Timeline

Nov 26, 2018:

8:17 PM Changeset in webkit [238545] by dbates@webkit.org

• 5 edits
  4 adds in trunk

REGRESSION (r238078): Do not draw caps lock indicator when Strong Password button is shown
​https://bugs.webkit.org/show_bug.cgi?id=191969
<rdar://problem/46247569>

Reviewed by Dean Jackson.

Source/WebCore:

Following r238078 we now support drawing the caps lock indicator in password fields on iOS.
However it is not meaningful to show the caps lock indicator when the Strong Password button
is visible because the password field is not editable. We should not paint the caps lock
indicator when the Strong Password button is visible.

Tests: fast/forms/auto-fill-button/caps-lock-indicator-should-be-visible-after-hiding-auto-fill-strong-password-button.html

fast/forms/auto-fill-button/caps-lock-indicator-should-not-be-visible-when-auto-fill-strong-password-button-is-visible.html

• html/TextFieldInputType.cpp:

(WebCore::TextFieldInputType::shouldDrawCapsLockIndicator const): Do not draw the caps
lock indicator when the password field has the Strong Password button.
(WebCore::TextFieldInputType::updateAutoFillButton): Call capsLockStateMayHaveChanged() to
update the visibility of the caps lock indicator when the auto fill button has changed.

LayoutTests:

Add tests to ensure that we show or hide the caps lock indicator depending on whether the
Strong Password button is visible.

• TestExpectations: Skip the tests below on all platforms. We will selectively enable them on Mac.
• fast/forms/auto-fill-button/caps-lock-indicator-should-be-visible-after-hiding-auto-fill-strong-password-button-expected.html: Added.
• fast/forms/auto-fill-button/caps-lock-indicator-should-be-visible-after-hiding-auto-fill-strong-password-button.html: Added.
• fast/forms/auto-fill-button/caps-lock-indicator-should-not-be-visible-when-auto-fill-strong-password-button-is-visible-expected.html: Added.
• fast/forms/auto-fill-button/caps-lock-indicator-should-not-be-visible-when-auto-fill-strong-password-button-is-visible.html: Added.
• platform/mac-wk2/TestExpectations: Mark the tests above as PASS so that we run them.

7:32 PM Changeset in webkit [238544] by Wenson Hsieh

• 5 edits
  1 add in trunk

[Cocoa] No way for clients to tell whether the content view is in the responder chain when the web view is
​https://bugs.webkit.org/show_bug.cgi?id=169212
<rdar://problem/30899656>

Reviewed by Tim Horton.

Source/WebKit:

Add an SPI hook to allow internal WKWebView clients to determine whether the WKWebView's content view is the
first responder. Intended for use by clients, such as Mail, that embed native text input views and other views
that may become first responder within the view hierarchy of the WKWebView.

• UIProcess/API/Cocoa/WKWebView.mm:

(-[WKWebView _contentViewIsFirstResponder]):

• UIProcess/API/Cocoa/WKWebViewPrivate.h:

Tools:

Add an API test to exercise the behavior of -_contentViewIsFirstResponder when an embedded text input becomes
and resigns first responder.

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

(-[FirstResponderTestingView initWithFrame:]):
(-[FirstResponderTestingView inputField]):
(TestWebKitAPI::TEST):

6:26 PM Changeset in webkit [238543] by Caio Lima

• 40 edits
  2 adds in trunk

Re-introduce op_bitnot
​https://bugs.webkit.org/show_bug.cgi?id=190923

Reviewed by Yusuke Suzuki.

JSTests:

• stress/bit-not-must-generate.js: Added.
• stress/bitwise-not-no-int32.js: Added.

Source/JavaScriptCore:

With the introduction of BigInt as a new type, we can't emit bitwise
not as x ^ -1 anymore, because this is incompatible with the new type.
Based on that, this Patch is adding op_bitnot as a new operation
into LLInt, as well as introducing ArithBitNot node into DFG to support
JIT compilation of such opcode. We will use the ValueProfile of this
intruction in the future to generate better code when its operand
is not Int32.

• assembler/MacroAssemblerARM64.h:

(JSC::MacroAssemblerARM64::not32):

• assembler/MacroAssemblerARMv7.h:

(JSC::MacroAssemblerARMv7::not32):

• assembler/MacroAssemblerMIPS.h:

(JSC::MacroAssemblerMIPS::not32):

• bytecode/BytecodeList.rb:
• bytecode/BytecodeUseDef.h:

(JSC::computeUsesForBytecodeOffset):
(JSC::computeDefsForBytecodeOffset):

• bytecode/CodeBlock.cpp:

(JSC::CodeBlock::finishCreation):

• bytecode/Opcode.h:

(JSC::padOpcodeName):

• bytecompiler/BytecodeGenerator.cpp:

(JSC::BytecodeGenerator::emitUnaryOp):

• bytecompiler/NodesCodegen.cpp:

(JSC::UnaryPlusNode::emitBytecode):
(JSC::BitwiseNotNode::emitBytecode): Deleted.

• dfg/DFGAbstractInterpreterInlines.h:

(JSC::DFG::AbstractInterpreter<AbstractStateType>::executeEffects):

• dfg/DFGBackwardsPropagationPhase.cpp:

(JSC::DFG::BackwardsPropagationPhase::propagate):

• dfg/DFGByteCodeParser.cpp:

(JSC::DFG::ByteCodeParser::parseBlock):

• dfg/DFGCapabilities.cpp:

(JSC::DFG::capabilityLevel):

• dfg/DFGClobberize.h:

(JSC::DFG::clobberize):

• dfg/DFGDoesGC.cpp:

(JSC::DFG::doesGC):

• dfg/DFGFixupPhase.cpp:

(JSC::DFG::FixupPhase::fixupNode):

• dfg/DFGNodeType.h:
• dfg/DFGOperations.cpp:
• dfg/DFGOperations.h:
• dfg/DFGPredictionPropagationPhase.cpp:
• dfg/DFGSafeToExecute.h:

(JSC::DFG::safeToExecute):

• dfg/DFGSpeculativeJIT.cpp:

(JSC::DFG::SpeculativeJIT::compileBitwiseNot):

• dfg/DFGSpeculativeJIT.h:
• dfg/DFGSpeculativeJIT32_64.cpp:

(JSC::DFG::SpeculativeJIT::compile):

• dfg/DFGSpeculativeJIT64.cpp:

(JSC::DFG::SpeculativeJIT::compile):

• ftl/FTLCapabilities.cpp:

(JSC::FTL::canCompile):

• ftl/FTLLowerDFGToB3.cpp:

(JSC::FTL::DFG::LowerDFGToB3::compileNode):
(JSC::FTL::DFG::LowerDFGToB3::compileArithBitNot):

• jit/JIT.cpp:

(JSC::JIT::privateCompileMainPass):
(JSC::JIT::privateCompileSlowCases):

• jit/JIT.h:
• jit/JITArithmetic.cpp:

(JSC::JIT::emit_op_bitnot):

• llint/LowLevelInterpreter32_64.asm:
• llint/LowLevelInterpreter64.asm:
• offlineasm/cloop.rb:
• parser/NodeConstructors.h:

(JSC::BitwiseNotNode::BitwiseNotNode):

• parser/Nodes.h:
• parser/ResultType.h:

(JSC::ResultType::bigIntOrInt32Type):
(JSC::ResultType::forBitOp):

• runtime/CommonSlowPaths.cpp:

(JSC::SLOW_PATH_DECL):

• runtime/CommonSlowPaths.h:

6:09 PM Changeset in webkit [238542] by Fujii Hironori

• 2 edits in trunk/Source/WebKitLegacy/win

[Clang][WebKitLegacy][Win] MarshallingHelpers.h(65,47): warning: extra qualification on member 'windowsEpochAbsoluteTime' [-Wmicrosoft-extra-qualification]
​https://bugs.webkit.org/show_bug.cgi?id=191959

Reviewed by Alex Christensen.

• MarshallingHelpers.h: Removed extra "MarshallingHelpers::" of windowsEpochAbsoluteTime declaration.

6:01 PM Changeset in webkit [238541] by Fujii Hironori

• 2 edits in trunk/Source/WTF

[Win][Clang] SOFT_LINK reports warning: implicit conversion between pointer-to-function and pointer-to-object is a Microsoft extension [-Wmicrosoft-cast]
​https://bugs.webkit.org/show_bug.cgi?id=191960

Reviewed by Alex Christensen.

• wtf/win/SoftLinking.h: Do reinterpret_cast<void*> a function

pointer argument of EncodePointer. Changed the type of stored
function pointer returned by EncodePointer.

5:59 PM Changeset in webkit [238540] by Ryan Haddad

• 5 edits
  4 deletes in trunk

Unreviewed, rolling out r238357.

One of the layout tests added with this change is failing on Mojave.

Reverted changeset:
"REGRESSION (r238078): Do not draw caps lock indicator when Strong Password button is shown"
​https://bugs.webkit.org/show_bug.cgi?id=191969
​https://trac.webkit.org/changeset/238513

5:54 PM Changeset in webkit [238539] by Fujii Hironori

• 7 edits in trunk

[CMake] Remove ENABLE_ACCESSIBILITY CMake variable
​https://bugs.webkit.org/show_bug.cgi?id=191831

Reviewed by Michael Catanzaro.

.:

ENABLE_ACCESSIBILITY is a bit confusing name because there is no
such ENABLE_ACCESSIBILITY macros in C++ source files.

• Source/cmake/OptionsGTK.cmake: Removed ENABLE_ACCESSIBILITY.
• Source/cmake/OptionsWPE.cmake: Ditto.
• Source/cmake/OptionsWin.cmake: Ditto.
• Source/cmake/WebKitFeatures.cmake: Ditto.

Tools:

• WebKitTestRunner/CMakeLists.txt:

5:37 PM Changeset in webkit [238538] by timothy_horton@apple.com

• 30 edits
  1 copy
  6 adds in trunk

Insert <attachment> elements under editable images to make their backing data accessible
​https://bugs.webkit.org/show_bug.cgi?id=191844
<rdar://problem/30900149>

Reviewed by Simon Fraser.

Source/WebCore:

Test: editing/images/editable-image-creates-attachment.html

• html/HTMLImageElement.cpp:

(WebCore::HTMLImageElement::parseAttribute):
(WebCore::HTMLImageElement::insertedIntoAncestor):
(WebCore::HTMLImageElement::removedFromAncestor):
When the x-apple-editable-image attribute changes, or the element is
moved into or out of a document, call updateEditableImage.

(WebCore::HTMLImageElement::editableImageViewID const):
Adopt EditableImageReference.

(WebCore::HTMLImageElement::updateEditableImage):
When the image element moves into a document, the setting is on, and
the appropriate attribute is applied, add an <attachment> into the
shadow DOM, and inform the UI process both of the editable image's
creation and that it should be associated with the new attachment.

Use an EditableImageReference to extend the lifetime of the
corresponding editable image in the UI process, and to communicate
the attachment association.

If the element was cloned from another editable image element, use the
EditableImageReference and attachmentID from the original; the embedded
view will be re-parented under this element's layer, and the attachment
will be cloned (with a new ID) by editing code if the element is parented.

(WebCore::HTMLImageElement::attachmentIdentifier const):
(WebCore::HTMLImageElement::copyNonAttributePropertiesFromElement):
Store the aforementioned bits of information when cloned so that we can
reconstitute the appropriate attachment data and embedded view.

• html/HTMLImageElement.h:
• page/ChromeClient.h:

• Sources.txt:
• WebCore.xcodeproj/project.pbxproj:
• page/EditableImageReference.cpp: Added.

(WebCore::EditableImageReference::EditableImageReference):
(WebCore::EditableImageReference::~EditableImageReference):
(WebCore::EditableImageReference::associateWithAttachment):

• page/EditableImageReference.h: Added.

(WebCore::EditableImageReference::create):
(WebCore::EditableImageReference::embeddedViewID const):
Add EditableImageReference, which manages the lifetime of the UI-side
EditableImage and helps clients communicate about it. It is refcounted
so that cloned <img> elements can potentially borrow the UI-side state
(in the case where they end up getting parented).

• page/NavigatorBase.cpp:

Fix an unrelated unified build failure that I exposed.

Source/WebKit:

• DerivedSources.make:
• SourcesCocoa.txt:
• UIProcess/API/APIAttachment.h:

fileWrapper() is no longer a trivial getter; it can now construct
the file wrapper from a file wrapper generator if necessary.

Add setFileWrapperGenerator() and invalidateGeneratedFileWrapper().

Make m_fileWrapper mutable so it can be adjusted inside its own getter.

