:]):
-        (TestWebKitAPI::TEST):
-        (testVideoData): Deleted.
-        (displayOptionsWithMode): Deleted.
-        (-[TestWKWebView synchronouslyInsertAttachmentWithFilename:contentType:data:options:]): Deleted.
-
-2018-08-14  Devin Rousso  <drousso@apple.com>
-
-        Unreviewed, change my emails.
-
-        * Scripts/webkitpy/common/config/contributors.json:
-        * Scripts/webkitpy/common/config/watchlist:
-
-2018-08-14  Saam barati  <sbarati@apple.com>
-
-        HashMap<Ref<P>, V> asserts when V is not zero for its empty value
-        https://bugs.webkit.org/show_bug.cgi?id=188582
-
-        Reviewed by Sam Weinig.
-
-        * TestWebKitAPI/Tests/WTF/HashMap.cpp:
-        (TestWebKitAPI::TEST):
-
-2018-08-14  Zalan Bujtas  <zalan@apple.com>
-
-        [LFC][Floating] Add support for negative clearance.
-        https://bugs.webkit.org/show_bug.cgi?id=188555
-
-        Reviewed by Simon Fraser.
-
-        * LayoutReloaded/misc/LFC-passing-tests.txt:
-
-2018-08-14 Sihui Liu <sihui_liu@apple.com>
-
-        Crash in WebKit::filterPreloadHSTSEntry via NetworkProcess::getHostNamesWithHSTSCache
-        https://bugs.webkit.org/show_bug.cgi?id=188576
-        <rdar://problem/43148977>
-
-        Reviewed by Alex Christensen.
-
-        * TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj:
-        * TestWebKitAPI/Tests/WebKitCocoa/WKWebsiteDatastore.mm: Added.
-        (TEST):
-
-2018-08-14  Alex Christensen  <achristensen@webkit.org>
-
-        isValidCSSSelector is unsafe to be called from a non-main thread
-        https://bugs.webkit.org/show_bug.cgi?id=188581
-        <rdar://problem/40517358>
-
-        Reviewed by Sam Weinig.
-
-        * TestWebKitAPI/Tests/WebCore/ContentExtensions.cpp:
-        (TestWebKitAPI::InMemoryCompiledContentExtension::create):
-        (TestWebKitAPI::checkCompilerError):
-
-2018-08-14  Ansh Shukla  <ansh_shukla@apple.com>
-
-        NSURLAuthenticationMethodOAuth challenges are surfaced to clients in -didReceiveAuthenticationChallenge as NSURLAuthenticationMethodDefault
-        https://bugs.webkit.org/show_bug.cgi?id=186870
-        <rdar://problem/41314410>
-
-        Reviewed by Alex Christensen.
-
-        * WebKitTestRunner/TestController.cpp:
-        (WTR::toString):
-        (WTR::TestController::canAuthenticateAgainstProtectionSpace): Expose type of authentication challenge so we can test OAuth.
-        (WTR::TestController::didReceiveAuthenticationChallenge):
-
-2018-08-14  Fujii Hironori  <Hironori.Fujii@sony.com>
-
-        [webkitpy][Win] LayoutTests: test names should be Unix style, separated by slash not backslash
-        https://bugs.webkit.org/show_bug.cgi?id=187973
-
-        Reviewed by Alex Christensen.
-
-        In LayoutTests, test names have been canonicalized in Unix style
-        since Bug 63597, for example 'fast/css/001.html'. But Bug 179219,
-        Bug 179572, Bug 180660, and Bug 181814 have changed to use
-        os.path.seq instead of slash if Windows Python is used.
-
-        Revert parts of those changes. Change relative_test_filename to
-        return a slash-separated test name as chromium_win.py used to do.
-
-        This change fixes all 41 test-webkitpy failures in WinCairo port.
-
-        * Scripts/webkitpy/layout_tests/models/test_expectations.py:
-        (TestExpectationParser._parse_line): Do not convert test names with normpath.
-        * Scripts/webkitpy/port/base.py:
-        (Port.normalize_test_name): Use TEST_PATH_SEPARATOR instead of os.path.sep.
-        (Port.relative_test_filename): Replace self.host.filesystem.sep with self.TEST_PATH_SEPARATOR.
-        (Port.abspath_for_test): Replace self.TEST_PATH_SEPARATOR with self.host.filesystem.sep.
-        * Scripts/webkitpy/port/driver.py:
-        (Driver): Use '/' instead of os.sep.
-        * Scripts/webkitpy/port/win.py:
-        (WinCairoPort): Do not override TEST_PATH_SEPARATOR.
-
-2018-08-14  Aakash Jain  <aakash_jain@apple.com>
-
-        [ews-build] Add build step to run WK1 layout-test
-        https://bugs.webkit.org/show_bug.cgi?id=188498
-
-        Reviewed by Lucas Forschler.
-
-        * BuildSlaveSupport/ews-build/steps.py:
-        (RunWebKit1Tests): Class to run WebKit1Tests.
-        * BuildSlaveSupport/ews-build/steps_unittest.py: Added unit-tests.
-
-2018-08-14  Patrick Griffis  <pgriffis@igalia.com>
-
-        [Flatpak] Fix --cmakeargs
-        https://bugs.webkit.org/show_bug.cgi?id=188567
-
-        Reviewed by Michael Catanzaro.
-
-        * flatpak/flatpakutils.py:
-        (WebkitFlatpak.setup_dev_env):
-
-2018-08-14  Darshan Kadu  <dkadu@igalia.com>
-
-        Enhancement request: Make export-w3c-test-changes add the PR to "See Also" links
-        https://bugs.webkit.org/show_bug.cgi?id=186140
-
-        Reviewed by Youenn Fablet.
-
-        Added an optional parameter see_also to post_comment_to_bug function and used it for adding see_also variable's content in "See Also"
-        * Scripts/webkitpy/common/net/bugzilla/bugzilla.py:
-        (Bugzilla.post_comment_to_bug):
-        Modified mock for see_also parameter
-        * Scripts/webkitpy/common/net/bugzilla/bugzilla_mock.py:
-        (MockBugzilla.post_comment_to_bug):
-        Made following unit tests to have see_also in bug comment
-        * Scripts/webkitpy/tool/bot/flakytestreporter_unittest.py:
-        * Scripts/webkitpy/tool/bot/sheriff_unittest.py:
-        (SheriffTest.test_post_blame_comment_on_bug):
-        * Scripts/webkitpy/tool/commands/applywatchlistlocal_unittest.py:
-        * Scripts/webkitpy/tool/commands/queues_unittest.py:
-        * Scripts/webkitpy/tool/commands/upload_unittest.py:
-        * Scripts/webkitpy/tool/steps/applywatchlist_unittest.py:
-        * Scripts/webkitpy/w3c/test_exporter.py:
-        (WebPlatformTestExporter.make_pull_request):
-        * Scripts/webkitpy/w3c/test_exporter_unittest.py:
-        (TestExporterTest.MockBugzilla.post_comment_to_bug):
-        (TestExporterTest.test_export):
-
-2018-08-14  Aakash Jain  <aakash_jain@apple.com>
-
-        [ews-build] Add support for max_builds parameter for workers
-        https://bugs.webkit.org/show_bug.cgi?id=188531
-
-        Reviewed by Lucas Forschler.
-
-        * BuildSlaveSupport/ews-build/loadConfig.py:
-        (loadBuilderConfig): Added support for max_builds.
-
-2018-08-14  Zalan Bujtas  <zalan@apple.com>
-
-        [LFC][Floating] Adjust vertical position with non-collapsing previous sibling margin.
-        https://bugs.webkit.org/show_bug.cgi?id=188543
-
-        Reviewed by Antti Koivisto.
-
-        * LayoutReloaded/misc/LFC-passing-tests.txt:
-
-2018-08-14  Carlos Garcia Campos  <cgarcia@igalia.com>
-
-        Unreviewed. Fix WebDriver tests after r234839.
-
-        New pytest requires to autoinstall more_itertools and six too.
-
-        * Scripts/webkitpy/thirdparty/__init__.py:
-        (AutoinstallImportHook._install_pytest):
-
-2018-08-14  Tomas Popela  <tpopela@redhat.com>
-
-        [GTK] Minibrowser: Add labels for buttons
-        https://bugs.webkit.org/show_bug.cgi?id=188549
-
-        Reviewed by Carlos Garcia Campos.
-
-        So they are accessible easier (i.e. while testing WebKitGTK+ through Minibrowser in Dogtail).
-
-        * MiniBrowser/gtk/BrowserWindow.c:
-        (webViewIsLoadingChanged):
-        (browserWindowSetupEditorToolbar):
-        (browser_window_init):
-
-2018-08-14  Carlos Garcia Campos  <cgarcia@igalia.com>
-
-        Unreviewed. Fix WebDriver tests after r234839.
-
-        New pytest requires to autoinstall atomicwrites too.
-
-        * Scripts/webkitpy/thirdparty/__init__.py:
-        (AutoinstallImportHook._install_pytest):
-
-2018-08-14  Carlos Garcia Campos  <cgarcia@igalia.com>
-
-        Unreviewed. Fix WebDriver tests after r234839.
-
-        New pytest requires to autoinstall pluggy and funcsigs too.
-
-        * Scripts/webkitpy/thirdparty/__init__.py:
-        (AutoinstallImportHook._install_pytest):
-
-2018-08-13  Carlos Garcia Campos  <cgarcia@igalia.com>
-
-        Unreviewed. Update W3C WebDriver imported tests.
-
-        Bump pytest version to 3.6.2.
-
-        * Scripts/webkitpy/thirdparty/__init__.py:
-        (AutoinstallImportHook._install_pytest):
-
-2018-08-13  Thomas Denney  <tdenney@apple.com>
-
-        Allow the substring 'me' in contributor names and email addresses
-        https://bugs.webkit.org/show_bug.cgi?id=188538
-
-        Reviewed by Tim Horton.
-
-        A test didn't permit the string "me" in email addresses or names of contributors.
-
-        * Scripts/webkitpy/common/config/committers_unittest.py:
-        (CommittersTest.test_contributors_by_fuzzy_match):
-
-2018-08-13  Commit Queue  <commit-queue@webkit.org>
-
-        Unreviewed, rolling out r234812.
-        https://bugs.webkit.org/show_bug.cgi?id=188536
-
-        broke the build (Requested by ThomasDenney on #webkit).
-
-        Reverted changeset:
-
-        "Added Thomas Denney to contributors.json."
-        https://bugs.webkit.org/show_bug.cgi?id=188525
-        https://trac.webkit.org/changeset/234812
-
-2018-08-13  Sihui Liu  <sihui_liu@apple.com>
-
-        Avoid timeout resulted from calling waitUntilDone when test is not running
-        https://bugs.webkit.org/show_bug.cgi?id=188389
-
-        Reviewed by Chris Dumez.
-
-        If test is not running, we should not set the waitUntilDone flag, or it may cause subsequent tests timeout.
-
-        * WebKitTestRunner/InjectedBundle/TestRunner.cpp:
-        (WTR::TestRunner::waitUntilDone):
-
-2018-08-13  Wenson Hsieh  <wenson_hsieh@apple.com>
-
-        [WK2] [macOS] Implement a mechanism to test drag and drop
-        https://bugs.webkit.org/show_bug.cgi?id=181898
-        <rdar://problem/39181698>
-
-        Reviewed by Simon Fraser.
-
-        Implements the currently stubbed DragAndDropSimulator on macOS, and introduces a new API test for r227266. See
-        comments below for more detail.
-
-        * TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj:
-        * TestWebKitAPI/Tests/WebKitCocoa/DragAndDropTests.mm: Copied from Tools/TestWebKitAPI/mac/DragAndDropSimulatorMac.mm.
-
-        Introduce a file for cross-platform drag and drop tests, currently for iOS and macOS. Additionally add a test
-        for r227266, which was fixed earlier this year but could not be tested due to a lack of testing mechanism on
-        macOS in WebKit2.
-
-        (TEST):
-        * TestWebKitAPI/Tests/WebKitCocoa/full-page-dropzone.html: Added.
-
-        Minor tweaks to this test page to add "dragover" and "drop" event handlers.
-
-        * TestWebKitAPI/Tests/WebKitCocoa/image-in-link-and-input.html:
-        * TestWebKitAPI/Tests/WebKitCocoa/link-in-iframe-and-input.html: Added.
-
-        Add a new test page that includes a link embedded within an iframe below a plain text input.
-
-        * TestWebKitAPI/Tests/mac/DragAndDropTestsMac.mm: Added.
-        (TEST):
-        * TestWebKitAPI/Tests/mac/LegacyDragAndDropTests.mm: Renamed from Tools/TestWebKitAPI/Tests/mac/DragAndDropPasteboardTests.mm.
-
-        Move only existing WebKit2 macOS drag and drop test (DragAndDropPasteboardTests.NumberOfValidItemsForDrop) out
-        of DragAndDropPasteboardTests.mm and into a new file, DragAndDropTestsMac.mm. Additionally, rename
-        DragAndDropPasteboardTests to LegacyDragAndDropTests, since it now only contains two legacy WebView tests for
-        drag and drop.
-
-        (+[FrameLoadCompletionListener listenerWithCompletionBlock:]):
-        (-[FrameLoadCompletionListener initWithCompletionBlock:]):
-        (-[FrameLoadCompletionListener webView:didFinishLoadForFrame:]):
-        (-[DragSource draggingSourceOperationMaskForLocal:]):
-        (-[DragInfo initWithImage:offset:pasteboard:source:destinationWindow:]):
-        (-[DragInfo lastMousePosition]):
-        (-[DragInfo setLastMousePosition:]):
-        (-[DragInfo draggingDestinationWindow]):
-        (-[DragInfo draggingSourceOperationMask]):
-        (-[DragInfo draggingLocation]):
-        (-[DragInfo draggedImageLocation]):
-        (-[DragInfo draggedImage]):
-        (-[DragInfo draggingPasteboard]):
-        (-[DragInfo draggingSource]):
-        (-[DragInfo draggingSequenceNumber]):
-        (-[DragInfo slideDraggedImageTo:]):
-        (-[DragInfo namesOfPromisedFilesDroppedAtDestination:]):
-        (-[DragInfo draggingFormation]):
-        (-[DragInfo setDraggingFormation:]):
-        (-[DragInfo animatesToDestination]):
-        (-[DragInfo setAnimatesToDestination:]):
-        (-[DragInfo numberOfValidItemsForDrop]):
-        (-[DragInfo setNumberOfValidItemsForDrop:]):
-        (-[DragInfo enumerateDraggingItemsWithOptions:forView:classes:searchOptions:usingBlock:]):
-        (-[DragInfo springLoadingHighlight]):
-        (-[DragInfo resetSpringLoading]):
-        (TestWebKitAPI::getTestImage):
-        (TestWebKitAPI::webViewAfterPerformingDragOperation):
-        (TestWebKitAPI::TEST):
-        * TestWebKitAPI/Tests/mac/full-page-dropzone.html: Removed.
-        * TestWebKitAPI/cocoa/DragAndDropSimulator.h:
-
-        Flesh out some of the DragAndDropSimulator API for macOS, exposing (among other things) the drag pasteboard,
-        the current NSDraggingInfo, the initial location of the drag image, and the drag image itself.
-
-        * TestWebKitAPI/cocoa/TestWKWebView.h:
-        * TestWebKitAPI/cocoa/TestWKWebView.mm:
-        (-[TestWKWebView mouseDownAtPoint:simulatePressure:]):
-        (-[TestWKWebView mouseUpAtPoint:]):
-        (-[TestWKWebView mouseMoveToPoint:withFlags:]):
-        (-[TestWKWebView sendClicksAtPoint:numberOfClicks:]):
-        (-[TestWKWebView mouseEnterAtPoint:]):
-        (-[TestWKWebView mouseExitAtPoint:]):
-        (-[TestWKWebView mouseDragToPoint:]):
-        (-[TestWKWebView _mouseEventWithType:atLocation:]):
-        (-[TestWKWebView _mouseEventWithType:atLocation:flags:timestamp:clickCount:]):
-
-        Add TestWKWebView helpers to send MouseMove, MouseEnter and MouseDrag NSEvents to the web view. Additionally,
-        rename parameter names to these helpers to make it more obvious that these locations are all in NSWindow
-        coordinates.
-
-        (-[TestWKWebView typeCharacter:]):
-
-        Drive-by style fix: put this opening brace on the beginning of the next line.
-
-        * TestWebKitAPI/ios/DragAndDropSimulatorIOS.mm:
-        (-[DragAndDropSimulator initWithWebViewFrame:]):
-        (-[DragAndDropSimulator initWithWebViewFrame:configuration:]):
-        (-[DragAndDropSimulator webView]):
-
-        Small iOS DragAndDropSimulator adjustments for new DragAndDropSimulator interfaces.
-
-        * TestWebKitAPI/mac/DragAndDropSimulatorMac.mm:
-        (-[DragAndDropTestWKWebView initWithFrame:configuration:simulator:]):
-
-        Introduce a WKWebView subclass for testing drag and drop that overrides `-dragImage:at:offset:…`, and instead
-        allows DragAndDropSimulator to take over the drag.
-
-        (-[DragAndDropTestWKWebView dragImage:at:offset:event:pasteboard:source:slideBack:]):
-
-        Override this entry point into drag and drop code, and instead call out to the DragAndDropSimulator to
-        coordinate the drag.
-
-        (-[DragAndDropTestWKWebView waitForPendingMouseEvents]):
-
-        Helper method to wait for the web process to finish handling all in-flight mouse events.
-
-        (defaultExternalDragImage):
-
-        Set this image as the default drag image when simulating an incoming drag session from outside of the web view.
-
-        (-[DragAndDropSimulator initWithWebViewFrame:]):
-        (-[DragAndDropSimulator initWithWebViewFrame:configuration:]):
-        (-[DragAndDropSimulator flipAboutXAxisInHostWindow:]):
-
-        Helper method to flip a given point about the X axis of the window.
-
-        (-[DragAndDropSimulator locationInViewForCurrentProgress]):
-
-        Map a progress value (between 0 and 1) to a drag location.
-
-        (-[DragAndDropSimulator initialProgressForMouseDrag]):
-
-        Determines the initial progress value when initiation a drag in web content. This is the initial progress
-        required to ensure that the first mouse drag event exceeds the drag distance hysteresis and causes any drag
-        (if applicable) to begin.
-
-        (-[DragAndDropSimulator runFrom:to:]):
-        (-[DragAndDropSimulator performDragInWebView:atLocation:withImage:pasteboard:source:]):
-
-        Helper to coordinate drag updates in both the cases where we're simulating a drag session entering from outside
-        of the web view, and in the case where we've initiated a drag from the web view itself.
-
-        (-[DragAndDropSimulator webView]):
-        (-[DragAndDropSimulator setExternalDragPasteboard:]):
-        (-[DragAndDropSimulator externalDragPasteboard]):
-
-        Just like its iOS counterpart (setExternalItemProviders:), setting an external drag pasteboard on macOS puts the
-        DragAndDropSimulator in a mode that simulates a drag coming in from outside the web view, using the given
-        pasteboard.
-
-        (-[DragAndDropSimulator setExternalDragImage:]):
-        (-[DragAndDropSimulator externalDragImage]):
-
-        May be optionally set when specifying an external drag pasteboard to specify the drag image used. If no external
-        drag image is specified, falls back to the default image returned by `defaultExternalDragImage()`.
-
-        (-[DragAndDropSimulator draggingInfo]):
-        (-[DragAndDropSimulator willEndDraggingHandler]):
-        (-[DragAndDropSimulator setWillEndDraggingHandler:]):
-
-        Hook to allow tests to run logic right before performing the drop (if the current drag operation is not none) or
-        ending the drag session without performing a drag operation.
-
-        (-[DragAndDropSimulator initWithWebView:]): Deleted.
-        (-[DragAndDropSimulator dealloc]): Deleted.
-        (-[DragAndDropSimulator phase]): Deleted.
-        * TestWebKitAPI/mac/TestDraggingInfo.h: Copied from Tools/TestWebKitAPI/mac/DragAndDropSimulatorMac.mm.
-        * TestWebKitAPI/mac/TestDraggingInfo.mm: Added.
-
-        Mock object conforming to NSDraggingInfo that is passed to WKWebView when invoking -draggingUpdated:,
-        -draggingEntered: and -draggingExited:.
-
-        (-[TestDraggingInfo draggingPasteboard]):
-        (-[TestDraggingInfo setDraggingPasteboard:]):
-        (-[TestDraggingInfo draggingSource]):
-        (-[TestDraggingInfo setDraggingSource:]):
-        (-[TestDraggingInfo enumerateDraggingItemsWithOptions:forView:classes:searchOptions:usingBlock:]):
-        (-[TestDraggingInfo draggingDestinationWindow]):
-        (-[TestDraggingInfo draggedImage]):
-        (-[TestDraggingInfo setDraggedImage:]):
-        (-[TestDraggingInfo slideDraggedImageTo:]):
-        (-[TestDraggingInfo namesOfPromisedFilesDroppedAtDestination:]):
-        (-[TestDraggingInfo resetSpringLoading]):
-
-        Empty method stubs, to be implemented in the future as needed.
-
-2018-08-13  Thomas Denney  <tdenney@apple.com>
-
-        Added Thomas Denney to contributors.json.
-        https://bugs.webkit.org/show_bug.cgi?id=188525
-
-        Reviewed by Myles C. Maxfield.
-
-        * Scripts/webkitpy/common/config/contributors.json:
-
-2018-08-13  Antti Koivisto  <antti@apple.com>
-
-        Meaning of OptionSet::contains is unclear when used with OptionSet argument
-        https://bugs.webkit.org/show_bug.cgi?id=188501
-
-        Reviewed by Anders Carlsson.
-
-        * TestWebKitAPI/Tests/WTF/OptionSet.cpp:
-        (TestWebKitAPI::TEST):
-
-2018-08-13  Commit Queue  <commit-queue@webkit.org>
-
-        Unreviewed, rolling out r234747.
-        https://bugs.webkit.org/show_bug.cgi?id=188524
-
-        plugin processes crash on launch (Requested by smfr on
-        #webkit).
-
-        Reverted changeset:
-
-        "We should cache the compiled sandbox profile in a data vault"
-        https://bugs.webkit.org/show_bug.cgi?id=184991
-        https://trac.webkit.org/changeset/234747
-
-2018-08-13  Wenson Hsieh  <wenson_hsieh@apple.com>
-
-        [iOS] Dragging a non-editable text selection into a plain text input inserts HTML markup
-        https://bugs.webkit.org/show_bug.cgi?id=188485
-        <rdar://problem/43168784>
-
-        Reviewed by Tim Horton.
-
-        Add a new API test to verify that dropping selected non-editable rich text into a textarea inserts text as
-        expected, rather than markup.
-
-        It's somewhat interesting to note that this particular use case isn't exercised by any existing tests; the
-        closest test is DragAndDropTests.ContentEditableToTextarea, which drags a rich text selection from a
-        contenteditable element to a text area. However, due to logic in `DragController::concludeEditDrag` that handles
-        drag and drop across editable content differently than drag and drop from non-editable to editable content, the
-        bug that is fixed here doesn't surface in that existing test.
-
-        * TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj:
-        * TestWebKitAPI/Tests/WebKitCocoa/selected-text-and-textarea.html: Added.
-        * TestWebKitAPI/Tests/ios/DragAndDropTestsIOS.mm:
-        (TestWebKitAPI::TEST):
-
-2018-08-13  Alex Christensen  <achristensen@webkit.org>
-
-        Use a 1-byte enum class for TextDirection
-        https://bugs.webkit.org/show_bug.cgi?id=188350
-
-        Reviewed by Simon Fraser.
-
-        * TestWebKitAPI/Tests/WebCore/ComplexTextController.cpp:
-        (TestWebKitAPI::TEST_F):
-        * TestWebKitAPI/Tests/WebCore/StringWithDirection.cpp:
-        (WebCore::operator<<):
-        (TestWebKitAPI::TEST):
-        * TestWebKitAPI/Tests/WebKitCocoa/CommandBackForward.mm:
-        (TEST_F):
-
-2018-08-13  Zalan Bujtas  <zalan@apple.com>
-
-        [LFC][Floating] Add basic clearance support
-        https://bugs.webkit.org/show_bug.cgi?id=188492
-
-        Reviewed by Antti Koivisto.
-
-        * LayoutReloaded/misc/LFC-passing-tests.txt:
-
-2018-08-12  Zalan Bujtas  <zalan@apple.com>
-
-        [LFC] Float prev/next sibling should prevent top/bottom margin collapsing with parent.
-        https://bugs.webkit.org/show_bug.cgi?id=188487
-
-        Reviewed by Antti Koivisto.
-
-        * LayoutReloaded/misc/LFC-passing-tests.txt:
-
-2018-08-11  Aakash Jain  <aakash_jain@apple.com>
-
-        [ews-build] Separate queues for Builders and Testers - iOS
-        https://bugs.webkit.org/show_bug.cgi?id=188396
-
-        Reviewed by Lucas Forschler.
-
-        * BuildSlaveSupport/ews-build/config.json: Splitted iOS queues into builders and testers.
-        Also added ews121 and ews122 to match with current EWS configuration.
-        * BuildSlaveSupport/ews-build/factories.py:
-        (TestsFactory): Base classes for various Test factories.
-        (TestsFactory.getProduct): Download and extract the archive.
-
-2018-08-10  Daniel Bates  <dabates@apple.com>
-
-        webkit-patch setup-git-clone should set Git core editor to commit-log-editor
-        https://bugs.webkit.org/show_bug.cgi?id=188473
-
-        Reviewed by Ryosuke Niwa.
-
-        Have "webkit-patch setup-git-clone" set the Git core editor to "perl Tools/Scripts/commit-log-editor --regenerate-log"
-        so that the Git commit message is generated from the ChangeLog for developers that manually
-        generate their ChangeLog using prepare-ChangeLog.
-
-        * Scripts/webkitpy/tool/commands/setupgitclone.py:
-        (SetupGitClone.execute):
-
-2018-08-10  Truitt Savell  <tsavell@apple.com>
-
-        Unreviewed, rolling out r234749.
-
-        Caused all perf tests to fail in Sierra
-
-        Reverted changeset:
-
-        "[webkitpy][Win] LayoutTests: test names should be Unix style,
-        separated by slash not backslash"
-        https://bugs.webkit.org/show_bug.cgi?id=187973
-        https://trac.webkit.org/changeset/234749
-
-2018-08-10  David Fenton  <david_fenton@apple.com>
-
-        Timeouts in TestWebKitAPI.PreferredAudioBufferSize.AudioWithWebAudio and TestWebKitAPI.PreferredAudioBufferSize.WebAudio.
-        https://bugs.webkit.org/show_bug.cgi?id=188470.
-
-        Unreviewed test gardening, disabling the tests.
-
-        * TestWebKitAPI/Tests/WebKitCocoa/PreferredAudioBufferSize.mm:
-        (TEST_F):
-
-2018-08-10  Joseph Pecoraro  <pecoraro@apple.com>
-
-        Eliminate remaining instances of "ElCapitan" from LayoutTests / Tools
-        https://bugs.webkit.org/show_bug.cgi?id=188458
-
-        Reviewed by Alex Christensen.
-
-        * Scripts/webkitpy/layout_tests/models/test_expectations.py:
-        (TestExpectationParser):
-        * TestResultServer/static-dashboards/flakiness_dashboard.js:
-
-2018-08-09  Fujii Hironori  <Hironori.Fujii@sony.com>
-
-        [webkitpy][Win] LayoutTests: test names should be Unix style, separated by slash not backslash
-        https://bugs.webkit.org/show_bug.cgi?id=187973
-
-        Reviewed by Alex Christensen.
-
-        In LayoutTests, test names have been canonicalized in Unix style
-        since Bug 63597, for example 'fast/css/001.html'. But, Bug 179219,
-        Bug 179572, Bug 180660 and Bug 181814 have changed to use
-        os.path.seq instead of slash if Windows Python is used.
-
-        Revert parts of those changes. Change relative_test_filename to
-        return a slash sperated test name as well as chromium_win.py used
-        to do.
-
-        This change fixes all 41 test-webkitpy failures in WinCairo port.
-
-        * Scripts/webkitpy/layout_tests/models/test_expectations.py:
-        (TestExpectationParser._parse_line): Do not convert a test name by using normpath.
-        * Scripts/webkitpy/port/base.py:
-        (Port.normalize_test_name): Use TEST_PATH_SEPARATOR instead of os.path.sep.
-        (Port.relative_test_filename): Replace self.host.filesystem.sep with self.TEST_PATH_SEPARATOR.
-        (Port.abspath_for_test): Split test_name with TEST_PATH_SEPARATOR.
-        * Scripts/webkitpy/port/driver.py:
-        (Driver): Use '/' instead of os.sep.
-        * Scripts/webkitpy/port/win.py:
-        (WinCairoPort): Do not override TEST_PATH_SEPARATOR.
-
-2018-08-09  Thomas Denney  <tdenney@apple.com>
-
-        [WHLSL] Add individual test durations to output
-        https://bugs.webkit.org/show_bug.cgi?id=188448
-
-        Reviewed by Myles C. Maxfield.
