Timeline


and

07/28/16: Today

08:43 Changeset [203818] by commit-queue@webkit.org
  • 18 edits
    6 moves in trunk

[Streams API] Replace ReadableStreamController by ReadableStreamDefaultController
https://bugs.webkit.org/show_bug.cgi?id=160242

Patch by Romain Bellessort <romain.bellessort@crf.canon.fr> on 2016-07-28
Reviewed by Youenn Fablet.

Replaced ReadableStreamController by ReadableStreamDefaultController to align
with updated Streams API specification.

No change in functionality.

Source/WebCore:

  • CMakeLists.txt:
  • DerivedSources.cpp:
  • DerivedSources.make:
  • Modules/streams/ReadableStream.js:

(initializeReadableStream):

  • Modules/streams/ReadableStreamDefaultController.idl: Renamed from Source/WebCore/Modules/streams/ReadableStreamController.idl.
  • Modules/streams/ReadableStreamDefaultController.js: Renamed from Source/WebCore/Modules/streams/ReadableStreamController.js.

(enqueue):
(error):
(close):
(desiredSize):

  • Modules/streams/ReadableStreamInternals.js:

(privateInitializeReadableStreamDefaultController):
(isReadableStreamDefaultController):

  • Modules/streams/ReadableStreamSource.h:

(WebCore::ReadableStreamSource::controller):
(WebCore::ReadableStreamSource::start):

  • Modules/streams/ReadableStreamSource.idl:
  • WebCore.xcodeproj/project.pbxproj:
  • bindings/js/JSDOMGlobalObject.cpp:

(WebCore::JSDOMGlobalObject::addBuiltinGlobals):

  • bindings/js/JSReadableStreamPrivateConstructors.cpp:

(WebCore::constructJSReadableStreamDefaultController):
(WebCore::JSBuiltinReadableStreamDefaultControllerPrivateConstructor::initializeExecutable):
(WebCore::createReadableStreamDefaultControllerPrivateConstructor):

  • bindings/js/JSReadableStreamPrivateConstructors.h:
  • bindings/js/JSReadableStreamSourceCustom.cpp:

(WebCore::JSReadableStreamSource::start):

  • bindings/js/ReadableStreamDefaultController.cpp: Renamed from Source/WebCore/bindings/js/ReadableStreamController.cpp.

(WebCore::callFunction):
(WebCore::ReadableStreamDefaultController::invoke):
(WebCore::ReadableStreamDefaultController::isControlledReadableStreamLocked):

  • bindings/js/ReadableStreamDefaultController.h: Renamed from Source/WebCore/bindings/js/ReadableStreamController.h.

(WebCore::ReadableStreamDefaultController::ReadableStreamDefaultController):
(WebCore::ReadableStreamDefaultController::close):
(WebCore::ReadableStreamDefaultController::error):
(WebCore::ReadableStreamDefaultController::enqueue):
(WebCore::ReadableStreamDefaultController::globalObject):
(WebCore::ReadableStreamDefaultController::error<String>):

  • bindings/js/WebCoreBuiltinNames.h:

LayoutTests:

  • streams/readable-stream-default-controller-error-expected.txt: Renamed from LayoutTests/streams/readable-stream-controller-error-expected.txt.
  • streams/readable-stream-default-controller-error.html: Renamed from LayoutTests/streams/readable-stream-controller-error.html.
  • streams/readable-stream-error-messages-expected.txt:
  • streams/readable-stream-error-messages.html:
  • streams/reference-implementation/readable-stream-expected.txt:
05:07 Changeset [203817] by ossy@webkit.org
  • 2 edits in trunk/Source/JavaScriptCore

[ARM] Typo fix after r121885
https://bugs.webkit.org/show_bug.cgi?id=160288

Reviewed by Zoltan Herczeg.

  • assembler/MacroAssemblerARM.h:

(JSC::MacroAssemblerARM::maxJumpReplacementSize):

02:28 Changeset [203816] by ossy@webkit.org
  • 2 edits in trunk/Source/JavaScriptCore

64-bit alignment check isn't necessary in ARMAssembler::prepareExecutableCopy after r202214
https://bugs.webkit.org/show_bug.cgi?id=159711

Reviewed by Mark Lam.

  • assembler/ARMAssembler.cpp:

(JSC::ARMAssembler::prepareExecutableCopy):

02:27 MathML/Early_2016_Refactoring edited by fred.wang@free.fr
(diff)
01:30 Changeset [203815] by commit-queue@webkit.org
  • 35 edits in trunk

Compute fetch response type in case of cross-origin requests
https://bugs.webkit.org/show_bug.cgi?id=158565

Patch by Youenn Fablet <youenn@apple.com> on 2016-07-28
Reviewed by Alex Christensen.

LayoutTests/imported/w3c:

Rebasing fetch API tests as filtering is now done.
Rebasing XHR tests as console messages are no longer available when trying to access non-exposed headers.

  • web-platform-tests/XMLHttpRequest/getresponseheader-cookies-and-more-expected.txt:
  • web-platform-tests/fetch/api/basic/mode-no-cors-expected.txt:
  • web-platform-tests/fetch/api/basic/mode-no-cors-worker-expected.txt:
  • web-platform-tests/fetch/api/cors/cors-basic-expected.txt:
  • web-platform-tests/fetch/api/cors/cors-filtering-expected.txt:
  • web-platform-tests/fetch/api/request/request-cache-expected.txt:

Source/WebCore:

Covered by rebased tests.

Implementing Response filtering based on Response tainting in ResourceResponse.
Refactoring code in FetchHeaders and CrossOriginAccessControl.cpp accordingly.

Computing response tainting in SubresourceLoader for all resources.
This is used by DocumentThreadableLoader which now filters responses accordingly for all its clients including fetch and XHR.

Response tainting notably allows computing the response type and filtering out headers in case of cross origin responses.

Removing the filtering implemented in XMLHttpRequest as this is done before it gets access to the headers.
This is triggering some rebasing in the XHR tests as error messages triggered by trying to access unsafe headers no longer happen.

This filtering currently requires creating a new ResourceResponse object from the one sent from CachedResource.
This is done so as the same ResourceResponse may be reused accross loads and may be filtered differently by given to two different DocumentThreadableLoader
This can be mitigated in the future by changing ThreadableLoaderClient API to pass a ResourceResponse&&.

  • Modules/fetch/FetchHeaders.cpp: Moving header checking in HTTParsers.h/.cpp

(WebCore::isForbiddenHeaderName): Deleted.
(WebCore::isForbiddenResponseHeaderName): Deleted.
(WebCore::isSimpleHeader): Deleted.

  • loader/CrossOriginAccessControl.cpp:

(WebCore::parseAccessControlExposeHeadersAllowList): Deleted.

  • loader/CrossOriginAccessControl.h: Moving header checking in HTTParsers.h/.cpp
  • loader/DocumentThreadableLoader.cpp:

(WebCore::DocumentThreadableLoader::responseReceived):
(WebCore::DocumentThreadableLoader::didReceiveResponse): Doing response filtering. Since underlying loaders are
not yet aware that fetch mode may be cors (it is always no-cors currently), the tainting needs to be updated.
(WebCore::DocumentThreadableLoader::loadRequest): Computing response tainting in case of synchronous calls to ensure headers are filtered for synchronous XHR.

  • loader/DocumentThreadableLoader.h:
  • loader/SubresourceLoader.cpp:

(WebCore::SubresourceLoader::init): Getting origin from its resource and setting response tainting accordingly
(WebCore::SubresourceLoader::willSendRequestInternal): Always calling checkRedirectionCrossOriginAccessControl
to ensure response tainting is computed, even for no-cors resources.
(WebCore::SubresourceLoader::didReceiveResponse):
(WebCore::SubresourceLoader::checkRedirectionCrossOriginAccessControl): Computing response tainting in case of redirection.

  • loader/SubresourceLoader.h:
  • loader/cache/CachedResource.cpp:

(WebCore::CachedResource::load): Computing resource origin from the HTTP headers or from the document if none is
set in the HTTP headers.
(WebCore::CachedResource::setCrossOrigin): Helper routine to set response tainting.
(WebCore::CachedResource::isCrossOrigin): Helper routine to know whether resource is cross origin
(WebCore::CachedResource::isClean):
(WebCore::CachedResource::setResponse): Removing m_responseType

  • loader/cache/CachedResource.h:

(WebCore::CachedResource::responseTainting):
(WebCore::CachedResource::origin):
(WebCore::CachedResource::setOpaqueRedirect): Deleted.

  • platform/network/HTTPParsers.cpp: Implementing safe response header checking

(WebCore::parseAccessControlExposeHeadersAllowList):
(WebCore::isForbiddenHeaderName):
(WebCore::isForbiddenResponseHeaderName):
(WebCore::isSimpleHeader):
(WebCore::isCrossOriginSafeHeader):

  • platform/network/HTTPParsers.h:
  • platform/network/ResourceRequestBase.cpp:

(WebCore::ResourceRequestBase::hasHTTPOrigin): Added.
(WebCore::ResourceRequestBase::clearHTTPOrigin):

  • platform/network/ResourceRequestBase.h:
  • platform/network/ResourceResponseBase.cpp: Implementation of response filtering.

(WebCore::ResourceResponseBase::filterResponse):

  • platform/network/ResourceResponseBase.h:
  • xml/XMLHttpRequest.cpp:

(WebCore::isSetCookieHeader): Deleted.
(WebCore::XMLHttpRequest::getAllResponseHeaders): Removing header filtering since DocumentThreadableLoader does it.
(WebCore::XMLHttpRequest::getResponseHeader): Ditto.

LayoutTests:

Rebasing fetch API tests as filtering is now done.
Rebasing XHR tests as console messages are no longer available when trying to access non-exposed headers.

  • http/tests/xmlhttprequest/access-control-basic-whitelist-response-headers-expected.txt:
  • http/tests/xmlhttprequest/access-control-response-with-expose-headers-expected.txt:
  • http/tests/xmlhttprequest/get-dangerous-headers-expected.txt:
  • http/tests/xmlhttprequest/getResponseHeader-expected.txt:
  • platform/ios-simulator-wk2/imported/w3c/web-platform-tests/fetch/api/basic/mode-no-cors-expected.txt:
  • platform/ios-simulator-wk2/imported/w3c/web-platform-tests/fetch/api/basic/mode-no-cors-worker-expected.txt:
  • platform/ios-simulator-wk2/imported/w3c/web-platform-tests/fetch/api/cors/cors-basic-expected.txt:
  • platform/mac-wk2/imported/w3c/web-platform-tests/fetch/api/basic/mode-no-cors-expected.txt:
  • platform/mac-wk2/imported/w3c/web-platform-tests/fetch/api/basic/mode-no-cors-worker-expected.txt:
  • platform/mac-wk2/imported/w3c/web-platform-tests/fetch/api/cors/cors-basic-expected.txt:

07/27/16: Yesterday

23:37 Changeset [203814] by commit-queue@webkit.org
  • 2 edits in trunk/Source/WebCore

[Streams API] Use makeThisTypeError in ReadableStreamDefaultReader.js
https://bugs.webkit.org/show_bug.cgi?id=160254

Patch by Romain Bellessort <romain.bellessort@crf.canon.fr> on 2016-07-27
Reviewed by Darin Adler.

Use makeThisTypeError and makeGetterTypeError in ReadableStreamDefaultReader.js

No change in functionality.

  • Modules/streams/ReadableStreamDefaultReader.js:

(cancel):
(read):
(releaseLock):
(closed):

23:28 Changeset [203813] by commit-queue@webkit.org
  • 3 edits in trunk/Source/WebCore

[soup] Incorrect usage of relaxAdoptionRequirement in the constructor of SocketStreamHandle
https://bugs.webkit.org/show_bug.cgi?id=160281

Patch by Fujii Hironori <Hironori.Fujii@sony.com> on 2016-07-27
Reviewed by Carlos Garcia Campos.

No new tests (No behavior change).

Incrementing refcount in a constructor causes an assertion failure
that it's not adopted yet. So, relaxAdoptionRequirement() was
used to avoid the problem in the constructors of
SocketStreamHandle. This is a incorrect solution. The correct
solution is to make SocketStreamHandle::create() increment the
refcount after calling the constructor.

  • platform/network/soup/SocketStreamHandle.h: Removed the second

constructor of SocketStreamHandle which is not used anymore.
Uninlined create() because this is not trivial anymore.

  • platform/network/soup/SocketStreamHandleSoup.cpp:

(WebCore::SocketStreamHandle::create): Do the rest of jobs which
was done by the constructors.
(WebCore::SocketStreamHandle::SocketStreamHandle): Move the jobs
after initialization to create().
Removed the second constructor.

22:59 Changeset [203812] by commit-queue@webkit.org
  • 3 edits in trunk/Source/JavaScriptCore

[JSC] Remove some unused code from FTL
https://bugs.webkit.org/show_bug.cgi?id=160285

Patch by Benjamin Poulain <bpoulain@apple.com> on 2016-07-27
Reviewed by Mark Lam.

All the liveness and swapping is done inside B3,
this code is no longer needed.

  • dfg/DFGEdge.h:

(JSC::DFG::Edge::doesNotKill): Deleted.

  • ftl/FTLLowerDFGToB3.cpp:

(JSC::FTL::DFG::LowerDFGToB3::doesKill): Deleted.

22:47 Changeset [203811] by ap@apple.com
  • 2 edits in trunk/Tools

LayoutTestRelay should wait for WebKitTestRunnerApp installation to complete
https://bugs.webkit.org/show_bug.cgi?id=160262

Reviewed by Daniel Bates.

  • LayoutTestRelay/LayoutTestRelay/LTRelayController.m:

(-[LTRelayController installApp]):

21:45 Changeset [203810] by ryanhaddad@apple.com
  • 3 edits in trunk/LayoutTests

Marking http/tests/loading/basic-credentials-sent-automatically.html as flaky on mac and ios-sim wk2
https://bugs.webkit.org/show_bug.cgi?id=159427

Unreivewed test gardening.

  • platform/ios-simulator-wk2/TestExpectations:
  • platform/mac-wk2/TestExpectations:
18:59 Changeset [203809] by aestes@apple.com
  • 4 edits in trunk/Source/WebKit2

[iOS] Add WKUIDelegate SPI for specifying that an attachment list is from a managed source
https://bugs.webkit.org/show_bug.cgi?id=160280
<rdar://problem/27471815>

Reviewed by Dan Bernstein.

  • Platform/spi/ios/UIKitSPI.h: Declared

UIPreviewItemTypeAttachment, UIPreviewDataAttachmentList, and UIPreviewDataAttachmentIndex.

  • UIProcess/API/Cocoa/WKUIDelegatePrivate.h: Declared -_attachmentListForWebView:sourceIsManaged:.
  • UIProcess/ios/WKContentViewInteraction.mm:

(-[WKContentView _dataForPreviewItemController:atPosition:type:]): Used UIPreviewItemTypeAttachment,
UIPreviewDataAttachmentList, and UIPreviewDataAttachmentIndex. If uiDelegate responds to
-_attachmentListForWebView:sourceIsManaged:, called it instead of -_attachmentListForWebView:.
Set sourceIsManaged as the value for the UIPreviewDataAttachmentListSourceIsManaged key in dataForPreview.

18:36 Changeset [203808] by commit-queue@webkit.org
  • 23 edits
    4 deletes in trunk/Source/JavaScriptCore

[JSC] DFG::Node should not have its own allocator
https://bugs.webkit.org/show_bug.cgi?id=160098

Patch by Benjamin Poulain <bpoulain@apple.com> on 2016-07-27
Reviewed by Geoffrey Garen.

We need some design changes for DFG::Node:
-Accessing the index must be fast. B3 uses indices for sets

and maps, it is a lot faster than hashing pointers.

-We should be able to subclass DFG::Node to specialize it.

  • CMakeLists.txt:
  • JavaScriptCore.xcodeproj/project.pbxproj:
  • dfg/DFGAllocator.h: Removed.

(JSC::DFG::Allocator::Region::size): Deleted.
(JSC::DFG::Allocator::Region::headerSize): Deleted.
(JSC::DFG::Allocator::Region::numberOfThingsPerRegion): Deleted.
(JSC::DFG::Allocator::Region::data): Deleted.
(JSC::DFG::Allocator::Region::isInThisRegion): Deleted.
(JSC::DFG::Allocator::Region::regionFor): Deleted.
(JSC::DFG::Allocator<T>::Allocator): Deleted.
(JSC::DFG::Allocator<T>::~Allocator): Deleted.
(JSC::DFG::Allocator<T>::allocate): Deleted.
(JSC::DFG::Allocator<T>::free): Deleted.
(JSC::DFG::Allocator<T>::freeAll): Deleted.
(JSC::DFG::Allocator<T>::reset): Deleted.
(JSC::DFG::Allocator<T>::indexOf): Deleted.
(JSC::DFG::Allocator<T>::allocatorOf): Deleted.
(JSC::DFG::Allocator<T>::bumpAllocate): Deleted.
(JSC::DFG::Allocator<T>::freeListAllocate): Deleted.
(JSC::DFG::Allocator<T>::allocateSlow): Deleted.
(JSC::DFG::Allocator<T>::freeRegionsStartingAt): Deleted.
(JSC::DFG::Allocator<T>::startBumpingIn): Deleted.

  • dfg/DFGByteCodeParser.cpp:

(JSC::DFG::ByteCodeParser::addToGraph):

  • dfg/DFGCPSRethreadingPhase.cpp:

(JSC::DFG::CPSRethreadingPhase::freeUnnecessaryNodes):
(JSC::DFG::CPSRethreadingPhase::addPhiSilently):

  • dfg/DFGCleanUpPhase.cpp:

(JSC::DFG::CleanUpPhase::run):

  • dfg/DFGConstantFoldingPhase.cpp:

(JSC::DFG::ConstantFoldingPhase::run):

  • dfg/DFGConstantHoistingPhase.cpp:
  • dfg/DFGDCEPhase.cpp:

(JSC::DFG::DCEPhase::fixupBlock):

  • dfg/DFGDriver.cpp:

(JSC::DFG::compileImpl):

  • dfg/DFGGraph.cpp:

(JSC::DFG::Graph::Graph):
(JSC::DFG::Graph::deleteNode):
(JSC::DFG::Graph::killBlockAndItsContents):
(JSC::DFG::Graph::~Graph): Deleted.

  • dfg/DFGGraph.h:

(JSC::DFG::Graph::addNode):

  • dfg/DFGLICMPhase.cpp:

(JSC::DFG::LICMPhase::attemptHoist):

  • dfg/DFGLongLivedState.cpp: Removed.

(JSC::DFG::LongLivedState::LongLivedState): Deleted.
(JSC::DFG::LongLivedState::~LongLivedState): Deleted.
(JSC::DFG::LongLivedState::shrinkToFit): Deleted.

  • dfg/DFGLongLivedState.h: Removed.
  • dfg/DFGNode.cpp:

(JSC::DFG::Node::index): Deleted.

  • dfg/DFGNode.h:

(JSC::DFG::Node::index):

  • dfg/DFGNodeAllocator.h: Removed.

(operator new ): Deleted.

  • dfg/DFGObjectAllocationSinkingPhase.cpp:
  • dfg/DFGPlan.cpp:

(JSC::DFG::Plan::compileInThread):
(JSC::DFG::Plan::compileInThreadImpl):

  • dfg/DFGPlan.h:
  • dfg/DFGSSAConversionPhase.cpp:

(JSC::DFG::SSAConversionPhase::run):

  • dfg/DFGWorklist.cpp:

(JSC::DFG::Worklist::runThread):

  • runtime/VM.cpp:

(JSC::VM::VM): Deleted.

  • runtime/VM.h:
16:54 Changeset [203807] by nvasilyev@apple.com
  • 4 edits in trunk/Source/WebInspectorUI

Web Inspector: Visual Styles Sidebar should have only one column when it's narrow
https://bugs.webkit.org/show_bug.cgi?id=159905
<rdar://problem/27413248>

Reviewed by Joseph Pecoraro.

Many CSS values often get clipped in the two colunm layout. Change the layout to one colunm
when visual style rows get too narrow.

  • UserInterface/Views/VisualStyleDetailsPanel.css:

(.sidebar > .panel.details.css-style .visual > .details-section .details-section > .content .group > .row):
(.sidebar > .panel.details.css-style .visual > .details-section .details-section > .content .group > .metric-section-row):
Wrap all rows except for position/padding/margin controls.

(.sidebar > .panel.details.css-style .visual > .details-section .details-section > .content .group > .row > .visual-style-property-container:not(.layout-reversed):last-child): Deleted.
Margin between the first and the second column doesn't make sense one column layout.
Set the margin in .visual-style-property-container instead.

  • UserInterface/Views/VisualStyleDetailsPanel.js:

(WebInspector.VisualStyleDetailsPanel.prototype._generateMetricSectionRows):

  • UserInterface/Views/VisualStylePropertyEditor.css:

(.visual-style-property-container):

16:50 Changeset [203806] by cdumez@apple.com
  • 9 edits in trunk

First parameter to HTMLMediaElement.canPlayType() should be mandatory
https://bugs.webkit.org/show_bug.cgi?id=160275

Reviewed by Ryosuke Niwa.

LayoutTests/imported/w3c:

Rebaseline w3c test now that more checks are passing.

  • web-platform-tests/html/dom/interfaces-expected.txt:

Source/WebCore:

First parameter to HTMLMediaElement.canPlayType() should be mandatory:

Firefox and Chrome agree with the specification.

No new tests, rebaselined existing tests.

  • html/HTMLMediaElement.idl:

LayoutTests:

Update existing tests to reflect behavior change.

  • media/encrypted-media/encrypted-media-can-play-type.html:
  • media/media-can-play-type-expected.txt:
  • media/media-can-play-type.html:
  • platform/mac/media/encrypted-media/encrypted-media-can-play-type-expected.txt:
16:50 Changeset [203805] by cdumez@apple.com
  • 5 edits in trunk

First parameter to setTimeout() / setInterval() should be mandatory
https://bugs.webkit.org/show_bug.cgi?id=160277

Reviewed by Darin Adler.

LayoutTests/imported/w3c:

Rebaseline W3C test now that more checks are passing.

  • web-platform-tests/html/dom/interfaces-expected.txt:

Source/WebCore:

First parameter to setTimeout() / setInterval() should be mandatory:

Firefox and Chrome agree with the specification.

No new tests, rebaselined existing test.

  • bindings/js/JSDOMWindowCustom.cpp:

(WebCore::JSDOMWindow::setTimeout):
(WebCore::JSDOMWindow::setInterval):

  • bindings/js/JSWorkerGlobalScopeCustom.cpp:

(WebCore::JSWorkerGlobalScope::setTimeout):
(WebCore::JSWorkerGlobalScope::setInterval):

16:36 Changeset [203804] by commit-queue@webkit.org
  • 2 edits in trunk/Source/WebInspectorUI

[Mac] Web Inspector: CodeMirror-based editor bindings for Home and End don't match system behavior
https://bugs.webkit.org/show_bug.cgi?id=160267
<rdar://problem/27575553>

Patch by Joseph Pecoraro <pecoraro@apple.com> on 2016-07-27
Reviewed by Brian Burg.

  • UserInterface/Views/CodeMirrorEditor.js:

(WebInspector.CodeMirrorEditor.create):
(WebInspector.CodeMirrorEditor):
Add some key map overrides for Home and End to better match system Mac
behavior. This scrolls to the start or end of a document and does not
change the cursor position.

16:28 Changeset [203803] by cdumez@apple.com
  • 3 edits
    2 adds in trunk

Parameters to insertAdjacentText() / insertAdjacentHTML() should be mandatory
https://bugs.webkit.org/show_bug.cgi?id=160274

Reviewed by Darin Adler.

Source/WebCore:

Parameters to insertAdjacentText() / insertAdjacentHTML() should be mandatory:

Firefox and Chrome agree with the specification (although Firefox does not
have insertAdjacentText()).

Test: fast/dom/Element/insertAdjacentText-parameters.html

  • html/HTMLElement.idl:

LayoutTests:

Add test coverage.

  • fast/dom/Element/insertAdjacentText-parameters-expected.txt: Added.
  • fast/dom/Element/insertAdjacentText-parameters.html: Added.
16:22 Changeset [203802] by benjamin@webkit.org
  • 7 edits in trunk/Source/JavaScriptCore

[JSC] Fix a bunch of use-after-free of DFG::Node
https://bugs.webkit.org/show_bug.cgi?id=160228

Patch by Benjamin Poulain <bpoulain@apple.com> on 2016-07-27
Reviewed by Mark Lam.

FTL had a few places where we use a node after it has been
deleted. The dangling pointers come from the SSA liveness information
kept on the basic blocks.

This patch fixes the issues I could find and adds liveness invalidation
to help finding dependencies like these.

  • dfg/DFGBasicBlock.h:

(JSC::DFG::BasicBlock::SSAData::invalidate):

  • dfg/DFGConstantFoldingPhase.cpp:

(JSC::DFG::ConstantFoldingPhase::run):
Constant folding phase was deleting nodes in the loop over basic blocks.
The problem is the deleted nodes can be referenced by other blocks.
When the abstract interpreter was manipulating the abstract values of those
it was doing so on the dead nodes.

  • dfg/DFGConstantHoistingPhase.cpp:

Just invalidation. Nothing wrong here since the useless nodes were
kept live while iterating the blocks.

  • dfg/DFGGraph.cpp:

(JSC::DFG::Graph::killBlockAndItsContents):
(JSC::DFG::Graph::killUnreachableBlocks):
(JSC::DFG::Graph::invalidateNodeLiveness):

  • dfg/DFGGraph.h:
  • dfg/DFGPlan.cpp:

(JSC::DFG::Plan::compileInThreadImpl):
We had a lot of use-after-free in LCIM because we were using the stale
live nodes deleted by previous phases.

15:34 Changeset [203801] by bdakin@apple.com
  • 4 edits in trunk/Source/WebCore

Add localizable strings for inserting list types
https://bugs.webkit.org/show_bug.cgi?id=160233
-and corresponding-
rdar://problem/26102954

Reviewed by Dan Bernstein.

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

(WebCore::insertListTypeNone):
(WebCore::insertListTypeNoneAccessibilityTitle):
(WebCore::insertListTypeBulleted):
(WebCore::insertListTypeBulletedAccessibilityTitle):
(WebCore::insertListTypeNumbered):
(WebCore::insertListTypeNumberedAccessibilityTitle):

  • platform/LocalizedStrings.h:
15:15 Changeset [203800] by cdumez@apple.com
  • 3 edits
    2 adds in trunk

Parameters to DOMParser.parseFromString() should be mandatory
https://bugs.webkit.org/show_bug.cgi?id=160268

Reviewed by Ryosuke Niwa.

Source/WebCore:

Parameters to DOMParser.parseFromString() should be mandatory:

Firefox and Chrome agree with the specification.

Test: fast/parser/DOMParser-parameters.html

  • xml/DOMParser.idl:

LayoutTests:

Add test coverage.

  • fast/parser/DOMParser-parameters-expected.txt: Added.
  • fast/parser/DOMParser-parameters.html: Added.
15:06 Changeset [203799] by eric.carlson@apple.com
  • 4 edits in trunk/Source/WebCore

Captions do not render in PiP window when element is hidden
https://bugs.webkit.org/show_bug.cgi?id=160265
<rdar://problem/27556788>

Reviewed by Simon Fraser.

  • html/shadow/MediaControlElements.cpp:

(WebCore::MediaControlTextTrackContainerElement::createTextTrackRepresentationImage): Pass new flag so
caption layers are always rendered.

  • rendering/RenderLayer.cpp:

(WebCore::RenderLayer::paintLayerContents): Paint non-visible layers when PaintLayerIgnoreVisibility
flag is set.

  • rendering/RenderLayer.h: Define PaintLayerIgnoreVisibility.
14:59 Changeset [203798] by keith_miller@apple.com
  • 2 edits
    1 add in trunk/Source/JavaScriptCore

concatAppendOne should allocate using the indexing type of the array if it cannot merge
https://bugs.webkit.org/show_bug.cgi?id=160261
<rdar://problem/27530122>

Reviewed by Mark Lam.

Before, if we could not merge the indexing types for copying, we would allocate the
the array as ArrayWithUndecided. Instead, we should allocate an array with the original
array's indexing type.

  • runtime/ArrayPrototype.cpp:

(JSC::concatAppendOne):

  • tests/stress/concat-append-one-with-sparse-array.js: Added.
14:51 Changeset [203797] by cdumez@apple.com
  • 11 edits in trunk

Parameter to named property getter should be mandatory
https://bugs.webkit.org/show_bug.cgi?id=160269

Reviewed by Ryosuke Niwa.

LayoutTests/imported/w3c:

Rebaseline W3C test now that more checks are passing.

  • web-platform-tests/html/dom/interfaces-expected.txt:

Source/WebCore:

Parameter to named property getter should be mandatory:

No new tests, rebaselined existing test.

  • Modules/mediastream/RTCStatsResponse.idl:
  • bindings/scripts/test/TestOverrideBuiltins.idl:
  • html/HTMLOptionsCollection.idl:
  • html/HTMLSelectElement.idl:
  • plugins/DOMMimeTypeArray.idl:
  • plugins/DOMPlugin.idl:
  • plugins/DOMPluginArray.idl:
14:36 Changeset [203796] by cdumez@apple.com
  • 4 edits in trunk

First parameter to Range.createContextualFragment() should be mandatory
https://bugs.webkit.org/show_bug.cgi?id=160266

Reviewed by Ryosuke Niwa.

LayoutTests/imported/w3c:

Rebaseline W3C test now that one more check is passing.

  • web-platform-tests/domparsing/createContextualFragment-expected.txt:

Source/WebCore:

First parameter to Range.createContextualFragment() should be mandatory:

Firefox and Chrome agree with the specification.

No new tests, rebaselined existing test.