• UIProcess/API/Cocoa/APIAttachmentCocoa.mm:

(API::Attachment::fileWrapper const):
If we have a fileWrapperGenerator and don't have a cached file wrapper,
create one before returning it.

(API::Attachment::invalidateGeneratedFileWrapper):
Invalidate the currently-cached file wrapper. The next time a client
requests the file wrapper it will be regenerated.

(API::Attachment::fileName const):
(API::Attachment::fileSizeForDisplay const):
(API::Attachment::enclosingImageData const):
(API::Attachment::isEmpty const):
(API::Attachment::createSerializedRepresentation const):
Make use of fileWrapper() instead of m_fileWrapper directly, to ensure
that it is created lazily if necessary.

(API::Attachment::setFileWrapperGenerator):

• UIProcess/RemoteLayerTree/ios/RemoteLayerTreeHostIOS.mm:

(WebKit::RemoteLayerTreeHost::createEmbeddedView):

• UIProcess/RemoteLayerTree/ios/RemoteLayerTreeViews.h:
• UIProcess/RemoteLayerTree/ios/RemoteLayerTreeViews.mm:

(-[WKEmbeddedView initWithEmbeddedViewID:]):
Defer to EditableImageController for creating WKDrawingViews for
editable images. This is done primarily so we don't have to pollute
Remote Layer Tree and DrawingArea interfaces with editable-image-specific messages.

• UIProcess/WebPageProxy.cpp:

(WebKit::m_editableImageController):
(WebKit::m_resetRecentCrashCountTimer): Deleted.

• UIProcess/WebPageProxy.h:

(WebKit::WebPageProxy::editableImageController):
Keep an EditableImageController on the WebPageProxy.

• UIProcess/ios/EditableImageController.h: Added.
• UIProcess/ios/EditableImageController.messages.in: Added.
• UIProcess/ios/EditableImageController.mm: Added.

(WebKit::EditableImageController::EditableImageController):
(WebKit::EditableImageController::~EditableImageController):
(WebKit::EditableImageController::ensureEditableImage):
(WebKit::EditableImageController::editableImage):
(WebKit::EditableImageController::didCreateEditableImage):
(WebKit::EditableImageController::didDestroyEditableImage):
(WebKit::EditableImageController::associateWithAttachment):
(WebKit::EditableImageController::invalidateAttachmentForEditableImage):
Add EditableImageController, which keeps track of EditableImages.
It can be messaged directly to create or destroy the UI-side state
of an editable image, and also to associate a WKDrawingView with
a particular attachment.

• UIProcess/ios/WKDrawingView.h:
• UIProcess/ios/WKDrawingView.mm:

(-[WKDrawingView initWithEmbeddedViewID:webPageProxy:]):
Store the WebPageProxy (weakly) so that we can get to the EditableImageController.

(-[WKDrawingView layoutSubviews]):
(-[WKDrawingView PNGRepresentation]):
Synchronously render the PKCanvasView to PNG.

(-[WKDrawingView drawingDidChange:]):
If the drawing changes, inform the APIAttachment that it needs
to discard its NSFileWrapper; a new one will be generated lazily.

(-[WKDrawingView init]): Deleted.

• WebKit.xcodeproj/project.pbxproj:
• WebProcess/WebCoreSupport/WebChromeClient.h:
• WebProcess/WebCoreSupport/ios/WebChromeClientIOS.mm:

(WebKit::WebChromeClient::associateEditableImageWithAttachment):
(WebKit::WebChromeClient::didCreateEditableImage):
(WebKit::WebChromeClient::didDestroyEditableImage):

Tools:

• DumpRenderTree/ios/UIScriptControllerIOS.mm:

(WTR::UIScriptController::attachmentInfo):

• TestRunnerShared/UIScriptContext/Bindings/UIScriptController.idl:
• TestRunnerShared/UIScriptContext/UIScriptController.cpp:

(WTR::UIScriptController::attachmentInfo):

• TestRunnerShared/UIScriptContext/UIScriptController.h:
• WebKitTestRunner/ios/UIScriptControllerIOS.mm:

(WTR::UIScriptController::attachmentInfo):
Add a UIScriptController mechanism to retrieve information about
a given attachment.

LayoutTests:

• editing/images/editable-image-creates-attachment-expected.txt: Added.
• editing/images/editable-image-creates-attachment.html: Added.
• resources/ui-helper.js:

(window.UIHelper.attachmentInfo):
(window.UIHelper):

5:15 PM Changeset in webkit [238537] by Kocsen Chung

• 7 edits in tags/Safari-607.1.14.3/Source

Versioning.

5:12 PM Changeset in webkit [238536] by Kocsen Chung

• 1 copy in tags/Safari-607.1.14.3

New tag.

5:05 PM Changeset in webkit [238535] by aakash_jain@apple.com

• 1 edit
  3 adds in trunk/Tools

[ews-app] Add support to communicate with Buildbot
​https://bugs.webkit.org/show_bug.cgi?id=191935

Reviewed by Lucas Forschler.

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

5:02 PM Changeset in webkit [238534] by Jonathan Bedard

• 2 edits in trunk/Tools

webkitpy: Race condition in ServerProcess tests
​https://bugs.webkit.org/show_bug.cgi?id=191989
<rdar://problem/45536844>

Reviewed by Lucas Forschler.

Serializing these tests to prevent the race conditions.

• Scripts/webkitpy/port/server_process_unittest.py:

(TestServerProcess.serial_test_basic):
(TestServerProcess.serial_test_read_after_process_exits):
(TestServerProcess.serial_test_process_crashing):
(TestServerProcess.serial_test_process_crashing_no_data):
(TestServerProcess.test_basic): Deleted.
(TestServerProcess.test_read_after_process_exits): Deleted.
(TestServerProcess.test_process_crashing): Deleted.
(TestServerProcess.test_process_crashing_no_data): Deleted.

4:57 PM Changeset in webkit [238533] by Alan Coon

• 6 edits
  3 adds in branches/safari-606.4.1.2-branch

Cherry-pick r238326. rdar://problem/46259210

All users of ArrayBuffer should agree on the same max size
​https://bugs.webkit.org/show_bug.cgi?id=191771

Reviewed by Mark Lam.

JSTests:

• stress/big-wasm-memory-grow-no-max.js: Added. (foo): (catch):
• stress/big-wasm-memory-grow.js: Added. (foo): (catch):
• stress/big-wasm-memory.js: Added. (foo): (catch):

Source/JavaScriptCore:

Array buffers cannot be larger than 0x7fffffff, because otherwise loading typedArray.length in the DFG/FTL would produce
a uint32 or would require a signedness check, neither of which sounds reasonable. It's better to just bound their max size
instead.

• runtime/ArrayBuffer.cpp: (JSC::ArrayBufferContents::ArrayBufferContents): (JSC::ArrayBufferContents::tryAllocate): (JSC::ArrayBufferContents::transferTo): (JSC::ArrayBufferContents::copyTo): (JSC::ArrayBufferContents::shareWith):
• runtime/ArrayBuffer.h:
• wasm/WasmMemory.cpp: (JSC::Wasm::Memory::tryCreate): (JSC::Wasm::Memory::grow):
• wasm/WasmPageCount.h:

git-svn-id: ​https://svn.webkit.org/repository/webkit/trunk@238326 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4:48 PM Changeset in webkit [238532] by aakash_jain@apple.com

• 2 edits in trunk/Tools

[ews-app] Add field in Patch object to indicate if it has been sent to Buildbot
​https://bugs.webkit.org/show_bug.cgi?id=191929

Reviewed by Lucas Forschler.

• BuildSlaveSupport/ews-app/ews/models/patch.py:

(Patch): Added sent_to_buildbot field.
(Patch.is_patch_sent_to_buildbot): Checks if the patch has been sent to Buildbot.

4:47 PM Changeset in webkit [238531] by Alan Coon

• 6 edits
  3 adds in branches/safari-606-branch

Cherry-pick r238326. rdar://problem/46259215

All users of ArrayBuffer should agree on the same max size
​https://bugs.webkit.org/show_bug.cgi?id=191771

Reviewed by Mark Lam.

JSTests:

• stress/big-wasm-memory-grow-no-max.js: Added. (foo): (catch):
• stress/big-wasm-memory-grow.js: Added. (foo): (catch):
• stress/big-wasm-memory.js: Added. (foo): (catch):

Source/JavaScriptCore:

Array buffers cannot be larger than 0x7fffffff, because otherwise loading typedArray.length in the DFG/FTL would produce
a uint32 or would require a signedness check, neither of which sounds reasonable. It's better to just bound their max size
instead.

• runtime/ArrayBuffer.cpp: (JSC::ArrayBufferContents::ArrayBufferContents): (JSC::ArrayBufferContents::tryAllocate): (JSC::ArrayBufferContents::transferTo): (JSC::ArrayBufferContents::copyTo): (JSC::ArrayBufferContents::shareWith):
• runtime/ArrayBuffer.h:
• wasm/WasmMemory.cpp: (JSC::Wasm::Memory::tryCreate): (JSC::Wasm::Memory::grow):
• wasm/WasmPageCount.h:

git-svn-id: ​https://svn.webkit.org/repository/webkit/trunk@238326 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4:29 PM Changeset in webkit [238530] by aakash_jain@apple.com

• 1 edit
  1 add in trunk/Tools

[ews-app] Add a config file
​https://bugs.webkit.org/show_bug.cgi?id=191933

Reviewed by Lucas Forschler.

• BuildSlaveSupport/ews-app/ews/config.py: Added.

4:21 PM Changeset in webkit [238529] by aakash_jain@apple.com

• 2 edits in trunk/Tools

[ews-app] Add methods to save patch to database
​https://bugs.webkit.org/show_bug.cgi?id=191928

Reviewed by Lucas Forschler.

• BuildSlaveSupport/ews-app/ews/models/patch.py:

(Patch.save_patch): Method to save the patch to database.
(Patch.save_patches): Method to save multiple patches to database.
(Patch.is_valid_patch_id): Checks if the patch id is valid.
(Patch.is_existing_patch_id): Checks if the patch id already exists in database.

4:02 PM Changeset in webkit [238528] by jer.noble@apple.com

• 15 edits in trunk/Source

Adopt -setOverrideRouteSharingPolicy:routingContextUID: SPI
​https://bugs.webkit.org/show_bug.cgi?id=190951
<rdar://problem/45213065>

Reviewed by Alex Christensen.

Source/WebCore:

Request the correct route policy and context from the VideoFullscreenModel.

• platform/cocoa/VideoFullscreenModel.h:

(WebCore::VideoFullscreenModel::requestRouteSharingPolicyAndContextUID):

• platform/cocoa/VideoFullscreenModelVideoElement.h:
• platform/cocoa/VideoFullscreenModelVideoElement.mm:

(WebCore::VideoFullscreenModelVideoElement::requestRouteSharingPolicyAndContextUID):

• platform/ios/VideoFullscreenInterfaceAVKit.h:
• platform/ios/VideoFullscreenInterfaceAVKit.mm:

(-[WebAVPlayerViewController setWebKitOverrideRouteSharingPolicy:routingContextUID:]):
(VideoFullscreenInterfaceAVKit::setVideoFullscreenModel):
(VideoFullscreenInterfaceAVKit::doSetup):

Source/WebCore/PAL:

• pal/spi/cocoa/AVKitSPI.h:

Source/WebKit:

Add an asyncronous reply request to VideoFullscreenManager.

• Platform/IPC/MessageSender.h:

(IPC::MessageSender::sendWithAsyncReply):

• UIProcess/Cocoa/VideoFullscreenManagerProxy.h:
• UIProcess/Cocoa/VideoFullscreenManagerProxy.mm:

(WebKit::VideoFullscreenModelContext::requestRouteSharingPolicyAndContextUID):

• WebProcess/cocoa/VideoFullscreenManager.h:
• WebProcess/cocoa/VideoFullscreenManager.messages.in:
• WebProcess/cocoa/VideoFullscreenManager.mm:

(WebKit::VideoFullscreenManager::requestRouteSharingPolicyAndContextUID):

3:06 PM Changeset in webkit [238527] by achristensen@apple.com

• 5 edits in trunk

Rename WKWebView._safeBrowsingWarningForTesting to _safeBrowsingWarning to use it for more than testing
​https://bugs.webkit.org/show_bug.cgi?id=191981

Reviewed by Tim Horton.

Source/WebKit:

• UIProcess/API/Cocoa/WKWebView.mm:

(-[WKWebView _safeBrowsingWarning]):
(-[WKWebView _safeBrowsingWarningForTesting]): Deleted.

