Timeline

Mar 13, 2019:

10:44 PM Changeset in webkit [242936] by benjamin@webkit.org

• 3 edits in trunk/Source/WebCore

Fix the argument type of RenderView::resumePausedImageAnimationsIfNeeded()
​https://bugs.webkit.org/show_bug.cgi?id=195659

Reviewed by Saam Barati.

The two callers of resumePausedImageAnimationsIfNeeded() both get the IntRect
as a reference. The rect was going on the stack then used as a reference again.

• rendering/RenderView.cpp:

(WebCore::RenderView::resumePausedImageAnimationsIfNeeded):

• rendering/RenderView.h:

9:49 PM Changeset in webkit [242935] by Alan Bujtas

• 5 edits in trunk/Source

[ContentChangeObserver] Stop content observation when content calls preventDefault() on touch events
​https://bugs.webkit.org/show_bug.cgi?id=195724
<rdar://problem/48873456>

Reviewed by Simon Fraser.

Source/WebCore:

Call willNotProceedWithClick on preventDefault. This is very similar to the long press case.

• page/ios/ContentChangeObserver.cpp:

(WebCore::willNotProceedWithClick):
(WebCore::ContentChangeObserver::didRecognizeLongPress):
(WebCore::ContentChangeObserver::didPreventDefaultForEvent):

• page/ios/ContentChangeObserver.h:

Source/WebKit:

• WebProcess/WebCoreSupport/ios/WebChromeClientIOS.mm:

(WebKit::WebChromeClient::didPreventDefaultForEvent):

8:59 PM Changeset in webkit [242934] by Wenson Hsieh

• 5 edits in trunk

Make -[_WKAttachment setFileWrapper:contentType:completion:] robust when given a nil completion handler
​https://bugs.webkit.org/show_bug.cgi?id=195725
<rdar://problem/48545062>

Reviewed by Tim Horton.

Source/WebKit:

Add a missing nil check before invoking the given completionHandler in the case where the attachment is invalid.
Tested by augmenting WKAttachmentTests.SetFileWrapperForPDFImageAttachment to exercise this scenario.

• UIProcess/API/APIAttachment.cpp:

(API::Attachment::invalidate):

Additionally make sure that an invalidated _WKAttachment is also considered to be disconnected.

• UIProcess/API/Cocoa/_WKAttachment.mm:

(-[_WKAttachment setFileWrapper:contentType:completion:]):

Tools:

Test that we don't crash when changing the file wrapper of an invalid attachment, if the given completion
handler is nil.

• TestWebKitAPI/Tests/WebKitCocoa/WKAttachmentTests.mm:

(TestWebKitAPI::TEST):

8:07 PM Changeset in webkit [242933] by Dewei Zhu

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

Test freshness page should improve the ability to correlating issues from same builder.
​https://bugs.webkit.org/show_bug.cgi?id=195242

Reviewed by Ryosuke Niwa.

Added the ability to highlight indicators with same builder when mouse is hovering over one indicator.
This is a very useful visualization for correlating issues specific to a builder.
Added tooltip with latest build link when hovering over an indicator.

• public/v3/components/freshness-indicator.js:

(FreshnessIndicator): Removed 'summary' field as it's no longer needed.
Added 'highlighted' field.
(FreshnessIndicator.prototype.update): Added 'highlighted' argument.
(FreshnessIndicator.prototype.didConstructShadowTree): Make indicator to dispatch mouse enter and leave
messages so that UI can highlight corresponding cells.
(FreshnessIndicator.prototype.render):
(FreshnessIndicator.cssTemplate):

• public/v3/pages/test-freshness-page.js: Added tooltip to show latest build time and build link.

Added logic to manually compute table body height.
(TestFreshnessPage):
(TestFreshnessPage.prototype.didConstructShadowTree):
(TestFreshnessPage.prototype._fetchTestResults):
(TestFreshnessPage.prototype.render):
(TestFreshnessPage.prototype._renderTooltip):
(TestFreshnessPage.prototype._constructTableCell):
(TestFreshnessPage.cssTemplate):

7:46 PM Changeset in webkit [242932] by commit-queue@webkit.org

• 5 edits
  1 add in trunk

Add utility function to allow easy reverse range-based iteration of a container
​https://bugs.webkit.org/show_bug.cgi?id=195542

Patch by Sam Weinig <​sam@webkit.org> on 2019-03-13
Reviewed by Antti Koivisto.

Source/WTF:

Add functions to create an IteratorRange<T> that will iterate a container backwards. It
works with any container that is compatible with std::rbegin() and std::rend(). It is
expected to be used in conjunction with range-based for-loops like so:

for (auto& value : WTF::makeReversedRange(myContainer))

...

• wtf/IteratorRange.h:

(WTF::makeReversedRange):

Tools:

• TestWebKitAPI/CMakeLists.txt:
• TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj:
• TestWebKitAPI/Tests/WTF/IteratorRange.cpp: Added.

(TestWebKitAPI::TEST):
Add test to ensure WTF::makeReversedRange() works correctly and uses the correct types.

6:42 PM Changeset in webkit [242931] by Justin Fan

• 37 edits
  3 copies in trunk

[Web GPU] Updates to GPUCommandBuffer for new GPUCommandQueue concept
​https://bugs.webkit.org/show_bug.cgi?id=195083
<rdar://problem/48423591>

Reviewed by Dean Jackson.

Source/WebCore:

WebGPUCommandBuffer now represents a completed GPUCommandBuffer that can only be used in queue submits. The previous WebGPUCommandBuffer
is now WebGPUCommandEncoder.

Affected Web GPU tests updated to match new API.

New files and symbols:

• CMakeLists.txt:
• DerivedSources.make:
• Sources.txt:
• WebCore.xcodeproj/project.pbxproj:
• bindings/js/WebCoreBuiltinNames.h:

Implement new WebGPUCommandBuffer, now just a DOM object carrier for a finished GPUCommandBuffer:

• Modules/webgpu/WebGPUCommandBuffer.cpp:

(WebCore::WebGPUCommandBuffer::create):
(WebCore::WebGPUCommandBuffer::WebGPUCommandBuffer):
(WebCore::WebGPUBufferCopyView::tryCreateGPUBufferCopyView const): Deleted.
(WebCore::WebGPUTextureCopyView::tryCreateGPUTextureCopyView const): Deleted.
(WebCore::WebGPUCommandBuffer::beginRenderPass): Deleted.
(WebCore::WebGPUCommandBuffer::copyBufferToBuffer): Deleted.
(WebCore::WebGPUCommandBuffer::copyBufferToTexture): Deleted.
(WebCore::WebGPUCommandBuffer::copyTextureToBuffer): Deleted.
(WebCore::WebGPUCommandBuffer::copyTextureToTexture): Deleted.

• Modules/webgpu/WebGPUCommandBuffer.h:

(WebCore::WebGPUCommandBuffer::commandBuffer):
(WebCore::WebGPUCommandBuffer::commandBuffer const): Deleted.

• Modules/webgpu/WebGPUCommandBuffer.idl:

Rename old WebGPUCommandBuffer to WebGPUCommandEncoder:

• Modules/webgpu/WebGPUCommandEncoder.cpp: Copied from Source/WebCore/Modules/webgpu/WebGPUCommandBuffer.cpp.

(WebCore::WebGPUBufferCopyView::tryCreateGPUBufferCopyView const):
(WebCore::WebGPUTextureCopyView::tryCreateGPUTextureCopyView const):
(WebCore::WebGPUCommandEncoder::create):
(WebCore::WebGPUCommandEncoder::WebGPUCommandEncoder):
(WebCore::WebGPUCommandEncoder::beginRenderPass):
(WebCore::WebGPUCommandEncoder::copyBufferToBuffer):
(WebCore::WebGPUCommandEncoder::copyBufferToTexture):
(WebCore::WebGPUCommandEncoder::copyTextureToBuffer):
(WebCore::WebGPUCommandEncoder::copyTextureToTexture):
(WebCore::WebGPUCommandEncoder::finish): Added. "Completes" this and invalidates it. Returns its GPUCommandBuffer, ready for submission.

• Modules/webgpu/WebGPUCommandEncoder.h: Copied from Source/WebCore/Modules/webgpu/WebGPUCommandBuffer.h.
• Modules/webgpu/WebGPUCommandEncoder.idl: Copied from Source/WebCore/Modules/webgpu/WebGPUCommandBuffer.idl.
• Modules/webgpu/WebGPUDevice.cpp:

(WebCore::WebGPUDevice::createCommandEncoder const): Renamed fom createCommandBuffer. Now returns non-nullable.
(WebCore::WebGPUDevice::createCommandBuffer const): Deleted.

• Modules/webgpu/WebGPUDevice.h:
• Modules/webgpu/WebGPUDevice.idl:
• Modules/webgpu/WebGPUProgrammablePassEncoder.cpp:

(WebCore::WebGPUProgrammablePassEncoder::WebGPUProgrammablePassEncoder):
(WebCore::WebGPUProgrammablePassEncoder::endPass): No longer returns the original WebGPUCommandBuffer.
(WebCore::WebGPUProgrammablePassEncoder::setBindGroup const):
(WebCore::WebGPUProgrammablePassEncoder::setPipeline):

• Modules/webgpu/WebGPUProgrammablePassEncoder.h:
• Modules/webgpu/WebGPUProgrammablePassEncoder.idl:
• Modules/webgpu/WebGPUQueue.cpp:

(WebCore::WebGPUQueue::submit): Replace unnecessary rvalue reference parameter.

• Modules/webgpu/WebGPUQueue.h:
• Modules/webgpu/WebGPUQueue.idl:
• Modules/webgpu/WebGPURenderPassEncoder.cpp:

(WebCore::WebGPURenderPassEncoder::create):
(WebCore::WebGPURenderPassEncoder::WebGPURenderPassEncoder):
(WebCore::WebGPURenderPassEncoder::setVertexBuffers):
(WebCore::WebGPURenderPassEncoder::draw):
(WebCore::WebGPURenderPassEncoder::passEncoder const): Now returns a pointer since it is properly backed by a RefPtr.

• Modules/webgpu/WebGPURenderPassEncoder.h:
• Modules/webgpu/WebGPUSwapChain.cpp:

(WebCore::WebGPUSwapChain::getCurrentTexture): No longer invalidates m_currentTexture. Doh!

• platform/graphics/gpu/GPUCommandBuffer.h: Missing includes for the *CopyView structs.
• platform/graphics/gpu/GPUDevice.cpp:

(WebCore::GPUDevice::tryCreateCommandBuffer const): Renamed from createCommandBuffer.
(WebCore::GPUDevice::createCommandBuffer): Deleted.

• platform/graphics/gpu/GPUDevice.h:
• platform/graphics/gpu/cocoa/GPUCommandBufferMetal.mm:

(WebCore::GPUCommandBuffer::tryCreate): Renamed from create.
(WebCore::GPUCommandBuffer::create): Deleted.

LayoutTests:

Update existing Web GPU tests for GPUCommandEncoder and new GPUCommandBuffer.

• webgpu/blit-commands.html:
• webgpu/buffer-command-buffer-races.html:
• webgpu/buffer-resource-triangles.html:
• webgpu/command-buffers-expected.txt:
• webgpu/command-buffers.html:
• webgpu/depth-enabled-triangle-strip.html:
• webgpu/js/webgpu-functions.js:

(beginBasicRenderPass):

• webgpu/render-command-encoding.html:
• webgpu/simple-triangle-strip.html:
• webgpu/texture-triangle-strip.html:
• webgpu/vertex-buffer-triangle-strip.html:

6:40 PM Changeset in webkit [242930] by timothy@apple.com

• 2 edits in trunk/Source/WebKit

REGRESSION (r242908): 'NSInvalidArgumentException', reason: '+[PKPaymentMerchantSession count]: unrecognized selector sent to class 0x1c0fae060'
​https://bugs.webkit.org/show_bug.cgi?id=195720

Reviewed by Andy Estes.

Add back decode(Decoder& decoder, Class allowedClass) for Apple Pay code.

• Shared/Cocoa/ArgumentCodersCocoa.h:

(IPC::decode): Added.

6:35 PM Changeset in webkit [242929] by Devin Rousso

• 2 edits in trunk/Source/WebInspectorUI

Web Inspector: Debugger: pausing in an inline script on a page with a URL query creates an Extra Script
​https://bugs.webkit.org/show_bug.cgi?id=195705
<rdar://problem/48853820>

Reviewed by Antoine Quint.

• UserInterface/Models/Script.js:

(WI.Script.prototype._resolveResource):
If the page's URL has a query parameter, the payload we receive for any inline <script>s
doesn't include the query parameter as part of its URL. As such, if there isn't an existing
resource with a URL that exactly matches the URL of the script and if the URL of the main
resource for the script's target starts with the URL of the script, we assume that the
script "belongs" to the target's main resource and associate the script with it as such.

6:09 PM Changeset in webkit [242928] by Caio Lima

• 6 edits in trunk/Source/JavaScriptCore

[JSC] CodeBlock::visitChildren is reporting extra memory even when its JITCode is singleton
​https://bugs.webkit.org/show_bug.cgi?id=195638

Reviewed by Mark Lam.

This patch introduces a m_isShared flag to track whether the
JITCode is shared between many CodeBlocks. This flag is used in
CodeBlock::setJITCode and CodeBlock::visitChildren to avoid
reporting duplicated extra memory for singleton JITCodes.
With those changes, we now stop counting singleton LLIntEntrypoints
as extra memory, since they are declared as static variables. This
change can potentially avoid unecessary GC pressure, because
extra memory is used by Heap::updateAllocationLimits() to update Heap
limits.
Even though it is hard to show performance difference for this change
(see results below), it is important to keep extra memory usage
correct. Otherwise, it can be a source of a complicated bug on
GC in the future.

Results from last run of Speedometer 2 comparing ToT and changes. We
collected those numbers running Minibrowser on a MacBook Pro 15-inch
with 2,6 GHz Intel Core i7. Both versions are with JIT disabled,
since these singleton JITCode are only used by this configuration:

Speedometer2 Run #1

ToT: 58.2 +- 1.1
changes: 57.9 +- 0.99

Speedometer2 Run #2

ToT: 58.5 +- 1.7
changes: 58.0 +- 1.5

Speedometer2 Run #2

ToT: 58.5 +- 0.99
changes: 57.1 +- 1.5

• bytecode/CodeBlock.cpp:

(JSC::CodeBlock::estimatedSize):
(JSC::CodeBlock::visitChildren):

• bytecode/CodeBlock.h:

(JSC::CodeBlock::setJITCode):

• jit/JITCode.cpp:

(JSC::JITCode::JITCode):
(JSC::JITCodeWithCodeRef::JITCodeWithCodeRef):
(JSC::DirectJITCode::DirectJITCode):
(JSC::NativeJITCode::NativeJITCode):

• jit/JITCode.h:

(JSC::JITCode::isShared const):

• llint/LLIntEntrypoint.cpp:

(JSC::LLInt::setFunctionEntrypoint):
(JSC::LLInt::setEvalEntrypoint):
(JSC::LLInt::setProgramEntrypoint):
(JSC::LLInt::setModuleProgramEntrypoint):

6:05 PM Changeset in webkit [242927] by Kocsen Chung

• 1 copy in tags/Safari-607.2.1

Tag Safari-607.2.1.

5:50 PM Changeset in webkit [242926] by timothy_horton@apple.com

• 2 edits in trunk/Source/WebKit

Stop using some deprecated SPI in WKDrawingView
​https://bugs.webkit.org/show_bug.cgi?id=195706
<rdar://problem/48062599>

Reviewed by Wenson Hsieh.

• UIProcess/ios/WKDrawingView.mm:

(-[WKDrawingView renderedDrawing]):
(-[WKDrawingView PNGRepresentation]):
(-[WKDrawingView loadDrawingFromPNGRepresentation:]):

5:38 PM Changeset in webkit [242925] by Ryan Haddad

• 2 edits in trunk/Source/WebKitLegacy/win

Unreviewed attempt to fix the Windows build after r242920.

• WebCoreSupport/WebFrameLoaderClient.h:

5:37 PM Changeset in webkit [242924] by aboya@igalia.com

• 3 edits in trunk/LayoutTests

[GTK] Unreviewed test gardening
​https://bugs.webkit.org/show_bug.cgi?id=195717

• platform/gtk/TestExpectations:
• platform/wpe/TestExpectations:

5:17 PM Changeset in webkit [242923] by timothy@apple.com

• 2 edits in trunk/Source/WebKit

Unreviewed speculative build fix for watchOS after r242908.

• Shared/Cocoa/ArgumentCodersCocoa.mm:

(IPC::decodeObject):

5:06 PM Changeset in webkit [242922] by mark.lam@apple.com

• 3 edits in trunk/JSTests

Gardening: reducing the variants on 2 tests to avoid timing out on JSC Debug queue.
​https://bugs.webkit.org/show_bug.cgi?id=195415

Not reviewed.

Changed these tests to only run the default configuration.
The ftl-no-cjit-validate-sampling-profiler variant was timing out.
There's no strong need to run this test on that variant.

• stress/dfg-to-string-on-int-does-gc.js:
• stress/dfg-to-string-on-string-or-string-object-does-not-gc.js:

4:46 PM Changeset in webkit [242921] by Alan Bujtas

• 3 edits in trunk/Source/WebCore

[WeakPtr] RenderListMarker::m_listItem should be a WeakPtr
​https://bugs.webkit.org/show_bug.cgi?id=195704
<rdar://problem/48486278>

Reviewed by Simon Fraser.

• rendering/RenderListMarker.cpp:

(WebCore::RenderListMarker::RenderListMarker):
(WebCore::RenderListMarker::paint):
(WebCore::RenderListMarker::layout):
(WebCore::RenderListMarker::updateContent):
(WebCore::RenderListMarker::computePreferredLogicalWidths):
(WebCore::RenderListMarker::lineHeight const):
(WebCore::RenderListMarker::baselinePosition const):
(WebCore::RenderListMarker::suffix const):
(WebCore::RenderListMarker::isInside const):
(WebCore::RenderListMarker::getRelativeMarkerRect):

• rendering/RenderListMarker.h:

4:37 PM Changeset in webkit [242920] by dino@apple.com

• 31 edits
  2 adds in trunk

Block all plugins smaller than 5x5px
​https://bugs.webkit.org/show_bug.cgi?id=195702
<rdar://problem/28435204>

Reviewed by Sam Weinig.

Source/WebCore:

Block all plugins that are smaller than a threshold, in this case
5px x 5px. Other browsers have implemented this for a while, and now
that we have Intersection Observers, small plugins are no longer
necessary.

Test: plugins/small-plugin-blocked.html

• en.lproj/Localizable.strings: New message for a small plugin.
• platform/LocalizedStrings.cpp:

(WebCore::pluginTooSmallText):

• platform/LocalizedStrings.h:

• html/HTMLPlugInElement.cpp: Helper function for Internals testing.

(WebCore::HTMLPlugInElement::isBelowSizeThreshold const):

• html/HTMLPlugInElement.h:

• loader/EmptyClients.cpp: Removed an unused function.

(WebCore::EmptyFrameLoaderClient::recreatePlugin): Deleted.

• loader/EmptyFrameLoaderClient.h:
• loader/FrameLoaderClient.h:

• page/Settings.yaml: Add flag for new feature.

• rendering/RenderEmbeddedObject.cpp: New unavailability reason for

embedded objects.
(WebCore::unavailablePluginReplacementText):

• rendering/RenderEmbeddedObject.h:

(WebCore::RenderEmbeddedObject::pluginUnavailabilityReason const):

• testing/Internals.cpp: Helper function for testing.

(WebCore::Internals::pluginIsBelowSizeThreshold):

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

Source/WebKit:

Block all plugins that are smaller than a threshold, in this case
5x5px. Other browsers have implemented this for a while, and now
that we have Intersection Observers, small plugins are no longer
necessary.

• Shared/WebPreferences.yaml: New setting for this feature.

• UIProcess/WebPageProxy.cpp: Handle new unavailability type.

(WebKit::WebPageProxy::unavailablePluginButtonClicked):

• WebProcess/WebCoreSupport/WebChromeClient.cpp:

(WebKit::WebChromeClient::shouldUnavailablePluginMessageBeButton const):

• WebProcess/WebCoreSupport/WebFrameLoaderClient.cpp: Removed this function

as it was never being called.
(WebKit::WebFrameLoaderClient::recreatePlugin): Deleted.

• WebProcess/WebCoreSupport/WebFrameLoaderClient.h:

• WebProcess/WebPage/WebPage.cpp:

(WebKit::pluginIsSmall): Checks the size of the plugin.
(WebKit::WebPage::createPlugin): If the plugin is too small, stop it from
launching.

Source/WebKitLegacy/mac:

Removed a function that was never being called.

• WebCoreSupport/WebFrameLoaderClient.h:
• WebCoreSupport/WebFrameLoaderClient.mm:

(WebFrameLoaderClient::recreatePlugin): Deleted.

LayoutTests:

New test for some small plugins. Updated existing tests
to create plugins bigger than a threshold if necessary.

• plugins/clicking-missing-plugin-fires-delegate.html:
• plugins/destroy-stream-twice.html:
• plugins/npruntime/npruntime.html:
• plugins/object-embed-plugin-scripting.html:
• plugins/small-plugin-blocked-expected.txt: Added.
• plugins/small-plugin-blocked.html: Added.
• platform/mac-wk1/TestExpectations: Skip new test on WK1.

4:36 PM Changeset in webkit [242919] by Alan Bujtas

• 4 edits
  2 adds in trunk

Use RenderBox::previousSiblingBox/nextSiblingBox in RenderMultiColumnFlow
​https://bugs.webkit.org/show_bug.cgi?id=195701
<rdar://problem/48448658>

Reviewed by Simon Fraser.

Source/WebCore:

It's safer to use existing RenderBox functions to get sibling boxes.

Test: fast/ruby/crash-when-paginated-ruby.html

• rendering/RenderMultiColumnFlow.cpp:

(WebCore::RenderMultiColumnFlow::nextColumnSetOrSpannerSiblingOf):
(WebCore::RenderMultiColumnFlow::previousColumnSetOrSpannerSiblingOf):

LayoutTests:

• fast/ruby/crash-when-paginated-ruby-expected.txt: Added.
• fast/ruby/crash-when-paginated-ruby.html: Added.

4:24 PM Changeset in webkit [242918] by Keith Rollin

• 16 edits
  1 add in trunk/Source

Add support for new StagedFrameworks layout
​https://bugs.webkit.org/show_bug.cgi?id=195543

Reviewed by Alexey Proskuryakov.

Source/JavaScriptCore:

When creating the WebKit layout for out-of-band Safari/WebKit updates,
use an optional path prefix when called for.

• Configurations/Base.xcconfig:

Source/ThirdParty/libwebrtc:

When creating the WebKit layout for out-of-band Safari/WebKit updates,
use an optional path prefix when called for.

• Configurations/Base.xcconfig:

Source/WebCore:

When creating the WebKit layout for out-of-band Safari/WebKit updates,
use an optional path prefix when called for.

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

• Configurations/WebCore.xcconfig:

Source/WebCore/PAL:

When creating the WebKit layout for out-of-band Safari/WebKit updates,
use an optional path prefix when called for.

• Configurations/PAL.xcconfig:

Source/WebInspectorUI:

When creating the WebKit layout for out-of-band Safari/WebKit updates,
use an optional path prefix when called for.

Opportunistic cleanup: remove unused
OTHER_LDFLAGS_VERSIONED_FRAMEWORK_PATH variable, which otherwise would
have needlessly been updated to also incorporate the new prefix.

• Configurations/Base.xcconfig:
• Configurations/WebKitTargetConditionals.xcconfig: Added.

Source/WebKit:

When creating the WebKit layout for out-of-band Safari/WebKit updates,
use an optional path prefix when called for.

Update the dyld_env path in OTHER_LDFLAGS_VERSIONED_FRAMEWORK_PATH to
also understand about this layout.

• Configurations/BaseTarget.xcconfig:

Source/WebKitLegacy/mac:

When creating the WebKit layout for out-of-band Safari/WebKit updates,
use an optional path prefix when called for.

