Timeline

Aug 29, 2017:

11:59 PM Changeset in webkit [221344] by Carlos Garcia Campos

• 7 edits in trunk

REGRESSION(r221064): [GTK] Editor not correctly working after r221064
​https://bugs.webkit.org/show_bug.cgi?id=176052

Reviewed by Michael Catanzaro.

Source/WebKit:

Since r221064 we are not always notified about typing attributes when editor state changes. didChangeSelection
no longer includes the typing attributes in EditorStateChange message, it's scheduled to be sent after the
compositing layer have been flushed, but that part is not implemented for GTK+ port.

Fixes test /webkit2/WebKitWebView/editor-state/typing-attributes.

• WebProcess/WebPage/AcceleratedDrawingArea.cpp:

(WebKit::AcceleratedDrawingArea::updateBackingStoreState): Call WebPage::flushPendingEditorStateUpdate() after
the layout.

• WebProcess/WebPage/CoordinatedGraphics/CoordinatedLayerTreeHost.cpp:

(WebKit::CoordinatedLayerTreeHost::layerFlushTimerFired): Call WebPage::flushPendingEditorStateUpdate() after
the display sync.

• WebProcess/WebPage/DrawingAreaImpl.cpp:

(WebKit::DrawingAreaImpl::display): Call WebPage::flushPendingEditorStateUpdate() after the layout.

• WebProcess/WebPage/gtk/WebPageGtk.cpp:

(WebKit::WebPage::platformEditorState const): Also return early, setting isMissingPostLayoutData to true, when
there's no frame view, the view needs a layout or the result is not for editable content.

Tools:

Run the test inside a window to ensure display updates happen.

• TestWebKitAPI/Tests/WebKitGtk/TestWebViewEditor.cpp:

(testWebViewEditorEditorStateTypingAttributes):

10:09 PM Changeset in webkit [221343] by Wenson Hsieh

• 6 edits
  2 adds in trunk

REGRESSION(r210287) On drop, event.dataTransfer.getData("text") returns an empty string when dragging an image
​https://bugs.webkit.org/show_bug.cgi?id=170637
<rdar://problem/31347248>

Reviewed by Ryosuke Niwa.

Source/WebCore:

In r210287, the behavior of DragData::containsFiles was changed to return true if NSFilesPromisePboardType is
present in the pasteboard. This means that we will consider images dragged from web content, for which we add
the NSFilesPromisePboardType UTI, as containing files on the pasteboard; this, in turn, means we'll initialize
the DataTransfer upon drop with m_forFileDrag set to true. Due to early returns in getData() and setData() to
deny data access when dropping a dragged file, this means the page won't ever get access to the URL in the
pasteboard due to the presence of the NSFilesPromisePboardType UTI.

To fix this, we replace the early m_forFileDrag returns in getData and setData, instead early returning the null
string if there are any file URLs present on the pasteboard (determined via readFilenames() retrieving a non-
empty result).

Test: editing/pasteboard/drag-drop-href-as-text-data.html

• dom/DataTransfer.cpp:

(WebCore::DataTransfer::DataTransfer):
(WebCore::DataTransfer::getData const):
(WebCore::DataTransfer::setData):

Rather than bail upon forFileDrag() (formerly, m_forFileDrag) being true, bail if there are any file URLs
present on the pasteboard. It seems like this was the intention of the early return in the first place, to
prevent the page from being able to ask for a real file URL when dragging a file.

(WebCore::DataTransfer::files const):
(WebCore::DataTransfer::setDragImage):
(WebCore::DataTransfer::setDropEffect):
(WebCore::DataTransfer::setEffectAllowed):

Swap m_forDrag and m_forFileDrag with forDrag() and forFileDrag(), respectively.

• dom/DataTransfer.h:

(WebCore::DataTransfer::forDrag const):
(WebCore::DataTransfer::forFileDrag const):

Instead of caching two bools to represent state (m_forDrag and m_forFileDrag), just remember the DataTransfer's
m_type and turn the flags into const helpers that check for the value of m_type.

LayoutTests:

Adds a new test to verify that upon dropping an image enclosed within an anchor, DataTransfer.getData() can be
used to grab the href of the enclosing anchor.

• TestExpectations:
• editing/pasteboard/drag-drop-href-as-text-data-expected.txt: Added.
• editing/pasteboard/drag-drop-href-as-text-data.html: Added.
• platform/mac-wk1/TestExpectations:

9:32 PM Changeset in webkit [221342] by commit-queue@webkit.org

• 4 edits in trunk

[Fetch API] Request should throw when keep alive is true and body is a ReadableStream
​https://bugs.webkit.org/show_bug.cgi?id=176083

Patch by Youenn Fablet <​youenn@apple.com> on 2017-08-29
Reviewed by Alex Christensen.

LayoutTests/imported/w3c:

• web-platform-tests/fetch/api/request/request-keepalive-expected.txt:

Source/WebCore:

Covered by rebased test.

• Modules/fetch/FetchRequest.cpp:

(WebCore::FetchRequest::setBody):

9:20 PM Changeset in webkit [221341] by commit-queue@webkit.org

• 6 edits in trunk/Source/JavaScriptCore

Unreviewed, rolling out r221317.
​https://bugs.webkit.org/show_bug.cgi?id=176090

"It broke a testing mode because we will never FTL compile a
function that repeatedly throws" (Requested by saamyjoon on
#webkit).

Reverted changeset:

"Throwing an exception in the DFG/FTL should not be a
jettison-able OSR exit"
​https://bugs.webkit.org/show_bug.cgi?id=176060
​http://trac.webkit.org/changeset/221317

9:07 PM Changeset in webkit [221340] by don.olmstead@sony.com

• 12 edits in trunk

[CMake] Use find_package for zlib
​https://bugs.webkit.org/show_bug.cgi?id=176075

Reviewed by Alex Christensen.

.:

• Source/cmake/OptionsAppleWin.cmake:
• Source/cmake/OptionsWinCairo.cmake:

Source/WebCore:

No new tests. No change in behavior.

• CMakeLists.txt:
• PlatformAppleWin.cmake:
• PlatformMac.cmake:
• PlatformWinCairo.cmake:

Source/WebKitLegacy:

• PlatformWin.cmake:

Tools:

• TestWebKitAPI/PlatformWin.cmake:

9:07 PM Changeset in webkit [221339] by aestes@apple.com

• 2 edits in trunk/Source/WebCore/PAL

[Xcode] Fix up file and group paths in PAL.xcodeproj
​https://bugs.webkit.org/show_bug.cgi?id=176073

Reviewed by Alex Christensen.

• PAL.xcodeproj/project.pbxproj:

9:04 PM Changeset in webkit [221338] by Matt Baker

• 22 edits
  1 copy in trunk/Source/WebInspectorUI

Web Inspector: Critical content browser toolbar buttons are hidden at narrow widths
​https://bugs.webkit.org/show_bug.cgi?id=175999

Reviewed by Devin Rousso.

This patch adds a VisibilityPriority concept to NavigationItems. If a
NavigationBar cannot fit all of its items in the available space, items
are hidden to make room, starting with the lowest priority item. Consecutive
dividers are then collapsed, as well as leading and trailing dividers.

• UserInterface/Main.html:

New file.

• UserInterface/Views/CanvasContentView.js:

(WI.CanvasContentView):

• UserInterface/Views/ConsoleDrawer.js:

(WI.ConsoleDrawer):

• UserInterface/Views/ContentBrowser.js:

(WI.ContentBrowser):

• UserInterface/Views/ContentBrowserTabContentView.js:

(WI.ContentBrowserTabContentView):

• UserInterface/Views/DOMTreeContentView.js:

(WI.DOMTreeContentView):
Set High and Low priorities, and group the back/forward buttons.

• UserInterface/Views/GroupNavigationItem.js: Added.

(WI.GroupNavigationItem):
(WI.GroupNavigationItem.prototype.get navigationItems):
(WI.GroupNavigationItem.prototype.get minimumWidth):
(WI.GroupNavigationItem.prototype.updateLayout):
(WI.GroupNavigationItem.prototype.didAttach):
(WI.GroupNavigationItem.prototype.didDetach):
NavigationItem groups. Grouped items are shown/hidden together.

• UserInterface/Views/HierarchicalPathNavigationItem.js:

(WI.HierarchicalPathNavigationItem.prototype.updateLayout):

• UserInterface/Views/ImageResourceContentView.js:

(WI.ImageResourceContentView):

• UserInterface/Views/IndexedDatabaseObjectStoreContentView.js:

(WI.IndexedDatabaseObjectStoreContentView):

• UserInterface/Views/LogContentView.js:

(WI.LogContentView):
Set High and Low priorities.

• UserInterface/Views/NavigationBar.css:

(.navigation-bar .item.force-hidden):
New hidden class, which must be tracked separately from ".hidden".
The former is an implementation detail of NavigationBar, while the
latter is set by the client.

• UserInterface/Views/NavigationBar.js:

(WI.NavigationBar.prototype.insertNavigationItem):
(WI.NavigationBar.prototype.removeNavigationItem):
(WI.NavigationBar.prototype.findNavigationItem.matchingSelfOrChild):
(WI.NavigationBar.prototype.findNavigationItem):
(WI.NavigationBar.prototype.layout.forceItemHidden):
(WI.NavigationBar.prototype.layout.isDivider):
(WI.NavigationBar.prototype.layout.calculateVisibleItemWidth):
(WI.NavigationBar.prototype.layout):
(WI.NavigationBar.prototype._calculateMinimumWidth):
(WI.NavigationBar.prototype.get _visibleNavigationItems):
(WI.NavigationBar):

• UserInterface/Views/NavigationItem.js:

Add support for visibility priority, an integer value that determines the
order in which items are hidden when the NavigationBar becomes too narrow
to fit all of items child items. NavigationIte3m defines constants for
Low, Normal (the default), and High priority.

(WI.NavigationItem):
(WI.NavigationItem.prototype.get minimumWidth):
(WI.NavigationItem.prototype.get width):
(WI.NavigationItem.prototype.get visibilityPriority):
(WI.NavigationItem.prototype.set visibilityPriority):
(WI.NavigationItem.prototype.updateLayout):
(WI.NavigationItem.prototype.didAttach):
(WI.NavigationItem.prototype.didDetach):
Encapsulate the setting of the parent NavigationBar. Needed so that
GroupNavigationItem can forward this action to its children.

• UserInterface/Views/NetworkGridContentView.js:

(WI.NetworkGridContentView):

• UserInterface/Views/RadioButtonNavigationItem.js:

(WI.RadioButtonNavigationItem.prototype.updateLayout):

• UserInterface/Views/RecordingContentView.js:

(WI.RecordingContentView):

• UserInterface/Views/ScriptContentView.js:

(WI.ScriptContentView):

• UserInterface/Views/TextContentView.js:

(WI.TextContentView):

• UserInterface/Views/TextResourceContentView.js:

(WI.TextResourceContentView):

• UserInterface/Views/TimelineRecordingContentView.js:

(WI.TimelineRecordingContentView):

• UserInterface/Views/TimelineTabContentView.js:

(WI.TimelineTabContentView):
Set High and Low priorities, and group Timeline view mode buttons.

7:54 PM Changeset in webkit [221337] by Brent Fulgham

• 10 edits in trunk

ResourceLoadStatistics logic does not understand custom WebsiteData stores
​https://bugs.webkit.org/show_bug.cgi?id=176037
<rdar://problem/33338238>

Reviewed by Alex Christensen.

Source/WebKit:

The NetworkProcess::updateCookiePartitioningForTopPrivatelyOwnedDomains always notifies the default website
data store about observations it has made. This should be revised so that WebKit clients that register
custom data stores through the WKWebsiteDataStores API can be assured that observations made in one session
do not manipulate data from another session.

• NetworkProcess/NetworkProcess.cpp:

(WebKit::NetworkProcess::updateCookiePartitioningForTopPrivatelyOwnedDomains): Use the passed sessionID to locate
the correct NetworkStorageSession to notify about the new partitioning data.

• NetworkProcess/NetworkProcess.h:
• NetworkProcess/NetworkProcess.messages.in: Accept new argument.
• UIProcess/WebsiteData/WebsiteDataStore.cpp:

(WebKit::WebsiteDataStore::updateCookiePartitioningForTopPrivatelyOwnedDomains): Include the WebsiteDataStore's
session ID in the message to update cookie partition data.

Tools:

• TestWebKitAPI/Tests/WebKitCocoa/WebsiteDataStoreCustomPaths.mm:

(TEST): Update for ResourceLoadStatistics directories.

7:40 PM Changeset in webkit [221336] by webkit@devinrousso.com

• 9 edits in trunk

CallTracingCallback should ignore readonly attribute
​https://bugs.webkit.org/show_bug.cgi?id=176070

Reviewed by Matt Baker.

Source/WebCore:

Update existing tests to remove calls to existing readonly attributes.

• bindings/scripts/CodeGeneratorJS.pm:

(GenerateAttributeGetterBodyDefinition):
If the attribute is marked readonly do not add logic for the CallTracingCallback.

• bindings/scripts/test/TestCallTracer.idl:
• bindings/scripts/test/JS/JSTestCallTracer.cpp:

(WebCore::JSTestCallTracerConstructor::initializeProperties):
(WebCore::jsTestCallTracerTestReadonlyAttributeGetter):
(WebCore::jsTestCallTracerTestReadonlyAttribute):

LayoutTests:

• inspector/canvas/recording-2d-expected.txt:
• inspector/canvas/recording-2d.html:
• inspector/canvas/recording-webgl-expected.txt:
• inspector/canvas/recording-webgl.html:

Remove calls to existing readonly attributes.

7:26 PM Changeset in webkit [221335] by commit-queue@webkit.org

• 3 edits
  2 adds in trunk

Assertion failure when opening a file with a missing tag closing bracket
​https://bugs.webkit.org/show_bug.cgi?id=176061

Patch by Said Abou-Hallawa <​sabouhallawa@apple.com> on 2017-08-29
Reviewed by Darin Adler.

Source/WebCore:

If a tag is missing its closing bracket, the tokenizer just needs to advance()
the character position without checking m_currentCharacter != '\n'. Newline
character is a valid ending for partially closed tags.

Test: fast/tokenizer/missing-script-tag-closing-bracket.html

• html/parser/HTMLTokenizer.cpp:

(WebCore::HTMLTokenizer::commitToPartialEndTag):

LayoutTests:

• fast/tokenizer/missing-script-tag-closing-bracket-expected.txt: Added.
• fast/tokenizer/missing-script-tag-closing-bracket.html: Added.

7:13 PM Changeset in webkit [221334] by beidson@apple.com

• 7 edits in trunk

Rename "potentionally trustworthy" to "potentially trustworthy"
​https://bugs.webkit.org/show_bug.cgi?id=176081

Reviewed by Daniel Bates.

Source/WebCore:

The Secure Context spec refers to this as "potentially trustworthy"

It's possible that this was done potentionally, but we should follow the spec to alleviate confusion.