  • dom/Range.idl:
14:33 Changeset [203795] by cdumez@apple.com
  • 5 edits
    4 adds in trunk

Align MediaList with the CSSOM specification
https://bugs.webkit.org/show_bug.cgi?id=160257

Reviewed by Ryosuke Niwa.

Source/WebCore:

Align MediaList with the CSSOM specification:

In particular, the parameter to item() / deleteMedium() and
appendMedium() is now mandatory.

Firefox and Chrome agree with the specification.

Test: fast/css/MediaList-mediaText-null.html

fast/css/MediaList-parameters.html

  • css/MediaList.idl:

LayoutTests:

  • fast/css/MediaList-mediaText-null-expected.txt: Added.
  • fast/css/MediaList-mediaText-null.html: Added.

Add test coverage for MediaList.mediaText to make sure it is not nullable
and treats null as the empty string. Our IDL did not match the
specification here but our behavior was correct. Therefore, this test is
passing with and without my change. I just wanted to make sure we had
good coverage since I updated our IDL to match the specification.

  • fast/css/MediaList-parameters-expected.txt: Added.
  • fast/css/MediaList-parameters.html: Added.

Add test coverage for mandatory parameters.

  • fast/dom/non-numeric-values-numeric-parameters-expected.txt:
  • fast/dom/script-tests/non-numeric-values-numeric-parameters.js:

Update existing test to reflect behavior change.

14:18 Changeset [203794] by ryanhaddad@apple.com
  • 2 edits in trunk/LayoutTests

Correct the syntax used to skip js/regress/script-tests/bigswitch-indirect-symbol.js

Unreviewed test gardening.

  • js/regress/script-tests/bigswitch-indirect-symbol.js:
14:11 Changeset [203793] by sbarati@apple.com
  • 2 edits
    1 add in trunk/Source/JavaScriptCore

We don't optimize for-in properly in baseline JIT (maybe other JITs too) with an object with symbols
https://bugs.webkit.org/show_bug.cgi?id=160211
<rdar://problem/27572612>

Reviewed by Geoffrey Garen.

The fast for-in iteration mode assumes all inline/out-of-line properties
can be iterated in linear order. This is not true if we have Symbols
because Symbols should not be iterated by for-in.

  • runtime/Structure.cpp:

(JSC::Structure::add):

  • tests/stress/symbol-should-not-break-for-in.js: Added.

(assert):
(foo):

14:10 Changeset [203792] by commit-queue@webkit.org
  • 2 edits in trunk/Source/WebCore

Fullscreen video zoom button does not work after rotating when aspect ratio matches display.
https://bugs.webkit.org/show_bug.cgi?id=160263
rdar://problem/27368872

Patch by Jeremy Jones <jeremyj@apple.com> on 2016-07-27
Reviewed by Eric Carlson.

When video and display aspect ratio match, and rotating from landscape to protrait, the transform used in layout
will be Identity. This means checking the transform for identity is an insufficient test to see if the bounds
need to be resolved.

Instead, always attempt to resolve the bounds and do a more accurate test while doing so.

  • platform/ios/WebVideoFullscreenInterfaceAVKit.mm:

(-[WebAVPlayerLayer layoutSublayers]):
(-[WebAVPlayerLayer resolveBounds]):

14:00 Changeset [203791] by commit-queue@webkit.org
  • 3 edits in trunk/Source/WebKit2

[iOS] Remove unused textContentType SPI from _WKFormInputSession
https://bugs.webkit.org/show_bug.cgi?id=160260

Patch by Chelsea Pugh <cpugh@apple.com> on 2016-07-27
Reviewed by Dan Bernstein.

  • UIProcess/API/Cocoa/_WKFormInputSession.h: Remove unused SPI.
  • UIProcess/ios/WKContentViewInteraction.mm:

(-[WKContentView textInputTraits]): Set textContentType of _traits to whatever we classify it as based
on the assisted node info. The default textContentType is nil, and that is our fallback in our method
for determining textContentType.
(-[WKFormInputSession textContentType]): Deleted.
(-[WKFormInputSession setTextContentType:]): Deleted.

13:55 Changeset [203790] by mark.lam@apple.com
  • 2 edits
    1 add in trunk/Source/JavaScriptCore

The second argument for Function.prototype.apply should be array-like or null/undefined.
https://bugs.webkit.org/show_bug.cgi?id=160212
<rdar://problem/27328525>

Reviewed by Filip Pizlo.

The spec for Function.prototype.apply says its second argument can only be null,
undefined, or must be array-like. See
https://tc39.github.io/ecma262/#sec-function.prototype.apply and
https://tc39.github.io/ecma262/#sec-createlistfromarraylike.

Our previous implementation was not handling this correctly for SymbolType.
This is now fixed.

  • interpreter/Interpreter.cpp:

(JSC::sizeOfVarargs):

  • tests/stress/apply-second-argument-must-be-array-like.js: Added.
13:52 Changeset [203789] by andersca@apple.com
  • 2 edits in trunk/Source/WebCore

Stop accepting the deprecated "requiredShippingAddressFields" and "requiredBillingAddressFields" properties
https://bugs.webkit.org/show_bug.cgi?id=160264
rdar://problem/27574519

Reviewed by Simon Fraser.

  • Modules/applepay/ApplePaySession.cpp:

(WebCore::createPaymentRequest):
(WebCore::isValidPaymentRequestPropertyName):

13:51 Changeset [203788] by cdumez@apple.com
  • 21 edits in trunk

First parameter to indexed property getters should be mandatory
https://bugs.webkit.org/show_bug.cgi?id=160259

Reviewed by Ryosuke Niwa.

LayoutTests/imported/w3c:

Rebaseline W3C test now that more checks are passing.

  • web-platform-tests/html/dom/interfaces-expected.txt:

Source/WebCore:

First parameter to indexed property getters should be mandatory:

No new tests, rebaselined existing tests.

  • Modules/gamepad/deprecated/GamepadList.idl:
  • bindings/js/JSHTMLAllCollectionCustom.cpp:

(WebCore::JSHTMLAllCollection::item):

  • css/CSSRuleList.idl:
  • css/CSSValueList.idl:
  • css/StyleSheetList.idl:
  • css/WebKitCSSFilterValue.idl:
  • css/WebKitCSSTransformValue.idl:
  • dom/ClientRectList.idl:
  • dom/DOMStringList.idl:
  • dom/DataTransferItemList.idl:
  • html/HTMLAllCollection.idl:
  • plugins/DOMMimeTypeArray.idl:
  • plugins/DOMPlugin.idl:
  • plugins/DOMPluginArray.idl:

LayoutTests:

Update existing tests to reflect behavior change.

  • fast/css/webkit-keyframes-crash.html:
  • fast/dom/non-numeric-values-numeric-parameters-expected.txt:
  • fast/dom/script-tests/non-numeric-values-numeric-parameters.js:
13:05 Changeset [203787] by ryanhaddad@apple.com
  • 2 edits in trunk/LayoutTests

Land test expectations for rdar://problem/27574303.

Unreviewed test gardening.

  • platform/mac-wk2/TestExpectations:
12:56 Changeset [203786] by sbarati@apple.com
  • 7 edits in trunk/Source/JavaScriptCore

MathICs should be able to emit only a jump along the inline path when they don't have any type data
https://bugs.webkit.org/show_bug.cgi?id=160110

Reviewed by Mark Lam.

This patch allows for MathIC fast-path generation to be delayed.
We delay when we don't see any observed type information for
the lhs/rhs operand, which implies that the MathIC has never
executed. This is profitable for two main reasons:

  1. If the math operation never executes, we emit much less code.
  2. Once we get type information for the lhs/rhs, we can emit better code.

To implement this, we just emit a jump to the slow path call
that will repatch on first execution.

New data for add:

| JetStream | Unity 3D |

------| -------------

Old | 148 bytes | 143 bytes |

------| -------------

New | 116 bytes | 113 bytes |

------------------------------------

New data for mul:

| JetStream | Unity 3D |

------| -------------

Old | 210 bytes | 185 bytes |

------| -------------

New | 170 bytes | 137 bytes |

------------------------------------

  • jit/JITAddGenerator.cpp:

(JSC::JITAddGenerator::generateInline):

  • jit/JITAddGenerator.h:

(JSC::JITAddGenerator::isLeftOperandValidConstant):
(JSC::JITAddGenerator::isRightOperandValidConstant):
(JSC::JITAddGenerator::arithProfile):

  • jit/JITMathIC.h:

(JSC::JITMathIC::generateInline):
(JSC::JITMathIC::generateOutOfLine):
(JSC::JITMathIC::finalizeInlineCode):

  • jit/JITMathICInlineResult.h:
  • jit/JITMulGenerator.cpp:

(JSC::JITMulGenerator::generateInline):

  • jit/JITMulGenerator.h:

(JSC::JITMulGenerator::isLeftOperandValidConstant):
(JSC::JITMulGenerator::isRightOperandValidConstant):
(JSC::JITMulGenerator::arithProfile):

  • jit/JITOperations.cpp:
11:46 Changeset [203785] by jfernandez@igalia.com
  • 2 edits in trunk/Source/WebCore

[css-grid] The isValidTransition function must not alter the GridSizingData object
https://bugs.webkit.org/show_bug.cgi?id=160252

Reviewed by Darin Adler.

It's not correct that a function which purpose is to check out the
validity of a state modifies such state.

That code was there to allow the extra row track sizing iteration
in the case of a grid container with indefinite height. We need to
do that in order to figure out its intrinsic height, which will be
used then to properly sizing the row tracks.

Since the intrinsic height computation uses directly the
computeUsedBreadthOfGridTracks function, it does not alter the
algorithm state-machine, hence, we can safely remove this code,
which was incorrect in any case.

No new tests, it's just a refactoring with no functionality change.

  • rendering/RenderGrid.cpp:

(WebCore::RenderGrid::GridSizingData::isValidTransition):

11:27 Changeset [203784] by cdumez@apple.com
  • 5 edits
    2 adds in trunk

First parameter to Document.execCommand() / queryCommand*() should be mandatory
https://bugs.webkit.org/show_bug.cgi?id=160258

Reviewed by Darin Adler.

LayoutTests/imported/w3c:

Rebaseline W3C test now that more checks are passing.

  • web-platform-tests/html/dom/interfaces-expected.txt:

Source/WebCore:

First parameter to Document.execCommand() / queryCommand*() should be mandatory:

Firefox and Chrome agree with the specification.

Test: fast/dom/Document/editing-parameters.html

  • dom/Document.idl:

LayoutTests:

Add layout test coverage.

  • fast/dom/Document/editing-parameters-expected.txt: Added.
  • fast/dom/Document/editing-parameters.html: Added.
11:04 Changeset [203783] by ryanhaddad@apple.com
  • 2 edits in trunk/LayoutTests

Fix a typo in TestExpectations.

Unreviewed test gardening.

  • platform/mac-wk1/TestExpectations:
11:04 Changeset [203782] by cdumez@apple.com
  • 8 edits
    2 adds in trunk

Align CSSSupportsRule with the specification
https://bugs.webkit.org/show_bug.cgi?id=160256

Reviewed by Darin Adler.

Source/WebCore:

Align CSSSupportsRule with the specification:

In particular:

  • Make the parameters to insertRule() / deleteRule() mandatory
  • Expose CSSSupportsRule on the global Window object

Both Firefox and Chrome agree with the specification here.

Test: fast/css/CSSSupportsRule-parameters.html

  • css/CSSSupportsRule.idl:

LayoutTests:

  • fast/css/CSSSupportsRule-parameters-expected.txt: Added.
  • fast/css/CSSSupportsRule-parameters.html: Added.

Add layout test coverage for mandatory parameters.

  • platform/efl/js/dom/global-constructors-attributes-expected.txt:
  • platform/gtk/js/dom/global-constructors-attributes-expected.txt:
  • platform/mac-yosemite/js/dom/global-constructors-attributes-expected.txt:
  • platform/mac/js/dom/global-constructors-attributes-expected.txt:
  • platform/win/js/dom/global-constructors-attributes-expected.txt:

Rebaseline existing test now that CSSSupportsRule is exposed on the
global Window object.

10:15 Changeset [203781] by ryanhaddad@apple.com
  • 2 edits in trunk/Tools

Disable WebCoreNSURLSessionTest API tests on ios-simulator
https://bugs.webkit.org/show_bug.cgi?id=160225

Reviewed by Alexey Proskuryakov.

  • TestWebKitAPI/Tests/WebCore/WebCoreNSURLSession.mm:
10:03 Changeset [203780] by cdumez@apple.com
  • 2 edits in trunk/Source/WebKit2

Fix m_isInBackground initialization for Safari View Services
https://bugs.webkit.org/show_bug.cgi?id=160255
<rdar://problem/27569255>

Reviewed by Tim Horton.

Fix m_isInBackground initialization for Safari View Services. The
code was using m_applicationStateMonitor without initializing it.
Instead, use the local applicationStateMonitor and make sure we
invalidate it before it gets released.

  • UIProcess/ApplicationStateTracker.mm:

(WebKit::ApplicationStateTracker::ApplicationStateTracker):

09:18 Changeset [203779] by commit-queue@webkit.org
  • 2 edits in trunk/LayoutTests

[GTK] Unreviewed gardening: update expectations after r203770
https://bugs.webkit.org/show_bug.cgi?id=160253

Unreviewed gardening.

Patch by Miguel Gomez <magomez@igalia.com> on 2016-07-27

  • platform/gtk/TestExpectations:
09:14 Changeset [203778] by bburg@apple.com
  • 2 edits in trunk/Source/WebInspectorUI

Regression(r203535): Uncaught Exception: TypeError: Not enough arguments at LayerTreeDataGridNode.js:47
https://bugs.webkit.org/show_bug.cgi?id=160187
<rdar://problem/27540435>

Reviewed by Eric Carlson.

After r203535, document.createTextNode() requires an argument.

  • UserInterface/Views/LayerTreeDataGridNode.js:

(WebInspector.LayerTreeDataGridNode.prototype.createCellContent):
Since this use-site is for creating a cell in an unknown column,
initialize it to '–'. Previously it would have been the string
"undefined" or empty.

09:00 Changeset [203777] by commit-queue@webkit.org
  • 7 edits
    5 adds in trunk

[GTK] Fix some video/canvas tests that should be passing
https://bugs.webkit.org/show_bug.cgi?id=160023

Patch by Miguel Gomez <magomez@igalia.com> on 2016-07-27
Reviewed by Carlos Garcia Campos.

Tools:

Add a platform identifier to the TestRunner's page user agent when the tests are run on the
EFL or GTK platforms.

  • WebKitTestRunner/efl/TestControllerEfl.cpp:

(WTR::TestController::platformConfigureViewForTest):

  • WebKitTestRunner/gtk/TestControllerGtk.cpp:

(WTR::TestController::platformConfigureViewForTest):

LayoutTests:

Modify the tests to use a tolerance of 6 when running on the GTK or EFL platforms. Also,
added new expectations for the tests that need them.

  • fast/canvas/canvas-createPattern-video-loading.html:
  • fast/canvas/canvas-createPattern-video-modify.html:
  • media/video-canvas-createPattern.html:
  • platform/efl/fast/canvas/canvas-createPattern-video-loading-expected.txt: Added.
  • platform/efl/fast/canvas/canvas-createPattern-video-modify-expected.txt: Added.
  • platform/gtk/fast/canvas/canvas-createPattern-video-loading-expected.txt: Added.
  • platform/gtk/fast/canvas/canvas-createPattern-video-modify-expected.txt: Added.
  • resources/platform-helper.js: Added.

(isGtk):
(isEfl):
(videoCanvasPixelComparisonTolerance):

08:58 Changeset [203776] by carlosgc@webkit.org
  • 13 edits in trunk/Source/WebKit2

[Coordinated Graphics] Improve scheduling of tasks between threads in CoordinatedGraphicsScene
https://bugs.webkit.org/show_bug.cgi?id=160238

Reviewed by Michael Catanzaro.

This patch makes the following improvements:

  • Avoid scheduling tasks to the main thread if the scene is detached.
  • Do not take references when not actually sending tasks to another threads.
  • Use Function instead of std::function on dispatch methods.
  • Remove purgeBackingStores that is actually never called. It's only scheduled from purgeGLResources() that is always called after detach.
  • Shared/CoordinatedGraphics/CoordinatedGraphicsScene.cpp:

(WebKit::CoordinatedGraphicsScene::dispatchOnMainThread):
(WebKit::CoordinatedGraphicsScene::dispatchOnClientRunLoop):
(WebKit::CoordinatedGraphicsScene::paintToCurrentGLContext):
(WebKit::CoordinatedGraphicsScene::updateViewport):
(WebKit::CoordinatedGraphicsScene::onNewBufferAvailable):
(WebKit::CoordinatedGraphicsScene::commitSceneState):
(WebKit::CoordinatedGraphicsScene::renderNextFrame):
(WebKit::CoordinatedGraphicsScene::purgeGLResources):
(WebKit::CoordinatedGraphicsScene::commitScrollOffset):
(WebKit::CoordinatedGraphicsScene::detach):
(WebKit::CoordinatedGraphicsScene::setActive):
(WebKit::CoordinatedGraphicsScene::dispatchCommitScrollOffset): Deleted.
(WebKit::CoordinatedGraphicsScene::purgeBackingStores): Deleted.

  • Shared/CoordinatedGraphics/CoordinatedGraphicsScene.h:
  • Shared/CoordinatedGraphics/threadedcompositor/ThreadedCompositor.cpp:

(WebKit::ThreadedCompositor::purgeBackingStores): Deleted.

  • Shared/CoordinatedGraphics/threadedcompositor/ThreadedCompositor.h:
  • UIProcess/CoordinatedGraphics/CoordinatedLayerTreeHostProxy.cpp:

(WebKit::CoordinatedLayerTreeHostProxy::purgeBackingStores): Deleted.

  • UIProcess/CoordinatedGraphics/CoordinatedLayerTreeHostProxy.h:
  • WebProcess/WebPage/CoordinatedGraphics/CompositingCoordinator.cpp:

(WebKit::CompositingCoordinator::invalidate):

  • WebProcess/WebPage/CoordinatedGraphics/CompositingCoordinator.h:
  • WebProcess/WebPage/CoordinatedGraphics/CoordinatedLayerTreeHost.cpp:

(WebKit::CoordinatedLayerTreeHost::invalidate):
(WebKit::CoordinatedLayerTreeHost::purgeBackingStores): Deleted.

  • WebProcess/WebPage/CoordinatedGraphics/CoordinatedLayerTreeHost.h:
  • WebProcess/WebPage/CoordinatedGraphics/CoordinatedLayerTreeHost.messages.in:
  • WebProcess/WebPage/CoordinatedGraphics/ThreadedCoordinatedLayerTreeHost.h:
08:52 Changeset [203775] by carlosgc@webkit.org
  • 2 edits in trunk/Source/WebCore

[Soup] Test http/tests/xmlhttprequest/auth-reject-protection-space.html fails since added in r203743
https://bugs.webkit.org/show_bug.cgi?id=160235

Reviewed by Michael Catanzaro.

It times out in release and crashes in debug due to an ASSERT_NOT_REACHED(). The soup backend doesn't really
support rejecting the protection space and continuing with the next one, but we can at least continue with the
request without crendential to make the test pass.

  • platform/network/soup/ResourceHandleSoup.cpp:

(WebCore::ResourceHandle::receivedChallengeRejection):

08:50 WebKitGTK/Gardening/Calendar edited by mcatanzaro@igalia.com
(diff)
08:49 Changeset [203774] by carlosgc@webkit.org
  • 2 edits in trunk/Source/WebKit2

[GTK] Remove network setup from web process
https://bugs.webkit.org/show_bug.cgi?id=160236

Reviewed by Michael Catanzaro.

We are still doing network init and finish in th web process. It's useless since we switched to mandatory
network process.

  • WebProcess/gtk/WebProcessMainGtk.cpp:
08:49 WebKitGTK/Gardening/Calendar edited by mcatanzaro@igalia.com
(diff)
08:30 Changeset [203773] by clopez@igalia.com
  • 2 edits in trunk/Tools

[Tools] The built product doesn't contains the dwo files when DEBUG_FISSION is enabled.
https://bugs.webkit.org/show_bug.cgi?id=160247

Reviewed by Michael Catanzaro.

  • BuildSlaveSupport/built-product-archive:

(createZip):
(dirContainsdwo):
(archiveBuiltProduct):

03:40 Changeset [203772] by commit-queue@webkit.org
  • 23 edits
    4 moves in trunk

[Streams API] Replace ReadableStreamReader by ReadableStreamDefaultReader
https://bugs.webkit.org/show_bug.cgi?id=160194

Patch by Romain Bellessort <romain.bellessort@crf.canon.fr> on 2016-07-27
Reviewed by Youenn Fablet.

Replaced ReadableStreamReader by ReadableStreamDefaultReader to align
with updated Streams API specification.

No change in functionality.

Source/WebCore:

  • CMakeLists.txt:
  • DerivedSources.cpp:
  • DerivedSources.make:
  • Modules/fetch/FetchInternals.js:

(consumeStream):

  • Modules/fetch/FetchResponse.js:

(body):

  • Modules/streams/ReadableStream.js:

(getReader):

  • Modules/streams/ReadableStreamDefaultReader.idl: Renamed from Source/WebCore/Modules/streams/ReadableStreamReader.idl.
  • Modules/streams/ReadableStreamDefaultReader.js: Renamed from Source/WebCore/Modules/streams/ReadableStreamReader.js.

(cancel):
(read):
(releaseLock):
(closed):

  • Modules/streams/ReadableStreamInternals.js:

(privateInitializeReadableStreamDefaultReader):
(teeReadableStream):
(teeReadableStreamPullFunction):
(isReadableStreamDefaultReader):
(closeReadableStream):
(readFromReadableStreamDefaultReader):

  • WebCore.xcodeproj/project.pbxproj:
  • bindings/js/JSDOMGlobalObject.cpp:

(WebCore::JSDOMGlobalObject::addBuiltinGlobals):
(WebCore::JSDOMGlobalObject::finishCreation):

  • bindings/js/JSReadableStreamPrivateConstructors.cpp:

(WebCore::constructJSReadableStreamDefaultReader):
(WebCore::JSBuiltinReadableStreamDefaultReaderPrivateConstructor::initializeExecutable):
(WebCore::createReadableStreamDefaultReaderPrivateConstructor):

  • bindings/js/JSReadableStreamPrivateConstructors.h:
  • bindings/js/WebCoreBuiltinNames.h:
  • features.json:

LayoutTests:

  • streams/brand-checks.html:
  • streams/readable-stream-controller-error-expected.txt:
  • streams/readable-stream-controller-error.html:
  • streams/readable-stream-default-reader-read-expected.txt: Renamed from LayoutTests/streams/readable-stream-reader-read-expected.txt.
  • streams/readable-stream-default-reader-read.html: Renamed from LayoutTests/streams/readable-stream-reader-read.html.
  • streams/readable-stream-error-messages-expected.txt:
  • streams/readable-stream-error-messages.html:
  • streams/reference-implementation/readable-stream-reader-expected.txt:
  • streams/shadowing-Promise-expected.txt:
  • streams/shadowing-Promise.html:
03:07 Changeset [203771] by jfernandez@igalia.com
  • 4 edits
    6 adds in trunk

[css-grid] Handle alignment with orthogonal flows
https://bugs.webkit.org/show_bug.cgi?id=159295

Reviewed by Darin Adler.

Now that grid sizing and positioning issues wrt orthogonal flows have
been clarified in the last spec draft, we can adapt now our alignment
logic to work with orthogonal flows.

Source/WebCore:

Even though basic alignment would work with orthogonal flows with
this patch, we still doesn't allow stretching in that case. I'll provide a
patch for that feature since it's a complex logic and better have an
isolated change.

Tests: fast/css-grid-layout/grid-item-alignment-with-orthogonal-flows-vertical-lr.html

fast/css-grid-layout/grid-item-alignment-with-orthogonal-flows-vertical-rl.html
fast/css-grid-layout/grid-item-alignment-with-orthogonal-flows.html

  • rendering/RenderGrid.cpp:

(WebCore::computeOverflowAlignmentOffset): Using 'size' instead of 'breadth' as concept.
(WebCore::RenderGrid::columnAxisPositionForChild): Dealing with orthogonal flow cases.
(WebCore::RenderGrid::rowAxisPositionForChild): Dealing with orthogonal flow cases.
(WebCore::RenderGrid::columnAxisOffsetForChild): Using 'size' instead of 'breadth' as concept.
(WebCore::RenderGrid::rowAxisOffsetForChild): Using 'size' instead of 'breadth' as concept.
(WebCore::RenderGrid::findChildLogicalPosition): Dealing with orthogonal flow cases.

LayoutTests:

These tests ensure that alignment works as expected in the cases where
grid and its children are orthogonal.

  • fast/css-grid-layout/grid-item-alignment-with-orthogonal-flows-expected.txt: Added.
  • fast/css-grid-layout/grid-item-alignment-with-orthogonal-flows-vertical-lr-expected.txt: Added.
  • fast/css-grid-layout/grid-item-alignment-with-orthogonal-flows-vertical-lr.html: Added.
  • fast/css-grid-layout/grid-item-alignment-with-orthogonal-flows-vertical-rl-expected.txt: Added.
  • fast/css-grid-layout/grid-item-alignment-with-orthogonal-flows-vertical-rl.html: Added.
  • fast/css-grid-layout/grid-item-alignment-with-orthogonal-flows.html: Added.
  • fast/css-grid-layout/resources/grid-alignment.css:

(.alignContentSpaceBetween):
(.alignContentSpaceAround):
(.alignContentSpaceEvenly):
(.alignContentStretch):
(.selfStart):
(.selfEnd):
(.selfCenter):
(.selfRight):
(.selfLeft):
(.selfSelfStart):
(.selfSelfEnd):
(.itemsSelfEnd): Deleted.

00:42 Changeset [203770] by achristensen@apple.com
  • 2 edits in trunk/Source/WebKit/win

Fix Win debug build after r203749.

  • WebView.cpp:
00:06 Changeset [203769] by carlosgc@webkit.org
  • 1 copy in releases/WebKitGTK/webkit-2.13.4

WebKitGTK+ 2.13.4

00:03 Changeset [203768] by carlosgc@webkit.org
  • 4 edits in trunk

Unreviewed. Update OptionsGTK.cmake and NEWS for 2.13.4 release.

.:

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

Source/WebKit2:

  • gtk/NEWS: Add release notes for 2.13.4.

07/26/16:

23:38 Changeset [203767] by commit-queue@webkit.org
  • 21 edits
    1 copy
    1 add in trunk

[Fetch API] Response constructor should be able to take a ReadableStream as body
https://bugs.webkit.org/show_bug.cgi?id=159804

Patch by Youenn Fablet <youenn@apple.com> on 2016-07-26
Reviewed by Alex Christensen.

LayoutTests/imported/w3c:

  • web-platform-tests/fetch/api/response/response-consume-empty-expected.txt:
  • web-platform-tests/fetch/api/response/response-consume-expected.txt:
  • web-platform-tests/fetch/api/response/response-consume.html: Updating test to exercice Response coonstructor with a ReadableStream.

Source/WebCore:

Covered by existing and updated tests.

Introduced FetchBodyConsumer to encapsulate the code responsible to adapt FetchBody internal data to the requests made by user scripts.
This refactoring eases the handling of internal data coming from ReadableStream.

FetchLoader is now delegating conversion from the data to its m_consumer field.
If m_consumer is null, FetchLoader calls FetchLoaderClient::didReceiveData (streaming reception mode).
Clients of FetchLoader needs to pass a FetchBodyConsumer to the FetchLoader to do the data adaptation at loader creation time.

Added support for body data passed as a ReadableStream to Response.
This requires to set @body internal slot of the Response object in the constructor initialization JS built-in.

To actually use that data, Body accessors are also implemented as JS built-in for Response.
Since there is no need to do so for Request, FetchResponse is no longer marked as implementing FetchBody but all
FetchBody IDL description is inlined in FetchResponse.idl.

For each body accessor (arrayBuffer, blob, json, text), the corresponding JS built-in checks whether @body internal slot is set.
If that is not the case, regular handling is done through a new private method called @consume.
If @body internal slot is set, chunks are pumped from the ReadableStream using ReadableStream internal built-ins functions.
Data handling is done in C++ through the private methods @startConsumingStream, @consumeChunk and @finishConsumingStream.

To support cloning of Response with bodies, clone method is also implemented as a JS built-in.
Main clone is done through @cloneFoJS private method implemented in C++.
JS built-in clone code does the teeing of the ReadableStream using ReadableStream JS built-in internal functions.

Introducing a new ReadableStream type for FetchBody to cope with body being stored in a ReadableStream.

Introducing a new Loaded type for FetchBody to cope with body being stored in the FetchBodyConsumer owned by FetchBody.
This allows removing the conversion of data to JSC::ArrayBuffer which was done by defaut at the end of Fetch loading if user script did not request data before.
At the end of a load, the data remains in FetchBodyConsumer and the body is marked as Loaded.
If the user wants to get the data after data finished being loaded, FetchBodyConsumer will do the required conversions.

Introducing DeferredWrapper::resolveWithNewValue to handle the case of resolving promises with new objects.
This allows directly using toJSNewlyCreated instead of toJS.

  • CMakeLists.txt: Adding FetchBodyConsumer.cpp. Removing WebCoreJSBuiltins.cpp from CMake list as it is not needed.
  • Modules/fetch/FetchBody.cpp: Moving data adaptation code to FetchBodyConsumer.

