WebKit-https.git
5 years agoTry to fix the Windows build after r174315.
timothy_horton@apple.com [Sat, 4 Oct 2014 19:55:49 +0000 (19:55 +0000)]
Try to fix the Windows build after r174315.

* WebKit.vcxproj/WebKitExportGenerator/WebKitExports.def.in:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@174321 268f45cc-cd09-0410-ab3c-d52691b4dbfc

5 years agoTry to fix the Windows build after r174315.
timothy_horton@apple.com [Sat, 4 Oct 2014 19:35:06 +0000 (19:35 +0000)]
Try to fix the Windows build after r174315.

* WebKit.vcxproj/WebKitExportGenerator/WebKitExports.def.in:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@174320 268f45cc-cd09-0410-ab3c-d52691b4dbfc

5 years agoWeb Inspector: timelines should not count time elapsed while paused in the debugger
burg@cs.washington.edu [Sat, 4 Oct 2014 19:18:38 +0000 (19:18 +0000)]
Web Inspector: timelines should not count time elapsed while paused in the debugger
https://bugs.webkit.org/show_bug.cgi?id=136351

Reviewed by Timothy Hatcher.

Source/JavaScriptCore:

Now that we have a stopwatch to provide pause-aware timing data, we can remove the
profiler's handling of debugger pause/continue callbacks. The timeline agent accounts
for debugger pauses by pausing and resuming the stopwatch.

* API/JSProfilerPrivate.cpp:
(JSStartProfiling): Use a fresh stopwatch when profiling from the JSC API.
* inspector/ScriptDebugServer.cpp:
(Inspector::ScriptDebugServer::handlePause):
* profiler/LegacyProfiler.cpp:
(JSC::LegacyProfiler::profiler): Use nullptr.
(JSC::LegacyProfiler::startProfiling): Hand off a stopwatch to the profile generator.
(JSC::LegacyProfiler::stopProfiling): Use nullptr.
(JSC::LegacyProfiler::didPause): Deleted.
(JSC::LegacyProfiler::didContinue): Deleted.
* profiler/LegacyProfiler.h:
* profiler/ProfileGenerator.cpp: Remove debugger pause/continue callbacks and the
timestamp member that was used to track time elapsed by the debugger. Just use the
stopwatch's elapsed times to generate start/elapsed times for function calls.
(JSC::ProfileGenerator::create):
(JSC::ProfileGenerator::ProfileGenerator):
(JSC::ProfileGenerator::beginCallEntry):
(JSC::ProfileGenerator::endCallEntry):
(JSC::ProfileGenerator::didPause): Deleted.
(JSC::ProfileGenerator::didContinue): Deleted.
* profiler/ProfileGenerator.h:

Source/WebCore:

To avoid counting time elapsed while the debugger is paused, timeline records should
keep track of time elapsed since the start of timeline capturing, rather than wall clock
timestamps. We can easily compute elapsed time by sharing Stopwatch instance among
all timeline record-generating code. The stopwatch is paused while the debugger is paused,
so subsequent time measurements will not include time elapsed while the debugger is paused.

Agents use the shared stopwatch to generate timestamps if the timeline agent is active
(i.e., a timeline recording is being captured). If not, use a zero timestamp since the timing data is only revealed through the Timeline interface.

This refactoring is safe because start and end times are only used to graph records; the
timestamp's actual value is irrelevant and is not displayed in the user interface. Date
timestamps are still included with network-related records as part of their header data.

No new tests, because we cannot reliably test timing changes induced by debugger pauses.
It is possible for records to accrue time before the debugger pauses or after it resumes.

* inspector/InspectorCSSAgent.cpp: Remove unused include.
* inspector/InspectorPageAgent.cpp: Use timestamps from the shared stopwatch.
(WebCore::InspectorPageAgent::timestamp):
(WebCore::InspectorPageAgent::domContentEventFired):
(WebCore::InspectorPageAgent::loadEventFired):
* inspector/InspectorPageAgent.h:
* inspector/InspectorResourceAgent.cpp: Use timestamps from the shared stopwatch.
(WebCore::InspectorResourceAgent::timestamp):
(WebCore::InspectorResourceAgent::willSendRequest):
(WebCore::InspectorResourceAgent::didReceiveResponse):
(WebCore::InspectorResourceAgent::didReceiveData):
(WebCore::InspectorResourceAgent::didFinishLoading):
(WebCore::InspectorResourceAgent::didFailLoading):
(WebCore::InspectorResourceAgent::didLoadResourceFromMemoryCache):
(WebCore::InspectorResourceAgent::willSendWebSocketHandshakeRequest):
(WebCore::InspectorResourceAgent::didReceiveWebSocketHandshakeResponse):
(WebCore::InspectorResourceAgent::didCloseWebSocket):
(WebCore::InspectorResourceAgent::didReceiveWebSocketFrame):
(WebCore::InspectorResourceAgent::didSendWebSocketFrame):
(WebCore::InspectorResourceAgent::didReceiveWebSocketFrameError):
* inspector/InspectorResourceAgent.h:
* inspector/InspectorTimelineAgent.cpp: Add calls to reset, start, and stop the stopwatch.
(WebCore::InspectorTimelineAgent::didCreateFrontendAndBackend):
(WebCore::InspectorTimelineAgent::internalStart):
(WebCore::InspectorTimelineAgent::internalStop):
(WebCore::startProfiling):
(WebCore::InspectorTimelineAgent::startFromConsole):
(WebCore::InspectorTimelineAgent::willCallFunction):
(WebCore::InspectorTimelineAgent::willEvaluateScript):
(WebCore::InspectorTimelineAgent::didPause):
(WebCore::InspectorTimelineAgent::didContinue):
(WebCore::InspectorTimelineAgent::InspectorTimelineAgent):
(WebCore::InspectorTimelineAgent::timestamp):
(WebCore::TimelineTimeConverter::reset): Deleted.
* inspector/InspectorTimelineAgent.h: Make timestamp() public, and remove old timepieces.
(WebCore::TimelineTimeConverter::TimelineTimeConverter): Deleted.
(WebCore::TimelineTimeConverter::fromMonotonicallyIncreasingTime): Deleted.
(WebCore::InspectorTimelineAgent::timeConverter): Deleted.
* inspector/TimelineRecordFactory.cpp:

Source/WebInspectorUI:

Don't update the timeline's current time when the debugger is paused.

Start and end times for timeline records are now in seconds elapsed since timeline
recording started, rather than milliseconds since the epoch. Add a workaround to
preserve compatibility with old backends.

* UserInterface/Controllers/TimelineManager.js:
(WebInspector.TimelineManager.prototype.capturingStarted):
(WebInspector.TimelineManager.prototype.eventRecorded.processRecord):
(WebInspector.TimelineManager.prototype.eventRecorded):
* UserInterface/Views/TimelineContentView.js:
(WebInspector.TimelineContentView.prototype._debuggerPaused):
(WebInspector.TimelineContentView.prototype._debuggerResumed):

Source/WTF:

* WTF.vcxproj/WTF.vcxproj:
* WTF.vcxproj/WTF.vcxproj.filters:
* WTF.xcodeproj/project.pbxproj:
* wtf/CMakeLists.txt:
* wtf/Stopwatch.h: Added. This implements a refcounted monotonic stopwatch.
(WTF::Stopwatch::reset):
(WTF::Stopwatch::start):
(WTF::Stopwatch::stop):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@174319 268f45cc-cd09-0410-ab3c-d52691b4dbfc

5 years agoFTL should sink PutLocals
fpizlo@apple.com [Sat, 4 Oct 2014 17:18:25 +0000 (17:18 +0000)]
FTL should sink PutLocals
https://bugs.webkit.org/show_bug.cgi?id=137168

Reviewed by Oliver Hunt.
Source/JavaScriptCore:

We've known for a while that our PutLocal situation was sub-optimal. We emit them anytime we
"pass" arguments to an inlined function call, because we need to enable the runtime to grab
those arguments when doing foo.arguments where foo is inlined: our engine doesn't deoptimize
in that case but rather just relies on the arguments being flushed (i.e. a copy of their
values is spilled) at a well-known place in a well-known format.

The PutLocals incur two costs: (1) they are store instructions and stores ain't free, and (2)
they look like escaping sites and so they inhibit object allocation sinking.

But in most cases, the PutLocals are unnecessary because the inlined code never performs any
side effect that could transitively lead to function.arguments. Even if the inlined code
could do such a side effect, it may be on a rare path so there is no need to penalize the
entire function.

This patch implements one solution to the PutLocal problem: it aggressively sinks PutLocals
to the latest possible point. This is even more aggressive than the object allocation
sinking. That sinking algorithm avoids creating situations where an object could be
materialized more than one along any path. PutLocal sinking, on the other hand, doesn't avoid
this at all - both to make the phase cheaper and simpler and to make it more aggressive.
Every PutLocal is sunk no matter what.

The upside of this patch is that it eliminates many PutLocals: many of them are sunk "past
their death", thus eliminating them completely. Others are sunk to rare paths. This enables a
lot of object allocation sinking and it removes a lot of pointless store instructions.

It also has downsites. Sinking PutLocals increases register pressure because it increases the
live ranges of things like inlined arguments.

This patch is a net performance win in its current form: 1% SunSpider regression, 2% OctaneV2
progression, 0.6% Kraken regression, 1% AsmBench progression, and 0.5% CompressionBench
regression. The biggest win is on Octane/raytrace, which improves by 27%.

Relanding after fixing internal builds. We have to be careful about implicit casts from int64
to int32.

* CMakeLists.txt:
* JavaScriptCore.vcxproj/JavaScriptCore.vcxproj:
* JavaScriptCore.xcodeproj/project.pbxproj:
* bytecode/CodeBlock.h:
* bytecode/Operands.h:
(JSC::Operands::dump): Deleted.
* bytecode/OperandsInlines.h:
(JSC::Traits>::dump):
* bytecode/VirtualRegister.h:
(JSC::VirtualRegister::isHeader):
* dfg/DFGByteCodeParser.cpp:
(JSC::DFG::ByteCodeParser::InlineStackEntry::InlineStackEntry):
* dfg/DFGClobberSet.h:
(JSC::DFG::ClobberSetAdd::operator()):
(JSC::DFG::ClobberSetOverlaps::operator()):
* dfg/DFGClobberize.h:
(JSC::DFG::clobberize):
(JSC::DFG::NoOpClobberize::operator()):
(JSC::DFG::CheckClobberize::operator()):
(JSC::DFG::AbstractHeapOverlaps::operator()):
(JSC::DFG::ReadMethodClobberize::operator()):
(JSC::DFG::WriteMethodClobberize::operator()):
(JSC::DFG::DefMethodClobberize::operator()):
* dfg/DFGFlushFormat.h:
(JSC::DFG::merge):
* dfg/DFGGraph.cpp:
(JSC::DFG::Graph::Graph):
* dfg/DFGGraph.h:
(JSC::DFG::Graph::capturedVarsFor):
* dfg/DFGObjectAllocationSinkingPhase.cpp:
(JSC::DFG::ObjectAllocationSinkingPhase::determineMaterializationPoints):
(JSC::DFG::ObjectAllocationSinkingPhase::placeMaterializationPoints):
* dfg/DFGPlan.cpp:
(JSC::DFG::Plan::compileInThreadImpl):
* dfg/DFGPreciseLocalClobberize.h: Added.
(JSC::DFG::PreciseLocalClobberizeAdaptor::PreciseLocalClobberizeAdaptor):
(JSC::DFG::PreciseLocalClobberizeAdaptor::read):
(JSC::DFG::PreciseLocalClobberizeAdaptor::write):
(JSC::DFG::PreciseLocalClobberizeAdaptor::def):
(JSC::DFG::PreciseLocalClobberizeAdaptor::callIfAppropriate):
(JSC::DFG::PreciseLocalClobberizeAdaptor::readTop):
(JSC::DFG::PreciseLocalClobberizeAdaptor::writeTop):
(JSC::DFG::forEachLocalReadByUnwind):
(JSC::DFG::preciseLocalClobberize):
* dfg/DFGPutLocalSinkingPhase.cpp: Added.
(JSC::DFG::performPutLocalSinking):
* dfg/DFGPutLocalSinkingPhase.h: Added.
* dfg/DFGSSACalculator.h:
(JSC::DFG::SSACalculator::computePhis):
* dfg/DFGValidate.cpp:

Source/WTF:

Make the set bits of a BitVector iterable.

* wtf/BitVector.h:
(WTF::BitVector::SetBitsIterable::SetBitsIterable):
(WTF::BitVector::SetBitsIterable::iterator::iterator):
(WTF::BitVector::SetBitsIterable::iterator::operator*):
(WTF::BitVector::SetBitsIterable::iterator::operator++):
(WTF::BitVector::SetBitsIterable::iterator::operator==):
(WTF::BitVector::SetBitsIterable::iterator::operator!=):
(WTF::BitVector::SetBitsIterable::begin):
(WTF::BitVector::SetBitsIterable::end):
(WTF::BitVector::setBits):

LayoutTests:

* js/regress/elidable-new-object-then-call-expected.txt: Added.
* js/regress/elidable-new-object-then-call.html: Added.
* js/regress/script-tests/elidable-new-object-then-call.js: Added.
(sumOfArithSeries):
(bar):
(foo):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@174318 268f45cc-cd09-0410-ab3c-d52691b4dbfc

5 years ago[jhbuild] Simplify platform detection in update-webkit-libs-jhbuild
ossy@webkit.org [Sat, 4 Oct 2014 10:04:47 +0000 (10:04 +0000)]
[jhbuild] Simplify platform detection in update-webkit-libs-jhbuild
https://bugs.webkit.org/show_bug.cgi?id=137390

Reviewed by Carlos Garcia Campos.

* Scripts/update-webkit-libs-jhbuild:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@174317 268f45cc-cd09-0410-ab3c-d52691b4dbfc

5 years agoUnreviewed. Add missing inspector files to compilation.
carlosgc@webkit.org [Sat, 4 Oct 2014 09:07:36 +0000 (09:07 +0000)]
Unreviewed. Add missing inspector files to compilation.

* PlatformGTK.cmake: Add javascript sources in
WebInspectorUI/UserInterface/External/Esprima to the list of
inspector files.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@174316 268f45cc-cd09-0410-ab3c-d52691b4dbfc

5 years agoMake it possible to test page overlays
timothy_horton@apple.com [Sat, 4 Oct 2014 08:47:04 +0000 (08:47 +0000)]
Make it possible to test page overlays
https://bugs.webkit.org/show_bug.cgi?id=137339

Reviewed by Dan Bernstein.

Tests: pageoverlay/overlay-installation.html
       pageoverlay/overlay-large-document-scrolled.html
       pageoverlay/overlay-large-document.html

Do some initial work to make page overlays testable.
There are many more tests to be written, but this lays the groundwork.

* WebCore.exp.in:
* WebCore.xcodeproj/project.pbxproj:
* WebCore.vcxproj/WebCoreTestSupport.vcxproj:
* WebCore.vcxproj/WebCoreTestSupport.vcxproj.filters:
* CMakeLists.txt:
Add MockPageOverlayClient.

* platform/graphics/GraphicsLayer.h:
* page/PageOverlayController.h:
* page/PageOverlayController.cpp:
(WebCore::PageOverlayController::shouldSkipLayerInDump):
Don't skip PageOverlayController-owned layers in the layer tree dump if the
layerTreeAsText behavior says that we should include them.

* platform/graphics/GraphicsLayer.cpp:
(WebCore::dumpChildren):

* platform/graphics/GraphicsLayerClient.h:
(WebCore::GraphicsLayerClient::shouldSkipLayerInDump):
* rendering/RenderLayerBacking.cpp:
(WebCore::RenderLayerBacking::shouldSkipLayerInDump):
* rendering/RenderLayerBacking.h:
Pass the LayerTreeAsTextBehavior through shouldSkipLayerInDump.

* testing/Internals.cpp:
(WebCore::Internals::resetToConsistentState):
Uninstall all page overlays when resetting to a consistent state.

(WebCore::Internals::installMockPageOverlay):
(WebCore::Internals::pageOverlayLayerTreeAsText):
Expose two new internals functions.
installMockPageOverlay takes "view" or "document", allowing script
to choose the appropriate overlay type.
pageOverlayLayerTreeAsText dumps both view and document-relative
overlays, from each overlay container layer.

* testing/Internals.h:
* testing/Internals.idl:
Fix some typos.

* testing/MockPageOverlayClient.cpp: Added.
(WebCore::MockPageOverlayClient::shared):
(WebCore::MockPageOverlayClient::MockPageOverlayClient):
(WebCore::MockPageOverlayClient::installOverlay):
(WebCore::MockPageOverlayClient::uninstallAllOverlays):
(WebCore::MockPageOverlayClient::layerTreeAsText):
(WebCore::MockPageOverlayClient::pageOverlayDestroyed):
(WebCore::MockPageOverlayClient::willMoveToPage):
(WebCore::MockPageOverlayClient::didMoveToPage):
When the overlay is parented, mark it as needing repaint.
Future tests might want to test overlays without content,
but for now this is a reasonable default behavior.

(WebCore::MockPageOverlayClient::drawRect):
Paint the overlay rects as described in the LayoutTests ChangeLog.

(WebCore::MockPageOverlayClient::mouseEvent):
(WebCore::MockPageOverlayClient::didScrollFrame):
(WebCore::MockPageOverlayClient::copyAccessibilityAttributeStringValueForPoint):
(WebCore::MockPageOverlayClient::copyAccessibilityAttributeBoolValueForPoint):
(WebCore::MockPageOverlayClient::copyAccessibilityAttributeNames):

* testing/MockPageOverlayClient.h: Added.

* pageoverlay/overlay-installation.html: Added.
Add a test which just installs one view-relative and one document-relative
overlay, and dumps the layer tree and pixel results. The mock overlay
strokes a blue rectangle (inset a little from the overlay bounds) for
view-relative overlays, and a green rectangle (inset a bit more) for
document-relative overlays.

* pageoverlay/overlay-large-document.html: Added.
* pageoverlay/overlay-large-document-scrolled.html: Added.
Tests that do the same as the -installation test, but with a larger document;
the view-relative overlay should be the same size (or slightly smaller,
because we have scrollbars now), but the document-relative overlay should
be the size of the document. The second test scrolls to maxX, maxY, and
the document-relative overlay should scroll along with it, while the
view-relative overlay should remain fixed.

* pageoverlay/overlay-installation-expected.txt: Added.
* pageoverlay/overlay-large-document-expected.txt: Added.
* pageoverlay/overlay-large-document-scrolled-expected.txt: Added.
"Normal" layer tree expectations.

* platform/mac-wk1/pageoverlay/overlay-installation-expected.png: Added.
* platform/mac-wk1/pageoverlay/overlay-installation-expected.txt: Added.
* platform/mac-wk1/pageoverlay/overlay-large-document-expected.png: Added.
* platform/mac-wk1/pageoverlay/overlay-large-document-expected.txt: Added.
* platform/mac-wk1/pageoverlay/overlay-large-document-scrolled-expected.png: Added.
* platform/mac-wk1/pageoverlay/overlay-large-document-scrolled-expected.txt: Added.
Expected results for WebKit1 differ for view-relative overlays, because
we don't currently hook up view-relative overlays for WebKit1.

* platform/mac/pageoverlay/overlay-installation-expected.png: Added.
* platform/mac/pageoverlay/overlay-large-document-expected.png: Added.
* platform/mac/pageoverlay/overlay-large-document-scrolled-expected.png: Added.
"Normal" pixel dump expectations.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@174315 268f45cc-cd09-0410-ab3c-d52691b4dbfc

5 years agoMove DragController::createFragmentFromDragData implementation to the editor
carlosgc@webkit.org [Sat, 4 Oct 2014 08:37:57 +0000 (08:37 +0000)]
Move DragController::createFragmentFromDragData implementation to the editor
https://bugs.webkit.org/show_bug.cgi?id=136819

Reviewed by Darin Adler.

Make Editor::webContentFromPasteboard() common to all ports and
add a implementation for every port.

* PlatformWin.cmake: Add new files to compilation.
* WebCore.vcxproj/WebCore.vcxproj: Ditto.
* editing/Editor.cpp:
(WebCore::Editor::pasteWithPasteboard): Move to EditorWin.cpp
since this is only used by windows port.
* editing/Editor.h:
* editing/efl/EditorEfl.cpp:
(WebCore::Editor::webContentFromPasteboard):
* editing/gtk/EditorGtk.cpp:
(WebCore::Editor::webContentFromPasteboard):
* editing/win/EditorWin.cpp: Added.
(WebCore::Editor::pasteWithPasteboard):
(WebCore::Editor::webContentFromPasteboard):
* page/DragController.cpp:
(WebCore::documentFragmentFromDragData): Use the editor instead of
createFragmentFromDragData() that has been removed. Also made this
method a static funtion again.
* page/DragController.h:
* page/efl/DragControllerEfl.cpp: Remove createFragmentFromDragData().
* page/gtk/DragControllerGtk.cpp: Ditto.
* page/mac/DragControllerMac.mm: Ditto.
* page/win/DragControllerWin.cpp: Ditto.
* platform/Pasteboard.h:
(WebCore::Pasteboard::dragDataMap): Add a getter for m_dragDataMap
as required by Editor::webContentFromPasteboard for windows.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@174314 268f45cc-cd09-0410-ab3c-d52691b4dbfc

5 years agoUse is<>() / downcast<>() for CSS StyleProperties
cdumez@apple.com [Sat, 4 Oct 2014 08:30:15 +0000 (08:30 +0000)]
Use is<>() / downcast<>() for CSS StyleProperties
https://bugs.webkit.org/show_bug.cgi?id=137398

Reviewed by Benjamin Poulain.

Use is<>() / downcast<>() for CSS StyleProperties subclasses.

No new tests, no behavior change.

* css/StyleProperties.cpp:
(WebCore::StyleProperties::immutableCopyIfNeeded):
(WebCore::MutableStyleProperties::MutableStyleProperties):
(WebCore::StyleProperties::hasCSSOMWrapper):
* css/StyleProperties.h:
(WebCore::StyleProperties::PropertyReference::propertyMetadata):
(WebCore::StyleProperties::PropertyReference::propertyValue):
(WebCore::StyleProperties::propertyCount):
(WebCore::StyleProperties::deref):
(WebCore::StyleProperties::findPropertyIndex):
(isType):
(WebCore::toMutableStyleProperties): Deleted.
(WebCore::toImmutableStyleProperties): Deleted.
* css/StyleRule.cpp:
(WebCore::StyleRule::mutableProperties):
(WebCore::StyleRulePage::mutableProperties):
(WebCore::StyleRuleFontFace::mutableProperties):
* css/WebKitCSSKeyframeRule.cpp:
(WebCore::StyleKeyframe::mutableProperties):
* dom/StyledElement.cpp:
(WebCore::StyledElement::ensureMutableInlineStyle):
(WebCore::StyledElement::setInlineStyleFromString):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@174313 268f45cc-cd09-0410-ab3c-d52691b4dbfc

5 years ago[Windows] Skip remaining failing media tests after adding audio devices to the bots.
roger_fong@apple.com [Sat, 4 Oct 2014 08:05:22 +0000 (08:05 +0000)]
[Windows] Skip remaining failing media tests after adding audio devices to the bots.

* platform/win/TestExpectations:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@174312 268f45cc-cd09-0410-ab3c-d52691b4dbfc

5 years ago[TexMap] Clean up GraphicsLayerTextureMapper::prepareBackingStoreIfNeeded()
zandobersek@gmail.com [Sat, 4 Oct 2014 07:56:23 +0000 (07:56 +0000)]
[TexMap] Clean up GraphicsLayerTextureMapper::prepareBackingStoreIfNeeded()
https://bugs.webkit.org/show_bug.cgi?id=137263

Reviewed by Darin Adler.

* platform/graphics/texmap/GraphicsLayerTextureMapper.cpp:
(WebCore::GraphicsLayerTextureMapper::prepareBackingStoreIfNeeded): Clean up
the method a bit. When the backing store will not be in use the change mask
should only get updated if there was a backing store present beforehand.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@174311 268f45cc-cd09-0410-ab3c-d52691b4dbfc

5 years ago[GTK] Do not generate webkit2gtk-tests-resources.gresource unconditionally
carlosgc@webkit.org [Sat, 4 Oct 2014 07:46:42 +0000 (07:46 +0000)]
[GTK] Do not generate webkit2gtk-tests-resources.gresource unconditionally
https://bugs.webkit.org/show_bug.cgi?id=137391

Reviewed by Martin Robinson.

Add a custom command to only re-generate
webkit2gtk-tests-resources.gresource when any of its dependencies changes.

* TestWebKitAPI/Tests/WebKit2Gtk/CMakeLists.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@174310 268f45cc-cd09-0410-ab3c-d52691b4dbfc

5 years ago[GTK] generate-gtkdoc crashes when generating HTML due to encoding issues
carlosgc@webkit.org [Sat, 4 Oct 2014 07:35:47 +0000 (07:35 +0000)]
[GTK] generate-gtkdoc crashes when generating HTML due to encoding issues
https://bugs.webkit.org/show_bug.cgi?id=135502

Reviewed by Martin Robinson.

When writing to stdout/stderr fails due to a UnicodeDecodeError
exception, try again without encoding the output.

* gtk/gtkdoc.py:
(GTKDoc._run_command):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@174309 268f45cc-cd09-0410-ab3c-d52691b4dbfc

5 years ago[TexMap] Remove QDoc comments in GraphicsLayerTextureMapper
zandobersek@gmail.com [Sat, 4 Oct 2014 07:14:35 +0000 (07:14 +0000)]
[TexMap] Remove QDoc comments in GraphicsLayerTextureMapper
https://bugs.webkit.org/show_bug.cgi?id=137264

Reviewed by Andreas Kling.

Remove all the QDoc comments in the class that were
left over by the Qt port.

* platform/graphics/texmap/GraphicsLayerTextureMapper.cpp:
(WebCore::GraphicsLayerTextureMapper::setNeedsDisplay):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@174308 268f45cc-cd09-0410-ab3c-d52691b4dbfc

5 years agoIntroduce Promise A+ tests into WebKit
utatane.tea@gmail.com [Sat, 4 Oct 2014 02:41:42 +0000 (02:41 +0000)]
Introduce Promise A+ tests into WebKit
https://bugs.webkit.org/show_bug.cgi?id=136878

Reviewed by Sam Weinig.

To ensure WebKit Promise stability, introducing Promise A+ tests[1] into WebKit.
It's also used to test V8 Promise[2].
So the tests guarantee that WebKit Promise's spec validity and interchangeability
to the other browser's Promise implementation.

Promise A+ tests assumes that it works on the Node.js environment. To work it on
LayoutTests correctly, we need to transform Node.js styled JavaScript code into
browser styled code. In this patch, we use browserify[3] to transform the test code.
Through transforming pipeline, some dependent JS code is bundled into the produced code.
So added SinonJS[4] and browserify's license files into the test directory.

And tests assume that it is used with Mocha[5] test runner. So introducing mocha.js and
added LICENSE file into this directory.

And since Promises/A+ 2.3.3 has many tests, it consumes much time (13s in GTK debug build).
So mark it as a Slow test.