• UIProcess/API/Cocoa/WKWebViewPrivate.h:

Tools:

• TestWebKitAPI/Tests/WebKitCocoa/SafeBrowsing.mm:

(TEST):
(safeBrowsingView):

2:57 PM Changeset in webkit [238526] by dbates@webkit.org

• 10 edits
  4 adds in trunk

REGRESSION (r237738): Command Down Arrow doesn't scroll to the end of a page anymore
​https://bugs.webkit.org/show_bug.cgi?id=191967
<rdar://problem/45976390>

Reviewed by Tim Horton.

Source/WebKit:

Fixes an issue where pressing Command + Down Arrow does not scroll the view to the end of the page.

Following r237738 the value of the enumerations used to identify modifier keys (e.g. Shift) changed
to match the values of the corresponding enumerations in GraphicsServices, which are the
enumerations UIKit uses to computes the modifier flags bitmask when instantiating a WebEvent to
pass to WebKit. Before r237738 WebKit was using enumerations whose values matched the values
of the corresponding UIKit public API UIKeyModifier* enumerations. For non-content editable elements,
WebKit intercepts UIKit events in -_handleKeyUIEvent, synthesizes and dispatches its own WebEvent.
However it was creating WebEvents with a modifier flags bitmask built from the UIKeyModifier* enumerations,
-_modifierFlags, as opposed to a bitmask from the GraphicsServices enumerations, -_gsModifierFlags.
Instead WebKit should call -_gsModifierFlags to compute the GraphicsServices-compatible modifier
flags bitmask when instantiating a WebEvent.

• Platform/spi/ios/UIKitSPI.h: Expose -_gsModifierFlags and remove -_modifierFlags.
• UIProcess/ios/WKContentViewInteraction.mm:

(-[WKContentView _handleKeyUIEvent:]): Remove unnecessary code to update the current modifier state.
This will be done by the WebProcess when it receives the keyboard event.
(-[WKContentView _didHandleKeyEvent:eventWasHandled:]): Do not pass modifier flags changed events
to the scrolling animator as it does not know how to handle these kinds of events and triggers an
assertion failure when it tries to read the input string from the event (calls -charactersIgnoringModifiers).
FlagsChanged WebEvents events do not have an input string just like a FlagsChanged NSEvent that
they model on Mac.

• UIProcess/ios/WKWebEvent.mm:

(-[WKWebEvent initWithEvent:]): Pass the value of -_gsModifierFlags for the modifier flags bitmask
instead of the value of -_modifierFlags.

Tools:

Add more test infrastructure to dispatch key up events.

• WebKitTestRunner/ios/HIDEventGenerator.h:
• WebKitTestRunner/ios/HIDEventGenerator.mm:

(createHIDKeyEvent): Renamed from createHIDKeyDownEvent() and modified to take a boolean as
to whether to create an event for a key down or key up.
(createHIDKeyDownEvent): Deleted; renamed to createHIDKeyEvent().

• WebKitTestRunner/ios/UIScriptControllerIOS.mm:

(WTR::createUIPhysicalKeyboardEvent): Added. Convenience function to create a UIPhysicalKeyboardEvent.
(WTR::UIScriptController::keyDown): Modified to dispatch a key up event in addition to dispatching
a key down event.

LayoutTests:

Add tests to ensure that Command + Down Arrow and Command + Up Arrow scroll to the end of
the page and the top of the page, respectively.

• fast/scrolling/ios/key-command-scroll-to-bottom-expected.html: Added.
• fast/scrolling/ios/key-command-scroll-to-bottom.html: Added.
• fast/scrolling/ios/key-command-scroll-to-top-expected.html: Added.
• fast/scrolling/ios/key-command-scroll-to-top.html: Added.
• resources/ui-helper.js:

(window.UIHelper.keyDown): Pass the modifiers array to EventSender.keyDown() to make
this function work on Mac.

2:54 PM Changeset in webkit [238525] by Chris Dumez

• 5 edits in trunk/Source/WebKit

[PSON] process pre-warming should not be on for everyone
​https://bugs.webkit.org/show_bug.cgi?id=191966
<rdar://problem/46138499>

Reviewed by Ryosuke Niwa.

If automatic process-prewarming is not explicitly turned on/off by the client, then we now only
turn it on automatically after the first process swap. Previously, it was always enabled if
process-swap on cross-site navigation was enabled (which is the case by default for all apps).

This is important because some apps do not actually browse (only display static content) or never
browse cross-site, and thus would not benefit from process prewarming and yet pay a memory cost.

• UIProcess/API/APIProcessPoolConfiguration.h:

Add m_clientWouldBenefitFromAutomaticProcessPrewarming flag which is false by default and is used
as a fallback if the client did not call setIsAutomaticProcessWarmingEnabled().

• UIProcess/API/C/WKContextConfigurationRef.cpp:

(WKContextConfigurationPrewarmsProcessesAutomatically):
(WKContextConfigurationSetPrewarmsProcessesAutomatically):
Add C API to toggle automatic process prewarming. It is needed for Safari to turn on this feature
by default for browsing. This way Safari does not have to wait for the first process swap to
start prewarming.

• UIProcess/API/C/WKContextConfigurationRef.h:
• UIProcess/WebProcessPool.cpp:

(WebKit::WebProcessPool::didReachGoodTimeToPrewarm):
Only do automatic process prewarming if process-swap on cross-site navigation is also enabled.
This is needed for clients like Safari that explicitly enable automatic process prewarming but
only want to have it enabled if PSON is also enabled via experimental features.

(WebKit::WebProcessPool::processForNavigation):
On first process swap, if the client did not explicitly turn on or off automatic process prewarming
then we enable it then. This avoids paying the cost of process prewarming in apps that would not
benefit from it.

2:48 PM Changeset in webkit [238524] by rniwa@webkit.org

• 19 edits
  26 adds in trunk

SVG use element inside a shadow tree cannot reference an element in the same tree
​https://bugs.webkit.org/show_bug.cgi?id=174977
<rdar://problem/33665636>

Reviewed by Zalan Bujtas.

Source/WebCore:

Make fragment URL references used by SVGelements within a shadow tree to refer to other elements
in the same shadow tree. To do this, this patch makes targetElementFromIRIString take a TreeScope
instead of a Document, and updates its call sites.

This patch updates the most uses of targetElementFromIRIString except CSS cursor image, altGraph,
and glyphRef since the cursor image isn't really a SVG feature, and there aren't really real world
use cases in which altGraph and glyphRef are used within shadow trees.

Tests: fast/shadow-dom/svg-animate-href-change-in-shadow-tree.html

fast/shadow-dom/svg-animate-href-in-shadow-tree.html
fast/shadow-dom/svg-feimage-href-in-shadow-tree.html
fast/shadow-dom/svg-linear-gradient-href-in-shadow-tree.html
fast/shadow-dom/svg-mpath-href-change-in-shadow-tree.html
fast/shadow-dom/svg-mpath-href-in-shadow-tree.html
fast/shadow-dom/svg-radial-gradient-href-in-shadow-tree.html
fast/shadow-dom/svg-text-path-href-change-in-shadow-tree.html
fast/shadow-dom/svg-text-path-href-in-shadow-tree.html
fast/shadow-dom/svg-thref-href-change-in-shadow-tree.html
fast/shadow-dom/svg-thref-href-in-shadow-tree.html
fast/shadow-dom/svg-use-href-change-in-shadow-tree.html
fast/shadow-dom/svg-use-href-in-shadow-tree.html

• accessibility/AccessibilitySVGElement.cpp:

(WebCore::AccessibilitySVGElement::targetForUseElement const):

• css/CSSCursorImageValue.cpp:
• rendering/svg/RenderSVGTextPath.cpp:

(WebCore::RenderSVGTextPath::layoutPath const):

• svg/SVGAltGlyphElement.cpp:

(WebCore::SVGAltGlyphElement::hasValidGlyphElements const):

• svg/SVGFEImageElement.cpp:

(WebCore::SVGFEImageElement::buildPendingResource):
(WebCore::SVGFEImageElement::build):

• svg/SVGGlyphRefElement.cpp:

(WebCore::SVGGlyphRefElement::hasValidGlyphElement const):

• svg/SVGLinearGradientElement.cpp:

(WebCore::SVGLinearGradientElement::collectGradientAttributes):

• svg/SVGMPathElement.cpp:

(WebCore::SVGMPathElement::buildPendingResource):
(WebCore::SVGMPathElement::pathElement):

• svg/SVGRadialGradientElement.cpp:

(WebCore::SVGRadialGradientElement::collectGradientAttributes):

• svg/SVGTRefElement.cpp:

(WebCore::SVGTRefElement::buildPendingResource):

• svg/SVGTextPathElement.cpp:

(WebCore::SVGTextPathElement::buildPendingResource):

• svg/SVGURIReference.cpp:

(WebCore::SVGURIReference::targetElementFromIRIString):

• svg/SVGURIReference.h:
• svg/SVGUseElement.cpp:

(WebCore::SVGUseElement::updateShadowTree):
(WebCore::SVGUseElement::findTarget const):

• svg/animation/SVGSMILElement.cpp:

(WebCore::SVGSMILElement::buildPendingResource):
(WebCore::SVGSMILElement::insertedIntoAncestor):

• svg/graphics/filters/SVGFEImage.cpp:

(WebCore::FEImage::FEImage):
(WebCore::FEImage::createWithIRIReference):
(WebCore::FEImage::referencedRenderer const):

• svg/graphics/filters/SVGFEImage.h:

LayoutTests:

Added tests for specifying a fragment URL to refer to other elements within shadow trees.

This patch doesn't add a test for updating a fragment URL of linearGradient and radialGradient
since it doesn't work at all even outside shadow trees. See webkit.org/b/191934.

• fast/shadow-dom/svg-animate-href-change-in-shadow-tree-expected.html: Added.
• fast/shadow-dom/svg-animate-href-change-in-shadow-tree.html: Added.
• fast/shadow-dom/svg-animate-href-in-shadow-tree-expected.html: Added.
• fast/shadow-dom/svg-animate-href-in-shadow-tree.html: Added.
• fast/shadow-dom/svg-feimage-href-in-shadow-tree-expected.html: Added.
• fast/shadow-dom/svg-feimage-href-in-shadow-tree.html: Added.
• fast/shadow-dom/svg-linear-gradient-href-in-shadow-tree-expected.html: Added.
• fast/shadow-dom/svg-linear-gradient-href-in-shadow-tree.html: Added.
• fast/shadow-dom/svg-mpath-href-change-in-shadow-tree-expected.html: Added.
• fast/shadow-dom/svg-mpath-href-change-in-shadow-tree.html: Added.
• fast/shadow-dom/svg-mpath-href-in-shadow-tree-expected.html: Added.
• fast/shadow-dom/svg-mpath-href-in-shadow-tree.html: Added.
• fast/shadow-dom/svg-radial-gradient-href-in-shadow-tree-expected.html: Added.
• fast/shadow-dom/svg-radial-gradient-href-in-shadow-tree.html: Added.
• fast/shadow-dom/svg-text-path-href-change-in-shadow-tree-expected.html: Added.
• fast/shadow-dom/svg-text-path-href-change-in-shadow-tree.html: Added.
• fast/shadow-dom/svg-text-path-href-in-shadow-tree-expected.html: Added.
• fast/shadow-dom/svg-text-path-href-in-shadow-tree.html: Added.
• fast/shadow-dom/svg-thref-href-change-in-shadow-tree-expected.html: Added.
• fast/shadow-dom/svg-thref-href-change-in-shadow-tree.html: Added.
• fast/shadow-dom/svg-thref-href-in-shadow-tree-expected.html: Added.
• fast/shadow-dom/svg-thref-href-in-shadow-tree.html: Added.
• fast/shadow-dom/svg-use-href-change-in-shadow-tree-expected.html: Added.
• fast/shadow-dom/svg-use-href-change-in-shadow-tree.html: Added.
• fast/shadow-dom/svg-use-href-in-shadow-tree-expected.html: Added.
• fast/shadow-dom/svg-use-href-in-shadow-tree.html: Added.

2:22 PM Changeset in webkit [238523] by Truitt Savell

• 3 edits in trunk/Source/WebCore

Unreviewed, rolling out r238357.

Casued three css tests to fail and crash on ios sim

Reverted changeset:

"Avoid triggering compositing updates when only the root layer
is composited"
​https://bugs.webkit.org/show_bug.cgi?id=191813
​https://trac.webkit.org/changeset/238357

2:15 PM Changeset in webkit [238522] by dbates@webkit.org

• 7 edits
  2 adds in trunk

Caret disappears at end of password field when caps lock indicator is shown; password field
not scrolled when caps lock indicator is shown
​https://bugs.webkit.org/show_bug.cgi?id=191164
<rdar://problem/45738179>

Reviewed by Dean Jackson.