(WebCore::FetchBody::extract):
(WebCore::FetchBody::arrayBuffer):
(WebCore::FetchBody::blob): Setting contentType in FetchBodyConsumer to handle proper creation of blob.
(WebCore::FetchBody::json):
(WebCore::FetchBody::text):
(WebCore::FetchBody::consume): Refactoring and added case of Loaded bodies.
(WebCore::FetchBody::consumeAsStream): Ditto.
(WebCore::FetchBody::consumeArrayBuffer):
(WebCore::FetchBody::consumeText):
(WebCore::FetchBody::consumeBlob):
(WebCore::FetchBody::loadingFailed):
(WebCore::FetchBody::loadingSucceeded):
(WebCore::FetchBody::loadingType): Deleted.
(WebCore::blobFromArrayBuffer): Deleted.
(WebCore::FetchBody::fulfillTextPromise): Deleted.
(WebCore::FetchBody::loadedAsArrayBuffer): Deleted.
(WebCore::FetchBody::loadedAsText): Deleted.

  • Modules/fetch/FetchBody.h:

(WebCore::FetchBody::consumer):

  • Modules/fetch/FetchBodyConsumer.cpp: Added, responsible of data adaptation.

(WebCore::blobFromData):
(WebCore::FetchBodyConsumer::resolveWithData):
(WebCore::FetchBodyConsumer::resolve):
(WebCore::FetchBodyConsumer::append):
(WebCore::FetchBodyConsumer::takeData):
(WebCore::FetchBodyConsumer::takeAsArrayBuffer):
(WebCore::FetchBodyConsumer::takeAsBlob):
(WebCore::FetchBodyConsumer::takeAsText):

  • Modules/fetch/FetchBodyConsumer.h: Added.

(WebCore::FetchBodyConsumer::FetchBodyConsumer):
(WebCore::FetchBodyConsumer::setContentType):
(WebCore::FetchBodyConsumer::setType):
(WebCore::FetchBodyConsumer::type):
(WebCore::FetchBodyConsumer::clean):
(WebCore::FetchBodyConsumer::hasData):

  • Modules/fetch/FetchBodyOwner.cpp:

(WebCore::FetchBodyOwner::loadBlob):
(WebCore::FetchBodyOwner::blobLoadingSucceeded):
(WebCore::FetchBodyOwner::loadedBlobAsText): Deleted.

  • Modules/fetch/FetchBodyOwner.h:

(WebCore::FetchBodyOwner::loadedBlobAsArrayBuffer): Deleted.

  • Modules/fetch/FetchInternals.js:

(consumeStream): Pump ReadableStream data and send it to FetchResponse to be converted according user need.

  • Modules/fetch/FetchLoader.cpp:

(WebCore::FetchLoader::FetchLoader): FetchLoader is simplified as it has two nodes: consumer mode in which case
data is sent to the consumer and no-consumer mode in which case data is passed to loader client. The second mode
is used to conveyy data to ReadableStream source.
(WebCore::FetchLoader::stop):
(WebCore::FetchLoader::startStreaming):
(WebCore::FetchLoader::didReceiveData):
(WebCore::FetchLoader::didFinishLoading): Deleted.

  • Modules/fetch/FetchLoader.h:
  • Modules/fetch/FetchLoaderClient.h:

(WebCore::FetchLoaderClient::didFinishLoadingAsText): Deleted.
(WebCore::FetchLoaderClient::didFinishLoadingAsArrayBuffer): Deleted.

  • Modules/fetch/FetchResponse.cpp:

(WebCore::FetchResponse::cloneForJS):
(WebCore::FetchResponse::BodyLoader::didSucceed):
(WebCore::FetchResponse::BodyLoader::start):
(WebCore::FetchResponse::consume): Introduced to consume data stored in body and not in ReadableStream.
(WebCore::FetchResponse::startConsumingStream): Introduced to start process of consuming data stored in the ReadableStream.
(WebCore::FetchResponse::consumeChunk): Reception of ReadableStream data.
(WebCore::FetchResponse::finishConsumingStream): Doing the final conversion.
(WebCore::FetchResponse::clone): Deleted.
(WebCore::FetchResponse::BodyLoader::didFinishLoadingAsArrayBuffer): Deleted.

  • Modules/fetch/FetchResponse.h:
  • Modules/fetch/FetchResponse.idl: Inlining of FetchBody methods/attributes and adding private methods.
  • Modules/fetch/FetchResponse.js:

(initializeFetchResponse):
(clone):
(arrayBuffer):
(blob):
(formData):
(json):
(text):

  • WebCore.xcodeproj/project.pbxproj:
  • bindings/js/JSDOMPromise.h:

(WebCore::DeferredWrapper::resolveWithNewValue):

  • bindings/js/WebCoreBuiltinNames.h:
23:37 Changeset [203766] by commit-queue@webkit.org
  • 10 edits
    4 adds in trunk

JS Built-ins should throw this-error messages consistently with binding generated code
https://bugs.webkit.org/show_bug.cgi?id=160191

Patch by Youenn Fablet <youennf@gmail.com> on 2016-07-26
Reviewed by Darin Adler.

Source/WebCore:

Introducing @makeThisTypeError and @makeGetterTypeError to create TypeError objects with a consistent error message.
Making use of these functions in streams API and fetch API related built-in code.

Refactored JSDOMBinding.cpp code by adding makeThisTypeErrorMessage and makeGetterTypeErrorMessage helper routines
These routines are used by both new built-in functions as well as binding generated code helper routine.

Tests: fetch/fetch-error-messages.html

streams/readable-stream-error-messages.html

  • Modules/fetch/FetchResponse.js:

(body): Adding an explicit check so that the error message is right. The previous error message was related to the call of Response.@isDisturbed.

  • Modules/streams/ReadableStream.js:

(cancel):
(getReader):
(pipeTo):
(tee):
(locked):

  • Modules/streams/ReadableStreamController.js:

(enqueue):
(error):
(close):
(desiredSize):

  • Modules/streams/ReadableStreamReader.js:

(cancel):
(read):
(releaseLock):
(closed):

  • bindings/js/JSDOMBinding.cpp:

(WebCore::makeGetterTypeErrorMessage):
(WebCore::throwGetterTypeError):
(WebCore::makeThisTypeErrorMessage):
(WebCore::throwThisTypeError):
(WebCore::throwSequenceTypeError): Deleted.
(WebCore::throwSetterTypeError): Deleted.

  • bindings/js/JSDOMBinding.h:
  • bindings/js/JSDOMGlobalObject.cpp:

(WebCore::makeThisTypeErrorForBuiltins):
(WebCore::makeGetterTypeErrorForBuiltins):
(WebCore::JSDOMGlobalObject::addBuiltinGlobals):

  • bindings/js/WebCoreBuiltinNames.h:

LayoutTests:

  • fetch/fetch-error-messages-expected.txt: Added.
  • fetch/fetch-error-messages.html: Added.
  • streams/readable-stream-error-messages-expected.txt: Added.
  • streams/readable-stream-error-messages.html: Added.
23:35 Changeset [203765] by carlosgc@webkit.org
  • 3 edits in trunk/Source/WebCore

Unreviewed. Fix GTK+ distcheck build.

wtf/spi/darwin/dyldSPI.h is not included in GTK+ release tarballs.

  • html/HTMLObjectElement.cpp: Include wtf/spi/darwin/dyldSPI.h only for iOS.
  • html/MediaElementSession.cpp: Ditto.
23:20 Changeset [203764] by mmaxfield@apple.com
  • 5 edits
    4 adds in trunk

[iOS] SF-Heavy is inaccessible by web content
https://bugs.webkit.org/show_bug.cgi?id=160186
<rdar://problem/27434423>

Reviewed by Dean Jackson.

Source/WebCore:

Once we create the system font, we need to modify it with the appropriate weight.
This is because the CoreText API we use to get the system font on iOS does not
let us choose the exact weight we want.

Test: fast/text/system-font-weight.html

  • platform/graphics/ios/FontCacheIOS.mm:

(WebCore::baseSystemFontDescriptor):
(WebCore::systemFontModificationAttributes):
(WebCore::systemFontDescriptor):
(WebCore::platformFontWithFamilySpecialCase):

  • platform/spi/cocoa/CoreTextSPI.h:

LayoutTests:

  • platform/ios-simulator/TestExpectations: system-font-weight-italic.html

is expected to fail on iOS 9.

  • fast/text/system-font-weight-italic-expected.txt: Added.
  • fast/text/system-font-weight-italic.html: Added.
  • fast/text/system-font-weight-expected.txt: Added.
  • fast/text/system-font-weight.html: Added.
22:31 Changeset [203763] by ryanhaddad@apple.com
  • 2 edits in trunk/LayoutTests

Skip failing JSC test regress/script-tests/bigswitch-indirect-symbol.js
https://bugs.webkit.org/show_bug.cgi?id=160035

Unreviewed test gardening.

  • js/regress/script-tests/bigswitch-indirect-symbol.js:
22:20 Changeset [203762] by commit-queue@webkit.org
  • 2 edits in trunk/Source/WebCore

[GTK] ASSERTION FAILED: !m_adoptionIsRequired when Inspector Server is connected
https://bugs.webkit.org/show_bug.cgi?id=160229

Patch by Fujii Hironori <Hironori.Fujii@sony.com> on 2016-07-26
Reviewed by Carlos Garcia Campos.

An assertion fails because refcount of SocketStreamHandle is
incremented before adoptRef, in the constructor of
SocketStreamHandle. The constructor of SocketStreamHandle needs
to increment recount because it passes this pointer to libsoup.

  • platform/network/soup/SocketStreamHandleSoup.cpp:

(WebCore::SocketStreamHandle::SocketStreamHandle): Do
relaxAdoptionRequirement() as well as the another constructor.

21:27 Changeset [203761] by cdumez@apple.com
  • 9 edits in trunk

Move 'dir' attribute from HTMLDocument to Document
https://bugs.webkit.org/show_bug.cgi?id=160231

Reviewed by Sam Weinig.

LayoutTests/imported/w3c:

Rebaseline W3C test now that more checks are passing.

  • web-platform-tests/html/dom/interfaces-expected.txt:

Source/WebCore:

Move 'dir' attribute from HTMLDocument to Document to match the
specification:

Both Firefox and Chrome have 'dir' on Document already.

No new tests, rebaselined existing test.

  • dom/Document.cpp:

(WebCore::Document::dir):
(WebCore::Document::setDir):

  • dom/Document.h:
  • dom/Document.idl:
  • html/HTMLDocument.cpp:

(WebCore::HTMLDocument::dir): Deleted.
(WebCore::HTMLDocument::setDir): Deleted.

  • html/HTMLDocument.h:
  • html/HTMLDocument.idl:
21:12 Changeset [203760] by cdumez@apple.com
  • 8 edits
    2 adds in trunk

Second parameter to History.pushState() / replaceState() should be mandatory
https://bugs.webkit.org/show_bug.cgi?id=160230

Reviewed by Sam Weinig.

LayoutTests/imported/w3c:

Rebaseline W3C test now that more checks are passing.

  • web-platform-tests/html/dom/interfaces-expected.txt:

Source/WebCore:

Second parameter to History.pushState() / replaceState() should be mandatory:

Firefox and Chrome agree with the specification.

No new tests, rebaselined existing test.

  • bindings/js/JSHistoryCustom.cpp:

(WebCore::JSHistory::pushState):
(WebCore::JSHistory::replaceState):

  • page/History.idl:

LayoutTests:

  • fast/history/state-api-parameters.html: Added.
  • fast/history/state-api-parameters-expected.txt: Added.

Add layout test coverage.

  • fast/history/replacestate-nocrash.html:
  • fast/loader/stateobjects/popstate-fires-with-page-cache.html:

Update existing tests to reflect behavior change.

20:42 Changeset [203759] by cdumez@apple.com
  • 4 edits in trunk

Align Node.isEqualNode() with the specification
https://bugs.webkit.org/show_bug.cgi?id=160224

Reviewed by Sam Weinig.

LayoutTests/imported/w3c:

Rebaseline W3C test now that one more check is passing. We are now
passing all the checks in this test like Firefox and Chrome.

  • web-platform-tests/dom/nodes/Node-isEqualNode-expected.txt:

Source/WebCore:

Align our implementation for Node.isEqualNode() to match more closely
the text of the specification:

This also fixes a bug where isEqualNode() would sometimes return false
wrongly for elements because we were comparing the value returned by
nodeName() which returns the tagName of elements. The issue was that
the tagName's case may differ depending on the element being in an
HTMLDocument or not. We now compare Element::tagQName() instead which
ends up comparing namespace / namespace prefix and localName, as per
the specification. The new behavior matches Firefox and Chrome and
helps us pass an extra W3C test.

No new tests, rebaselined existing test.

  • dom/Node.cpp:

(WebCore::Node::isEqualNode):

18:23 Changeset [203758] by sbarati@apple.com
  • 10 edits in trunk/Source/JavaScriptCore

rollout r203666
https://bugs.webkit.org/show_bug.cgi?id=160226

Unreviewed rollout.

  • b3/B3BasicBlock.h:

(JSC::B3::BasicBlock::successorBlock):

  • b3/B3LowerToAir.cpp:

(JSC::B3::Air::LowerToAir::createGenericCompare):

  • b3/B3LowerToAir.h:
  • b3/air/AirArg.cpp:

(JSC::B3::Air::Arg::isRepresentableAs):
(JSC::B3::Air::Arg::usesTmp):

  • b3/air/AirArg.h:

(JSC::B3::Air::Arg::isRepresentableAs):
(JSC::B3::Air::Arg::asNumber):
(JSC::B3::Air::Arg::castToType): Deleted.

  • b3/air/AirCode.h:

(JSC::B3::Air::Code::size):
(JSC::B3::Air::Code::at):

  • b3/air/AirOpcode.opcodes:
  • b3/air/AirValidate.h:
  • b3/air/opcode_generator.rb:
  • b3/testb3.cpp:

(JSC::B3::compileAndRun):
(JSC::B3::testSomeEarlyRegister):
(JSC::B3::zero):
(JSC::B3::run):
(JSC::B3::lowerToAirForTesting): Deleted.
(JSC::B3::testBranchBitAndImmFusion): Deleted.

17:45 Changeset [203757] by ryanhaddad@apple.com
  • 2 edits in trunk/LayoutTests

Marking inspector/memory/tracking.html as flaky on El Capitan Debug WK1
https://bugs.webkit.org/show_bug.cgi?id=160227

Unreviewed test gardening.

  • platform/mac-wk1/TestExpectations:
17:44 Changeset [203756] by simon.fraser@apple.com
  • 4 edits in trunk/Source/WebKit

Fix the Windows debug build.

WebResourceLoadScheduler's logging was a holdover from it being in WebCore,
and prior to r203749 WebKit was actually using a WebCore log channel.

For some reason this doesn't build on Windows debug, so just remove
this logging for now.

Source/WebKit:

  • WebCoreSupport/WebResourceLoadScheduler.cpp:

(WebResourceLoadScheduler::scheduleLoad): Deleted.
(WebResourceLoadScheduler::servePendingRequests): Deleted.
(WebResourceLoadScheduler::scheduleServePendingRequests): Deleted.
(WebResourceLoadScheduler::requestTimerFired): Deleted.
(WebResourceLoadScheduler::HostInformation::addLoadInProgress): Deleted.

Source/WebKit/mac:

  • Misc/WebKitLogging.h:
16:40 Changeset [203755] by achristensen@apple.com
  • 2 edits in trunk/Tools

Fix tests after r203743.
https://bugs.webkit.org/show_bug.cgi?id=156947

  • WebKitTestRunner/TestController.cpp:

(WTR::TestController::resetStateToConsistentValues):
Reset the new boolean m_rejectsProtectionSpaceAndContinueForAuthenticationChallenges.

16:39 Changeset [203754] by simon.fraser@apple.com
  • 6 edits in trunk/Source

Sort the project files.

Source/WebCore:

  • WebCore.xcodeproj/project.pbxproj:

Source/WebKit:

  • WebKit.xcodeproj/project.pbxproj:

Source/WebKit2:

  • WebKit2.xcodeproj/project.pbxproj:
16:37 Changeset [203753] by cdumez@apple.com
  • 4 edits
    4 adds in trunk

Align CSSKeyframesRule with the specification
https://bugs.webkit.org/show_bug.cgi?id=160219

Reviewed by Darin Adler.

Source/WebCore:

Align CSSKeyframesRule with the specification:

In particular, the parameter to insertRule() / appendRule() /
deleteRule() / findRule() should be mandatory. Both Firefox and Chrome
agree with the specification here.

Also, the CSSKeyframesRule.name attribute should not be nullable.
Chrome agrees with the specification. However, Firefox, has the
attribute nullable. This patch aligns our behavior with Chrome and
the specification.

Tests: animations/CSSKeyframesRule-name-null.html

animations/CSSKeyframesRule-parameters.html

  • css/CSSKeyframesRule.h:

(WebCore::StyleRuleKeyframes::name):
(WebCore::StyleRuleKeyframes::setName):

  • css/CSSKeyframesRule.idl:

LayoutTests:

Add layout test coverage.

  • animations/CSSKeyframesRule-name-null-expected.txt: Added.
  • animations/CSSKeyframesRule-name-null.html: Added.
  • animations/CSSKeyframesRule-parameters-expected.txt: Added.
  • animations/CSSKeyframesRule-parameters.html: Added.
16:36 Changeset [203752] by mmaxfield@apple.com
  • 28 edits
    2 adds in trunk

[iPhone] Playing a video on tudou.com plays only sound, no video
https://bugs.webkit.org/show_bug.cgi?id=160178
<rdar://problem/27535468>

Source/WebCore:

Reviewed by Eric Carlson and Dan Bernstein.

This patch re-implements r203520 in a much simpler way which doesn't involve a new SPI.
The biggest problem with r203520 is that it make it impossible for a WKWebView to match
MobileSafari's behavior. Instead of adding this new SPI, a simple version check should
be used to keep old apps working.

The new behavior is characterized by the following table:

| iOS | Non-iOS

=============================================================================================
requiresPlayInlineAttribute == true | Old app: honor -webkit-playsinline | honor playsinline

| New app: honor playsinline | honor playsinline


requiresPlayInlineAttribute == false | Always inline | Always inline

Specifically, this patch reverts r203545 which is the commit which actually removes
the old SPI. As soon as Safari is migrated back to this old SPI, I'll remove the two
new SPIs added in r203520.

Tests: media/video-playsinline.html

media/video-webkit-playsinline.html

  • html/MediaElementSession.cpp:

(WebCore::MediaElementSession::requiresFullscreenForVideoPlayback):

  • page/Settings.cpp:
  • page/Settings.in:
  • testing/InternalSettings.cpp:

(WebCore::InternalSettings::Backup::Backup):
(WebCore::InternalSettings::Backup::restoreTo):
(WebCore::InternalSettings::setInlineMediaPlaybackRequiresPlaysInlineAttribute):

  • testing/InternalSettings.h:
  • testing/InternalSettings.idl:

Source/WebKit/mac:

Reviewed by Eric Carlson and Dan Bernstein.

  • WebView/WebPreferenceKeysPrivate.h:
  • WebView/WebPreferences.mm:

(+[WebPreferences initialize]):
(-[WebPreferences inlineMediaPlaybackRequiresPlaysInlineAttribute]):
(-[WebPreferences setInlineMediaPlaybackRequiresPlaysInlineAttribute:]):

  • WebView/WebPreferencesPrivate.h:
  • WebView/WebView.mm:

(-[WebView _preferencesChanged:]):

Source/WebKit2:

Reviewed by Eric Carlson and Dan Bernstein.

  • Shared/WebPreferencesDefinitions.h:
  • UIProcess/API/C/WKPreferences.cpp:

(WKPreferencesSetInlineMediaPlaybackRequiresPlaysInlineAttribute):
(WKPreferencesGetInlineMediaPlaybackRequiresPlaysInlineAttribute):

  • UIProcess/API/C/WKPreferencesRefPrivate.h:
  • UIProcess/API/Cocoa/WKWebView.mm:

(-[WKWebView _initializeWithConfiguration:]):

  • UIProcess/API/Cocoa/WKWebViewConfiguration.mm:

(-[WKWebViewConfiguration init]):
(-[WKWebViewConfiguration copyWithZone:]):
(-[WKWebViewConfiguration _inlineMediaPlaybackRequiresPlaysInlineAttribute]):
(-[WKWebViewConfiguration _setInlineMediaPlaybackRequiresPlaysInlineAttribute:]):

  • UIProcess/API/Cocoa/WKWebViewConfigurationPrivate.h:
  • WebProcess/WebPage/WebPage.cpp:

(WebKit::WebPage::updatePreferences):

Tools:

Reviewed by Dan Bernstein.

  • DumpRenderTree/mac/DumpRenderTree.mm:

(setDefaultsToConsistentValuesForTesting):

  • TestWebKitAPI/Tests/WebKit2Cocoa/RequiresUserActionForPlayback.mm:

(RequiresUserActionForPlaybackTest::SetUp):

  • WebKitTestRunner/TestController.cpp:

(WTR::TestController::resetPreferencesToConsistentValues):

  • WebKitTestRunner/cocoa/TestControllerCocoa.mm:

(WTR::initializeWebViewConfiguration):

LayoutTests:

Reviewed by Eric Carlson and Dan Bernstein.

  • media/video-playsinline-expected.txt:
  • media/video-playsinline.html:
  • media/video-webkit-playsinline-expected.txt: Added.
  • media/video-webkit-playsinline.html: Added.
16:33 Changeset [203751] by zalan@apple.com
  • 4 edits in trunk/Source/WebCore

Move RenderView::shouldDisableLayoutStateForSubtree to SubtreeLayoutStateMaintainer.
https://bugs.webkit.org/show_bug.cgi?id=160215

Reviewed by Darin Adler and Simon Fraser.

No change in functionality.

  • page/FrameView.cpp:

(WebCore::SubtreeLayoutStateMaintainer::SubtreeLayoutStateMaintainer):

  • rendering/RenderView.cpp:

(WebCore::RenderView::shouldDisableLayoutStateForSubtree):

  • rendering/RenderView.h:
16:31 Changeset [203750] by commit-queue@webkit.org
  • 5 edits
    2 deletes in trunk/Source/WebKit2

Remove unused DownloadAuthenticationClient
https://bugs.webkit.org/show_bug.cgi?id=160220

Patch by Alex Christensen <achristensen@webkit.org> on 2016-07-26
Reviewed by Darin Adler.

  • CMakeLists.txt:
  • NetworkProcess/Downloads/Download.cpp:
  • NetworkProcess/Downloads/Download.h:
  • NetworkProcess/Downloads/DownloadAuthenticationClient.cpp: Removed.
  • NetworkProcess/Downloads/DownloadAuthenticationClient.h: Removed.
  • WebKit2.xcodeproj/project.pbxproj:
16:30 Changeset [203749] by simon.fraser@apple.com
  • 28 edits
    1 copy
    2 adds in trunk

Allow LOG macros to be used outside the namespace, and other logging cleanup
https://bugs.webkit.org/show_bug.cgi?id=160216

Reviewed by Anders Carlsson.
Source/WebCore:

Fix some issues with the LOG macros.

First, they were not usable outside the WebKit namespace in WebKit2 code. Fix by moving
its log channels outside of the namespace (they are protected by a unique prefix anyway).

Second, allow LOG_WITH_STREAM to be used in WebKit2 by moving that and a helper macro
into LogMacros.h, which is exported from WebCore as a private header.

Third, split the Logging.h header into two. Logging.h remains for framework-internal
log channels and log macros. Add LogInitialization.h which is external, and used to
initialize the channels.

Finally unify "initializeLogChannels" and "initializeLoggingChannels" terminology everywhere.

  • WebCore.xcodeproj/project.pbxproj:
  • platform/LogInitialization.h: Added.
  • platform/LogMacros.h: Added.
  • platform/Logging.cpp:

(WebCore::initializeLogChannelsIfNecessary):
(WebCore::initializeLoggingChannelsIfNecessary): Deleted.

  • platform/Logging.h:
  • testing/js/WebCoreTestSupport.cpp:

(WebCoreTestSupport::initializeLogChannelsIfNecessary):
(WebCoreTestSupport::initializeLoggingChannelsIfNecessary): Deleted.

  • testing/js/WebCoreTestSupport.h:

Source/WebKit:

Fix some issues with the LOG macros.

First, they were not usable outside the WebKit namespace in WebKit2 code. Fix by moving
its log channels outside of the namespace (they are protected by a unique prefix anyway).

Second, allow LOG_WITH_STREAM to be used in WebKit2 by moving that and a helper macro
into LogMacros.h, which is exported from WebCore as a private header.

Third, split the Logging.h header into two. Logging.h remains for framework-internal
log channels and log macros. Add LogInitialization.h which is external, and used to
initialize the channels.

Finally unify "initializeLogChannels" and "initializeLoggingChannels" terminology everywhere.

  • WebCoreSupport/WebResourceLoadScheduler.cpp:

Source/WebKit/mac:

Fix some issues with the LOG macros.

First, they were not usable outside the WebKit namespace in WebKit2 code. Fix by moving
its log channels outside of the namespace (they are protected by a unique prefix anyway).

Second, allow LOG_WITH_STREAM to be used in WebKit2 by moving that and a helper macro
into LogMacros.h, which is exported from WebCore as a private header.

Third, split the Logging.h header into two. Logging.h remains for framework-internal
log channels and log macros. Add LogInitialization.h which is external, and used to
initialize the channels.

Finally unify "initializeLogChannels" and "initializeLoggingChannels" terminology everywhere.

  • Misc/WebKitLogging.h:
  • Misc/WebKitLogging.m:
  • WebView/WebView.mm:

(-[WebView _commonInitializationWithFrameName:groupName:]):

Source/WebKit/win:

Fix some issues with the LOG macros.

First, they were not usable outside the WebKit namespace in WebKit2 code. Fix by moving
its log channels outside of the namespace (they are protected by a unique prefix anyway).

Second, allow LOG_WITH_STREAM to be used in WebKit2 by moving that and a helper macro
into LogMacros.h, which is exported from WebCore as a private header.

Third, split the Logging.h header into two. Logging.h remains for framework-internal
log channels and log macros. Add LogInitialization.h which is external, and used to
initialize the channels.

Finally unify "initializeLogChannels" and "initializeLoggingChannels" terminology everywhere.

  • WebKitLogging.cpp:
  • WebKitLogging.h:
  • WebView.cpp:

(WebView::initWithFrame):

Source/WebKit2:

Fix some issues with the LOG macros.

First, they were not usable outside the WebKit namespace in WebKit2 code. Fix by moving
its log channels outside of the namespace (they are protected by a unique prefix anyway).

Second, allow LOG_WITH_STREAM to be used in WebKit2 by moving that and a helper macro
into LogMacros.h, which is exported from WebCore as a private header.

Third, split the Logging.h header into two. Logging.h remains for framework-internal
log channels and log macros. Add LogInitialization.h which is external, and used to
initialize the channels.

Finally unify "initializeLogChannels" and "initializeLoggingChannels" terminology everywhere.

  • NetworkProcess/NetworkProcess.cpp:
  • Platform/LogInitialization.h: Copied from Source/WebKit2/Platform/foundation/LoggingFoundation.mm.
  • Platform/Logging.cpp:

(WebKit::initializeLogChannelsIfNecessary):

  • Platform/Logging.h:
  • Platform/foundation/LoggingFoundation.mm:
  • Shared/WebKit2Initialize.cpp:

(WebKit::InitializeWebKit2):

  • UIProcess/API/Cocoa/WKWebView.mm:

(-[WKWebView _updateContentRectsWithState:]):
(-[WKWebView _navigationGestureDidBegin]):

  • UIProcess/WebProcessPool.cpp:

(WebKit::m_hiddenPageThrottlingTimer):

  • WebKit2.xcodeproj/project.pbxproj:

Tools:

initializeLoggingChannelsIfNecessary -> initializeLogChannelsIfNecessary

  • DumpRenderTree/TestRunner.cpp:
  • DumpRenderTree/mac/DumpRenderTree.mm:

(resetWebViewToConsistentStateBeforeTesting):

16:15 Changeset [203748] by rniwa@webkit.org
  • 2 edits in trunk/Websites/perf.webkit.org

REGRESSION: Tooltip for analysis tasks doesn't show up on charts
https://bugs.webkit.org/show_bug.cgi?id=160221

Rubber-stamped by Chris Dumez.

The bug was caused by ChartPaneBase resetting annotation bars every time the current point has moved.
Avoid doing this in ChartPaneBase's _renderAnnotations().

  • public/v3/components/chart-pane-base.js:

(ChartPaneBase):
(ChartPaneBase.prototype.fetchAnalysisTasks):
(ChartPaneBase.prototype._renderAnnotations):

16:06 Changeset [203747] by caitp@igalia.com
  • 3 edits
    2 moves in trunk/Source/JavaScriptCore

[JSC] Object.getOwnPropertyDescriptors should not add undefined props to result
https://bugs.webkit.org/show_bug.cgi?id=159409

Reviewed by Geoffrey Garen.

  • runtime/ObjectConstructor.cpp:

(JSC::objectConstructorGetOwnPropertyDescriptors):

  • tests/es6.yaml:
  • tests/es6/Object_static_methods_Object.getOwnPropertyDescriptors-proxy.js:

(testPropertiesIndexedSetterOnPrototypeThrows.set get var): Deleted.
(testPropertiesIndexedSetterOnPrototypeThrows): Deleted.

  • tests/stress/Object_static_methods_Object.getOwnPropertyDescriptors-proxy.js: Renamed from Source/JavaScriptCore/tests/es6/Object_static_methods_Object.getOwnPropertyDescriptors-proxy.js.
  • tests/stress/Object_static_methods_Object.getOwnPropertyDescriptors.js: Renamed from Source/JavaScriptCore/tests/es6/Object_static_methods_Object.getOwnPropertyDescriptors.js.
15:56 Changeset [203746] by andersca@apple.com
  • 2 edits in trunk/Source/WebCore

onpaymentauthorized callback not received when authorizing for a second time
https://bugs.webkit.org/show_bug.cgi?id=160218
rdar://problem/27527151

Reviewed by Tim Horton.