-
-        * WebGPUShadingLanguageRI/Test.js: Add timing output.
-
-2018-08-09  Ben Richards  <benton_richards@apple.com>
-
-        We should cache the compiled sandbox profile in a data vault
-        https://bugs.webkit.org/show_bug.cgi?id=184991
-
-        Reviewed by Ryosuke Niwa.
-
-        Added trace points for sandbox initialization
-
-        * Tracing/SystemTracePoints.plist:
-
-2018-08-09  Jer Noble  <jer.noble@apple.com>
-
-        Video playback is using more power
-        https://bugs.webkit.org/show_bug.cgi?id=188452
-        <rdar://problem/42298937>
-
-        Reviewed by Eric Carlson.
-
-        * TestWebKitAPI/Tests/WebKitCocoa/PreferredAudioBufferSize.mm:
-        (PreferredAudioBufferSize::createView):
-        (PreferredAudioBufferSize::preferredAudioBufferSize const):
-        (TEST_F):
-
-2018-08-09  Alex Christensen  <achristensen@webkit.org>
-
-        REGRESSION(234640) Loading stalls in environments without SafariSafeBrowsing framework
-        https://bugs.webkit.org/show_bug.cgi?id=188453
-        <rdar://problem/43102553>
-
-        Reviewed by Chris Dumez
-
-        * TestWebKitAPI/Tests/WebKitCocoa/SafeBrowsing.mm:
-        (TEST):
-        (+[NullLookupContext sharedLookupContext]):
-        (TestWebKitAPI::TEST): Deleted.
-
-2018-08-09  Alex Christensen  <achristensen@webkit.org>
-
-        WKURLSchemeHandler crashes when sent errors with sync XHR
-        https://bugs.webkit.org/show_bug.cgi?id=188358
-
-        Reviewed by Chris Dumez.
-
-        * TestWebKitAPI/Tests/WebKitCocoa/WKURLSchemeHandler-1.mm:
-        (-[SyncErrorScheme webView:startURLSchemeTask:]):
-        (-[SyncErrorScheme webView:stopURLSchemeTask:]):
-        (-[SyncErrorScheme webView:runJavaScriptAlertPanelWithMessage:initiatedByFrame:completionHandler:]):
-
-2018-08-09  Per Arne Vollan  <pvollan@apple.com>
-
-        REGRESSION(r234652): fast/scrolling/rtl-scrollbars-animation-property.html is failing
-        https://bugs.webkit.org/show_bug.cgi?id=188397
-
-        Reviewed by Tim Horton.
-
-        After r234652, [NSScroller preferredScrollerStyle] is called in the UI process, and we need to set the same
-        user defaults related to scrollbars in the UI process as we do in the WebProcess.
-
-        * WebKitTestRunner/mac/main.mm:
-        (setDefaultsToConsistentValuesForTesting):
-
-2018-08-09  Wenson Hsieh  <wenson_hsieh@apple.com>
-
-        [iOS WK2] Rename and refactor DataInteractionTests and DataInteractionSimulator
-        https://bugs.webkit.org/show_bug.cgi?id=188439
-
-        Reviewed by Andy Estes and Megan Gardner.
-
-        Refactor DataInteractionSimulator (currently used for drag and drop testing on iOS) so that it exists across
-        macOS and iOS, and rename it to DragAndDropSimulator. Additionally, rename DataInteractionTests to
-        DragAndDropTests.
-
-        * TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj:
-        * TestWebKitAPI/Tests/WebKitCocoa/WKAttachmentTests.mm:
-        (TestWebKitAPI::TEST):
-        * TestWebKitAPI/Tests/ios/DragAndDropTestsIOS.mm: Renamed from Tools/TestWebKitAPI/Tests/ios/DataInteractionTests.mm.
-        (testIconImage):
-        (testZIPArchive):
-        (-[UIItemProvider registerDataRepresentationForTypeIdentifier:withData:]):
-        (-[TestWKWebView editorContainsImageElement]):
-        (-[TestWKWebView editorValue]):
-        (makeCGRectValue):
-        (checkCGRectIsEqualToCGRectWithLogging):
-        (checkSelectionRectsWithLogging):
-        (checkRichTextTypePrecedesPlainTextType):
-        (checkFirstTypeIsPresentAndSecondTypeIsMissing):
-        (checkTypeIdentifierIsRegisteredAtIndex):
-        (checkEstimatedSize):
-        (checkSuggestedNameAndEstimatedSize):
-        (checkStringArraysAreEqual):
-        (checkDragCaretRectIsContainedInRect):
-        (checkJSONWithLogging):
-        (testIconImageData):
-        (runTestWithTemporaryTextFile):
-        (runTestWithTemporaryFolder):
-        (TestWebKitAPI::TEST):
-        (TestWebKitAPI::setUpTestWebViewForDataTransferItems):
-        (TestWebKitAPI::testDragAndDropOntoTargetElements):
-        (TestWebKitAPI::testIconImageData):
-        * TestWebKitAPI/cocoa/DragAndDropSimulator.h: Renamed from Tools/TestWebKitAPI/ios/DataInteractionSimulator.h.
-        * TestWebKitAPI/ios/DragAndDropSimulatorIOS.mm: Renamed from Tools/TestWebKitAPI/ios/DataInteractionSimulator.mm.
-        (-[WKWebView dropInteractionDelegate]):
-        (-[WKWebView dragInteractionDelegate]):
-        (-[WKWebView dropInteraction]):
-        (-[WKWebView dragInteraction]):
-        (-[MockDragDropSession initWithItems:location:window:allowMove:]):
-        (-[MockDragDropSession allowsMoveOperation]):
-        (-[MockDragDropSession isRestrictedToDraggingApplication]):
-        (-[MockDragDropSession hasItemsConformingToTypeIdentifiers:]):
-        (-[MockDragDropSession canLoadObjectsOfClass:]):
-        (-[MockDragDropSession canLoadObjectsOfClasses:]):
-        (-[MockDragDropSession items]):
-        (-[MockDragDropSession setItems:]):
-        (-[MockDragDropSession addItems:]):
-        (-[MockDragDropSession locationInView:]):
-        (-[MockDropSession initWithProviders:location:window:allowMove:]):
-        (-[MockDropSession isLocal]):
-        (-[MockDropSession progress]):
-        (-[MockDropSession setProgressIndicatorStyle:]):
-        (-[MockDropSession progressIndicatorStyle]):
-        (-[MockDropSession operationMask]):
-        (-[MockDropSession localDragSession]):
-        (-[MockDropSession hasItemsConformingToTypeIdentifier:]):
-        (-[MockDropSession canCreateItemsOfClass:]):
-        (-[MockDropSession loadObjectsOfClass:completion:]):
-        (-[MockDragSession initWithWindow:allowMove:]):
-        (-[MockDragSession localOperationMask]):
-        (-[MockDragSession externalOperationMask]):
-        (-[MockDragSession session]):
-        (-[MockDragSession localContext]):
-        (-[MockDragSession setLocalContext:]):
-        (dragAndDropEventNames):
-        (-[DragAndDropSimulatorApplication _cancelAllTouches]):
-        (-[DragAndDropSimulator initWithWebView:]):
-        (-[DragAndDropSimulator dealloc]):
-        (-[DragAndDropSimulator _resetSimulatedState]):
-        (-[DragAndDropSimulator observedEventNames]):
-        (-[DragAndDropSimulator simulateAllTouchesCanceled:]):
-        (-[DragAndDropSimulator runFrom:to:]):
-        (-[DragAndDropSimulator runFrom:to:additionalItemRequestLocations:]):
-        (-[DragAndDropSimulator finalSelectionRects]):
-        (-[DragAndDropSimulator _concludeDropAndPerformOperationIfNecessary]):
-        (-[DragAndDropSimulator _enqueuePendingAdditionalItemRequestLocations]):
-        (-[DragAndDropSimulator _sendQueuedAdditionalItemRequest]):
-        (-[DragAndDropSimulator _advanceProgress]):
-        (-[DragAndDropSimulator _currentLocation]):
-        (-[DragAndDropSimulator _scheduleAdvanceProgress]):
-        (-[DragAndDropSimulator sourceItemProviders]):
-        (-[DragAndDropSimulator externalItemProviders]):
-        (-[DragAndDropSimulator setExternalItemProviders:]):
-        (-[DragAndDropSimulator phase]):
-        (-[DragAndDropSimulator liftPreviews]):
-        (-[DragAndDropSimulator lastKnownDragCaretRect]):
-        (-[DragAndDropSimulator waitForInputSession]):
-        (-[DragAndDropSimulator insertedAttachments]):
-        (-[DragAndDropSimulator removedAttachments]):
-        (-[DragAndDropSimulator endDataTransfer]):
-        (-[DragAndDropSimulator _webView:dataInteractionOperationWasHandled:forSession:itemProviders:]):
-        (-[DragAndDropSimulator _webView:willUpdateDataInteractionOperationToOperation:forSession:]):
-        (-[DragAndDropSimulator _webView:adjustedDataInteractionItemProvidersForItemProvider:representingObjects:additionalData:]):
-        (-[DragAndDropSimulator _webView:showCustomSheetForElement:]):
-        (-[DragAndDropSimulator _webView:willPerformDropWithSession:]):
-        (-[DragAndDropSimulator _webView:didInsertAttachment:]):
-        (-[DragAndDropSimulator _webView:didRemoveAttachment:]):
-        (-[DragAndDropSimulator _webView:focusShouldStartInputSession:]):
-        (-[DragAndDropSimulator _webView:didStartInputSession:]):
-        * TestWebKitAPI/ios/UIKitSPI.h:
-        * TestWebKitAPI/mac/DragAndDropSimulatorMac.mm: Added.
-        (-[DragAndDropSimulator initWithWebView:]):
-        (-[DragAndDropSimulator dealloc]):
-        (-[DragAndDropSimulator runFrom:to:]):
-        (-[DragAndDropSimulator insertedAttachments]):
-        (-[DragAndDropSimulator removedAttachments]):
-
-2018-08-09  Sihui Liu  <sihui_liu@apple.com>
-
-        REGRESSION (r232083): WKWebView loses first-party cookies on iOS
-        https://bugs.webkit.org/show_bug.cgi?id=188443
-        <rdar://problem/42991584>
-
-        Reviewed by Chris Dumez.
-
-        Skip WebKit.WKHTTPCookieStoreWithoutProcessPool on iOS.
-
-        * TestWebKitAPI/Tests/WebKitCocoa/WKHTTPCookieStore.mm:
-
-2018-08-09  Ross Kirsling  <ross.kirsling@sony.com>
-
-        Unreviewed follow-up fix for r234720.
-
-        * Scripts/webkitpy/bindings/main.py:
-        (BindingsTests.close_and_remove):
-        (BindingsTests.main):
-        Return to the original approach, since this seems to be the simplest cross-platform solution.
-
-2018-08-09  Ali Juma  <ajuma@chromium.org>
-
-        Import WPTs for IntersectionObserver
-        https://bugs.webkit.org/show_bug.cgi?id=188416
-
-        Reviewed by Simon Fraser.
-
-        Enable IntersectionObserver in DumpRenderTree. 
-
-        * DumpRenderTree/mac/DumpRenderTree.mm:
-        (enableExperimentalFeatures):
-        (setWebPreferencesForTestOptions):
-        * WebKitTestRunner/TestController.cpp:
-        (WTR::TestController::resetPreferencesToConsistentValues):
-
-2018-08-08  Ross Kirsling  <ross.kirsling@sony.com>
-
-        run-bindings-tests is not Win32-compatible
-        https://bugs.webkit.org/show_bug.cgi?id=188424
-
-        Reviewed by Fujii Hironori.
-
-        * Scripts/webkitpy/bindings/main.py:
-        (BindingsTests.main):
-        Stop leaking file descriptors.
-        (See https://www.logilab.org/blogentry/17873 for details, though the solution here is even simpler.)
-
-2018-08-08  Alex Christensen  <achristensen@webkit.org>
-
-        Fix possible null dereference in WebBackForwardList::restoreFromState
-        https://bugs.webkit.org/show_bug.cgi?id=188418
-        <rdar://problem/42531726>
-
-        Reviewed by Chris Dumez.
-
-        * TestWebKitAPI/Tests/WebKit/RestoreSessionState.cpp:
-        (TestWebKitAPI::TEST):
-
-2018-08-08  Ross Kirsling  <ross.kirsling@sony.com>
-
-        run-builtins-generator-tests does not correctly handle CRLFs from stderr
-        https://bugs.webkit.org/show_bug.cgi?id=188392
-
-        Reviewed by Fujii Hironori.
-
-        Background:
-        file.write converts LF to os.linesep, so if the input string contains CRLF, we end up writing CRCRLF to the file.
-
-        * Scripts/webkitpy/common/system/executive.py:
-        (Executive.run_command):
-        Normalize CRLF to LF in decoded stdout/stderr data, so that we don't return consumers a platform-specific string.
-
-2018-08-08  Aditya Keerthi  <akeerthi@apple.com>
-
-        Unreviewed, add myself to committers list.
-
-        * Scripts/webkitpy/common/config/contributors.json:
-
-2018-08-08  Commit Queue  <commit-queue@webkit.org>
-
-        Unreviewed, rolling out r234314, r234320, and r234321.
-        https://bugs.webkit.org/show_bug.cgi?id=188414
-
-        Caused email sign in issue (Requested by ryanhaddad on
-        #webkit).
-
-        Reverted changesets:
-
-        "Remove unused WKNavigationDelegatePrivate
-        decidePolicyForNavigationAction SPI"
-        https://bugs.webkit.org/show_bug.cgi?id=188077
-        https://trac.webkit.org/changeset/234314
-
-        "Fix API tests after r234314"
-        https://bugs.webkit.org/show_bug.cgi?id=188077
-        https://trac.webkit.org/changeset/234320
-
-        "Fix API tests after r234314"
-        https://bugs.webkit.org/show_bug.cgi?id=188077
-        https://trac.webkit.org/changeset/234321
-
-2018-08-08  Wenson Hsieh  <wenson_hsieh@apple.com>
-
-        [iOS] fast/events/ios/contenteditable-autocapitalize.html is a flaky failure
-        https://bugs.webkit.org/show_bug.cgi?id=188401
-        <rdar://problem/32542300>
-
-        Reviewed by Ryosuke Niwa.
-
-        When run individually, fast/events/ios/contenteditable-autocapitalize.html passes consistently; however, when
-        run right after another layout test that finishes while the keyboard is shown, this test sometimes fails. This
-        is because each of the three steps of this test ends when UIScriptController's `didHideKeyboardCallback` is
-        invoked, and if the keyboard only begins to dismiss after the previous test completes, we have a race. When the
-        keyboard finishes dismissing after the UI script is evaluated, it will trigger UI script completion early and
-        skip over one of the steps in the layout test, resulting in a text diff failure.
-
-        To fix this, add a mechanism in WebKitTestRunner to wait until the keyboard is dismissed (with a short timeout)
-        as a part of resetting test controller state, before moving on to the next layout test.
-
-        * WebKitTestRunner/cocoa/TestRunnerWKWebView.h:
-        * WebKitTestRunner/cocoa/TestRunnerWKWebView.mm:
-        (-[TestRunnerWKWebView didStartFormControlInteraction]):
-        (-[TestRunnerWKWebView didEndFormControlInteraction]):
-
-        Use these hooks to keep track of whether the previous test is presenting any form input UI.
-
-        (-[TestRunnerWKWebView isInteractingWithFormControl]):
-        * WebKitTestRunner/ios/TestControllerIOS.mm:
-        (WTR::handleKeyboardWillHideNotification):
-        (WTR::handleKeyboardDidHideNotification):
-        (WTR::TestController::platformInitialize):
-        (WTR::TestController::platformDestroy):
-
-        Register during initialization (and unregister during teardown) for keyboard hiding notifications, to keep track
-        of when the keyboard dismissal animation ends.
-
-        (WTR::TestController::platformResetStateToConsistentValues):
-
-        Make a couple of tweaks here: (1) if form input UI is being shown, tell the web view to resign first responder,
-        which causes the field to lose focus. (2) If necessary, wait for the current keyboard dismissal animation to
-        finish. This includes any keyboard dismissal animations triggered as a result of step (1).
-
-2018-08-05  Darin Adler  <darin@apple.com>
-
-        [Cocoa] More tweaks and refactoring to prepare for ARC
-        https://bugs.webkit.org/show_bug.cgi?id=188245
-
-        Reviewed by Dan Bernstein.
-
-        * DumpRenderTree/TestNetscapePlugIn/PluginObject.h: Use const void* for a
-        CFTypeRef instead of void*.
-
-        * DumpRenderTree/TestNetscapePlugIn/PluginObjectMac.mm:
-        (createCoreAnimationLayer): Use CFBridgingRetain.
-
-        * DumpRenderTree/mac/AccessibilityUIElementMac.mm:
-        (AccessibilityUIElement::lineTextMarkerRangeForTextMarker): Use bridging casts.
-        (AccessibilityUIElement::textMarkerRangeForElement): Ditto.
-        (AccessibilityUIElement::selectedTextMarkerRange): Ditto.
-        (AccessibilityUIElement::textMarkerRangeLength): Ditto.
-        (AccessibilityUIElement::attributedStringForTextMarkerRangeContainsAttribute): Ditto.
-        (AccessibilityUIElement::indexForTextMarker): Ditto.
-        (AccessibilityUIElement::textMarkerForIndex): Ditto.
-        (AccessibilityUIElement::isTextMarkerValid): Ditto.
-        (AccessibilityUIElement::previousTextMarker): Ditto.
-        (AccessibilityUIElement::nextTextMarker): Ditto.
-        (AccessibilityUIElement::stringForTextMarkerRange): Ditto.
-        (createJSStringRef): Ditto.
-        (AccessibilityUIElement::attributedStringForTextMarkerRange): Ditto.
-        (AccessibilityUIElement::attributedStringForTextMarkerRangeWithOptions): Ditto.
-        (AccessibilityUIElement::textMarkerRangeForMarkers): Ditto.
-        (AccessibilityUIElement::startTextMarkerForTextMarkerRange): Ditto.
-        (AccessibilityUIElement::endTextMarkerForTextMarkerRange): Ditto.
-        (AccessibilityUIElement::endTextMarkerForBounds): Ditto.
-        (AccessibilityUIElement::startTextMarkerForBounds): Ditto.
-        (AccessibilityUIElement::textMarkerForPoint): Ditto.
-        (AccessibilityUIElement::accessibilityElementForTextMarker): Ditto.
-        (AccessibilityUIElement::startTextMarker): Ditto.
-        (AccessibilityUIElement::endTextMarker): Ditto.
-        (AccessibilityUIElement::setSelectedVisibleTextRange): Ditto.
-        (AccessibilityUIElement::leftWordTextMarkerRangeForTextMarker): Ditto.
-        (AccessibilityUIElement::rightWordTextMarkerRangeForTextMarker): Ditto.
-        (AccessibilityUIElement::previousWordStartTextMarkerForTextMarker): Ditto.
-        (AccessibilityUIElement::nextWordEndTextMarkerForTextMarker): Ditto.
-        (AccessibilityUIElement::paragraphTextMarkerRangeForTextMarker): Ditto.
-        (AccessibilityUIElement::previousParagraphStartTextMarkerForTextMarker): Ditto.
-        (AccessibilityUIElement::nextParagraphEndTextMarkerForTextMarker): Ditto.
-        (AccessibilityUIElement::sentenceTextMarkerRangeForTextMarker): Ditto.
-        (AccessibilityUIElement::previousSentenceStartTextMarkerForTextMarker): Ditto.
-        (AccessibilityUIElement::nextSentenceEndTextMarkerForTextMarker): Ditto.
-
-        * DumpRenderTree/mac/DumpRenderTree.mm:
-        (resetWebPreferencesToConsistentValues): Use bridging cast.
-        (dumpFramesAsText): Ditto.
-        (dumpBackForwardListForAllWindows): Use a modern for loop instead of
-        CFArray functions to iterate the array.
-        (dump): Use bridging cast.
-        (runTest): Ditto.
-
-        * DumpRenderTree/mac/DumpRenderTreePasteboard.mm:
-        (+[LocalPasteboard alloc]): Use class_createInstance instead of NSAllocateObject.
-        (toUTI): Return a RetainPtr<CFStringRef>.
-        (-[LocalPasteboard types]): Use the above and bridging casts.
-        (-[LocalPasteboard availableTypeFromArray:]): Ditto.
-        (-[LocalPasteboard setData:forType:]): Ditto.
-        (-[LocalPasteboard dataForType:]): Ditto.
-        (-[LocalPasteboard setPropertyList:forType:]): Use NSPropertyListSerialization
-        instead of CFPropertyList.
-        (-[LocalPasteboard setString:forType:]): Use NSString method instead of
-        CFStringCreateExternalRepresentation.
-
-        * DumpRenderTree/mac/MockGeolocationProvider.h: Put CFTypeRef in a HashSet
-        instead of an Objective-C object pointer type.
-        * DumpRenderTree/mac/MockGeolocationProvider.mm:
-        (-[MockGeolocationProvider registerWebView:]): Update for above.
-        (-[MockGeolocationProvider unregisterWebView:]): Ditto.
-        (-[MockGeolocationProvider timerFired]): Ditto.
-
-        * DumpRenderTree/mac/MockWebNotificationProvider.h: Use CFTypeRef in
-        HashMap and HashSet instead of an Objective-C object pointer type.
-        * DumpRenderTree/mac/MockWebNotificationProvider.mm:
-        (-[MockWebNotificationProvider registerWebView:]): Update for above.
-        (-[MockWebNotificationProvider unregisterWebView:]): Ditto.
-        (-[MockWebNotificationProvider showNotification:fromWebView:]): Ditto.
-        (-[MockWebNotificationProvider cancelNotification:]): Ditto.
-        (-[MockWebNotificationProvider simulateWebNotificationClick:]): Ditto.
-
-        * DumpRenderTree/mac/ObjCPlugin.m:
-        (-[JSObjC retainObject:]): Use CFRetain instead of -[NSObject retain].
-
-        * DumpRenderTree/mac/TestRunnerMac.mm:
-        (TestRunner::addDisallowedURL): Use bridging cast.
-        (TestRunner::applicationCacheDiskUsageForOrigin): Ditto.
-        (TestRunner::clearApplicationCacheForOrigin): Ditto.
-        (originsArrayToJS): Ditto.
-        (TestRunner::copyDecodedHostName): Ditto.
-        (TestRunner::copyEncodedHostName): Ditto.
-        (TestRunner::queueLoad): Ditto.
-        (TestRunner::setDomainRelaxationForbiddenForURLScheme): Ditto.
-        (TestRunner::setMockGeolocationPositionUnavailableError): Ditto.
-        (TestRunner::setUserStyleSheetLocation): Ditto.
-        (TestRunner::setValueForUser): Ditto.
-        (TestRunner::overridePreference): Ditto.
-        (TestRunner::execCommand): Ditto.
-        (TestRunner::findString): Ditto.
-        (TestRunner::isCommandEnabled): Ditto.
-        (TestRunner::addOriginAccessWhitelistEntry): Ditto.
-        (TestRunner::removeOriginAccessWhitelistEntry): Ditto.
-        (TestRunner::addUserScript): Ditto.
-        (TestRunner::addUserStyleSheet): Ditto.
-        (TestRunner::evaluateInWebInspector): Ditto.
-        (TestRunner::evaluateScriptInIsolatedWorld): Ditto.
-        (TestRunner::authenticateSession): Ditto.
-        (TestRunner::grantWebNotificationPermission): Ditto.
-        (TestRunner::denyWebNotificationPermission): Ditto.
-        * TestRunnerShared/cocoa/LayoutTestSpellChecker.mm:
-        (-[LayoutTestSpellChecker setResultsFromJSObject:inContext:]): Ditto.
-
-        * WebKitTestRunner/InjectedBundle/mac/AccessibilityUIElementMac.mm:
-        (WTR::AccessibilityUIElement::lineTextMarkerRangeForTextMarker): Use bridging cast.
-        (WTR::AccessibilityUIElement::textMarkerRangeForElement): Ditto.
-        (WTR::AccessibilityUIElement::previousTextMarker): Ditto.
-        (WTR::AccessibilityUIElement::nextTextMarker): Ditto.
-        (WTR::AccessibilityUIElement::textMarkerRangeForMarkers): Ditto.
-        (WTR::AccessibilityUIElement::selectedTextMarkerRange): Ditto.
-        (WTR::AccessibilityUIElement::startTextMarkerForTextMarkerRange): Ditto.
-        (WTR::AccessibilityUIElement::endTextMarkerForTextMarkerRange): Ditto.
-        (WTR::AccessibilityUIElement::endTextMarkerForBounds): Ditto.
-        (WTR::AccessibilityUIElement::startTextMarkerForBounds): Ditto.
-        (WTR::AccessibilityUIElement::textMarkerForPoint): Ditto.
-        (WTR::AccessibilityUIElement::textMarkerForIndex): Ditto.
-        (WTR::AccessibilityUIElement::startTextMarker): Ditto.
-        (WTR::AccessibilityUIElement::endTextMarker): Ditto.
-        (WTR::AccessibilityUIElement::leftWordTextMarkerRangeForTextMarker): Ditto.
-        (WTR::AccessibilityUIElement::rightWordTextMarkerRangeForTextMarker): Ditto.
-        (WTR::AccessibilityUIElement::previousWordStartTextMarkerForTextMarker): Ditto.
-        (WTR::AccessibilityUIElement::nextWordEndTextMarkerForTextMarker): Ditto.
-        (WTR::AccessibilityUIElement::paragraphTextMarkerRangeForTextMarker): Ditto.
-        (WTR::AccessibilityUIElement::previousParagraphStartTextMarkerForTextMarker): Ditto.
-        (WTR::AccessibilityUIElement::nextParagraphEndTextMarkerForTextMarker): Ditto.
-        (WTR::AccessibilityUIElement::sentenceTextMarkerRangeForTextMarker): Ditto.
-        (WTR::AccessibilityUIElement::previousSentenceStartTextMarkerForTextMarker): Ditto.
-        (WTR::AccessibilityUIElement::nextSentenceEndTextMarkerForTextMarker): Ditto.
-
-        * WebKitTestRunner/PlatformMac.cmake: Added WebKitTestRunnerWindow.h and .mm files.
-        * WebKitTestRunner/WebKitTestRunner.xcodeproj/project.pbxproj: Ditto.
-
-        * WebKitTestRunner/mac/EventSenderProxy.mm: Add include of WebKitTestRunnerWindow.h
-        so we can use the type and know it inherits from NSWindow.
-
-        * WebKitTestRunner/mac/PlatformWebViewMac.mm: Add include of WebKitTestRunnerWindow.h
-        so we can use the type and know it inherits from NSWindow. Moved WebKitTestRunnerWindow
-        class into a separate file.
-        (WTR::PlatformWebView::keyWindow): Call +[WebKitTestRunnerWindow _WTR_keyWindow]
-        instead of have a second copy of the implementation.
-
-        * WebKitTestRunner/mac/WebKitTestRunnerEvent.mm: Add include of WebKitTestRunnerWindow.h
-        so we can use the type and know it inherits from NSWindow. Also removed "using namespace".
-        (+[WebKitTestRunnerEvent mouseLocation]): Updated for above.
-
-        * WebKitTestRunner/mac/WebKitTestRunnerPasteboard.mm:
-        (+[LocalPasteboard alloc]): Use class_createInstance instead of NSAllocateObject.
-        (-[LocalPasteboard availableTypeFromArray:]): Use modern for loop.
-        (-[LocalPasteboard setData:forType:]): Reordered method to be a tiny bit more efficient.
-        (-[LocalPasteboard setPropertyList:forType:]): Use NSPropertyListSerialization
-        instead of CFPropertyList.
-        (-[LocalPasteboard setString:forType:]): Use NSString method instead of
-        CFStringCreateExternalRepresentation.
-
-        * WebKitTestRunner/mac/WebKitTestRunnerWindow.h: Moved declaration of
-        WebKitTestRunnerWindow here so it can be included by files that need to know
-        what the type is and that it inherits from NSWindow.
-
-        * WebKitTestRunner/mac/WebKitTestRunnerWindow.mm: Added. Moved implementation here.
-        (+[WebKitTestRunnerWindow _WTR_keyWindow]): Use modern for loop instead of the old,
-        messier implementation.
-
-2018-08-07  Alex Christensen  <achristensen@webkit.org>
-
-        Add a test for using SafeBrowsing
-        https://bugs.webkit.org/show_bug.cgi?id=188290
-
-        Reviewed by Chris Dumez.
-
-        Swizzle out SSBLookupContext to write tests for how WebKit responds when it responds in different ways.
-        We don't do anything with its results in WebKit yet, but when we do I'll add the expected behavior to this test.
-
-        * TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj:
-        * TestWebKitAPI/Tests/WebKitCocoa/SafeBrowsing.mm: Added.
