⚠ Archived content — this site is no longer maintained.   Current WebKit documentation is at docs.webkit.org.

Timeline



Sep 6, 2013:

11:45 PM Changeset in webkit [155248] by commit-queue@webkit.org
  • 6 edits in trunk/Source/WebCore

Get MEDIA_STREAM compiling for other ports (EFL and GTK)
​https://bugs.webkit.org/show_bug.cgi?id=120811

Patch by Thiago de Barros Lacerda <​thiago.lacerda@openbossa.org> on 2013-09-06
Reviewed by Eric Carlson.

  • CMakeLists.txt:
  • GNUmakefile.list.am:
  • Modules/mediastream/LocalMediaStream.h:
  • Modules/mediastream/MediaStreamRegistry.cpp:

(WebCore::MediaStreamRegistry::lookupMediaStreamDescriptor):

  • Modules/mediastream/RTCStatsResponse.h:
11:26 PM Changeset in webkit [155247] by commit-queue@webkit.org
  • 2 edits in trunk/Source/WebInspectorUI

Web Inspector: Download Archive button incorrectly styled when disabled and tree element is not-selected
​https://bugs.webkit.org/show_bug.cgi?id=120866

Patch by Joseph Pecoraro <Joseph Pecoraro> on 2013-09-06
Reviewed by Timothy Hatcher.

  • UserInterface/TreeElementStatusButton.css:

(.item > .status > .status-button.disabled > svg *):

11:25 PM Changeset in webkit [155246] by zandobersek@gmail.com
  • 2 edits in trunk

REGRESSION(r155143): Build failures on GTK port with Clang and libstdc++ < 4.8.1
​https://bugs.webkit.org/show_bug.cgi?id=120896

Reviewed by Anders Carlsson.

The GTK port currently only permits using the libstdc++ standard library when compiling with
Clang. After r155143, build failures are occurring when using Clang and libstdc++ that predates
the 4.8.0 release due to the use of std::is_trivially_destructible that isn't available in
libstdc++ < 4.8.0.

To not add additional special casing, the GTK port should move onto requiring libstdc++ >= 4.8.1
when compiling with the Clang compiler. Version 4.8.1 was chosen since it's C++11 feature-complete.
This strict requirement is possible as compiling the GTK port with the Clang compiler is not really
widespread, so we can afford to adjust the required dependencies to match other ports' progression
instead of modifying the code.

  • Source/autotools/CheckSystemAndBasicDependencies.m4: If the detected compiler is Clang, also check

that the libstdc++ standard library is used by testing for the GLIBCXX macro that should be defined
to the value lesser than the '20130531', the date stamp used by the 4.8.1 release of libstdc++. Since
possible future releases of the 4.6 or 4.7 series of libstdc++ will also match this check due to a newer
date stamp contained in GLIBCXX, the std::is_trivially_destructible struct is also used so the
compilation will fail if the libstdc++ that's used is older than allowed (and therefor does not support
the feature). If the check fails, a fatal error is thrown, describing the requirement. Everything carries
on as normal otherwise.

11:13 PM Changeset in webkit [155245] by zandobersek@gmail.com
  • 2 edits in trunk

[GTK] Bump the required Clang version to 3.2
​https://bugs.webkit.org/show_bug.cgi?id=112537

Reviewed by Gustavo Noronha Silva.

  • Source/autotools/CheckSystemAndBasicDependencies.m4: Require the Clang 3.2 stack

when the Clang compiler is specified.

11:05 PM Changeset in webkit [155244] by Bem Jones-Bey
  • 20 edits
    2 adds in trunk

[CSS Shapes] Floats with shape-outside aren't painting in the correct order
​https://bugs.webkit.org/show_bug.cgi?id=118492

Reviewed by Darin Adler.

Source/WebCore:

When removing the old positioning behavior, the code that makes a
float with shape-outside gain a layer was left behind. This patch
removes that, so now floats with shape-outside are painted just like
floats without shape-outside.

Also remove the isFloatingWithShapeOutside helper method because its
use is now somewhat redundant, and the check it does is now only
needed in one place.

Test: fast/shapes/shape-outside-floats/shape-outside-floats-not-a-layer.html

  • rendering/RenderBox.h:

(WebCore::RenderBox::shapeOutsideInfo):

  • rendering/RenderObject.h:
  • rendering/shapes/ShapeOutsideInfo.cpp:

(WebCore::ShapeOutsideInfo::isEnabledFor):

LayoutTests:

Update tests to reflect correct painting order. Add a new test for the
correct painting behavior.

  • csswg/submitted/shapes/shape-outside/shape-outside-floats-circle-000-expected.html:
  • csswg/submitted/shapes/shape-outside/shape-outside-floats-circle-000.html:
  • csswg/submitted/shapes/shape-outside/shape-outside-floats-ellipse-000-expected.html:
  • csswg/submitted/shapes/shape-outside/shape-outside-floats-ellipse-000.html:
  • csswg/submitted/shapes/shape-outside/shape-outside-floats-inset-rectangle-001-expected.html:
  • csswg/submitted/shapes/shape-outside/shape-outside-floats-inset-rectangle-001.html:
  • csswg/submitted/shapes/shape-outside/shape-outside-floats-rounded-rectangle-001-expected.html:
  • csswg/submitted/shapes/shape-outside/shape-outside-floats-rounded-rectangle-001.html:
  • csswg/submitted/shapes/shape-outside/shape-outside-floats-rounded-rectangle-002-expected.html:
  • csswg/submitted/shapes/shape-outside/shape-outside-floats-rounded-rectangle-002.html:
  • csswg/submitted/shapes/shape-outside/shape-outside-floats-rounded-rectangle-003-expected.html:
  • csswg/submitted/shapes/shape-outside/shape-outside-floats-rounded-rectangle-003.html:
  • csswg/submitted/shapes/shape-outside/shape-outside-floats-rounded-rectangle-004-expected.html:
  • csswg/submitted/shapes/shape-outside/shape-outside-floats-rounded-rectangle-004.html:
  • csswg/submitted/shapes/shape-outside/shape-outside-floats-stacked-000.html:
  • fast/shapes/shape-outside-floats/shape-outside-floats-not-a-layer-expected.html: Added.
  • fast/shapes/shape-outside-floats/shape-outside-floats-not-a-layer.html: Added.
10:47 PM Changeset in webkit [155243] by fpizlo@apple.com
  • 7 edits in trunk/Source/JavaScriptCore

FTL should support Call/Construct in the worst way possible
​https://bugs.webkit.org/show_bug.cgi?id=120916

Reviewed by Oliver Hunt.

This adds support for Call/Construct by just calling out to C code that uses
the JSC::call/JSC::construct runtime functions for making calls. This is slow
and terrible, but it dramatically extends FTL coverage.

Supporting calls in a meaningful way meant also supporting
GlobalVarWatchpoint.

The extension of coverage helped to find a bunch of bugs:

  • ObjectOrOtherUse was claimed to be supported in the FTL but speculate() didn't support it. That means that any node with an ObjectOrOtherUse edge that got DCE'd would cause the FTL to ICE.


  • There was a bad fall-through compileCompareStrictEq() that led to ICE.


  • The OSR exit reconstruction code was assuming it could do fast checks on node->child1() before even determining the type of node; that crashes if the node is HasVarArgs. Fixed by checking HasVarArgs first.


  • The OSR exit compiler was using the wrong peekOffset for CArgumentGetter. The default is 1, which assumes that you didn't push anything onto the stack after getting called. The OSR exit thunks push FP, so the offset should be 2.


This passes stress tests and is probably huge performance regression if you
--useExperimentalFTL=true. The regression will be fixed in
​https://bugs.webkit.org/show_bug.cgi?id=113621.

  • dfg/DFGOperations.cpp:
  • dfg/DFGOperations.h:
  • ftl/FTLCapabilities.cpp:

(JSC::FTL::canCompile):

  • ftl/FTLIntrinsicRepository.h:
  • ftl/FTLLowerDFGToLLVM.cpp:

(JSC::FTL::LowerDFGToLLVM::compileNode):
(JSC::FTL::LowerDFGToLLVM::compileGlobalVarWatchpoint):
(JSC::FTL::LowerDFGToLLVM::compileCompareStrictEq):
(JSC::FTL::LowerDFGToLLVM::compileCallOrConstruct):
(JSC::FTL::LowerDFGToLLVM::speculate):
(JSC::FTL::LowerDFGToLLVM::speculateObjectOrOther):
(JSC::FTL::LowerDFGToLLVM::addExitArgumentForNode):

  • ftl/FTLOSRExitCompiler.cpp:

(JSC::FTL::compileStub):

10:36 PM Changeset in webkit [155242] by andersca@apple.com
  • 11 edits
    1 add in trunk/Source

Add a new smart pointer type for fastMalloc'ed memory
​https://bugs.webkit.org/show_bug.cgi?id=120911

Reviewed by Andreas Kling.

Source/WebCore:

Update for WTF changes.

  • platform/network/cf/FormDataStreamCFNet.cpp:

(WebCore::advanceCurrentStream):

Source/WTF:

Due to an oversight on my part, Vector::releaseBuffer() currently returns an OwnPtr
which means that its data will be freed with operator delete instead of fastFree.

Fix this by introducing a new MllocPtr smart pointer class and change Vector::releaseBuffer() to return it instead.

  • WTF.pro:
  • WTF.vcxproj/WTF.vcxproj:
  • WTF.vcxproj/WTF.vcxproj.filters:
  • WTF.xcodeproj/project.pbxproj:
  • wtf/CMakeLists.txt:
  • wtf/FastAllocBase.h:
  • wtf/MallocPtr.h: Added.
  • wtf/Vector.h:
  • wtf/text/StringBuffer.h:
  • wtf/text/StringImpl.h:
10:34 PM Changeset in webkit [155241] by commit-queue@webkit.org
  • 43 edits in trunk/Source/WebInspectorUI

Web Inspector: Lint JavaScript and CSS for minor issues
​https://bugs.webkit.org/show_bug.cgi?id=120884

Patch by Joseph Pecoraro <Joseph Pecoraro> on 2013-09-06
Reviewed by Timothy Hatcher.

JS, caught a bunch of missing semicolons but at least one possible
issue with duplicate object literal keys in CSSKeywordCompletions.

CSS, cause some duplicate properties, an invalid property value,
and some "0px" => "0" cleanup.

10:28 PM Changeset in webkit [155240] by krit@webkit.org
  • 5 edits in trunk

Interpolate between CSS filter() and cached images
​https://bugs.webkit.org/show_bug.cgi?id=120602

Reviewed by Simon Fraser.

Source/WebCore:

Interpolate from a filter() image function to a url() image reference and
the other way around, if the input image on the filter function is the same
as the referenced image on url().

  • page/animation/CSSPropertyAnimation.cpp: Replaced old

filterBlend function with blendFilter function which is more generic
and can be reused in multiple places.

(WebCore::blendFilterOperations):
(WebCore::blendFunc):
(WebCore::blendFilter):

LayoutTests:

Add aditional test for interpolation from url to a -webkit-filter function
and the other way around. Added an additional test to make sure that different
input images don't make start and end interpolate.

  • fast/filter-image/filter-image-animation.html:
10:25 PM Changeset in webkit [155239] by fpizlo@apple.com
  • 2 edits in trunk/Source/JavaScriptCore

jsc shell should destroy VM as a workaround for LLVM's exit-time destructors
​https://bugs.webkit.org/show_bug.cgi?id=120921

Reviewed by Oliver Hunt.

LLVM's exit-time destructors will fire when we exit. If there is an on-going
FTL compile at exit, which will happen if the VM that triggered the compile
isn't shut down, then we will crash.

We should get rid of LLVM's exit-time destructors. But before we do that, we
should just do a clean VM shutdown to suppress spurious crashes. This will
help in expanding LLVM coverage for now.

  • jsc.cpp:

(jscmain):

10:24 PM Changeset in webkit [155238] by rniwa@webkit.org
  • 2 edits in trunk/LayoutTests

REGRESSION(r155100): animations/cross-fade-background-image.html fails
​https://bugs.webkit.org/show_bug.cgi?id=120923

Add the failing test expectation.

  • platform/mac/TestExpectations:
10:18 PM Changeset in webkit [155237] by rniwa@webkit.org
  • 2 edits in trunk/LayoutTests

compositing/overlap-blendingchildren-opacity-huge.html and children-opacity-no-overlap.html are failing
​https://bugs.webkit.org/show_bug.cgi?id=120922

Add the failing test expectations.

  • platform/mac/TestExpectations:
9:59 PM Changeset in webkit [155236] by jinwoo7.song@samsung.com
  • 2 edits
    1 add in trunk/LayoutTests

[EFL] Unreviewed gardening.

  • platform/efl/TestExpectations: Removed passing test.
  • platform/efl/svg/text/text-overflow-ellipsis-svgfont-kerning-ligatures-expected.txt: Added.
9:40 PM Changeset in webkit [155235] by rniwa@webkit.org
  • 2 edits in trunk/Tools

Unreviewed, rolling out r155196.
​http://trac.webkit.org/changeset/155196
​https://bugs.webkit.org/show_bug.cgi?id=120919

Broke 16 tests on build.webkit.org (Requested by rniwa on
#webkit).

Patch by Commit Queue <​commit-queue@webkit.org> on 2013-09-06

  • Scripts/webkitpy/layout_tests/controllers/manager.py:

(Manager._set_up_run):

9:15 PM Changeset in webkit [155234] by rniwa@webkit.org
  • 4 edits in trunk/Source/WebCore

Elements retained by AccessibilityImageMapLink leak
​https://bugs.webkit.org/show_bug.cgi?id=120917

Reviewed by Chris Fleizach.

Merge ​https://chromium.googlesource.com/chromium/blink/+/df0bd288efe65e5e38425fbdcdfc68b5a2a79e42

Unfortunately, the test added by the Blink change doesn't reproduce a leak in WebKit.

  • accessibility/AccessibilityImageMapLink.cpp:

(WebCore::AccessibilityImageMapLink::detachFromParent):

  • accessibility/AccessibilityImageMapLink.h:
  • accessibility/AccessibilityMockObject.h:

(WebCore::AccessibilityMockObject::detachFromParent):

7:38 PM Changeset in webkit [155233] by akling@apple.com
  • 16 edits in trunk/Source/WebCore

Make Obj-C bindings generator less #include-happy.
<​https://webkit.org/b/120913>

Reviewed by Anders Carlsson.

Knock out an old FIXME in CodeGeneratorObjC.pm and avoid including unrelated
interface declarations from implementation files where they are not needed.

  • bindings/scripts/CodeGeneratorObjC.pm:

(GenerateImplementation):

Only include DOMFooInternal.h relevant to the current interface.

  • bindings/scripts/test/ObjC/*

Rebaseline Obj-C bindings tests now that they include fewer things.

7:35 PM Changeset in webkit [155232] by akling@apple.com
  • 2 edits in trunk/Source/WTF

Remove old workaround for Nokia's Harmattan GCC on Qt/ARM.
<​https://webkit.org/b/37253>

Reviewed by Anders Carlsson.

Apparently this was a workaround for GCC getting stuck in an infinite loop
when building for ARM in a Nokia SDK over 2 years ago.

I think it's safe to remove this now, but if I'm wrong, and you're reading
this ChangeLog wondering why I did this to you, feel free to put it back in.

  • wtf/PassRefPtr.h:

(WTF::refIfNotNull):
(WTF::derefIfNotNull):

7:34 PM Changeset in webkit [155231] by akling@apple.com
  • 2 edits in trunk/Source/WTF

Remove unused RefPtr(PlacementNewAdopt) constructor.
<​https://webkit.org/b/120914>

Reviewed by Anders Carlsson.

  • wtf/RefPtr.h:
7:32 PM Changeset in webkit [155230] by akling@apple.com
  • 29 edits in trunk/Source/WebCore

Rebaseline bindings tests after Swedish breakage.

7:28 PM Changeset in webkit [155229] by akling@apple.com
  • 18 edits
    1 add in trunk/Source

Don't include Document.h from JSDOMBinding.h
<​https://webkit.org/b/120909>

Reviewed by Anders Carlsson.

Move DOMConstructorWithDocument out of JSDOMBinding.h to its own header.
This was the only part of JSDOMBinding.h that needed Document's definition.

This reduces the include dependency graph of many bindings files.

  • WebCore.xcodeproj/project.pbxproj:
  • bindings/js/DOMConstructorWithDocument.h:
  • bindings/js/JSDOMBinding.cpp:
  • bindings/js/JSDOMBinding.h:
  • bindings/scripts/CodeGeneratorJS.pm:
  • bindings/scripts/CodeGeneratorGObject.pm:
7:08 PM Changeset in webkit [155228] by rniwa@webkit.org
  • 21 edits in trunk/Source

Range::ownerDocument should return Document&
​https://bugs.webkit.org/show_bug.cgi?id=120908

Reviewed by Andreas Kling.

Source/WebCore:

Return Document& in Range::ownerDocument(). Also make setDocument take Document& so that this invariant is self-evident.

  • bindings/objc/DOM.mm:

(-[DOMRange boundingBox]):
(-[DOMRange renderedImageForcingBlackText:]):
(-[DOMRange textRects]):

  • dom/Range.cpp:

(WebCore::Range::setDocument):
(WebCore::Range::setStart):
(WebCore::Range::setEnd):
(WebCore::Range::selectNode):
(WebCore::Range::selectNodeContents):

  • dom/Range.h:

(WebCore::Range::ownerDocument):

  • editing/AlternativeTextController.cpp:

(WebCore::AlternativeTextController::handleAlternativeTextUIResult):

  • editing/Editor.cpp:

(WebCore::Editor::avoidIntersectionWithDeleteButtonController):
(WebCore::isFrameInRange):
(WebCore::Editor::countMatchesForText):

  • editing/EditorCommand.cpp:

(WebCore::unionDOMRanges):

  • editing/TextCheckingHelper.cpp:

(WebCore::TextCheckingHelper::unifiedTextCheckerEnabled):

  • editing/TextIterator.cpp:

(WebCore::plainText):
(WebCore::findPlainText):

  • editing/htmlediting.cpp:

(WebCore::extendRangeToWrappingNodes):

  • editing/markup.cpp:

(WebCore::createMarkup):
(WebCore::createFragmentFromText):

  • page/DragController.cpp:

(WebCore::documentFragmentFromDragData):
(WebCore::DragController::concludeEditDrag):

  • page/Page.cpp:

(WebCore::Page::rangeOfString):

  • rendering/RenderObject.cpp:

(WebCore::RenderObject::absoluteBoundingBoxRectForRange):

  • testing/Internals.cpp:

(WebCore::Internals::addTextMatchMarker):

Source/WebKit/blackberry:

  • WebKitSupport/InPageSearchManager.cpp:

(BlackBerry::WebKit::InPageSearchManager::findNextString):
(BlackBerry::WebKit::InPageSearchManager::findAndMarkText):
(BlackBerry::WebKit::InPageSearchManager::clearTextMatches):
(BlackBerry::WebKit::InPageSearchManager::setActiveMatchAndMarker):
(BlackBerry::WebKit::InPageSearchManager::frameUnloaded):
(BlackBerry::WebKit::InPageSearchManager::scopeStringMatches):

Source/WebKit/mac:

  • WebView/WebPDFView.mm:

(isFrameInRange):

Source/WebKit2:

  • WebProcess/InjectedBundle/API/mac/WKDOMRange.mm:

(-[WKDOMRange textRects]):

4:50 PM Changeset in webkit [155227] by Lucas Forschler
  • 1 copy in tags/Safari-534.59.10

New Tag.

4:46 PM Changeset in webkit [155226] by Brent Fulgham
  • 5 edits
    2 adds in trunk/Source/WebCore

[Windows] StructuredExceptionHandlerSuppressor Causes Bad Interactions with Support Libraries
​https://bugs.webkit.org/show_bug.cgi?id=120901

Reviewed by Anders Carlsson.

Change from old "crash on any exception" implementation to a new version that leaves normal
exception handling infrastructure in place for use by support libraries. We check exceptions
and decide whether to abort or not based on the failure type.