• Configurations/WebKitLegacy.xcconfig:

Source/WTF:

Opportunistic cleanup: remove unused JAVASCRIPTCORE_FRAMEWORKS_DIR
variable.

• Configurations/Base.xcconfig:

4:18 PM Changeset in webkit [242917] by Wenson Hsieh

• 4 edits
  2 adds in trunk

Fix an edge case where HTMLFormElement::removeFormElement is invoked twice with the same element
​https://bugs.webkit.org/show_bug.cgi?id=195663
<rdar://problem/48576391>

Reviewed by Ryosuke Niwa.

Source/WebCore:

Currently, it's possible for HTMLFormControlElement's destructor to be reentrant. This may happen if the form
control element is ref'd while carrying out its destructor's logic. This may happen in two places in
HTMLFormControlElement (didChangeForm and resetDefaultButton), both of which actually don't require ensuring a
protected reference to the form control element since they should never result in any script execution.

To fix the bug, convert these strong references into raw pointers, and add ScriptDisallowedScope to ensure that
we don't change these codepaths in the future, such that they trigger arbitrary script execution.

Test: fast/forms/remove-associated-element-after-gc.html

• html/HTMLFormControlElement.cpp:

(WebCore::HTMLFormControlElement::didChangeForm):

• html/HTMLFormElement.cpp:

(WebCore::HTMLFormElement::resetDefaultButton):

LayoutTests:

Add a layout test to exercise the scenario described in the WebCore ChangeLog.

• fast/forms/remove-associated-element-after-gc-expected.txt: Added.
• fast/forms/remove-associated-element-after-gc.html: Added.

4:07 PM Changeset in webkit [242916] by yoshiaki.jitsukawa@sony.com

• 3 edits in trunk/Source/bmalloc

[bmalloc] Use MADV_FREE on FreeBSD
​https://bugs.webkit.org/show_bug.cgi?id=195665

Reviewed by Geoffrey Garen.

• bmalloc/BPlatform.h:

Introduce BOS_FREEBSD, which is equivalent to WTF_OS_FREEBSD

• bmalloc/VMAllocate.h:

(bmalloc::vmDeallocatePhysicalPages):

Use MADV_FREE instead of MADV_DONTNEED if BOS(FREEBSD), since on FreeBSD,
unlike on Linux, MADV_DONTNEED doesn't let the OS discard the contents of
the pages.

3:26 PM Changeset in webkit [242915] by dbates@webkit.org

• 3 edits in trunk/Source/WebCore

Remove some unnecessary !USE(UIKIT_KEYBOARD_ADDITIONS) guards
​https://bugs.webkit.org/show_bug.cgi?id=195703

Reviewed by Tim Horton.

Remove out-of-date comment and unncessary !USE(UIKIT_KEYBOARD_ADDITIONS) guards. Following
r240604 we now make use of WebCore::windowsKeyCodeForCharCode() even for hardware key events
when USE(UIKIT_KEYBOARD_ADDITIONS) is enabled.

No functionality changed. So, no new tests.

• platform/ios/KeyEventIOS.mm:

(WebCore::windowsKeyCodeForCharCode):

• platform/ios/WebEvent.mm:

(normalizedStringWithAppKitCompatibilityMapping):

3:09 PM Changeset in webkit [242914] by Nikita Vasilyev

• 3 edits
  2 adds in trunk

REGRESSION(r240946): Web Inspector: Styles: removing selected property doesn't update overridden status
​https://bugs.webkit.org/show_bug.cgi?id=195389
<rdar://problem/48658929>

Reviewed by Matt Baker.

Source/WebInspectorUI:

• UserInterface/Models/DOMNodeStyles.js:

(WI.DOMNodeStyles.prototype.changeStyleText):
Call DOMNodeStyles.prototype.refresh after the callback. No updates
to CSSStyleDeclaration happen until the callback is called.

LayoutTests:

• inspector/css/overridden-property-expected.txt: Added.
• inspector/css/overridden-property.html: Added.

3:06 PM Changeset in webkit [242913] by Simon Fraser

• 31 edits
  1 copy
  6 adds in trunk

Source/WebCore:
Scrolling tree should reposition non-stacking order descendents of overflow:scroll
​https://bugs.webkit.org/show_bug.cgi?id=195608

Reviewed by Zalan Bujtas.

Step 1: add scrolling tree positioning nodes classes (but don't create them yet).

Add Scrolling{State,Tree}PositionedNode to track composited layers that have to be repositioned when
an async overflow:scroll scrolls. There are two instances in which this is necessary, reflected by
the values of ScrollPositioningBehavior:

ScrollPositioningBehavior::Moves - a composited layer whose containing block chain includes an

async overflow scroller, but whose composited (i.e. z-order) parent is outside of the overflow.
When the overflow scrolls, this layer has to move along with the overflow.

ScrollPositioningBehavior::Stationary - a composited layer whose containing block chain skips the

overflow scroller, but whose compositing (z-order) parent is the scroller, or inside the scroller.
This only applies to position:absolute, on, for example, an overflow:scroll ith opacity.

PositionedNodes are modeled after Fixed/Sticky nodes, with a new type of layout constraint just called LayoutConstraints.

This patch adds support for PositionedNodes in the scrolling trees, but RenderLayerCompositor::computeCoordinatedPositioningForLayer()
is just a stub so the new node types aren't created yet.

RenderLayerBacking stores a ScrollingNodeID for the positioning role (just like the other roles). Since the Positioning
role is about position relative to ancestors, a node with both Positioning and FrameHosting or Scrolling roles treats
the Positioning node as the parent of the other types. A node should never have both Positioning and ViewportConstrained roles.

Test: scrollingcoordinator/scrolling-tree/positioned-nodes.html

• Sources.txt:
• SourcesCocoa.txt:
• WebCore.xcodeproj/project.pbxproj:
• page/scrolling/AsyncScrollingCoordinator.cpp:

(WebCore::AsyncScrollingCoordinator::setPositionedNodeGeometry):
(WebCore::AsyncScrollingCoordinator::setRelatedOverflowScrollingNodes):

• page/scrolling/AsyncScrollingCoordinator.h:
• page/scrolling/ScrollingConstraints.cpp:

(WebCore::operator<<):

• page/scrolling/ScrollingConstraints.h:

(WebCore::LayoutConstraints::LayoutConstraints):
(WebCore::LayoutConstraints::operator== const):
(WebCore::LayoutConstraints::operator!= const):
(WebCore::LayoutConstraints::alignmentOffset const):
(WebCore::LayoutConstraints::setAlignmentOffset):
(WebCore::LayoutConstraints::layerPositionAtLastLayout const):
(WebCore::LayoutConstraints::setLayerPositionAtLastLayout):
(WebCore::LayoutConstraints::scrollPositioningBehavior const):
(WebCore::LayoutConstraints::setScrollPositioningBehavior):

• page/scrolling/ScrollingCoordinator.cpp:

(WebCore::operator<<):

• page/scrolling/ScrollingCoordinator.h:

(WebCore::ScrollingCoordinator::setPositionedNodeGeometry):
(WebCore::ScrollingCoordinator::setRelatedOverflowScrollingNodes):

• page/scrolling/ScrollingCoordinatorTypes.h:
• page/scrolling/ScrollingStateNode.h:

(WebCore::ScrollingStateNode::isPositionedNode const):

• page/scrolling/ScrollingStatePositionedNode.cpp: Added.

(WebCore::ScrollingStatePositionedNode::create):
(WebCore::ScrollingStatePositionedNode::ScrollingStatePositionedNode):
(WebCore::ScrollingStatePositionedNode::clone):
(WebCore::ScrollingStatePositionedNode::setAllPropertiesChanged):
(WebCore::ScrollingStatePositionedNode::setRelatedOverflowScrollingNodes):
(WebCore::ScrollingStatePositionedNode::updateConstraints):
(WebCore::ScrollingStatePositionedNode::dumpProperties const):

• page/scrolling/ScrollingStatePositionedNode.h: Added.
• page/scrolling/ScrollingStateTree.cpp:

(WebCore::ScrollingStateTree::createNode):

• page/scrolling/ScrollingTree.cpp:

(WebCore::ScrollingTree::updateTreeFromStateNode):

• page/scrolling/ScrollingTree.h:
• page/scrolling/ScrollingTreeNode.h:

(WebCore::ScrollingTreeNode::isPositionedNode const):

• page/scrolling/cocoa/ScrollingTreePositionedNode.h: Copied from Source/WebCore/page/scrolling/cocoa/ScrollingTreeStickyNode.h.
• page/scrolling/cocoa/ScrollingTreePositionedNode.mm: Added.

(WebCore::ScrollingTreePositionedNode::create):
(WebCore::ScrollingTreePositionedNode::ScrollingTreePositionedNode):
(WebCore::ScrollingTreePositionedNode::~ScrollingTreePositionedNode):
(WebCore::ScrollingTreePositionedNode::commitStateBeforeChildren):
(WebCore::ScrollingTreePositionedNode::applyLayerPositions):
(WebCore::ScrollingTreePositionedNode::relatedNodeScrollPositionDidChange):
(WebCore::ScrollingTreePositionedNode::dumpProperties const):

• page/scrolling/cocoa/ScrollingTreeStickyNode.h:
• page/scrolling/mac/ScrollingTreeMac.cpp:

(ScrollingTreeMac::createScrollingTreeNode):

• platform/ScrollTypes.h:
• rendering/RenderLayerBacking.cpp:

(WebCore::RenderLayerBacking::coordinatedScrollingRoles const):
(WebCore::RenderLayerBacking::detachFromScrollingCoordinator):

• rendering/RenderLayerBacking.h:
• rendering/RenderLayerCompositor.cpp:

(WebCore::RenderLayerCompositor::computeCoordinatedPositioningForLayer const):
(WebCore::scrollCoordinationRoleForNodeType):
(WebCore::RenderLayerCompositor::updateScrollCoordinationForLayer):
(WebCore::RenderLayerCompositor::updateScrollingNodeForViewportConstrainedRole):
(WebCore::RenderLayerCompositor::updateScrollingNodeLayers):
(WebCore::RenderLayerCompositor::updateScrollingNodeForPositioningRole):

• rendering/RenderLayerCompositor.h:

Source/WebKit:
Scrolling tree should reposition non-stacking order descendents of overflow:scroll.
​https://bugs.webkit.org/show_bug.cgi?id=195608

Reviewed by Zalan Bujtas.

Step 1: add scrolling tree positioning nodes classes (but don't create them yet).

Add Scrolling{State,Tree}PositionedNode to track composited layers that have to be repositioned when
an async overflow:scroll scrolls. There are two instances in which this is necessary, reflected by
the values of ScrollPositioningBehavior:

ScrollPositioningBehavior::Moves - a composited layer whose containing block chain includes an

async overflow scroller, but whose composited (i.e. z-order) parent is outside of the overflow.
When the overflow scrolls, this layer has to move along with the overflow.

ScrollPositioningBehavior::Stationary - a composited layer whose containing block chain skips the

overflow scroller, but whose compositing (z-order) parent is the scroller, or inside the scroller.
This only applies to position:absolute, on, for example, an overflow:scroll ith opacity.

PositionedNodes are modeled after Fixed/Sticky nodes, with a new type of layout constraint just called LayoutConstraints.

This patch adds support for PositionedNodes in the scrolling trees, but RenderLayerCompositor::computeCoordinatedPositioningForLayer()
is just a stub so the new node types aren't created yet.

RenderLayerBacking stores a ScrollingNodeID for the positioning role (just like the other roles). Since the Positioning
role is about position relative to ancestors, a node with both Positioning and FrameHosting or Scrolling roles treats
the Positioning node as the parent of the other types. A node should never have both Positioning and ViewportConstrained roles.

• Shared/RemoteLayerTree/RemoteScrollingCoordinatorTransaction.cpp:

(ArgumentCoder<ScrollingStateFrameScrollingNode>::encode):
(ArgumentCoder<ScrollingStatePositionedNode>::encode):
(ArgumentCoder<ScrollingStatePositionedNode>::decode):
(WebKit::encodeNodeAndDescendants):
(WebKit::RemoteScrollingCoordinatorTransaction::decode):
(WebKit::dump):

• Shared/WebCoreArgumentCoders.cpp:

(IPC::ArgumentCoder<LayoutConstraints>::encode):
(IPC::ArgumentCoder<LayoutConstraints>::decode):
(IPC::ArgumentCoder<StickyPositionViewportConstraints>::decode):

• Shared/WebCoreArgumentCoders.h:
• UIProcess/RemoteLayerTree/RemoteScrollingCoordinatorProxy.cpp:

(WebKit::RemoteScrollingCoordinatorProxy::connectStateNodeLayers):

• UIProcess/RemoteLayerTree/RemoteScrollingTree.cpp:

(WebKit::RemoteScrollingTree::createScrollingTreeNode):

• UIProcess/RemoteLayerTree/ios/RemoteScrollingCoordinatorProxyIOS.mm:

(WebKit::RemoteScrollingCoordinatorProxy::connectStateNodeLayers):

LayoutTests:
Scrolling tree should reposition non-stacking order descendents of overflow:scroll
​https://bugs.webkit.org/show_bug.cgi?id=195608

Reviewed by Zalan Bujtas.

The results of this test will change when we enable positioned nodes. It tests various
combinations of overflow and stacking.

• platform/ios-wk2/scrollingcoordinator/scrolling-tree/positioned-nodes-expected.txt: Added.
• scrollingcoordinator/scrolling-tree/positioned-nodes-expected.txt: Added.
• scrollingcoordinator/scrolling-tree/positioned-nodes.html: Added.

3:05 PM Changeset in webkit [242912] by mark.lam@apple.com

• 4 edits in trunk/Source/JavaScriptCore

Remove unneeded --tradeDestructorBlocks option.
​https://bugs.webkit.org/show_bug.cgi?id=195698
<rdar://problem/39681388>

Reviewed by Yusuke Suzuki.

There's no reason why we would ever want --tradeDestructorBlocks to be false.

Also, there was an assertion in BlockDirectory::endMarking() for when
(!Options::tradeDestructorBlocks() && needsDestruction()). This assertion is
outdated because the BlockDirectory's m_empty set used to mean the set of all
blocks that have no live (as in not reachable by GC) objects and dead objects
also do not require destructors to be called on them. The current meaning of
m_empty is that it is the set of all blocks that have no live objects,
independent of whether they needs destructors to be called on them or not.
The assertion is no longer valid for the new meaning of m_empty as m_empty may
now contain destructible blocks. This assertion is now removed as part of this
patch.

• heap/BlockDirectory.cpp:

(JSC::BlockDirectory::endMarking):

• heap/LocalAllocator.cpp:

(JSC::LocalAllocator::tryAllocateWithoutCollecting):

• runtime/Options.h:

2:47 PM Changeset in webkit [242911] by youenn@apple.com

• 27 edits
  2 copies
  2 adds in trunk

Check IDB quota usage through QuotaManager
​https://bugs.webkit.org/show_bug.cgi?id=195302

Reviewed by Chris Dumez.

Source/WebCore:

For every write operation, compute an estimate size and check for quota before proceeding.
When proceeding, store the estimate size in a map.
If size of the database is to be computed when the task is not done,
the estimate size will be added to the current size of the databases.
At the end of the task, the estimate size is removed from the map,
and the databases size is refreshed.

This patch implements size estimation for write tasks.
Put/add operations might overestimate the size
when an old value will be replaced by a new value.
In that case, we do not substract the old value size since we do not know it.

This patch implements database opening by adding a fixed small cost,
as we do not know whether the database is new or not.

For the first IDB request, we have not computed the size of the database.
To do so, we need to go to a background thread and do that file size computation.
For that purpose, we add support for being-initialized quota user.
Quota manager is calling whenInitialized on its quota user and will
delay any quota check requests until its quota user is answering this callback.

For in process IDB, use the default storage quota per origin and do not increase it.
Future work should move it to NetworkProcess and implement some quota checking.

Cache API and IDB quota management are not yet fully unified.
If IDB is used on start-up, we should check for Cache API storage size.
Conversely, on Cache API first wite task, even if IDB is not being used,
we should compute the size of the IDB data for the given origin.

Test: http/tests/IndexedDB/storage-limit.https.html

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

(WebCore::IDBServer::IDBServer::create):
(WebCore::IDBServer::IDBServer::IDBServer):
(WebCore::IDBServer::m_quotaManagerGetter):
(WebCore::IDBServer::IDBServer::QuotaUser::QuotaUser):
(WebCore::IDBServer::IDBServer::QuotaUser::~QuotaUser):
(WebCore::IDBServer::IDBServer::QuotaUser::clearSpaceUsed):
(WebCore::IDBServer::IDBServer::QuotaUser::whenInitialized):
(WebCore::IDBServer::IDBServer::QuotaUser::initializeSpaceUsed):
(WebCore::IDBServer::IDBServer::quotaUser):
(WebCore::IDBServer::IDBServer::startComputingSpaceUsedForOrigin):
(WebCore::IDBServer::IDBServer::computeSpaceUsedForOrigin):
(WebCore::IDBServer::IDBServer::finishComputingSpaceUsedForOrigin):
(WebCore::IDBServer::IDBServer::requestSpace):
(WebCore::IDBServer::IDBServer::clearSpaceUsed):
(WebCore::IDBServer::IDBServer::setSpaceUsed):
(WebCore::IDBServer::IDBServer::increasePotentialSpaceUsed):
(WebCore::IDBServer::IDBServer::decreasePotentialSpaceUsed):

• Modules/indexeddb/server/IDBServer.h:

(WebCore::IDBServer::IDBServer::create):

• Modules/indexeddb/server/MemoryIDBBackingStore.cpp:

(WebCore::IDBServer::MemoryIDBBackingStore::databasesSizeForOrigin const):

• Modules/indexeddb/server/MemoryIDBBackingStore.h:
• Modules/indexeddb/server/SQLiteIDBBackingStore.cpp:

(WebCore::IDBServer::SQLiteIDBBackingStore::databasesSizeForFolder):
(WebCore::IDBServer::SQLiteIDBBackingStore::databasesSizeForOrigin const):
(WebCore::IDBServer::SQLiteIDBBackingStore::maximumSize const):

• Modules/indexeddb/server/SQLiteIDBBackingStore.h:
• Modules/indexeddb/server/UniqueIDBDatabase.cpp:

(WebCore::IDBServer::estimateSize):
(WebCore::IDBServer::UniqueIDBDatabase::UniqueIDBDatabase):
(WebCore::IDBServer::quotaErrorMessageName):
(WebCore::IDBServer::UniqueIDBDatabase::requestSpace):
(WebCore::IDBServer::UniqueIDBDatabase::performCurrentOpenOperation):
(WebCore::IDBServer::UniqueIDBDatabase::storeCallbackOrFireError):
(WebCore::IDBServer::UniqueIDBDatabase::createObjectStore):
(WebCore::IDBServer::UniqueIDBDatabase::createObjectStoreAfterQuotaCheck):
(WebCore::IDBServer::UniqueIDBDatabase::renameObjectStore):
(WebCore::IDBServer::UniqueIDBDatabase::renameObjectStoreAfterQuotaCheck):
(WebCore::IDBServer::UniqueIDBDatabase::createIndex):
(WebCore::IDBServer::UniqueIDBDatabase::createIndexAfterQuotaCheck):
(WebCore::IDBServer::UniqueIDBDatabase::renameIndex):
(WebCore::IDBServer::UniqueIDBDatabase::renameIndexAfterQuotaCheck):
(WebCore::IDBServer::UniqueIDBDatabase::putOrAdd):
(WebCore::IDBServer::UniqueIDBDatabase::putOrAddAfterQuotaCheck):
(WebCore::IDBServer::UniqueIDBDatabase::postDatabaseTaskReply):
(WebCore::IDBServer::UniqueIDBDatabase::immediateCloseForUserDelete):
(WebCore::IDBServer::UniqueIDBDatabase::updateSpaceUsedIfNeeded):
(WebCore::IDBServer::UniqueIDBDatabase::performErrorCallback):
(WebCore::IDBServer::UniqueIDBDatabase::performKeyDataCallback):

• Modules/indexeddb/server/UniqueIDBDatabase.h:

(WebCore::IDBServer::UniqueIDBDatabase::server):

• Modules/indexeddb/shared/InProcessIDBServer.cpp:

(WebCore::InProcessIDBServer::create):
(WebCore::InProcessIDBServer::quotaManager):
(WebCore::storageQuotaManagerGetter):
(WebCore::InProcessIDBServer::InProcessIDBServer):

• Modules/indexeddb/shared/InProcessIDBServer.h:
• loader/EmptyClients.cpp:
• storage/StorageQuotaManager.cpp:

(WebCore::StorageQuotaManager::addUser):
(WebCore::StorageQuotaManager::requestSpace):

• storage/StorageQuotaManager.h:

(WebCore::StorageQuotaManager::defaultQuota):
(WebCore::StorageQuotaManager::removeUser):

• storage/StorageQuotaUser.h:

(WebCore::StorageQuotaUser::whenInitialized):

Source/WebKit:

Set the quota manager getter for IDBServer at creation time.

• NetworkProcess/NetworkProcess.cpp:

(WebKit::NetworkProcess::createIDBServer):
(WebKit::NetworkProcess::idbServer):

• NetworkProcess/NetworkProcess.h:
• WebProcess/Databases/WebDatabaseProvider.cpp:

(WebKit::WebDatabaseProvider::idbConnectionToServerForSession):

Source/WebKitLegacy:

• Storage/WebDatabaseProvider.cpp:

(WebDatabaseProvider::idbConnectionToServerForSession):

LayoutTests:

Update IDB quota test according quota limit of 400ko.
Update WK1 test expectations to skip quota check tests.

• http/tests/IndexedDB/resources/shared.js: Added.
• http/tests/IndexedDB/resources/storage-limit.js: Added.
• http/tests/IndexedDB/storage-limit.https-expected.txt: Added.
• http/tests/IndexedDB/storage-limit.https.html: Added.
• platform/mac-wk1/TestExpectations:
• platform/win/TestExpectations:
• storage/indexeddb/resources/storage-limit.js:
• storage/indexeddb/storage-limit-expected.txt:

2:42 PM Changeset in webkit [242910] by dinfuehr@igalia.com

• 6 edits
  2 adds in trunk

String overflow when using StringBuilder in JSC::createError
​https://bugs.webkit.org/show_bug.cgi?id=194957

Reviewed by Mark Lam.

JSTests:

Add test string-overflow-createError-bulder.js that overflows
StringBuilder in notAFunctionSourceAppender. The second new test
string-overflow-createError-fit.js has an error message that doesn't
overflow, it still failed since the String's capacity can't be doubled.
Run test string-overflow-createError.js only in the default
configuration to reduce memory consumption when running the test
in all configurations on multiple CPUs in parallel.

• stress/string-overflow-createError-builder.js: Copied from JSTests/stress/string-overflow-createError.js.

(catch):

• stress/string-overflow-createError-fit.js: Copied from JSTests/stress/string-overflow-createError.js.

(catch):

• stress/string-overflow-createError.js:

Source/JavaScriptCore:

StringBuilder in notAFunctionSourceAppender didn't check
for overflows but just failed.

• runtime/ExceptionHelpers.cpp:

(JSC::notAFunctionSourceAppender):

Source/WTF:

When calculating the new capacity of a StringBuilder object,
use a limit of MaxLength instead of MaxLength+1. Allocating
a string of size MaxLength+1 always fails. This means that expanding
a StringBuilder only worked when the newly doubled capacity is less or
equal to MaxLength.

• wtf/text/StringBuilder.cpp:

2:22 PM Changeset in webkit [242909] by Chris Dumez

• 4 edits in trunk/Source

Better build fix after r242901.

Reviewed by Jer Noble.

Source/WebCore:

• platform/audio/cocoa/MediaSessionManagerCocoa.mm:

(MediaSessionManagerCocoa::sessionWillBeginPlayback):
(MediaSessionManagerCocoa::updateNowPlayingInfo):

Source/WTF:

• wtf/Logger.h:

(WTF::LogArgument::toString):

2:18 PM Changeset in webkit [242908] by timothy@apple.com

• 16 edits
  2 deletes in trunk/Source

Consolidate ArgumentCodersMac and ArgumentCodersCocoa.
​https://bugs.webkit.org/show_bug.cgi?id=195636
rdar://problem/45055697

Reviewed by Ryosuke Niwa.