Only null out the active session if the status is a final state status.

  • Modules/applepay/PaymentCoordinator.cpp:

(WebCore::PaymentCoordinator::completePaymentSession):

15:42 Changeset [203745] by cdumez@apple.com
  • 4 edits in trunk

Range.prototype.compareBoundaryPoints.length should be 2
https://bugs.webkit.org/show_bug.cgi?id=160217

Reviewed by Sam Weinig.

LayoutTests/imported/w3c:

Rebaseline W3C test now that one more check is passing.

  • web-platform-tests/dom/interfaces-expected.txt:

Source/WebCore:

Range.prototype.compareBoundaryPoints.length:

We had a bug in our IDL which caused length to be 0 even though
both parameters were effectively mandatory.

No new tests, rebaselined existing test.

  • dom/Range.idl:
15:38 Changeset [203744] by cdumez@apple.com
  • 6 edits
    4 adds in trunk

Align CSSStyleDeclaration with the specification
https://bugs.webkit.org/show_bug.cgi?id=160214

Reviewed by Darin Adler.

Source/WebCore:

Align CSSStyleDeclaration with the specification:

In particular, the parameters to removeProperty() / item() and
getPropertyPriority() should be mandatory.

Firefox and Chrome match the specification.

Tests: fast/css/CSSStyleDeclaration-cssText-null.html

fast/css/CSSStyleDeclaration-parameters.html

  • bindings/js/JSCSSStyleDeclarationCustom.cpp:

(WebCore::JSCSSStyleDeclaration::getPropertyCSSValue):

  • css/CSSStyleDeclaration.idl:

LayoutTests:

  • fast/css/CSSStyleDeclaration-cssText-null-expected.txt: Added.
  • fast/css/CSSStyleDeclaration-cssText-null.html: Added.

Add layout test coverage for setting cssText to null. This test
passes in WebKit, Firefox and Chrome, with or without my change.
Our IDL wrongly reported the cssText attribute as nullable but
WebKit was already behaving correctly.

  • fast/css/CSSStyleDeclaration-parameters-expected.txt: Added.
  • fast/css/CSSStyleDeclaration-parameters.html: Added.

Add testing for omitting CSSStyleDeclaration API parameters, to
make sure they are mandatory. This test passes in Firefox and
Chrome.

  • fast/dom/non-numeric-values-numeric-parameters-expected.txt:
  • fast/dom/script-tests/non-numeric-values-numeric-parameters.js:

Update existing test to reflect behavior change.

15:31 Changeset [203743] by ddkilzer@apple.com
  • 16 edits
    2 adds in trunk

Networking process crash due to missing -[WebCoreAuthenticationClientAsChallengeSender performDefaultHandlingForAuthenticationChallenge:] implementation
https://bugs.webkit.org/show_bug.cgi?id=156947
<rdar://problem/23325160>

Reviewed by Alex Christensen.

Source/WebCore:

Test: http/tests/xmlhttprequest/auth-reject-protection-space.html

  • platform/network/mac/AuthenticationMac.mm:

(-[WebCoreAuthenticationClientAsChallengeSender performDefaultHandlingForAuthenticationChallenge:]): Added.
(-[WebCoreAuthenticationClientAsChallengeSender rejectProtectionSpaceAndContinueWithChallenge:]): Added.

Source/WebKit2:

  • UIProcess/API/C/WKAuthenticationDecisionListener.cpp:

(WKAuthenticationDecisionListenerRejectProtectionSpaceAndContinue):

  • UIProcess/API/C/WKAuthenticationDecisionListener.h:

Added new SPI for testing corresponding to calling the completion handler of
WKWebView.didReceiveAuthenticationChallenge with NSURLSessionAuthChallengeRejectProtectionSpace.

Tools:

  • DumpRenderTree/TestRunner.cpp:

(TestRunner::TestRunner):
(setRejectsProtectionSpaceAndContinueForAuthenticationChallengesCallback):

  • DumpRenderTree/TestRunner.h:

(TestRunner::rejectsProtectionSpaceAndContinueForAuthenticationChallenges):
(TestRunner::setRejectsProtectionSpaceAndContinueForAuthenticationChallenges):

  • DumpRenderTree/mac/ResourceLoadDelegate.mm:

(-[ResourceLoadDelegate webView:resource:didReceiveAuthenticationChallenge:fromDataSource:]):

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

(WTR::TestRunner::queueNonLoadingScript):
(WTR::TestRunner::setRejectsProtectionSpaceAndContinueForAuthenticationChallenges):

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

(WTR::TestController::didReceiveAuthenticationChallenge):

  • WebKitTestRunner/TestController.h:

(WTR::TestController::setRejectsProtectionSpaceAndContinueForAuthenticationChallenges):

  • WebKitTestRunner/TestInvocation.cpp:

(WTR::TestInvocation::didReceiveMessageFromInjectedBundle):
Add TestRunner.setRejectsProtectionSpaceAndContinueForAuthenticationChallenges to use for testing.

LayoutTests:

  • http/tests/xmlhttprequest/auth-reject-protection-space-expected.txt: Added.
  • http/tests/xmlhttprequest/auth-reject-protection-space.html: Added.
15:04 Changeset [203742] by ddkilzer@apple.com
  • 2 edits in trunk/Tools

check-for-exit-time-destructors should be usable outside Xcode
<https://webkit.org/b/160195>

Reviewed by Darin Adler.

  • Scripts/check-for-exit-time-destructors: Update to parse
-hhelp switch, or to take one argument to a binary to check

for exit time destructors on the command-line. The clang
compiler will find these at compile-time with the
-Wexit-time-destructors switch, but this script will check for
them after-the-fact.

15:03 Changeset [203741] by andersca@apple.com
  • 3 edits in trunk/Source/WebKit2

Payment session does not end if user closes all Safari windows
https://bugs.webkit.org/show_bug.cgi?id=160213
rdar://problem/27480873

Reviewed by Tim Horton.

Listen for the NSWindowWillCloseNotification of the sheet window and hide the payment UI when
the sheet window is going to be closed.

  • UIProcess/ApplePay/WebPaymentCoordinatorProxy.h:
  • UIProcess/ApplePay/mac/WebPaymentCoordinatorProxyMac.mm:

(WebKit::WebPaymentCoordinatorProxy::platformShowPaymentUI):
(WebKit::WebPaymentCoordinatorProxy::hidePaymentUI):

15:00 Changeset [203740] by cdumez@apple.com
  • 11 edits
    2 adds in trunk

Parameters to CSSStyleSheet.insertRule() / deleteRule() should be mandatory
https://bugs.webkit.org/show_bug.cgi?id=160210

Reviewed by Darin Adler.

Source/WebCore:

Parameters to CSSStyleSheet.insertRule() / deleteRule() should be mandatory:

They are mandatory in Firefox.
They are mandatory in Chrome except for the second parameter of insertRule()
which merely logs a deprecation warning.

This patch aligns our behavior with Chrome to move towards to specification
while limiting the risk of breakage.

Test: fast/css/stylesheet-parameters.html

  • css/CSSStyleSheet.cpp:

(WebCore::CSSStyleSheet::deprecatedInsertRule):

  • css/CSSStyleSheet.h:
  • css/CSSStyleSheet.idl:

LayoutTests:

  • fast/css/stylesheet-parameters-expected.txt: Added.
  • fast/css/stylesheet-parameters.html: Added.

Add layout test coverage.

  • editing/selection/first-letter-selection-crash.html:
  • fast/css/counters/asterisk-counter-update-after-layout-crash.html:
  • fast/dom/HTMLElement/dynamic-editability-change.html:
  • fast/dom/non-numeric-values-numeric-parameters-expected.txt:
  • fast/dom/script-tests/non-numeric-values-numeric-parameters.js:

Update existing tests to reflect the behavior change.

14:52 Changeset [203739] by commit-queue@webkit.org
  • 21 edits
    2 adds in trunk

HTMLVideoElement frames do not update on iOS when src is a MediaStream blob
https://bugs.webkit.org/show_bug.cgi?id=159833
<rdar://problem/27379487>

Patch by George Ruan <gruan@apple.com> on 2016-07-26
Reviewed by Eric Carlson.

Source/WebCore:

Test: fast/mediastream/MediaStream-video-element-displays-buffer.html

  • WebCore.xcodeproj/project.pbxproj:
  • platform/cf/CoreMediaSoftLink.cpp: Add CMSampleBufferCreateReadyWithImageBuffer and CMVideoFormatDescriptionCreateForImageBuffer

softlink.

  • platform/cf/CoreMediaSoftLink.h: Ditto.
  • platform/cocoa/CoreVideoSoftLink.cpp: Add CVPixelBufferCreate, kCVPixelBufferCGBitmapContextCompatibilityKey, and

kCVPixelBufferCGImageCompatibilityKey.

  • platform/cocoa/CoreVideoSoftLink.h: Ditto.
  • platform/graphics/avfoundation/MediaSampleAVFObjC.h: Change create to return a Ref<T> instead

of RefPtr<T>.

  • platform/graphics/avfoundation/objc/MediaPlayerPrivateMediaStreamAVFObjC.h: Make observer of

MediaStreamTrackPrivate and make MediaPlayer use an AVSampleBufferDisplayLayer instead of CALayer.

  • platform/graphics/avfoundation/objc/MediaPlayerPrivateMediaStreamAVFObjC.mm: Ditto.

(WebCore::MediaPlayerPrivateMediaStreamAVFObjC::~MediaPlayerPrivateMediaStreamAVFObjC): Clean up
observers and AVSampleBufferDisplayLayer.
(WebCore::MediaPlayerPrivateMediaStreamAVFObjC::isAvailable): Ensures AVSampleBufferDisplayLayer
is available.
(WebCore::MediaPlayerPrivateMediaStreamAVFObjC::enqueueAudioSampleBufferFromTrack): Placeholder.
(WebCore::MediaPlayerPrivateMediaStreamAVFObjC::enqueueVideoSampleBufferFromTrack): Responsible
for enqueuing sample buffers to the active video track.
(WebCore::MediaPlayerPrivateMediaStreamAVFObjC::ensureLayer): Ensures that an AVSampleBufferDisplayLayer
exists.
(WebCore::MediaPlayerPrivateMediaStreamAVFObjC::destroyLayer): Destroys the AVSampleBufferDisplayLayer.
(WebCore::MediaPlayerPrivateMediaStreamAVFObjC::platformLayer): Replace CALayer with AVSampleBufferDisplayLayer.
(WebCore::MediaPlayerPrivateMediaStreamAVFObjC::currentDisplayMode): Ditto.
(WebCore::MediaPlayerPrivateMediaStreamAVFObjC::play): Call updateReadyState as a deferred task.
(WebCore::MediaPlayerPrivateMediaStreamAVFObjC::currentReadyState): readyState is bumped to HAVE_ENOUGH_DATA
only when the MediaPlayerPrivateMediaStreamAVFObjC has received a media sample.
(WebCore::MediaPlayerPrivateMediaStreamAVFObjC::sampleBufferUpdated): Called from MediaStreamTrackPrivate when a
new SampleBuffer is available.
(WebCore::updateTracksOfType): Manage adding and removing self as observer from tracks.
(WebCore::MediaPlayerPrivateMediaStreamAVFObjC::updateTracks): Replace CALayer with AVSampleBufferDisplayLayer
(WebCore::MediaPlayerPrivateMediaStreamAVFObjC::acceleratedRenderingStateChanged): Copied from
MediaPlayerPrivateMediaSourceAVFObjC.mm
(WebCore::MediaPlayerPrivateMediaStreamAVFObjC::load): Deleted CALayer.
(WebCore::MediaPlayerPrivateMediaStreamAVFObjC::updateDisplayMode): Deleted process of updating CALayer.
(WebCore::MediaPlayerPrivateMediaStreamAVFObjC::updateIntrinsicSize): Deleted CALayer.
(WebCore::MediaPlayerPrivateMediaStreamAVFObjC::createPreviewLayers): Deleted.

  • platform/mediastream/MediaStreamPrivate.cpp:

(WebCore::MediaStreamPrivate::updateActiveVideoTrack): Remove redundant check.

  • platform/mediastream/MediaStreamTrackPrivate.cpp:

(WebCore::MediaStreamTrackPrivate::sourceHasMoreMediaData): Called from RealtimeMediaSource when a new SampleBuffer
is available.

  • platform/mediastream/MediaStreamTrackPrivate.h:

(WebCore::MediaStreamTrackPrivate::Observer::sampleBufferUpdated): Relays to MediaPlayerPrivateMediaStreamAVFObjC that
a new SampleBuffer is available to enqueue to the AVSampleBufferDisplayLayer.

  • platform/mediastream/RealtimeMediaSource.cpp:

(WebCore::RealtimeMediaSource::settingsDidChange): Fix grammatical mistake in function name settingsDidChanged().
(WebCore::RealtimeMediaSource::mediaDataUpdated): Relays to all observers that a new SampleBuffer is available.
(WebCore::RealtimeMediaSource::settingsDidChanged): Deleted.

  • platform/mediastream/RealtimeMediaSource.h:
  • platform/mediastream/mac/AVVideoCaptureSource.mm:

(WebCore::AVVideoCaptureSource::processNewFrame): Calls mediaDataUpdated when a new SampleBuffer is captured.

  • platform/mediastream/mac/MockRealtimeVideoSourceMac.h:
  • platform/mediastream/mac/MockRealtimeVideoSourceMac.mm:

(WebCore::MockRealtimeVideoSourceMac::CMSampleBufferFromPixelBuffer): Convert CVPixelBuffer to CMSampleBuffer.
(WebCore::MockRealtimeVideoSourceMac::pixelBufferFromCGImage): Convert CGImage to CVPixelBuffer.
(WebCore::MockRealtimeVideoSourceMac::updateSampleBuffer): Creates a CMSampleBuffer from current imageBuffer and
sends the CMSampleBuffer to MediaPlayerPrivateMediaStreamAVFObjC

  • platform/mock/MockRealtimeVideoSource.cpp:

(WebCore::MockRealtimeVideoSource::setFrameRate): Fix grammar of settingsDidChanged() to settingsDidChange().
(WebCore::MockRealtimeVideoSource::setSize): Ditto.
(WebCore::MockRealtimeVideoSource::generateFrame): Call updateSampleBuffer().

  • platform/mock/MockRealtimeVideoSource.h: Change elapsedTime() from private to protected.

(WebCore::MockRealtimeVideoSource::updateSampleBuffer): Overriden by MockRealtimeVideoSourceMac.

LayoutTests:

  • fast/mediastream/MediaStream-video-element-displays-buffer-expected.txt: Added.
  • fast/mediastream/MediaStream-video-element-displays-buffer.html: Added. Checks that

a video element with a mediastream source displays frames that are neither black or transparent.

  • fast/mediastream/resources/getUserMedia-helper.js:

(setupVideoElementWithStream): Sets up video element with global variable mediastream.

14:49 Changeset [203738] by zalan@apple.com
  • 5 edits in trunk/Source/WebCore

Move ControlStates HashMap to RenderBox.
https://bugs.webkit.org/show_bug.cgi?id=160206

Reviewed by Simon Fraser.

Move and modernize it.

No change in functionality.

  • platform/ControlStates.h:

(WebCore::ControlStates::ControlStates): Deleted.

  • rendering/RenderBox.cpp:

(WebCore::controlStatesRendererMap):
(WebCore::controlStatesForRenderer):
(WebCore::removeControlStatesForRenderer):
(WebCore::RenderBox::~RenderBox):
(WebCore::RenderBox::paintBoxDecorations):

  • rendering/RenderElement.cpp:

(WebCore::controlStatesRendererMap): Deleted.
(WebCore::RenderElement::hasControlStatesForRenderer): Deleted.
(WebCore::RenderElement::controlStatesForRenderer): Deleted.
(WebCore::RenderElement::removeControlStatesForRenderer): Deleted.
(WebCore::RenderElement::addControlStatesForRenderer): Deleted.

  • rendering/RenderElement.h:
14:35 Changeset [203737] by eric.carlson@apple.com
  • 2 edits in trunk/Source/WebCore

Occasional crash in WebCore::RenderVTTCue::initializeLayoutParameters
https://bugs.webkit.org/show_bug.cgi?id=160208

Reviewed by Darin Adler.

  • rendering/RenderVTTCue.cpp:

(WebCore::RenderVTTCue::initializeLayoutParameters): Return when firstChild is NULL so a
release build will not crash.

14:01 Changeset [203736] by rniwa@webkit.org
  • 2 edits in trunk/Websites/perf.webkit.org

REGRESSION: The arrow indicating the current page doesn't get updated
https://bugs.webkit.org/show_bug.cgi?id=160185

Reviewed by Chris Dumez.

The bug was caused by Heading's render() function not updating the DOM more than once. I don't understand
how this has ever worked. Fixed the bug by rendering DOM whenever the current page has changed.

  • public/v3/pages/heading.js:

(Heading):
(Heading.prototype.render):

13:59 Changeset [203735] by rniwa@webkit.org
  • 6 edits
    1 delete in trunk/LayoutTests

Remove the tests for legacy custom elements API
https://bugs.webkit.org/show_bug.cgi?id=160209

Reviewed by Chris Dumez.

Removed the tests for legacy custom elements v0 API. The tests for the new v1 API is at fast/custom-elements.

  • fast/dom/custom: Removed.
  • fast/dom/custom/document-register-basic-expected.txt: Removed.
  • fast/dom/custom/document-register-basic.html: Removed.
  • fast/dom/custom/document-register-namespace-expected.txt: Removed.
  • fast/dom/custom/document-register-namespace.html: Removed.
  • fast/dom/custom/document-register-reentrant-null-constructor-expected.txt: Removed.
  • fast/dom/custom/document-register-reentrant-null-constructor.html: Removed.
  • fast/dom/custom/document-register-reentrant-returning-fake-expected.txt: Removed.
  • fast/dom/custom/document-register-reentrant-returning-fake.html: Removed.
  • fast/dom/custom/document-register-reentrant-throwing-constructor-expected.txt: Removed.
  • fast/dom/custom/document-register-reentrant-throwing-constructor.html: Removed.
  • fast/dom/custom/document-register-type-extensions-expected.txt: Removed.
  • fast/dom/custom/document-register-type-extensions.html: Removed.
  • fast/dom/custom/lifecycle-ready-createElement-recursion-expected.txt: Removed.
  • fast/dom/custom/lifecycle-ready-createElement-recursion.html: Removed.
  • fast/dom/custom/lifecycle-ready-createElement-reentrancy-expected.txt: Removed.
  • fast/dom/custom/lifecycle-ready-createElement-reentrancy.html: Removed.
  • fast/dom/custom/lifecycle-ready-creation-api-expected.txt: Removed.
  • fast/dom/custom/lifecycle-ready-creation-api.html: Removed.
  • fast/dom/custom/lifecycle-ready-innerHTML-expected.txt: Removed.
  • fast/dom/custom/lifecycle-ready-innerHTML.html: Removed.
  • fast/dom/custom/lifecycle-ready-parser-only-expected.html: Removed.
  • fast/dom/custom/lifecycle-ready-parser-only.html: Removed.
  • fast/dom/custom/lifecycle-ready-parser-script-expected.txt: Removed.
  • fast/dom/custom/lifecycle-ready-parser-script.html: Removed.
  • fast/dom/custom/lifecycle-ready-paste-expected.txt: Removed.
  • fast/dom/custom/lifecycle-ready-paste.html: Removed.
  • fast/dom/custom/resources: Removed.
  • fast/dom/custom/resources/document-register-fuzz.js: Removed.
  • platform/efl/TestExpectations:
  • platform/gtk/TestExpectations:
  • platform/ios-simulator/TestExpectations:
  • platform/mac/TestExpectations:
  • platform/win/TestExpectations:
13:57 Changeset [203734] by cdumez@apple.com
  • 4 edits in trunk

Parameters to CustomEvent.initCustomEvent() should be mandatory
https://bugs.webkit.org/show_bug.cgi?id=160205

Reviewed by Ryosuke Niwa.

LayoutTests/imported/w3c:

Rebaseline W3C test now that more checks are passing.

  • web-platform-tests/dom/interfaces-expected.txt:

Source/WebCore:

Parameters to CustomEvent.initCustomEvent() should be mandatory:

Firefox and Chrome agree with the specification.

No new tests, rebaselined existing test.

  • dom/CustomEvent.idl:
13:54 Changeset [203733] by cdumez@apple.com
  • 13 edits in trunk

Second parameter to Range.isPointInRange() / comparePoint() should be mandatory
https://bugs.webkit.org/show_bug.cgi?id=160202

Reviewed by Ryosuke Niwa.

LayoutTests/imported/w3c:

Rebaseline W3C test now that more checks are passing.

  • web-platform-tests/dom/interfaces-expected.txt:

Source/WebCore:

Second parameter to Range.isPointInRange() / comparePoint() should be mandatory
and be of type "unsigned long":

Firefox and Chrome agree with the specification.

No new tests, rebaselined existing tests.

  • accessibility/AXObjectCache.cpp:

(WebCore::AXObjectCache::traverseToOffsetInRange):

  • dom/DocumentMarkerController.cpp:

(WebCore::DocumentMarkerController::removeMarkers):
(WebCore::DocumentMarkerController::markersInRange):
(DocumentMarkerController::setMarkersActive):

  • dom/Range.cpp:

(WebCore::Range::isPointInRange):
(WebCore::Range::comparePoint):
(WebCore::Range::compareBoundaryPoints):
(WebCore::Range::toString):
(WebCore::Range::absoluteTextRects):
(WebCore::Range::absoluteTextQuads):
(WebCore::boundaryTextNodesMerged):
(WebCore::Range::getBorderAndTextQuads):

  • dom/Range.h:

(WebCore::Range::startOffset):
(WebCore::Range::endOffset):

  • dom/Range.idl:
  • dom/RangeBoundaryPoint.h:

(WebCore::RangeBoundaryPoint::ensureOffsetIsValid):
(WebCore::RangeBoundaryPoint::toPosition):
(WebCore::RangeBoundaryPoint::offset):
(WebCore::RangeBoundaryPoint::setOffset):
(WebCore::RangeBoundaryPoint::setToBeforeChild):
(WebCore::RangeBoundaryPoint::setToAfterChild):
(WebCore::RangeBoundaryPoint::setToEndOfNode):
(WebCore::RangeBoundaryPoint::childBeforeWillBeRemoved):
(WebCore::RangeBoundaryPoint::invalidateOffset):

LayoutTests:

Update existing test to reflect behavior change.

  • fast/dom/non-numeric-values-numeric-parameters-expected.txt:
  • fast/dom/script-tests/non-numeric-values-numeric-parameters.js:
13:18 Changeset [203732] by commit-queue@webkit.org
  • 23 edits
    8 copies
    11 adds in trunk

[Fetch API] Add support for fetch mode, in particular cors
https://bugs.webkit.org/show_bug.cgi?id=156753

Patch by Youenn Fablet <youenn@apple.com> on 2016-07-26
Reviewed by Darin Adler.

LayoutTests/imported/w3c:

Rebasing tests.

  • web-platform-tests/fetch/api/basic/integrity-expected.txt:
  • web-platform-tests/fetch/api/basic/integrity-worker-expected.txt:
  • web-platform-tests/fetch/api/basic/mode-no-cors-expected.txt:
  • web-platform-tests/fetch/api/basic/mode-no-cors-worker-expected.txt:
  • web-platform-tests/fetch/api/cors/cors-basic-expected.txt:
  • web-platform-tests/fetch/api/cors/cors-basic-worker-expected.txt:
  • web-platform-tests/fetch/api/cors/cors-cookies-expected.txt:
  • web-platform-tests/fetch/api/cors/cors-cookies-worker-expected.txt:
  • web-platform-tests/fetch/api/cors/cors-filtering-expected.txt:
  • web-platform-tests/fetch/api/cors/cors-filtering-worker-expected.txt:
  • web-platform-tests/fetch/api/cors/cors-multiple-origins-expected.txt:
  • web-platform-tests/fetch/api/cors/cors-multiple-origins-worker-expected.txt:
  • web-platform-tests/fetch/api/cors/cors-preflight-expected.txt:
  • web-platform-tests/fetch/api/cors/cors-preflight-referrer-worker-expected.txt:
  • web-platform-tests/fetch/api/cors/cors-preflight-worker-expected.txt:
  • web-platform-tests/fetch/api/cors/cors-redirect-credentials-expected.txt:
  • web-platform-tests/fetch/api/cors/cors-redirect-credentials-worker-expected.txt:
  • web-platform-tests/fetch/api/credentials/authentication-basic-worker-expected.txt:
  • web-platform-tests/fetch/api/request/request-cache-expected.txt:

Source/WebCore:

Covered by rebased tests.

  • Modules/fetch/FetchLoader.cpp:

(WebCore::FetchLoader::start): Passing fetch mode to ThreadableLoader.
Disabling as a temp fix credentials in case of CORS mode as credential options is not yet supported and would make several tests fail.

LayoutTests:

Rebasing specific expectations as Maci/iOS WK2 does not like https tests.

  • platform/ios-simulator-wk2/imported/w3c/web-platform-tests/fetch/api/basic/mode-no-cors-expected.txt: Added.
  • platform/ios-simulator-wk2/imported/w3c/web-platform-tests/fetch/api/basic/mode-no-cors-worker-expected.txt: Added.
  • platform/ios-simulator-wk2/imported/w3c/web-platform-tests/fetch/api/cors/cors-basic-expected.txt: Added.
  • platform/ios-simulator-wk2/imported/w3c/web-platform-tests/fetch/api/cors/cors-basic-worker-expected.txt: Added.
  • platform/mac-wk2/imported/w3c/web-platform-tests/fetch/api/basic/mode-no-cors-expected.txt: Added.
  • platform/mac-wk2/imported/w3c/web-platform-tests/fetch/api/basic/mode-no-cors-worker-expected.txt: Added.
  • platform/mac-wk2/imported/w3c/web-platform-tests/fetch/api/cors/cors-basic-expected.txt: Added.
  • platform/mac-wk2/imported/w3c/web-platform-tests/fetch/api/cors/cors-basic-worker-expected.txt: Added.
12:09 Changeset [203731] by cdumez@apple.com
  • 10 edits
    2 adds
    4 deletes in trunk

Align NamedNodeMap with the specification
https://bugs.webkit.org/show_bug.cgi?id=160204

Reviewed by Darin Adler.

LayoutTests/imported/w3c:

Rebaseline W3C test now that more checks are passing.

  • web-platform-tests/dom/interfaces-expected.txt:

Source/WebCore:

Align NamedNodeMap with the specification:

In particular, mark parameters as mandatory when they should be, and
use tighter "Attr" typing instead of Node.

Chrome and Firefox agree with the specification.

Test: fast/dom/NamedNodeMap-parameters.html

  • dom/NamedNodeMap.cpp:

(WebCore::NamedNodeMap::getNamedItem):
(WebCore::NamedNodeMap::getNamedItemNS):
(WebCore::NamedNodeMap::removeNamedItem):
(WebCore::NamedNodeMap::removeNamedItemNS):
(WebCore::NamedNodeMap::setNamedItem):
(WebCore::NamedNodeMap::item):

  • dom/NamedNodeMap.h:
  • dom/NamedNodeMap.idl:

LayoutTests:

  • dom/html/level2/core/hc_namednodemapinvalidtype1-expected.txt: Removed.
  • dom/html/level2/core/hc_namednodemapinvalidtype1.html: Removed.
  • dom/xhtml/level2/core/hc_namednodemapinvalidtype1-expected.txt: Removed.
  • dom/xhtml/level2/core/hc_namednodemapinvalidtype1.xhtml: Removed.

Drop outdated DOM level 2 tests that expect the wrong exception type to
be thrown when passing a non-Attr node in.

  • fast/dom/NamedNodeMap-parameters-expected.txt: Added.
  • fast/dom/NamedNodeMap-parameters.html: Added.

Add layout test coverage. I have verified that this test is passing in
both Firefox and Chrome.

  • fast/dom/NamedNodeMap-setNamedItem-crash-expected.txt:
  • fast/dom/non-numeric-values-numeric-parameters-expected.txt:
  • fast/dom/script-tests/non-numeric-values-numeric-parameters.js:

Update / rebaseline existing tests to reflect behavior change.

11:58 Changeset [203730] by mark.lam@apple.com
  • 2 edits in trunk/Source/JavaScriptCore

Remove unused DEBUG_WITH_BREAKPOINT configuration.
https://bugs.webkit.org/show_bug.cgi?id=160203

Reviewed by Keith Miller.

  • bytecompiler/BytecodeGenerator.cpp:

(JSC::BytecodeGenerator::emitDebugHook):

11:58 Changeset [203729] by commit-queue@webkit.org
  • 4 edits
    2 adds in trunk

Infinite Canvas context save() causes WebKit to crash
https://bugs.webkit.org/show_bug.cgi?id=159586
<rdar://problem/26759984>

Patch by Said Abou-Hallawa <sabouhallawa@apple.com> on 2016-07-26
Reviewed by Simon Fraser.

Source/WebCore:

Limit the size of the canvas context state stack to 1024 * 16 saves. All
the saves which come after that limit will stay unrealized. The restore()
should not have any effect till there is no unrealized saves.

Test: fast/canvas/canvas-context-save-limit.html

  • html/canvas/CanvasRenderingContext2D.cpp:

(WebCore::CanvasRenderingContext2D::realizeSaves):
(WebCore::CanvasRenderingContext2D::realizeSavesLoop):

  • html/canvas/CanvasRenderingContext2D.h:

LayoutTests:

  • fast/canvas/canvas-context-save-limit-expected.txt: Added.
  • fast/canvas/canvas-context-save-limit.html: Added.
11:00 Changeset [203728] by commit-queue@webkit.org
  • 5 edits
    3 adds in trunk

DOMTokenList should be iterable
https://bugs.webkit.org/show_bug.cgi?id=160183

Patch by Youenn Fablet <youenn@apple.com> on 2016-07-26
Reviewed by Chris Dumez.

