⚠ Archived content — this site is no longer maintained.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

Timeline



Oct 9, 2013:

9:24 PM Changeset in webkit [157209] by fpizlo@apple.com
  • 27 edits
    4 adds in trunk/Source/JavaScriptCore

FTL should be able to do simple OSR exits using llvm.webkit.stackmap
​https://bugs.webkit.org/show_bug.cgi?id=122538

Reviewed by Oliver Hunt.

This gives the FTL the ability to OSR exit using the llvm.webkit.stackmap intrinsic.

  • The FTL compiles all OSR exit calls as calls to llvm.webkit.stackmap with a unique ID, passing a requested size that is big enough for own jump replacement.


  • After LLVM compilation, we parse the new LLVM stackmap section.


  • For all llvm.webkit.stackmaps that we used for OSR exits, we do a jumpReplacement, which targets exit thunks that we generate.


  • If an exit thunk fires, it causes JSC to compile an exit off-ramp that uses a combination of the JSC-internal OSR exit accounting (FTL::ExitValue and friends) and LLVM stackmap's accounting of where data actually ended up (register, indirect, constant) to reconstruct bytecode state.


This still has shortcomings; for example it cannot handle XMM or YMM registers. Handling
YMM registers will require adding some basic YMM support to our assemblers - really we
just need the ability to move a YMM's value into a GPR.

This patch preserves all of the old, intrinsic-less, FTL OSR exit support. Hence it
manages to pass all existing FTL tests even despite its incompleteness. I think that's
the right way to go since this is already a big patch, and anyway it would be great to
keep the intrinsic-less FTL OSR exit support so long as the LLVM side of this hasn't
landed.

  • JavaScriptCore.xcodeproj/project.pbxproj:
  • assembler/AbstractMacroAssembler.h:

(JSC::AbstractMacroAssembler::firstRegister):
(JSC::AbstractMacroAssembler::lastRegister):

  • assembler/MacroAssembler.h:

(JSC::MacroAssembler::isStackRelated):
(JSC::MacroAssembler::firstRealRegister):
(JSC::MacroAssembler::nextRegister):
(JSC::MacroAssembler::secondRealRegister):

  • assembler/MacroAssemblerX86Common.h:
  • assembler/X86Assembler.h:

(JSC::X86Assembler::firstRegister):
(JSC::X86Assembler::lastRegister):

  • dfg/DFGPlan.cpp:

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

  • ftl/FTLCArgumentGetter.cpp:

(JSC::FTL::CArgumentGetter::loadNextAndBox):

  • ftl/FTLCArgumentGetter.h:

(JSC::FTL::CArgumentGetter::loadNextDoubleIntoGPR):

  • ftl/FTLCompile.cpp:

(JSC::FTL::mmAllocateCodeSection):
(JSC::FTL::mmAllocateDataSection):
(JSC::FTL::dumpDataSection):
(JSC::FTL::fixFunctionBasedOnStackMaps):
(JSC::FTL::compile):

  • ftl/FTLExitThunkGenerator.cpp:

(JSC::FTL::ExitThunkGenerator::emitThunk):
(JSC::FTL::ExitThunkGenerator::emitThunks):

  • ftl/FTLExitThunkGenerator.h:
  • ftl/FTLExitValue.h:

(JSC::FTL::ExitValue::isInJSStackSomehow):
(JSC::FTL::ExitValue::valueFormat):

  • ftl/FTLFail.cpp:

(JSC::FTL::fail):

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

(JSC::FTL::generateExitThunks):
(JSC::FTL::LowerDFGToLLVM::LowerDFGToLLVM):
(JSC::FTL::LowerDFGToLLVM::appendOSRExit):
(JSC::FTL::LowerDFGToLLVM::emitOSRExitCall):
(JSC::FTL::LowerDFGToLLVM::linkOSRExitsAndCompleteInitializationBlocks):

  • ftl/FTLOSRExit.h:
  • ftl/FTLOSRExitCompilationInfo.h:

(JSC::FTL::OSRExitCompilationInfo::OSRExitCompilationInfo):

  • ftl/FTLOSRExitCompiler.cpp:

(JSC::FTL::compileStubWithOSRExitStackmap):
(JSC::FTL::compileStubWithoutOSRExitStackmap):
(JSC::FTL::compileFTLOSRExit):

  • ftl/FTLSaveRestore.cpp: Added.

