Timeline

Aug 14, 2014:

9:24 PM Changeset in webkit [172621] by akling@apple.com

• 3 edits in trunk/Source/WebCore

Remove two unused Range functions.
<​https://webkit.org/b/135944>

Reviewed by Geoffrey Garen.

• dom/Range.cpp:

(WebCore::Range::maxStartOffset): Deleted.
(WebCore::Range::maxEndOffset): Deleted.

• dom/Range.h:

9:10 PM Changeset in webkit [172620] by dbates@webkit.org

• 2 edits in trunk/Source/WebCore

Preserve editing style for -webkit-tap-highlight-color on all ports that enable ENABLE_TOUCH_EVENTS
​https://bugs.webkit.org/show_bug.cgi?id=135963

Reviewed by Andy Estes.

Currently we only preserve the CSS property -webkit-tap-highlight-color during editing operations
for iOS. Instead we should preserve it for all ports that enable ENABLE_TOUCH_EVENTS as this CSS
property isn't specific to iOS.

Also added FIXME comment with regards to the CSS property -webkit-composition-fill-color.

• editing/EditingStyle.cpp:

9:06 PM Changeset in webkit [172619] by benjamin@webkit.org

• 8 edits
  8 adds in trunk

CSS JIT: compile the :empty pseudo class
​https://bugs.webkit.org/show_bug.cgi?id=135958

Reviewed by Andreas Kling.

Source/WebCore:

Compile :empty, nothing fancy.

Tests: fast/selectors/empty-basics.html

fast/selectors/empty-long-adjacent-backtracking.html
fast/selectors/empty-adjacent-style-update.html
fast/selectors/empty-style-update.html

• cssjit/SelectorCompiler.cpp:

(WebCore::SelectorCompiler::addPseudoClassType):
(WebCore::SelectorCompiler::SelectorCodeGenerator::generateElementMatching):
(WebCore::SelectorCompiler::jumpIfElementIsNotEmpty):
(WebCore::SelectorCompiler::setElementStyleIsAffectedByEmpty):
(WebCore::SelectorCompiler::setElementStyleFromContextIsAffectedByEmptyAndUpdateRenderStyleIfNecessary):
(WebCore::SelectorCompiler::setElementStyleIsAffectedByEmptyAndUpdateRenderStyleIfNecessary):
(WebCore::SelectorCompiler::SelectorCodeGenerator::generateElementIsEmpty):

• dom/CharacterData.h:

(WebCore::CharacterData::dataMemoryOffset):

• dom/ContainerNode.h:

(WebCore::ContainerNode::firstChildMemoryOffset):

• dom/Node.h:

(WebCore::Node::flagIsText):

Source/WTF:

• wtf/text/StringImpl.h:

(WTF::StringImpl::lengthMemoryOffset):

LayoutTests:

Add some test coverage, :empty had very little testing.

The test empty-adjacent-style-update expose some problems with style update,
I will look into them separately.

• fast/selectors/empty-adjacent-style-update-expected.txt: Added.
• fast/selectors/empty-adjacent-style-update.html: Added.
• fast/selectors/empty-basics-expected.txt: Added.
• fast/selectors/empty-basics.html: Added.
• fast/selectors/empty-long-adjacent-backtracking-expected.txt: Added.
• fast/selectors/empty-long-adjacent-backtracking.html: Added.
• fast/selectors/empty-style-update-expected.txt: Added.
• fast/selectors/empty-style-update.html: Added.

8:22 PM Changeset in webkit [172618] by akling@apple.com

• 3 edits in trunk/Source/JavaScriptCore

Allocate the whole RegExpMatchesArray backing store up front.
<​https://webkit.org/b/135217>

We were using the generic array backing store allocation path for
RegExpMatchesArray which meant starting with 4 slots and then growing
it dynamically as we append. Since we always know the final number of
entries up front, allocate a perfectly-sized backing store right away.

~2% progression on Octane/regexp.

Reviewed by Geoffrey Garen.

• runtime/JSArray.h:

(JSC::createArrayButterflyWithExactLength):

• runtime/RegExpMatchesArray.cpp:

(JSC::RegExpMatchesArray::create):

5:30 PM Changeset in webkit [172617] by commit-queue@webkit.org

• 3 edits in trunk/Source/WebInspectorUI

Web Inspector: Dragging Undocked inspector by window title has broken behavior
​https://bugs.webkit.org/show_bug.cgi?id=135950

Patch by Joseph Pecoraro <Joseph Pecoraro> on 2014-08-14
Reviewed by Timothy Hatcher.

• UserInterface/Base/Main.js:

On Mac ports where we the inspector content goes everywhere, window dragging behavior
still exists for the top 22px of the window. So don't do our JavaScript based window
dragging for this special region.

• UserInterface/Models/KeyboardShortcut.js:

(WebInspector.KeyboardShortcut.Modifier.get CommandOrControl):
(WebInspector.KeyboardShortcut.prototype.get displayName):
Instead of calling a host function, use the value already in the frontend.

5:21 PM Changeset in webkit [172616] by commit-queue@webkit.org

• 4 edits
  2 adds in trunk/Source/WebCore

Implement scroll snapping animations on Mac
​https://bugs.webkit.org/show_bug.cgi?id=135768

Patch by Wenson Hsieh <Wenson Hsieh> on 2014-08-14
Reviewed by Beth Dakin.

Implementing the scroll snap animations required for snapping in both overflow and mainframe areas on Mac.
Since we receive a series of discrete wheel events, we need to predict the distance we need to traverse,
compute the appropriate snap point, and then compute an animation curve to that location. The snap animations
are split into two types: snapping, which handles letting go of the trackpad with zero velocity, and gliding,
which handles flick gestures. In both cases, sinusoidal curves are used to ease animation to the target
location. In the former case, the initial velocity is low, and increases to a maximum value during the middle
of the animation before decreasing to 0. In the latter case, the curve is computed such that the initial
velocity matches the user's scroll velocity, and the final velocity matches a lower final velocity that is
somewhat arbitrarily computed. (See the FIXME in AxisScrollSnapOffsets::initializeGlideParameters). How the
equations and constants were chosen is described in greater detail in the comments above
AxisScrollSnapOffsets::computeGlideDelta and AxisScrollSnapOffsets::computeSnapDelta. Note that the final
velocity should ideally be equal to 0. However, with this particular curve, this caused the animation to feel
too slow near the snap point.

No tests, since there is no observable change in behavior. Tests will be included when iOS and Mac scrolling
areas hook into this code.

• PlatformMac.cmake:
• WebCore.xcodeproj/project.pbxproj:
• page/scrolling/AxisScrollSnapOffsets.cpp:

(WebCore::closestSnapOffset): Computes the closest snap offset given velocity and prdicted destination.

• page/scrolling/AxisScrollSnapOffsets.h:
• platform/mac/AxisScrollSnapAnimator.h: Added.

(WebCore::AxisScrollSnapAnimatorClient::~AxisScrollSnapAnimatorClient):

• platform/mac/AxisScrollSnapAnimator.mm: Added.

(WebCore::toWheelEventStatus): Converts a pair of PlatformWheelEventPhases to a WheelEventStatus.
(WebCore::projectedInertialScrollDistance): Attempts to predict the distance covered by the inertial scrolling phase.
(WebCore::AxisScrollSnapAnimator::AxisScrollSnapAnimator):
(WebCore::AxisScrollSnapAnimator::handleWheelEvent): Updates the internal state of the AxisScrollSnapAnimator, given a PlatformWheelEvent
(WebCore::AxisScrollSnapAnimator::shouldOverrideWheelEvent): Determines whether or not we should override a wheel event given the animator's internal state.
(WebCore::AxisScrollSnapAnimator::scrollSnapAnimationUpdate): Updates a single loop of the scroll snapping animation.
(WebCore::AxisScrollSnapAnimator::beginScrollSnapAnimation):
(WebCore::AxisScrollSnapAnimator::endScrollSnapAnimation):
(WebCore::AxisScrollSnapAnimator::computeSnapDelta): See comments for more information.
(WebCore::AxisScrollSnapAnimator::computeGlideDelta): See comments for more information.
(WebCore::AxisScrollSnapAnimator::initializeGlideParameters):
(WebCore::AxisScrollSnapAnimator::pushInitialWheelDelta):
(WebCore::AxisScrollSnapAnimator::averageInitialWheelDelta):
(WebCore::AxisScrollSnapAnimator::clearInitialWheelDeltaWindow):

5:18 PM Changeset in webkit [172615] by commit-queue@webkit.org

• 2 edits in trunk/Source/WebInspectorUI

Web Inspector: Split console toggle button wrong - Images/Images/SplitToggleUp.svg does not exist
​https://bugs.webkit.org/show_bug.cgi?id=135957

Patch by Joseph Pecoraro <Joseph Pecoraro> on 2014-08-14
Reviewed by Timothy Hatcher.

• UserInterface/Views/LogContentView.js:

(WebInspector.LogContentView):
Wrong image path passed to platformImagePath.

4:59 PM Changeset in webkit [172614] by commit-queue@webkit.org

• 39 edits in trunk/Source

Allow high fidelity type profiling to be enabled and disabled.
​https://bugs.webkit.org/show_bug.cgi?id=135423

Patch by Saam Barati <​sbarati@apple.com> on 2014-08-14
Reviewed by Geoffrey Garen.

Source/JavaScriptCore:

• Merged op_put_to_scope_with_profile and op_get_from_scope_with_profile into op_profile_types_with_high_fidelity by adding extra arguments to the opcode.
• Altered SymbolTable to use less memory by adding a rare data structure for type profiling.
• Created an interface to turn on and off type profiling from the Web Inspector.
• Refactored how entries are written to HighFidelityLog to make it easier to inline when generating machine code.
• Implemented op_profile_types_with_high_fidelity in the baseline JIT by inlining the process of writing to the log and doing a small amount of type inference optimizations.

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

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

• bytecode/CodeBlock.cpp:

(JSC::CodeBlock::dumpBytecode):
(JSC::CodeBlock::CodeBlock):
(JSC::CodeBlock::finalizeUnconditionally):
(JSC::CodeBlock::scopeDependentProfile): Deleted.

• bytecode/CodeBlock.h:
• bytecode/TypeLocation.h:

(JSC::TypeLocation::TypeLocation):

• bytecompiler/BytecodeGenerator.cpp:

(JSC::BytecodeGenerator::generate):
(JSC::BytecodeGenerator::emitMove):
(JSC::BytecodeGenerator::emitProfileTypesWithHighFidelity):
(JSC::BytecodeGenerator::emitGetFromScopeWithProfile): Deleted.
(JSC::BytecodeGenerator::emitPutToScopeWithProfile): Deleted.