-        (-[SafeBrowsingDelegate webView:didFinishNavigation:]):
-        (TestWebKitAPI::TEST):
-
-2018-08-07  Lucas Forschler  <lforschler@apple.com>
-
-        If there's a Radar in the ChangeLog, webkit-patch upload/create-bug should put the radar in the bug and set InRadar
-        https://bugs.webkit.org/show_bug.cgi?id=188235
-
-        Reviewed by Daniel Bates, Kocsen Chung, and Aakash Jain.
-
-        * Scripts/webkitpy/common/checkout/changelog.py: 
-        (ChangeLogEntry): Teach the ChangeLog tools how to parse a radar_id
-        * Scripts/webkitpy/common/checkout/changelog_unittest.py: Test out the new _parse_radar_id functionality
-        * Scripts/webkitpy/common/net/bugzilla/bugzilla.py:
-        (Bugzilla.add_keyword_to_bug): Add logic to add keywords using the changeform
-        * Scripts/webkitpy/tool/commands/upload.py:
-        (Upload): Teach the upload step how to add radar information when creating a new bug.
-        * Scripts/webkitpy/tool/steps/__init__.py:
-        * Scripts/webkitpy/tool/steps/addradar.py: Added.
-        (AddRadar):
-        (AddRadar.run): Contains the logic to add the radar information to bugzilla.
-        * Scripts/webkitpy/tool/steps/createbug.py:
-        (CreateBug.run): set a "created_new_bug" state when creating a new bugzilla bug.
-
-2018-08-07  Wenson Hsieh  <wenson_hsieh@apple.com>
-
-        Post-review feedback after r234614.
-        https://bugs.webkit.org/show_bug.cgi?id=188322
-        <rdar://problem/42455270>
-
-        Reviewed by Sam Weinig.
-
-        Use std::round instead of roundl — there's no reason to go from double to long double precision when rounding.
-
-        * TestWebKitAPI/Tests/ios/KeyboardInputTestsIOS.mm:
-        (rounded):
-
-2018-08-07  Don Olmstead  <don.olmstead@sony.com>
-
-        [MiniBrowser][Win] Allow selection of recent user agents
-        https://bugs.webkit.org/show_bug.cgi?id=188371
-
-        Reviewed by Fujii Hironori.
-
-        Updates the user agent spoofing strings to the latest released
-        versions and adds Microsoft Edge's user agent.
-
-        Moves the menu item selection to the common base class, BrowserWindow
-        to support user agent spoofing in a WebKitLegacyBrowserWindow and
-        WebKitBrowserWindow.
-
-        * MiniBrowser/win/BrowserWindow.cpp: Added.
-        (BrowserWindow::setUserAgent):
-        * MiniBrowser/win/BrowserWindow.h:
-        * MiniBrowser/win/CMakeLists.txt:
-        * MiniBrowser/win/MainWindow.cpp:
-        (MainWindow::toggleMenuItem):
-        * MiniBrowser/win/MiniBrowserLib.rc:
-        * MiniBrowser/win/MiniBrowserLibResource.h:
-        * MiniBrowser/win/WebKitBrowserWindow.cpp:
-        * MiniBrowser/win/WebKitBrowserWindow.h:
-        * MiniBrowser/win/WebKitLegacyBrowserWindow.cpp:
-        (WebKitLegacyBrowserWindow::setUserAgent):
-        * MiniBrowser/win/WebKitLegacyBrowserWindow.h:
-
-2018-08-07  Yusuke Suzuki  <utatane.tea@gmail.com>
-
-        Unreviewed, add a new address to my configuration
-
-        * Scripts/webkitpy/common/config/contributors.json:
-
-2018-08-06  David Quesada  <david_quesada@apple.com>
-
-        webkitdirs.pm should default to iPhone SE for 64-bit testing
-        https://bugs.webkit.org/show_bug.cgi?id=188361
-
-        Reviewed by Daniel Bates.
-
-        Use iPhone SE as the default iOS Simulator device type. This corresponds to the change to webkitpy in r234467.
-
-        * Scripts/webkitdirs.pm:
-        (findOrCreateSimulatorForIOSDevice):
-        * Scripts/webkitpy/layout_tests/run_webkit_tests.py:
-        (parse_args):
-            Update the help text for --device-type to reflect the new default device.
-
-2018-08-06  Wenson Hsieh  <wenson_hsieh@apple.com>
-
-        [iOS] Caret disappears after resigning and becoming first responder if active focus state is retained
-        https://bugs.webkit.org/show_bug.cgi?id=188322
-        <rdar://problem/42455270>
-
-        Reviewed by Tim Horton.
-
-        Fixes an API test added in r234600, which had incorrect test expectations — namely, the width of the caret rect
-        was not in content view coordinates, unlike the x, y, and height.
-
-        * TestWebKitAPI/Tests/ios/KeyboardInputTestsIOS.mm:
-        (rounded):
-        (-[TestWKWebView waitForCaretViewFrameToBecome:]):
-        (-[TestWKWebView waitForSelectionViewRectsToBecome:]):
-        (TestWebKitAPI::TEST):
-
-2018-08-06  Ryan Haddad  <ryanhaddad@apple.com>
-
-        Unreviewed, rolling out r234569.
-
-        Breaks internal builds.
-
-        Reverted changeset:
-
-        "We should cache the compiled sandbox profile in a data vault"
-        https://bugs.webkit.org/show_bug.cgi?id=184991
-        https://trac.webkit.org/changeset/234569
-
-2018-08-06  Fujii Hironori  <Hironori.Fujii@sony.com>
-
-        [Win][DLLLauncher] Remove --highDPI switch
-        https://bugs.webkit.org/show_bug.cgi?id=188255
-
-        Reviewed by Yusuke Suzuki.
-
-        Bug 149417 has added `--highDPI` switch to make MiniBrowser DPI
-        aware. But, Bug 158733 has made MiniBrowser always DPI aware. It
-        became useless since the change. Remove the switch.
-
-        This change fixes the issue all dump render tree tests are failing
-        in WinCairo port if high DPI display is connected.
-
-        * MiniBrowser/win/Common.cpp:
-        (parseCommandLine): Removed `--highDPI` switch.
-        * win/DLLLauncher/DLLLauncherMain.cpp:
-        (wWinMain): Do not call SetProcessDPIAware.
-        (shouldUseHighDPI): Deleted.
-
-2018-08-06  Wenson Hsieh  <wenson_hsieh@apple.com>
-
-        [iOS] Tests that try to interact with the QuickType bar time out on iOS 11+
-        https://bugs.webkit.org/show_bug.cgi?id=188335
-        <rdar://problem/32542437> and <rdar://problem/32542433>
-
-        Reviewed by Tim Horton.
-
-        Remove a UIScriptController helper method that attempted to wait for QuickType buttons to appear, and then send
-        touch events to the buttons. These were only used for the two tests which this patch refactors. See the
-        LayoutTests ChangeLog for more detail.
-
-        * DumpRenderTree/ios/UIScriptControllerIOS.mm:
-        (WTR::UIScriptController::selectTextCandidateAtIndex): Deleted.
-        * TestRunnerShared/UIScriptContext/Bindings/UIScriptController.idl:
-        * TestRunnerShared/UIScriptContext/UIScriptController.cpp:
-        (WTR::UIScriptController::selectTextCandidateAtIndex): Deleted.
-        (WTR::UIScriptController::waitForTextPredictionsViewAndSelectCandidateAtIndex): Deleted.
-        * TestRunnerShared/UIScriptContext/UIScriptController.h:
-        * TestRunnerShared/spi/UIKitTestSPI.h:
-
-        Remove an internal class declaration that is now unnecessary.
-
-        * WebKitTestRunner/ios/UIScriptControllerIOS.mm:
-        (WTR::UIScriptController::selectTextCandidateAtIndex): Deleted.
-        (WTR::UIScriptController::waitForTextPredictionsViewAndSelectCandidateAtIndex): Deleted.
-
-2018-08-06  Wenson Hsieh  <wenson_hsieh@apple.com>
-
-        [iOS] Caret disappears after resigning and becoming first responder if active focus state is retained
-        https://bugs.webkit.org/show_bug.cgi?id=188322
-        <rdar://problem/42455270>
-
-        Reviewed by Tim Horton.
-
-        Adds plumbing in UIScriptController to grab the start and end selection handle rects for use in the new layout
-        test. Also adds new API tests to verify that when a web view resigns first responder, both caret and range
-        selection views are hidden, and when first responder status is restored, both caret and range selection views
-        are made visible again.
-
-        * DumpRenderTree/ios/UIScriptControllerIOS.mm:
-        (WTR::UIScriptController::selectionStartGrabberViewRect const):
-        (WTR::UIScriptController::selectionEndGrabberViewRect const):
-        * TestRunnerShared/UIScriptContext/Bindings/UIScriptController.idl:
-        * TestRunnerShared/UIScriptContext/UIScriptController.cpp:
-        (WTR::UIScriptController::selectionStartGrabberViewRect const):
-        (WTR::UIScriptController::selectionEndGrabberViewRect const):
-        * TestRunnerShared/UIScriptContext/UIScriptController.h:
-        * TestWebKitAPI/Tests/ios/KeyboardInputTestsIOS.mm:
-        (-[TestWKWebView waitForCaretViewFrameToBecome:]):
-        (-[TestWKWebView waitForSelectionViewRectsToBecome:]):
-        (webViewWithAutofocusedInput):
-
-        Pull out some common logic for creating a web view that allows programmatic focus to present the keyboard, and
-        immediately loading a web page with an autofocusing text field.
-
-        (TestWebKitAPI::TEST):
-        * TestWebKitAPI/cocoa/TestWKWebView.h:
-        * TestWebKitAPI/cocoa/TestWKWebView.mm:
-        (-[TestWKWebView caretViewRectInContentCoordinates]):
-        (-[TestWKWebView selectionViewRectsInContentCoordinates]):
-        * WebKitTestRunner/ios/UIScriptControllerIOS.mm:
-        (WTR::UIScriptController::selectionStartGrabberViewRect const):
-        (WTR::UIScriptController::selectionEndGrabberViewRect const):
-
-2018-08-06  Wenson Hsieh  <wenson_hsieh@apple.com>
-
-        [iOS] Layout tests that send HID events cause WebKitTestRunner to crash on recent SDKs
-        https://bugs.webkit.org/show_bug.cgi?id=188334
-        <rdar://problem/40630074>
-
-        Reviewed by Tim Horton.
-
-        To mark the end of previously dispatched IOHID events, HIDEventGenerator currently sends a vendor-defined event
-        and stores the completion callback ID for the previously dispatched events as vendor-defined data. When this
-        vendor-defined marker event is handled by the application, we then read the callback ID back from the event, map
-        it to a completion block, and invoke the completion block to signal that the previous HID event has been
-        processed.
-
-        This callback ID is an unsigned, so we tell IOKit that we need `sizeof(unsigned)` (4 bytes) to store it. On
-        shipping software, IOKit clamps this to a minimum of 8 bytes, i.e. `sizeof(CFIndex)`. When we later call
-        IOHIDEventGetIntegerValue to read the value of our vendor-defined data as a CFIndex, we get our expected
-        callback ID because the buffer was clamped to 8 bytes.
-
-        However, on recent iOS SDKs that contain the fix for <rdar://problem/20082284>, IOKit no longer clamps the size
-        of the vendor-defined data buffer to 8 bytes. This means that when we try to use IOHIDEventGetIntegerValue to
-        read our callback ID back, we end up getting a CFIndex where the lower 4 bytes are the callback ID we wrote, and
-        the upper 4 bytes are garbage. In the case where any of these upper 4 bytes are non-zero, we fail to map the
-        callback ID to a completion handler, and so we never finish dispatching the HID event, causing an exception to
-        be thrown.
-
-        To fix this, we adjust callback ID to be a CFIndex, which matches IOHIDEventGetIntegerValue's return type.
-
-        * WebKitTestRunner/ios/HIDEventGenerator.mm:
-        (+[HIDEventGenerator nextEventCallbackID]):
-        (-[HIDEventGenerator _sendMarkerHIDEventWithCompletionBlock:]):
-
-        Also refactor a bit of `-_sendMarkerHIDEventWithCompletionBlock:` by using auto and move semantics.
-
-2018-08-03  Ben Richards  <benton_richards@apple.com>
-
-        We should cache the compiled sandbox profile in a data vault
-        https://bugs.webkit.org/show_bug.cgi?id=184991
-
-        Reviewed by Ryosuke Niwa.
-
-        Added trace points for sandbox initialization
-
-        * Tracing/SystemTracePoints.plist:
-
-2018-08-03  David Fenton  <david_fenton@apple.com>
-
-        Unreviewed, rolling out r234517.
-
-        Caused API test failures on iOS
-
-        Reverted changeset:
-
-        "[Cocoa] setCookie API fails to set session cookies for
-        defaultDataStore if processPool created but not used"
-        https://bugs.webkit.org/show_bug.cgi?id=188209
-        https://trac.webkit.org/changeset/234517
-
-2018-08-03  Ben Richards  <benton_richards@apple.com>
-
-        Add configuration for automatic process pre-warming
-        https://bugs.webkit.org/show_bug.cgi?id=187108
-
-        Reviewed by Ryosuke Niwa.
-
-        Added new test case for setting maximum prewarmed process count and updated test case for process swap on navigation to set the maximum prewarmed process count to be 1 where relevant.
-
-        * TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj:
-        * TestWebKitAPI/Tests/WebKitCocoa/InitialWarmedProcessUsed.mm:
-        (TEST):
-        * TestWebKitAPI/Tests/WebKitCocoa/ProcessSwapOnNavigation.mm:
-        * TestWebKitAPI/Tests/WebKitCocoa/SetMaximumPrewarmedProcessCount.mm:
-        (TEST):
-
-2018-08-03  Aakash Jain  <aakash_jain@apple.com>
-
-        [ews-build] Add build step to Check Patch Relevance
-        https://bugs.webkit.org/show_bug.cgi?id=188295
-
-        Reviewed by Lucas Forschler.
-
-        * BuildSlaveSupport/ews-build/steps.py:
-        (CheckPatchRelevance): Added step to check patch relevance.
-        (CheckPatchRelevance._patch_is_relevant): Checks if the patch is relevant.
-        (CheckPatchRelevance._get_patch): Retrieves the patch from buildbot.
-        (CheckPatchRelevance._addToLog): Add the log message.
-        (CheckPatchRelevance.start):
-        * BuildSlaveSupport/ews-build/factories.py: Added CheckPatchRelevance step appropriately.
-
-2018-08-03  Jer Noble  <jer.noble@apple.com>
-
-        Unreviewed gardening; r234534 changed the policy for what audio session is set when a
-        media element is paused. Updating the test to reflect the new policy.
-
-        * TestWebKitAPI/Tests/WebKitLegacy/ios/AudioSessionCategoryIOS.mm:
-        (TestWebKitAPI::TEST):
-
-2018-08-03  Zalan Bujtas  <zalan@apple.com>
-
-        [LFC][Floating] Now that the document renderer belongs to "complicated cases", adjust viewport stretching.
-        https://bugs.webkit.org/show_bug.cgi?id=188306
-
-        Reviewed by Antti Koivisto.
-
-        * LayoutReloaded/misc/LFC-passing-tests.txt:
-
-2018-08-03  Carlos Garcia Campos  <cgarcia@igalia.com>
-
-        [WPE] Use the new key mapper API from WPEBackend
-        https://bugs.webkit.org/show_bug.cgi?id=188216
-
-        Reviewed by Žan Doberšek.
-
-        * WebKitTestRunner/wpe/EventSenderProxyWPE.cpp:
-        (WTR::EventSenderProxy::keyDown): Use wpe_input_xkb_context_get_entries_for_key_code() to get the hardware key
-        code.
-        * wpe/backends/WindowViewBackend.cpp:
-        (WPEToolingBackends::WindowViewBackend::WindowViewBackend):
-        (WPEToolingBackends::WindowViewBackend::handleKeyEvent):
-        (WPEToolingBackends::WindowViewBackend::modifiers const):
-        * wpe/backends/WindowViewBackend.h:
-
-2018-08-03  Carlos Garcia Campos  <cgarcia@igalia.com>
-
-        [WPE] Implement MouseEvent.buttons
-        https://bugs.webkit.org/show_bug.cgi?id=187998
-
-        Reviewed by Žan Doberšek.
-
-        Pass modifiers to mouse events.
-
-        * WebKitTestRunner/EventSenderProxy.h:
-        * WebKitTestRunner/wpe/EventSenderProxyWPE.cpp:
-        (WTR::senderButtonToWPEButton):
-        (WTR::modifierForButton):
-        (WTR::EventSenderProxy::mouseDown):
-        (WTR::EventSenderProxy::mouseUp):
-        (WTR::EventSenderProxy::mouseMoveTo):
-        (WTR::EventSenderProxy::mouseScrollBy):
-        (WTR::wkEventModifiersToWPE):
-        (WTR::wpeKeySymForKeyRef):
-        (WTR::EventSenderProxy::keyDown):
-        (WTR::EventSenderProxy::prepareAndDispatchTouchEvent):
-        * wpe/backends/WindowViewBackend.cpp:
-        (WPEToolingBackends::WindowViewBackend::handleKeyEvent):
-        (WPEToolingBackends::WindowViewBackend::modifiers const):
-        * wpe/backends/WindowViewBackend.h:
-
-2018-08-03  Carlos Garcia Campos  <cgarcia@igalia.com>
-
-        [WPE] Use WPE key symbols and new API instead of xkbcommon and the key mapper
-        https://bugs.webkit.org/show_bug.cgi?id=188093
-
-        Reviewed by Žan Doberšek.
-
-        * MiniBrowser/wpe/main.cpp: Use WPE key symbols.
-        * WebKitTestRunner/wpe/EventSenderProxyWPE.cpp:
-        (WTR::wpeKeySymForKeyRef): Use WPE key symbols and wpe_unicode_to_key_code().
-        (WTR::EventSenderProxy::keyDown): Do not set unicode char to the event since it was removed.
-        * flatpak/org.webkit.WPE.yaml: Bump WPEBackend and WPEBackend-fdo versions.
-        * wpe/backends/WindowViewBackend.cpp:
-        (WPEToolingBackends::WindowViewBackend::handleKeyEvent): Do not set unicode char to the event since it was removed.
-        * wpe/jhbuild.modules: Bump WPEBackend and WPEBackend-fdo versions.
-
-2018-08-02  Nan Wang  <n_wang@apple.com>
-
-        AX: [iOS] add support to return the attributed string under the element
-        https://bugs.webkit.org/show_bug.cgi?id=188276
-        <rdar://problem/42872357>
-
-        Reviewed by Chris Fleizach.
-
-        * DumpRenderTree/AccessibilityUIElement.cpp:
-        (attributedStringForElementCallback):
-        (AccessibilityUIElement::getJSClass):
-        * DumpRenderTree/AccessibilityUIElement.h:
-        * DumpRenderTree/ios/AccessibilityUIElementIOS.mm:
-        (AccessibilityUIElement::attributedStringForElement):
-        * WebKitTestRunner/InjectedBundle/AccessibilityUIElement.cpp:
-        (WTR::AccessibilityUIElement::attributedStringForElement):
-        * WebKitTestRunner/InjectedBundle/AccessibilityUIElement.h:
-        * WebKitTestRunner/InjectedBundle/Bindings/AccessibilityUIElement.idl:
-        * WebKitTestRunner/InjectedBundle/ios/AccessibilityUIElementIOS.mm:
-        (WTR::AccessibilityUIElement::attributedStringForElement):
-
-2018-08-02  Ryan Haddad  <ryanhaddad@apple.com>
-
-        Skip API test DataInteractionTests.CustomActionSheetPopover on current SDKs.
-        https://bugs.webkit.org/show_bug.cgi?id=175204
-
-        Unreviewed test gardening.
-
-        * TestWebKitAPI/Tests/ios/DataInteractionTests.mm:
-
-2018-08-02  Sihui Liu  <sihui_liu@apple.com>
-
-        [Cocoa] setCookie API fails to set session cookies for defaultDataStore if processPool created but not used
-        https://bugs.webkit.org/show_bug.cgi?id=188209
-
-        Reviewed by Geoffrey Garen.
-
-        * TestWebKitAPI/Tests/WebKitCocoa/WKHTTPCookieStore.mm:
-        (TEST):
-
-2018-08-02  Dan Bernstein  <mitz@apple.com>
-
-        Fixed WebKit.AttrStyle timing out on build.webkit.org.
-
-        * TestWebKitAPI/Tests/WebKitCocoa/AttrStyle.html: Changed from a URL that requires network
-          access to one that doesn’t.
-        * TestWebKitAPI/Tests/WebKitCocoa/AttrStyle.mm:
-        (TEST): Updated expected value.
-
-2018-08-02  Wenson Hsieh  <wenson_hsieh@apple.com>
-
-        [iOS] Keyboard becomes unresponsive after pressing delete while pressing down on a character key with accents
-        https://bugs.webkit.org/show_bug.cgi?id=188251
-        <rdar://problem/37842108>
-
-        Reviewed by Tim Horton.
-
-        Adds a new API test to verify that clients (in this case, UIKit) is allowed to invoke
-        -handleKeyWebEvent:withCompletionHandler: within the completion block of a prior invocation.
-
-        * TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj:
-        * TestWebKitAPI/Tests/ios/KeyboardInputTestsIOS.mm: Added.
-        (TestWebKitAPI::TEST):
-        * TestWebKitAPI/Tests/ios/TestInputDelegate.h: Added.
-
-        Pull some logic used to force an input session to start out from an existing API test file
-        (WKWebViewAutofillTests) and into a separate helper class that is used by both the existing API tests and the
-        new keyboard input test.
-
-        * TestWebKitAPI/Tests/ios/WKWebViewAutofillTests.mm:
-        (-[AutofillTestView initWithFrame:]):
-        (TestWebKitAPI::TEST):
-        (-[TestInputDelegate init]): Deleted.
-        (-[TestInputDelegate _webView:focusShouldStartInputSession:]): Deleted.
-        * TestWebKitAPI/ios/UIKitSPI.h:
-
-        Add some UIKit SPI utilized by the new API test.
-
-2018-08-01  Dan Bernstein  <mitz@apple.com>
-
-        Optionally expose Attr::style to JavaScript
-        https://bugs.webkit.org/show_bug.cgi?id=188226
-        <rdar://problem/42818113>
-
-        Reviewed by Darin Adler.
-
-        * TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj:
-        * TestWebKitAPI/Tests/WebKitCocoa/AttrStyle.html: Added.
-        * TestWebKitAPI/Tests/WebKitCocoa/AttrStyle.mm: Added.
-        (TEST):
-
-2018-08-02  David Fenton  <david_fenton@apple.com>
-
-        Unreviewed, rolling out r234489.
-
-        Caused 50+ crashes and 60+ API failures on iOS
-
-        Reverted changeset:
-
-        "[WTF] Rename String::format to String::deprecatedFormat"
-        https://bugs.webkit.org/show_bug.cgi?id=188191
-        https://trac.webkit.org/changeset/234489
-
-2018-08-01  Ross Kirsling  <ross.kirsling@sony.com>
-
-        TestWTF.WTF_NeverDestroyed.Construct output differs for MSVC in Debug mode
-        https://bugs.webkit.org/show_bug.cgi?id=188244
-
-        Reviewed by Daniel Bates.
-
-        Follow-up to r234179. MSVC chooses not to inline the lambda in Debug mode. 
-
-        * TestWebKitAPI/Tests/WTF/NeverDestroyed.cpp:
-
-2018-08-01  Tomas Popela  <tpopela@redhat.com>
-
-        [WTF] Rename String::format to String::deprecatedFormat
-        https://bugs.webkit.org/show_bug.cgi?id=188191
-
-        Reviewed by Darin Adler.
-
-        It should be replaced with string concatenation.
-
-        * DumpRenderTree/win/DumpRenderTree.cpp:
-        (applicationId):
-        (main):
-        * TestWebKitAPI/win/PlatformUtilitiesWin.cpp:
-        (TestWebKitAPI::Util::createURLForResource):
-        * WebKitTestRunner/InjectedBundle/TestRunner.cpp:
-        (WTR::cacheTestRunnerCallback):
-        * WebKitTestRunner/InjectedBundle/atk/AccessibilityNotificationHandlerAtk.cpp:
-        (WTR::AccessibilityNotificationHandler::connectAccessibilityCallbacks):
-        * WebKitTestRunner/InjectedBundle/atk/AccessibilityUIElementAtk.cpp:
-        (WTR::AccessibilityUIElement::attributedStringForRange):
-        (WTR::AccessibilityUIElement::url):
-        (WTR::stringAtOffset):
-        * WebKitTestRunner/TestController.cpp:
-        (WTR::TestController::didReceiveAuthenticationChallenge):
-        (WTR::TestController::downloadDidFail):
-        (WTR::originUserVisibleName):
-        (WTR::userMediaOriginHash):
-        (WTR::TestController::didNavigateWithNavigationData):
-        (WTR::TestController::didPerformClientRedirect):
-        (WTR::TestController::didPerformServerRedirect):
-        (WTR::TestController::didUpdateHistoryTitle):
-
-2018-08-01  Nan Wang  <n_wang@apple.com>
-
-        AX: AOM: Add ARIA IDL Attribute Reflection
-        https://bugs.webkit.org/show_bug.cgi?id=184676
-        <rdar://problem/39476882>
-
-        Reviewed by Chris Fleizach.
-
-        * DumpRenderTree/mac/DumpRenderTree.mm:
-        (enableExperimentalFeatures):
-        * WebKitTestRunner/TestController.cpp:
-        (WTR::TestController::resetPreferencesToConsistentValues):
-
-2018-08-01  Jonathan Bedard  <jbedard@apple.com>
-
-        Use iPhone SE as the default simulated device
-        https://bugs.webkit.org/show_bug.cgi?id=188227
-
-        Reviewed by Aakash Jain.
-
-        * Scripts/webkitpy/port/ios_simulator.py:
-        (IOSSimulatorPort):
-
-2018-08-01  Commit Queue  <commit-queue@webkit.org>
-
-        Unreviewed, rolling out r234443 and r234445.
-        https://bugs.webkit.org/show_bug.cgi?id=188224
-
-        Revision caused 3 api failures across all platforms.
-        (Requested by Truitt on #webkit).
-
-        Reverted changesets:
-
-        "Add configuration for automatic process pre-warming"
-        https://bugs.webkit.org/show_bug.cgi?id=187108
-        https://trac.webkit.org/changeset/234443
-
-        "Add configuration for automatic process pre-warming"
-        https://bugs.webkit.org/show_bug.cgi?id=187108
-        https://trac.webkit.org/changeset/234445
-
-2018-07-31  Ben Richards  <benton_richards@apple.com>
-
-        Add configuration for automatic process pre-warming
-        https://bugs.webkit.org/show_bug.cgi?id=187108
-
-        Reviewed by Ryosuke Niwa.
-
-        Added new test case for setting maximum prewarmed process count and updated test case for process swap on navigation to set the maximum prewarmed process count to be 1 where relevant.
-
-        * TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj:
-        * TestWebKitAPI/Tests/WebKitCocoa/ProcessSwapOnNavigation.mm:
-        * TestWebKitAPI/Tests/WebKitCocoa/SetMaximumPrewarmedProcessCount.mm: Added.
-        (TEST):
-
-2018-07-31  John Wilander  <wilander@apple.com>
-
-        Resource Load Statistics: Remove partitioned cookies for reduced complexity, lower memory footprint, and ability to support more platforms
-        https://bugs.webkit.org/show_bug.cgi?id=188109
-        <rdar://problem/42664391>
-
-        Reviewed by Brent Fulgham, Chris Dumez, and Alex Christensen.
-
-        This patch removes cookie partitioning which reduces the model to just
-        blocked cookies (in third-party contexts) and first-party cookie access.
-
-        Several of the changes are renaming to reflect that there are no more
-        cookie partitions.
-
-        * WebKitTestRunner/InjectedBundle/Bindings/TestRunner.idl:
-           Removed or renamed functions to reflect that there are no more
-           partitioned cookies.
-        * WebKitTestRunner/InjectedBundle/InjectedBundle.cpp:
-        (WTR::InjectedBundle::didReceiveMessageToPage):
-        * WebKitTestRunner/InjectedBundle/TestRunner.cpp:
-        (WTR::TestRunner::statisticsUpdateCookieBlocking):
-        (WTR::TestRunner::statisticsCallDidSetBlockCookiesForHostCallback):
-        (WTR::TestRunner::setCookieStoragePartitioningEnabled): Deleted.
-        (WTR::TestRunner::setStatisticsHasHadNonRecentUserInteraction): Deleted.
-        (WTR::TestRunner::setStatisticsTimeToLiveCookiePartitionFree): Deleted.
-        (WTR::TestRunner::statisticsUpdateCookiePartitioning): Deleted.
-        (WTR::TestRunner::statisticsSetShouldPartitionCookiesForHost): Deleted.
-        (WTR::TestRunner::statisticsCallDidSetPartitionOrBlockCookiesForHostCallback): Deleted.
-        * WebKitTestRunner/InjectedBundle/TestRunner.h:
-        * WebKitTestRunner/TestController.cpp:
-        (WTR::TestController::statisticsUpdateCookieBlocking):
-        (WTR::TestController::setStatisticsHasHadNonRecentUserInteraction): Deleted.