[1]: https://github.com/promises-aplus/promises-tests
[2]: https://code.google.com/p/v8/source/browse#svn%2Ftrunk%2Ftest%2Fpromises-aplus
[3]: http://browserify.org/
[4]: http://sinonjs.org/
[5]: http://visionmedia.github.io/mocha/

* TestExpectations:
* js/promises-tests/README: Added.
* js/promises-tests/browserify-entry-point.js: Added.
(.):
* js/promises-tests/browserify.LICENSE: Added.
* js/promises-tests/mocha.LICENSE: Added.
* js/promises-tests/promises-tests-2-1-2-expected.txt: Added.
* js/promises-tests/promises-tests-2-1-2.html: Added.
* js/promises-tests/promises-tests-2-1-3-expected.txt: Added.
* js/promises-tests/promises-tests-2-1-3.html: Added.
* js/promises-tests/promises-tests-2-2-1-expected.txt: Added.
* js/promises-tests/promises-tests-2-2-1.html: Added.
* js/promises-tests/promises-tests-2-2-2-expected.txt: Added.
* js/promises-tests/promises-tests-2-2-2.html: Added.
* js/promises-tests/promises-tests-2-2-3-expected.txt: Added.
* js/promises-tests/promises-tests-2-2-3.html: Added.
* js/promises-tests/promises-tests-2-2-4-expected.txt: Added.
* js/promises-tests/promises-tests-2-2-4.html: Added.
* js/promises-tests/promises-tests-2-2-5-expected.txt: Added.
* js/promises-tests/promises-tests-2-2-5.html: Added.
* js/promises-tests/promises-tests-2-2-6-expected.txt: Added.
* js/promises-tests/promises-tests-2-2-6.html: Added.
* js/promises-tests/promises-tests-2-2-7-expected.txt: Added.
* js/promises-tests/promises-tests-2-2-7.html: Added.
* js/promises-tests/promises-tests-2-3-1-expected.txt: Added.
* js/promises-tests/promises-tests-2-3-1.html: Added.
* js/promises-tests/promises-tests-2-3-2-expected.txt: Added.
* js/promises-tests/promises-tests-2-3-2.html: Added.
* js/promises-tests/promises-tests-2-3-3-expected.txt: Added.
* js/promises-tests/promises-tests-2-3-3.html: Added.
* js/promises-tests/promises-tests-2-3-4-expected.txt: Added.
* js/promises-tests/promises-tests-2-3-4.html: Added.
* js/promises-tests/promises-tests.LICENSE: Added.
* js/promises-tests/promises-tests.js: Added.
(.):
* js/promises-tests/promises-tests/.gitignore: Added.
* js/promises-tests/promises-tests/.jshintrc: Added.
* js/promises-tests/promises-tests/.npmignore: Added.
* js/promises-tests/promises-tests/LICENSE.txt: Added.
* js/promises-tests/promises-tests/README.md: Added.
* js/promises-tests/promises-tests/lib/cli.js: Added.
(getAdapterFilePath):
(adapterObjectFromFilePath):
* js/promises-tests/promises-tests/lib/getMochaOpts.js: Added.
(module.exports):
* js/promises-tests/promises-tests/lib/programmaticRunner.js: Added.
(normalizeAdapter.adapter.resolved):
(normalizeAdapter.adapter.rejected):
(normalizeAdapter):
(.cb):
(.):
(module.exports):
(module.exports.mocha):
* js/promises-tests/promises-tests/lib/tests/2.1.2.js: Added.
(.):
* js/promises-tests/promises-tests/lib/tests/2.1.3.js: Added.
(.):
* js/promises-tests/promises-tests/lib/tests/2.2.1.js: Added.
* js/promises-tests/promises-tests/lib/tests/2.2.2.js: Added.
(.):
* js/promises-tests/promises-tests/lib/tests/2.2.3.js: Added.
(.):
* js/promises-tests/promises-tests/lib/tests/2.2.4.js: Added.
(.):
* js/promises-tests/promises-tests/lib/tests/2.2.5.js: Added.
* js/promises-tests/promises-tests/lib/tests/2.2.6.js: Added.
(.return.if):
(callbackAggregator):
(.):
* js/promises-tests/promises-tests/lib/tests/2.2.7.js: Added.
(.):
* js/promises-tests/promises-tests/lib/tests/2.3.1.js: Added.
(.):
* js/promises-tests/promises-tests/lib/tests/2.3.2.js: Added.
(.):
(testPromiseResolution):
* js/promises-tests/promises-tests/lib/tests/2.3.3.js: Added.
(.):
(testPromiseResolution):
(testCallingResolvePromiseRejectsWith):
(testCallingRejectPromiseRejectsWith):
(describe):
* js/promises-tests/promises-tests/lib/tests/2.3.4.js: Added.
(.):
* js/promises-tests/promises-tests/lib/tests/helpers/reasons.js: Added.
(exports.string_appeared_here):
(exports.string_appeared_here.return.then):
* js/promises-tests/promises-tests/lib/tests/helpers/testThreeCases.js: Added.
(.specify):
(.):
(exports.testFulfilled):
(exports.testRejected):
* js/promises-tests/promises-tests/lib/tests/helpers/thenables.js: Added.
(exports.fulfilled.string_appeared_here.return.then):
(exports.fulfilled.string_appeared_here):
(exports.fulfilled.string_appeared_here.):
(exports.fulfilled.get string_appeared_here.return.then):
(exports.fulfilled.get string_appeared_here):
(exports.rejected.string_appeared_here.return.then):
(exports.rejected.string_appeared_here):
(exports.rejected.string_appeared_here.):
(exports.rejected.get string_appeared_here.return.then):
(exports.rejected.get string_appeared_here):
* js/promises-tests/promises-tests/package.json: Added.
* js/promises-tests/promises-tests/test/getMochaOptsTest.js: Added.
* js/promises-tests/resources/adapter.js: Added.
(LayoutTestsReporter.runner):
(window.adapter.resolved):
(window.adapter.rejected):
(window.adapter.deferred):
* js/promises-tests/resources/mocha.js: Added.
(.):
* js/promises-tests/sinonjs.LICENSE: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@174307 268f45cc-cd09-0410-ab3c-d52691b4dbfc

5 years agoBugzilla status bubble displays an exception showing attachment 175738
ap@apple.com [Fri, 3 Oct 2014 23:48:52 +0000 (23:48 +0000)]
Bugzilla status bubble displays an exception showing attachment 175738
https://bugs.webkit.org/show_bug.cgi?id=137410

Reviewed by Tim Horton.

* QueueStatusServer/app.yaml: Update version.

* QueueStatusServer/handlers/statusbubble.py: (StatusBubble._build_bubble):
Don't fail even if expected status message is missing.

* QueueStatusServer/model/patchlog.py: (PatchLog): Quick unrelated fix - define
the field in the same way it's defined in QueueStatus table.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@174304 268f45cc-cd09-0410-ab3c-d52691b4dbfc

5 years agoBugzilla status bubble displays an exception showing attachment 175738
ap@apple.com [Fri, 3 Oct 2014 23:36:54 +0000 (23:36 +0000)]
Bugzilla status bubble displays an exception showing attachment 175738
https://bugs.webkit.org/show_bug.cgi?id=137410

Reviewed by Tim Horton.

* QueueStatusServer/app.yaml: Update version.

* QueueStatusServer/handlers/statusbubble.py: (StatusBubble._build_bubble):
Don't fail even if expected status message is missing.

* QueueStatusServer/model/patchlog.py: (PatchLog): Quick unrelated fix - define
the field in the same way it's defined in QueueStatus table.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@174303 268f45cc-cd09-0410-ab3c-d52691b4dbfc

5 years agoRecord latest message in PatchLog for quick access
ap@apple.com [Fri, 3 Oct 2014 23:27:52 +0000 (23:27 +0000)]
Record latest message in PatchLog for quick access
https://bugs.webkit.org/show_bug.cgi?id=137405

Fixing a typo.

* QueueStatusServer/app.yaml:
* QueueStatusServer/handlers/releasepatch.py:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@174302 268f45cc-cd09-0410-ab3c-d52691b4dbfc

5 years agoREGRESSION(r174216): CodeBlock::dumpByteCodes crashes on op_push_name_scope
msaboff@apple.com [Fri, 3 Oct 2014 23:22:21 +0000 (23:22 +0000)]
REGRESSION(r174216): CodeBlock::dumpByteCodes crashes on op_push_name_scope
https://bugs.webkit.org/show_bug.cgi?id=137412

Reviewed by Mark Lam.

Added support for the JSNameScope::type opcode parameter in dumpBytecode().

* bytecode/CodeBlock.cpp:
(JSC::CodeBlock::dumpBytecode):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@174301 268f45cc-cd09-0410-ab3c-d52691b4dbfc

5 years agoUse is<>() / downcast<>() for CSSValue subclasses
cdumez@apple.com [Fri, 3 Oct 2014 23:17:00 +0000 (23:17 +0000)]
Use is<>() / downcast<>() for CSSValue subclasses
https://bugs.webkit.org/show_bug.cgi?id=137381

Reviewed by Benjamin Poulain.

Use is<>() / downcast<>() for CSSValue subclasses.

No new tests, no behavior change.

* css/CSSAspectRatioValue.h:
* css/CSSBorderImageSliceValue.h:
* css/CSSCalculationValue.cpp:
(WebCore::CSSCalcExpressionNodeParser::parseValue):
* css/CSSCalculationValue.h:
* css/CSSCanvasValue.h:
* css/CSSComputedStyleDeclaration.cpp:
(WebCore::getBorderRadiusShorthandValue):
(WebCore::ComputedStyleExtractor::propertyMatches):
* css/CSSCrossfadeValue.cpp:
(WebCore::subimageKnownToBeOpaque):
* css/CSSCrossfadeValue.h:
* css/CSSCursorImageValue.cpp:
(WebCore::CSSCursorImageValue::~CSSCursorImageValue):
(WebCore::CSSCursorImageValue::updateIfSVGCursorIsUsed):
(WebCore::CSSCursorImageValue::cachedImage):
(WebCore::CSSCursorImageValue::cachedOrPendingImage):
(WebCore::CSSCursorImageValue::isSVGCursor):
* css/CSSCursorImageValue.h:
* css/CSSFilterImageValue.h:
* css/CSSFontFaceSrcValue.h:
* css/CSSFontFeatureValue.h:
* css/CSSFontSelector.cpp:
(WebCore::CSSFontSelector::addFontFaceRule):
* css/CSSFontValue.h:
* css/CSSFunctionValue.h:
* css/CSSGradientValue.cpp:
(WebCore::CSSGradientValue::image):
(WebCore::CSSGradientValue::gradientWithStylesResolved):
* css/CSSGradientValue.h:
* css/CSSGridLineNamesValue.h:
* css/CSSGridTemplateAreasValue.h:
* css/CSSImageGeneratorValue.cpp:
(WebCore::CSSImageGeneratorValue::image):
(WebCore::CSSImageGeneratorValue::isFixedSize):
(WebCore::CSSImageGeneratorValue::fixedSize):
(WebCore::CSSImageGeneratorValue::isPending):
(WebCore::CSSImageGeneratorValue::knownToBeOpaque):
(WebCore::CSSImageGeneratorValue::loadSubimages):
(WebCore::CSSImageGeneratorValue::subimageIsPending):
(WebCore::CSSImageGeneratorValue::cachedImageForCSSValue):
* css/CSSImageGeneratorValue.h:
* css/CSSImageSetValue.cpp:
(WebCore::CSSImageSetValue::fillImageSet):
* css/CSSImageSetValue.h:
* css/CSSImageValue.h:
* css/CSSInheritedValue.h:
* css/CSSInitialValue.h:
* css/CSSLineBoxContainValue.h:
* css/CSSParser.cpp:
(WebCore::CSSParser::parseColor):
(WebCore::CSSParser::addFillValue):
(WebCore::CSSParser::addAnimationValue):
(WebCore::CSSParser::parse3ValuesFillPosition):
(WebCore::CSSParser::parseFillPosition):
(WebCore::CSSParser::parseFillRepeat):
(WebCore::gridMissingGridPositionValue):
(WebCore::CSSParser::parseGridTemplateRowsAndAreas):
(WebCore::CSSParser::parseBasicShapeCircle):
(WebCore::CSSParser::parseBasicShapeEllipse):
(WebCore::CSSParser::parseDeprecatedGradient):
(WebCore::CSSParser::parseDeprecatedRadialGradient):
(WebCore::CSSParser::parseRadialGradient):
(WebCore::CSSParser::createFontFaceRule):
* css/CSSPrimitiveValue.h:
* css/CSSReflectValue.h:
* css/CSSShadowValue.h:
* css/CSSTimingFunctionValue.h:
* css/CSSToStyleMap.cpp:
(WebCore::CSSToStyleMap::mapFillAttachment):
(WebCore::CSSToStyleMap::mapFillClip):
(WebCore::CSSToStyleMap::mapFillComposite):
(WebCore::CSSToStyleMap::mapFillBlendMode):
(WebCore::CSSToStyleMap::mapFillOrigin):
(WebCore::CSSToStyleMap::mapFillRepeatX):
(WebCore::CSSToStyleMap::mapFillRepeatY):
(WebCore::CSSToStyleMap::mapFillSize):
(WebCore::CSSToStyleMap::mapFillXPosition):
(WebCore::CSSToStyleMap::mapFillYPosition):
(WebCore::CSSToStyleMap::mapFillMaskSourceType):
(WebCore::CSSToStyleMap::mapAnimationDelay):
(WebCore::CSSToStyleMap::mapAnimationDirection):
(WebCore::CSSToStyleMap::mapAnimationDuration):
(WebCore::CSSToStyleMap::mapAnimationFillMode):
(WebCore::CSSToStyleMap::mapAnimationIterationCount):
(WebCore::CSSToStyleMap::mapAnimationName):
(WebCore::CSSToStyleMap::mapAnimationPlayState):
(WebCore::CSSToStyleMap::mapAnimationProperty):
(WebCore::CSSToStyleMap::mapAnimationTimingFunction):
(WebCore::CSSToStyleMap::mapNinePieceImage):
(WebCore::CSSToStyleMap::mapNinePieceImageSlice):
(WebCore::CSSToStyleMap::mapNinePieceImageQuad):
(WebCore::CSSToStyleMap::mapNinePieceImageRepeat):
* css/CSSUnicodeRangeValue.h:
* css/CSSValue.cpp:
(WebCore::CSSValue::isImplicitInitialValue):
(WebCore::CSSValue::addSubresourceStyleURLs):
(WebCore::CSSValue::hasFailedOrCanceledSubresources):
(WebCore::CSSValue::equals):
(WebCore::CSSValue::cssText):
(WebCore::CSSValue::destroy):
(WebCore::CSSValue::cloneForCSSOM):
* css/CSSValue.h:
* css/CSSValueList.h:
(WebCore::CSSValueListInspector::CSSValueListInspector):
* css/DeprecatedStyleBuilder.cpp:
(WebCore::ApplyPropertyDefault::applyValue):
(WebCore::ApplyPropertyNumber::applyValue):
(WebCore::ApplyPropertyAuto::applyValue):
(WebCore::ApplyPropertyClip::applyValue):
(WebCore::ApplyPropertyColor::applyValue):
(WebCore::ApplyPropertyLength::applyValue):
(WebCore::ApplyPropertyString::applyValue):
(WebCore::ApplyPropertyBorderRadius::applyValue):
(WebCore::ApplyPropertyFillLayer::applyValue):
(WebCore::ApplyPropertyComputeLength::applyValue):
(WebCore::ApplyPropertyFont::applyValue):
(WebCore::ApplyPropertyFontFamily::applyValue):
(WebCore::ApplyPropertyFontSize::applyValue):
(WebCore::ApplyPropertyFontWeight::applyValue):
(WebCore::ApplyPropertyFontVariantLigatures::applyValue):
(WebCore::ApplyPropertyCounter::applyValue):
(WebCore::ApplyPropertyCursor::applyValue):
(WebCore::ApplyPropertyTextAlign::applyValue):
(WebCore::ApplyPropertyTextDecoration::applyValue):
(WebCore::ApplyPropertyTextDecorationSkip::applyValue):
(WebCore::ApplyPropertyMarqueeIncrement::applyValue):
(WebCore::ApplyPropertyMarqueeRepetition::applyValue):
(WebCore::ApplyPropertyMarqueeSpeed::applyValue):
(WebCore::ApplyPropertyTextUnderlinePosition::applyValue):
(WebCore::ApplyPropertyLineHeight::applyValue):
(WebCore::ApplyPropertyLineHeightForIOSTextAutosizing::applyValue):
(WebCore::ApplyPropertyWordSpacing::applyValue):
(WebCore::ApplyPropertyPageSize::applyValue):
(WebCore::ApplyPropertyTextEmphasisStyle::applyValue):
(WebCore::ApplyPropertyTextEmphasisPosition::applyValue):
(WebCore::ApplyPropertyResize::applyValue):
(WebCore::ApplyPropertyVerticalAlign::applyValue):
(WebCore::ApplyPropertyAspectRatio::applyValue):
(WebCore::ApplyPropertyZoom::applyValue):
(WebCore::ApplyPropertyDisplay::applyValue):
(WebCore::ApplyPropertyClipPath::applyValue):
(WebCore::ApplyPropertyShape::applyValue):
(WebCore::ApplyPropertyImageResolution::applyValue):
(WebCore::ApplyPropertyTextIndent::applyValue):
* css/MediaList.cpp:
(WebCore::reportMediaQueryWarningIfNeeded):
* css/MediaQueryEvaluator.cpp:
(WebCore::compareAspectRatioValue):
(WebCore::numberValue):
(WebCore::orientationMediaFeatureEval):
(WebCore::evalResolution):
(WebCore::device_pixel_ratioMediaFeatureEval):
(WebCore::resolutionMediaFeatureEval):
(WebCore::computeLength):
(WebCore::view_modeMediaFeatureEval):
(WebCore::pointerMediaFeatureEval):
* css/SVGCSSComputedStyleDeclaration.cpp:
(WebCore::strokeDashArrayToCSSValueList):
(WebCore::ComputedStyleExtractor::svgPropertyValue):
* css/SVGCSSStyleSelector.cpp:
(WebCore::StyleResolver::applySVGProperty):
* css/SourceSizeList.cpp:
(WebCore::computeLength):
* css/StyleProperties.cpp:
(WebCore::StyleProperties::getLayeredShorthandValue):
* css/StyleResolver.cpp:
(WebCore::extractDirectionAndWritingMode):
(WebCore::createGridTrackSize):
(WebCore::createGridTrackList):
(WebCore::createGridPosition):
(WebCore::StyleResolver::parseSnapCoordinate):
(WebCore::StyleResolver::parseSnapPoints):
(WebCore::StyleResolver::applyProperty):
(WebCore::StyleResolver::styleImage):
(WebCore::StyleResolver::generatedOrPendingFromValue):
(WebCore::StyleResolver::createFilterOperations):
* css/TransformFunctions.cpp:
(WebCore::transformsForValue):
* css/ViewportStyleResolver.cpp:
(WebCore::ViewportStyleResolver::getViewportArgumentValue):
* css/WebKitCSSFilterValue.h:
* css/WebKitCSSMatrix.cpp:
(WebCore::WebKitCSSMatrix::setMatrixValue):
* css/WebKitCSSTransformValue.h:
(WebCore::toWebKitCSSTransformValue): Deleted.
* editing/ApplyStyleCommand.cpp:
(WebCore::ApplyStyleCommand::computedFontSize):
* editing/EditingStyle.cpp:
(WebCore::identifierForStyleProperty):
(WebCore::HTMLElementEquivalent::valueIsPresentInStyle):
(WebCore::HTMLTextDecorationEquivalent::valueIsPresentInStyle):
(WebCore::cssValueToRGBA):
(WebCore::EditingStyle::extractFontSizeDelta):
(WebCore::EditingStyle::textDirection):
(WebCore::EditingStyle::prepareToApplyAt):
(WebCore::EditingStyle::mergeStyle):
(WebCore::EditingStyle::mergeStyleFromRulesForSerialization):
(WebCore::EditingStyle::legacyFontSize):
(WebCore::EditingStyle::textDirectionForSelection):
(WebCore::StyleChange::extractTextStyles):
(WebCore::diffTextDecorations):
(WebCore::fontWeightIsBold):
(WebCore::isTransparentColorValue):
* editing/EditorCommand.cpp:
(WebCore::executeToggleStyleInList):
* editing/cocoa/HTMLConverter.mm:
(stringFromCSSValue):
(HTMLConverterCaches::floatPropertyValueForNode):
(HTMLConverterCaches::colorPropertyValueForNode):
* editing/markup.cpp:
(WebCore::propertyMissingOrEqualToNone):
* page/PageSerializer.cpp:
(WebCore::PageSerializer::retrieveResourcesForProperties):
* page/animation/CSSPropertyAnimation.cpp:
(WebCore::blendFunc):
* rendering/RenderTextControl.cpp:
(WebCore::RenderTextControl::adjustInnerTextStyle):
* rendering/style/StylePendingImage.h:
(WebCore::StylePendingImage::cssImageValue):
(WebCore::StylePendingImage::cssCursorImageValue):
(WebCore::StylePendingImage::cssImageSetValue):
* svg/SVGColor.h:
* svg/SVGFontFaceElement.cpp:
(WebCore::SVGFontFaceElement::rebuildFontFace):
* svg/SVGPaint.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@174300 268f45cc-cd09-0410-ab3c-d52691b4dbfc

5 years agoRemove a workaround for a bug that's been fixed
andersca@apple.com [Fri, 3 Oct 2014 23:05:07 +0000 (23:05 +0000)]
Remove a workaround for a bug that's been fixed
https://bugs.webkit.org/show_bug.cgi?id=137411

Reviewed by Dan Bernstein.

* mac/rewrite-availability-macros.sh:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@174299 268f45cc-cd09-0410-ab3c-d52691b4dbfc

5 years agoImplement op_profile_type in the 32-bit baseline JIT
saambarati1@gmail.com [Fri, 3 Oct 2014 22:19:33 +0000 (22:19 +0000)]
Implement op_profile_type in the 32-bit baseline JIT
https://bugs.webkit.org/show_bug.cgi?id=137181

Reviewed by Michael Saboff.

Generate inline code to write to the TypeProfilerLog inside the 32-bit
baseline JIT instead of unconditionally bailing out to the slow path
for op_profile_type.

* jit/JITOpcodes32_64.cpp:
(JSC::JIT::emit_op_profile_type):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@174298 268f45cc-cd09-0410-ab3c-d52691b4dbfc

5 years agoClean up interface to Font::expansionOpportunityCount()
mmaxfield@apple.com [Fri, 3 Oct 2014 22:07:42 +0000 (22:07 +0000)]
Clean up interface to Font::expansionOpportunityCount()
https://bugs.webkit.org/show_bug.cgi?id=137355

Reviewed by Dean Jackson.

There are two overloads of Font::expansionOpportunityCount() which perform the same
operation. One overload takes a UChar*, the other takes an LChar*, and they both
take a length. This is the abstraction that StringView was designed to be. Instead
of forcing each caller to take a different overload based on if their data is
8 bit or not, allow the caller to construct a StringView and pass that into
Font::expansionOpportunityCount() instead of a raw pointer/length.

No new tests because there is no behavior change.

* platform/graphics/Font.cpp:
(WebCore::Font::expansionOpportunityCountInternal): Original two functions,
renamed.
(WebCore::Font::expansionOpportunityCount): Takes a StringView, calls
expansionOpportunityCountInternal().
* platform/graphics/Font.h: Update signatures.
* platform/graphics/WidthIterator.cpp:
(WebCore::WidthIterator::WidthIterator): Use new signature to
Font::expansionOpportunityCount().
* platform/graphics/mac/ComplexTextController.cpp:
(WebCore::ComplexTextController::ComplexTextController): Ditto.
* rendering/RenderBlockLineLayout.cpp:
(WebCore::RenderBlockFlow::computeInlineDirectionPositionsForSegment): Ditto.
* rendering/RenderText.cpp:
(WebCore::RenderText::stringView): Accessor to encapsulate character pointer
and length.
* rendering/RenderText.h: Signature of new accessor.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@174297 268f45cc-cd09-0410-ab3c-d52691b4dbfc

5 years agoUnreviewed, rolling out r174275.
bdakin@apple.com [Fri, 3 Oct 2014 22:05:22 +0000 (22:05 +0000)]
Unreviewed, rolling out r174275.
https://bugs.webkit.org/show_bug.cgi?id=137408

Build failures on the internal bots. (Requested by dethbakin
on #webkit).

Reverted changeset:

"FTL should sink PutLocals"
https://bugs.webkit.org/show_bug.cgi?id=137168
http://trac.webkit.org/changeset/174275

Patch by Commit Queue <commit-queue@webkit.org> on 2014-10-03

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@174296 268f45cc-cd09-0410-ab3c-d52691b4dbfc

5 years ago[Win] Unreviewed build fix for MSVC 2013 SP 3.
bfulgham@apple.com [Fri, 3 Oct 2014 22:03:59 +0000 (22:03 +0000)]
[Win] Unreviewed build fix for MSVC 2013 SP 3.

The std::array initializer is not fully implemented in SP3 and causes a
build error.

* platform/graphics/transforms/AffineTransform.cpp: Use old style assignment
initialization for MSVC builds until this bug is fixed.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@174295 268f45cc-cd09-0410-ab3c-d52691b4dbfc

5 years agotearoff_arguments should always refer to the unmodified arguments register
oliver@apple.com [Fri, 3 Oct 2014 21:46:14 +0000 (21:46 +0000)]
tearoff_arguments should always refer to the unmodified arguments register
https://bugs.webkit.org/show_bug.cgi?id=137406

Reviewed by Michael Saboff.

To simplify subsequent work, and remove unnecessary work from
actual execution this patch simply ensures that tear_off_arguments
refers to the actual unmodified arguments register.

* bytecompiler/BytecodeGenerator.cpp:
(JSC::BytecodeGenerator::emitReturn):
* dfg/DFGByteCodeParser.cpp:
(JSC::DFG::ByteCodeParser::parseBlock):
* jit/JITOpcodes.cpp:
(JSC::JIT::emit_op_tear_off_arguments):
* jit/JITOpcodes32_64.cpp:
(JSC::JIT::emit_op_tear_off_arguments):
* llint/LLIntSlowPaths.cpp:
(JSC::LLInt::LLINT_SLOW_PATH_DECL):
* llint/LowLevelInterpreter32_64.asm:
* llint/LowLevelInterpreter64.asm:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@174294 268f45cc-cd09-0410-ab3c-d52691b4dbfc

5 years agoAdd TypeCasts.h to xcodeproject
cdumez@apple.com [Fri, 3 Oct 2014 21:43:18 +0000 (21:43 +0000)]
Add TypeCasts.h to xcodeproject
https://bugs.webkit.org/show_bug.cgi?id=137403

Reviewed by Benjamin Poulain.

Add TypeCasts.h to xcodeproject.

* WTF.xcodeproj/project.pbxproj:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@174293 268f45cc-cd09-0410-ab3c-d52691b4dbfc

5 years agoWeb Inspector: Move the computation that results in UI strings from JSC to the Web...
saambarati1@gmail.com [Fri, 3 Oct 2014 21:35:20 +0000 (21:35 +0000)]
Web Inspector: Move the computation that results in UI strings from JSC to the Web Inspector
https://bugs.webkit.org/show_bug.cgi?id=137295

Reviewed by Timothy Hatcher.

Source/JavaScriptCore:

Remove unnecessary functions and properties from JSC that are
now being computed inside the Web Inspector.

* inspector/agents/InspectorRuntimeAgent.cpp:
(Inspector::InspectorRuntimeAgent::getRuntimeTypesForVariablesAtOffsets):
* inspector/protocol/Runtime.json:
* runtime/TypeSet.cpp:
(JSC::TypeSet::allPrimitiveTypeNames): Deleted.
* runtime/TypeSet.h:

Source/WebInspectorUI:

The inspector now computes a set of properties that was previously
computed inside JSC.

This patch also addresses the weird experience of the
type of an expression claiming to be both a Number and
an Integer. This could be a bit confusing because Number
is supposed to represent Double|Integer and the previous
UI implied that Number was equivalent to Double. The new UI only
ever shows either the type of Number or Integer when showing
the type information of an expression. It never shows both Number
and Integer simultaneously to describe a single expression.

* UserInterface/Models/TypeSet.js:
(WebInspector.TypeSet):
(WebInspector.TypeSet.prototype.isContainedIn):
(WebInspector.TypeSet.prototype.get primitiveTypeNames):
* UserInterface/Views/TypePropertiesSection.js:
(WebInspector.TypePropertiesSection):
(WebInspector.TypePropertiesSection.prototype.onpopulate):
* UserInterface/Views/TypeTokenView.js:
(WebInspector.TypeTokenView):
(WebInspector.TypeTokenView.prototype.update):
(WebInspector.TypeTokenView.prototype._shouldShowPopover):
(WebInspector.TypeTokenView.prototype._displayTypeName):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@174292 268f45cc-cd09-0410-ab3c-d52691b4dbfc

5 years agoRecord latest message in PatchLog for quick access
ap@apple.com [Fri, 3 Oct 2014 21:35:16 +0000 (21:35 +0000)]
Record latest message in PatchLog for quick access
https://bugs.webkit.org/show_bug.cgi?id=137405

Reviewed by Ryosuke Niwa.

It's no SQL...

* QueueStatusServer/handlers/releasepatch.py:
* QueueStatusServer/handlers/updatestatus.py:
* QueueStatusServer/loggers/recordpatchevent.py:
* QueueStatusServer/model/patchlog.py:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@174291 268f45cc-cd09-0410-ab3c-d52691b4dbfc

5 years ago-asan-blacklist option also renamed to generic -fsanitize-blacklist
dfarler@apple.com [Fri, 3 Oct 2014 21:18:12 +0000 (21:18 +0000)]
-asan-blacklist option also renamed to generic -fsanitize-blacklist
https://bugs.webkit.org/show_bug.cgi?id=137377

Reviewed by Daniel Bates.