Source/WebCore:

DOMTokenList should be iterable as per https://dom.spec.whatwg.org/#interface-domtokenlist

Test: fast/dom/domTokenListIterator.html

  • html/DOMTokenList.idl: Added iterable to the interface description.

LayoutTests:

  • fast/dom/domTokenListIterator-expected.txt: Added.
  • fast/dom/domTokenListIterator.html: Added.
  • fast/dom/iterable-tests.js: Added.
  • fast/dom/nodeListIterator-expected.txt:
  • fast/dom/nodeListIterator.html: Making use of iterable-tests.js
10:26 Changeset [203727] by carlosgc@webkit.org
  • 2 edits in trunk/Source/WebKit2

[Threaded Compositor] ASSERTION FAILED: canAccessThreadLocalDataForThread(m_thread) after r203718
https://bugs.webkit.org/show_bug.cgi?id=160201

Reviewed by Michael Catanzaro.

I forgot to call purgeGLResources() before invalidating the scene in the compositing thread.

  • Shared/CoordinatedGraphics/threadedcompositor/ThreadedCompositor.cpp:

(WebKit::ThreadedCompositor::invalidate):

10:13 Changeset [203726] by commit-queue@webkit.org
  • 21 edits
    2 deletes in trunk

Unreviewed, rolling out r203719.
https://bugs.webkit.org/show_bug.cgi?id=160200

It is breaking win build (Requested by youenn on #webkit).

Reverted changeset:

"[Fetch API] Response constructor should be able to take a
ReadableStream as body"
https://bugs.webkit.org/show_bug.cgi?id=159804
http://trac.webkit.org/changeset/203719

09:57 Changeset [203725] by wilander@apple.com
  • 3 edits
    6 adds in trunk

Stop supporting compressed character sets BOCU-1 and SCSU
https://bugs.webkit.org/show_bug.cgi?id=159581

Reviewed by Brent Fulgham.

WebKit should not support the compressed character sets BOCU-1 and SCSU.
Chrome and Firefox don't and these old formats may pass server-side character
filters while still rendering in WebKit.

The HTML specification says "The above prohibits supporting, for example,
CESU-8, UTF-7, BOCU-1, SCSU, EBCDIC, and UTF-32."
https://html.spec.whatwg.org/#character-encodings

Source/WebCore:

Tests: http/tests/misc/char-encoding-bocu-1-blacklisted.html

http/tests/misc/char-encoding-scsu-blacklisted.html

  • platform/text/TextEncodingRegistry.cpp:

Blacklisted BOCU-1 and SCSU character sets.

LayoutTests:

  • http/tests/misc/char-encoding-bocu-1-blacklisted-expected.txt: Added.
  • http/tests/misc/char-encoding-bocu-1-blacklisted.html: Added.
  • http/tests/misc/char-encoding-scsu-blacklisted-expected.txt: Added.
  • http/tests/misc/char-encoding-scsu-blacklisted.html: Added.
  • http/tests/misc/resources/bocu-1-cyrillic.php: Added.
  • http/tests/misc/resources/scsu-cyrillic.php: Added.
09:32 Changeset [203724] by enrica@apple.com
  • 2 edits in trunk/Source/WebKit2

Support configurable autocapitalization.
https://bugs.webkit.org/show_bug.cgi?id=158860
rdar://problem/27536113

Reviewed by Tim Horton.

Autocapitalization should be enabled/disabled regardless of whether
we are using advance spelling feature.

  • UIProcess/mac/TextCheckerMac.mm:

(WebKit::TextChecker::checkTextOfParagraph):
(WebKit::TextChecker::getGuessesForWord):

09:29 Changeset [203723] by carlosgc@webkit.org
  • 2 edits in trunk/Source/WebCore
ASSERTION FAILED: !m_frame->page()->defersLoading()
InspectorInstrumentation::isDebuggerPaused(m_frame)

https://bugs.webkit.org/show_bug.cgi?id=160193

Reviewed by Michael Catanzaro.

This is happening in the GTK+ Debug bot when running test loader/load-defer.html (note that the assert is inside
a !USE(CF) block).
The test is creating an iframe with load deferred, then in a timeout it disables the deferred load and checks
that the load actually happens. What happens is that the initial empty document is what calls
DocumentLoader::finishedLoading() when load is still deferred. The onload handler is not called because load
events are disabled for the initial empty document in SubframeLoader::loadSubframe(), but
DocumentLoader::finishedLoading() is called unconditionally from maybeLoadEmpty(). I think it's fine to call
DocumentLoader::finishedLoading() for the initial empty document even when load is deferred, so we can simply
update the assert to handle that case.

  • loader/DocumentLoader.cpp:

(WebCore::DocumentLoader::finishedLoading): Do not assert if called for the initial empty document when load is
deferred.

09:26 Changeset [203722] by carlosgc@webkit.org
  • 3 edits in trunk/Source/WebKit2

[Coordinated Graphics] Test fast/fixed-layout/fixed-layout.html crashes in debug
https://bugs.webkit.org/show_bug.cgi?id=160117

Reviewed by Michael Catanzaro.

The problem is that WebPage has its own m_useFixedLayout that is only updated when changed from the UI
process. However, layout tests doing internals.setUseFixedLayout() change the frame view directly, and the
WebPage doesn't notice it.

  • WebProcess/WebPage/WebPage.cpp:

(WebKit::WebPage::setFixedVisibleContentRect): Deleted.
(WebKit::WebPage::sendViewportAttributesChanged): Change the assert to check the main FrameView is in fixed
layout mode.

  • WebProcess/WebPage/WebPage.h:
09:25 Changeset [203721] by carlosgc@webkit.org
  • 5 edits in trunk/Source/WebKit2

[Threaded Compositor] ASSERTION FAILED: isMainThread() when ThreadedCompositor is destroyed since r203718
https://bugs.webkit.org/show_bug.cgi?id=160197

Reviewed by Žan Doberšek.

ThreadedCompositor can be destroyed from a secondary thread, for example, when a task takes a reference and the
main threads derefs it, when the task finishes in the secondary thread the lambda ends up deleting the threaded
compositor. This is ok for the Threaded compositor but not for the CompositingRunLoop class. this was not a
problem before r203718 because the CompositingRunLoop object was created and destroyed in the same thread
always, but now it's part of the ThreadedCompositor class. This patch uses std:unique_ptr again to explicitly
create the CompositingRunLoop in the ThreadedCompositor constructor and delete in the invalidate() method to
make sure it happens in the main thread in both cases.

  • Shared/CoordinatedGraphics/threadedcompositor/CompositingRunLoop.cpp:

(WebKit::WorkQueuePool::invalidate):
(WebKit::WorkQueuePool::getOrCreateWorkQueueForContext):

  • Shared/CoordinatedGraphics/threadedcompositor/CompositingRunLoop.h:
  • Shared/CoordinatedGraphics/threadedcompositor/ThreadedCompositor.cpp:

(WebKit::ThreadedCompositor::ThreadedCompositor):
(WebKit::ThreadedCompositor::invalidate):
(WebKit::ThreadedCompositor::setNativeSurfaceHandleForCompositing):
(WebKit::ThreadedCompositor::setDeviceScaleFactor):
(WebKit::ThreadedCompositor::setDrawsBackground):
(WebKit::ThreadedCompositor::didChangeViewportSize):
(WebKit::ThreadedCompositor::didChangeViewportAttribute):
(WebKit::ThreadedCompositor::didChangeContentsSize):
(WebKit::ThreadedCompositor::scrollTo):
(WebKit::ThreadedCompositor::scrollBy):
(WebKit::ThreadedCompositor::updateViewport):
(WebKit::ThreadedCompositor::scheduleDisplayImmediately):
(WebKit::ThreadedCompositor::forceRepaint):

  • Shared/CoordinatedGraphics/threadedcompositor/ThreadedCompositor.h:
09:23 Changeset [203720] by commit-queue@webkit.org
  • 25 edits in trunk

Remove ClientCredentialPolicy cross-origin option from ResourceLoaderOptions
https://bugs.webkit.org/show_bug.cgi?id=159413

Patch by Youenn Fablet <youenn@apple.com> on 2016-07-26
Reviewed by Alex Christensen.

LayoutTests/imported/w3c:

Below test changes as ResourceLoader is now computing whether to request credentials to clients if:

  • request is authorized to request credentials (DocumentThreadableLoader only allows same-origin to make such thing)
  • credential policy is Include or Same-Origin and request is same-origin.

This test changes as current fetch loader sets the credential mode to Omit, thus disabling credential request.

To be noted that only fetch API is allowing to disable credentials sending for same-origin request using "Omit"
credential mode.

  • web-platform-tests/fetch/api/credentials/authentication-basic-expected.txt: Rebasing test.

Source/WebCore:

ClientCredentialPolicy had three values (not ask, ask, ask only for same origin).
The distinction between allowing cross-origin or same-origin credentials is misleading as it is not supported
for synchronous loads and not supported by Network process.
It is best replaced by a boolean option (ask or not ask).

Same-origin ClientCredentialPolicy option was only used by DocumentThreadableLoader for asynchronous loads.
Since DocumentThreadableLoader is already computing whether the request is cross-origin, it can also compute
whether credentials may be requested or not. In case of cross-origin redirections, credentials are omitted, thus
disabling any possibility for requesting credentials for cross-origin resources after redirections.

Moving ClientCredentialPolicy to ResourceLoaderOptions since it is not used by platform code except for some
mac-specific code that is already using ResourceLoaderOptions.

Covered by existing tests.

  • loader/CrossOriginPreflightChecker.cpp:

(WebCore::CrossOriginPreflightChecker::startPreflight): Setting clearly credential mode to Omit credentials.
(WebCore::CrossOriginPreflightChecker::doPreflight): Ditto.

  • loader/DocumentLoader.cpp:

(WebCore::DocumentLoader::startLoadingMainResource): AskClientForAllCredentials ->
ClientCredentialPolicy::MayAskClientForCredentials.

  • loader/DocumentThreadableLoader.cpp:

(WebCore::DocumentThreadableLoader::loadRequest): Disabling requesting credentials for any cross-origin request.

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

(WebCore::MediaResourceLoader::requestResource): AskClientForAllCredentials -> ClientCredentialPolicy::MayAskClientForCredentials.

  • loader/NetscapePlugInStreamLoader.cpp:

(WebCore::NetscapePlugInStreamLoader::NetscapePlugInStreamLoader): Ditto.

  • loader/ResourceLoader.cpp:

(WebCore::ResourceLoader::isAllowedToAskUserForCredentials): Disabling client credential request if ClientCredentialPolicy is CannotAskClientForCredentials.
Otherwise, returns true if fetch credentials mode allows it.

  • loader/ResourceLoaderOptions.h:

(WebCore::ResourceLoaderOptions::ResourceLoaderOptions):
(WebCore::ResourceLoaderOptions::clientCredentialPolicy): Deleted.
(WebCore::ResourceLoaderOptions::setClientCredentialPolicy): Deleted.

  • loader/cache/CachedResourceLoader.cpp:

(WebCore::CachedResourceLoader::requestUserCSSStyleSheet): AskClientForAllCredentials -> ClientCredentialPolicy::MayAskClientForCredentials.
(WebCore::CachedResourceLoader::defaultCachedResourceOptions): AskClientForAllCredentials -> ClientCredentialPolicy::MayAskClientForCredentials.

  • loader/icon/IconLoader.cpp:

(WebCore::IconLoader::startLoading): DoNotAskClientForAnyCredentials -> ClientCredentialPolicy::CannotAskClientForCredentials.

  • platform/graphics/avfoundation/cf/WebCoreAVCFResourceLoader.cpp:

(WebCore::WebCoreAVCFResourceLoader::startLoading): DoNotAskClientForCrossOriginCredentials -> ClientCredentialPolicy::CannotAskClientForCredentials.
This is ok as credentials mode is omit and stored credentials are not allowed.

  • platform/graphics/avfoundation/objc/WebCoreAVFResourceLoader.mm:

(WebCore::WebCoreAVFResourceLoader::startLoading): Ditto.

  • platform/network/ResourceHandleTypes.h:
  • xml/XSLTProcessorLibxslt.cpp: DoNotAskClientForCrossOriginCredentials -> ClientCredentialPolicy::MayAskClientForCredentials.

This is ok as this is a synchronous load.
(WebCore::docLoaderFunc):

  • xml/parser/XMLDocumentParserLibxml2.cpp:

(WebCore::openFunc): DoNotAskClientForCrossOriginCredentials -> ClientCredentialPolicy::MayAskClientForCredentials.
This is ok as this is a synchronous load.

Source/WebKit2:

Renaming of ClientCredentialPolicy values.

  • NetworkProcess/Downloads/DownloadManager.cpp:

(WebKit::DownloadManager::startDownload):

  • NetworkProcess/NetworkLoad.cpp:

(WebKit::NetworkLoad::continueCanAuthenticateAgainstProtectionSpace):
(WebKit::NetworkLoad::didReceiveAuthenticationChallenge):

  • NetworkProcess/NetworkLoadParameters.h:
  • NetworkProcess/NetworkResourceLoadParameters.cpp:

(WebKit::NetworkResourceLoadParameters::NetworkResourceLoadParameters):

  • NetworkProcess/NetworkResourceLoader.cpp:

(WebKit::NetworkResourceLoader::NetworkResourceLoader):

  • WebProcess/Network/WebLoaderStrategy.cpp:

(WebKit::WebLoaderStrategy::scheduleLoad):

  • WebProcess/Network/WebResourceLoader.cpp:

(WebKit::WebResourceLoader::willSendRequest):

LayoutTests:

  • platform/mac-wk1/imported/w3c/web-platform-tests/fetch/api/credentials/authentication-basic-expected.txt: Removed.
09:03 Changeset [203719] by commit-queue@webkit.org
  • 21 edits
    1 copy
    1 add in trunk

[Fetch API] Response constructor should be able to take a ReadableStream as body
https://bugs.webkit.org/show_bug.cgi?id=159804

Patch by Youenn Fablet <youenn@apple.com> on 2016-07-26
Reviewed by Alex Christensen.

LayoutTests/imported/w3c:

  • web-platform-tests/fetch/api/response/response-consume-empty-expected.txt:
  • web-platform-tests/fetch/api/response/response-consume-expected.txt:
  • web-platform-tests/fetch/api/response/response-consume.html: Updating test to exercice Response coonstructor with a ReadableStream.

Source/WebCore:

Covered by existing and updated tests.

Introduced FetchBodyConsumer to encapsulate the code responsible to adapt FetchBody internal data to the requests made by user scripts.
This refactoring eases the handling of internal data coming from ReadableStream.

FetchLoader is now delegating conversion from the data to its m_consumer field.
If m_consumer is null, FetchLoader calls FetchLoaderClient::didReceiveData (streaming reception mode).
Clients of FetchLoader needs to pass a FetchBodyConsumer to the FetchLoader to do the data adaptation at loader creation time.

Added support for body data passed as a ReadableStream to Response.
This requires to set @body internal slot of the Response object in the constructor initialization JS built-in.

To actually use that data, Body accessors are also implemented as JS built-in for Response.
Since there is no need to do so for Request, FetchResponse is no longer marked as implementing FetchBody but all
FetchBody IDL description is inlined in FetchResponse.idl.

For each body accessor (arrayBuffer, blob, json, text), the corresponding JS built-in checks whether @body internal slot is set.
If that is not the case, regular handling is done through a new private method called @consume.
If @body internal slot is set, chunks are pumped from the ReadableStream using ReadableStream internal built-ins functions.
Data handling is done in C++ through the private methods @startConsumingStream, @consumeChunk and @finishConsumingStream.

To support cloning of Response with bodies, clone method is also implemented as a JS built-in.
Main clone is done through @cloneFoJS private method implemented in C++.
JS built-in clone code does the teeing of the ReadableStream using ReadableStream JS built-in internal functions.

Introducing a new ReadableStream type for FetchBody to cope with body being stored in a ReadableStream.

Introducing a new Loaded type for FetchBody to cope with body being stored in the FetchBodyConsumer owned by FetchBody.
This allows removing the conversion of data to JSC::ArrayBuffer which was done by defaut at the end of Fetch loading if user script did not request data before.
At the end of a load, the data remains in FetchBodyConsumer and the body is marked as Loaded.
If the user wants to get the data after data finished being loaded, FetchBodyConsumer will do the required conversions.

Introducing DeferredWrapper::resolveWithNewValue to handle the case of resolving promises with new objects.
This allows directly using toJSNewlyCreated instead of toJS.

  • CMakeLists.txt: Adding FetchBodyConsumer.
  • Modules/fetch/FetchBody.cpp: Moving data adaptation code to FetchBodyConsumer.

(WebCore::FetchBody::extract):
(WebCore::FetchBody::arrayBuffer):
(WebCore::FetchBody::blob): Setting contentType in FetchBodyConsumer to handle proper creation of blob.
(WebCore::FetchBody::json):
(WebCore::FetchBody::text):
(WebCore::FetchBody::consume): Refactoring and added case of Loaded bodies.
(WebCore::FetchBody::consumeAsStream): Ditto.
(WebCore::FetchBody::consumeArrayBuffer):
(WebCore::FetchBody::consumeText):
(WebCore::FetchBody::consumeBlob):
(WebCore::FetchBody::loadingFailed):
(WebCore::FetchBody::loadingSucceeded):
(WebCore::FetchBody::loadingType): Deleted.
(WebCore::blobFromArrayBuffer): Deleted.
(WebCore::FetchBody::fulfillTextPromise): Deleted.
(WebCore::FetchBody::loadedAsArrayBuffer): Deleted.
(WebCore::FetchBody::loadedAsText): Deleted.

  • Modules/fetch/FetchBody.h:

(WebCore::FetchBody::consumer):

  • Modules/fetch/FetchBodyConsumer.cpp: Added, responsible of data adaptation.

(WebCore::blobFromData):
(WebCore::FetchBodyConsumer::resolveWithData):
(WebCore::FetchBodyConsumer::resolve):
(WebCore::FetchBodyConsumer::append):
(WebCore::FetchBodyConsumer::takeData):
(WebCore::FetchBodyConsumer::takeAsArrayBuffer):
(WebCore::FetchBodyConsumer::takeAsBlob):
(WebCore::FetchBodyConsumer::takeAsText):

  • Modules/fetch/FetchBodyConsumer.h: Added.

(WebCore::FetchBodyConsumer::FetchBodyConsumer):
(WebCore::FetchBodyConsumer::setContentType):
(WebCore::FetchBodyConsumer::setType):
(WebCore::FetchBodyConsumer::type):
(WebCore::FetchBodyConsumer::clean):
(WebCore::FetchBodyConsumer::hasData):

  • Modules/fetch/FetchBodyOwner.cpp:

(WebCore::FetchBodyOwner::loadBlob):
(WebCore::FetchBodyOwner::blobLoadingSucceeded):
(WebCore::FetchBodyOwner::loadedBlobAsText): Deleted.

  • Modules/fetch/FetchBodyOwner.h:

(WebCore::FetchBodyOwner::loadedBlobAsArrayBuffer): Deleted.

  • Modules/fetch/FetchInternals.js:

(consumeStream): Pump ReadableStream data and send it to FetchResponse to be converted according user need.

  • Modules/fetch/FetchLoader.cpp:

(WebCore::FetchLoader::FetchLoader): FetchLoader is simplified as it has two nodes: consumer mode in which case
data is sent to the consumer and no-consumer mode in which case data is passed to loader client. The second mode
is used to conveyy data to ReadableStream source.
(WebCore::FetchLoader::stop):
(WebCore::FetchLoader::startStreaming):
(WebCore::FetchLoader::didReceiveData):
(WebCore::FetchLoader::didFinishLoading): Deleted.

  • Modules/fetch/FetchLoader.h:
  • Modules/fetch/FetchLoaderClient.h:

(WebCore::FetchLoaderClient::didFinishLoadingAsText): Deleted.
(WebCore::FetchLoaderClient::didFinishLoadingAsArrayBuffer): Deleted.

  • Modules/fetch/FetchResponse.cpp:

(WebCore::FetchResponse::cloneForJS):
(WebCore::FetchResponse::BodyLoader::didSucceed):
(WebCore::FetchResponse::BodyLoader::start):
(WebCore::FetchResponse::consume): Introduced to consume data stored in body and not in ReadableStream.
(WebCore::FetchResponse::startConsumingStream): Introduced to start process of consuming data stored in the ReadableStream.
(WebCore::FetchResponse::consumeChunk): Reception of ReadableStream data.
(WebCore::FetchResponse::finishConsumingStream): Doing the final conversion.
(WebCore::FetchResponse::clone): Deleted.
(WebCore::FetchResponse::BodyLoader::didFinishLoadingAsArrayBuffer): Deleted.

  • Modules/fetch/FetchResponse.h:
  • Modules/fetch/FetchResponse.idl: Inlining of FetchBody methods/attributes and adding private methods.
  • Modules/fetch/FetchResponse.js:

(initializeFetchResponse):
(clone):
(arrayBuffer):
(blob):
(formData):
(json):
(text):

  • WebCore.xcodeproj/project.pbxproj:
  • bindings/js/JSDOMPromise.h:

(WebCore::DeferredWrapper::resolveWithNewValue):

  • bindings/js/WebCoreBuiltinNames.h:
06:26 Changeset [203718] by carlosgc@webkit.org
  • 5 edits in trunk/Source/WebKit2

[Threaded Compositor] Crashes and deadlocks in single web process mode
https://bugs.webkit.org/show_bug.cgi?id=160160

Reviewed by Žan Doberšek.

Every WebPage has its own threaded compositor that runs its own compositing thread. That means that when there's
more than one WebPage in the same process, we are running OpenGL stuff in different secondary threads. That's
causing crashes and deadlocks in X and graphics drivers. We should ensure there's a single compositing thread
per process when multiple threads is not supported. This is causing unit test
WebKit2.WKPageGetScaleFactorNotZero to time out since we switched to the threaded compositor. That test is
creating two pages in the same web process, and most of the times the web process crashes or deadlocks causing
the test to never finish and time out.
This patch makes CompositingRunLoop use a thread pool that spawns the compositing threads and schedules the tasks
there.

  • Shared/CoordinatedGraphics/threadedcompositor/CompositingRunLoop.cpp:

(WebKit::WorkQueuePool::singleton):
(WebKit::WorkQueuePool::dispatch):
(WebKit::WorkQueuePool::runLoop):
(WebKit::WorkQueuePool::invalidate):
(WebKit::WorkQueuePool::WorkQueuePool):
(WebKit::WorkQueuePool::getOrCreateWorkQueueForContext):
(WebKit::CompositingRunLoop::CompositingRunLoop):
(WebKit::CompositingRunLoop::~CompositingRunLoop):
(WebKit::CompositingRunLoop::performTask):
(WebKit::CompositingRunLoop::performTaskSync):

  • Shared/CoordinatedGraphics/threadedcompositor/CompositingRunLoop.h:
  • Shared/CoordinatedGraphics/threadedcompositor/ThreadedCompositor.cpp:

(WebKit::ThreadedCompositor::ThreadedCompositor):
(WebKit::ThreadedCompositor::invalidate):
(WebKit::ThreadedCompositor::setNativeSurfaceHandleForCompositing):
(WebKit::ThreadedCompositor::setDeviceScaleFactor):
(WebKit::ThreadedCompositor::setDrawsBackground):
(WebKit::ThreadedCompositor::didChangeViewportSize):
(WebKit::ThreadedCompositor::didChangeViewportAttribute):
(WebKit::ThreadedCompositor::didChangeContentsSize):
(WebKit::ThreadedCompositor::scrollTo):
(WebKit::ThreadedCompositor::scrollBy):
(WebKit::ThreadedCompositor::purgeBackingStores):
(WebKit::ThreadedCompositor::renderNextFrame):
(WebKit::ThreadedCompositor::commitScrollOffset):
(WebKit::ThreadedCompositor::updateViewport):
(WebKit::ThreadedCompositor::scheduleDisplayImmediately):
(WebKit::ThreadedCompositor::forceRepaint):
(WebKit::ThreadedCompositor::tryEnsureGLContext): Deleted.
(WebKit::ThreadedCompositor::glContext): Deleted.
(WebKit::ThreadedCompositor::updateSceneState): Deleted.

  • Shared/CoordinatedGraphics/threadedcompositor/ThreadedCompositor.h:
04:08 Changeset [203717] by svillar@igalia.com
  • 19 edits in trunk

[css-grid] repeat() syntax should take a <track-list> argument
https://bugs.webkit.org/show_bug.cgi?id=160162

Reviewed by Darin Adler.

Source/WebCore:

The repeat() notation used to allow just 1 <track-size> as second argument. Specs have been
recently modified so that a <track-list> is now supported, meaning that we can pass an
arbitrary number of track sizes and line numbers.

It has been working for some time for repeat() if the first argument was a positive integer,
but it requires some changes for the auto repeat cases (auto-fill and auto-fit).

  • css/CSSComputedStyleDeclaration.cpp:

(WebCore::OrderedNamedLinesCollector::OrderedNamedLinesCollector): Store the total number of
auto repeat tracks and the length of a single repetition instead of the number of
repetitions.
(WebCore::OrderedNamedLinesCollector::collectLineNamesForIndex): Do not assume that there is
only 1 repeat track.
(WebCore::valueForGridTrackList):

  • css/CSSParser.cpp:

(WebCore::CSSParser::parseGridTrackRepeatFunction): Allow multiple tracks in repeat().

  • rendering/RenderGrid.cpp:

(WebCore::RenderGrid::rawGridTrackSize): Renamed repetitions -> autoRepeatTracksCount.
(WebCore::RenderGrid::computeAutoRepeatTracksCount): Use all the repeat tracks to compute
the total track size of a single repetition.
(WebCore::RenderGrid::computeEmptyTracksForAutoRepeat):

  • rendering/style/GridPositionsResolver.cpp:

(WebCore::NamedLineCollection::NamedLineCollection): Renamed m_repetitions ->
m_autoRepeatTotalTracks. Added m_autoRepeatTrackListLength (it was always 1 before).
(WebCore::NamedLineCollection::find): Resolve lines inside multiple repeat tracks.
(WebCore::NamedLineCollection::firstPosition): Ditto.

  • rendering/style/GridPositionsResolver.h:

LayoutTests:

Added new test cases with multiple tracks inside repeat() notation, both for fixed an
automatic (auto-fill & auto-fit) repetitions.

  • fast/css-grid-layout/grid-auto-fill-columns-expected.txt:
  • fast/css-grid-layout/grid-auto-fill-columns.html:
  • fast/css-grid-layout/grid-auto-fill-rows-expected.txt:
  • fast/css-grid-layout/grid-auto-fill-rows.html:
  • fast/css-grid-layout/grid-auto-fit-columns-expected.txt:
  • fast/css-grid-layout/grid-auto-fit-columns.html:
  • fast/css-grid-layout/grid-auto-fit-rows-expected.txt:
  • fast/css-grid-layout/grid-auto-fit-rows.html:
  • fast/css-grid-layout/grid-element-auto-repeat-get-set-expected.txt:
  • fast/css-grid-layout/grid-element-auto-repeat-get-set.html:
  • fast/css-grid-layout/grid-element-repeat-get-set-expected.txt:
  • fast/css-grid-layout/grid-element-repeat-get-set.html:
02:37 Changeset [203716] by svillar@igalia.com
  • 15 edits in trunk

[css-grid] grid-auto-flow|row should take a <track-size>+
https://bugs.webkit.org/show_bug.cgi?id=160158

Reviewed by Darin Adler.

Source/WebCore:

  • css/CSSComputedStyleDeclaration.cpp:

(WebCore::valueForGridTrackSizeList):
(WebCore::ComputedStyleExtractor::propertyValue): Return a list of <track-size> instead of
just one.

  • css/CSSParser.cpp:

(WebCore::CSSParser::parseValue): Use the new values of TrackListType;
(WebCore::CSSParser::parseGridTemplateRowsAndAreasAndColumns): Ditto.
(WebCore::CSSParser::parseGridTemplateShorthand): Ditto.
(WebCore::CSSParser::parseGridShorthand): Ditto.
(WebCore::CSSParser::parseGridTrackList): Changed behavior depending on the value of
TrackSizeList.

  • css/CSSParser.h: TrackListType has now 3 different values which determine the behavior of

parseGridTrackList.

  • css/CSSPropertyNames.in: Use a new converter for lists.
  • css/StyleBuilderConverter.h:

(WebCore::StyleBuilderConverter::convertGridTrackSizeList):

  • rendering/RenderGrid.cpp:

(WebCore::RenderGrid::rawGridTrackSize): Resolve the size of the auto track.

  • rendering/style/RenderStyle.h:

(WebCore::RenderStyle::gridAutoColumns): Return a Vector.
(WebCore::RenderStyle::gridAutoRows): Ditto.
(WebCore::RenderStyle::setGridAutoColumns): Store a Vector.
(WebCore::RenderStyle::setGridAutoRows): Ditto.
(WebCore::RenderStyle::initialGridAutoColumns): Return a Vector with one auto track.
(WebCore::RenderStyle::initialGridAutoRows): Ditto.

  • rendering/style/StyleGridData.h: Store a Vector of GridTrackSize instead of just one.

LayoutTests:

  • fast/css-grid-layout/grid-auto-columns-rows-get-set-expected.txt:
  • fast/css-grid-layout/grid-auto-columns-rows-get-set.html:
  • fast/css-grid-layout/grid-shorthand-get-set-expected.txt:
  • fast/css-grid-layout/grid-shorthand-get-set.html:
  • svg/css/getComputedStyle-basic-expected.txt: CSSPrimitiveValue -> CSSValueList.
01:06 Changeset [203715] by lforschler@apple.com
  • 1 edit in trunk/Tools/ChangeLog

Test svn.webkit.org functionality after maintenance.