Source/WebCore:

Fixes an issue where the caret may be occluded by- or paint on top of- the caps lock indicator on
Mac and iOS, respectively.

If there has not been a previous selection in a focused password field, including a caret
selection made by pressing the arrow keys, then we never scroll the password field to reveal
the current selection when the caps lock indicator is made visible. When the caps lock indicator
is made visible or hidden the size of the inner text renderer changes as it shrinks or expands
to make space for the caps lock indicator or to fill the void of the now hidden caps lock indicator,
respectively. We should detect such size changes and schedule an update and reveal of the current
selection after layout.

Test: fast/forms/password-scrolled-after-caps-lock-toggled.html

• editing/FrameSelection.cpp:

(WebCore::FrameSelection::setNeedsSelectionUpdate): Modified to take an enum to override the current
selection reveal mode for the next update.

• editing/FrameSelection.h:
• rendering/RenderTextControlSingleLine.cpp:

(WebCore::RenderTextControlSingleLine::layout): Schedule post-layout a selection update that
reveals the current selection. We pass FrameSelection::RevealSelectionAfterUpdate::Forced to ensure
that the scheduled selection update scrolls to the reveal the current selection regardless of selection
reveal mode. This is necessary because typing into a password field does not change the current
selection reveal mode.

LayoutTests:

Add a test to ensure that we scroll the password field when caps lock is toggled.

• TestExpectations: Skip the test on all platforms as we only support toggling Caps Lock in

WebKit2 on Mac at the moment.

• fast/forms/password-scrolled-after-caps-lock-toggled-expected.txt: Added.
• fast/forms/password-scrolled-after-caps-lock-toggled.html: Added.
• platform/mac-wk2/TestExpectations: Mark the test as PASS so that we run it.

1:59 PM Changeset in webkit [238521] by Kocsen Chung

• 16 edits in tags/Safari-607.1.14.2

Cherry-pick r238501. rdar://problem/46140542

[Win] Reduce the use of WKSI library calls: CoreAnimation
​https://bugs.webkit.org/show_bug.cgi?id=191777
<rdar://problem/46140542>

Reviewed by Zalan Bujtas.

Update the Windows build of WebKit to refer to the SPI headers, rather than WebKitSystemInterface.
Move a small amount of glue code from WKSI to WebCore, and remove any includes or link
directives for WebKitSystemInterface.lib.
Source/WebCore:

No new tests. No change in behavior.

• platform/graphics/BitmapImage.cpp:
• platform/graphics/ca/win/CACFLayerTreeHost.h:
• platform/graphics/ca/win/PlatformCALayerWin.cpp: (layerTreeHostForLayer):
• platform/graphics/cg/GraphicsContextCG.cpp:
• platform/graphics/cg/PathCG.cpp:
• platform/graphics/cg/PatternCG.cpp:
• platform/graphics/win/WKCAImageQueue.cpp:

Source/WebKitLegacy:

• PlatformWin.cmake:

Source/WebKitLegacy/win:

• FullscreenVideoController.cpp:
• WebKitGraphics.cpp:
• WebPreferences.cpp:

Tools:

• TestWebKitAPI/PlatformWin.cmake:

git-svn-id: ​https://svn.webkit.org/repository/webkit/trunk@238501 268f45cc-cd09-0410-ab3c-d52691b4dbfc

1:55 PM Changeset in webkit [238520] by dbates@webkit.org

• 2 edits in trunk/LayoutTests

Placeholder text is not repainted after caps lock indicator is hidden
​https://bugs.webkit.org/show_bug.cgi?id=191968
<rdar://problem/46247234>

Substitute UIHelper.toggleCapsLock() for testRunner.toggleCapsLock() as the former replaces
the latter following r238512.

• fast/repaint/placeholder-after-caps-lock-hidden.html:

1:50 PM Changeset in webkit [238519] by dbates@webkit.org

• 5 edits
  2 adds in trunk

Placeholder text is not repainted after caps lock indicator is hidden
​https://bugs.webkit.org/show_bug.cgi?id=191968
<rdar://problem/46247234>

Reviewed by Zalan Bujtas.

Source/WebCore:

Fixes an issue where the placeholder text in a password field is not repainted when the
caps lock indicator is hidden.

The placeholder renderer is special. It is an excluded child renderer and does not take
part in normal flow layout. It is also created and destroyed as needed. The caps lock
indicator is also special in that it is implemented as a RenderImage and we do not know
its dimensions before it is loaded and the load happens asynchronously. As a result we
detect when the inner text size changes and mark the placeholder as dirty as a way to
keep the dimensions of the placeholder in sync with the dimensions of the inner text.

Test: fast/repaint/placeholder-after-caps-lock-hidden.html

• rendering/RenderTextControlSingleLine.cpp:

(WebCore::RenderTextControlSingleLine::layout): Mark the placeholder as needing layout
the size of the inner text changes.

LayoutTests:

Add a test to ensure to that the placeholder text is repainted when the caps lock indicator is hidden.

• TestExpectations: Skip the test on all platforms as we only support toggling Caps Lock in

WebKit2 on Mac at the moment.

• fast/repaint/placeholder-after-caps-lock-hidden.html: Added.
• platform/mac-wk2/TestExpectations: Mark the test as PASS so that we run it.

1:44 PM Changeset in webkit [238518] by aakash_jain@apple.com

• 3 edits in trunk/Tools

[ews-app] Set Foreign Keys in EWS Django app models
​https://bugs.webkit.org/show_bug.cgi?id=191790

Reviewed by Lucas Forschler.

• BuildSlaveSupport/ews-app/ews/models/builds.py: Set Foreign Key.
• BuildSlaveSupport/ews-app/ews/models/steps.py: Ditto.

1:39 PM Changeset in webkit [238517] by aakash_jain@apple.com

• 2 edits in trunk/Tools

[ews-app] Do not store patch content in database
​https://bugs.webkit.org/show_bug.cgi?id=191788

Reviewed by Lucas Forschler.

• BuildSlaveSupport/ews-app/ews/models/patch.py: Removed content field.

1:19 PM Changeset in webkit [238516] by commit-queue@webkit.org

• 2 edits in trunk/Source/WebCore

Picture-in-picture window size changes unnecesarily when URL changes.
​https://bugs.webkit.org/show_bug.cgi?id=191787

Patch by Jeremy Jones <​jeremyj@apple.com> on 2018-11-26
Reviewed by Jer Noble.

When loading a new URL, the video dimensions are temporarily 0,0. Instead of
defaulting back to 4:3 sized pip window temporarily, keep the old dimensions until
there is a new valid size.

• platform/ios/VideoFullscreenInterfaceAVKit.mm:

(VideoFullscreenInterfaceAVKit::videoDimensionsChanged):

1:13 PM Changeset in webkit [238515] by aestes@apple.com

• 20 edits
  1 copy
  1 add in trunk/Source

[Cocoa] Make it easier to encode NSObjects
​https://bugs.webkit.org/show_bug.cgi?id=191948

Reviewed by Dean Jackson.

Source/WebCore:

• Modules/applepay/Payment.h: Changed the PKPayment * constructor to take a

RetainPtr<PKPayment>&& instead.

• Modules/applepay/PaymentContact.h: Ditto for PKContact.
• Modules/applepay/PaymentMerchantSession.h: Ditto for PKPaymentMerchantSession.
• Modules/applepay/PaymentMethod.h: Ditto for PKPaymentMethod.
• Modules/applepay/cocoa/PaymentCocoa.mm:

(WebCore::Payment::Payment): Moved definition out-of-line.
(WebCore::Payment::pkPayment const): Ditto.

• Modules/applepay/cocoa/PaymentContactCocoa.mm:

(WebCore::PaymentContact::PaymentContact): Ditto.
(WebCore::PaymentContact::pkContact const): Ditto.

• Modules/applepay/cocoa/PaymentMethodCocoa.mm:

(WebCore::PaymentMethod::PaymentMethod): Ditto.
(WebCore::PaymentMethod::pkPaymentMethod const): Ditto.

Source/WebCore/PAL:

• pal/spi/cocoa/PassKitSPI.h: Declared some classes that secretly conform to NSSecureCoding.

Source/WebKit:

It should be easier to encode objects that conform to NSSecureCoding for WebKit IPC. Right
now, several argument coders duplicate the logic for encoding and decoding objects using
NSKeyedArchiver and NSKeyedUnarchiver.

This patch adds encodeObject() and decodeObject() primitives for encoding and decoding using
Foundation keyed archiving. It then partially specializes ArgumentCoder for raw pointers and
RetainPtrs whose pointee conforms to NSSecureCoding to call these new primitives.

This allows us to use Encoder::operator<< to encode raw pointers or RetainPtrs and use
Decoder::operator>> to decode to an optional RetainPtr. By default, IPC::decode infers the
allowed class for decoding by calling [T class]. We can also specify the allowed classes
explicitly for containers and classes that are loaded at runtime.

• Shared/Cocoa/ArgumentCodersCocoa.h: Added.
• Shared/Cocoa/ArgumentCodersCocoa.mm: Added.
• Shared/Cocoa/DataDetectionResult.mm:

(WebKit::DataDetectionResult::encode const): Removed custom encoding code and used ArgumentCodersCocoa instead.
(WebKit::DataDetectionResult::decode): Ditto.

• Shared/Cocoa/WebCoreArgumentCodersCocoa.mm:

(IPC::ArgumentCoder<WebCore::Payment>::encode): Ditto.
(IPC::ArgumentCoder<WebCore::Payment>::decode): Ditto.
(IPC::ArgumentCoder<WebCore::PaymentAuthorizationResult>::decode): Ditto.
(IPC::ArgumentCoder<WebCore::PaymentContact>::encode): Ditto.
(IPC::ArgumentCoder<WebCore::PaymentContact>::decode): Ditto.
(IPC::ArgumentCoder<WebCore::PaymentMerchantSession>::encode): Ditto.
(IPC::ArgumentCoder<WebCore::PaymentMerchantSession>::decode): Ditto.
(IPC::ArgumentCoder<WebCore::PaymentMethod>::encode): Ditto.
(IPC::ArgumentCoder<WebCore::PaymentMethod>::decode): Ditto.
(IPC::ArgumentCoder<ApplePaySessionPaymentRequest>::decode): Ditto.

• Shared/RemoteLayerTree/RemoteLayerTreeTransaction.mm:

(WebKit::RemoteLayerTreeTransaction::description const): Added some needed namespaces due to unified source shuffling.

• Shared/WebCoreArgumentCoders.h: Modernized several decoders.
• Shared/ios/InteractionInformationAtPosition.mm:

(WebKit::InteractionInformationAtPosition::encode const): Removed custom encoding code and used ArgumentCodersCocoa instead.
(WebKit::InteractionInformationAtPosition::decode): Ditto.

• Shared/mac/WebCoreArgumentCodersMac.mm:

(IPC::ArgumentCoder<WebCore::ProtectionSpace>::encodePlatformData): Ditto.
(IPC::ArgumentCoder<WebCore::ProtectionSpace>::decodePlatformData): Ditto.
(IPC::ArgumentCoder<WebCore::Credential>::encodePlatformData): Ditto.
(IPC::ArgumentCoder<WebCore::Credential>::decodePlatformData): Ditto.
(IPC::ArgumentCoder<WebCore::MediaPlaybackTargetContext>::encodePlatformData): Ditto.
(IPC::ArgumentCoder<WebCore::MediaPlaybackTargetContext>::decodePlatformData): Ditto.
(IPC::deviceContextKey): Deleted.

• Shared/mac/WebHitTestResultData.mm:

(WebKit::WebHitTestResultData::platformEncode const): Ditto.
(WebKit::WebHitTestResultData::platformDecode): Ditto.

• SourcesCocoa.txt: Added ArgumentCodersCocoa.mm.
• WebKit.xcodeproj/project.pbxproj: Ditto.

12:55 PM Changeset in webkit [238514] by Jonathan Bedard

• 2 edits in trunk/Tools

webkitpy: Retry app installation on watchOS simulators (Follow-up fix)
​https://bugs.webkit.org/show_bug.cgi?id=191769
<rdar://problem/46139850>

Unreviewed infrastructure fix.

• Scripts/webkitpy/xcode/simulated_device.py:

(SimulatedDevice): Increase retry attempts.

12:42 PM Changeset in webkit [238513] by dbates@webkit.org

• 5 edits
  4 adds in trunk

REGRESSION (r238078): Do not draw caps lock indicator when Strong Password button is shown
​https://bugs.webkit.org/show_bug.cgi?id=191969
<rdar://problem/46247569>

Reviewed by Dean Jackson.