* asan/asan.xcconfig:
-asan-blacklist -> -fsanitize-blacklist

http://llvm.org/viewvc/llvm-project?view=revision&revision=212505

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@174290 268f45cc-cd09-0410-ab3c-d52691b4dbfc

5 years agoFTL should sink PutLocals
betravis@adobe.com [Fri, 3 Oct 2014 20:47:25 +0000 (20:47 +0000)]
FTL should sink PutLocals
https://bugs.webkit.org/show_bug.cgi?id=137168

Patch by Filip Pizlo <fpizlo@apple.com> on 2014-10-02
Reviewed by Oliver Hunt.

* js/regress/elidable-new-object-then-call-expected.txt: Added.
* js/regress/elidable-new-object-then-call.html: Added.
* js/regress/script-tests/elidable-new-object-then-call.js: Added.
(sumOfArithSeries):
(bar):
(foo):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@174289 268f45cc-cd09-0410-ab3c-d52691b4dbfc

5 years agoRevert an accidental change introduced by r173318.
jer.noble@apple.com [Fri, 3 Oct 2014 20:45:29 +0000 (20:45 +0000)]
Revert an accidental change introduced by r173318.

Rubber-stamped by Tim Horton.

* WebProcess/WebPage/mac/TiledCoreAnimationDrawingArea.mm:
(WebKit::TiledCoreAnimationDrawingArea::suspendPainting):
(WebKit::TiledCoreAnimationDrawingArea::resumePainting):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@174288 268f45cc-cd09-0410-ab3c-d52691b4dbfc

5 years agoWKWebView snapshot of Daring Fireball has the wrong color in the obscured inset
timothy_horton@apple.com [Fri, 3 Oct 2014 20:36:51 +0000 (20:36 +0000)]
WKWebView snapshot of Daring Fireball has the wrong color in the obscured inset
https://bugs.webkit.org/show_bug.cgi?id=137383
<rdar://problem/18535439>

Reviewed by Sam Weinig.

* WebProcess/WebPage/WebPage.cpp:
(WebKit::WebPage::snapshotAtSize):
Use documentBackgroundColor instead of baseBackgroundColor if backgroundShouldExtendBeyondPage is set.

* rendering/RenderView.cpp:
(WebCore::RenderView::paintBoxDecorations):
Use documentBackgroundColor instead of baseBackgroundColor if backgroundShouldExtendBeyondPage is set.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@174287 268f45cc-cd09-0410-ab3c-d52691b4dbfc

5 years ago[iOS] Highlight clicked links in WKPDFView
aestes@apple.com [Fri, 3 Oct 2014 20:32:34 +0000 (20:32 +0000)]
[iOS] Highlight clicked links in WKPDFView
https://bugs.webkit.org/show_bug.cgi?id=137400

Reviewed by Tim Horton.

Show a _UIHighlightView on top of clicked links for 200 ms before starting the navigation. This matches the behavior of UIWebPDFView.

* UIProcess/ios/WKPDFView.mm:
(-[WKPDFView _createHighlightViewWithFrame:]): Created a _UIHighlightView with a color and border radius matching the values used by UIWebPDFView.
(-[WKPDFView annotation:wasTouchedAtPoint:controller:]): Displayed the highlight, then removed it after the navigation began.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@174286 268f45cc-cd09-0410-ab3c-d52691b4dbfc

5 years ago[Windows] Include batch size argument to layout tests on Windows.
roger_fong@apple.com [Fri, 3 Oct 2014 20:29:29 +0000 (20:29 +0000)]
[Windows] Include batch size argument to layout tests on Windows.

* BuildSlaveSupport/build.webkit.org-config/master.cfg:
(RunWebKitTests.start):
This is a speculative fix for the the inconsistent crashes we are seeing on Windows.
These crashes seem to happen on random tests though generally in a spread out fashion.
It seems possible that the DRT instance is being clobbered somehow after running too many
tests in a row so we will try to mitigate this by just forcing DRT to restart periodically.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@174285 268f45cc-cd09-0410-ab3c-d52691b4dbfc

5 years agoREGRESSION (r165896): UIWebView PDFs have a forced black background
timothy_horton@apple.com [Fri, 3 Oct 2014 20:28:04 +0000 (20:28 +0000)]
REGRESSION (r165896): UIWebView PDFs have a forced black background
https://bugs.webkit.org/show_bug.cgi?id=137387
<rdar://problem/17948411>

Reviewed by David Kilzer.

* WebView/WebPDFViewPlaceholder.h:
* WebView/WebPDFViewPlaceholder.mm:
(-[WebPDFViewPlaceholder layout]):
Remove an ill-advised attempt to force the PDF background color from WebKit.
The idea was that we would maintain Safari's black background except if there
were no pages in the PDF, in which case we would use white instead to avoid
having an all-black view. However, this broke clients who didn't want a black
background, because we were now forcing it to black. Also, we don't need to
maintain the black color in Safari because it doesn't use WebPDFViewPlaceholder anymore.
This wasn't an important part of the patch, in any case.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@174284 268f45cc-cd09-0410-ab3c-d52691b4dbfc

5 years agowebkit-appearance: default-button styling does not set the button font color to white
dino@apple.com [Fri, 3 Oct 2014 20:27:28 +0000 (20:27 +0000)]
webkit-appearance: default-button styling does not set the button font color to white
https://bugs.webkit.org/show_bug.cgi?id=137399
rdar://problem/17788616

Reviewed by Dave Hyatt.

When a button is styled with -webkit-appearance: default-button, it should
use the CSSValueActivebuttontext value for its text color when the window
is active, and the normal value when the window is inactive.

Since activating (focus/blur) windows doesn't cause a style recalculation, this
is applied as a paint-time operation. However, the render tree children that
paint the text don't know that they are contained within a RenderButton
and that the button is a default type. So I added an inherited flag, similar
to what we use for visited links, to remember if we're in a button. And then
computeTextPaintStyle chooses the correct value.

We can't test this because our LayoutTest system doesn't have
an active window, and thus the default button renders like a normal
button. This should cause no regressions though. Also, the appearance is
protected behind an SPI (or Setting), so this is not exposed to the Web.

* rendering/RenderTheme.cpp:
(WebCore::RenderTheme::adjustStyle): Set the flag if we're inside a default button.
* rendering/TextPaintStyle.cpp:
(WebCore::computeTextPaintStyle): If we're in a default button inside an active window,
use the CSSValueActivebuttontext color.
* rendering/style/RenderStyle.cpp:
(WebCore::RenderStyle::RenderStyle): Add and expose the new inherited flag: ._insideDefaultButton.
(WebCore::RenderStyle::changeRequiresRepaint): Ditto.
* rendering/style/RenderStyle.h: Ditto.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@174283 268f45cc-cd09-0410-ab3c-d52691b4dbfc

5 years agoREGRESSION (r173265): UIWebView PDF loading is completely broken
timothy_horton@apple.com [Fri, 3 Oct 2014 20:26:41 +0000 (20:26 +0000)]
REGRESSION (r173265): UIWebView PDF loading is completely broken
https://bugs.webkit.org/show_bug.cgi?id=137386
<rdar://problem/18524717>

Reviewed by David Kilzer.

* WebView/WebPDFViewPlaceholder.h:
* WebView/WebPDFViewPlaceholder.mm:
(-[WebPDFViewPlaceholder layout]):
(-[WebPDFViewPlaceholder finishedLoadingWithDataSource:]):
Restore important code that creates the CGDataProvider and sets _didFinishLoad
and calls _doPostLoadOrUnlockTasks after finishing a load, which was haphazardly
removed in the aforementioned revision.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@174282 268f45cc-cd09-0410-ab3c-d52691b4dbfc

5 years ago[SVG -> OTF Converter] Support non-BMP codepoints
mmaxfield@apple.com [Fri, 3 Oct 2014 20:10:56 +0000 (20:10 +0000)]
[SVG -> OTF Converter] Support non-BMP codepoints
https://bugs.webkit.org/show_bug.cgi?id=137097

Patch by Myles C. Maxfield <mmaxfield@apple.com> on 2014-09-28
Reviewed by Darin Adler.

This patch has two pieces:

1) Moves write16(), overwrite32() and overwrite16() inside SVGToOTFFontConverter. This
is so that we don't have to keep passing around references to m_result everywhere. Instead,
put the output vector inside the class as an instance member variable. In addition, rename
write32() and write16() to append32() and append16(), and provide a copy of append32 inside
SVGToOTFFontConverter.

2) Change the "Codepoint" typedef to be a String, and rename it to Codepoints. This is so that
we can implement ligatures later on. Update all the places that use this variable to work with
a String of multiple codepoints.

This patch updates the CMAP table to use format 12 instead of format 4. It currently only maps
single codepoints to glyph IDs and skips all multi-codepoint glyphs. We will implement
ligatures in another patch.

Test: svg/custom/glyph-selection-non-bmp.svg

* svg/SVGToOTFFontConversion.cpp:
(WebCore::append32): Append a result to a given vector. This is for the path transcoder.
(WebCore::SVGToOTFFontConverter::releaseResult): WTF::move()'s the result.
(WebCore::SVGToOTFFontConverter::GlyphData::GlyphData): Update to take Codepoints instead of Codepoint.
(WebCore::SVGToOTFFontConverter::append32): Piece 1 above.
(WebCore::SVGToOTFFontConverter::append16): Ditto.
(WebCore::SVGToOTFFontConverter::append4ByteCode): Convenience function.
(WebCore::SVGToOTFFontConverter::overwrite32): Piece 1 above.
(WebCore::SVGToOTFFontConverter::overwrite16): Ditto.
(WebCore::SVGToOTFFontConverter::appendCMAPTable): Update to format 12.
(WebCore::SVGToOTFFontConverter::appendHEADTable): Passing in a result vector is no longer necessary.
(WebCore::SVGToOTFFontConverter::appendHHEATable): Ditto.
(WebCore::SVGToOTFFontConverter::appendHMTXTable): Ditto.
(WebCore::SVGToOTFFontConverter::appendMAXPTable): Ditto.
(WebCore::SVGToOTFFontConverter::appendNAMETable): Ditto.
(WebCore::SVGToOTFFontConverter::appendOS2Table): Ditto.
(WebCore::SVGToOTFFontConverter::appendPOSTTable): Ditto.
(WebCore::isValidStringForCFF): Ditto.
(WebCore::SVGToOTFFontConverter::appendCFFValidString): Ditto.
(WebCore::SVGToOTFFontConverter::appendCFFTable): Ditto.
(WebCore::SVGToOTFFontConverter::appendVORGTable): Ditto.
(WebCore::SVGToOTFFontConverter::appendVHEATable): Ditto.
(WebCore::SVGToOTFFontConverter::appendVMTXTable): Ditto.
(WebCore::SVGToOTFFontConverter::addCodepointRanges): Update for Codepoints.
(WebCore::SVGToOTFFontConverter::addCodepoints): Ditto.
(WebCore::SVGToOTFFontConverter::addGlyphNames): Ditto.
(WebCore::SVGToOTFFontConverter::appendKERNSubtable): Ditto.
(WebCore::SVGToOTFFontConverter::appendKERNTable): Ditto.
(WebCore::writeCFFEncodedNumber): Passing in a result vector is no longer necessary.
(WebCore::SVGToOTFFontConverter::appendGlyphData): Update for Codepoints.
(WebCore::SVGToOTFFontConverter::processGlyphElement): Update for Codepoints.
(WebCore::SVGToOTFFontConverter::compareCodepointsLexicographically): Comparing Codepoints
(WebCore::SVGToOTFFontConverter::SVGToOTFFontConverter): Update for Codepoints.
(WebCore::SVGToOTFFontConverter::calculateChecksum): Passing in a result vector is no longer necessary.
(WebCore::SVGToOTFFontConverter::appendTable): Ditto.
(WebCore::SVGToOTFFontConverter::convertSVGToOTFFont): Ditto.
(WebCore::convertSVGToOTFFont): Move out of the result vector.
(WebCore::write32): Deleted.
(WebCore::write16): Deleted.
(WebCore::overwrite32): Deleted.
(WebCore::appendCFFValidString): Deleted.
(WebCore::transcodeGlyphPaths): Deleted.
(WebCore::calculateChecksum): Deleted.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@174279 268f45cc-cd09-0410-ab3c-d52691b4dbfc

5 years agoUnreviewed, rolling out r174270.
bdakin@apple.com [Fri, 3 Oct 2014 20:08:32 +0000 (20:08 +0000)]
Unreviewed, rolling out r174270.
https://bugs.webkit.org/show_bug.cgi?id=137397

crashes on the bots (Requested by dethbakin on #webkit).

Reverted changeset:

"EventSender dispatches should be per-Document"
https://bugs.webkit.org/show_bug.cgi?id=136051
http://trac.webkit.org/changeset/174270

Patch by Commit Queue <commit-queue@webkit.org> on 2014-10-03

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@174278 268f45cc-cd09-0410-ab3c-d52691b4dbfc

5 years ago[iOS] Back-to-back fullscreen videos result in incorrectly sized videos
jer.noble@apple.com [Fri, 3 Oct 2014 20:07:01 +0000 (20:07 +0000)]
[iOS] Back-to-back fullscreen videos result in incorrectly sized videos
https://bugs.webkit.org/show_bug.cgi?id=137380

Reviewed by Eric Carlson.

During a normal workflow, a <video> element will create an AVPlayerLayer, then it will get a
fullscreen (hosted) parent layer, then it will be told an explicit fullscreen size. But when
a <video> element creates an AVPlayerLayer after the <video> element is already in
fullscreen mode, it sets that AVPlayerLayer's frame using the size of the hosted parent
layer, rather than the explicit fullscreen size. Instead, use the (already set) explicit
fullscreen size after creating the AVPlayerLayer.

* platform/graphics/avfoundation/objc/MediaPlayerPrivateAVFoundationObjC.mm:
(WebCore::MediaPlayerPrivateAVFoundationObjC::ensureAVPlayerLayer):
(WebCore::MediaPlayerPrivateAVFoundationObjC::platformLayer):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@174277 268f45cc-cd09-0410-ab3c-d52691b4dbfc

5 years ago[Windows] Settle for 2 chid processes running NRWT on Windows.
roger_fong@apple.com [Fri, 3 Oct 2014 19:58:52 +0000 (19:58 +0000)]
[Windows] Settle for 2 chid processes running NRWT on Windows.
* Scripts/webkitpy/port/win.py:
(WinPort.default_child_processes): 3 child processes reintroduces the sporadic crashing, will settle for 2.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@174276 268f45cc-cd09-0410-ab3c-d52691b4dbfc

5 years agoFTL should sink PutLocals
fpizlo@apple.com [Fri, 3 Oct 2014 19:38:06 +0000 (19:38 +0000)]
FTL should sink PutLocals
https://bugs.webkit.org/show_bug.cgi?id=137168

Reviewed by Oliver Hunt.
Source/JavaScriptCore:

We've known for a while that our PutLocal situation was sub-optimal. We emit them anytime we
"pass" arguments to an inlined function call, because we need to enable the runtime to grab
those arguments when doing foo.arguments where foo is inlined: our engine doesn't deoptimize
in that case but rather just relies on the arguments being flushed (i.e. a copy of their
values is spilled) at a well-known place in a well-known format.

The PutLocals incur two costs: (1) they are store instructions and stores ain't free, and (2)
they look like escaping sites and so they inhibit object allocation sinking.

But in most cases, the PutLocals are unnecessary because the inlined code never performs any
side effect that could transitively lead to function.arguments. Even if the inlined code
could do such a side effect, it may be on a rare path so there is no need to penalize the
entire function.

This patch implements one solution to the PutLocal problem: it aggressively sinks PutLocals
to the latest possible point. This is even more aggressive than the object allocation
sinking. That sinking algorithm avoids creating situations where an object could be
materialized more than one along any path. PutLocal sinking, on the other hand, doesn't avoid
this at all - both to make the phase cheaper and simpler and to make it more aggressive.
Every PutLocal is sunk no matter what.

The upside of this patch is that it eliminates many PutLocals: many of them are sunk "past
their death", thus eliminating them completely. Others are sunk to rare paths. This enables a
lot of object allocation sinking and it removes a lot of pointless store instructions.

It also has downsites. Sinking PutLocals increases register pressure because it increases the
live ranges of things like inlined arguments.

This patch is a net performance win in its current form: 1% SunSpider regression, 2% OctaneV2
progression, 0.6% Kraken regression, 1% AsmBench progression, and 0.5% CompressionBench
regression. The biggest win is on Octane/raytrace, which improves by 27%.

* CMakeLists.txt:
* JavaScriptCore.vcxproj/JavaScriptCore.vcxproj:
* JavaScriptCore.xcodeproj/project.pbxproj:
* bytecode/CodeBlock.h:
* bytecode/Operands.h:
(JSC::Operands::dump): Deleted.
* bytecode/OperandsInlines.h:
(JSC::Traits>::dump):
* bytecode/VirtualRegister.h:
(JSC::VirtualRegister::isHeader):
* dfg/DFGByteCodeParser.cpp:
(JSC::DFG::ByteCodeParser::InlineStackEntry::InlineStackEntry):
* dfg/DFGClobberSet.h:
(JSC::DFG::ClobberSetAdd::operator()):
(JSC::DFG::ClobberSetOverlaps::operator()):
* dfg/DFGClobberize.h:
(JSC::DFG::clobberize):
(JSC::DFG::NoOpClobberize::operator()):
(JSC::DFG::CheckClobberize::operator()):
(JSC::DFG::AbstractHeapOverlaps::operator()):
(JSC::DFG::ReadMethodClobberize::operator()):
(JSC::DFG::WriteMethodClobberize::operator()):
(JSC::DFG::DefMethodClobberize::operator()):
* dfg/DFGFlushFormat.h:
(JSC::DFG::merge):
* dfg/DFGGraph.cpp:
(JSC::DFG::Graph::Graph):
* dfg/DFGGraph.h:
(JSC::DFG::Graph::capturedVarsFor):
* dfg/DFGObjectAllocationSinkingPhase.cpp:
(JSC::DFG::ObjectAllocationSinkingPhase::determineMaterializationPoints):
(JSC::DFG::ObjectAllocationSinkingPhase::placeMaterializationPoints):
* dfg/DFGPlan.cpp:
(JSC::DFG::Plan::compileInThreadImpl):
* dfg/DFGPreciseLocalClobberize.h: Added.
(JSC::DFG::PreciseLocalClobberizeAdaptor::PreciseLocalClobberizeAdaptor):
(JSC::DFG::PreciseLocalClobberizeAdaptor::read):
(JSC::DFG::PreciseLocalClobberizeAdaptor::write):
(JSC::DFG::PreciseLocalClobberizeAdaptor::def):
(JSC::DFG::PreciseLocalClobberizeAdaptor::callIfAppropriate):
(JSC::DFG::PreciseLocalClobberizeAdaptor::readTop):
(JSC::DFG::PreciseLocalClobberizeAdaptor::writeTop):
(JSC::DFG::forEachLocalReadByUnwind):
(JSC::DFG::preciseLocalClobberize):
* dfg/DFGPutLocalSinkingPhase.cpp: Added.
(JSC::DFG::performPutLocalSinking):
* dfg/DFGPutLocalSinkingPhase.h: Added.
* dfg/DFGSSACalculator.h:
(JSC::DFG::SSACalculator::computePhis):
* dfg/DFGValidate.cpp:

Source/WTF:

Make the set bits of a BitVector iterable.

* wtf/BitVector.h:
(WTF::BitVector::SetBitsIterable::SetBitsIterable):
(WTF::BitVector::SetBitsIterable::iterator::iterator):
(WTF::BitVector::SetBitsIterable::iterator::operator*):
(WTF::BitVector::SetBitsIterable::iterator::operator++):
(WTF::BitVector::SetBitsIterable::iterator::operator==):
(WTF::BitVector::SetBitsIterable::iterator::operator!=):
(WTF::BitVector::SetBitsIterable::begin):
(WTF::BitVector::SetBitsIterable::end):
(WTF::BitVector::setBits):

LayoutTests:

* js/regress/elidable-new-object-then-call-expected.txt: Added.
* js/regress/elidable-new-object-then-call.html: Added.
* js/regress/script-tests/elidable-new-object-then-call.js: Added.
(sumOfArithSeries):
(bar):
(foo):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@174275 268f45cc-cd09-0410-ab3c-d52691b4dbfc

5 years ago[GTK] Bump up and patch dependencies to allow building for aarch64
commit-queue@webkit.org [Fri, 3 Oct 2014 18:38:40 +0000 (18:38 +0000)]
[GTK] Bump up and patch dependencies to allow building for aarch64
https://bugs.webkit.org/show_bug.cgi?id=137147

Patch by Akos Kiss <akiss@inf.u-szeged.hu> on 2014-10-03
Reviewed by Martin Robinson.

The currently used versions of pixman, libffi, librsvg, glib,
glib-networking, fontconfig, libxml2, gst-plugins-base fail to configure
and/or build for aarch64, so they are bumped up and patched where
necessary. Wherever possible, the same versions and patches were used as
for the EFL port bump-up
(https://bugs.webkit.org/show_bug.cgi?id=135885).

After the bump-up, a hitherto hidden bug in gst-plugins-good caused a
regression by crashing http/tests/media/video-redirect.html . A patch
already in upstream fixes that bug.

This change does not touch xserver, which also needs bumping up, but
will be dealt with later.

* gtk/jhbuild.modules:
Specified versions and patches necessary to correctly build for aarch64.
* gtk/patches/gst-events-arent-gstobjects.patch:
Added. Patches gst-plugins-good to fix a bug in .mov file demuxing.
* gtk/patches/gst-prevent-neon-check-in-configure-from-passing-under-aarch64.patch:
Added. Patches gst-plugins-base to configure correctly for aarch64.
* gtk/patches/librsvg-2.36.1-bump-up-config.guess-to-support-aarch64.patch:
Added. Bumps up config.guess and config.sub in librsvg to recognize
aarch64.
* gtk/patches/libxml2-2.9.0-dtd.patch:
Removed. Not necessary for libxml2 v2.9.1 anymore.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@174274 268f45cc-cd09-0410-ab3c-d52691b4dbfc

5 years agoREGRESSION (r173531): Use after free in WebCore::RenderStyle::fontMetrics /
betravis@adobe.com [Fri, 3 Oct 2014 18:35:53 +0000 (18:35 +0000)]
REGRESSION (r173531): Use after free in WebCore::RenderStyle::fontMetrics /
WebCore::CSSPrimitiveValue::computeLengthDouble
https://bugs.webkit.org/show_bug.cgi?id=136864

Reviewed by Andreas Kling.

Source/WebCore:

FontLoader previously called updateDocumentStyleIfNeeded,
which would reset styles currently in use as part of
the tabIndex calculation. The FontLoader should instead
wait for pending stylesheets to load.

Tests: fast/css/fontloader-tab-index.html

* css/FontLoader.cpp:
(WebCore::FontLoader::notifyWhenFontsReady): Do not immediately
call loadingDone().
(WebCore::FontLoader::loadingDone): Wait for stylesheets to
finish loading rather than updating document styles.
* css/FontLoader.h:
(WebCore::FontLoader::loading): Include JS font loads when testing
for the loading state.

LayoutTests:

Test that getting the tab index on a body element with
font-relative measurements to a local @font-face do not
cause a crash.

* fast/css/fontloader-tab-index-expected.html: Added.
* fast/css/fontloader-tab-index.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@174273 268f45cc-cd09-0410-ab3c-d52691b4dbfc

5 years agoUnreviewed build fix.
mmaxfield@apple.com [Fri, 3 Oct 2014 17:55:17 +0000 (17:55 +0000)]
Unreviewed build fix.

Fixing the Windows build after r174269.

Patch by Myles C. Maxfield <mmaxfield@apple.com> on 2014-10-03

* platform/graphics/win/UniscribeController.cpp:
(WebCore::UniscribeController::shapeAndPlaceItem):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@174272 268f45cc-cd09-0410-ab3c-d52691b4dbfc

5 years agoSupport modern for loops over StringViews
mmaxfield@apple.com [Fri, 3 Oct 2014 17:43:32 +0000 (17:43 +0000)]
Support modern for loops over StringViews
https://bugs.webkit.org/show_bug.cgi?id=137165

Patch by Myles C. Maxfield <mmaxfield@apple.com> on 2014-10-03
Reviewed by Darin Adler.

This patch adds two functions, codePoints() and codeUnits(), on StringView.
These two functions return small objects which have begin() and end() functions,
which means it can be used by the modern for loop style. This small class also
has an inner iterator class which can be incremented, dereferenced, and
compared.

Using these new objects looks like this:
for (UChar codeunit : stringView.codeUnits()) { } and
for (UChar32 codepoint : stringView.codePoints()) { }.

* wtf/text/StringView.h:
(WTF::StringView::codepoints):
(WTF::StringView::codeunits):
(WTF::StringViewCodePointIterator::StringViewCodePointIterator):
(WTF::StringViewCodePointIterator::Iterator::Iterator):
(WTF::StringViewCodePointIterator::Iterator::operator*):
(WTF::StringViewCodePointIterator::Iterator::operator!=):
(WTF::StringViewCodeUnitIterator::StringViewCodeUnitIterator):
(WTF::StringViewCodeUnitIterator::Iterator::Iterator):
(WTF::StringViewCodeUnitIterator::Iterator::operator*):
(WTF::StringViewCodeUnitIterator::Iterator::operator!=):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@174271 268f45cc-cd09-0410-ab3c-d52691b4dbfc

5 years agoEventSender dispatches should be per-Document
burg@cs.washington.edu [Fri, 3 Oct 2014 17:15:51 +0000 (17:15 +0000)]
EventSender dispatches should be per-Document
https://bugs.webkit.org/show_bug.cgi?id=136051

Reviewed by Andreas Kling.

EventSender uses a Timer to asynchronously dispatch the same event to multiple
elements of the same type on a future run loop. However, we previously dispatched events
to elements from any Document rather than the one that requested the event send.

This patch changes EventSender to be associated with a single Document. EventSenders
for each Document are lazily allocated. Further per-Document or global optimizations
are possible.

This change also lays the groundwork for making EventSender-related DOM events
deterministic during web replay, tracked at <https://webkit.org/b/137090>.

No new tests, covered by fast/text/svg-font-trigger-load-event.html.

* CMakeLists.txt:
* WebCore.vcxproj/WebCore.vcxproj:
* WebCore.vcxproj/WebCore.vcxproj.filters:
* WebCore.xcodeproj/project.pbxproj:
* dom/DOMAllInOne.cpp:
* dom/Document.cpp:
(WebCore::Document::Document):
(WebCore::Document::implicitClose):
Dispatch pending events for this document only. Rename local variable f to frame.

* dom/Document.h:
(WebCore::Document::sharedEventSenders):
* dom/EventSender.h:
(WebCore::EventSender::EventSender):
(WebCore::EventSender::hasPendingEvents):
(WebCore::EventSender<T>::dispatchEventSoon):
(WebCore::EventSender<T>::cancelEvent):
(WebCore::EventSender<T>::dispatchPendingEvents):
(WebCore::EventSender<T>::EventSender): Deleted. The type argument was not necessary.
* dom/SharedEventSenders.cpp: Added. Lazily creates event senders as they are requested.
(WebCore::SharedEventSenders::linkLoadEventSender):
(WebCore::SharedEventSenders::styleLoadEventSender):
(WebCore::SharedEventSenders::imageBeforeloadEventSender):
(WebCore::SharedEventSenders::imageLoadEventSender):
(WebCore::SharedEventSenders::imageErrorEventSender):
* dom/SharedEventSenders.h: Added.
(WebCore::SharedEventSenders::SharedEventSenders):
* html/HTMLLinkElement.cpp:
(WebCore::HTMLLinkElement::~HTMLLinkElement):
(WebCore::HTMLLinkElement::dispatchPendingEvent):
(WebCore::HTMLLinkElement::notifyLoadedSheetAndAllCriticalSubresources):
(WebCore::linkLoadEventSender): Deleted.
(WebCore::HTMLLinkElement::dispatchPendingLoadEvents): Deleted.
* html/HTMLLinkElement.h:
* html/HTMLStyleElement.cpp:
(WebCore::HTMLStyleElement::~HTMLStyleElement):
(WebCore::HTMLStyleElement::dispatchPendingEvent):
(WebCore::HTMLStyleElement::notifyLoadedSheetAndAllCriticalSubresources):
(WebCore::styleLoadEventSender): Deleted.
(WebCore::HTMLStyleElement::dispatchPendingLoadEvents): Deleted.
* html/HTMLStyleElement.h:
* loader/ImageLoader.cpp:
(WebCore::ImageLoader::beforeLoadEventSender):
(WebCore::ImageLoader::loadEventSender):
(WebCore::ImageLoader::errorEventSender):
(WebCore::ImageLoader::dispatchPendingEvent):
(WebCore::beforeLoadEventSender): Deleted.
(WebCore::loadEventSender): Deleted.
(WebCore::errorEventSender): Deleted.
(WebCore::ImageLoader::dispatchPendingBeforeLoadEvents): Deleted.
(WebCore::ImageLoader::dispatchPendingLoadEvents): Deleted.
(WebCore::ImageLoader::dispatchPendingErrorEvents): Deleted.
* loader/ImageLoader.h:
* xml/parser/XMLDocumentParser.cpp:
(WebCore::XMLDocumentParser::append): Remove an early call to the EventSender for image
beforeload events. This was a performance optimization, but is no longer safe since the
document (and thus the event sender) may not be available in a synchronous document write.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@174270 268f45cc-cd09-0410-ab3c-d52691b4dbfc

5 years agoTextRun::length() should return an unsigned
mmaxfield@apple.com [Fri, 3 Oct 2014 16:57:10 +0000 (16:57 +0000)]
TextRun::length() should return an unsigned
https://bugs.webkit.org/show_bug.cgi?id=137354

Patch by Myles C. Maxfield <mmaxfield@apple.com> on 2014-10-03
Reviewed by Antti Koivisto.