00:51 MathML/Early_2016_Refactoring edited by fred.wang@free.fr
(diff)
00:46 Changeset [203714] by fred.wang@free.fr
  • 6 edits
    3 adds in trunk

MathOperator: Add a mapping from combining to non-combining equivalents
https://bugs.webkit.org/show_bug.cgi?id=159513

Patch by Frederic Wang <fwang@igalia.com> on 2016-07-25
Reviewed by Darin Adler.

Source/WebCore:

Many math fonts provide stretch variants and assemblies for combining characters but not for
their non-combining equivalent. In the MathML recommendation, it is suggested to use
non-combining charaters, so we allow the operator stretching code to look for constructions
associated to these non-combining characters in order to still be able to stretch the
combining ones.

Test: mathml/presentation/bug159513.html

  • rendering/mathml/MathOperator.cpp:

(WebCore::MathOperator::getGlyph): New function extending getBaseGlyph to retrieve the glyph
data for an arbitrary character.
(WebCore::MathOperator::getMathVariantsWithFallback): This helper function calls
getMathVariants for the base glyph. If no constructions are available, it calls
getMathVariants for the glyph associated to equivalent fallback characters as listed in the
small characterFallback table.
(WebCore::MathOperator::calculateStretchyData): Call getMathVariantsWithFallback instead of
getMathVariants. Note that we do not need to do that for calculateDisplayStyleLargeOperator
as we do not use fallback for large operators.

  • rendering/mathml/MathOperator.h:

(WebCore::MathOperator::getBaseGlyph): Use getGlyph to implement this function.

LayoutTests:

  • mathml/presentation/bug159513.html: Added.
  • platform/gtk/mathml/presentation/bug159513-expected.png: Added.
  • platform/gtk/mathml/presentation/bug159513-expected.txt: Added.
  • platform/ios-simulator/TestExpectations: Skip this test on iOS.
  • platform/mac/TestExpectations: Skip this test on Mac.

07/25/16:

22:46 Changeset [203713] by cdumez@apple.com
  • 12 edits in trunk

Second parameter to Range.setStart() / setEnd() should be mandatory
https://bugs.webkit.org/show_bug.cgi?id=160184

Reviewed by Ryosuke Niwa.

LayoutTests/imported/w3c:

Rebaseline W3C test now that more checks are passing.

  • web-platform-tests/dom/interfaces-expected.txt:

Source/WebCore:

Second parameter to Range.setStart() / setEnd() should be mandatory:

Also use "unsigned long" instead of "long" type for the parameter,
as per the specification.

Firefox and Chrome agree with the specification.

No new tests, rebaselined existing test.

  • dom/Range.cpp:

(WebCore::Range::setStart):
(WebCore::Range::setEnd):
(WebCore::Range::checkNodeWOffset):

  • dom/Range.h:
  • dom/Range.idl:
  • dom/RangeBoundaryPoint.h:

(WebCore::RangeBoundaryPoint::set):

LayoutTests:

Update tests to reflect behavior change.

  • editing/deleting/delete-uneditable-style.html:
  • fast/dom/non-numeric-values-numeric-parameters-expected.txt:
  • fast/dom/script-tests/non-numeric-values-numeric-parameters.js:
  • fast/regions/simplified-layout-no-regions.html:
22:27 Changeset [203712] by cdumez@apple.com
  • 4 edits in trunk

DOMTokenList.prototype.toString should be enumerable
https://bugs.webkit.org/show_bug.cgi?id=160182

Reviewed by Ryosuke Niwa.

LayoutTests/imported/w3c:

Rebaseline W3C test now that one more check is passing.

  • web-platform-tests/dom/interfaces-expected.txt:

Source/WebCore:

DOMTokenList.prototype.toString should be enumerable:

Firefox and Chrome agree with the specification.

No new tests, rebaselined existing test.

  • html/DOMTokenList.idl:
22:27 Changeset [203711] by n_wang@apple.com
  • 7 edits in trunk

AX: Expose autoFillButtonType to accessibility
https://bugs.webkit.org/show_bug.cgi?id=160179

Reviewed by Chris Fleizach.

Source/WebCore:

Added a new attribute on Mac to expose the auto-fill button type.

Changes are covered in modified test.

  • accessibility/AccessibilityObject.cpp:

(WebCore::AccessibilityObject::isValueAutofillAvailable):
(WebCore::AccessibilityObject::valueAutofillButtonType):
(WebCore::AccessibilityObject::isValueAutofilled):

  • accessibility/AccessibilityObject.h:

(WebCore::AccessibilityObject::passwordFieldValue):

  • accessibility/mac/WebAccessibilityObjectWrapperMac.mm:

(-[WebAccessibilityObjectWrapper accessibilityAttributeValue:]):

LayoutTests:

  • accessibility/auto-fill-types-expected.txt:
  • accessibility/auto-fill-types.html:
20:04 Changeset [203710] by rniwa@webkit.org
  • 2 edits in trunk/Websites/perf.webkit.org

Build fix for Safari 9.

  • public/v3/models/build-request.js:

(BuildRequest.prototype.waitingTime): Don't use "const" in strict mode.

19:45 Changeset [203709] by rniwa@webkit.org
  • 15 edits
    2 adds in trunk/Websites/perf.webkit.org

Perf dashboard should show the list of a pending A/B testing jobs
https://bugs.webkit.org/show_bug.cgi?id=160138

Rubber-stamped by Chris Dumez.

Add a page to show the list of A/B testing build requests per triggerable. Ideally, we would like to
see a queue per builder but that would require changes to database tables and syncing scripts.

Because this page is most useful when the analysis task with which each build request is associated,
JSON API at /api/build-requests/ has been modified to return the analysis task ID for each request.

Also streamlined the page that shows the list of analysis tasks per Chris' feedback by consolidating
"Bisecting" and "Identified" into "Investigated" and moving the toolbar from the upper left corner
inside the heading to right beneath the heading above the table. Also made the category page serialize
the filter an user had typed in so that reloading the page doesn't clear it.

  • public/api/analysis-tasks.php:

(fetch_associated_data_for_tasks): Removed 'category' from the list of columns returned as the notion
of 'category' is only relevant in UI, and it's better computed in the front-end.
(format_task): Ditto.
(determine_category): Deleted.

  • public/api/test-groups.php:

(main):

  • public/include/build-requests-fetcher.php:

(BuildRequestsFetcher::fetch_for_task): Include the analysis task ID in the rows.
(BuildRequestsFetcher::fetch_for_group): Ditto. Ditto.
(BuildRequestsFetcher::fetch_incomplete_requests_for_triggerable): Ditto.
(BuildRequestsFetcher::results_internal): Ditto.

  • public/v3/index.html:
  • public/v3/main.js:

(main): Create a newly introduced BuildRequestQueuePage as a subpage of AnalysisCategoryPage.

  • public/v3/components/ratio-bar-graph.js:

(RatioBarGraph.prototype.update): Fixed a bogus assertion here. ratio can be any number. The coercion
into [-1, 1] is done inside RatioBarGraph's render() function.

  • public/v3/models/analysis-task.js:

(AnalysisTask.prototype.category): Moved the code to compute the analysis task's category from
determine_category in analysis-tasks.php. Also merged "bisecting" and "identified" into "investigated".
(AnalysisTask.categories): Merged "bisecting" and "identified" into "investigated".

  • public/v3/models/build-request.js:

(BuildRequest): Remember the triggerable and the analysis task associated with this request as well as
the time at when this request was created.
(BuildRequest.prototype.analysisTaskId): Added.
(BuildRequest.prototype.statusLabel): Use a shorter label: "Waiting" for "pending" status.
(BuildRequest.prototype.createdAt): Added.
(BuildRequest.prototype.waitingTime): Added. Returns a human readable time duration since the creation
of this build request such as "2 hours 21 minutes" against a reference time.
(BuildRequest.fetchTriggerables): Added.
(BuildRequest.cachedRequestsForTriggerableID): Added. Used when navigating back to

  • public/v3/pages/analysis-category-page.js:

(AnalysisCategoryPage): Construct AnalysisCategoryToolbar and store it in this._categoryToolbar since it
no longer inherits from Toolbar class, which PageWithHeading recognizes and stores.
(AnalysisCategoryPage.prototype.title):
(AnalysisCategoryPage.prototype.serializeState): Added.
(AnalysisCategoryPage.prototype.stateForCategory): Added. Include the filter in the serialization.
(AnalysisCategoryPage.prototype.updateFromSerializedState): Restore the filter from the URL state.
(AnalysisCategoryPage.prototype.filterDidChange): Added. Called by AnalysisCategoryToolbar to update
the URL state in addition to calling render() as done previously via setFilterCallback.
(AnalysisCategoryPage.prototype.render): Always call _categoryToolbar.render() since the hyperlinks for
the category pages now include the filter, which can be updated in each call.
(AnalysisCategoryPage.cssTemplate):

  • public/v3/pages/analysis-category-toolbar.js:

(AnalysisCategoryToolbar): Inherits from ComponentBase instead of Toolbar since AnalysisCategoryToolbar
no longer works with Heading class unlike other subclasses of Toolbar class.
(AnalysisCategoryToolbar.prototype.setCategoryPage): Added.
(AnalysisCategoryToolbar.prototype.setFilterCallback): Deleted.
(AnalysisCategoryToolbar.prototype.setFilter): Added. Used to restore from a serialized URL state.
(AnalysisCategoryToolbar.prototype.render): Don't recreate the input element as it clears the value as
well as the selection of the element. Also use AnalysisCategoryPage's stateForCategory to serialize the
category name and the current filter for each hyperlink.
(AnalysisCategoryToolbar.prototype._filterMayHaveChanged): Now takes an boolean argument specifying
whether the URL state should be updated or not. We update the URL only when a change event is fired to
avoid constantly updating it while an user is still typing.
(AnalysisCategoryToolbar.cssTemplate): Added.
(AnalysisCategoryToolbar.htmlTemplate): Added a button to open the newly added queue page.

  • public/v3/pages/build-request-queue-page.js:

(BuildRequestQueuePage): Added.
(BuildRequestQueuePage.prototype.routeName): Added.
(BuildRequestQueuePage.prototype.pageTitle): Added.
(BuildRequestQueuePage.prototype.open): Added. Fetch open build requests for every triggerables using
the same API as the syncing scripts.
(BuildRequestQueuePage.prototype.render): Added.
(BuildRequestQueuePage.prototype._constructBuildRequestTable): Added. Construct a table for the list of
pending, scheduled or running build requests in the order syncing scripts would see. Note that the list
of build requests returned by /api/build-requests/* can contain completed, canceled, or failed requests
since the JSON returns all build requests associated with each test group if one of the requests of the
group have not finished. This helps syncing scripts picking the right builder for A/B testing when it
had previously been unloaded or crashed in the middle of processing a test group. This characteristics
of the API actually helps us here because we can reliably compute the total number of build requests in
the group. The first half of this function does this counting as well as collapses all but the first
unfinished build requests into a "contraction" row, which just shows the number of build requests that
are remaining in the group.
(BuildRequestQueuePage.cssTemplate): Added.
(BuildRequestQueuePage.htmlTemplate): Added.

  • public/v3/pages/summary-page.js:

(SummaryPage.prototype.open): Use one-day median instead of seven-day median to compute the status.
(SummaryPageConfigurationGroup): Initialize _ratio to NaN. This was causing assertion failures in
RatioBarGraph's update() while measurement sets are being fetched.

  • server-tests/api-build-requests-tests.js: Updated the tests per change in BuildRequest's statusLabel.
  • unit-tests/analysis-task-tests.js: Ditto.
  • unit-tests/test-groups-tests.js: Ditto.
  • unit-tests/build-request-tests.js: Added tests for BuildRequest's waitingTime.
19:40 Changeset [203708] by zalan@apple.com
  • 15 edits in trunk/Source/WebCore

Cleanup RenderTable*::createAnonymous*
https://bugs.webkit.org/show_bug.cgi?id=160175

Reviewed by Simon Fraser.

This patch

  1. tightens the type on createAnonymousBoxWithSameTypeAs, createAnonymousWithParentRendererAndDisplay and

createAnonymousWithParentRenderer from RenderObject to the appropriate type.

  1. changes the return type of create* function from raw pointer to std::unique_ptr<>
  2. decouples createAnonymousBoxWithSameTypeAs and createAnonymousWithParentRenderer.

createAnonymousBoxWithSameTypeAs misleadingly calls createAnonymousWithParentRenderer
while it is never the parent in case of table items.
(std::unique_ptr::release() vs. WTFMove() will be addressed in a separate patch)

No change in functionality.

  • rendering/RenderBlock.cpp:

(WebCore::RenderBlock::createAnonymousBoxWithSameTypeAs):
(WebCore::RenderBlock::createAnonymousWithParentRendererAndDisplay):

  • rendering/RenderBlock.h:

(WebCore::RenderBlock::createAnonymousBlock):

  • rendering/RenderBox.cpp:

(WebCore::RenderBox::layoutOverflowRectForPropagation):

  • rendering/RenderBox.h:

(WebCore::RenderBox::createAnonymousBoxWithSameTypeAs):

  • rendering/RenderElement.cpp:

(WebCore::RenderElement::addChild):

  • rendering/RenderInline.cpp:

(WebCore::RenderInline::splitFlow):

  • rendering/RenderTable.cpp:

(WebCore::RenderTable::addChild):
(WebCore::RenderTable::createTableWithStyle):
(WebCore::RenderTable::createAnonymousWithParentRenderer):

  • rendering/RenderTable.h:

(WebCore::RenderTable::createAnonymousBoxWithSameTypeAs):

  • rendering/RenderTableCell.cpp:

(WebCore::RenderTableCell::createTableCellWithStyle):
(WebCore::RenderTableCell::createAnonymousWithParentRenderer):

  • rendering/RenderTableCell.h:

(WebCore::RenderTableCell::createAnonymousBoxWithSameTypeAs):

  • rendering/RenderTableRow.cpp:

(WebCore::RenderTableRow::addChild):
(WebCore::RenderTableRow::createTableRowWithStyle):
(WebCore::RenderTableRow::createAnonymousWithParentRenderer):

  • rendering/RenderTableRow.h:

(WebCore::RenderTableRow::createAnonymousBoxWithSameTypeAs):

  • rendering/RenderTableSection.cpp:

(WebCore::RenderTableSection::addChild):
(WebCore::RenderTableSection::createTableSectionWithStyle):
(WebCore::RenderTableSection::createAnonymousWithParentRenderer):

  • rendering/RenderTableSection.h:

(WebCore::RenderTableSection::createAnonymousBoxWithSameTypeAs):

19:24 Changeset [203707] by cdumez@apple.com
  • 3 edits
    2 adds in trunk

Touch properties should be on the prototype
https://bugs.webkit.org/show_bug.cgi?id=160174

Reviewed by Ryosuke Niwa.

Source/WebCore:

Touch properties should be on the prototype:

Chrome agrees with the specification.

Test: platform/ios-simulator/ios/touch/Touch-attributes-prototype.html

  • bindings/scripts/CodeGeneratorJS.pm:

(InterfaceRequiresAttributesOnInstanceForCompatibility): Deleted.

LayoutTests:

Add layout test coverage.

  • platform/ios-simulator/ios/touch/Touch-attributes-prototype-expected.txt: Added.
  • platform/ios-simulator/ios/touch/Touch-attributes-prototype.html: Added.
19:18 Changeset [203706] by commit-queue@webkit.org
  • 2 edits in trunk/Source/WebCore

Set MediaRemote playback state based on MediaSession playback state.
https://bugs.webkit.org/show_bug.cgi?id=160177

Patch by Jeremy Jones <jeremyj@apple.com> on 2016-07-25
Reviewed by Eric Carlson.

Use playback session state to update media remote playback state instead of
unconditionally setting it to playing.

  • platform/audio/mac/MediaSessionManagerMac.mm:

(WebCore::MediaSessionManagerMac::updateNowPlayingInfo):

19:15 Changeset [203705] by zalan@apple.com
  • 9 edits in trunk/Source/WebCore

RenderBox::haveSameDirection is used only by table items.
https://bugs.webkit.org/show_bug.cgi?id=160141

Reviewed by Simon Fraser.

Remove RenderBox::haveSameDirection() since it's used only by RenderTable*
classes. The new stand alone function (with 2 arguments) now checks if both of
the objects are valid.

No change in functionality.

  • rendering/RenderBox.h:

(WebCore::RenderBox::hasSameDirectionAs): Deleted.

  • rendering/RenderTable.cpp:

(WebCore::RenderTable::tableStartBorderAdjoiningCell):
(WebCore::RenderTable::tableEndBorderAdjoiningCell):

  • rendering/RenderTable.h:

(WebCore::haveSameDirection):

  • rendering/RenderTableCell.cpp:

(WebCore::RenderTableCell::hasStartBorderAdjoiningTable):
(WebCore::RenderTableCell::hasEndBorderAdjoiningTable):

  • rendering/RenderTableCell.h:

(WebCore::RenderTableCell::borderAdjoiningTableStart):
(WebCore::RenderTableCell::borderAdjoiningTableEnd):

  • rendering/RenderTableRow.h:

(WebCore::RenderTableRow::borderAdjoiningTableStart):
(WebCore::RenderTableRow::borderAdjoiningTableEnd):

  • rendering/RenderTableSection.cpp:

(WebCore::RenderTableSection::borderAdjoiningStartCell):
(WebCore::RenderTableSection::borderAdjoiningEndCell):
(WebCore::RenderTableSection::firstRowCellAdjoiningTableStart):
(WebCore::RenderTableSection::firstRowCellAdjoiningTableEnd):

  • rendering/RenderTableSection.h:

(WebCore::RenderTableSection::borderAdjoiningTableStart):
(WebCore::RenderTableSection::borderAdjoiningTableEnd):

18:05 Changeset [203704] by benjamin@webkit.org
  • 23 edits
    4 copies in trunk/Source/JavaScriptCore

Unreviewed, rolling out r203703.

It breaks some internal tests

Reverted changeset:

"[JSC] DFG::Node should not have its own allocator"
https://bugs.webkit.org/show_bug.cgi?id=160098
http://trac.webkit.org/changeset/203703

17:27 Changeset [203703] by commit-queue@webkit.org
  • 23 edits
    4 deletes in trunk/Source/JavaScriptCore

[JSC] DFG::Node should not have its own allocator
https://bugs.webkit.org/show_bug.cgi?id=160098

Patch by Benjamin Poulain <bpoulain@apple.com> on 2016-07-25
Reviewed by Geoffrey Garen.

We need some design changes for DFG::Node:
-Accessing the index must be fast. B3 uses indices for sets

and maps, it is a lot faster than hashing pointers.

-We should be able to subclass DFG::Node to specialize it.

  • CMakeLists.txt:
  • JavaScriptCore.xcodeproj/project.pbxproj:
  • dfg/DFGAllocator.h: Removed.

(JSC::DFG::Allocator::Region::size): Deleted.
(JSC::DFG::Allocator::Region::headerSize): Deleted.
(JSC::DFG::Allocator::Region::numberOfThingsPerRegion): Deleted.
(JSC::DFG::Allocator::Region::data): Deleted.
(JSC::DFG::Allocator::Region::isInThisRegion): Deleted.
(JSC::DFG::Allocator::Region::regionFor): Deleted.
(JSC::DFG::Allocator<T>::Allocator): Deleted.
(JSC::DFG::Allocator<T>::~Allocator): Deleted.
(JSC::DFG::Allocator<T>::allocate): Deleted.
(JSC::DFG::Allocator<T>::free): Deleted.
(JSC::DFG::Allocator<T>::freeAll): Deleted.
(JSC::DFG::Allocator<T>::reset): Deleted.
(JSC::DFG::Allocator<T>::indexOf): Deleted.
(JSC::DFG::Allocator<T>::allocatorOf): Deleted.
(JSC::DFG::Allocator<T>::bumpAllocate): Deleted.
(JSC::DFG::Allocator<T>::freeListAllocate): Deleted.
(JSC::DFG::Allocator<T>::allocateSlow): Deleted.
(JSC::DFG::Allocator<T>::freeRegionsStartingAt): Deleted.
(JSC::DFG::Allocator<T>::startBumpingIn): Deleted.

  • dfg/DFGByteCodeParser.cpp:

(JSC::DFG::ByteCodeParser::addToGraph):

  • dfg/DFGCPSRethreadingPhase.cpp:

(JSC::DFG::CPSRethreadingPhase::freeUnnecessaryNodes):
(JSC::DFG::CPSRethreadingPhase::addPhiSilently):

  • dfg/DFGCleanUpPhase.cpp:

(JSC::DFG::CleanUpPhase::run):

  • dfg/DFGConstantFoldingPhase.cpp:

(JSC::DFG::ConstantFoldingPhase::run):

  • dfg/DFGConstantHoistingPhase.cpp:
  • dfg/DFGDCEPhase.cpp:

(JSC::DFG::DCEPhase::fixupBlock):

  • dfg/DFGDriver.cpp:

(JSC::DFG::compileImpl):

  • dfg/DFGGraph.cpp:

(JSC::DFG::Graph::Graph):
(JSC::DFG::Graph::deleteNode):
(JSC::DFG::Graph::killBlockAndItsContents):
(JSC::DFG::Graph::~Graph): Deleted.

  • dfg/DFGGraph.h:

(JSC::DFG::Graph::addNode):

  • dfg/DFGLICMPhase.cpp:

(JSC::DFG::LICMPhase::attemptHoist):

  • dfg/DFGLongLivedState.cpp: Removed.

(JSC::DFG::LongLivedState::LongLivedState): Deleted.
(JSC::DFG::LongLivedState::~LongLivedState): Deleted.
(JSC::DFG::LongLivedState::shrinkToFit): Deleted.

  • dfg/DFGLongLivedState.h: Removed.
  • dfg/DFGNode.cpp:

(JSC::DFG::Node::index): Deleted.

  • dfg/DFGNode.h:

(JSC::DFG::Node::index):

  • dfg/DFGNodeAllocator.h: Removed.

(operator new ): Deleted.

  • dfg/DFGObjectAllocationSinkingPhase.cpp:
  • dfg/DFGPlan.cpp:

(JSC::DFG::Plan::compileInThread):
(JSC::DFG::Plan::compileInThreadImpl):

  • dfg/DFGPlan.h:
  • dfg/DFGSSAConversionPhase.cpp:

(JSC::DFG::SSAConversionPhase::run):

  • dfg/DFGWorklist.cpp:

(JSC::DFG::Worklist::runThread):

  • runtime/VM.cpp:

(JSC::VM::VM): Deleted.

  • runtime/VM.h:
17:21 Changeset [203702] by cdumez@apple.com
  • 8 edits
    5 adds in trunk

ClientRect properties should be on the prototype
https://bugs.webkit.org/show_bug.cgi?id=160165

Reviewed by Geoffrey Garen.

Source/WebCore:

Move ClientRect properties from the instance to the prototype. This
matches the specification, Firefox and Chrome.

Also add a serializer to ClientRect in order to match the specification:

This avoids breaking content that relies on JSON.stringify() to
serialize ClientRect objects.

Tests: fast/css/ClientRect-attributes-prototype.html

fast/css/ClientRect-serialization.html

  • CMakeLists.txt:
  • WebCore.xcodeproj/project.pbxproj:
  • bindings/js/JSBindingsAllInOne.cpp:
  • bindings/js/JSClientRectCustom.cpp: Added.

(WebCore::JSClientRect::toJSON):

  • bindings/scripts/CodeGeneratorJS.pm:
  • dom/ClientRect.idl:

LayoutTests:

  • fast/css/ClientRect-attributes-prototype-expected.txt: Added.
  • fast/css/ClientRect-attributes-prototype.html: Added.

Add layout test to check that ClientRect's properties are on the
prototype.

  • fast/css/ClientRect-serialization-expected.txt: Added.
  • fast/css/ClientRect-serialization.html: Added.

Add layout test to check that ClientRect has a serializer.

14:53 Changeset [203701] by cdumez@apple.com
  • 8 edits
    2 adds in trunk

Parameters to DOMImplementation.createDocumentType() should be mandatory and non-nullable
https://bugs.webkit.org/show_bug.cgi?id=160167

Reviewed by Ryosuke Niwa.

LayoutTests/imported/w3c:

Rebaseline a W3C test now that more checks are passing.

  • web-platform-tests/dom/interfaces-expected.txt:

Source/WebCore:

Parameters to DOMImplementation.createDocumentType() should be mandatory
and non-nullable:

Firefox and Chrome both agree with the specification. However, those
parameters were nullable and optional in WebKit.

Test: fast/dom/DOMImplementation/createDocumentType-parameters.html

  • dom/DOMImplementation.idl:

LayoutTests:

  • editing/selection/script-tests/DOMSelection-DocumentType.js:
  • fast/dom/DOMImplementation/createDocumentType-err-expected.txt:
  • fast/dom/DOMImplementation/script-tests/createDocumentType-err.js:

Update existing tests to reflect the behavior change.

  • fast/dom/DOMImplementation/createDocumentType-parameters-expected.txt: Added.
  • fast/dom/DOMImplementation/createDocumentType-parameters.html: Added.

Add layout test coverage. I have verified that this test passes on both
Firefox and Chrome.

14:38 Changeset [203700] by beidson@apple.com
  • 7 edits in trunk/Tools

Modern IDB: Make sure IndexedDB works from file:// url documents by default
https://bugs.webkit.org/show_bug.cgi?id=153783

Reviewed by Alex Christensen.

Previously, to grant IndexedDB access to file:// urls for testing purposes,
we had to call the SPI [WKWebViewConfiguration _setAllowUniversalAccessFromFileURLs:].

As of https://trac.webkit.org/changeset/203695 this is no longer required.

Change the relevant API tests to make sure this continues to be no longer required.

  • TestWebKitAPI/Tests/WebKit2Cocoa/IDBDeleteRecovery.mm:
  • TestWebKitAPI/Tests/WebKit2Cocoa/IndexedDBDatabaseProcessKill.mm:
  • TestWebKitAPI/Tests/WebKit2Cocoa/IndexedDBMultiProcess.mm:
  • TestWebKitAPI/Tests/WebKit2Cocoa/IndexedDBPersistence.mm:
  • TestWebKitAPI/Tests/WebKit2Cocoa/StoreBlobThenDelete.mm:
  • TestWebKitAPI/Tests/WebKit2Cocoa/WebProcessKillIDBCleanup.mm:
14:03 Changeset [203699] by fpizlo@apple.com
  • 7 edits in trunk/Source/JavaScriptCore

AssemblyHelpers should own all of the cell allocation methods
https://bugs.webkit.org/show_bug.cgi?id=160171

Reviewed by Saam Barati.

Prior to this change we had some code in DFGSpeculativeJIT.h and some code in JIT.h that
did cell allocation.

This change moves all of that code into AssemblyHelpers.h.

  • dfg/DFGSpeculativeJIT.h:

(JSC::DFG::SpeculativeJIT::emitAllocateJSCell):
(JSC::DFG::SpeculativeJIT::emitAllocateJSObject):
(JSC::DFG::SpeculativeJIT::emitAllocateJSObjectWithKnownSize):
(JSC::DFG::SpeculativeJIT::emitAllocateVariableSizedJSObject):
(JSC::DFG::SpeculativeJIT::emitAllocateDestructibleObject):

  • jit/AssemblyHelpers.h:

(JSC::AssemblyHelpers::emitAllocate):
(JSC::AssemblyHelpers::emitAllocateJSCell):
(JSC::AssemblyHelpers::emitAllocateJSObject):
(JSC::AssemblyHelpers::emitAllocateJSObjectWithKnownSize):
(JSC::AssemblyHelpers::emitAllocateVariableSized):
(JSC::AssemblyHelpers::emitAllocateVariableSizedJSObject):
(JSC::AssemblyHelpers::emitAllocateDestructibleObject):

  • jit/JIT.h:
  • jit/JITInlines.h:

(JSC::JIT::isOperandConstantChar):
(JSC::JIT::emitValueProfilingSite):
(JSC::JIT::emitAllocateJSObject): Deleted.

  • jit/JITOpcodes.cpp:

(JSC::JIT::emit_op_new_object):
(JSC::JIT::emit_op_create_this):

  • jit/JITOpcodes32_64.cpp:

(JSC::JIT::emit_op_new_object):
(JSC::JIT::emit_op_create_this):

13:51 Changeset [203698] by wenson_hsieh@apple.com
  • 6 edits
    2 adds in trunk

Media controls should not be displayed for a video until it starts playing
https://bugs.webkit.org/show_bug.cgi?id=160092
<rdar://problem/26986673>

Reviewed by Beth Dakin.

Source/WebCore:

For videos that have never played back yet, we should not show media controls. To ensure this
behavior, we ensure that the playback behavior restriction is set upon creating the media
element. This restriction is then removed when the media element begins to play.

Added two new WebKit API tests.

  • html/HTMLMediaElement.cpp:

(WebCore::HTMLMediaElement::HTMLMediaElement):

Tools:

Verify that multiple videos do or don't show the media controller depending on whether videos
are playing. Also tweaks an existing API test (VideoControlsManagerSingleLargeVideo) that was
passing because we were always showing media controls for large videos with audio, even if they
had not played back yet. This change ensures that large videos with audio show media controls
only after they begin to play back, and not by virtue of being large enough for main content.

  • TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj:
  • TestWebKitAPI/Tests/WebKit2Cocoa/VideoControlsManager.mm:

(TestWebKitAPI::TEST):

  • TestWebKitAPI/Tests/WebKit2Cocoa/large-video-with-audio.html:
  • TestWebKitAPI/Tests/WebKit2Cocoa/large-videos-with-audio-autoplay.html: Added.
  • TestWebKitAPI/Tests/WebKit2Cocoa/large-videos-with-audio.html: Added.
13:40 Changeset [203697] by sbarati@apple.com
  • 8 edits in trunk/Source/JavaScriptCore

MathICs should be able to take and dump stats about code size
https://bugs.webkit.org/show_bug.cgi?id=160148

Reviewed by Filip Pizlo.