-        (WTR::TestController::setStatisticsTimeToLiveCookiePartitionFree): Deleted.
-        (WTR::TestController::statisticsUpdateCookiePartitioning): Deleted.
-        (WTR::TestController::statisticsSetShouldPartitionCookiesForHost): Deleted.
-        * WebKitTestRunner/TestController.h:
-        * WebKitTestRunner/TestInvocation.cpp:
-        (WTR::TestInvocation::didReceiveSynchronousMessageFromInjectedBundle):
-        (WTR::TestInvocation::didSetBlockCookiesForHost):
-        (WTR::TestInvocation::didSetHasHadUserInteraction):
-        (WTR::TestInvocation::didSetPartitionOrBlockCookiesForHost): Deleted.
-        (WTR::TestInvocation::didSetHasHadNonRecentUserInteraction): Deleted.
-        * WebKitTestRunner/TestInvocation.h:
-        * WebKitTestRunner/cocoa/TestControllerCocoa.mm:
-        (WTR::initializeWebViewConfiguration):
-
-2018-07-31  Alex Christensen  <achristensen@webkit.org>
-
-        Build fix.  Remove unused variable.
-
-        * TestWebKitAPI/Tests/WebCore/TransformationMatrix.cpp:
-        (TestWebKitAPI::TEST):
-        transform was never used. It was introduced in r205871
-        Apparently I'm the first one to build in a certain configuration that found it.
-
-2018-07-31  Fujii Hironori  <Hironori.Fujii@sony.com>
-
-        server_process.py should print returncode in debug log if the process crashed
-        https://bugs.webkit.org/show_bug.cgi?id=188160
-
-        Reviewed by Konstantin Tokarev.
-
-        * Scripts/webkitpy/port/server_process.py:
-        (ServerProcess.has_crashed): Put self._proc.returncode into the debug log message.
-
-2018-07-31  Tomas Popela  <tpopela@redhat.com>
-
-        [WTF] String::formatWithArguments() is unused
-        https://bugs.webkit.org/show_bug.cgi?id=187955
-
-        Reviewed by Darin Adler.
-
-        Remove tests for WTF::String::formatWithArguments() as it's unused and
-        we are removing it.
-
-        * TestWebKitAPI/Tests/WTF/WTFString.cpp:
-
-2018-07-30  Aakash Jain  <aakash_jain@apple.com>
-
-        [ews-build] Fix typo in builder name for Bindings-tests-EWS.
-
-        * BuildSlaveSupport/ews-build/config.json: Changed 'tests' to 'Tests' to make it consistent with
-        other builders.
-
-2018-07-30 Aakash Jain  <aakash_jain@apple.com>
-
-        [ews-build] Add support for API tests in OpenSource EWS
-        https://bugs.webkit.org/show_bug.cgi?id=188156
-
-        Reviewed by Lucas Forschler.
-
-        * BuildSlaveSupport/ews-build/config.json:
-        * BuildSlaveSupport/ews-build/factories.py:
-        (BuildFactory.__init__): Added triggers parameter.
-        (BuildFactory): If trigger is defined, create and upload archive and trigger appropriate queues.
-        (BuildFactory.propertiesToPassToTriggers): Pass all the required properties to triggered queue.
-        (APITestsFactory): Factory for running API tests.
-        * BuildSlaveSupport/ews-build/steps.py:
-        (ConfigureBuild.start): Set the property only if property is defined in config.json. Also set the 
-        source of the property.
-
-2018-07-30  Aakash Jain  <aakash_jain@apple.com>
-
-        [ews-build] loadConfig should ensure that the triggers are valid
-        https://bugs.webkit.org/show_bug.cgi?id=188134
-
-        Reviewed by Lucas Forschler.
-
-        * BuildSlaveSupport/ews-build/loadConfig.py:
-        (checkValidBuilder): Added a check to ensure that the builder doesn't refernce non-existing scheduler.
-        (checkValidSchedulers): Ensures that the scheduler is not un-used.
-        * BuildSlaveSupport/ews-build/loadConfig_unittest.py: Updated unit-tests.
-
-2018-07-27  Michael Catanzaro  <mcatanzaro@igalia.com>
-
-        Unreviewed, fix typo in test expectations and mark another test as timing out
-
-        * TestWebKitAPI/glib/TestExpectations.json:
-
-2018-07-27  Michael Catanzaro  <mcatanzaro@igalia.com>
-
-        More unreviewed GTK test gardening
-
-        * TestWebKitAPI/glib/TestExpectations.json:
-
-2018-07-27  Simon Fraser  <simon.fraser@apple.com>
-
-        Be more conservative with compositing layer creation when memory is low
-        https://bugs.webkit.org/show_bug.cgi?id=187866
-        rdar://problem/42366345
-
-        Reviewed by Zalan Bujtas.
-        
-        When process physical footprint is above a fraction of the jetsam limit, be more conservative in making
-        compositing layers. We avoid compositing for these situations:
-        1. Layers with 3D transforms which are affine (like translateZ(0)).
-        2. Layers with will-change
-        3. Layers for canvases (other than WebGL/WebGPU)
-        
-        We reuse some macOS code in MemoryPressureHandler() but choose different thresholds for iOS,
-        falling into "conservative mode" at 50% of jetsam limit, and "strict mode" at 65%.
-        Compositing chooses to be more conservative in either "conservative" or "strict" memory modes.
-        
-        Plumb through a "compositingPolicyOverride" both so that on-device testing isn't
-        flakily falling into a different mode, and so that we can impose the conservative
-        mode for testing.
-
-        * WebKitTestRunner/InjectedBundle/InjectedBundlePage.cpp:
-        (WTR::InjectedBundlePage::prepare):
-
-2018-07-27  Michael Catanzaro  <mcatanzaro@igalia.com>
-
-        Unreviewed GTK test gardening
-
-        * TestWebKitAPI/glib/TestExpectations.json:
-
-2018-07-27  Alex Christensen  <achristensen@webkit.org>
-
-        Fix API tests after r234314
-        https://bugs.webkit.org/show_bug.cgi?id=188077
-
-        Make API tests use new delegate method.
-
-        * TestWebKitAPI/Tests/WebKitCocoa/WebsitePolicies.mm:
-        (-[WebsitePoliciesWebsiteDataStoreDelegate _webView:decidePolicyForNavigationAction:userInfo:decisionHandler:]):
-        (-[WebsitePoliciesWebsiteDataStoreDelegate _webView:decidePolicyForNavigationAction:decisionHandler:]): Deleted.
-
-2018-07-27  Alex Christensen  <achristensen@webkit.org>
-
-        Fix API tests after r234314
-        https://bugs.webkit.org/show_bug.cgi?id=188077
-
-        Make API tests use new delegate method.
-
-        * TestWebKitAPI/Tests/WebKitCocoa/WebsitePolicies.mm:
-        (-[ContentBlockingWebsitePoliciesDelegate _webView:decidePolicyForNavigationAction:userInfo:decisionHandler:]):
-        (-[AutoplayPoliciesDelegate _webView:decidePolicyForNavigationAction:userInfo:decisionHandler:]):
-        (-[AsyncAutoplayPoliciesDelegate _webView:decidePolicyForNavigationAction:userInfo:decisionHandler:]):
-        (-[CustomHeaderFieldsDelegate _webView:decidePolicyForNavigationAction:userInfo:decisionHandler:]):
-        (-[PopUpPoliciesDelegate _webView:decidePolicyForNavigationAction:userInfo:decisionHandler:]):
-        (-[ContentBlockingWebsitePoliciesDelegate _webView:decidePolicyForNavigationAction:decisionHandler:]): Deleted.
-        (-[AutoplayPoliciesDelegate _webView:decidePolicyForNavigationAction:decisionHandler:]): Deleted.
-        (-[AsyncAutoplayPoliciesDelegate _webView:decidePolicyForNavigationAction:decisionHandler:]): Deleted.
-        (-[CustomHeaderFieldsDelegate _webView:decidePolicyForNavigationAction:decisionHandler:]): Deleted.
-        (-[PopUpPoliciesDelegate _webView:decidePolicyForNavigationAction:decisionHandler:]): Deleted.
-
-2018-07-26  Andy VanWagoner  <andy@vanwagoner.family>
-
-        [INTL] Remove INTL sub-feature compile flags
-        https://bugs.webkit.org/show_bug.cgi?id=188081
-
-        Reviewed by Michael Catanzaro.
-
-        * Scripts/webkitperl/FeatureList.pm:
-        * TestWebKitAPI/Configurations/FeatureDefines.xcconfig:
-
-2018-07-26  Chris Dumez  <cdumez@apple.com>
-
-        Loading a file URL and then issuing a reload right away causes the load to fail due to sandboxing
-        https://bugs.webkit.org/show_bug.cgi?id=188078
-        <rdar://problem/42562493>
-
-        Reviewed by Geoff Garen.
-
-        Add API test coverage. It exercises the right code path but unfortunately is not yet a regression
-        test because TestWebKitAPI does not appear to be sandboxed (rdar://problem/42638129).
-
-        * TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj:
-        * TestWebKitAPI/Tests/WebKitCocoa/LoadFileThenReload.mm: Added.
-        (-[LoadFileThenReloadDelegate webView:didFinishNavigation:]):
-        (-[LoadFileThenReloadDelegate webView:didFailProvisionalNavigation:withError:]):
-        (-[LoadFileThenReloadDelegate webView:didFailNavigation:withError:]):
-        (TEST):
-
-2018-07-26  Aaron Chu  <aaron_chu@apple.com>
-
-        Add Aaron Chu as a committer to contributors.json
-        https://bugs.webkit.org/show_bug.cgi?id=188066
-
-        Unreviwed. Added myself as committer.
-
-        * Scripts/webkitpy/common/config/contributors.json:
-
-2018-07-26  Aakash Jain  <aakash_jain@apple.com>
-
-        [ews-build] Add build step to run API Tests
-        https://bugs.webkit.org/show_bug.cgi?id=188079
-
-        Reviewed by Lucas Forschler.
-
-        * BuildSlaveSupport/ews-build/steps.py:
-        (TestWithFailureCount): Common class for various tests with failure count.
-        (TestWithFailureCount.start): Configure Buildbot's logobserver.
-        (TestWithFailureCount.commandComplete): Updates the failure count.
-        (TestWithFailureCount.evaluateCommand): Set the step status appropriately.
-        (TestWithFailureCount.getResultSummary): Set the step summary appropriately.
-        (RunAPITests): Build step to run API tests.
-        (RunAPITests.countFailures): Count the failures by parsing the output.
-        * BuildSlaveSupport/ews-build/steps_unittest.py: Added unit-tests.
-
-2018-07-26  Ross Kirsling  <ross.kirsling@sony.com>
-
-        String(View) should have a splitAllowingEmptyEntries function instead of a flag parameter
-        https://bugs.webkit.org/show_bug.cgi?id=187963
-
-        Reviewed by Alex Christensen.
-
-        * TestWebKitAPI/Tests/WTF/WTFString.cpp:
-        Add tests for String::split and String::splitAllowingEmptyEntries.
-
-        * TestWebKitAPI/Tests/WTF/StringView.cpp:
-        * WebKitTestRunner/TestController.cpp:
-        (WTR::updateTestOptionsFromTestHeader):
-        Update split/splitAllowingEmptyEntries usage.
-
-2018-07-26  Commit Queue  <commit-queue@webkit.org>
-
-        Unreviewed, rolling out r234181 and r234189.
-        https://bugs.webkit.org/show_bug.cgi?id=188075
-
-        These are not needed right now (Requested by thorton on
-        #webkit).
-
-        Reverted changesets:
-
-        "Enable Web Content Filtering on watchOS"
-        https://bugs.webkit.org/show_bug.cgi?id=187979
-        https://trac.webkit.org/changeset/234181
-
-        "HAVE(PARENTAL_CONTROLS) should be true on watchOS"
-        https://bugs.webkit.org/show_bug.cgi?id=187985
-        https://trac.webkit.org/changeset/234189
-
-2018-07-04  Darin Adler  <darin@apple.com>
-
-        Improve WebGPU implementation, including using Metal Objective-C protocols more simply and correctly
-        https://bugs.webkit.org/show_bug.cgi?id=187333
-
-        Reviewed by Sam Weinig.
-
-        * TestWebKitAPI/Tests/WebCore/mac/GPUBuffer.mm: Updated.
-        * TestWebKitAPI/Tests/WebCore/mac/GPUCommandQueue.mm: Updated.
-        * TestWebKitAPI/Tests/WebCore/mac/GPUDevice.mm: Updated.
-        * TestWebKitAPI/Tests/WebCore/mac/GPUFunction.mm: Updated.
-        * TestWebKitAPI/Tests/WebCore/mac/GPULibrary.mm: Updated.
-
-2018-07-25  Aakash Jain  <aakash_jain@apple.com>
-
-        [ews-build] Add build steps DownloadBuiltProduct and ExtractBuiltProduct
-        https://bugs.webkit.org/show_bug.cgi?id=188031
-
-        Reviewed by Lucas Forschler.
-
-        * BuildSlaveSupport/ews-build/steps.py:
-        (DownloadBuiltProduct): Added build step DownloadBuiltProduct.
-        (ExtractBuiltProduct): Added build step ExtractBuiltProduct.
-        * BuildSlaveSupport/ews-build/steps_unittest.py: Added unit-tests.
-
-2018-07-25  Aakash Jain  <aakash_jain@apple.com>
-
-        [ews-build] Add build step UploadBuiltProduct
-        https://bugs.webkit.org/show_bug.cgi?id=188016
-
-        Reviewed by Lucas Forschler.
-
-        * BuildSlaveSupport/ews-build/steps.py:
-        (UploadBuiltProduct): Added build step UploadBuiltProduct.
-        * BuildSlaveSupport/ews-build/steps_unittest.py: Added unit-test.
-
-2018-07-25  Aakash Jain  <aakash_jain@apple.com>
-
-        [ews-build] Add build step ArchiveBuiltProduct
-        https://bugs.webkit.org/show_bug.cgi?id=188014
-
-        Reviewed by Lucas Forschler.
-
-        * BuildSlaveSupport/ews-build/steps.py:
-        (ArchiveBuiltProduct): Added build step ArchiveBuiltProduct.
-        * BuildSlaveSupport/ews-build/steps_unittest.py: Added unit-tests.
-
-2018-07-25  Chris Dumez  <cdumez@apple.com>
-
-        navigator.userAgent may return outdated value after webView.customUserAgent is set
-        https://bugs.webkit.org/show_bug.cgi?id=188009
-        <rdar://problem/42566456>
-
-        Reviewed by Alex Christensen.
-
-        Add API test coverage.
-
-        * TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj:
-        * TestWebKitAPI/Tests/WebKitCocoa/CustomUserAgent.mm: Added.
-        (TEST):
-
-2018-07-25  Charlie Turner  <cturner@igalia.com>
-
-        [Flatpak] Pass more environment variables to sandbox
-        https://bugs.webkit.org/show_bug.cgi?id=188002
-
-        Reviewed by Michael Catanzaro.
-
-        * flatpak/flatpakutils.py:
-        (WebkitFlatpak.run_in_sandbox):
-
-2018-07-25  Zan Dobersek  <zdobersek@igalia.com>
-
-        [WPE] Switch ViewBackend to using wpe_view_backend_exportable_fdo_egl APIs
-        https://bugs.webkit.org/show_bug.cgi?id=187997
-
-        Reviewed by Carlos Garcia Campos.
-
-        Instead of the raw wpe_view_backend_exportable_fdo API, switch to using
-        the EGL-oriented API that simply serves EGLImageKHR objects instead of
-        raw buffer resources. This allows us not to wrangle with creating and
-        managing lifetime of the same EGLImageKHR by ourselves.
-
-        ViewBackend now defines a wpe_view_backend_exportable_fdo_egl_client
-        struct and uses wpe_view_backend_exportable_fdo_egl_create() to create
-        the exportable object. This means that for each frame we're served an
-        EGLImageKHR object that is passed on to the displayBuffer()
-        implementation.
-
-        HeadlessViewBackend now just manages the pending and committed image
-        objects. Width and height values used during pixel data retrieval are
-        gathered from the ViewBackend's size that was specified when
-        constructed.
-
-        WindowViewBackend similarly only manages the committed image object,
-        not having to create an equivalent EGLImageKHR object from the raw
-        buffer resource anymore.
-
-        Unused EGL entrypoints in both HeadlessViewBackend and WindowViewBackend
-        classes are removed, and destructors for both classes are made virtual.
-
-        The WPEBackend-fdo version is bumped in order to have the EGL-specific
-        export API available.
-
-        * flatpak/org.webkit.WPE.yaml:
-        * wpe/backends/HeadlessViewBackend.cpp:
-        (WPEToolingBackends::HeadlessViewBackend::HeadlessViewBackend):
-        (WPEToolingBackends::HeadlessViewBackend::~HeadlessViewBackend):
-        (WPEToolingBackends::HeadlessViewBackend::createSnapshot):
-        (WPEToolingBackends::HeadlessViewBackend::performUpdate):
-        (WPEToolingBackends::HeadlessViewBackend::displayBuffer):
-        * wpe/backends/HeadlessViewBackend.h:
-        * wpe/backends/ViewBackend.cpp:
-        (WPEToolingBackends::ViewBackend::initialize):
-        * wpe/backends/ViewBackend.h:
-        * wpe/backends/WindowViewBackend.cpp:
-        (WPEToolingBackends::WindowViewBackend::WindowViewBackend):
-        (WPEToolingBackends::WindowViewBackend::~WindowViewBackend):
-        (WPEToolingBackends::WindowViewBackend::displayBuffer):
-        * wpe/backends/WindowViewBackend.h:
-        * wpe/jhbuild.modules:
-
-2018-07-24  Thomas Denney  <tdenney@apple.com>
-
-        Remove support for the double type from WHLSL
-        https://bugs.webkit.org/show_bug.cgi?id=187977
-
-        Reviewed by Myles C. Maxfield.
-
-        * WebGPUShadingLanguageRI/All.js: Removed DoubleLiteral.js and DoubleLiteralType.js include.
-        * WebGPUShadingLanguageRI/DoubleLiteral.js: Removed.
-        * WebGPUShadingLanguageRI/DoubleLiteralType.js: Removed.
-        * WebGPUShadingLanguageRI/Intrinsics.js: Removed double arithmetic functions.
-        * WebGPUShadingLanguageRI/Parse.js:
-        (parseTerm): Removed parsing of double literals.
-        * WebGPUShadingLanguageRI/SPIRV.html: Removed DoubleLiteral.js and DoubleLiteralType.js include.
-        * WebGPUShadingLanguageRI/SPIRVCodegen.js: Removed double type logic.
-        * WebGPUShadingLanguageRI/StandardLibrary.js: Removed double type declarations.
-        * WebGPUShadingLanguageRI/Test.html: Removed DoubleLiteral.js and DoubleLiteralType.js include.
-        * WebGPUShadingLanguageRI/Test.js: Removed tests that test doubles or ported them to use floats.
-        * WebGPUShadingLanguageRI/index.html: Removed DoubleLiteral.js and DoubleLiteralType.js include.
-
-2018-07-24  Tim Horton  <timothy_horton@apple.com>
-
-        Enable Web Content Filtering on watchOS
-        https://bugs.webkit.org/show_bug.cgi?id=187979
-        <rdar://problem/42559346>
-
-        Reviewed by Wenson Hsieh.
-
-        * TestWebKitAPI/Configurations/FeatureDefines.xcconfig:
-
-2018-07-24  Daniel Bates  <dabates@apple.com>
-
-        Move-constructing NeverDestroyed should move construct underlying object instead of copy constructing it
-        https://bugs.webkit.org/show_bug.cgi?id=187971
-
-        Reviewed by Saam Barati.
-
-        Add unit tests to cover move constructing a NeverDestroyed with a non-const and const data type
-        as well as invoking makeNeverDestroyed() with the result of a function that returns a move-only
-        data type.
-
-        * TestWebKitAPI/CMakeLists.txt:
-        * TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj:
-        * TestWebKitAPI/Tests/WTF/MoveOnlyLifecycleLogger.cpp: Added.
-        (TestWebKitAPI::MoveOnlyLifecycleLogger::MoveOnlyLifecycleLogger):
-        (TestWebKitAPI::MoveOnlyLifecycleLogger::~MoveOnlyLifecycleLogger):
-        (TestWebKitAPI::MoveOnlyLifecycleLogger::operator=):
-        (TestWebKitAPI::MoveOnlyLifecycleLogger::setName):
-        (TestWebKitAPI::TEST):
-        * TestWebKitAPI/Tests/WTF/MoveOnlyLifecycleLogger.h: Added.
-        * TestWebKitAPI/Tests/WTF/NeverDestroyed.cpp:
-        (TestWebKitAPI::TEST):
-
-2018-07-24  Thomas Denney  <tdenney@apple.com>
-        Minor changes to the WHLSL interpreter to support the new compiler
-        https://bugs.webkit.org/show_bug.cgi?id=187728
-
-        Reviewed by Myles C. Maxfield.
-
-        This bug contains small changes that are make minor changes to the
-        behaviour of the existing interpreter (the tests are unaffected and all
-        still pass).
-
-        * WebGPUShadingLanguageRI/EnumType.js:
-        (EnumType.prototype.get isEnum): Property added to EnumType and Type
-        * WebGPUShadingLanguageRI/Evaluator.js: Correct typo
-        * WebGPUShadingLanguageRI/Func.js:
-        (Func.prototype.get isEntryPoint): Add boolean property
-        * WebGPUShadingLanguageRI/FuncInstantiator.js:
-        (FuncInstantiator.prototype.get instances): Public accessor for existing
-        private field
-        * WebGPUShadingLanguageRI/FunctionLikeBlock.js: Refactored the class so
-        that it directly wraps a Func instance (therefore the only interface
-        change is to the constructor and the addition of the func accessor).
-        This change allows the de-inlining of functions.
-        (FunctionLikeBlock):
-        (FunctionLikeBlock.prototype.get func):
-        (FunctionLikeBlock.prototype.get returnType):
-        (FunctionLikeBlock.prototype.get parameters):
-        (FunctionLikeBlock.prototype.get body):
-        * WebGPUShadingLanguageRI/Inliner.js: Use the new FunctionLikeBlock
-        constructor
-        (Inliner.prototype.visitCallExpression):
-        (Inliner):
-        * WebGPUShadingLanguageRI/Intrinsics.js: Migrate the swizzle operators
-        to native functions (with additional support in StandardLibrary.js).
-        This reduces first-time compile-time from around 4s to around 0.1s.
-        (Intrinsics):
-        * WebGPUShadingLanguageRI/Lexer.js: Ensure that the parsing of multiline
-        comments completes correctly
-        (Lexer.prototype.next):
-        (Lexer):
-        * WebGPUShadingLanguageRI/MakeArrayRefExpression.js: Addition of type
-        property, which is based on the type of the lValue and the semantics of
-        the language (the @ operator yields an array reference with the thread
-        scope).
-        (MakeArrayRefExpression):
-        (MakeArrayRefExpression.prototype.get type):
-        * WebGPUShadingLanguageRI/NativeFunc.js: Add setter/getter for
-        implementationData to support swizzle operator change
-        (NativeFunc.prototype.get implementationData):
-        (NativeFunc.prototype.set implementationData):
-        * WebGPUShadingLanguageRI/Rewriter.js: Use new FunctionLikeBlock
-        constructor
-        (Rewriter.prototype.visitFunctionLikeBlock):
-        * WebGPUShadingLanguageRI/StandardLibrary.js:
-        (_generateSwizzle): Deleted.
-        * WebGPUShadingLanguageRI/SwizzleOp.js: Introduces a new class for
-        carrying metadata about swizzle operators so that they are implemented
-        as native functions rather than generated at compile time.
-        (SwizzleOp):
-        (SwizzleOp.prototype.get outSize):
-        (SwizzleOp.prototype.get components):
-        (SwizzleOp.prototype.get inSize):
-        (SwizzleOp.prototype.toString):
-        (SwizzleOp.allSwizzleOperators.):
-        (SwizzleOp.allSwizzleOperators):
-        * WebGPUShadingLanguageRI/StandardLibrary.js: +        (SwizzleOp):
-        (SwizzleOp.prototype.get outSize):
-        (SwizzleOp.prototype.get components):
-        (SwizzleOp.prototype.get inSize):
-        (SwizzleOp.prototype.toString):
-        (generateSwizzles):
-        (_generateSwizzle): Deleted.
-        * WebGPUShadingLanguageRI/Type.js: Addition of isEnum property, as
-        above.
-        (Type.prototype.get isEnum):
-
-2018-07-24  Ryan Haddad  <ryanhaddad@apple.com>
-
-        Unreviewed, rolling out r234121.
-
-        Caused perf test failures.
-
-        Reverted changeset:
-
-        "We should cache the compiled sandbox profile in a data vault"
-        https://bugs.webkit.org/show_bug.cgi?id=184991
-        https://trac.webkit.org/changeset/234121
-
-2018-07-24  Thibault Saunier  <tsaunier@igalia.com>
-
-        [Flatpak] Add openh264 as needed by WebRTC
-        https://bugs.webkit.org/show_bug.cgi?id=187956
-
-        Reviewed by Philippe Normand.
-
-        * flatpak/org.webkit.WebKit.yaml:
-
-2018-07-24  Ross Kirsling  <ross.kirsling@sony.com>
-
-        WebCore::URL::hostIsIPAddress needs a Windows implementation
-        https://bugs.webkit.org/show_bug.cgi?id=187859
-
-        Reviewed by Fujii Hironori.
-
-        * TestWebKitAPI/Tests/WebCore/URL.cpp:
-        Check for a few other ways an IP address could be invalid.
-
-2018-07-24  Charlie Turner  <cturner@igalia.com>
-
-        [Flatpak] Avoid consuming unknown arguments in flatpak wrapper
-        https://bugs.webkit.org/show_bug.cgi?id=187951
-
-        When running run-webkit-tests --wpe --debug, it was observed that
-        both self.platform != 'wpe' and self.debug != True. This turned
-        out to be because of the "args" argument consuming the remainder
-        of future arguments that came after it. Since the first argument
-        to load_from_args was the test script to run, it would end up
-        never processing the other arguments supplied to the wrapper
-        script.
-
-        Tested this by rebuilding GTK in debug mode, running the
-        minibrowser with various extra options, and running some tests
-        with various extra options.
-
-        Reviewed by Michael Catanzaro.
-
-        * flatpak/flatpakutils.py:
-        (WebkitFlatpak.load_from_args):
-
-2018-07-24  Thibault Saunier  <tsaunier@igalia.com>
-
-        [Flatpak] Remove now unnecessary 'hacks' for running tests
-        https://bugs.webkit.org/show_bug.cgi?id=187608
-
-        Reviewed by Philippe Normand.
-
-        We used to need to not share X11 with the sandbox when running tests
-        but it is not needed anymore as we do not mount `/tmp` on `/tmp` in the
-        sandbox (we mount it on `/run/host/tmp` now).
-
-        This has the advantage of making running the MiniBrowser to display test
-        results working again.
-
-        * flatpak/flatpakutils.py:
-        (WebkitFlatpak.clean_args):
-        (WebkitFlatpak.run_in_sandbox):
-        (WebkitFlatpak._cleanup_faltpak_args_for_tests_if_needed): Deleted.
-
-2018-07-23  Fujii Hironori  <Hironori.Fujii@sony.com>
-
-        webkitpy.port.server_process_unittest.TestServerProcess.test_basic failed on Windows Python
-        https://bugs.webkit.org/show_bug.cgi?id=187581
-
-        Reviewed by Daniel Bates.
-
-        There are two failures in this test case:
-        1. proc.poll() doesn't return 0.
-        2. stderr is not output.
-
-        For failure #1, this is expected. the process should not exit at
-        the time. proc.poll() should return None because the process is
-        still alive.
-
-        This change added a new test to check proc.poll() becomes 0 after
-        the process successfully exits.
-
-        For failure #2, stderr is not flushed even though stdout is
-        flushed. This change uses '-u' command switch to force stdin,
-        stdout and stderr to be totally unbuffered.
-
-        * Scripts/webkitpy/port/server_process_unittest.py:
-        (TestServerProcess.test_basic): Added -u command switch. Do not
-        flush stdout. Removed the special condition for Windows. Add a new
-        test to check proc.poll() returns 0.
-        (TestServerProcess.test_process_crashing): Added -u command
-        switch. Do not flush stdout.
-
-2018-07-23  Ross Kirsling  <ross.kirsling@sony.com>
-
-        WTF::StringView::split should have an allowEmptyEntries flag
-        https://bugs.webkit.org/show_bug.cgi?id=187864
-
-        Reviewed by Konstantin Tokarev.
-
-        * TestWebKitAPI/Tests/WTF/StringView.cpp: Add test.
-
-2018-07-23  Ben Richards  <benton_richards@apple.com>
-
-        We should cache the compiled sandbox profile in a data vault
-        https://bugs.webkit.org/show_bug.cgi?id=184991
-
-        Reviewed by Ryosuke Niwa.