Currently, the m_len instance variable that backs the TextRun::length() function is
an unsigned, but the length() function implicitly casts that to an int when
returning it. This patch makes the function return an unsigned, and makes the
do any casting if necessary.

No new tests because there is no behavior change.

* platform/graphics/Font.cpp:
(WebCore::Font::drawText): Cast result to int.
(WebCore::Font::drawEmphasisMarks): Ditto.
(WebCore::Font::adjustSelectionRectForText): Ditto.
(WebCore::computeUnderlineType): Update internal type to be unsigned.
* platform/graphics/GlyphBuffer.h:
(WebCore::GlyphBuffer::add): Update sentinel value to be unsigned.
(WebCore::GlyphBuffer::saveOffsetsInString): Ditto.
* platform/graphics/GraphicsContext.cpp: Update internal type to be unsigned.
* platform/graphics/TextRun.h:
(WebCore::TextRun::length): Update return type.
(WebCore::TextRun::charactersLength): Ditto.
* platform/graphics/WidthIterator.cpp:
(WebCore::WidthIterator::advanceInternal): Cast result to int.
* rendering/svg/SVGTextMetricsBuilder.cpp:
(WebCore::SVGTextMetricsBuilder::currentCharacterStartsSurrogatePair): Remove
unnecessary casts.
(WebCore::SVGTextMetricsBuilder::advance): Ditto.
* svg/SVGFontData.cpp:
(WebCore::SVGFontData::applySVGGlyphSelection): Cast result to int.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@174269 268f45cc-cd09-0410-ab3c-d52691b4dbfc

5 years agoRemove RefPtrHashMap
weinig@apple.com [Fri, 3 Oct 2014 16:47:06 +0000 (16:47 +0000)]
Remove RefPtrHashMap
https://bugs.webkit.org/show_bug.cgi?id=137200

Reviewed by Darin Adler.

Source/WTF:

HashMap now has all the features of the RefPtr specialization of HashMap
(which lived in RefPtrHashMap.h) and is in fact more efficient when moving
items into it. For instance:

For HashMap<RefPtr<Foo>, Bar> map:

RefPtr<Foo> foo = ...;
map.add(WTF::move(foo), Bar());

will now not cause a ref() / deref(), but will successfully move the item in
without churn. The same is true if foo.release() was used.

* WTF.vcxproj/WTF.vcxproj:
* WTF.vcxproj/WTF.vcxproj.filters:
* WTF.xcodeproj/project.pbxproj:
* wtf/CMakeLists.txt:
Remove references to RefPtrHashMap.h.

* wtf/RefPtrHashMap.h: Removed.

* wtf/HashMap.h:
Remove #include of RefPtrHashMap.h and add inlineGet to match the interface
exposed by the RefPtrHashMap specialization.

Tools:

* TestWebKitAPI/Tests/WTF/HashMap.cpp:
(TestWebKitAPI::TEST):
Add tests for using a HashMap with RefPtr keys, focusing on ensuring that
a minimum of refcount churn occurs.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@174268 268f45cc-cd09-0410-ab3c-d52691b4dbfc

5 years ago[GTK] Unreviewed GTK gardening.
clopez@igalia.com [Fri, 3 Oct 2014 16:43:06 +0000 (16:43 +0000)]
[GTK] Unreviewed GTK gardening.

* platform/gtk/TestExpectations: Update test name for bug 136673 after r173418.
* platform/gtk/css1/box_properties/clear_float-expected.txt: Rebaseline after r174233.
* platform/gtk/css1/box_properties/float_on_text_elements-expected.txt: Rebaseline after r174233.
* platform/gtk/css2.1/t0905-c5525-fltcont-00-d-g-expected.txt: Rebaseline after r174233.
* platform/gtk/css2.1/t0905-c5525-fltwidth-00-c-g-expected.txt: Rebaseline after r174233.
* platform/gtk/css2.1/t1602-c546-txt-align-00-b-expected.txt: Rebaseline after r174233.
* platform/gtk/fast/multicol/float-avoidance-expected.txt: Rebaseline after r174233.
* platform/gtk/fast/multicol/overflow-across-columns-expected.txt: Rebaseline after r174233.
* platform/gtk/fast/multicol/overflow-across-columns-percent-height-expected.txt: Rebaseline after r174233.
* platform/gtk/fast/multicol/overflow-unsplittable-expected.txt: Rebaseline after r174233.
* platform/gtk/fast/multicol/positive-leading-expected.txt: Rebaseline after r174233.
* platform/gtk/fast/multicol/vertical-lr/float-avoidance-expected.txt: Rebaseline after r174233.
* platform/gtk/fast/multicol/vertical-rl/float-avoidance-expected.txt: Rebaseline after r174233.
* platform/gtk/fast/regions/multiple-directionality-changes-in-variable-width-regions-expected.txt: Rebaseline after r174233.
* platform/gtk/fast/regions/repaint/line-flow-with-floats-in-regions-expected.txt: Rebaseline after r174233.
* platform/gtk/fast/repaint/line-flow-with-floats-1-expected.txt: Rebaseline after r174233.
* platform/gtk/fast/repaint/line-flow-with-floats-10-expected.txt: Rebaseline after r174233.
* platform/gtk/fast/repaint/line-flow-with-floats-2-expected.txt: Rebaseline after r174233.
* platform/gtk/fast/repaint/line-flow-with-floats-3-expected.txt: Rebaseline after r174233.
* platform/gtk/fast/repaint/line-flow-with-floats-4-expected.txt: Rebaseline after r174233.
* platform/gtk/fast/repaint/line-flow-with-floats-5-expected.txt: Rebaseline after r174233.
* platform/gtk/fast/repaint/line-flow-with-floats-6-expected.txt: Rebaseline after r174233.
* platform/gtk/fast/repaint/line-flow-with-floats-7-expected.txt: Rebaseline after r174233.
* platform/gtk/fast/repaint/line-flow-with-floats-8-expected.txt: Rebaseline after r174233.
* platform/gtk/fast/repaint/line-flow-with-floats-9-expected.txt: Rebaseline after r174233.
* platform/gtk/fast/text/justified-selection-at-edge-expected.txt: Rebaseline after r174233.
* platform/gtk/fast/text/justify-ideograph-leading-expansion-expected.txt: Rebaseline after r174233.
* platform/gtk/fast/text/justify-nbsp-expected.txt: Added. Rebaseline after r174233.
* platform/gtk/fast/text/trailing-white-space-2-expected.txt: Rebaseline after r174233.
* platform/gtk/tables/mozilla/marvin/x_tbody_align_justify-expected.txt: Rebaseline after r174233.
* platform/gtk/tables/mozilla/marvin/x_td_align_justify-expected.txt: Rebaseline after r174233.
* platform/gtk/tables/mozilla/marvin/x_tfoot_align_justify-expected.txt: Rebaseline after r174233.
* platform/gtk/tables/mozilla/marvin/x_th_align_justify-expected.txt: Rebaseline after r174233.
* platform/gtk/tables/mozilla/marvin/x_thead_align_justify-expected.txt: Rebaseline after r174233.
* platform/gtk/tables/mozilla/marvin/x_tr_align_justify-expected.txt: Rebaseline after r174233.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@174267 268f45cc-cd09-0410-ab3c-d52691b4dbfc

5 years agoRemove workarounds for versions of Python before 2.7
evab.u-szeged@partner.samsung.com [Fri, 3 Oct 2014 15:03:27 +0000 (15:03 +0000)]
Remove workarounds for versions of Python before 2.7
https://bugs.webkit.org/show_bug.cgi?id=137338

Reviewed by Csaba Osztrogon√°c.

* Scripts/webkitpy/common/system/autoinstall.py:
(AutoInstaller._extract_targz):
(AutoInstaller._unzip):
(AutoInstaller): Deleted.
(AutoInstaller._extract_all): Deleted.
* Scripts/webkitpy/common/webkitunittest.py:
(TestCase.setUp): Deleted.
* Scripts/webkitpy/layout_tests/models/test_configuration.py:
(TestConfigurationConverter.collapse_macros):
(TestConfigurationConverter.to_specifiers_list.try_collapsing):
(TestConfigurationConverter.to_specifiers_list.try_abbreviating):
(TestConfigurationConverter):
(TestConfigurationConverter.combinations): Deleted.
* Scripts/webkitpy/layout_tests/models/test_expectations_unittest.py:
* Scripts/webkitpy/port/base.py:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@174266 268f45cc-cd09-0410-ab3c-d52691b4dbfc

5 years agoUnreviewed, rolling out r174264.
evab.u-szeged@partner.samsung.com [Fri, 3 Oct 2014 12:26:32 +0000 (12:26 +0000)]
Unreviewed, rolling out r174264.
https://bugs.webkit.org/show_bug.cgi?id=137392

It broke run-webkit-tests (Requested by ebalazsfalvi on
#webkit).

Reverted changeset:

"Remove workarounds for versions of Python before 2.7"
https://bugs.webkit.org/show_bug.cgi?id=137338
http://trac.webkit.org/changeset/174264

Patch by Commit Queue <commit-queue@webkit.org> on 2014-10-03

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@174265 268f45cc-cd09-0410-ab3c-d52691b4dbfc

5 years agoRemove workarounds for versions of Python before 2.7
evab.u-szeged@partner.samsung.com [Fri, 3 Oct 2014 10:14:36 +0000 (10:14 +0000)]
Remove workarounds for versions of Python before 2.7
https://bugs.webkit.org/show_bug.cgi?id=137338

Reviewed by Csaba Osztrogon√°c.

* Scripts/webkitpy/common/system/autoinstall.py:
(AutoInstaller._extract_targz):
(AutoInstaller._unzip):
(AutoInstaller): Deleted.
(AutoInstaller._extract_all): Deleted.
* Scripts/webkitpy/common/webkitunittest.py:
(TestCase.setUp): Deleted.
* Scripts/webkitpy/layout_tests/models/test_configuration.py:
(TestConfigurationConverter.collapse_macros):
(TestConfigurationConverter.to_specifiers_list.try_collapsing):
(TestConfigurationConverter.to_specifiers_list.try_abbreviating):
(TestConfigurationConverter):
(TestConfigurationConverter.combinations): Deleted.
* Scripts/webkitpy/layout_tests/models/test_expectations_unittest.py:
* Scripts/webkitpy/port/base.py:
* Scripts/webkitpy/port/driver.py:
(ContentBlock.__init__):
* Scripts/webkitpy/port/server_process.py:
(ServerProcess._reset):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@174264 268f45cc-cd09-0410-ab3c-d52691b4dbfc

5 years agoRemove autoinstaller workaround for python 2.6
evab.u-szeged@partner.samsung.com [Fri, 3 Oct 2014 10:13:41 +0000 (10:13 +0000)]
Remove autoinstaller workaround for python 2.6
https://bugs.webkit.org/show_bug.cgi?id=137343

Reviewed by Daniel Bates.

Reverting <http://trac.webkit.org/changeset/124928>, neither Python
2.6, nor Snow Leopard is supported anymore, so the workaround is not
needed anymore.

* Scripts/webkitpy/common/system/autoinstall.py:
(AutoInstaller.install):
* Scripts/webkitpy/test/main.py:
(Tester._run_tests):
* Scripts/webkitpy/thirdparty/__init__.py:
(AutoinstallImportHook._install_mechanize):
(AutoinstallImportHook._install_pep8):
(AutoinstallImportHook._install_pylint):
(AutoinstallImportHook._install_buildbot):
(AutoinstallImportHook._install_coverage):
(AutoinstallImportHook._install_eliza):
(AutoinstallImportHook._install_irc):
(AutoinstallImportHook._install):
(autoinstall_everything):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@174263 268f45cc-cd09-0410-ab3c-d52691b4dbfc

5 years ago[GTK] URTBF after r174231.
ossy@webkit.org [Fri, 3 Oct 2014 09:25:08 +0000 (09:25 +0000)]
[GTK] URTBF after r174231.

* WebProcess/InjectedBundle/API/gtk/WebKitWebPage.cpp:
(webkit_web_page_get_dom_document):
* WebProcess/WebPage/DrawingAreaImpl.cpp:
(WebKit::DrawingAreaImpl::didInstallPageOverlay): Deleted.
(WebKit::DrawingAreaImpl::didUninstallPageOverlay): Deleted.
(WebKit::DrawingAreaImpl::setPageOverlayNeedsDisplay): Deleted.
(WebKit::DrawingAreaImpl::setPageOverlayOpacity): Deleted.
* WebProcess/WebPage/DrawingAreaImpl.h:
* WebProcess/WebPage/gtk/LayerTreeHostGtk.cpp:
(WebKit::LayerTreeHostGtk::createPageOverlayLayer):
(WebKit::LayerTreeHostGtk::destroyPageOverlayLayer):
(WebKit::LayerTreeHostGtk::didInstallPageOverlay): Deleted.
(WebKit::LayerTreeHostGtk::didUninstallPageOverlay): Deleted.
(WebKit::LayerTreeHostGtk::setPageOverlayNeedsDisplay): Deleted.
* WebProcess/WebPage/gtk/LayerTreeHostGtk.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@174262 268f45cc-cd09-0410-ab3c-d52691b4dbfc

5 years ago[EFL] Fix build break since r174231 and r174256
gyuyoung.kim@samsung.com [Fri, 3 Oct 2014 08:00:43 +0000 (08:00 +0000)]
[EFL] Fix build break since r174231 and r174256
https://bugs.webkit.org/show_bug.cgi?id=137384

Unreviewed, EFL build fix.

r174231 and r174256 didn't consider EFL port and CoordinatedGraphics area.

Source/WebCore:

* css/WebKitCSSViewportRule.cpp:
(WebCore::WebKitCSSViewportRule::WebKitCSSViewportRule):
* css/WebKitCSSViewportRule.h:

Source/WebKit2:

* WebProcess/WebPage/CoordinatedGraphics/CoordinatedDrawingArea.cpp:
(WebKit::CoordinatedDrawingArea::mainFrameContentSizeChanged):
* WebProcess/WebPage/CoordinatedGraphics/CoordinatedLayerTreeHost.cpp:
(WebKit::CoordinatedLayerTreeHost::setRootCompositingLayer):
(WebKit::CoordinatedLayerTreeHost::didFlushRootLayer):
(WebKit::CoordinatedLayerTreeHost::deviceOrPageScaleFactorChanged):
* WebProcess/WebPage/TapHighlightController.cpp:
(WebKit::TapHighlightController::highlight):
(WebKit::TapHighlightController::hideHighlight):
* WebProcess/WebPage/TapHighlightController.h: Add WebCore:: namespace prefix to PageOverlay argument type.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@174261 268f45cc-cd09-0410-ab3c-d52691b4dbfc

5 years agoChange how 32-bit JSValues check if they are a Boolean
saambarati1@gmail.com [Fri, 3 Oct 2014 07:53:08 +0000 (07:53 +0000)]
Change how 32-bit JSValues check if they are a Boolean

Rubber stamped by Filip Pizlo.

32-bit JSValue::isBoolean can simply check if its tag corresponds
to the boolean tag instead of checking if it's either true or false.

* runtime/JSCJSValueInlines.h:
(JSC::JSValue::isBoolean):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@174260 268f45cc-cd09-0410-ab3c-d52691b4dbfc

5 years agoCSS Selectors Level 4: Add parsing for :matches
utatane.tea@gmail.com [Fri, 3 Oct 2014 07:13:02 +0000 (07:13 +0000)]
CSS Selectors Level 4: Add parsing for :matches
https://bugs.webkit.org/show_bug.cgi?id=137348

Reviewed by Benjamin Poulain.

Source/WebCore:

Add initial parsing functionality of :matches pseudo class in Selectors Level 4.
It accepts selector list as a parameter. It can contain pseudo elements.
And now at the parsing phase, multiple pseudo elements are allowed.
Currently :not doesn't accept any functional pseudo classes, :not(:matches(...)) is rejected.
And currently, :matches(:visited, :link) is allowed in the parsing phase.

* css/CSSGrammar.y.in:
* css/CSSParser.cpp:
(WebCore::CSSParser::detectFunctionTypeToken):
* css/CSSSelector.cpp:
(WebCore::appendSelectorList):
(WebCore::CSSSelector::selectorText):
* css/CSSSelector.h:
* css/SelectorChecker.cpp:
(WebCore::SelectorChecker::checkOne):
* css/SelectorPseudoClassAndCompatibilityElementMap.in:
* cssjit/SelectorCompiler.cpp:
(WebCore::SelectorCompiler::addPseudoClassType):

LayoutTests:

* fast/css/css-selector-text-expected.txt:
* fast/css/css-selector-text.html:
* fast/css/css-set-selector-text-expected.txt:
* fast/css/css-set-selector-text.html:
Cover CSSOM tests; canonicalizing :matches text representation.

* fast/css/parsing-css-matches-1-expected.txt: Added.
* fast/css/parsing-css-matches-1.html: Added.
* fast/css/parsing-css-matches-2-expected.txt: Added.
* fast/css/parsing-css-matches-2.html: Added.
* fast/css/parsing-css-matches-3-expected.txt: Added.
* fast/css/parsing-css-matches-3.html: Added.
* fast/css/parsing-css-matches-4-expected.txt: Added.
* fast/css/parsing-css-matches-4.html: Added.
* fast/css/parsing-css-matches-5-expected.txt: Added.
* fast/css/parsing-css-matches-5.html: Added.
* fast/css/parsing-css-matches-6-expected.txt: Added.
* fast/css/parsing-css-matches-6.html: Added.
Valid :matches are tested in the above tests. Selector lists with 1, 2, 3 elements are tested.

* fast/css/parsing-css-matches-7-expected.txt: Added.
* fast/css/parsing-css-matches-7.html: Added.
* fast/css/parsing-css-matches-8-expected.txt: Added.
* fast/css/parsing-css-matches-8.html: Added.
Invalid :matches are tested in the above tests. :matches() etc. are tested.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@174259 268f45cc-cd09-0410-ab3c-d52691b4dbfc

5 years agoURTBF after r174231 to fix non Apple builds.
ossy@webkit.org [Fri, 3 Oct 2014 05:39:09 +0000 (05:39 +0000)]
URTBF after r174231 to fix non Apple builds.

* WebProcess/WebPage/WebPage.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@174257 268f45cc-cd09-0410-ab3c-d52691b4dbfc

5 years agoCSSOM rule wrappers always have a corresponding internal style object.
akling@apple.com [Fri, 3 Oct 2014 04:38:19 +0000 (04:38 +0000)]
CSSOM rule wrappers always have a corresponding internal style object.
<https://webkit.org/b/137379>

Make the CSSOM rule wrapper classes use references & Ref instead of
raw pointers and RefPtr for pointing to the internal style rule.

Also changed the signature of reattach() to take a StyleRuleBase&
since it's never called with a null pointer.

Finally marked the rule classes final.

Reviewed by Geoffrey Garen.

* css/CSSCharsetRule.h:
* css/CSSFontFaceRule.cpp:
(WebCore::CSSFontFaceRule::CSSFontFaceRule):
(WebCore::CSSFontFaceRule::reattach):
* css/CSSFontFaceRule.h:
(WebCore::CSSFontFaceRule::create): Deleted.
* css/CSSGroupingRule.cpp:
(WebCore::CSSGroupingRule::CSSGroupingRule):
(WebCore::CSSGroupingRule::reattach):
* css/CSSGroupingRule.h:
* css/CSSImportRule.cpp:
(WebCore::CSSImportRule::CSSImportRule):
(WebCore::CSSImportRule::href):
(WebCore::CSSImportRule::media):
(WebCore::CSSImportRule::cssText):
(WebCore::CSSImportRule::styleSheet):
(WebCore::CSSImportRule::reattach):
* css/CSSImportRule.h:
(WebCore::CSSImportRule::create): Deleted.
* css/CSSMediaRule.cpp:
(WebCore::CSSMediaRule::CSSMediaRule):
(WebCore::CSSMediaRule::mediaQueries):
(WebCore::CSSMediaRule::reattach):
* css/CSSMediaRule.h:
(WebCore::CSSMediaRule::create): Deleted.
* css/CSSPageRule.cpp:
(WebCore::CSSPageRule::CSSPageRule):
(WebCore::CSSPageRule::reattach):
* css/CSSPageRule.h:
(WebCore::CSSPageRule::create): Deleted.
* css/CSSRule.h:
* css/CSSStyleRule.cpp:
(WebCore::CSSStyleRule::CSSStyleRule):
(WebCore::CSSStyleRule::reattach):
* css/CSSStyleRule.h:
(WebCore::CSSStyleRule::create): Deleted.
(WebCore::CSSStyleRule::styleRule): Deleted.
* css/CSSStyleSheet.cpp:
(WebCore::CSSStyleSheet::reattachChildRuleCSSOMWrappers):
* css/CSSSupportsRule.cpp:
(WebCore::CSSSupportsRule::CSSSupportsRule):
(WebCore::CSSSupportsRule::conditionText):
* css/CSSSupportsRule.h:
(WebCore::CSSSupportsRule::create):
* css/CSSUnknownRule.h:
* css/StyleRule.cpp:
(WebCore::StyleRuleBase::createCSSOMWrapper):
* css/StyleRuleImport.h:
(WebCore::StyleRuleImport::mediaQueries):
* css/WebKitCSSKeyframeRule.cpp:
(WebCore::WebKitCSSKeyframeRule::WebKitCSSKeyframeRule):
(WebCore::WebKitCSSKeyframeRule::reattach):
* css/WebKitCSSKeyframeRule.h:
* css/WebKitCSSKeyframesRule.cpp:
(WebCore::WebKitCSSKeyframesRule::WebKitCSSKeyframesRule):
(WebCore::WebKitCSSKeyframesRule::item):
(WebCore::WebKitCSSKeyframesRule::reattach):
* css/WebKitCSSKeyframesRule.h:
(WebCore::WebKitCSSKeyframesRule::create):
* css/WebKitCSSRegionRule.cpp:
(WebCore::WebKitCSSRegionRule::WebKitCSSRegionRule):
(WebCore::WebKitCSSRegionRule::cssText):
* css/WebKitCSSRegionRule.h:
(WebCore::WebKitCSSRegionRule::create):
* css/WebKitCSSViewportRule.cpp:
(WebCore::WebKitCSSViewportRule::reattach):
* css/WebKitCSSViewportRule.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@174256 268f45cc-cd09-0410-ab3c-d52691b4dbfc

5 years agoAdd back debugging/testing code that I accidentally removed.
andersca@apple.com [Fri, 3 Oct 2014 02:49:15 +0000 (02:49 +0000)]
Add back debugging/testing code that I accidentally removed.

* wtf/text/StringConcatenate.h:
(WTF::StringTypeAdapter<String>::writeTo):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@174255 268f45cc-cd09-0410-ab3c-d52691b4dbfc

5 years agoWeb Inspector: Cannot add a new CSS Rule to elements with 3 or more class names
commit-queue@webkit.org [Fri, 3 Oct 2014 02:07:17 +0000 (02:07 +0000)]
Web Inspector: Cannot add a new CSS Rule to elements with 3 or more class names
https://bugs.webkit.org/show_bug.cgi?id=137375

Patch by Joseph Pecoraro <pecoraro@apple.com> on 2014-10-02
Reviewed by Timothy Hatcher.

Correctly merge all class names with dots instead of just the first two.

* UserInterface/Models/DOMNode.js:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@174254 268f45cc-cd09-0410-ab3c-d52691b4dbfc

5 years agoUse is<>() / downcast<>() for CSSRule subclasses
cdumez@apple.com [Fri, 3 Oct 2014 01:40:12 +0000 (01:40 +0000)]
Use is<>() / downcast<>() for CSSRule subclasses
https://bugs.webkit.org/show_bug.cgi?id=137364

Reviewed by Benjamin Poulain.

Use is<>() / downcast<>() functions for CSSRule subclasses.

No new tests, no behavior change.

* css/CSSFontFaceRule.h:
* css/CSSImportRule.h:
* css/CSSMediaRule.h:
* css/CSSRule.h:
* css/CSSStyleRule.h:
* css/CSSSupportsRule.h:
* css/InspectorCSSOMWrappers.cpp:
(WebCore::InspectorCSSOMWrappers::collect):
* css/WebKitCSSKeyframesRule.h:
* css/WebKitCSSRegionRule.h:
* inspector/InspectorCSSAgent.cpp:
(WebCore::InspectorCSSAgent::asCSSStyleRule):
(WebCore::InspectorCSSAgent::collectStyleSheets):
(WebCore::InspectorCSSAgent::buildArrayForRuleList):
* inspector/InspectorStyleSheet.cpp:
(WebCore::asCSSRuleList):
(WebCore::fillMediaListChain):
(WebCore::InspectorStyleSheet::addRule):
(WebCore::InspectorStyleSheet::collectFlatRules):
* page/PageSerializer.cpp:
(WebCore::PageSerializer::serializeCSSStyleSheet):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@174253 268f45cc-cd09-0410-ab3c-d52691b4dbfc

5 years ago<rdar://problem/18531395> REGRESSION (r173272): Crash in ResourceResponse::platformSu...
mitz@apple.com [Fri, 3 Oct 2014 00:37:47 +0000 (00:37 +0000)]
<rdar://problem/18531395> REGRESSION (r173272): Crash in ResourceResponse::platformSuggestedFilename() when called on the null response
https://bugs.webkit.org/show_bug.cgi?id=137239

Reviewed by Pratik Solanki.

* platform/network/cf/ResourceResponseCFNet.cpp:
(WebCore::ResourceResponse::platformSuggestedFilename): Return the null String if
m_cfResponse is null.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@174252 268f45cc-cd09-0410-ab3c-d52691b4dbfc

5 years agoUse version number 101000 instead of 10100 for OS X Yosemite
dbates@webkit.org [Fri, 3 Oct 2014 00:30:42 +0000 (00:30 +0000)]
Use version number 101000 instead of 10100 for OS X Yosemite
https://bugs.webkit.org/show_bug.cgi?id=137372

Reviewed by Mark Rowe.

The version number for OS X Yosemite is 101000.

Source/WebKit/mac:

* WebView/WebView.mm:
(-[WebView _preferencesChanged:]):

Tools:

* DumpRenderTree/mac/TextInputController.m:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@174251 268f45cc-cd09-0410-ab3c-d52691b4dbfc

5 years ago[iOS] Create an SPI wrapper for _UIHighlightView and use it in WKContentView
aestes@apple.com [Thu, 2 Oct 2014 23:56:32 +0000 (23:56 +0000)]
[iOS] Create an SPI wrapper for _UIHighlightView and use it in WKContentView
https://bugs.webkit.org/show_bug.cgi?id=137370

Reviewed by Tim Horton.

Source/WebCore:

Added _UIHighlightViewSPI.h. When building against the internal SDK it imports <UIKit/_UIHighlightView.h>. Otherwise, it redeclares the SPI we use in WebKit2.

* WebCore.xcodeproj/project.pbxproj:
* platform/spi/ios/_UIHighlightViewSPI.h: Added.

Source/WebKit2:

* UIProcess/ios/WKContentViewInteraction.mm: Included <WebCore/_UIHighlightViewSPI.h> instead of <UIKit/_UIHighlightView.h>.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@174250 268f45cc-cd09-0410-ab3c-d52691b4dbfc

5 years agoupdate-work-items should never delete items
ap@apple.com [Thu, 2 Oct 2014 23:49:30 +0000 (23:49 +0000)]
update-work-items should never delete items
https://bugs.webkit.org/show_bug.cgi?id=137366

Reviewed by Ryosuke Niwa.

As we don't just replace the whole list any more, indicate which items are high
priority, and which are not. Hight priority ones will be prepended to the queue,
others will be appended.