• bytecompiler/BytecodeGenerator.h:
• bytecompiler/NodesCodegen.cpp:

(JSC::ThisNode::emitBytecode):
(JSC::ResolveNode::emitBytecode):
(JSC::BracketAccessorNode::emitBytecode):
(JSC::DotAccessorNode::emitBytecode):
(JSC::FunctionCallValueNode::emitBytecode):
(JSC::FunctionCallResolveNode::emitBytecode):
(JSC::FunctionCallBracketNode::emitBytecode):
(JSC::FunctionCallDotNode::emitBytecode):
(JSC::CallFunctionCallDotNode::emitBytecode):
(JSC::ApplyFunctionCallDotNode::emitBytecode):
(JSC::PostfixNode::emitResolve):
(JSC::PostfixNode::emitBracket):
(JSC::PostfixNode::emitDot):
(JSC::PrefixNode::emitResolve):
(JSC::PrefixNode::emitBracket):
(JSC::PrefixNode::emitDot):
(JSC::ReadModifyResolveNode::emitBytecode):
(JSC::AssignResolveNode::emitBytecode):
(JSC::AssignDotNode::emitBytecode):
(JSC::ReadModifyDotNode::emitBytecode):
(JSC::AssignBracketNode::emitBytecode):
(JSC::ReadModifyBracketNode::emitBytecode):
(JSC::ReturnNode::emitBytecode):
(JSC::FunctionBodyNode::emitBytecode):

• inspector/agents/InspectorRuntimeAgent.cpp:

(Inspector::InspectorRuntimeAgent::InspectorRuntimeAgent):
(Inspector::InspectorRuntimeAgent::getRuntimeTypesForVariablesAtOffsets):
(Inspector::TypeRecompiler::operator()):
(Inspector::recompileAllJSFunctionsForTypeProfiling):
(Inspector::InspectorRuntimeAgent::willDestroyFrontendAndBackend):
(Inspector::InspectorRuntimeAgent::enableHighFidelityTypeProfiling):
(Inspector::InspectorRuntimeAgent::disableHighFidelityTypeProfiling):
(Inspector::InspectorRuntimeAgent::setHighFidelityTypeProfilingEnabledState):

• inspector/agents/InspectorRuntimeAgent.h:
• inspector/agents/JSGlobalObjectRuntimeAgent.cpp:

(Inspector::JSGlobalObjectRuntimeAgent::willDestroyFrontendAndBackend):

• inspector/protocol/Runtime.json:
• jit/JIT.cpp:

(JSC::JIT::privateCompileMainPass):
(JSC::JIT::privateCompile):

• jit/JIT.h:
• jit/JITOpcodes.cpp:

(JSC::JIT::emit_op_profile_types_with_high_fidelity):

• jit/JITOpcodes32_64.cpp:

(JSC::JIT::emit_op_profile_types_with_high_fidelity):

• jit/JITOperations.cpp:
• jit/JITOperations.h:
• llint/LLIntSlowPaths.cpp:

(JSC::LLInt::LLINT_SLOW_PATH_DECL):
(JSC::LLInt::getFromScopeCommon): Deleted.
(JSC::LLInt::putToScopeCommon): Deleted.

• llint/LLIntSlowPaths.h:
• llint/LowLevelInterpreter.asm:
• runtime/CodeCache.cpp:

(JSC::CodeCache::getGlobalCodeBlock):

• runtime/CommonSlowPaths.cpp:

(JSC::SLOW_PATH_DECL):

• runtime/CommonSlowPaths.h:
• runtime/HighFidelityLog.cpp:

(JSC::HighFidelityLog::initializeHighFidelityLog):
(JSC::HighFidelityLog::~HighFidelityLog):
(JSC::HighFidelityLog::processHighFidelityLog):

• runtime/HighFidelityLog.h:

(JSC::HighFidelityLog::LogEntry::structureIDOffset):
(JSC::HighFidelityLog::LogEntry::valueOffset):
(JSC::HighFidelityLog::LogEntry::locationOffset):
(JSC::HighFidelityLog::recordTypeInformationForLocation):
(JSC::HighFidelityLog::logEndPtr):
(JSC::HighFidelityLog::logStartOffset):
(JSC::HighFidelityLog::currentLogEntryOffset):

• runtime/HighFidelityTypeProfiler.cpp:

(JSC::HighFidelityTypeProfiler::logTypesForTypeLocation):
(JSC::descriptorMatchesTypeLocation):

• runtime/HighFidelityTypeProfiler.h:
• runtime/SymbolTable.cpp:

(JSC::SymbolTable::SymbolTable):
(JSC::SymbolTable::cloneCapturedNames):
(JSC::SymbolTable::prepareForHighFidelityTypeProfiling):
(JSC::SymbolTable::uniqueIDForVariable):
(JSC::SymbolTable::uniqueIDForRegister):
(JSC::SymbolTable::globalTypeSetForRegister):
(JSC::SymbolTable::globalTypeSetForVariable):

• runtime/SymbolTable.h:

(JSC::SymbolTable::add):
(JSC::SymbolTable::set):

• runtime/TypeLocationCache.cpp:

(JSC::TypeLocationCache::getTypeLocation):

• runtime/TypeSet.cpp:

(JSC::TypeSet::getRuntimeTypeForValue):
(JSC::TypeSet::addTypeInformation):
(JSC::TypeSet::allPrimitiveTypeNames):
(JSC::TypeSet::addTypeForValue): Deleted.

• runtime/TypeSet.h:
• runtime/VM.cpp:

(JSC::VM::VM):
(JSC::VM::nextTypeLocation):
(JSC::VM::enableHighFidelityTypeProfiling):
(JSC::VM::disableHighFidelityTypeProfiling):
(JSC::VM::dumpHighFidelityProfilingTypes):

• runtime/VM.h:

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

Source/WebCore:

PageRuntimeAgent and WorkerRuntimeAgent now call their super
class's (InspectorRuntimeAgent) implementation of willDestroyFrontendAndBackend
to give InspectorRuntimeAgent a chance to recompile all JavaScript
functions, if necessary, for type profiling.

• inspector/PageRuntimeAgent.cpp:

(WebCore::PageRuntimeAgent::willDestroyFrontendAndBackend):

• inspector/WorkerRuntimeAgent.cpp:

(WebCore::WorkerRuntimeAgent::willDestroyFrontendAndBackend):

4:50 PM Changeset in webkit [172613] by Lucas Forschler

• 5 edits in branches/safari-600.1-branch/Source

Versioning.

4:43 PM Changeset in webkit [172612] by Lucas Forschler

• 1 copy in tags/Safari-600.1.10

New Tag.

3:59 PM Changeset in webkit [172611] by Lucas Forschler

• 2 edits in branches/safari-600.1-branch/Source/WebCore

Merged r172609. <rdar://problem/18023620>

3:56 PM Changeset in webkit [172610] by Lucas Forschler

• 5 edits in branches/safari-600.1-branch

Merged r172604. <rdar://problem/18022221>

3:22 PM Changeset in webkit [172609] by Brent Fulgham

• 2 edits in trunk/Source/WebCore

[Win] Correct build when DerivedSources location not supplied by build environment
​https://bugs.webkit.org/show_bug.cgi?id=135954
<rdar://problem/18023620>

Reviewed by Joseph Pecoraro.

The DerivedSourcesDirectory is not always defined, depending on what
solution is being used to build this project. Instead, just define
it directly, since we already know where it needs to go.

Also:

1. Stop redirecting the copy output to NUL so I can see if this fails in the future.
2. Make sure the target directory exists before copying to it.

• WebCore.vcxproj/copyForwardingHeaders.cmd: Properly specify

the directories for copying the JS file.

2:39 PM Changeset in webkit [172608] by ggaren@apple.com

• 2 edits in trunk/Source/bmalloc

Fixed a bmalloc crash seen on the EWS bot
​https://bugs.webkit.org/show_bug.cgi?id=135955

Reviewed by Andreas Kling.

• bmalloc/Syscall.h: Some CG APIs vm_copy their input buffers. If the

input buffer is a malloc region, that region will get marked Copy-On-Write
by the kernel. Calls to madvise() for COW regions fail and return EINVAL
on older OS X's. In 10.10, they still fail, but they do not return
EINVAL.

So, we can only ASSERT that our syscalls succeed starting with 10.10.

2:36 PM Changeset in webkit [172607] by zoltan@webkit.org

• 2 edits in trunk/Source/WebCore

[CSS3-Text] Cosmetics on RenderBlockFlow::textAlignmentForLine
​https://bugs.webkit.org/show_bug.cgi?id=135940

Reviewed by Darin Adler.

Processing the not CSS3_TEXT case first, and removing an extra whitespace after r172524.

• rendering/RenderBlockLineLayout.cpp:

(WebCore::RenderBlockFlow::textAlignmentForLine):

2:34 PM Changeset in webkit [172606] by commit-queue@webkit.org

• 2 edits in trunk/Source/WebInspectorUI

Web Inspector: Avoid re-binding the same function over and over
​https://bugs.webkit.org/show_bug.cgi?id=135949

Patch by Joseph Pecoraro <Joseph Pecoraro> on 2014-08-14
Reviewed by Timothy Hatcher.

• UserInterface/Controllers/TimelineManager.js:

(WebInspector.TimelineManager.prototype._startAutoCapturing):
(WebInspector.TimelineManager.prototype._stopAutoRecordingSoon):
(WebInspector.TimelineManager.prototype._resetAutoRecordingDeadTimeTimeout):
Stash a bound copy of the function once, and reuse it.

2:17 PM Changeset in webkit [172605] by ggaren@apple.com

• 2 edits in trunk/Source/bmalloc

Fixed the bmalloc build
​https://bugs.webkit.org/show_bug.cgi?id=135953

Reviewed by Andreas Kling.

• bmalloc.xcodeproj/project.pbxproj: Marked a few headers as private.

These headers are used, so they must be available outside the project.

2:09 PM Changeset in webkit [172604] by Brent Fulgham

• 5 edits in trunk

[Win] Extend auto-version and version-stamp for RC_ProjectBuildVersion
​https://bugs.webkit.org/show_bug.cgi?id=135948
<rdar://problem/18022221>

Reviewed by David Kilzer.

Tools:

• Scripts/webkitperl/auto-version_unittest/autoVersionTests.pl: Add tests

for RC_ProjectBuildVersion.

WebKitLibraries:

• win/tools/scripts/auto-version.pl:

(splitVersion): Use RC_ProjectBuildVersion (or RC_PROJECTBUILDVERSION) if set
in the environment as the 'Build Version', unless the multi-tuple version
supplied by RC_ProjectSourceVersion already has a fourth component.

• win/tools/scripts/version-stamp.pl: Check for RC_ProjectBuildVersion and

use it in preference to RC_PROJECTBUILDVERSION.

2:04 PM Changeset in webkit [172603] by mhock@apple.com

• 3 edits
  3 adds in trunk

IndexedDB should respect SchemeRegistry's database access setting.
​https://bugs.webkit.org/show_bug.cgi?id=135890
<rdar://problem/17989469>

Reviewed by Alexey Proskuryakov.

Source/WebCore:

Test: storage/indexeddb/open-db-private-browsing.html

• Modules/indexeddb/IDBFactory.cpp:

LayoutTests:

• storage/indexeddb/open-db-private-browsing-expected.txt: Added.
• storage/indexeddb/open-db-private-browsing.html: Added. Simple test to open database in private browsing mode.
• storage/indexeddb/resources/open-db-private-browsing.js: Added.

1:58 PM Changeset in webkit [172602] by dfarler@apple.com

• 2 edits in trunk/Tools

[iOS] run-webkit-tests runs webkit-build-directory on every test
​https://bugs.webkit.org/show_bug.cgi?id=135409

Reviewed by Daniel Bates.

• Scripts/webkitpy/port/ios.py:

(IOSSimulatorPort.init): Cache Mac build directory.
(IOSSimulatorPort.relay_path): Use cached build directory.
(IOSSimulatorPort._path_to_image_diff): Use cached build directory.

1:40 PM Changeset in webkit [172601] by commit-queue@webkit.org

• 3 edits in trunk/Source/WebInspectorUI

Web Inspector: Timelines can get caught in requestAnimationFrame loop when not recording
​https://bugs.webkit.org/show_bug.cgi?id=135946

Patch by Joseph Pecoraro <Joseph Pecoraro> on 2014-08-14
Reviewed by Timothy Hatcher.