Source/WebCore:

• editing/DictionaryPopupInfo.h:

(WebCore::DictionaryPopupInfo::encodingRequiresPlatformData const): Added.

• editing/FontAttributes.h:

(WebCore::FontAttributes::encodingRequiresPlatformData const): Added.

Source/WebKit:

Merge the two similar encoders and decoders. This avoids issues where
one encoder could be used and the other decoder, which caused a crash.
It also stops handling NSAttributedString specifically and just uses
the NSSecureCoding path to handle more complex attributes.

Some WebCore encoders code needed to move to platform files, since
ArgumentCodersCocoa.h requires an ObjectiveC++ implementation to work.

• Shared/Cocoa/ArgumentCodersCocoa.h:

(IPC::encode):
(IPC::decode):
(IPC::ArgumentCoder<RetainPtr<T>>::encode):
(IPC::ArgumentCoder<RetainPtr<T>>::decode):

• Shared/Cocoa/ArgumentCodersCocoa.mm:

(IPC::typeFromObject):
(IPC::isSerializableFont):
(IPC::isSerializableValue):
(IPC::encodeObject):
(IPC::decodeObject):

• Shared/Cocoa/LoadParametersCocoa.mm:
• Shared/Cocoa/WebCoreArgumentCodersCocoa.mm:

(IPC::ArgumentCoder<WebCore::DictionaryPopupInfo>::encodePlatformData):
(IPC::ArgumentCoder<WebCore::DictionaryPopupInfo>::decodePlatformData):
(IPC::ArgumentCoder<WebCore::FontAttributes>::encodePlatformData):
(IPC::ArgumentCoder<WebCore::FontAttributes>::decodePlatformData):

• Shared/WebCoreArgumentCoders.cpp:

(IPC::ArgumentCoder<DictionaryPopupInfo>::encode):
(IPC::ArgumentCoder<DictionaryPopupInfo>::decode):
(IPC::ArgumentCoder<FontAttributes>::encode):
(IPC::ArgumentCoder<FontAttributes>::decode):

• Shared/WebCoreArgumentCoders.h:
• Shared/mac/ArgumentCodersMac.h: Removed.
• Shared/mac/ArgumentCodersMac.mm: Removed.
• Shared/mac/AttributedString.mm:

(WebKit::AttributedString::encode const):

• Shared/mac/ObjCObjectGraph.mm:
• SourcesCocoa.txt:
• WebKit.xcodeproj/project.pbxproj:

2:15 PM Changeset in webkit [242907] by aestes@apple.com

• 1 edit in trunk/Source/WTF/wtf/FeatureDefines.h

Try again to fix the Mac build.

• wtf/FeatureDefines.h:

2:05 PM Changeset in webkit [242906] by Chris Dumez

• 2 edits in trunk/Source/WebCore

Unreviewed build fix after r242901.

• platform/audio/cocoa/MediaSessionManagerCocoa.mm:

(MediaSessionManagerCocoa::updateNowPlayingInfo):

1:51 PM Changeset in webkit [242905] by Chris Dumez

• 26 edits in trunk

Use a ServiceWorker process per registrable domain
​https://bugs.webkit.org/show_bug.cgi?id=195649

Reviewed by Youenn Fablet.

Source/WebCore:

Use a ServiceWorker process per registrable domain instead of one per security origin. This is
more in line with PSON and avoids launching too many processes.

• page/ClientOrigin.h:

(WebCore::ClientOrigin::clientRegistrableDomain const):

• workers/service/server/SWServer.cpp:

(WebCore::SWServer::tryInstallContextData):
(WebCore::SWServer::serverToContextConnectionCreated):
(WebCore::SWServer::runServiceWorkerIfNecessary):
(WebCore::SWServer::markAllWorkersForRegistrableDomainAsTerminated):
(WebCore::SWServer::registerServiceWorkerClient):
(WebCore::SWServer::unregisterServiceWorkerClient):
(WebCore::SWServer::needsServerToContextConnectionForRegistrableDomain const):

• workers/service/server/SWServer.h:
• workers/service/server/SWServerToContextConnection.cpp:

(WebCore::SWServerToContextConnection::SWServerToContextConnection):
(WebCore::SWServerToContextConnection::~SWServerToContextConnection):
(WebCore::SWServerToContextConnection::connectionForRegistrableDomain):

• workers/service/server/SWServerToContextConnection.h:

(WebCore::SWServerToContextConnection::registrableDomain const):

• workers/service/server/SWServerWorker.cpp:

(WebCore::SWServerWorker::contextConnection):

• workers/service/server/SWServerWorker.h:

(WebCore::SWServerWorker::registrableDomain const):

Source/WebKit:

Use a ServiceWorker process per registrable domain instead of one per security origin. This is
more in line with PSON and avoids launching too many processes.

• NetworkProcess/NetworkProcess.cpp:

(WebKit::NetworkProcess::createNetworkConnectionToWebProcess):
(WebKit::NetworkProcess::connectionToContextProcessWasClosed):
(WebKit::NetworkProcess::needsServerToContextConnectionForRegistrableDomain const):
(WebKit::NetworkProcess::serverToContextConnectionForRegistrableDomain):
(WebKit::NetworkProcess::createServerToContextConnection):
(WebKit::NetworkProcess::swContextConnectionMayNoLongerBeNeeded):

• NetworkProcess/NetworkProcess.h:
• NetworkProcess/NetworkProcess.messages.in:
• NetworkProcess/ServiceWorker/WebSWServerConnection.cpp:

(WebKit::WebSWServerConnection::scheduleJobInServer):

• NetworkProcess/ServiceWorker/WebSWServerToContextConnection.cpp:

(WebKit::WebSWServerToContextConnection::WebSWServerToContextConnection):

• NetworkProcess/ServiceWorker/WebSWServerToContextConnection.h:
• UIProcess/Network/NetworkProcessProxy.cpp:

(WebKit::NetworkProcessProxy::getNetworkProcessConnection):
(WebKit::NetworkProcessProxy::establishWorkerContextConnectionToNetworkProcess):
(WebKit::NetworkProcessProxy::establishWorkerContextConnectionToNetworkProcessForExplicitSession):

• UIProcess/Network/NetworkProcessProxy.h:
• UIProcess/Network/NetworkProcessProxy.messages.in:
• UIProcess/ServiceWorkerProcessProxy.cpp:

(WebKit::ServiceWorkerProcessProxy::create):
(WebKit::ServiceWorkerProcessProxy::ServiceWorkerProcessProxy):
(WebKit::ServiceWorkerProcessProxy::getLaunchOptions):

• UIProcess/ServiceWorkerProcessProxy.h:
• UIProcess/WebProcessPool.cpp:

(WebKit::WebProcessPool::establishWorkerContextConnectionToNetworkProcess):
(WebKit::WebProcessPool::disconnectProcess):
(WebKit::WebProcessPool::updateProcessAssertions):

• UIProcess/WebProcessPool.h:

Tools:

Update API test coverage.

• TestWebKitAPI/Tests/WebKitCocoa/ServiceWorkerBasic.mm:

1:33 PM Changeset in webkit [242904] by aestes@apple.com

• 2 edits in trunk/Source

Try to fix the Mac build after r242356.

Source/WebKit:

• Shared/ApplePay/cocoa/WebPaymentCoordinatorProxyCocoa.mm:

(WebKit::WebPaymentCoordinatorProxy::platformPaymentRequest):

Source/WTF:

• wtf/FeatureDefines.h:

1:32 PM Changeset in webkit [242903] by Chris Dumez

• 13 edits in trunk

REGRESSION(PSON, r240660): Navigation over process boundary is flashy when using Cmd-left/right arrow to navigate
​https://bugs.webkit.org/show_bug.cgi?id=195684
<rdar://problem/48294714>

Reviewed by Antti Koivisto.

Source/WebCore:

The issue was caused by us failing to suspend the current page on navigation because the source and
target WebBackForwardListItem are identical. The source WebBackForwardListItem was wrong.

When a navigation is triggered by the WebContent process (and not the UIProcess), we create the Navigation
object in WebPageProxy::decidePolicyForNavigationAction(). For the navigation's targetItem, we use the
target item identifier provided by the WebContent process via the NavigationActionData. However,
for the source item, we would use the WebBackForwardList's currentItem in the UIProcess. The issue
is that the WebBackForwardList's currentItem usually has already been updated to be the target
item via a WebPageProxy::BackForwardGoToItem() synchronous IPC.

To avoid raciness and given that the current history management is fragile (as it is managed by
both the UIProcess and the WebProcess), I am now passing the source item identifier in
addition to the target item identifier in the NavigationActionData that is sent by the WebProcess.
This is a lot less error prone, the WebProcess knows more accurately which history items it is going
from and to.

• loader/FrameLoader.cpp:

(WebCore::FrameLoader::loadURLIntoChildFrame):
(WebCore::FrameLoader::loadDifferentDocumentItem):
(WebCore::FrameLoader::loadItem):
(WebCore::FrameLoader::retryAfterFailedCacheOnlyMainResourceLoad):

• loader/FrameLoader.h:
• loader/HistoryController.cpp:

(WebCore::HistoryController::recursiveGoToItem):

• loader/NavigationAction.cpp:

(WebCore::NavigationAction::setSourceBackForwardItem):

• loader/NavigationAction.h:

(WebCore::NavigationAction::sourceBackForwardItemIdentifier const):

Source/WebKit:

• Shared/NavigationActionData.cpp:

(WebKit::NavigationActionData::encode const):
(WebKit::NavigationActionData::decode):

• Shared/NavigationActionData.h:
• UIProcess/WebPageProxy.cpp:

(WebKit::WebPageProxy::decidePolicyForNavigationAction):
(WebKit::WebPageProxy::backForwardAddItem):

• WebProcess/WebCoreSupport/WebFrameLoaderClient.cpp:

(WebKit::WebFrameLoaderClient::dispatchDecidePolicyForNavigationAction):

Tools:

Add API test coverage.

• TestWebKitAPI/Tests/WebKitCocoa/ProcessSwapOnNavigation.mm:

1:27 PM Changeset in webkit [242902] by ysuzuki@apple.com

• 5 edits in trunk/Source/JavaScriptCore

[JSC] Move species watchpoint installation from ArrayPrototype to JSGlobalObject
​https://bugs.webkit.org/show_bug.cgi?id=195593

Reviewed by Keith Miller.

This patch moves watchpoints installation and watchpoints themselves from ArrayPrototype to JSGlobalObject because of the following two reasons.

1. ArrayPrototype configures finalizer because of std::unique_ptr<> for watchpoints. If we move them from ArrayPrototype to JSGlobalObject, we do not need to set finalizer. And we can avoid unnecessary WeakBlock allocation.

2. This code lazily configures watchpoints instead of setting watchpoints eagerly in JSGlobalObject::init. We would like to expand this mechanism to other watchpoints which are eagerly configured in JSGlobalObject::init. Putting these code in JSGlobalObject instead of scattering them in each XXXPrototype / XXXConstructor can encourage the reuse of the code.

• runtime/ArrayPrototype.cpp:

(JSC::ArrayPrototype::create):
(JSC::speciesWatchpointIsValid):
(JSC::ArrayPrototype::destroy): Deleted.
(JSC::ArrayPrototype::tryInitializeSpeciesWatchpoint): Deleted.
(JSC::ArrayPrototypeAdaptiveInferredPropertyWatchpoint::ArrayPrototypeAdaptiveInferredPropertyWatchpoint): Deleted.
(JSC::ArrayPrototypeAdaptiveInferredPropertyWatchpoint::handleFire): Deleted.

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

(JSC::JSGlobalObject::tryInstallArraySpeciesWatchpoint): Instead of using ArrayPrototypeAdaptiveInferredPropertyWatchpoint,
we use ObjectPropertyChangeAdaptiveWatchpoint. We create watchpoints after touching WatchpointSet since ObjectPropertyChangeAdaptiveWatchpoint
requires WatchpointSet is IsWatched state.

• runtime/JSGlobalObject.h:

1:18 PM Changeset in webkit [242901] by jer.noble@apple.com

• 13 edits
  1 add in trunk/Source

Add AggregateLogger, a Logger specialization for singleton classes.
​https://bugs.webkit.org/show_bug.cgi?id=195644

Reviewed by Eric Carlson.

Source/WebCore:

Convert debug logging over to release logging through the use of AggregateLogger.

• platform/audio/PlatformMediaSession.h:

(WebCore::PlatformMediaSession::client const):

• platform/audio/PlatformMediaSessionManager.cpp:

(WebCore::PlatformMediaSessionManager::PlatformMediaSessionManager):
(WebCore::PlatformMediaSessionManager::beginInterruption):
(WebCore::PlatformMediaSessionManager::endInterruption):
(WebCore::PlatformMediaSessionManager::addSession):
(WebCore::PlatformMediaSessionManager::removeSession):
(WebCore::PlatformMediaSessionManager::sessionWillBeginPlayback):
(WebCore::PlatformMediaSessionManager::sessionWillEndPlayback):
(WebCore::PlatformMediaSessionManager::setCurrentSession):
(WebCore::PlatformMediaSessionManager::applicationWillBecomeInactive const):
(WebCore::PlatformMediaSessionManager::applicationDidBecomeActive const):
(WebCore::PlatformMediaSessionManager::applicationDidEnterBackground const):
(WebCore::PlatformMediaSessionManager::applicationWillEnterForeground const):
(WebCore::PlatformMediaSessionManager::logChannel const):

• platform/audio/PlatformMediaSessionManager.h:
• platform/audio/cocoa/MediaSessionManagerCocoa.mm:

(MediaSessionManagerCocoa::updateSessionState):
(MediaSessionManagerCocoa::sessionWillBeginPlayback):
(MediaSessionManagerCocoa::removeSession):
(MediaSessionManagerCocoa::sessionWillEndPlayback):
(MediaSessionManagerCocoa::clientCharacteristicsChanged):
(MediaSessionManagerCocoa::updateNowPlayingInfo):

• platform/audio/ios/MediaSessionManagerIOS.mm:

(WebCore::MediaSessionManageriOS::resetRestrictions):
(WebCore::MediaSessionManageriOS::configureWireLessTargetMonitoring):

Source/WTF:

Add a new class, AggregateLogger, which will log messages to each of its aggregated loggers.

Drive-by fixes: allow "const void*" to be directly logged by converting the pointer to a hex string.

• WTF.xcodeproj/project.pbxproj:
• wtf/AggregateLogger.h: Added.

(WTF::AggregateLogger::create):
(WTF::AggregateLogger::addLogger):
(WTF::AggregateLogger::removeLogger):
(WTF::AggregateLogger::logAlways const):
(WTF::AggregateLogger::error const):
(WTF::AggregateLogger::warning const):
(WTF::AggregateLogger::info const):
(WTF::AggregateLogger::debug const):
(WTF::AggregateLogger::willLog const):
(WTF::AggregateLogger::AggregateLogger):
(WTF::AggregateLogger::log const):

• wtf/CMakeLists.h:
• wtf/Logger.cpp:

(WTF::>::toString):

• wtf/Logger.h:

1:03 PM Changeset in webkit [242900] by Truitt Savell

• 2 edits in trunk/LayoutTests

(r242595) Layout Tests in imported/w3c/web-platform-tests/html/semantics/embedded-content/media-elements/* are failing
​https://bugs.webkit.org/show_bug.cgi?id=195466

Unreviewed test gardening.

• platform/mac/TestExpectations:

12:53 PM Changeset in webkit [242899] by Chris Dumez

• 13 edits in trunk/Source

Drop legacy WebCore::toRegistrableDomain() utility function
​https://bugs.webkit.org/show_bug.cgi?id=195637

Reviewed by Geoffrey Garen.

Drop legacy toRegistrableDomain() / registrableDomainAreEqual() utility functions.
Update call sites to use modern RegistrableDomain type instead.

Source/WebCore:

• loader/CrossOriginAccessControl.cpp:

(WebCore::shouldCrossOriginResourcePolicyCancelLoad):

• loader/FrameLoader.cpp:

(WebCore::FrameLoader::setFirstPartyForCookies):

• page/DOMWindow.cpp:

(WebCore::DOMWindow::failedToRegisterDeviceMotionEventListener):

• platform/RegistrableDomain.h:

(WebCore::registrableDomainsAreEqual):

• platform/network/ResourceRequestBase.h:
• platform/network/cf/ResourceRequestCFNet.cpp:
• platform/network/cocoa/ResourceRequestCocoa.mm:

Source/WebKit:

• UIProcess/WebProcessPool.cpp:

(WebKit::WebProcessPool::processForNavigationInternal):

12:28 PM Changeset in webkit [242898] by Devin Rousso

• 2 edits in trunk/Source/WebInspectorUI

Web Inspector: Protocol Logging: log messages with backtrace if inspector^{2 is open
​https://bugs.webkit.org/show_bug.cgi?id=195687}

Reviewed by Joseph Pecoraro.

• UserInterface/Protocol/LoggingProtocolTracer.js:

(WI.LoggingProtocolTracer.prototype._processEntry):

12:19 PM Changeset in webkit [242897] by Devin Rousso

• 2 edits in trunk/Source/WebInspectorUI

Web Inspector: REGRESSION(r242737): unnecessary semicolon added when populating WI.TreeOutline stylesheet
​https://bugs.webkit.org/show_bug.cgi?id=195689

Reviewed by Joseph Pecoraro.

• UserInterface/Views/TreeOutline.js:

(WI.TreeOutline._generateStyleRulesIfNeeded):

12:04 PM Changeset in webkit [242896] by Joseph Pecoraro

• 5 edits in trunk

Web Inspector: Network - HAR Export duplicates blocked/send time if there was no dns/connect block
​https://bugs.webkit.org/show_bug.cgi?id=195655
<rdar://problem/48831152>

Reviewed by Devin Rousso.

Source/WebInspectorUI:

• UserInterface/Controllers/HARBuilder.js:

(WI.HARBuilder.timings):

LayoutTests:

• http/tests/inspector/network/har/har-basic-expected.txt:
• http/tests/inspector/network/har/har-basic.html:

11:45 AM Changeset in webkit [242895] by achristensen@apple.com

• 4 edits in trunk/Source/WebCore

Prevent checked_cf_cast crashes in ResourceResponse::platformCertificateInfo
​https://bugs.webkit.org/show_bug.cgi?id=195686

Reviewed by Tim Horton.

This covers up a type confusion bug on some OSes until rdar://problem/48853137 is resolved.

• SourcesCocoa.txt:
• WebCore.xcodeproj/project.pbxproj:
• platform/network/cocoa/ResourceResponseCocoa.mm:

11:25 AM Changeset in webkit [242894] by commit-queue@webkit.org

• 2 edits in trunk/Source/WebCore

[GStreamer][WebRTC] Add support for omxh264enc encoder
​https://bugs.webkit.org/show_bug.cgi?id=195676

Patch by Thibault Saunier <​tsaunier@igalia.com> on 2019-03-13
Reviewed by Philippe Normand.

• platform/mediastream/libwebrtc/GStreamerVideoEncoder.cpp:

(setup_omxh264enc):
(set_bitrate_bit_per_sec):
(gst_webrtc_video_encoder_class_init):

11:14 AM Changeset in webkit [242893] by graouts@webkit.org

• 3 edits in trunk

REGRESSION(r240634): Element::hasPointerCapture() passes a JS-controlled value directly into a HashMap as a key
​https://bugs.webkit.org/show_bug.cgi?id=195683
<rdar://problem/48659950>

Reviewed by Alex Christensen.

Source/WebCore:

While PointerID is defined as int32_t, we now use int64_t as the key of the HashMap mapping PointerID to CapturingData so that we use
a value outside of the int32_t range as a safe empty and removed values, allowing any int32_t to be provided through the API for
lookup in this HashMap.

Test: pointerevents/pointer-id-crash.html

• page/PointerCaptureController.h:

LayoutTests:

Add a new test which would crash in debug builds prior to this fix.

• pointerevents/pointer-id-crash-expected.txt: Added.
• pointerevents/pointer-id-crash.html: Added.

11:09 AM Changeset in webkit [242892] by commit-queue@webkit.org

• 3 edits
  1 add in trunk/Source/bmalloc

Fix testbmalloc build
​https://bugs.webkit.org/show_bug.cgi?id=195660

Patch by Sam Weinig <​sam@webkit.org> on 2019-03-13
Reviewed by Geoffrey Garen.

• bmalloc.xcodeproj/project.pbxproj:

Link Foundation in when building testbmalloc. Since bmalloc requires Foundation, and is a static
library, all clients of bmalloc are required to link it themselves.

• bmalloc/IsoPageInlines.h:
• bmalloc/StdLibExtras.h: Added.

(bmalloc::bitwise_cast):
Add bitwise_cast implementation, and use it in IsoPageInlines.h. It is a layering violation
to expect the one from WTF to be available, as seems to have been the case.

10:57 AM Changeset in webkit [242891] by beidson@apple.com

• 17 edits
  2 adds in trunk/Source

Take UnboundedNetworking assertion when a file upload is in progress.
​https://bugs.webkit.org/show_bug.cgi?id=195497

Reviewed by Geoff Garen.

Source/WebCore:

• platform/network/ResourceRequestBase.cpp:

(WebCore::ResourceRequestBase::hasUpload const):

• platform/network/ResourceRequestBase.h:

Source/WebKit:

This patch implements whole bunch of bookkeeping in both the Networking and UI processes.

The TLDR of that bookkeeping is:

• Whenever any uploads are in progress, take an assertion for both Networking and UI processes.
• Whenever a particular WebProcess has an upload in progress, take an assertion for it.

• NetworkProcess/NetworkConnectionToWebProcess.cpp:

(WebKit::NetworkConnectionToWebProcess::NetworkConnectionToWebProcess):
(WebKit::NetworkConnectionToWebProcess::setProcessIdentifier):
(WebKit::NetworkConnectionToWebProcess::setConnectionHasUploads):
(WebKit::NetworkConnectionToWebProcess::clearConnectionHasUploads):

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

• NetworkProcess/NetworkResourceLoadMap.cpp: Added.

(WebKit::NetworkResourceLoadMap::add):
(WebKit::NetworkResourceLoadMap::remove):
(WebKit::NetworkResourceLoadMap::get const):

• NetworkProcess/NetworkResourceLoadMap.h: Added.

(WebKit::NetworkResourceLoadMap::NetworkResourceLoadMap):
(WebKit::NetworkResourceLoadMap::isEmpty const):
(WebKit::NetworkResourceLoadMap::contains const):
(WebKit::NetworkResourceLoadMap::begin):
(WebKit::NetworkResourceLoadMap::values):

• NetworkProcess/NetworkSession.cpp:

• Scripts/webkit/messages.py:

• Sources.txt:

• UIProcess/Network/NetworkProcessProxy.cpp:

(WebKit::NetworkProcessProxy::takeUploadAssertion):
(WebKit::NetworkProcessProxy::clearUploadAssertion):

• UIProcess/Network/NetworkProcessProxy.h:

• UIProcess/WebProcessPool.cpp:

(WebKit::WebProcessPool::setWebProcessHasUploads):
(WebKit::WebProcessPool::clearWebProcessHasUploads):

• UIProcess/WebProcessPool.h:
• UIProcess/WebProcessPool.messages.in:

• WebKit.xcodeproj/project.pbxproj:

• WebProcess/WebProcess.cpp:

(WebKit::WebProcess::ensureNetworkProcessConnection):

10:53 AM Changeset in webkit [242890] by Kocsen Chung

• 7 edits in trunk/Source

Versioning.

10:48 AM Changeset in webkit [242889] by Chris Dumez

• 11 edits in trunk/Source/WebKit

Use new RegistrableDomain type in PSON code
​https://bugs.webkit.org/show_bug.cgi?id=195634

Reviewed by Youenn Fablet.

Use new RegistrableDomain type in PSON code instead of more error-prone String type.