-
-        Added trace points for sandbox initialization
-
-        * Tracing/SystemTracePoints.plist:
-
-2018-07-23  Zalan Bujtas  <zalan@apple.com>
-
-        DRT relies on [webView display] to call -viewWillDraw on the WebHTMLView
-        https://bugs.webkit.org/show_bug.cgi?id=187924
-        <rdar://problem/41745884>
-
-        Reviewed by Simon Fraser.
-
-        WebKit relies on -viewWillDraw getting called on WebHTMLView, even though that view has not been marked as needing display.
-
-        * DumpRenderTree/mac/PixelDumpSupportMac.mm:
-        (createBitmapContextFromWebView):
-
-2018-07-23  Chris Dumez  <cdumez@apple.com>
-
-        WebResourceLoadStatisticsStore fails to unregister itself as a MessageReceiver in its destructor
-        https://bugs.webkit.org/show_bug.cgi?id=187910
-        <rdar://problem/42356526>
-
-        Reviewed by Brent Fulgham.
-
-        Add API test coverage.
-
-        * TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj:
-        * TestWebKitAPI/Tests/WebKitCocoa/ResourceLoadStatistics.mm:
-        (-[DisableITPDuringNavigationDelegate webView:didCommitNavigation:]):
-        (-[DisableITPDuringNavigationDelegate webView:didFinishNavigation:]):
-        (TEST):
-        * TestWebKitAPI/Tests/WebKitCocoa/notify-resourceLoadObserver.html: Added.
-
-2018-07-23  Daniel Bates  <dabates@apple.com>
-
-        test-webkitpy should take configuration command line options
-        https://bugs.webkit.org/show_bug.cgi?id=187872
-
-        Reviewed by David Kilzer.
-
-        Add --debug and --release optional command line options and teach the test bots to invoke test-webkitpy
-        with the appropriate configuration. At the moment these options are only meaningful on Mac since they
-        effect the configuration lldbWebKitTester is built with and we only build lldbWebKitTester on Mac.
-
-        * BuildSlaveSupport/build.webkit.org-config/steps.py:
-        (RunPythonTests): Pass the configuration to test-webkitpy. On Mac this will effect
-        how lldbWebKitTester is built. All other ports do not make use of the specified configuration.
-        * Scripts/webkitpy/test/main.py:
-        (_build_lldb_webkit_tester): Modified to take the configuration to build lldbWebKitTester. If
-        the specified configuration is None then we fall back to using the default configuration (as set
-        by script set-webkit-configuration).
-        (Tester._parse_args): Add command line options --debug and --release to build lldbWebKitTester
-        with a Debug configuration and a Release configuration, respectively
-        (Tester._run_tests): Pass the configuration to build lldbWebKitTester that we parsed from
-        the command line options. It may be None if neither --debug nor --release were passed.
-
-2018-07-23  Commit Queue  <commit-queue@webkit.org>
-
-        Unreviewed, rolling out r233030.
-        https://bugs.webkit.org/show_bug.cgi?id=187904
-
-        Broke tarball builds (Requested by mcatanzaro on #webkit).
-
-        Reverted changeset:
-
-        "run-gtk-tests (glib/common.py) cannot determine build
-        directory when webKitBranchBuild=true"
-        https://bugs.webkit.org/show_bug.cgi?id=185643
-        https://trac.webkit.org/changeset/233030
-
-2018-07-21  Simon Fraser  <simon.fraser@apple.com>
-
-        Fix lldb summarizers for HashMaps and HashSets
-        https://bugs.webkit.org/show_bug.cgi?id=187883
-
-        Reviewed by Dan Bates.
-        
-        HashMap and HashSet summarizers were broken in various ways.
-        
-        The 'type summary add -x' regexp needs to match at the start and end of the string
-        to avoid finding types nested inside other templates.
-        
-        The HashTable synthetic child code incorrectly used the type of the key, rather than
-        the value when synthesizing children.
-        
-        Add HashMap and HashSet summarizers.
-
-        I tried to get synthetic children working directly for HashMap and HashSet, but was unable.
-
-        * lldb/lldbWebKitTester/main.cpp:
-        (testSummaryProviders):
-        * lldb/lldb_webkit.py:
-        (__lldb_init_module):
-        (__lldb_init_module.lldb_webkit):
-        (WTFHashMap_SummaryProvider):
-        (WTFHashSet_SummaryProvider):
-        (WebCoreColorProvider._to_string_extended):
-        (WebCoreURLProvider.to_string):
-        (WTFHashMapProvider):
-        (WTFHashMapProvider.__init__):
-        (WTFHashMapProvider.tableSize):
-        (WTFHashMapProvider.keyCount):
-        (WTFHashSetProvider):
-        (WTFHashSetProvider.__init__):
-        (WTFHashSetProvider.tableSize):
-        (WTFHashSetProvider.keyCount):
-        (WTFHashTableProvider.tableSize):
-        (WTFHashTableProvider):
-        (WTFHashTableProvider.keyCount):
-        (WTFHashTableProvider.get_child_at_index):
-        (WTFHashTableProvider.update):
-        * lldb/lldb_webkit_unittest.py:
-        (TestSummaryProviders.serial_test_WTFVectorProvider_empty_vector):
-        (TestSummaryProviders.serial_test_WTFVectorProvider_vector_size_and_capacity):
-        (TestSummaryProviders):
-        (TestSummaryProviders.serial_test_WTFHashMap_tablesize_and_size):
-        (TestSummaryProviders.serial_test_WTFHashMap_of_vectors_tablesize_and_size):
-        (TestSummaryProviders.serial_test_WTFHashSet_tablesize_and_size):
-
-2018-07-21  Simon Fraser  <simon.fraser@apple.com>
-
-        Add lldb formatters for WebCore geometry types
-        https://bugs.webkit.org/show_bug.cgi?id=187884
-
-        Reviewed by Daniel Bates.
-        
-        Add lldb formatters for IntPoint, IntSize, IntRect, FloatPoint, FloatSize, FloatRect
-        and LayoutRect.
-        
-        Testing is blocked by webkit.org/b/187872.
-
-        * lldb/lldbWebKitTester/main.cpp:
-        (testSummaryProviders):
-        * lldb/lldb_webkit.py:
-        (__lldb_init_module.lldb_webkit):
-        (__lldb_init_module):
-        (WebCoreLayoutRect_SummaryProvider):
-        (WebCoreIntSize_SummaryProvider):
-        (WebCoreIntPoint_SummaryProvider):
-        (WebCoreFloatSize_SummaryProvider):
-        (WebCoreFloatPoint_SummaryProvider):
-        (WebCoreIntRect_SummaryProvider):
-        (WebCoreFloatRect_SummaryProvider):
-        (WebCoreLayoutRectProvider):
-        (WebCoreLayoutRectProvider.__init__):
-        (WebCoreLayoutRectProvider.get_x):
-        (WebCoreLayoutRectProvider.get_y):
-        (WebCoreLayoutRectProvider.get_width):
-        (WebCoreLayoutRectProvider.get_height):
-        (WebCoreIntPointProvider):
-        (WebCoreIntPointProvider.__init__):
-        (WebCoreIntPointProvider.get_x):
-        (WebCoreIntPointProvider.get_y):
-        (WebCoreIntSizeProvider):
-        (WebCoreIntSizeProvider.__init__):
-        (WebCoreIntSizeProvider.get_width):
-        (WebCoreIntSizeProvider.get_height):
-        (WebCoreIntRectProvider):
-        (WebCoreIntRectProvider.__init__):
-        (WebCoreIntRectProvider.get_x):
-        (WebCoreIntRectProvider.get_y):
-        (WebCoreIntRectProvider.get_width):
-        (WebCoreIntRectProvider.get_height):
-        (WebCoreFloatPointProvider):
-        (WebCoreFloatPointProvider.__init__):
-        (WebCoreFloatPointProvider.get_x):
-        (WebCoreFloatPointProvider.get_y):
-        (WebCoreFloatSizeProvider):
-        (WebCoreFloatSizeProvider.__init__):
-        (WebCoreFloatSizeProvider.get_width):
-        (WebCoreFloatSizeProvider.get_height):
-        (WebCoreFloatRectProvider):
-        (WebCoreFloatRectProvider.__init__):
-        (WebCoreFloatRectProvider.get_x):
-        (WebCoreFloatRectProvider.get_y):
-        (WebCoreFloatRectProvider.get_width):
-        (WebCoreFloatRectProvider.get_height):
-
-2018-07-20  John Wilander  <wilander@apple.com>
-
-        Resource Load Statistics: Enable basic functionality in experimental debug mode
-        https://bugs.webkit.org/show_bug.cgi?id=187835
-        <rdar://problem/42408590>
-
-        Reviewed by Chris Dumez.
-
-        This patch makes the experimental ITP Debug Mode feature work, at least
-        to a basic level. All of the changes to TestRunner are to support the
-        layout tests for the functionality.
-
-        * WebKitTestRunner/InjectedBundle/Bindings/TestRunner.idl:
-        * WebKitTestRunner/InjectedBundle/InjectedBundle.cpp:
-        (WTR::InjectedBundle::didReceiveMessageToPage):
-        * WebKitTestRunner/InjectedBundle/TestRunner.cpp:
-        (WTR::TestRunner::setStatisticsDebugMode):
-        (WTR::TestRunner::statisticsCallDidSetDebugModeCallback):
-        (WTR::TestRunner::setStatisticsPrevalentResourceForDebugMode):
-        (WTR::TestRunner::statisticsCallDidSetPrevalentResourceForDebugModeCallback):
-        * WebKitTestRunner/InjectedBundle/TestRunner.h:
-        * WebKitTestRunner/TestController.cpp:
-        (WTR::TestController::setStatisticsDebugMode):
-        (WTR::TestController::setStatisticsPrevalentResourceForDebugMode):
-        * WebKitTestRunner/TestController.h:
-        * WebKitTestRunner/TestInvocation.cpp:
-        (WTR::TestInvocation::didReceiveSynchronousMessageFromInjectedBundle):
-        (WTR::TestInvocation::didSetStatisticsDebugMode):
-        (WTR::TestInvocation::didSetPrevalentResourceForDebugMode):
-        * WebKitTestRunner/TestInvocation.h:
-
-2018-07-20  Ryosuke Niwa  <rniwa@apple.com>
-
-        Picking a color from the color panel for typing attributes needs to inverse transform through color-filter
-        https://bugs.webkit.org/show_bug.cgi?id=187846
-
-        Reviewed by Simon Fraser.
-
-        Fixed the bug that testRunner.execCommand was using the second argument as the value.
-
-        The second argument, aShowDefaultUI, should always be ignored in testRunner.execCommand,
-        and the third argument should be used as the value. DumpRenderTree's implementation does this already.
-
-        * WebKitTestRunner/InjectedBundle/Bindings/TestRunner.idl:
-        * WebKitTestRunner/InjectedBundle/TestRunner.cpp:
-        (WTR::TestRunner::execCommand):
-        * WebKitTestRunner/InjectedBundle/TestRunner.h:
-
-2018-07-20  Dean Jackson  <dino@apple.com>
-
-        Accidentally checked in the wrong version of these files.
-
-        * lldb/lldbWebKitTester/lldbWebKitTester.xcodeproj/project.pbxproj:
-        * lldb/lldbWebKitTester/main.cpp:
-        (testSummaryProviders):
-        * lldb/lldb_webkit_unittest.py:
-        (TestSummaryProviders.serial_test_WTFVectorProvider_vector_size_and_capacity):
-        (TestSummaryProviders.serial_test_WebCoreColorProvider_invalid_color): Deleted.
-        (TestSummaryProviders.serial_test_WebCoreColorProvider_extended_color): Deleted.
-        (TestSummaryProviders.serial_test_WebCoreColorProvider_rgb_color): Deleted.
-        (TestSummaryProviders.serial_test_WebCoreColorProvider_rgba_color): Deleted.
-        (TestSummaryProviders.serial_test_WebCoreColorProvider_semantic_color): Deleted.
-
-2018-07-20  Dean Jackson  <dino@apple.com>
-
-        Provide an lldb type summary for WebCore::Color
-        https://bugs.webkit.org/show_bug.cgi?id=187776
-
-        Reviewed by Dan Bates.
-
-        This time without the tests because we don't want to build
-        all of WebKit just for python tests.
-
-        * lldb/lldb_webkit.py: Add a Color summary provider.
-        (__lldb_init_module):
-        (WebCoreColor_SummaryProvider):
-        (WebCoreColorProvider):
-        (WebCoreColorProvider.__init__):
-        (WebCoreColorProvider._is_extended):
-        (WebCoreColorProvider._is_valid):
-        (WebCoreColorProvider._is_semantic):
-        (WebCoreColorProvider._to_string_extended):
-        (WebCoreColorProvider.to_string):
-
-2018-07-20  Brady Eidson  <beidson@apple.com>
-
-        Add WKNavigation/WKNavigationAction related SPI.
-        https://bugs.webkit.org/show_bug.cgi?id=187826
-
-        Reviewed by Chris Dumez.
-
-        * TestWebKitAPI/Tests/WebKitCocoa/Navigation.mm:
-        (-[NavigationActionHasNavigationDelegate webView:didFinishNavigation:]):
-        (-[NavigationActionHasNavigationDelegate webView:decidePolicyForNavigationAction:decisionHandler:]):
-        (TEST):
-
-2018-07-20  Commit Queue  <commit-queue@webkit.org>
-
-        Unreviewed, rolling out r233943, r233988, r234004, and
-        r234012.
-        https://bugs.webkit.org/show_bug.cgi?id=187856
-
-        webkitpy tests are failing to build lldbWebKitTester
-        (Requested by ryanhaddad on #webkit).
-
-        Reverted changesets:
-
-        "Provide an lldb type summary for WebCore::Color"
-        https://bugs.webkit.org/show_bug.cgi?id=187776
-        https://trac.webkit.org/changeset/233943
-
-        "Provide an lldb type summary for WebCore::Color"
-        https://bugs.webkit.org/show_bug.cgi?id=187776
-        https://trac.webkit.org/changeset/233988
-
-        "Provide an lldb type summary for WebCore::Color"
-        https://bugs.webkit.org/show_bug.cgi?id=187776
-        https://trac.webkit.org/changeset/234004
-
-        "Provide an lldb type summary for WebCore::Color"
-        https://bugs.webkit.org/show_bug.cgi?id=187776
-        https://trac.webkit.org/changeset/234012
-
-2018-07-20  Philippe Normand  <pnormand@igalia.com>
-
-        [GTK][WPE] enable-media-capabilities websetting
-        https://bugs.webkit.org/show_bug.cgi?id=187847
-
-        Reviewed by Carlos Garcia Campos.
-
-        * TestWebKitAPI/Tests/WebKitGLib/TestWebKitSettings.cpp:
-        (testWebKitSettings): Unit test for the new enable-media-capabilities setting.
-
-2018-07-19  Carlos Garcia Campos  <cgarcia@igalia.com>
-
-        [GLIB] jsc_context_evaluate_in_object() should receive an instance when a JSCClass is given
-        https://bugs.webkit.org/show_bug.cgi?id=187798
-
-        Reviewed by Michael Catanzaro.
-
-        Update test cases to the new API and use a destroy function and vtable in the test case of calling
-        jsc_context_evaluate_in_object() with a JSCClass.
-
-        * TestWebKitAPI/Tests/JavaScriptCore/glib/TestJSC.cpp:
-        (testJSCEvaluateInObject):
-
-2018-07-19  Ross Kirsling  <ross.kirsling@sony.com>
-
-        TestWTF.WTF.StringConcatenate_Unsigned fails for unsigned short on Windows
-        https://bugs.webkit.org/show_bug.cgi?id=187712
-
-        Reviewed by Fujii Hironori.
-
-        * TestWebKitAPI/Tests/WTF/StringConcatenate.cpp:
-        Mark result of unsigned short test case as platform-specific,
-        since Windows' behavior is actually *less* surprising here.
-
-2018-07-19  Lucas Forschler  <lforschler@apple.com>
-
-        Teach the AWS Lambda to use the [minified]-platforms database
-        https://bugs.webkit.org/show_bug.cgi?id=187817
-        
-        Reviewed by David Kilzer.
-
-        * WebKitArchiveSupport/lambda/register-archive-in-dynamodb.py:
-        (lambda_handler): Add support for the platforms.webkit.org database
-        * WebKitArchiveSupport/lambda/register-minified-s3-archive-in-dynamodb.py:
-        (lambda_handler): Add support for the minified-platforms.webkit.org database
-
-2018-07-19  Dean Jackson  <dino@apple.com>
-
-        Provide an lldb type summary for WebCore::Color
-        https://bugs.webkit.org/show_bug.cgi?id=187776
-
-        Dan Bates helped me with this attempted fix. Fingers crossed!
-
-        * Scripts/build-lldbwebkittester:
-
-2018-07-19  Stephan Szabo  <stephan.szabo@sony.com>
-
-        [WinCairo] Support DEVELOPER_MODE for allowing inspection of web inspector
-        https://bugs.webkit.org/show_bug.cgi?id=187786
-
-        Reviewed by Fujii Hironori.
-
-        * Scripts/webkitdirs.pm:
-        (generateBuildSystemFromCMakeProject): Turn on DEVELOPER_MODE
-        for WinCairo builds.
-
-2018-07-19  Dean Jackson  <dino@apple.com>
-
-        Provide an lldb type summary for WebCore::Color
-        https://bugs.webkit.org/show_bug.cgi?id=187776
-
-        Another attempted build fix for Debug.
-
-        * lldb/lldbWebKitTester/Configurations/Base.xcconfig:
-        * lldb/lldbWebKitTester/lldbWebKitTester.xcodeproj/project.pbxproj:
-
-2018-07-19  Thomas Denney  <tdenney@apple.com>
-
-        [WHLSL] The interpreter doesn't support boolean short-circuiting
-        https://bugs.webkit.org/show_bug.cgi?id=187779
-
-        Reviewed by Alex Christensen.
-
-        * WebGPUShadingLanguageRI/Evaluator.js:
-        (Evaluator.prototype.visitLogicalExpression): RHS is only evaluated when necessary
-        * WebGPUShadingLanguageRI/Test.js:
-        (tests.booleanShortcircuiting): Adds 4 tests for the evaluation of logical expresions
-
-2018-07-19  Dean Jackson  <dino@apple.com>
-
-        Provide an lldb type summary for WebCore::Color
-        https://bugs.webkit.org/show_bug.cgi?id=187776
-
-        Attempted build fix.
-
-        * lldb/lldbWebKitTester/Configurations/Base.xcconfig:
-
-2018-07-19  Brady Eidson  <beidson@apple.com>
-
-        Add an SPI policy action to allow clients to explicitly ask for a new process on a navigation.
-        https://bugs.webkit.org/show_bug.cgi?id=187789
-
-        Reviewed by Andy Estes.
-
-        * TestWebKitAPI/Tests/WebKitCocoa/ProcessSwapOnNavigation.mm:
-        (-[PSONNavigationDelegate init]):
-        (-[PSONNavigationDelegate webView:decidePolicyForNavigationAction:decisionHandler:]):
-
-2018-07-19  Thibault Saunier  <tsaunier@igalia.com>
-
-        [Flatpak] Let flatpak process handle SIGINT
-        https://bugs.webkit.org/show_bug.cgi?id=187521
-
-        Reviewed by Philippe Normand.
-
-        Ensuring that flatpak process ends properly and that the sandbox is teard down.
-        It also avoids showing usless backtrace to the end user and makes gdb
-        much more usable.
-
-        Also make flatpakutils not verbose by default
-
-        * flatpak/flatpakutils.py:
-        (disable_signals):
-        (WebkitFlatpak.run_in_sandbox):
-        (WebkitFlatpak.run_gdb):
-
-2018-07-18  Stephan Szabo  <stephan.szabo@sony.com>
-
-        Unreviewed. Fix contributors entry.
-
-        * Scripts/webkitpy/common/config/contributors.json:
-
-2018-07-18  Dean Jackson  <dino@apple.com>
-
-        Provide an lldb type summary for WebCore::Color
-        https://bugs.webkit.org/show_bug.cgi?id=187776
-
-        Reviewed by Dan Bates.
-
-        * lldb/lldbWebKitTester/lldbWebKitTester.xcodeproj/project.pbxproj: Link against WebKit
-            to get to WebCore.
-        * lldb/lldbWebKitTester/main.cpp:
-        (testSummaryProviders): Create some Color objects for testing.
-
-        * lldb/lldb_webkit.py: Add a Color summary provider.
-        (__lldb_init_module):
-        (WebCoreColor_SummaryProvider):
-        (WebCoreColorProvider):
-        (WebCoreColorProvider.__init__):
-        (WebCoreColorProvider._is_extended):
-        (WebCoreColorProvider._is_valid):
-        (WebCoreColorProvider._is_semantic):
-        (WebCoreColorProvider._to_string_extended):
-        (WebCoreColorProvider.to_string):
-
-        * lldb/lldb_webkit_unittest.py: Tests.
-        (TestSummaryProviders.serial_test_WTFVectorProvider_vector_size_and_capacity):
-        (TestSummaryProviders):
-        (TestSummaryProviders.serial_test_WebCoreColorProvider_invalid_color):
-        (TestSummaryProviders.serial_test_WebCoreColorProvider_extended_color):
-        (TestSummaryProviders.serial_test_WebCoreColorProvider_rgb_color):
-        (TestSummaryProviders.serial_test_WebCoreColorProvider_rgba_color):
-
-2018-07-18  Truitt Savell  <tsavell@apple.com>
-
-        Unreviewed, rolling out r233934.
-
-        Revision caused 5 webkitpy failures on Mac and broke webkitpy
-        testing on iOS simulator
-
-        Reverted changeset:
-
-        "Provide an lldb type summary for WebCore::Color"
-        https://bugs.webkit.org/show_bug.cgi?id=187776
-        https://trac.webkit.org/changeset/233934
-
-2018-07-18  Chris Dumez  <cdumez@apple.com>
-
-        REGRESSION (r233897): CrashTracer: com.apple.WebKit.WebContent.Development at com.apple.WebKit: WKBooleanGetValue + 9
-        https://bugs.webkit.org/show_bug.cgi?id=187784
-        <rdar://problem/42329230>
-
-        Reviewed by Brady Eidson.
-
-        When process swapping on navigation, WebPageProxy::reattachToWebProcess() unregisters the page as a MessageReceiver
-        from the old WebProcessProxy and registers itself as a MessageReceiver for the new WebProcessProxy instead. This
-        means that after this point, IPC sent by the previous WebProcess to its WebPageProxy will fail.
-
-        When we process swap, we also navigate the page in the old WebProcess to about:blank, when the navigation to
-        about:blank would complete, the WebKitTestRunner's injected bundle would try and send IPC to the WebPageProxy
-        which would fail and would cause WKBundlePagePostSynchronousMessageForTesting() to return a null result. WKTR
-        would crash when dereferencing this null result. This patch addresses this by dealing with the potential null
-        result.
-
-        * WebKitTestRunner/InjectedBundle/InjectedBundle.cpp:
-        (WTR::InjectedBundle::shouldProcessWorkQueue const):
-
-2018-07-18  Dean Jackson  <dino@apple.com>
-
-        Provide an lldb type summary for WebCore::Color
-        https://bugs.webkit.org/show_bug.cgi?id=187776
-
-        Reviewed by Dan Bates.
-
-        * lldb/lldbWebKitTester/lldbWebKitTester.xcodeproj/project.pbxproj: Link against WebKit
-            to get to WebCore.
-        * lldb/lldbWebKitTester/main.cpp:
-        (testSummaryProviders): Create some Color objects for testing.
-
-        * lldb/lldb_webkit.py: Add a Color summary provider.
-        (__lldb_init_module):
-        (WebCoreColor_SummaryProvider):
-        (WebCoreColorProvider):
-        (WebCoreColorProvider.__init__):
-        (WebCoreColorProvider._is_extended):
-        (WebCoreColorProvider._is_valid):
-        (WebCoreColorProvider._is_semantic):
-        (WebCoreColorProvider._to_string_extended):
-        (WebCoreColorProvider.to_string):
-
-        * lldb/lldb_webkit_unittest.py: Tests.
-        (TestSummaryProviders.serial_test_WTFVectorProvider_vector_size_and_capacity):
-        (TestSummaryProviders):
-        (TestSummaryProviders.serial_test_WebCoreColorProvider_invalid_color):
-        (TestSummaryProviders.serial_test_WebCoreColorProvider_extended_color):
-        (TestSummaryProviders.serial_test_WebCoreColorProvider_rgb_color):
-        (TestSummaryProviders.serial_test_WebCoreColorProvider_rgba_color):
-
-2018-07-18  Stephan Szabo  <stephan.szabo@sony.com>
-
-        Adding myself to contributors.json
-
-        Unreviewed contributors.json change.
-
-        * Scripts/webkitpy/common/config/contributors.json:
-
-2018-07-18  Jer Noble  <jer.noble@apple.com>
-
-        -_beginAnimatedResizeWithUpdates: can leave view in bad state if called during an existing animation
-        https://bugs.webkit.org/show_bug.cgi?id=187739
-        <rdar://problem/42304518>
-
-        Reviewed by Tim Horton.
-
-        * TestWebKitAPI/Tests/WebKitCocoa/AnimatedResize.mm:
-        (TEST):
-
-2018-07-18  Jer Noble  <jer.noble@apple.com>
-
-        PiP from Element Fullscreen should match AVKit's behavior
-        https://bugs.webkit.org/show_bug.cgi?id=187623
-        <rdar://problem/41212379>
-
-        Reviewed by Jon Lee.
-
-        * TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj:
-        * TestWebKitAPI/Tests/WebKitCocoa/ExitFullscreenOnEnterPiP.html: Added.
-        * TestWebKitAPI/Tests/WebKitCocoa/ExitFullscreenOnEnterPiP.mm: Added.
-        (-[ExitFullscreenOnEnterPiPUIDelegate _webView:hasVideoInPictureInPictureDidChange:]):
-        (-[ExitFullscreenOnEnterPiPUIDelegate _webViewDidEnterFullscreen:]):
-        (-[ExitFullscreenOnEnterPiPUIDelegate _webViewDidExitFullscreen:]):
-        (TestWebKitAPI::TEST):
-
-2018-07-18  Aakash Jain  <aakash_jain@apple.com>
-
-        [ews-build] Add build step to run layout-test
-        https://bugs.webkit.org/show_bug.cgi?id=187674
-
-        Reviewed by Lucas Forschler.
-
-        * BuildSlaveSupport/ews-build/steps.py:
-        (RunWebKitTests): Added build step to run layout-tests. This is similar to current EWS.
-        * BuildSlaveSupport/ews-build/steps_unittest.py: Added unit-tests.
-        * BuildSlaveSupport/ews-build/factories.py: Added build-step to run layout-test in iOSSimulatorFactory.
-
-2018-07-18  Wenson Hsieh  <wenson_hsieh@apple.com>
-
-        Add SPI to defer running async script until after document load
-        https://bugs.webkit.org/show_bug.cgi?id=187748
-        <rdar://problem/42317378>
-
-        Reviewed by Ryosuke Niwa and Tim Horton.
-
-        Add an API test to verify that when the deferred async script configuration is set, async scripts will be
-        executed after the DOMContentLoaded event.
-
-        * TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj:
-        * TestWebKitAPI/Tests/WebKitCocoa/RunScriptAfterDocumentLoad.mm: Added.
-        (TEST):
-
-2018-07-18  Charlie Turner  <cturner@igalia.com>
-
-        [WPE] Update WPEBackend in flatpak
-        https://bugs.webkit.org/show_bug.cgi?id=187753
-
-        r233763 updated WPEBackend for the jhbuild, but not the flatpak. This
-        caused WPE_BACKEND_CHECK_VERSION to not be defined, triggering a bug
-        in the version guard,
-
-        #if defined(WPE_BACKEND_CHECK_VERSION) && WPE_BACKEND_CHECK_VERSION(0, 2, 0)
-
-        This doesn't work as intended, since the C preprocessor first expands
-        all macro arguments in the #if expression before evaluating the
-        expression. When WPE_BACKEND_CHECK_VERSION is not defined, this will
-        lead to a syntax error and some head scratching.
-
-        A future patch should address the buggy macro check, when it is
-        decided whether the assume the macro is always defined and bump
-        the API requirements on WPEBackend, or to move the check into two
-        conditionals to avoid the expansion problem.
-
-        Unreviewed build fix.
-        
-        * flatpak/org.webkit.WPE.yaml:
-
-2018-07-18  Simon Fraser  <simon.fraser@apple.com>
-
-        Fix the parsing of colors in attributed string tests, and make it possible to dump the typing attributes
-        https://bugs.webkit.org/show_bug.cgi?id=187747
-
-        Reviewed by Ryosuke Niwa.
-        
-        Add textInputController.attributedStringForTyping(), which returns a one-character
-        attributed string whose attributes are the typing attributes, making it possible to
-        test -[WebView typingAttributes].
-        
-        Sadly WebCore's convertObjcValueToValue() doesn't know how to convert NSDictionary,
-        so we can't return -typingAttributes directly.