Each TimelineContentView was listening for the TimelineManager's start/stop
capturing events. However, only the active recording could possibly update.
Likewise, if a pre-existing recording started and stopped we did not properly
stop in the case where we were waiting for an event to update the current time,
which would result in an infinite rAF loop of no updates.

• UserInterface/Models/TimelineRecording.js:

(WebInspector.TimelineRecording.prototype.unloaded):
When a timeline recording is unloaded trigger an event.

• UserInterface/Views/TimelineContentView.js:

(WebInspector.TimelineContentView.prototype._startUpdatingCurrentTime):
Assert we were not already waiting.

(WebInspector.TimelineContentView.prototype._stopUpdatingCurrentTime):
Properly handle the case of starting/stoping an already existing recording
without a new event coming in.

(WebInspector.TimelineContentView.prototype._recordingTimesUpdated):
(WebInspector.TimelineContentView.prototype._recordingReset):
Convert to a boolean instead of deleting state.

(WebInspector.TimelineContentView.prototype._recordingUnloaded):
Stop listening to timeline manager events that no longer affect unloaded recordings.

12:46 PM Changeset in webkit [172600] by yoav@yoav.ws

• 2 edits in trunk/Source/WebCore

srcset with w descriptor doesn't behave as expected when sizes is not supported.
​https://bugs.webkit.org/show_bug.cgi?id=135935

Reviewed by Dean Jackson.

In current implementation, if sizes is not supported, when authors would write
markup such as <img src=fallback.jpg srcset="100px.jpg 100w, 400px.jpg 400w">
the first candidate in srcset would be picked, regardless of its dimensions
That is likely to be confusing for authors.
Dropping these "not yet supported" candidates is likely to be less confusing,
and will result in the "fallback.jpg" candidate being picked.

No new tests since this change only concerns builds that are built
with the PICTURE_SIZES compile flag turned off.

• html/parser/HTMLSrcsetParser.cpp:

(WebCore::parseDescriptors):
Drop candidates that include either 'w' or 'h' descriptors when the
sizes feature is not supported.

12:38 PM Changeset in webkit [172599] by Antti Koivisto

• 2 edits in trunk/Source/WebCore

Make inherited style and parent renderer references
​https://bugs.webkit.org/show_bug.cgi?id=135857

Reviewed by Andreas Kling.

We don't recurse into non-rendered subtrees anymore so they can't be null.
Also stop using Element::renderStyle() as it differs from e->renderer().style() in
some rare cases (that should be fixed).

• style/StyleResolveTree.cpp:

(WebCore::Style::RenderTreePosition::parent):
(WebCore::Style::RenderTreePosition::RenderTreePosition):
(WebCore::Style::RenderTreePosition::canInsert):
(WebCore::Style::RenderTreePosition::insert):
(WebCore::Style::RenderTreePosition::computeNextSibling):
(WebCore::Style::shouldCreateRenderer):
(WebCore::Style::styleForElement):
(WebCore::Style::createRendererIfNeeded):
(WebCore::Style::textRendererIsNeeded):
(WebCore::Style::createTextRendererIfNeeded):
(WebCore::Style::updateTextRendererAfterContentChange):
(WebCore::Style::attachChildren):
(WebCore::Style::attachDistributedChildren):
(WebCore::Style::attachShadowRoot):
(WebCore::Style::resetStyleForNonRenderedDescendants):
(WebCore::Style::attachBeforeOrAfterPseudoElementIfNeeded):
(WebCore::Style::attachRenderTree):
(WebCore::Style::resolveLocal):
(WebCore::Style::resolveShadowTree):
(WebCore::Style::updateBeforeOrAfterPseudoElement):
(WebCore::Style::resolveTree):

12:28 PM Changeset in webkit [172598] by oliver@apple.com

• 7 edits in trunk/Source/JavaScriptCore

Update scope resolution to assume that the parent activation is always there
​https://bugs.webkit.org/show_bug.cgi?id=135947

Reviewed by Andreas Kling.

Another incremental step in removing the idea of lazily created
activations.

• dfg/DFGSpeculativeJIT32_64.cpp:

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

• dfg/DFGSpeculativeJIT64.cpp:

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

• jit/JITPropertyAccess.cpp:

(JSC::JIT::emitResolveClosure):

• jit/JITPropertyAccess32_64.cpp:

(JSC::JIT::emitResolveClosure):

• llint/LowLevelInterpreter32_64.asm:
• llint/LowLevelInterpreter64.asm:

11:54 AM Changeset in webkit [172597] by Antti Koivisto

• 2 edits in trunk/Source/WebCore

[Forms] We should share RenderStyle object for optgroup and option element
​https://bugs.webkit.org/show_bug.cgi?id=88405

Reviewed by Andreas Kling.

We can remove this special case now.

• css/StyleResolver.cpp:

(WebCore::StyleResolver::canShareStyleWithElement):

11:36 AM Changeset in webkit [172596] by enrica@apple.com

• 2 edits in trunk/Source/WebKit2

[Services with UI] Selection services don't work inside <iframes>.
​https://bugs.webkit.org/show_bug.cgi?id=135941
<rdar://problem/17957690>

Reviewed by Tim Horton.

Need to map the selection rectangles using the correct FrameView.
When handling the click, we must use the selection from the focused frame.

• WebProcess/WebPage/mac/ServicesOverlayController.mm:

(WebKit::ServicesOverlayController::buildSelectionHighlight):
(WebKit::ServicesOverlayController::handleClick):

11:17 AM Changeset in webkit [172595] by mrowe@apple.com

• 6 edits in trunk/Source

<​https://webkit.org/b/135909> Move helper applications out of the root of the framework.

As described in <​https://developer.apple.com/library/mac/technotes/tn2206/>, for bundles containing
a Versions directory there may be no other content at the top level of the bundle other than symlinks.
Upcoming changes to code signing will prevent bundles that violate this rule from being signed.

Reviewed by Sam Weinig.

Source/WebKit:

• WebKit.xcodeproj/project.pbxproj: Add the symlink to WebKitPluginHost.app in the Versions/A

directory of the framework rather than at the top level.

Source/WebKit2:

• Configurations/Base.xcconfig: Define a configuration setting that points to the content directory

of the framework. On OS X this is Versions/A. On iOS, where frameworks are shallow, this is the top level.

• Configurations/BaseLegacyProcess.xcconfig: Install the legacy processes in the content directory

of the framework.

• WebKit2.xcodeproj/project.pbxproj: Copy the legacy processes into the content directory of the

framework during engineering builds. Generate symlinks for the legacy processes to their locations
in Versions/Current. This is necessary because -[NSBundle pathForAuxiliaryExecutable:] only looks
at the top level of the framework wrapper.

11:07 AM Changeset in webkit [172594] by oliver@apple.com

• 7 edits in trunk/Source/JavaScriptCore

Create activations eagerly
​https://bugs.webkit.org/show_bug.cgi?id=135942

Reviewed by Geoffrey Garen.

Prepare to rewrite activation objects into a more
sane implementation. Step 1 is reverting to eager
creation of the activation object. This results in
a 1.35x regression in earley, but otherwise has a
minimal performance impact.

The earley regression is being tracked by bug #135943

• bytecompiler/BytecodeGenerator.cpp:

(JSC::BytecodeGenerator::BytecodeGenerator):
(JSC::BytecodeGenerator::emitNewFunctionInternal):
(JSC::BytecodeGenerator::emitNewFunctionExpression):
(JSC::BytecodeGenerator::emitCallEval):
(JSC::BytecodeGenerator::emitPushWithScope):
(JSC::BytecodeGenerator::emitPushCatchScope):
(JSC::BytecodeGenerator::createActivationIfNecessary): Deleted.

• bytecompiler/BytecodeGenerator.h:
• jit/JITOpcodes.cpp:

(JSC::JIT::emit_op_create_activation):

• jit/JITOpcodes32_64.cpp:

(JSC::JIT::emit_op_create_activation):

• llint/LowLevelInterpreter32_64.asm:
• llint/LowLevelInterpreter64.asm:

10:59 AM Changeset in webkit [172593] by achristensen@apple.com

• 1 edit
  1 delete in trunk/Source/WebKit

Unreviewed. Removing empty directories.

• qt: Removed.
• qt/declarative: Removed.
• qt/docs: Removed.
• qt/docs/webkitsnippets: Removed.
• qt/examples: Removed.
• qt/tests: Removed.
• qt/tests/MIMESniffing: Removed.
• qt/tests/benchmarks: Removed.
• qt/tests/qgraphicswebview: Removed.
• qt/tests/qwebelement: Removed.
• qt/tests/qwebframe: Removed.
• qt/tests/qwebhistory: Removed.
• qt/tests/qwebpage: Removed.
• qt/tests/qwebsecurityorigin: Removed.
• qt/tests/qwebview: Removed.

10:52 AM Changeset in webkit [172592] by commit-queue@webkit.org

• 2 edits in trunk/Source/WebKit2

WebKit2GTK - WebKitWebProcess assertion fails when dragging and dropping a file into the view
​https://bugs.webkit.org/show_bug.cgi?id=127576

Patch by Michael Catanzaro <Michael Catanzaro> on 2014-08-14
Reviewed by Carlos Garcia Campos.

• UIProcess/WebPageProxy.cpp:

(WebKit::WebPageProxy::performDragControllerAction): Assume read access
to any file that has been dragged into the web view when compiling for
GTK, since we don't support sandbox extensions.

10:49 AM Changeset in webkit [172591] by Antti Koivisto

• 10 edits in trunk/Source/WebCore

Remove nonRendererStyle
​https://bugs.webkit.org/show_bug.cgi?id=135938

Reviewed by Andreas Kling.

It is used to support styling of non-renderer option elements. We have a better
mechanism for this, Element::computedStyle().

• dom/Element.cpp:

(WebCore::Element::resetComputedStyle):
(WebCore::Element::willResetComputedStyle):

Add a virtual callback for computed style reset.

• dom/Element.h:
• dom/Node.h:

(WebCore::Node::nonRendererStyle): Deleted.

• dom/NodeRenderStyle.h:

(WebCore::Node::renderStyle):

renderStyle() now always matches renderer()->style()

• html/HTMLOptGroupElement.cpp:

(WebCore::HTMLOptGroupElement::HTMLOptGroupElement):
(WebCore::HTMLOptGroupElement::isFocusable):
(WebCore::HTMLOptGroupElement::didAttachRenderers): Deleted.
(WebCore::HTMLOptGroupElement::willDetachRenderers): Deleted.
(WebCore::HTMLOptGroupElement::updateNonRenderStyle): Deleted.
(WebCore::HTMLOptGroupElement::nonRendererStyle): Deleted.
(WebCore::HTMLOptGroupElement::customStyleForRenderer): Deleted.

• html/HTMLOptGroupElement.h:
• html/HTMLOptionElement.cpp:

(WebCore::HTMLOptionElement::isFocusable):

Use computedStyle.

(WebCore::HTMLOptionElement::willResetComputedStyle):
(WebCore::HTMLOptionElement::didAttachRenderers): Deleted.
(WebCore::HTMLOptionElement::willDetachRenderers): Deleted.
(WebCore::HTMLOptionElement::updateNonRenderStyle): Deleted.
(WebCore::HTMLOptionElement::nonRendererStyle): Deleted.
(WebCore::HTMLOptionElement::customStyleForRenderer): Deleted.
(WebCore::HTMLOptionElement::didRecalcStyle): Deleted.

• html/HTMLOptionElement.h:
• rendering/RenderMenuList.cpp:

(RenderMenuList::updateOptionsWidth):
(RenderMenuList::setTextFromOption):

Use computedStyle.