• dom/Document.cpp:

(WebCore::Document::isSecureContext const):

• page/SecurityOrigin.cpp:

(WebCore::shouldTreatAsPotentiallyTrustworthy):
(WebCore::shouldTreatAsPotentionallyTrustworthy): Deleted.

• page/SecurityOrigin.h:

(WebCore::SecurityOrigin::isPotentiallyTrustworthy const):
(WebCore::SecurityOrigin::isPotentionallyTrustworthy const): Deleted.

• workers/WorkerGlobalScope.cpp:

(WebCore::WorkerGlobalScope::isSecureContext const):

Tools:

• TestWebKitAPI/Tests/WebCore/SecurityOrigin.cpp:

(TestWebKitAPI::TEST_F):

6:40 PM WebKitGTK/Gardening/Calendar edited by Adrian Perez de Castro

(diff)

6:39 PM WebKitGTK/Gardening/Calendar edited by Adrian Perez de Castro

(diff)

6:38 PM Changeset in webkit [221333] by commit-queue@webkit.org

• 5 edits in trunk/Source/WebKit

Automatically determine if a class has a modern decoder
​https://bugs.webkit.org/show_bug.cgi?id=176084

Patch by Alex Christensen <​achristensen@webkit.org> on 2017-08-29
Reviewed by Sam Weinig.

Instead of determining if ModernDecoder is defined in the class,
just check the signature of the decode member function.

• Platform/IPC/ArgumentCoder.h:

(): Deleted.

• Shared/WebPageCreationParameters.h:
• Shared/WebPageGroupData.h:
• Shared/WebsitePolicies.h:

6:29 PM Changeset in webkit [221332] by commit-queue@webkit.org

• 3 edits in trunk/Source/WebInspectorUI

REGRESSION(r220235): Web Inspector: Global search should not happen incrementally
​https://bugs.webkit.org/show_bug.cgi?id=176063

Patch by Joseph Pecoraro <Joseph Pecoraro> on 2017-08-29
Reviewed by Devin Rousso.

• UserInterface/Base/Main.js:

(WI.contentLoaded):

• UserInterface/Views/SearchBar.js:

(WI.SearchBar):
(WI.SearchBar.prototype._handleKeydownEvent):
Remove delegate and simplify setting incremental.

6:27 PM Changeset in webkit [221331] by commit-queue@webkit.org

• 2 edits in trunk/Tools

[Win][JSCOnly] Fix call for updating wincairo-libs in build-jsc
​https://bugs.webkit.org/show_bug.cgi?id=176080

Patch by Stephan Szabo <​stephan.szabo@sony.com> on 2017-08-29
Reviewed by Brent Fulgham.

• Scripts/build-jsc:

6:22 PM Changeset in webkit [221330] by Yusuke Suzuki

• 4 edits
  1 add in trunk/Source/WTF

[JSC] Use table based approach for JSON.stringify's Quote
​https://bugs.webkit.org/show_bug.cgi?id=176044

Reviewed by Darin Adler.

We change escape operation of JSON Quote from branch-based to table-based.
This patch partially adopts SpiderMonkey's change to StringBuilderJSON.cpp
to optimize this escaping operation. We separate changes from StringBuilder.cpp
to apply MPL to StringBuilderJSON.cpp file. Since WebKit already adopts MPL in
some files (like, DateMath.h), it is acceptable.

Kraken json-stringify-tinderbox shows 7.2% improvement.

baseline patched

json-stringify-tinderbox 40.429+-0.771 ^{37.693+-0.862} definitely 1.0726x faster

• WTF.xcodeproj/project.pbxproj:
• wtf/CMakeLists.txt:
• wtf/text/StringBuilder.cpp:

(WTF::appendQuotedJSONStringInternalSlow): Deleted.
(WTF::appendQuotedJSONStringInternal): Deleted.
(WTF::StringBuilder::appendQuotedJSONString): Deleted.

• wtf/text/StringBuilderJSON.cpp: Added.

(WTF::appendQuotedJSONStringInternal):
(WTF::StringBuilder::appendQuotedJSONString):

5:39 PM Changeset in webkit [221329] by commit-queue@webkit.org

• 18 edits
  1 add in trunk

Add support for FetchRequest.body
​https://bugs.webkit.org/show_bug.cgi?id=176066

Patch by Youenn Fablet <​youenn@apple.com> on 2017-08-29
Reviewed by Alex Christensen.

LayoutTests/imported/w3c:

• web-platform-tests/fetch/api/request/request-disturbed-expected.txt:
• web-platform-tests/fetch/api/request/request-idl-expected.txt:

Source/WebCore:

Covered by existing tests.

Adding support for body attribute getter for FetchRequest.
To do so, FetchRequest will need to create a ReadableStream.
Adding support for DOM based creation of ReadableStream and conversion to JS values.
Small refactoring to make names more consistent.

• CMakeLists.txt:
• Modules/beacon/NavigatorBeacon.cpp:

(WebCore::NavigatorBeacon::sendBeacon):

• Modules/cache/Cache.cpp:

(WebCore::Cache::put):

• Modules/fetch/FetchBody.cpp:

(WebCore::FetchBody::extract):

• Modules/fetch/FetchBody.h:

(WebCore::FetchBody::hasReadableStream const):
(WebCore::FetchBody::readableStream):
(WebCore::FetchBody::setReadableStream):
(WebCore::FetchBody::FetchBody):
(WebCore::FetchBody::isReadableStream const): Deleted.
(WebCore::FetchBody::setAsReadableStream): Deleted.

• Modules/fetch/FetchBody.idl:
• Modules/fetch/FetchBodyOwner.cpp:

(WebCore::FetchBodyOwner::readableStream):

• Modules/fetch/FetchBodyOwner.h:

(WebCore::FetchBodyOwner::hasReadableStreamBody const):
(WebCore::FetchBodyOwner::isReadableStreamBody const): Deleted.

• Modules/fetch/FetchRequest.cpp:

(WebCore::FetchRequest::setBody):

• Modules/fetch/FetchRequest.h:
• Modules/fetch/FetchResponse.cpp:

(WebCore::FetchResponse::setBodyAsReadableStream):
(WebCore::FetchResponse::fetch):

• Modules/fetch/FetchResponse.h:
• WebCore.xcodeproj/project.pbxproj:
• bindings/js/ReadableStream.h:

(WebCore::JSConverter<IDLInterface<ReadableStream>>::convert):

5:29 PM Changeset in webkit [221328] by Yusuke Suzuki

• 4 edits in trunk/Source/JavaScriptCore

[DFG] Add constant folding rule to convert CompareStrictEq(Untyped, Untyped [with non string cell constant]) to CompareEqPtr(Untyped)
​https://bugs.webkit.org/show_bug.cgi?id=175895

Reviewed by Saam Barati.

We have bucket === @sentinelMapBucket code in builtin. Since @sentinelMapBucket and bucket
are MapBucket cell (SpecCellOther), we do not have any good fixup for CompareStrictEq.
But rather than introducing a special fixup edge (like, NonStringCellUse), converting
CompareStrictEq(Untyped, Untyped) to CompareEqPtr is simpler.
In constant folding phase, we convert CompareStrictEq(Untyped, Untyped) to CompareEqPtr(Untyed)
if one side of the children is constant non String cell.

This slightly optimizes map/set iteration.

set-for-each 4.5064+-0.3072 ^{3.2862+-0.2098} definitely 1.3713x faster
large-map-iteration 56.2583+-1.6640 53.6798+-2.0097 might be 1.0480x faster
set-for-of 8.8058+-0.5953 ^{7.5832+-0.3805} definitely 1.1612x faster
map-for-each 4.2633+-0.2694 ^{3.3967+-0.3013} definitely 1.2551x faster
map-for-of 13.1556+-0.5707 12.4911+-0.6004 might be 1.0532x faster

• dfg/DFGAbstractInterpreterInlines.h:

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

• dfg/DFGConstantFoldingPhase.cpp:

(JSC::DFG::ConstantFoldingPhase::foldConstants):

• dfg/DFGNode.h:

(JSC::DFG::Node::convertToCompareEqPtr):

5:06 PM Changeset in webkit [221327] by Yusuke Suzuki

• 55 edits
  3 adds in trunk

[JSC] Use reifying system for "name" property of builtin JSFunction
​https://bugs.webkit.org/show_bug.cgi?id=175260

Reviewed by Saam Barati.

JSTests:

• stress/accessors-get-set-prefix.js:
• stress/builtin-function-name.js: Added.

(shouldBe):
(shouldThrow):
(shouldBe.JSON.stringify.Object.getOwnPropertyDescriptor):
(shouldBe.JSON.stringify.Object.getOwnPropertyNames.Array.prototype.filter.sort):

Source/JavaScriptCore:

Currently builtin JSFunction uses direct property for "name", which is different
from usual JSFunction. Usual JSFunction uses reifying system for "name". We would like
to apply this reifying mechanism to builtin JSFunction to simplify code and drop
JSFunction::createBuiltinFunction.

We would like to store the "correct" name in FunctionExecutable. For example,
we would like to store the name like "get [Symbol.species]" to FunctionExecutable
instead of specifying name when creating JSFunction. To do so, we add a new
annotations, @getter and @overriddenName. When @getter is specified, the name of
the function becomes "get xxx". And when @overriddenName="xxx" is specified,
the name of the function becomes "xxx".

• Scripts/builtins/builtins_generate_combined_header.py:

(generate_section_for_code_table_macro):

• Scripts/builtins/builtins_generate_combined_implementation.py:

(BuiltinsCombinedImplementationGenerator.generate_secondary_header_includes):

• Scripts/builtins/builtins_generate_separate_header.py:

(generate_section_for_code_table_macro):

• Scripts/builtins/builtins_generate_separate_implementation.py:

(BuiltinsSeparateImplementationGenerator.generate_secondary_header_includes):

• Scripts/builtins/builtins_model.py:

(BuiltinFunction.init):
(BuiltinFunction.fromString):

• Scripts/builtins/builtins_templates.py:
• Scripts/tests/builtins/JavaScriptCore-Builtin.prototype-Combined.js:

(overriddenName.string_appeared_here.match):
(intrinsic.RegExpTestIntrinsic.test):

• Scripts/tests/builtins/JavaScriptCore-Builtin.prototype-Separate.js:

(overriddenName.string_appeared_here.match):
(intrinsic.RegExpTestIntrinsic.test):

• Scripts/tests/builtins/expected/JavaScriptCore-Builtin.Promise-Combined.js-result:
• Scripts/tests/builtins/expected/JavaScriptCore-Builtin.Promise-Separate.js-result:
• Scripts/tests/builtins/expected/JavaScriptCore-Builtin.prototype-Combined.js-result:
• Scripts/tests/builtins/expected/JavaScriptCore-Builtin.prototype-Separate.js-result:
• Scripts/tests/builtins/expected/JavaScriptCore-BuiltinConstructor-Combined.js-result:
• Scripts/tests/builtins/expected/JavaScriptCore-BuiltinConstructor-Separate.js-result:
• Scripts/tests/builtins/expected/JavaScriptCore-InternalClashingNames-Combined.js-result:
• Scripts/tests/builtins/expected/WebCore-AnotherGuardedInternalBuiltin-Separate.js-result:
• Scripts/tests/builtins/expected/WebCore-ArbitraryConditionalGuard-Separate.js-result:
• Scripts/tests/builtins/expected/WebCore-GuardedBuiltin-Separate.js-result:
• Scripts/tests/builtins/expected/WebCore-GuardedInternalBuiltin-Separate.js-result:
• Scripts/tests/builtins/expected/WebCore-UnguardedBuiltin-Separate.js-result:
• Scripts/tests/builtins/expected/WebCore-xmlCasingTest-Separate.js-result:
• builtins/BuiltinExecutables.cpp:

(JSC::BuiltinExecutables::BuiltinExecutables):

• builtins/BuiltinExecutables.h:
• builtins/FunctionPrototype.js:

(symbolHasInstance): Deleted.

• builtins/GlobalOperations.js:

(globalPrivate.speciesGetter): Deleted.

• builtins/IteratorPrototype.js:

(symbolIteratorGetter): Deleted.

• builtins/RegExpPrototype.js:

(match): Deleted.
(replace): Deleted.
(search): Deleted.
(split): Deleted.

• jsc.cpp:

(functionCreateBuiltin):

• runtime/FunctionPrototype.cpp:

(JSC::FunctionPrototype::addFunctionProperties):

• runtime/IteratorPrototype.cpp:

(JSC::IteratorPrototype::finishCreation):

• runtime/JSFunction.cpp:

(JSC::JSFunction::getOwnNonIndexPropertyNames):
(JSC::JSFunction::reifyLazyBoundNameIfNeeded):
(JSC::JSFunction::createBuiltinFunction): Deleted.

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

(JSC::JSGlobalObject::init):

• runtime/JSObject.cpp:

(JSC::JSObject::putDirectBuiltinFunction):
(JSC::JSObject::putDirectBuiltinFunctionWithoutTransition):

• runtime/JSTypedArrayViewPrototype.cpp:

(JSC::JSTypedArrayViewPrototype::finishCreation):

• runtime/Lookup.cpp:

(JSC::reifyStaticAccessor):

• runtime/RegExpPrototype.cpp:

(JSC::RegExpPrototype::finishCreation):

Source/WebCore:

Use @getter for JSBuiltin getters.

• Modules/fetch/FetchResponse.js:

(bodyUsed): Deleted.
(body): Deleted.

• Modules/streams/ReadableByteStreamController.js:

(byobRequest): Deleted.
(desiredSize): Deleted.

• Modules/streams/ReadableStream.js:

(locked): Deleted.

• Modules/streams/ReadableStreamBYOBReader.js:

(closed): Deleted.

• Modules/streams/ReadableStreamBYOBRequest.js:

(view): Deleted.

• Modules/streams/ReadableStreamDefaultController.js:

(desiredSize): Deleted.

• Modules/streams/ReadableStreamDefaultReader.js:

(closed): Deleted.

• Modules/streams/WritableStream.js:

(closed): Deleted.
(ready): Deleted.
(state): Deleted.

• bindings/js/JSDOMBuiltinConstructor.h:

(WebCore::JSDOMBuiltinConstructor<JSClass>::finishCreation):

LayoutTests:

• js/dom/builtin-getter-name-expected.txt: Added.
• js/dom/builtin-getter-name.html: Added.

4:50 PM Changeset in webkit [221326] by pvollan@apple.com

• 2 edits in trunk/LayoutTests

Rebaseline editing/style/5084241.html after r220706.

Unreviewed test gardening.

• platform/win/editing/style/5084241-expected.txt:

4:47 PM Changeset in webkit [221325] by Matt Lewis

• 2 edits
  1 copy
  3 adds in trunk/LayoutTests

Rebaselined http/tests/websocket/tests/hybi/secure-cookie-secure-connection.pl for ios-11.

Unreviewed test gardening.