• UIProcess/SuspendedPageProxy.cpp:

(WebKit::SuspendedPageProxy::SuspendedPageProxy):

• UIProcess/SuspendedPageProxy.h:
• UIProcess/WebProcessCache.cpp:

(WebKit::WebProcessCache::canCacheProcess const):
(WebKit::WebProcessCache::addProcessIfPossible):
(WebKit::WebProcessCache::takeProcess):
(WebKit::WebProcessCache::clearAllProcessesForSession):

• UIProcess/WebProcessCache.h:
• UIProcess/WebProcessPool.cpp:

(WebKit::WebProcessPool::addProcessToOriginCacheSet):
(WebKit::WebProcessPool::removeProcessFromOriginCacheSet):
(WebKit::WebProcessPool::processForNavigationInternal):
(WebKit::WebProcessPool::findReusableSuspendedPageProcess):
(WebKit::WebProcessPool::didCollectPrewarmInformation):
(WebKit::WebProcessPool::tryPrewarmWithDomainInformation):

• UIProcess/WebProcessPool.h:
• UIProcess/WebProcessProxy.cpp:

(WebKit::WebProcessProxy::canBeAddedToWebProcessCache const):
(WebKit::WebProcessProxy::maybeShutDown):
(WebKit::WebProcessProxy::didCollectPrewarmInformation):
(WebKit::WebProcessProxy::didStartProvisionalLoadForMainFrame):

• UIProcess/WebProcessProxy.h:

(WebKit::WebProcessProxy::registrableDomain const):

• UIProcess/WebProcessProxy.messages.in:
• WebProcess/WebProcess.cpp:

(WebKit::WebProcess::sendPrewarmInformation):

10:28 AM Changeset in webkit [242888] by bshafiei@apple.com

• 6 edits in branches/safari-607-branch/Source/WebKit

Apply patch. rdar://problem/48839383

Cherry-pick r242230. rdar://problem/48839383

Revert r232263: it caused processes to crash because process was suspended with locked file
​https://bugs.webkit.org/show_bug.cgi?id=195122
<rdar://problem/48444599>

Reviewed by Geoffrey Garen.

• NetworkProcess/NetworkProcess.cpp: (WebKit::NetworkProcess::NetworkProcess):
• NetworkProcess/NetworkProcess.h:
• Shared/WebSQLiteDatabaseTracker.cpp: (WebKit::WebSQLiteDatabaseTracker::WebSQLiteDatabaseTracker): (WebKit::WebSQLiteDatabaseTracker::hysteresisUpdated):
• Shared/WebSQLiteDatabaseTracker.h:
• WebProcess/WebProcess.cpp: (WebKit::m_webSQLiteDatabaseTracker): (WebKit::m_nonVisibleProcessCleanupTimer): Deleted.
• WebProcess/WebProcess.h:

git-svn-id: ​http://svn.webkit.org/repository/webkit/trunk@242230 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10:25 AM Changeset in webkit [242887] by Shawn Roberts

• 2 edits in trunk/LayoutTests

http/tests/websocket/tests/hybi/handshake-ok-with-legacy-sec-websocket-response-headers.html is a flaky failure on Mac WK2
​https://bugs.webkit.org/show_bug.cgi?id=173041

Unreviewed test gardening.

• platform/mac/TestExpectations: Marking as flaky until a fix lands

10:14 AM Changeset in webkit [242886] by youenn@apple.com

• 10 edits in trunk/Source

Enable libwebrtc logging control through WebCore
​https://bugs.webkit.org/show_bug.cgi?id=195658

Reviewed by Eric Carlson.

Source/ThirdParty/libwebrtc:

Add a callback to get access to libwebrtc log messages.

• Configurations/libwebrtc.iOS.exp:
• Configurations/libwebrtc.iOSsim.exp:
• Configurations/libwebrtc.mac.exp:
• Source/webrtc/rtc_base/logging.cc:
• Source/webrtc/rtc_base/logging.h:

Source/WebCore:

Add support for WebCore logging of libwebrtc messages.
This is controlled by WebRTC log channel state and level.
In case of private browsing mode, any logging is disabled.
This will stay for the lifetime of the process.
No change of behavior.

• Modules/mediastream/RTCPeerConnection.cpp:

(WebCore::RTCPeerConnection::create):

• platform/mediastream/libwebrtc/LibWebRTCProvider.cpp:

(WebCore::doReleaseLogging):
(WebCore::setLogging):
(WebCore::computeLogLevel):
(WebCore::initializePeerConnectionFactoryAndThreads):
(WebCore::LibWebRTCProvider::setEnableLogging):

• platform/mediastream/libwebrtc/LibWebRTCProvider.h:

10:11 AM Changeset in webkit [242885] by aakash_jain@apple.com

• 2 edits
  1 delete in trunk/Tools

[ews-app] Remove unused patch view
​https://bugs.webkit.org/show_bug.cgi?id=195669

Reviewed by Lucas Forschler.

• BuildSlaveSupport/ews-app/ews/urls.py:
• BuildSlaveSupport/ews-app/ews/views/patch.py: Removed.

10:03 AM Changeset in webkit [242884] by aakash_jain@apple.com

• 3 edits in trunk/Tools

[ews-app] Use Buildbot result code variables
​https://bugs.webkit.org/show_bug.cgi?id=195668

Reviewed by Alexey Proskuryakov.

• BuildSlaveSupport/ews-app/ews/common/buildbot.py:

(Buildbot): Added Buildbot result code variables.

• BuildSlaveSupport/ews-app/ews/views/statusbubble.py:

(StatusBubble._build_bubble): Used Buildbot result code variables.

9:08 AM Changeset in webkit [242883] by commit-queue@webkit.org

• 2 edits in trunk/Source/WebCore

[GStreamer][WebRTC] Do not sync encoder on the clock
​https://bugs.webkit.org/show_bug.cgi?id=195673

we should encode as fast as possible and totally ignore timestamp while
doing so.

Patch by Thibault Saunier <​tsaunier@igalia.com> on 2019-03-13
Reviewed by Philippe Normand.

• platform/mediastream/libwebrtc/GStreamerVideoEncoderFactory.cpp:

(WebCore::GStreamerVideoEncoder::InitEncode):

8:33 AM Changeset in webkit [242882] by commit-queue@webkit.org

• 3 edits in trunk/Tools

[Flatpak] Expand submodule recursively
​https://bugs.webkit.org/show_bug.cgi?id=195672

Building WPE with flatpak was impossible without that as it was raising an exception.
Also update wpebackend-fdo to match what is built in jhbuild

Patch by Thibault Saunier <​tsaunier@igalia.com> on 2019-03-13
Reviewed by Philippe Normand.

• flatpak/flatpakutils.py:

(expand_submodules_recurse):
(expand_manifest):

• flatpak/org.webkit.WPEModules.yaml:

8:28 AM Changeset in webkit [242881] by commit-queue@webkit.org

• 2 edits in trunk/Source/WebCore

[GStreamer][WebRTC]: Use codec setting video height/width as fallback
​https://bugs.webkit.org/show_bug.cgi?id=195675

Patch by Thibault Saunier <​tsaunier@igalia.com> on 2019-03-13
Reviewed by Philippe Normand.

In some cases the frame height and width is not set (not sure why/ in
what conditions but it happens) so make sure to get the information from
the VideoCodec when configuring the encoder.

• platform/mediastream/libwebrtc/GStreamerVideoDecoderFactory.cpp:

(WebCore::GStreamerVideoDecoder::GStreamerVideoDecoder):
(WebCore::GStreamerVideoDecoder::GetCapsForFrame):

4:27 AM Changeset in webkit [242880] by aakash_jain@apple.com

• 2 edits in trunk/Tools

[ews-app] status bubble should be hidden for certain builds
​https://bugs.webkit.org/show_bug.cgi?id=194597

Reviewed by Dewei Zhu.

• BuildSlaveSupport/ews-app/ews/views/statusbubble.py:

(StatusBubble._build_bubble):
(StatusBubble._should_show_bubble_for_build): Hide bubble for builds which were skipped
because the patch didn't have relevant changes.
(StatusBubble._should_show_bubble_for_queue): Hide bubbles for queues which are not deployed
in production yet.

2:47 AM Changeset in webkit [242879] by Carlos Garcia Campos

• 1 copy in releases/WebKitGTK/webkit-2.24.0

WebKitGTK 2.24.0

2:47 AM Changeset in webkit [242878] by Carlos Garcia Campos

• 4 edits in releases/WebKitGTK/webkit-2.24

Unreviewed. Update OptionsGTK.cmake and NEWS for 2.24.0 release

.:

• Source/cmake/OptionsGTK.cmake: Bump version numbers.

Source/WebKit:

• gtk/NEWS: Add release notes for 2.24.0.

2:25 AM Changeset in webkit [242877] by Carlos Garcia Campos

• 6 edits in releases/WebKitGTK/webkit-2.24/Source/WebCore

Merge r242793 - [GStreamer][v4l2] Synchronous video texture flushing support
​https://bugs.webkit.org/show_bug.cgi?id=195453

Reviewed by Xabier Rodriguez-Calvar.

The v4l2 video decoder currently requires that downstream users of
the graphics resources complete any pending draw call and release
resources before returning from the DRAIN query.

To accomplish this the player monitors the pipeline and whenever a
v4l2 decoder is added, synchronous video texture flushing support
is enabled. Additionally and for all decoder configurations, a
flush is performed before disposing of the player.

• platform/graphics/gstreamer/MediaPlayerPrivateGStreamer.cpp:

(WebCore::MediaPlayerPrivateGStreamer::playbinDeepElementAddedCallback):
Monitor elements added to the decodebin bin.
(WebCore::MediaPlayerPrivateGStreamer::decodebinElementAdded): Set
a flag if a v4l2 decoder was added in decodebin.
(WebCore::MediaPlayerPrivateGStreamer::createGSTPlayBin): Connect
to the deep-element-added signal so as to monitor pipeline
topology updates.

• platform/graphics/gstreamer/MediaPlayerPrivateGStreamer.h:
• platform/graphics/gstreamer/MediaPlayerPrivateGStreamerBase.cpp:

(WebCore::MediaPlayerPrivateGStreamerBase::~MediaPlayerPrivateGStreamerBase):
Flush video texture before disposing of the player.
(WebCore::MediaPlayerPrivateGStreamerBase::flushCurrentBuffer):
Synchronously flush if the pipeline contains a v4l2 decoder.
(WebCore::MediaPlayerPrivateGStreamerBase::createGLAppSink): Monitor push events only.

• platform/graphics/gstreamer/MediaPlayerPrivateGStreamerBase.h:
• platform/graphics/texmap/TextureMapperPlatformLayerProxy.cpp:

(WebCore::TextureMapperPlatformLayerProxy::pushNextBuffer): New
boolean flag used mostly to trigger synchronous flush conditions.
(WebCore::TextureMapperPlatformLayerProxy::dropCurrentBufferWhilePreservingTexture):
Optionally drop the current buffer in a synchronous manner. By
default the method keeps operating asynchronously.

• platform/graphics/texmap/TextureMapperPlatformLayerProxy.h:

2:25 AM Changeset in webkit [242876] by Carlos Garcia Campos

• 3 edits in releases/WebKitGTK/webkit-2.24/Tools

Merge r242637 - [GTK] Make Tools/gtkdoc python3 compatible
​https://bugs.webkit.org/show_bug.cgi?id=195359

Reviewed by Carlos Garcia Campos.

• gtkdoc/generate-gtkdoc:

ConfigParser was reworked in Python 3.2 so we have adapt the code to
work with Python 2 and 3.
(get_gtkdoc_module_paths):
The iteritems() was removed in Python 3, so let's use items() that's
available in Python 2 and 3.
(get_generator_for_config):

• gtkdoc/gtkdoc.py:

(GTKDoc._run_command):
The sys.stdout.write() is expecting str in Python 3 and not bytes
(that are coming from stdout.encode()). Use sys.stdout.buffer.write()
for passing the bytes there.

2:25 AM Changeset in webkit [242875] by Carlos Garcia Campos

• 2 edits in releases/WebKitGTK/webkit-2.24/Source/WebCore

Merge r242640 - GLContextEGL: desired EGL config should search for 8-bit components by default
​https://bugs.webkit.org/show_bug.cgi?id=195413

Reviewed by Carlos Garcia Campos.

The EGL config search in GLContextEGL should by default look for
RGBA8888 configurations while allowing RGB565 as an alternative.
This prevents from accidentally landing on an RGBA1010102
configuration that is available with some graphics stacks, and which is
not expected in e.g. window snapshotting that's done for layout test
output comparison.

• platform/graphics/egl/GLContextEGL.cpp:

(WebCore::GLContextEGL::getEGLConfig): EGL config search should by
default request 8-bit color channels.

2:25 AM Changeset in webkit [242874] by Carlos Garcia Campos

• 2 edits in releases/WebKitGTK/webkit-2.24/Source/WebCore/platform/gtk/po

Merge r242863 - [l10n] Updated Italian translation of WebKitGTK+
​https://bugs.webkit.org/show_bug.cgi?id=195620

Patch by Milo Casagrande <​milo@milo.name> on 2019-03-13
Rubber-stamped by Carlos Garcia Campos.

• it.po:

2:25 AM Changeset in webkit [242873] by Carlos Garcia Campos

• 2 edits in releases/WebKitGTK/webkit-2.24/Source/WebCore

Merge r242864 - [CoordinatedGraphics] Null dereference in CoordinatedGraphicsLayer::setCoordinatorIncludingSubLayersIfNeeded
​https://bugs.webkit.org/show_bug.cgi?id=195615

Reviewed by Carlos Garcia Campos.

Exit early if we don't receive a valid coordinator.

• platform/graphics/texmap/coordinated/CoordinatedGraphicsLayer.cpp:

(WebCore::CoordinatedGraphicsLayer::setCoordinatorIncludingSubLayersIfNeeded):

2:25 AM Changeset in webkit [242872] by Carlos Garcia Campos

• 3 edits
  2 adds in releases/WebKitGTK/webkit-2.24

Merge r242826 - [WebGL] WebGLBuffer can be too large
​https://bugs.webkit.org/show_bug.cgi?id=195068
<rdar://problem/48414289>

Reviewed by Antoine Quint.

Source/WebCore:

When creating an element array buffer, make sure to
test against the maximum size of an ArrayBuffer, rather
than just assume it can be created.

Test: fast/canvas/webgl/largeBuffer.html

• html/canvas/WebGLBuffer.cpp:

(WebCore::WebGLBuffer::associateBufferDataImpl):

LayoutTests:

• fast/canvas/webgl/largeBuffer-expected.txt: Added.
• fast/canvas/webgl/largeBuffer.html: Added.

2:25 AM Changeset in webkit [242871] by Carlos Garcia Campos

• 3 edits
  1 add in releases/WebKitGTK/webkit-2.24

Merge r242810 - The HasIndexedProperty node does GC.
​https://bugs.webkit.org/show_bug.cgi?id=195559
<rdar://problem/48767923>

Reviewed by Yusuke Suzuki.

JSTests:

• stress/HasIndexedProperty-does-gc.js: Added.

Source/JavaScriptCore:

HasIndexedProperty can call the slow path operationHasIndexedPropertyByInt(),
which can eventually call JSString::getIndex(), which can resolve a rope.

• dfg/DFGDoesGC.cpp:

(JSC::DFG::doesGC):

2:25 AM Changeset in webkit [242870] by Carlos Garcia Campos

• 3 edits
  2 adds in releases/WebKitGTK/webkit-2.24

Merge r242515 - SVGPathSegList.insertItemBefore() should fail if the newItem belongs to an animating animPathSegList
​https://bugs.webkit.org/show_bug.cgi?id=195333
<rdar://problem/48475802>

Reviewed by Simon Fraser.

Source/WebCore:

Because the SVG1.1 specs states that the newItem should be removed from
its original list before adding it to another list,
SVGPathSegList.insertItemBefore() should fail if the new item belongs to
an animating animPathSegList since it is read-only.

Test: svg/dom/SVGPathSegList-insert-from-animating-animPathSegList.svg

• svg/SVGPathSegList.cpp:

(WebCore::SVGPathSegList::processIncomingListItemValue):

LayoutTests:

• svg/dom/SVGPathSegList-insert-from-animating-animPathSegList-expected.txt: Added.
• svg/dom/SVGPathSegList-insert-from-animating-animPathSegList.svg: Added.

2:25 AM Changeset in webkit [242869] by Carlos Garcia Campos

• 2 edits in releases/WebKitGTK/webkit-2.24/Source/WebCore

Merge r242639 - Use a thread safe refcounter for FilterOperation.
​https://bugs.webkit.org/show_bug.cgi?id=194149

Reviewed by Carlos Garcia Campos.

Use a thread safe refcounter for FilterOperation.

• platform/graphics/filters/FilterOperation.h:

2:25 AM Changeset in webkit [242868] by Carlos Garcia Campos

• 3 edits
  1 add in releases/WebKitGTK/webkit-2.24

Merge r242667 - Stack overflow crash in JSC::JSObject::hasInstance.
​https://bugs.webkit.org/show_bug.cgi?id=195458
<rdar://problem/48710195>

Reviewed by Yusuke Suzuki.

JSTests:

• stress/stack-overflow-in-custom-hasInstance.js: Added.

Source/JavaScriptCore:

• runtime/JSObject.cpp:

(JSC::JSObject::hasInstance):

2:25 AM WebKitGTK/2.24.x edited by Carlos Garcia Campos

(diff)

2:25 AM Changeset in webkit [242867] by Carlos Garcia Campos

• 6 edits
  2 adds in releases/WebKitGTK/webkit-2.24

Merge r242587 - Crash when attempting to change input type while dismissing datalist suggestions
​https://bugs.webkit.org/show_bug.cgi?id=195384
<rdar://problem/48563718>

Reviewed by Brent Fulgham.

Source/WebCore:

When closing a datalist suggestion menu, WebPageProxy sends a message to WebPage instructing it to tell its
active datalist suggestions picker to close. However, for a myriad of reasons, the suggestions picker (kept
alive by its text input type) may have already gone away by this point. To mitigate this, make WebPage weakly
reference its active datalist suggestions picker.

Test: fast/forms/datalist/change-input-type-after-closing-datalist-suggestions.html

• platform/DataListSuggestionPicker.h:

Make DataListSuggestionPicker capable of being weakly referenced. Additionally, fix some minor preexisting
issues in this header (#imports instead of #includes, as well as an unnecessary include of IntRect.h).

Source/WebKit:

See WebCore ChangeLog for more details.

• WebProcess/WebPage/WebPage.cpp:

(WebKit::WebPage::setActiveDataListSuggestionPicker):
(WebKit::WebPage::didSelectDataListOption):
(WebKit::WebPage::didCloseSuggestions):

• WebProcess/WebPage/WebPage.h:

Turn m_activeDataListSuggestionPicker from a raw pointer into a WeakPtr.

LayoutTests:

Add a new layout test to exercise this scenario.

• fast/forms/datalist/change-input-type-after-closing-datalist-suggestions-expected.txt: Added.
• fast/forms/datalist/change-input-type-after-closing-datalist-suggestions.html: Added.

2:24 AM Changeset in webkit [242866] by Carlos Garcia Campos

• 6 edits
  1 add in releases/WebKitGTK/webkit-2.24

Merge r242569 - Air::reportUsedRegisters must padInterference
​https://bugs.webkit.org/show_bug.cgi?id=195303
<rdar://problem/48270343>

Reviewed by Keith Miller.

JSTests:

• stress/optional-def-arg-width-should-be-both-early-and-late-use.js: Added.

Source/JavaScriptCore:

reportUsedRegisters uses reg liveness to eliminate loads/moves into dead
registers. However, liveness can report incorrect results in certain
scenarios when considering liveness at instruction boundaries. For example,
it can go wrong when an Inst has a LateUse of a register and the following
Inst has an EarlyDef of that same register. Such a scenario could lead us
to incorrectly say the register is not live-in to the first Inst. Pad
interference inserts Nops between such instruction boundaries that cause
this issue.

The test with this patch fixes the issue in reportUsedRegisters. This patch
also conservatively makes it so that lowerAfterRegAlloc calls padInterference
since it also reasons about liveness.

• b3/air/AirLowerAfterRegAlloc.cpp:

(JSC::B3::Air::lowerAfterRegAlloc):

• b3/air/AirPadInterference.h:
• b3/air/AirReportUsedRegisters.cpp:

(JSC::B3::Air::reportUsedRegisters):

• b3/testb3.cpp:

(JSC::B3::testReportUsedRegistersLateUseNotDead):
(JSC::B3::run):

2:24 AM Changeset in webkit [242865] by Carlos Garcia Campos

• 3 edits
  1 add in releases/WebKitGTK/webkit-2.24

Merge r242276 - DFG: Loop-invariant code motion (LICM) should not hoist dead code
​https://bugs.webkit.org/show_bug.cgi?id=194945
<rdar://problem/48311657>

Reviewed by Saam Barati.

• dfg/DFGLICMPhase.cpp:

(JSC::DFG::LICMPhase::run):

1:55 AM Changeset in webkit [242864] by magomez@igalia.com

• 2 edits in trunk/Source/WebCore

[CoordinatedGraphics] Null dereference in CoordinatedGraphicsLayer::setCoordinatorIncludingSubLayersIfNeeded
​https://bugs.webkit.org/show_bug.cgi?id=195615

Reviewed by Carlos Garcia Campos.

Exit early if we don't receive a valid coordinator.

• platform/graphics/texmap/coordinated/CoordinatedGraphicsLayer.cpp:

(WebCore::CoordinatedGraphicsLayer::setCoordinatorIncludingSubLayersIfNeeded):

1:39 AM Changeset in webkit [242863] by Carlos Garcia Campos

• 2 edits in trunk/Source/WebCore/platform/gtk/po

[l10n] Updated Italian translation of WebKitGTK+
​https://bugs.webkit.org/show_bug.cgi?id=195620

Patch by Milo Casagrande <​milo@milo.name> on 2019-03-13
Rubber-stamped by Carlos Garcia Campos.

• it.po:

1:24 AM Changeset in webkit [242862] by bshafiei@apple.com

• 3 edits
  1 add in branches/safari-607-branch

Cherry-pick r242810. rdar://problem/48839349

The HasIndexedProperty node does GC.
​https://bugs.webkit.org/show_bug.cgi?id=195559
<rdar://problem/48767923>

Reviewed by Yusuke Suzuki.

JSTests:

• stress/HasIndexedProperty-does-gc.js: Added.

Source/JavaScriptCore:

HasIndexedProperty can call the slow path operationHasIndexedPropertyByInt(),
which can eventually call JSString::getIndex(), which can resolve a rope.

• dfg/DFGDoesGC.cpp: (JSC::DFG::doesGC):

git-svn-id: ​https://svn.webkit.org/repository/webkit/trunk@242810 268f45cc-cd09-0410-ab3c-d52691b4dbfc

1:24 AM Changeset in webkit [242861] by bshafiei@apple.com

• 4 edits
  2 adds in branches/safari-607-branch

Cherry-pick r242749. rdar://problem/48839358

[macOS] Dispatching reentrant "contextmenu" events may cause crashes
​https://bugs.webkit.org/show_bug.cgi?id=195571
<rdar://problem/48086046>

Reviewed by Andy Estes.

Source/WebCore:

Make ContextMenuController::handleContextMenuEvent robust against reentrancy by guarding it with a boolean flag.
As demonstrated in the test case, it is currently possible to force WebKit into a bad state by dispatching a
synthetic "contextmenu" event from within the scope of one of the "before(copy|cut|paste)" events triggered as
a result of handling a context menu event.

Test: fast/events/contextmenu-reentrancy-crash.html

• page/ContextMenuController.cpp: (WebCore::ContextMenuController::handleContextMenuEvent):
• page/ContextMenuController.h:

LayoutTests:

Add a test to verify that triggering reentrant "contextmenu" events from script does not cause a crash.