  • WebCore.vcxproj/WebCore.vcxproj: Add new implementation files.
  • WebCore.vcxproj/WebCore.vcxproj.filters: Ditto.
  • platform/graphics/ca/win/LayerChangesFlusher.cpp:

(WebCore::LayerChangesFlusher::hookCallback): Allocate handler struct on stack.

  • platform/win/StructuredExceptionHandlerSuppressor.cpp: Added.

(exceptionShouldTerminateProgram): New.
(exceptionHandler): New.
(WebCore::StructuredExceptionHandlerSuppressor::StructuredExceptionHandlerSuppressor): Moved
from header, and updated to add our new exception handler.
(WebCore::StructuredExceptionHandlerSuppressor::~StructuredExceptionHandlerSuppressor): Moved
from header.

  • platform/win/StructuredExceptionHandlerSuppressor.h: Move implementation of constructor and

destructor from header file.

  • platform/win/makesafeseh.asm: Added.
4:38 PM Changeset in webkit [155225] by Lucas Forschler
  • 5 edits in branches/safari-534.59-branch/Source

Versioning.

4:35 PM Changeset in webkit [155224] by vivek.vg@samsung.com
  • 5 edits in trunk/Source/WebCore

Refactor XMLDocumentParser to defer creation of XMLErrors until error occurs
​https://bugs.webkit.org/show_bug.cgi?id=120850

Reviewed by Darin Adler.

XMLErrors is created as soon as XMLDocumentParser is constructed. It may not be required
for all the valid XML documents. Deferring its creation until an error for the invalid
XML document.

No new tests as code refactoring.

  • xml/parser/XMLDocumentParser.cpp:

(WebCore::XMLDocumentParser::handleError):
(WebCore::XMLDocumentParser::insertErrorMessageBlock):

  • xml/parser/XMLDocumentParser.h:
  • xml/parser/XMLDocumentParserLibxml2.cpp:

(WebCore::XMLDocumentParser::XMLDocumentParser):

  • xml/parser/XMLDocumentParserQt.cpp:

(WebCore::XMLDocumentParser::XMLDocumentParser):

4:30 PM Changeset in webkit [155223] by ddkilzer@apple.com
  • 2 edits in trunk/Source/WTF

BUILD FIX (r155171): INSTALL_PATH_PREFIX should no longer be used beneath BUILT_PRODUCTS_DIR in HEADER_SEARCH_PATHS for iOS

  • Configurations/Base.xcconfig:

(HEADER_SEARCH_PATHS): Change
$(BUILT_PRODUCTS_DIR)/$(INSTALL_PATH_PREFIX)/usr/local/include to
$(BUILT_PRODUCTS_DIR)/usr/local/include.

4:05 PM Changeset in webkit [155222] by fpizlo@apple.com
  • 2 edits in trunk/Source/JavaScriptCore

FTL ArithMod Int32Use doesn't check for negative zero correctly
​https://bugs.webkit.org/show_bug.cgi?id=120905

Reviewed by Mark Hahnenberg.

  • ftl/FTLLowerDFGToLLVM.cpp:

(JSC::FTL::LowerDFGToLLVM::compileArithMod):

4:04 PM Changeset in webkit [155221] by andersca@apple.com
  • 5 edits in trunk/Source

Remove fastNew/fastDelete
​https://bugs.webkit.org/show_bug.cgi?id=120904

Reviewed by Andreas Kling.

Source/WebCore:

Call operator delete directly.

  • css/CSSSelectorList.cpp:

(WebCore::CSSSelectorList::adoptSelectorVector):

Source/WTF:

Remove the various variants of fastNew and fastDelete.

  • wtf/FastAllocBase.h:
3:34 PM Changeset in webkit [155220] by fpizlo@apple.com
  • 4 edits in trunk

FTL ArithNeg Int32Use doesn't check negative zero
​https://bugs.webkit.org/show_bug.cgi?id=120900

Reviewed by Mark Hahnenberg.

Source/JavaScriptCore:

  • ftl/FTLLowerDFGToLLVM.cpp:

(JSC::FTL::LowerDFGToLLVM::compileArithNegate):

LayoutTests:

  • fast/js/regress/script-tests/negative-zero-modulo.js:
3:32 PM Changeset in webkit [155219] by andersca@apple.com
  • 14 edits in trunk/Source

Stop using fastNew/fastDelete in JavaScriptCore
​https://bugs.webkit.org/show_bug.cgi?id=120898

Reviewed by Oliver Hunt.

Source/JavaScriptCore:

Change all the hash table members in ExecState to be OwnPtrs and use
adoptPtr instead. Also, since none of the hash tables can be null, change their getters
to return references and propagate the reference types wherever we know that a HashTable can't be null.

  • interpreter/CallFrame.h:

(JSC::ExecState::arrayConstructorTable):
(JSC::ExecState::arrayPrototypeTable):
(JSC::ExecState::booleanPrototypeTable):
(JSC::ExecState::dataViewTable):
(JSC::ExecState::dateTable):
(JSC::ExecState::dateConstructorTable):
(JSC::ExecState::errorPrototypeTable):
(JSC::ExecState::globalObjectTable):
(JSC::ExecState::jsonTable):
(JSC::ExecState::numberConstructorTable):
(JSC::ExecState::numberPrototypeTable):
(JSC::ExecState::objectConstructorTable):
(JSC::ExecState::privateNamePrototypeTable):
(JSC::ExecState::regExpTable):
(JSC::ExecState::regExpConstructorTable):
(JSC::ExecState::regExpPrototypeTable):
(JSC::ExecState::stringConstructorTable):
(JSC::ExecState::promisePrototypeTable):
(JSC::ExecState::promiseConstructorTable):
(JSC::ExecState::promiseResolverPrototypeTable):

  • runtime/ClassInfo.h:

(JSC::ClassInfo::propHashTable):

  • runtime/Lookup.h:

(JSC::getStaticPropertySlot):
(JSC::getStaticFunctionSlot):
(JSC::getStaticValueSlot):
(JSC::lookupPut):

  • runtime/VM.cpp:

(JSC::VM::VM):
(JSC::VM::~VM):

  • runtime/VM.h:

Source/WebCore:

Update for changes to JavaScriptCore.

  • bindings/js/DOMObjectHashTableMap.h:

(WebCore::DOMObjectHashTableMap::get):

  • bindings/js/JSDOMBinding.cpp:

(WebCore::getHashTableForGlobalData):

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

(WebCore::JSDOMWindow::put):

  • bindings/js/JSPluginElementFunctions.h:

(WebCore::pluginElementCustomGetOwnPropertySlot):

  • bindings/js/JSStorageCustom.cpp:

(WebCore::JSStorage::deleteProperty):
(WebCore::JSStorage::putDelegate):

  • bindings/scripts/CodeGeneratorJS.pm:

(hashTableAccessor):
(prototypeHashTableAccessor):
(constructorHashTableAccessor):
(GenerateGetOwnPropertySlotBody):
(GenerateImplementation):
(GenerateConstructorHelperMethods):

3:26 PM Changeset in webkit [155218] by fpizlo@apple.com
  • 2 edits in trunk/Tools

The 'failed' file should report collectionName/testName rather than just
testName.

Rubber stamped by Mark Hahnenberg.

  • Scripts/run-jsc-stress-tests:
3:08 PM Changeset in webkit [155217] by akling@apple.com
  • 31 edits
    1 add
    1 delete in trunk/Source

Include Frame.h in fewer places.
<​https://webkit.org/b/120868>

Reviewed by Anders Carlsson.

  • Move JSMainThreadExecState::instrumentFunctionCall() to a separate header. This avoids including InspectorInstrumentation.h from JSMainThreadExecState.h which caused all JS and Objective-C bindings to depend on Frame.h
  • Out-of-line the three methods on FrameView that were operating on Frames. Most of the call sites were in FrameView.cpp so they will still get inlined.
  • Removed GenericBinding.h, nothing was using it anyway.

After this, touching Frame.h and doing a (debug) rebuild goes from 9m30 to 3m30
on my laptop.

  • GNUmakefile.list.am:
  • WebCore.exp.in:
  • WebCore.vcxproj/WebCore.vcxproj:
  • WebCore.vcxproj/WebCore.vcxproj.filters:
  • WebCore.xcodeproj/project.pbxproj:
  • bindings/generic/GenericBinding.h: Removed.
  • bindings/js/JSCallbackData.cpp:
  • bindings/js/JSEventListener.cpp:
  • bindings/js/JSMainThreadExecState.h:
  • bindings/js/JSMainThreadExecStateInstrumentation.h: Added.

(WebCore::JSMainThreadExecState::instrumentFunctionCall):

  • bindings/js/JSMutationCallback.cpp:
  • bindings/js/ScheduledAction.cpp:
  • css/DeprecatedStyleBuilder.cpp:
  • css/ElementRuleCollector.cpp:
  • css/MediaQueryEvaluator.cpp:
  • css/SelectorChecker.h:
  • dom/StyledElement.cpp:
  • html/HTMLPlugInImageElement.cpp:
  • inspector/InspectorPageAgent.h:
  • inspector/NetworkResourcesData.h:
  • inspector/PageRuntimeAgent.cpp:
  • loader/cache/CachedImage.cpp:
  • page/FrameView.cpp:

(WebCore::FrameView::renderView):
(WebCore::FrameView::mapFromLayoutToCSSUnits):
(WebCore::FrameView::mapFromCSSToLayoutUnits):

  • page/FrameView.h:
  • page/TouchAdjustment.cpp:
  • rendering/RenderBoxModelObject.cpp:
  • rendering/RenderLayer.cpp:
  • rendering/RenderLayerCompositor.h:
  • rendering/RenderText.cpp:
  • svg/graphics/SVGImage.cpp:
3:04 PM Changeset in webkit [155216] by fpizlo@apple.com
  • 2 edits in trunk/Tools

Fix run-jsc-stress-tests to run on bots with old Ruby.

Also make it create a repro file.

Rubber stamped by Mark Hahnenberg.

  • Scripts/run-jsc-stress-tests:
3:04 PM WebKitGTK/2.2.x edited by berto@igalia.com
(diff)
3:03 PM WebKitGTK/2.0.x edited by berto@igalia.com
(diff)
2:58 PM Changeset in webkit [155215] by Brent Fulgham
  • 4 edits
    1 move in trunk/Source/WebCore

[Windows] Unreviewed gardening: Spell 'Suppressor' properly

  • WebCore.vcxproj/WebCore.vcxproj: Change StructuredExceptionHandlerSupressor to StructuredExceptionHandlerSuppressor
  • WebCore.vcxproj/WebCore.vcxproj.filters: Ditto
  • platform/graphics/ca/win/LayerChangesFlusher.cpp:

(WebCore::LayerChangesFlusher::hookCallback): Ditto

  • platform/win/StructuredExceptionHandlerSuppressor.h: Copied from platform/win/StructuredExceptionHandlerSupressor.h.
  • platform/win/StructuredExceptionHandlerSupressor.h: Removed.
2:53 PM Changeset in webkit [155214] by Lucas Forschler
  • 5 edits in branches/safari-537-branch/Source

Versioning.

2:39 PM Changeset in webkit [155213] by Lucas Forschler
  • 1 copy in tags/Safari-537.70

New Tag.

2:32 PM Changeset in webkit [155212] by roger_fong@apple.com
  • 3 edits in trunk/Source/WebKit/win

[Windows] Implement text offset methods of IAccessibleText interface.
​https://bugs.webkit.org/show_bug.cgi?id=120820.
<rdar://problem/14925242>

Reviewed by Brenet Fulgham.

  • AccessibleTextImpl.cpp:

(AccessibleText::get_textBeforeOffset):
(AccessibleText::get_textAfterOffset):
(AccessibleText::get_textAtOffset):
(AccessibleText::isInRange):

  • AccessibleTextImpl.h:
1:54 PM Changeset in webkit [155211] by hyatt@apple.com
  • 73 edits
    2 adds in trunk/Source/WebCore

Add new RenderBlockFlow class.
​https://bugs.webkit.org/show_bug.cgi?id=120777

Reviewed by Beth Dakin.

This patch adds a new RenderBlockFlow class. The idea is to get all of the line/float/margin
code out of RenderBlock and into the subclass, and then we can make RenderBlock an abstract
base. This will save memory for all of the classes that derive from RenderBlock but have
their own layout systems (e.g., flexbox, grid, table), and it will let us tighten up
types to make it more clear when we're talking about block layout code vs. a more generic
block-level object that can have any kind of layout.

This patch adds the new class and it makes sure all the objects that should be block flows
inherit from the new class.

isBlockFlow() proved to be a problematic method (and was the source of regressions when I
landed this patch the first time). When the method was first implemented, it was added as
"I'm a block but not a table." and at the time table was the only subclass of block. However
when new subclasses were added, this function was never patched to account for them. This
means, for example, that flexible boxes and grids claim to be block flows in the codebase
and have for years. This is obviously wrong, since flexible boxes are not in fact block
flows, but because of this mistake, code has started relying on the incorrect behavior of
isBlockFlow() (form controls mostly).

What I've done to deal with this problem is renamed isBlockFlow() to be
isBlockFlowFlexBoxOrGrid() in order to make it obvious that this method is returning true
for pretty much all blocks other than tables. This makes a lot of call sites look pretty
stupid, since they obviously never intended to include flexible boxes or grids, but at least
it is clear what the method is doing now.

The plan is to land this patch and then start converting call sites from
isBlockFlowFlexBoxOrGrid() to isRenderBlockFlow() one by one, and that way we can figure
out which code is making bad assumptions and change those callers to be isRenderBlock()
instead. Ultimately this isBlockFlowFlexBoxOrGrid() method will go away with callers either
changing to use isRenderBlock() or isRenderBlockFlow().

  • CMakeLists.txt:
  • GNUmakefile.list.am:
  • Target.pri:
  • WebCore.vcxproj/WebCore.vcxproj:
  • WebCore.xcodeproj/project.pbxproj:
  • accessibility/AccessibilityRenderObject.cpp:

(WebCore::AccessibilityRenderObject::computeAccessibilityIsIgnored):
(WebCore::AccessibilityRenderObject::determineAccessibilityRole):

  • dom/Position.cpp:

(WebCore::Position::isCandidate):
(WebCore::Position::getInlineBoxAndOffset):
(WebCore::Position::primaryDirection):

  • dom/PositionIterator.cpp:

(WebCore::PositionIterator::isCandidate):

  • editing/ApplyStyleCommand.cpp:

(WebCore::ApplyStyleCommand::applyInlineStyleToPushDown):

  • editing/CompositeEditCommand.cpp:

(WebCore::CompositeEditCommand::addBlockPlaceholderIfNeeded):

  • editing/Editor.cpp:

(WebCore::Editor::baseWritingDirectionForSelectionStart):

  • editing/FrameSelection.cpp:

(WebCore::caretRenderer):

  • editing/TextIterator.cpp:

(WebCore::TextIterator::shouldRepresentNodeOffsetZero):

  • editing/TypingCommand.cpp:

(WebCore::TypingCommand::makeEditableRootEmpty):

  • editing/htmlediting.cpp:

(WebCore::isBlockFlowElement):

  • html/HTMLDetailsElement.cpp:

(WebCore::HTMLDetailsElement::createRenderer):

  • html/HTMLSummaryElement.cpp:

(WebCore::HTMLSummaryElement::createRenderer):

  • html/shadow/SliderThumbElement.cpp:

(WebCore::RenderSliderThumb::RenderSliderThumb):

  • html/shadow/SliderThumbElement.h:
  • rendering/LayoutState.cpp:

(WebCore::LayoutState::LayoutState):

  • rendering/RenderBlock.cpp:

(WebCore::RenderBlock::createAnonymous):
(WebCore::RenderBlock::layoutBlock):
(WebCore::RenderBlock::collapseMargins):
(WebCore::RenderBlock::selectionGaps):
(WebCore::RenderBlock::firstLineBoxBaseline):
(WebCore::RenderBlock::lastLineBoxBaseline):
(WebCore::RenderBlock::firstLineBlock):
(WebCore::findFirstLetterBlock):
(WebCore::shouldCheckLines):
(WebCore::RenderBlock::adjustForBorderFit):

  • rendering/RenderBlock.h:
  • rendering/RenderBox.cpp:

(WebCore::RenderBox::computeRectForRepaint):
(WebCore::RenderBox::positionForPoint):

  • rendering/RenderBoxModelObject.h:

(WebCore::RenderBoxModelObject::canHaveBoxInfoInRegion):

  • rendering/RenderDeprecatedFlexibleBox.cpp:

(WebCore::RenderDeprecatedFlexibleBox::applyLineClamp):
(WebCore::RenderDeprecatedFlexibleBox::clearLineClamp):

  • rendering/RenderDetailsMarker.cpp:

(WebCore::RenderDetailsMarker::RenderDetailsMarker):

  • rendering/RenderDetailsMarker.h:
  • rendering/RenderFieldset.cpp:

(WebCore::RenderFieldset::RenderFieldset):

  • rendering/RenderFieldset.h:
  • rendering/RenderFileUploadControl.cpp:

(WebCore::RenderFileUploadControl::RenderFileUploadControl):

  • rendering/RenderFileUploadControl.h:
  • rendering/RenderFlowThread.cpp:

(WebCore::RenderFlowThread::RenderFlowThread):

  • rendering/RenderFlowThread.h:
  • rendering/RenderFullScreen.cpp:
  • rendering/RenderInline.cpp:

(WebCore::RenderInline::computeRectForRepaint):

  • rendering/RenderLineBoxList.cpp:

(WebCore::RenderLineBoxList::dirtyLinesFromChangedChild):

  • rendering/RenderListItem.cpp:

(WebCore::RenderListItem::RenderListItem):
(WebCore::RenderListItem::willBeDestroyed):
(WebCore::RenderListItem::willBeRemovedFromTree):
(WebCore::RenderListItem::layout):
(WebCore::RenderListItem::addOverflowFromChildren):
(WebCore::RenderListItem::computePreferredLogicalWidths):
(WebCore::RenderListItem::paint):

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

(WebCore::RenderMeter::RenderMeter):

  • rendering/RenderMeter.h:
  • rendering/RenderMultiColumnBlock.cpp:

(WebCore::RenderMultiColumnBlock::RenderMultiColumnBlock):

  • rendering/RenderMultiColumnBlock.h:
  • rendering/RenderObject.cpp:

(WebCore::RenderObject::createObject):
(WebCore::RenderObject::computeRectForRepaint):
(WebCore::RenderObject::styleWillChange):
(WebCore::firstLineStyleForCachedUncachedType):

  • rendering/RenderObject.h:

(WebCore::RenderObject::isBR):
(WebCore::RenderObject::isRenderBlockFlow):
(WebCore::RenderObject::isBlockFlowFlexBoxOrGrid):

  • rendering/RenderObjectChildList.cpp:

(WebCore::RenderObjectChildList::insertChildNode):

  • rendering/RenderProgress.cpp:

(WebCore::RenderProgress::RenderProgress):

  • rendering/RenderProgress.h:
  • rendering/RenderRegion.cpp:

(WebCore::RenderRegion::RenderRegion):

  • rendering/RenderRegion.h:
  • rendering/RenderRuby.cpp:

(WebCore::RenderRubyAsBlock::RenderRubyAsBlock):

  • rendering/RenderRuby.h:
  • rendering/RenderRubyBase.cpp:

(WebCore::RenderRubyBase::RenderRubyBase):

  • rendering/RenderRubyBase.h:
  • rendering/RenderRubyRun.cpp:

(WebCore::RenderRubyRun::RenderRubyRun):

  • rendering/RenderRubyRun.h:
  • rendering/RenderRubyText.cpp:

(WebCore::RenderRubyText::RenderRubyText):

  • rendering/RenderRubyText.h:
  • rendering/RenderTableCaption.cpp:

(WebCore::RenderTableCaption::RenderTableCaption):

  • rendering/RenderTableCaption.h:
  • rendering/RenderTableCell.cpp:

(WebCore::RenderTableCell::RenderTableCell):

  • rendering/RenderTableCell.h:
  • rendering/RenderTextControl.cpp:

(WebCore::RenderTextControl::RenderTextControl):

  • rendering/RenderTextControl.h:
  • rendering/RenderTextControlSingleLine.h:

(WebCore::RenderTextControlInnerBlock::RenderTextControlInnerBlock):

  • rendering/RenderTextTrackCue.cpp:

(WebCore::RenderTextTrackCue::RenderTextTrackCue):

  • rendering/RenderTextTrackCue.h:
  • rendering/RenderView.cpp:

(WebCore::RenderView::RenderView):

  • rendering/RenderView.h:
  • rendering/RenderingAllInOne.cpp:
  • rendering/mathml/RenderMathMLOperator.cpp:

(WebCore::RenderMathMLOperator::createGlyph):

  • rendering/svg/RenderSVGBlock.cpp:

(WebCore::RenderSVGBlock::RenderSVGBlock):

  • rendering/svg/RenderSVGBlock.h:
1:31 PM Changeset in webkit [155210] by berto@igalia.com
  • 2 edits in trunk/Source/ThirdParty

GTest: add support for FreeBSD and Hurd
​https://bugs.webkit.org/show_bug.cgi?id=120870

Patch by Emilio Pozuelo Monfort <​pochu@debian.org>
Reviewed by Gustavo Noronha Silva.