This will make testing changes on MathIC going forward much easier.
We will be able to easily see if modifications to MathIC will lead
to us generating smaller code. We now only dump average size when we
regenerate any MathIC. This works out for large tests/pages, but is not
great for testing small programs. We can add more dump points later if
we find that we want to dump stats while running small small programs.

  • bytecode/CodeBlock.cpp:

(JSC::CodeBlock::jitSoon):
(JSC::CodeBlock::dumpMathICStats):

  • bytecode/CodeBlock.h:

(JSC::CodeBlock::isStrictMode):
(JSC::CodeBlock::ecmaMode):

  • dfg/DFGSpeculativeJIT.cpp:

(JSC::DFG::SpeculativeJIT::compileMathIC):

  • ftl/FTLLowerDFGToB3.cpp:

(JSC::FTL::DFG::LowerDFGToB3::compileMathIC):

  • jit/JITArithmetic.cpp:

(JSC::JIT::emitMathICFast):
(JSC::JIT::emitMathICSlow):

  • jit/JITMathIC.h:

(JSC::JITMathIC::finalizeInlineCode):
(JSC::JITMathIC::codeSize):

  • jit/JITOperations.cpp:
13:35 Changeset [203696] by jiewen_tan@apple.com
  • 16 edits
    10 moves
    1 add in trunk

Rename SubtleCrypto to WebKitSubtleCrypto
https://bugs.webkit.org/show_bug.cgi?id=160067
<rdar://problem/27483617>

Reviewed by Brent Fulgham.

Source/WebCore:

Tests: crypto/webkitSubtle/gc-2.html

crypto/webkitSubtle/gc-3.html
crypto/webkitSubtle/gc.html

Rename Class SubtleCrypto to WebKitSubtleCrypto, and Crypto.subtle to Crypto.webkitSubtle in order
to let the new implementation to reuse the name SubtleCrypto. This renaming should match what our
current JSBindings use, and therefore should not introduce any change of behavoir.

  • CMakeLists.txt:

Revise project files for for new file names.

  • DerivedSources.cpp:
  • DerivedSources.make:
  • PlatformEfl.cmake:
  • PlatformGTK.cmake:
  • PlatformMac.cmake:
  • WebCore.xcodeproj/project.pbxproj:

Revise project files for for new file names.

  • bindings/js/JSWebKitSubtleCryptoCustom.cpp: Renamed from Source/WebCore/bindings/js/JSSubtleCryptoCustom.cpp.
  • crypto/WebKitSubtleCrypto.cpp: Renamed from Source/WebCore/crypto/SubtleCrypto.cpp.
  • crypto/WebKitSubtleCrypto.h: Renamed from Source/WebCore/crypto/SubtleCrypto.h.
  • crypto/WebKitSubtleCrypto.idl: Renamed from Source/WebCore/crypto/SubtleCrypto.idl.
  • page/Crypto.cpp:

(WebCore::Crypto::webkitSubtle):
(WebCore::Crypto::subtle): Deleted.

  • page/Crypto.h:
  • page/Crypto.idl:

LayoutTests:

Move tests involving crypto.webkitSubtle from crypto/subtle to crypto/webkitSubtle.

  • crypto/webkitSubtle/gc-2-expected.txt: Renamed from LayoutTests/crypto/subtle/gc-2-expected.txt.
  • crypto/webkitSubtle/gc-2.html: Renamed from LayoutTests/crypto/subtle/gc-2.html.
  • crypto/webkitSubtle/gc-3-expected.txt: Renamed from LayoutTests/crypto/subtle/gc-3-expected.txt.
  • crypto/webkitSubtle/gc-3.html: Renamed from LayoutTests/crypto/subtle/gc-3.html.
  • crypto/webkitSubtle/gc-expected.txt: Renamed from LayoutTests/crypto/subtle/gc-expected.txt.
  • crypto/webkitSubtle/gc.html: Renamed from LayoutTests/crypto/subtle/gc.html.
  • platform/efl/TestExpectations:
  • platform/gtk/TestExpectations:
  • platform/ios-simulator-wk1/TestExpectations:
  • platform/win/TestExpectations:
12:57 Changeset [203695] by beidson@apple.com
  • 3 edits
    2 moves
    1 add
    1 delete in trunk

Allow LocalStorage by default for file URLs.
https://bugs.webkit.org/show_bug.cgi?id=160169

Reviewed by Brent Fulgham.

Source/WebCore:

Test: storage/domstorage/localstorage/file-can-access.html

  • page/SecurityOrigin.cpp:

(WebCore::SecurityOrigin::canAccessStorage): Remove the m_universalAccess check for local URLs.

LayoutTests:

  • storage/domstorage/localstorage/blocked-file-access-expected.txt: Removed.
  • storage/domstorage/localstorage/file-can-access-expected.txt: Added.
  • storage/domstorage/localstorage/file-can-access.html: Renamed from LayoutTests/storage/domstorage/localstorage/blocked-file-access.html.
  • storage/domstorage/localstorage/resources/unblocked-example.html: Renamed from LayoutTests/storage/domstorage/localstorage/resources/blocked-example.html.
12:15 Changeset [203694] by n_wang@apple.com
  • 3 edits
    2 adds in trunk

AX: AccessibilityRenderObject is adding duplicated children when CSS first-letter is being used.
https://bugs.webkit.org/show_bug.cgi?id=160155

Reviewed by Chris Fleizach.

Source/WebCore:

We were adding the same text node twice if CSS first-letter selector was being used. Added a
check for the inline continuation so that we only add it once.

Test: accessibility/mac/css-first-letter-children.html

  • accessibility/AccessibilityRenderObject.cpp:

(WebCore::firstChildConsideringContinuation):

LayoutTests:

  • accessibility/mac/css-first-letter-children-expected.txt: Added.
  • accessibility/mac/css-first-letter-children.html: Added.
12:04 Changeset [203693] by sbarati@apple.com
  • 17 edits in trunk/Source/JavaScriptCore

op_mul/ArithMul(Untyped,Untyped) should be an IC
https://bugs.webkit.org/show_bug.cgi?id=160108

Reviewed by Mark Lam.

This patch makes Mul a type based IC in much the same way that we made
Add a type-based IC. I implemented Mul in the same way. I abstracted the
implementation of the Add IC in the various JITs to allow for it to
work over arbitrary IC snippets. This will make adding Div/Sub/Pow in the
future easy. This patch also adds a new boolean argument to the various
snippet generateFastPath() methods to indicate if we should emit result profiling.
I added this because we want this profiling to be emitted for Mul in
the baseline, but not in the DFG. We used to indicate this through passing
in a nullptr for the ArithProfile, but we no longer do that in the upper
JIT tiers. So we are passing an explicit request from the JIT tier about
whether or not it's worth it for the IC to emit profiling.

We now emit much less code for Mul. Here is some data on the average
Mul snippet/IC size:

| JetStream | Unity 3D |

------| -------------

Old | ~280 bytes | ~280 bytes |

------| -------------

New | 210 bytes | 185 bytes |

------------------------------------

  • bytecode/CodeBlock.cpp:

(JSC::CodeBlock::addJITAddIC):
(JSC::CodeBlock::addJITMulIC):
(JSC::CodeBlock::findStubInfo):

  • bytecode/CodeBlock.h:

(JSC::CodeBlock::stubInfoBegin):
(JSC::CodeBlock::stubInfoEnd):

  • dfg/DFGSpeculativeJIT.cpp:

(JSC::DFG::GPRTemporary::adopt):
(JSC::DFG::FPRTemporary::FPRTemporary):
(JSC::DFG::SpeculativeJIT::compileValueAdd):
(JSC::DFG::SpeculativeJIT::compileMathIC):
(JSC::DFG::SpeculativeJIT::compileArithMul):

  • dfg/DFGSpeculativeJIT.h:

(JSC::DFG::SpeculativeJIT::callOperation):
(JSC::DFG::GPRTemporary::GPRTemporary):
(JSC::DFG::GPRTemporary::operator=):
(JSC::DFG::FPRTemporary::~FPRTemporary):
(JSC::DFG::FPRTemporary::fpr):

  • ftl/FTLLowerDFGToB3.cpp:

(JSC::FTL::DFG::LowerDFGToB3::compileToThis):
(JSC::FTL::DFG::LowerDFGToB3::compileValueAdd):
(JSC::FTL::DFG::LowerDFGToB3::compileMathIC):
(JSC::FTL::DFG::LowerDFGToB3::compileArithMul):

  • jit/JIT.h:

(JSC::JIT::getSlowCase):

  • jit/JITAddGenerator.cpp:

(JSC::JITAddGenerator::generateInline):
(JSC::JITAddGenerator::generateFastPath):

  • jit/JITAddGenerator.h:

(JSC::JITAddGenerator::JITAddGenerator):
(JSC::JITAddGenerator::isLeftOperandValidConstant):
(JSC::JITAddGenerator::isRightOperandValidConstant):

  • jit/JITArithmetic.cpp:

(JSC::JIT::emit_op_add):
(JSC::JIT::emitSlow_op_add):
(JSC::JIT::emitMathICFast):
(JSC::JIT::emitMathICSlow):
(JSC::JIT::emit_op_mul):
(JSC::JIT::emitSlow_op_mul):
(JSC::JIT::emit_op_sub):

  • jit/JITInlines.h:

(JSC::JIT::callOperation):

  • jit/JITMathIC.h:

(JSC::JITMathIC::slowPathStartLocation):
(JSC::JITMathIC::slowPathCallLocation):
(JSC::JITMathIC::isLeftOperandValidConstant):
(JSC::JITMathIC::isRightOperandValidConstant):
(JSC::JITMathIC::generateInline):
(JSC::JITMathIC::generateOutOfLine):

  • jit/JITMathICForwards.h:
  • jit/JITMulGenerator.cpp:

(JSC::JITMulGenerator::generateInline):
(JSC::JITMulGenerator::generateFastPath):

  • jit/JITMulGenerator.h:

(JSC::JITMulGenerator::JITMulGenerator):
(JSC::JITMulGenerator::isLeftOperandValidConstant):
(JSC::JITMulGenerator::isRightOperandValidConstant):
(JSC::JITMulGenerator::didEmitFastPath): Deleted.
(JSC::JITMulGenerator::endJumpList): Deleted.
(JSC::JITMulGenerator::slowPathJumpList): Deleted.

  • jit/JITOperations.cpp:
  • jit/JITOperations.h:
11:45 Changeset [203692] by weinig@apple.com
  • 2 edits in trunk/Source/WebKit2

Fix assertion.

  • NetworkProcess/cache/NetworkCacheCodersCocoa.cpp:

(WebKit::NetworkCache::encodeCertificateChain):

11:01 Changeset [203691] by commit-queue@webkit.org
  • 5 edits
    2 adds in trunk/Source/WebKit2

Split platform specific parts of NetworkCacheCoders.cpp into separate files
https://bugs.webkit.org/show_bug.cgi?id=160151

Patch by Sam Weinig <sam@webkit.org> on 2016-07-25
Reviewed by Alex Christensen.

  • NetworkProcess/cache/NetworkCacheCoders.cpp:

(WebKit::NetworkCache::Coder<WebCore::CertificateInfo>::encode): Deleted.
(WebKit::NetworkCache::Coder<WebCore::CertificateInfo>::decode): Deleted.

  • NetworkProcess/cache/NetworkCacheCodersCocoa.cpp: Copied from Source/WebKit2/NetworkProcess/cache/NetworkCacheCoders.cpp.

(WebKit::NetworkCache::encodeCFData): Moved.
(WebKit::NetworkCache::decodeCFData): Moved.
(WebKit::NetworkCache::encodeSecTrustRef): Moved.
(WebKit::NetworkCache::decodeSecTrustRef): Moved.
(WebKit::NetworkCache::encodeCertificateChain): Moved.
(WebKit::NetworkCache::decodeCertificateChain): Moved.
(WebKit::NetworkCache::Coder<WebCore::CertificateInfo>::encode): Moved.
(WebKit::NetworkCache::Coder<WebCore::CertificateInfo>::decode): Moved.

  • NetworkProcess/cache/NetworkCacheCodersSoup.cpp: Copied from Source/WebKit2/NetworkProcess/cache/NetworkCacheCoders.cpp.

(WebKit::NetworkCache::Coder<WebCore::CertificateInfo>::encode): Moved.
(WebKit::NetworkCache::Coder<WebCore::CertificateInfo>::decode): Moved.

  • PlatformGTK.cmake:
  • PlatformMac.cmake:
  • WebKit2.xcodeproj/project.pbxproj:

Add new files.

10:49 Changeset [203690] by wenson_hsieh@apple.com
  • 7 edits
    3 adds in trunk

Media controls on apple.com don't disappear when movie finishes playing
https://bugs.webkit.org/show_bug.cgi?id=160068
<rdar://problem/26668526>

Reviewed by Darin Adler.

Source/WebCore:

When a video ends, it should cause media controls to hide. While current logic
mostly accounts for this, it does not account for programmatic seeks causing
the video to lose its 'ended' status before querying for whether or not to
show media controls.

Three new API tests: large-video-seek-after-ending.html
large-video-hides-controls-after-seek-to-end.html
large-video-seek-to-beginning-and-play-after-ending.html

  • html/HTMLMediaElement.cpp:

(WebCore::HTMLMediaElement::mediaPlayerTimeChanged):
(WebCore::HTMLMediaElement::setPlaying):

  • html/MediaElementSession.cpp:

(WebCore::MediaElementSession::canControlControlsManager):

  • html/MediaElementSession.h:

Tools:

Adds new API tests. Please see WebCore ChangeLog for more details.

  • TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj:
  • TestWebKitAPI/Tests/WebKit2Cocoa/VideoControlsManager.mm:

(-[MediaPlaybackMessageHandler initWithWKWebView:finalMessageString:]):
(-[MediaPlaybackMessageHandler userContentController:didReceiveScriptMessage:]):
(TestWebKitAPI::TEST):
(-[DidPlayMessageHandler initWithWKWebView:]): Deleted.
(-[DidPlayMessageHandler userContentController:didReceiveScriptMessage:]): Deleted.

  • TestWebKitAPI/Tests/WebKit2Cocoa/large-video-hides-controls-after-seek-to-end.html: Added.
  • TestWebKitAPI/Tests/WebKit2Cocoa/large-video-seek-after-ending.html: Added.
  • TestWebKitAPI/Tests/WebKit2Cocoa/large-video-seek-to-beginning-and-play-after-ending.html: Added.
10:21 Changeset [203689] by bshafiei@apple.com
  • 5 edits in branches/safari-602-branch/Source

Versioning.

10:13 Changeset [203688] by commit-queue@webkit.org
  • 9 edits
    2 copies in trunk/Source/WebCore

Introduce a MathMLOperatorElement class
https://bugs.webkit.org/show_bug.cgi?id=160034

Patch by Frederic Wang <fwang@igalia.com> on 2016-07-25
Reviewed by Darin Adler.

No new tests, rendering is unchaned.

  • CMakeLists.txt: Add MathMLOperatorElement to the build file.
  • WebCore.xcodeproj/project.pbxproj: Ditto.
  • mathml/MathMLAllInOne.cpp: Ditto.
  • mathml/MathMLOperatorElement.cpp: New DOM class for <mo> element.

(WebCore::MathMLOperatorElement::MathMLOperatorElement):
(WebCore::MathMLOperatorElement::create):
(WebCore::MathMLOperatorElement::parseAttribute): Handle mo attributes.
(WebCore::MathMLOperatorElement::createElementRenderer): Create RenderMathMLOperator.

  • mathml/MathMLOperatorElement.h: Declare a class deriving from MathMLTextElement.
  • mathml/MathMLTextElement.cpp: Remove all the RenderMathMLOperator parts.

(WebCore::MathMLTextElement::MathMLTextElement): Remove inline keyword so that the class can
be overriden.
(WebCore::MathMLTextElement::parseAttribute): Remove code handled in MathMLOperatorElement.
(WebCore::MathMLTextElement::createElementRenderer): Ditto.

  • mathml/MathMLTextElement.h: Make class and members overridable.
  • mathml/mathtags.in: Map mo to MathMLOperatorElement.
  • rendering/mathml/RenderMathMLOperator.cpp:

(WebCore::RenderMathMLOperator::RenderMathMLOperator): Make the constructor take a
MathMLOperatorElement.

  • rendering/mathml/RenderMathMLOperator.h: Ditto.
09:57 Changeset [203687] by cdumez@apple.com
  • 4 edits in trunk/Source/WebKit2

[iOS] Make sure we call the ProcessAssertion invalidation handler on the main thread
https://bugs.webkit.org/show_bug.cgi?id=160140
<rdar://problem/27399998>

Reviewed by Darin Adler.

Based on crash traces, it appears BKSProcessAssertion is calling our
invalidation handler on a background thread. This was not anticipated
and therefore, this would lead to thread safety issues and crashes.

We now make sure to call our invalidation handler on the main thread.
We also use a WeakPtr to ensure that the ProcessAssertion is still
alive once on the main thread and before calling the invalidation
handler.

  • UIProcess/ProcessAssertion.cpp:

(WebKit::ProcessAssertion::ProcessAssertion):

  • UIProcess/ProcessAssertion.h:

(WebKit::ProcessAssertion::ProcessAssertion):
(WebKit::ProcessAssertion::createWeakPtr):

  • UIProcess/ios/ProcessAssertionIOS.mm:

(WebKit::ProcessAssertion::ProcessAssertion):
(WebKit::ProcessAssertion::markAsInvalidated):

09:37 Changeset [203686] by darin@apple.com
  • 4 edits in trunk/Source

Speed up make process slightly by improving "list of files" idiom
https://bugs.webkit.org/show_bug.cgi?id=160164

Reviewed by Mark Lam.

  • DerivedSources.make: Change rules that build lists of files to only run when

DerivedSources.make has been modified since the last time they were run. Since the
list of files are inside this file, this is safe, and this is faster than always
comparing and regenerating the file containing the list of files each time.

09:30 Changeset [203685] by philn@webkit.org
  • 2 edits in trunk/Tools

Unreviewed, fix test-webkitpy after r203674.

  • Scripts/webkitpy/port/linux_get_crash_log_unittest.py:

(GDBCrashLogGeneratorTest.test_generate_crash_log):

09:03 Changeset [203684] by wenson_hsieh@apple.com
  • 4 edits
    2 adds in trunk

The web process hangs when computing elements-based snap points for a container with large max scroll offset
https://bugs.webkit.org/show_bug.cgi?id=152605
<rdar://problem/25353661>

Reviewed by Simon Fraser.

Source/WebCore:

Fixes a bug in the computation of axis snap points. The ScrollSnapPoints object, which tracks
snap points along a particular axis, has two flags, hasRepeat and usesElements. For elements-
based snapping, both flags would be turned on, since StyleBuilderConverter::convertScrollSnapPoints
short-circuits for elements-based snapping and does not default usesRepeat to false. To address this,
we make ScrollSnapPoints not repeat(100%) by default.

Test: css3/scroll-snap/scroll-snap-elements-container-larger-than-children.html

  • css/StyleBuilderConverter.h:

(WebCore::StyleBuilderConverter::convertScrollSnapPoints): Deleted.

  • rendering/style/StyleScrollSnapPoints.cpp:

(WebCore::ScrollSnapPoints::ScrollSnapPoints):

LayoutTests:

Adds a scroll snap offset computation test case that would have previously
caused the web process to hang before this patch.

  • css3/scroll-snap/scroll-snap-elements-container-larger-than-children-expected.txt: Added.
  • css3/scroll-snap/scroll-snap-elements-container-larger-than-children.html: Added.
08:02 Changeset [203683] by carlosgc@webkit.org
  • 2 edits in trunk/Source/WebCore

REGRESSION(r200931): Invalid cast in highestAncestorToWrapMarkup()
https://bugs.webkit.org/show_bug.cgi?id=160163

Reviewed by Michael Catanzaro.

Since r200931 the result of enclosingNodeOfType() in highestAncestorToWrapMarkup() is downcasted to Element, but
the result of enclosingNodeOfType() can be a Node that is not an Element, in this case is Text. The cast is not
needed at all since that node is passed to editingIgnoresContent() and selectionFromContentsOfNode() and both
receive a Node not an Element.

  • editing/markup.cpp:

(WebCore::highestAncestorToWrapMarkup): Remove invalid cast.

08:01 Changeset [203682] by carlosgc@webkit.org
  • 2 edits in trunk/Source/WebCore

[Coordinated Graphics] ASSERTION FAILED: m_coordinator->isFlushingLayerChanges() in fast/repaint/animation-after-layer-scroll.html
https://bugs.webkit.org/show_bug.cgi?id=160156

Reviewed by Michael Catanzaro.

So, we fixed an assertion in r203663, but now is hitting the next one. As explained in bug #160142, flush
compositing state can be triggered in tests by RenderLayerCompositor::layerTreeAsText(), without the coordinator
even noticing it, so the assert can be just removed.

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

(WebCore::CoordinatedGraphicsLayer::flushCompositingStateForThisLayerOnly): Remove incorrect assert.

07:33 Changeset [203681] by zalan@apple.com
  • 3 edits in trunk/Source/WebCore

EllipsisBox ctor's isVertical parameter should read isHorizontal.
https://bugs.webkit.org/show_bug.cgi?id=160153

Reviewed by Andreas Kling.

It indicates whether the ellipsis box is horizontal. (both the callsites
and the parent class use isHorizontal)

No change in functionality.

  • rendering/EllipsisBox.cpp:

(WebCore::EllipsisBox::EllipsisBox):

  • rendering/EllipsisBox.h:
05:25 MathML/Early_2016_Refactoring edited by fred.wang@free.fr
(diff)
01:57 Changeset [203680] by svillar@igalia.com
  • 21 edits
    1 copy
    2 moves
    5 adds in trunk

[css-grid] Implement repeat(auto-fit)
https://bugs.webkit.org/show_bug.cgi?id=159771

Reviewed by Darin Adler.

Source/WebCore:

The auto-fit keyword works exactly as the already implemented auto-fill except that all
empty tracks collapse (became 0px). Absolutely positioned items do not participate on the
layout of the grid so they are not considered (a grid with only absolutely positioned items
is considered an empty grid).

Whenever a track collapses the gutters on either side do also collapse. When a collapsed
track's gutters collapse, they coincide exactly. If one side of a collapsed track does not
have a gutter then collapsing its gutters results in no gutter on either "side" of the
collapsed track.

In practice this means that is not possible to know the gap between 2 consecutive auto
repeat tracks without examining some others whenever there are collapsed tracks.

Uncommented the auto-fit cases from Mozilla tests. They have to be adapted as the reftest
machinery requires all the content to be rendered in the original 800x600 viewport.

Tests: fast/css-grid-layout/grid-auto-fit-columns.html
fast/css-grid-layout/grid-auto-fit-rows.html
fast/css-grid-layout/mozilla/grid-repeat-auto-fill-fit-005-part-1.html
fast/css-grid-layout/mozilla/grid-repeat-auto-fill-fit-005-part-2.html

  • css/CSSComputedStyleDeclaration.cpp:

(WebCore::valueForGridTrackList): Use the newly added trackSizesForComputedStyle().

  • rendering/RenderGrid.cpp:

(WebCore::RenderGrid::computeTrackBasedLogicalHeight):
(WebCore::RenderGrid::computeTrackSizesForDirection):
(WebCore::RenderGrid::isEmptyAutoRepeatTrack):
(WebCore::RenderGrid::gridGapForDirection): Returns the gap directly from the style.
(WebCore::RenderGrid::guttersSize): Computes the gap between a startLine and an
endLine. This method may need to inspect some other surrounding tracks to compute the gap.
(WebCore::RenderGrid::computeIntrinsicLogicalWidths):
(WebCore::RenderGrid::computeIntrinsicLogicalHeight):
(WebCore::RenderGrid::computeUsedBreadthOfGridTracks):
(WebCore::RenderGrid::gridTrackSize):
(WebCore::RenderGrid::resolveContentBasedTrackSizingFunctionsForItems):
(WebCore::RenderGrid::computeAutoRepeatTracksCount):
(WebCore::RenderGrid::computeEmptyTracksForAutoRepeat): Returns a Vector with the auto
repeat tracks that are going to be collapsed because they're empty.
(WebCore::RenderGrid::placeItemsOnGrid):
(WebCore::RenderGrid::trackSizesForComputedStyle): Used by ComputedStyle logic to print the
size of tracks. Added in order to hide the actual contents of m_columnPositions and
m_rowPositions to the outter world.
(WebCore::RenderGrid::offsetAndBreadthForPositionedChild):
(WebCore::RenderGrid::gridAreaBreadthForChild):
(WebCore::RenderGrid::populateGridPositionsForDirection): Added some extra code to compute
gaps as they cannot be directly added between tracks in case of having collapsed tracks.
(WebCore::RenderGrid::columnAxisOffsetForChild):
(WebCore::RenderGrid::rowAxisOffsetForChild):
(WebCore::RenderGrid::offsetBetweenTracks): Deleted.

  • rendering/RenderGrid.h: Made some API private. Added new required methods/attributes.
  • css/CSSComputedStyleDeclaration.cpp:

(WebCore::valueForGridTrackList):

  • rendering/RenderGrid.cpp:

(WebCore::RenderGrid::computeTrackBasedLogicalHeight):
(WebCore::RenderGrid::computeTrackSizesForDirection):
(WebCore::RenderGrid::hasAutoRepeatEmptyTracks):
(WebCore::RenderGrid::isEmptyAutoRepeatTrack):
(WebCore::RenderGrid::gridGapForDirection):
(WebCore::RenderGrid::guttersSize):
(WebCore::RenderGrid::computeIntrinsicLogicalWidths):
(WebCore::RenderGrid::computeIntrinsicLogicalHeight):
(WebCore::RenderGrid::computeUsedBreadthOfGridTracks):
(WebCore::RenderGrid::gridTrackSize):
(WebCore::RenderGrid::resolveContentBasedTrackSizingFunctionsForItems):
(WebCore::RenderGrid::computeAutoRepeatTracksCount):
(WebCore::RenderGrid::computeEmptyTracksForAutoRepeat):
(WebCore::RenderGrid::placeItemsOnGrid):
(WebCore::RenderGrid::trackSizesForComputedStyle):
(WebCore::RenderGrid::offsetAndBreadthForPositionedChild):
(WebCore::RenderGrid::assumedRowsSizeForOrthogonalChild):
(WebCore::RenderGrid::gridAreaBreadthForChild):
(WebCore::RenderGrid::populateGridPositionsForDirection):
(WebCore::RenderGrid::columnAxisOffsetForChild):
(WebCore::RenderGrid::rowAxisOffsetForChild):
(WebCore::RenderGrid::offsetBetweenTracks): Deleted.

  • rendering/RenderGrid.h:

LayoutTests:

Uncommented the auto-fit cases. Split the Mozilla's 005 test in two because it was not
possible to fit all the content in a viewport without scrollbars.

  • fast/css-grid-layout/grid-auto-fit-columns-expected.txt: Added.
  • fast/css-grid-layout/grid-auto-fit-columns.html: Added.
  • fast/css-grid-layout/grid-auto-fit-rows-expected.txt: Added.
  • fast/css-grid-layout/grid-auto-fit-rows.html: Added.
  • fast/css-grid-layout/grid-element-auto-repeat-get-set-expected.txt:
  • fast/css-grid-layout/grid-element-auto-repeat-get-set.html:
  • fast/css-grid-layout/grid-only-abspos-item-computed-style-crash-expected.txt:
  • fast/css-grid-layout/grid-only-abspos-item-computed-style-crash.html:
  • fast/css-grid-layout/grid-positioned-items-padding-expected.txt:
  • fast/css-grid-layout/grid-positioned-items-padding.html:
  • fast/css-grid-layout/grid-template-columns-rows-computed-style-gaps-content-alignment-expected.txt:
  • fast/css-grid-layout/grid-template-columns-rows-computed-style-gaps-content-alignment.html:
  • fast/css-grid-layout/mozilla/grid-repeat-auto-fill-fit-001-expected.html:
  • fast/css-grid-layout/mozilla/grid-repeat-auto-fill-fit-001.html:
  • fast/css-grid-layout/mozilla/grid-repeat-auto-fill-fit-002-expected.html:
  • fast/css-grid-layout/mozilla/grid-repeat-auto-fill-fit-002.html:
  • fast/css-grid-layout/mozilla/grid-repeat-auto-fill-fit-003-expected.html:
  • fast/css-grid-layout/mozilla/grid-repeat-auto-fill-fit-003.html:
  • fast/css-grid-layout/mozilla/grid-repeat-auto-fill-fit-004-expected.html:
  • fast/css-grid-layout/mozilla/grid-repeat-auto-fill-fit-004.html:
  • fast/css-grid-layout/mozilla/grid-repeat-auto-fill-fit-005-part-1-expected.html: Renamed from LayoutTests/fast/css-grid-layout/mozilla/grid-repeat-auto-fill-fit-005-expected.html.
  • fast/css-grid-layout/mozilla/grid-repeat-auto-fill-fit-005-part-1.html: Copied from LayoutTests/fast/css-grid-layout/mozilla/grid-repeat-auto-fill-fit-005.html.
  • fast/css-grid-layout/mozilla/grid-repeat-auto-fill-fit-005-part-2-expected.html: Added.
  • fast/css-grid-layout/mozilla/grid-repeat-auto-fill-fit-005-part-2.html: Renamed from LayoutTests/fast/css-grid-layout/mozilla/grid-repeat-auto-fill-fit-005.html.
00:21 MathML/Early_2016_Refactoring edited by fred.wang@free.fr
(diff)
00:18 Changeset [203679] by fred.wang@free.fr
  • 12 edits in trunk/Source/WebCore

Move parsing of display, displaystyle and mathvariant attributes into MathML element classes
https://bugs.webkit.org/show_bug.cgi?id=159623

Patch by Frederic Wang <fwang@igalia.com> on 2016-07-24
Reviewed by Brent Fulgham.