This creates a bit of unfairness in that high priority item queue becomes a LIFO.
But hopefully we will never have many rollouts competing like that.

* QueueStatusServer/app.yaml: Update version.

* QueueStatusServer/handlers/updateworkitems.py: Never remove items. Pass high
priority and regular items separately. Removed some error handling that used to
end up in returning status 500 - an uncaught exception does that automatically.

* QueueStatusServer/main.py: Removed unnecessary regexps from URL matching code.

* QueueStatusServer/model/workitems.py: Added a way to add high priority items.

* QueueStatusServer/templates/updateworkitems.html: Added a field for high
priority items.

* Scripts/webkitpy/common/net/bugzilla/bugzilla.py:
(BugzillaQueries._parse_attachment_ids_request_query): Removed an annoying log
line that complicated testing.

* Scripts/webkitpy/common/net/statusserver.py: Pass high priority items separately.

* Scripts/webkitpy/tool/bot/feeders.py: For commit queue, split high and regular
priority items into separate lists.

* Scripts/webkitpy/common/net/statusserver_mock.py:
* Scripts/webkitpy/tool/bot/feeders_unittest.py:
* Scripts/webkitpy/tool/commands/queues_unittest.py:
Updated tests.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@174248 268f45cc-cd09-0410-ab3c-d52691b4dbfc

5 years agoAnnoying build warnings in WTFString API tests
b.long@cablelabs.com [Thu, 2 Oct 2014 23:41:16 +0000 (23:41 +0000)]
Annoying build warnings in WTFString API tests
https://bugs.webkit.org/show_bug.cgi?id=137360

Reviewed by Darin Adler.

* TestWebKitAPI/Tests/WTF/WTFString.cpp:
(TestWebKitAPI::TEST): Change EXPECT_EQ(false, ok) to EXPECT_FALSE(ok) and EXPECT_EQ(true, ok) to EXPECT_TRUE(ok).

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@174246 268f45cc-cd09-0410-ab3c-d52691b4dbfc

5 years agoCSS JIT: add the initial implementation of :nth-child(An+B of selector)
benjamin@webkit.org [Thu, 2 Oct 2014 23:41:01 +0000 (23:41 +0000)]
CSS JIT: add the initial implementation of :nth-child(An+B of selector)
https://bugs.webkit.org/show_bug.cgi?id=137285

Reviewed by Andreas Kling.

Source/WebCore:

This patch adds the baseline support for JIT compiling :nth-child(An+B of selector).

The simple optimizations from the classic :nth-child() are carried over to :nth-child(An+B of selector),
but the new code generation is pretty much left completely unoptimized.

The main challenge introduced with the new functional pseudo classes is the possibility of backtracking
while already inside a backtracking chain.

For example:
    :nth-child(odd of a+a+a~a)+b+b~b
has to backtrack the chain a+a+a while already in the backtracking chain b+b+b.

The easiest way to solve this would have been to have each subselector generate a new function.
The generator for :nth-child() would simply call the function call for each sibling.

In the example, we would have:
    functionA(): a+a+a~a
    selectorChecker(): :nth-child(odd of functionA())+b+b~b

I decided against explicit functions because I believe the would be counterproductive for the common cases.
I expect most use of :nth-child(of) to be done with a light filter without combinators, things like:
    :nth-child(odd of .class)
    :nth-child(odd of type.class)
    :nth-child(odd of type[attribute=value])
    etc.

For simple cases, inlining the internal filter will greatly simplify the generated code.

In this particular patch, the selector checkers are inlined, but we do not attempt to have good code generation.
The optimizations will come later.

To implement backtracking in the subselectors, the backtracking implementation was abstracted into the structure
BacktrackingLevel. Every time the code generator starts a new subselector, a new BacktrackingLevel is pushed onto
the multilevel backtracking stack. The code generation always use the latest BacktrackingLevel for code generation
until the subselector is finished, and code generation can continue with the existing state.

Tests: fast/selectors/nth-child-of-backtracking-adjacent-2.html
       fast/selectors/nth-child-of-backtracking-adjacent-optimized.html
       fast/selectors/nth-child-of-backtracking-adjacent.html
       fast/selectors/nth-child-of-boundaries-1.html
       fast/selectors/nth-child-of-boundaries-2.html
       fast/selectors/nth-child-of-boundaries-3.html
       fast/selectors/nth-child-of-chained-2.html
       fast/selectors/nth-child-of-has-parent.html
       fast/selectors/nth-child-of-never-matching-selector.html
       fast/selectors/nth-child-of-with-invalid-An+B.html

* cssjit/SelectorCompiler.cpp:
(WebCore::SelectorCompiler::addPseudoClassType):
(WebCore::SelectorCompiler::SelectorCodeGenerator::SelectorCodeGenerator):
(WebCore::SelectorCompiler::minimumRegisterRequirements):
(WebCore::SelectorCompiler::computeBacktrackingMemoryRequirements):
Previously, we only needed stack space for the adjacent backtracking entry.

Now we can have several backtracking level active at the same time. We compute how much stack we need
initially and each backtracking level take stack references as needed.

(WebCore::SelectorCompiler::computeBacktrackingInformation):
The entire backtracking infrastructure already works great for multilevel backtracking. It is actually
amazing how everything fits nicely together :)

The only addition here is logging facility in case we run into a bug.

(WebCore::SelectorCompiler::SelectorCodeGenerator::generateSelectorChecker):
(WebCore::SelectorCompiler::SelectorCodeGenerator::generateSelectorCheckerExcludingPseudoElements):
generateSelectorChecker() has handling for pseudo elements, which should not be used
for :nth-child(of). The core generator was moved to generateSelectorCheckerExcludingPseudoElements()
to be used for code generation of subselectors.

(WebCore::SelectorCompiler::SelectorCodeGenerator::generateParentElementTreeWalker):
(WebCore::SelectorCompiler::SelectorCodeGenerator::generateAncestorTreeWalker):
(WebCore::SelectorCompiler::SelectorCodeGenerator::generateDirectAdjacentTreeWalker):
(WebCore::SelectorCompiler::SelectorCodeGenerator::generateIndirectAdjacentTreeWalker):
(WebCore::SelectorCompiler::SelectorCodeGenerator::linkFailures):
(WebCore::SelectorCompiler::SelectorCodeGenerator::generateAdjacentBacktrackingTail):
(WebCore::SelectorCompiler::SelectorCodeGenerator::generateDescendantBacktrackingTail):
The two main changes are:
1) Adopt the BacktrackingLevel to handle multilevel generation.
2) Descendant backtracking can use the stack if we have multi level descendant backtracking.
   In that case, we just use the stack like for adjacent backtracking.

(WebCore::SelectorCompiler::SelectorCodeGenerator::generateElementIsNthChildOf):
This is very similar to generateElementIsNthChild(). The main difference is we cannot use
the cache in this case.

(WebCore::SelectorCompiler::SelectorCodeGenerator::generateElementMatching):
(WebCore::SelectorCompiler::nthFilterIsAlwaysSatisified):
(WebCore::SelectorCompiler::SelectorCodeGenerator::generateElementIsNthChild):
(WebCore::SelectorCompiler::SelectorCodeGenerator::generateNthFilterTest):

LayoutTests:

Those are new tests on top of the existing tests for :nth-child(An+B of selector).
They cover mostly areas related to the JIT optimizations.

The areas with the least test coverage for correctness is multilevel backtracking.
I expect we will create more of those tests while extending :not() and adding :matches().

* fast/selectors/nth-child-of-backtracking-adjacent-expected.txt: Added.
* fast/selectors/nth-child-of-backtracking-adjacent-optimized.html: Added.
* fast/selectors/nth-child-of-backtracking-adjacent-expected.html: Added.
* fast/selectors/nth-child-of-backtracking-adjacent.html: Added.
* fast/selectors/nth-child-of-backtracking-adjacent-2-expected.txt: Added.
* fast/selectors/nth-child-of-backtracking-adjacent-2.html: Added.
Those tests are based on Yusuke's backtracking test suite for adjacent combinators.

The tests in nth-child-of-backtracking-adjacent.html use :nth-child(of) inside a regular
backtracking chain.

The tests in nth-child-of-backtracking-adjacent-2.html test backtracking inside :nth-chlid(of).

* fast/selectors/nth-child-of-boundaries-1-expected.txt: Added.
* fast/selectors/nth-child-of-boundaries-1.html: Added.
* fast/selectors/nth-child-of-boundaries-2-expected.txt: Added.
* fast/selectors/nth-child-of-boundaries-2.html: Added.
* fast/selectors/nth-child-of-boundaries-3-expected.txt: Added.
* fast/selectors/nth-child-of-boundaries-3.html: Added.
Test various abuse of :nth-child(of).

* fast/selectors/nth-child-of-chained-2-expected.txt: Added.
* fast/selectors/nth-child-of-chained-2.html: Added.
Test that disjoint :nth-child(of) are not agregated, unlike the old :nth-child().

* fast/selectors/nth-child-of-has-parent-expected.txt: Added.
* fast/selectors/nth-child-of-has-parent.html: Added.
Make sure we don't optimize out the parent check when skipping filters that always match.

* fast/selectors/nth-child-of-never-matching-selector-expected.txt: Added.
* fast/selectors/nth-child-of-never-matching-selector.html: Added.
Test code generation for things that never match.

* fast/selectors/nth-child-of-with-invalid-An+B-expected.txt: Added.
* fast/selectors/nth-child-of-with-invalid-An+B.html: Added.
Test for useless variation of An+B that can never match anything.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@174245 268f45cc-cd09-0410-ab3c-d52691b4dbfc

5 years ago[iOS] Networking process stops loading web pages while running Alexa test with random...
psolanki@apple.com [Thu, 2 Oct 2014 23:39:34 +0000 (23:39 +0000)]
[iOS] Networking process stops loading web pages while running Alexa test with random URL list
https://bugs.webkit.org/show_bug.cgi?id=137362
<rdar://problem/18507382>

Reviewed by Alexey Proskuryakov.

ResourceHandleCFURLConnectionDelegateWithOperationQueue needs to signal threads waiting on
its semaphore when the handle is being destroyed. Otherwise, we can leave dispatch threads
hanging around waiting for a response. This can happen when the Web Content process dies.
Any thread/queue waiting on a response from that web process will leak and stay around
forever. If we reach the dispatch queue limit of 64, then all networking will cease to
happen in the Networking process. Fix this by signalling waiting threads and clearing out
our state in much the same way that -[WebCoreResourceHandleAsOperationQueueDelegate
detachHandle] does for Mac.

* platform/network/cf/ResourceHandleCFURLConnectionDelegate.h:
* platform/network/cf/ResourceHandleCFURLConnectionDelegateWithOperationQueue.cpp:
(WebCore::ResourceHandleCFURLConnectionDelegateWithOperationQueue::releaseHandle):
* platform/network/cf/ResourceHandleCFURLConnectionDelegateWithOperationQueue.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@174244 268f45cc-cd09-0410-ab3c-d52691b4dbfc

5 years ago[Windows] The number of processes did not affect the crash inconsistency, increase...
roger_fong@apple.com [Thu, 2 Oct 2014 23:36:34 +0000 (23:36 +0000)]
[Windows] The number of processes did not affect the crash inconsistency, increase number of child processes again.
* Scripts/webkitpy/port/win.py:
(WinPort.default_child_processes): Ambitiously try 3 processes the testers are too slow now.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@174243 268f45cc-cd09-0410-ab3c-d52691b4dbfc

5 years agoMove ASan arguments from LLVM (-mllvm) to the frontend
dfarler@apple.com [Thu, 2 Oct 2014 23:32:06 +0000 (23:32 +0000)]
Move ASan arguments from LLVM (-mllvm) to the frontend
https://bugs.webkit.org/show_bug.cgi?id=137365

Reviewed by David Kilzer.

* asan/asan.xcconfig:
Remove the -mllvm prefix to the asan-blacklist argument, moving it
to the clang frontend.

http://llvm.org/viewvc/llvm-project?view=revision&revision=212505

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@174242 268f45cc-cd09-0410-ab3c-d52691b4dbfc

5 years ago[Windows] Skip a small handful of tests that fail consistently on Windows bots Part 2.
roger_fong@apple.com [Thu, 2 Oct 2014 23:30:04 +0000 (23:30 +0000)]
[Windows] Skip a small handful of tests that fail consistently on Windows bots Part 2.
* platform/win/TestExpectations: Typo, meant to skip the whole folder on Debug tests.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@174241 268f45cc-cd09-0410-ab3c-d52691b4dbfc

5 years agoMarking flaky on Mavericks.
bdakin@apple.com [Thu, 2 Oct 2014 23:27:26 +0000 (23:27 +0000)]
Marking flaky on Mavericks.

* platform/mac/TestExpectations:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@174240 268f45cc-cd09-0410-ab3c-d52691b4dbfc

5 years ago[Windows] Skip a small handful of tests that fail consistently on Windows bots.
roger_fong@apple.com [Thu, 2 Oct 2014 23:14:23 +0000 (23:14 +0000)]
[Windows] Skip a small handful of tests that fail consistently on Windows bots.
* platform/win/TestExpectations:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@174239 268f45cc-cd09-0410-ab3c-d52691b4dbfc

5 years agoFix the iOS build.
aestes@apple.com [Thu, 2 Oct 2014 23:03:35 +0000 (23:03 +0000)]
Fix the iOS build.

* WebProcess/WebPage/ios/WebPageIOS.mm:
(WebKit::WebPage::blurAssistedNode):
(WebKit::WebPage::setAssistedNodeValue):
(WebKit::WebPage::setAssistedNodeValueAsNumber):
(WebKit::WebPage::setAssistedNodeSelectedIndex):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@174238 268f45cc-cd09-0410-ab3c-d52691b4dbfc

5 years ago[Windows] Unskip media tests now that audio devices are installed on the bots.
roger_fong@apple.com [Thu, 2 Oct 2014 23:01:30 +0000 (23:01 +0000)]
[Windows] Unskip media tests now that audio devices are installed on the bots.

* platform/win/TestExpectations: There will likely still be a handful of failures left, which we will then skip.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@174237 268f45cc-cd09-0410-ab3c-d52691b4dbfc

5 years agoUse is<>() / downcast<>() for CSSBasicShape objects
cdumez@apple.com [Thu, 2 Oct 2014 22:53:18 +0000 (22:53 +0000)]
Use is<>() / downcast<>() for CSSBasicShape objects
https://bugs.webkit.org/show_bug.cgi?id=137331

Reviewed by Andreas Kling.

Use is<>() / downcast<>() for CSSBasicShape objects.

No new tests, no behavior change.

* css/BasicShapeFunctions.cpp:
(WebCore::basicShapeForValue):
* css/CSSBasicShapes.cpp:
(WebCore::CSSBasicShapeCircle::equals):
(WebCore::CSSBasicShapeEllipse::equals):
(WebCore::CSSBasicShapePolygon::equals):
(WebCore::CSSBasicShapeInset::equals):
* css/CSSBasicShapes.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@174236 268f45cc-cd09-0410-ab3c-d52691b4dbfc

5 years agoXMLHttpRequestProgressEventThrottle shouldn't throttle / defer progress events if...
cdumez@apple.com [Thu, 2 Oct 2014 22:49:06 +0000 (22:49 +0000)]
XMLHttpRequestProgressEventThrottle shouldn't throttle / defer progress events if there are no listeners
https://bugs.webkit.org/show_bug.cgi?id=137346

Reviewed by Alexey Proskuryakov.

Previously XMLHttpRequestProgressEventThrottle would throttle / defer
progress events, which involves queueing events and starting timers,
even if there were no listeners for those events (which is fairly
common case).

This patch updates XMLHttpRequestProgressEventThrottle to make sure
there are actual progress event listeners *before* attempting to
throttle and do extra processing, to avoid doing unnecessary work.

This patch also makes XMLHttpRequestProgressEventThrottle::dispatchEvent()
private as this method is only called from other dispatch methods of
the class and never from the outside.

No new tests, no behavior change.

* xml/XMLHttpRequestProgressEventThrottle.cpp:
(WebCore::XMLHttpRequestProgressEventThrottle::dispatchThrottledProgressEvent):
(WebCore::XMLHttpRequestProgressEventThrottle::dispatchProgressEvent):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@174235 268f45cc-cd09-0410-ab3c-d52691b4dbfc

5 years agoSimplify StringTypeAdapter templates
andersca@apple.com [Thu, 2 Oct 2014 22:48:48 +0000 (22:48 +0000)]
Simplify StringTypeAdapter templates
https://bugs.webkit.org/show_bug.cgi?id=137356

Reviewed by Andreas Kling.

* wtf/text/StringConcatenate.h:
Use StringView for copying characters. Use inheritance for char* vs const char* etc. Make all StringAdapter member functions const.

* wtf/text/StringView.h:
Move enough functions out of line so we can include WTFString.h after the StringView class definition.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@174234 268f45cc-cd09-0410-ab3c-d52691b4dbfc

5 years ago[Subpixel] Use floats instead of ints for text justification expansion
mmaxfield@apple.com [Thu, 2 Oct 2014 22:25:50 +0000 (22:25 +0000)]
[Subpixel] Use floats instead of ints for text justification expansion
https://bugs.webkit.org/show_bug.cgi?id=137327

Reviewed by Zalan Bujtas.

Source/WebCore:

Use a float instead of an int for the expansion variable in InlineBox. The int was inside an inner class
that held a collection of bitfields, and this patch removes that inner-inner variable.

No new tests because there is plenty of existing coverage and existing tests are rebaselined.

* rendering/InlineBox.cpp: Update static_assert for the size of InlineBox.
* rendering/InlineBox.h: Use a float for the justification expansion amount instead of an int bitfield.
(WebCore::InlineBox::InlineBoxBitfields::InlineBoxBitfields): Remove int bitfield.
(WebCore::InlineBox::InlineBox): Initialize float expansion.
(WebCore::InlineBox::expansion): Return a float.
(WebCore::InlineBox::setExpansion): Take a float.
(WebCore::InlineBox::InlineBoxBitfields::expansion): Deleted. Moved to InlineBox.
(WebCore::InlineBox::InlineBoxBitfields::setExpansion): Deleted. Ditto.
* rendering/InlineTextBox.h:
(WebCore::InlineTextBox::setExpansion): Take a float.

LayoutTests:

Rebaselining tests due to subpixel differences.

* platform/mac/css1/box_properties/clear_float-expected.txt:
* platform/mac/css1/box_properties/float_on_text_elements-expected.txt:
* platform/mac/css2.1/t0905-c5525-fltcont-00-d-g-expected.txt:
* platform/mac/css2.1/t0905-c5525-fltwidth-00-c-g-expected.txt:
* platform/mac/css2.1/t1602-c546-txt-align-00-b-expected.txt:
* platform/mac/fast/multicol/float-avoidance-expected.txt:
* platform/mac/fast/multicol/overflow-across-columns-expected.txt:
* platform/mac/fast/multicol/overflow-across-columns-percent-height-expected.txt:
* platform/mac/fast/multicol/overflow-unsplittable-expected.txt:
* platform/mac/fast/multicol/positive-leading-expected.txt:
* platform/mac/fast/multicol/vertical-lr/float-avoidance-expected.txt:
* platform/mac/fast/multicol/vertical-rl/float-avoidance-expected.txt:
* platform/mac/fast/regions/multiple-directionality-changes-in-variable-width-regions-expected.txt:
* platform/mac/fast/regions/repaint/line-flow-with-floats-in-regions-expected.txt:
* platform/mac/fast/repaint/line-flow-with-floats-1-expected.txt:
* platform/mac/fast/repaint/line-flow-with-floats-10-expected.txt:
* platform/mac/fast/repaint/line-flow-with-floats-2-expected.txt:
* platform/mac/fast/repaint/line-flow-with-floats-3-expected.txt:
* platform/mac/fast/repaint/line-flow-with-floats-4-expected.txt:
* platform/mac/fast/repaint/line-flow-with-floats-5-expected.txt:
* platform/mac/fast/repaint/line-flow-with-floats-6-expected.txt:
* platform/mac/fast/repaint/line-flow-with-floats-7-expected.txt:
* platform/mac/fast/repaint/line-flow-with-floats-8-expected.txt:
* platform/mac/fast/repaint/line-flow-with-floats-9-expected.txt:
* platform/mac/fast/ruby/bopomofo-rl-expected.txt:
* platform/mac/fast/text/hyphenate-character-expected.txt:
* platform/mac/fast/text/hyphens-expected.txt:
* platform/mac/fast/text/justified-selection-at-edge-expected.txt:
* platform/mac/fast/text/justify-ideograph-leading-expansion-expected.txt:
* platform/mac/fast/text/justify-nbsp-expected.txt: Added.
* platform/mac/fast/text/trailing-white-space-2-expected.txt:
* platform/mac/fast/text/trailing-white-space-expected.txt:
* platform/mac/tables/mozilla/marvin/x_tbody_align_justify-expected.txt:
* platform/mac/tables/mozilla/marvin/x_td_align_justify-expected.txt:
* platform/mac/tables/mozilla/marvin/x_tfoot_align_justify-expected.txt:
* platform/mac/tables/mozilla/marvin/x_th_align_justify-expected.txt:
* platform/mac/tables/mozilla/marvin/x_thead_align_justify-expected.txt:
* platform/mac/tables/mozilla/marvin/x_tr_align_justify-expected.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@174233 268f45cc-cd09-0410-ab3c-d52691b4dbfc

5 years ago[iOS] Teach WKPDFView to navigate to pageNumber links
aestes@apple.com [Thu, 2 Oct 2014 21:51:14 +0000 (21:51 +0000)]
[iOS] Teach WKPDFView to navigate to pageNumber links
https://bugs.webkit.org/show_bug.cgi?id=137358

Reviewed by Tim Horton.

Treat PDF pageNumber annotations as if they were fragment navigations in an HTML document. For a given page
number N, create a fragment called "#pageN" and tell WebKit to navigate to it. By doing this we create history
entries for each pageNumber navigation and allow for proper back/forward. When navigating to a page, reset to
the initial scale factor and scroll to the beginning of the Nth page.

* UIProcess/API/Cocoa/WKWebView.mm:
(-[WKWebView _zoomToPoint:atScale:animated:]): Added an animated parameter. If animated is NO, use a duration of 0.
(-[WKWebView _zoomToRect:atScale:origin:animated:]): Added an animated parameter and passed it to _zoomToPoint:atScale:animated:.
(-[WKWebView _zoomOutWithOrigin:animated:]): Ditto.
(-[WKWebView _zoomToRect:withOrigin:fitEntireRect:minimumScale:maximumScale:minimumScrollDistance:]): Called _zoomToRect:atScale:origin:animated:,
setting animated to YES.
(-[WKWebView _didSameDocumentNavigationForMainFrame:]): Called web_didSameDocumentNavigation: on _customContentView.
(-[WKWebView _zoomToPoint:atScale:]): Deleted.
(-[WKWebView _zoomToRect:atScale:origin:]): Deleted.
(-[WKWebView _zoomOutWithOrigin:]): Deleted.
* UIProcess/API/Cocoa/WKWebViewInternal.h:
* UIProcess/Cocoa/WKWebViewContentProvider.h:
* UIProcess/ios/PageClientImplIOS.mm:
(WebKit::PageClientImpl::didSameDocumentNavigationForMainFrame): Called _didSameDocumentNavigationForMainFrame on m_webView.
* UIProcess/ios/WKContentView.mm:
(-[WKContentView _zoomOutWithOrigin:]): Called _zoomOutWithOrigin:animated: on m_webView, setting animated to YES.
* UIProcess/ios/WKPDFPageNumberIndicator.h:
* UIProcess/ios/WKPDFPageNumberIndicator.mm:
(-[WKPDFPageNumberIndicator hide]): Added a method to hide the page number indicator.
* UIProcess/ios/WKPDFView.mm:
(-[WKPDFView web_setContentProviderData:suggestedFilename:]): Added a FIXME for restoring scroll position and page scale when loading from the back/forward list.
(-[WKPDFView scrollViewDidScroll:]): Stopped showing the page number indicator if a same-document navigation is occurring (to match UIWebPDFView).
(-[WKPDFView _updatePageNumberIndicator]): Ditto.
(-[WKPDFView web_didSameDocumentNavigation:]):
(-[WKPDFView _resetZoomAnimated:]): For same-document navigations of type kWKSameDocumentNavigationSessionStatePop, extracted the page index from the URL fragment
identifier, hid the page number indicator, reset the zoom (without an animation to match UIWebPDFView), and scrolled to the beginning of the given page.
(-[WKPDFView resetZoom:]): Called _resetZoomAnimated:, setting animated to YES.
(-[WKPDFView annotation:wasTouchedAtPoint:controller:]): If there is a non-zero pageNumber in the link annotation, construct a "#pageN" fragment and navigate to it.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@174232 268f45cc-cd09-0410-ab3c-d52691b4dbfc

5 years agoMove PageOverlay[Controller] to WebCore
timothy_horton@apple.com [Thu, 2 Oct 2014 21:06:14 +0000 (21:06 +0000)]
Move PageOverlay[Controller] to WebCore
https://bugs.webkit.org/show_bug.cgi?id=137164
<rdar://problem/18508258>

Reviewed by Anders Carlsson.

* CMakeLists.txt:
* WebCore.vcxproj/WebCore.vcxproj:
* WebCore.vcxproj/WebCore.vcxproj.filters:
* WebCore.xcodeproj/project.pbxproj:
Make the appropriate build system changes.

* WebCore.exp.in:
Export the new PageOverlay[Controller] symbols from WebCore.

* loader/EmptyClients.h:
* page/ChromeClient.h:
(WebCore::ChromeClient::documentOverlayLayerForFrame): Deleted.
Add attachViewOverlayGraphicsLayer, which allows PageOverlayController
to push view-relative page overlay root layers down to WebKit to be
installed just inside the view.

Remove documentOverlayLayerForFrame because RenderLayerCompositor can now
talk directly to PageOverlayController.

* page/EventHandler.cpp:
(WebCore::EventHandler::handleMousePressEvent):
(WebCore::EventHandler::mouseMoved):
(WebCore::EventHandler::handleMouseReleaseEvent):
When the mouse is moved, pressed, or released, give PageOverlayController
the first shot at handling the event.

* page/FrameView.cpp:
(WebCore::FrameView::setFrameRect):
Inform PageOverlayController that the main FrameView's size changed,
so it can update the size of view-relative overlays.

(WebCore::FrameView::setContentsSize):
Inform PageOverlayController that the main FrameView's contents size changed,
so it can update the size of document-relative overlays.

(WebCore::FrameView::setFixedVisibleContentRect):
(WebCore::FrameView::didChangeScrollOffset):
(WebCore::FrameView::scrollTo):
(WebCore::FrameView::wheelEvent):
Inform PageOverlayController that something scrolled.

(WebCore::FrameView::setExposedRect):
Inform PageOverlayController that the exposed rect changed, so it can push
the new exposed rect down to the overlays.

* page/FrameView.h:
Add didChangeScrollOffset.

* page/MainFrame.cpp:
(WebCore::MainFrame::MainFrame):
* page/MainFrame.h:
Keep one PageOverlayController per MainFrame.

* page/Page.cpp:
(WebCore::Page::setDeviceScaleFactor):
Inform PageOverlayController that the device scale factor changed.

(WebCore::Page::setSessionID):
Remove a blank line.

* page/PageOverlay.cpp: Renamed from Source/WebKit2/WebProcess/WebPage/PageOverlay.cpp.
Move PageOverlay from WebKit2 to WebCore.
A few overarching changes that I won't detail in each place:
    - references instead of pointers in many places
    - WebCore types instead of WebKit2 types everywhere

(WebCore::PageOverlay::bounds):
It is possible to install an overlay before we have a FrameView.
We will eventually get a view/contents size changed notification and try again.

(WebCore::PageOverlay::copyAccessibilityAttributeStringValueForPoint):
(WebCore::PageOverlay::copyAccessibilityAttributeBoolValueForPoint):
(WebCore::PageOverlay::copyAccessibilityAttributeNames):
Split copyAccessibilityAttribute into "StringValueForPoint" and "BoolValueForPoint"
variants, because we don't have anything like WKTypeRef here.

* page/PageOverlay.h: Renamed from Source/WebKit2/WebProcess/WebPage/PageOverlay.h.
Make PageOverlay a normal RefCounted object instead of a WebKit2 API object.
Leave a comment noting that we should move the PageOverlay's GraphicsLayer
to PageOverlay instead of a map on PageOverlayController.

* page/PageOverlayController.cpp: Renamed from Source/WebKit2/WebProcess/WebPage/PageOverlayController.cpp.
Move PageOverlayController from WebKit2 to WebCore.
A few overarching changes that I won't detail in each place:
    - references instead of pointers in many places
    - WebCore types instead of WebKit2 types everywhere

Leave a FIXME that didChangeSettings is not currently called when settings change.
This is not a problem for normal use of overlays, only for dynamically changing
layer border/repaint counter settings.