  • gtest/include/gtest/internal/gtest-port.h:
1:20 PM Changeset in webkit [155209] by fpizlo@apple.com
  • 2 edits in trunk/Source/JavaScriptCore

Concurrent FTL causes !hasOptimizedReplacement() asserts in cti_optimize
​https://bugs.webkit.org/show_bug.cgi?id=120890

Reviewed by Mark Hahnenberg.

Don't install an FTL code block if the DFG code block has already been jettisoned.

  • dfg/DFGToFTLDeferredCompilationCallback.cpp:

(JSC::DFG::ToFTLDeferredCompilationCallback::compilationDidComplete):

1:04 PM Changeset in webkit [155208] by ap@apple.com
  • 1 edit
    3 adds in trunk/LayoutTests

Revalidation header blacklisting should be case-insensitive.
​https://bugs.webkit.org/show_bug.cgi?id=120832

Reviewed by Alexey Proskuryakov.

Adds a new test to verify that a CSS file served initially as type
'text/css' maintains that type even if a revalidation response sends
'text/plain' along with a 304 response.

  • http/tests/cache/content-type-ignored-during-revalidation-expected.txt: Added.
  • http/tests/cache/content-type-ignored-during-revalidation.html: Added.
  • http/tests/cache/resources/stylesheet304-bad-content-type.php: Added.
1:00 PM Changeset in webkit [155207] by vivek.vg@samsung.com
  • 2 edits in trunk/Source/WebCore

Calling window.close() should indicate failure with warning message
​https://bugs.webkit.org/show_bug.cgi?id=120156

Reviewed by Darin Adler.

Closing of window is allowed for the tests for the automation.

  • page/DOMWindow.cpp:

(WebCore::DOMWindow::close):

12:40 PM Changeset in webkit [155206] by fpizlo@apple.com
  • 2 edits in trunk/Tools

Actually increment $numFailures

Reviewed by Mark Hahnenberg.

  • Scripts/run-jsc-stress-tests:
12:36 PM Changeset in webkit [155205] by fpizlo@apple.com
  • 2 edits in trunk/Tools

Fix run-jsc-stress-tests to use Pathname instead of File.realpath.

Also make it uses system() instead of popen(); that means that test output
will go to the console. This reduces error detection flakiness. It's fine
because stress tests don't print() unless they really have to.

Rubber stamped by Mark Hahnenberg.

  • Scripts/run-jsc-stress-tests:
12:32 PM Changeset in webkit [155204] by Lucas Forschler
  • 2 edits in branches/safari-537-branch/Source/WebCore

Merged r155203. <rdar://problem/14929843>

12:18 PM Changeset in webkit [155203] by ap@apple.com
  • 2 edits in trunk/Source/WebCore

2013-09-06 Mike West <​mkwst@chromium.org>

Revalidation header blacklisting should be case-insensitive.
​https://bugs.webkit.org/show_bug.cgi?id=120832

Reviewed by Alexey Proskuryakov.

Headers like 'content-type' should be ignored for 304 responses,
even if they are delivered as 'Content-Type', or 'CoNtEnT-TyPe', etc.

I broke this behavior in ​http://trac.webkit.org/changeset/142068
("Entity-header extension headers honored on 304 responses"). Pages like
​https://learndev.unm.edu/ currently break on reload, as they incorrectly
send 'Content-Type: text/plain' for 304 responses for resources like
CSS and JavaScript. The browser should drop these headers, but because
we're comparing in a case-sensitive fashion, we don't.

​https://code.google.com/p/chromium/issues/detail?id=246875 documents the
Blink-side fix; this patch is a port of that patch.

Test: http/tests/cache/content-type-ignored-during-revalidation.html

  • loader/cache/CachedResource.cpp: (WebCore::shouldUpdateHeaderAfterRevalidation): Compare the provided AtomicString 'header' to the revalidation blacklists in a case-insensitive fashion.
12:16 PM WebKitGTK/2.2.x edited by berto@igalia.com
(diff)
12:15 PM WebKitGTK/2.0.x edited by berto@igalia.com
(diff)
12:14 PM Changeset in webkit [155202] by fpizlo@apple.com
  • 4 edits in trunk/LayoutTests

fast/js/dfg-* tests should wait for the concurrent JIT
​https://bugs.webkit.org/show_bug.cgi?id=120723

Rubber stamped by Oliver Hunt.

Convert more tests.

  • fast/js/dfg-check-structure-elimination-for-non-cell-expected.txt:
  • fast/js/dfg-check-two-structures-expected.txt:
  • fast/js/dfg-compare-final-object-to-final-object-or-other-expected.txt:
  • fast/js/dfg-compare-final-object-to-final-object-or-other-when-both-proven-final-object-expected.txt:
  • fast/js/dfg-compare-final-object-to-final-object-or-other-when-proven-final-object-expected.txt:
  • fast/js/dfg-constant-fold-first-local-read-after-block-merge-expected.txt:
  • fast/js/dfg-constant-fold-logical-not-branch-expected.txt:
  • fast/js/dfg-constant-fold-misprediction-expected.txt:
  • fast/js/dfg-constant-fold-uncaptured-variable-that-is-later-captured-expected.txt:
  • fast/js/dfg-convert-this-dom-window-expected.txt:
  • fast/js/dfg-convert-this-object-then-exit-on-other-expected.txt:
  • fast/js/dfg-convert-this-other-then-exit-on-object-expected.txt:
  • fast/js/dfg-convert-this-polymorphic-object-then-exit-on-other-expected.txt:
  • fast/js/dfg-convert-this-polymorphic-object-then-exit-on-string-expected.txt:
  • fast/js/dfg-create-inlined-arguments-in-closure-inline-expected.txt:
  • fast/js/dfg-cross-global-object-inline-new-array-literal-expected.txt:
  • fast/js/dfg-cross-global-object-inline-new-array-literal-with-variables-expected.txt:
  • fast/js/jsc-test-list:
  • fast/js/script-tests/dfg-check-structure-elimination-for-non-cell.js:
  • fast/js/script-tests/dfg-check-two-structures.js:
  • fast/js/script-tests/dfg-compare-final-object-to-final-object-or-other-when-both-proven-final-object.js:
  • fast/js/script-tests/dfg-compare-final-object-to-final-object-or-other-when-proven-final-object.js:
  • fast/js/script-tests/dfg-compare-final-object-to-final-object-or-other.js:
  • fast/js/script-tests/dfg-constant-fold-first-local-read-after-block-merge.js:
  • fast/js/script-tests/dfg-constant-fold-logical-not-branch.js:
  • fast/js/script-tests/dfg-constant-fold-misprediction.js:
  • fast/js/script-tests/dfg-constant-fold-uncaptured-variable-that-is-later-captured.js:
  • fast/js/script-tests/dfg-convert-this-dom-window.js:
  • fast/js/script-tests/dfg-convert-this-object-then-exit-on-other.js:
  • fast/js/script-tests/dfg-convert-this-other-then-exit-on-object.js:
  • fast/js/script-tests/dfg-convert-this-polymorphic-object-then-exit-on-other.js:
  • fast/js/script-tests/dfg-convert-this-polymorphic-object-then-exit-on-string.js:
  • fast/js/script-tests/dfg-create-inlined-arguments-in-closure-inline.js:
  • fast/js/script-tests/dfg-cross-global-object-inline-new-array-literal-with-variables.js:

(done):
(doit):

  • fast/js/script-tests/dfg-cross-global-object-inline-new-array-literal.js:

(done):
(doit):

12:05 PM WebKitGTK/2.0.x edited by berto@igalia.com
(diff)
12:01 PM Changeset in webkit [155201] by fpizlo@apple.com
  • 46 edits in trunk

REGRESSION(149636, merged in 153145): ToThis conversion doesn't work in the DFG
​https://bugs.webkit.org/show_bug.cgi?id=120781

Reviewed by Mark Hahnenberg.

Roll this back in with a build fix.

  • Use some method table hacks to detect if the CheckStructure optimization is valid for to_this.


  • Introduce a FinalObjectUse and use it for ToThis->Identity conversion.


This looks like it might be perf-neutral on the major benchmarks, but it
introduces some horrible performance cliffs. For example if you add methods to
the Array prototype, you'll get horrible performance cliffs. As in virtual calls
to C++ every time you call a JS function even if it's inlined.
LongSpider/3d-cube appears to hit this.

  • dfg/DFGAbstractInterpreterInlines.h:

(JSC::DFG::::executeEffects):

  • dfg/DFGByteCodeParser.cpp:

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

  • dfg/DFGFixupPhase.cpp:

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

  • dfg/DFGRepatch.cpp:

(JSC::DFG::emitPutTransitionStub):

  • dfg/DFGSafeToExecute.h:

(JSC::DFG::SafeToExecuteEdge::operator()):

  • dfg/DFGSpeculativeJIT.cpp:

(JSC::DFG::SpeculativeJIT::speculateFinalObject):
(JSC::DFG::SpeculativeJIT::speculate):

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

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

  • dfg/DFGSpeculativeJIT64.cpp:

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

  • dfg/DFGUseKind.cpp:

(WTF::printInternal):

  • dfg/DFGUseKind.h:

(JSC::DFG::typeFilterFor):
(JSC::DFG::isCell):

11:51 AM Changeset in webkit [155200] by fpizlo@apple.com
  • 44 edits
    1 add in trunk

Introduce a way to run benchmarks and JSRegress as stress tests with different jsc command-line options
​https://bugs.webkit.org/show_bug.cgi?id=120808

Source/JavaScriptCore:

Reviewed by Mark Hahnenberg and rubber stamped by Geoffrey Garen.

Allow --useExperimentalFTL=true even if FTL isn't built since this simplifies
testing.

  • dfg/DFGTierUpCheckInjectionPhase.cpp:

(JSC::DFG::TierUpCheckInjectionPhase::run):

Tools:

Reviewed by Mark Hahnenberg and rubber stamped by Geoffrey Garen.

Add a script for running stress tests. A stress test is a .js file that is run
through different configurations of JSC. It can control which configurations it
runs by using "@ <ruby code>" to guide the script.

This script is now run as part of run-javascriptcore-tests, on Mac only.

  • Scripts/run-javascriptcore-tests:
  • Scripts/run-jsc-stress-tests: Added.

LayoutTests:

Reviewed by Mark Hahnenberg and rubber stamped by Geoffrey Garen.

Make it so that long-running tests aren't run as part of the stress test (they
will still run as part of benchmarks and LayoutTests) or run them with fewer
configurations.

  • fast/js/regress/script-tests/ArrayBuffer-DataView-alloc-large-long-lived.js:
  • fast/js/regress/script-tests/ArrayBuffer-DataView-alloc-long-lived.js:
  • fast/js/regress/script-tests/ArrayBuffer-Int32Array-byteOffset.js:
  • fast/js/regress/script-tests/ArrayBuffer-Int8Array-alloc-huge-long-lived.js:
  • fast/js/regress/script-tests/ArrayBuffer-Int8Array-alloc-large-long-lived-fragmented.js:
  • fast/js/regress/script-tests/ArrayBuffer-Int8Array-alloc-large-long-lived.js:
  • fast/js/regress/script-tests/ArrayBuffer-Int8Array-alloc-long-lived-buffer.js:
  • fast/js/regress/script-tests/ArrayBuffer-Int8Array-alloc-long-lived.js:
  • fast/js/regress/script-tests/ArrayBuffer-Int8Array-alloc.js:
  • fast/js/regress/script-tests/DataView-custom-properties.js:
  • fast/js/regress/script-tests/Float32Array-to-Float64Array-set.js:
  • fast/js/regress/script-tests/Float64Array-alloc-long-lived.js:
  • fast/js/regress/script-tests/Float64Array-to-Int16Array-set.js:
  • fast/js/regress/script-tests/HashMap-put-get-iterate-keys.js:
  • fast/js/regress/script-tests/HashMap-put-get-iterate.js:
  • fast/js/regress/script-tests/HashMap-string-put-get-iterate.js:
  • fast/js/regress/script-tests/Int16Array-alloc-long-lived.js:
  • fast/js/regress/script-tests/Int16Array-to-Int32Array-set.js:
  • fast/js/regress/script-tests/Int32Array-alloc-huge-long-lived.js:
  • fast/js/regress/script-tests/Int32Array-alloc-huge.js:
  • fast/js/regress/script-tests/Int32Array-alloc-large-long-lived.js:
  • fast/js/regress/script-tests/Int32Array-alloc-large.js:
  • fast/js/regress/script-tests/Int32Array-alloc-long-lived.js:
  • fast/js/regress/script-tests/Int32Array-alloc.js:
  • fast/js/regress/script-tests/Int8Array-alloc-long-lived.js:
  • fast/js/regress/script-tests/array-nonarray-polymorhpic-access.js:
  • fast/js/regress/script-tests/basic-set.js:
  • fast/js/regress/script-tests/emscripten-memops.js:
  • fast/js/regress/script-tests/inline-arguments-local-escape.js:
  • fast/js/regress/script-tests/method-on-number.js:
  • fast/js/regress/script-tests/nested-function-parsing-random.js:
  • fast/js/regress/script-tests/nested-function-parsing.js:
  • fast/js/regress/script-tests/new-array-buffer-push.js:
  • fast/js/regress/script-tests/new-array-push.js:
  • fast/js/regress/script-tests/poly-stricteq.js:
  • fast/js/regress/script-tests/splice-to-remove.js:
  • fast/js/regress/script-tests/string-equality.js:
  • fast/js/regress/script-tests/string-repeat-arith.js:
  • fast/js/regress/script-tests/string-sub.js:
11:28 AM Changeset in webkit [155199] by eric.carlson@apple.com
  • 5 edits in trunk

[MediaStream API] Allow empty MediaStreams
​https://bugs.webkit.org/show_bug.cgi?id=120815

Reviewed by Jer Noble.

The spec is a bit unclear ecactly what should happen with newly created MediaStreams
that doesn't have any tracks. The current implementation is too strict and the spec
is being worked upon to be clearer. In the meantime this patch doesn't set eneded to
true in the constructor.

Source/WebCore:

No new tests, existing test updated.

(WebCore::MediaStreamRegistry::lookupMediaStreamDescriptor):

  • platform/mediastream/MediaStreamDescriptor.h:

(WebCore::MediaStreamDescriptor::MediaStreamDescriptor):

LayoutTests:

  • fast/mediastream/MediaStreamConstructor-expected.txt:
  • fast/mediastream/MediaStreamConstructor.html:
11:21 AM Changeset in webkit [155198] by andersca@apple.com
  • 5 edits in trunk/Source/WebCore

Stop using fastNew/fastDelete in WebCore
​https://bugs.webkit.org/show_bug.cgi?id=120867

Reviewed by Geoffrey Garen.

Using fastNew/fastDelete can be dangerous, especially when put into a smart pointer
such as OwnPtr which uses regular delete. Because of this I'd like to remove fastNew/fastDelete.
Turns out it's only used in a couple of places in WebCore, so just use new/delete here instead.

  • platform/audio/FFTFrame.h:
  • platform/audio/gstreamer/FFTFrameGStreamer.cpp:

(WebCore::FFTFrame::FFTFrame):
(WebCore::FFTFrame::~FFTFrame):
(WebCore::FFTFrame::doFFT):
(WebCore::FFTFrame::doInverseFFT):

  • platform/graphics/gstreamer/MediaPlayerPrivateGStreamerBase.cpp:

(WebCore::MediaPlayerPrivateGStreamerBase::MediaPlayerPrivateGStreamerBase):
(WebCore::MediaPlayerPrivateGStreamerBase::~MediaPlayerPrivateGStreamerBase):

  • platform/graphics/gstreamer/VideoSinkGStreamer.cpp:

(webkitVideoSinkDispose):

11:19 AM Changeset in webkit [155197] by commit-queue@webkit.org
  • 3 edits in trunk/Tools

[GTK] Desktop proxy settings are ignored inside the internal jhbuild
​https://bugs.webkit.org/show_bug.cgi?id=120833

Patch by Mario Sanchez Prada <​mario.prada@samsung.com> on 2013-09-06
Reviewed by Gustavo Noronha Silva.

Add dconf as an optional module for the internal jhbuild, along
with vala 0.17 as a dependency, and make sure we do not explicitly
set always the 'memory' backend, so we can have an easy way to
allow apps access desktop settings from the jhbuild. This is
useful for example to be able to run apps such as MiniBrowser
inside the jhbuild (by means of run-launcher, for instance) while
working behind a proxy, since those network settings have to be
read from the system global configuration in org.gnome.system.proxy.

  • gtk/jhbuild-optional.modules: Added dconf 0.14 and vala 0.17.
  • gtk/jhbuildrc: Do not explicitly choose the GSettings memory

backend here, since that will make more difficult to use the dconf
backend from inside the jhbuild if we need to do it so.

11:11 AM Changeset in webkit [155196] by ap@apple.com
  • 2 edits in trunk/Tools

Lots of pixel tests are failing on first run, so tests cannot finish
​https://bugs.webkit.org/show_bug.cgi?id=120755

Reviewed by Ryosuke Niwa.

For whatever reason, many reftests are color profile sensitive, and we only set
color profile to a known one for pixel tests, not for reftests.

Pixel testing mode is also enabled when retrying tests, so these failures used to
be essentially hidden - the tests were "flaky", but that was not reported loudly enough.
But now, so many tests fail that it makes run-webkit-tests stop early, and never retry.

This probably shouldn't happen, as reftests should not be color profile sensitive.
The fix is a stopgap measure to unblock EWS until a better fix can be investigated.

10:33 AM Changeset in webkit [155195] by andersca@apple.com
  • 2 edits in trunk/Source/WebCore

Fix fastMalloc/delete mismatches in CSS parser
​https://bugs.webkit.org/show_bug.cgi?id=120864

Reviewed by Andreas Kling.

Stop putting fastMalloced memory inside an OwnPtr.

  • css/CSSParserValues.cpp:

(WebCore::CSSParserSelector::CSSParserSelector):
(WebCore::CSSParserSelector::adoptSelectorVector):

10:23 AM Changeset in webkit [155194] by akling@apple.com
  • 23 edits in trunk/Source

FrameLoader::subresourceLoader() should return a reference.
<​https://webkit.org/b/120830>

Reviewed by Anders Carlsson.

There is always a SubresourceLoader, so this method could never return null.
Also made m_subresourceLoader an OwnPtr to reduce header dependencies.

9:48 AM Changeset in webkit [155193] by berto@igalia.com
  • 2 edits in trunk/Source/WTF

[GTK] GlibUtilities: getCurrentExecutablePath() does not compile on GNU/Hurd
​https://bugs.webkit.org/show_bug.cgi?id=120793

Reviewed by Gustavo Noronha Silva.