(JSC::FTL::bytesForGPRs):
(JSC::FTL::requiredScratchMemorySizeInBytes):
(JSC::FTL::offsetOfGPR):
(JSC::FTL::saveAllRegisters):
(JSC::FTL::restoreAllRegisters):

  • ftl/FTLSaveRestore.h: Added.
  • ftl/FTLStackMaps.cpp: Added.

(JSC::FTL::readObject):
(JSC::FTL::StackMaps::Constant::parse):
(JSC::FTL::StackMaps::Constant::dump):
(JSC::FTL::StackMaps::Location::parse):
(JSC::FTL::StackMaps::Location::dump):
(JSC::FTL::StackMaps::Location::involvesGPR):
(JSC::FTL::StackMaps::Location::isGPR):
(JSC::FTL::StackMaps::Location::gpr):
(JSC::FTL::StackMaps::Location::restoreInto):
(JSC::FTL::StackMaps::Record::parse):
(JSC::FTL::StackMaps::Record::dump):
(JSC::FTL::StackMaps::parse):
(JSC::FTL::StackMaps::dump):
(JSC::FTL::StackMaps::dumpMultiline):
(JSC::FTL::StackMaps::getRecordMap):
(WTF::printInternal):

  • ftl/FTLStackMaps.h: Added.
  • ftl/FTLState.h:
  • ftl/FTLThunks.cpp:

(JSC::FTL::osrExitGenerationThunkGenerator):

  • ftl/FTLValueFormat.cpp:

(JSC::FTL::reboxAccordingToFormat):

  • ftl/FTLValueFormat.h:
  • runtime/DataView.cpp:

(JSC::DataView::create):

  • runtime/DataView.h:

(JSC::DataView::read):

  • runtime/Options.h:
9:24 PM Changeset in webkit [157208] by fpizlo@apple.com
  • 4 edits in trunk/Source/WTF

FTL should be able to do simple OSR exits using llvm.webkit.stackmap
​https://bugs.webkit.org/show_bug.cgi?id=122538

Reviewed by Oliver Hunt.

  • wtf/PrintStream.cpp:

(WTF::printInternal):

  • wtf/PrintStream.h:
  • wtf/RefCountedArray.h:

(WTF::RefCountedArray::byteSize):

8:26 PM Changeset in webkit [157207] by rniwa@webkit.org
  • 2 edits in trunk/Source/WebCore

Another build fix after r157204.

  • dom/EventDispatcher.cpp:

(WebCore::EventDispatcher::dispatch):

8:19 PM Changeset in webkit [157206] by rniwa@webkit.org
  • 2 edits in trunk/Source/WebCore

Build fix after r157204.

  • dom/EventDispatcher.cpp:

(WebCore::EventDispatcher::dispatch):

7:46 PM Changeset in webkit [157205] by Darin Adler
  • 22 edits in trunk/Source/WebCore

Use unique_ptr instead of deleteAllValues in XPath
​https://bugs.webkit.org/show_bug.cgi?id=121082

Reviewed by Anders Carlsson.

  • dom/Attr.h: Made the namespaceURI function public so it can be called by

code in XPath. Since the class is FINAL, calls won't do virtual dispatch.

  • xml/XPathExpression.cpp:

(WebCore::XPathExpression::XPathExpression): Added. Use std::move.
(WebCore::XPathExpression::createExpression): Use the new Parser::parseStatement
and more std::move.
(WebCore::XPathExpression::~XPathExpression): Removed now-unneeded delete.

  • xml/XPathExpression.h: Removed unused create function. Use std::unique_ptr.
  • xml/XPathExpressionNode.cpp:

(WebCore::XPath::Expression::evaluationContext): Use NeverDestroyed instead of
DEFINE_STATIC_LOCAL.
(WebCore::XPath::Expression::setSubexpressions): Added. Used to set all the
subexpressions at once rather than adding one at a time.

  • xml/XPathExpressionNode.h: Use plain unsigned instead of unsigned long.

Eliminated the ParseNode class, used only by the parser to delete objects,
which we now do with %destructor. Made more functions protected. Fixed the
capitalization of the word "subexpression". Made m_subexpressions be a Vector
of std::unique_ptr.