• http/tests/websocket/tests/hybi/secure-cookie-secure-connection-expected.txt:

4:45 PM Changeset in webkit [221324] by Nikita Vasilyev

• 2 edits in trunk/Source/WebInspectorUI

Web Inspector: Convert all methods in Utilities.js to ECMAScript 2015 shorthand syntax
​https://bugs.webkit.org/show_bug.cgi?id=176067

Reviewed by Joseph Pecoraro.

• UserInterface/Base/Utilities.js:

(get return):
(addStringToken):
(prettyFunctionName):
(warn):
(error):
(isEmptyObject):
(isEnterKey):
(resolveDotsInPath):
(parseMIMEType):
(simpleGlobStringToRegExp):
(defaultComparator):
(get Object):
(appendWebInspectorSourceURL):
(appendWebInspectorConsoleEvaluationSourceURL):
(isWebInspectorInternalScript):
(isWebInspectorConsoleEvaluationScript):
(isWebKitInjectedScript):
(isWebKitInternalScript):
(isFunctionStringNativeCode):
(isTextLikelyMinified):
(doubleQuotedString):
(insertionIndexForObjectInListSortedByFunction):
(insertObjectIntoSortedArray):
(decodeBase64ToBlob):

4:35 PM Changeset in webkit [221323] by pvollan@apple.com

• 2 edits in trunk/Source/WTF

[Win] Crash under WorkQueue::performWorkOnRegisteredWorkThread in layout tests.
​https://bugs.webkit.org/show_bug.cgi?id=176064

Reviewed by Saam Barati.

The crash log indicates that the function pointer is null in this case.

• wtf/win/WorkQueueWin.cpp:

(WTF::WorkQueue::dispatch):

4:32 PM Changeset in webkit [221322] by Matt Lewis

• 4 edits in trunk/LayoutTests

Rebaselined js/dom/global-constructors-attributes.html after r221258.

Unreviewed test gardening.

• platform/mac-elcapitan-wk2/js/dom/global-constructors-attributes-expected.txt:
• platform/mac-wk1/js/dom/global-constructors-attributes-expected.txt:
• platform/mac/js/dom/global-constructors-attributes-expected.txt:

4:32 PM Changeset in webkit [221321] by aestes@apple.com

• 2 edits in trunk/Source/WebCore

REGRESSION (r215290): "Where From" metadata is empty when dragging an image out of Safari
​https://bugs.webkit.org/show_bug.cgi?id=176068

Reviewed by Wenson Hsieh.

If you drag an image out of Safari, drop it on the Desktop, and open "Get Info", there is a
"Where from:" field under "More Info" that should list the image's URL. After r215290, this
field is blank due to a mixup with the string passed to URLWithUserTypedString().

Fix this by passing metadataURLString to URLWithUserTypedString() instead of urlString.

Not possible to reliably test this since WebKit adds metadata on a global concurrent queue.

• platform/mac/FileSystemMac.mm:

(WebCore::setMetadataURL):

4:10 PM Changeset in webkit [221320] by pvollan@apple.com

• 2 edits in trunk/LayoutTests

Rebaseline editing/style/5065910.html after r220706.

Unreviewed test gardening.

• platform/win/editing/style/5065910-expected.txt:

3:49 PM Changeset in webkit [221319] by achristensen@apple.com

• 10 edits in trunk/Source/WebKit

Begin transition to modern IPC decoding
​https://bugs.webkit.org/show_bug.cgi?id=176043

Reviewed by JF Bastien.

Right now, if a class is decoded from IPC we must have a default constructor.
This prevents us from having Ref or C++ references in such types, which is cluttering up our code.
This is because IPC::decode makes a default-constructed object, fills it, and returns a bool indicating success.
Making IPC::decode instead return a std::optional makes it so we do not need to call an empty constructor.
This could also enable us to add IPC::Decoder::operator>> and other fun things!
I also modernized two arbitrary classes, WebsitePolicies and WebPageGroupData with more to come.
There's no good way to update the actual generated IPC code until each class has been transitioned.

• Platform/IPC/ArgumentCoder.h:

(IPC::ArgumentCoder::decode):

• Platform/IPC/Decoder.h:

(IPC::Decoder::decode):

• Shared/WebPageCreationParameters.cpp:

(WebKit::WebPageCreationParameters::decode):

• Shared/WebPageCreationParameters.h:
• Shared/WebPageGroupData.cpp:

(WebKit::WebPageGroupData::decode):

• Shared/WebPageGroupData.h:
• Shared/WebsitePolicies.h:

(WebKit::WebsitePolicies::decode):

3:45 PM WebInspectorTests edited by Nikita Vasilyev

List assertion matchers. They are hard to find otherwise. (diff)

3:34 PM Changeset in webkit [221318] by commit-queue@webkit.org

• 2 edits in trunk/Source/WebCore

CacheStorage does not need to sort its caches
​https://bugs.webkit.org/show_bug.cgi?id=176059

Patch by Youenn Fablet <​youenn@apple.com> on 2017-08-29
Reviewed by Chris Dumez.

Covered by existing tests.

• Modules/cache/CacheStorage.cpp:

(WebCore::CacheStorage::retrieveCaches):

2:55 PM Changeset in webkit [221317] by sbarati@apple.com

• 6 edits in trunk/Source/JavaScriptCore

Throwing an exception in the DFG/FTL should not be a jettison-able OSR exit
​https://bugs.webkit.org/show_bug.cgi?id=176060

Reviewed by Michael Saboff.

OSR exitting when we throw an exception is expected behavior. We should
not count these exits towards our jettison OSR exit threshold.

• bytecode/ExitKind.cpp:

(JSC::exitKindToString):
(JSC::exitKindMayJettison):

• bytecode/ExitKind.h:
• dfg/DFGSpeculativeJIT32_64.cpp:

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

• dfg/DFGSpeculativeJIT64.cpp:

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

• ftl/FTLLowerDFGToB3.cpp:

(JSC::FTL::DFG::LowerDFGToB3::compileThrow):

2:41 PM Changeset in webkit [221316] by Adrian Perez de Castro

• 2 edits in trunk/Tools

[WPE] Avoid using autogen for JHBuild modules, trying to fix the EWS

Unreviewed build fix.

This also makes the JHBuild process faster, because there are less
modules which will trigger slow, ancient Perl and M4 based tooling.

• wpe/jhbuild.modules:

2:39 PM Changeset in webkit [221315] by commit-queue@webkit.org

• 14 edits in trunk/Source

Setting the cache storage engine root path according the session WebsiteDataStore
​https://bugs.webkit.org/show_bug.cgi?id=176055

Patch by Youenn Fablet <​youenn@apple.com> on 2017-08-29
Reviewed by Alex Christensen.

Source/WebCore:

No test right now as the engine is not making any read/write operations.

• platform/network/NetworkStorageSession.h:

(WebCore::NetworkStorageSession::cacheStorageDirectory const):
(WebCore::NetworkStorageSession::setCacheStorageDirectory):

Source/WebKit:

Adding disk read/write capacities to CacheStorage engine.
This is used to store per-origin cache names in a file.
Making Engine a thread safe refcounted object so that it does read/write in a background thread.

Setting the root path of the engine to the directory set for the session in use according the WebsiteDataStore.
For each WebsiteDataStore, a path and a sandbox extension is passed to the network process if persistency of the cache storage is wanted.
Ephemeral sessions will set the path of the engine to null so that no read/write access is performed.

• NetworkProcess/NetworkConnectionToWebProcess.cpp:

(WebKit::NetworkConnectionToWebProcess::ensureLegacyPrivateBrowsingSession):

• NetworkProcess/NetworkProcess.cpp:

(WebKit::NetworkProcess::initializeNetworkProcess):
(WebKit::NetworkProcess::cacheStorageDirectory const):

• NetworkProcess/NetworkProcess.h:
• NetworkProcess/cache/CacheStorageEngine.cpp:

(WebKit::CacheStorage::Engine::from):
(WebKit::CacheStorage::Engine::defaultEngine):
(WebKit::CacheStorage::Engine::Engine):
(WebKit::CacheStorage::Engine::writeFile):

• NetworkProcess/cache/CacheStorageEngine.h:

(WebKit::CacheStorage::Engine::create):
(WebKit::CacheStorage::Engine::shouldPersist const):

• NetworkProcess/cocoa/NetworkProcessCocoa.mm:

(WebKit::NetworkProcess::platformInitializeNetworkProcessCocoa):

• NetworkProcess/mac/RemoteNetworkingContext.mm:

(WebKit::RemoteNetworkingContext::ensureWebsiteDataStoreSession):

• Shared/WebsiteDataStoreParameters.cpp:

(WebKit::WebsiteDataStoreParameters::encode const):
(WebKit::WebsiteDataStoreParameters::decode):

• Shared/WebsiteDataStoreParameters.h:
• UIProcess/WebsiteData/Cocoa/WebsiteDataStoreCocoa.mm:

(WebKit::WebsiteDataStore::parameters):

2:35 PM Changeset in webkit [221314] by jmarcell@apple.com

• 7 edits in branches/safari-604-branch/Source

Versioning.

2:32 PM Changeset in webkit [221313] by jmarcell@apple.com

• 1 copy in tags/Safari-604.2.6

Tag Safari-604.2.6.

2:28 PM Changeset in webkit [221312] by rniwa@webkit.org

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

Build fix. OS X "revision" can have a space.

• public/include/commit-log-fetcher.php:

2:18 PM Changeset in webkit [221311] by rniwa@webkit.org

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

Make it possible to specify A/B testing revision with a partial hash
​https://bugs.webkit.org/show_bug.cgi?id=176047

Rubber-stamped by Chris Dumez.

Added the support for specifying a partial hash in A/B testing instead of the full hash.

• public/include/commit-log-fetcher.php:

(CommitLogFetcher::find_commit_id_by_revision): Extracted from associate-commit.php.

• public/privileged-api/associate-commit.php:

(main):

• public/privileged-api/create-test-group.php:

(main): Use find_commit_id_by_revision here to support scheduling an A/B testing with a partial hash.

• server-tests/privileged-api-create-test-group-tests.js:

(createAnalysisTask): Make it possible to customize revision string in some test cases.

• server-tests/resources/test-server.js:

(TestServer.prototype._stopApache): Fixed the bug that cleanup step always fails whenever the test file
runs more than 8s.

2:00 PM Changeset in webkit [221310] by Matt Lewis

• 2 edits in trunk/LayoutTests

Rebaselined http/tests/websocket/tests/hybi/secure-cookie-secure-connection.pl for ios-11.

Unreviewed test gardening.

• http/tests/websocket/tests/hybi/secure-cookie-secure-connection-expected.txt:

1:53 PM Changeset in webkit [221309] by don.olmstead@sony.com

• 12 edits in trunk

[CMake] Use find_package for libxml2 and libxstl
​https://bugs.webkit.org/show_bug.cgi?id=176042

Reviewed by Konstantin Tokarev.

.:

• Source/cmake/OptionsAppleWin.cmake:
• Source/cmake/OptionsWinCairo.cmake:

Source/WebCore:

No new tests. No change in behavior.

• CMakeLists.txt:
• PlatformGTK.cmake:
• PlatformWPE.cmake:
• PlatformWinCairo.cmake:

Source/WebKitLegacy:

• PlatformWin.cmake:

Tools:

• TestWebKitAPI/PlatformWin.cmake:

12:25 PM Changeset in webkit [221308] by Nikita Vasilyev

• 8 edits in trunk

Web Inspector: Optimize View.prototype.removeSubview
​https://bugs.webkit.org/show_bug.cgi?id=176041

Reviewed by Matt Baker.

Source/WebInspectorUI:

Look up a subview in an array only once, not twice.

• UserInterface/Base/Utilities.js:

(Array.prototype.removeAll):
(Array.prototype.remove):
Split Array.propotype.remove(value, onlyFirst) into Array.propotype.removeAll(value) and
Array.propotype.remove(value).

• UserInterface/Controllers/DebuggerManager.js:

(WI.DebuggerManager.prototype._debuggerBreakpointOptions):

• UserInterface/Views/ContentViewContainer.js:

(WI.ContentViewContainer.prototype._clearTombstonesForContentView):
(WI.ContentViewContainer.prototype._disassociateFromContentView):

• UserInterface/Views/View.js:

(WI.View.prototype.removeSubview):

LayoutTests:

Split Array.prototype.remove(value, onlyFirst) into Array.prototype.removeAll(value) and Array.prototype.remove(value).

• inspector/unit-tests/array-utilities-expected.txt:
• inspector/unit-tests/array-utilities.html:

11:34 AM Changeset in webkit [221307] by pvollan@apple.com

• 2 edits in trunk/Tools

[Win] A WTF test is timing out in run-api-tests.
​https://bugs.webkit.org/show_bug.cgi?id=176056

Reviewed by Brent Fulgham.

The test SynchronizedFixedQueue.ProduceOnly relies on that putting a thread to sleep for
1ms will wake it up before another thread which is put to sleep for 10ms. I believe this
is not neccessarily true on Windows where the thread scheduling does not have millisecond
accuracy. Fix the test timeout by adjusting the sleep time slightly.

• TestWebKitAPI/Tests/WTF/SynchronizedFixedQueue.cpp:

(TestWebKitAPI::ToUpperConverter::startProducing):

11:08 AM Changeset in webkit [221306] by dbates@webkit.org

• 2 edits in trunk/Source/WebCore

CacheQueryOptions::isolatedCopy() copies the cache name twice
​https://bugs.webkit.org/show_bug.cgi?id=175974

Reviewed by Youenn Fablet.

Currently CacheQueryOptions has a user-defined constructor that calls String.isolatedCopy()
on the passed cache name. CacheQueryOptions::isolatedCopy() also calls String.isolatedCopy()
on the cache name before passing the result to the user-defined constructor; => we malloc
and copy the cache name twice. Ideally we would remove the user-defined constructors and
have callers use aggregate initializer syntax to instantiate a CacheQueryOptions. Unfortunately
we cannot do this until we upgrade from Visual Studio 2015 to Visual Studio 2017 as the former
does not support non-static data member initializers (NSDMI) for aggregates and CacheQueryOptions
has some. Therefore we modify the user-defined, non-default, constructor to take a String&&
and conditionally compile the the constructors when building with compilers that do not
support NSDMI for aggregates.

• Modules/cache/CacheQueryOptions.h:

(WebCore::CacheQueryOptions::CacheQueryOptions):

10:43 AM Changeset in webkit [221305] by commit-queue@webkit.org

• 9 edits in trunk

CanvasCaptureMediaStreamTrack clone is not a CanvasCaptureMediaStreamTrack
​https://bugs.webkit.org/show_bug.cgi?id=176036

Patch by Youenn Fablet <​youenn@apple.com> on 2017-08-29
Reviewed by Eric Carlson.

Source/WebCore:

Covered by existing tests.

Instead of creating a MediaStreamTrack clone when cloning a CanvasCaptureMediaStreamTrack,
we are now creating a CanvasCaptureMediaStreamTrack.
This allows exposing the expected methods and having the expected functionality.