Original patch by Svante Signell <​svante.signell@telia.com>.

PATH_MAX is not defined in GNU/Hurd, this patch adds a dummy
implementation to make it compile.

  • wtf/gobject/GlibUtilities.cpp:

(getCurrentExecutablePath):

9:32 AM Changeset in webkit [155192] by commit-queue@webkit.org
  • 8 edits in trunk

[GTK] AccessibilityUIElement::addNotificationListener() crashes on debug build
​https://bugs.webkit.org/show_bug.cgi?id=120416

Patch by Denis Nomiyama <​d.nomiyama@samsung.com> on 2013-09-06
Reviewed by Mario Sanchez Prada.

Tools:

Fixed crashes when running debug DRT. Simplified loops at AccessibilityCallbackAtk.cpp where the HashMap
iterator was removed inside a loop. Fixed AccessibilityUIElement::addNotificationListener() where
m_notificationHandler expected RefPtr.

The global notification handler was stored in the HashMap with key 0. And this caused an assertion when
HashMap::add() or find() are called. To fix it, moved the global handler to a separated pointer.

  • DumpRenderTree/atk/AccessibilityCallbacks.h: Removed the global notification key.
  • DumpRenderTree/atk/AccessibilityCallbacksAtk.cpp:

(axObjectEventListener): Simplified the code by using HashMap::find() and a separate pointer for the
global notification handler.
(addAccessibilityNotificationHandler): Simplified the code by using HashMap::find() and a separate pointer
for the global notification handler.
(removeAccessibilityNotificationHandler): Added the removal for the global notification handler.

  • DumpRenderTree/atk/AccessibilityNotificationHandlerAtk.h:

(AccessibilityNotificationHandler::create): Added static function to create
AccessibilityNotificationHandler.

  • DumpRenderTree/atk/AccessibilityUIElementAtk.cpp:

(AccessibilityUIElement::addNotificationListener): Assigned m_notificationHandler with
AccessibilityNotificationHandler::create().

LayoutTests:

Fixed crashes when running debug DRT on tests that require an a11y notification handler.

  • platform/gtk/TestExpectations: Unskipped some a11y tests that were crashing before this fix.

Updated the expectation of accessibility/notification-listeners.html to expect failure on Debug build
since it is not crashing anymore. This other issue is tracked in a separate bug (bug 120669).

9:13 AM Changeset in webkit [155191] by Chris Fleizach
  • 4 edits
    2 adds in trunk

AX: aria-relevant does not expose AXARIARelevant
​https://bugs.webkit.org/show_bug.cgi?id=120373

Reviewed by Darin Adler.

Source/WebCore:

When global ARIA attributes are present on an Element we need to ignore
a presentational role, if set on that Element.

Test: accessibility/global-aria-attributes-invalidate-presentational.html

  • accessibility/AccessibilityNodeObject.cpp:

(WebCore::AccessibilityNodeObject::determineAriaRoleAttribute):

  • accessibility/AccessibilityObject.cpp:

(WebCore::AccessibilityObject::supportsARIAAttributes):

LayoutTests:

  • accessibility/global-aria-attributes-invalidate-presentational-expected.txt: Added.
  • accessibility/global-aria-attributes-invalidate-presentational.html: Added.
9:12 AM Changeset in webkit [155190] by commit-queue@webkit.org
  • 3 edits in trunk/Source/WebCore

[Qt][BlackBerry] Text track localized string functions are named wrong
​https://bugs.webkit.org/show_bug.cgi?id=120788

Patch by Brendan Long <​b.long@cablelabs.com> on 2013-09-06
Reviewed by Philippe Normand.

This renames textTrackOffText() to textTrackOffMenuItemText(), and
adds textTrackAutomaticMenuItemText(), to make them match the
definitions in LocalizedStrings.h.

  • platform/blackberry/LocalizedStringsBlackBerry.cpp:
  • platform/qt/LocalizedStringsQt.cpp:
9:03 AM Changeset in webkit [155189] by commit-queue@webkit.org
  • 2 edits in trunk/Source/ThirdParty/ANGLE

[Win][WebGL] WebGL rendering is slow.
​https://bugs.webkit.org/show_bug.cgi?id=120841

Patch by ​peavo@outlook.com <​peavo@outlook.com> on 2013-09-06
Reviewed by Brent Fulgham.

Rendering WebGL content is slow on Windows, because lots of debug information is written to debug.txt.

  • ANGLE.vcxproj/ANGLERelease.props: Disable logging in release builds.
6:33 AM Changeset in webkit [155188] by Csaba Osztrogonác
  • 5 edits in trunk/Tools

[Qt] REGRESSION(r155140) Pixel tests is still broken on Qt with QT_WEBKIT_DISABLE_UIPROCESS_DUMPPIXELS=1
​https://bugs.webkit.org/show_bug.cgi?id=120847

Patch by Gabor Abraham <​abrhm@inf.u-szeged.hu> on 2013-09-06
Reviewed by Csaba Osztrogonác.

Move the force repaint code back to if (PlatformWebView::windowSnapshotEnabled()) block to support Qt.
Typo fixed: windowShapshotEnabled -> windowSnapshotEnabled.
Add a default PlatformWebView::windowSnapshotEnabled() to the !PLATFORM(QT).

  • WebKitTestRunner/PlatformWebView.h:

(WTR::PlatformWebView::windowSnapshotEnabled):

  • WebKitTestRunner/TestInvocation.cpp:

(WTR::TestInvocation::dumpResults):

  • WebKitTestRunner/qt/PlatformWebViewQt.cpp:

(WTR::WrapperWindow::handleStatusChanged):
(WTR::PlatformWebView::windowSnapshotEnabled):

  • WebKitTestRunner/qt/TestInvocationQt.cpp:

(WTR::TestInvocation::dumpPixelsAndCompareWithExpected):

5:07 AM Changeset in webkit [155187] by allan.jensen@digia.com
  • 2 edits in trunk/Tools

REGRESSION(r155140) Broke pixel tests on EFL/GTK/Qt
​https://bugs.webkit.org/show_bug.cgi?id=120843

Reviewed by Csaba Osztrogonác.

Call notifyDone like the platform specific forceRepaintDoneCallback methods
it replaced did. The method is empty on the Apple ports.

  • WebKitTestRunner/TestInvocation.cpp:

(WTR::TestInvocation::forceRepaintDoneCallback):

4:43 AM Changeset in webkit [155186] by commit-queue@webkit.org
  • 1 edit
    9 deletes in trunk/LayoutTests

[CSS Regions] Remove orphan test results
​https://bugs.webkit.org/show_bug.cgi?id=120845

Patch by Radu Stavila <​stavila@adobe.com> on 2013-09-06
Reviewed by Antti Koivisto.

Removed remaining orphan test results after ​https://bugs.webkit.org/show_bug.cgi?id=120760.

  • platform/efl/fast/regions/overflow-not-moving-below-floats-in-variable-width-regions-expected.png: Removed.
  • platform/efl/fast/regions/overflow-not-moving-below-floats-in-variable-width-regions-expected.txt: Removed.
  • platform/gtk/fast/regions/overflow-not-moving-below-floats-in-variable-width-regions-expected.png: Removed.
  • platform/gtk/fast/regions/overflow-not-moving-below-floats-in-variable-width-regions-expected.txt: Removed.
  • platform/mac/fast/regions/overflow-not-moving-below-floats-in-variable-width-regions-expected.png: Removed.
  • platform/mac/fast/regions/overflow-not-moving-below-floats-in-variable-width-regions-expected.txt: Removed.
  • platform/qt-wk2/fast/regions/overflow-not-moving-below-floats-in-variable-width-regions-expected.png: Removed.
  • platform/qt/fast/regions/overflow-not-moving-below-floats-in-variable-width-regions-expected.png: Removed.
  • platform/qt/fast/regions/overflow-not-moving-below-floats-in-variable-width-regions-expected.txt: Removed.
4:21 AM Changeset in webkit [155185] by jinwoo7.song@samsung.com
  • 1 edit
    3 adds in trunk/LayoutTests

[EFL] Unreviewed gardening. Added missing results.

  • platform/efl/fast/css-generated-content/details-summary-before-after-expected.txt: Added.
  • platform/efl/fast/regions/region-dynamic-after-before-expected.txt: Added.
  • platform/efl/fast/regions/region-generated-content-before-after-expected.txt: Added.
4:04 AM Changeset in webkit [155184] by simon.pena@samsung.com
  • 4 edits
    3 adds
    2 deletes in trunk/LayoutTests

[GTK] Unreviewed gardening. Update TestExpectations.

  • platform/gtk-wk2/TestExpectations: Skip speech input tests, mark as failing drag and drop tests.
  • platform/gtk-wk2/plugins/npruntime/object-from-destroyed-plugin-expected.txt: Added.
  • platform/gtk-wk2/plugins/npruntime/object-from-destroyed-plugin-in-subframe-expected.txt: Added.
  • platform/gtk/TestExpectations: Updated failing tests.
  • platform/gtk/animations/cross-fade-background-image-expected.png: Removed.
  • platform/gtk/animations/cross-fade-background-image-expected.txt: Removed.
  • platform/gtk/editing/spelling/spellcheck-paste-expected.txt: Rebaseline after r154869.
3:11 AM Changeset in webkit [155183] by allan.jensen@digia.com
  • 2 edits in trunk/Source/WebCore

[Qt][WK2] Cached credentials not cleared between tests
​https://bugs.webkit.org/show_bug.cgi?id=120775

Reviewed by Jocelyn Turcotte.

Clear the access credentials when resetting the state of a page for testing.

  • testing/Internals.cpp:

(WebCore::Internals::resetToConsistentState):

2:56 AM Changeset in webkit [155182] by allan.jensen@digia.com
  • 2 edits in trunk/Tools

[Qt][WK2] Window used by WebKitTestRunner loses active status
​https://bugs.webkit.org/show_bug.cgi?id=120774

Reviewed by Jocelyn Turcotte.

When focusing the PlatformWebView which is done at the start of each test, we should
ensure the host window of PlatformWebView is also made active, otherwise we will
not receive the keyboard events send by EventSenderProxy.

  • WebKitTestRunner/qt/PlatformWebViewQt.cpp:

(WTR::PlatformWebView::focus):

1:04 AM Changeset in webkit [155181] by svillar@igalia.com
  • 6 edits
    2 adds in trunk

[CSS Grid Layout] Resolve named grid lines
​https://bugs.webkit.org/show_bug.cgi?id=119632

Reviewed by Andreas Kling.

From Blink r152659, r153794 by <​jchaffraix@chromium.org>

Source/WebCore:

Test: fast/css-grid-layout/grid-item-named-grid-line-resolution.html

This change adds the code which translates from a raw named grid
line (stored in RenderStyle) to its actual grid position. This
even includes named grid resolution for grid lines with the 'span'
keyword.

  • rendering/RenderGrid.cpp:

(WebCore::RenderGrid::explicitGridSizeForSide):
(WebCore::adjustGridPositionForRowEndColumnEndSide):
(WebCore::adjustGridPositionForSide):
(WebCore::RenderGrid::resolveNamedGridLinePositionFromStyle):
(WebCore::RenderGrid::resolveGridPositionFromStyle):
(WebCore::RenderGrid::resolveGridPositionAgainstOppositePosition):
(WebCore::RenderGrid::resolveNamedGridLinePositionAgainstOppositePosition):
(WebCore::RenderGrid::resolveRowStartColumnStartNamedGridLinePositionAgainstOppositePosition):
(WebCore::RenderGrid::resolveRowEndColumnEndNamedGridLinePositionAgainstOppositePosition):

  • rendering/RenderGrid.h:

LayoutTests:

Added a new test to check named grid lines resolution. This also
provides some more new test cases with named grid lines and
negative positions.

  • fast/css-grid-layout/grid-item-named-grid-line-resolution-expected.txt: Added.
  • fast/css-grid-layout/grid-item-named-grid-line-resolution.html: Added.
  • fast/css-grid-layout/grid-item-negative-position-resolution-expected.txt:
  • fast/css-grid-layout/grid-item-negative-position-resolution.html:
12:18 AM Changeset in webkit [155180] by zandobersek@gmail.com
  • 2 edits in trunk/Source/JavaScriptCore

Unreviewed build fix for the GTK port when building with FTL JIT enabled.

  • GNUmakefile.list.am: Add the missing files to the build.
12:02 AM Changeset in webkit [155179] by krit@webkit.org
  • 7 edits in trunk

Allow string as image reference on filter()
​https://bugs.webkit.org/show_bug.cgi?id=120829

Reviewed by Andreas Kling.

Source/WebCore:

Beside <image> (which already includes url(), gradients, cross-fade() and
image-set()), the filter() image function should support <string> as image
reference.

​http://dev.w3.org/fxtf/filters/#FilterCSSImageValue

  • css/CSSParser.cpp:

(WebCore::CSSParser::parseFilterImage): Create CSSImageValue from string.

LayoutTests:

Check that string can be used as image reference, that we
interpolate between the same images once referenced by string
and once by url() and that other tokens are not excepted.

  • fast/filter-image/filter-image-animation-expected.txt:
  • fast/filter-image/filter-image-animation.html:
  • fast/filter-image/parse-filter-image-expected.txt:
  • fast/filter-image/parse-filter-image.html:

Sep 5, 2013:

11:00 PM Changeset in webkit [155178] by krit@webkit.org
  • 3 edits
    2 adds in trunk

Support SVG filters on -webkit-filter() function
​https://bugs.webkit.org/show_bug.cgi?id=120806

Reviewed by Darin Adler.

Source/WebCore:

Add support for SVG Filters on CSS -webkit-filter() image function.

Test: fast/filter-image/filter-image-svg.html

  • css/CSSFilterImageValue.cpp:

(WebCore::CSSFilterImageValue::image): Pass renderer to filter builder.

LayoutTests:

Test that the -webkit-filter() function can reference a SVG Filter
from the document.

  • fast/filter-image/filter-image-svg-expected.html: Added.
  • fast/filter-image/filter-image-svg.html: Added.
10:50 PM Changeset in webkit [155177] by oliver@apple.com
  • 5 edits in trunk/Source/JavaScriptCore

Make it simpler to introduce new data types to the global object
​https://bugs.webkit.org/show_bug.cgi?id=120801

Reviewed by Gavin Barraclough.

Add an iterator macro that lists all the "simple" ES types (e.g. type
consists of instance, constructor, and prototype classes). So that
we don't need to have every new type litter JSGlobalObject.{cpp,h} with
members, accessors, and manual GC visiting.

  • runtime/JSGlobalObject.cpp:

(JSC::JSGlobalObject::visitChildren):

  • runtime/JSGlobalObject.h:
10:48 PM Changeset in webkit [155176] by commit-queue@webkit.org
  • 5 edits in trunk/Source/WebCore

DRY out srcset related deviceScaleFactor calculations
​https://bugs.webkit.org/show_bug.cgi?id=120791

Identical deviceScaleFactor calculations were performed in 3 different locations.
I've added that calculation as a method of Document.
Previous calculations are replaced by calls to this method.

Patch by Yoav Weiss <​yoav@yoav.ws> on 2013-09-05
Reviewed by Andreas Kling.

No new tests since this is a refactoring change. No functionality have changed.

  • dom/Document.cpp:

(WebCore::Document::deviceScaleFactor):

  • dom/Document.h:
  • html/HTMLImageElement.cpp:

(WebCore::HTMLImageElement::parseAttribute):

  • html/parser/HTMLDocumentParser.cpp:

(WebCore::HTMLDocumentParser::pumpTokenizer):
(WebCore::HTMLDocumentParser::insert):

10:47 PM Changeset in webkit [155175] by commit-queue@webkit.org
  • 4 edits in trunk

[Qt] DefaultFullScreenVideoHandler and PlatformVideoWindow are included in the build when they are disabled
​https://bugs.webkit.org/show_bug.cgi?id=117206

Patch by Brendan Long <​b.long@cablelabs.com> on 2013-09-05
Reviewed by Philippe Normand.

.:

  • Source/widgetsapi.pri: Don't include DefaultFullScreenVideoHandler when it's disabled.

Source/WebCore:

No new tests because this just fixes a build warning.

  • Target.pri: Don't include PlatformVideoWindow* files in the build when they are disabled.
9:49 PM Changeset in webkit [155174] by akling@apple.com
  • 8 edits in trunk/Source

FrameView: Constructor should take Frame&.
<​https://webkit.org/b/120824>

Reviewed by Anders Carlsson.

  • WebCore.exp.in:
  • inspector/InspectorOverlay.cpp:

(WebCore::InspectorOverlay::overlayPage):

  • page/Frame.cpp:

(WebCore::Frame::createView):

  • page/FrameView.cpp:

(WebCore::FrameView::FrameView):
(WebCore::FrameView::create):

  • page/FrameView.h:
  • svg/graphics/SVGImage.cpp:

(WebCore::SVGImage::dataChanged):

8:57 PM Changeset in webkit [155173] by jinwoo7.song@samsung.com
  • 2 edits in trunk/LayoutTests

[EFL] Unreviewed gardening.

  • platform/efl-wk1/TestExpectations: Failing after r155014.
8:50 PM Changeset in webkit [155172] by mitz@apple.com
  • 2 edits in trunk/Source/ThirdParty/ANGLE

INSTALL_PATH_PREFIX is used even when not installing
​https://bugs.webkit.org/show_bug.cgi?id=120818

Reviewed by Darin Adler.

  • Configurations/ANGLE.xcconfig: Changed the definition of PRIVATE_HEADERS_FOLDER_PATH

to include INSTALL_PATH_PREFIX only when installing.

8:50 PM Changeset in webkit [155171] by mitz@apple.com
  • 3 edits in trunk/Source/WTF

INSTALL_PATH_PREFIX is used even when not installing
​https://bugs.webkit.org/show_bug.cgi?id=120810

Reviewed by Andy Estes.

  • Configurations/CopyWTFHeaders.xcconfig: Removed INSTALL_PATH_PREFIX from

the definition of PRIVATE_HEADERS_FOLDER_PATH. Also removed the leading
slash, because the Xcode definition of this build setting is relative.

  • WTF.xcodeproj/project.pbxproj: Prepend INSTALL_PATH_PREFIX here when installing.
8:43 PM Changeset in webkit [155170] by jinwoo7.song@samsung.com
  • 3 edits in trunk/Source/WebCore

Fix build warnings by unused parameter
​https://bugs.webkit.org/show_bug.cgi?id=120821

Reviewed by Anders Carlsson.

Fix warning by removing unused parameter.

  • platform/graphics/gstreamer/TextSinkGStreamer.cpp:

(webkitTextSinkGetProperty):
(webkitTextSinkSetProperty):

  • platform/network/soup/SynchronousLoaderClientSoup.cpp:

(WebCore::SynchronousLoaderClient::didReceiveAuthenticationChallenge):

8:33 PM Changeset in webkit [155169] by commit-queue@webkit.org
  • 2 edits in trunk/Source/WebInspectorUI

Web Inspector: Breakpoint Action JS Editor has line wrapping issues with long no-space lines
​https://bugs.webkit.org/show_bug.cgi?id=120822

Patch by Joseph Pecoraro <Joseph Pecoraro> on 2013-09-05
Reviewed by Timothy Hatcher.

  • UserInterface/BreakpointActionView.css:

(.breakpoint-action-eval-editor > .CodeMirror):
(.breakpoint-action-eval-editor > .CodeMirror-scroll):

6:34 PM Changeset in webkit [155168] by akling@apple.com
  • 2 edits in trunk/Source/WebCore

FrameLoader: Convert three little loops to using Vector<Ref<Frame>>.
<​https://webkit.org/b/120798>

Reviewed by Antti Koivisto.

Also apply a bit of inline capacity + reserveInitialCapacity + uncheckedAppend
in FrameLoader::detachChildren() since we know the final size from the start.