9:39 AM Changeset in webkit [172590] by Lucas Forschler

• 3 edits in branches/safari-600.1-branch/Source/WebCore

Merged r172584. <rdar://problem/18015428>

9:37 AM Changeset in webkit [172589] by Lucas Forschler

• 1 edit in branches/safari-600.1-branch/Source/WebInspectorUI/UserInterface/Views/DashboardContainerView.css

Merge patch from <rdar://problem/18016176>.

9:27 AM Changeset in webkit [172588] by Lucas Forschler

• 2 edits in branches/safari-600.1-branch/Source/WebCore

Merged r172506. <rdar://problem/17810998>

9:21 AM Changeset in webkit [172587] by commit-queue@webkit.org

• 4 edits in trunk

Add support for ppc, ppc64, ppc64le, s390, s390x into the CMake build
​https://bugs.webkit.org/show_bug.cgi?id=135937

Patch by Tomas Popela <​tpopela@redhat.com> on 2014-08-14
Reviewed by Carlos Garcia Campos.

• CMakeLists.txt:

9:08 AM Changeset in webkit [172586] by zandobersek@gmail.com

• 3 edits in trunk/Tools

Initialize a separate variable (TestWTF_SOURCES) with the list of files
that are built into the TestWTF executable and move it above the platform-specific
config file inclusion. This makes it possible to add platform-specific files to that
list (e.g. GUniquePtr.cpp for the GTK port).

Rubber-stamped by Carlos Garcia Campos.

• TestWebKitAPI/CMakeLists.txt:
• TestWebKitAPI/PlatformGTK.cmake:

9:06 AM WebKitGTK/KeepingTheTreeGreen edited by clopez@igalia.com

(diff)

8:58 AM Changeset in webkit [172585] by clopez@igalia.com

• 4 edits
  1 add
  1 delete in trunk/LayoutTests

[GTK] Unreviewed GTK gardening.

• platform/gtk/TestExpectations: Update expectations with new identified failures and flakiness.
• platform/gtk/fast/ruby/ruby-base-merge-block-children-crash-2-expected.txt: Rebaseline after r172517.
• platform/gtk/fast/ruby/ruby/ruby-base-merge-block-children-crash-2-expected.png: Removed. File was in wrong path.
• platform/gtk/fast/ruby/ruby/ruby-base-merge-block-children-crash-2-expected.txt: Removed. File was in wrong path.
• platform/gtk/inspector-protocol/dom/getAccessibilityPropertiesForNode-expected.txt: Rebaseline after r172136.
• platform/gtk/scrollbars/scrollbar-selectors-expected.txt: Added. Rebaseline after r172220.

8:13 AM Changeset in webkit [172584] by jer.noble@apple.com

• 3 edits in trunk/Source/WebCore

[MSE] Altering the quality of a YouTube video will cause the video to distort and display an error message
​https://bugs.webkit.org/show_bug.cgi?id=135931

Reviewed by Eric Carlson.

When removing samples from the TrackBuffer's sample map, also remove those samples from the
TrackBuffer's decode queue. Otherwise, removed samples may persist in the decode queue and
either break sync-sample dependencies or cause decoding artifacts.

Pull the code which removes samples from a track buffer into its own utility function, and
use this function both from removeCodedFrames(), and also when samples are removed due to
overlapping appends in sourceBufferPrivateDidReceiveSample(). In order to reference
TrackBuffers outside of SourceBuffer (and in the static removeSamplesFromTrackBuffer()
function), make TrackBuffer a public forward declaration.

• Modules/mediasource/SourceBuffer.cpp:

(WebCore::removeSamplesFromTrackBuffer):
(WebCore::SourceBuffer::removeCodedFrames):
(WebCore::SourceBuffer::sourceBufferPrivateDidReceiveSample):

• Modules/mediasource/SourceBuffer.h:

7:44 AM WebKitGTK/KeepingTheTreeGreen edited by clopez@igalia.com

(diff)

6:10 AM Changeset in webkit [172583] by Carlos Garcia Campos

• 3 edits in trunk

[GTK] REGRESSION(r171598, CMake): Introspection files are installed at the wrong path
​https://bugs.webkit.org/show_bug.cgi?id=135934

Reviewed by Gustavo Noronha Silva.

Since r171598 gir files are installed under $datadir/webkitgtk-4.0/gir-1.0/.
The gir files should installed in $datadir/gir-1.0/ together with all other
gir files. We don't need to install introspection files in a versioned
directory because their filenames already contain the binary version. But before
r171598, the files were only installed to the right directory if the
gobject-instrospection pkg-config file was in the same prefix than the one we
wanted to install, because the gir and typelibs directories were extracted from
the pkg-config file. We should define INTROSPECTION_INSTALL_GIRDIR and
INTROSPECTION_INSTALL_TYPELIBDIR in OptionsGTK.cmake based on $datadir and $libdir
like we do in the autotools build.

• Source/cmake/FindGObjectIntrospection.cmake: Do not define

INTROSPECTION_INSTALL_GIRDIR and INTROSPECTION_INSTALL_TYPELIBDIR.

• Source/cmake/OptionsGTK.cmake: Define

INTROSPECTION_INSTALL_GIRDIR and INTROSPECTION_INSTALL_TYPELIBDIR.

4:13 AM Changeset in webkit [172582] by mihnea@adobe.com

• 1 edit
  8 moves
  1 add in trunk/LayoutTests

[CSSRegions] Move svg tests into fast/regions/svg
​https://bugs.webkit.org/show_bug.cgi?id=135883

Reviewed by Andrei Bucur.

Move tests and adjust paths accordingly.

• fast/regions/svg/svg-doc-fragment-not-collected-expected.html: Renamed from LayoutTests/fast/regions/svg-doc-fragment-not-collected-expected.html.
• fast/regions/svg/svg-doc-fragment-not-collected.html: Renamed from LayoutTests/fast/regions/svg-doc-fragment-not-collected.html.
• fast/regions/svg/svg-element-not-collected-expected.html: Renamed from LayoutTests/fast/regions/svg-element-not-collected-expected.html.
• fast/regions/svg/svg-element-not-collected.html: Renamed from LayoutTests/fast/regions/svg-element-not-collected.html.
• fast/regions/svg/svg-root-element-collected-expected.txt: Renamed from LayoutTests/fast/regions/svg-root-element-collected-expected.txt.
• fast/regions/svg/svg-root-element-collected.html: Renamed from LayoutTests/fast/regions/svg-root-element-collected.html.
• fast/regions/svg/symbol-in-named-flow-crash-expected.txt: Renamed from LayoutTests/fast/regions/symbol-in-named-flow-crash-expected.txt.
• fast/regions/svg/symbol-in-named-flow-crash.svg: Renamed from LayoutTests/fast/regions/symbol-in-named-flow-crash.svg.

3:31 AM Changeset in webkit [172581] by Carlos Garcia Campos

• 2 edits in trunk

[GTK] WebKit DOM documentation (webkitdomgtk) not installed after make install
​https://bugs.webkit.org/show_bug.cgi?id=135836

Reviewed by Philippe Normand.

• Source/PlatformGTK.cmake: Add install command to also install

the GObject DOM bindings API docs.

2:35 AM Changeset in webkit [172580] by zandobersek@gmail.com

• 2 edits in trunk/Source/WebCore

ImageBufferDataCairo.h is missing header guards
​https://bugs.webkit.org/show_bug.cgi?id=135933

Reviewed by Philippe Normand.

• platform/graphics/cairo/ImageBufferDataCairo.h:

Add the missing head guards to avoid problems when re-inclusion occurs.

2:33 AM Changeset in webkit [172579] by zandobersek@gmail.com

• 2 edits in trunk

Unreviewed. In r165709 I mistakenly appended the -fno-exceptions
and -fno-strict-aliasing compiler flags to CMAKE_CXX_FLAGS and then
re-set CMAKE_C_FLAGS with the new string. The two flags should really
be appended to CMAKE_C_FLAGS and the same variable re-set with the
new string.

• Source/cmake/OptionsCommon.cmake:

1:49 AM Changeset in webkit [172578] by commit-queue@webkit.org

• 2 edits in trunk/Source/JavaScriptCore

Fix JSC::ARM64Assembler::LinkRecord::RealTypes
​https://bugs.webkit.org/show_bug.cgi?id=135906

Patch by Akos Kiss <​akiss@inf.u-szeged.hu> on 2014-08-14
Reviewed by Michael Saboff.

JSC::ARM64Assembler::LinkRecord::RealTypes::m_compareRegister is defined
to occupy 5 bits but JSC::ARM64Assembler::RegisterID needs 6 bits. So,
increase the size of the bit field and also reorganize the struct to
better align with word boundaries.

• assembler/ARM64Assembler.h:

Aug 13, 2014:

10:09 PM Changeset in webkit [172577] by dbates@webkit.org

• 2 edits in trunk/Source/bmalloc

Attempt to fix the build following <​http://trac.webkit.org/changeset/172576>
(​https://bugs.webkit.org/show_bug.cgi?id=135895)

Substitute PerThreadStorage<T>::initSharedKeyIfNeeded() for initSharedKeyIfNeeded() in
implementation of PerThread<T>::getFastCase().

• bmalloc/PerThread.h:

(bmalloc::PerThread<T>::getFastCase):

10:00 PM Changeset in webkit [172576] by dbates@webkit.org

• 3 edits in trunk/Source/bmalloc

Make bmalloc::PerThread work without C++ thread local storage
​https://bugs.webkit.org/show_bug.cgi?id=135895

Reviewed by Geoffrey Garen.

Implement support for building bmalloc without C++ thread local storage.

• bmalloc/BPlatform.h: Remove macro define BPLATFORM_IOS_SIMULATOR. Added macro function

BCOMPILER_SUPPORTS() and macro define BCOMPILER_SUPPORTS_CXX_THREAD_LOCAL that can be used
to determine whether the compiler supports C++ thread local storage.

• bmalloc/PerThread.h:

(bmalloc::PerThreadStorage::get): Modified to call pthread_getspecific() when building
without C++ thread local storage.
(bmalloc::PerThreadStorage::initSharedKeyIfNeeded): Added.
(bmalloc::PerThreadStorage::init): Moved logic to initialize shared Pthread key from here to
PerThreadStorage::initSharedKeyIfNeeded().
(bmalloc::PerThread<T>::getFastCase): Modified to call PerThreadStorage::initSharedKeyIfNeeded()
before querying PerThreadStorage::get() when building without C++ thread local storage so as to
ensure that the shared key has been initialized.
(_pthread_setspecific_direct): Deleted.
(_pthread_getspecific_direct): Deleted.