-
-        * DumpRenderTree/mac/TextInputControllerMac.m:
-        (+[TextInputController isSelectorExcludedFromWebScript:]):
-        (-[TextInputController attributedStringForTyping]):
-
-2018-07-18  Carlos Garcia Campos  <cgarcia@igalia.com>
-
-        [GLIB] Add jsc_context_check_syntax() to GLib API
-        https://bugs.webkit.org/show_bug.cgi?id=187694
-
-        Reviewed by Yusuke Suzuki.
-
-        Add a new test case.
-
-        * TestWebKitAPI/Tests/JavaScriptCore/glib/TestJSC.cpp:
-        (testJSCCheckSyntax):
-        (main):
-
-2018-07-17  Sihui Liu  <sihui_liu@apple.com>
-
-        REGRESSION: [macOS Sierra] TestWebKitAPI.WebKit.WebsiteDataStoreCustomPaths is a flaky failure
-        https://bugs.webkit.org/show_bug.cgi?id=187066
-        <rdar://problem/41609065>
-
-        Reviewed by Chris Dumez.
-
-        In r232668 we started to do reload if web process crashes by default. As we killed the web 
-        process explicitly in this test, if web page got reloaded, messages would be sent again, 
-        and flag set in message handler could keep the test continue to evaluate the expectation 
-        without waiting for removeDataOfTypes to finish.
-
-        * TestWebKitAPI/Tests/WebKitCocoa/WebsiteDataStoreCustomPaths.mm:
-        (-[WebsiteDataStoreCustomPathsMessageHandler webViewWebContentProcessDidTerminate:]):
-        (TEST):
-
-2018-07-17  Chris Dumez  <cdumez@apple.com>
-
-        Turn on PSON in WebKitTestRunner
-        https://bugs.webkit.org/show_bug.cgi?id=186542
-
-        Reviewed by Brady Eidson.
-
-        Turn on PSON by default in WebKitTestRunner.
-
-        * WebKitTestRunner/TestOptions.h:
-
-2018-07-17  Wenson Hsieh  <wenson_hsieh@apple.com>
-
-        Add an SPI hook to allow clients to yield document parsing and script execution
-        https://bugs.webkit.org/show_bug.cgi?id=187682
-        <rdar://problem/42207453>
-
-        Reviewed by Ryosuke Niwa.
-
-        Add a few tests to exercise the new document yield token SPI, verifying that clients can use the SPI to defer
-        document load, and that doing so doesn't cause deferred `script` to execute in the wrong order (i.e. before
-        synchronous script, or after "DOMContentLoaded").
-
-        * TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj:
-        * TestWebKitAPI/Tests/WebKitCocoa/ParserYieldTokenPlugIn.mm: Added.
-        (-[ParserYieldTokenPlugIn takeDocumentParserTokenAfterCommittingLoad]):
-        (-[ParserYieldTokenPlugIn releaseDocumentParserToken]):
-        (-[ParserYieldTokenPlugIn webProcessPlugInBrowserContextController:didCommitLoadForFrame:]):
-        (-[ParserYieldTokenPlugIn webProcessPlugIn:didCreateBrowserContextController:]):
-        (-[ParserYieldTokenPlugIn webProcessPlugInBrowserContextController:didFinishDocumentLoadForFrame:]):
-        (-[ParserYieldTokenPlugIn webProcessPlugInBrowserContextController:didFinishLoadForFrame:]):
-
-        Add an injected bundle object that knows how to take and release multiple document parser yield tokens.
-
-        * TestWebKitAPI/Tests/WebKitCocoa/ParserYieldTokenTests.h: Added.
-        * TestWebKitAPI/Tests/WebKitCocoa/ParserYieldTokenTests.mm: Added.
-        (+[ParserYieldTokenTestWebView webView]):
-        (-[ParserYieldTokenTestWebView bundle]):
-        (-[ParserYieldTokenTestWebView schemeHandler]):
-        (-[ParserYieldTokenTestWebView didFinishDocumentLoad]):
-        (-[ParserYieldTokenTestWebView didFinishLoad]):
-        (waitForDelay):
-        (TEST):
-        * TestWebKitAPI/Tests/WebKitCocoa/TestURLSchemeHandler.h: Added.
-        * TestWebKitAPI/Tests/WebKitCocoa/TestURLSchemeHandler.mm: Added.
-        (-[TestURLSchemeHandler webView:startURLSchemeTask:]):
-        (-[TestURLSchemeHandler webView:stopURLSchemeTask:]):
-        (-[TestURLSchemeHandler setStartURLSchemeTaskHandler:]):
-        (-[TestURLSchemeHandler startURLSchemeTaskHandler]):
-        (-[TestURLSchemeHandler setStopURLSchemeTaskHandler:]):
-        (-[TestURLSchemeHandler stopURLSchemeTaskHandler]):
-
-        Add a new test helper class to handle custom schemes via a block-based API.
-
-        * TestWebKitAPI/Tests/WebKitCocoa/text-with-async-script.html: Added.
-
-        New test HTML page that contains a deferred script element, a synchronous script element, another deferred
-        script element, and then some text, images, and links.
-
-        * TestWebKitAPI/Tests/WebKitCocoa/text-with-deferred-script.html: Added.
-
-2018-07-17  Truitt Savell  <tsavell@apple.com>
-
-        Adding myself to Contributors.json
-
-        Unreviewed
-
-        * Scripts/webkitpy/common/config/contributors.json:
-
-2018-07-17  Aditya Keerthi  <akeerthi@apple.com>
-
-        [macOS] TestWebKitAPI.PictureInPicture.WKUIDelegate is timing out
-        https://bugs.webkit.org/show_bug.cgi?id=187730
-
-        Reviewed by Jer Noble.
-
-        This regression was introduced by r233865. PIP can now only be initiated from a
-        window that is on screen.
-
-        * TestWebKitAPI/Tests/WebKitCocoa/PictureInPictureDelegate.mm:
-        (TestWebKitAPI::TEST):
-
-2018-07-17  John Wilander  <wilander@apple.com>
-
-        Add completion handlers to TestRunner functions setStatisticsLastSeen(), setStatisticsPrevalentResource(), setStatisticsVeryPrevalentResource(), setStatisticsHasHadUserInteraction(), and setStatisticsHasHadNonRecentUserInteraction()
-        https://bugs.webkit.org/show_bug.cgi?id=187710
-        <rdar://problem/42252757>
-
-        Reviewed by Chris Dumez.
-
-        * WebKitTestRunner/InjectedBundle/Bindings/TestRunner.idl:
-        * WebKitTestRunner/InjectedBundle/InjectedBundle.cpp:
-        (WTR::InjectedBundle::didReceiveMessageToPage):
-        * WebKitTestRunner/InjectedBundle/TestRunner.cpp:
-        (WTR::TestRunner::setStatisticsLastSeen):
-        (WTR::TestRunner::statisticsCallDidSetLastSeenCallback):
-        (WTR::TestRunner::setStatisticsPrevalentResource):
-        (WTR::TestRunner::statisticsCallDidSetPrevalentResourceCallback):
-        (WTR::TestRunner::setStatisticsVeryPrevalentResource):
-        (WTR::TestRunner::statisticsCallDidSetVeryPrevalentResourceCallback):
-        (WTR::TestRunner::setStatisticsHasHadUserInteraction):
-        (WTR::TestRunner::setStatisticsHasHadNonRecentUserInteraction):
-        (WTR::TestRunner::statisticsCallDidSetHasHadUserInteractionCallback):
-        * WebKitTestRunner/InjectedBundle/TestRunner.h:
-        * WebKitTestRunner/TestController.cpp:
-        (WTR::TestController::setStatisticsLastSeen):
-        (WTR::TestController::setStatisticsPrevalentResource):
-        (WTR::TestController::setStatisticsVeryPrevalentResource):
-        (WTR::TestController::setStatisticsHasHadUserInteraction):
-        (WTR::TestController::setStatisticsHasHadNonRecentUserInteraction):
-        * WebKitTestRunner/TestInvocation.cpp:
-        (WTR::TestInvocation::didSetLastSeen):
-        (WTR::TestInvocation::didSetPrevalentResource):
-        (WTR::TestInvocation::didSetVeryPrevalentResource):
-        (WTR::TestInvocation::didSetHasHadUserInteraction):
-        (WTR::TestInvocation::didSetHasHadNonRecentUserInteraction):
-        * WebKitTestRunner/TestInvocation.h:
-
-2018-07-16  Jeremy Jones  <jeremyj@apple.com>
-
-        Fix API Test failures introduced by r233865
-        https://bugs.webkit.org/show_bug.cgi?id=187720
-
-        Unreviewed APITest fix after r233865.
-
-        Fullscreen can now only be initiated from a window that is on screen.
-
-        * TestWebKitAPI/Tests/WebKitCocoa/FullscreenDelegate.mm:
-        (TestWebKitAPI::TEST):
-        * TestWebKitAPI/Tests/WebKitCocoa/FullscreenTopContentInset.mm:
-        (TestWebKitAPI::TEST):
-
-2018-07-11  Dean Jackson  <dino@apple.com>
-
-        Allow removal of white backgrounds
-        https://bugs.webkit.org/show_bug.cgi?id=187574
-        <rdar://problem/41146792>
-
-        Reviewed by Simon Fraser.
-
-        Add a new menu item for punching out white backgrounds in MiniBrowser.
-        In WebKitTestRunner, expose the new setting and hook that up to
-        drawing a background in the WebView.
-
-        * MiniBrowser/mac/AppDelegate.m:
-        (defaultConfiguration): Add _punchOutWhiteBackgroundsInDarkMode.
-
-        * MiniBrowser/mac/SettingsController.h: Ditto.
-        * MiniBrowser/mac/SettingsController.m:
-        (-[SettingsController _populateMenu]):
-        (-[SettingsController validateMenuItem:]):
-        (-[SettingsController togglePunchOutWhiteBackgroundsInDarkMode:]):
-        (-[SettingsController punchOutWhiteBackgroundsInDarkMode]):
-
-        * MiniBrowser/mac/WK1BrowserWindowController.m:
-        (-[WK1BrowserWindowController didChangeSettings]): Set the new preference.
-
-        * WebKitTestRunner/PlatformWebView.h: Expose a drawsBackground property.
-        * WebKitTestRunner/gtk/PlatformWebViewGtk.cpp: Null implementation.
-        (WTR::PlatformWebView::drawsBackground const):
-        (WTR::PlatformWebView::setDrawsBackground):
-        * WebKitTestRunner/wpe/PlatformWebViewWPE.cpp: Ditto.
-        (WTR::PlatformWebView::drawsBackground const):
-        (WTR::PlatformWebView::setDrawsBackground):
-        * WebKitTestRunner/ios/PlatformWebViewIOS.mm: Call into the WKWebView and
-        set its SPI.
-        (WTR::PlatformWebView::drawsBackground const):
-        (WTR::PlatformWebView::setDrawsBackground):
-        * WebKitTestRunner/mac/PlatformWebViewMac.mm: Ditto.
-        (WTR::PlatformWebView::drawsBackground const):
-        (WTR::PlatformWebView::setDrawsBackground):
-
-        * WebKitTestRunner/TestController.cpp: Reset and copy the new preference.
-        (WTR::TestController::resetPreferencesToConsistentValues):
-        (WTR::updateTestOptionsFromTestHeader):
-        * WebKitTestRunner/TestOptions.h:
-        (WTR::TestOptions::hasSameInitializationOptions const):
-
-        * WebKitTestRunner/cocoa/TestControllerCocoa.mm:
-        (WTR::TestController::platformCreateWebView): If the option for punching
-        out the background was set, tell the WebView to not draw its background.
-
-2018-07-16  Aditya Keerthi  <akeerthi@apple.com>
-
-        [Datalist][macOS] Add suggestions UI for TextFieldInputTypes
-        https://bugs.webkit.org/show_bug.cgi?id=186531
-
-        Reviewed by Tim Horton.
-
-        Added isShowingDatalistSuggestions testing hook in order to enable testing of the
-        visibility of the suggestions view.
-
-        * DumpRenderTree/mac/UIScriptControllerMac.mm:
-        (WTR::UIScriptController::isShowingDataListSuggestions const):
-        * TestRunnerShared/UIScriptContext/Bindings/UIScriptController.idl:
-        * TestRunnerShared/UIScriptContext/UIScriptController.cpp:
-        (WTR::UIScriptController::isShowingDataListSuggestions const):
-        * TestRunnerShared/UIScriptContext/UIScriptController.h:
-        * WebKitTestRunner/mac/UIScriptControllerMac.mm:
-        (WTR::UIScriptController::isShowingDataListSuggestions const):
-
-2018-07-16  Yusuke Suzuki  <utatane.tea@gmail.com>
-
-        Add --target-path option to dump-class-layout
-        https://bugs.webkit.org/show_bug.cgi?id=187687
-
-        Reviewed by Simon Fraser.
-
-        We add an escape hatch to dump-class-layout for specifying target path directly.
-        This `--target-path` allows us to use dump-class-layout in the other ports
-        like JSCOnly.
-
-        We can dump class layout if we build the target with clang by using the following command.
-
-        Tools/Scripts/dump-class-layout \
-            --architecture=x86_64 \
-            --target-path=path/to/libJavaScriptCore.so \
-            JavaScriptCore \
-            ScopeNode
-
-        * Scripts/dump-class-layout:
-        (main):
-
-2018-07-16  Sihui Liu  <sihui_liu@apple.com>
-
-        IndexedDB: closeAndDeleteDatabasesForOrigins should remove all databases for those origins
-        https://bugs.webkit.org/show_bug.cgi?id=187631
-        <rdar://problem/42164227>
-
-        Reviewed by Brady Eidson.
-
-        * TestWebKitAPI/Tests/WebKitCocoa/WebsiteDataStoreCustomPaths.mm:
-        (TEST):
-
-2018-07-15  Carlos Garcia Campos  <cgarcia@igalia.com>
-
-        [GLIB] Add API to evaluate code using a given object to store global symbols
-        https://bugs.webkit.org/show_bug.cgi?id=187639
-
-        Reviewed by Michael Catanzaro.
-
-        Add a new test case.
-
-        * TestWebKitAPI/Tests/JavaScriptCore/glib/TestJSC.cpp:
-        (testJSCEvaluateInObject):
-        (testJSCExceptions):
-        (main):
-
-2018-07-15  Zan Dobersek  <zdobersek@igalia.com>
-
-        [webkitpy] run-web-platform-tests should allow specifying custom WPT metadata directories
-        https://bugs.webkit.org/show_bug.cgi?id=187353
-
-        Reviewed by Youenn Fablet.
-
-        When using run-web-platform-tests, allow specifying custom WPT metadata
-        directory. This will avoid generating such metadata-filled directory
-        from the port-specific JSON expectations file, and instead will search
-        for the metadata files and the include manifest inside the specified
-        directory. The MANIFEST.json file will also be generated under there.
-
-        Rationale for this change is prototyping using custom metadata
-        directories and avoiding generating all this content from a JSON file.
-        Using this by default in the future would avoid sporradic changes in how
-        WPT handles metadata .ini files and releases us from the obligation of
-        adjusting the generator for every such change. This would also allow
-        managing this metadata content in a separate repository, avoiding
-        polluting the main webkit.org repository with thousands of .ini files.
-
-        If this turns out to be the preferable way of managing the metadata
-        content, the JSON files under WebPlatformTests/ and the related code in
-        wpt_runner.py would be removed.
-
-        * Scripts/webkitpy/w3c/wpt_runner.py:
-        (parse_args):
-        (WPTRunner._wpt_run_paths):
-        (WPTRunner.run):
-        * Scripts/webkitpy/w3c/wpt_runner_unittest.py:
-        (WPTRunnerTest.test_run_with_specified_options):
-
-2018-07-13  Ross Kirsling  <ross.kirsling@sony.com>
-
-        [WinCairo] run-api-tests is timing out for almost all test cases
-        https://bugs.webkit.org/show_bug.cgi?id=187547
-
-        Reviewed by Fujii Hironori.
-
-        * Scripts/webkitpy/port/server_process.py:
-        (ServerProcess._wait_for_data_and_update_buffers_using_win32_apis):
-        Still attempt to read from stdout/stderr even if the process has exited, as long as it's before the deadline.
-
-        * Scripts/webkitpy/port/server_process_unittest.py:
-        (TestServerProcess.test_read_after_process_exits):
-        Add test case.
-
-2018-07-13  Ryan Haddad  <ryanhaddad@apple.com>
-
-        Add TestExpectations and baselines for iOS 12
-        https://bugs.webkit.org/show_bug.cgi?id=187628
-
-        Reviewed by Alexey Proskuryakov.
-
-        * Scripts/webkitpy/port/ios.py:
-        (IOSPort): Bump current version to 12.
-
-2018-07-13  Ryan Haddad  <ryanhaddad@apple.com>
-
-        Add TestExpectations and baselines for Mojave.
-        https://bugs.webkit.org/show_bug.cgi?id=187620
-
-        Reviewed by Alexey Proskuryakov.
-
-        * Scripts/webkitpy/port/mac.py:
-        (MacPort): Bump current version to 10.14.
-
-2018-07-13  Timothy Hatcher  <timothy@apple.com>
-
-        Add _drawsBackground to WKWebViewConfiguration.
-        https://bugs.webkit.org/show_bug.cgi?id=187665
-        rdar://problem/42182268
-
-        Reviewed by Tim Horton.
-
-        * TestWebKitAPI/Tests/WebKitCocoa/Configuration.mm:
-        (TestWebKitAPI.WebKit.ConfigurationDrawsBackground): Added.
-
-2018-07-13  Chris Dumez  <cdumez@apple.com>
-
-        Add a FrameLoaderClient willInjectUserScriptForFrame callback
-        https://bugs.webkit.org/show_bug.cgi?id=187565
-        <rdar://problem/42095701>
-
-        Unreviewed WKTR fix after r233782.
-
-        * WebKitTestRunner/TestInvocation.cpp:
-        (WTR::TestInvocation::didReceiveSynchronousMessageFromInjectedBundle):
-
-2018-07-13  Chris Dumez  <cdumez@apple.com>
-
-        Reduce size of WebCore::URL
-        https://bugs.webkit.org/show_bug.cgi?id=186820
-        <rdar://problem/42087508>
-
-        Unreviewed, re-generates Service worker registrations database used by API tests after r233789.
-
-        * TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj:
-        * TestWebKitAPI/Tests/WebKitCocoa/SimpleServiceWorkerRegistrations-2.sqlite3: Removed.
-        * TestWebKitAPI/Tests/WebKitCocoa/SimpleServiceWorkerRegistrations-2.sqlite3-shm: Removed.
-        * TestWebKitAPI/Tests/WebKitCocoa/SimpleServiceWorkerRegistrations-3.sqlite3: Renamed from Tools/TestWebKitAPI/Tests/WebKitCocoa/SimpleServiceWorkerRegistrations-2.sqlite3-wal.
-        * TestWebKitAPI/Tests/WebKitCocoa/WebsiteDataStoreCustomPaths.mm:
-        (TEST):
-
-2018-07-12  Wenson Hsieh  <wenson_hsieh@apple.com>
-
-        Make it easier to hit the significant rendered text layout milestone on pages with main article elements
-        https://bugs.webkit.org/show_bug.cgi?id=187578
-        <rdar://problem/42104637>
-
-        Reviewed by Ryosuke Niwa.
-
-        Tweak an existing layout test to additionally check that the significant text layout milestone is fired on a
-        page with an article element that is clearly the "main" content on the page (i.e. much taller than all other
-        articles).
-
-        * TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj:
-        * TestWebKitAPI/Tests/WebKitCocoa/significant-text-milestone-article.html: Copied from Tools/TestWebKitAPI/Tests/WebKitCocoa/significant-text-milestone.html.
-        * TestWebKitAPI/Tests/WebKitCocoa/significant-text-milestone.html:
-        * TestWebKitAPI/Tests/ios/RenderingProgressTests.mm:
-        (TEST):
-
-2018-07-12  Fujii Hironori  <Hironori.Fujii@sony.com>
-
-        Change my status to be a WebKit reviewer.
-
-        Unreviewed status update.
-
-        * Scripts/webkitpy/common/config/contributors.json:
-
-2018-07-12  Fujii Hironori  <Hironori.Fujii@sony.com>
-
-        Unreviewed contributors.json update
-
-        Just run "validate-committer-lists -c" to canonicalize the style.
-
-        * Scripts/webkitpy/common/config/contributors.json:
-
-2018-07-12  Alex Christensen  <achristensen@webkit.org>
-
-        Reduce size of WebCore::URL
-        https://bugs.webkit.org/show_bug.cgi?id=186820
-
-        Reviewed by Yusuke Suzuki and Youenn Fablet.
-
-        * TestWebKitAPI/Tests/WebCore/URLParser.cpp:
-        (TestWebKitAPI::TEST_F):
-
-2018-07-12  Brady Eidson  <beidson@apple.com>
-
-        Make process-swap-on-navigation an experimental feature.
-        <rdar://problem/41995053> and https://bugs.webkit.org/show_bug.cgi?id=187558
-
-        Reviewed by Chris Dumez.
-
-        Make sure WKTR doesn't enable ProcessSwapOnNavigation by default simply because it's an experimental feature.
-
-        * WebKitTestRunner/TestController.cpp:
-        (WTR::TestController::createWebViewWithOptions):
-        (WTR::TestController::resetPreferencesToConsistentValues):
-
-        * WebKitTestRunner/TestOptions.h:
-        (WTR::TestOptions::shouldEnableProcessSwapOnNavigation const):
-
-2018-07-12  Youenn Fablet  <youenn@apple.com>
-
-        Add a FrameLoaderClient willInjectUserScriptForFrame callback
-        https://bugs.webkit.org/show_bug.cgi?id=187565
-
-        Reviewed by Alex Christensen.
-
-        Added new test runner API to check for willInjectUserScriptForFrame callback.
-
-        * WebKitTestRunner/InjectedBundle/Bindings/TestRunner.idl:
-        * WebKitTestRunner/InjectedBundle/InjectedBundle.h:
-        (WTR::InjectedBundle::resetUserScriptInjectedCount):
-        (WTR::InjectedBundle::increaseUserScriptInjectedCount):
-        (WTR::InjectedBundle::userScriptInjectedCount const):
-        * WebKitTestRunner/InjectedBundle/InjectedBundlePage.cpp:
-        (WTR::InjectedBundlePage::InjectedBundlePage):
-        (WTR::InjectedBundlePage::resetAfterTest):
-        (WTR::InjectedBundlePage::willInjectUserScriptForFrame):
-        * WebKitTestRunner/InjectedBundle/InjectedBundlePage.h:
-        * WebKitTestRunner/InjectedBundle/TestRunner.cpp:
-        (WTR::TestRunner::userScriptInjectedCount const):
-        (WTR::TestRunner::injectUserScript):
-        * WebKitTestRunner/InjectedBundle/TestRunner.h:
-        * WebKitTestRunner/TestController.cpp:
-        (WTR::TestController::injectUserScript):
-        * WebKitTestRunner/TestController.h:
-        * WebKitTestRunner/TestInvocation.cpp:
-        (WTR::TestInvocation::didReceiveSynchronousMessageFromInjectedBundle):
-        * WebKitTestRunner/cocoa/TestControllerCocoa.mm:
-        (WTR::TestController::injectUserScript):
-
-2018-07-12  Sihui Liu  <sihui_liu@apple.com>
-
-        IndexedDB: database file of subframe cannot be removed
-        https://bugs.webkit.org/show_bug.cgi?id=187564
-
-        Reviewed by Youenn Fablet.
-
-        * TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj:
-        * TestWebKitAPI/Tests/WebKitCocoa/WebsiteDataStoreCustomPaths.mm:
-        (TEST):
-
-2018-07-12  Timothy Hatcher  <timothy@apple.com>
-
-        Make WebKit.LinkColorWithSystemAppearance work across system versions.
-        https://bugs.webkit.org/show_bug.cgi?id=187573
-
-        Reviewed by Wenson Hsieh.
-
-        * TestWebKitAPI/Tests/WebKitCocoa/SystemColors.mm:
-        (TestWebKitAPI.WebKit.LinkColorWithSystemAppearance): Dynamically generate the expected string.
-
-2018-07-11  Carlos Garcia Campos  <cgarcia@igalia.com>
-
-        [WPE] Pass the backend library name as command line parameter to the web process
-        https://bugs.webkit.org/show_bug.cgi?id=186841
-
-        Reviewed by Žan Doberšek.
-
-        * MiniBrowser/wpe/main.cpp:
-        (main): Stop using WPE_BACKEND_LIBRARY env var.
-        * Scripts/run-wpe-tests:
-        (WPETestRunner.__init__): Ditto.
-        * Scripts/webkitpy/port/wpe.py:
-        (WPEPort.setup_environ_for_server): Ditto.
-        * wpe/backends/ViewBackend.cpp:
-        (WPEToolingBackends::ViewBackend::ViewBackend): Call wpe_loader_init() to ensure WPEBackend-fdo is used.
-        * wpe/jhbuild.modules: Upgrade WPEBackend to version 0.2.0.
-
-2018-07-11  Carlos Garcia Campos  <cgarcia@igalia.com>
-
-        [GLIB] Add API to allow creating variadic functions
-        https://bugs.webkit.org/show_bug.cgi?id=187517
-
-        Reviewed by Michael Catanzaro.
-
-        Add test cases.
-
-        * TestWebKitAPI/Tests/JavaScriptCore/glib/TestJSC.cpp:
-        (testJSCFunction):
-        (fooCreateWithFooV):
-        (multiplyFooV):
-
-2018-07-11  Carlos Garcia Campos  <cgarcia@igalia.com>
-
-        [GLIB] Add jsc_context_get_global_object() to GLib API
-        https://bugs.webkit.org/show_bug.cgi?id=187515
-
-        Reviewed by Michael Catanzaro.
-
-        Add test cases.
-
-        * TestWebKitAPI/Tests/JavaScriptCore/glib/TestJSC.cpp:
-        (testJSCGlobalObject):
-        (main):
-
-2018-07-11  Carlos Garcia Campos  <cgarcia@igalia.com>
-
-        [GLIB] Handle G_TYPE_STRV in glib API
-        https://bugs.webkit.org/show_bug.cgi?id=187512
-
-        Reviewed by Michael Catanzaro.
-
-        Add test cases.
-
-        * TestWebKitAPI/Tests/JavaScriptCore/glib/TestJSC.cpp:
-        (testJSCTypes):
-        (joinFunction):
-        (testJSCFunction):
-
-2018-07-11  Aakash Jain  <aakash_jain@apple.com>
-
-        [ews-build] EWS should unapply patch and re-run jsc tests when patch fails jsc tests
-        https://bugs.webkit.org/show_bug.cgi?id=187549
-
-        Reviewed by Alexey Proskuryakov.
-
-        * BuildSlaveSupport/ews-build/factories.py:
-        (JSCTestsFactory.__init__):
-        * BuildSlaveSupport/ews-build/steps.py:
-        (CompileJSCOnlyToT): Build ToT without patch.
-        (CompileJSCOnlyToT.doStepIf): Run this step only if patch failed to build.
-        (RunJavaScriptCoreTests.evaluateCommand): Set patchFailedJSCTests property appropriately.
-        (ReRunJavaScriptCoreTests): Re-runs JSC tests.
-        (ReRunJavaScriptCoreTests.doStepIf): Run this step only if patch failed JSC tests.
-        (ReRunJavaScriptCoreTests.evaluateCommand): Set patchFailedJSCTests property appropriately.
-        (RunJavaScriptCoreTestsToT): Run JSC tests on ToT without patch.
-        (RunJavaScriptCoreTestsToT.doStepIf): Run this step only if patch failed JSC tests.
-        * BuildSlaveSupport/ews-build/steps_unittest.py: Added unit-tests.
-
-2018-07-11  Commit Queue  <commit-queue@webkit.org>
-
-        Unreviewed, rolling out r233742.
-        https://bugs.webkit.org/show_bug.cgi?id=187577
-
-        Binary incompatible change with respect to Service Worker
-        registration map (Requested by dydz on #webkit).
-
-        Reverted changeset:
-
-        "Reduce size of WebCore::URL"
-        https://bugs.webkit.org/show_bug.cgi?id=186820
-        https://trac.webkit.org/changeset/233742
-
-2018-07-11  Aditya Keerthi  <akeerthi@apple.com>
-
-        REGRESSION (231276): Attempting to copy an image fails
-        https://bugs.webkit.org/show_bug.cgi?id=187212
-
-        Reviewed by Ryosuke Niwa.
-        <rdar://problem/41540074>
-
-        Augmented test to include a URL where the lastPathComponent looks similar to a
-        file URL.
-
-        * TestWebKitAPI/Tests/mac/ContextMenuCanCopyURL.html:
-        * TestWebKitAPI/Tests/mac/ContextMenuCanCopyURL.mm:
-        (TestWebKitAPI::TEST):
-
-2018-07-11  Alex Christensen  <achristensen@webkit.org>
-
-        Add SPI for immediate injection of user scripts
-        https://bugs.webkit.org/show_bug.cgi?id=173342
-        <rdar://problem/29202285>
-
-        Reviewed by Brady Eidson, Youenn Fablet, and Geoff Garen.