• Modules/mediastream/CanvasCaptureMediaStreamTrack.cpp:

(WebCore::CanvasCaptureMediaStreamTrack::clone):

• Modules/mediastream/CanvasCaptureMediaStreamTrack.h:
• Modules/mediastream/MediaStreamTrack.cpp:

(WebCore::MediaStreamTrack::clone):

• Modules/mediastream/MediaStreamTrack.h:

LayoutTests:

• fast/mediacapturefromelement/CanvasCaptureMediaStream-clone-track-expected.txt:
• fast/mediastream/captureStream/canvas2d.html:
• fast/mediastream/captureStream/canvas2d-expected.txt:

10:40 AM Changeset in webkit [221304] by pvollan@apple.com

• 2 edits in trunk/LayoutTests

Skip editing/pasteboard/paste-empty-startcontainer.html on Windows.
It is skipped on all other platforms.

Unreviewed test gardening.

• platform/win/TestExpectations:

10:35 AM Changeset in webkit [221303] by Matt Lewis

• 2 edits in trunk/LayoutTests

Marked fast/events/mouse-cursor-no-mousemove.html as flaky.
​https://bugs.webkit.org/show_bug.cgi?id=169238

Unreviewed test gardening.

• platform/mac-wk2/TestExpectations:

10:00 AM Changeset in webkit [221302] by Chris Dumez

• 30 edits
  8 copies
  2 adds in trunk

Add initial support for dataTransferItem.webkitGetAsEntry()
​https://bugs.webkit.org/show_bug.cgi?id=176038
<rdar://problem/34121095>

Reviewed by Wenson Hsieh.

Source/JavaScriptCore:

Add CommonIdentifier needed by [EnabledAtRuntime].

• runtime/CommonIdentifiers.h:

Source/WebCore:

Add initial support for dataTransferItem.webkitGetAsEntry() as per:

• ​https://wicg.github.io/entries-api/#html-data

The method works as expected and returns a FileSystemEntry. The entry
can be either a file or a directory. All of the attributes exposed on
the FileSystemEntry return correct values.

What remains to be implemented for this new API to actually be useful:

• directory.getFile() / directory.getDirectory()
• file.file()
• entry.getParent()
• directoryReader.readEntries()

Test: editing/pasteboard/datatransfer-items-drop-getAsEntry.html

• CMakeLists.txt:
• DerivedSources.make:
• WebCore.xcodeproj/project.pbxproj:
• bindings/js/JSBindingsAllInOne.cpp:

Add new files to projects.

• Modules/entriesapi/DOMFileSystem.cpp:

(WebCore::DOMFileSystem::DOMFileSystem):

• Modules/entriesapi/DOMFileSystem.h:

(WebCore::DOMFileSystem::create):
DOMFileSystem factory now takes a name parameter. This name is a UUID, as suggested in
the specification and implemented by Firefox.

• Modules/entriesapi/ErrorCallback.cpp: Added.
• Modules/entriesapi/ErrorCallback.h:

Add scheduleCallback() convenience method as we often need to call this callback
asynchronously.

• Modules/entriesapi/FileSystemDirectoryEntry.cpp:

(WebCore::FileSystemDirectoryEntry::FileSystemDirectoryEntry):
(WebCore::FileSystemDirectoryEntry::createReader):
(WebCore::FileSystemDirectoryEntry::getFile):
(WebCore::FileSystemDirectoryEntry::getDirectory):

• Modules/entriesapi/FileSystemDirectoryEntry.h:
• Add implementation for createReader().
• Update getFile() / getDirectory() to call the error callback for now as they are unimplemented.
• Have the factory take the virtual path of the entry so that it can be exposed via the fullPath attribute in IDL.

• Modules/entriesapi/FileSystemDirectoryEntry.idl:

Expose createReader() now that we have stubs for FileSystemDirectoryReader.

• Modules/entriesapi/FileSystemDirectoryReader.cpp: Added.
• Modules/entriesapi/FileSystemDirectoryReader.h: Added.
• Modules/entriesapi/FileSystemDirectoryReader.idl: Added.
• Modules/entriesapi/FileSystemEntriesCallback.cpp: Added.
• Modules/entriesapi/FileSystemEntriesCallback.h: Added.
• Modules/entriesapi/FileSystemEntriesCallback.idl: Added.

Land stubs for FileSystemDirectoryReader and its FileSystemEntriesCallback.

• Modules/entriesapi/FileSystemEntry.cpp:

(WebCore::FileSystemEntry::FileSystemEntry):

• Modules/entriesapi/FileSystemEntry.h:

(WebCore::FileSystemEntry::virtualPath const):

• Modules/entriesapi/FileSystemEntry.idl:

Update constructor to take a virtual path, which is exposed via fullPath
attribute in IDL.

• Modules/entriesapi/FileSystemFileEntry.cpp:

(WebCore::FileSystemFileEntry::FileSystemFileEntry):
(WebCore::FileSystemFileEntry::file):

• Modules/entriesapi/FileSystemFileEntry.h:

(isType):

• Modules/entriesapi/FileSystemFileEntry.idl:
• Update file() to call the error callback for now as they are unimplemented.
• Have the factory take the virtual path of the entry so that it can be exposed via the fullPath attribute in IDL.

• bindings/js/JSFileSystemEntryCustom.cpp: Added.

(WebCore::toJSNewlyCreated):
(WebCore::toJS):
Add custom implementation of FileSystemEntry's toJS() function. This is
needed because FileSystemEntry has 2 subclasses and we need to type check
at runtime to construct the right wrapper.

• dom/DataTransferItem.cpp:

(WebCore::DataTransferItem::getAsEntry const):

• dom/DataTransferItem.h:
• dom/DataTransferItem.idl:

Add initial implementation for getAsEntry(), currently exposed as
webkitGetAsEntry().

• dom/DataTransferItemList.cpp:

(WebCore::DataTransferItemList::ensureItems const):
Construct a DataTransferItem if the file is a directory now that we
support getAsEntry().

• fileapi/File.cpp:

(WebCore::File::isDirectory const):

• fileapi/File.h:

Add convenience method to query if a File is a directory.

Source/WebKit:

Make dataTransfer.items an experimental feature so that it can be easily
turned on from the Develop menu, for testing convenience.

• Shared/WebPreferencesDefinitions.h:

LayoutTests:

• editing/pasteboard/datatransfer-items-drop-getAsEntry-expected.txt: Added.
• editing/pasteboard/datatransfer-items-drop-getAsEntry.html: Added.

Add layout test coverage.

• http/wpt/entries-api/interfaces-expected.txt:

Rebaseline test now that we support more of the API.

• platform/wk2/TestExpectations:

Skip new test on WK2 because it relies on beginDragWithFiles().

10:00 AM Changeset in webkit [221301] by eric.carlson@apple.com

• 2 edits in trunk/Tools

Unreviewed, update my watchlist filters.

• Scripts/webkitpy/common/config/watchlist:

9:52 AM Changeset in webkit [221300] by Matt Lewis

• 2 edits in trunk/LayoutTests

Marked media/video-controls-no-display-with-text-track.html and media/video-controls-in-media-document.html as flaky.
​https://bugs.webkit.org/show_bug.cgi?id=176054

Unreviewed test gardening.

• platform/mac-wk2/TestExpectations:

8:21 AM Changeset in webkit [221299] by pvollan@apple.com

• 2 edits in trunk/LayoutTests

Rebaseline fast/doctypes/002.html after r220858.

Unreviewed test gardening.

• platform/win/fast/doctypes/002-expected.txt:

7:57 AM Changeset in webkit [221298] by pvollan@apple.com

• 2 edits in trunk/LayoutTests

Rebaseline fast/lists/marker-before-empty-inline.html after r220858.

Unreviewed test gardening.

• platform/win/fast/lists/marker-before-empty-inline-expected.txt:

7:33 AM Changeset in webkit [221297] by pvollan@apple.com

• 2 edits in trunk/LayoutTests

Rebaseline css2.1/t0805-c5520-brdr-b-01-e.html after r220858.

Unreviewed test gardening.

• platform/win/css2.1/t0805-c5520-brdr-b-01-e-expected.txt:

7:20 AM Changeset in webkit [221296] by commit-queue@webkit.org

• 2 edits in trunk/LayoutTests

[GTK] Remove some obsolete expectations.
​https://bugs.webkit.org/show_bug.cgi?id=176051

Unreviewed test gardening.

Patch by Ms2ger <​Ms2ger@igalia.com> on 2017-08-29

• platform/gtk/TestExpectations:
  • fast/animation/request-animation-frame-throttling-detached-iframe.html: passing since r215259.
  • fast/animation/request-animation-frame-throttling-lowPowerMode.html: passing since r215259.
  • fast/events/drag-parent-node.html: passing since r217670-r217679.
  • fast/events/keydown-leftright-keys.html: passing since r210504.
  • fast/events/selectionchange-iframe.html: passing since r205908-r205913.

7:01 AM Changeset in webkit [221295] by Carlos Garcia Campos

• 3 edits in trunk/Tools

Unreviewed. REGRESSION(r221219): Fix unit tests using resources from WebKit directory.

The WebKit2 directory was renamed as WebKit in r221219, but neither run-gtk-tests script nor
Test::getResourcesDir() were updated to use the new name.

• Scripts/run-gtk-tests:

(TestRunner._setup_testing_environment):

• TestWebKitAPI/glib/WebKitGLib/TestMain.h:

(Test::getResourcesDir):

4:30 AM Changeset in webkit [221294] by zandobersek@gmail.com

• 13 edits in trunk/Source/WebCore

[WebCrypto] Push WorkQueue dispatches for HKDF, HMAC, PBKDF2 algorithms into shared code
​https://bugs.webkit.org/show_bug.cgi?id=175620

Reviewed by Jiewen Tan.

Push the WorkQueue dispatch code and other code duplicated between the
Mac and libgcrypt implementations of Web Crypto into the shared layer.
This patch focuses on the HKDF, HMAC and PBKDF2 algorithms.

Functions with platform-specific implementations that are invoked from
these asynchronous dispatches are made static and return an ExceptionOr
value. CryptoAlgorithmParameters objects are passed through non-const
references because data getters could lazily construct the underlying
Vector objects. CryptoKey objects are passed through const references.
Implementations can then manually retrieve and further validate any key
or parameter data, as required for that specific implementation. Input
data is passed through const references to the original Vector objects.

No new tests -- no changes in behavior that's covered by existing tests.

• crypto/algorithms/CryptoAlgorithmHKDF.cpp:

(WebCore::CryptoAlgorithmHKDF::deriveBits):

• crypto/algorithms/CryptoAlgorithmHKDF.h:
• crypto/algorithms/CryptoAlgorithmHMAC.cpp:

(WebCore::CryptoAlgorithmHMAC::sign):
(WebCore::CryptoAlgorithmHMAC::verify):

• crypto/algorithms/CryptoAlgorithmHMAC.h:
• crypto/algorithms/CryptoAlgorithmPBKDF2.cpp:

(WebCore::CryptoAlgorithmPBKDF2::deriveBits):

• crypto/algorithms/CryptoAlgorithmPBKDF2.h:
• crypto/gcrypt/CryptoAlgorithmHKDFGCrypt.cpp:

(WebCore::CryptoAlgorithmHKDF::platformDeriveBits):

• crypto/gcrypt/CryptoAlgorithmHMACGCrypt.cpp:

(WebCore::CryptoAlgorithmHMAC::platformSign):
(WebCore::CryptoAlgorithmHMAC::platformVerify):

• crypto/gcrypt/CryptoAlgorithmPBKDF2GCrypt.cpp:

(WebCore::CryptoAlgorithmPBKDF2::platformDeriveBits):

• crypto/mac/CryptoAlgorithmHKDFMac.cpp:

(WebCore::CryptoAlgorithmHKDF::platformDeriveBits):

• crypto/mac/CryptoAlgorithmHMACMac.cpp:

(WebCore::CryptoAlgorithmHMAC::platformSign):
(WebCore::CryptoAlgorithmHMAC::platformVerify):

• crypto/mac/CryptoAlgorithmPBKDF2Mac.cpp:

(WebCore::CryptoAlgorithmPBKDF2::platformDeriveBits):

4:29 AM Changeset in webkit [221293] by zandobersek@gmail.com

• 10 edits in trunk/Source/WebCore

[WebCrypto] Push WorkQueue dispatches for EC algorithms into shared code
​https://bugs.webkit.org/show_bug.cgi?id=175619

Reviewed by Jiewen Tan.

Push the WorkQueue dispatch code duplicated between the Mac and libgcrypt
implementations of Web Crypto into the shared layer. This patch focuses on
the EC-based algorithms.

Functions with platform-specific implementations that are invoked from
these asynchronous dispatches are made static and return an ExceptionOr or
std::optional value. CryptoKey and CryptoAlgorithmParameters objects are
passed to those through const references. Input data is passed through
const references to the original Vector objects.

No new tests -- no changes in behavior that's covered by existing tests.

• crypto/algorithms/CryptoAlgorithmECDH.cpp:

(WebCore::CryptoAlgorithmECDH::deriveBits):

• crypto/algorithms/CryptoAlgorithmECDH.h:
• crypto/algorithms/CryptoAlgorithmECDSA.cpp:

(WebCore::CryptoAlgorithmECDSA::sign):
(WebCore::CryptoAlgorithmECDSA::verify):

• crypto/algorithms/CryptoAlgorithmECDSA.h:
• crypto/gcrypt/CryptoAlgorithmECDHGCrypt.cpp:

(WebCore::CryptoAlgorithmECDH::platformDeriveBits):

• crypto/gcrypt/CryptoAlgorithmECDSAGCrypt.cpp:

(WebCore::CryptoAlgorithmECDSA::platformSign):
(WebCore::CryptoAlgorithmECDSA::platformVerify):

• crypto/keys/CryptoKeyEC.h:
• crypto/mac/CryptoAlgorithmECDHMac.cpp:

(WebCore::CryptoAlgorithmECDH::platformDeriveBits):

• crypto/mac/CryptoAlgorithmECDSAMac.cpp:

(WebCore::CryptoAlgorithmECDSA::platformSign):
(WebCore::CryptoAlgorithmECDSA::platformVerify):

2:32 AM Changeset in webkit [221292] by svillar@igalia.com

• 5 edits
  8 adds in trunk

[SVG] Leak in SVGAnimatedListPropertyTearOff
​https://bugs.webkit.org/show_bug.cgi?id=172545

Reviewed by Darin Adler.

Source/WebCore:

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

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

In order to effectively break the cycle without freeing too many wrappers we should take two
measures:
1) Break the reference cycle by storing raw pointers in the m_wrappers Vector
2) Remove the ListItemTearOff which is being deleted (it notifies the animated property by
calling propertyWillBeDeleted) from the m_wrappers Vector.

This is a re-land of r219334 which caused early releases of custom data attribute objects
added to SVG elements (wkb.ug/175023).