Source/WebCore:

Following r238078 we now support drawing the caps lock indicator in password fields on iOS.
However it is not meaningful to show the caps lock indicator when the Strong Password button
is visible because the password field is not editable. We should not paint the caps lock
indicator when the Strong Password button is visible.

Tests: fast/forms/auto-fill-button/caps-lock-indicator-should-be-visible-when-after-hiding-auto-fill-strong-password-button.html

fast/forms/auto-fill-button/caps-lock-indicator-should-not-be-visible-when-auto-fill-strong-password-button-is-visible.html

• html/TextFieldInputType.cpp:

(WebCore::TextFieldInputType::shouldDrawCapsLockIndicator const): Do not draw the caps
lock indicator when the password field has the Strong Password button.
(WebCore::TextFieldInputType::updateAutoFillButton): Call capsLockStateMayHaveChanged() to
update the visibility of the caps lock indicator when the auto fill button has changed.

LayoutTests:

Add tests to ensure that we show or hide the caps lock indicator depending on whether the
Strong Password button is visible.

• TestExpectations: Skip the tests below on all platforms. We will selectively enable them on Mac.
• fast/forms/auto-fill-button/caps-lock-indicator-should-be-visible-when-after-hiding-auto-fill-strong-password-button-expected.html: Added.
• fast/forms/auto-fill-button/caps-lock-indicator-should-be-visible-when-after-hiding-auto-fill-strong-password-button.html: Added.
• fast/forms/auto-fill-button/caps-lock-indicator-should-not-be-visible-when-auto-fill-strong-password-button-is-visible-expected.html: Added.
• fast/forms/auto-fill-button/caps-lock-indicator-should-not-be-visible-when-auto-fill-strong-password-button-is-visible.html: Added.
• platform/mac-wk2/TestExpectations: Mark the tests above as PASS so that we run them.

12:41 PM Changeset in webkit [238512] by dbates@webkit.org

• 17 edits in trunk

Move testRunner.toggleCapsLock() to uiController
​https://bugs.webkit.org/show_bug.cgi?id=191972

Reviewed by Tim Horton.

Tools:

Move testRunner.toggleCapsLock() to uiController as uiController is the preferred JavaScript
object for UI test functions. Having this functionality be on uiController makes it an
asynchronous function naturally and complements use of onkeydown, onkeyup listeners to know
when the Caps Lock key event is dispatched. It also facilitates its use directly as part of
scripts with other uiController functions invocations that are passed in a single invocation
of testRunner.runUIScript().

• DumpRenderTree/ios/UIScriptControllerIOS.mm:

(WTR::UIScriptController::toggleCapsLock):

• DumpRenderTree/mac/UIScriptControllerMac.mm:

(WTR::UIScriptController::toggleCapsLock):
Added stub functions that invoke the callback. We do not support toggling caps lock in Legacy
WebKit at the moment. Legacy WebKit reads the caps lock key state directly from the OS. Modern
WebKit caches the caps lock state in the WebProcess as a natural side effect of the fact that
the UIProcess is the only process capable of querying the caps lock key state from the OS and
hence the UIProcess must send over this state to the WebProcess.

• TestRunnerShared/UIScriptContext/Bindings/UIScriptController.idl: Add IDL for new function.
• TestRunnerShared/UIScriptContext/UIScriptController.cpp:

(WTR::UIScriptController::toggleCapsLock): Added empty implementation for ports non-Cocoa ports.

• TestRunnerShared/UIScriptContext/UIScriptController.h:

• WebKitTestRunner/InjectedBundle/Bindings/TestRunner.idl:
• WebKitTestRunner/InjectedBundle/TestRunner.cpp:

(WTR::TestRunner::toggleCapsLock): Deleted.

• WebKitTestRunner/InjectedBundle/TestRunner.h:
• WebKitTestRunner/TestController.h:
• WebKitTestRunner/TestInvocation.cpp:

(WTR::TestInvocation::didReceiveSynchronousMessageFromInjectedBundle):

• WebKitTestRunner/cocoa/TestControllerCocoa.mm:

(WTR::TestController::toggleCapsLock): Deleted; moved to UIScriptController::toggleCapsLock().
Removed logic to handle testRunner.toggleCapsLock().

• WebKitTestRunner/ios/UIScriptControllerIOS.mm:

(WTR::UIScriptController::toggleCapsLock): Add stub function that invokes the callback
and a FIXME comment that explains that we will implement this function in <​https://bugs.webkit.org/show_bug.cgi?id=191815>.

• WebKitTestRunner/mac/UIScriptControllerMac.mm:

(WTR::UIScriptController::toggleCapsLock): Moved the implementation from TestController::toggleCapsLock().

LayoutTests:

Updated existing test to use UIHelper.toggleCapsLock(), which calls uiController.toggleCapsLock(),
now that testRunner.toggleCapsLock() was removed.

• fast/events/detect-caps-lock.html:
• resources/ui-helper.js:

(window.UIHelper.toggleCapsLock): Added. Convenience function to call uiController.toggleCapsLock()
and return a Promise that is resolved once the UIProcess has dispatched the NSEvent to simulate
pressing the caps lock key.

12:29 PM Changeset in webkit [238511] by sbarati@apple.com

• 3 edits
  1 add in trunk

InPlaceAbstractState::endBasicBlock rule for SetLocal should filter the value based on the flush format
​https://bugs.webkit.org/show_bug.cgi?id=191956
<rdar://problem/45665806>

Reviewed by Yusuke Suzuki.

JSTests:

• stress/end-basic-block-set-local-should-filter-type.js: Added.

(bar):
(foo):

Source/JavaScriptCore:

This is a similar bug to what Keith fixed in r232134. The issue is if we have
a program like this:

a: JSConstant(jsNumber(0))
b: SetLocal(Int32:@a, loc1, FlushedInt32)
c: ArrayifyToStructure(Cell:@a)
d: Jump(...)

At the point in the program right after the Jump, a GetLocal for loc1
would return whatever the ArrayifyToStructure resulting type is. This breaks
the invariant that a GetLocal must return a value that is a subtype of its
FlushFormat. InPlaceAbstractState::endBasicBlock will know if a SetLocal is
the final node touching a local slot. If so, it'll see if any nodes later
in the block may have refined the type of the value stored in that slot. If
so, endBasicBlock() further refines the type to ensure that any GetLocals
loading from the same slot will result in having this more refined type.
However, we must ensure that this logic only considers types within the
hierarchy of the variable access data's FlushFormat, otherwise, we may
break the invariant that a GetLocal's type is a subtype of its FlushFormat.

• dfg/DFGInPlaceAbstractState.cpp:

(JSC::DFG::InPlaceAbstractState::endBasicBlock):

12:14 PM Changeset in webkit [238510] by sbarati@apple.com

• 4 edits
  1 add in trunk

Object allocation sinking phase needs to iterate each scope offset instead of just iterating the symbol table's hashmap when handling an activation
​https://bugs.webkit.org/show_bug.cgi?id=191958
<rdar://problem/46221877>

Reviewed by Yusuke Suzuki.

JSTests:

• stress/object-allocation-sinking-phase-needs-to-write-to-each-scope-offset.js: Added.

(x):
(foo):

Source/JavaScriptCore:

There may be more entries in an activation than unique variables
in a symbol table's hashmap. For example, if you have two parameters
to a function, and they both are the same name, and the function
uses eval, we'll end up with two scope slots, but only a single
entry in the hashmap in the symbol table. Object allocation sinking
phase was previously iterating over the hashmap, assuming these
values were equivalent. This is wrong in the above case. Instead,
we need to iterate over each scope offset.

• dfg/DFGObjectAllocationSinkingPhase.cpp:
• runtime/GenericOffset.h:

(JSC::GenericOffset::operator+=):
(JSC::GenericOffset::operator-=):

12:06 PM Changeset in webkit [238509] by mark.lam@apple.com

• 3 edits
  1 add in trunk

NaNs read from Wasm code needs to be be purified.
​https://bugs.webkit.org/show_bug.cgi?id=191056
<rdar://problem/45660341>

Reviewed by Filip Pizlo.

JSTests:

• wasm/regress/regress-191056.js: Added.

Source/JavaScriptCore:

• wasm/js/WebAssemblyModuleRecord.cpp:

(JSC::WebAssemblyModuleRecord::link):

11:59 AM Changeset in webkit [238508] by msaboff@apple.com

• 2 edits in trunk/JSTests

32-bit JSC test failure: stress/regexp-compile-oom.js
​https://bugs.webkit.org/show_bug.cgi?id=191375

Reviewed by Mark Lam.

Disabled the test for 32 bit platforms.

• stress/regexp-compile-oom.js:

11:55 AM Changeset in webkit [238507] by Wenson Hsieh

• 6 edits in trunk/Source/WebKit

CompletionHandler-based async IPC messages only work when the completion handler takes a single argument
​https://bugs.webkit.org/show_bug.cgi?id=191965

Reviewed by Tim Horton.

Teach messages.py to handle the case where an async IPC completion handler takes no arguments, or takes more
than a single argument. Currently, the generated code attempts to wrap all arguments in a WTFMove(*~), but
this either results in WTFMove(*) in the case where there are no arguments, or WTFMove(*foo, *bar, *baz) in
the case where there are several arguments. Both of these results fail to compile.

Instead, emit completionHandler() when there are no arguments, and
completionHandler(WTFMove(*foo), WTFMove(*bar), WTFMove(*baz)) when there are multiple arguments.

Tests: TestAsyncMessageWithNoArguments

TestAsyncMessageWithMultipleArguments

• Scripts/webkit/MessageReceiverSuperclass-expected.cpp:

(Messages::WebPage::TestAsyncMessageWithNoArguments::callReply):
(Messages::WebPage::TestAsyncMessageWithNoArguments::cancelReply):
(Messages::WebPage::TestAsyncMessageWithNoArguments::send):
(Messages::WebPage::TestAsyncMessageWithMultipleArguments::callReply):
(Messages::WebPage::TestAsyncMessageWithMultipleArguments::cancelReply):
(Messages::WebPage::TestAsyncMessageWithMultipleArguments::send):
(WebKit::WebPage::didReceiveMessage):

• Scripts/webkit/MessagesSuperclass-expected.h:

(Messages::WebPage::TestAsyncMessageWithNoArguments::receiverName):
(Messages::WebPage::TestAsyncMessageWithNoArguments::name):
(Messages::WebPage::TestAsyncMessageWithNoArguments::asyncMessageReplyName):
(Messages::WebPage::TestAsyncMessageWithNoArguments::arguments const):
(Messages::WebPage::TestAsyncMessageWithMultipleArguments::receiverName):
(Messages::WebPage::TestAsyncMessageWithMultipleArguments::name):
(Messages::WebPage::TestAsyncMessageWithMultipleArguments::asyncMessageReplyName):
(Messages::WebPage::TestAsyncMessageWithMultipleArguments::arguments const):

• Scripts/webkit/messages.py:
• Scripts/webkit/messages_unittest.py:

Add new messages.py unit tests to cover these cases.

• Scripts/webkit/test-superclass-messages.in:

11:51 AM Changeset in webkit [238506] by Kocsen Chung

• 7 edits in tags/Safari-607.1.14.2/Source

Versioning.

11:37 AM Changeset in webkit [238505] by Kocsen Chung

• 1 copy in tags/Safari-607.1.14.2

New tag.

11:29 AM Changeset in webkit [238504] by commit-queue@webkit.org

• 7 edits in trunk/Source

Streamline ListHashSet use in floating object code
​https://bugs.webkit.org/show_bug.cgi?id=191957

Patch by Sam Weinig <​sam@webkit.org> on 2018-11-26
Reviewed by Alex Christensen.

Source/WebCore:

Simplify use of ListHashSet by using new raw pointer overloads and
making use of reversed order of template arguments in the find() and
contains() overloads that take hash translators.

• rendering/FloatingObjects.cpp:

(WebCore::FloatingObjects::remove):
Use raw pointer overloads of contains and remove. Remove seperate call
to find / check agains end() which is unnecessary as remove() already
does that.

• rendering/FloatingObjects.h:

(WebCore::FloatingObjectHashFunctions::hash):
(WebCore::FloatingObjectHashFunctions::equal):
(WebCore::FloatingObjectHashTranslator::hash):
(WebCore::FloatingObjectHashTranslator::equal):
Add hash()/equal() overloads for the raw pointer cases. As the FIXME
notes, this could be simplified by changing PtrHashBase to use designated
bottleneck functions for hash() and equal().

• rendering/RenderBlockFlow.cpp:

(WebCore::RenderBlockFlow::containsFloat const):
(WebCore::RenderBlockFlow::insertFloatingObject):
(WebCore::RenderBlockFlow::removeFloatingObject):
(WebCore::RenderBlockFlow::hasOverhangingFloat):
(WebCore::RenderBlockFlow::addIntrudingFloats):
Use simplified calls.