9:25 PM Changeset in webkit [172575] by Lucas Forschler

• 2 edits in branches/safari-600.1-branch/Source/WebKit2

Merged r172546. <rdar://problem/17998929>

9:23 PM Changeset in webkit [172574] by Lucas Forschler

• 2 edits in branches/safari-600.1-branch/Source/WebKit2

Merged r172539. <rdar://problem/17998706>

9:21 PM Changeset in webkit [172573] by Lucas Forschler

• 2 edits in branches/safari-600.1-branch/Source/WebCore

Merged r172538. <rdar://problem/17665229>

9:18 PM Changeset in webkit [172572] by Lucas Forschler

• 3 edits
  2 copies in branches/safari-600.1-branch

Merged r172534. <rdar://problem/17942035>

9:16 PM Changeset in webkit [172571] by Lucas Forschler

• 2 edits in branches/safari-600.1-branch/Source/WebInspectorUI

Merged r172533. <rdar://problem/18009039>

9:15 PM Changeset in webkit [172570] by Lucas Forschler

• 2 edits in branches/safari-600.1-branch/Source/WebInspectorUI

Merged r172531. <rdar://problem/18008775>

9:13 PM Changeset in webkit [172569] by Lucas Forschler

• 2 edits in branches/safari-600.1-branch/Source/WebInspectorUI

Merged r172530. <rdar://problem/18008584>

9:11 PM Changeset in webkit [172568] by Lucas Forschler

• 3 edits in branches/safari-600.1-branch/Source/WebKit2

Merged r172528. <rdar://problem/17982341>

9:10 PM Changeset in webkit [172567] by Lucas Forschler

• 3 edits in branches/safari-600.1-branch/Source/WebInspectorUI

Merged r172527. <rdar://problem/17999511>

9:07 PM Changeset in webkit [172566] by Lucas Forschler

• 9 edits in branches/safari-600.1-branch/Source

Merged r172526. <rdar://problem/18005074>

9:05 PM Changeset in webkit [172565] by Lucas Forschler

• 2 edits in branches/safari-600.1-branch/Source/WebInspectorUI

Merged r172525. <rdar://problem/17997324>

9:03 PM Changeset in webkit [172564] by Lucas Forschler

• 2 edits in branches/safari-600.1-branch/Source/WebInspectorUI

Merged r172515. <rdar://problem/17997717>

9:01 PM Changeset in webkit [172563] by Lucas Forschler

• 2 edits in branches/safari-600.1-branch/Source/WebKit2

Merged r172512. <rdar://problem/17982275>

8:59 PM Changeset in webkit [172562] by Lucas Forschler

• 3 edits in branches/safari-600.1-branch/Source/WebCore

Merged r172507. <rdar://problem/17979262>

8:57 PM Changeset in webkit [172561] by Lucas Forschler

• 5 edits in branches/safari-600.1-branch/Source

Merged r172501. <rdar://problem/17992795>

8:55 PM Changeset in webkit [172560] by Lucas Forschler

• 12 edits in branches/safari-600.1-branch/Source/WebKit2

Merged r172500. <rdar://problem/17986556>

8:53 PM Changeset in webkit [172559] by Lucas Forschler

• 2 edits in branches/safari-600.1-branch/Source/WebKit2

Merged r172498. <rdar://problem/17996339>

8:49 PM Changeset in webkit [172558] by Lucas Forschler

• 2 edits in branches/safari-600.1-branch/Source/WebKit2

Merged r172497. <rdar://problem/17994679>

8:47 PM Changeset in webkit [172557] by Lucas Forschler

• 2 edits in branches/safari-600.1-branch/Source/WebInspectorUI

Merged r172488. <rdar://problem/17994473>

8:45 PM Changeset in webkit [172556] by Lucas Forschler

• 3 edits in branches/safari-600.1-branch/Source/WebKit2

Merged r172484. <rdar://problem/17582099>

8:43 PM Changeset in webkit [172555] by Lucas Forschler

• 7 edits in branches/safari-600.1-branch/Source/WebKit2

Merged r172483. r<rdar://problem/17935736>

8:41 PM Changeset in webkit [172554] by Lucas Forschler

• 2 edits in branches/safari-600.1-branch/Source/WebKit2

Merged r172395. <rdar://problem/17837670>

8:38 PM Changeset in webkit [172553] by Lucas Forschler

• 9 edits in branches/safari-600.1-branch/Source

Merged r172383. <rdar://problem/17971057>

8:19 PM Changeset in webkit [172552] by Lucas Forschler

• 3 edits in branches/safari-600.1-branch/Source/WebKit2

Merged r172382. <rdar://problem/17935736>

8:12 PM Changeset in webkit [172551] by Lucas Forschler

• 3 edits in branches/safari-600.1-branch/Source/WebInspectorUI/UserInterface/Views

Merge patch for <rdar://problem/17999231>.

7:53 PM Changeset in webkit [172550] by commit-queue@webkit.org

• 2 edits in trunk/Source/JavaScriptCore

Add ARM64 support to CMake-based builds
​https://bugs.webkit.org/show_bug.cgi?id=135912

Patch by Akos Kiss <​akiss@inf.u-szeged.hu> on 2014-08-13
Reviewed by Gyuyoung Kim.

This patch ensures that CMake does not fail with Unknown CPU error when
building for ARM64.

• CMakeLists.txt:

7:52 PM Changeset in webkit [172549] by commit-queue@webkit.org

• 2 edits
  1 add in trunk/Tools

[EFL] Bump up and patch dependencies to allow building for aarch64
​https://bugs.webkit.org/show_bug.cgi?id=135885

Patch by Akos Kiss <​akiss@inf.u-szeged.hu> on 2014-08-13
Reviewed by Gyuyoung Kim.

The currently used versions of pixman, libffi, glib, glib-networking,
and fontconfig fail to configure for aarch64, so they are bumped up to
the latest stable version.

Gstreamer-plugins-base configures for aarch64 but incorrectly recogizes
ARM NEON assembly support, so it is patched with a change that has been
accepted by upstream.

• efl/jhbuild.modules:
• efl/patches/gst-prevent-neon-check-in-configure-from-passing-under-aarch64.patch: Added.

6:22 PM Changeset in webkit [172548] by commit-queue@webkit.org

• 8 edits in trunk/Source

Enable CSS_SCROLL_SNAP for iOS
​https://bugs.webkit.org/show_bug.cgi?id=135915

Turn on CSS_SCROLL_SNAP for iOS and the iOS simulator.

Patch by Wenson Hsieh <Wenson Hsieh> on 2014-08-13
Reviewed by Tim Horton.

• Configurations/FeatureDefines.xcconfig:

6:13 PM Changeset in webkit [172547] by jonowells@apple.com

• 1 edit
  2 adds in trunk/Source/WebInspectorUI

Web Inspector: eslint configuration should be stored as .eslintrc
​https://bugs.webkit.org/show_bug.cgi?id=135916

.eslintrc file added to allow for ESLint solutions, within editors and otherwise,
to be automatically configured based on the WebKit JavaScript coding style. The
initial file is incomplete.

Reviewed by Joseph Pecoraro.

• .eslintrc: Added.
• External/.eslintrc: Added.

5:39 PM Changeset in webkit [172546] by enrica@apple.com

• 2 edits in trunk/Source/WebKit2

[Services with UI] Use a longer delay duration for editable content.
​https://bugs.webkit.org/show_bug.cgi?id=135918
<rdar://problem/17998929>

Reviewed by Tim Horton.

Use a 1 second delay for selections in editable content.

• WebProcess/WebPage/mac/ServicesOverlayController.mm:

(WebKit::ServicesOverlayController::remainingTimeUntilHighlightShouldBeShown):

5:23 PM Changeset in webkit [172545] by Simon Fraser

• 2 edits
  1 add in trunk/LayoutTests

Land correct results for compositing/fixed-image-loading.html on WK1 and WK2.

• compositing/fixed-image-loading-expected.txt:
• platform/mac-wk2/compositing/fixed-image-loading-expected.txt: Added.

5:23 PM Changeset in webkit [172544] by timothy@apple.com

• 2 edits in trunk/Source/WebInspectorUI

Web Inspector: Missing semicolon after console.assert in TimelineSidebarPanel.js
​https://bugs.webkit.org/show_bug.cgi?id=135917

Reviewed by Joseph Pecoraro.

• UserInterface/Views/TimelineSidebarPanel.js:

5:16 PM Changeset in webkit [172543] by commit-queue@webkit.org

• 8 edits
  2 adds in trunk/Source/WebCore

Compute and store snap point positions
​https://bugs.webkit.org/show_bug.cgi?id=135268

Patch by Wenson Hsieh <Wenson Hsieh> on 2014-08-13
Reviewed by Simon Fraser.

Working towards implementing a way to extract snap positions (i.e. LayoutUnits) from the parsed snap point properties from CSS,
as well as a way of storing the information. Supports both mainframe and overflow scrolling.

No changes in behavior. I will write tests when I implement snapping for iOS and Mac.

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

(WebCore::FrameView::updateSnapOffsets): Invokes the static routine in AxisScrollSnapOffsets to update snap offsets for mainframe.

• page/FrameView.h:
• page/scrolling/AxisScrollSnapOffsets.cpp: Added.

(WebCore::appendAndSortChildSnapOffsets):
(WebCore::updateFromStyle):
(WebCore::updateSnapOffsetsForScrollableArea): Update the ScrollableArea's snap axes.

• page/scrolling/AxisScrollSnapOffsets.h: Added.
• platform/ScrollableArea.cpp:

(WebCore::ScrollableArea::setHorizontalSnapOffsets):
(WebCore::ScrollableArea::setVerticalSnapOffsets):
(WebCore::ScrollableArea::clearHorizontalSnapOffsets): Resets the unique_ptr for horizontal snap points.
(WebCore::ScrollableArea::clearVerticalSnapOffsets): Resets the unique_ptr for vertical snap points.

• platform/ScrollableArea.h:

(WebCore::ScrollableArea::horizontalSnapOffsets): Returns a pointer to the ScrollableArea's horizontal snap points.
(WebCore::ScrollableArea::verticalSnapOffsets): Returns a pointer to the ScrollableArea's vertical snap points.
(WebCore::ScrollableArea::updateSnapOffsets):

• rendering/RenderLayer.cpp:

(WebCore::RenderLayer::updateSnapOffsets): Invokes the static routine in AxisScrollSnapOffsets to update snap offsets for an overflow area.

• rendering/RenderLayer.h:

4:55 PM Changeset in webkit [172542] by mrowe@apple.com

• 9 edits in trunk

<​https://webkit.org/b/135911> WebKit should build on Yosemite with the public SDK.

Reviewed by Darin Adler.

Source/WebCore:

• rendering/RenderThemeMac.mm: Fix the forward-declaration of NSServicesRolloverButtonCell.

Source/WebKit/mac:

• Misc/WebSharingServicePickerController.mm: Forward-declare some details related to NSSharingServicePicker.

Source/WebKit2:

• Platform/IPC/mac/ImportanceAssertion.h: Forward-declare the new assertion functions we use.
• UIProcess/mac/WebContextMenuProxyMac.mm: Forward-declare some details related to NSSharingServicePicker.

Tools:

• DumpRenderTree/mac/TextInputController.m: Don't use extern "C" in a non-C++ file.

4:35 PM Changeset in webkit [172541] by ap@apple.com

• 2 edits in trunk/Source/WebKit2

iOS build fix.

• UIProcess/WebContext.cpp:

(WebKit::WebContext::ensureNetworkProcess):
(WebKit::WebContext::createNewWebProcess):

3:53 PM Changeset in webkit [172540] by achristensen@apple.com

• 17 edits in trunk

Progress towards CMake on Mac.
​https://bugs.webkit.org/show_bug.cgi?id=135819

Reviewed by Laszlo Gombos.

.:

• Source/cmake/OptionsMac.cmake:

Disable some more features temporarily to get CMake working.

• Source/cmake/WebKitMacros.cmake:

Added an extension parameter to GENERATE_BINDINGS to allow ObjC bindings to generate .mm files.

Source/JavaScriptCore:

• CMakeLists.txt:

Add the remote inspector headers to the forwarding headers list.

Source/WebCore:

• CMakeLists.txt:

Include derived sources directory, which is needed by Apple ports.
Separated svg and non-svg idls for ObjC bindings.
Don't build the new gamepad API right now. It's causing naming conflicts and it's not done.
Process CSSUnknownRule.idl.

• PlatformGTK.cmake:

Added cpp as the extension of the generated bindings.