Tests: svg/animations/animation-leak-list-property-instances.html

svg/dom/SVGAnimatedListPropertyTearOff-crash-2.html
svg/dom/SVGAnimatedListPropertyTearOff-crash.html
svg/dom/SVGAnimatedListPropertyTearOff-leak.html

• svg/properties/SVGAnimatedListPropertyTearOff.h:
• svg/properties/SVGListProperty.h:

(WebCore::SVGListProperty::getItemValuesAndWrappers):

• svg/properties/SVGListPropertyTearOff.h:

(WebCore::SVGListPropertyTearOff::removeItemFromList):

LayoutTests:

The list of new added tests includes the one for the original bug, a new test for the
regression and a couple of tests imported from Blink which verify that
SVGAnimatedListPropertyTearOff does not crash after the context element goes out of scope.

• svg/animations/animation-leak-list-property-instances-expected.txt: Added.
• svg/animations/animation-leak-list-property-instances.html: Added.
• svg/dom/SVGAnimatedListPropertyTearOff-crash-2-expected.txt: Added. Imported from Blink.
• svg/dom/SVGAnimatedListPropertyTearOff-crash-2.html: Added. Imported from Blink.
• svg/dom/SVGAnimatedListPropertyTearOff-crash-expected.txt: Added. Imported from Blink.
• svg/dom/SVGAnimatedListPropertyTearOff-crash.html: Added. Imported from Blink.
• svg/dom/SVGAnimatedListPropertyTearOff-leak-expected.txt: Added.
• svg/dom/SVGAnimatedListPropertyTearOff-leak.html: Added.

2:10 AM Changeset in webkit [221291] by aestes@apple.com

• 14 edits
  1 add in trunk/Source

[Mac] Upstream WKSetMetadataURL() from WebKitSystemInterface
​https://bugs.webkit.org/show_bug.cgi?id=176046

Reviewed by Alex Christensen.

Source/WebCore:

• platform/FileSystem.cpp:

(WebCore::setMetadataURL):

• platform/FileSystem.h:

(WebCore::setMetadataURL):

• platform/mac/FileSystemMac.mm:

(WebCore::setMetadataURL):

• platform/mac/WebCoreSystemInterface.h:
• platform/mac/WebCoreSystemInterface.mm:

Source/WebCore/PAL:

• PAL.xcodeproj/project.pbxproj:
• pal/spi/mac/MetadataSPI.h: Added.

Source/WebKit:

• UIProcess/mac/WebPageProxyMac.mm:

(WebKit::WebPageProxy::savePDFToTemporaryFolderAndOpenWithNativeApplicationRaw):

• WebProcess/WebCoreSupport/mac/WebSystemInterface.mm:

(InitWebCoreSystemInterface):

Source/WebKitLegacy/mac:

Removed the pthread logic from -_webkit_setMetadataURL:referrer:atPath: and replaced with
WebCore::setMetadataURL(), which uses dispatch_async() for the same purpose as the pthread.

• Misc/WebNSFileManagerExtras.mm:

(-[NSFileManager _webkit_setMetadataURL:referrer:atPath:]):
(setMetaData): Deleted.

• WebCoreSupport/WebSystemInterface.mm:

(InitWebCoreSystemInterface):

12:46 AM Changeset in webkit [221290] by zandobersek@gmail.com

• 2 edits in trunk/Source/WebCore

Unreviewed WPE debug build fix.

• platform/graphics/gstreamer/mse/AppendPipeline.cpp:

(WebCore::AppendPipeline::dispatchPendingDecryptionStructure):
AppendState is an enum class, enum values have to be accessed
using scope resolution.

Aug 28, 2017:

9:33 PM Changeset in webkit [221289] by Brent Fulgham

• 3 edits in trunk/Source/WebKitLegacy/win

Unreviewed build fix #2 after r221275.

• WebCoreSupport/WebPlatformStrategies.cpp:

(WebPlatformStrategies::cookieRequestHeaderFieldValue): Correct return type and arguments
to match new API.

• WebCoreSupport/WebPlatformStrategies.h: Update signatures.

7:59 PM Changeset in webkit [221288] by eric.carlson@apple.com

• 4 edits in trunk

Logger should use makeString instead of String::format
​https://bugs.webkit.org/show_bug.cgi?id=176035

Reviewed by Jer Noble.

Source/WebCore/PAL:

• pal/Logger.h:

(PAL::LogArgument::toString):
(PAL::Logger::logAlways):
(PAL::Logger::error):
(PAL::Logger::warning):
(PAL::Logger::notice):
(PAL::Logger::info):
(PAL::Logger::debug):
(PAL::Logger::MethodAndPointer::MethodAndPointer):
(PAL::Logger::log):
(PAL::LogArgument<Logger::MethodAndPointer>::toString):

Tools:

• TestWebKitAPI/Tests/WebCore/Logging.cpp:

(TestWebKitAPI::TEST_F): Update test.

7:18 PM Changeset in webkit [221287] by pvollan@apple.com

• 2 edits in trunk/LayoutTests

Rebaseline accessibility/menu-list-crash2.html after r220930.

Unreviewed test gardening.

• platform/win/accessibility/menu-list-crash2-expected.txt:

5:55 PM Changeset in webkit [221286] by aestes@apple.com

• 15 edits
  2 adds in trunk/Source

[Cocoa] Upstream WKGetWebDefaultCFStringEncoding()
​https://bugs.webkit.org/show_bug.cgi?id=176039

Reviewed by Alex Christensen.

Source/WebCore:

• PlatformMac.cmake:
• WebCore.xcodeproj/project.pbxproj:
• platform/ios/WebCoreSystemInterfaceIOS.mm:
• platform/mac/WebCoreSystemInterface.h:
• platform/mac/WebCoreSystemInterface.mm:
• platform/text/TextEncodingRegistry.cpp:

(WebCore::defaultTextEncodingNameForSystemLanguage):

• platform/text/TextEncodingRegistry.h:
• platform/text/ios/TextEncodingRegistryIOS.mm: Added.

(WebCore::webDefaultCFStringEncoding):

• platform/text/mac/TextEncodingRegistryMac.mm: Added.

(WebCore::webDefaultCFStringEncoding):

Source/WebKit:

• WebProcess/WebCoreSupport/mac/WebSystemInterface.mm:

(InitWebCoreSystemInterface):

Source/WebKitLegacy/mac:

• WebCoreSupport/WebSystemInterface.mm:

(InitWebCoreSystemInterface):

• WebView/WebPreferences.mm:

(+[WebPreferences _systemCFStringEncoding]):

Source/WTF:

• wtf/spi/cf/CFStringSPI.h:

5:42 PM Changeset in webkit [221285] by Ryan Haddad

• 2 edits
  1 copy
  3 adds in trunk/LayoutTests

Rebaseline http/tests/websocket/tests/hybi/secure-cookie-secure-connection.pl.

Unreviewed test gardening.

• http/tests/websocket/tests/hybi/secure-cookie-secure-connection-expected.txt:
• platform/mac-highsierra/http/tests/websocket/tests/hybi/secure-cookie-secure-connection-expected.txt: Copied from LayoutTests/http/tests/websocket/tests/hybi/secure-cookie-secure-connection-expected.txt.

5:40 PM Changeset in webkit [221284] by Michael Catanzaro

• 5 edits in trunk/Tools

[GStreamer] The glvideoflip GStreamer element isn't available. Video mirroring and rotation functionalities are thus disabled.
​https://bugs.webkit.org/show_bug.cgi?id=175576

Reviewed by Carlos Alberto Lopez Perez.

Build graphene to enable the glvideoflip element.

• gstreamer/jhbuild.modules:
• gtk/jhbuild.modules:
• jhbuild/jhbuild-wrapper:
• wpe/jhbuild.modules:

5:27 PM Changeset in webkit [221283] by commit-queue@webkit.org

• 2 edits in trunk/Source/WebCore

REGRESSION(r220278): Web Inspector: ContextMenu items are not getting triggered
​https://bugs.webkit.org/show_bug.cgi?id=176034

Patch by Joseph Pecoraro <Joseph Pecoraro> on 2017-08-28
Reviewed by Devin Rousso.

• inspector/InspectorFrontendHost.cpp:

(WebCore::InspectorFrontendHost::showContextMenu):
Responses go through InspectorFrontendAPI not InspectorFrontendHost.

5:21 PM Changeset in webkit [221282] by Matt Lewis

• 2 edits in branches/safari-604.1.38.0-branch/LayoutTests

Unreviewed test gardening. <rdar://problem/33381269> <rdar://problem/33903582>

• platform/ios-11/TestExpectations:

4:56 PM Changeset in webkit [221281] by Megan Gardner

• 2 edits in trunk/Source/WebKit

Fix incorrect enum in atBoundaryOfGranularity call
​https://bugs.webkit.org/show_bug.cgi?id=176004

Reviewed by Wenson Hsieh and Enrica Casucci

Enum that was being used was incorrect, as that case in not covered in called function.
Resulted in a no-op. Should now have correct behavior around line boundaries.

Not possible to add tests for selections with velocity. Will attempt to add this behavior later.

• WebProcess/WebPage/ios/WebPageIOS.mm:

(WebKit::rangeAtWordBoundaryForPosition):

4:29 PM Changeset in webkit [221280] by Matt Lewis

• 2 edits in trunk/LayoutTests

Unreviewed test gardening. <rdar://problem/34051745>

• platform/ios-11/TestExpectations:

4:29 PM Changeset in webkit [221279] by commit-queue@webkit.org

• 1 edit
  1 add in trunk/Source/WebKit

[WinCairo] Add PlatformWin cmake for Webkit
​https://bugs.webkit.org/show_bug.cgi?id=176027

Patch by Stephan Szabo <​stephan.szabo@sony.com> on 2017-08-28
Reviewed by Brent Fulgham.

• PlatformWin.cmake: Added.

3:46 PM Changeset in webkit [221278] by Brent Fulgham

• 2 edits in trunk/Source/WebKitLegacy/win

Unreviewed build fix after r221275.

• WebCoreSupport/WebPlatformStrategies.cpp:

(WebPlatformStrategies::cookieRequestHeaderFieldValue): Correct return type and arguments
to match new API.

3:38 PM Changeset in webkit [221277] by commit-queue@webkit.org

• 9 edits in trunk

WebRTC MediaStream created without tracks does not update active state after tracks are added
​https://bugs.webkit.org/show_bug.cgi?id=175434

Patch by Youenn Fablet <​youenn@apple.com> on 2017-08-28
Reviewed by Eric Carlson.

Source/WebCore:

Covered by updated test.

Removing active/inactive events.
active attribute is still kept and is updated synchronously on every track change.

• Modules/mediastream/MediaStream.cpp:

(WebCore::MediaStream::MediaStream):
(WebCore::MediaStream::activeStatusChanged):
(WebCore::MediaStream::internalAddTrack):
(WebCore::MediaStream::updateActiveState):
(WebCore::MediaStream::hasPendingActivity const):
(WebCore::MediaStream::scheduleActiveStateChange): Deleted.
(WebCore::MediaStream::activityEventTimerFired): Deleted.

• Modules/mediastream/MediaStream.h:
• Modules/mediastream/MediaStream.idl:

LayoutTests:

• fast/mediacapturefromelement/CanvasCaptureMediaStream-clone-track-expected.txt:
• fast/mediacapturefromelement/CanvasCaptureMediaStream-creation-expected.txt:
• fast/mediastream/MediaStream-add-remove-tracks-expected.txt:
• fast/mediastream/MediaStream-add-remove-tracks.html:

3:23 PM Changeset in webkit [221276] by aestes@apple.com

• 20 edits in trunk/Source

[Cocoa] Upstream CFNetwork-related WebKitSystemInterface functions
​https://bugs.webkit.org/show_bug.cgi?id=176032

Reviewed by Alex Christensen.

Source/WebCore:

• platform/ios/WebCoreSystemInterfaceIOS.mm:
• platform/mac/WebCoreSystemInterface.h:
• platform/mac/WebCoreSystemInterface.mm:
• platform/network/NetworkStorageSession.h:
• platform/network/cf/NetworkStorageSessionCFNet.cpp:

(WebCore::NetworkStorageSession::switchToNewTestingSession):
(WebCore::NetworkStorageSession::ensureSession):

• platform/network/cf/ResourceRequest.h:
• platform/network/cocoa/NetworkStorageSessionCocoa.mm:

(WebCore::createPrivateStorageSession):

• platform/network/cocoa/ResourceRequestCocoa.mm:

(WebCore::ResourceRequest::setStorageSession):
(WebCore::copyRequestWithStorageSession):
(WebCore::cachedResponseForRequest):

• platform/network/mac/CookieJarMac.mm:

(WebCore::httpCookies):
(WebCore::setHTTPCookiesForURL):
(WebCore::deleteHTTPCookie):
(WebCore::deleteAllHTTPCookies):
(WebCore::setCookiesFromDOM):
(WebCore::httpCookieAcceptPolicy):
(WebCore::cookiesEnabled):
(WebCore::deleteCookie):
(WebCore::getHostnamesWithCookies):
(WebCore::deleteAllCookies):
(WebCore::deleteCookiesForHostnames):

• platform/network/mac/ResourceHandleMac.mm:

(WebCore::ResourceHandle::createNSURLConnection):

Source/WebCore/PAL:

• pal/spi/cf/CFNetworkSPI.h:

Source/WebKit:

• UIProcess/Cocoa/WebProcessPoolCocoa.mm:

(WebKit::privateBrowsingSession):

• WebProcess/WebCoreSupport/mac/WebSystemInterface.mm:

(InitWebCoreSystemInterface):

• WebProcess/WebPage/mac/WebPageMac.mm:

(WebKit::WebPage::platformHasLocalDataForURL):
(WebKit::cachedResponseForURL):

Source/WebKitLegacy/mac:

• WebCoreSupport/WebSystemInterface.mm:

(InitWebCoreSystemInterface):

• WebView/WebView.mm:

(-[WebView _cachedResponseForURL:]):

3:08 PM Changeset in webkit [221275] by Brent Fulgham

• 26 edits
  4 adds in trunk

Disable access to secure cookies if an HTTPS site loads mixed content (Part 2: Header Requests)
​https://bugs.webkit.org/show_bug.cgi?id=175992
<rdar://problem/34086613>

Reviewed by Daniel Bates.

Source/WebCore:

The original work in Bug 157053 did not properly handle the case of websockets. This patch completes
the changes to secure cookie handling to make sure websockets are also protected.

Tests: http/tests/websocket/tests/hybi/secure-cookie-insecure-connection.pl

http/tests/websocket/tests/hybi/secure-cookie-secure-connection.pl

• Modules/websockets/WebSocketChannel.cpp:

(WebCore::WebSocketChannel::clientHandshakeRequest): Remove 'const' declaration so we can work with
a mutable Document object. This allows us to mark the Document as having accessed secure cookies.
(WebCore::WebSocketChannel::clientHandshakeRequest const): Deleted.

• Modules/websockets/WebSocketChannel.h:
• Modules/websockets/WebSocketHandshake.cpp:

(WebCore::WebSocketHandshake::clientHandshakeMessage): Ditto.
(WebCore::WebSocketHandshake::clientHandshakeRequest): Ditto.
(WebCore::WebSocketHandshake::clientHandshakeMessage const): Deleted.
(WebCore::WebSocketHandshake::clientHandshakeRequest const): Deleted.