• rendering/RenderBlockLineLayout.cpp:

(WebCore::RenderBlockFlow::layoutRunsAndFloatsInRange):
(WebCore::RenderBlockFlow::linkToEndLineIfNeeded):
Use simplified calls.

Source/WTF:

• wtf/ListHashSet.h:

Reverses the order of the template arguments for the find() and contains()
overload that allow specifying a hash translator to allow the compiler to
deduce type T. This simplifies call sites and matches other WTF containers.

11:15 AM Changeset in webkit [238503] by commit-queue@webkit.org

• 5 edits in trunk/Source

Use Full Screen consistently in localizable strings.
​https://bugs.webkit.org/show_bug.cgi?id=190363
rdar://problem/43882333

Patch by Jeremy Jones <​jeremyj@apple.com> on 2018-11-26
Reviewed by Jon Lee.

Source/WebCore:

No new tests because only change is to localizable strings.

Rename "Fullscreen" to "Full Screen" in localizable strings for consistency.

• English.lproj/Localizable.strings:
• platform/LocalizedStrings.cpp:

(WebCore::contextMenuItemTagEnterVideoFullscreen):
(WebCore::contextMenuItemTagExitVideoFullscreen):
(WebCore::localizedMediaControlElementHelpText):

Source/WebKit:

Rename "Fullscreen" to "Full Screen" in localizable strings for consistency.

• UIProcess/ios/fullscreen/WKFullScreenViewController.mm:

(-[WKFullScreenViewController _showPhishingAlert]):

11:01 AM Changeset in webkit [238502] by Devin Rousso

• 10 edits in trunk/Source/WebInspectorUI

Web Inspector: "No Filter Results" in navigation sidebar should have a button to clear filters
​https://bugs.webkit.org/show_bug.cgi?id=191916

Reviewed by Joseph Pecoraro.

• UserInterface/Views/NavigationSidebarPanel.js:

(WI.NavigationSidebarPanel.prototype.resetFilter): Added.
(WI.NavigationSidebarPanel.prototype._checkForEmptyFilterResults.checkTreeOutlineForEmptyFilterResults):

• UserInterface/Views/FilterBar.js:

(WI.FilterBar.prototype.clear):

• UserInterface/Views/FilterBarButton.js:

(WI.FilterBarButton.prototype.toggle):

• UserInterface/Views/ResourceSidebarPanel.js:

(WI.ResourceSidebarPanel.prototype.resetFilter): Added.

• UserInterface/Views/StorageSidebarPanel.js:

(WI.StorageSidebarPanel.prototype.resetFilter): Added.

• UserInterface/Views/AuditTestContentView.js:

(WI.AuditTestContentView.prototype.showFilteredPlaceholder):

• UserInterface/Views/NetworkTableContentView.js:

(WI.NetworkTableContentView.prototype._showEmptyFilterResultsMessage):

• UserInterface/Views/Main.css:

(.message-text-view > .message + button): Added.

• Localizations/en.lproj/localizedStrings.js:

10:26 AM Changeset in webkit [238501] by Brent Fulgham

• 16 edits in trunk

[Win] Reduce the use of WKSI library calls: CoreAnimation
​https://bugs.webkit.org/show_bug.cgi?id=191777
<rdar://problem/46140542>

Reviewed by Zalan Bujtas.

Update the Windows build of WebKit to refer to the SPI headers, rather than WebKitSystemInterface.
Move a small amount of glue code from WKSI to WebCore, and remove any includes or link
directives for WebKitSystemInterface.lib.
Source/WebCore:

No new tests. No change in behavior.

• platform/graphics/BitmapImage.cpp:
• platform/graphics/ca/win/CACFLayerTreeHost.h:
• platform/graphics/ca/win/PlatformCALayerWin.cpp:

(layerTreeHostForLayer):

• platform/graphics/cg/GraphicsContextCG.cpp:
• platform/graphics/cg/PathCG.cpp:
• platform/graphics/cg/PatternCG.cpp:
• platform/graphics/win/WKCAImageQueue.cpp:

Source/WebKitLegacy:

• PlatformWin.cmake:

Source/WebKitLegacy/win:

• FullscreenVideoController.cpp:
• WebKitGraphics.cpp:
• WebPreferences.cpp:

Tools:

• TestWebKitAPI/PlatformWin.cmake:

10:20 AM Changeset in webkit [238500] by Devin Rousso

• 3 edits in trunk/Source/WebInspectorUI

Web Inspector: Audit: automatically add to tab bar when the experimental setting is enabled
​https://bugs.webkit.org/show_bug.cgi?id=191659

Reviewed by Joseph Pecoraro.

• UserInterface/Views/SettingsTabContentView.js:

(WI.SettingsTabContentView.prototype._createExperimentalSettingsView):

• UserInterface/Base/Main.js:

(WI.loaded):
Drive-by: use existing constants instead of hard-coded strings for consistency.

10:15 AM Changeset in webkit [238499] by Tadeu Zagallo

• 3 edits
  1 add in trunk

ASSERTION FAILED: m_outOfLineJumpTargets.contains(bytecodeOffset)
​https://bugs.webkit.org/show_bug.cgi?id=191716
<rdar://problem/45723878>

Reviewed by Saam Barati.

JSTests:

• stress/regress-187373.js: Added.

(async.fn):

Source/JavaScriptCore:

After ​https://bugs.webkit.org/show_bug.cgi?id=187373, when updating
jump targets during generatorification, we only stored the new jump
target when it changed. However, the out-of-line jump targets are
cleared at the beginning of the pass, so we need to store it
unconditionally.

• bytecode/PreciseJumpTargetsInlines.h:

(JSC::extractStoredJumpTargetsForInstruction):
(JSC::updateStoredJumpTargetsForInstruction):

9:42 AM Changeset in webkit [238498] by dbates@webkit.org

• 4 edits in trunk/Source/WebKit

Wire up ChromeClient::takeFocus() on iOS
​https://bugs.webkit.org/show_bug.cgi?id=191763
<rdar://problem/18584508>

Reviewed by Dan Bernstein.

Expose existing Mac SPI for use on iOS. This will allow Safari to be notified when
the engine has cycled through all the tab focusable elements on the page.

• UIProcess/API/Cocoa/WKUIDelegatePrivate.h:
• UIProcess/Cocoa/UIDelegate.h:
• UIProcess/Cocoa/UIDelegate.mm:

(WebKit::UIDelegate::setDelegate):

9:11 AM Changeset in webkit [238497] by Wenson Hsieh

• 3 edits in trunk/Source/WebKit

Unreviewed, fix the internal 32-bit macOS 10.13 build after r238471

r238471 added an #include WebPageMessages.h in WebPage.h, which causes the 32-bit macOS build using an
internal macOS SDK ≤ 10.13 to fail. To address this, move the #include back under PLATFORM(IOS_FAMILY) by
changing the parameters of WebPage::removeDataDetectedLinks and WebPage::detectDataInAllFrames to completion
handlers rather than async IPC replies.

• WebProcess/WebPage/WebPage.cpp:

(WebKit::WebPage::removeDataDetectedLinks):
(WebKit::WebPage::detectDataInAllFrames):

• WebProcess/WebPage/WebPage.h:

8:49 AM Changeset in webkit [238496] by Alan Bujtas

• 1 edit
  1 add in trunk/LayoutTests

Unreviewed rebaseline after r238493.

• platform/ios/fast/inline/inline-content-with-float-and-margin-expected.txt: Added.

8:24 AM Changeset in webkit [238495] by Alan Bujtas

• 2 edits in trunk/LayoutTests

Unreviewed test gardening after r238280.

• fast/content-observation/click-event-suppression-on-content-change.html:

7:30 AM Changeset in webkit [238494] by Alan Bujtas

• 1 edit
  1 add in trunk/LayoutTests

Unreviewed rebaseline after r238460.

• platform/ios/fast/inline/inline-content-with-image-simple-expected.txt: Added.

7:23 AM Changeset in webkit [238493] by Alan Bujtas

• 5 edits
  2 adds in trunk

[LFC][IFC] Use float box's margin box to adjust the line constraints.
​https://bugs.webkit.org/show_bug.cgi?id=191961

Reviewed by Antti Koivisto.

Source/WebCore:

Test: fast/inline/inline-content-with-float-and-margin.html

(WebCore::Layout::outputMismatchingComplexLineInformationIfNeeded):
(WebCore::Layout::LayoutState::verifyAndOutputMismatchingLayoutTree const):

• layout/inlineformatting/InlineFormattingContext.cpp:

(WebCore::Layout::InlineFormattingContext::layoutInlineContent const):

Tools:

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

LayoutTests:

• fast/inline/inline-content-with-float-and-margin-expected.txt: Added.
• fast/inline/inline-content-with-float-and-margin.html: Added.

6:18 AM Changeset in webkit [238492] by Carlos Garcia Campos

• 2 edits in trunk/Source/WebCore

[GTK] Accessing default web context before gtk_init results in drawing failure (Gtk-WARNING : drawing failure for widget 'WebKitWebView': invalid value for an input Visual*)
​https://bugs.webkit.org/show_bug.cgi?id=150303

Reviewed by Michael Catanzaro.

Ensure gtk has been initialized before trying to get the default display.

• platform/graphics/PlatformDisplay.cpp:

(WebCore::PlatformDisplay::createPlatformDisplay):

4:09 AM Changeset in webkit [238491] by jfernandez@igalia.com

• 7 edits in trunk

[css-grid] absolute positioned child is sized wrongly when using auto-fit, generating spurious collapsed tracks
​https://bugs.webkit.org/show_bug.cgi?id=191938

Reviewed by Manuel Rego Casasnovas.

LayoutTests/imported/w3c:

This change makes several cases of the following tests to pass now.

• web-platform-tests/css/css-grid/abspos/grid-positioned-items-and-autofit-tracks-004-expected.txt:
• web-platform-tests/css/css-grid/abspos/grid-positioned-items-and-autofit-tracks-005-expected.txt:
• web-platform-tests/css/css-grid/abspos/grid-positioned-items-and-autofit-tracks-006-expected.txt:
• web-platform-tests/css/css-grid/abspos/grid-positioned-items-and-autofit-tracks-007-expected.txt:

Source/WebCore:

The guttersSize function has a complex logic to compute the gaps in a
specific GridSpan, considering different scenarios of collapsed tracks
for such span.

The first case is avoiding the duplicated gap because of trailing
collapsed tracks.

The second case considered is looking for non-empty tracks before the
GridSpan end, if it points to an empty track, so we must add this gap.

The last case is to consider the gap of non-empty tracks after the
GridSpan end line, if it points to an empty track.

There are several cases that are not considered or incorrectly computed.
This patch addresses those cases; basically, we should only consider gaps
when there are non-empty tracks before and after the collapsed tracks.
Additionally, we should avoid duplicating the gaps size adding both,
before and after non-empty track's gap.

No new tests, this change is covered by current tests and make several cases to pass now.

• rendering/RenderGrid.cpp:

(WebCore::RenderGrid::guttersSize const):

1:40 AM Changeset in webkit [238490] by Antti Koivisto

• 6 edits in trunk/Source/WebKit

Clean up layer tree freezing logic in WebPage
​https://bugs.webkit.org/show_bug.cgi?id=191826

Reviewed by Dean Jackson.

Use OptionSet<LayerTreeFreezeReason> to track various reasons that can cause layer tree to get frozen.

• WebProcess/WebPage/WebPage.cpp:

(WebKit::WebPage::reinitializeWebPage):
(WebKit::WebPage::freezeLayerTree):
(WebKit::WebPage::unfreezeLayerTree):
(WebKit::WebPage::updateDrawingAreaLayerTreeFreezeState):

Layer tree is unfrozen when there there are no reasons to freeze it.

(WebKit::WebPage::didStartPageTransition):
(WebKit::WebPage::didCompletePageTransition):
(WebKit::WebPage::beginPrinting):
(WebKit::WebPage::endPrinting):
(WebKit::WebPage::setIsSuspended):
(WebKit::WebPage::setLayerTreeStateIsFrozen): Deleted.

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

(WebKit::WebPage::applicationDidEnterBackground):
(WebKit::WebPage::applicationWillEnterForeground):

• WebProcess/WebProcess.cpp:

(WebKit::WebProcess::actualPrepareToSuspend):
(WebKit::WebProcess::cancelPrepareToSuspend):
(WebKit::WebProcess::freezeAllLayerTrees):
(WebKit::WebProcess::unfreezeAllLayerTrees):
(WebKit::WebProcess::processDidResume):
(WebKit::WebProcess::setAllLayerTreeStatesFrozen): Deleted.