• fast/events/contextmenu-reentrancy-crash-expected.txt: Added.
• fast/events/contextmenu-reentrancy-crash.html: Added.

git-svn-id: ​https://svn.webkit.org/repository/webkit/trunk@242749 268f45cc-cd09-0410-ab3c-d52691b4dbfc

1:24 AM Changeset in webkit [242860] by bshafiei@apple.com

• 23 edits in branches/safari-607-branch

Cherry-pick r242735. rdar://problem/48780112

Add a WKContentRuleList variant that uses copied memory instead of mmap'd shared memory for class A containerized apps
​https://bugs.webkit.org/show_bug.cgi?id=195511
<rdar://problem/44873269>

Patch by Alex Christensen <​achristensen@webkit.org> on 2019-03-11
Reviewed by Darin Adler.

Source/WebKit:

• NetworkProcess/NetworkContentRuleListManager.cpp: (WebKit::NetworkContentRuleListManager::addContentRuleLists):
• NetworkProcess/NetworkContentRuleListManager.h:
• NetworkProcess/cache/NetworkCacheFileSystem.cpp: (WebKit::NetworkCache::pathRegisteredAsUnsafeToMemoryMapForTesting): (WebKit::NetworkCache::registerPathAsUnsafeToMemoryMapForTesting): (WebKit::NetworkCache::isSafeToUseMemoryMapForPath):
• NetworkProcess/cache/NetworkCacheFileSystem.h:
• Shared/WebCompiledContentRuleList.cpp: (WebKit::WebCompiledContentRuleList::usesCopiedMemory const): (WebKit::WebCompiledContentRuleList::conditionsApplyOnlyToDomain const): (WebKit::WebCompiledContentRuleList::filtersWithoutConditionsBytecode const): (WebKit::WebCompiledContentRuleList::filtersWithConditionsBytecode const): (WebKit::WebCompiledContentRuleList::topURLFiltersBytecode const): (WebKit::WebCompiledContentRuleList::actions const):
• Shared/WebCompiledContentRuleList.h:
• Shared/WebCompiledContentRuleListData.cpp: (WebKit::WebCompiledContentRuleListData::size const): (WebKit::WebCompiledContentRuleListData::dataPointer const): (WebKit::WebCompiledContentRuleListData::encode const): (WebKit::WebCompiledContentRuleListData::decode):
• Shared/WebCompiledContentRuleListData.h: (WebKit::WebCompiledContentRuleListData::WebCompiledContentRuleListData):
• UIProcess/API/APIContentRuleList.cpp: (API::ContentRuleList::usesCopiedMemory const):
• UIProcess/API/APIContentRuleList.h:
• UIProcess/API/APIContentRuleListStore.cpp: (API::getData): (API::decodeContentRuleListMetaData): (API::ContentRuleListStore::readContentsOfFile): (API::MappedOrCopiedData::dataPointer const): (API::openAndMapOrCopyContentRuleList): (API::compiledToFile): (API::createExtension): (API::ContentRuleListStore::lookupContentRuleList): (API::ContentRuleListStore::compileContentRuleList): (API::ContentRuleListStore::getContentRuleListSource): (API::openAndMapContentRuleList): Deleted.
• UIProcess/API/APIContentRuleListStore.h:
• UIProcess/API/Cocoa/APIContentRuleListStoreCocoa.mm: (API::ContentRuleListStore::readContentsOfFile):
• UIProcess/API/Cocoa/WKContentRuleListStore.mm: (+[WKContentRuleListStore _registerPathAsUnsafeToMemoryMapForTesting:]):
• UIProcess/API/Cocoa/WKContentRuleListStorePrivate.h:
• UIProcess/API/Cocoa/_WKUserContentFilter.mm: (-[_WKUserContentFilter usesCopiedMemory]):
• UIProcess/API/Cocoa/_WKUserContentFilterPrivate.h:
• WebProcess/UserContent/WebUserContentController.cpp: (WebKit::WebUserContentController::addContentRuleLists):
• WebProcess/UserContent/WebUserContentController.h:
• WebProcess/WebPage/WebPage.cpp: (WebKit::m_hostFileDescriptor):

Tools:

• TestWebKitAPI/Tests/WebKitCocoa/WKContentExtensionStore.mm: (-[TestSchemeHandlerSubresourceShouldBeBlocked webView:startURLSchemeTask:]): (-[TestSchemeHandlerSubresourceShouldBeBlocked webView:stopURLSchemeTask:]): (TEST_F):

git-svn-id: ​https://svn.webkit.org/repository/webkit/trunk@242735 268f45cc-cd09-0410-ab3c-d52691b4dbfc

1:24 AM Changeset in webkit [242859] by bshafiei@apple.com

• 3 edits
  1 add in branches/safari-607-branch

Cherry-pick r242667. rdar://problem/48839311

Stack overflow crash in JSC::JSObject::hasInstance.
​https://bugs.webkit.org/show_bug.cgi?id=195458
<rdar://problem/48710195>

Reviewed by Yusuke Suzuki.

JSTests:

• stress/stack-overflow-in-custom-hasInstance.js: Added.

Source/JavaScriptCore:

• runtime/JSObject.cpp: (JSC::JSObject::hasInstance):

git-svn-id: ​https://svn.webkit.org/repository/webkit/trunk@242667 268f45cc-cd09-0410-ab3c-d52691b4dbfc

1:24 AM Changeset in webkit [242858] by bshafiei@apple.com

• 6 edits
  2 adds in branches/safari-607-branch

Cherry-pick r242587. rdar://problem/48839354

Crash when attempting to change input type while dismissing datalist suggestions
​https://bugs.webkit.org/show_bug.cgi?id=195384
<rdar://problem/48563718>

Reviewed by Brent Fulgham.

Source/WebCore:

When closing a datalist suggestion menu, WebPageProxy sends a message to WebPage instructing it to tell its
active datalist suggestions picker to close. However, for a myriad of reasons, the suggestions picker (kept
alive by its text input type) may have already gone away by this point. To mitigate this, make WebPage weakly
reference its active datalist suggestions picker.

Test: fast/forms/datalist/change-input-type-after-closing-datalist-suggestions.html

• platform/DataListSuggestionPicker.h:

Make DataListSuggestionPicker capable of being weakly referenced. Additionally, fix some minor preexisting
issues in this header (#imports instead of #includes, as well as an unnecessary include of IntRect.h).

Source/WebKit:

See WebCore ChangeLog for more details.

• WebProcess/WebPage/WebPage.cpp: (WebKit::WebPage::setActiveDataListSuggestionPicker): (WebKit::WebPage::didSelectDataListOption): (WebKit::WebPage::didCloseSuggestions):
• WebProcess/WebPage/WebPage.h:

Turn m_activeDataListSuggestionPicker from a raw pointer into a WeakPtr.

LayoutTests:

Add a new layout test to exercise this scenario.

• fast/forms/datalist/change-input-type-after-closing-datalist-suggestions-expected.txt: Added.
• fast/forms/datalist/change-input-type-after-closing-datalist-suggestions.html: Added.

git-svn-id: ​https://svn.webkit.org/repository/webkit/trunk@242587 268f45cc-cd09-0410-ab3c-d52691b4dbfc

1:24 AM Changeset in webkit [242857] by bshafiei@apple.com

• 6 edits
  1 add in branches/safari-607-branch

Cherry-pick r242569. rdar://problem/48839273

Air::reportUsedRegisters must padInterference
​https://bugs.webkit.org/show_bug.cgi?id=195303
<rdar://problem/48270343>

Reviewed by Keith Miller.

JSTests:

• stress/optional-def-arg-width-should-be-both-early-and-late-use.js: Added.

Source/JavaScriptCore:

reportUsedRegisters uses reg liveness to eliminate loads/moves into dead
registers. However, liveness can report incorrect results in certain
scenarios when considering liveness at instruction boundaries. For example,
it can go wrong when an Inst has a LateUse of a register and the following
Inst has an EarlyDef of that same register. Such a scenario could lead us
to incorrectly say the register is not live-in to the first Inst. Pad
interference inserts Nops between such instruction boundaries that cause
this issue.

The test with this patch fixes the issue in reportUsedRegisters. This patch
also conservatively makes it so that lowerAfterRegAlloc calls padInterference
since it also reasons about liveness.

• b3/air/AirLowerAfterRegAlloc.cpp: (JSC::B3::Air::lowerAfterRegAlloc):
• b3/air/AirPadInterference.h:
• b3/air/AirReportUsedRegisters.cpp: (JSC::B3::Air::reportUsedRegisters):
• b3/testb3.cpp: (JSC::B3::testReportUsedRegistersLateUseNotDead): (JSC::B3::run):

git-svn-id: ​https://svn.webkit.org/repository/webkit/trunk@242569 268f45cc-cd09-0410-ab3c-d52691b4dbfc

1:24 AM Changeset in webkit [242856] by bshafiei@apple.com

• 3 edits
  2 adds in branches/safari-607-branch

Cherry-pick r242515. rdar://problem/48839275

SVGPathSegList.insertItemBefore() should fail if the newItem belongs to an animating animPathSegList
​https://bugs.webkit.org/show_bug.cgi?id=195333
<rdar://problem/48475802>

Reviewed by Simon Fraser.

Source/WebCore:

Because the SVG1.1 specs states that the newItem should be removed from
its original list before adding it to another list,
SVGPathSegList.insertItemBefore() should fail if the new item belongs to
an animating animPathSegList since it is read-only.

Test: svg/dom/SVGPathSegList-insert-from-animating-animPathSegList.svg

• svg/SVGPathSegList.cpp: (WebCore::SVGPathSegList::processIncomingListItemValue):

LayoutTests:

• svg/dom/SVGPathSegList-insert-from-animating-animPathSegList-expected.txt: Added.
• svg/dom/SVGPathSegList-insert-from-animating-animPathSegList.svg: Added.

git-svn-id: ​https://svn.webkit.org/repository/webkit/trunk@242515 268f45cc-cd09-0410-ab3c-d52691b4dbfc

1:24 AM Changeset in webkit [242855] by bshafiei@apple.com

• 3 edits
  1 add in branches/safari-607-branch

Cherry-pick r242276. rdar://problem/48839389

DFG: Loop-invariant code motion (LICM) should not hoist dead code
​https://bugs.webkit.org/show_bug.cgi?id=194945
<rdar://problem/48311657>

Reviewed by Saam Barati.

• dfg/DFGLICMPhase.cpp: (JSC::DFG::LICMPhase::run):

git-svn-id: ​https://svn.webkit.org/repository/webkit/trunk@242276 268f45cc-cd09-0410-ab3c-d52691b4dbfc

1:24 AM Changeset in webkit [242854] by bshafiei@apple.com

• 5 edits
  1 add in branches/safari-607-branch

Cherry-pick r242114. rdar://problem/48839372

wasmToJS() should purify incoming NaNs.
​https://bugs.webkit.org/show_bug.cgi?id=194807
<rdar://problem/48189132>

Reviewed by Saam Barati.

JSTests:

• wasm/regress/wasmToJS-should-purify-NaNs.js: Added.

Source/JavaScriptCore:

• runtime/JSCJSValue.h: (JSC::jsNumber):
• runtime/TypedArrayAdaptors.h: (JSC::IntegralTypedArrayAdaptor::toJSValue):
• wasm/js/WasmToJS.cpp: (JSC::Wasm::wasmToJS):

git-svn-id: ​https://svn.webkit.org/repository/webkit/trunk@242114 268f45cc-cd09-0410-ab3c-d52691b4dbfc

1:24 AM Changeset in webkit [242853] by bshafiei@apple.com

• 2 edits in branches/safari-607-branch/Source/WebCore

Cherry-pick r241632. rdar://problem/48839379

Sample domainsVisited diagnostic logging
​https://bugs.webkit.org/show_bug.cgi?id=194657

Reviewed by Ryosuke Niwa.

Sample domainsVisited diagnostic logging, we are getting a lot of data from
this key and this is hurting our other keys.

• page/Page.cpp: (WebCore::Page::logNavigation):

git-svn-id: ​https://svn.webkit.org/repository/webkit/trunk@241632 268f45cc-cd09-0410-ab3c-d52691b4dbfc

1:24 AM Changeset in webkit [242852] by bshafiei@apple.com

• 4 edits
  4 adds in branches/safari-607-branch

Cherry-pick r241615. rdar://problem/48839366

SamplingProfiler::stackTracesAsJSON() should escape strings.
​https://bugs.webkit.org/show_bug.cgi?id=194649
<rdar://problem/48072386>

Reviewed by Saam Barati.

JSTests:

• stress/sampling-profiler-stack-trace-with-double-quote-in-function-name.js: Added.
• stress/type-profiler-with-double-quote-in-constructor-name.js: Added.
• stress/type-profiler-with-double-quote-in-field-name.js: Added.
• stress/type-profiler-with-double-quote-in-optional-field-name.js: Added.

Source/JavaScriptCore:

Ditto for TypeSet::toJSONString() and TypeSet::toJSONString().

• runtime/SamplingProfiler.cpp: (JSC::SamplingProfiler::stackTracesAsJSON):
• runtime/TypeSet.cpp: (JSC::TypeSet::toJSONString const): (JSC::StructureShape::toJSONString const):

git-svn-id: ​https://svn.webkit.org/repository/webkit/trunk@241615 268f45cc-cd09-0410-ab3c-d52691b4dbfc

1:24 AM Changeset in webkit [242851] by bshafiei@apple.com

• 4 edits in branches/safari-607-branch

Cherry-pick r241608. rdar://problem/48839277

[WebVTT] Inline WebVTT styles should start with '::cue'
​https://bugs.webkit.org/show_bug.cgi?id=194227

Reviewed by Eric Carlson.

Source/WebCore:

The original fix in r241203 is not sufficient, since it only checks if the CSS string starts
with '::cue'. Before accepting a CSS string from a WebVTT file, it should be checked that
all selectors starts with '::cue'.

Test: media/track/track-cue-css.html

• html/track/WebVTTParser.cpp: (WebCore::WebVTTParser::checkAndStoreStyleSheet):

LayoutTests:

Add invalid 'STYLE' blocks which the WebVTT parser should reject.

• media/track/captions-webvtt/css-styling.vtt:

git-svn-id: ​https://svn.webkit.org/repository/webkit/trunk@241608 268f45cc-cd09-0410-ab3c-d52691b4dbfc

1:24 AM Changeset in webkit [242850] by bshafiei@apple.com

• 3 edits in branches/safari-607-branch/Source/WebCore

Cherry-pick r241468. rdar://problem/48839280

REGRESSION: [ Mac Debug WK2 ] Layout Test storage/indexeddb/key-type-infinity-private.html is a flaky crash
​https://bugs.webkit.org/show_bug.cgi?id=194413
<rdar://problem/47897254>

Reviewed by Brady Eidson.

IDB clients expected transaction operations to be executed in order, but in
UniqueIDBDatabase::immediateCloseForUserDelete, callbacks in callback map were errored out randomly.
This patch added a callback queue to UniqueIDBDatabase to make sure callbacks will be called in the same order
as IDB Server receives the request.

• Modules/indexeddb/server/UniqueIDBDatabase.cpp: (WebCore::IDBServer::UniqueIDBDatabase::storeCallbackOrFireError): (WebCore::IDBServer::UniqueIDBDatabase::immediateCloseForUserDelete): (WebCore::IDBServer::UniqueIDBDatabase::performErrorCallback): (WebCore::IDBServer::UniqueIDBDatabase::performKeyDataCallback): (WebCore::IDBServer::UniqueIDBDatabase::performGetResultCallback): (WebCore::IDBServer::UniqueIDBDatabase::performGetAllResultsCallback): (WebCore::IDBServer::UniqueIDBDatabase::performCountCallback): (WebCore::IDBServer::UniqueIDBDatabase::forgetErrorCallback):
• Modules/indexeddb/server/UniqueIDBDatabase.h:

git-svn-id: ​https://svn.webkit.org/repository/webkit/trunk@241468 268f45cc-cd09-0410-ab3c-d52691b4dbfc

1:24 AM Changeset in webkit [242849] by bshafiei@apple.com

• 2 edits in branches/safari-607-branch/Source/WebKit

Cherry-pick r241452. rdar://problem/48839390

Make WebRTCUnifiedPlanEnabled true by default
​https://bugs.webkit.org/show_bug.cgi?id=194595

Reviewed by Eric Carlson.

• Shared/WebPreferences.yaml:

git-svn-id: ​https://svn.webkit.org/repository/webkit/trunk@241452 268f45cc-cd09-0410-ab3c-d52691b4dbfc

1:24 AM Changeset in webkit [242848] by bshafiei@apple.com

• 2 edits in branches/safari-607-branch/Source/WebKit

Cherry-pick r241002. rdar://problem/48839377

UserMediaPermissionRequestManagerProxy lambdas should check for 'this' being valid
​https://bugs.webkit.org/show_bug.cgi?id=194246

Reviewed by Eric Carlson.

With PSON enabled, the manager proxy can be destroyed.
It is thus important for its callbacks to check for 'this' to be valid.

• UIProcess/UserMediaPermissionRequestManagerProxy.cpp: (WebKit::UserMediaPermissionRequestManagerProxy::captureDevicesChanged): (WebKit::UserMediaPermissionRequestManagerProxy::requestUserMediaPermissionForFrame): (WebKit::UserMediaPermissionRequestManagerProxy::enumerateMediaDevicesForFrame):

git-svn-id: ​https://svn.webkit.org/repository/webkit/trunk@241002 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12:48 AM Changeset in webkit [242847] by Fujii Hironori

• 5 edits in trunk/Source

[Win][PlayStation] Remove WebCore::standardUserAgentForURL
​https://bugs.webkit.org/show_bug.cgi?id=195662

Reviewed by Ryosuke Niwa.

Source/WebCore:

WebCore::standardUserAgentForURL is just a stub in Windows port.

No new tests because there is no behavior change.

• platform/win/UserAgentWin.cpp:

(WebCore::standardUserAgentForURL): Deleted.

• platform/playstation/UserAgentPlayStation.cpp:

(WebCore::standardUserAgentForURL): Deleted.

Source/WebKit:

• WebProcess/WebPage/win/WebPageWin.cpp:

(WebKit::WebPage::platformUserAgent const): Return an empty string as well as mac/ios ports.

12:05 AM Changeset in webkit [242846] by bshafiei@apple.com

• 2 edits in branches/safari-607-branch/Source/WebCore

Cherry-pick r242770. rdar://problem/48795265

REGRESSION(r236281): YouTube Movies fail with "video format" error
​https://bugs.webkit.org/show_bug.cgi?id=195598
<rdar://problem/48782842>

Reviewed by Jon Lee.

Partially revert r236281 for YouTube.com.

• page/Quirks.cpp: (WebCore::Quirks::hasBrokenEncryptedMediaAPISupportQuirk const):

git-svn-id: ​https://svn.webkit.org/repository/webkit/trunk@242770 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12:05 AM Changeset in webkit [242845] by bshafiei@apple.com

• 2 edits in branches/safari-607-branch/Source/WebKit

Cherry-pick r242394. rdar://problem/48591282

Cherry-pick r242394. rdar://problem/48591281

Unreviewed build fix after r242378

• UIProcess/ios/EditableImageController.mm: (WebKit::EditableImageController::associateWithAttachment):

git-svn-id: ​https://svn.webkit.org/repository/webkit/trunk@242394 268f45cc-cd09-0410-ab3c-d52691b4dbfc

git-svn-id: ​https://svn.webkit.org/repository/webkit/branches/safari-607.1.40.0-branch@242502 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12:05 AM Changeset in webkit [242844] by bshafiei@apple.com

• 5 edits in branches/safari-607-branch/Source/WebKit

Cherry-pick r242378. rdar://problem/48591282

Cherry-pick r242378. rdar://problem/48591281

Check contextIDs when handling WebContent messages
​https://bugs.webkit.org/show_bug.cgi?id=195289
<rdar://problem/48475870>

Reviewed by Alex Christensen.

The WebContent process is untrusted because it handles arbitrary markup and javascript from untrusted sources.
We should handle its messages with suspicion, and make sure the arguments are valid and usable before honoring them.

This patch hardens the message passing layer by performing MESSAGE_CHECK in places that had been overlooked.

• UIProcess/Cocoa/PlaybackSessionManagerProxy.mm: (WebKit::PlaybackSessionManagerProxy::setUpPlaybackControlsManagerWithID): (WebKit::PlaybackSessionManagerProxy::currentTimeChanged): (WebKit::PlaybackSessionManagerProxy::bufferedTimeChanged): (WebKit::PlaybackSessionManagerProxy::seekableRangesVectorChanged): (WebKit::PlaybackSessionManagerProxy::canPlayFastReverseChanged): (WebKit::PlaybackSessionManagerProxy::audioMediaSelectionOptionsChanged): (WebKit::PlaybackSessionManagerProxy::legibleMediaSelectionOptionsChanged): (WebKit::PlaybackSessionManagerProxy::audioMediaSelectionIndexChanged): (WebKit::PlaybackSessionManagerProxy::legibleMediaSelectionIndexChanged): (WebKit::PlaybackSessionManagerProxy::externalPlaybackPropertiesChanged): (WebKit::PlaybackSessionManagerProxy::wirelessVideoPlaybackDisabledChanged): (WebKit::PlaybackSessionManagerProxy::mutedChanged): (WebKit::PlaybackSessionManagerProxy::volumeChanged): (WebKit::PlaybackSessionManagerProxy::durationChanged): (WebKit::PlaybackSessionManagerProxy::playbackStartedTimeChanged): (WebKit::PlaybackSessionManagerProxy::rateChanged): (WebKit::PlaybackSessionManagerProxy::pictureInPictureSupportedChanged): (WebKit::PlaybackSessionManagerProxy::pictureInPictureActiveChanged): (WebKit::PlaybackSessionManagerProxy::handleControlledElementIDResponse const):
• UIProcess/Cocoa/UserMediaCaptureManagerProxy.cpp: (WebKit::UserMediaCaptureManagerProxy::createMediaSourceForCaptureDeviceWithConstraints): (WebKit::UserMediaCaptureManagerProxy::startProducingData): (WebKit::UserMediaCaptureManagerProxy::stopProducingData): (WebKit::UserMediaCaptureManagerProxy::end): (WebKit::UserMediaCaptureManagerProxy::capabilities): (WebKit::UserMediaCaptureManagerProxy::setMuted): (WebKit::UserMediaCaptureManagerProxy::applyConstraints):
• UIProcess/Cocoa/VideoFullscreenManagerProxy.mm: (WebKit::VideoFullscreenManagerProxy::setupFullscreenWithID): (WebKit::VideoFullscreenManagerProxy::setHasVideo): (WebKit::VideoFullscreenManagerProxy::setVideoDimensions): (WebKit::VideoFullscreenManagerProxy::enterFullscreen): (WebKit::VideoFullscreenManagerProxy::exitFullscreen): (WebKit::VideoFullscreenManagerProxy::exitFullscreenWithoutAnimationToMode): (WebKit::VideoFullscreenManagerProxy::setInlineRect): (WebKit::VideoFullscreenManagerProxy::setHasVideoContentLayer): (WebKit::VideoFullscreenManagerProxy::cleanupFullscreen): (WebKit::VideoFullscreenManagerProxy::preparedToReturnToInline): (WebKit::VideoFullscreenManagerProxy::preparedToExitFullscreen):
• UIProcess/ios/EditableImageController.mm: (WebKit::EditableImageController::didCreateEditableImage): (WebKit::EditableImageController::didDestroyEditableImage): (WebKit::EditableImageController::associateWithAttachment):

git-svn-id: ​https://svn.webkit.org/repository/webkit/trunk@242378 268f45cc-cd09-0410-ab3c-d52691b4dbfc

git-svn-id: ​https://svn.webkit.org/repository/webkit/branches/safari-607.1.40.0-branch@242501 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12:05 AM Changeset in webkit [242843] by bshafiei@apple.com

• 2 edits in branches/safari-607-branch/Source/WebKit

Cherry-pick r242629. rdar://problem/48716552

Crash under RemoteLayerTreePropertyApplier::applyProperties
​https://bugs.webkit.org/show_bug.cgi?id=195448
<rdar://problem/48588226>

Reviewed by Simon Fraser.