(WebCore::PageOverlayController::PageOverlayController):
(WebCore::PageOverlayController::createRootLayersIfNeeded):
We will now lazily initialize the root layers when we first try to use them,
because otherwise we try to create them before the ChromeClient's GraphicsLayerFactory
is hooked up, in some cases, and that can cause us to create the wrong kind of GraphicsLayers.

(WebCore::PageOverlayController::installPageOverlay):
Enter compositing mode when installing a PageOverlay.
Avoid pushing the Page to the PageOverlay until after we've set up
its root layer, so that the installation process can make use of that layer;
clients may, for example, expect to be able to setNeedsDisplay() in the
didMoveToPage callback.
Avoid updateOverlayGeometry until the layer is created and we've pushed
the Page down, so that e.g. bounds() will use the right Page.

(WebCore::PageOverlayController::didChangeSettings):
Leave a FIXME about a future improvement to didChangeSettings.

(WebCore::PageOverlayController::deviceScaleFactor):
(WebCore::PageOverlayController::notifyFlushRequired):
Null-check Page; these can get called when tearing down the Frame.

* page/PageOverlayController.h: Renamed from Source/WebKit2/WebProcess/WebPage/PageOverlayController.h.

* platform/graphics/GraphicsLayerClient.h:
Add wtf/Forward.h, because GraphicsLayerClient.h uses String.

* rendering/RenderLayer.cpp:
(WebCore::RenderLayer::scrollTo):
Inform PageOverlayController that something scrolled.

* rendering/RenderLayerCompositor.cpp:
(WebCore::RenderLayerCompositor::hasAnyAdditionalCompositedLayers):
Make hasAnyAdditionalCompositedLayers return true if there are any page overlays,
because we need to avoid falling out of compositing mode if we still have
page overlays.

(WebCore::RenderLayerCompositor::updateCompositingLayers):
Enter compositing mode if we have page overlays.

(WebCore::RenderLayerCompositor::appendOverlayLayers): Deleted.
(WebCore::RenderLayerCompositor::appendDocumentOverlayLayers):
Rename appendOverlayLayers -> appendDocumentOverlayLayers.

(WebCore::RenderLayerCompositor::attachRootLayer):
(WebCore::RenderLayerCompositor::detachRootLayer):
Attach/detach the view overlay root layer when attaching/detaching the
normal compositing root layer.

(WebCore::RenderLayerCompositor::rootLayerAttachmentChanged):
* rendering/RenderLayerCompositor.h:

* CMakeLists.txt:
* WebKit2.xcodeproj/project.pbxproj:
Make the appropriate build system changes.

* WebProcess/InjectedBundle/API/c/WKBundleAPICast.h:
Make WKBundlePageOverlayRef map to WebPageOverlay instead of PageOverlay.

* WebProcess/InjectedBundle/API/c/WKBundlePage.cpp:
(WKBundlePageInstallPageOverlay):
(WKBundlePageUninstallPageOverlay):
(WKBundlePageInstallPageOverlayWithAnimation):
(WKBundlePageUninstallPageOverlayWithAnimation):
Forward PageOverlay installation/uninstallation to WebCore.

* WebProcess/InjectedBundle/API/c/WKBundlePageOverlay.cpp:
(WKBundlePageOverlayGetTypeID):
(WKBundlePageOverlayCreate):
(WKBundlePageOverlaySetAccessibilityClient):
Use WebPageOverlay instead of PageOverlay, and references instead of pointers.
We forward client callbacks through from PageOverlay::Client to WebPageOverlay::Client.

(PageOverlayClientImpl::copyAccessibilityAttributeValue):
(PageOverlayClientImpl::copyAccessibilityAttributeNames):
Convert from wtf types to WebKit2 types.

* WebProcess/WebCoreSupport/WebChromeClient.cpp:
(WebKit::WebChromeClient::graphicsLayerFactory):
If we don't have a drawing area, don't dereference it!

(WebKit::WebChromeClient::attachViewOverlayGraphicsLayer):
Forward attachViewOverlayGraphicsLayer to the DrawingArea.

* WebProcess/WebCoreSupport/WebChromeClient.h:

* WebProcess/WebCoreSupport/WebInspectorClient.cpp:
* WebProcess/WebCoreSupport/WebInspectorClient.h:
* WebProcess/WebPage/FindController.cpp:
* WebProcess/WebPage/FindController.h:
* WebProcess/WebPage/ios/FindControllerIOS.mm:
* WebProcess/WebPage/ios/FindIndicatorOverlayClientIOS.h:
* WebProcess/WebPage/mac/ServicesOverlayController.mm:
* WebProcess/WebPage/ServicesOverlayController.h:
Mechanical changes to adopt to the fact that PageOverlay is in WebCore now.

* WebProcess/WebPage/LayerTreeHost.h:
* WebProcess/WebPage/WebPage.cpp:
* WebProcess/WebPage/WebPage.h:
Remove support for WebKit2-based page overlays.
Return MainFrame instead of Frame from mainFrame().

* WebProcess/WebPage/WebPageOverlay.cpp: Added.
* WebProcess/WebPage/WebPageOverlay.h: Added.
(WebKit::WebPageOverlay::create):
(WebKit::WebPageOverlay::WebPageOverlay):
(WebKit::WebPageOverlay::~WebPageOverlay):
(WebKit::WebPageOverlay::setNeedsDisplay):
(WebKit::WebPageOverlay::clear):
(WebKit::WebPageOverlay::pageOverlayDestroyed):
(WebKit::WebPageOverlay::willMoveToPage):
(WebKit::WebPageOverlay::didMoveToPage):
(WebKit::WebPageOverlay::drawRect):
(WebKit::WebPageOverlay::mouseEvent):
(WebKit::WebPageOverlay::didScrollFrame):
(WebKit::WebPageOverlay::copyAccessibilityAttributeStringValueForPoint):
(WebKit::WebPageOverlay::copyAccessibilityAttributeBoolValueForPoint):
(WebKit::WebPageOverlay::copyAccessibilityAttributeNames):
WebPageOverlay exists to be our API object, but forwards everything
to its WebCore::PageOverlay, and implements WebCore::PageOverlay::Client, and
forwards client callbacks to its WebPageOverlay::Client (which
WKBundlePageOverlay implements).

* WebProcess/WebPage/mac/WKAccessibilityWebPageObjectMac.mm:
(-[WKAccessibilityWebPageObject accessibilityParameterizedAttributeNames]):
(-[WKAccessibilityWebPageObject accessibilityAttributeValue:forParameter:]):
Forward the three AX properties that PageOverlays are ever queried for
to the appropriate PageOverlayController function.

* WebProcess/WebPage/mac/RemoteLayerTreeDrawingArea.h:
* WebProcess/WebPage/mac/RemoteLayerTreeDrawingArea.mm:
(WebKit::RemoteLayerTreeDrawingArea::RemoteLayerTreeDrawingArea):
(WebKit::RemoteLayerTreeDrawingArea::updateRootLayers):
Factor updateRootLayers out of setRootCompositingLayer.

(WebKit::RemoteLayerTreeDrawingArea::attachViewOverlayGraphicsLayer):
If we attach/detach the view-relative page overlay root layer,
reattach the root compositing layer's children.

(WebKit::RemoteLayerTreeDrawingArea::setRootCompositingLayer):
Ditto for the root compositing layer.

(WebKit::RemoteLayerTreeDrawingArea::updateScrolledExposedRect):
(WebKit::RemoteLayerTreeDrawingArea::flushLayers):
(WebKit::RemoteLayerTreeDrawingArea::mainFrameContentSizeChanged):
Informing PageOverlayController of exposed rect/document size changes is now
WebCore's job.

* WebProcess/WebPage/mac/TiledCoreAnimationDrawingArea.h:
* WebProcess/WebPage/mac/TiledCoreAnimationDrawingArea.mm:
(WebKit::TiledCoreAnimationDrawingArea::TiledCoreAnimationDrawingArea):
(WebKit::TiledCoreAnimationDrawingArea::updateRootLayers):
Factor updateRootLayers out of setRootCompositingLayer.

(WebKit::TiledCoreAnimationDrawingArea::attachViewOverlayGraphicsLayer):
(WebKit::TiledCoreAnimationDrawingArea::setRootCompositingLayer):
Make use of updateRootLayers when we attach/detach root compositing layer/view overlay layer.

(WebKit::TiledCoreAnimationDrawingArea::mainFrameContentSizeChanged):
(WebKit::TiledCoreAnimationDrawingArea::flushLayers):
(WebKit::TiledCoreAnimationDrawingArea::updateScrolledExposedRect):
Informing PageOverlayController of exposed rect/document size changes is now
WebCore's job.

* WebProcess/WebPage/DrawingArea.h:
(WebKit::DrawingArea::attachViewOverlayGraphicsLayer):

* Shared/WebRenderLayer.cpp:
* Shared/WebRenderObject.cpp:
* WebProcess/InjectedBundle/API/c/WKBundleAPICast.h:
* WebProcess/InjectedBundle/API/mac/WKWebProcessPlugInBrowserContextController.mm:
* WebProcess/WebCoreSupport/mac/WebEditorClientMac.mm:
* WebProcess/WebProcess.cpp:
Include MainFrame.h because WebPage::mainFrame returns a MainFrame now.

* WebCoreSupport/WebChromeClient.h:
* WebCoreSupport/WebChromeClient.mm:
(WebChromeClient::attachViewOverlayGraphicsLayer):
Add an empty ChromeClient override.

* WebCoreSupport/WebChromeClient.cpp:
(WebChromeClient::attachViewOverlayGraphicsLayer):
* WebCoreSupport/WebChromeClient.h:
Add an empty ChromeClient override.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@174231 268f45cc-cd09-0410-ab3c-d52691b4dbfc

5 years agoWKDownloadCopyRequest() returns 0 until the download starts
jeffm@apple.com [Thu, 2 Oct 2014 21:01:08 +0000 (21:01 +0000)]
WKDownloadCopyRequest() returns 0 until the download starts
https://bugs.webkit.org/show_bug.cgi?id=137322

Reviewed by Anders Carlsson.

The m_request ivar in DownloadProxy was only being initialized when didStart() was called after the
download starts. Since clients have access to the ResourceRequest when the DownloadProxy is created,
there's no reason not to pass that along to DownloadProxy::create() so it can be returned if
WKDownloadCopyRequest() is called before didStart().

* UIProcess/Downloads/DownloadProxy.cpp:
(WebKit::DownloadProxy::create):
Add ResourceRequest parameter, and pass it to DownloadProxy constructor.

(WebKit::DownloadProxy::DownloadProxy):
Initialize m_request.

* UIProcess/Downloads/DownloadProxy.h:
Add ResourceRequest parameter to create() and constructor.

* UIProcess/Downloads/DownloadProxyMap.cpp:
(WebKit::DownloadProxyMap::createDownloadProxy):
Add ResourceRequest parameter and pass it to DownloadProxy::create().

* UIProcess/Downloads/DownloadProxyMap.h:
Add ResourceRequest parameter to createDownloadProxy().

* UIProcess/Network/NetworkProcessProxy.cpp:
(WebKit::NetworkProcessProxy::createDownloadProxy):
Add ResourceRequest parameter and pass it to DownloadProxy::create().

* UIProcess/Network/NetworkProcessProxy.h:
Add ResourceRequest parameter to createDownloadProxy().

* UIProcess/WebContext.cpp:
(WebKit::WebContext::download):
Pass ResourceRequest to createDownloadProxy().

(WebKit::WebContext::createDownloadProxy):
Add ResourceRequest parameter and pass it to createDownloadProxy() functions.

* UIProcess/WebContext.h:
Add ResourceRequest parameter to createDownloadProxy().

* UIProcess/WebPageProxy.cpp:
(WebKit::WebPageProxy::receivedPolicyDecision):
Pass m_decidePolicyForResponseRequest to WebContext::createDownloadProxy().

* UIProcess/WebProcessProxy.cpp:
(WebKit::WebProcessProxy::createDownloadProxy):
Add ResourceRequest parameter and pass it to DownloadProxyMap::create().

* UIProcess/WebProcessProxy.h:
Add ResourceRequest parameter to createDownloadProxy().

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@174230 268f45cc-cd09-0410-ab3c-d52691b4dbfc

5 years agoBack TextRun with a StringView
mmaxfield@apple.com [Thu, 2 Oct 2014 20:54:21 +0000 (20:54 +0000)]
Back TextRun with a StringView
https://bugs.webkit.org/show_bug.cgi?id=137352

Reviewed by Antti Koivisto.

TextRun contains a union of a UChar* and an LChar* as well as a length, which is the abstraction
that StringView is intended to be. This patch cleans up TextRun by replacing the union and length
with a StringView instance member.

No new tests because there is no behavior change.

* platform/graphics/TextRun.h:
(WebCore::TextRun::TextRun):
(WebCore::TextRun::subRun):
(WebCore::TextRun::operator[]):
(WebCore::TextRun::data8):
(WebCore::TextRun::data16):
(WebCore::TextRun::characters8):
(WebCore::TextRun::characters16):
(WebCore::TextRun::is8Bit):
(WebCore::TextRun::length):
(WebCore::TextRun::string):
(WebCore::TextRun::setText):
(WebCore::TextRun::stringView):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@174228 268f45cc-cd09-0410-ab3c-d52691b4dbfc

5 years agoMarking test as flaky.
bdakin@apple.com [Thu, 2 Oct 2014 20:50:08 +0000 (20:50 +0000)]
Marking test as flaky.

* TestExpectations:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@174227 268f45cc-cd09-0410-ab3c-d52691b4dbfc

5 years agoDo all closed variable access through the local lexical object
oliver@apple.com [Thu, 2 Oct 2014 20:35:58 +0000 (20:35 +0000)]
Do all closed variable access through the local lexical object
https://bugs.webkit.org/show_bug.cgi?id=136869

Reviewed by Filip Pizlo.

This patch makes all reads and writes from captured registers
go through the lexical record, and by doing so removes the
need for record tearoff.

To keep the patch simple we still number variables as though
they are local stack allocated registers, but ::local() will
fail. When local fails we perform a generic resolve, and in
that resolve we now use a ResolveScopeInfo struct to pass
around information about whether a lookup is a statically
known captured variable, and its location in the activation.
To ensure correct behaviour during codeblock linking we also
add a LocalClosureVariable resolution type.

To ensure correct semantics for the Arguments object, we now
have to eagerly create the Arguments object for any function
that uses both the Arguments object and requires a lexical
record.

* bytecode/BytecodeList.json:
* bytecode/BytecodeUseDef.h:
(JSC::computeUsesForBytecodeOffset):
(JSC::computeDefsForBytecodeOffset):
* bytecode/CodeBlock.cpp:
(JSC::CodeBlock::dumpBytecode):
(JSC::CodeBlock::CodeBlock):
(JSC::CodeBlock::finalizeUnconditionally):
* bytecompiler/BytecodeGenerator.cpp:
(JSC::BytecodeGenerator::BytecodeGenerator):
(JSC::BytecodeGenerator::initializeCapturedVariable):
  During the entry to a function we are not yet in a position
  to allocate temporaries so we directly use the lexical
  environment register.
(JSC::BytecodeGenerator::resolveCallee):
(JSC::BytecodeGenerator::emitMove):
(JSC::BytecodeGenerator::local):
(JSC::BytecodeGenerator::constLocal):
(JSC::BytecodeGenerator::emitResolveScope):
(JSC::BytecodeGenerator::emitResolveConstantLocal):
  The two resolve scope operations could technically skip
  the op_resolve_scope, and simply perform
      op_mov dst, recordRegister
  but for now it seemed best to maintain the same basic
  behaviour.
(JSC::BytecodeGenerator::emitGetFromScope):
(JSC::BytecodeGenerator::emitPutToScope):
(JSC::BytecodeGenerator::createArgumentsIfNecessary):
  If we have an environment we've already created Arguments
  so no need to check again.
(JSC::BytecodeGenerator::emitReturn):
  Don't need to emit tearoff_environment
* bytecompiler/BytecodeGenerator.h:
(JSC::Local::Local):
(JSC::Local::operator bool):
(JSC::Local::get):
(JSC::Local::isReadOnly):
(JSC::Local::isSpecial):
(JSC::ResolveScopeInfo::ResolveScopeInfo):
(JSC::ResolveScopeInfo::isLocal):
(JSC::ResolveScopeInfo::localIndex):
(JSC::BytecodeGenerator::shouldCreateArgumentsEagerly):
(JSC::Local::isCaptured): Deleted.
(JSC::Local::captureMode): Deleted.
* bytecompiler/NodesCodegen.cpp:
(JSC::ResolveNode::emitBytecode):
(JSC::EvalFunctionCallNode::emitBytecode):
(JSC::FunctionCallResolveNode::emitBytecode):
(JSC::PostfixNode::emitResolve):
(JSC::DeleteResolveNode::emitBytecode):
(JSC::TypeOfResolveNode::emitBytecode):
(JSC::PrefixNode::emitResolve):
(JSC::ReadModifyResolveNode::emitBytecode):
(JSC::AssignResolveNode::emitBytecode):
(JSC::ConstDeclNode::emitCodeSingle):
(JSC::EmptyVarExpression::emitBytecode):
(JSC::ForInNode::tryGetBoundLocal):
(JSC::ForInNode::emitLoopHeader):
(JSC::ForOfNode::emitBytecode):
(JSC::BindingNode::bindValue):
* dfg/DFGAbstractInterpreterInlines.h:
(JSC::DFG::AbstractInterpreter<AbstractStateType>::executeEffects):
* dfg/DFGByteCodeParser.cpp:
(JSC::DFG::ByteCodeParser::parseBlock):
* dfg/DFGCapabilities.cpp:
(JSC::DFG::capabilityLevel):
* dfg/DFGClobberize.h:
(JSC::DFG::clobberize):
* dfg/DFGDoesGC.cpp:
(JSC::DFG::doesGC):
* dfg/DFGFixupPhase.cpp:
(JSC::DFG::FixupPhase::fixupNode):
* dfg/DFGGraph.cpp:
(JSC::DFG::Graph::tryGetRegisters):
* dfg/DFGNodeType.h:
* dfg/DFGPredictionPropagationPhase.cpp:
(JSC::DFG::PredictionPropagationPhase::propagate):
* dfg/DFGSafeToExecute.h:
(JSC::DFG::safeToExecute):
* dfg/DFGSpeculativeJIT32_64.cpp:
(JSC::DFG::SpeculativeJIT::compile):
* dfg/DFGSpeculativeJIT64.cpp:
(JSC::DFG::SpeculativeJIT::compile):
* ftl/FTLCapabilities.cpp:
(JSC::FTL::canCompile):
* interpreter/Interpreter.cpp:
(JSC::unwindCallFrame):
* jit/JIT.cpp:
(JSC::JIT::privateCompileMainPass):
(JSC::JIT::privateCompileSlowCases):
* jit/JIT.h:
* jit/JITOpcodes.cpp:
(JSC::JIT::emit_op_captured_mov): Deleted.
(JSC::JIT::emit_op_tear_off_lexical_environment): Deleted.
(JSC::JIT::emitSlow_op_captured_mov): Deleted.
* jit/JITOpcodes32_64.cpp:
(JSC::JIT::emit_op_captured_mov): Deleted.
(JSC::JIT::emit_op_tear_off_lexical_environment): Deleted.
* jit/JITOperations.cpp:
* jit/JITOperations.h:
* jit/JITPropertyAccess.cpp:
(JSC::JIT::emit_op_resolve_scope):
(JSC::JIT::emit_op_get_from_scope):
(JSC::JIT::emitPutClosureVar):
(JSC::JIT::emit_op_put_to_scope):
(JSC::JIT::emitSlow_op_put_to_scope):
* jit/JITPropertyAccess32_64.cpp:
(JSC::JIT::emit_op_resolve_scope):
(JSC::JIT::emit_op_get_from_scope):
(JSC::JIT::emitPutClosureVar):
(JSC::JIT::emit_op_put_to_scope):
(JSC::JIT::emitSlow_op_put_to_scope):
* llint/LLIntData.cpp:
(JSC::LLInt::Data::performAssertions):
* llint/LLIntSlowPaths.cpp:
(JSC::LLInt::LLINT_SLOW_PATH_DECL):
* llint/LLIntSlowPaths.h:
* llint/LowLevelInterpreter.asm:
* llint/LowLevelInterpreter32_64.asm:
* llint/LowLevelInterpreter64.asm:
* runtime/Arguments.cpp:
(JSC::Arguments::tearOff):
* runtime/Arguments.h:
(JSC::Arguments::argument):
* runtime/CommonSlowPaths.cpp:
(JSC::SLOW_PATH_DECL): Deleted.
* runtime/CommonSlowPaths.h:
* runtime/JSLexicalEnvironment.cpp:
(JSC::JSLexicalEnvironment::visitChildren):
(JSC::JSLexicalEnvironment::symbolTableGet):
(JSC::JSLexicalEnvironment::symbolTablePut):
(JSC::JSLexicalEnvironment::getOwnNonIndexPropertyNames):
(JSC::JSLexicalEnvironment::getOwnPropertySlot):
(JSC::JSLexicalEnvironment::argumentsGetter):
* runtime/JSLexicalEnvironment.h:
(JSC::JSLexicalEnvironment::create):
(JSC::JSLexicalEnvironment::JSLexicalEnvironment):
(JSC::JSLexicalEnvironment::tearOff): Deleted.
(JSC::JSLexicalEnvironment::isTornOff): Deleted.
* runtime/JSScope.cpp:
(JSC::resolveTypeName):
* runtime/JSScope.h:
(JSC::makeType):
(JSC::needsVarInjectionChecks):
* runtime/WriteBarrier.h:
(JSC::WriteBarrier<Unknown>::WriteBarrier):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@174226 268f45cc-cd09-0410-ab3c-d52691b4dbfc

5 years agoHave is<>(T*) function do a null check on the pointer argument
cdumez@apple.com [Thu, 2 Oct 2014 19:39:41 +0000 (19:39 +0000)]
Have is<>(T*) function do a null check on the pointer argument
https://bugs.webkit.org/show_bug.cgi?id=137333

Reviewed by Gavin Barraclough.

Have is<>(T*) function do a null check on the argument instead of a
simple assertion. This makes sense for 2 reasons:
1. It is more consistent with downcast<>(T*), which will succeed even
   if the argument is a nullptr.
2. It simplifies the code a bit as it gets rid of a lot of explicit
   null checks.

Source/WebCore:

No new tests, no behavior change.