-
-        * TestWebKitAPI/Tests/WebKit2Cocoa/UserContentController.mm:
-        (waitForMessage):
-        (TEST):
-
-2018-07-11  Myles C. Maxfield  <mmaxfield@apple.com>
-
-        Move WHLSL spec work to the W3C GPUWeb GitHub repository
-
-        See https://github.com/gpuweb/WHLSL
-
-        * WebGPUShadingLanguageRI/SpecWork/Makefile: Removed.
-        * WebGPUShadingLanguageRI/SpecWork/WSL.g4: Removed.
-        * WebGPUShadingLanguageRI/SpecWork/WSL.ott: Removed.
-        * WebGPUShadingLanguageRI/SpecWork/source/conf.py: Removed.
-        * WebGPUShadingLanguageRI/SpecWork/source/index.rst: Removed.
-
-2018-07-11  Dean Jackson  <dino@apple.com>
-
-        Force WKTR to use the light appearance
-        https://bugs.webkit.org/show_bug.cgi?id=187572
-        <rdar://problem/42097805>
-
-        Reviewed by Simon Fraser.
-
-        Set the NSAppearance on the NSWindow to light mode.
-
-        * WebKitTestRunner/mac/PlatformWebViewMac.mm:
-        (WTR::PlatformWebView::PlatformWebView): Use the Aqua appearance.
-
-2018-07-11  Alex Christensen  <achristensen@webkit.org>
-
-        Reduce size of WebCore::URL
-        https://bugs.webkit.org/show_bug.cgi?id=186820
-
-        Reviewed by Yusuke Suzuki.
-
-        * TestWebKitAPI/Tests/WebCore/URLParser.cpp:
-        (TestWebKitAPI::TEST_F):
-
-2018-07-11  Youenn Fablet  <youenn@apple.com>
-
-        Fix remaining Cross-Origin-Resource-Policy failures, if any
-        https://bugs.webkit.org/show_bug.cgi?id=186761
-        <rdar://problem/41209829>
-
-        Reviewed by Alex Christensen.
-
-        * TestWebKitAPI/Tests/WebCore/HTTPParsers.cpp:
-        (TestWebKitAPI::TEST):
-
- 2018-07-11  Aakash Jain  <aakash_jain@apple.com>
-        [ews-build] EWS should unapply the patch and build ToT when patch fails to build
-        https://bugs.webkit.org/show_bug.cgi?id=187358
-        Reviewed by Alexey Proskuryakov.
-        * BuildSlaveSupport/ews-build/factories.py:
-        (BuildFactory.__init__):
-        * BuildSlaveSupport/ews-build/steps.py:
-        (CompileWebKit): Set haltOnFailure to false so as to run next steps (e.g.: un-apply patch, compile ToT).
-        (CompileWebKit.evaluateCommand): Set patchFailedToBuild property.
-        (CompileWebKitToT): Build ToT without patch.
-        (CompileWebKitToT.doStepIf): Run this step only if patch failed to build.
-        * BuildSlaveSupport/ews-build/steps_unittest.py: Added unit-tests. Also lower-cased Release and Debug.
-
-2018-07-11  Robin Morisset  <rmorisset@apple.com>
-
-        [WSL] WIP, more of the spec
-        https://bugs.webkit.org/show_bug.cgi?id=186310
-
-        Rubberstamped by Filip Pizlo.
-
-        Fixed a few issues with the formal rules (such as adding the restrictions on pointers for compilation to SpirV),
-        and started writing some of the english in the spec.
-
-        * WebGPUShadingLanguageRI/SpecWork/WSL.ott:
-        * WebGPUShadingLanguageRI/SpecWork/source/index.rst:
-
-2018-07-10  Aakash Jain  <aakash_jain@apple.com>
-
-        [ews-build] Add build step to UnApply patch
-        https://bugs.webkit.org/show_bug.cgi?id=187357
-
-        Reviewed by Alexey Proskuryakov.
-
-        * BuildSlaveSupport/ews-build/steps.py:
-        (UnApplyPatchIfRequired.__init__): Pass alwaysUseLatest parameter in order to unapply the patch.
-        (UnApplyPatchIfRequired.doStepIf): Run this step only if patch failed to build or failed test. These
-        properties will be set by build-steps for build and test.
-        (UnApplyPatchIfRequired.hideStepIf): Hide this step if it is not run.
-
-2018-07-10  Dewei Zhu  <dewei_zhu@apple.com>
-
-        Update ARES-6.patch file accordingly for r233631.
-        https://bugs.webkit.org/show_bug.cgi?id=187534
-
-        Reviewed by Saam Barati.
-
-        Update ARES-6.patch file acoordingly after ARES-6 plan file change to
-        avoid warning while applying the patch.
-
-        * Scripts/webkitpy/benchmark_runner/data/patches/webserver/ARES-6.patch:
-
-2018-07-10  Tim Horton  <timothy_horton@apple.com>
-
-        doAfterNextPresentationUpdate should not be called while content is hidden due to animated resize
-        https://bugs.webkit.org/show_bug.cgi?id=187500
-        <rdar://problem/41294139>
-
-        Reviewed by Simon Fraser.
-
-        * TestWebKitAPI/Tests/WebKitCocoa/AnimatedResize.mm:
-        (-[AnimatedResizeWebView _endAnimatedResize]):
-        Set didEndAnimatedResize before calling super, because it makes the
-        new test easier and doesn't affect any of the existing ones.
-
-        (AnimatedResizeBlocksViewportFitChanges):
-        This doAfterNextPresentationUpdate is always called during animated resize,
-        and then synchronously waited for, so it /must/ use
-        _doAfterNextPresentationUpdateWithoutWaitingForAnimatedResizeWithTesting.
-        
-        (AnimatedResizeBlocksDoAfterNextPresentationUpdate):
-        Add a test ensuring that doAfterNextPresentationUpdate is deferred
-        until endAnimatedResize is called.
-
-        * TestWebKitAPI/cocoa/TestNavigationDelegate.mm:
-        (-[WKWebView _test_waitForDidFinishNavigation]):
-        _test_waitForDidFinishNavigation is sometimes called during animated resize,
-        and synchronously waits for a doAfterNextPresentationUpdate, so
-        it cannot wait for the animated resize to complete.
-
-2018-07-10  Fujii Hironori  <Hironori.Fujii@sony.com>
-
-        [GTK][WPE] TestWTF and TestJSC fail to compile due to missing forwarding headers
-        https://bugs.webkit.org/show_bug.cgi?id=184441
-
-        Reviewed by Michael Catanzaro.
-
-        All executables of TestWebKitAPI need the forwarding headers of
-        WebCore and PAL because config.h includes
-        <WebCore/PlatformExportMacros.h> and <pal/ExportMacros.h>.
-
-        The forwarding headers of WebCore are generated by
-        generate-forwarding-headers.pl in GTK and WPE ports. The
-        forwarding headers of PAL are generated by
-        WEBKIT_MAKE_FORWARDING_HEADERS in all CMake ports.
-
-        * TestWebKitAPI/PlatformGTK.cmake: Added PALForwardingHeaders to
-        ForwardingHeadersForTestWebKitAPI_NAME. Added dependencies of
-        ${ForwardingHeadersForTestWebKitAPI_NAME} to TestJSC.
-        * TestWebKitAPI/PlatformWPE.cmake: Ditto.
-
-2018-07-10  Stephan Szabo  <stephan.szabo@sony.com>
-
-        [WinCairo] Support display of webinspector ui on non-legacy minibrowser
-        https://bugs.webkit.org/show_bug.cgi?id=187415
-
-        Reviewed by Brian Burg.
-
-        * MiniBrowser/win/WebKitBrowserWindow.cpp:
-        (WebKitBrowserWindow::WebKitBrowserWindow):
-
-2018-07-10  John Wilander  <wilander@apple.com>
-
-        Resource Load Statistics: Make testRunner.statisticsResetToConsistentState() take a completion handler
-        https://bugs.webkit.org/show_bug.cgi?id=187499
-        <rdar://problem/41999431>
-
-        Reviewed by Chris Dumez.
-
-        * WebKitTestRunner/InjectedBundle/Bindings/TestRunner.idl:
-        * WebKitTestRunner/InjectedBundle/InjectedBundle.cpp:
-        (WTR::InjectedBundle::didReceiveMessageToPage):
-        * WebKitTestRunner/InjectedBundle/TestRunner.cpp:
-        (WTR::TestRunner::statisticsResetToConsistentState):
-        (WTR::TestRunner::statisticsCallDidResetToConsistentStateCallback):
-        * WebKitTestRunner/InjectedBundle/TestRunner.h:
-        * WebKitTestRunner/TestController.cpp:
-        (WTR::TestController::statisticsResetToConsistentState):
-        * WebKitTestRunner/TestInvocation.cpp:
-        (WTR::TestInvocation::didResetStatisticsToConsistentState):
-        * WebKitTestRunner/TestInvocation.h:
-
-2018-07-10  Ryosuke Niwa  <rniwa@webkit.org>
-
-        Disable cross-origin-window-policy by default
-        https://bugs.webkit.org/show_bug.cgi?id=187509
-
-        Reviewed by Chris Dumez.
-
-        Enable the feature for testing purposes. DumpRenderTree does this already in enableExperimentalFeatures.
-
-        * WebKitTestRunner/TestController.cpp:
-        (WTR::TestController::resetPreferencesToConsistentValues):
-
-2018-07-10  Thibault Saunier  <tsaunier@igalia.com>
-
-        [Flatpak] Feature options to build-webkit and --makeargs are not passed inside the sandbox
-        https://bugs.webkit.org/show_bug.cgi?id=187470
-
-        Reviewed by Philippe Normand.
-
-        Never force building, the user is responsible to make that decision and the workflow is
-        now based on external scripts so it doesn't make sense anymore. Doing that leads
-        to ignoring the options the user passed in.
-
-        * flatpak/flatpakutils.py:
-        (WebkitFlatpak.setup_dev_env):
-
-2018-07-10  Thibault Saunier  <tsaunier@igalia.com>
-
-        [Flatpak] Fix running wpt tests inside flatpak
-        https://bugs.webkit.org/show_bug.cgi?id=187490
-
-        Reviewed by Philippe Normand.
-
-        Still not *all* passing but a big majority do.
-
-        * Scripts/webkitpy/w3c/wpt_runner.py:
-        (main): Generate the /etc/hosts and mount it in flatpak as required.
-        (WPTRunner.prepare_wpt_checkout): Renamed as we need to first checkout WPT code before generating /etc/hosts.
-        (WPTRunner._prepare_wpt_checkout): Deleted.
-        * flatpak/flatpakutils.py: Added a stdout= argument to run_in_sandbox() and
-        added a way to pass extra env to flatpak itself (so we can mount /etc/hosts).
-        Stopped passing `--device=all` as it is not required since r233638 (Bug 187400).
-        * flatpak/org.webkit.WebKit.yaml: Added virtualenv for python2 as required by wpt.
-
-2018-07-10  Thibault Saunier  <tsaunier@igalia.com>
-
-        [Flatpak] Use logging to print debug information
-        https://bugs.webkit.org/show_bug.cgi?id=187496
-
-        Reviewed by Philippe Normand.
-
-        * flatpak/flatpakutils.py:
-        (WebkitFlatpak.load_from_args):
-        (WebkitFlatpak.clean_args):
-        (WebkitFlatpak.run_in_sandbox):
-
-2018-07-09  Timothy Hatcher  <timothy@apple.com>
-
-        Semantic colors don't update when accessibility Increase Contrast mode is enabled.
-        https://bugs.webkit.org/show_bug.cgi?id=187425
-        rdar://problem/39948240
-        rdar://problem/41796865
-
-        Reviewed by Tim Horton.
-
-        * TestWebKitAPI/Tests/WebKitCocoa/SystemColors.mm:
-        (TestWebKitAPI.WebKit.LinkColorWithSystemAppearance): Use new color value.
-
-2018-07-09  Simon Fraser  <simon.fraser@apple.com>
-
-        Fix dump-class-layout to show bit padding, and fix issues with padding offsets
-        https://bugs.webkit.org/show_bug.cgi?id=187442
-
-        Reviewed by Daniel Bates.
-        
-        Synthetic padding members were shown with the wrong offset because they used the
-        absolute offset rather than the class-relative offset. This didn't change the padding
-        math, but made the output confusing.
-        
-        Also add support for showing empty bits in bitfields, and fix computation of padding
-        after bitfields. Empty bits are computed by inserting a bit padding member after
-        a bitfield that is not followed by another bitfield (making the assumption that bit
-        padding will fill to the next byte boundary).
-        
-        The computation of padding after bitfields was also wrong, since lldb's member_type.GetByteSize()
-        just reports the size of the type without the bitfield modifier (e.g. for "unsigned : 2" it returned 4).
-        Fix by setting the byte size for bitfield fields to the number of bits rounded up to the next byte;
-        this allows byte padding following the bitfield to be computed correctly.
-        
-        Add or modify test to cover these issues.
-
-        * lldb/dump_class_layout_unittest.py:
-        (serial_test_ClassWithPaddedBitfields):
-        (serial_test_MemberHasBitfieldPadding):
-        (serial_test_InheritsFromClassWithPaddedBitfields):
-        * lldb/lldbWebKitTester/DumpClassLayoutTesting.cpp:
-        (avoidClassDeadStripping):
-        * lldb/lldb_dump_class_layout.py:
-        (ClassLayoutBase):
-        (ClassLayoutBase._to_string_recursive):
-        (ClassLayout._parse):
-        (ClassLayout._compute_padding_recursive):
-
-2018-07-09  Youenn Fablet  <youenn@apple.com>
-
-        StringView operator==(char*) should check the length of the string
-        https://bugs.webkit.org/show_bug.cgi?id=187422
-
-        Reviewed by Chris Dumez.
-
-        * TestWebKitAPI/Tests/WTF/StringView.cpp:
-        (TestWebKitAPI::equal2):
-        (TestWebKitAPI::TEST):
-
-2018-07-09  Ross Kirsling  <rkirsling@gmail.com>
-
-        [WinCairo] httpd service install needs to precede server start
-        https://bugs.webkit.org/show_bug.cgi?id=187283
-
-        Reviewed by Daniel Bates.
-
-        * Scripts/webkitpy/port/base.py:
-        (Port.check_build):
-        Remove unused parameter `needs_http`.
-        
-        (Port.check_sys_deps):
-        (Port.start_http_server):
-        Move `check_httpd` call.
-
-        * Scripts/webkitpy/port/win.py:
-        (WinPort.check_httpd): Added.
-        (WinCairoPort.check_httpd): Deleted.
-        Move `httpd -k install` logic to common Windows location.
-
-        * Scripts/webkitpy/layout_tests/controllers/manager.py:
-        (Manager._set_up_run):
-        * Scripts/webkitpy/performance_tests/perftestsrunner.py:
-        * Scripts/webkitpy/port/gtk.py:
-        (GtkPort.check_sys_deps):
-        * Scripts/webkitpy/port/ios_simulator.py:
-        (IOSSimulatorPort.check_sys_deps):
-        * Scripts/webkitpy/port/mock_drt.py:
-        (MockDRTPort.check_build):
-        (MockDRTPort.check_sys_deps):
-        * Scripts/webkitpy/port/port_testcase.py:
-        (PortTestCase.integration_test_check_sys_deps):
-        (test_check_build):
-        * Scripts/webkitpy/port/test.py:
-        * Scripts/webkitpy/port/wpe.py:
-        (WPEPort.check_sys_deps):
-        * Scripts/webkitpy/tool/commands/perfalizer.py:
-        (PerfalizerTask.run):
-        Remove `needs_http` arguments.
-
-2018-07-09  Philippe Normand  <pnormand@igalia.com>
-
-        [Flatpak] Unreviewed, switch to https for the gst-libav tarball.
-        
-        * flatpak/org.webkit.WebKit.yaml:
-
-2018-07-09  Thibault Saunier  <tsaunier@igalia.com>
-
-        [Flatpak] HTTP layout tests broken due to missing CGI.pm in the sandbox
-        https://bugs.webkit.org/show_bug.cgi?id=187384
-
-        Reviewed by Philippe Normand.
-
-        And make it usable inside the sandbox even when
-        perl is launched "tainted" (as the PERL5LIB won't
-        work otherwise).
-
-        * flatpak/flatpakutils.py:
-        (WebkitFlatpak.run_in_sandbox):
-        * flatpak/org.webkit.WebKit.yaml:
-
-2018-07-09  Thibault Saunier  <tsaunier@igalia.com>
-
-        [WPE][GTK] Mount /dev/shm on /run/shm in sandbox
-        https://bugs.webkit.org/show_bug.cgi?id=187400
-
-        Reviewed by Philippe Normand.
-
-        This is where it is expected to be when it is a symlink,
-        without mounting it the websocket python server can't work
-        inside the sandbox.
-
-        * flatpak/flatpakutils.py:
-        (WebkitFlatpak.run_in_sandbox):
-
-2018-07-08  Filip Pizlo  <fpizlo@apple.com>
-
-        run-benchmark should run ARES-6 1.0.1
-        https://bugs.webkit.org/show_bug.cgi?id=187452
-
-        Reviewed by Sam Weinig.
-        
-        We forgot to update this when we released 1.0.1!
-
-        * Scripts/webkitpy/benchmark_runner/data/plans/ares6.plan:
-
-2018-07-07  Youenn Fablet  <youenn@apple.com>
-
-        WPT importer should create dummy HTML files for *.window.js script files
-        https://bugs.webkit.org/show_bug.cgi?id=187426
-
-        Reviewed by Ryosuke Niwa.
-
-        Some WPT tests use *.window.js tests, which should be accessed as *.window.html.
-        Do the same strategy as done for worker.js/any.js files.
-
-        * Scripts/webkitpy/w3c/test_importer.py:
-        (TestImporter.write_html_files_for_templated_js_tests):
-
-2018-07-07  Wenson Hsieh  <wenson_hsieh@apple.com>
-
-        Introduce a layout milestone to track when the document contains a large number of rendered characters
-        https://bugs.webkit.org/show_bug.cgi?id=187412
-        <rdar://problem/41744338>
-
-        Reviewed by Ryosuke Niwa.
-
-        Add a new API test to check that we observe the new rendering milestone in an article-like document.
-
-        * TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj:
-        * TestWebKitAPI/Tests/WebKitCocoa/significant-text-milestone.html: Added.
-        * TestWebKitAPI/Tests/ios/RenderingProgressTests.mm:
-        (TEST):
-
-2018-06-30  Simon Fraser  <simon.fraser@apple.com>
-
-        dump-class-layout reports wrong padding in many cases
-        https://bugs.webkit.org/show_bug.cgi?id=185801
-
-        Reviewed by Dan Bates.
-        
-        Enhance dump-class-layout, fixing bugs and adding unit tests. This also includes
-        the patch from bug 187141.
-        
-        Move the code that uses lldb to fetch the class layout to its own module,
-        lldb_dump_class_layout.py, so it can be unit tested. Change this code
-        to build up a data structure for the class. This is necessary since correct
-        padding computation is easier with second traversal over the data structure.
-        
-        Try to deal with the empty base class optimization, which is necessary to report
-        correct padding in classes using std::unique_ptr, and correctly handle virtual base classes.
-        
-        The ClassLayoutBase class knows how to test for equality and generate a string
-        or string list representation of itself. The derived ClassLayout class knows
-        how to build up the class layout using the lldb Python bindings.
-        
-        Finally we wrap the lldb debugger instance in LLDBDebuggerInstance so that we don't
-        have to make a new one for each unit test. The tests have to run serially.
-        
-        Add to lldbWebKitTester a .cpp file that contains a set of C++ classes to unit-test
-        dump-class-layout.
-
-        * Scripts/dump-class-layout:
-        (main):
-        * lldb/dump_class_layout_unittest.py:
-        (destroy_cached_debug_session):
-        (TestDumpClassLayout):
-        (TestDumpClassLayout.setUpClass):
-        (TestDumpClassLayout.setUp):
-        (TestDumpClassLayout.serial_test_BasicClassLayout):
-        (serial_test_PaddingBetweenClassMembers):
-        (serial_test_BoolPaddingClass):
-        (serial_test_ClassWithEmptyClassMembers):
-        (serial_test_SimpleVirtualClass):
-        (serial_test_SimpleVirtualClassWithNonVirtualBase):
-        (serial_test_InterleavedVirtualNonVirtual):
-        (serial_test_ClassWithTwoVirtualBaseClasses):
-        (serial_test_ClassWithClassMembers):
-        (serial_test_ClassWithBitfields):
-        (serial_test_ClassWithUniquePtrs):
-        (serial_test_ClassWithOptionals):
-        (TestDumpClassLayout.test_BasicClassLayout): Deleted.
-        * lldb/lldbWebKitTester/DumpClassLayoutTesting.cpp:
-        (EmptyClass::doStuff):
-        (VirtualBaseClass::~VirtualBaseClass):
-        (VirtualBaseClass2::~VirtualBaseClass2):
-        (SimpleVirtualClass::~SimpleVirtualClass):
-        (SimpleVirtualClass::doStuff):
-        (SimpleVirtualClassWithNonVirtualBase::~SimpleVirtualClassWithNonVirtualBase):
-        (SimpleVirtualClassWithNonVirtualBase::doStuff):
-        (avoidClassDeadStripping):
-        * lldb/lldb_dump_class_layout.py:
-        (ansi_colors):
-        (ClassLayoutBase):
-        (ClassLayoutBase.__init__):
-        (ClassLayoutBase.__ne__):
-        (ClassLayoutBase.__eq__):
-        (ClassLayoutBase._to_string_recursive):
-        (ClassLayoutBase.as_string_list):
-        (ClassLayoutBase.as_string):
-        (ClassLayoutBase.dump):
-        (ClassLayoutExpected):
-        (ClassLayoutExpected.__init__):
-        (ClassLayout):
-        (ClassLayout.__init__):
-        (ClassLayout._has_polymorphic_base_class):
-        (ClassLayout._parse):
-        (ClassLayout._probably_has_empty_base_class_optimization):
-        (ClassLayout._compute_padding_recursive):
-        (ClassLayout._compute_padding):
-        (LLDBDebuggerInstance):
-        (LLDBDebuggerInstance.__init__):
-        (LLDBDebuggerInstance.__del__):
-        (LLDBDebuggerInstance._get_first_file_architecture):
-        (LLDBDebuggerInstance.layout_for_classname):
-        (ClassLayoutDumper): Deleted.
-        (ClassLayoutDumper.__init__): Deleted.
-        (ClassLayoutDumper._get_first_file_architecture): Deleted.
-        (ClassLayoutDumper.verify_type): Deleted.
-        (ClassLayoutDumper.verify_type_recursive): Deleted.
-        (ClassLayoutDumper._class_layout_as_string): Deleted.
-        (ClassLayoutDumper.dump_to_string): Deleted.
-        (ClassLayoutDumper.dump): Deleted.
-
-2018-07-06  Christopher Reid  <chris.reid@sony.com>
-
-        [WinCairo] WebKit MiniBrowser crashes when attempting to navigate to certain URLs
-        https://bugs.webkit.org/show_bug.cgi?id=187167
-
-        Reviewed by Alex Christensen.
-
-        A null byte was written past the end of the buffer causing the crash.
-        Some of the heap allocated buffers were also not getting deleted.
-
-        * MiniBrowser/win/WebKitBrowserWindow.cpp:
-
-2018-07-06  Thibault Saunier  <tsaunier@igalia.com>
-
-        [WPE][GTK] Build php zlib plugin and enable sockets in flatpak
-        https://bugs.webkit.org/show_bug.cgi?id=187401
-
-        Reviewed by Philippe Normand.
-
-        * flatpak/org.webkit.WebKit.yaml:
-
-2018-07-06  Aakash Jain  <aakash_jain@apple.com>
-
-        [ews-build] EWS should run javascriptcore tests
-        https://bugs.webkit.org/show_bug.cgi?id=187365
-
-        Reviewed by David Kilzer.
-
-        * BuildSlaveSupport/ews-build/steps.py: Added step to run JavaScriptCoreTests.
-        * BuildSlaveSupport/ews-build/steps_unittest.py: Added unit-tests.
-        * BuildSlaveSupport/ews-build/factories.py:
-
-2018-07-06  Carlos Garcia Campos  <cgarcia@igalia.com>
-
-        [GTK][WPE] REGRESSION(r233111): Test /webkit/WebKitUserContentManager/script-message-received fails
-        https://bugs.webkit.org/show_bug.cgi?id=187383
-
-        Reviewed by Žan Doberšek.
-
-        This is actually a problem in the test itself. The problem is that we run js, connect to message-received and
-        quite the main loop when the message is received, which happens before the run js operation has been
-        completed. The test finishes early and since r233111 an IPC message is still pending (I guess the run js
-        callback) and the web page is still protected. Just adding a run loop operation before finishing the test would
-        fix it. We should wait for the js run to complete, instead of the message.
-
-        * TestWebKitAPI/Tests/WebKitGLib/TestWebKitUserContentManager.cpp:
-        (UserScriptMessageTest::scriptMessageReceived):
-        (UserScriptMessageTest::waitUntilMessageReceived):
-        (UserScriptMessageTest::runJavaScriptFinished):
-        (UserScriptMessageTest::postMessageAndWaitUntilReceived):
-
-2018-07-05  Youenn Fablet  <youenn@apple.com>
-
-        REGRESSION (r230843): Flash doesn't work; Produces blue box on page
-        https://bugs.webkit.org/show_bug.cgi?id=187346
-        <rdar://problem/41773974>
-
-        Reviewed by Ryosuke Niwa.
-
-        * WebKitTestRunner/TestController.cpp:
-        (WTR::TestController::resetStateToConsistentValues):
-        (WTR::TestController::setBlockAllPlugins):
-        When not blocked, set the load policy to allow once.
-        * WebKitTestRunner/TestController.h:
-
-2018-07-05  Aakash Jain  <aakash_jain@apple.com>
-
-        [ews-build] Bindings-tests-EWS should output result in json
-        https://bugs.webkit.org/show_bug.cgi?id=187330
-
-        Reviewed by David Kilzer.
-
-        * BuildSlaveSupport/ews-build/steps.py:
-        (RunBindingsTests):  Passed --json-output parameter to run-bindings-tests.
-        * BuildSlaveSupport/ews-build/steps_unittest.py: Updated unit-tests.
-
-2018-07-05  Aakash Jain  <aakash_jain@apple.com>
-
-        [ews-build] WebKitPy-Tests-EWS should output result in json
-        https://bugs.webkit.org/show_bug.cgi?id=187331
-
-        Follow-up fix to r233527. Extracting jsonFileName as a variable and re-using it.
-
-2018-07-05  Aakash Jain  <aakash_jain@apple.com>
-
-        [ews-build] WebKitPy-Tests-EWS should output result in json
-        https://bugs.webkit.org/show_bug.cgi?id=187331
-
-        Reviewed by David Kilzer.
-
-        * BuildSlaveSupport/ews-build/steps.py:
-        (RunWebKitPyTests): Passed --json-output parameter to test-webkitpy.
-        * BuildSlaveSupport/ews-build/steps_unittest.py: Updated unit-tests.
-
-2018-07-05  Philippe Normand  <pnormand@igalia.com>
-
-        [Flatpak] Add gst-libav to dependencies
-        https://bugs.webkit.org/show_bug.cgi?id=187340
-
-        Reviewed by Žan Doberšek.
-
-        * flatpak/org.webkit.WebKit.yaml: Use tarballs instead of git
-        clones because the FFMPeg git server doesn't support shallow
-        clones.
-
-2018-07-05  Thibault Saunier  <tsaunier@igalia.com>
-
-        [WPE][GTK]: Call gtk platform 'gtk' by default as this is what PortFactory expects
-        https://bugs.webkit.org/show_bug.cgi?id=187322
-
-        Reviewed by Žan Doberšek.
-
-        * flatpak/flatpakutils.py:
-        (WebkitFlatpak.__init__):
-
-2018-07-05  Zan Dobersek  <zdobersek@igalia.com>
-
-        Unreviewed, rolling out r233511.
-
-        Proper fix provided in bug #187322
-
-        Reverted changeset:
-
-        "Unreviewed. Fixing a call in flatpakutils.py to
-        PortFactory.get(),"
-        https://trac.webkit.org/changeset/233511
-
-2018-07-04  Zan Dobersek  <zdobersek@igalia.com>
-
-        Unreviewed. Fixing a call in flatpakutils.py to PortFactory.get(),
-        enforcing the passed platform argument to always be lowercase, as is
-        expected from the PortFactory code.
-
-        * flatpak/flatpakutils.py:
-        (WebkitFlatpak.clean_args):
-
-2018-07-04  Tim Horton  <timothy_horton@apple.com>
-
-        Introduce PLATFORM(IOSMAC)
-        https://bugs.webkit.org/show_bug.cgi?id=187315
-
-        Reviewed by Dan Bernstein.
-
-        * TestWebKitAPI/Configurations/Base.xcconfig:
-        * TestWebKitAPI/Configurations/FeatureDefines.xcconfig:
-
-2018-07-03  Brady Eidson  <beidson@apple.com>
-
-        WebKit.ResponsivenessTimerCrash sometimes fails.