• UIProcess/RemoteLayerTree/RemoteLayerTreeHost.mm: (WebKit::RemoteLayerTreeHost::updateLayerTree): Under some currently-unknown circumstances, the UI process is receiving commits referring to layers that it does not know about. One understood case of this was fixed in r241899, but there seem to be cases remaining that are not understood. Also, add a release log so that we can identify any downstream effects.

git-svn-id: ​https://svn.webkit.org/repository/webkit/trunk@242629 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Mar 12, 2019:

9:24 PM Changeset in webkit [242842] by Ross Kirsling

• 31 edits in trunk

[Win] Fix a slew of simple clang-cl warnings.
​https://bugs.webkit.org/show_bug.cgi?id=195652

Reviewed by Don Olmstead.

Source/WebCore:

• page/AutoscrollController.cpp:

(WebCore::AutoscrollController::handleMouseReleaseEvent): -Wswitch

• platform/network/curl/CurlContext.cpp:

(WebCore::CurlHandle::willSetupSslCtx):
(WebCore::CurlHandle::appendRequestHeaders): -Wunused-variable

• platform/network/curl/CurlFormDataStream.cpp:

(WebCore::CurlFormDataStream::computeContentLength): -Wunused-variable

• platform/network/curl/CurlRequest.cpp:

(WebCore::CurlRequest::CurlRequest): -Wreorder
(WebCore::CurlRequest::setupTransfer): -Wunused-variable

• platform/network/curl/CurlSSLVerifier.cpp:

(WebCore::CurlSSLVerifier::CurlSSLVerifier):

• platform/network/curl/CurlSSLVerifier.h: -Wunused-private-field
• platform/win/LoggingWin.cpp:

(WebCore::logLevelString): -Wwritable-strings

• rendering/RenderThemeWin.cpp: -Wunused-const-variable (x2)

(WebCore::RenderThemeWin::getThemeData): -Wswitch

Source/WebCore/PAL:

• pal/win/LoggingWin.cpp:

(PAL::logLevelString): -Wwritable-strings

Source/WebKit:

• NetworkProcess/curl/NetworkDataTaskCurl.cpp:

(WebKit::NetworkDataTaskCurl::invokeDidReceiveResponse): -Wswitch

• Platform/IPC/win/ConnectionWin.cpp:

(IPC::Connection::readEventHandler): -Wunused-variable

• Platform/win/LoggingWin.cpp:

(WebKit::logLevelString): -Wwritable-strings

• UIProcess/Launcher/win/ProcessLauncherWin.cpp:

(WebKit::ProcessLauncher::launchProcess): -Wunused-variable

Source/WebKitLegacy/win:

• Interfaces/AccessibleComparable.idl: -Wmissing-braces
• Plugins/PluginDatabaseWin.cpp: -Wunused-function

(WebCore::addJavaPluginDirectory): Deleted.

• WebCoreSupport/AcceleratedCompositingContext.cpp:

(AcceleratedCompositingContext::AcceleratedCompositingContext): -Wreorder

• WebCoreSupport/WebEditorClient.cpp:

(WebEditorUndoCommand::WebEditorUndoCommand): -Wreorder
(undoNameForEditAction): -Wswitch

• WebCoreSupport/WebFrameLoaderClient.cpp:

(WebFrameLoaderClient::WebFrameLoaderClient): -Wswitch
(WebFrameLoaderClient::updateGlobalHistoryRedirectLinks): -Wunused-variable

• WebCoreSupport/WebInspectorClient.cpp:

(WebInspectorFrontendClient::WebInspectorFrontendClient): -Wreorder

• WebCoreSupport/WebInspectorClient.h: -Winconsistent-missing-override
• WebError.cpp:

(WebError::WebError): -Wreorder

• WebFrame.cpp:

(WebFrame::stringByEvaluatingJavaScriptInScriptWorld): -Wlogical-op-parentheses

• WebHistory.cpp: -Wunused-function (x5)

(areEqualOrClose): Deleted.
(addDayToSystemTime): Deleted.
(getDayBoundaries): Deleted.
(beginningOfDay): Deleted.
(dateKey): Deleted.

• WebNotificationCenter.cpp:

(WebNotificationCenter::removeObserver): -Wunused-variable

• WebView.cpp: -Wunused-function, -Wreorder

(WebView::addVisitedLinks): -Wunused-variable

Tools:

• WebKitTestRunner/win/PlatformWebViewWin.cpp:

(WTR::PlatformWebView::windowSnapshotImage): -Wunused-variable

7:34 PM Changeset in webkit [242841] by ysuzuki@apple.com

• 3 edits
  1 add in trunk

[JSC] OSR entry should respect abstract values in addition to flush formats
​https://bugs.webkit.org/show_bug.cgi?id=195653

Reviewed by Mark Lam.

JSTests:

• stress/osr-entry-locals-none.js: Added.

Source/JavaScriptCore:

Let's consider the following graph.

Block #0

...
27:< 2:loc13> JSConstant(JS|UseAsOther, StringIdent, Strong:String (atomic) (identifier): , StructureID: 42679, bc#10, ExitValid)
...
28:< 2:loc13> ArithPow(DoubleRep:@437<Double>, Int32:@27, Double|UseAsOther, BytecodeDouble, Exits, bc#10, ExitValid)
29:<!0:-> MovHint(DoubleRep:@28<Double>, MustGen, loc7, W:SideState, ClobbersExit, bc#10, ExitValid)
30:< 1:-> SetLocal(DoubleRep:@28<Double>, loc7(M<Double>/FlushedDouble), machine:loc6, W:Stack(-8), bc#10, exit: bc#14, ExitValid) predicting BytecodeDouble
...
73:<!0:-> Jump(MustGen, T:#1, W:SideState, bc#71, ExitValid)

Block #1 (bc#71): (OSR target) pred, #0

...

102:<!2:loc15> GetLocal(Check:Untyped:@400, Double|MustGen|PureInt, BytecodeDouble, loc7(M<Double>/FlushedDouble), machine:loc6, R:Stack(-8), bc#120, ExitValid) predicting BytecodeDouble

...

CFA at @28 says it is invalid since there are type contradiction (Int32:@27 v.s. StringIdent). So, of course, we do not propagate #0's type information to #1 since we become invalid state.
However, #1 is still reachable since it is an OSR target. Since #0 was only the predecessor of #1, loc7's type information becomes None at the head of #1.
Since loc7's AbstractValue is None, @102 GetLocal emits breakpoint. It is OK as long as OSR entry fails because AbstractValue validation requires the given value is None type.

The issue here is that we skipped AbstractValue validation when we have FlushFormat information. Since loc7 has FlushedDouble format, DFG OSR entry code does not validate it against AbstractValue,
which is None. Then, we hit the breakpoint emitted by @102.

This patch performs AbstractValue validation against values even if we have FlushFormat. We should correctly configure AbstractValue for OSR entry's locals too to avoid unnecessary OSR entry
failures in the future but anyway validating locals with AbstractValue is correct behavior here since DFGSpeculativeJIT relies on that.

• dfg/DFGOSREntry.cpp:

(JSC::DFG::prepareOSREntry):

7:31 PM Changeset in webkit [242840] by rniwa@webkit.org

• 12 edits in trunk/Source

Move the code for determining the need for touch bar quirks to Quirks class
​https://bugs.webkit.org/show_bug.cgi?id=195654

Reviewed by Brent Fulgham.

Source/WebCore:

Moved the code to determine whether the touch bar quirks are needed or not from WebKit2.

• WebCore.xcodeproj/project.pbxproj:
• page/Quirks.cpp:

(WebCore::Quirks::isTouchBarUpdateSupressedForHiddenContentEditable const):
(WebCore::Quirks::isNeverRichlyEditableForTouchBar const):

• page/Quirks.h:

Source/WebKit:

Moved the code to determine whether touch bar quirks are needed or not to WebCore.

Also renamed HiddenContentEditableQuirk to IsTouchBarUpdateSupressedForHiddenContentEditable
and PlainTextQuirk to NeverRichlyEditableForTouchBar.

• UIProcess/Cocoa/WebViewImpl.h:

(WebKit::WebViewImpl::isRichlyEditableForTouchBar): Renamed.

• UIProcess/Cocoa/WebViewImpl.mm:

(WebKit::WebViewImpl::updateTouchBar):
(WebKit::WebViewImpl::candidateListTouchBarItem const):
(WebKit::WebViewImpl::isRichlyEditableForTouchBar const): Renamed from isRichlyEditable.
(WebKit::WebViewImpl::textTouchBar const):
(WebKit::WebViewImpl::updateTextTouchBar):

• UIProcess/WebPageProxy.cpp:

(WebKit::WebPageProxy::setIsTouchBarUpdateSupressedForHiddenContentEditable): Renamed from
setNeedsHiddenContentEditableQuirk.
(WebKit::WebPageProxy::setIsNeverRichlyEditableForTouchBar): Renamed from setNeedsPlainTextQuirk.

• UIProcess/WebPageProxy.h:

(WebKit::WebPageProxy::isTouchBarUpdateSupressedForHiddenContentEditable const): Renamed from
needsHiddenContentEditableQuirk.
(WebKit::WebPageProxy::isNeverRichlyEditableForTouchBar const): Renamed from needsPlainTextQuirk.

• UIProcess/WebPageProxy.messages.in: Renamed the IPC messages.
• WebProcess/WebPage/WebPage.cpp:

(WebKit::needsHiddenContentEditableQuirk): Deleted. Moved to WebCore.
(WebKit::needsPlainTextQuirk): Deleted. Moved to WebCore.
(WebKit::WebPage::didStartPageTransition):
(WebKit::WebPage::didChangeSelection):

• WebProcess/WebPage/WebPage.h:

(WebKit::WebPage):

6:33 PM Changeset in webkit [242839] by Justin Fan

• 53 edits
  1 copy
  5 moves
  6 deletes in trunk

[Web GPU] Update GPURenderPipelineDescriptor and add GPUColorStateDescriptor.format
​https://bugs.webkit.org/show_bug.cgi?id=195518
<rdar://problem/46322356>

Reviewed by Myles C. Maxfield.

Source/WebCore:

Upgrade the implementation of GPURenderPipelineDescriptor and GPURenderPipeline and match the updated Web GPU API.
Add stubs for GPUColorStateDescriptor so attachment format can be provided by GPURenderPipelineDescriptor.

All affected Web GPU tests updated to cover existing behavior.

Update file names and symbols:

• CMakeLists.txt:
• DerivedSources-input.xcfilelist:
• DerivedSources-output.xcfilelist:
• DerivedSources.make:
• Sources.txt:
• WebCore.xcodeproj/project.pbxproj:
• bindings/js/WebCoreBuiltinNames.h:

Updates to GPURenderPipeline, GPURenderPipelineDescriptor, and its components:

• Modules/webgpu/GPUColorStateDescriptor.idl: Added. Provide the expected texture format of the render pipeline's color attachments.
• Modules/webgpu/GPUInputStateDescriptor.idl: Renamed from Source/WebCore/Modules/webgpu/WebGPUInputStateDescriptor.idl.
• Modules/webgpu/GPUTextureFormat.idl: Update the existing values to match the new style.
• Modules/webgpu/GPUVertexAttributeDescriptor.idl: Renamed from Source/WebCore/Modules/webgpu/WebGPUVertexAttributeDescriptor.idl.
• Modules/webgpu/GPUVertexInputDescriptor.idl: Renamed from Source/WebCore/Modules/webgpu/WebGPUVertexInputDescriptor.idl.
• Modules/webgpu/WebGPUDevice.cpp:

(WebCore::WebGPUDevice::createPipelineLayout const): Remove unnecessary comment.
(WebCore::WebGPUDevice::createRenderPipeline const): Descriptor validation and conversion moved into WebGPURenderPipelineDescriptor.
(WebCore::validateAndConvertPipelineStage): Deleted.

• Modules/webgpu/WebGPUDevice.h: create* functions should not return nullable.
• Modules/webgpu/WebGPUIndexFormat.h: Removed. Moved into GPUInputStateDescriptor.idl.
• Modules/webgpu/WebGPUInputStateDescriptor.h: Removed.
• Modules/webgpu/WebGPUInputStepMode.idl: Removed. Moved into GPUVertexInputDescriptor.idl.
• Modules/webgpu/WebGPUPipelineLayout.h:

(WebCore::WebGPUPipelineLayout::pipelineLayout): Getters should return raw references.

• Modules/webgpu/WebGPUPipelineStageDescriptor.h: Now shares a common base with GPUPipelineStageDescriptor.
• Modules/webgpu/WebGPURenderPipeline.cpp:

(WebCore::WebGPURenderPipeline::create):
(WebCore::WebGPURenderPipeline::WebGPURenderPipeline):

• Modules/webgpu/WebGPURenderPipeline.h: Now internally nullable.

(WebCore::WebGPURenderPipeline::renderPipeline const):
(WebCore::WebGPURenderPipeline::renderPipeline): Deleted.

• Modules/webgpu/WebGPURenderPipelineDescriptor.cpp:

(WebCore::WebGPUPipelineStageDescriptor::asGPUPipelineStageDescriptor const): Validate and convert a WebGPUPipelineStageDescriptor to GPU version.
(WebCore::WebGPURenderPipelineDescriptor::asGPURenderPipelineDescriptor const): Ditto for WebGPURenderPipelineDescriptor.

• Modules/webgpu/WebGPURenderPipelineDescriptor.h: Now shares a base class and some instance variables with GPURenderPipelineDescriptor.
• Modules/webgpu/WebGPURenderPipelineDescriptor.idl: Update GPUPrimitiveTopology for new style and add colorStates.
• Modules/webgpu/WebGPUShaderModule.idl: Small pilot to test using InterfaceName to easily rename DOM-facing interfaces.
• Modules/webgpu/WebGPUVertexAttributeDescriptor.h: Removed.
• Modules/webgpu/WebGPUVertexFormat.idl: Removed. Moved and updated in GPUVertexAttributeDescriptor.idl.
• Modules/webgpu/WebGPUVertexInputDescriptor.h: Removed.
• platform/graphics/gpu/GPUInputStateDescriptor.h:
• platform/graphics/gpu/GPUPipelineStageDescriptor.h:

(WebCore::GPUPipelineStageDescriptor::GPUPipelineStageDescriptor):

• platform/graphics/gpu/GPURenderPipelineDescriptor.h: Add shared base class for Web/GPURenderPipelineDescriptor.

(WebCore::GPURenderPipelineDescriptor::GPURenderPipelineDescriptor):

• platform/graphics/gpu/GPUTextureFormat.h:
• platform/graphics/gpu/GPUVertexAttributeDescriptor.h:
• platform/graphics/gpu/GPUVertexInputDescriptor.h:
• platform/graphics/gpu/cocoa/GPURenderPipelineMetal.mm:

(WebCore::setFunctionsForPipelineDescriptor): Make fragment required since descriptor validation fails if fragment function is not found right now.
(WebCore::mtlVertexFormatForGPUVertexFormat): Renamed from validateAndConvertVertexFormatToMTLVertexFormat.
(WebCore::mtlStepFunctionForGPUInputStepMode): Renamed from validateAndConvertStepModeToMTLStepFunction
(WebCore::trySetInputStateForPipelineDescriptor):
(WebCore::trySetColorStatesForColorAttachmentArray):
(WebCore::tryCreateMtlRenderPipelineState):
(WebCore::GPURenderPipeline::create):
(WebCore::validateAndConvertVertexFormatToMTLVertexFormat): Deleted.
(WebCore::validateAndConvertStepModeToMTLStepFunction): Deleted.

• platform/graphics/gpu/cocoa/GPUUtilsMetal.mm:

(WebCore::platformTextureFormatForGPUTextureFormat):

• platform/graphics/gpu/GPUColorStateDescriptor.h: Added.

Misc:

• Modules/webgpu/WebGPUProgrammablePassEncoder.cpp:

(WebCore::WebGPUProgrammablePassEncoder::setPipeline):

• Modules/webgpu/WebGPUProgrammablePassEncoder.h:
• platform/graphics/gpu/GPUProgrammablePassEncoder.h:
• platform/graphics/gpu/GPURenderPassEncoder.h:
• platform/graphics/gpu/GPURenderPipeline.h:

(WebCore::GPURenderPipeline::primitiveTopology const):

• platform/graphics/gpu/cocoa/GPURenderPassEncoderMetal.mm:

(WebCore::GPURenderPassEncoder::setPipeline):
(WebCore::primitiveTypeForGPUPrimitiveTopology):

Tools:

• DumpRenderTree/DerivedSources-input.xcfilelist:
• DumpRenderTree/DerivedSources-output.xcfilelist:

LayoutTests:

Update all tests with new enum styles and re-namings, and specify the format of the color attachment.

• webgpu/blit-commands.html:
• webgpu/buffer-command-buffer-races.html:
• webgpu/buffer-resource-triangles.html:
• webgpu/depth-enabled-triangle-strip.html:
• webgpu/js/webgpu-functions.js:

(createBasicSwapChain):
(createBasicDepthTexture):

• webgpu/render-pipelines-expected.txt:
• webgpu/render-pipelines.html: Remove error cases, as createRenderPipeline no longer returns a null value on failure.
• webgpu/shader-modules.html:
• webgpu/texture-triangle-strip.html:
• webgpu/textures-textureviews.html:
• webgpu/vertex-buffer-triangle-strip.html:

6:26 PM Changeset in webkit [242838] by msaboff@apple.com

• 3 edits
  1 add in trunk

REGRESSION (iOS 12.2): Webpage using CoffeeScript crashes
​https://bugs.webkit.org/show_bug.cgi?id=195613

Reviewed by Mark Lam.

JSTests:

New regression test.

• stress/regexp-backref-inbounds.js: Added.

(testRegExp):

Source/JavaScriptCore:

The bug here is in Yarr JIT backreference matching code. We are incorrectly
using a checkedOffset / inputPosition correction when checking for the available
length left in a string. It is improper to do these corrections as a backreference's
match length is based on what was matched in the referenced capture group and not
part of the checkedOffset and inputPosition computed when we compiled the RegExp.
In some cases, the resulting incorrect calculation would allow us to go past
the subject string's length. Removed these adjustments.

After writing tests for the first bug, found another bug where the non-greedy
backreference backtracking code didn't do an "are we at the end of the input?" check.
This caused an infinite loop as we'd jump from the backtracking code back to
try matching one more backreference, fail and then backtrack.

• yarr/YarrJIT.cpp:

(JSC::Yarr::YarrGenerator::generateBackReference):
(JSC::Yarr::YarrGenerator::backtrackBackReference):

6:19 PM Changeset in webkit [242837] by bshafiei@apple.com

• 1 copy in tags/Safari-607.1.40.1.4

Tag Safari-607.1.40.1.4.

5:36 PM Changeset in webkit [242836] by jiewen_tan@apple.com

• 6 edits in trunk/Source/WebKit

Add WebFrameProxy::loadData
​https://bugs.webkit.org/show_bug.cgi?id=195647
<rdar://problem/48826856>

Reviewed by Youenn Fablet.

This patch adds WebFrameProxy::loadData which is a simplified version of WebPageProxy::loadData that
loads substitute data to an iframe. This is needed by the Load Optimizer.

• UIProcess/WebFrameProxy.cpp:

(WebKit::WebFrameProxy::loadData):

• UIProcess/WebFrameProxy.h:
• WebProcess/WebPage/WebPage.cpp:

(WebKit::WebPage::loadDataInFrame):

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

5:15 PM Changeset in webkit [242835] by rniwa@webkit.org

• 2 edits in trunk/Source/WebKitLegacy/mac

Remove a site specific hack for AppleConnect plugin
​https://bugs.webkit.org/show_bug.cgi?id=195643

Reviewed by Simon Fraser.

r66437 added a workaround for AppleConnect plugin.

Remove this code since Safari doesn't even use WebKitLegacy anymore,
and other WebKit clients support AppleConnect plugins.

• WebCoreSupport/WebFrameLoaderClient.mm:

(WebFrameLoaderClient::createPlugin):

5:12 PM Changeset in webkit [242834] by Alan Bujtas

• 3 edits in trunk/Source/WebCore

[ContentChangeObserver] Reset state when touchStart does not turn into click.
​https://bugs.webkit.org/show_bug.cgi?id=195603
<rdar://problem/48796582>

Reviewed by Simon Fraser.

Add reset() function to assert and reset the current state.

• page/ios/ContentChangeObserver.cpp:

(WebCore::ContentChangeObserver::willNotProceedWithClick):
(WebCore::ContentChangeObserver::adjustObservedState):

• page/ios/ContentChangeObserver.h:

4:51 PM Changeset in webkit [242833] by Wenson Hsieh

• 9 edits
  2 adds in trunk

[iOS] Input view sometimes flickers when blurring and refocusing an element
​https://bugs.webkit.org/show_bug.cgi?id=195639
<rdar://problem/48735337>

Reviewed by Tim Horton.

Source/WebKit:

On iOS, if a focused element is blurred and immediately refocused in the scope of user interaction, we will end
up reloading interaction state (input views, autocorrection contexts, etc.) in the UI process. On certain well-
trafficked websites, this results in the input view and input accessory view flickering (or more egregiously,
scrolling to re-reveal the focused element) when changing selection.

To fix the issue, this patch refactors some focus management logic to suppress sending focused element updates
to the UI process in the case where the same element is being blurred and immediately refocused. To do this, we
track the most recently blurred element and bail when the recently blurred element is identical to the newly
focused element. See below for more detail.

Test: fast/forms/ios/keyboard-stability-when-refocusing-element.html

• UIProcess/WebPageProxy.h:
• UIProcess/ios/WKContentViewInteraction.mm:

(-[WKContentView _elementDidFocus:userIsInteracting:blurPreviousNode:changingActivityState:userObject:]):
(-[WKContentView _elementDidBlur]):

Update the web process' notion of whether an input view is showing. Importantly, this accounts for decisions
made by _WKUIDelegate. See below for more details.

(isAssistableInputType): Deleted.

Removed this helper function; this was only used in one place as a sanity check that the focused element's type
is not none, right before attempting to show an input view. Instead, we can just check the focused element's
type directly against InputType::None in the if statement of the early return.

• UIProcess/ios/WebPageProxyIOS.mm:

(WebKit::WebPageProxy::setIsShowingInputViewForFocusedElement):

Add a hook to notify the web process when an input view is showing or not (see below for more detail).

• WebProcess/WebPage/WebPage.cpp:

(WebKit::WebPage::didStartPageTransition):
(WebKit::WebPage::elementDidRefocus):
(WebKit::WebPage::shouldDispatchUpdateAfterFocusingElement const):

Add a helper to determine whether we notify the UI process about a newly focused element. On macOS, this is true
only when the new focused element is neither the currently focused element, nor the focused element that was
just blurred. On iOS, we have an additional constraint that when the input view is not showing, we still need to
notify the UI process, since the UI process might want to begin showing the keyboard for an element that has
only been programmatically focused, for which we aren't currently showing the input view.

(WebKit::WebPage::elementDidFocus):
(WebKit::WebPage::elementDidBlur):

Replace a couple of existing member variables in WebPage used for focus management:

• Replace m_hasPendingBlurNotification with m_recentlyBlurredElement, a RefPtr to the Element that is being

blurred. Behavior here is the same as before (i.e. having a pending blur notification is equivalent to
having recently blurred a focused element). However, this allows us to check newly focused elements against
the recently blurred element in WebPage::elementDidFocus().

• Replace m_isFocusingElementDueToUserInteraction with m_isShowingInputViewForFocusedElement. The flag

m_isFocusingElementDueToUserInteraction was originally added to fix <webkit.org/b/146735>, by ensuring that
we don't send redundant ElementDidFocus (formerly, StartAssistingNode) messages to the UI process even when
the keyboard is already up. In these simpler times, user interaction when focusing an element was equivalent
to showing an input view for the focused element. However, in today's world, there are a variety of reasons
why we might or might not show an input view for a given element (including, but not limited to activity
state changes and decisions made by _WKInputDelegate). As such, it doesn't make sense to continue relying on
m_isFocusingElementDueToUserInteraction in this early return. Instead, have the UI process propagate a
message back to the web process, to let it know whether there is a keyboard showing, and use this flag
instead.

• WebProcess/WebPage/WebPage.h:
• WebProcess/WebPage/WebPage.messages.in:
• WebProcess/WebPage/ios/WebPageIOS.mm:

(WebKit::WebPage::setIsShowingInputViewForFocusedElement):

LayoutTests:

Add a test to ensure that the form control interaction doesn't stop and start again when blurring and focusing
an editable element.