* Modules/indexeddb/IDBFactory.cpp:
* Modules/webaudio/OfflineAudioContext.cpp:
(WebCore::OfflineAudioContext::create):
* Modules/websockets/ThreadableWebSocketChannel.cpp:
(WebCore::ThreadableWebSocketChannel::create):
* Modules/websockets/WebSocket.cpp:
(WebCore::WebSocket::connect):
* Modules/websockets/WebSocketHandshake.cpp:
(WebCore::WebSocketHandshake::clientHandshakeMessage):
* accessibility/AXObjectCache.cpp:
(WebCore::AXObjectCache::focusedUIElementForPage):
(WebCore::createFromRenderer):
(WebCore::AXObjectCache::getOrCreate):
(WebCore::AXObjectCache::handleLiveRegionCreated):
(WebCore::AXObjectCache::handleAttributeChanged):
(WebCore::AXObjectCache::labelChanged):
(WebCore::AXObjectCache::rootAXEditableElement):
(WebCore::isNodeAriaVisible):
* accessibility/AccessibilityListBoxOption.cpp:
(WebCore::AccessibilityListBoxOption::isEnabled):
(WebCore::AccessibilityListBoxOption::isSelected):
(WebCore::AccessibilityListBoxOption::canSetSelectedAttribute):
(WebCore::AccessibilityListBoxOption::stringValue):
(WebCore::AccessibilityListBoxOption::listBoxOptionParentNode):
* accessibility/AccessibilityMediaControls.cpp:
(WebCore::AccessibilityMediaTimeline::valueDescription):
* accessibility/AccessibilityMenuListPopup.cpp:
(WebCore::AccessibilityMenuListPopup::menuListOptionAccessibilityObject):
* accessibility/AccessibilityNodeObject.cpp:
(WebCore::AccessibilityNodeObject::determineAccessibilityRole):
(WebCore::AccessibilityNodeObject::canvasHasFallbackContent):
(WebCore::AccessibilityNodeObject::isNativeTextControl):
(WebCore::AccessibilityNodeObject::isNativeImage):
(WebCore::AccessibilityNodeObject::isInputImage):
(WebCore::AccessibilityNodeObject::isEnabled):
(WebCore::AccessibilityNodeObject::isPressed):
(WebCore::AccessibilityNodeObject::isHovered):
(WebCore::AccessibilityNodeObject::isReadOnly):
(WebCore::AccessibilityNodeObject::isRequired):
(WebCore::AccessibilityNodeObject::valueForRange):
(WebCore::AccessibilityNodeObject::maxValueForRange):
(WebCore::AccessibilityNodeObject::minValueForRange):
(WebCore::AccessibilityNodeObject::isControl):
(WebCore::AccessibilityNodeObject::anchorElement):
(WebCore::isNodeActionElement):
(WebCore::AccessibilityNodeObject::mouseButtonListener):
(WebCore::AccessibilityNodeObject::labelForElement):
(WebCore::AccessibilityNodeObject::titleElementText):
(WebCore::AccessibilityNodeObject::alternativeText):
(WebCore::AccessibilityNodeObject::visibleText):
(WebCore::AccessibilityNodeObject::accessibilityDescription):
(WebCore::AccessibilityNodeObject::helpText):
(WebCore::AccessibilityNodeObject::hierarchicalLevel):
(WebCore::AccessibilityNodeObject::textUnderElement):
(WebCore::AccessibilityNodeObject::title):
(WebCore::AccessibilityNodeObject::colorValue):
(WebCore::accessibleNameForNode):
(WebCore::AccessibilityNodeObject::canSetFocusAttribute):
* accessibility/AccessibilityObject.cpp:
(WebCore::AccessibilityObject::press):
(WebCore::AccessibilityObject::hasTagName):
(WebCore::AccessibilityObject::hasAttribute):
(WebCore::AccessibilityObject::element):
(WebCore::AccessibilityObject::classList):
* accessibility/AccessibilityRenderObject.cpp:
(WebCore::AccessibilityRenderObject::isFileUploadButton):
(WebCore::AccessibilityRenderObject::anchorElement):
(WebCore::AccessibilityRenderObject::helpText):
(WebCore::AccessibilityRenderObject::labelElementContainer):
(WebCore::AccessibilityRenderObject::internalLinkElement):
(WebCore::AccessibilityRenderObject::addRadioButtonGroupMembers):
(WebCore::AccessibilityRenderObject::titleUIElement):
(WebCore::AccessibilityRenderObject::computeAccessibilityIsIgnored):
(WebCore::AccessibilityRenderObject::accessKey):
(WebCore::AccessibilityRenderObject::url):
(WebCore::AccessibilityRenderObject::setElementAttributeValue):
(WebCore::AccessibilityRenderObject::setFocused):
(WebCore::AccessibilityRenderObject::setValue):
(WebCore::AccessibilityRenderObject::getDocumentLinks):
(WebCore::AccessibilityRenderObject::accessibilityImageMapHitTest):
(WebCore::AccessibilityRenderObject::accessibilityHitTest):
(WebCore::AccessibilityRenderObject::correspondingLabelForControlElement):
(WebCore::AccessibilityRenderObject::renderObjectIsObservable):
(WebCore::AccessibilityRenderObject::determineAccessibilityRole):
(WebCore::AccessibilityRenderObject::inheritsPresentationalRole):
(WebCore::AccessibilityRenderObject::addTextFieldChildren):
(WebCore::AccessibilityRenderObject::remoteSVGRootElement):
(WebCore::AccessibilityRenderObject::setAccessibleName):
(WebCore::AccessibilityRenderObject::stringValueForMSAA):
(WebCore::AccessibilityRenderObject::isLinked):
(WebCore::AccessibilityRenderObject::stringRoleForMSAA):
(WebCore::AccessibilityRenderObject::isMathElement):
* accessibility/AccessibilitySearchFieldButtons.cpp:
(WebCore::AccessibilitySearchFieldCancelButton::press):
* accessibility/AccessibilityTable.cpp:
(WebCore::AccessibilityTable::tableElement):
(WebCore::AccessibilityTable::isDataTable):
(WebCore::AccessibilityTable::isTableExposableThroughAccessibility):
(WebCore::AccessibilityTable::title):
* accessibility/atk/AXObjectCacheAtk.cpp:
(WebCore::notifyChildrenSelectionChange):
* accessibility/atk/WebKitAccessibleInterfaceTable.cpp:
(webkitAccessibleTableGetCaption):
* accessibility/atk/WebKitAccessibleWrapperAtk.cpp:
(webkitAccessibleGetName):
(webkitAccessibleGetDescription):
(webkitAccessibleGetAttributes):
* bindings/gobject/WebKitDOMPrivate.cpp:
(WebKit::wrap):
* bindings/js/JSAudioContextCustom.cpp:
(WebCore::constructJSAudioContext):
* bindings/js/JSDOMGlobalObject.cpp:
(WebCore::toJSDOMGlobalObject):
* bindings/js/JSDOMWindowCustom.cpp:
(WebCore::namedItemGetter):
(WebCore::JSDOMWindow::getOwnPropertySlot):
(WebCore::JSDOMWindow::getOwnPropertySlotByIndex):
* bindings/js/JSElementCustom.cpp:
(WebCore::toJSNewlyCreated):
* bindings/js/JSErrorHandler.cpp:
(WebCore::JSErrorHandler::handleEvent):
* bindings/js/JSEventListener.cpp:
(WebCore::JSEventListener::handleEvent):
* bindings/js/JSHTMLDocumentCustom.cpp:
(WebCore::JSHTMLDocument::nameGetter):
* bindings/js/JSLazyEventListener.cpp:
(WebCore::JSLazyEventListener::initializeJSFunction):
* bindings/js/JSNodeCustom.cpp:
(WebCore::isReachableFromDOM):
(WebCore::createWrapperInline):
* bindings/js/ScheduledAction.cpp:
(WebCore::ScheduledAction::execute):
* bindings/js/ScriptState.cpp:
(WebCore::frameFromExecState):
* bindings/objc/DOM.mm:
(kitClass):
* css/CSSCursorImageValue.cpp:
(WebCore::resourceReferencedByCursorElement):
* css/CSSStyleSheet.cpp:
(WebCore::isAcceptableCSSStyleSheetParent):
* css/SelectorChecker.cpp:
(WebCore::SelectorChecker::checkOne):
* css/SelectorCheckerTestFunctions.h:
(WebCore::isAutofilled):
(WebCore::isDisabled):
(WebCore::isEnabled):
(WebCore::isChecked):
(WebCore::matchesLangPseudoClass):
(WebCore::matchesFutureCuePseudoClass):
(WebCore::matchesPastCuePseudoClass):
* css/StyleResolver.cpp:
(WebCore::StyleResolver::State::initElement):
(WebCore::StyleResolver::locateCousinList):
(WebCore::elementHasDirectionAuto):
(WebCore::StyleResolver::canShareStyleWithElement):
(WebCore::StyleResolver::findSiblingForStyleSharing):
(WebCore::StyleResolver::adjustRenderStyle):
(WebCore::StyleResolver::useSVGZoomRulesForLength):
* css/StyleSheetList.cpp:
(WebCore::StyleSheetList::getNamedItem):
* dom/Attr.cpp:
(WebCore::Attr::style):
* dom/CharacterData.cpp:
(WebCore::CharacterData::parserAppendData):
(WebCore::CharacterData::setDataAndUpdate):
* dom/ContainerNode.cpp:
(WebCore::checkAcceptChild):
(WebCore::ContainerNode::notifyChildRemoved):
* dom/ContainerNodeAlgorithms.cpp:
(WebCore::ChildNodeInsertionNotifier::notifyDescendantInsertedIntoTree):
(WebCore::ChildNodeRemovalNotifier::notifyDescendantRemovedFromTree):
* dom/CurrentScriptIncrementer.h:
(WebCore::CurrentScriptIncrementer::CurrentScriptIncrementer):
* dom/DataTransfer.cpp:
(WebCore::DataTransfer::setDragImage):
* dom/Document.cpp:
(WebCore::Document::elementFromPoint):
(WebCore::Document::setTitle):
(WebCore::Document::iconURLs):
(WebCore::eventTargetElementForDocument):
(WebCore::Document::updateHoverActiveState):
* dom/DocumentStyleSheetCollection.cpp:
(WebCore::DocumentStyleSheetCollection::collectActiveStyleSheets):
* dom/Element.cpp:
(WebCore::Element::computeInheritedLanguage):
* dom/Element.h:
(WebCore::Node::parentElement):
* dom/ElementTraversal.h:
(WebCore::Traversal<Element>::nextTemplate):
(WebCore::ElementTraversal::previousIncludingPseudo):
(WebCore::ElementTraversal::nextIncludingPseudo):
(WebCore::ElementTraversal::nextIncludingPseudoSkippingChildren):
(WebCore::ElementTraversal::pseudoAwarePreviousSibling):
* dom/EventDispatcher.cpp:
(WebCore::WindowEventContext::WindowEventContext):
(WebCore::EventDispatcher::dispatchEvent):
(WebCore::nodeOrHostIfPseudoElement):
(WebCore::EventPath::EventPath):
* dom/EventTarget.cpp:
(WebCore::EventTarget::fireEventListeners):
* dom/KeyboardEvent.cpp:
(WebCore::findKeyboardEvent):
* dom/LiveNodeList.cpp:
(WebCore::LiveNodeList::namedItem):
* dom/MessagePort.cpp:
(WebCore::MessagePort::dispatchMessages):
* dom/MouseEvent.cpp:
(WebCore::SimulatedMouseEvent::SimulatedMouseEvent):
* dom/NamedNodeMap.cpp:
(WebCore::NamedNodeMap::setNamedItem):
* dom/Node.cpp:
(WebCore::Node::dumpStatistics):
(WebCore::Node::pseudoAwarePreviousSibling):
(WebCore::Node::pseudoAwareNextSibling):
(WebCore::Node::computedStyle):
(WebCore::Node::parentOrShadowHostElement):
(WebCore::Node::rootEditableElement):
(WebCore::Node::ancestorElement):
(WebCore::Node::compareDocumentPosition):
(WebCore::appendAttributeDesc):
(WebCore::Node::showNodePathForThis):
(WebCore::Node::enclosingLinkEventParentOrSelf):
(WebCore::Node::dispatchEvent):
(WebCore::Node::defaultEventHandler):
* dom/NodeRenderingTraversal.cpp:
(WebCore::NodeRenderingTraversal::nodeCanBeDistributed):
(WebCore::NodeRenderingTraversal::traverseParent):
* dom/NodeTraversal.cpp:
(WebCore::NodeTraversal::previousIncludingPseudo):
(WebCore::NodeTraversal::nextIncludingPseudo):
(WebCore::NodeTraversal::nextIncludingPseudoSkippingChildren):
* dom/Position.cpp:
(WebCore::Position::element):
(WebCore::endsOfNodeAreVisuallyDistinctPositions):
(WebCore::Position::isRenderedCharacter):
(WebCore::Position::leadingWhitespacePosition):
* dom/Range.cpp:
(WebCore::Range::insertNode):
(WebCore::Range::getBorderAndTextQuads):
* dom/ScriptElement.cpp:
(WebCore::toScriptElementIfPossible):
* dom/ShadowRoot.h:
(WebCore::Node::parentOrShadowHostNode):
* dom/Text.cpp:
(WebCore::earliestLogicallyAdjacentTextNode):
(WebCore::latestLogicallyAdjacentTextNode):
(WebCore::isSVGShadowText):
* dom/TextNodeTraversal.cpp:
(WebCore::TextNodeTraversal::contentsAsString):
* dom/TextNodeTraversal.h:
(WebCore::TextNodeTraversal::firstTextChildTemplate):
(WebCore::TextNodeTraversal::firstTextWithinTemplate):
(WebCore::TextNodeTraversal::traverseNextTextTemplate):
(WebCore::TextNodeTraversal::nextSibling):
* dom/TreeScopeAdopter.cpp:
(WebCore::TreeScopeAdopter::moveTreeToNewScope):
(WebCore::TreeScopeAdopter::moveNodeToNewDocument):
* editing/ApplyBlockElementCommand.cpp:
(WebCore::isNewLineAtPosition):
(WebCore::ApplyBlockElementCommand::endOfNextParagrahSplittingTextNodesIfNeeded):
* editing/ApplyStyleCommand.cpp:
(WebCore::isLegacyAppleStyleSpan):
(WebCore::isStyleSpanOrSpanWithOnlyStyleAttribute):
(WebCore::isSpanWithoutAttributesOrUnstyledStyleSpan):
(WebCore::isEmptyFontTag):
(WebCore::ApplyStyleCommand::applyBlockStyle):
(WebCore::ApplyStyleCommand::applyRelativeFontStyleChange):
(WebCore::dummySpanAncestorForNode):
(WebCore::ApplyStyleCommand::splitAncestorsWithUnicodeBidi):
(WebCore::ApplyStyleCommand::removeEmbeddingUpToEnclosingBlock):
(WebCore::ApplyStyleCommand::highestAncestorWithConflictingInlineStyle):
(WebCore::ApplyStyleCommand::applyInlineStyleToPushDown):
(WebCore::ApplyStyleCommand::splitTextAtEnd):
(WebCore::ApplyStyleCommand::splitTextElementAtEnd):
(WebCore::ApplyStyleCommand::shouldSplitTextElement):
(WebCore::ApplyStyleCommand::surroundNodeRangeWithElement):
(WebCore::ApplyStyleCommand::applyInlineStyleChange):
(WebCore::ApplyStyleCommand::joinChildTextNodes):
* editing/BreakBlockquoteCommand.cpp:
(WebCore::BreakBlockquoteCommand::doApply):
* editing/CompositeEditCommand.cpp:
(WebCore::CompositeEditCommand::isRemovableBlock):
(WebCore::CompositeEditCommand::insertNodeAt):
(WebCore::CompositeEditCommand::canRebalance):
(WebCore::CompositeEditCommand::prepareWhitespaceAtPositionForSplit):
(WebCore::CompositeEditCommand::deleteInsignificantText):
(WebCore::CompositeEditCommand::removePlaceholderAt):
(WebCore::CompositeEditCommand::cleanupAfterDeletion):
(WebCore::CompositeEditCommand::moveParagraphs):
* editing/DeleteButtonController.cpp:
(WebCore::enclosingDeletableElement):
* editing/DeleteSelectionCommand.cpp:
(WebCore::DeleteSelectionCommand::handleGeneralDelete):
(WebCore::DeleteSelectionCommand::fixupWhitespace):
(WebCore::DeleteSelectionCommand::doApply):
* editing/EditingStyle.cpp:
(WebCore::EditingStyle::wrappingStyleForSerialization):
* editing/Editor.cpp:
(WebCore::imageElementFromImageDocument):
(WebCore::Editor::shouldInsertFragment):
(WebCore::Editor::setBaseWritingDirection):
(WebCore::Editor::setComposition):
(WebCore::Editor::isSpellCheckingEnabledFor):
(WebCore::Editor::applyEditingStyleToElement):
* editing/FormatBlockCommand.cpp:
(WebCore::isElementForFormatBlock):
* editing/FrameSelection.cpp:
(WebCore::removingNodeRemovesPosition):
(WebCore::CaretBase::paintCaret):
(WebCore::FrameSelection::debugRenderer):
(WebCore::FrameSelection::selectAll):
(WebCore::scanForForm):
* editing/InsertLineBreakCommand.cpp:
(WebCore::InsertLineBreakCommand::doApply):
* editing/InsertListCommand.cpp:
(WebCore::InsertListCommand::mergeWithNeighboringLists):
* editing/InsertParagraphSeparatorCommand.cpp:
(WebCore::InsertParagraphSeparatorCommand::doApply):
* editing/InsertTextCommand.cpp:
(WebCore::InsertTextCommand::insertTab):
* editing/ReplaceSelectionCommand.cpp:
(WebCore::haveSameTagName):
(WebCore::ReplaceSelectionCommand::removeRedundantStylesAndKeepStyleSpanInline):
(WebCore::ReplaceSelectionCommand::removeUnrenderedTextNodesAtEnds):
(WebCore::ReplaceSelectionCommand::shouldPerformSmartReplace):
(WebCore::ReplaceSelectionCommand::addSpacesForSmartReplace):
(WebCore::ReplaceSelectionCommand::mergeTextNodesAroundPosition):
(WebCore::ReplaceSelectionCommand::insertAsListItems):
(WebCore::ReplaceSelectionCommand::performTrivialReplace):
* editing/SpellChecker.cpp:
(WebCore::SpellChecker::isCheckable):
* editing/SplitTextNodeContainingElementCommand.cpp:
(WebCore::SplitTextNodeContainingElementCommand::doApply):
* editing/TextIterator.cpp:
(WebCore::isRendererReplacedElement):
* editing/VisibleSelection.cpp:
(WebCore::VisibleSelection::isInPasswordField):
* editing/VisibleUnits.cpp:
(WebCore::startPositionForLine):
(WebCore::endPositionForLine):
(WebCore::startOfParagraph):
(WebCore::endOfParagraph):
* editing/cocoa/HTMLConverter.mm:
(HTMLConverter::aggregatedAttributesForAncestors):
(HTMLConverter::aggregatedAttributesForElementAndItsAncestors):
(WebCore::editingAttributedStringFromRange):
* editing/htmlediting.cpp:
(WebCore::enclosingBlock):
(WebCore::enclosingElementWithTag):
(WebCore::enclosingAnchorElement):
(WebCore::enclosingList):
(WebCore::lineBreakExistsAtPosition):
(WebCore::areIdenticalElements):
(WebCore::isNonTableCellHTMLBlockElement):
(WebCore::deprecatedEnclosingBlockFlowElement):
* editing/ios/EditorIOS.mm:
(WebCore::Editor::setTextAlignmentForChangedBaseWritingDirection):
* editing/markup.cpp:
(WebCore::StyledMarkupAccumulator::appendText):
(WebCore::ancestorToRetainStructureAndAppearanceForBlock):
(WebCore::isPlainTextMarkup):
* fileapi/FileReader.cpp:
(WebCore::FileReader::readAsArrayBuffer):
(WebCore::FileReader::readAsBinaryString):
(WebCore::FileReader::readAsText):
(WebCore::FileReader::readAsDataURL):
* html/BaseChooserOnlyDateAndTimeInputType.cpp:
(WebCore::BaseChooserOnlyDateAndTimeInputType::updateAppearance):
* html/FTPDirectoryDocument.cpp:
(WebCore::FTPDirectoryDocumentParser::loadDocumentTemplate):
* html/FormAssociatedElement.cpp:
(WebCore::FormAssociatedElement::findAssociatedForm):
* html/HTMLAnchorElement.cpp:
(WebCore::appendServerMapMousePosition):
(WebCore::HTMLAnchorElement::defaultEventHandler):
(WebCore::HTMLAnchorElement::eventType):
(WebCore::isEnterKeyKeydownEvent):
(WebCore::isLinkClick):
* html/HTMLAreaElement.cpp:
(WebCore::HTMLAreaElement::imageElement):
* html/HTMLBodyElement.cpp:
(WebCore::HTMLBodyElement::insertedInto):
* html/HTMLButtonElement.cpp:
(WebCore::HTMLButtonElement::defaultEventHandler):
* html/HTMLCollection.cpp:
(WebCore::HTMLCollection::namedItem):
(WebCore::HTMLCollection::updateNamedElementCache):
* html/HTMLDocument.cpp:
(WebCore::HTMLDocument::bgColor):
(WebCore::HTMLDocument::setBgColor):
(WebCore::HTMLDocument::fgColor):
(WebCore::HTMLDocument::setFgColor):
(WebCore::HTMLDocument::alinkColor):
(WebCore::HTMLDocument::setAlinkColor):
(WebCore::HTMLDocument::linkColor):
(WebCore::HTMLDocument::setLinkColor):
(WebCore::HTMLDocument::vlinkColor):
(WebCore::HTMLDocument::setVlinkColor):
(WebCore::HTMLDocument::isFrameSet):
* html/HTMLElement.cpp:
(WebCore::HTMLElement::matchesReadWritePseudoClass):
(WebCore::mergeWithNextTextNode):
(WebCore::HTMLElement::setOuterHTML):
(WebCore::HTMLElement::setOuterText):
(WebCore::HTMLElement::insertAdjacentElement):
(WebCore::contextElementForInsertion):
(WebCore::HTMLElement::directionality):
(WebCore::HTMLElement::dirAttributeChanged):
* html/HTMLEmbedElement.cpp:
(WebCore::HTMLEmbedElement::rendererIsNeeded):
* html/HTMLFieldSetElement.cpp:
(WebCore::updateFromControlElementsAncestorDisabledStateUnder):
(WebCore::HTMLFieldSetElement::disabledStateChanged):
* html/HTMLFormControlElement.cpp:
(WebCore::HTMLFormControlElement::computeIsDisabledByFieldsetAncestor):
(WebCore::shouldAutofocus):
(WebCore::HTMLFormControlElement::enclosingFormControlElement):
* html/HTMLFormElement.cpp:
(WebCore::HTMLFormElement::rendererIsNeeded):
(WebCore::HTMLFormElement::submitImplicitly):
(WebCore::submitElementFromEvent):
(WebCore::HTMLFormElement::validateInteractively):
(WebCore::HTMLFormElement::submit):
(WebCore::HTMLFormElement::reset):
(WebCore::HTMLFormElement::defaultButton):
(WebCore::HTMLFormElement::documentDidResumeFromPageCache):
* html/HTMLFrameOwnerElement.cpp:
(WebCore::HTMLFrameOwnerElement::getSVGDocument):
* html/HTMLFrameSetElement.cpp:
(WebCore::HTMLFrameSetElement::defaultEventHandler):
* html/HTMLInputElement.cpp:
(WebCore::HTMLInputElement::defaultEventHandler):
(WebCore::HTMLInputElement::dataList):
* html/HTMLLabelElement.cpp:
(WebCore::nodeAsSupportedLabelableElement):
* html/HTMLLegendElement.cpp:
(WebCore::HTMLLegendElement::virtualForm):
* html/HTMLMediaElement.cpp:
(WebCore::HTMLMediaElement::parseAttribute):
* html/HTMLObjectElement.cpp:
(WebCore::HTMLObjectElement::hasFallbackContent):
(WebCore::HTMLObjectElement::updateDocNamedItem):
* html/HTMLOptGroupElement.cpp:
(WebCore::HTMLOptGroupElement::recalcSelectOptions):
(WebCore::HTMLOptGroupElement::ownerSelectElement):
* html/HTMLOptionElement.cpp:
(WebCore::HTMLOptionElement::setText):
(WebCore::HTMLOptionElement::index):
(WebCore::HTMLOptionElement::ownerDataListElement):
(WebCore::HTMLOptionElement::ownerSelectElement):
(WebCore::HTMLOptionElement::textIndentedToRespectGroupLabel):
(WebCore::HTMLOptionElement::isDisabledFormControl):
(WebCore::HTMLOptionElement::collectOptionInnerText):
* html/HTMLPlugInImageElement.cpp:
(WebCore::addPlugInsFromNodeListMatchingPlugInOrigin):
(WebCore::HTMLPlugInImageElement::defaultEventHandler):
* html/HTMLScriptElement.cpp:
(WebCore::HTMLScriptElement::setText):
* html/HTMLSelectElement.cpp:
(WebCore::HTMLSelectElement::nextValidIndex):
(WebCore::HTMLSelectElement::saveLastSelection):
(WebCore::HTMLSelectElement::setActiveSelectionAnchorIndex):
(WebCore::HTMLSelectElement::updateListBoxSelection):
(WebCore::HTMLSelectElement::listBoxOnChange):
(WebCore::HTMLSelectElement::recalcListItems):
(WebCore::HTMLSelectElement::selectedIndex):
(WebCore::HTMLSelectElement::selectOption):
(WebCore::HTMLSelectElement::optionToListIndex):
(WebCore::HTMLSelectElement::listToOptionIndex):
(WebCore::HTMLSelectElement::deselectItemsWithoutValidation):
(WebCore::HTMLSelectElement::saveFormControlState):
(WebCore::HTMLSelectElement::searchOptionsForValue):
(WebCore::HTMLSelectElement::restoreFormControlState):
(WebCore::HTMLSelectElement::appendFormData):
(WebCore::HTMLSelectElement::reset):
(WebCore::HTMLSelectElement::menuListDefaultEventHandler):
(WebCore::HTMLSelectElement::updateSelectedState):
(WebCore::HTMLSelectElement::listBoxDefaultEventHandler):
(WebCore::HTMLSelectElement::defaultEventHandler):
(WebCore::HTMLSelectElement::lastSelectedListIndex):
(WebCore::HTMLSelectElement::optionAtIndex):
(WebCore::HTMLSelectElement::accessKeySetSelectedIndex):
(WebCore::HTMLSelectElement::length):
* html/HTMLSourceElement.cpp:
(WebCore::HTMLSourceElement::insertedInto):
(WebCore::HTMLSourceElement::removedFrom):
* html/HTMLSummaryElement.cpp:
(WebCore::isClickableControl):
(WebCore::HTMLSummaryElement::defaultEventHandler):
* html/HTMLTableElement.cpp:
(WebCore::HTMLTableElement::caption):
* html/HTMLTablePartElement.cpp:
(WebCore::HTMLTablePartElement::findParentTable):
* html/HTMLTableRowElement.cpp:
(WebCore::HTMLTableRowElement::rowIndex):
* html/HTMLTableRowsCollection.cpp:
(WebCore::HTMLTableRowsCollection::rowAfter):
(WebCore::HTMLTableRowsCollection::lastRow):
* html/HTMLTextAreaElement.cpp:
(WebCore::HTMLTextAreaElement::defaultEventHandler):
* html/HTMLTextFormControlElement.cpp:
(WebCore::HTMLTextFormControlElement::innerTextValue):
(WebCore::positionForIndex):
(WebCore::HTMLTextFormControlElement::indexForPosition):
(WebCore::HTMLTextFormControlElement::valueWithHardLineBreaks):
* html/HTMLTitleElement.cpp:
(WebCore::HTMLTitleElement::setText):
* html/HTMLTrackElement.cpp:
(WebCore::HTMLTrackElement::mediaElement):
* html/ImageDocument.cpp:
(WebCore::ImageEventListener::handleEvent):
* html/InputType.cpp:
(WebCore::InputType::shouldSubmitImplicitly):
* html/MediaDocument.cpp:
(WebCore::ancestorVideoElement):
(WebCore::MediaDocument::defaultEventHandler):
* html/RadioInputType.cpp:
(WebCore::RadioInputType::handleKeydownEvent):
(WebCore::RadioInputType::isKeyboardFocusable):
* html/TextFieldInputType.cpp:
(WebCore::TextFieldInputType::shouldSubmitImplicitly):
* html/parser/HTMLConstructionSite.cpp:
(WebCore::hasImpliedEndTag):
(WebCore::HTMLConstructionSite::attachLater):
(WebCore::HTMLConstructionSite::insertHTMLFormElement):
(WebCore::HTMLConstructionSite::insertTextNode):
(WebCore::HTMLConstructionSite::ownerDocumentForCurrentNode):
(WebCore::HTMLConstructionSite::findFosterSite):
* html/parser/HTMLElementStack.cpp:
(WebCore::HTMLNames::isScopeMarker):
(WebCore::HTMLNames::isTableScopeMarker):
(WebCore::HTMLNames::isSelectScopeMarker):
* html/parser/HTMLTreeBuilder.cpp:
(WebCore::HTMLTreeBuilder::HTMLTreeBuilder):
(WebCore::HTMLTreeBuilder::processStartTagForInBody):
(WebCore::HTMLTreeBuilder::processStartTag):
(WebCore::HTMLTreeBuilder::resetInsertionModeAppropriately):
(WebCore::HTMLTreeBuilder::processEndTag):
(WebCore::HTMLTreeBuilder::processCharacterBuffer):
* html/shadow/InsertionPoint.h:
(WebCore::isActiveInsertionPoint):
(WebCore::parentNodeForDistribution):
(WebCore::parentElementForDistribution):
* html/shadow/MediaControlElementTypes.cpp:
(WebCore::parentMediaElement):
(WebCore::mediaControlElementType):
(WebCore::MediaControlVolumeSliderElement::defaultEventHandler):
* html/shadow/MediaControlElements.cpp:
(WebCore::MediaControlPanelElement::defaultEventHandler):
(WebCore::MediaControlVolumeSliderContainerElement::defaultEventHandler):
(WebCore::MediaControlClosedCaptionsTrackListElement::defaultEventHandler):
(WebCore::MediaControlTimelineElement::defaultEventHandler):
* html/shadow/MediaControls.cpp:
(WebCore::MediaControls::containsRelatedTarget):
* html/shadow/MediaControlsApple.cpp:
(WebCore::MediaControlsAppleEventListener::handleEvent):
* html/shadow/SliderThumbElement.cpp:
(WebCore::SliderThumbElement::defaultEventHandler):
* html/shadow/SpinButtonElement.cpp:
(WebCore::SpinButtonElement::defaultEventHandler):
(WebCore::SpinButtonElement::forwardEvent):
* html/shadow/TextControlInnerElements.cpp:
(WebCore::SearchFieldResultsButtonElement::defaultEventHandler):
(WebCore::SearchFieldCancelButtonElement::defaultEventHandler):
* html/track/VTTCue.cpp:
(WebCore::VTTCue::copyWebVTTNodeToDOMTree):
(WebCore::VTTCue::markFutureAndPastNodes):
* inspector/InspectorCSSAgent.cpp:
(WebCore::InspectorCSSAgent::elementForId):
(WebCore::InspectorCSSAgent::buildObjectForAttributesStyle):
* inspector/InspectorDOMAgent.cpp:
(WebCore::InspectorDOMAgent::unbind):
(WebCore::InspectorDOMAgent::assertDocument):
(WebCore::InspectorDOMAgent::assertElement):
(WebCore::InspectorDOMAgent::querySelector):
(WebCore::InspectorDOMAgent::querySelectorAll):
(WebCore::InspectorDOMAgent::setNodeName):
(WebCore::InspectorDOMAgent::buildObjectForNode):
(WebCore::InspectorDOMAgent::buildObjectForAccessibilityProperties):
(WebCore::InspectorDOMAgent::innerParentNode):
* inspector/InspectorFrontendHost.cpp:
(WebCore::InspectorFrontendHost::dispatchEventAsContextMenuEvent):
* inspector/InspectorInstrumentation.cpp:
(WebCore::frameForScriptExecutionContext):
(WebCore::InspectorInstrumentation::instrumentingAgentsForNonDocumentContext):
* inspector/InspectorInstrumentation.h:
(WebCore::InspectorInstrumentation::instrumentingAgentsForContext):
* inspector/InspectorNodeFinder.cpp:
(WebCore::InspectorNodeFinder::searchUsingDOMTreeTraversal):
(WebCore::InspectorNodeFinder::searchUsingXPath):
(WebCore::InspectorNodeFinder::searchUsingCSSSelectors):
* inspector/InspectorOverlay.cpp:
(WebCore::buildObjectForElementInfo):
* inspector/InspectorStyleSheet.cpp:
(WebCore::InspectorStyleSheet::inlineStyleSheetText):
* loader/FormSubmission.cpp:
(WebCore::FormSubmission::create):
* loader/FrameLoader.cpp:
(WebCore::FrameLoader::handleFallbackContent):
* loader/SubframeLoader.cpp:
(WebCore::SubframeLoader::loadPlugin):
* loader/ThreadableLoader.cpp:
(WebCore::ThreadableLoader::create):
(WebCore::ThreadableLoader::loadResourceSynchronously):
* loader/cache/MemoryCache.cpp:
(WebCore::MemoryCache::removeRequestFromCache):
(WebCore::MemoryCache::removeRequestFromSessionCaches):
* mathml/MathMLElement.cpp:
(WebCore::MathMLElement::attributeChanged):
* mathml/MathMLSelectElement.cpp:
(WebCore::MathMLSelectElement::getSelectedSemanticsChild):
* page/Chrome.cpp:
(WebCore::Chrome::setToolTip):
* page/ContextMenuController.cpp:
(WebCore::ContextMenuController::maybeCreateContextMenu):
(WebCore::ContextMenuController::populate):
* page/DOMTimer.cpp:
(WebCore::DOMTimer::install):
(WebCore::DOMTimer::fired):
* page/DragController.cpp:
(WebCore::elementUnderMouse):
(WebCore::DragController::operationForLoad):
(WebCore::DragController::canProcessDrag):
(WebCore::DragController::draggableElement):
* page/EventHandler.cpp:
(WebCore::EventHandler::handleMousePressEvent):
(WebCore::isSubmitImage):
(WebCore::targetIsFrame):
(WebCore::findDropZone):
(WebCore::EventHandler::updateDragAndDrop):
(WebCore::EventHandler::updateMouseEventTargetNode):
(WebCore::EventHandler::handleTextInputEvent):
* page/FocusController.cpp:
(WebCore::FocusNavigationScope::owner):
(WebCore::FocusController::findFocusableElementDescendingDownIntoFrameDocument):
(WebCore::FocusController::findFocusableElementAcrossFocusScope):
(WebCore::FocusController::findElementWithExactTabIndex):
(WebCore::nextElementWithGreaterTabIndex):
(WebCore::previousElementWithLowerTabIndex):
(WebCore::FocusController::nextFocusableElement):
(WebCore::FocusController::previousFocusableElement):
(WebCore::clearSelectionIfNeeded):
(WebCore::FocusController::advanceFocusDirectionally):
* page/Frame.cpp:
(WebCore::Frame::searchForLabelsBeforeElement):
(WebCore::Frame::setPageAndTextZoomFactors):
* page/FrameView.cpp:
(WebCore::FrameView::init):
(WebCore::FrameView::scrollToAnchor):
* page/PageSerializer.cpp:
(WebCore::PageSerializer::serializeFrame):
* page/SpatialNavigation.cpp:
(WebCore::FocusCandidate::FocusCandidate):
(WebCore::scrollInDirection):
(WebCore::scrollableEnclosingBoxOrParentFrameForNodeInDirection):
(WebCore::canScrollInDirection):
(WebCore::nodeRectInAbsoluteCoordinates):
(WebCore::areElementsOnSameLine):
* page/ios/FrameIOS.mm:
(WebCore::ancestorRespondingToClickEvents):
* platform/network/FormData.cpp:
(WebCore::FormData::appendKeyValuePairItems):
* rendering/HitTestResult.cpp:
(WebCore::HitTestResult::setInnerNode):
(WebCore::HitTestResult::setInnerNonSharedNode):
(WebCore::HitTestResult::title):
(WebCore::HitTestResult::innerTextIfTruncated):
(WebCore::HitTestResult::mediaSupportsFullscreen):
(WebCore::HitTestResult::enterFullscreenForVideo):
(WebCore::HitTestResult::mediaIsVideo):
(WebCore::HitTestResult::innerElement):
(WebCore::HitTestResult::innerNonSharedElement):
* rendering/RenderBlockFlow.cpp:
(WebCore::resizeTextPermitted):
* rendering/RenderBox.cpp:
(WebCore::RenderBox::sizesLogicalWidthToFitContent):
* rendering/RenderCounter.cpp:
(WebCore::planCounter):
* rendering/RenderDetailsMarker.cpp:
(WebCore::RenderDetailsMarker::isOpen):
* rendering/RenderEmbeddedObject.cpp:
(WebCore::RenderEmbeddedObject::handleUnavailablePluginIndicatorEvent):
* rendering/RenderFileUploadControl.cpp:
(WebCore::RenderFileUploadControl::uploadButton):
* rendering/RenderImage.cpp:
(WebCore::RenderImage::paintAreaElementFocusRing):
(WebCore::RenderImage::paintIntoRect):
(WebCore::RenderImage::imageMap):
* rendering/RenderLayer.cpp:
(WebCore::RenderLayer::scrollRectToVisible):
(WebCore::RenderLayer::resize):
(WebCore::RenderLayer::updateSnapOffsets):
* rendering/RenderLayerBacking.cpp:
(WebCore::isRestartedPlugin):
* rendering/RenderListBox.cpp:
(WebCore::RenderListBox::updateFromElement):
(WebCore::RenderListBox::addFocusRingRects):
(WebCore::RenderListBox::paintItemForeground):
(WebCore::RenderListBox::paintItemBackground):
* rendering/RenderMarquee.cpp:
(WebCore::RenderMarquee::marqueeSpeed):
* rendering/RenderMenuList.cpp:
(WebCore::selectedOptionCount):
(RenderMenuList::updateOptionsWidth):
(RenderMenuList::setTextFromOption):
(RenderMenuList::itemText):
(RenderMenuList::itemIsEnabled):
(RenderMenuList::itemIsLabel):
(RenderMenuList::itemIsSelected):
* rendering/RenderMeter.cpp:
(WebCore::RenderMeter::meterElement):
* rendering/RenderNamedFlowThread.cpp:
(WebCore::RenderNamedFlowThread::isChildAllowed):
* rendering/RenderObject.cpp:
(WebCore::RenderObject::addPDFURLRect):
(WebCore::RenderObject::shouldRespectImageOrientation):
(WebCore::RenderObject::updateDragState):
(WebCore::RenderObject::getUncachedPseudoStyle):
(WebCore::RenderObject::getTextDecorationColors):
* rendering/RenderProgress.cpp:
(WebCore::RenderProgress::progressElement):
* rendering/RenderSnapshottedPlugIn.cpp:
(WebCore::RenderSnapshottedPlugIn::handleEvent):
* rendering/RenderTable.cpp:
(WebCore::RenderTable::convertStyleLogicalWidthToComputedWidth):
(WebCore::RenderTable::convertStyleLogicalHeightToComputedHeight):
* rendering/RenderTableCell.cpp:
(WebCore::RenderTableCell::parseColSpanFromDOM):
(WebCore::RenderTableCell::parseRowSpanFromDOM):
* rendering/RenderTheme.cpp:
(WebCore::RenderTheme::isEnabled):
(WebCore::RenderTheme::isFocused):
(WebCore::RenderTheme::isPressed):
(WebCore::RenderTheme::isSpinUpButtonPartPressed):
(WebCore::RenderTheme::isReadOnlyControl):
(WebCore::RenderTheme::isHovered):
(WebCore::RenderTheme::isSpinUpButtonPartHovered):
(WebCore::RenderTheme::paintSliderTicks):
* rendering/RenderThemeGtk.cpp:
(WebCore::getMediaElementFromRenderObject):
(WebCore::nodeHasClass):
* rendering/RenderThemeMac.mm:
(WebCore::RenderThemeMac::updatePressedState):
(WebCore::RenderThemeMac::paintSliderThumb):
(WebCore::RenderThemeMac::paintSearchFieldCancelButton):
(WebCore::RenderThemeMac::paintSnapshottedPluginOverlay):
* rendering/RenderTreeAsText.cpp:
(WebCore::isEmptyOrUnstyledAppleStyleSpan):
(WebCore::RenderTreeAsText::writeRenderObject):
* rendering/TextAutosizer.cpp:
(WebCore::TextAutosizer::isAutosizingContainer):
(WebCore::TextAutosizer::containerContainsOneOfTags):
* rendering/mathml/RenderMathMLBlock.cpp:
(WebCore::RenderMathMLBlock::isChildAllowed):
* rendering/svg/RenderSVGGradientStop.cpp:
(WebCore::RenderSVGGradientStop::gradientElement):
* rendering/svg/RenderSVGTransformableContainer.cpp:
(WebCore::RenderSVGTransformableContainer::calculateLocalTransform):
* rendering/svg/SVGRenderingContext.cpp:
(WebCore::SVGRenderingContext::prepareToRenderSVGContent):
* style/StyleResolveTree.cpp:
(WebCore::Style::invalidateWhitespaceOnlyTextSiblingsAfterAttachIfNeeded):
(WebCore::Style::attachChildren):
(WebCore::Style::attachDistributedChildren):
(WebCore::Style::detachDistributedChildren):
(WebCore::Style::detachChildren):
(WebCore::Style::resolveShadowTree):
(WebCore::Style::resolveTree):
* svg/SVGAElement.cpp:
(WebCore::SVGAElement::defaultEventHandler):
* svg/SVGAltGlyphElement.cpp:
(WebCore::SVGAltGlyphElement::hasValidGlyphElements):
* svg/SVGAnimateMotionElement.cpp:
(WebCore::SVGAnimateMotionElement::hasValidAttributeType):
* svg/SVGDocument.cpp:
(WebCore::SVGDocument::rootElement):
* svg/SVGElement.cpp:
(WebCore::SVGElement::isOutermostSVGSVGElement):
(WebCore::SVGElement::ownerSVGElement):
(WebCore::SVGElement::viewportElement):
* svg/SVGFontData.cpp:
(WebCore::SVGFontData::applySVGGlyphSelection):
* svg/SVGFontFaceElement.cpp:
(WebCore::SVGFontFaceElement::associatedFontElement):
(WebCore::SVGFontFaceElement::rebuildFontFace):
* svg/SVGFontFaceSrcElement.cpp:
(WebCore::SVGFontFaceSrcElement::childrenChanged):
* svg/SVGGlyphElement.cpp:
(WebCore::SVGGlyphElement::invalidateGlyphCache):
* svg/SVGHKernElement.cpp:
(WebCore::SVGHKernElement::insertedInto):
(WebCore::SVGHKernElement::removedFrom):
* svg/SVGLengthContext.cpp:
(WebCore::SVGLengthContext::determineViewport):
* svg/SVGLinearGradientElement.cpp:
(WebCore::SVGLinearGradientElement::collectGradientAttributes):
* svg/SVGLocatable.cpp:
(WebCore::isViewportElement):
(WebCore::SVGLocatable::getTransformToElement):
* svg/SVGMPathElement.cpp:
(WebCore::SVGMPathElement::pathElement):
(WebCore::SVGMPathElement::notifyParentOfPathChange):
* svg/SVGPathElement.cpp:
(WebCore::SVGPathElement::invalidateMPathDependencies):
* svg/SVGPatternElement.cpp:
(WebCore::SVGPatternElement::collectPatternAttributes):
* svg/SVGRadialGradientElement.cpp:
(WebCore::SVGRadialGradientElement::collectGradientAttributes):
* svg/SVGSVGElement.cpp:
(WebCore::SVGSVGElement::setupInitialView):
* svg/SVGTextContentElement.cpp:
(WebCore::SVGTextContentElement::elementFromRenderer):
* svg/SVGUseElement.cpp:
(WebCore::dumpInstanceTree):
(WebCore::SVGUseElement::toClipPath):
(WebCore::SVGUseElement::expandUseElementsInShadowTree):
(WebCore::SVGUseElement::expandSymbolElementsInShadowTree):
* svg/SVGVKernElement.cpp:
(WebCore::SVGVKernElement::insertedInto):
(WebCore::SVGVKernElement::removedFrom):
* svg/animation/SVGSMILElement.cpp:
(WebCore::SVGSMILElement::buildPendingResource):
* svg/graphics/SVGImageCache.cpp:
(WebCore::SVGImageCache::imageForRenderer):
* testing/Internals.cpp:
(WebCore::Internals::shadowRootType):
(WebCore::Internals::visiblePlaceholder):
(WebCore::Internals::selectColorInColorChooser):
(WebCore::Internals::wasLastChangeUserEdit):
(WebCore::Internals::updateLayoutIgnorePendingStylesheetsAndRunPostLayoutTasks):
(WebCore::Internals::isSelectPopupVisible):
* testing/js/WebCoreTestSupport.cpp:
(WebCoreTestSupport::injectInternalsObject):
* xml/XMLHttpRequest.cpp:
(WebCore::XMLHttpRequest::open):
* xml/XPathFunctions.cpp:
(WebCore::XPath::expandedNameLocalPart):
(WebCore::XPath::FunLang::evaluate):
* xml/XPathNodeSet.cpp:
(WebCore::XPath::sortBlock):
(WebCore::XPath::NodeSet::sort):
(WebCore::XPath::findRootNode):
(WebCore::XPath::NodeSet::traversalSort):
* xml/parser/XMLDocumentParserLibxml2.cpp:
(WebCore::XMLDocumentParser::XMLDocumentParser):