• WebProcess/WebProcess.h:

12:54 AM Changeset in webkit [238489] by berto@igalia.com

• 2 edits in trunk

[GTK] [2.22.0] Fails to build in armel
​https://bugs.webkit.org/show_bug.cgi?id=188862

Disable the JIT and enable C_LOOP on ARM CPUs without Thumb2
support.

Reviewed by Michael Catanzaro.

• Source/cmake/WebKitFeatures.cmake:

12:06 AM Changeset in webkit [238488] by Manuel Rego Casasnovas

• 9 edits
  2 adds in trunk

[css-grid] Fix grid container sizing under min-content height
​https://bugs.webkit.org/show_bug.cgi?id=191889

Reviewed by Javier Fernandez.

LayoutTests/imported/w3c:

Imported tests from WPT.

• web-platform-tests/css/css-grid/grid-model/grid-container-sizing-constraints-001-expected.txt: Added.
• web-platform-tests/css/css-grid/grid-model/grid-container-sizing-constraints-001.html: Added.
• web-platform-tests/css/css-grid/grid-model/w3c-import.log:

Source/WebCore:

The spec is quite clear
(​https://drafts.csswg.org/css-sizing/#valdef-width-min-content):

"min-content

If specified for the inline axis, use the min-content inline size;
otherwise behaves as the property’s initial value."

So if a grid container has "height: min-content" it should behave
the same than with "height: auto".

The patch removes computeIntrinsicLogicalContentHeightUsing() in
RenderGrid as we don't need a custom one anymore.
We can also get rid of m_minContentHeight and m_maxContentHeight
attributes that were only used for this logic.

Test: fast/css-grid-layout/grid-track-sizing-with-orthogonal-flows.html

fast/css-grid-layout/maximize-tracks-definite-indefinite-height.html
imported/w3c/web-platform-tests/css/css-grid/grid-model/grid-container-sizing-constraints-001.html

• rendering/RenderGrid.cpp:

(WebCore::RenderGrid::layoutBlock):
(WebCore::RenderGrid::computeIntrinsicLogicalWidths const):
(WebCore::RenderGrid::computeTrackSizesForIndefiniteSize const):

• rendering/RenderGrid.h:

LayoutTests:

Update tests results to the new behavior.
Also modified a little bit the test to keep them in sync with Blink versions.

• fast/css-grid-layout/grid-track-sizing-with-orthogonal-flows-expected.txt:
• fast/css-grid-layout/grid-track-sizing-with-orthogonal-flows.html:
• fast/css-grid-layout/maximize-tracks-definite-indefinite-height.html:

Nov 25, 2018:

10:55 PM Changeset in webkit [238487] by Ryan Haddad

• 2 edits in trunk/LayoutTests

Unreviewed test gardening, rebaseline webgpu/queue-creation.html.

• webgpu/queue-creation-expected.txt:

5:06 PM Changeset in webkit [238486] by Alan Bujtas

• 2 edits in trunk/Tools

Unreviewed test fix after r238467.

• TestWebKitAPI/Tests/WTF/ListHashSet.cpp:

(TestWebKitAPI::TEST):

2:59 PM Changeset in webkit [238485] by Wenson Hsieh

• 2 edits in trunk/Tools

Unreviewed, fix the internal iOS 12 build after r238471

• TestWebKitAPI/Tests/WebKitCocoa/DataDetection.mm:

(TEST):

2:11 PM Changeset in webkit [238484] by Devin Rousso

• 20 edits
  3 copies
  8 adds in trunk/Source/WebInspectorUI

Web Inspector: Timelines: add Media timeline
​https://bugs.webkit.org/show_bug.cgi?id=191625

Reviewed by Matt Baker.

Add a new timeline to Timelines for media related events (e.g. event/fullscreen/low-power).
Mimics what is visible by using the same instrumentation points as the Network tab when
"Group Media Requests" is enabled.

• UserInterface/Models/MediaInstrument.js: Added.

(WI.MediaInstrument):
(WI.MediaInstrument.supported):
(WI.MediaInstrument.prototype.get timelineRecordType):
(WI.MediaInstrument.prototype.startInstrumentation):
(WI.MediaInstrument.prototype.stopInstrumentation):

• UserInterface/Models/MediaTimelineRecord.js: Added.

(WI.MediaTimelineRecord):
(WI.MediaTimelineRecord.prototype.get eventType):
(WI.MediaTimelineRecord.prototype.get domNode):
(WI.MediaTimelineRecord.prototype.get domEvent):
(WI.MediaTimelineRecord.prototype.get isLowPower):
(WI.MediaTimelineRecord.prototype.get displayName):
(WI.MediaTimelineRecord.prototype.saveIdentityToCookie):

• UserInterface/Views/MediaTimelineDataGridNode.js: Added.

(WI.MediaTimelineDataGridNode):
(WI.MediaTimelineDataGridNode.prototype.get records):
(WI.MediaTimelineDataGridNode.prototype.get data):
(WI.MediaTimelineDataGridNode.prototype.createCellContent):
(WI.MediaTimelineDataGridNode.prototype.iconClassNames):
(WI.MediaTimelineDataGridNode.prototype.filterableDataForColumn):

• UserInterface/Views/MediaTimelineOverviewGraph.js: Added.

(WI.MediaTimelineOverviewGraph):
(WI.MediaTimelineOverviewGraph.prototype.reset):
(WI.MediaTimelineOverviewGraph.prototype.shown):
(WI.MediaTimelineOverviewGraph.prototype.hidden):
(WI.MediaTimelineOverviewGraph.prototype.layout):
(WI.MediaTimelineOverviewGraph.prototype.updateSelectedRecord):
(WI.MediaTimelineOverviewGraph.prototype._handleRecordAdded):

• UserInterface/Views/MediaTimelineOverviewGraph.css: Added.

(.timeline-overview-graph.media > .timeline-record-bar):
(.timeline-overview-graph.media > .timeline-record-bar > .segment):

• UserInterface/Views/MediaTimelineView.js: Added.

(WI.MediaTimelineView):
(WI.MediaTimelineView.prototype.get secondsPerPixel):
(WI.MediaTimelineView.prototype.get selectionPathComponents):
(WI.MediaTimelineView.prototype.closed):
(WI.MediaTimelineView.prototype.reset):
(WI.MediaTimelineView.prototype.dataGridSortComparator.compareDOMNodes):
(WI.MediaTimelineView.prototype.dataGridSortComparator.):
(WI.MediaTimelineView.prototype.dataGridSortComparator):
(WI.MediaTimelineView.prototype.layout):
(WI.MediaTimelineView.prototype._processPendingRecords):
(WI.MediaTimelineView.prototype._handleRecordAdded):
(WI.MediaTimelineView.prototype._handleSelectionPathComponentSiblingSelected):

• UserInterface/Views/MediaTimelineView.css: Added.

(.timeline-view.media > .data-grid):

• UserInterface/Views/DOMEventsBreakdownView.js:

(WI.DOMEventsBreakdownView.prototype.layout):

• UserInterface/Views/DOMNodeEventsContentView.js:

(WI.DOMNodeEventsContentView.prototype.initialLayout):

• UserInterface/Views/NetworkDOMNodeDetailView.js:

(WI.NetworkDOMNodeDetailView):
(WI.NetworkDOMNodeDetailView.prototype.showContentViewForIdentifier):

• UserInterface/Views/NetworkTableContentView.js:

(WI.NetworkTableContentView):
(WI.NetworkTableContentView.prototype.get filterNavigationItems):
(WI.NetworkTableContentView.prototype._populateWaterfallGraph):
(WI.NetworkTableContentView.prototype._showDetailView):
(WI.NetworkTableContentView.prototype._waterfallPopoverContentForNodeEntry):
Remove passing of startTimestamp to WI.DOMEventsBreakdownView, as we should be showing
absolute timestamps for each event, not relative to the start of the recording.

• UserInterface/Views/TimelineIcons.css:

(.media-icon .icon): Added.
(.dom-event-record .icon): Added.
(.dom-event-record.fullscreen .icon): Added.
(.low-power-record .icon): Added.

• UserInterface/Controllers/TimelineManager.js:

(WI.TimelineManager.availableTimelineTypes):
(WI.TimelineManager.prototype.capturingStarted):
(WI.TimelineManager.prototype.capturingStopped):
(WI.TimelineManager.prototype._updateAutoCaptureInstruments):
(WI.TimelineManager.prototype._handleDOMNodeDidFireEvent): Added.
(WI.TimelineManager.prototype._handleDOMNodeLowPowerChanged): Added.

• UserInterface/Views/TimelineRuler.js:

(WI.TimelineRuler.prototype.clearMarkers):
(WI.TimelineRuler.prototype._handleClick):

• UserInterface/Views/TimelineOverview.js:

(WI.TimelineOverview.prototype._timelineRulerMouseClicked):

• UserInterface/Models/TimelineRecording.js:

(WI.TimelineRecording.prototype.reset):
(WI.TimelineRecording.prototype.addEventMarker):
(WI.TimelineRecording.prototype.addRecord):
(WI.TimelineRecording.prototype._keyForRecord):
Drive-by: rework the logic for "click" event pass-through to the graph underneath.

• UserInterface/Models/Instrument.js:

(WI.Instrument.createForTimelineType):

• UserInterface/Models/TimelineRecord.js:

• UserInterface/Views/TimelineTabContentView.js:

(WI.TimelineTabContentView.displayNameForTimelineType):
(WI.TimelineTabContentView.iconClassNameForTimelineType):
(WI.TimelineTabContentView.genericClassNameForTimelineType):
(WI.TimelineTabContentView.iconClassNameForRecord):
(WI.TimelineTabContentView.displayNameForRecord):

• UserInterface/Views/TimelineRecordBar.css:

(.timeline-record-bar.timeline-record-type-media > .segment): Added.

• UserInterface/Views/ContentView.js:

(WI.ContentView.createFromRepresentedObject):

• UserInterface/Views/TimelineOverviewGraph.js:

(WI.TimelineOverviewGraph.createForTimeline):

• UserInterface/Models/ScriptTimelineRecord.js:

(WI.ScriptTimelineRecord.EventType.displayName):

• UserInterface/Main.html:
• UserInterface/Test.html:
• Localizations/en.lproj/localizedStrings.js:
• UserInterface/Images/DOMEvent.svg: Added.
• UserInterface/Images/DOMEventFullscreen.svg: Added.
• UserInterface/Images/LowPower.svg: Added.
• UserInterface/Images/MediaInstrument.svg: Added.

1:44 PM Changeset in webkit [238483] by Matt Baker

• 14 edits in trunk/Source/WebInspectorUI

Web Inspector: Remove parameters from TreeOutline SelectionDidChange event
​https://bugs.webkit.org/show_bug.cgi?id=191770

Reviewed by Devin Rousso.

• UserInterface/Controllers/CallFrameTreeController.js:

(WI.CallFrameTreeController.prototype._treeSelectionDidChange):

• UserInterface/Views/AuditNavigationSidebarPanel.js:

(WI.AuditNavigationSidebarPanel.prototype._treeSelectionDidChange):

• UserInterface/Views/CanvasSidebarPanel.js:

(WI.CanvasSidebarPanel):
(WI.CanvasSidebarPanel.prototype._treeSelectionDidChange):
(WI.CanvasSidebarPanel.prototype._treeOutlineSelectionDidChange): Deleted.

• UserInterface/Views/CanvasTabContentView.js:

(WI.CanvasTabContentView.prototype._canvasTreeOutlineSelectionDidChange):

• UserInterface/Views/DebuggerSidebarPanel.js:

(WI.DebuggerSidebarPanel.prototype._treeSelectionDidChange):

• UserInterface/Views/OpenResourceDialog.js:

(WI.OpenResourceDialog.prototype._treeSelectionDidChange):

• UserInterface/Views/ResourceSidebarPanel.js:

(WI.ResourceSidebarPanel.prototype._treeSelectionDidChange):

• UserInterface/Views/SearchSidebarPanel.js:

(WI.SearchSidebarPanel.prototype._treeSelectionDidChange):

• UserInterface/Views/StorageSidebarPanel.js:

(WI.StorageSidebarPanel._treeSelectionDidChange):

• UserInterface/Views/TimelineDataGrid.js:

(WI.TimelineDataGrid.prototype._popoverCallStackTreeSelectionDidChange):
(WI.TimelineDataGrid):

• UserInterface/Views/TimelineTabContentView.js:

(WI.TimelineTabContentView.prototype._recordingsTreeSelectionDidChange):
Use event.target.selectedTreeElement instead of event.data.selectedElement.