• Modules/websockets/WebSocketHandshake.h:
• loader/CookieJar.cpp:

(WebCore::cookies): Small naming cleanup.
(WebCore::cookieRequestHeaderFieldValue): Pass a flag indicating whether secure cookies should be
included in the result or not. Set the document flag indicating secure cookies were accessed (if
they were), and return the resulting cookie string.

• loader/CookieJar.h:
• platform/CookiesStrategy.h:
• platform/network/CacheValidation.cpp:

(WebCore::headerValueForVary): Revise to pass a flag indicating whether secure cookies should be included
or not.

• platform/network/PlatformCookieJar.h:
• platform/network/cf/CookieJarCFNet.cpp:

(WebCore::doesContainSecureCookies): Added helper method.
(WebCore::cookiesForDOM): Revise to use new helper function.
(WebCore::cookieRequestHeaderFieldValue): Revise to accept a flag indicating if secure cookies should
be included in the result. Return a pair consisting of the cookie string, and a flag indicating whether
secure cookies were accessed or not.

• platform/network/curl/CookieJarCurl.cpp:

(WebCore::CookieJarCurlFileSystem::cookieRequestHeaderFieldValue): Ditto.
(WebCore::cookieRequestHeaderFieldValue): Ditto.

• platform/network/curl/CookieJarCurl.h:
• platform/network/mac/CookieJarMac.mm:

(WebCore::cookiesForSession): Ditto.
(WebCore::cookiesForDOM): Update for new 'cookiesForSession' signature.
(WebCore::cookieRequestHeaderFieldValue): Ditto.

• platform/network/soup/CookieJarSoup.cpp:

(WebCore::cookieRequestHeaderFieldValue): Revise to accept a flag indicating if secure cookies should
be included in the result. Return a pair consisting of the cookie string, and a flag indicating whether
secure cookies were accessed or not.

Source/WebKit:

• NetworkProcess/NetworkConnectionToWebProcess.cpp:

(WebKit::NetworkConnectionToWebProcess::cookiesForDOM): Use a better label than 'result'.
(WebKit::NetworkConnectionToWebProcess::cookieRequestHeaderFieldValue): Modify to accept a flag
indicating if secure cookies should be included, and return a pair containing the resulting
cookie string and a boolean indicating if secure cookies were accessed.

• NetworkProcess/NetworkConnectionToWebProcess.h:
• NetworkProcess/NetworkConnectionToWebProcess.messages.in: Ditto.
• Shared/mac/CookieStorageShim.mm:

(WebKit::webKitCookieStorageCopyRequestHeaderFieldsForURL): Ditto.

• WebProcess/WebCoreSupport/WebPlatformStrategies.cpp:

(WebKit::WebPlatformStrategies::cookieRequestHeaderFieldValue): Ditto.

• WebProcess/WebCoreSupport/WebPlatformStrategies.h:

Source/WebKitLegacy/mac:

• WebCoreSupport/WebPlatformStrategies.h:
• WebCoreSupport/WebPlatformStrategies.mm:

(WebPlatformStrategies::cookieRequestHeaderFieldValue): Modify to accept a flag indicating if secure
cookies should be included, and return a pair containing the resulting cookie string and a boolean
indicating if secure cookies were accessed.

LayoutTests:

• http/tests/websocket/tests/hybi/secure-cookie-insecure-connection-expected.txt: Added.
• http/tests/websocket/tests/hybi/secure-cookie-insecure-connection.pl: Added.
• http/tests/websocket/tests/hybi/secure-cookie-secure-connection-expected.txt: Added.
• http/tests/websocket/tests/hybi/secure-cookie-secure-connection.pl: Added.

2:57 PM Changeset in webkit [221274] by Ryan Haddad

• 2 edits in trunk/LayoutTests

Unreviewed, remove duplicated TestExpectation entry.

• platform/mac-highsierra-wk2/TestExpectations:

2:48 PM Changeset in webkit [221273] by jmarcell@apple.com

• 2 edits in branches/safari-604-branch/Source/WebKit

Cherry-pick r221161. rdar://problem/34113784

2:48 PM Changeset in webkit [221272] by jmarcell@apple.com

• 4 edits in branches/safari-604-branch

Cherry-pick r221153. rdar://problem/34113728

2:14 PM Changeset in webkit [221271] by pvollan@apple.com

• 2 edits in trunk/LayoutTests

Before r220970, these ssl tests were failing. Now they are timing out.

Unreviewed test gardening.

• platform/win/TestExpectations:

2:13 PM Changeset in webkit [221270] by Ryan Haddad

• 2 edits in trunk/LayoutTests

Mark tiled-drawing/scrolling/fast-scroll-iframe-latched-mainframe-with-handler.html as flaky.
​https://bugs.webkit.org/show_bug.cgi?id=172544

Unreviewed test gardening.

• platform/mac-wk2/TestExpectations:

1:57 PM Changeset in webkit [221269] by jmarcell@apple.com

• 3 edits in branches/safari-604-branch/Source/WebKit

Cherry-pick r221138. rdar://problem/34113574

1:11 PM Changeset in webkit [221268] by pvollan@apple.com

• 2 edits in trunk/Source/WebKitLegacy/win

[Win] The test http/tests/misc/policy-delegate-called-twice.html is crashing.
​https://bugs.webkit.org/show_bug.cgi?id=176031

Reviewed by Brent Fulgham.

When the function in the FramePolicyFunction parameter is called, it has already been moved
to an internal member by the setUpPolicyListener method, and is no longer valid. Invoke the
function by using the stored member instead.

• WebCoreSupport/WebFrameLoaderClient.cpp:

(WebFrameLoaderClient::dispatchDecidePolicyForResponse):
(WebFrameLoaderClient::dispatchDecidePolicyForNewWindowAction):
(WebFrameLoaderClient::dispatchDecidePolicyForNavigationAction):

1:10 PM Changeset in webkit [221267] by Matt Lewis

• 2 edits in trunk/LayoutTests

Marked svg/in-html/by-reference.html as flaky.
​https://bugs.webkit.org/show_bug.cgi?id=175781

Unreviewed test gardening.

• platform/mac-wk2/TestExpectations:

12:56 PM Changeset in webkit [221266] by commit-queue@webkit.org

• 2 edits in trunk/Source/WebInspectorUI

Web Inspector: Remove some unused DataGrid code
​https://bugs.webkit.org/show_bug.cgi?id=176029

Patch by Joseph Pecoraro <Joseph Pecoraro> on 2017-08-28
Reviewed by Matt Baker.

• UserInterface/Views/DataGrid.js:

(WI.DataGrid.prototype.get copyTextDelimiter):
(WI.DataGrid.prototype.set copyTextDelimiter):
(WI.DataGrid.prototype.resizerDragging):
(WI.DataGrid.prototype.columnWidthsMap): Deleted.
(WI.DataGrid.prototype.applyColumnWidthsMap): Deleted.
(WI.DataGrid.prototype.get resizeMethod): Deleted.
(WI.DataGrid.prototype.set resizeMethod): Deleted.

12:48 PM Changeset in webkit [221265] by pvollan@apple.com

• 2 edits in trunk/LayoutTests

Skip service workers tests on Windows.

Unreviewed test gardening.

• platform/win/TestExpectations:

12:05 PM Changeset in webkit [221264] by Matt Lewis

• 2 edits in trunk/LayoutTests

Fixed expectations for http/tests/loading/basic-auth-remove-credentials.html by removing double expectations

Unreviewed test gardening.

• platform/mac-wk2/TestExpectations:

11:40 AM Changeset in webkit [221263] by Ryan Haddad

• 3 edits in trunk/LayoutTests

Rebaseline js/dom/global-constructors-attributes.html after r221209.

Unreviewed test gardening..

• platform/mac-highsierra-wk1/js/dom/global-constructors-attributes-expected.txt:
• platform/mac-highsierra/js/dom/global-constructors-attributes-expected.txt:

11:40 AM WebKitIDL edited by Konstantin Tokarev

JSCustomSetter -> CustomSetter (diff)

11:26 AM WebKitIDL edited by Konstantin Tokarev

Remove uses of obsolete module MODULE_NAME { } blocks which were … (diff)

11:21 AM WebKitIDL edited by Konstantin Tokarev

Remove mentions of CPP bindings which are long gone (diff)

11:20 AM Changeset in webkit [221262] by achristensen@apple.com

• 4 edits in trunk/Source/WebKit

Add WKUIDelegatePrivate equivalent of WKPageUIClient's didExceedBackgroundResourceLimitWhileInForeground
​https://bugs.webkit.org/show_bug.cgi?id=176028
<rdar://problem/29270035>

Reviewed by Andy Estes.

An API test for this did not succeed with either the memory or CPU limit calls,
possibly because the app must be in the foreground for this call to be made.

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

(WebKit::UIDelegate::setDelegate):
(WebKit::toWKResourceLimit):
(WebKit::UIDelegate::UIClient::didExceedBackgroundResourceLimitWhileInForeground):

10:31 AM WebKitGTK/Gardening/Calendar edited by Michael Catanzaro

(diff)

10:30 AM WebKitGTK/Gardening/Calendar edited by Michael Catanzaro

(diff)

9:52 AM Changeset in webkit [221261] by commit-queue@webkit.org

• 5 edits
  1 delete in trunk/LayoutTests

Fix document-create-touch.html.
​https://bugs.webkit.org/show_bug.cgi?id=172813

Patch by Ms2ger <​Ms2ger@igalia.com> on 2017-08-28
Reviewed by Michael Catanzaro.

The document.createTouch() call is supposed to throw an exception (as
defined by WebIDL). It has been doing this since r203941, but this
wasn't noticed because the entire fast/events/touch folder is skipped
on mac, ios and win.

• fast/events/touch/document-create-touch-expected.txt:
• fast/events/touch/document-create-touch.html:
• platform/gtk/TestExpectations:
• platform/gtk/fast/events/touch/document-create-touch-expected.txt: Removed.
• platform/wpe/TestExpectations:

9:37 AM Changeset in webkit [221260] by clopez@igalia.com

• 3 edits
  1 move in trunk/Source/WebCore

[GTK] [WPE] Rename EventHandlerGlib to EventHandlerGLib
​https://bugs.webkit.org/show_bug.cgi?id=175864

Unreviewed cosmetic follow-up patch after r221075.

No new tests. No change in behavior.

• PlatformGTK.cmake:
• PlatformWPE.cmake:
• platform/glib/EventHandlerGLib.cpp: Renamed from Source/WebCore/platform/glib/EventHandlerGlib.cpp.

(WebCore::EventHandler::tabsToAllFormControls const):
(WebCore::EventHandler::focusDocumentView):
(WebCore::EventHandler::passWidgetMouseDownEventToWidget):
(WebCore::EventHandler::passMouseDownEventToWidget):
(WebCore::EventHandler::eventActivatedView const):
(WebCore::EventHandler::widgetDidHandleWheelEvent):
(WebCore::EventHandler::createDraggingDataTransfer const):
(WebCore::EventHandler::passMousePressEventToSubframe):
(WebCore::EventHandler::passMouseMoveEventToSubframe):
(WebCore::EventHandler::passMouseReleaseEventToSubframe):
(WebCore::EventHandler::accessKeyModifiers):
(WebCore::EventHandler::shouldTurnVerticalTicksIntoHorizontal const):

8:54 AM Changeset in webkit [221259] by Ryan Haddad

• 2 edits in branches/safari-604.1.38.0-branch/Tools

Cherry-pick r221117.

8:50 AM Changeset in webkit [221258] by Ryan Haddad

• 4 edits in trunk/LayoutTests

Rebaseline js/dom/global-constructors-attributes.html after r221209.

Unreviewed test gardening.

• platform/mac-elcapitan-wk2/js/dom/global-constructors-attributes-expected.txt:
• platform/mac-wk1/js/dom/global-constructors-attributes-expected.txt:
• platform/mac/js/dom/global-constructors-attributes-expected.txt:

8:29 AM Changeset in webkit [221257] by Michael Catanzaro

• 4 edits
  1 add in trunk/LayoutTests

Unreviewed GTK/WPE test gardening

• platform/gtk/TestExpectations:
• platform/gtk/compositing/iframes/compositing-for-scrollable-iframe-expected.txt: Added.
• platform/gtk/js/dom/global-constructors-attributes-expected.txt:
• platform/wpe/js/dom/global-constructors-attributes-expected.txt:

7:38 AM Changeset in webkit [221256] by Carlos Garcia Campos

• 10 edits in releases/WebKitGTK/webkit-2.18/Source/JavaScriptCore

Merge r221018 - We are using valueProfileForBytecodeOffset when there may not be a value profile
​https://bugs.webkit.org/show_bug.cgi?id=175812

Reviewed by Michael Saboff.

This patch uses the type system to aid the code around CodeBlock's ValueProfile
accessor methods. valueProfileForBytecodeOffset used to return ValueProfile*,
so there were callers of this that thought it could return nullptr when there
was no such ValueProfile. This was not the case, it always returned a non-null
pointer. This patch changes valueProfileForBytecodeOffset to return ValueProfile&
and adds a new tryGetValueProfileForBytecodeOffset method that returns ValueProfile*
and does the right thing if there is no such ValueProfile.

This patch also changes the other ValueProfile accessors on CodeBlock to
return ValueProfile& instead of ValueProfile*. Some callers handled the null
case unnecessarily, and using the type system to specify the result can't be
null removes these useless branches.

• bytecode/CodeBlock.cpp:

(JSC::CodeBlock::updateAllPredictionsAndCountLiveness):
(JSC::CodeBlock::dumpValueProfiles):
(JSC::CodeBlock::tryGetValueProfileForBytecodeOffset):
(JSC::CodeBlock::valueProfileForBytecodeOffset):
(JSC::CodeBlock::validate):

• bytecode/CodeBlock.h:

(JSC::CodeBlock::valueProfileForArgument):
(JSC::CodeBlock::valueProfile):
(JSC::CodeBlock::valueProfilePredictionForBytecodeOffset):
(JSC::CodeBlock::getFromAllValueProfiles):

• dfg/DFGByteCodeParser.cpp:

(JSC::DFG::ByteCodeParser::handleInlining):

• dfg/DFGGraph.cpp:

(JSC::DFG::Graph::methodOfGettingAValueProfileFor):

• dfg/DFGPredictionInjectionPhase.cpp:

(JSC::DFG::PredictionInjectionPhase::run):

• jit/JIT.h:
• jit/JITInlines.h:

(JSC::JIT::emitValueProfilingSite):

• profiler/ProfilerBytecodeSequence.cpp:

(JSC::Profiler::BytecodeSequence::BytecodeSequence):

• tools/HeapVerifier.cpp:

(JSC::HeapVerifier::validateJSCell):

6:55 AM Changeset in webkit [221255] by Carlos Garcia Campos

• 17 edits
  2 adds in trunk/Source

WebDriver: implement screen capture commands
​https://bugs.webkit.org/show_bug.cgi?id=174615

Reviewed by Brian Burg.