Source/WebKit/ios:

* WebCoreSupport/WebChromeClientIOS.mm:
(WebChromeClientIOS::focusedElementChanged):

Source/WebKit/mac:

* WebCoreSupport/WebChromeClient.mm:
(WebChromeClient::focusedElementChanged):
* WebCoreSupport/WebEditorClient.mm:
(WebEditorClient::textFieldDidBeginEditing):
(WebEditorClient::textFieldDidEndEditing):
(WebEditorClient::textDidChangeInTextField):
(WebEditorClient::doTextFieldCommandFromEvent):
(WebEditorClient::textWillBeDeletedInTextField):
(WebEditorClient::textDidChangeInTextArea):
* WebView/WebHTMLRepresentation.mm:
(inputElementFromDOMElement):
(searchForLabelsBeforeElement):
* WebView/WebRenderLayer.mm:
(+[WebRenderLayer nameForLayer:]):

Source/WebKit/win:

* DOMCoreClasses.cpp:
(DOMElement::createInstance):
* DOMHTMLClasses.cpp:
(DOMHTMLDocument::body):
(DOMHTMLDocument::forms):
(DOMHTMLElement::idName):
(DOMHTMLElement::innerText):
(DOMHTMLElement::setInnerText):
(DOMHTMLFormElement::action):
(DOMHTMLFormElement::method):
(DOMHTMLOptionElement::text):
(DOMHTMLOptionElement::label):
(DOMHTMLInputElement::form):
(DOMHTMLInputElement::disabled):
(DOMHTMLInputElement::readOnly):
(DOMHTMLInputElement::setType):
(DOMHTMLInputElement::value):
(DOMHTMLInputElement::setValue):
(DOMHTMLInputElement::setValueForUser):
(DOMHTMLInputElement::select):
(DOMHTMLInputElement::setSelectionStart):
(DOMHTMLInputElement::selectionStart):
(DOMHTMLInputElement::setSelectionEnd):
(DOMHTMLInputElement::selectionEnd):
(DOMHTMLInputElement::isTextField):
(DOMHTMLInputElement::rectOnScreen):
(DOMHTMLInputElement::replaceCharactersInRange):
(DOMHTMLInputElement::selectedRange):
(DOMHTMLInputElement::setAutofilled):
(DOMHTMLInputElement::isAutofilled):
(DOMHTMLInputElement::isUserEdited):
(DOMHTMLTextAreaElement::form):
(DOMHTMLTextAreaElement::value):
(DOMHTMLTextAreaElement::setValue):
(DOMHTMLTextAreaElement::select):
(DOMHTMLTextAreaElement::isUserEdited):
* WebFrame.cpp:
(formElementFromDOMElement):
(inputElementFromDOMElement):
(WebFrame::elementWithName):

Source/WebKit2:

* Shared/WebRenderObject.cpp:
(WebKit::WebRenderObject::WebRenderObject):
* WebProcess/InjectedBundle/InjectedBundleHitTestResult.cpp:
(WebKit::InjectedBundleHitTestResult::mediaType):
* WebProcess/Plugins/PluginView.cpp:
(WebKit::PluginView::handleEvent):
* WebProcess/WebCoreSupport/WebChromeClient.cpp:
(WebKit::WebChromeClient::focusedElementChanged):
* WebProcess/WebCoreSupport/WebEditorClient.cpp:
(WebKit::WebEditorClient::textFieldDidBeginEditing):
(WebKit::WebEditorClient::textFieldDidEndEditing):
(WebKit::WebEditorClient::textDidChangeInTextField):
(WebKit::WebEditorClient::textDidChangeInTextArea):
(WebKit::WebEditorClient::doTextFieldCommandFromEvent):
(WebKit::WebEditorClient::textWillBeDeletedInTextField):
* WebProcess/WebPage/WebFrame.cpp:
(WebKit::WebFrame::containsAnyFormElements):
(WebKit::WebFrame::containsAnyFormControls):
* WebProcess/WebPage/WebPage.cpp:
(WebKit::WebPage::pdfDocumentForPrintingFrame):
(WebKit::WebPage::determinePrimarySnapshottedPlugIn):
* WebProcess/WebPage/ios/WebPageIOS.mm:
(WebKit::WebPage::blurAssistedNode):
(WebKit::WebPage::setAssistedNodeValue):
(WebKit::WebPage::setAssistedNodeValueAsNumber):
(WebKit::WebPage::setAssistedNodeSelectedIndex):
(WebKit::WebPage::getPositionInformation):
(WebKit::WebPage::performActionOnElement):
(WebKit::isAssistableNode):
(WebKit::WebPage::getAssistedNodeInformation):

Source/WTF:

* wtf/TypeCasts.h:
(WTF::is):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@174225 268f45cc-cd09-0410-ab3c-d52691b4dbfc

5 years agoObject allocation sinking should have a sound story for picking materialization points
fpizlo@apple.com [Thu, 2 Oct 2014 19:38:08 +0000 (19:38 +0000)]
Object allocation sinking should have a sound story for picking materialization points
https://bugs.webkit.org/show_bug.cgi?id=137315

Reviewed by Oliver Hunt.

The only missing piece was having the object allocation sinking phase locate materialization
points that were at CFG edges.

The logic for how and why this "just works" relies on some properties of critical edge
breaking, so I was fairly careful in how I did this. Also, this requires inserting things at
the "first origin node" of a block - that is the first node in a block that has a NodeOrigin
and therefore is allowed to exit. We basically had support for such a notion before, but
didn't close the loop on it; this patch does that.

Also I added the ability to provide a BasicBlock* as context for a DFG_ASSERT().

* dfg/DFGBasicBlock.cpp:
(JSC::DFG::BasicBlock::firstOriginNode):
(JSC::DFG::BasicBlock::firstOrigin):
* dfg/DFGBasicBlock.h:
* dfg/DFGCriticalEdgeBreakingPhase.cpp:
(JSC::DFG::CriticalEdgeBreakingPhase::breakCriticalEdge):
* dfg/DFGGraph.cpp:
(JSC::DFG::crash):
(JSC::DFG::Graph::handleAssertionFailure):
* dfg/DFGGraph.h:
* dfg/DFGLoopPreHeaderCreationPhase.cpp:
(JSC::DFG::createPreHeader):
* dfg/DFGNodeOrigin.h:
(JSC::DFG::NodeOrigin::isSet):
* dfg/DFGObjectAllocationSinkingPhase.cpp:
(JSC::DFG::ObjectAllocationSinkingPhase::determineMaterializationPoints):
(JSC::DFG::ObjectAllocationSinkingPhase::placeMaterializationPoints):
(JSC::DFG::ObjectAllocationSinkingPhase::promoteSunkenFields):
(JSC::DFG::ObjectAllocationSinkingPhase::createMaterialize):
* dfg/DFGValidate.cpp:
(JSC::DFG::Validate::validate):
* runtime/Options.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@174224 268f45cc-cd09-0410-ab3c-d52691b4dbfc

5 years agoUnreviewed, rolling out r174120.
commit-queue@webkit.org [Thu, 2 Oct 2014 19:14:57 +0000 (19:14 +0000)]
Unreviewed, rolling out r174120.
https://bugs.webkit.org/show_bug.cgi?id=137350

While number of failures is reasonable with 2 processes, test
results seem to be inconsistent between runs (Requested by
rfong on #webkit).

Reverted changeset:

"[Windows] Back to 2 child processes for NRWT on Windows."
http://trac.webkit.org/changeset/174120

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@174223 268f45cc-cd09-0410-ab3c-d52691b4dbfc

5 years agoClean up loops in CSSGradientValue.cpp
cabanier@adobe.com [Thu, 2 Oct 2014 18:35:30 +0000 (18:35 +0000)]
Clean up loops in CSSGradientValue.cpp
https://bugs.webkit.org/show_bug.cgi?id=137332

Reviewed by Andreas Kling.

This patch cleans up a couple of for loops by using C++11 syntax and
fixes some grammar in a comment.

No new tests, no behavior change.

* css/CSSGradientValue.cpp:
(WebCore::CSSGradientValue::gradientWithStylesResolved): Move for loops to C++11.
(WebCore::CSSGradientValue::addStops): Update grammar in comment for gradient midpoints.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@174222 268f45cc-cd09-0410-ab3c-d52691b4dbfc

5 years agoDumpRenderTree and WebKitTestRunner should use WebKitystemInterfaceYosemite when...
jeffm@apple.com [Thu, 2 Oct 2014 17:39:01 +0000 (17:39 +0000)]
DumpRenderTree and WebKitTestRunner should use WebKitystemInterfaceYosemite when building on Yosemite
https://bugs.webkit.org/show_bug.cgi?id=137347

Reviewed by Dan Bernstein.

* DumpRenderTree/mac/Configurations/DebugRelease.xcconfig:
Use WebKitSystemInterfaceYosemite on OS X 10.10.

* WebKitTestRunner/Configurations/DebugRelease.xcconfig:
Ditto.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@174221 268f45cc-cd09-0410-ab3c-d52691b4dbfc

5 years agoClean up: Move XPC forward declarations in JavaScriptCore to WTF SPI wrapper header
dbates@webkit.org [Thu, 2 Oct 2014 16:24:10 +0000 (16:24 +0000)]
Clean up: Move XPC forward declarations in JavaScriptCore to WTF SPI wrapper header
https://bugs.webkit.org/show_bug.cgi?id=137277

Reviewed by Alexey Proskuryakov.

Source/JavaScriptCore:

Use wtf/spi/darwin/XPCSPI.h instead of including the corresponding XPC headers/
forward declaring XPC functions.

* inspector/remote/RemoteInspector.mm:
* inspector/remote/RemoteInspectorXPCConnection.h:
* inspector/remote/RemoteInspectorXPCConnection.mm:

Source/WebKit2:

Use wtf/spi/darwin/XPCSPI.h instead of including the corresponding XPC headers/
forward declaring XPC functions.

* Platform/IPC/Connection.h:
* Platform/IPC/mac/ConnectionMac.mm:
* Shared/EntryPointUtilities/mac/XPCService/XPCServiceMain.Development.mm:
* Shared/EntryPointUtilities/mac/XPCService/XPCServiceMain.mm:
* UIProcess/Launcher/mac/ProcessLauncherMac.mm:

Source/WTF:

Towards centralizing the declarations of XPC functions used in JavaScriptCore
and WebKit2, add a header that wraps these function declarations called XPCSPI.h
and use it instead of including the corresponding XPC API/SPI headers/forward
declaring XPC functions. This will also aid in the effort to make WebKit2 build
for iOS with the public iOS SDK.

* WTF.xcodeproj/project.pbxproj:
* wtf/spi/darwin/XPCSPI.h: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@174220 268f45cc-cd09-0410-ab3c-d52691b4dbfc

5 years agoUse variadic templates for jsMakeNontrivialString
andersca@apple.com [Thu, 2 Oct 2014 16:14:25 +0000 (16:14 +0000)]
Use variadic templates for jsMakeNontrivialString
https://bugs.webkit.org/show_bug.cgi?id=137325

Reviewed by Sam Weinig.

* runtime/JSString.h:
(JSC::jsNontrivialString):
Add an overload that takes an rvalue reference to a String so we can transfer ownership easily.

* runtime/JSStringBuilder.h:
(JSC::jsMakeNontrivialString):
Make this a variadic function template, with a single-parameter version that can steal the string if it's OK to do so.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@174219 268f45cc-cd09-0410-ab3c-d52691b4dbfc

5 years ago[ Windows ] More Windows test gardening to get bots green Part 4.
roger_fong@apple.com [Thu, 2 Oct 2014 15:57:49 +0000 (15:57 +0000)]
[ Windows ] More Windows test gardening to get bots green Part 4.
* platform/win/TestExpectations: Hopefully the last batch of indiscriminate mass skipping.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@174218 268f45cc-cd09-0410-ab3c-d52691b4dbfc

5 years agoAX: Default orientation for aria scrollbars should be vertical
k.czech@samsung.com [Thu, 2 Oct 2014 15:32:01 +0000 (15:32 +0000)]
AX: Default orientation for aria scrollbars should be vertical
https://bugs.webkit.org/show_bug.cgi?id=137341

Reviewed by Chris Fleizach.

Source/WebCore:

Default orientation for aria scrollbars regarding w3c spec should be vertical

No new tests, covering by existing one.

* accessibility/AccessibilityRenderObject.cpp:
(WebCore::AccessibilityRenderObject::orientation):

LayoutTests:

Update expectation and test for having new test case.

* accessibility/aria-scrollbar-role-expected.txt:
* accessibility/aria-scrollbar-role.html:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@174217 268f45cc-cd09-0410-ab3c-d52691b4dbfc

5 years agoFixed the Inspector to be able to properly distinguish between scope types.
mark.lam@apple.com [Thu, 2 Oct 2014 15:15:03 +0000 (15:15 +0000)]
Fixed the Inspector to be able to properly distinguish between scope types.
<https://webkit.org/b/137279>

Source/JavaScriptCore:

Reviewed by Geoffrey Garen.

The pre-existing code incorrectly labels Catch Scopes and Function Name Scopes
as With Scopes.  This patch will fix this.

* bytecode/BytecodeList.json:
* bytecompiler/BytecodeGenerator.cpp:
(JSC::BytecodeGenerator::emitPushFunctionNameScope):
(JSC::BytecodeGenerator::emitPushCatchScope):
- These now passes stores the desired JSNameScope::Type in a bytecode operand.
* debugger/DebuggerScope.cpp:
(JSC::DebuggerScope::isCatchScope):
(JSC::DebuggerScope::isFunctionNameScope):
- Added queries to be able to explicitly test if the scope is a CatchScope
  or FunctionNameScope.  The FunctionNameScope is the case where the
  NameScope is used to capture the function name of a function expression.
* debugger/DebuggerScope.h:
* inspector/InjectedScriptSource.js:
* inspector/JSJavaScriptCallFrame.cpp:
(Inspector::JSJavaScriptCallFrame::scopeType):
* inspector/JSJavaScriptCallFrame.h:
* inspector/JSJavaScriptCallFramePrototype.cpp:
(Inspector::JSJavaScriptCallFramePrototype::finishCreation):
(Inspector::jsJavaScriptCallFrameConstantFUNCTION_NAME_SCOPE):
* inspector/protocol/Debugger.json:
* jit/CCallHelpers.h:
(JSC::CCallHelpers::setupArgumentsWithExecState):
* jit/JIT.h:
* jit/JITInlines.h:
(JSC::JIT::callOperation):
* jit/JITOpcodes.cpp:
(JSC::JIT::emit_op_push_name_scope):
* jit/JITOpcodes32_64.cpp:
(JSC::JIT::emit_op_push_name_scope):
* jit/JITOperations.cpp:
* jit/JITOperations.h:
* llint/LLIntSlowPaths.cpp:
(JSC::LLInt::LLINT_SLOW_PATH_DECL):
* llint/LowLevelInterpreter.asm:
* runtime/JSFunction.cpp:
(JSC::JSFunction::addNameScopeIfNeeded):
* runtime/JSNameScope.h:
(JSC::JSNameScope::create):
(JSC::JSNameScope::isFunctionNameScope):
(JSC::JSNameScope::isCatchScope):
(JSC::JSNameScope::JSNameScope):
- Now stores the JSNameScope::Type in a field.

Source/WebInspectorUI:

Reviewed by Geoffrey Garen and Joseph Pecoraro.

* Localizations/en.lproj/localizedStrings.js:
* UserInterface/Controllers/DebuggerManager.js:
* UserInterface/Models/ScopeChainNode.js:
* UserInterface/Views/ScopeChainDetailsSidebarPanel.js:
- Added handling of the FunctionNameScope case.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@174216 268f45cc-cd09-0410-ab3c-d52691b4dbfc

5 years ago[GTK] [JHBuild] Switch sourceware.org mirror to http.
clopez@igalia.com [Thu, 2 Oct 2014 12:30:00 +0000 (12:30 +0000)]
[GTK] [JHBuild] Switch sourceware.org mirror to http.

Unreviewed.

The ftp access is recently failing to work as expected, and the
bots fail to download the file.
The same mirror is available via http and seems to work better.

* gtk/jhbuild.modules:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@174209 268f45cc-cd09-0410-ab3c-d52691b4dbfc

5 years ago[EFL][WK2] test_ewk2_view is failed
ryuan.choi@gmail.com [Thu, 2 Oct 2014 11:14:43 +0000 (11:14 +0000)]
[EFL][WK2] test_ewk2_view is failed
https://bugs.webkit.org/show_bug.cgi?id=132983

Reviewed by Gyuyoung Kim.

Enable test_ewk2_view because test cases of test_ewk2_view are passed except
for few tests related to device_pixel_ratio.
This patch adds macro guard for ailed test case to CAIRO_SURFACE_SET_DEVICE_SCALE macro.
In addition, fixed some warnings while building with test_ewk2_view.

* PlatformEfl.cmake: Enables test_ewk2_view.
* UIProcess/API/efl/tests/UnitTestUtils/EWK2UnitTestBase.cpp: Added const keyword for argument to fix the warnings.
(EWK2UnitTest::EWK2UnitTestBase::keyDown):
(EWK2UnitTest::EWK2UnitTestBase::keyUp):
* UIProcess/API/efl/tests/UnitTestUtils/EWK2UnitTestBase.h:
* UIProcess/API/efl/tests/test_ewk2_view.cpp:
(TEST_F):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@174208 268f45cc-cd09-0410-ab3c-d52691b4dbfc