• UserInterface/Views/TimelineOverview.js:

(WI.TimelineOverview.prototype._timelinesTreeSelectionDidChange):
(WI.TimelineOverview.prototype._timelinesTreeSelectionDidChange.updateGraphSelectedState): Deleted.
Refactor away the single place event.data.deselectedElement was used.

• UserInterface/Views/TreeElement.js:

(WI.TreeElement.prototype.select):
(WI.TreeElement.prototype.deselect):
Remove selectedElement and deselectedElement from event data.

1:35 PM Changeset in webkit [238482] by Alan Bujtas

• 10 edits in trunk/Source/WebCore

[LFC] Remove PointInContainingBlock and PositionInContainingBlock
​https://bugs.webkit.org/show_bug.cgi?id=191954

Reviewed by Antti Koivisto.

Use Point and Position instead. Points and positions are by default in the containing block's coordinate system.

• layout/FormattingContextGeometry.cpp:

(WebCore::Layout::contentHeightForFormattingContextRoot):

• layout/LayoutUnits.h:

(WebCore::Layout::Position::operator LayoutUnit const):
(WebCore::Layout::operator<):
(WebCore::Layout::operator==):

• layout/floats/FloatAvoider.cpp:

(WebCore::Layout::FloatAvoider::FloatAvoider):
(WebCore::Layout::FloatAvoider::setHorizontalConstraints):
(WebCore::Layout::FloatAvoider::horizontalPositionCandidate):
(WebCore::Layout::FloatAvoider::initialHorizontalPosition const):

• layout/floats/FloatBox.cpp:

(WebCore::Layout::FloatBox::horizontalPositionCandidate):
(WebCore::Layout::FloatBox::verticalPositionCandidate):
(WebCore::Layout::FloatBox::initialVerticalPosition const):

• layout/floats/FloatingContext.cpp:

(WebCore::Layout::FloatingContext::positionForFloat const):
(WebCore::Layout::FloatingContext::positionForFloatAvoiding const):
(WebCore::Layout::FloatingContext::verticalPositionWithClearance const):
(WebCore::Layout::FloatingContext::floatingPosition const):
(WebCore::Layout::FloatingPair::horizontalConstraints const):
(WebCore::Layout::FloatingPair::bottom const):

• layout/floats/FloatingContext.h:
• layout/floats/FloatingState.cpp:

(WebCore::Layout::FloatingState::constraints const):
(WebCore::Layout::FloatingState::bottom const):

• layout/floats/FloatingState.h:

(WebCore::Layout::FloatingState::FloatItem::bottom const):

• layout/inlineformatting/InlineFormattingContext.cpp:

(WebCore::Layout::InlineFormattingContext::initializeNewLine const):

1:32 PM Changeset in webkit [238481] by Alan Bujtas

• 7 edits in trunk/Source/WebCore

[LFC] Rename Layout::Position to Layout::Point
​https://bugs.webkit.org/show_bug.cgi?id=191950

Reviewed by Antti Koivisto.

It actually represents a point.

• layout/FormattingContext.cpp:

(WebCore::Layout::FormattingContext::mapTopLeftToAncestor):
(WebCore::Layout::FormattingContext::mapCoordinateToAncestor):

• layout/FormattingContext.h:
• layout/LayoutUnits.h:

(WebCore::Layout::Point::Point):
(WebCore::Layout::Point::moveBy):
(WebCore::Layout::Position::operator LayoutPoint const): Deleted.
(WebCore::Layout::Position::Position): Deleted.
(WebCore::Layout::Position::moveBy): Deleted.

• layout/blockformatting/BlockFormattingContext.h:
• layout/blockformatting/BlockFormattingContextGeometry.cpp:

(WebCore::Layout::BlockFormattingContext::Geometry::staticPosition):

• layout/floats/FloatingState.cpp:

(WebCore::Layout::FloatingState::constraints const):

1:30 PM Changeset in webkit [238480] by Alan Bujtas

• 3 edits in trunk/Source/WebCore

[LFC] Floating code should use typed positions (PositionInContextRoot).
​https://bugs.webkit.org/show_bug.cgi?id=191949

Reviewed by Antti Koivisto.

Use PositionInContextRoot instead of LayoutUnit.

• layout/floats/FloatingState.cpp:

(WebCore::Layout::FloatingState::constraints const):
(WebCore::Layout::FloatingState::bottom const):

• layout/floats/FloatingState.h:

(WebCore::Layout::FloatingState::leftBottom const):
(WebCore::Layout::FloatingState::rightBottom const):
(WebCore::Layout::FloatingState::bottom const):

1:28 PM Changeset in webkit [238479] by Alan Bujtas

• 4 edits in trunk/Source/WebCore

Fix build after r238472.

• layout/inlineformatting/InlineFormattingContext.cpp:

(WebCore::Layout::InlineFormattingContext::splitInlineRunIfNeeded const):

• layout/inlineformatting/InlineFormattingContextGeometry.cpp:

(WebCore::Layout::InlineFormattingContext::Geometry::runWidth):

• layout/inlineformatting/InlineLineBreaker.cpp:

(WebCore::Layout::InlineLineBreaker::textWidth const):

10:22 AM Changeset in webkit [238478] by Michael Catanzaro

• 2 edits in trunk/Source/WTF

CRASH() should call abort() except on Darwin and in developer builds
​https://bugs.webkit.org/show_bug.cgi?id=184408

Reviewed by Daniel Bates.

CRASH() should call abort() except on Darwin and in developer builds, as discussed on
webkit-dev. This should be slightly nicer than dereferencing 0xbadbeef.

On Darwin, CRASH() uses a breakpoint trap, which seems to corrupt the stack on Linux, so we
can't do that.

Continue to call WTFCrash() in developer mode, and make no changes to WTFCrash(), since it
is reportedly useful in nightmare scenarios where core dumps are unavailable.

We also have to define CRASH_UNDER_CONSTEXPR_CONTEXT(). It's a bit odd that it's possible to
use a non-constexpr function here, but it works. Currently this macro uses WTFCrash(), which
is also non-constexpr.

• wtf/Assertions.h:

8:22 AM Changeset in webkit [238477] by Michael Catanzaro

• 2 edits in trunk/Source/WTF

Unreviewed, rolling out r238469.

Broke the build

Reverted changeset:

"CRASH() should call abort() except on Darwin and in developer
builds"
​https://bugs.webkit.org/show_bug.cgi?id=184408
​https://trac.webkit.org/changeset/238469

4:19 AM Changeset in webkit [238476] by Antti Koivisto

• 8 edits in trunk/Source/WebKit

RemoteLayerTreeNode construction cleanups
​https://bugs.webkit.org/show_bug.cgi?id=191951

Reviewed by Tim Horton.

• UIProcess/RemoteLayerTree/RemoteLayerTreeHost.h:
• UIProcess/RemoteLayerTree/RemoteLayerTreeHost.mm:

(WebKit::RemoteLayerTreeHost::updateLayerTree):

Drop unused LayerProperties argument.

(WebKit::RemoteLayerTreeHost::createLayer):

Split the actual construction out to makeNode.
Map updating is shared between platforms.
Move setting of layerID and clearing default actions to RemoteLayerTreeNode.

(WebKit::RemoteLayerTreeHost::makeNode):

Returns RemoteLayerTreeNode.

(WebKit::RemoteLayerTreeHost::setLayerID): Deleted.
(WebKit::RemoteLayerTreeHost::layerID): Deleted.

Move to RemoteLayerTreeNode.

• UIProcess/RemoteLayerTree/RemoteLayerTreeNode.h:
• UIProcess/RemoteLayerTree/RemoteLayerTreeNode.mm:

(WebKit::RemoteLayerTreeNode::RemoteLayerTreeNode):
(WebKit::RemoteLayerTreeNode::setLayerID):
(WebKit::RemoteLayerTreeNode::layerID):

• UIProcess/RemoteLayerTree/ios/RemoteLayerTreeHostIOS.mm:

(WebKit::RemoteLayerTreeHost::makeNode):
(WebKit::RemoteLayerTreeHost::createEmbeddedView):
(WebKit::RemoteLayerTreeHost::createLayer): Deleted.

• UIProcess/RemoteLayerTree/ios/RemoteLayerTreeViews.mm:

(-[WKCompositingView description]):
(-[WKUIRemoteView description]):
(-[WKBackdropView description]):

• WebProcess/WebPage/RemoteLayerTree/PlatformCAAnimationRemote.mm:

(WebKit::addAnimationToLayer):

4:11 AM Changeset in webkit [238475] by timothy_horton@apple.com

• 25 edits
  1 copy
  7 adds in trunk

Make it possible to insert editable images with a gesture
​https://bugs.webkit.org/show_bug.cgi?id=191937

Reviewed by Wenson Hsieh.

Source/WebCore:

Tests:

editing/images/redo-insert-editable-image-maintains-strokes.html,
editing/images/undo-insert-editable-image.html,
editing/images/basic-editable-image-from-execCommand.html

• Sources.txt:
• WebCore.xcodeproj/project.pbxproj:
• Source/WebCore/editing/EditorCommand.cpp:
• Source/WebCore/en.lproj/Localizable.strings:
• editing/EditAction.h:
• editing/Editor.cpp:

(WebCore::Editor::insertEditableImage):

• editing/Editor.h:
• editing/InsertEditableImageCommand.cpp: Added.

(WebCore::InsertEditableImageCommand::InsertEditableImageCommand):
(WebCore::InsertEditableImageCommand::doApply):

• editing/InsertEditableImageCommand.h: Added.

(WebCore::InsertEditableImageCommand::create):

• editing/VisibleSelection.cpp:

Add an editor command that inserts an editable image.
It will likely get a bit more complicated, but for now it just inserts
a 100% by 300px editable image.

Source/WebKit:

• UIProcess/API/Cocoa/WKWebView.mm:

(-[WKWebView _stylusTapGestureShouldCreateEditableImage]):

• UIProcess/API/Cocoa/WKWebViewInternal.h:

Add a internal getter for a WKWebViewConfiguration property.

• UIProcess/WebEditCommandProxy.cpp:

(WebKit::WebEditCommandProxy::nameForEditAction):
Add a undo name.

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

(-[WKContentView setupInteraction]):
(-[WKContentView cleanupInteraction]):
(-[WKContentView _removeDefaultGestureRecognizers]):
(-[WKContentView _addDefaultGestureRecognizers]):
Add a single-stylus-tap gesture recognizer.

(-[WKContentView _stylusSingleTapRecognized:]):
If allowed, request to insert an editable image when a stylus tap occurs.

• UIProcess/ios/WebPageProxyIOS.mm:

(WebKit::WebPageProxy::handleStylusSingleTapAtPoint):

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

(WebKit::WebPage::handleStylusSingleTapAtPoint):
Do a hit test, select the hit position, insert an editable image, and
then de-assist any assisted node (to make the keyboard go away).
For now, we'll only insert if we hit non-replaced elements,
though this heuristic will need to be enhanced significantly once we
decide on a design.

Source/WebKitLegacy/mac:

• WebCoreSupport/WebEditorClient.mm:

(undoNameForEditAction):
Add a undo name.

Tools:

• WebKitTestRunner/ios/UIScriptControllerIOS.mm:

(WTR::UIScriptController::drawSquareInEditableImage):
If the canvas already has a drawing, draw a new stroke on top of it
instead of removing the existing stroke.

LayoutTests:

• editing/images/basic-editable-image-from-execCommand-expected.txt: Added.
• editing/images/basic-editable-image-from-execCommand.html: Added.

Add a test that ensures that editable images also work when
inserted via the editor command.

• editing/images/redo-insert-editable-image-maintains-strokes-expected.txt: Added.
• editing/images/redo-insert-editable-image-maintains-strokes.html: Added.

Add a test that ensures that strokes are maintained when re-doing an
un-done editable image insertion.

• editing/images/undo-insert-editable-image-expected.txt: Added.
• editing/images/undo-insert-editable-image.html: Added.

Add a test that ensures that the selection stays in a sensible place
when undoing and redoing editable image insertion.

4:05 AM Changeset in webkit [238474] by timothy_horton@apple.com

• 4 edits in trunk/Source/WebKit

Scrolling and drawing compete for incoming gestures
​https://bugs.webkit.org/show_bug.cgi?id=191940

Reviewed by Wenson Hsieh.

• Platform/spi/ios/PencilKitSPI.h:
• UIProcess/API/Cocoa/WKWebView.mm:

(-[WKWebView _initializeWithConfiguration:]):

• UIProcess/ios/WKDrawingView.mm:

(-[WKDrawingView initWithEmbeddedViewID:webPageProxy:]):
Make scrolling and drawing mutually exclusive if editable images
are enabled.