  • loader/FrameLoader.cpp:

(WebCore::FrameLoader::detachChildren):
(WebCore::FrameLoader::checkLoadComplete):
(WebCore::FrameLoader::shouldClose):

6:24 PM Changeset in webkit [155167] by akling@apple.com
  • 2 edits in trunk/Source/WebCore

ContainerNode: Apply Ref<T> to some popular DOM functions.
<​https://webkit.org/b/120800>

Reviewed by Antti Koivisto.

Use Ref to avoid a bunch of null checks in some popular DOM functions.

  • dom/ContainerNode.cpp:

(WebCore::ContainerNode::insertBefore):

'refChild' is never null here, so store it in a Ref.

(WebCore::ContainerNode::replaceChild):

'removedChild' is never null here, so store it in a Ref.

(WebCore::ContainerNode::removeChild):

'child' is never null here, so store it in a Ref.
Also return "true" at the end of the function since we can't
null-check 'child' anymore (not that it would ever be false!)

(WebCore::dispatchChildInsertionEvents):
(WebCore::dispatchChildRemovalEvents):

Use Ref<Document> in both of these methods.

6:08 PM Changeset in webkit [155166] by mrowe@apple.com
  • 11 edits in trunk/Source/JavaScriptCore

Roll out r155149 since it broke the build.

5:55 PM Changeset in webkit [155165] by commit-queue@webkit.org
  • 2 edits in trunk/Source/WebInspectorUI

Web Inspector: Edit Breakpoint popover sometimes appears misplaced in top left
​https://bugs.webkit.org/show_bug.cgi?id=120804

Patch by Joseph Pecoraro <Joseph Pecoraro> on 2013-09-05
Reviewed by Timothy Hatcher.

Grab the bounding rect immediately, instead of grabbing it from the element
after the user has selected the "Edit Breakpoint" context menu item. When
the popover was misplaced it was when using an element had been removed or
replaced in the DOM, and caused a bogus bounding client rect.

  • UserInterface/Breakpoint.js:

(WebInspector.Breakpoint.prototype.editBreakpoint):
(WebInspector.Breakpoint.prototype._showEditBreakpointPopover):

5:35 PM Changeset in webkit [155164] by akling@apple.com
  • 9 edits in trunk/Source

ScrollView::children() should return a reference.
<​https://webkit.org/b/120795>

Reviewed by Anders Carlsson.

Source/WebCore:

This function was already just returning the address of a member variable.
Modernized some loops that were using it.

  • bindings/js/PageScriptDebugServer.cpp:

(WebCore::PageScriptDebugServer::setJavaScriptPaused):

  • page/FrameView.cpp:

(WebCore::FrameView::hasCustomScrollbars):
(WebCore::FrameView::updateLayoutAndStyleIfNeededRecursive):

  • page/Page.cpp:

(WebCore::Page::pluginViews):

  • page/scrolling/ScrollingCoordinator.cpp:

(WebCore::ScrollingCoordinator::computeNonFastScrollableRegion):

  • platform/ScrollView.h:

(WebCore::ScrollView::children):

  • plugins/IFrameShimSupport.cpp:

(WebCore::getPluginOcclusions):

Source/WebKit/mac:

  • WebView/WebView.mm:

(-[WebView _addScrollerDashboardRegionsForFrameView:dashboardRegions:]):

5:23 PM Changeset in webkit [155163] by ryuan.choi@samsung.com
  • 8 edits in trunk

[CMAKE] Add c++0x into CXX_FLAGS as a default
​https://bugs.webkit.org/show_bug.cgi?id=120812

Unreviewed build fix for EFL ports.

.:

  • Source/cmake/OptionsCommon.cmake: Moved c++0x option here from WebKitHelpers.
  • Source/cmake/WebKitHelpers.cmake:

Source/WebKit/efl:

  • tests/UnitTestUtils/EWKTestConfig.h: Add space between strings.

Source/WebKit2:

  • UIProcess/API/efl/tests/UnitTestUtils/EWK2UnitTestBase.h: Added constexpr for in-class initialization.
  • UIProcess/API/efl/tests/UnitTestUtils/EWK2UnitTestEnvironment.cpp: Added space between strings.

(EWK2UnitTest::EWK2UnitTestEnvironment::defaultTestPageUrl):
(EWK2UnitTest::EWK2UnitTestEnvironment::defaultTheme):
(EWK2UnitTest::EWK2UnitTestEnvironment::urlForResource):

5:18 PM Changeset in webkit [155162] by Beth Dakin
  • 73 edits
    2 deletes in trunk/Source/WebCore

Reverting revisions 155139, 155141, 155142, and 155145 since they appear to have
caused about 50 new test failures.

  • CMakeLists.txt:
  • GNUmakefile.list.am:
  • Target.pri:
  • WebCore.vcxproj/WebCore.vcxproj:
  • WebCore.xcodeproj/project.pbxproj:
  • accessibility/AccessibilityRenderObject.cpp:

(WebCore::AccessibilityRenderObject::computeAccessibilityIsIgnored):
(WebCore::AccessibilityRenderObject::determineAccessibilityRole):

  • dom/Position.cpp:

(WebCore::Position::isCandidate):
(WebCore::Position::getInlineBoxAndOffset):
(WebCore::Position::primaryDirection):

  • dom/PositionIterator.cpp:

(WebCore::PositionIterator::isCandidate):

  • editing/ApplyStyleCommand.cpp:

(WebCore::ApplyStyleCommand::applyInlineStyleToPushDown):

  • editing/CompositeEditCommand.cpp:

(WebCore::CompositeEditCommand::addBlockPlaceholderIfNeeded):

  • editing/Editor.cpp:

(WebCore::Editor::baseWritingDirectionForSelectionStart):

  • editing/FrameSelection.cpp:

(WebCore::caretRenderer):

  • editing/TextIterator.cpp:

(WebCore::TextIterator::shouldRepresentNodeOffsetZero):

  • editing/TypingCommand.cpp:

(WebCore::TypingCommand::makeEditableRootEmpty):

  • editing/htmlediting.cpp:

(WebCore::isBlockFlowElement):

  • html/HTMLDetailsElement.cpp:

(WebCore::HTMLDetailsElement::createRenderer):

  • html/HTMLSummaryElement.cpp:

(WebCore::HTMLSummaryElement::createRenderer):

  • html/shadow/SliderThumbElement.cpp:

(WebCore::RenderSliderThumb::RenderSliderThumb):

  • html/shadow/SliderThumbElement.h:
  • rendering/LayoutState.cpp:

(WebCore::LayoutState::LayoutState):

  • rendering/RenderBlock.cpp:

(WebCore::RenderBlock::createAnonymous):
(WebCore::RenderBlock::layoutBlock):
(WebCore::RenderBlock::collapseMargins):
(WebCore::RenderBlock::selectionGaps):
(WebCore::RenderBlock::firstLineBoxBaseline):
(WebCore::RenderBlock::lastLineBoxBaseline):
(WebCore::RenderBlock::firstLineBlock):
(WebCore::findFirstLetterBlock):
(WebCore::shouldCheckLines):
(WebCore::RenderBlock::adjustForBorderFit):

  • rendering/RenderBlock.h:
  • rendering/RenderBlockFlow.cpp: Removed.
  • rendering/RenderBlockFlow.h: Removed.
  • rendering/RenderBox.cpp:

(WebCore::RenderBox::computeRectForRepaint):
(WebCore::RenderBox::positionForPoint):

  • rendering/RenderBoxModelObject.h:

(WebCore::RenderBoxModelObject::canHaveBoxInfoInRegion):

  • rendering/RenderDeprecatedFlexibleBox.cpp:

(WebCore::RenderDeprecatedFlexibleBox::applyLineClamp):
(WebCore::RenderDeprecatedFlexibleBox::clearLineClamp):

  • rendering/RenderDetailsMarker.cpp:

(WebCore::RenderDetailsMarker::RenderDetailsMarker):

  • rendering/RenderDetailsMarker.h:
  • rendering/RenderFieldset.cpp:

(WebCore::RenderFieldset::RenderFieldset):

  • rendering/RenderFieldset.h:
  • rendering/RenderFileUploadControl.cpp:

(WebCore::RenderFileUploadControl::RenderFileUploadControl):

  • rendering/RenderFileUploadControl.h:
  • rendering/RenderFlowThread.cpp:

(WebCore::RenderFlowThread::RenderFlowThread):

  • rendering/RenderFlowThread.h:
  • rendering/RenderFullScreen.cpp:
  • rendering/RenderInline.cpp:

(WebCore::RenderInline::computeRectForRepaint):

  • rendering/RenderLineBoxList.cpp:

(WebCore::RenderLineBoxList::dirtyLinesFromChangedChild):

  • rendering/RenderListItem.cpp:

(WebCore::RenderListItem::RenderListItem):
(WebCore::RenderListItem::willBeDestroyed):
(WebCore::RenderListItem::willBeRemovedFromTree):
(WebCore::RenderListItem::layout):
(WebCore::RenderListItem::addOverflowFromChildren):
(WebCore::RenderListItem::computePreferredLogicalWidths):
(WebCore::RenderListItem::paint):

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

(WebCore::RenderMeter::RenderMeter):

  • rendering/RenderMeter.h:
  • rendering/RenderMultiColumnBlock.cpp:

(WebCore::RenderMultiColumnBlock::RenderMultiColumnBlock):

  • rendering/RenderMultiColumnBlock.h:
  • rendering/RenderObject.cpp:

(WebCore::RenderObject::createObject):
(WebCore::RenderObject::computeRectForRepaint):
(WebCore::RenderObject::styleWillChange):
(WebCore::firstLineStyleForCachedUncachedType):

  • rendering/RenderObject.h:

(WebCore::RenderObject::isBlockFlow):
(WebCore::RenderObject::isRenderBlock):

  • rendering/RenderObjectChildList.cpp:

(WebCore::RenderObjectChildList::insertChildNode):

  • rendering/RenderProgress.cpp:

(WebCore::RenderProgress::RenderProgress):

  • rendering/RenderProgress.h:
  • rendering/RenderRegion.cpp:

(WebCore::RenderRegion::RenderRegion):

  • rendering/RenderRegion.h:
  • rendering/RenderRuby.cpp:

(WebCore::RenderRubyAsBlock::RenderRubyAsBlock):

  • rendering/RenderRuby.h:
  • rendering/RenderRubyBase.cpp:

(WebCore::RenderRubyBase::RenderRubyBase):

  • rendering/RenderRubyBase.h:
  • rendering/RenderRubyRun.cpp:

(WebCore::RenderRubyRun::RenderRubyRun):

  • rendering/RenderRubyRun.h:
  • rendering/RenderRubyText.cpp:

(WebCore::RenderRubyText::RenderRubyText):

  • rendering/RenderRubyText.h:
  • rendering/RenderTableCaption.cpp:

(WebCore::RenderTableCaption::RenderTableCaption):

  • rendering/RenderTableCaption.h:
  • rendering/RenderTableCell.cpp:

(WebCore::RenderTableCell::RenderTableCell):

  • rendering/RenderTableCell.h:
  • rendering/RenderTextControl.cpp:

(WebCore::RenderTextControl::RenderTextControl):

  • rendering/RenderTextControl.h:
  • rendering/RenderTextControlSingleLine.h:

(WebCore::RenderTextControlInnerBlock::RenderTextControlInnerBlock):

  • rendering/RenderTextTrackCue.cpp:

(WebCore::RenderTextTrackCue::RenderTextTrackCue):

  • rendering/RenderTextTrackCue.h:
  • rendering/RenderView.cpp:

(WebCore::RenderView::RenderView):

  • rendering/RenderView.h:
  • rendering/RenderingAllInOne.cpp:
  • rendering/mathml/RenderMathMLOperator.cpp:

(WebCore::RenderMathMLOperator::createGlyph):

  • rendering/svg/RenderSVGBlock.cpp:

(WebCore::RenderSVGBlock::RenderSVGBlock):

  • rendering/svg/RenderSVGBlock.h:
5:16 PM Changeset in webkit [155161] by ap@apple.com
  • 2 edits in trunk/Source/WebCore

Flaky Test: platform/mac/editing/deleting/deletionUI-single-instance.html
​https://bugs.webkit.org/show_bug.cgi?id=114181

Reviewed by Ryosuke Niwa.

Editor survives navigations, and so did DeleteButtonController. But the state it holds
is document specific (like nodes in the old document).

  • editing/Editor.cpp: (WebCore::Editor::clear): Replace DeleteButtonController.
4:30 PM Changeset in webkit [155160] by Seokju Kwon
  • 2 edits in trunk/Tools

Unreviewed build fix. VectorReverse.cpp was removed after r155151.

  • TestWebKitAPI/CMakeLists.txt:
4:27 PM Changeset in webkit [155159] by msaboff@apple.com
  • 3 edits in trunk/Source/JavaScriptCore

Cleanup formatting of byte code debug output
Source/JavaScriptCore/ChangeLog

Rubber stamped by Filip Pizlo.

Put the formatting of the byte code offset and operation into one common function to
simplify and unify formatting. Changed CodeBlock::registerName() to return
"thist" for argument register 0, "argN" for other argument registers and "locN" for
local registers.

  • bytecode/CodeBlock.cpp:

(JSC::CodeBlock::registerName):
(JSC::CodeBlock::printUnaryOp):
(JSC::CodeBlock::printBinaryOp):
(JSC::CodeBlock::printConditionalJump):
(JSC::CodeBlock::printGetByIdOp):
(JSC::CodeBlock::printCallOp):
(JSC::CodeBlock::printPutByIdOp):
(JSC::CodeBlock::dumpBytecode):

  • bytecode/CodeBlock.h:

(JSC::CodeBlock::printLocationAndOp):
(JSC::CodeBlock::printLocationOpAndRegisterOperand):

4:13 PM Changeset in webkit [155158] by andersca@apple.com
  • 2 edits in trunk/Source/WTF

Fix build with older versions of clang from Xcode.

  • wtf/Compiler.h:
4:13 PM Changeset in webkit [155157] by Lucas Forschler
  • 5 edits in branches/safari-537-branch/Source

Versioning.

4:11 PM Changeset in webkit [155156] by Lucas Forschler
  • 1 copy in tags/Safari-537.69

New Tag.

3:57 PM Changeset in webkit [155155] by gyuyoung.kim@samsung.com
  • 34 edits in trunk/Source/WebCore

Generate toFooElement() functions from tagname data.
​https://bugs.webkit.org/show_bug.cgi?id=120676

Reviewed by Andreas Kling.

As r154965 introduced the isFooElement() generation, this patch adds generations of toFooElement() function
to *generateTypeChecking* attribute. To represent isFooElement() and toFooElement(), *generateTypeChecking*
is changed to *generateTypeHelpers*.

Removed all the hand-written toFooElement() that can be covered by the *generateTypeHelpers*.

  • GNUmakefile.am:
  • GNUmakefile.list.am:
  • WebCore.vcxproj/WebCore.vcxproj:
  • WebCore.xcodeproj/project.pbxproj:
  • dom/make_names.pl:

(defaultTagPropertyHash):
(printTypeHelpers):
(printTypeHelpersHeaderFile):

  • html/HTMLAnchorElement.h:
  • html/HTMLAreaElement.h:
  • html/HTMLAudioElement.h:
  • html/HTMLCanvasElement.h:
  • html/HTMLElement.h:
  • html/HTMLFormElement.h:
  • html/HTMLFrameSetElement.h:
  • html/HTMLImageElement.h:
  • html/HTMLInputElement.h:
  • html/HTMLLabelElement.h:
  • html/HTMLMapElement.h:
  • html/HTMLMeterElement.h:
  • html/HTMLOptGroupElement.h:
  • html/HTMLOptionElement.h:
  • html/HTMLProgressElement.h:
  • html/HTMLScriptElement.h:
  • html/HTMLStyleElement.h:
  • html/HTMLTableElement.h:
  • html/HTMLTableRowElement.h:
  • html/HTMLTagNames.in:
  • html/HTMLTextAreaElement.h:
  • html/HTMLTitleElement.h:
  • svg/SVGElement.h:
  • svg/SVGFontElement.h:
  • svg/SVGFontFaceElement.h:
  • svg/SVGImageElement.h:
  • svg/SVGScriptElement.h:
  • svg/svgtags.in:
3:11 PM Changeset in webkit [155154] by commit-queue@webkit.org
  • 2 edits in trunk/Source/WebInspectorUI

Web Inspector: Remove harmless assert, triggered hitting breakpoint in auto prettyprinted code on reload
​https://bugs.webkit.org/show_bug.cgi?id=120332

Patch by Joseph Pecoraro <Joseph Pecoraro> on 2013-09-05
Reviewed by Timothy Hatcher.

  • UserInterface/TextEditor.js:
3:06 PM Changeset in webkit [155153] by ap@apple.com
  • 4 edits in trunk/Source/WebKit2

WKFullScreenWindowController extends lifetime of WKView, deleting it at a wrong time
​https://bugs.webkit.org/show_bug.cgi?id=120792
<rdar://problem/14884666>

Reviewed by Jer Noble.

  • UIProcess/API/mac/WKView.mm: (-[WKView fullScreenWindowController]): Use a newly minted initializer for the controller.
  • UIProcess/mac/WKFullScreenWindowController.h: Removed unused web view accessors. Changed the class to take web view at initialization time.
  • UIProcess/mac/WKFullScreenWindowController.mm: (-[WKFullScreenWindowController initWithWindow:webView:]): Initialize the controller inone step. (-[WKFullScreenWindowController dealloc]): WebView is now a raw pointer, no need to zero it. (-[WKFullScreenWindowController close]): Make sure to not leave a dangling WKView pointer (this method is indirectly but inevitably called when WKView is deallocated).
2:52 PM Changeset in webkit [155152] by andersca@apple.com
  • 4 edits in trunk

Make Vector::uncheckedAppend work with move-only types
​https://bugs.webkit.org/show_bug.cgi?id=120799

Reviewed by Andreas Kling.

Source/WTF:

  • wtf/Vector.h:

(WTF::::uncheckedAppend):
Use std::forward to invoke the move constructor when possible.

Tools:

  • TestWebKitAPI/Tests/WTF/Vector.cpp:

(TestWebKitAPI::MoveOnly):
Add a move-only class.

(TestWebKitAPI::TEST):
Add a test for Vector<MoveOnly>::uncheckedAppend.

2:34 PM Changeset in webkit [155151] by andersca@apple.com
  • 8 edits
    2 deletes in trunk/Tools

Move all Vector tests into Vector.cpp
​https://bugs.webkit.org/show_bug.cgi?id=120797

Reviewed by Andreas Kling.

  • TestWebKitAPI/CMakeLists.txt:
  • TestWebKitAPI/GNUmakefile.am:
  • TestWebKitAPI/TestWebKitAPI.vcxproj/TestWebKitAPI.vcxproj:
  • TestWebKitAPI/TestWebKitAPI.vcxproj/TestWebKitAPI.vcxproj.filters:
  • TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj:
  • TestWebKitAPI/Tests/WTF/Vector.cpp:

(TestWebKitAPI::TEST):

  • TestWebKitAPI/Tests/WTF/VectorBasic.cpp: Removed.
  • TestWebKitAPI/Tests/WTF/VectorReverse.cpp: Removed.
  • TestWebKitAPI/Tests/WTF/WTF.pro:
2:25 PM Changeset in webkit [155150] by akling@apple.com
  • 10 edits in trunk/Source/WebCore

Cached Page and Frame don't need to be ref-counted.
<​https://webkit.org/b/120710>

Reviewed by Anders Carlsson.

  • CachedPage is owned by HistoryItem.
  • CachedFrame is owned by CachedPage.

Remove the ref counting from these objects to make the code less confusing.

Added a new method:

  • PassOwnPtr<CachedPage> PageCache::take(HistoryItem*)

..which is what it looks like - a combined get() and remove() that transfers
ownership of the CachedPage to the caller.

This is used by commitProvisionalLoad() and invalidateCurrentItemCachedPage()
to accomplish in one swoop what they used to do in awkwardly spaced steps.