-        https://bugs.webkit.org/show_bug.cgi?id=187314
-
-        Reviewed by Tim Horton.
-
-        Actually unregister observers from objects we're about to destroy before destroying them.
-
-        * TestWebKitAPI/Tests/WebKit/ResponsivenessTimerCrash.mm:
-        (-[ResponsivenessTimerObserver observeValueForKeyPath:ofObject:change:context:]): 
-
-2018-07-03  Fujii Hironori  <Hironori.Fujii@sony.com>
-
-        update-webkitgtk-flatpak fails if running with python2
-        https://bugs.webkit.org/show_bug.cgi?id=187240
-
-        Reviewed by Michael Catanzaro.
-
-        urllib.urlretrieve of Python2 sends invalid requests. Use
-        urllib2.urlopen instead.
-
-        * flatpak/flatpakutils.py: Import urllib2.urlopen for Python2,
-        urllib.request.urlopen for Python3.
-        (FlatpakRepo.repo_file): Copy the content of urlopen result to the
-        temporary file.
-
-2018-07-03  Chris Dumez  <cdumez@apple.com>
-
-        Make CallbackMap::invalidate() safe to re-enter
-        https://bugs.webkit.org/show_bug.cgi?id=187298
-        <rdar://problem/41057167>
-
-        Reviewed by Geoffrey Garen.
-
-        Add API test coverage.
-
-        * TestWebKitAPI/Tests/WebKitCocoa/WebContentProcessDidTerminate.mm:
-        (TEST):
-
-2018-07-03  Chris Dumez  <cdumez@apple.com>
-
-        [Cocoa] Disable vnode guard related simulated crashes for WKTR / DRT and WebSQL
-        https://bugs.webkit.org/show_bug.cgi?id=187270
-        <rdar://problem/40674034>
-
-        Reviewed by Geoffrey Garen.
-
-        Disable vnode guard related simulated crashes for WKTR / DRT by setting the
-        expected environment variable in the test driver.
-
-        * Scripts/webkitpy/port/driver.py:
-        (Driver._setup_environ_for_driver):
-
-2018-07-03  Eric Carlson  <eric.carlson@apple.com>
-
-        Video sometimes flickers when playing to AppleTV
-        https://bugs.webkit.org/show_bug.cgi?id=187193
-        <rdar://problem/40153865>
-
-        Unreviewed, fix test broken by r233435.
-
-        * TestWebKitAPI/Tests/WebKitLegacy/ios/AudioSessionCategoryIOS.mm:
-        (TestWebKitAPI::waitUntilAudioSessionCategoryIsEqualTo): Poll for a change in audio session 
-        category because it doesn't necessarily happen synchronously.
-        (TestWebKitAPI::TEST):
-
-2018-07-03  Thibault Saunier  <tsaunier@igalia.com>
-
-        Unreviewed, add myself as a WebKit committer.
-
-        And canonicalize the file as suggested by `check-webkit-style`
-
-        * Scripts/webkitpy/common/config/contributors.json:
-
-2018-07-03  Basuke Suzuki  <Basuke.Suzuki@sony.com>
-
-        check-webkit-style: Run the tool without SVN client.
-        https://bugs.webkit.org/show_bug.cgi?id=187294
-
-        Reviewed by Daniel Bates.
-
-        When invoking check-webkit-style without SVN client, the exception was raised and terminated.
-        Using Git repository is acceptable workflow according to the wording on:
-        <https://webkit.org/getting-the-code/#checking-out-with-git>
-        and using Subversion is considered optional by following link:
-        <https://webkit.org/webkit-on-windows/#installing-developer-tools>.
-        This patch makes it so that people with only a Git checkout of WebKit that commit using
-        the commit-queue (or another machine with an SVN checkout) can run check-webkit-style.
-
-        * Scripts/webkitpy/common/checkout/scm/svn.py:
-        (SVN.in_working_directory):
-
-2018-07-03  Aakash Jain  <aakash_jain@apple.com>
-
-        Remove old JSCTestsFactory
-
-        Follow-up fix to r233471.
-
-2018-07-03  Aakash Jain  <aakash_jain@apple.com>
-
-        [ews-build] Add support for JSC-Tests-EWS
-        https://bugs.webkit.org/show_bug.cgi?id=187201
-
-        Reviewed by Lucas Forschler.
-
-        * BuildSlaveSupport/ews-build/factories.py:
-        (JSCTestsFactory): Added JSCTestsFactory.
-        * BuildSlaveSupport/ews-build/steps.py:
-        (CompileJSCOnly): Command to compile JSC.
-        * BuildSlaveSupport/ews-build/steps_unittest.py: Updated unit-tests.
-        * BuildSlaveSupport/ews-build/config.json: Changed platform to jsc-only. This matches with build.webkit.org configuration.
-
-2018-07-03  Zalan Bujtas  <zalan@apple.com>
-
-        [LFC] Add test-list file for regression testing
-        https://bugs.webkit.org/show_bug.cgi?id=187296
-
-        Reviewed by Antti Koivisto.
-
-        run-webkit-tests --test-list=./Tools/LayoutReloaded/misc/LFC-passing-tests.txt
-
-        * LayoutReloaded/misc/LFC-passing-tests.txt: Added.
-
-2018-07-03  John Wilander  <wilander@apple.com>
-
-        Resource Load Statistics: Make WebsiteDataStore::getAllStorageAccessEntries() call the right network process instead of iterating over the process pools
-        https://bugs.webkit.org/show_bug.cgi?id=187277
-        <rdar://problem/41745510>
-
-        Reviewed by Chris Dumez.
-
-        * WebKitTestRunner/cocoa/TestControllerCocoa.mm:
-        (WTR::TestController::getAllStorageAccessEntries):
-
-2018-07-02  Wenson Hsieh  <wenson_hsieh@apple.com>
-
-        Fix a leak in ActionSheetTests.DismissingActionSheetShouldNotDismissPresentingViewController
-        https://bugs.webkit.org/show_bug.cgi?id=187280
-
-        Reviewed by Tim Horton.
-
-        Add a missing `adoptNS()`.
-
-        * TestWebKitAPI/Tests/ios/ActionSheetTests.mm:
-        (TestWebKitAPI::TEST):
-
-2018-07-02  Myles C. Maxfield  <mmaxfield@apple.com>
-
-        [Cocoa] Google Fonts doesn't work if the user has the requested font locally-installed
-        https://bugs.webkit.org/show_bug.cgi?id=187228
-        <rdar://problem/40967280>
-
-        Reviewed by Brent Fulgham.
-
-        The test only fails before the patch if the lookup for Helvetica2 is allowed to occur.
-
-        * WebKitTestRunner/mac/TestControllerMac.mm:
-        (WTR::allowedFontFamilySet):
-
-2018-07-02  Ryosuke Niwa  <rniwa@webkit.org>
-
-        run-benchmark should include the version number of MotionMark's results
-        https://bugs.webkit.org/show_bug.cgi?id=187274
-
-        Reviewed by Jon Lee.
-
-        Append the version number as a part of the suite name.
-
-        * Scripts/webkitpy/benchmark_runner/data/patches/webserver/MotionMark.patch:
-
-2018-07-02  Wenson Hsieh  <wenson_hsieh@apple.com>
-
-        [WK1] editing/spelling/markers.html is failing on recent builds of macOS Mojave
-        https://bugs.webkit.org/show_bug.cgi?id=187253
-
-        Reviewed by Tim Horton.
-
-        Enhances and refactors LayoutTestSpellChecker, which was introduced in r233412.
-
-        * DumpRenderTree/TestRunner.cpp:
-        (setSpellCheckerResultsCallback):
-
-        Rename setSpellCheckerTextReplacements to setSpellCheckerResults.
-
-        (TestRunner::staticFunctions):
-        (setSpellCheckerTextReplacementsCallback): Deleted.
-        * DumpRenderTree/TestRunner.h:
-        * DumpRenderTree/mac/TestRunnerMac.mm:
-        (TestRunner::setSpellCheckerResults):
-        (TestRunner::setSpellCheckerTextReplacements): Deleted.
-        * DumpRenderTree/win/TestRunnerWin.cpp:
-        (TestRunner::setSpellCheckerResults):
-        (TestRunner::setSpellCheckerTextReplacements): Deleted.
-        * TestRunnerShared/cocoa/LayoutTestSpellChecker.h:
-        * TestRunnerShared/cocoa/LayoutTestSpellChecker.mm:
-        (-[LayoutTestTextCheckingResult initWithType:range:replacement:details:]):
-        (-[LayoutTestTextCheckingResult grammarDetails]):
-        (-[LayoutTestSpellChecker reset]):
-        (-[LayoutTestSpellChecker results]):
-        (-[LayoutTestSpellChecker setResults:]):
-        (-[LayoutTestSpellChecker setResultsFromJSObject:inContext:]):
-
-        Add support for passing in a list of grammar correction detail ranges. Necessary for simulating grammar errors.
-
-        (-[LayoutTestSpellChecker checkString:range:types:options:inSpellDocumentWithTag:orthography:wordCount:]):
-
-        Tweaked to always call the superclass method. This ensures that we set the `orthography` and `wordCount`
-        outpointers if applicable.
-
-        (-[LayoutTestSpellChecker requestCheckingOfString:range:types:options:inSpellDocumentWithTag:completionHandler:]):
-
-        Added support for simulating asynchronous spell checking.
-
-        (-[LayoutTestTextCheckingResult initWithType:range:replacement:]): Deleted.
-        (-[LayoutTestSpellChecker replacements]): Deleted.
-        (-[LayoutTestSpellChecker setReplacements:]): Deleted.
-        (-[LayoutTestSpellChecker setReplacementsFromJSObject:inContext:]): Deleted.
-
-2018-07-02  Brady Eidson  <beidson@apple.com>
-
-        Another unreviewed followup to:
-        Crash notifying observers of responsiveness state change
-
-        * TestWebKitAPI/cocoa/TestWKWebView.h:
-        * TestWebKitAPI/cocoa/TestWKWebView.mm:
-        (-[TestWKWebView typeCharacter:]): Move this back to the Mac-only category, as we won't be running this
-          test on iOS for other reasons.
-
-2018-07-02  Brady Eidson  <beidson@apple.com>
-
-        Unreviewed followup to:
-        Crash notifying observers of responsiveness state change
-
-        * TestWebKitAPI/Tests/WebKit/ResponsivenessTimerCrash.mm: Need the C-SPI for this test.
-
-2018-07-02  Brady Eidson  <beidson@apple.com>
-
-        Crash notifying observers of responsiveness state change
-        <rdar://problem/41267796> and https://bugs.webkit.org/show_bug.cgi?id=187262
-
-        Reviewed by Tim Horton.
-
-        * TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj:
-        * TestWebKitAPI/Tests/WebKit/ResponsivenessTimerCrash.mm: Added.
-        (-[RTObserver observeValueForKeyPath:ofObject:change:context:]):
-        (TestWebKitAPI::TEST):
-        * TestWebKitAPI/cocoa/TestWKWebView.h:
-        * TestWebKitAPI/cocoa/TestWKWebView.mm:
-        (-[TestWKWebView typeCharacter:]):
-        (-[TestWKWebView hostWindow]):
-
-2018-07-02  Michael Catanzaro  <mcatanzaro@igalia.com>
-
-        [WPE][GTK] flatpakutils.py should respect set-webkit-configuration build type
-        https://bugs.webkit.org/show_bug.cgi?id=187218
-
-        Reviewed by Philippe Normand.
-
-        Using the webkitpy Config module, if the build type is not passed on the command line, we
-        read it from WebKitBuild/Configuration and respect it.
-
-        * flatpak/flatpakutils.py:
-        (WebkitFlatpak.__init__):
-        (WebkitFlatpak.clean_args):
-
-2018-07-02  Zan Dobersek  <zdobersek@igalia.com>
-
-        [WPE] Add touch support to WindowViewBackend
-        https://bugs.webkit.org/show_bug.cgi?id=187245
-
-        Reviewed by Carlos Garcia Campos.
-
-        Add touch input support to the WindowViewBackend implementation,
-        plugging into the Wayland protocol in order to get properly notified
-        about these events, and then dispatching them against the appropriate
-        wpe_view_backend object so that WebKit can process them.
-
-        * wpe/backends/ViewBackend.cpp:
-        (WPEToolingBackends::ViewBackend::dispatchInputTouchEvent):
-        * wpe/backends/ViewBackend.h:
-        * wpe/backends/WindowViewBackend.cpp:
-        * wpe/backends/WindowViewBackend.h:
-
-2018-07-02  Rob Buis  <rbuis@igalia.com>
-
-        [GTK] ASSERTION FAILED: url == m_string in UserAgentQuirks test
-        https://bugs.webkit.org/show_bug.cgi?id=186838
-
-        This URL constructor expects canonicalized input, so fix the passed URL. I missed
-        this one some weeks ago since I had an older checkout.
-
-        Reviewed by Michael Catanzaro.
-
-        * TestWebKitAPI/Tests/WebCore/UserAgentQuirks.cpp:
-        (TestWebKitAPI::TEST):
-
-2018-07-01  Wenson Hsieh  <wenson_hsieh@apple.com>
-
-        [macOS] Text replacements that end with symbols are expanded immediately
-        https://bugs.webkit.org/show_bug.cgi?id=187225
-        <rdar://problem/41112433>
-
-        Reviewed by Darin Adler.
-
-        Adds testing support for mocking NSSpellChecker's text checking results. See below for more details, and the new
-        layout test for an example of its usage.
-
-        * DumpRenderTree/DumpRenderTree.xcodeproj/project.pbxproj:
-        * DumpRenderTree/TestRunner.cpp:
-        (setSpellCheckerTextReplacementsCallback):
-
-        Add bindings support in TestRunner to specify a set of text replacement mappings. Each entry in the dictionary
-        maps a string representing an input to the NSSpellChecker to some information describing the spell checking
-        result that LayoutTestSpellChecker will return.
-
-        (TestRunner::staticFunctions):
-        * DumpRenderTree/TestRunner.h:
-        * DumpRenderTree/mac/DumpRenderTree.mm:
-        (resetWebViewToConsistentStateBeforeTesting):
-
-        Restore the original shared NSSpellchecker before transitioning to the next layout test, if needed.
-
-        * DumpRenderTree/mac/DumpRenderTreeSpellChecker.mm: Removed.
-        * DumpRenderTree/mac/TestRunnerMac.mm:
-        (TestRunner::setSpellCheckerLoggingEnabled):
-        (TestRunner::setSpellCheckerTextReplacements):
-
-        These testRunner methods ensure that `-[NSSpellChecker sharedSpellChecker]` is swizzled to return our mock
-        LayoutTestSpellChecker instance before calling into it.
-
-        * DumpRenderTree/win/TestRunnerWin.cpp:
-        (TestRunner::setSpellCheckerTextReplacements):
-        * TestRunnerShared/cocoa/LayoutTestSpellChecker.h: Renamed from Tools/DumpRenderTree/mac/DumpRenderTreeSpellChecker.h.
-        * TestRunnerShared/cocoa/LayoutTestSpellChecker.mm: Added.
-
-        Moved DumpRenderTreeSpellChecker to LayoutTestSpellChecker, and made it compile for both WebKitTestRunner
-        (modern WebKit) and DumpRenderTree (legacy WebKit) by putting it in TestRunnerShared.
-
-        (existingGlobalLayoutTestSpellChecker):
-        (ensureGlobalLayoutTestSpellChecker):
-        (stringForCorrectionResponse):
-        (nsTextCheckingType):
-        (-[LayoutTestTextCheckingResult initWithType:range:replacement:]):
-        (-[LayoutTestTextCheckingResult range]):
-        (-[LayoutTestTextCheckingResult resultType]):
-        (-[LayoutTestTextCheckingResult replacementString]):
-        (-[LayoutTestTextCheckingResult description]):
-
-        LayoutTestTextCheckingResult represents a fake NSTextCheckingResult containing spell checking results supplied
-        by the layout test.
-
-        (+[LayoutTestSpellChecker installIfNecessary]):
-        (+[LayoutTestSpellChecker uninstallAndReset]):
-
-        Helper methods to begin and end swizzling the shared NSSpellChecker.
-
-        (-[LayoutTestSpellChecker reset]):
-
-        Resets the state of the LayoutTestSpellChecker (this entails clearing the fake replacements dictionary and
-        turning off logging for recorded spellchecking responses).
-
-        (-[LayoutTestSpellChecker replacements]):
-        (-[LayoutTestSpellChecker setReplacements:]):
-        (-[LayoutTestSpellChecker setReplacementsFromJSObject:inContext:]):
-
-        Helper method to take a `JSObjectRef` supplied by the test runner and transform it into a map of spell checking
-        string inputs to LayoutTestTextCheckingResults.
-
-        (-[LayoutTestSpellChecker checkString:range:types:options:inSpellDocumentWithTag:orthography:wordCount:]):
-
-        Consult the text replacement map and bail early if a match is found, before calling into real NSSpellChecker
-        logic to perform spellchecking.
-
-        (-[LayoutTestSpellChecker recordResponse:toCorrection:forWord:language:inSpellDocumentWithTag:]):
-
-        Reimplement the original functionality in DumpRenderTreeSpellChecker to make `-recordResponse:…` print to stdout
-        by overriding the method and printing if the `-spellCheckerLoggingEnabled` property has been set to YES.
-
-        * WebKitTestRunner/WebKitTestRunner.xcodeproj/project.pbxproj:
-
-2018-07-01  Thibault Saunier  <tsaunier@igalia.com>
-
-        [WPE][GTK] Fix retrieving backtrace from within flatpak sandbox in test runner
-        https://bugs.webkit.org/show_bug.cgi?id=187232
-
-        Reviewed by Michael Catanzaro.
-
-        The place where host tmp folder is mounted changed to /run/host/ we
-        need to take that into account.
-
-        Also explicitely close temporary files as relying on garbage collection to
-        remove them is a bad idea.
-
-        Also minor optimization avoiding to compile regex for each backtrace
-        retrieval.
-
-        * Scripts/webkitpy/port/linux_get_crash_log.py:
-        (GDBCrashLogGenerator):
-        (GDBCrashLogGenerator._get_tmp_file_name):
-        (GDBCrashLogGenerator._get_trace_from_systemd):
-
-2018-06-30  Daniel Bates  <dabates@apple.com>
-
-        test-webkitpy never rebuilds lldbWebKitTester
-        https://bugs.webkit.org/show_bug.cgi?id=187229
-
-        Reviewed by David Kilzer.
-
-        Have test-webkitpy always call script build-lldbwebkittester to build lldbWebKitTester, if needed.
-
-        Currently test-webkitpy only calls script build-lldbwebkittester if lldbWebKitTester does not
-        exist in the build products directory for the active configuration (e.g. Debug); => we do not
-        build lldbWebKitTester again if its source changes (say, as a result of hacking on lldbWebKitTester).
-        Instead we should have test-webkitpy unconditionally call build-lldbwebkittester to ensure
-        that we have an up-to-date build of lldbWebKitTester.
-
-        * Scripts/webkitpy/test/main.py:
-        (Tester._run_tests):
-        (_find_lldb_webkit_tester): Deleted.
-
-2018-06-30  Michael Catanzaro  <mcatanzaro@igalia.com>
-
-        [CMake] build-webkit should be able to enable address sanitizer
-        https://bugs.webkit.org/show_bug.cgi?id=187222
-
-        Reviewed by Daniel Bates.
-
-        If 'set-webkit-configuration --asan' has been run, then build-webkit should pass
-        -DENABLE_ADDRESS_SANITIZER=ON to CMake, so that it actually does something.
-
-        This also works for build-jsc.
-
-        * Scripts/webkitdirs.pm:
-        (generateBuildSystemFromCMakeProject):
-
-2018-06-30  Thibault Saunier  <tsaunier@igalia.com>
-
-        Make check_flatpak() not verbose when only trying to run command inside flatpak
-        https://bugs.webkit.org/show_bug.cgi?id=187187
-
-        Reviewed by Michael Catanzaro.
-
-        * flatpak/flatpakutils.py:
-        (run_in_sandbox_if_available):
-
-2018-06-29  Daniel Bates  <dabates@apple.com>
-
-        build-webkit: Perl "use of uninitialized value $previousContents"
-        https://bugs.webkit.org/show_bug.cgi?id=185776
-
-        Reviewed by Michael Catanzaro.
-
-        Fixes an issue where reading an empty cached argument file would cause
-        Perl "uninitialized value" warnings of the form:
-            Use of uninitialized value $previousContents in chomp at C:/WebKit-BuildWorker/wincairo-wkl-debug/build/Tools/Scripts/webkitdirs.pm line 1969.
-            Use of uninitialized value $previousContents in string ne at C:/WebKit-BuildWorker/wincairo-wkl-debug/build/Tools/Scripts/webkitdirs.pm line 1972.
-
-        * Scripts/webkitdirs.pm:
-        (isCachedArgumentfileOutOfDate):
-
-2018-06-29  Daniel Bates  <dabates@apple.com>
-
-        Perl uninitialized value $isEnabled when running build-jsc using a CMake build
-        https://bugs.webkit.org/show_bug.cgi?id=187208
-
-        Reviewed by Tim Horton.
-
-        Share logic for computing the feature flags to enable in CMake with both script
-        build-webkit and script build-jsc instead of duplicating it. This also fixes a
-        bug in the version of this logic in build-jsc that was inadvertently not updated
-        in r222394. In r222394 we removed the notion of a default value for a feature flag
-        listed in webkitperl::FeatureList.
-
-        We keep the current behavior of build-jsc and leave it up to the build system
-        to determine whether to enable or disable ENABLE_EXPERIMENTAL_FEATURES.
-
-        * Scripts/build-jsc:
-        (buildMyProject):
-        (cMakeArgsFromFeatures): Deleted.
-        * Scripts/build-webkit:
-        (cMakeArgsFromFeatures): Deleted; moved to webkitdirs.pm.
-        * Scripts/webkitdirs.pm:
-        (cmakeArgsFromFeatures): Moved code from script build-webkit so that it can be shared
-        with build-jsc.
-
-2018-06-29  Tim Horton  <timothy_horton@apple.com>
-
-        Add -apple-color-filter and system appearance toggles to MiniBrowser
-        https://bugs.webkit.org/show_bug.cgi?id=187210
-
-        Reviewed by Simon Fraser.
-
-        * MiniBrowser/mac/AppDelegate.m:
-        (defaultConfiguration):
-        * MiniBrowser/mac/SettingsController.h:
-        * MiniBrowser/mac/SettingsController.m:
-        (-[SettingsController _populateMenu]):
-        (-[SettingsController validateMenuItem:]):
-        (-[SettingsController toggleAppleColorFilterEnabled:]):
-        (-[SettingsController appleColorFilterEnabled]):
-        (-[SettingsController toggleUseSystemAppearance:]):
-        (-[SettingsController useSystemAppearance]):
-        * MiniBrowser/mac/WK1BrowserWindowController.m:
-        (-[WK1BrowserWindowController didChangeSettings]):
-        * MiniBrowser/mac/WK2BrowserWindowController.m:
-        (-[WK2BrowserWindowController didChangeSettings]):
-
-2018-06-29  Aditya Keerthi  <akeerthi@apple.com>
-
-        [macOS] Do not crash if there is an attempt to copy a file URL to the clipboard
-        https://bugs.webkit.org/show_bug.cgi?id=187183
-
-        Reviewed by Wenson Hsieh.
-
-        Added a 'runSingly' option to the TestOptions struct. Setting this option to true
-        will force a new PlatformWebView to be created before running a test. This ensures
-        that any state set by previous tests are not preserved.
-
-        One example of the importance of having this ability is in the case where we want
-        to test functionality that deals with read access to files. If a test were to
-        load a valid file URL, universal read access will be granted in WebProcessProxy.
-        This prevents us from testing cases that rely on us not have universal read
-        access. Now, if we write the test using the 'runSingly' option, any state set
-        by previous tests is cleared. Consequently, our test will behave as expected.
-
-        * WebKitTestRunner/PlatformWebView.h:
-        (WTR::PlatformWebView::viewSupportsOptions const):
-        * WebKitTestRunner/TestController.cpp:
-        (WTR::updateTestOptionsFromTestHeader):
-        * WebKitTestRunner/TestOptions.h:
-        (WTR::TestOptions::hasSameInitializationOptions const):
-
-2018-06-29  Daniel Bates  <dabates@apple.com>
-
-        [lldb-webkit] Non-empty strings may be pretty-printed as empty
-        https://bugs.webkit.org/show_bug.cgi?id=187185
-
-        Reviewed by Simon Fraser.
-
-        For some reason lldb(1) sometimes has an issue accessing members of WTF::StringImplShape
-        via a WTF::StringImpl pointer (why?). Explicitly casting a WTF::StringImpl* to a
-        WTF::StringImplShape* before accessing such members makes LLDB happy.
-
-        I tried writing a test for this both for the LLVM project and to add to our lldb_webkit unit
-        tests to no avail. I have only been able to reproduce this bug sporadically during my WebCore/WebKit
-        debugging sessions so far.
-
-        * lldb/lldb_webkit.py:
-        (WTFStringImplProvider.__init__): Explicitly cast the WTF::StringImpl* to WTF::StringImplShape*.
-        (WTFStringImplProvider.get_data8): Update code now that we are directly accessing WTF::StringImplShape*.
-        (WTFStringImplProvider.get_data16): Ditto.
-
-2018-06-29  Chris Dumez  <cdumez@apple.com>
-
-        WebKitLegacy: Can trigger recursive loads triggering debug assertions
-        https://bugs.webkit.org/show_bug.cgi?id=187121
-        <rdar://problem/41259430>
-
-        Reviewed by Brent Fulgham.
-
-        Add API test coverage.
-
-        * TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj:
-        * TestWebKitAPI/Tests/mac/StartLoadInDidFailProvisionalLoad.mm: Added.
-        (-[StartLoadInDidFailProvisionalLoadDelegate webView:didFailProvisionalLoadWithError:forFrame:]):
-        (-[StartLoadInDidFailProvisionalLoadDelegate webView:didFinishLoadForFrame:]):
-        (TestWebKitAPI::TEST):
-
-2018-06-29  Lucas Forschler  <lforschler@apple.com>
-
-        Teach bisect-builds to retrieve supported platforms from the rest api.
-        https://bugs.webkit.org/show_bug.cgi?id=187195
-        
-        This patch updates bisect-builds to use v2_2 of the API. v2_2 adds two api endpoints: /platforms, and /minified-platforms.
-        These endpoints scan a dynamoDB database, returning the identifier key for all available platforms.
-        The identifier keys are in the format: platform-architecture-configuration.
-
-        Reviewed by Aakash Jain.
-
-        * Scripts/bisect-builds:
-        (get_api_archive_url): refactored from get_api_url.
-        (get_platforms): helper function to retrieve from the platforms api.
-        (minified_platforms): returns a list of minified platforms currently in the database
-        (unminified_platforms): returns a list of unminified platforms currently in the database
-        (is_supported_platform): checks if the platform-architecture-configuration combo is supported
-        (fetch_revision_list): refactored to use get_api_archive_url
-        (get_api_url): Deleted/refactored to get_api_archive_url.
-
-2018-06-29  Keith Miller  <keith_miller@apple.com>
-
-        run-jsc should print when jsc exits with non-zero status
-        https://bugs.webkit.org/show_bug.cgi?id=187192
-
-        Reviewed by Saam Barati.
-
-        * Scripts/run-jsc:
-
-2018-06-29  Aakash Jain  <aakash_jain@apple.com>
-
-        [ews-build] Add timeout to webkitpy and webkitperl tests
-        https://bugs.webkit.org/show_bug.cgi?id=187191
-
-        Reviewed by Lucas Forschler.
-
-        * BuildSlaveSupport/ews-build/steps.py:
-        (RunWebKitPerlTests.__init__): Set timeout of 2 minutes.
-        (RunWebKitPyTests.__init__): Ditto.
-        * BuildSlaveSupport/ews-build/steps_unittest.py: Updated unit-tests.
-
-2018-06-29  Carlos Alberto Lopez Perez  <clopez@igalia.com>
-
-        [WPE]: Fix exception handling when flatpak is not installed
-        https://bugs.webkit.org/show_bug.cgi?id=186771
-
-        Unreviewed followup-fix after r233362
-
-        This was causing failures when flatpak is not installed.
-
-        * flatpak/flatpakutils.py:
-        (check_flatpak):
-
-2018-06-29  Ross Kirsling  <ross.kirsling@sony.com>
-
-        [JSCOnly] Restore Windows build.
-        https://bugs.webkit.org/show_bug.cgi?id=187127
-
-        Reviewed by Michael Catanzaro.
-
-        * Scripts/build-jsc:
-        Fix condition for disabling FTL JIT on Windows.
-
-        * Scripts/webkitdirs.pm:
-        (executableProductDir):
-        JSCOnly should still use bin64 on Windows (for consistency with full WebKit build).
-        (determineIsWin64):
-        JSCOnly already defaults to 64-bit on Windows with Ninja -- ensure that this is true even with MSBuild.