• PlatformMac.cmake:

Include more directories and sources.
Add forwarding headers.
Generate ObjC bindings.

• bindings/objc/DOMUIKitExtensions.mm:

Moved config.h to recognize the PLATFORM macro.

• bindings/scripts/CodeGeneratorObjC.pm:

(ReadPublicInterfaces):
(GenerateInterface):
Added some workarounds for CMake.

• bindings/scripts/IDLAttributes.txt:

Added missing attribute names.

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

Added PLATFORM(IOS) to protect IOS_AIRPLAY code.

• platform/mac/PasteboardMac.mm:

Added necessary DRAG_SUPPORT protection.

• platform/text/mac/TextCodecMac.cpp:

Moved config.h to recognize the PLATFORM macro.

Tools:

• WebKitTestRunner/CMakeLists.txt:

Added new cpp parameter for the extension of the generated files.

3:30 PM Changeset in webkit [172539] by commit-queue@webkit.org

• 2 edits in trunk/Source/WebKit2

Opening Web Inspector causes a large amount of sandbox violations
​https://bugs.webkit.org/show_bug.cgi?id=135908

Patch by Joseph Pecoraro <Joseph Pecoraro> on 2014-08-13
Reviewed by Timothy Hatcher.

• WebProcess/com.apple.WebProcess.sb.in:

Permit the WebContent process to create file read extensions for the
system WebInspectorUI.framework which the Network process can already
read anyways.

3:15 PM Changeset in webkit [172538] by jer.noble@apple.com

• 2 edits in trunk/Source/WebCore

[MSE] Replacing existing samples with overlapping new samples will cause playback to stutter.
​https://bugs.webkit.org/show_bug.cgi?id=135902

Reviewed by Eric Carlson.

Only mark the TrackBuffer as needing re-enqueueing if the samples removed by overlapping samples
themselves overlap with potentially enqueued but not yet displayed frames.

• Modules/mediasource/SourceBuffer.cpp:

(WebCore::SourceBuffer::sourceBufferPrivateDidReceiveSample):

2:39 PM Changeset in webkit [172537] by Simon Fraser

• 2 edits in trunk/LayoutTests

Fix the test result for MountainLion and Mavericks, which don't composite fixed position.

• compositing/fixed-image-loading-expected.txt:

2:06 PM Changeset in webkit [172536] by akling@apple.com

• 20 edits in trunk/Source/WebCore

CSSValueList should never contain null values.
<​https://webkit.org/b/135887>

It's not legal for a CSSValueList to contain null CSSValues, so make
it store a Vector of Ref<CSSValue> internally instead.

This allows the compiler to avoid emitting a number of null checks.
It will also allow a future patch to make CSSValueList vend CSSValue&
instead of CSSValue*.

Most of the patch is fixing up things on their way into a CSSValueList
to not be null at the call site, now that append() and prepend() are
taking PassRef instead of PassRefPtr.

Reviewed by Sam Weinig.

• css/BasicShapeFunctions.cpp:

(WebCore::valueForBasicShape):

• css/BasicShapeFunctions.h:
• css/CSSBorderImage.cpp:

(WebCore::createBorderImageValue):

• css/CSSBorderImage.h:
• css/CSSComputedStyleDeclaration.cpp:

(WebCore::valueForImageSliceSide):
(WebCore::valueForNinePieceImageSlice):
(WebCore::valueForNinePieceImageQuad):
(WebCore::valueForNinePieceImage):
(WebCore::getBorderRadiusShorthandValue):
(WebCore::ComputedStyleExtractor::valueForShadow):
(WebCore::ComputedStyleExtractor::valueForFilter):
(WebCore::specifiedValueForGridTrackSize):
(WebCore::addValuesForNamedGridLinesAtIndex):
(WebCore::scrollSnapCoordinates):
(WebCore::createTimingFunctionValue):
(WebCore::fillSourceTypeToCSSValue):
(WebCore::contentToCSSValue):
(WebCore::shapePropertyValue):
(WebCore::ComputedStyleExtractor::propertyValue):
(WebCore::ComputedStyleExtractor::getCSSPropertyValuesForShorthandProperties):
(WebCore::ComputedStyleExtractor::getCSSPropertyValuesForSidesShorthand):
(WebCore::ComputedStyleExtractor::getCSSPropertyValuesForGridShorthand):
(WebCore::ComputedStyleExtractor::getBackgroundShorthandValue):

• css/CSSComputedStyleDeclaration.h:
• css/CSSFontSelector.cpp:

(WebCore::CSSFontSelector::addFontFaceRule):

• css/CSSParser.cpp:

(WebCore::createPrimitiveValuePair):
(WebCore::parseTranslateTransformValue):
(WebCore::CSSParser::createPrimitiveNumericValue):
(WebCore::CSSParser::createPrimitiveStringValue):
(WebCore::CSSParser::addFillValue):
(WebCore::CSSParser::parseNonElementSnapPoints):
(WebCore::CSSParser::parseScrollSnapDestination):
(WebCore::CSSParser::parseFillShorthand):
(WebCore::CSSParser::addAnimationValue):
(WebCore::CSSParser::parseAnimationShorthand):
(WebCore::CSSParser::parseTransitionShorthand):
(WebCore::CSSParser::parseQuotes):
(WebCore::CSSParser::parseContent):
(WebCore::CSSParser::parseFillProperty):
(WebCore::CSSParser::parseAnimationProperty):
(WebCore::CSSParser::parseGridPosition):
(WebCore::CSSParser::parseGridTemplateRowsAndAreas):
(WebCore::CSSParser::parseGridLineNames):
(WebCore::CSSParser::parseGridTrackList):
(WebCore::CSSParser::parseGridTrackRepeatFunction):
(WebCore::CSSParser::parseGridTrackSize):
(WebCore::CSSParser::parseBasicShapeAndOrBox):
(WebCore::CSSParser::parseFontVariant):
(WebCore::CSSParser::parseTransform):
(WebCore::CSSParser::parseBuiltinFilterArguments):
(WebCore::CSSParser::parseFilter):
(WebCore::CSSParser::parseTextEmphasisStyle):

• css/CSSParser.h:
• css/CSSProperty.cpp:

(WebCore::CSSProperty::wrapValueInCommaSeparatedList):

• css/CSSValueList.cpp:

(WebCore::CSSValueList::CSSValueList):
(WebCore::CSSValueList::removeAll):
(WebCore::CSSValueList::hasValue):
(WebCore::CSSValueList::copy):
(WebCore::CSSValueList::customCSSText):
(WebCore::CSSValueList::equals):
(WebCore::CSSValueList::addSubresourceStyleURLs):
(WebCore::CSSValueList::hasFailedOrCanceledSubresources):

• css/CSSValueList.h:

(WebCore::CSSValueList::item):
(WebCore::CSSValueList::itemWithoutBoundsCheck):
(WebCore::CSSValueList::append):
(WebCore::CSSValueList::prepend):

• css/SVGCSSComputedStyleDeclaration.cpp:

(WebCore::paintOrder):

• css/SVGCSSParser.cpp:

(WebCore::CSSParser::parsePaintOrder):

• editing/EditingStyle.cpp:

(WebCore::mergeTextDecorationValues):

• editing/EditorCommand.cpp:

(WebCore::executeToggleStyleInList):

• svg/SVGFontFaceSrcElement.cpp:

(WebCore::SVGFontFaceSrcElement::srcValue):

• svg/SVGLength.cpp:

(WebCore::SVGLength::toCSSPrimitiveValue):

• svg/SVGLength.h:

1:31 PM Changeset in webkit [172535] by achristensen@apple.com

• 1 edit
  2 deletes in trunk/Source/WebCore

Removed dead plugins code.
​https://bugs.webkit.org/show_bug.cgi?id=135862

Reviewed by Alexey Proskuryakov.

• plugins/PluginDataNone.cpp: Removed.
• plugins/mac: Removed.
• plugins/mac/PluginPackageMac.cpp: Removed.
• plugins/mac/PluginViewMac.mm: Removed.

1:21 PM Changeset in webkit [172534] by Simon Fraser

• 3 edits
  2 adds in trunk

[WK1] A fixed-position <img> may not show on page load
​https://bugs.webkit.org/show_bug.cgi?id=135893

Reviewed by Tim Horton.

Source/WebCore:

A position:fixed image could fail to display because we didn't make a
compositing layer for it under some circumstances. This can happen if
RenderLayerCompositor::requiresCompositingForPosition() runs before the
image is loaded, in which cause the layer is zero-sized, so does not get
composited.

When the image loads, there was no code path that ensured that compositing
would be re-evaluated (unlike size changes due to style updates). Fix by
having RenderLayer::contentChanged() also check for ImageChanged.

Test: compositing/fixed-image-loading.html

• rendering/RenderLayer.cpp:

(WebCore::RenderLayer::contentChanged):

LayoutTests:

Test with a position:fixed image. Note that the test has to force layout
early on to test the bug.

• compositing/fixed-image-loading-expected.txt: Added.
• compositing/fixed-image-loading.html: Added.

1:14 PM Changeset in webkit [172533] by commit-queue@webkit.org

• 2 edits in trunk/Source/WebInspectorUI

Web Inspector: Inline DataGrid inside details sidebar has double thick top border with legacy styles
​https://bugs.webkit.org/show_bug.cgi?id=135894

Patch by Joseph Pecoraro <Joseph Pecoraro> on 2014-08-13
Reviewed by Timothy Hatcher.

• UserInterface/Views/DetailsSection.css:

(body.mac-platform.legacy .details-section:not(.collapsed) .data-grid.inline):
In legacy styles the details-section header always has a 1px bottom border.
In all styles, an inline data grid had a 1px border everywhere. So, handle
this specific case by removing the top border of an inline data grid
that is inside an expanded details section.

12:43 PM Changeset in webkit [172532] by dbates@webkit.org

• 8 edits in trunk/Source

[iOS] Make JavaScriptCore and bmalloc build with the public SDK
​https://bugs.webkit.org/show_bug.cgi?id=135848

Reviewed by Geoffrey Garen.

Source/bmalloc:

• bmalloc/BPlatform.h: Added macro BPLATFORM_IOS_SIMULATOR, which evaluates to true

when building for the iOS Simulator.

• bmalloc/PerThread.h: Use pthread_machdep.h code path when building for iOS Simulator

using the public SDK.
(_pthread_setspecific_direct): Added; only defined when building for the iOS Simulator
using the public SDK.
(_pthread_getspecific_direct): Added; only defined when building for the iOS Simulator
using the public SDK.

Source/JavaScriptCore:

• API/JSBase.h: Declare NSMap functions with external linkage when building for iOS without the

header <Foundation/NSMapTablePriv.h>.

• inspector/remote/RemoteInspector.mm: Define XPC functions with external linkage when building

without the system header <xpc/xpc.h>.

• inspector/remote/RemoteInspectorXPCConnection.h: Define xpc_connection_t and xpc_object_t when building

without the system header <xpc/xpc.h>.

• inspector/remote/RemoteInspectorXPCConnection.mm: Declare XPC functions with external linkage when

building without without the system header <xpc/xpc.h>.
(Inspector::RemoteInspectorXPCConnection::closeOnQueue): Fix code style; use nullptr instead of NULL.
(Inspector::RemoteInspectorXPCConnection::sendMessage): Ditto.

12:25 PM Changeset in webkit [172531] by commit-queue@webkit.org

• 2 edits in trunk/Source/WebInspectorUI

Web Inspector: "No Filter Results" view only shows once, does not show again
​https://bugs.webkit.org/show_bug.cgi?id=135892

Patch by Joseph Pecoraro <Joseph Pecoraro> on 2014-08-13
Reviewed by Timothy Hatcher.