• fast/forms/ios/keyboard-stability-when-refocusing-element-expected.txt: Added.
• fast/forms/ios/keyboard-stability-when-refocusing-element.html: Added.

3:49 PM Changeset in webkit [242832] by commit-queue@webkit.org

• 3 edits in trunk/Tools

Unreviewed, rolling out r242825.
​https://bugs.webkit.org/show_bug.cgi?id=195648

"Broke webkitpy tests with my change to
lldb_dump_class_layout.py" (Requested by rmorisset on
#webkit).

Reverted changeset:

"Alter Tools/Scripts/dump-class-layout to be able to dump all
classes with suspicious padding"
​https://bugs.webkit.org/show_bug.cgi?id=195573
​https://trac.webkit.org/changeset/242825

3:42 PM Changeset in webkit [242831] by timothy@apple.com

• 5 edits in trunk/Source/WebCore

Expose document attributes and body background color through HTMLConverter.
​https://bugs.webkit.org/show_bug.cgi?id=195636
rdar://problem/45055697

Reviewed by Tim Horton.

Source/WebCore:

• editing/cocoa/HTMLConverter.h:
• editing/cocoa/HTMLConverter.mm:

(HTMLConverter::convert):
(WebCore::attributedStringFromRange):
(WebCore::attributedStringFromSelection):
(WebCore::attributedStringBetweenStartAndEnd):

Source/WebCore/PAL:

• pal/spi/cocoa/NSAttributedStringSPI.h:

(NSBackgroundColorDocumentAttribute): Added.

3:35 PM Changeset in webkit [242830] by Antti Koivisto

• 3 edits in trunk/Source/WebCore

Compositing layer that renders two positioned elements should not hit test
​https://bugs.webkit.org/show_bug.cgi?id=195371
<rdar://problem/48649586>

Reviewed by Simon Fraser.

Followup to fix the test case (fast/scrolling/ios/overflow-scroll-overlap-2.html)

• platform/graphics/ca/GraphicsLayerCA.cpp:

(WebCore::GraphicsLayerCA::setEventRegion):

Revert a last minute change (that was done to fix a Mac displaylist test).

• rendering/RenderLayerBacking.cpp:

(WebCore::RenderLayerBacking::paintIntoLayer):

Compute the region on iOS only for now (it is not used on other platforms).

2:47 PM Changeset in webkit [242829] by Devin Rousso

• 7 edits
  1 copy in trunk/Source/WebInspectorUI

Web Inspector: Sources: allow image collections to be filtered by type
​https://bugs.webkit.org/show_bug.cgi?id=195630

Reviewed by Matt Baker.

• UserInterface/Views/ResourceCollectionContentView.js:

(WI.ResourceCollectionContentView):
(WI.ResourceCollectionContentView.prototype.get navigationItems): Added.
(WI.ResourceCollectionContentView.prototype.contentViewAdded):
(WI.ResourceCollectionContentView.prototype.contentViewRemoved): Added.
(WI.ResourceCollectionContentView.prototype._updateImageTypeScopeBar): Added.
(WI.ResourceCollectionContentView.prototype._handleImageTypeSelectionChanged): Added.

• UserInterface/Views/ResourceCollectionContentView.css: Asdded.

(.resource-collection-image-type-scope-bar.default-item-selected):

• UserInterface/Views/CollectionContentView.css:

(.content-view.collection > .content-view[hidden]): Added.

• UserInterface/Views/ScopeBarItem.js:

(WI.ScopeBarItem.prototype.set hidden):

• UserInterface/Views/MultipleScopeBarItem.js:

(WI.MultipleScopeBarItem.prototype.set scopeBarItems):
(WI.MultipleScopeBarItem.prototype.set selectedScopeBarItem):
(WI.MultipleScopeBarItem.prototype.get _visibleScopeBarItems): Added.
(WI.MultipleScopeBarItem.prototype._selectElementSelectionChanged):
(WI.MultipleScopeBarItem.prototype._handleItemHiddenChanged): Added.
Dispatch an event when an item is hidden so that any owner WI.MultipleScopeBarItem can
rerender it's <select> without that item.

• Localizations/en.lproj/localizedStrings.js:
• UserInterface/Main.html:

2:03 PM Changeset in webkit [242828] by commit-queue@webkit.org

• 2 edits in trunk/Source/WTF

Unreviewed, rolling out r242747.
​https://bugs.webkit.org/show_bug.cgi?id=195641

Performance measurement is difficult in this period, rolling
out it and rolling in later to isolate it from the other
sensitive patches (Requested by yusukesuzuki on #webkit).

Reverted changeset:

"[JSC] Make StaticStringImpl & StaticSymbolImpl actually
static"
​https://bugs.webkit.org/show_bug.cgi?id=194212
​https://trac.webkit.org/changeset/242747

1:58 PM Changeset in webkit [242827] by aakash_jain@apple.com

• 2 edits in trunk/Tools

[ews-build] Show status bubbles while the patch is waiting in queue
​https://bugs.webkit.org/show_bug.cgi?id=195618

Reviewed by Lucas Forschler.

• BuildSlaveSupport/ews-app/ews/views/statusbubble.py:

(StatusBubble): Added ALL_QUEUES and ENABLED_QUEUES. Only certain queues are
enabled in initial deployment.
(StatusBubble._build_bubble): Display bubble even when build hasn't started.
(StatusBubble.get_latest_build_for_queue): Get latest build for a given queue.
(StatusBubble.get_builds_for_queue): Get all builds for a given queue.
(StatusBubble._should_show_bubble_for): Display bubble for only ENABLED_QUEUES for now.
(StatusBubble._build_bubbles_for_patch):

1:57 PM Changeset in webkit [242826] by dino@apple.com

• 3 edits
  2 adds in trunk

[WebGL] WebGLBuffer can be too large
​https://bugs.webkit.org/show_bug.cgi?id=195068
<rdar://problem/48414289>

Reviewed by Antoine Quint.

Source/WebCore:

When creating an element array buffer, make sure to
test against the maximum size of an ArrayBuffer, rather
than just assume it can be created.

Test: fast/canvas/webgl/largeBuffer.html

• html/canvas/WebGLBuffer.cpp:

(WebCore::WebGLBuffer::associateBufferDataImpl):

LayoutTests:

• fast/canvas/webgl/largeBuffer-expected.txt: Added.
• fast/canvas/webgl/largeBuffer.html: Added.

1:55 PM Changeset in webkit [242825] by rmorisset@apple.com

• 3 edits in trunk/Tools

Alter Tools/Scripts/dump-class-layout to be able to dump all classes with suspicious padding
​https://bugs.webkit.org/show_bug.cgi?id=195573

Reviewed by Simon Fraser.

Also modified the script so that when multiple types match a given name it shows them all and not arbitrarily pick one.

• Scripts/dump-class-layout:

(main):

• lldb/lldb_dump_class_layout.py:

(ClassLayout.init):
(ClassLayout._compute_padding_recursive):
(LLDBDebuggerInstance.dump_layout_for_classname):
(LLDBDebuggerInstance):
(LLDBDebuggerInstance.dump_all_wasteful_layouts):
(LLDBDebuggerInstance.layout_for_classname): Deleted.

1:40 PM Changeset in webkit [242824] by timothy_horton@apple.com

• 2 edits in trunk/Source/WebKit

More attempts at build fixing.

• UIProcess/ios/WKActionSheetAssistant.mm:

Yet more.

1:32 PM Changeset in webkit [242823] by Chris Dumez

• 7 edits
  1 add in trunk

Device Orientation access permission should be denied unless explicitly granted by the client
​https://bugs.webkit.org/show_bug.cgi?id=195625

Reviewed by Youenn Fablet.

Source/WebKit:

Device Orientation access permission should be denied unless explicitly granted by the client.
Previously, it was granted by default.

• UIProcess/API/APIUIClient.h:

(API::UIClient::shouldAllowDeviceOrientationAndMotionAccess):

• UIProcess/API/C/WKPage.cpp:

(WKPageSetPageUIClient):

• UIProcess/Cocoa/UIDelegate.mm:

(WebKit::UIDelegate::UIClient::shouldAllowDeviceOrientationAndMotionAccess):

Tools:

add API test coverage.

• TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj:
• TestWebKitAPI/Tests/WebKitCocoa/DeviceOrientation.mm: Added.

(-[DeviceOrientationMessageHandler userContentController:didReceiveScriptMessage:]):
(-[DeviceOrientationPermissionUIDelegate initWithHandler:]):
(-[DeviceOrientationPermissionUIDelegate _webView:shouldAllowDeviceOrientationAndMotionAccessRequestedByFrame:decisionHandler:]):
(runDeviceOrientationTest):
(TEST):

• TestWebKitAPI/Tests/WebKitCocoa/WebsitePolicies.mm:

(-[WebsitePoliciesDeviceOrientationUIDelegate _webView:shouldAllowDeviceOrientationAndMotionAccessRequestedByFrame:decisionHandler:]):

1:17 PM Changeset in webkit [242822] by Kocsen Chung

• 2 edits in branches/safari-607.1.40.0-branch/Source/WebCore

Apply patch. rdar://problem/48795264

1:10 PM Changeset in webkit [242821] by Nikita Vasilyev

• 2 edits in trunk/Source/WebInspectorUI

Web Inspector: Keyboard shortcut for settings tab too greedy on non-US keyboards
​https://bugs.webkit.org/show_bug.cgi?id=192947
<rdar://problem/46886779>

Reviewed by Devin Rousso.

• UserInterface/Base/Main.js:

(WI._showSettingsTab):

1:05 PM Changeset in webkit [242820] by Devin Rousso

• 3 edits in trunk/Source/WebInspectorUI

Web Inspector: Elements: provide node context menu items for event listeners sorted by node
​https://bugs.webkit.org/show_bug.cgi?id=195633

Reviewed by Matt Baker.

• UserInterface/Base/DOMUtilities.js:

(WI.linkifyNodeReferenceElement):
(WI.bindInteractionsForNodeToElement): Added.
Split logic for adding event listeners into a separate function so it can be used on
existing DOM without modifying it.

• UserInterface/Views/DOMNodeDetailsSidebarPanel.js:

(WI.DOMNodeDetailsSidebarPanel.prototype._refreshEventListeners.generateGroupsByNode):

12:37 PM Changeset in webkit [242819] by jer.noble@apple.com

• 2 edits in trunk/Tools

Flaky API Test TestWebKitAPI.WebKitLegacy.ScrollingDoesNotPauseMedia
​https://bugs.webkit.org/show_bug.cgi?id=195137
<rdar://problem/48810307>

Reviewed by Eric Carlson.

• TestWebKitAPI/Tests/WebKitLegacy/ios/ScrollingDoesNotPauseMedia.mm:

(TestWebKitAPI::TEST):

12:36 PM Changeset in webkit [242818] by sihui_liu@apple.com

• 7 edits in trunk

Layout Test imported/w3c/web-platform-tests/IndexedDB/fire-*-event-exception.html are failing
​https://bugs.webkit.org/show_bug.cgi?id=195581

LayoutTests/imported/w3c:

Updated test expectations to PASS.

Reviewed by Brady Eidson.

• web-platform-tests/IndexedDB/fire-error-event-exception-expected.txt:
• web-platform-tests/IndexedDB/fire-success-event-exception-expected.txt:
• web-platform-tests/IndexedDB/fire-upgradeneeded-event-exception-expected.txt:

Source/WebCore:

Reviewed by Brady Eidson.

Uncaught exceptions should be handled after IDBRequest dispatches events so that IDBTransaction would stay
active during event dispatch.

• Modules/indexeddb/IDBRequest.cpp:

(WebCore::IDBRequest::dispatchEvent):
(WebCore::IDBRequest::uncaughtExceptionInEventHandler):

• Modules/indexeddb/IDBRequest.h:

12:34 PM Changeset in webkit [242817] by aakash_jain@apple.com

• 4 edits in trunk/Tools

[ews-build] Change urls from uat to production
​https://bugs.webkit.org/show_bug.cgi?id=195566

Reviewed by Lucas Forschler.

• BuildSlaveSupport/ews-app/ews/config.py:
• BuildSlaveSupport/ews-build/events.py:
• BuildSlaveSupport/ews-build/steps.py:

12:32 PM Changeset in webkit [242816] by aakash_jain@apple.com

• 2 edits in trunk/Tools

[ews-build] change max_builds for local-worker to 1
​https://bugs.webkit.org/show_bug.cgi?id=195568

Reviewed by Lucas Forschler.

• BuildSlaveSupport/ews-build/loadConfig.py:

12:28 PM Changeset in webkit [242815] by sihui_liu@apple.com

• 4 edits in trunk

Layout Test imported/w3c/web-platform-tests/IndexedDB/transaction-abort-request-error.html is failing
​https://bugs.webkit.org/show_bug.cgi?id=195570

Reviewed by Brady Eidson.

LayoutTests/imported/w3c:

Updated test expectation to PASS.

• web-platform-tests/IndexedDB/transaction-abort-request-error-expected.txt:

Source/WebCore:

IDBRequest result should be undefined if it is never set.

• Modules/indexeddb/IDBRequest.cpp:

(WebCore::IDBRequest::IDBRequest):

12:18 PM Changeset in webkit [242814] by Antti Koivisto

• 3 edits in trunk/Source/WebKit

[iOS] Enable asynchronous frame scrolling by default
​https://bugs.webkit.org/show_bug.cgi?id=195622
<rdar://problem/48658028>

Reviewed by Simon Fraser

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

12:16 PM Changeset in webkit [242813] by Ryan Haddad

• 2 edits in trunk/LayoutTests

Unreviewed, fix a typo in TestExpecations.

• platform/mac/TestExpectations:

12:07 PM Changeset in webkit [242812] by rmorisset@apple.com

• 35 edits in trunk/Source

A lot more classes have padding that can be reduced by reordering their fields
​https://bugs.webkit.org/show_bug.cgi?id=195579

Reviewed by Mark Lam.

Source/bmalloc:

• bmalloc/Heap.h:
• bmalloc/Scavenger.h:

Source/JavaScriptCore:

• assembler/LinkBuffer.h:
• dfg/DFGArrayifySlowPathGenerator.h:

(JSC::DFG::ArrayifySlowPathGenerator::ArrayifySlowPathGenerator):

• dfg/DFGCallArrayAllocatorSlowPathGenerator.h:

(JSC::DFG::CallArrayAllocatorSlowPathGenerator::CallArrayAllocatorSlowPathGenerator):
(JSC::DFG::CallArrayAllocatorWithVariableSizeSlowPathGenerator::CallArrayAllocatorWithVariableSizeSlowPathGenerator):

• dfg/DFGGraph.h:
• dfg/DFGNode.h:

(JSC::DFG::SwitchData::SwitchData):

• dfg/DFGPlan.cpp:

(JSC::DFG::Plan::Plan):

• dfg/DFGPlan.h:
• dfg/DFGSlowPathGenerator.h:

(JSC::DFG::CallSlowPathGenerator::CallSlowPathGenerator):

• dfg/DFGSpeculativeJIT.cpp:

(JSC::DFG::SpeculativeJIT::SpeculativeJIT):

• dfg/DFGSpeculativeJIT.h:
• domjit/DOMJITSignature.h:

(JSC::DOMJIT::Signature::Signature):
(JSC::DOMJIT::Signature::effect):
(JSC::DOMJIT::Signature::argumentCount): Deleted.

• heap/MarkingConstraintSolver.h:
• heap/SlotVisitor.h:
• jit/CallFrameShuffleData.h:
• jit/JITDivGenerator.h:
• jit/SpillRegistersMode.h:
• parser/Nodes.h:
• profiler/ProfilerOSRExit.cpp:

(JSC::Profiler::OSRExit::OSRExit):

• profiler/ProfilerOSRExit.h:
• runtime/ArrayBufferView.h:
• runtime/SamplingProfiler.cpp:

(JSC::SamplingProfiler::SamplingProfiler):

• runtime/SamplingProfiler.h:
• runtime/TypeSet.cpp:

(JSC::StructureShape::StructureShape):

• runtime/TypeSet.h:
• runtime/Watchdog.h:

Source/WTF:

• wtf/CrossThreadQueue.h:
• wtf/Logger.h:
• wtf/MemoryPressureHandler.h:
• wtf/MetaAllocator.h:
• wtf/Threading.cpp:

12:03 PM Changeset in webkit [242811] by Devin Rousso

• 3 edits in trunk/Source/WebInspectorUI

Web Inspector: DOM Debugger: remove left padding when the last DOM breakpoint is removed
​https://bugs.webkit.org/show_bug.cgi?id=195522

Reviewed by Matt Baker.

• UserInterface/Views/DOMTreeContentView.js:

(WI.DOMTreeContentView.prototype._updateBreakpointStatus):

• UserInterface/Views/DOMTreeElement.js:

(WI.DOMTreeElement.prototype.get hasBreakpoint): Added.

12:02 PM Changeset in webkit [242810] by mark.lam@apple.com

• 3 edits
  1 add in trunk

The HasIndexedProperty node does GC.
​https://bugs.webkit.org/show_bug.cgi?id=195559
<rdar://problem/48767923>

Reviewed by Yusuke Suzuki.

JSTests:

• stress/HasIndexedProperty-does-gc.js: Added.

Source/JavaScriptCore:

HasIndexedProperty can call the slow path operationHasIndexedPropertyByInt(),
which can eventually call JSString::getIndex(), which can resolve a rope.

• dfg/DFGDoesGC.cpp:

(JSC::DFG::doesGC):

11:59 AM Changeset in webkit [242809] by Devin Rousso

• 12 edits
  2 adds in trunk

Web Inspector: Canvas: export recording as HTML
​https://bugs.webkit.org/show_bug.cgi?id=195311
<rdar://problem/48588673>

Reviewed by Joseph Pecoraro.

Source/WebInspectorUI:

• UserInterface/Models/Recording.js:

(WI.Recording.prototype.async swizzle):
(WI.Recording.prototype.toHTML): Added.
(WI.Recording.prototype.toHTML.escapeHTML): Added.
(WI.Recording.prototype.toHTML.processObject): Added.
(WI.Recording.prototype.toHTML.processValue): Added.

• UserInterface/Views/RecordingContentView.js:

(WI.RecordingContentView):
(WI.RecordingContentView.prototype._exportRecording):
(WI.RecordingContentView.prototype._exportReduction): Added.
(WI.RecordingContentView.prototype._updateExportButton): Added.
(WI.RecordingContentView.prototype._handleExportNavigationItemClicked): Added.
(WI.RecordingContentView.prototype._handleRecordingProcessedAction):

• UserInterface/Views/CanvasOverviewContentView.js:

(WI.CanvasOverviewContentView):

• UserInterface/Views/HeapAllocationsTimelineView.js:

(WI.HeapAllocationsTimelineView):

• UserInterface/Views/HeapSnapshotContentView.js:

(WI.HeapSnapshotContentView):
Drive-by: s/toolTip/tooltip.

• Localizations/en.lproj/localizedStrings.js:

LayoutTests:

• inspector/canvas/recording-html-2d.html: Added.
• inspector/canvas/recording-html-2d-expected.txt: Added.
• inspector/canvas/recording-2d-expected.txt: Added.
• inspector/canvas/resources/recording-utilities.js:

(TestPage.registerInitializer.log):
(TestPage.registerInitializer.window.startRecording):

• platform/ios-wk1/TestExpectations:
• platform/mac-wk1/TestExpectations:

11:56 AM Changeset in webkit [242808] by Devin Rousso

• 12 edits
  2 adds in trunk

Web Inspector: Audit: there should be a centralized place for reusable code
​https://bugs.webkit.org/show_bug.cgi?id=195265
<rdar://problem/47040673>

Reviewed by Joseph Pecoraro.

Source/JavaScriptCore:

• inspector/protocol/Audit.json:

Increment version.

Source/WebInspectorUI:

• UserInterface/Controllers/AuditManager.js:

(WI.AuditManager.prototype.async start):
(WI.AuditManager.prototype._topLevelTestForTest): Added.
(WI.AuditManager.prototype._topLevelTestForTest.walk): Added.

• UserInterface/Models/AuditTestBase.js:

(WI.AuditTestBase):
(WI.AuditTestBase.prototype.async setup): Added.
(WI.AuditTestBase.toJSON):

• UserInterface/Models/AuditTestCase.js:

(WI.AuditTestCase.async.fromPayload):
(WI.AuditTestCase.prototype.async run.async parseResponse):
Allow additional data to be passed back to the result's data for testing.

• UserInterface/Models/AuditTestGroup.js:

(WI.AuditTestGroup.async.fromPayload):

LayoutTests:

• inspector/audit/manager-start-setup.html: Added.
• inspector/audit/manager-start-setup-expected.txt: Added.
• inspector/model/auditTestCase.html:
• inspector/model/auditTestCase-expected.txt:
• inspector/model/auditTestGroup.html:
• inspector/model/auditTestGroup-expected.txt:

11:51 AM Changeset in webkit [242807] by pvollan@apple.com

• 9 edits in trunk/Source/WebKit

[iOS] Block access to backboardd service
​https://bugs.webkit.org/show_bug.cgi?id=195484

Reviewed by Brent Fulgham.

This patch is addressing blocking the backboardd service "com.apple.backboard.hid.services". Getting the
backlight level in the WebContent process will initiate a connection with this service. To be able to
block the service, the backlight level is queried in the UI process and sent to the WebContent process
when the WebContent process is started, and when the backlight level is changed. On the WebContent side,
the method getting the backlight level is swizzled to return the value sent from the UI process.

• Resources/SandboxProfiles/ios/com.apple.WebKit.WebContent.sb:
• UIProcess/Cocoa/WebProcessPoolCocoa.mm:

(WebKit::WebProcessPool::displayBrightness):
(WebKit::WebProcessPool::backlightLevelDidChangeCallback):
(WebKit::WebProcessPool::registerNotificationObservers):
(WebKit::WebProcessPool::unregisterNotificationObservers):

• UIProcess/WebProcessPool.cpp:

(WebKit::WebProcessPool::initializeNewWebProcess):

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

(WebKit::currentBacklightLevel):
(WebKit::WebProcess::backlightLevelDidChange):

11:25 AM Changeset in webkit [242806] by timothy_horton@apple.com

• 2 edits in trunk/Source/WebKit

Fix the build after 242801

• UIProcess/ios/WKActionSheetAssistant.mm:

(-[WKActionSheetAssistant _elementActionForDDAction:]):
More.

11:23 AM Changeset in webkit [242805] by timothy_horton@apple.com

• 2 edits in trunk/Source/WebKit

Fix the build after 242801

• UIProcess/ios/WKActionSheetAssistant.mm:

(-[WKActionSheetAssistant _elementActionForDDAction:]):

11:08 AM Changeset in webkit [242804] by youenn@apple.com

• 8 edits in trunk/Source/WebCore

Rename originsMatch in originSerializationsMatch
​https://bugs.webkit.org/show_bug.cgi?id=195572

Reviewed by Jiewen Tan.

In addition to renaming, make use of SecurityOrigin::isSameOriginAs
where it makes more sense than to compare origin serialization.
The main difference is that isSameOriginAs will return false for two different unique origins
while originsSerializationsMatch will not.

• Modules/credentialmanagement/CredentialsContainer.cpp:

(WebCore::CredentialsContainer::doesHaveSameOriginAsItsAncestors):

• Modules/mediastream/RTCController.cpp:

(WebCore::matchDocumentOrigin):

• Modules/mediastream/RTCPeerConnection.cpp:

(WebCore::RTCPeerConnection::certificatesFromConfiguration):

• loader/DocumentLoader.cpp:

(WebCore::DocumentLoader::shouldOpenExternalURLsPolicyToPropagate const):

• loader/cache/CachedResourceLoader.cpp:

(WebCore::CachedResourceLoader::shouldUpdateCachedResourceWithCurrentRequest):

• page/SecurityOrigin.cpp:

(WebCore::serializedOriginsMatch):
(WebCore::originsMatch): Deleted.