  • history/CachedFrame.h:

(WebCore::CachedFrame::create):

  • history/CachedPage.cpp:

(WebCore::CachedPage::create):

  • history/CachedPage.h:
  • history/HistoryItem.h:
  • history/PageCache.cpp:

(WebCore::PageCache::take):

  • history/PageCache.h:
  • loader/FrameLoader.cpp:

(WebCore::FrameLoader::commitProvisionalLoad):
(WebCore::FrameLoader::transitionToCommitted):

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

(WebCore::HistoryController::invalidateCurrentItemCachedPage):

2:19 PM Changeset in webkit [155149] by fpizlo@apple.com
  • 11 edits in trunk/Source/JavaScriptCore

REGRESSION(149636, merged in 153145): ToThis conversion doesn't work in the DFG
​https://bugs.webkit.org/show_bug.cgi?id=120781

Reviewed by Mark Hahnenberg.

  • Use some method table hacks to detect if the CheckStructure optimization is valid for to_this.


  • Introduce a FinalObjectUse and use it for ToThis->Identity conversion.


This looks like it might be perf-neutral on the major benchmarks, but it
introduces some horrible performance cliffs. For example if you add methods to
the Array prototype, you'll get horrible performance cliffs. As in virtual calls
to C++ every time you call a JS function even if it's inlined.
LongSpider/3d-cube appears to hit this.

  • dfg/DFGAbstractInterpreterInlines.h:

(JSC::DFG::::executeEffects):

  • dfg/DFGByteCodeParser.cpp:

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

  • dfg/DFGFixupPhase.cpp:

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

  • dfg/DFGSafeToExecute.h:

(JSC::DFG::SafeToExecuteEdge::operator()):

  • dfg/DFGSpeculativeJIT.cpp:

(JSC::DFG::SpeculativeJIT::speculateFinalObject):
(JSC::DFG::SpeculativeJIT::speculate):

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

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

  • dfg/DFGSpeculativeJIT64.cpp:

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

  • dfg/DFGUseKind.cpp:

(WTF::printInternal):

  • dfg/DFGUseKind.h:

(JSC::DFG::typeFilterFor):
(JSC::DFG::isCell):

2:17 PM Changeset in webkit [155148] by andersca@apple.com
  • 2 edits in trunk/Source/WTF

Add COMPILER_SUPPORTS(CXX_AUTO_TYPE) and #error if it's 0
​https://bugs.webkit.org/show_bug.cgi?id=120794

Reviewed by Andreas Kling.

People have already begun using C++11 auto in WebCore, so let's make it a prerequisite.

  • wtf/Compiler.h:
2:07 PM Changeset in webkit [155147] by ddkilzer@apple.com
  • 2 edits in trunk/Source/WebCore

BUILD FIX (r155108): Add SynchronousLoaderClientCFNet.cpp to Xcode project

  • WebCore.xcodeproj/project.pbxproj: The iOS port defines

USE(CFNETWORK), so it needs to compile
SynchronousLoaderClientCFNet.cpp to build successfully.

2:01 PM Changeset in webkit [155146] by andersca@apple.com
  • 2 edits in trunk/Source/WTF

Clean up wtf/Compiler.h
​https://bugs.webkit.org/show_bug.cgi?id=120790

Reviewed by Andreas Kling.

  • Get rid of the CLANG_PRAGMA define, it's not used anywhere.
  • Remove a workaround for a bug in the version of clang that came with Xcode 4.2.
  • Replace WTF_COMPILER_SUPPORTS_CXX_FINAL_CONTROL with a compiler quirk for versions of clang that have a buggy final implementation and fix a bug in the macro where final would be disabled for versions of clang where clang_minor is less than 2, regardless of the major version.
  • Fail if someone tries to compile the WebKit stack with a compiler that doesn't support rvalue references or static_assert.
  • wtf/Compiler.h:
1:54 PM Changeset in webkit [155145] by hyatt@apple.com
  • 1 edit in trunk/Source/WebCore/rendering/mathml/RenderMathMLOperator.cpp

Fix build bustage.

1:17 PM Changeset in webkit [155144] by andersca@apple.com
  • 2 edits in trunk/Source/WebKit2

Replace uses of CLANG_PRAGMA with #pragma

Rubber-stamped by Andreas Kling.

This file is only compiled by clang anyway.

  • WebProcess/InjectedBundle/mac/InjectedBundleMac.mm:

(WebKit::InjectedBundle::load):

1:12 PM Changeset in webkit [155143] by andersca@apple.com
  • 49 edits in trunk/Source

GCAssertions.h should use STL type traits and static_assert
​https://bugs.webkit.org/show_bug.cgi?id=120785

Reviewed by Andreas Kling.

Source/JavaScriptCore:

There's no need to rely on compiler specific support to figure out if a class is trivially destructable,
we can just use type traits from STL. Do this, fix the assert macro to use static_assert directly and
rename it from ASSERT_HAS_TRIVIAL_DESTRUCTOR to STATIC_ASSERT_IS_TRIVIALLY_DESTRUCTIBLE to clarify that
it's a static assert and to match the STL nomenclature.

  • API/JSCallbackFunction.cpp:
  • debugger/DebuggerActivation.cpp:
  • heap/GCAssertions.h:
  • runtime/ArrayConstructor.cpp:
  • runtime/BooleanConstructor.cpp:
  • runtime/BooleanObject.cpp:
  • runtime/BooleanPrototype.cpp:
  • runtime/DateConstructor.cpp:
  • runtime/ErrorConstructor.cpp:
  • runtime/ErrorInstance.cpp:
  • runtime/ErrorPrototype.cpp:
  • runtime/ExceptionHelpers.cpp:
  • runtime/FunctionConstructor.cpp:
  • runtime/FunctionPrototype.cpp:
  • runtime/GetterSetter.cpp:
  • runtime/InternalFunction.cpp:
  • runtime/JSAPIValueWrapper.cpp:
  • runtime/JSArray.cpp:
  • runtime/JSCell.cpp:
  • runtime/JSNotAnObject.cpp:
  • runtime/JSONObject.cpp:
  • runtime/JSObject.cpp:
  • runtime/JSPromiseConstructor.cpp:
  • runtime/JSPromisePrototype.cpp:
  • runtime/JSPromiseResolverConstructor.cpp:
  • runtime/JSPromiseResolverPrototype.cpp:
  • runtime/JSProxy.cpp:
  • runtime/JSScope.cpp:
  • runtime/JSWrapperObject.cpp:
  • runtime/MathObject.cpp:
  • runtime/NameConstructor.cpp:
  • runtime/NativeErrorConstructor.cpp:
  • runtime/NumberConstructor.cpp:
  • runtime/NumberObject.cpp:
  • runtime/NumberPrototype.cpp:
  • runtime/ObjectConstructor.cpp:
  • runtime/ObjectPrototype.cpp:
  • runtime/RegExpObject.cpp:
  • runtime/StrictEvalActivation.cpp:
  • runtime/StringConstructor.cpp:
  • runtime/StringObject.cpp:
  • runtime/StringPrototype.cpp:

Source/WebCore:

Update for JavaScriptCore changes.

  • bindings/js/JSDOMBinding.cpp:
  • bindings/js/JSImageConstructor.cpp:

Source/WebKit2:

Update for JavaScriptCore changes.

  • WebProcess/Plugins/Netscape/JSNPMethod.cpp:
12:57 PM Changeset in webkit [155142] by hyatt@apple.com
  • 3 edits in trunk/Source/WebCore

Make RenderListItem inherit from RenderBlockFlow.
​https://bugs.webkit.org/show_bug.cgi?id=120789

Reviewed by Beth Dakin.

  • rendering/RenderListItem.cpp:

(WebCore::RenderListItem::RenderListItem):
(WebCore::RenderListItem::willBeDestroyed):
(WebCore::RenderListItem::willBeRemovedFromTree):
(WebCore::RenderListItem::layout):
(WebCore::RenderListItem::addOverflowFromChildren):
(WebCore::RenderListItem::computePreferredLogicalWidths):
(WebCore::RenderListItem::paint):

  • rendering/RenderListItem.h:
12:47 PM Changeset in webkit [155141] by hyatt@apple.com
  • 6 edits in trunk/Source/WebCore

Make RenderBlock have a protected constructor..
​https://bugs.webkit.org/show_bug.cgi?id=120787

Reviewed by Beth Dakin.

Make sure nobody can create a RenderBlock directly. Make sure RenderBlockFlows
get made when a display:block/inline-block/compact/run-in is seen.

  • html/HTMLDetailsElement.cpp:

(WebCore::HTMLDetailsElement::createRenderer):

  • html/HTMLSummaryElement.cpp:

(WebCore::HTMLSummaryElement::createRenderer):

  • rendering/RenderBlock.cpp:

(WebCore::RenderBlock::createAnonymous):

  • rendering/RenderBlock.h:
  • rendering/RenderObject.cpp:

(WebCore::RenderObject::createObject):

12:38 PM Changeset in webkit [155140] by Simon Fraser
  • 5 edits in trunk/Tools

Expected result images are sometimes blank in WKTR
​https://bugs.webkit.org/show_bug.cgi?id=120715

Reviewed by Tim Horton.

In WebKitTestRunner, snapshots obtained via windowSnapshotImage() were
sometimes blank if a previous test triggered compositing mode, and the
current test or reference did not require compositing. This happened
because the UI process didn't wait for the web process to complete
its compositing mode switch before snapshotting. Fix by calling
WKPageForceRepaint() before we take the snapshot; this is async,
so we have to spin the runloop for a while.

Remove the Qt/EFL code that does the same thing.

  • WebKitTestRunner/TestInvocation.cpp:

(WTR::TestInvocation::forceRepaintDoneCallback):
(WTR::TestInvocation::dumpResults):

  • WebKitTestRunner/TestInvocation.h:
  • WebKitTestRunner/cairo/TestInvocationCairo.cpp:

(WTR::TestInvocation::dumpPixelsAndCompareWithExpected):

  • WebKitTestRunner/qt/TestInvocationQt.cpp:

(WTR::TestInvocation::dumpPixelsAndCompareWithExpected):

12:32 PM Changeset in webkit [155139] by hyatt@apple.com
  • 68 edits
    2 adds in trunk/Source/WebCore

Add new RenderBlockFlow class
​https://bugs.webkit.org/show_bug.cgi?id=120777

Reviewed by Simon Fraser.

This patch adds a new RenderBlockFlow class. The idea is to get all of the line/float/margin
code out of RenderBlock and into the subclass, and then we can make RenderBlock an abstract
base. This will save memory for all of the classes that derive from RenderBlock but have
their own layout systems (e.g., flexbox, grid, table), and it will let us tighten up
types to make it more clear when we're talking about block layout code vs. a more generic
block-level object that can have any kind of layout.

The only real change here is that isBlockFlow() has been renamed to isRenderBlockFlow() and
is now checking if the object is a RenderBlockFlow. All of the block subclasses have been patched
to be a RenderBlock or RenderBlockFlow.

  • CMakeLists.txt:
  • GNUmakefile.list.am:
  • Target.pri:
  • WebCore.vcxproj/WebCore.vcxproj:
  • WebCore.xcodeproj/project.pbxproj:
  • accessibility/AccessibilityRenderObject.cpp:

(WebCore::AccessibilityRenderObject::computeAccessibilityIsIgnored):
(WebCore::AccessibilityRenderObject::determineAccessibilityRole):

  • dom/Position.cpp:

(WebCore::Position::isCandidate):
(WebCore::Position::getInlineBoxAndOffset):
(WebCore::Position::primaryDirection):

  • dom/PositionIterator.cpp:

(WebCore::PositionIterator::isCandidate):

  • editing/ApplyStyleCommand.cpp:

(WebCore::ApplyStyleCommand::applyInlineStyleToPushDown):

  • editing/CompositeEditCommand.cpp:

(WebCore::CompositeEditCommand::addBlockPlaceholderIfNeeded):

  • editing/Editor.cpp:

(WebCore::Editor::baseWritingDirectionForSelectionStart):

  • editing/FrameSelection.cpp:

(WebCore::caretRenderer):

  • editing/TextIterator.cpp:

(WebCore::TextIterator::shouldRepresentNodeOffsetZero):

  • editing/TypingCommand.cpp:

(WebCore::TypingCommand::makeEditableRootEmpty):

  • editing/htmlediting.cpp:

(WebCore::isBlockFlowElement):

  • html/shadow/SliderThumbElement.cpp:

(WebCore::RenderSliderThumb::RenderSliderThumb):

  • html/shadow/SliderThumbElement.h:
  • rendering/LayoutState.cpp:

(WebCore::LayoutState::LayoutState):

  • rendering/RenderBlock.cpp:

(WebCore::RenderBlock::layoutBlock):
(WebCore::RenderBlock::collapseMargins):
(WebCore::RenderBlock::selectionGaps):
(WebCore::RenderBlock::firstLineBoxBaseline):
(WebCore::RenderBlock::lastLineBoxBaseline):
(WebCore::RenderBlock::firstLineBlock):
(WebCore::findFirstLetterBlock):
(WebCore::shouldCheckLines):
(WebCore::RenderBlock::adjustForBorderFit):

  • rendering/RenderBlock.h:
  • rendering/RenderBlockFlow.cpp: Added.

(WebCore::RenderBlockFlow::RenderBlockFlow):
(WebCore::RenderBlockFlow::~RenderBlockFlow):

  • rendering/RenderBlockFlow.h: Added.

(WebCore::toRenderBlockFlow):

  • rendering/RenderBox.cpp:

(WebCore::RenderBox::computeRectForRepaint):
(WebCore::RenderBox::positionForPoint):

  • rendering/RenderBoxModelObject.h:

(WebCore::RenderBoxModelObject::canHaveBoxInfoInRegion):

  • rendering/RenderDeprecatedFlexibleBox.cpp:

(WebCore::RenderDeprecatedFlexibleBox::applyLineClamp):
(WebCore::RenderDeprecatedFlexibleBox::clearLineClamp):

  • rendering/RenderDetailsMarker.cpp:

(WebCore::RenderDetailsMarker::RenderDetailsMarker):

  • rendering/RenderDetailsMarker.h:
  • rendering/RenderFieldset.cpp:

(WebCore::RenderFieldset::RenderFieldset):

  • rendering/RenderFieldset.h:
  • rendering/RenderFileUploadControl.cpp:

(WebCore::RenderFileUploadControl::RenderFileUploadControl):

  • rendering/RenderFileUploadControl.h:
  • rendering/RenderFlowThread.cpp:

(WebCore::RenderFlowThread::RenderFlowThread):

  • rendering/RenderFlowThread.h:
  • rendering/RenderFullScreen.cpp:
  • rendering/RenderInline.cpp:

(WebCore::RenderInline::computeRectForRepaint):

  • rendering/RenderLineBoxList.cpp:

(WebCore::RenderLineBoxList::dirtyLinesFromChangedChild):

  • rendering/RenderMeter.cpp:

(WebCore::RenderMeter::RenderMeter):

  • rendering/RenderMeter.h:
  • rendering/RenderMultiColumnBlock.cpp:

(WebCore::RenderMultiColumnBlock::RenderMultiColumnBlock):

  • rendering/RenderMultiColumnBlock.h:
  • rendering/RenderObject.cpp:

(WebCore::RenderObject::computeRectForRepaint):
(WebCore::RenderObject::styleWillChange):
(WebCore::firstLineStyleForCachedUncachedType):

  • rendering/RenderObject.h:

(WebCore::RenderObject::isBR):
(WebCore::RenderObject::isRenderBlockFlow):

  • rendering/RenderObjectChildList.cpp:

(WebCore::RenderObjectChildList::insertChildNode):

  • rendering/RenderProgress.cpp:

(WebCore::RenderProgress::RenderProgress):

  • rendering/RenderProgress.h:
  • rendering/RenderRegion.cpp:

(WebCore::RenderRegion::RenderRegion):

  • rendering/RenderRegion.h:
  • rendering/RenderRuby.cpp:

(WebCore::RenderRubyAsBlock::RenderRubyAsBlock):

  • rendering/RenderRuby.h:
  • rendering/RenderRubyBase.cpp:

(WebCore::RenderRubyBase::RenderRubyBase):

  • rendering/RenderRubyBase.h:
  • rendering/RenderRubyRun.cpp:

(WebCore::RenderRubyRun::RenderRubyRun):

  • rendering/RenderRubyRun.h:
  • rendering/RenderRubyText.cpp:

(WebCore::RenderRubyText::RenderRubyText):

  • rendering/RenderRubyText.h:
  • rendering/RenderTableCaption.cpp:

(WebCore::RenderTableCaption::RenderTableCaption):

  • rendering/RenderTableCaption.h:
  • rendering/RenderTableCell.cpp:

(WebCore::RenderTableCell::RenderTableCell):

  • rendering/RenderTableCell.h:
  • rendering/RenderTextControl.cpp:

(WebCore::RenderTextControl::RenderTextControl):

  • rendering/RenderTextControl.h:
  • rendering/RenderTextControlSingleLine.h:

(WebCore::RenderTextControlInnerBlock::RenderTextControlInnerBlock):

  • rendering/RenderTextTrackCue.cpp:

(WebCore::RenderTextTrackCue::RenderTextTrackCue):

  • rendering/RenderTextTrackCue.h:
  • rendering/RenderView.cpp:

(WebCore::RenderView::RenderView):

  • rendering/RenderView.h:
  • rendering/svg/RenderSVGBlock.cpp:

(WebCore::RenderSVGBlock::RenderSVGBlock):

  • rendering/svg/RenderSVGBlock.h:
12:25 PM Changeset in webkit [155138] by timothy_horton@apple.com
  • 2 edits in trunk/Source/WebCore

iOS build broke with change to make Mac use PDFKit for PDFDocumentImage
​https://bugs.webkit.org/show_bug.cgi?id=120771

Reviewed by Anders Carlsson.

  • platform/graphics/cg/PDFDocumentImage.cpp:

(WebCore::PDFDocumentImage::createPDFDocument):
Use the (removed in r155069) PLATFORM(MAC) way of making a CGDataProvider from a SharedBuffer
everywhere; this is what iOS was expecting to use (where I accidentally changed behavior),
and it should also work for Windows, the only other USE(CG) platform.

Since we only createPDFDocument() after all data has been received, I'm not
restoring the comment about the SharedBuffer being secretly written to behind its back.

11:58 AM Changeset in webkit [155137] by dino@apple.com
  • 8 edits in trunk

Rename supportsContext to probablySupportsContext
​https://bugs.webkit.org/show_bug.cgi?id=120716
<rdar://problem/14914499>

Reviewed by Antoine Quint.

As decided in: ​http://lists.whatwg.org/htdig.cgi/whatwg-whatwg.org/2013-September/040680.html

Source/WebCore:

  • bindings/js/JSHTMLCanvasElementCustom.cpp:

(WebCore::JSHTMLCanvasElement::probablySupportsContext):

  • html/HTMLCanvasElement.cpp:

(WebCore::HTMLCanvasElement::probablySupportsContext):

  • html/HTMLCanvasElement.h:
  • html/HTMLCanvasElement.idl:

LayoutTests:

  • fast/canvas/webgl/canvas-supports-context-expected.txt:
  • fast/canvas/webgl/canvas-supports-context.html:
11:29 AM Changeset in webkit [155136] by Joseph Pecoraro
  • 2 edits in trunk/Source/WebCore

Fix build warning about unused variable.

Unreviewed build fix.

  • bindings/js/ScriptDebugServer.cpp:

(WebCore::ScriptDebugServer::evaluateBreakpointAction):

11:23 AM Changeset in webkit [155135] by andersca@apple.com
  • 4 edits in trunk/Source

Enable C++11 wherever we build C++ files that include wtf/Platform.h
​https://bugs.webkit.org/show_bug.cgi?id=120782

Reviewed by Andreas Kling.

Source/ThirdParty:

Set CLANG_CXX_LANGUAGE_STANDARD to gnu++0x.

  • gtest/xcode/Config/General.xcconfig:

Source/WebCore:

Pass -std=gnu++11 when preprocessing wtf/Platform.h.