• UserInterface/Views/NavigationSidebarPanel.js:

(WebInspector.NavigationSidebarPanel.prototype.showEmptyContentPlaceholder):
We were checking at the wrong level for a parent. Check if the root
placeholder element has a parent or not.

12:12 PM Changeset in webkit [172530] by commit-queue@webkit.org

• 2 edits in trunk/Source/WebInspectorUI

Web Inspector: Incorrect border colors in event listeners sections in details sidebar
​https://bugs.webkit.org/show_bug.cgi?id=135891

Patch by Joseph Pecoraro <Joseph Pecoraro> on 2014-08-13
Reviewed by Timothy Hatcher.

• UserInterface/Views/DetailsSection.css:

(.details-section .details-section > .header):
(.details-section .details-section:not(.collapsed) > .header):
Remove box-shadow and replace with a bottom border.

(body.mac-platform.legacy .details-section .details-section.collapsed > .header):
(.details-section .details-section.collaspsed > .header): Deleted.
Collapsed was spelled incorrectly so this was not applying properly before.
But correct this for legacy because legacy still uses a box-shadow.

12:05 PM Changeset in webkit [172529] by Bem Jones-Bey

• 5 edits in trunk/Source/WebCore

Refactor ShapeOutsideInfo so it isn't mutated for each line
​https://bugs.webkit.org/show_bug.cgi?id=135781

Reviewed by Zoltan Horvath.

Encapsulate the per line state into a ShapeOutsideDeltas object.
updateDeltasForContainingBlockLine has been renamed to
computeDeltasForContainingBlockLine, and it returns an instance of the
ShapeOutsideDeltas object for that line. This object is cached, but
none of the functionality of ShapeOutsideInfo is dependant on any line
specific data anymore.

No new tests, no behavior change.

• rendering/FloatingObjects.cpp:

(WebCore::ComputeFloatOffsetForFloatLayoutAdapter<FloatingObject::FloatLeft>::updateOffsetIfNeeded):

Update to take a FloatingObject reference because the FloatingObject cannot be null.

(WebCore::ComputeFloatOffsetForFloatLayoutAdapter<FloatingObject::FloatRight>::updateOffsetIfNeeded):

Ditto.

(WebCore::ComputeFloatOffsetAdapter<FloatTypeValue>::collectIfNeeded): Pass FloatingObject to

updateOffsetIfNeeded as a reference, since it cannot be null.

(WebCore::ComputeFloatOffsetForLineLayoutAdapter<FloatingObject::FloatLeft>::updateOffsetIfNeeded):

Use ShapeOutsideDeltas object to calculate the offset and take the FloatingObject as a
reference.

(WebCore::ComputeFloatOffsetForLineLayoutAdapter<FloatingObject::FloatRight>::updateOffsetIfNeeded):

Ditto.

(WebCore::shapeInfoForFloat): Deleted.

• rendering/line/LineWidth.cpp:

(WebCore::LineWidth::shrinkAvailableWidthForNewFloatIfNeeded): Update to use ShapeOutsideDeltas object.

• rendering/shapes/ShapeOutsideInfo.cpp:

(WebCore::ShapeOutsideInfo::computeDeltasForContainingBlockLine): Return a ShaoeOutsideDeltas object

instead of storing line specific data in instance variables.

(WebCore::ShapeOutsideInfo::updateDeltasForContainingBlockLine): Deleted.

• rendering/shapes/ShapeOutsideInfo.h:

12:03 PM Changeset in webkit [172528] by timothy_horton@apple.com

• 3 edits in trunk/Source/WebKit2

Avoid making new active service overlay highlights while the mouse is down
​https://bugs.webkit.org/show_bug.cgi?id=135872
<rdar://problem/17982341>

Reviewed by Enrica Casucci.

• WebProcess/WebPage/ServicesOverlayController.h:
• WebProcess/WebPage/mac/ServicesOverlayController.mm:

(WebKit::ServicesOverlayController::remainingTimeUntilHighlightShouldBeShown):
(WebKit::ServicesOverlayController::mouseEvent):
If the mouse is pressed or it's been less than 200ms since the mouse went up,
don't allow the highlight to change. We apply the mouse-is-pressed rule to telephone
number highlights as well, unlike the rest of the hysteresis logic.

12:00 PM Changeset in webkit [172527] by commit-queue@webkit.org

• 3 edits in trunk/Source/WebInspectorUI

Web Inspector: Timeline Close buttons can use polish for new and legacy styles
​https://bugs.webkit.org/show_bug.cgi?id=135870

Patch by Joseph Pecoraro <Joseph Pecoraro> on 2014-08-13
Reviewed by Timothy Hatcher.

• UserInterface/Images/Legacy/Close.svg:

Ensure this has fill colors in tree selections.

• UserInterface/Views/TimelineSidebarPanel.css:

(.sidebar > .panel.navigation.timeline > .timelines-content .close-button):
(body.mac-platform.legacy .sidebar > .panel.navigation.timeline > .timelines-content .close-button):
(body.mac-platform.legacy .sidebar > .panel.navigation.timeline > .content .item > .status > .status-button):
Size the close-button for the top overviews differently in new styles (14x14) and legacy (12x12).

11:58 AM Changeset in webkit [172526] by timothy@apple.com

• 9 edits in trunk/Source

Web Inspector: Workaround a NSWindow change to the title bar.
​https://bugs.webkit.org/show_bug.cgi?id=135880

Reviewed by Joseph Pecoraro.

Source/WebInspectorUI:

• UserInterface/Base/Main.js:

(WebInspector._updateToolbarHeight): Don't call in legacy mode.

• UserInterface/Views/DashboardContainerView.css:

(.toolbar .dashboard-container):
(.toolbar .dashboard):
Drive-by fix some errant drop-shadows. The transparent border was pushing the inset top
drop shadow down when it shouldn't. Remove duplicate background and shadow.

• UserInterface/Views/Main.css:

(body): Set the background white, never needs to be transparent now.
(body.mac-platform.legacy:not(.docked)): Set background to transparent when not docked in legacy.
(body.docked.bottom): Update border color.
(body.docked.right): Ditto.
(body.docked.bottom.mac-platform.legacy): Keep old border color.
(body.docked.right.mac-platform.legacy): Ditto.

• UserInterface/Views/Toolbar.css:

(.toolbar): Set a default border and background gradient.
(body.window-inactive .toolbar): No gradient but use a solid color when inactive.
(body.mac-platform:not(.legacy):not(.docked) .toolbar): Padding on the top for the title bar.
(body.mac-platform:not(.legacy):not(.docked) .toolbar.icon-and-label-vertical): Account for title bar height.
(body.mac-platform:not(.legacy):not(.docked) .toolbar.icon-and-label-vertical.small-size): Ditto.
(body.mac-platform:not(.legacy):not(.docked) .toolbar.icon-only): Ditto.
(body.mac-platform:not(.legacy):not(.docked) .toolbar.label-only): Ditto.
(body.mac-platform.legacy .toolbar): No inset shadow in legacy when not docked.
(body.mac-platform.legacy.docked .toolbar): Restore legacy border color.
(body.mac-platform.legacy:not(.docked) .toolbar .item.button > .label): Only disable sub-pixel antialiasing on legacy!
(body.mac-platform.legacy.window-inactive .toolbar): Restore legacy border color.

Source/WebKit/mac:

• WebCoreSupport/WebInspectorClient.mm:

(WebInspectorFrontendClient::setToolbarHeight): Only call setContentBorderThickness: on 10.9 and earlier.
(-[WebInspectorWindowController init]): Only call setDrawsBackground: on 10.9 and earlier.
(-[WebInspectorWindowController window]): Set titlebarAppearsTransparent on 10.10.
Only call border thickness APIs on 10.9 and earlier.

Source/WebKit2:

• UIProcess/mac/WebInspectorProxyMac.mm:

(WebKit::WebInspectorProxy::createInspectorWindow): Set titlebarAppearsTransparent on 10.10.
Only call border thickness APIs on 10.9 and earlier.
(WebKit::WebInspectorProxy::platformCreateInspectorPage): Only call setDrawsBackground: on 10.9 and earlier.
(WebKit::WebInspectorProxy::platformSetToolbarHeight): Only call setContentBorderThickness: on 10.9 and earlier.

11:51 AM Changeset in webkit [172525] by timothy@apple.com

• 2 edits in trunk/Source/WebInspectorUI

Web Inspector: Restore Helvetica as the dashboard font in legacy mode
​https://bugs.webkit.org/show_bug.cgi?id=135886

Reviewed by Joseph Pecoraro.

• UserInterface/Views/DefaultDashboardView.css:

(body.mac-platform.legacy .toolbar .dashboard.default > .item):

11:00 AM Changeset in webkit [172524] by zoltan@webkit.org

• 4 edits
  2 adds in trunk

[CSS3-Text] Add rendering support for the none value of text-justify property
​https://bugs.webkit.org/show_bug.cgi?id=135349

Reviewed by David Hyatt.

Source/WebCore:

Spec: ​http://dev.w3.org/csswg/css-text-3/#valdef-text-justify.none

Test: fast/css3-text/css3-text-justify/text-justify-none.html

• rendering/RenderBlockLineLayout.cpp:

(WebCore::RenderBlockFlow::computeInlineDirectionPositionsForSegment):

LayoutTests:

• fast/css3-text/css3-text-justify/text-justify-none-expected.html: Added.
• fast/css3-text/css3-text-justify/text-justify-none.html: Added.
• platform/mac/TestExpectations:

10:24 AM Changeset in webkit [172523] by eric.carlson@apple.com

• 6 edits
  2 adds in trunk

Media document sends incorrect referer header
​https://bugs.webkit.org/show_bug.cgi?id=135839

Reviewed by Jer Noble.

Source/WebCore:

Test: http/tests/media/media-document-referer.html

• html/MediaDocument.cpp:

(WebCore::MediaDocumentParser::MediaDocumentParser): Stash referrer.
(WebCore::MediaDocumentParser::createDocumentStructure): Set outgoing referrer.
(WebCore::MediaDocument::MediaDocument): Stash referrer.

• html/MediaDocument.h:

LayoutTests:

• http/tests/media/media-document-referer-expected.txt: Added.
• http/tests/media/media-document-referer.html: Added.
• http/tests/media/resources/video-referer-check-referer.php: Take expected referer as a

parameter instead of hard coding it.

• http/tests/media/video-referer.html: Pass referer to loader script.

8:45 AM Changeset in webkit [172522] by mihnea@adobe.com

• 2 edits
  50 moves
  1 add in trunk/LayoutTests

[CSSRegions] Move scrolling tests under fast/regions/scrolling
​https://bugs.webkit.org/show_bug.cgi?id=135881

Reviewed by Andrei Bucur.