Source/WebDriver:

Implement takeScreenshot and takeElementScreenshot commands.

19. Screen Capture.

​https://w3c.github.io/webdriver/webdriver-spec.html#screen-capture

• CommandResult.cpp:

(WebDriver::CommandResult::CommandResult): Handle ScreenshotError protocol error.
(WebDriver::CommandResult::httpStatusCode const): Add UnableToCaptureScreen.
(WebDriver::CommandResult::errorString const): Ditto.

• CommandResult.h:
• Session.cpp:

(WebDriver::Session::takeScreenshot):

• Session.h:
• WebDriverService.cpp:

(WebDriver::WebDriverService::takeScreenshot):
(WebDriver::WebDriverService::takeElementScreenshot):

• WebDriverService.h:

Source/WebKit:

Extend takeScreenshot command to optionally take a screenshot of an element. When no element is provided, the
screenshot is taken from the page visible area.

• PlatformGTK.cmake: Add WebAutomationSessionCairo.cpp to compilation.
• PlatformWPE.cmake: Ditto.
• UIProcess/Automation/Automation.json: Add ScreenshotError and several optional parameters to takeScreenshot.
• UIProcess/Automation/WebAutomationSession.cpp:

(WebKit::WebAutomationSession::takeScreenshot): Receive optional frame, node and scrollIntoView that are
checked and passed to the web process.

• UIProcess/Automation/WebAutomationSession.h:
• UIProcess/Automation/cairo/WebAutomationSessionCairo.cpp: Added.

(WebKit::WebAutomationSession::platformGetBase64EncodedPNGData): Cairo implementation.

• UIProcess/Automation/gtk/WebAutomationSessionGtk.cpp:
• WebProcess/Automation/WebAutomationSessionProxy.cpp:

(WebKit::snapshotRectForScreenshot): Helper to get the rectangle to be used for a screenshot.
(WebKit::WebAutomationSessionProxy::takeScreenshot): If a node handle is provided take the snapshot using the
element rectangle, otherwise use the page visible content rectangle.

• WebProcess/Automation/WebAutomationSessionProxy.h:
• WebProcess/Automation/WebAutomationSessionProxy.messages.in: Update TakeSnapshot message.

6:39 AM Changeset in webkit [221254] by Carlos Garcia Campos

• 2 edits in releases/WebKitGTK/webkit-2.18/Source/WebCore

Merge r221008 - GLContext: zero-initialize the GLContext pointer in ThreadGlobalGLContext
​https://bugs.webkit.org/show_bug.cgi?id=175819

Reviewed by Xabier Rodriguez-Calvar.

• platform/graphics/GLContext.cpp: The ThreadGlobalGLContext object is

allocated on heap, so the embedded GLContext pointer can contain a
non-null value that can cause problems when e.g. checking for a current
GLContext on some specific thread on which a GLContext hasn't yet been
made current. Zero-initializing this pointer will avoid false positives
that can occur in these circumstances.

6:38 AM Changeset in webkit [221253] by Carlos Garcia Campos

• 3 edits
  1 move
  1 add in releases/WebKitGTK/webkit-2.18/Source/WebKit

Merge r221007 - [GTK][WPE] Rename StorageProcessMainGtk.cpp to StorageProcessMainGLib.cpp
​https://bugs.webkit.org/show_bug.cgi?id=175814

Reviewed by Gyuyoung Kim.

Both GTK+ and WPE ports already compile the StorageProcessMainGtk.cpp
file, but it should be renamed to StorageProcessMainGLib and moved into
the StorageProcess/glib/ directory.

• PlatformGTK.cmake:
• PlatformWPE.cmake:
• StorageProcess/glib/StorageProcessMainGLib.cpp: Renamed from Source/WebKit/StorageProcess/gtk/StorageProcessMainGtk.cpp.

6:30 AM Changeset in webkit [221252] by Carlos Garcia Campos

• 5 edits in releases/WebKitGTK/webkit-2.18/Source/WebCore

Merge r220988 - Cleanup TextPainter
​https://bugs.webkit.org/show_bug.cgi?id=175782

Reviewed by Myles C. Maxfield.

Remove redundant mention of "text" in the name of TextPainter member functions, add separate
setters for selection and non-selection shadows, and do some other minor cleanups.

• rendering/InlineTextBox.cpp:

(WebCore::InlineTextBox::paint): Update code for renamed functions.

• rendering/SimpleLineLayoutFunctions.cpp:

(WebCore::SimpleLineLayout::paintFlow): Ditto.

• rendering/TextPainter.cpp: Include ShadowData.h.

(WebCore::ShadowApplier::ShadowApplier): Use C++11 brace-initialization syntax.
(WebCore::ShadowApplier::isLastShadowIteration): Moved from TextPainter.h.
(WebCore::ShadowApplier::shadowIsCompletelyCoveredByText): Ditto.
(WebCore::TextPainter::paintTextOrEmphasisMarks): Renamed; formerly named drawTextOrEmphasisMarks.
(WebCore::TextPainter::paintTextWithShadows): Update code for renamed functions.
(WebCore::TextPainter::paintRange): Renamed; formerly named paintTextInRange.
(WebCore::TextPainter::paint): Renamed; formerly named paintText.
(WebCore::TextPainter::drawTextOrEmphasisMarks): Renamed to paintTextOrEmphasisMarks() to match
the naming convention of all the other paint functions in this class.
(WebCore::TextPainter::paintTextInRange): Renamed to paintRange.
(WebCore::TextPainter::paintText): Renamed to paint.

• rendering/TextPainter.h: Forward declare ShadowData.

(WebCore::TextPainter::setStyle): Renamed; formerly named setTextPaintStyle.
(WebCore::TextPainter::setSelectionStyle): Renamed; formerly named setSelectionPaintStyle.
(WebCore::TextPainter::setShadow): Added.
(WebCore::TextPainter::setSelectionShadow): Added.
(WebCore::TextPainter::setEmphasisMark): Renamed; formerly named addEmphasis.
(WebCore::TextPainter::setTextPaintStyle): Renamed to setStyle.
(WebCore::TextPainter::setSelectionPaintStyle): Renamed to setSelectionStyle.
(WebCore::TextPainter::addEmphasis): Renamed to setEmphasisMark.
(WebCore::TextPainter::addTextShadow): Split functionality into setShadow and setSelectionShadow.
(WebCore::ShadowApplier::isLastShadowIteration): Moved to TextPainter.cpp.
(WebCore::ShadowApplier::shadowIsCompletelyCoveredByText): Ditto.

6:27 AM Changeset in webkit [221251] by Carlos Garcia Campos

• 4 edits
  1 copy in releases/WebKitGTK/webkit-2.18/LayoutTests

Merge r220985 - Stop media/video-controls-toggling.html from timing out.
​https://bugs.webkit.org/show_bug.cgi?id=116266

Patch by Ms2ger <​Ms2ger@gmail.com> on 2017-08-21
Reviewed by Dean Jackson.

Use getComputedStyle to check the display property. Element::style only
contains properties set in the style attribute, but the display property
is set through CSS using the 'hidden' class.

• media/video-controls-toggling-expected.txt:
• media/video-controls-toggling.html:
• platform/gtk/TestExpectations:
• platform/gtk/media/video-controls-toggling-expected.txt: Copied from LayoutTests/media/video-controls-toggling-expected.txt. I'm not sure where the console message comes from, but it seems to be in the GTK-specific code.

6:22 AM Changeset in webkit [221250] by Carlos Garcia Campos

• 4 edits in releases/WebKitGTK/webkit-2.18

Merge r220982 - StringView could use a function to strip leading/trailing characters without allocation
​https://bugs.webkit.org/show_bug.cgi?id=175757

Reviewed by Darin Adler.

Source/WTF:

There are many places in WebCore/WebKit that we call functions like,
WebCore::stripLeadingAndTrailingHTMLSpaces, or String::stripWhiteSpace() only to use
the allocated String as a temporary for either another transformation or a comparison.
Now that we have StringView, we can avoid that extra allocation, by having returning a
StringView substring in these scenarios.

For instance, the check (from ScriptElement.cpp:287):

if (!stripLeadingAndTrailingHTMLSpaces(sourceURL).isEmpty()) {

...

}

currently allocates a string just to make this check. With a new
stripLeadingAndTrailingHTMLSpaces such as:

StringView stripLeadingAndTrailingHTMLSpaces(StringView stringView)
{

return stringView.stripLeadingAndTrailingMatchedCharacters([] (auto c) {

return isHTMLSpace(c);

});

}

We could instead have exact same code from ScriptElement.cpp now avoid an allocation.

• wtf/text/StringView.h:

(WTF::StringView::stripLeadingAndTrailingMatchedCharacters):

Tools:

• TestWebKitAPI/Tests/WTF/StringView.cpp:

Add tests for StringView::stripLeadingAndTrailingMatchedCharacters().

5:46 AM Changeset in webkit [221249] by Carlos Garcia Campos

• 2 edits in releases/WebKitGTK/webkit-2.18/Source/WebCore

Merge r220978 - [GTK][WPE][GSTREAMER_GL] Overriden virtual function not marked “override” in MediaPlayerPrivateGStreamerBase.h
​https://bugs.webkit.org/show_bug.cgi?id=175780

Reviewed by Konstantin Tokarev.

• platform/graphics/gstreamer/MediaPlayerPrivateGStreamerBase.h: Add missing "override" qualifier.

5:44 AM Changeset in webkit [221248] by Carlos Garcia Campos

• 5 edits in releases/WebKitGTK/webkit-2.18/LayoutTests

Merge r220967 - [GTK][WPE] Rebaseline tests with list item markers.
​https://bugs.webkit.org/show_bug.cgi?id=175771

Unreviewed test gardening.

The output was changed in r220858 and rebaselined for ios and mac only.
They're changes in render tree dumps that don't affect rendering.

Patch by Ms2ger <​Ms2ger@igalia.com> on 2017-08-21

• platform/gtk/css2.1/t0805-c5520-brdr-b-01-e-expected.txt:
• platform/gtk/fast/doctypes/002-expected.txt:
• platform/gtk/fast/lists/marker-before-empty-inline-expected.txt:
• platform/wpe/css2.1/t0805-c5520-brdr-b-01-e-expected.txt:

5:43 AM Changeset in webkit [221247] by Carlos Garcia Campos

• 2 edits in releases/WebKitGTK/webkit-2.18/Source/WebKit

Merge r220963 - MemoryCache::setCapacities assertion failure maxDeadBytes <= totalBytes
​https://bugs.webkit.org/show_bug.cgi?id=175571

Patch by Charlie Turner <​cturner@igalia.com> on 2017-08-21
Reviewed by Antti Koivisto.

• Shared/CacheModel.cpp:

(WebKit::calculateMemoryCacheSizes): Ensure cacheTotalCapacity is
set to a reasonable value even in low-memory environments.

5:32 AM Changeset in webkit [221246] by Carlos Garcia Campos

• 2 edits in trunk/Source/WebKit

Automation: takeScreenshot should use the visible content rect not the document rect
​https://bugs.webkit.org/show_bug.cgi?id=175665

Reviewed by Brian Burg.

According to the spec, we should get the toplevel browsing context document rectangle and take a screenshot of
it using the current viewport width and height. We are currently using the document size.

19. Screen Capture.

​https://w3c.github.io/webdriver/webdriver-spec.html#dfn-draw-a-bounding-box-from-the-framebuffer

• WebProcess/Automation/WebAutomationSessionProxy.cpp:

(WebKit::WebAutomationSessionProxy::takeScreenshot): Use FrameView::visibleContentRect().

5:07 AM Changeset in webkit [221245] by Carlos Garcia Campos

• 8 edits in releases/WebKitGTK/webkit-2.18/Source/WebCore

Merge r220934 - Followup (r220289): RenderImageResourceStyleImage code clean up
​https://bugs.webkit.org/show_bug.cgi?id=175444

Patch by Said Abou-Hallawa <​sabouhallawa@apple.com> on 2017-08-18
Reviewed by Darin Adler.

RenderImageResourceStyleImage may be created with a StyleImage of type
StyleGeneratedImage. It may also be associated with a CachedImage which
is loaded through a source URL. In this case, adding and removing m_renderer
as a client of the CachedImage will be done through
RenderImageResource::setCachedImage().

RenderImageResource::setCachedImage() is already called from
ImageLoader::updateRenderer() when the CachedImage finishes loading. This
call adds m_renderer to the clients of the CachedImage.
RenderImageResource::setCachedImage() will also be called from
RenderImageResourceStyleImage::shutdown() via RenderImageResource::shutdown()
to remove m_renderer from the clients of CachedImage by passing a null pointer.

• rendering/RenderImage.cpp:

(WebCore::RenderImage::styleWillChange):

• rendering/RenderImageResource.cpp:

(WebCore::RenderImageResource::initialize):
(WebCore::RenderImageResource::shutdown):
(WebCore::RenderImageResource::setCachedImage):
(WebCore::RenderImageResource::resetAnimation):
(WebCore::RenderImageResource::image const):
(WebCore::RenderImageResource::setContainerSizeForRenderer):
(WebCore::RenderImageResource::imageSize const):
(WebCore::RenderImageResource::~RenderImageResource): Deleted.
(WebCore::RenderImageResource::errorOccurred const): Deleted.
(WebCore::RenderImageResource::imageHasRelativeWidth const): Deleted.
(WebCore::RenderImageResource::imageHasRelativeHeight const): Deleted.
(WebCore::RenderImageResource::intrinsicSize const): Deleted.
(WebCore::RenderImageResource::getImageSize const): Deleted.

• rendering/RenderImageResource.h:

(WebCore::RenderImageResource::initialize):
(WebCore::RenderImageResource::renderer const):
(WebCore::RenderImageResource::errorOccurred const):
(WebCore::RenderImageResource::imageHasRelativeWidth const):
(WebCore::RenderImageResource::imageHasRelativeHeight const):
(WebCore::RenderImageResource::imageSize const):
(WebCore::RenderImageResource::intrinsicSize const):
(WebCore::RenderImageResource::imagePtr const):

• rendering/RenderImageResourceStyleImage.cpp:

(WebCore::RenderImageResourceStyleImage::initialize):
(WebCore::RenderImageResourceStyleImage::shutdown):
(WebCore::RenderImageResourceStyleImage::image const):
(WebCore::RenderImageResourceStyleImage::setContainerSizeForRenderer):
(WebCore::RenderImageResourceStyleImage::~RenderImageResourceStyleImage): Deleted.

• rendering/RenderImageResourceStyleImage.h:
• rendering/RenderSnapshottedPlugIn.cpp:

(WebCore::RenderSnapshottedPlugIn::RenderSnapshottedPlugIn):

• rendering/svg/RenderSVGImage.cpp:

(WebCore::RenderSVGImage::RenderSVGImage):

5:02 AM Changeset in webkit [221244] by Carlos Garcia Campos

• 3 edits in releases/WebKitGTK/webkit-2.18/Source/WebKit

Merge r220931 - REGRESSION (r220601): Crash when closing google doc after switching the order of tabs in safari
​https://bugs.webkit.org/show_bug.cgi?id=175721
<rdar://problem/33928369>

Reviewed by Geoffrey Garen.