  • DerivedSources.make:
11:01 AM Changeset in webkit [155134] by Joseph Pecoraro
  • 7 edits
    2 adds in trunk/Source/WebInspectorUI

Web Inspector: Breakpoint Actions
​https://bugs.webkit.org/show_bug.cgi?id=120576

Reviewed by Timothy Hatcher.

Give the CodeMirror editor for JavaScript Breakpoint Actions JS Runtime
completions (like window, document, etc.).

The logic was inside of JavaScriptLogViewController but was already
entirely independent. Factor it out into its own class and plug it into
CodeMirrorCompletionController as a "CompletionsProvider".

Because the class hooks into a global event to reset some state, make
it a singleton, so new instances are not leaked.

  • UserInterface/Main.html:
  • UserInterface/Main.js:

(WebInspector.loaded):

  • UserInterface/RuntimeManager.js: Added.

(WebInspector.RuntimeManager):
(WebInspector.RuntimeManager.prototype.evalCallback):
(WebInspector.RuntimeManager.prototype.evaluateInInspectedWindow):
Move eval in window to new RuntimeManager. A slightly cleaner place to this
since it is used both by the console and JS completions provider.

  • UserInterface/CodeMirrorCompletionController.js:

(WebInspector.CodeMirrorCompletionController):
(WebInspector.CodeMirrorCompletionController.prototype.setExtendedCompletionProvider):
(WebInspector.CodeMirrorCompletionController.prototype._completeAtCurrentPosition):
Allow a completions provider plugin to completions controller. If one
is available, this supercedes the delegate extend completions path.

  • UserInterface/BreakpointActionView.js:
  • UserInterface/ConsolePrompt.js:

(WebInspector.ConsolePrompt):
Use the new JS completions provider to get runtime JS completions.
NOTE: ConsolePrompt still needs the delegate path for SQL completions.

  • UserInterface/JavaScriptLogViewController.js:

(WebInspector.JavaScriptLogViewController):
(WebInspector.JavaScriptLogViewController.prototype.consolePromptTextCommitted):

  • UserInterface/JavaScriptRuntimeCompletionProvider.js: Added.

(WebInspector.JavaScriptRuntimeCompletionProvider):
(get WebInspector.JavaScriptRuntimeCompletionProvider.prototype.):
(get WebInspector.JavaScriptRuntimeCompletionProvider.prototype.completionControllerCompletionsNeeded):
Move the JS completions code to a new class. No changes needed.

11:00 AM Changeset in webkit [155133] by Joseph Pecoraro
  • 8 edits
    5 adds in trunk/Source/WebInspectorUI

Web Inspector: Breakpoint Actions
​https://bugs.webkit.org/show_bug.cgi?id=120576

Reviewed by Timothy Hatcher.

  • UserInterface/Breakpoint.js:

(WebInspector.Breakpoint):
(WebInspector.Breakpoint.prototype.get actions):
(WebInspector.Breakpoint.prototype.get options):
(WebInspector.Breakpoint.prototype.get info):
(WebInspector.Breakpoint.prototype._serializableActions):
Saving, restoring, and serialization.

(WebInspector.Breakpoint.prototype.createAction):
(WebInspector.Breakpoint.prototype.recreateAction):
(WebInspector.Breakpoint.prototype.removeAction):
(WebInspector.Breakpoint.prototype.breakpointActionDidChange):
Modifying _actions ivar.

(WebInspector.Breakpoint.prototype._editBreakpointPopoverContentElement):
(WebInspector.Breakpoint.prototype._popoverActionsCreateAddActionButton):
(WebInspector.Breakpoint.prototype._popoverActionsAddActionButtonClicked):
(WebInspector.Breakpoint.prototype._popoverActionsInsertBreakpointActionView):
(WebInspector.Breakpoint.prototype.breakpointActionViewAppendActionView):
(WebInspector.Breakpoint.prototype.breakpointActionViewRemoveActionView):
(WebInspector.Breakpoint.prototype.breakpointActionViewResized):
Popover UI containing breakpoint action views and converting to and from
having no actions.

(WebInspector.Breakpoint.prototype.willDismissPopover):
Cleanup some ivars when the popover dismisses.

  • UserInterface/Breakpoint.css:

(.popover .edit-breakpoint-popover-content.wide):
Wider setting when there are actions in the popover.

(.popover .edit-breakpoint-popover-content > table > tr > th):
Keep the title labels at the top of their section.

  • UserInterface/BreakpointAction.js:

(WebInspector.BreakpointAction):
(WebInspector.BreakpointAction.prototype.get breakpoint):
(WebInspector.BreakpointAction.prototype.get type):
(WebInspector.BreakpointAction.prototype.get data):
(WebInspector.BreakpointAction.prototype.set data):
(WebInspector.BreakpointAction.prototype.get info):
Very basic BreakpointAction class. Notifies Breakpoint on data changes.
Type changes are not allowed, instead just create a new BreakpointAction.

  • UserInterface/BreakpointActionView.css: Added.

(.breakpoint-action-block-header):
(.breakpoint-action-remove-button):
(.breakpoint-action-remove-button:active):
(.breakpoint-action-append-button):
(.breakpoint-action-block-body):
(.breakpoint-action-block-body > input):
(.breakpoint-action-eval-editor):
(.breakpoint-action-eval-editor > .CodeMirror):
(.breakpoint-action-eval-editor > .CodeMirror-scroll):

  • UserInterface/BreakpointActionView.js: Added.

(WebInspector.BreakpointActionView):
(WebInspector.BreakpointActionView.displayStringForType):
(WebInspector.BreakpointActionView.prototype.get action):
(WebInspector.BreakpointActionView.prototype.get element):
(WebInspector.BreakpointActionView.prototype._pickerChanged):
(WebInspector.BreakpointActionView.prototype._appendActionButtonClicked):
(WebInspector.BreakpointActionView.prototype._removeActionButtonClicked):
(WebInspector.BreakpointActionView.prototype._updateBody.switch.break):
(WebInspector.BreakpointActionView.prototype._logInputChanged):
(WebInspector.BreakpointActionView.prototype._codeMirrorBlurred):
(WebInspector.BreakpointActionView.prototype._codeMirrorViewportChanged):
UI and form controls for individual breakpoint actions. The view
wraps a single BreakpointAction, and for any actions that could affect
other views the delegate is notified.

  • UserInterface/DebuggerManager.js:

(WebInspector.DebuggerManager):
(WebInspector.DebuggerManager.prototype._debuggerBreakpointActionType):
When sending to the backend, make sure to send protocol enums.

  • UserInterface/Popover.js:

(WebInspector.Popover.prototype.update):
Update the popover when content changes.

  • UserInterface/Images/BreakpointActionAdd.svg: Added.
  • UserInterface/Images/BreakpointActionRemove.svg: Added.
  • UserInterface/InspectorBackendCommands.js:
  • UserInterface/Main.html:
  • Localizations/en.lproj/localizedStrings.js:

Misc. New strings, images, includes.

11:00 AM Changeset in webkit [155132] by Joseph Pecoraro
  • 8 edits
    6 adds in trunk

Web Inspector: Breakpoint Actions
​https://bugs.webkit.org/show_bug.cgi?id=120576

Reviewed by Timothy Hatcher.

Source/WebCore:

Tests: inspector-protocol/debugger/setBreakpoint-actions.html

inspector-protocol/debugger/setBreakpoint-options-exception.html

  • inspector/ScriptBreakpoint.h:

(WebCore::ScriptBreakpointAction::ScriptBreakpointAction):
(WebCore::ScriptBreakpoint::ScriptBreakpoint):
Struct for a breakpoint action. Type and string of data.

  • bindings/js/ScriptDebugServer.h:
  • bindings/js/ScriptDebugServer.cpp:

(WebCore::ScriptDebugServer::hasBreakpoint):
(WebCore::ScriptDebugServer::evaluateBreakpointAction):
(WebCore::ScriptDebugServer::evaluateBreakpointActions):
(WebCore::ScriptDebugServer::updateCallFrameAndPauseIfNeeded):
(WebCore::ScriptDebugServer::pauseIfNeeded):
When a breakpoint is triggered evaluate each of its breakpoint
actions in order. If there are exceptions, log them.

  • inspector/Inspector.json:
  • inspector/InspectorDebuggerAgent.cpp:

(WebCore::buildObjectForBreakpointCookie):
(WebCore::InspectorDebuggerAgent::didParseSource):
Save and restore breakpoint actions where needed.

(WebCore::breakpointActionTypeForString):
(WebCore::breakpointActionsFromProtocol):
(WebCore::InspectorDebuggerAgent::setBreakpointByUrl):
(WebCore::InspectorDebuggerAgent::setBreakpoint):
Parse optional breakpoint actions from the protocol into
ScriptBreakpointAction objects.

LayoutTests:

  • inspector-protocol/debugger/resources/breakpoint.js:
  • inspector-protocol/debugger/setBreakpoint-actions-expected.txt: Added.
  • inspector-protocol/debugger/setBreakpoint-actions.html: Added.
  • inspector-protocol/debugger/setBreakpoint-options-exception-expected.txt: Added.
  • inspector-protocol/debugger/setBreakpoint-options-exception.html: Added.
  • inspector-protocol/resources/console-helper.js: Added.

(ConsoleHelper.simplifiedConsoleMessage):

10:56 AM Changeset in webkit [155131] by Simon Fraser
  • 4 edits
    2 adds in trunk

Source/WebCore: Fix repaint issue on "paints into ancestor" filtered layers
​https://bugs.webkit.org/show_bug.cgi?id=120780

Reviewed by Tim Horton.

When a repaint happened on a layer with a filter, and which paints into
its compositing ancestor, we'd repaint the wrong layer (and assert).

Fix by ensuring that RenderLayer::enclosingFilterLayer() takes paintsIntoCompositedAncestor()
into account, by adding a function that we share between three callers who
check isComposited() && !paintsIntoCompositedAncestor(). I didn't use a function
on RenderLayer, because I wanted it to be inline but to not #include RenderLayerBacking
in RenderLayer.h.

Test: compositing/filters/opacity-change-on-filtered-paints-into-ancestor.html

  • dom/Node.cpp: Drive-by removal of #include "RenderLayer.h"
  • rendering/RenderLayer.cpp:

(WebCore::compositedWithOwnBackingStore):
(WebCore::RenderLayer::enclosingCompositingLayerForRepaint):
(WebCore::RenderLayer::enclosingFilterRepaintLayer):
(WebCore::RenderLayer::clippingRootForPainting):

LayoutTests: Fix repaint issue on "paints into ancestor" filtered layers
​https://bugs.webkit.org/show_bug.cgi?id=120780
<rdar://problem/14884148>

Reviewed by Tim Horton.

Ref test for opacity change on a filtered layer which paints into its compositing ancestor.

  • compositing/filters/opacity-change-on-filtered-paints-into-ancestor-expected.html: Added.
  • compositing/filters/opacity-change-on-filtered-paints-into-ancestor.html: Added.
10:54 AM Changeset in webkit [155130] by Brent Fulgham
  • 5 edits in trunk/Source/WebCore

[Windows] Unreviewed build/link improvement after r155127 to be
consistent about using debug-build support libraries.

Now that <rdar://problem/9898937> is resolved, we should be linking
to debug AVFoundationCF as well.

  • page/CaptionUserPreferencesMediaAF.cpp: Link to AVFoundationCF_debug

when building with the DEBUG_ALL target.

  • platform/graphics/avfoundation/InbandTextTrackPrivateAVF.cpp: Ditto.
  • platform/graphics/avfoundation/cf/AVFoundationCFSoftLinking.h: Ditto.
10:35 AM Changeset in webkit [155129] by Lucas Forschler
  • 3 edits
    2 copies in branches/safari-537-branch

Merged r154037. <rdar://problem/14705502>

10:28 AM Changeset in webkit [155128] by commit-queue@webkit.org
  • 2 edits
    4 adds in trunk/LayoutTests

[GTK] improving tests expectations related to trackmenu
​https://bugs.webkit.org/show_bug.cgi?id=120699

Adding tests expectations to the current GTK's trackmenu
implementation.

Also adding a in-band-track test expectation as failure since we don't
support in-band tracks yet.

Patch by Danilo Cesar Lemes de Paula <​danilo.cesar@collabora.co.uk> on 2013-09-05
Reviewed by Gustavo Noronha Silva.

  • platform/gtk/TestExpectations:
  • platform/gtk/media/video-controls-captions-trackmenu-expected.txt: Added.
  • platform/gtk/media/video-controls-captions-trackmenu-hide-on-click-expected.txt: Added.
  • platform/gtk/media/video-controls-captions-trackmenu-localized-expected.txt: Added.
  • platform/gtk/media/video-controls-captions-trackmenu-sorted-expected.txt: Added.
10:21 AM Changeset in webkit [155127] by roger_fong@apple.com
  • 2 edits in trunk/Source/WebCore

Link to libdispatch_debug.lib when using DebugSuffix configuration.
​https://bugs.webkit.org/show_bug.cgi?id=120712.
<rdar://problem/14883216>.

Reviewed by Brent Fulgham.

  • platform/graphics/avfoundation/cf/MediaPlayerPrivateAVFoundationCF.cpp:
10:11 AM Changeset in webkit [155126] by andersca@apple.com
  • 2 edits in trunk/Source/WTF

Change StringBuffer back to not use an OwnPtr in an attempt to fix ports.

  • wtf/text/StringBuffer.h:
10:10 AM Changeset in webkit [155125] by berto@igalia.com
  • 3 edits in trunk/Source/WebKit2

[WK2] [GTK] Remove the test to disable AC under Wayland from WebKitWebViewGroup
​https://bugs.webkit.org/show_bug.cgi?id=120762

Reviewed by Gustavo Noronha Silva.

The code to disable AC under Wayland is not run if we're using
WebKitTestRunner, so it would be better to move it to
WebKitWebViewBase.

However it's not really necessary to do that, instead it's much
simpler to disable accelerated compositing in all cases unless we
know we can support it.

  • UIProcess/API/gtk/WebKitWebViewBase.cpp:

(webkitWebViewBaseUpdatePreferences):

  • UIProcess/API/gtk/WebKitWebViewGroup.cpp:

(webkitWebViewGroupAttachSettingsToPageGroup):

9:42 AM Changeset in webkit [155124] by Csaba Osztrogonác
  • 3 edits in trunk/Tools

Make build.webkit.org report the number of failing run-fast-jsc tests
​https://bugs.webkit.org/show_bug.cgi?id=120766

Reviewed by Filip Pizlo.

  • BuildSlaveSupport/build.webkit.org-config/master.cfg:

(RunJavaScriptCoreTests.commandComplete):
(RunJavaScriptCoreTests.getText2):

  • BuildSlaveSupport/build.webkit.org-config/mastercfg_unittest.py:

(RunJavaScriptCoreTestsTest): Added.
(RunJavaScriptCoreTestsTest.assertResults):
(RunJavaScriptCoreTestsTest.test_no_regressions_old_output):
(RunJavaScriptCoreTestsTest.test_mozilla_failure_old_output):
(RunJavaScriptCoreTestsTest.test_mozilla_failure_new_output):
(RunJavaScriptCoreTestsTest.test_fast_js_failure_new_output):
(RunJavaScriptCoreTestsTest.test_fast_js_crash_new_output):
(RunJavaScriptCoreTestsTest.test_mozilla_and_fast_js_failure_new_output):

9:28 AM Changeset in webkit [155123] by Brent Fulgham
  • 2 edits in trunk/Source/JavaScriptCore

[Windows] Unreviewed build fix for DebugSuffix target.

  • JavaScriptCore.vcxproj/JavaScriptCore.vcxproj.filters: Don't build 64-bit assembly in 32-bit build.

Also correct 'filters' file so that files appear in categories that match their on-disk locations.

8:54 AM Changeset in webkit [155122] by fpizlo@apple.com
  • 28 edits in trunk/LayoutTests

fast/js/dfg-* tests should wait for the concurrent JIT
​https://bugs.webkit.org/show_bug.cgi?id=120723

Rubber stamped by Oliver Hunt.

Convert more tests.

  • fast/js/dfg-branch-logical-not-peephole-around-osr-exit-expected.txt:
  • fast/js/dfg-branch-not-fail-expected.txt:
  • fast/js/dfg-byte-array-put-expected.txt:
  • fast/js/dfg-call-function-hit-watchpoint-expected.txt:
  • fast/js/dfg-call-method-hit-watchpoint-expected.txt:
  • fast/js/dfg-captured-var-get-local-expected.txt:
  • fast/js/dfg-cfa-merge-with-dead-use-at-tail-expected.txt:
  • fast/js/dfg-cfg-simplify-eliminate-set-local-type-check-then-branch-not-null-and-decrement-expected.txt:
  • fast/js/dfg-cfg-simplify-eliminate-set-local-type-check-then-branch-not-null-expected.txt:
  • fast/js/dfg-cfg-simplify-eliminate-set-local-type-check-then-typeof-expected.txt:
  • fast/js/dfg-cfg-simplify-phantom-get-local-on-same-block-set-local-expected.txt:
  • fast/js/dfg-cfg-simplify-redundant-dead-get-local-expected.txt:
  • fast/js/dfg-check-function-change-structure-expected.txt:
  • fast/js/jsc-test-list:
  • fast/js/script-tests/dfg-branch-logical-not-peephole-around-osr-exit.js:
  • fast/js/script-tests/dfg-branch-not-fail.js:
  • fast/js/script-tests/dfg-byte-array-put.js:
  • fast/js/script-tests/dfg-call-function-hit-watchpoint.js:
  • fast/js/script-tests/dfg-call-method-hit-watchpoint.js:
  • fast/js/script-tests/dfg-captured-var-get-local.js:
  • fast/js/script-tests/dfg-cfa-merge-with-dead-use-at-tail.js:
  • fast/js/script-tests/dfg-cfg-simplify-eliminate-set-local-type-check-then-branch-not-null-and-decrement.js:
  • fast/js/script-tests/dfg-cfg-simplify-eliminate-set-local-type-check-then-branch-not-null.js:
  • fast/js/script-tests/dfg-cfg-simplify-eliminate-set-local-type-check-then-typeof.js:
  • fast/js/script-tests/dfg-cfg-simplify-phantom-get-local-on-same-block-set-local.js:
  • fast/js/script-tests/dfg-cfg-simplify-redundant-dead-get-local.js:
  • fast/js/script-tests/dfg-check-function-change-structure.js:
8:20 AM Changeset in webkit [155121] by andersca@apple.com
  • 6 edits in trunk/Source

Vector::releaseBuffer should return an OwnPtr
​https://bugs.webkit.org/show_bug.cgi?id=120718

Reviewed by Andreas Kling.

Source/WebCore:

Change FormStreamFields::currentData to an OwnPtr.

  • platform/network/cf/FormDataStreamCFNet.cpp:

(WebCore::closeCurrentStream):
(WebCore::advanceCurrentStream):
(WebCore::formCreate):

Source/WTF:

Change Vector::releaseBuffer() to return an OwnPtr. I intentionally chose
to use an OwnPtr over a PassOwnPtr since we're trying to move away from PassOwnPtr objects.

Fix fallout from this change by adopting OwnPtr/PassOwnPtr in StringBuffer and the two StringImpl
constructors that adopt the passed in pointer.

  • wtf/Vector.h:
  • wtf/text/StringBuffer.h:
  • wtf/text/StringImpl.h:
8:17 AM Changeset in webkit [155120] by akling@apple.com
  • 7 edits in trunk/Source/WebCore

Reverting "Cached Page and Frame don't need to be ref-counted.
<​https://webkit.org/b/120758>

This didn't work so well on Qt bots. Will have to take bigger steps
here instead of iterating.

  • history/CachedFrame.h:

(WebCore::CachedFrame::create):

  • history/CachedPage.cpp:

(WebCore::CachedPage::create):

  • history/CachedPage.h:
  • history/HistoryItem.h:
  • loader/FrameLoader.cpp:

(WebCore::FrameLoader::commitProvisionalLoad):
(WebCore::FrameLoader::transitionToCommitted):

  • loader/FrameLoader.h:
8:15 AM Changeset in webkit [155119] by andersca@apple.com
  • 11 edits in trunk/Source/WebCore

Animations in an AnimationList are never null
​https://bugs.webkit.org/show_bug.cgi?id=120720

Reviewed by Andreas Kling.