No new tests, already covered by existing tests.

  • mathml/MathMLElement.cpp:

(WebCore::MathMLElement::parseMathVariantAttribute): Move helper function to parse the
mathvariant attribute.
(WebCore::MathMLElement::getSpecifiedDisplayStyle): Helper function to set the displaystyle
value from the attribute specified on the MathML element.
(WebCore::MathMLElement::getSpecifiedMathVariant): Helper function to set the mathvariant
value from the attribute specified on the MathML element.

  • mathml/MathMLElement.h: Move the enum for mathvariant values and declare new members.

(WebCore::MathMLElement::acceptsDisplayStyleAttribute): Indicate whether the element accepts
displaystyle attribute (false for most of them).
(WebCore::MathMLElement::acceptsMathVariantAttribute): Indicate whether the element accepts
mathvariant attribute (false for most of them).

  • mathml/MathMLInlineContainerElement.cpp:

(WebCore::MathMLInlineContainerElement::acceptsDisplayStyleAttribute): Add mstyle and mtable
to the list of elements accepting the displaystyle attribute.
(WebCore::MathMLInlineContainerElement::acceptsMathVariantAttribute): Add mstyle to the list
of elements accepting the mathvariant attribute.
(WebCore::MathMLInlineContainerElement::parseAttribute): Mark displaystyle and mathvariant
dirty if necessary. Also use the new accepts*Attribute function.

  • mathml/MathMLInlineContainerElement.h: Declare overridden accepts*Attribute members.
  • mathml/MathMLMathElement.cpp:

(WebCore::MathMLMathElement::getSpecifiedDisplayStyle): Override acceptsDisplayStyleAttribute
so that the display attribute is also used to set the default value if the displaystyle
attribute is absent.
(WebCore::MathMLMathElement::parseAttribute): Mark displaystyle and mathvariant dirty if
necessary. We directly MathMLElement::parseAttribute to avoid duplicate work.

  • mathml/MathMLMathElement.h: Add the math tag to the list of elements accepting the

displaystyle and mathvariant attributes. Declare overridden getSpecifiedDisplayStyle.

  • mathml/MathMLTextElement.cpp:

(WebCore::MathMLTextElement::parseAttribute): Mark mathvariant as dirty.

  • mathml/MathMLTextElement.h: Add token elements to the list of elements accepting the

mathvariant attribute.

  • rendering/mathml/MathMLStyle.cpp:

(WebCore::MathMLStyle::updateStyleIfNeeded): Use the new MathMLElement::MathVariant enum.
(WebCore::MathMLStyle::resolveMathMLStyle): We no longer parse the display value to
initialize the default value on the math tag, because this is handled in
getSpecifiedDisplayStyle. In general, we also just call getSpecifiedDisplayStyle and
getSpecifiedMathVariant on the MathML elements instead of parsing the displaystyle and
mathvariant attributes here.
(WebCore::MathMLStyle::parseMathVariant): Deleted. This is moved into MathMLElement.

  • rendering/mathml/MathMLStyle.h: Use the new MathMLElement::MathVariant enum.
  • rendering/mathml/RenderMathMLToken.cpp: Ditto.

(WebCore::mathVariant): Ditto.
(WebCore::RenderMathMLToken::updateMathVariantGlyph): Ditto.

00:10 Changeset [203678] by carlosgc@webkit.org
  • 3 edits in trunk/Source/WebCore

Unreviewed. Remove unneeded header includes from CoordinatedGraphicsLayer.

Not only thjey are not needed, they are a layer violation, CoordinatedGraphicsLayer shouldn't know anything
about Page, Frame and FrameView.

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

07/24/16:

23:34 Changeset [203677] by philn@webkit.org
  • 1 edit
    1 add in trunk/Tools

Unreviewed, forgot to commit this file in r203674.

  • Scripts/process-linux-coredump: Added.

(main):

23:33 Changeset [203676] by carlosgc@webkit.org
  • 3 edits in trunk/Source/WebKit2

[GTK][Threaded Compositor] ASSERTION FAILED: !!handle !!m_nativeSurfaceHandle with several layout tests
https://bugs.webkit.org/show_bug.cgi?id=160143

Reviewed by Michael Catanzaro.

We have a message to set the native surface handle and another one for destroying it, the former is a normal
message while the latter is sync. This assertion happens if the web view is realized before the web process is
launched. This is the sequence:

1.- DrawingAreaProxyImpl sends SetNativeSurfaceHandleForCompositing message to the web process, since the
process hasn't been launched yet, the message is queued.
2.- Web process is launched and queued messages are now sent to the web process.
3.- The page is closed right after the web process is launched, and DrawingAreaProxyImpl sends
DestroyNativeSurfaceHandleForCompositing to the web process.
4.- The web process processes incoming messages, and DestroyNativeSurfaceHandleForCompositing is processed before
SetNativeSurfaceHandleForCompositing because it's sync.
5.- The web process processes SetNativeSurfaceHandleForCompositing message.

This is not only producing the assertion, it's also setting a handle for a X window already destroyed in the UI
process, so this could be producing the X errors we have seen in other tests. So, we need to make sure
SetNativeSurfaceHandleForCompositing and DestroyNativeSurfaceHandleForCompositing are handled in order by the
web process. We could make SetNativeSurfaceHandleForCompositing sync as well, but sync messages are just ignored
when sent before the web process has been launched (only normal messages are queued for obvious reasons). The
other option is sending the SetNativeSurfaceHandleForCompositing message with the
IPC::DispatchMessageEvenWhenWaitingForSyncReply flag. In this case the message is queued and dispatched on
process launch, but it's dispatched before other messages also queued without that flag, like
CreateWebPage. Since there's no WebPage the web process doesn't find a valid message receiver for it and
it's discarded. We need to ensure the DrawinArea object has been created before sending the
SetNativeSurfaceHandleForCompositing with the PC::DispatchMessageEvenWhenWaitingForSyncReply flag.

  • UIProcess/DrawingAreaProxyImpl.cpp:

(WebKit::DrawingAreaProxyImpl::didUpdateBackingStoreState): If we have received the first update and there's a
SetNativeSurfaceHandleForCompositing message pending, send it.
(WebKit::DrawingAreaProxyImpl::setNativeSurfaceHandleForCompositing): Do not send the message before the first
update is received.
(WebKit::DrawingAreaProxyImpl::destroyNativeSurfaceHandleForCompositing): If there was a
SetNativeSurfaceHandleForCompositing message pending, just ignore this destroy since the web process never
received the handle.

  • UIProcess/DrawingAreaProxyImpl.h:
23:28 Changeset [203675] by commit-queue@webkit.org
  • 27 edits
    3 copies
    1 delete in trunk

[Fetch API] Request should be created with any HeadersInit data
https://bugs.webkit.org/show_bug.cgi?id=159672

Patch by Youenn Fablet <youenn@apple.com> on 2016-07-24
Reviewed by Sam Weinig.

LayoutTests/imported/w3c:

  • imported/w3c/web-platform-tests/fetch/api/credentials/authentication-basic-expected.txt: Test was previously

failing due to lack of support for HeadersInit data in Request constructor.

  • platform/mac-wk1/imported/w3c/web-platform-tests/fetch/api/credentials/authentication-basic-expected.txt: Removed.
  • web-platform-tests/fetch/api/request/request-headers-expected.txt:
  • web-platform-tests/fetch/api/request/request-headers.html: Adding new test.

Source/JavaScriptCore:

  • Scripts/builtins/builtins_generator.py:

(WK_lcfirst): Synchronized with CodeGenerator.pm version.

Source/WebCore:

Made Request use JSBuiltinConstructor.
This allows initializing newly created Request with a JS built-in function, initializeFetchRequest.
initializeFetchRequest can call @fillFetchHeaders internal built-in to handle any HeadersInit data.
Future effort should be made to migrate more initialization code in initializeFetchRequest.

Made window and worker fetch function as a JS built-in.
This becomes more handy as these new functions can construct the Request object.
They can then call a single private function that takes a Request object as input.
Updated DOMWindowFetch and WorkerGlobalScopeFetch code accordingly.

To enable this, the binding generator is updated to support runtime-enabled JS built-in functions and
private functions atttached to global objects.

Covered by existing and modified tests.
Binding generator test covered by updated binding tests.

  • CMakeLists.txt: Adding DOMWindowFetch.js, FetchRequest.js and WorkerGlobalScopeFetch.js built-in files.
  • DerivedSources.make: Ditto.
  • Modules/fetch/DOMWindowFetch.cpp: Removed overloaded fetch and updated according new signature.

(WebCore::DOMWindowFetch::fetch):

  • Modules/fetch/DOMWindowFetch.h: Ditto.
  • Modules/fetch/DOMWindowFetch.idl: Making fetch a JS built-in and adding a @fetchRequest private function.
  • Modules/fetch/DOMWindowFetch.js: Added.

(fetch):

  • Modules/fetch/FetchHeaders.h:

(WebCore::FetchHeaders::setGuard): Used by FetchRequest when initializing headers.

  • Modules/fetch/FetchRequest.cpp:

(WebCore::buildHeaders): Removed as implemented in JS.
(WebCore::FetchRequest::initializeOptions): Added to handle most of the dictionary initialization.
(WebCore::FetchRequest::initializeWith): Method called from built-in constructor function.
(WebCore::FetchRequest::setBody): Corresponding to @setBody private method.
(WebCore::buildBody): Deleted.

  • Modules/fetch/FetchRequest.h:
  • Modules/fetch/FetchRequest.idl:
  • Modules/fetch/FetchRequest.js: Added.

(initializeFetchRequest): Implements fetch Request(input, init) constructor.

  • Modules/fetch/FetchResponse.cpp:

(WebCore::FetchResponse::fetch): Removed the construction of FetchRequest in fetch method since it is done by JS built-in code.

  • Modules/fetch/FetchResponse.h:
  • Modules/fetch/WorkerGlobalScopeFetch.cpp: Removed overloaded fetch and updated according new signature.

(WebCore::WorkerGlobalScopeFetch::fetch):

  • Modules/fetch/WorkerGlobalScopeFetch.h: Ditto.
  • Modules/fetch/WorkerGlobalScopeFetch.idl: Making fetch a JS built-in and adding a @fetchRequest private function.
  • Modules/fetch/WorkerGlobalScopeFetch.js: Added.

(fetch):

  • bindings/js/WebCoreBuiltinNames.h: Adding fetchRequest, setBody and Request private identifiers.
  • bindings/scripts/CodeGenerator.pm:

(WK_lcfirst): Replacing dOM by dom.

  • bindings/scripts/CodeGeneratorJS.pm:

(GenerateImplementation): Adding support for runtime-enabled built-in methods and private methods.

  • bindings/scripts/test/JS/JSTestGlobalObject.cpp:

(WebCore::JSTestGlobalObject::finishCreation):
(WebCore::jsTestGlobalObjectInstanceFunctionTestPrivateFunction):

  • bindings/scripts/test/ObjC/DOMTestGlobalObject.mm:

(-[DOMTestGlobalObject testJSBuiltinFunction]):

  • bindings/scripts/test/TestGlobalObject.idl: Adding tests for runtime-enabled global built-in methods and private methods.
23:07 Changeset [203674] by philn@webkit.org
  • 16 edits in trunk

Improve GDB backtrace generation for GTK/EFL
https://bugs.webkit.org/show_bug.cgi?id=128928

Reviewed by Carlos Garcia Campos.

Source/WebKit2:

Move the Web, Database and Network ProcessIdentifier functions to
the cross-platform WKContext and WKPage implementations.

  • UIProcess/API/C/WKContext.cpp:

(WKContextGetNetworkProcessIdentifier):
(WKContextGetDatabaseProcessIdentifier):

  • UIProcess/API/C/WKContextPrivate.h:
  • UIProcess/API/C/WKPage.cpp:

(WKPageGetProcessIdentifier):

  • UIProcess/API/C/WKPagePrivate.h:
  • UIProcess/API/C/mac/WKContextPrivateMac.h:
  • UIProcess/API/C/mac/WKContextPrivateMac.mm:
  • UIProcess/API/C/mac/WKPagePrivateMac.h:
  • UIProcess/API/C/mac/WKPagePrivateMac.mm:

Tools:

The PID of the crashed process is now correctly supplied to the
crash log reporter. The kernel core_pattern needs to be updated
after this change to something like:

echo "|/home/phil/WebKit/Tools/Scripts/process-linux-coredump /tmp/core-pid_%p.dump" > /proc/sys/kernel/core_pattern

  • Scripts/process-linux-coredump: Added.

(main): Minimal python script reading coredump data on stdin and writing it to a file in /tmp/.

  • Scripts/webkitpy/port/efl.py:

(EflPort._get_crash_log): Supply path of the process-linux-coredump script.

  • Scripts/webkitpy/port/gtk.py:

(GtkPort._get_crash_log): Ditto.

  • Scripts/webkitpy/port/linux_get_crash_log.py:

(GDBCrashLogGenerator.init): New argument for supplying the path of a coredump processor script.
(GDBCrashLogGenerator.generate_crash_log): Update error message, the core_pattern should now be set to pipe coredumps to a script.
(GDBCrashLogGenerator): Deleted.

  • Scripts/webkitpy/port/linux_get_crash_log_unittest.py:

(GDBCrashLogGeneratorTest.test_generate_crash_log): Update test expectations.

  • WebKitTestRunner/TestController.cpp:

(WTR::TestController::networkProcessDidCrash): Supply PID of crash process.
(WTR::TestController::databaseProcessDidCrash): Ditto.
(WTR::TestController::processDidCrash): Ditto.

20:42 Changeset [203673] by mattbaker@apple.com
  • 2 edits in trunk/Source/WebInspectorUI

Web Inspector: Filtering is broken in the Overview timeline view
https://bugs.webkit.org/show_bug.cgi?id=160152
<rdar://problem/27517481>

Reviewed by Joseph Pecoraro.

  • UserInterface/Views/SourceCodeTimelineTimelineDataGridNode.js:

(WebInspector.SourceCodeTimelineTimelineDataGridNode.prototype.filterableDataForColumn):
Non-resource nodes should be filtered based on their display name.

18:14 Changeset [203672] by bshafiei@apple.com
  • 3 edits in branches/safari-602-branch/LayoutTests

Merge r203665. rdar://problem/27453479

17:48 Changeset [203671] by weinig@apple.com
  • 3 edits in trunk/Source/WebKit2

Add specialization for encoding/decoding WebCore::CertificateInfos in the Network Cache
<rdar://problem/27409315>
https://bugs.webkit.org/show_bug.cgi?id=160144

Reviewed by Chris Dumez.

  • NetworkProcess/cache/NetworkCacheCoders.cpp:

(WebKit::NetworkCache::encodeCFData):
(WebKit::NetworkCache::decodeCFData):
(WebKit::NetworkCache::encodeSecTrustRef):
(WebKit::NetworkCache::decodeSecTrustRef):
(WebKit::NetworkCache::encodeCertificateChain):
(WebKit::NetworkCache::decodeCertificateChain):
(WebKit::NetworkCache::Coder<WebCore::CertificateInfo>::encode):
(WebKit::NetworkCache::Coder<WebCore::CertificateInfo>::decode):

  • NetworkProcess/cache/NetworkCacheStorage.h:

(WebKit::NetworkCache::Storage::version):
Bump the version and lastStableVersion to account for the format change.

13:33 Changeset [203670] by fpizlo@apple.com
  • 33 edits
    2 adds in trunk

B3 should support multiple entrypoints
https://bugs.webkit.org/show_bug.cgi?id=159391

Reviewed by Saam Barati.
Source/JavaScriptCore:


This teaches B3 how to compile procedures with multiple entrypoints in the best way ever.

Multiple entrypoints are useful. We could use them to reduce the cost of compiling OSR
entrypoints. We could use them to implement better try/catch.

Multiple entrypoints are hard to support. All of the code that assumed that the root block
is the entrypoint would have to be changed. Transformations like moveConstants() would have
to do crazy things if the existence of multiple entrypoints prevented it from finding a
single common dominator.

Therefore, we want to add multiple entrypoints without actually teaching the compiler that
there is such a thing. That's sort of what this change does.

This adds a new opcode to both B3 and Air called EntrySwitch. It's a terminal that takes
one or more successors and no value children. The number of successors must match
Procedure::numEntrypoints(), which could be arbitrarily large. The semantics of EntrySwitch
are:

  • Each of the entrypoints sets a hidden Entry variable to that entrypoint's index and jumps to the procedure's root block.


  • An EntrySwitch is a switch statement over this hidden Entry variable.


The way that we actually implement this is that Air has a very late phase - after all
register and stack layout - that clones all code where the Entry variable is live; i.e all
code in the closure over predecessors of all blocks that do EntrySwitch.

Usually, you would use this by creating an EntrySwitch in the root block, but you don't
have to do that. Just remember that the code before EntrySwitch gets cloned for each
entrypoint. We allow cloning of an arbitrarily large amount of code because restricting it,
and so restricing the placement of EntrySwitches, would be unelegant. It would be hard to
preserve this invariant. For example we wouldn't be able to lower any value before an
EntrySwitch to a control flow diamond.

This patch gives us an easy-to-use way to use B3 to compile code with multiple entrypoints.
Inside the compiler, only code that runs very late in Air has to know about this feature.
We get the best of both worlds!

Also, I finally got rid of the requirement that you explicitly cast BasicBlock* to
FrequentedBlock. I can no longer remember why I thought that was a good idea. Removing it
doesn't cause any problems and it makes code easier to write.

  • CMakeLists.txt:
  • JavaScriptCore.xcodeproj/project.pbxproj:
  • b3/B3BasicBlockUtils.h:

(JSC::B3::updatePredecessorsAfter):
(JSC::B3::clearPredecessors):
(JSC::B3::recomputePredecessors):

  • b3/B3FrequencyClass.h:

(JSC::B3::maxFrequency):

  • b3/B3Generate.h:
  • b3/B3LowerToAir.cpp:

(JSC::B3::Air::LowerToAir::lower):

  • b3/B3MoveConstants.cpp:
  • b3/B3Opcode.cpp:

(WTF::printInternal):

  • b3/B3Opcode.h:
  • b3/B3Procedure.cpp:

(JSC::B3::Procedure::isFastConstant):
(JSC::B3::Procedure::entrypointLabel):
(JSC::B3::Procedure::addDataSection):

  • b3/B3Procedure.h:

(JSC::B3::Procedure::numEntrypoints):
(JSC::B3::Procedure::setNumEntrypoints):
(JSC::B3::Procedure::setLastPhaseName):

  • b3/B3Validate.cpp:
  • b3/B3Value.cpp:

(JSC::B3::Value::effects):
(JSC::B3::Value::typeFor):

  • b3/B3Value.h:
  • b3/air/AirCode.cpp:

(JSC::B3::Air::Code::cCallSpecial):
(JSC::B3::Air::Code::isEntrypoint):
(JSC::B3::Air::Code::resetReachability):
(JSC::B3::Air::Code::dump):

  • b3/air/AirCode.h:

(JSC::B3::Air::Code::setFrameSize):
(JSC::B3::Air::Code::numEntrypoints):
(JSC::B3::Air::Code::entrypoints):
(JSC::B3::Air::Code::entrypoint):
(JSC::B3::Air::Code::setEntrypoints):
(JSC::B3::Air::Code::entrypointLabel):
(JSC::B3::Air::Code::setEntrypointLabels):
(JSC::B3::Air::Code::calleeSaveRegisters):

  • b3/air/AirCustom.h:

(JSC::B3::Air::PatchCustom::isTerminal):
(JSC::B3::Air::PatchCustom::hasNonArgEffects):
(JSC::B3::Air::PatchCustom::hasNonArgNonControlEffects):
(JSC::B3::Air::PatchCustom::generate):
(JSC::B3::Air::CommonCustomBase::hasNonArgEffects):
(JSC::B3::Air::CCallCustom::forEachArg):
(JSC::B3::Air::ColdCCallCustom::forEachArg):
(JSC::B3::Air::ShuffleCustom::forEachArg):
(JSC::B3::Air::EntrySwitchCustom::forEachArg):
(JSC::B3::Air::EntrySwitchCustom::isValidFormStatic):
(JSC::B3::Air::EntrySwitchCustom::isValidForm):
(JSC::B3::Air::EntrySwitchCustom::admitsStack):
(JSC::B3::Air::EntrySwitchCustom::isTerminal):
(JSC::B3::Air::EntrySwitchCustom::hasNonArgNonControlEffects):
(JSC::B3::Air::EntrySwitchCustom::generate):

  • b3/air/AirGenerate.cpp:

(JSC::B3::Air::prepareForGeneration):
(JSC::B3::Air::generate):

  • b3/air/AirLowerEntrySwitch.cpp: Added.

(JSC::B3::Air::lowerEntrySwitch):

  • b3/air/AirLowerEntrySwitch.h: Added.
  • b3/air/AirOpcode.opcodes:
  • b3/air/AirOptimizeBlockOrder.cpp:

(JSC::B3::Air::blocksInOptimizedOrder):

  • b3/air/AirSpecial.cpp:

(JSC::B3::Air::Special::isTerminal):
(JSC::B3::Air::Special::hasNonArgEffects):
(JSC::B3::Air::Special::hasNonArgNonControlEffects):

  • b3/air/AirSpecial.h:
  • b3/air/AirValidate.cpp:
  • b3/air/opcode_generator.rb:
  • b3/testb3.cpp:

Source/WTF:

  • wtf/GraphNodeWorklist.h: Expose some handy functionality.

(WTF::GraphNodeWorklist::pop):
(WTF::GraphNodeWorklist::saw):
(WTF::GraphNodeWorklist::seen):

  • wtf/VectorTraits.h: Fix a bug! Otherwise filling a vector of byte-sized enum classes doesn't work.

Websites/webkit.org:


Update some statements about ControlValue (which doesn't exist anymore) and add a blurb
about EntrySwitch.

  • docs/b3/index.html:
  • docs/b3/intermediate-representation.html:
12:53 Changeset [203669] by n_wang@apple.com
  • 3 edits
    2 adds in trunk

AX: Video Controls: Volume cannot be adjusted using VO.
https://bugs.webkit.org/show_bug.cgi?id=160107

Reviewed by Dean Jackson.

Source/WebCore:

The volume slider in video tag had 0.01 step which caused the screen reader adjusting it slowly.
Changed the step to 0.05 and added the aria-valuetext attribute to the slider, so that the value
is spoken in percentage.

Test: accessibility/mac/video-volume-slider-accessibility.html

  • Modules/mediacontrols/mediaControlsApple.js:

(Controller.prototype.createControls):
(Controller.prototype.handleVolumeSliderInput):
(Controller.prototype.updateVolume):

LayoutTests:

  • accessibility/mac/video-volume-slider-accessibility-expected.txt: Added.
  • accessibility/mac/video-volume-slider-accessibility.html: Added.
11:59 Changeset [203668] by fpizlo@apple.com
  • 3 edits in trunk/Source/JavaScriptCore

Unreviewed, fix broken test. I don't know why I goofed this up without seeing it before landing.

  • b3/air/AirOpcode.opcodes:
  • b3/testb3.cpp:

(JSC::B3::run):

10:47 Changeset [203667] by ddkilzer@apple.com
  • 3 edits
    2 adds in trunk

REGRESSION (r203106): Crash in WebCore::MathMLElement::parseMathMLLength()
<https://webkit.org/b/160111>
<rdar://problem/27506489>

Reviewed by Chris Dumez.

Source/WebCore:

Test: mathml/mpadded-crash.html

  • mathml/MathMLElement.cpp:

(WebCore::skipLeadingAndTrailingWhitespace): Change to take
StringView parameter instead of String to avoid creating a
temporary String that's released on return.

LayoutTests:

  • mathml/mpadded-crash-expected.txt: Added.
  • mathml/mpadded-crash.html: Added.
10:36 Changeset [203666] by fpizlo@apple.com
  • 11 edits in trunk/Source/JavaScriptCore

[B3] Fusing immediates into test instructions should work again
https://bugs.webkit.org/show_bug.cgi?id=160073

Reviewed by Sam Weinig.

When we introduced BitImm, we forgot to change the Branch(BitAnd(value, constant))
fusion. This emits test instructions, so it should use BitImm for the constant. But it
was still using Imm! This meant that isValidForm() always returned false.

This fixes the code path to use BitImm, and turns off our use of BitImm64 on x86 since
it provides no benefit on x86 and has some risk (the code appears to play fast and loose
with the scratch register).

This is not an obvious progression on anything, so I added comprehensive tests to
testb3, which check that we selected the optimal instruction in a variety of situations.
We should add more tests like this!

  • b3/B3BasicBlock.h:

(JSC::B3::BasicBlock::successorBlock):

  • b3/B3LowerToAir.cpp:

(JSC::B3::Air::LowerToAir::createGenericCompare):

  • b3/B3LowerToAir.h:
  • b3/air/AirArg.cpp:

(JSC::B3::Air::Arg::isRepresentableAs):
(JSC::B3::Air::Arg::usesTmp):

  • b3/air/AirArg.h:

(JSC::B3::Air::Arg::isRepresentableAs):
(JSC::B3::Air::Arg::castToType):
(JSC::B3::Air::Arg::asNumber):

  • b3/air/AirCode.h:

(JSC::B3::Air::Code::size):
(JSC::B3::Air::Code::at):

  • b3/air/AirOpcode.opcodes:
  • b3/air/AirValidate.h:
  • b3/air/opcode_generator.rb:
  • b3/testb3.cpp:

(JSC::B3::compile):
(JSC::B3::compileAndRun):
(JSC::B3::lowerToAirForTesting):
(JSC::B3::testSomeEarlyRegister):
(JSC::B3::testBranchBitAndImmFusion):
(JSC::B3::zero):
(JSC::B3::run):

10:25 Changeset [203665] by mmaxfield@apple.com
  • 3 edits in trunk/LayoutTests

Test gardening after r203626.
https://bugs.webkit.org/show_bug.cgi?id=160102
<rdar://problem/27453479>

Unreviewed.

  • platform/ios-simulator/editing/deleting/delete-emoji-expected.txt:
  • platform/mac-yosemite/editing/deleting/delete-emoji-expected.txt:
08:53 Changeset [203664] by utatane.tea@gmail.com
  • 3 edits in trunk/Source/JavaScriptCore

Unreviewed, update the exponentiation expression error message
https://bugs.webkit.org/show_bug.cgi?id=159969

Follow up patch for r203499.

  • parser/Parser.cpp:

(JSC::Parser<LexerType>::parseBinaryExpression):

  • tests/stress/pow-expects-update-expression-on-lhs.js:

(throw.new.Error):

06:08 Changeset [203663] by commit-queue@webkit.org
  • 3 edits in trunk/Source/WebCore

[Coordinated Graphics] ASSERTION FAILED: !m_flushingLayers in fast/repaint/animation-after-layer-scroll.html
https://bugs.webkit.org/show_bug.cgi?id=160142

Patch by Carlos Garcia Campos <cgarcia@igalia.com> on 2016-07-24
Reviewed by Michael Catanzaro.

This only happens in layout tests, because it happens when RenderLayerCompositor::layerTreeAsText() is
called. The thing is that CoordinatedGraphicsLayer::flushCompositingState() calls notifyFlushRequired() that
checks if the coordinator is flusing layers and if not it calls RenderLayerCompositor::notifyFlushRequired() and
returns early. This normally works because the coodinator is the one starting the layer flush, so that when
RenderLayerCompositor::flushPendingLayerChanges() is called the coordinator is always flusing layers. But
RenderLayerCompositor::layerTreeAsText() calls RenderLayerCompositor::flushPendingLayerChanges() directly, so at
that moment the coordinator is not flusing layers, what causes that
CoordinatedGraphicsLayer::flushCompositingState() ends up calling RenderLayerCompositor::notifyFlushRequired()
that schedules a new flush while flusing layers causing the
assertion. CoordinatedGraphicsLayer::flushCompositingState() is always called from
CompositingCoordinator::flushPendingLayerChanges() or RenderLayerCompositor::flushPendingLayerChanges() so we
never need to call RenderLayerCompositor::notifyFlushRequired() from there.

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

(WebCore::CoordinatedGraphicsLayer::notifyFlushRequired): This is void now since the return value is not checked anywhere.
(WebCore::CoordinatedGraphicsLayer::flushCompositingState): Remove the call to notifyFlushRequired().

  • platform/graphics/texmap/coordinated/CoordinatedGraphicsLayer.h:
06:04 Changeset [203662] by mcatanzaro@igalia.com
  • 3 edits in trunk/LayoutTests

[GTK] Layout test security/contentSecurityPolicy/plugins-types-allows-quicktime-plugin-replacement.html timing out
https://bugs.webkit.org/show_bug.cgi?id=160128

Unreviewed, skip the tests.

  • platform/efl/TestExpectations:
  • platform/gtk/TestExpectations:
04:10 Changeset [203661] by darin@apple.com
  • 3 edits
    2 deletes in trunk/Source

Adding a new WebCore JavaScript built-in source file does not trigger rebuild of WebCoreJSBuiltins*
https://bugs.webkit.org/show_bug.cgi?id=160115

Reviewed by Youenn Fablet.

Source/JavaScriptCore:

  • make-generated-sources.sh: Removed. Was unused.

Source/WebCore:

  • DerivedSources.make: Added a missing dependency so the rule that builds WebCore_BUILTINS_WRAPPERS

kicks in when the list of WebCore_BUILTINS_SOURCES is modified. Also added another missing dependency
so that changes to the JavaScript built-ins Python scripts will also trigger WebCore_BUILTINS_WRAPPERS.

  • make-generated-sources.sh: Removed. Was unused.
Note: See TracTimeline for information about the timeline view.