Make sure WebProcess::markAllLayersVolatile() does not call WTFMove() multiple times
on the same completion handler. Use a RefCounter to hold on to the completion handler
and make sure the handler gets called when the RefCounter's value becomes 0.

• WebProcess/WebProcess.cpp:

(WebKit::WebProcess::cancelPrepareToSuspend):
(WebKit::WebProcess::markAllLayersVolatile):
(WebKit::WebProcess::cancelMarkAllLayersVolatile):

• WebProcess/WebProcess.h:

4:57 AM Changeset in webkit [221243] by Carlos Garcia Campos

• 8 edits
  2 adds in releases/WebKitGTK/webkit-2.18/Source/WebCore

Merge r220916 - Factor render tree mutation code from RenderListItem to RenderTreeUpdater
​https://bugs.webkit.org/show_bug.cgi?id=175718

Reviewed by Andreas Kling.

We already stopped doing layout time mutations. We can now move the code out too.

• WebCore.xcodeproj/project.pbxproj:
• rendering/RenderListItem.cpp:

(WebCore::isHTMLListElement):
(WebCore::getParentOfFirstLineBox): Deleted.
(WebCore::firstNonMarkerChild): Deleted.
(WebCore::RenderListItem::updateMarkerRenderer): Deleted.

Moved to RenderTreeUpdater::ListItem.

• rendering/RenderListItem.h:
• rendering/RenderListMarker.cpp:

(WebCore::RenderListMarker::willBeDestroyed):

• rendering/TextAutoSizing.cpp:

(WebCore::TextAutoSizingValue::adjustTextNodeSizes):

• style/RenderTreeUpdater.cpp:

(WebCore::RenderTreeUpdater::pushParent):
(WebCore::RenderTreeUpdater::popParent):
(WebCore::RenderTreeUpdater::updateBeforeDescendants):
(WebCore::RenderTreeUpdater::updateAfterDescendants):

Factor pre/post update into functions.

(WebCore::RenderTreeUpdater::updateBeforeOrAfterPseudoElement):

• style/RenderTreeUpdater.h:
• style/RenderTreeUpdaterListItem.cpp: Added.

Mutation functions move here.

(WebCore::getParentOfFirstLineBox):
(WebCore::firstNonMarkerChild):
(WebCore::RenderTreeUpdater::ListItem::updateMarker):

• style/RenderTreeUpdaterListItem.h: Added.

4:53 AM Changeset in webkit [221242] by Carlos Garcia Campos

• 4 edits in releases/WebKitGTK/webkit-2.18

Merge r220910 - [GTK] Show controls if a video element isn't allowed to play inline.
​https://bugs.webkit.org/show_bug.cgi?id=141705

Patch by Ms2ger <​Ms2ger@gmail.com> on 2017-08-18
Reviewed by Xabier Rodriguez-Calvar.

Source/WebCore:

Test: covered by media/video-fullscreeen-only-controls.html [sic].

• Modules/mediacontrols/mediaControlsGtk.js:

(ControllerGtk.prototype.shouldHaveControls):

LayoutTests:

• platform/gtk/TestExpectations: enable test for this case.

4:52 AM Changeset in webkit [221241] by Carlos Garcia Campos

• 7 edits in trunk/Source/WebDriver

WebDriver: implement cookies commands
​https://bugs.webkit.org/show_bug.cgi?id=174613

Reviewed by Brian Burg.

Add cookies commands.

16. Cookies.

​https://w3c.github.io/webdriver/webdriver-spec.html#cookies

• CommandResult.cpp:

(WebDriver::CommandResult::httpStatusCode const): Add NoSuchCookie error.
(WebDriver::CommandResult::errorString const): Ditto.

• CommandResult.h:
• Session.cpp:

(WebDriver::parseAutomationCookie): Parse JSON cookie object returned by automation and convert it to a Cookie struct.
(WebDriver::builtAutomationCookie): Build a JSON cookie object as expected by automation from a Cookie struct.
(WebDriver::serializeCookie): Serialize a Cookie struct into a JSON cookie object according to the WebDriver spec.
(WebDriver::Session::getAllCookies):
(WebDriver::Session::getNamedCookie):
(WebDriver::Session::addCookie):
(WebDriver::Session::deleteCookie):
(WebDriver::Session::deleteAllCookies):

• Session.h:
• WebDriverService.cpp:

(WebDriver::WebDriverService::getAllCookies):
(WebDriver::WebDriverService::getNamedCookie):
(WebDriver::deserializeCookie):
(WebDriver::WebDriverService::addCookie):
(WebDriver::WebDriverService::deleteCookie):
(WebDriver::WebDriverService::deleteAllCookies):

• WebDriverService.h:

4:46 AM Changeset in webkit [221240] by commit-queue@webkit.org

• 2 edits in trunk/Tools

Ensure that the alert function is called in TestWebExtensions.
​https://bugs.webkit.org/show_bug.cgi?id=175880

Patch by Ms2ger <​Ms2ger@gmail.com> on 2017-08-28
Reviewed by Carlos Garcia Campos.

The test relied on a bug in the HTMLDocument JavaScript bindings: when calling
document.open() with three arguments, it would call any function set as the
"open" property on the window object, rather than just the real Window::open()
method. This bug was fixed by the removal of custom bindings for HTMLDocument
in r218437. As a result, the test no longer called alert(), which caused it to
fail.

Instead, call the window.open() function directly.

• TestWebKitAPI/Tests/WebKitGLib/TestWebExtensions.cpp:

(testWebExtensionIsolatedWorld):

4:23 AM Changeset in webkit [221239] by Michael Catanzaro

• 2 edits
  1 delete in trunk/LayoutTests

New baseline for insecure-audio-video-in-main-frame.html should apply to all ports

The global baseline was never updated because the test has been failing on macOS.

• http/tests/security/mixedContent/insecure-audio-video-in-main-frame-expected.txt:
• platform/gtk/http/tests/security/mixedContent/insecure-audio-video-in-main-frame-expected.txt: Removed.

4:18 AM Changeset in webkit [221238] by Carlos Garcia Campos

• 3 edits in trunk/Source/WebKit

[GTK][WPE] ASSERTION FAILED: !isOpen() in WebKit::IconDatabase::~IconDatabase()
​https://bugs.webkit.org/show_bug.cgi?id=175719

Reviewed by Michael Catanzaro.

This is happening always when running /webkit2/WebKitFaviconDatabase/favicon-database-test in debug builds. The
last step we do is removing all icons, then the test finishes, which destroys the WebKitFaviconDatabase object
that closes the icon database on dispose. The problem is that removing all icons schedules a main thread
notification and IconDatabase is not considered closed until all main thread callbacks have been dispatched. This
is never going to happen in the test, because the main loop is no longer running at that point. I don't think
it's worth it to consider the database open while main thread callbacks are pending, they are just notifications
and the client is no longer insterested on them afer closing the database. I think it's bettter and simpler to
simply cancel the pending callbacks on database close. That ensures that isOpen() after close() is always
false. This patch adds a helper private class to schedule notifications to the main thread that can be cancelled
on database close. It also removes the didClose() notification because it was unused and because it's pointless
now that we know the database is closed after close().

• UIProcess/API/glib/IconDatabase.cpp:

(WebKit::IconDatabase::open): Mark the main thread notifier as active.
(WebKit::IconDatabase::close): Mark the main thread notifier as not active.
(WebKit::IconDatabase::IconDatabase): Remove m_mainThreadCallbackCount initialization.
(WebKit::IconDatabase::isOpen const): Do what isOpenBesidesMainThreadCallbacks() used to do.
(WebKit::IconDatabase::removeAllIconsOnThread): Remove the notification because it's currently unused.
(WebKit::IconDatabase::dispatchDidImportIconURLForPageURLOnMainThread): Use MainThreadNotifier.
(WebKit::IconDatabase::dispatchDidImportIconDataForPageURLOnMainThread): Ditto.
(WebKit::IconDatabase::dispatchDidFinishURLImportOnMainThread): Ditto.
(WebKit::IconDatabase::isOpenBesidesMainThreadCallbacks const): Deleted.
(WebKit::IconDatabase::checkClosedAfterMainThreadCallback): Deleted.
(WebKit::IconDatabase::dispatchDidRemoveAllIconsOnMainThread): Deleted.

• UIProcess/API/glib/IconDatabase.h:

(WebKit::IconDatabaseClient::didChangeIconForPageURL):
(WebKit::IconDatabaseClient::didFinishURLImport):
(WebKit::IconDatabase::MainThreadNotifier::MainThreadNotifier):
(WebKit::IconDatabase::MainThreadNotifier::setActive):
(WebKit::IconDatabase::MainThreadNotifier::notify):
(WebKit::IconDatabase::MainThreadNotifier::stop):
(WebKit::IconDatabase::MainThreadNotifier::timerFired):
(WebKit::IconDatabaseClient::didRemoveAllIcons): Deleted.
(WebKit::IconDatabaseClient::didClose): Deleted.

4:02 AM Changeset in webkit [221237] by commit-queue@webkit.org

• 2 edits
  2 adds in trunk/LayoutTests

[GTK] Rebaseline insecure-audio-video-in-main-frame.html.
​https://bugs.webkit.org/show_bug.cgi?id=142482

Unreviewed test gardening.

The test is failing due to a different console message. It seems better
to add a platform-specific baseline so we can catch actual failures.

Patch by Ms2ger <​Ms2ger@igalia.com> on 2017-08-28

• platform/gtk/TestExpectations:
• platform/gtk/http/tests/security/mixedContent/insecure-audio-video-in-main-frame-expected.txt: Added.

3:20 AM Changeset in webkit [221236] by commit-queue@webkit.org

• 2 edits in trunk/LayoutTests

[GTK] Enable userAgentShadowDOM tests.
​https://bugs.webkit.org/show_bug.cgi?id=176015

Unreviewed test gardening.

Patch by Ms2ger <​Ms2ger@igalia.com> on 2017-08-28

• platform/gtk/TestExpectations:

1:16 AM Changeset in webkit [221235] by zandobersek@gmail.com

• 21 edits in trunk/Source/WebCore

[WebCrypto] Push WorkQueue dispatches for AES algorithms into shared code
​https://bugs.webkit.org/show_bug.cgi?id=175539

Reviewed by Sam Weinig.

Push the WorkQueue dispatch code and other code duplicated between the
Mac and libgcrypt implementations of Web Crypto into the shared layer.
This patch focuses on the AES-based algorithms.

Functions with platform-specific implementations that are invoked from
these asynchronous dispatches are made static and return an ExceptionOr
value. CryptoAlgorithmParameters objects are passed through non-const
references because data getters could lazily construct the underlying
Vector objects. CryptoKey objects are passed through const references.
Implementations can then manually retrieve and further validate any key
or parameter data, as required for that specific implementation. Input
data is passed through const references to the original Vector objects.

No new tests -- no changes in behavior that's covered by existing tests.

• crypto/algorithms/CryptoAlgorithmAES_CBC.cpp:

(WebCore::CryptoAlgorithmAES_CBC::encrypt):
(WebCore::CryptoAlgorithmAES_CBC::decrypt):

• crypto/algorithms/CryptoAlgorithmAES_CBC.h:
• crypto/algorithms/CryptoAlgorithmAES_CFB.cpp:

(WebCore::CryptoAlgorithmAES_CFB::encrypt):
(WebCore::CryptoAlgorithmAES_CFB::decrypt):

• crypto/algorithms/CryptoAlgorithmAES_CFB.h:
• crypto/algorithms/CryptoAlgorithmAES_CTR.cpp:

(WebCore::CryptoAlgorithmAES_CTR::encrypt):
(WebCore::CryptoAlgorithmAES_CTR::decrypt):

• crypto/algorithms/CryptoAlgorithmAES_CTR.h:
• crypto/algorithms/CryptoAlgorithmAES_GCM.cpp:

(WebCore::CryptoAlgorithmAES_GCM::encrypt):
(WebCore::CryptoAlgorithmAES_GCM::decrypt):

• crypto/algorithms/CryptoAlgorithmAES_GCM.h:
• crypto/algorithms/CryptoAlgorithmAES_KW.cpp:

(WebCore::CryptoAlgorithmAES_KW::wrapKey):
(WebCore::CryptoAlgorithmAES_KW::unwrapKey):

• crypto/algorithms/CryptoAlgorithmAES_KW.h:
• crypto/gcrypt/CryptoAlgorithmAES_CBCGCrypt.cpp:

(WebCore::CryptoAlgorithmAES_CBC::platformEncrypt):
(WebCore::CryptoAlgorithmAES_CBC::platformDecrypt):

• crypto/gcrypt/CryptoAlgorithmAES_CFBGCrypt.cpp:

(WebCore::CryptoAlgorithmAES_CFB::platformEncrypt):
(WebCore::CryptoAlgorithmAES_CFB::platformDecrypt):

• crypto/gcrypt/CryptoAlgorithmAES_CTRGCrypt.cpp:

(WebCore::CryptoAlgorithmAES_CTR::platformEncrypt):
(WebCore::CryptoAlgorithmAES_CTR::platformDecrypt):

• crypto/gcrypt/CryptoAlgorithmAES_GCMGCrypt.cpp:

(WebCore::gcryptEncrypt):
(WebCore::gcryptDecrypt):
(WebCore::CryptoAlgorithmAES_GCM::platformEncrypt):
(WebCore::CryptoAlgorithmAES_GCM::platformDecrypt):

• crypto/gcrypt/CryptoAlgorithmAES_KWGCrypt.cpp:

(WebCore::CryptoAlgorithmAES_KW::platformWrapKey):
(WebCore::CryptoAlgorithmAES_KW::platformUnwrapKey):

• crypto/mac/CryptoAlgorithmAES_CBCMac.cpp:

(WebCore::CryptoAlgorithmAES_CBC::platformEncrypt):
(WebCore::CryptoAlgorithmAES_CBC::platformDecrypt):

• crypto/mac/CryptoAlgorithmAES_CFBMac.cpp:

(WebCore::CryptoAlgorithmAES_CFB::platformEncrypt):
(WebCore::CryptoAlgorithmAES_CFB::platformDecrypt):

• crypto/mac/CryptoAlgorithmAES_CTRMac.cpp:

(WebCore::CryptoAlgorithmAES_CTR::platformEncrypt):
(WebCore::CryptoAlgorithmAES_CTR::platformDecrypt):

• crypto/mac/CryptoAlgorithmAES_GCMMac.cpp:

(WebCore::CryptoAlgorithmAES_GCM::platformEncrypt):
(WebCore::CryptoAlgorithmAES_GCM::platformDecrypt):

• crypto/mac/CryptoAlgorithmAES_KWMac.cpp:

(WebCore::CryptoAlgorithmAES_KW::platformWrapKey):
(WebCore::CryptoAlgorithmAES_KW::platformUnwrapKey):