• fast/regions/scrolling/scroll-into-view-expected.html: Renamed from LayoutTests/fast/regions/scroll-into-view-expected.html.
• fast/regions/scrolling/scroll-into-view-relative-expected.html: Renamed from LayoutTests/fast/regions/scroll-into-view-relative-expected.html.
• fast/regions/scrolling/scroll-into-view-relative.html: Renamed from LayoutTests/fast/regions/scroll-into-view-relative.html.
• fast/regions/scrolling/scroll-into-view.html: Renamed from LayoutTests/fast/regions/scroll-into-view.html.
• fast/regions/scrolling/scrollable-first-region-expected.html: Renamed from LayoutTests/fast/regions/scrollable-first-region-expected.html.
• fast/regions/scrolling/scrollable-first-region.html: Renamed from LayoutTests/fast/regions/scrollable-first-region.html.
• fast/regions/scrolling/scrollable-last-region-expected.html: Renamed from LayoutTests/fast/regions/scrollable-last-region-expected.html.
• fast/regions/scrolling/scrollable-last-region.html: Renamed from LayoutTests/fast/regions/scrollable-last-region.html.
• fast/regions/scrolling/scrollable-paragraph-unsplittable-expected.html: Renamed from LayoutTests/fast/regions/scrollable-paragraph-unsplittable-expected.html.
• fast/regions/scrolling/scrollable-paragraph-unsplittable.html: Renamed from LayoutTests/fast/regions/scrollable-paragraph-unsplittable.html.
• fast/regions/scrolling/scrollable-region-content-size-expected.html: Renamed from LayoutTests/fast/regions/scrollable-region-content-size-expected.html.
• fast/regions/scrolling/scrollable-region-content-size-horiz-bt-expected.html: Renamed from LayoutTests/fast/regions/scrollable-region-content-size-horiz-bt-expected.html.
• fast/regions/scrolling/scrollable-region-content-size-horiz-bt.html: Renamed from LayoutTests/fast/regions/scrollable-region-content-size-horiz-bt.html.
• fast/regions/scrolling/scrollable-region-content-size-no-overflow-expected.html: Renamed from LayoutTests/fast/regions/scrollable-region-content-size-no-overflow-expected.html.
• fast/regions/scrolling/scrollable-region-content-size-no-overflow-horiz-bt-expected.html: Renamed from LayoutTests/fast/regions/scrollable-region-content-size-no-overflow-horiz-bt-expected.html.
• fast/regions/scrolling/scrollable-region-content-size-no-overflow-horiz-bt.html: Renamed from LayoutTests/fast/regions/scrollable-region-content-size-no-overflow-horiz-bt.html.
• fast/regions/scrolling/scrollable-region-content-size-no-overflow-vert-rl-expected.html: Renamed from LayoutTests/fast/regions/scrollable-region-content-size-no-overflow-vert-rl-expected.html.
• fast/regions/scrolling/scrollable-region-content-size-no-overflow-vert-rl.html: Renamed from LayoutTests/fast/regions/scrollable-region-content-size-no-overflow-vert-rl.html.
• fast/regions/scrolling/scrollable-region-content-size-no-overflow.html: Renamed from LayoutTests/fast/regions/scrollable-region-content-size-no-overflow.html.
• fast/regions/scrolling/scrollable-region-content-size-relative-expected.html: Renamed from LayoutTests/fast/regions/scrollable-region-content-size-relative-expected.html.
• fast/regions/scrolling/scrollable-region-content-size-relative.html: Renamed from LayoutTests/fast/regions/scrollable-region-content-size-relative.html.
• fast/regions/scrolling/scrollable-region-content-size-vert-rl-expected.html: Renamed from LayoutTests/fast/regions/scrollable-region-content-size-vert-rl-expected.html.
• fast/regions/scrolling/scrollable-region-content-size-vert-rl.html: Renamed from LayoutTests/fast/regions/scrollable-region-content-size-vert-rl.html.
• fast/regions/scrolling/scrollable-region-content-size-vertical-lr-expected.html: Renamed from LayoutTests/fast/regions/scrollable-region-content-size-vertical-lr-expected.html.
• fast/regions/scrolling/scrollable-region-content-size-vertical-lr.html: Renamed from LayoutTests/fast/regions/scrollable-region-content-size-vertical-lr.html.
• fast/regions/scrolling/scrollable-region-content-size.html: Renamed from LayoutTests/fast/regions/scrollable-region-content-size.html.
• fast/regions/scrolling/scrollable-region-scrollable-absolute-content-background-expected.html: Renamed from LayoutTests/fast/regions/scrollable-region-scrollable-absolute-content-background-expected.html.
• fast/regions/scrolling/scrollable-region-scrollable-absolute-content-background.html: Renamed from LayoutTests/fast/regions/scrollable-region-scrollable-absolute-content-background.html.
• fast/regions/scrolling/scrollable-region-scrollable-content-background-expected.html: Renamed from LayoutTests/fast/regions/scrollable-region-scrollable-content-background-expected.html.
• fast/regions/scrolling/scrollable-region-scrollable-content-background.html: Renamed from LayoutTests/fast/regions/scrollable-region-scrollable-content-background.html.
• fast/regions/scrolling/scrollable-region-scrollable-content-expected.html: Renamed from LayoutTests/fast/regions/scrollable-region-scrollable-content-expected.html.
• fast/regions/scrolling/scrollable-region-scrollable-content.html: Renamed from LayoutTests/fast/regions/scrollable-region-scrollable-content.html.
• fast/regions/scrolling/scrollable-regions-hit-testing-inside-expected.html: Renamed from LayoutTests/fast/regions/scrollable-regions-hit-testing-inside-expected.html.
• fast/regions/scrolling/scrollable-regions-hit-testing-inside.html: Renamed from LayoutTests/fast/regions/scrollable-regions-hit-testing-inside.html.
• fast/regions/scrolling/scrollable-single-region-bt-expected.html: Renamed from LayoutTests/fast/regions/scrollable-single-region-bt-expected.html.
• fast/regions/scrolling/scrollable-single-region-bt.html: Renamed from LayoutTests/fast/regions/scrollable-single-region-bt.html.
• fast/regions/scrolling/scrollable-single-region-expected.html: Renamed from LayoutTests/fast/regions/scrollable-single-region-expected.html.
• fast/regions/scrolling/scrollable-single-region-lr-expected.html: Renamed from LayoutTests/fast/regions/scrollable-single-region-lr-expected.html.
• fast/regions/scrolling/scrollable-single-region-lr.html: Renamed from LayoutTests/fast/regions/scrollable-single-region-lr.html.
• fast/regions/scrolling/scrollable-single-region-relative-element-expected.html: Renamed from LayoutTests/fast/regions/scrollable-single-region-relative-element-expected.html.
• fast/regions/scrolling/scrollable-single-region-relative-element.html: Renamed from LayoutTests/fast/regions/scrollable-single-region-relative-element.html.
• fast/regions/scrolling/scrollable-single-region-rl-expected.html: Renamed from LayoutTests/fast/regions/scrollable-single-region-rl-expected.html.
• fast/regions/scrolling/scrollable-single-region-rl.html: Renamed from LayoutTests/fast/regions/scrollable-single-region-rl.html.
• fast/regions/scrolling/scrollable-single-region.html: Renamed from LayoutTests/fast/regions/scrollable-single-region.html.
• fast/regions/scrolling/wheel-scroll-abspos-expected.html: Renamed from LayoutTests/fast/regions/wheel-scroll-abspos-expected.html.
• fast/regions/scrolling/wheel-scroll-abspos.html: Renamed from LayoutTests/fast/regions/wheel-scroll-abspos.html.
• fast/regions/scrolling/wheel-scroll-expected.html: Renamed from LayoutTests/fast/regions/wheel-scroll-expected.html.
• fast/regions/scrolling/wheel-scroll.html: Renamed from LayoutTests/fast/regions/wheel-scroll.html.
• fast/regions/scrolling/wheel-scrollable-single-region-expected.html: Renamed from LayoutTests/fast/regions/wheel-scrollable-single-region-expected.html.
• fast/regions/scrolling/wheel-scrollable-single-region.html: Renamed from LayoutTests/fast/regions/wheel-scrollable-single-region.html.
• platform/gtk/TestExpectations: Fix path for previously moved tests.

7:31 AM Changeset in webkit [172521] by commit-queue@webkit.org

• 2 edits
  2 adds in trunk/Tools

Make check-webkit-style run sort-export-file
​https://bugs.webkit.org/show_bug.cgi?id=135877

Patch by Renato Nagy <​nagy.renato@stud.u-szeged.hu> on 2014-08-13
Reviewed by Csaba Osztrogonác.

• Scripts/webkitpy/style/checker.py:

(_all_categories):
(FileType):
(CheckerDispatcher._file_type):
(CheckerDispatcher._create_checker):

• Scripts/webkitpy/style/checkers/exportfile.py: Added.

(ExportFileChecker):
(ExportFileChecker.init):
(ExportFileChecker.check):

• Scripts/webkitpy/style/checkers/exportfile_unittest.py: Added.

(handle_style_error):
(MockErrorHandler):
(MockErrorHandler.init):
(MockErrorHandler.turn_off_line_filtering):
(MockErrorHandler.call):
(ExportFileTest):
(ExportFileTest.setUp):
(ExportFileTest.tearDown):
(ExportFileTest.test_sorted):
(ExportFileTest.test_non_sorted):
(ExportFileTest.test_parse_error):

6:12 AM Changeset in webkit [172520] by ryuan.choi@samsung.com

• 7 edits in trunk/Source/WebKit2

[EFL] Add API to set application name for the user agent
​https://bugs.webkit.org/show_bug.cgi?id=135640

Reviewed by Gyuyoung Kim.

Add ewk_application_name_for_user_agent_set to make default user agent string
with application name which application passes.
It's useful for application to set only application information without knowledge
of legacy user agent components.

• UIProcess/API/efl/EwkView.cpp:

(EwkView::EwkView):
(EwkView::setApplicationNameForUserAgent):

• UIProcess/API/efl/EwkView.h:

(EwkView::applicationNameForUserAgent):

• UIProcess/API/efl/ewk_view.cpp:

(ewk_view_application_name_for_user_agent_set):
(ewk_view_application_name_for_user_agent_get):

• UIProcess/API/efl/ewk_view.h:
• UIProcess/API/efl/tests/test_ewk2_view.cpp:

(TEST_F):

• UIProcess/efl/WebPageProxyEfl.cpp:

(WebKit::WebPageProxy::standardUserAgent):

5:10 AM Changeset in webkit [172519] by jinwoo7.song@samsung.com

• 2 edits in trunk/Tools

[EFL] Sets the persistent storage for cookies in MiniBrowser
​https://bugs.webkit.org/show_bug.cgi?id=135875

Reviewed by Gyuyoung Kim.

In case of EWK_COOKIE_ACCEPT_POLICY_ALWAYS, sets the persistent storage
for non-session cookies can be stored persistently.

• MiniBrowser/efl/main.c:

(elm_main):

4:41 AM EFLWebKit edited by gyuyoung.kim@samsung.com

(diff)

4:38 AM EFLWebKitBuildBots edited by gyuyoung.kim@samsung.com

(diff)

4:36 AM EFLWebKitBuildBots edited by gyuyoung.kim@samsung.com

(diff)

4:20 AM Changeset in webkit [172518] by Antti Koivisto

• 3 edits in trunk/LayoutTests

Don't recurse into non-rendered subtrees when computing style
​https://bugs.webkit.org/show_bug.cgi?id=135844

Reviewed by Andreas Kling.

Rebase.

• http/tests/security/video-poster-cross-origin-crash-expected.txt:
• platform/mac/fast/ruby/ruby-base-merge-block-children-crash-2-expected.txt:

3:13 AM Changeset in webkit [172517] by Antti Koivisto

• 2 edits in trunk/Source/WebCore

Don't recurse into non-rendered subtrees when computing style
​https://bugs.webkit.org/show_bug.cgi?id=135844

Reviewed by Andreas Kling.

• style/StyleResolveTree.cpp:

(WebCore::Style::resetStyleForNonRenderedDescendants):

Do a simple reset of the style dirty bits and any computed style in non-renderer subtrees.

(WebCore::Style::attachRenderTree):
(WebCore::Style::resolveTree):

Don't recurse into descendants if the element does not create a renderer.
We didn't compute style anyway in such subtrees.