• page/SecurityOrigin.h:

10:47 AM Changeset in webkit [242803] by eric.carlson@apple.com

• 2 edits in trunk/Source/WebKit

[iOS] Sandbox must allow mach lookup required to compress video
​https://bugs.webkit.org/show_bug.cgi?id=195627
<rdar://problem/48811072>

Reviewed by Youenn Fablet.

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

10:42 AM Changeset in webkit [242802] by rmorisset@apple.com

• 3 edits in trunk/Source/JavaScriptCore

blocksInPreOrder and blocksInPostOrder should reserve the right capacity for their result vector
​https://bugs.webkit.org/show_bug.cgi?id=195595

Reviewed by Saam Barati.

Also change BlockList from being Vector<BasicBlock*, 5> to Vector<BasicBlock*>

• dfg/DFGBasicBlock.h:
• dfg/DFGGraph.cpp:

(JSC::DFG::Graph::blocksInPreOrder):
(JSC::DFG::Graph::blocksInPostOrder):

10:41 AM Changeset in webkit [242801] by commit-queue@webkit.org

• 6 edits in trunk/Source

Check whether to launch a default action instead of action sheet
​https://bugs.webkit.org/show_bug.cgi?id=195225
<rdar://problem/47715544>

Patch by Jennifer Moore <​jennifer.moore@apple.com> on 2019-03-12
Source/WebCore/PAL:

Reviewed by Daniel Bates.

Add new SPI declarations.

• pal/spi/ios/DataDetectorsUISPI.h:

Source/WebKit:

Reviewed by Daniel Bates and Tim Horton.

Notify DataDetectors at the start of a touch on a link, and check whether to immediately
launch the default action instead of an action sheet.

• UIProcess/ios/WKActionSheetAssistant.h:
• UIProcess/ios/WKActionSheetAssistant.mm:

(-[WKActionSheetAssistant interactionDidStart]):
(-[WKActionSheetAssistant _createSheetWithElementActions:defaultTitle:showLinkTitle:]):
(-[WKActionSheetAssistant showImageSheet]):
(-[WKActionSheetAssistant showLinkSheet]):
(-[WKActionSheetAssistant showDataDetectorsSheet]):
(-[WKActionSheetAssistant _createSheetWithElementActions:showLinkTitle:]): Deleted.

• UIProcess/ios/WKContentViewInteraction.mm:

(-[WKContentView _webTouchEventsRecognized:]):

10:23 AM Changeset in webkit [242800] by dbates@webkit.org

• 2 edits
  1 add in trunk/LayoutTests

[iOS] Add test to ensure that a web page can prevent the default for Command + A
​https://bugs.webkit.org/show_bug.cgi?id=192425

Reviewed by Wenson Hsieh.

Skip the test until we have the UIKit fix for <rdar://problem/46430796>.

• fast/events/ios/key-command-select-all-prevent-default.html: Added.
• platform/ios/TestExpectations:

10:22 AM Changeset in webkit [242799] by timothy_horton@apple.com

• 3 edits in trunk/Tools

Fix the build

• TestRunnerShared/spi/PencilKitTestSPI.h:
• TestWebKitAPI/ios/PencilKitTestSPI.h:

10:15 AM Changeset in webkit [242798] by Alan Bujtas

• 8 edits
  1 copy
  1 add in trunk

[Synthetic Click] Dispatch mouseout soon after mouseup
​https://bugs.webkit.org/show_bug.cgi?id=195575
<rdar://problem/47093049>

Reviewed by Simon Fraser.

Source/WebCore:

Let's fire a mouseout event when a click is submitted as the result of a tap. It helps to dismiss content which would otherwise require you to move the mouse (cases like control bar on youtube.com).

Test: fast/events/touch/ios/content-observation/mouse-out-event-should-fire-on-click.html

• page/EventHandler.h:
• page/ios/EventHandlerIOS.mm:

(WebCore::EventHandler::dispatchFakeMouseOut):

Source/WebKit:

• WebProcess/WebPage/ios/WebPageIOS.mm:

(WebKit::WebPage::completeSyntheticClick):

LayoutTests:

• fast/events/touch/ios/content-observation/mouse-out-event-should-fire-on-click-expected.txt: Added.
• fast/events/touch/ios/content-observation/mouse-out-event-should-fire-on-click.html: Added.

9:50 AM Changeset in webkit [242797] by rniwa@webkit.org

• 2 edits in trunk/Source/WebCore

In CachedFrame's constructor, release-assert that DOMWindow still has a frame after page-caching subframes
​https://bugs.webkit.org/show_bug.cgi?id=195609

Reviewed by Chris Dumez.

r242677 added release assertions to DOMWindow::suspendForPageCache. But when the first release assert in
that function is hit, we still can't tell whether active DOM objects are detaching frames, or if creating
CachedFrame's on one of subframes is causing the frame to go way.

Add a release assertion immediately after creating CachedFrame on subframes to detect this case.

• history/CachedFrame.cpp:

(WebCore::CachedFrame::CachedFrame):

9:40 AM Changeset in webkit [242796] by Alan Bujtas

• 10 edits in trunk/Source

[ContentChangeObserver] Stop content change observation when the touch event turns into long press
​https://bugs.webkit.org/show_bug.cgi?id=195601
<rdar://problem/48796324>

Reviewed by Wenson Hsieh.

Source/WebCore:

Cancel the ongoing content observation (started at touchStart) when the touch event does not turn into a tap gesture.

Not testable because any subsequent tap would reset the state anyway (though it might be measurable through some code triggering heavy content change).

• page/ios/ContentChangeObserver.cpp:

(WebCore::ContentChangeObserver::didRecognizeLongPress):
(WebCore::ContentChangeObserver::willNotProceedWithClick):

• page/ios/ContentChangeObserver.h:

Source/WebKit:

Add didRecognizeLongPress() message to be able to cancel content observation (started at touchStart).

• UIProcess/WebPageProxy.h:
• UIProcess/ios/WKContentViewInteraction.mm:

(-[WKContentView _longPressRecognized:]):

• UIProcess/ios/WebPageProxyIOS.mm:

(WebKit::WebPageProxy::didRecognizeLongPress):

• WebProcess/WebPage/WebPage.h:
• WebProcess/WebPage/WebPage.messages.in:
• WebProcess/WebPage/ios/WebPageIOS.mm:

(WebKit::WebPage::didRecognizeLongPress):

8:07 AM Changeset in webkit [242795] by aestes@apple.com

• 3 edits in trunk/Source/WebKit

[Mac] Ensure Apple Pay is unavailable when PassKit.framework is missing
​https://bugs.webkit.org/show_bug.cgi?id=195583
<rdar://problem/48420224>

Reviewed by Daniel Bates.

PassKit.framework is optionally soft-linked on Mac because it is missing from the Recovery
Partition. We need to check if the framework is available before calling into it.

• Shared/ApplePay/cocoa/WebPaymentCoordinatorProxyCocoa.mm:

(WebKit::WebPaymentCoordinatorProxy::platformCanMakePayments):
(WebKit::WebPaymentCoordinatorProxy::platformCanMakePaymentsWithActiveCard):
(WebKit::WebPaymentCoordinatorProxy::platformOpenPaymentSetup):
(WebKit::WebPaymentCoordinatorProxy::platformAvailablePaymentNetworks):

• Shared/ApplePay/mac/WebPaymentCoordinatorProxyMac.mm:

(WebKit::WebPaymentCoordinatorProxy::platformShowPaymentUI):

7:45 AM Changeset in webkit [242794] by Antti Koivisto

• 25 edits
  2 adds in trunk

Compositing layer that renders two positioned elements should not hit test
​https://bugs.webkit.org/show_bug.cgi?id=195371
<rdar://problem/48649586>

Reviewed by Simon Fraser.

Source/WebCore:

Compute and pass an event region for layers if it differs from layer bounds.

This patch fixes various block overflow and layer expansion cases. It does not handle
overflowing line boxes yet (it adds tests for those too).

Test: fast/scrolling/ios/overflow-scroll-overlap-2.html

• platform/graphics/GraphicsLayer.cpp:

(WebCore::GraphicsLayer::setEventRegion):

• platform/graphics/GraphicsLayer.h:

(WebCore::GraphicsLayer::eventRegion):

• platform/graphics/Region.h:
• platform/graphics/ca/GraphicsLayerCA.cpp:

(WebCore::GraphicsLayerCA::setEventRegion):
(WebCore::GraphicsLayerCA::commitLayerChangesBeforeSublayers):
(WebCore::GraphicsLayerCA::updateEventRegion):

Pass the region via the main platform layer of the graphics layer.

• platform/graphics/ca/GraphicsLayerCA.h:
• platform/graphics/ca/PlatformCALayer.h:
• platform/graphics/ca/cocoa/PlatformCALayerCocoa.h:
• rendering/PaintInfo.h:
• rendering/PaintPhase.h:

Add EventRegion paint phase that computes the region instead of painting anything.

• rendering/RenderBlock.cpp:

(WebCore::RenderBlock::paintObject):

• rendering/RenderLayer.cpp:

(WebCore::RenderLayer::paintForegroundForFragments):
(WebCore::RenderLayer::paintForegroundForFragmentsWithPhase):

Invoke EventRegion paint phase.

• rendering/RenderLayer.h:
• rendering/RenderLayerBacking.cpp:

(WebCore::RenderLayerBacking::paintIntoLayer):

Request event region when pointing a layer.

Source/WebKit:

• Shared/RemoteLayerTree/RemoteLayerTreePropertyApplier.mm:

(WebKit::RemoteLayerTreePropertyApplier::applyProperties):

• Shared/RemoteLayerTree/RemoteLayerTreeTransaction.h:
• Shared/RemoteLayerTree/RemoteLayerTreeTransaction.mm:

(WebKit::RemoteLayerTreeTransaction::LayerProperties::LayerProperties):
(WebKit::RemoteLayerTreeTransaction::LayerProperties::encode const):
(WebKit::RemoteLayerTreeTransaction::LayerProperties::decode):

Pass event region to UI process.

• UIProcess/RemoteLayerTree/RemoteLayerTreeNode.h:

(WebKit::RemoteLayerTreeNode::layerID const):
(WebKit::RemoteLayerTreeNode::eventRegion const):

• UIProcess/RemoteLayerTree/RemoteLayerTreeNode.mm:

(WebKit::RemoteLayerTreeNode::RemoteLayerTreeNode):
(WebKit::RemoteLayerTreeNode::~RemoteLayerTreeNode):
(WebKit::RemoteLayerTreeNode::setEventRegion):

Maintain event region in RemoteLayerTreeNodes.

(WebKit::RemoteLayerTreeNode::initializeLayer):
(WebKit::RemoteLayerTreeNode::layerID):
(WebKit::RemoteLayerTreeNode::forCALayer):

Move layerID to RemoteLayerTreeNode and store RemoteLayerTreeNode pointer to CALayers instead.
This makes it easy to find the matching RemoteLayerTreeNode from a layer, globally.

(WebKit::RemoteLayerTreeNode::setLayerID): Deleted.

• UIProcess/RemoteLayerTree/ios/RemoteLayerTreeViews.mm:

(WebKit::collectDescendantViewsAtPoint):

If we have event region, use it for hit testing.

(-[UIView _web_findDescendantViewAtPoint:withEvent:]):
(collectDescendantViewsAtPoint): Deleted.

• WebProcess/WebPage/RemoteLayerTree/PlatformCALayerRemote.cpp:

(WebKit::PlatformCALayerRemote::eventRegion const):
(WebKit::PlatformCALayerRemote::setEventRegion):

• WebProcess/WebPage/RemoteLayerTree/PlatformCALayerRemote.h:

LayoutTests:

• fast/scrolling/ios/overflow-scroll-overlap-2-expected.txt: Added.
• fast/scrolling/ios/overflow-scroll-overlap-2.html: Added.

7:18 AM WebKitGTK/2.24.x edited by Philippe Normand

(diff)

7:17 AM Changeset in webkit [242793] by Philippe Normand

• 6 edits in trunk/Source/WebCore

[GStreamer][v4l2] Synchronous video texture flushing support
​https://bugs.webkit.org/show_bug.cgi?id=195453

Reviewed by Xabier Rodriguez-Calvar.

The v4l2 video decoder currently requires that downstream users of
the graphics resources complete any pending draw call and release
resources before returning from the DRAIN query.

To accomplish this the player monitors the pipeline and whenever a
v4l2 decoder is added, synchronous video texture flushing support
is enabled. Additionally and for all decoder configurations, a
flush is performed before disposing of the player.

• platform/graphics/gstreamer/MediaPlayerPrivateGStreamer.cpp:

(WebCore::MediaPlayerPrivateGStreamer::playbinDeepElementAddedCallback):
Monitor elements added to the decodebin bin.
(WebCore::MediaPlayerPrivateGStreamer::decodebinElementAdded): Set
a flag if a v4l2 decoder was added in decodebin.
(WebCore::MediaPlayerPrivateGStreamer::createGSTPlayBin): Connect
to the deep-element-added signal so as to monitor pipeline
topology updates.

• platform/graphics/gstreamer/MediaPlayerPrivateGStreamer.h:
• platform/graphics/gstreamer/MediaPlayerPrivateGStreamerBase.cpp:

(WebCore::MediaPlayerPrivateGStreamerBase::~MediaPlayerPrivateGStreamerBase):
Flush video texture before disposing of the player.
(WebCore::MediaPlayerPrivateGStreamerBase::flushCurrentBuffer):
Synchronously flush if the pipeline contains a v4l2 decoder.
(WebCore::MediaPlayerPrivateGStreamerBase::createGLAppSink): Monitor push events only.

• platform/graphics/gstreamer/MediaPlayerPrivateGStreamerBase.h:
• platform/graphics/texmap/TextureMapperPlatformLayerProxy.cpp:

(WebCore::TextureMapperPlatformLayerProxy::pushNextBuffer): New
boolean flag used mostly to trigger synchronous flush conditions.
(WebCore::TextureMapperPlatformLayerProxy::dropCurrentBufferWhilePreservingTexture):
Optionally drop the current buffer in a synchronous manner. By
default the method keeps operating asynchronously.

• platform/graphics/texmap/TextureMapperPlatformLayerProxy.h:

4:38 AM Changeset in webkit [242792] by rniwa@webkit.org

• 7 edits in trunk/Source/WebCore

Move the remaining code to decide whether site specific quirks are needed to Quirks class
​https://bugs.webkit.org/show_bug.cgi?id=195610

Reviewed by Antti Koivisto.

Moved the remaining code scattered across WebCore to decide whether a site specific quirk
is needed or not to Quirks class introduced in r236818.

• Modules/fetch/FetchRequest.cpp:

(WebCore::needsSignalQuirk): Deleted.
(WebCore::processInvalidSignal):

• html/HTMLFormControlElement.cpp:

(WebCore::HTMLFormControlElement::needsMouseFocusableQuirk const):

• html/HTMLMediaElement.cpp:

(WebCore::needsAutoplayPlayPauseEventsQuirk): Deleted.
(WebCore::HTMLMediaElement::dispatchPlayPauseEventsIfNeedsQuirks):
(WebCore::needsSeekingSupportQuirk): Deleted.
(WebCore::HTMLMediaElement::supportsSeeking const):

• html/MediaElementSession.cpp:

(WebCore::needsArbitraryUserGestureAutoplayQuirk): Deleted.
(WebCore::needsPerDocumentAutoplayBehaviorQuirk): Deleted.
(WebCore::MediaElementSession::playbackPermitted const):

• page/Quirks.cpp:

(WebCore::allowedAutoplayQuirks): Added.
(WebCore::Quirks::needsQuirks const): Added.
(WebCore::Quirks::shouldIgnoreInvalidSignal const): Added.
(WebCore::Quirks::needsFormControlToBeMouseFocusable const): Added.
(WebCore::Quirks::needsAutoplayPlayPauseEvents const): Added.
(WebCore::Quirks::needsSeekingSupportDisabled const): Addd.
(WebCore::Quirks::needsPerDocumentAutoplayBehavior const): Added.
(WebCore::Quirks::shouldAutoplayForArbitraryUserGesture const): Added.
(WebCore::Quirks::hasBrokenEncryptedMediaAPISupportQuirk const): Added.
(WebCore::Quirks::hasWebSQLSupportQuirk const): Fixed the coding style.

• page/Quirks.h:

4:09 AM Changeset in webkit [242791] by eocanha@igalia.com

• 7 edits in trunk

[Media][MSE] Don't emit timeUpdate after play() if currentTime hasn't changed
​https://bugs.webkit.org/show_bug.cgi?id=195454

Reviewed by Jer Noble.

Source/WebCore:

This change fixes YouTube 2019 MSE Conformance Tests "26. SFRPausedAccuracy"
and "27. HFRPausedAccuracy".

The first timeUpdate event after play() is omitted, because currentTime
doesn't actually change in that scenario.

Tests 26 and 27 measure the time drift (real time vs. media time) on playback
and start counting since the first timeUpdate event. In WebKit, that event
happens at play(), before the pipeline has completed the transition to playing.
Therefore, the real time inherits this startup delay and the test thinks that
the player has drifted.

• html/HTMLMediaElement.cpp:

(WebCore::HTMLMediaElement::playInternal): Don't emit a timeUpdated event unless currentTime has changed.

LayoutTests:

This patch removes expectations for the first timeUpdate event after
play(), because currentTime doesn't actually change in that scenario
and the spec[1] states that a timeupdate event is fired if "The current
playback position changed as part of normal playback or in an
especially interesting way, for example discontinuously."

[1] ​https://www.w3.org/TR/html52/semantics-embedded-content.html#eventdef-media-timeupdate

• media/video-paused-0-rate.html: Don't require the timeUpdate event when currentTime=0 to pass the test.
• media/video-play-pause-events-expected.txt: Ditto, and changed test description.
• media/video-play-pause-events.html: Changed test description to reflect the new behaviour.
• media/video-play-pause-exception-expected.txt: Don't require the timeUpdate event.

3:59 AM Changeset in webkit [242790] by eocanha@igalia.com

• 2 edits in trunk/Source/WebCore

[EME][GStreamer] Speculative build fix
​https://bugs.webkit.org/show_bug.cgi?id=195614

Unreviewed speculative WPE build fix after r242776.

• platform/graphics/gstreamer/eme/WebKitCommonEncryptionDecryptorGStreamer.h: Added missing include.

3:25 AM Changeset in webkit [242789] by Philippe Normand

• 4 edits in trunk/Source/WebCore

[GStreamer] remove legacy GStreamer version checks
​https://bugs.webkit.org/show_bug.cgi?id=195552

Reviewed by Xabier Rodriguez-Calvar.

We require GStreamer 1.8.x so version checks below that make
little sense. Also checks for odd minor version numbers make sense
only for the latest GStreamer git development version.

• platform/graphics/gstreamer/GStreamerCommon.cpp:

(WebCore::initializeGStreamerAndRegisterWebKitElements):

• platform/graphics/gstreamer/MediaPlayerPrivateGStreamer.cpp:

(WebCore::MediaPlayerPrivateGStreamer::createAudioSink):
(WebCore::MediaPlayerPrivateGStreamer::createGSTPlayBin):

• platform/graphics/gstreamer/MediaPlayerPrivateGStreamerBase.cpp:

(WebCore::MediaPlayerPrivateGStreamerBase::ensureGstGLContext):

2:08 AM WebKitGTK/2.24.x edited by Philippe Normand

(diff)

1:30 AM Changeset in webkit [242788] by Carlos Garcia Campos

• 12 edits in trunk

[WPE][GTK] Load events may occur in unexpected order when JS redirects page before subresource load finishes
​https://bugs.webkit.org/show_bug.cgi?id=194131

Source/WebKit:

Reviewed by Michael Catanzaro.

Ensure we emit the load-failed and load-changed with finished event when there's still an ongoing load when a
new provisional load strarts. Previous load fails with cancelled error.

• UIProcess/API/glib/WebKitWebView.cpp:

(webkitWebViewWillStartLoad): Call webkitWebViewLoadFailed() if current page load state is not finished.

• UIProcess/API/glib/WebKitWebViewPrivate.h:
• UIProcess/API/gtk/PageClientImpl.cpp:

(WebKit::PageClientImpl::didStartProvisionalLoadForMainFrame): Call webkitWebViewWillStartLoad().

• UIProcess/API/wpe/APIViewClient.h:

(API::ViewClient::willStartLoad): Add willStartLoad() to API::ViewClient

• UIProcess/API/wpe/PageClientImpl.cpp:

(WebKit::PageClientImpl::didStartProvisionalLoadForMainFrame): Call WPEView::willStartLoad().

• UIProcess/API/wpe/PageClientImpl.h:
• UIProcess/API/wpe/WPEView.cpp:

(WKWPE::View::willStartLoad): Call API::ViewClient::willStartLoad().

• UIProcess/API/wpe/WPEView.h:
• UIProcess/PageLoadState.h:

(WebKit::PageLoadState::isProvisional const):
(WebKit::PageLoadState::isCommitted const):
(WebKit::PageLoadState::isFinished const):

Tools:

Patch by Michael Catanzaro <Michael Catanzaro> on 2019-03-12
Reviewed by Michael Catanzaro.

• TestWebKitAPI/Tests/WebKitGLib/TestLoaderClient.cpp:

(uriChanged):
(testUnfinishedSubresourceLoad):
(serverCallback):
(beforeAll):

1:29 AM Changeset in webkit [242787] by calvaris@igalia.com

• 2 edits in trunk/Source/WebCore

[EME] generateRequest was not using the sanitized init data
​https://bugs.webkit.org/show_bug.cgi?id=195555

Reviewed by Jer Noble.

• Modules/encryptedmedia/MediaKeySession.cpp:

(WebCore::MediaKeySession::generateRequest): Use sanitized init
data instead of the original one.

1:13 AM Changeset in webkit [242786] by commit-queue@webkit.org

• 7 edits in trunk

Implement further CORS restrictions
​https://bugs.webkit.org/show_bug.cgi?id=188644

Patch by Rob Buis <​rbuis@igalia.com> on 2019-03-12
Reviewed by Darin Adler.

LayoutTests/imported/w3c:

Update improved test results.

• web-platform-tests/fetch/api/cors/cors-preflight-not-cors-safelisted.any-expected.txt:
• web-platform-tests/fetch/api/cors/cors-preflight-not-cors-safelisted.any.worker-expected.txt:
• web-platform-tests/fetch/api/headers/headers-no-cors.window-expected.txt:

Source/WebCore:

Verify that header value length is not greater than 128 [1]. Also implement
Step 5 of [2] to append values to existing header value when calling
Headers.append.

Tests: fetch/api/cors/cors-preflight-not-cors-safelisted.any.html

fetch/api/cors/cors-preflight-not-cors-safelisted.any.worker.html
fetch/api/headers/headers-no-cors.window.html

[1] ​https://fetch.spec.whatwg.org/#cors-safelisted-request-header
[2] ​https://fetch.spec.whatwg.org/#concept-headers-append

• Modules/fetch/FetchHeaders.cpp:

(WebCore::canWriteHeader):
(WebCore::appendToHeaderMap):
(WebCore::FetchHeaders::remove):
(WebCore::FetchHeaders::set):
(WebCore::FetchHeaders::filterAndFill):

• platform/network/HTTPParsers.cpp:

(WebCore::isCrossOriginSafeRequestHeader): verify that header length is not greater than 128

12:10 AM Changeset in webkit [242785] by commit-queue@webkit.org

• 3 edits in trunk/Source/WebInspectorUI

Web Inspector: Timelines - Improve handling of past recordings (readonly)
​https://bugs.webkit.org/show_bug.cgi?id=195594

Patch by Joseph Pecoraro <Joseph Pecoraro> on 2019-03-12
Reviewed by Devin Rousso.

• UserInterface/Views/TimelineRecordingContentView.js:

(WI.TimelineRecordingContentView.prototype._clearTimeline):
Don't allow clearing of a readonly recording.

• UserInterface/Views/TimelineTabContentView.js:

(WI.TimelineTabContentView.prototype._toggleRecordingOnSpacebar):
Don't do anything when viewing a readonly recording.

(WI.TimelineTabContentView.prototype._recordButtonClicked):
Start a new recording if viewing a readonly recording.