Change AnimationList::operator[] to return an Animation& and fix up related code to also take references.

  • css/CSSComputedStyleDeclaration.cpp:
  • css/DeprecatedStyleBuilder.cpp:
  • page/animation/CompositeAnimation.cpp:
  • page/animation/KeyframeAnimation.cpp:
  • platform/animation/Animation.h:
  • platform/animation/AnimationList.cpp:
  • platform/animation/AnimationList.h:
  • rendering/RenderLayerBacking.cpp:
8:11 AM Changeset in webkit [155118] by Csaba Osztrogonác
  • 3 edits in trunk/Tools

Make run-fast-jsc script bash, dash and Darwin's /bin/sh friendly
​https://bugs.webkit.org/show_bug.cgi?id=120759

Reviewed by Filip Pizlo.

  • Scripts/run-fast-jsc: Remove bashisms to make bash, dash and Darwin's /bin/sh happy.
  • Scripts/run-javascriptcore-tests: Enable run-fast-jsc tests everywhere, except Windows.
8:01 AM Changeset in webkit [155117] by fpizlo@apple.com
  • 2 edits in trunk/Tools

Unreviewed, green the Mac bots by boosting the timeout on JSC tests. One of the
Mozilla tests appears to run for a longer time.

I think we want to continue to allow a few tests here and there to be
longer-running, rather than excluding them. I seem to recall that those tests
catch good bugs.

  • Scripts/run-javascriptcore-tests:
7:45 AM Changeset in webkit [155116] by Antti Koivisto
  • 2 edits in trunk/Source/WebCore

Call createTextRenderersForSiblingsAfterAttachIfNeeded only for the attach root
​https://bugs.webkit.org/show_bug.cgi?id=120770

Reviewed by Andreas Kling.

There is no need to call this during recursive attach as siblings are going to be attached normally anyway.
Move call sites to attach roots only.

  • style/StyleResolveTree.cpp:

(WebCore::Style::createTextRenderersForSiblingsAfterAttachIfNeeded):

Factor to take reference and do the inital tests itself.

(WebCore::Style::createTextRendererIfNeeded):
(WebCore::Style::updateTextRendererAfterContentChange):
(WebCore::Style::attachChildren):

Also tightened the condition where previously attached children may be encountered.

(WebCore::Style::attachRenderTree):
(WebCore::Style::resolveLocal):
(WebCore::Style::updateTextStyle):
(WebCore::Style::reattachRenderTree):

7:31 AM Changeset in webkit [155115] by mikhail.pozdnyakov@intel.com
  • 3 edits in trunk/Source/WTF

Remove String(RefPtr<StringImpl>) constructor
​https://bugs.webkit.org/show_bug.cgi?id=120767

Reviewed by Anders Carlsson.

The String(RefPtr<StringImpl>) constructor should be removed for the following reasons:

  • It almost does not have clients, so it is unneeded.
  • RefPtr should never be passed by value
  • wtf/text/StringBuilder.cpp:

(WTF::StringBuilder::shrinkToFit):

  • wtf/text/WTFString.h:
7:23 AM Changeset in webkit [155114] by commit-queue@webkit.org
  • 7 edits
    6 adds
    45 deletes in trunk/LayoutTests

[CSS Regions] Convert existing pixel tests in ref tests
​https://bugs.webkit.org/show_bug.cgi?id=120760

Patch by Mihai Maerean <Mihai Maerean> on 2013-09-05
Reviewed by Antti Koivisto.

I have also removed text-align:justify from the tests because its not relevant to the tested functionality and
it makes creating the expected results very difficult because of the way fragmented text is rendered when
it's aligned justified.

  • fast/regions/overflow-in-uniform-regions-dynamic-expected.html: Added.
  • fast/regions/overflow-in-uniform-regions-dynamic.html:
  • fast/regions/overflow-in-variable-width-regions-expected.html: Added.
  • fast/regions/overflow-in-variable-width-regions.html:
  • fast/regions/overflow-moving-below-floats-in-variable-width-regions-expected.html: Added.
  • fast/regions/overflow-moving-below-floats-in-variable-width-regions.html:
  • fast/regions/overflow-not-moving-below-floats-in-variable-width-regions-expected.html: Added.
  • fast/regions/overflow-not-moving-below-floats-in-variable-width-regions.html:
  • fast/regions/overflow-rtl-in-variable-width-regions-expected.html: Added.
  • fast/regions/overflow-rtl-in-variable-width-regions.html:
  • fast/regions/overflow-size-change-in-variable-width-regions-expected.html: Added.
  • fast/regions/overflow-size-change-in-variable-width-regions.html:
  • platform/efl/fast/regions/overflow-in-uniform-regions-dynamic-expected.png: Removed.
  • platform/efl/fast/regions/overflow-in-uniform-regions-dynamic-expected.txt: Removed.
  • platform/efl/fast/regions/overflow-in-variable-width-regions-expected.png: Removed.
  • platform/efl/fast/regions/overflow-in-variable-width-regions-expected.txt: Removed.
  • platform/efl/fast/regions/overflow-moving-below-floats-in-variable-width-regions-expected.png: Removed.
  • platform/efl/fast/regions/overflow-moving-below-floats-in-variable-width-regions-expected.txt: Removed.
  • platform/efl/fast/regions/overflow-rtl-in-variable-width-regions-expected.png: Removed.
  • platform/efl/fast/regions/overflow-rtl-in-variable-width-regions-expected.txt: Removed.
  • platform/efl/fast/regions/overflow-size-change-in-variable-width-regions-expected.png: Removed.
  • platform/efl/fast/regions/overflow-size-change-in-variable-width-regions-expected.txt: Removed.
  • platform/gtk/fast/regions/overflow-in-uniform-regions-dynamic-expected.png: Removed.
  • platform/gtk/fast/regions/overflow-in-uniform-regions-dynamic-expected.txt: Removed.
  • platform/gtk/fast/regions/overflow-in-variable-width-regions-expected.png: Removed.
  • platform/gtk/fast/regions/overflow-in-variable-width-regions-expected.txt: Removed.
  • platform/gtk/fast/regions/overflow-moving-below-floats-in-variable-width-regions-expected.png: Removed.
  • platform/gtk/fast/regions/overflow-moving-below-floats-in-variable-width-regions-expected.txt: Removed.
  • platform/gtk/fast/regions/overflow-rtl-in-variable-width-regions-expected.png: Removed.
  • platform/gtk/fast/regions/overflow-rtl-in-variable-width-regions-expected.txt: Removed.
  • platform/gtk/fast/regions/overflow-size-change-in-variable-width-regions-expected.png: Removed.
  • platform/gtk/fast/regions/overflow-size-change-in-variable-width-regions-expected.txt: Removed.
  • platform/mac/fast/regions/overflow-in-uniform-regions-dynamic-expected.png: Removed.
  • platform/mac/fast/regions/overflow-in-uniform-regions-dynamic-expected.txt: Removed.
  • platform/mac/fast/regions/overflow-in-variable-width-regions-expected.png: Removed.
  • platform/mac/fast/regions/overflow-in-variable-width-regions-expected.txt: Removed.
  • platform/mac/fast/regions/overflow-moving-below-floats-in-variable-width-regions-expected.png: Removed.
  • platform/mac/fast/regions/overflow-moving-below-floats-in-variable-width-regions-expected.txt: Removed.
  • platform/mac/fast/regions/overflow-rtl-in-variable-width-regions-expected.png: Removed.
  • platform/mac/fast/regions/overflow-rtl-in-variable-width-regions-expected.txt: Removed.
  • platform/mac/fast/regions/overflow-size-change-in-variable-width-regions-expected.png: Removed.
  • platform/mac/fast/regions/overflow-size-change-in-variable-width-regions-expected.txt: Removed.
  • platform/qt-wk2/fast/regions/overflow-in-uniform-regions-dynamic-expected.png: Removed.
  • platform/qt-wk2/fast/regions/overflow-in-variable-width-regions-expected.png: Removed.
  • platform/qt-wk2/fast/regions/overflow-moving-below-floats-in-variable-width-regions-expected.png: Removed.
  • platform/qt-wk2/fast/regions/overflow-rtl-in-variable-width-regions-expected.png: Removed.
  • platform/qt-wk2/fast/regions/overflow-size-change-in-variable-width-regions-expected.png: Removed.
  • platform/qt/fast/regions/overflow-in-uniform-regions-dynamic-expected.png: Removed.
  • platform/qt/fast/regions/overflow-in-uniform-regions-dynamic-expected.txt: Removed.
  • platform/qt/fast/regions/overflow-in-variable-width-regions-expected.png: Removed.
  • platform/qt/fast/regions/overflow-in-variable-width-regions-expected.txt: Removed.
  • platform/qt/fast/regions/overflow-moving-below-floats-in-variable-width-regions-expected.png: Removed.
  • platform/qt/fast/regions/overflow-moving-below-floats-in-variable-width-regions-expected.txt: Removed.
  • platform/qt/fast/regions/overflow-rtl-in-variable-width-regions-expected.png: Removed.
  • platform/qt/fast/regions/overflow-rtl-in-variable-width-regions-expected.txt: Removed.
  • platform/qt/fast/regions/overflow-size-change-in-variable-width-regions-expected.png: Removed.
  • platform/qt/fast/regions/overflow-size-change-in-variable-width-regions-expected.txt: Removed.
7:15 AM Changeset in webkit [155113] by commit-queue@webkit.org
  • 2 edits in trunk/Source/WebCore

[GTK] Assertion fails when the mouse pointer is styled with a custom cursor
​https://bugs.webkit.org/show_bug.cgi?id=118905

Patch by Enrique Ocaña González <​eocanha@igalia.com> on 2013-09-05
Reviewed by Philippe Normand.

Checked that the custom cursor image and its pixbuf is valid and create a null
cursor otherwise.

  • platform/gtk/CursorGtk.cpp:

(WebCore::createCustomCursor):

7:08 AM WebKitGTK/2.2.x edited by andrunko@gmail.com
(diff)
7:08 AM WebKitGTK/2.0.x edited by andrunko@gmail.com
(diff)
6:25 AM Changeset in webkit [155112] by commit-queue@webkit.org
  • 6 edits in trunk/Source/WebCore

Precision updates in WebAudio.
​https://bugs.webkit.org/show_bug.cgi?id=119739

Patch by Praveen R Jadhav <​praveen.j@samsung.com> on 2013-09-05
Reviewed by Philippe Normand.

WebAudio Specification suggests to use 'double' datatypes
for minDecibels, maxDecibels and smoothingTimeConstant in
AnalyserNode.idl and currentTime in AudioContext.idl.
Current WebKit implementation has declared these attributes
as 'float' type.

Spec: ​https://dvcs.w3.org/hg/audio/raw-file/tip/webaudio/specification.html#AudioContext-section

​https://dvcs.w3.org/hg/audio/raw-file/tip/webaudio/specification.html#AnalyserNode

No new tests, covered by existing tests.

  • Modules/webaudio/AnalyserNode.cpp:

(WebCore::AnalyserNode::setMinDecibels):
(WebCore::AnalyserNode::setMaxDecibels):
(WebCore::AnalyserNode::setSmoothingTimeConstant):

  • Modules/webaudio/AnalyserNode.h:

(WebCore::AnalyserNode::minDecibels):
(WebCore::AnalyserNode::maxDecibels):
(WebCore::AnalyserNode::smoothingTimeConstant):

  • Modules/webaudio/AnalyserNode.idl:
  • Modules/webaudio/AudioContext.idl:
  • Modules/webaudio/RealtimeAnalyser.h:

(WebCore::RealtimeAnalyser::setMinDecibels):
(WebCore::RealtimeAnalyser::minDecibels):
(WebCore::RealtimeAnalyser::setMaxDecibels):
(WebCore::RealtimeAnalyser::maxDecibels):
(WebCore::RealtimeAnalyser::setSmoothingTimeConstant):
(WebCore::RealtimeAnalyser::smoothingTimeConstant):

6:22 AM Changeset in webkit [155111] by commit-queue@webkit.org
  • 2 edits in trunk/LayoutTests

[GTK] media/track/in-band/ layout tests introduced in r154908 are failing
​https://bugs.webkit.org/show_bug.cgi?id=120665

Patch by Brendan Long <​b.long@cablelabs.com> on 2013-09-05
Reviewed by Philippe Normand.

  • platform/efl/TestExpectations: Skip new media/track/in-band tests.
5:34 AM Changeset in webkit [155110] by zarvai@inf.u-szeged.hu
  • 3 edits in trunk/LayoutTests

[Qt] Unreviewed gardening.

  • platform/qt-wk1/TestExpectations: Unskipping after 155104.
  • platform/qt/TestExpectations: Unskipping after r155061.
3:33 AM Changeset in webkit [155109] by mihnea@adobe.com
  • 6 edits in trunk/Source/WebCore

Replace node() calls with generatingNode() for RenderRegion code
​https://bugs.webkit.org/show_bug.cgi?id=120397

Reviewed by Darin Adler.

As regions will become anonymous blocks nested inside elements with the flow-from property, a first step
would be to change the calls to node() with calls to generatingNode(). In the case of regions, generatingNode()
will always return an Element* and not a Node*. Therefore, to get rid of toElement() casts, i created
a specific method RenderRegion::generatingElement that will return the generating node as Element.
No new tests as no functional change was introduced.

Original patch by Catalin Badea.

  • css/ElementRuleCollector.cpp:

(WebCore::ElementRuleCollector::collectMatchingRulesForRegion):

  • dom/WebKitNamedFlow.cpp:

(WebCore::WebKitNamedFlow::getRegionsByContent):
(WebCore::WebKitNamedFlow::getRegions):

  • rendering/RenderRegion.cpp:

(WebCore::RenderRegion::regionOversetState):
(WebCore::RenderRegion::setRegionOversetState):
(WebCore::RenderRegion::checkRegionStyle):

  • rendering/RenderRegion.h:

(WebCore::RenderRegion::generatingElement):

  • rendering/RenderTreeAsText.cpp:

(WebCore::writeRenderRegionList): For pseudo-elements as regions, i want to print
{pseudo} instead of {div} marker for the parent.

2:51 AM Changeset in webkit [155108] by Csaba Osztrogonác
  • 7 edits
    1 copy
    1 add in trunk/Source/WebCore

Fix SynchronousLoaderClient class for !USE(CFNETWORK) platforms
​https://bugs.webkit.org/show_bug.cgi?id=120532

Reviewed by Darin Adler.

  • CMakeLists.txt:
  • PlatformEfl.cmake:
  • PlatformGTK.cmake:
  • WebCore.vcxproj/WebCore.vcxproj:
  • WebCore.vcxproj/WebCore.vcxproj.filters:
  • platform/network/SynchronousLoaderClient.cpp:
  • platform/network/cf/SynchronousLoaderClientCFNet.cpp: Added.

(WebCore::SynchronousLoaderClient::didReceiveAuthenticationChallenge):
(WebCore::SynchronousLoaderClient::platformBadResponseError):

  • platform/network/soup/SynchronousLoaderClientSoup.cpp: Added.

(WebCore::SynchronousLoaderClient::didReceiveAuthenticationChallenge):
(WebCore::SynchronousLoaderClient::platformBadResponseError):

2:01 AM Changeset in webkit [155107] by jinwoo7.song@samsung.com
  • 2 edits in trunk/Source/WebCore

Fix unused parameter warning in WebKitAccessibleInterfaceText.cpp
<​https://webkit.org/b/120761>

Reviewed by Gyuyoung Kim.

Fix warning by removing unused parameter.

  • accessibility/atk/WebKitAccessibleInterfaceText.cpp:

(sentenceAtPositionForAtkBoundary):

1:41 AM Changeset in webkit [155106] by ljaehun.lim@samsung.com
  • 2 edits in trunk/Source/WTF

Unreviewed. Fix build after r155083

Source/WTF/wtf/Vector.h:1210:48: error: ‘>>’ should be ‘> >’ within a nested template argument list

  • wtf/Vector.h: Add missing space.
12:50 AM WebKitGTK/2.2.x edited by zandobersek@gmail.com
Add r155032 as a proposed merge for 2.1.91. (diff)
12:38 AM Changeset in webkit [155105] by krit@webkit.org
  • 5 edits in trunk

NULL ptr in WebCore::RefCountedPropertyWrapper<WebCore::ClipPathOperation>::blend
​https://bugs.webkit.org/show_bug.cgi?id=105408

Reviewed by Dean Jackson.

Source/WebCore:

Adding an early return if from or to clip-path values are 'none'. According to the
specification we shall just interpolate between two basic shapes.

​http://dev.w3.org/csswg/css-shapes/#basic-shape-interpolation

  • page/animation/CSSPropertyAnimation.cpp:

(WebCore::blendFunc):

LayoutTests:

Test that animation from none to a basic shape on -webkit-clip-path doesn't crash.

  • css3/masking/clip-path-animation-expected.txt:
  • css3/masking/clip-path-animation.html:
12:33 AM Changeset in webkit [155104] by commit-queue@webkit.org
  • 2 edits in trunk/Source/WebCore

[Qt][WK1] REGRESSION(r154988): compositing/video/video-with-invalid-source.html
​https://bugs.webkit.org/show_bug.cgi?id=120683

Patch by Andre Moreira Magalhaes <Andre Moreira Magalhaes> on 2013-09-05
Reviewed by Philippe Normand.

Do not set pipeline state to NULL on MediaPlayerPrivateGStreamer::loadingFailed()
otherwise the bus is flushed and we never get a GST_MESSAGE_ERROR when failing to
load uris.
Also restore previous behaviour (before r154988) of not invoking loadingFailed() for
all failed manual state change attempts.

  • platform/graphics/gstreamer/MediaPlayerPrivateGStreamer.cpp:

(WebCore::MediaPlayerPrivateGStreamer::changePipelineState):
Do not call loadingFailed() if state change fails as all manual state changes are
now done with changePipelineState().
(WebCore::MediaPlayerPrivateGStreamer::play):
(WebCore::MediaPlayerPrivateGStreamer::pause):
(WebCore::MediaPlayerPrivateGStreamer::seek):
(WebCore::MediaPlayerPrivateGStreamer::handleMessage):
Restore previous behaviour (before changeset r154988) when calling changePipelineState().
(WebCore::MediaPlayerPrivateGStreamer::updateStates):
Do nothing if changing to READY on EOS (same behaviour as setting to NULL as it was before
changeset r154988).
(WebCore::MediaPlayerPrivateGStreamer::loadingFailed):
Do not set pipeline state to NULL so we properly get GST_MESSAGE_ERROR on loading failures.

12:23 AM Changeset in webkit [155103] by fpizlo@apple.com
  • 2 edits in trunk/Tools

Unreviewed, don't run fast/js tests in run-javascriptcore-tests on platforms that
can't support it because if non-bourne shells. This fixes part #2 of the Linux
bot breakage.

  • Scripts/run-javascriptcore-tests:
12:22 AM Changeset in webkit [155102] by berto@igalia.com
  • 2 edits in trunk/Source/WebCore

[GTK] MediaControlsGtk: fix warning in constructor due to incorrect order of attributes
​https://bugs.webkit.org/show_bug.cgi?id=120703

Reviewed by Carlos Garcia Campos.

Fix member initialization so it matches their declaration [-Wreorder].

  • html/shadow/MediaControlsGtk.cpp:

(WebCore::MediaControlsGtk::MediaControlsGtk):

12:18 AM Changeset in webkit [155101] by fpizlo@apple.com
  • 2 edits in trunk/Tools

Unreviewed, don't rely on File::Slurp. This fixes part #1 of Linux bot breakage.

  • Scripts/run-javascriptcore-tests:

(readAllLines):

12:14 AM WebKitGTK/2.2.x edited by berto@igalia.com
(diff)
12:06 AM WebKitGTK/2.0.x edited by berto@igalia.com
(diff)
Note: See TracTimeline for information about the timeline view.