  • xml/XPathFunctions.cpp: Marked all the classes FINAL and all their virtual

functions OVERRIDE. Reduced some of the boilerplate.
(WebCore::XPath::Function::setArguments): Passed in the name so we don't have
to store the function names in all the function objects.
(WebCore::XPath::FunId::evaluate): Use std::move instead of Value::adopt.
(WebCore::XPath::FunLocalName::evaluate): Use emptyString instead of "".
(WebCore::XPath::FunNamespaceURI::evaluate): Ditto.
(WebCore::XPath::FunName::evaluate): Ditto.
(WebCore::XPath::FunSubstringBefore::evaluate): Ditto.
(WebCore::XPath::FunSubstringAfter::evaluate): Ditto.
(WebCore::XPath::FunSubstring::evaluate): Ditto.
(WebCore::XPath::FunLang::evaluate): Use Value(false) so we can make the
constructor that takes a boolean explicit.
(WebCore::XPath::FunFalse::evaluate): Ditto.
(WebCore::XPath::populateFunctionMap): Changed idiom slightly to match other
maps we set up.
(WebCore::XPath::Function::create): Refactored createFunction into three
member functions of this name.

  • xml/XPathFunctions.h: Made all the Function members private or protected

except for Function::create.

  • xml/XPathGrammar.y: Changed the parser to use a reference instead of

a pointer, and to pass the reference through to yylex as well. Break up
the union into pieces and add %destructor as needed to make sure everything
gets deallocated if parsing fails. Added a new top level rule "Top" so that
we would not try to set the parse result multiple times during parsing.
Call setParseResult instead of setting m_topExpr directly. Use std::unique_ptr
to adopt pointers from the union. Removed all the register/unregister calls
that used to be used to manage storage. Also changed the four different node
types to be separate tokens instead of using string matching at this level
for them.

  • xml/XPathNodeSet.cpp:

(WebCore::XPath::NodeSet::sort): Removed const_cast since m_isSorted and
m_nodes are now both marked mutable. Also set m_isSorted after sorting; this
was an oversight that hurt performance before.
(WebCore::XPath::NodeSet::traversalSort): Ditto.
(WebCore::XPath::NodeSet::firstNode): Use nullptr.
(WebCore::XPath::NodeSet::anyNode): Ditto.

  • xml/XPathNodeSet.h: Removed unneeded extra includes. Removed the

WTF_MAKE_FAST_ALLOCATED for NodeSet since we never allocate these on the heap.
Added a constructor that takes a single node. Deleted the swap function, since
we now move instead of swap. Deleted the append function that takes a raw Node*,
since calling the PassRefPtr<Node> overload does the same thing. Removed the
unused reverse function. Marked both m_isSorted and m_nodes mutable so the
sort function can be const.

  • xml/XPathParser.cpp: Moved the Token struct in here instead of the header.

(WebCore::XPath::populateAxisNamesMap):Renamed to match our normal naming scheme,
and changed to use add instead of set, use WTF_ARRAY_LENGTH, and not use a typedef.
(WebCore::XPath::parseAxisName): Related style changes, including renaming to
better reflect the way this works with a boolean.
(WebCore::XPath::Parser::nextTokenInternal): Updated to call parseAxisName,
and to produce four separate tokens for the four node type functions.
(WebCore::XPath::Parser::nextToken): Renamed local variable.
(WebCore::XPath::Parser::Parser): Made this a real constructor that takes arguments
and initializes all the fields. Deleted the unneeded reset function.
(WebCore::XPath::Parser::lex): Changed to take an argument of an appropriate type
instead of casting from void*. Also changed the string code to leak a StringImpl,
which is now what the grammar code expects.
(WebCore::XPath::Parser::expandQualifiedName): Renamed from expandQName. Changed
to set m_sawNamespaceError instead of relying on code in the grammar to do it.
(WebCore::XPath::Parser::parseStatement): Removed most of the code in this function.
Changed to a much simpler model. Also made this a static member function so it
takes care of creating the parser itself and does not need to call reset. Also
changed return value to be a std::unique_ptr to make ownership more clear.

  • xml/XPathParser.h: Added a declaration of YYSTYPE. Removed unneeded forward

declarations and moved Token into the cpp file. Deleted most public functions,
leaving only parseStatement, the entry point, and the three functions needed by
the grammar, lex, expandQualifiedName, and setParseResult.

  • xml/XPathPath.cpp:

(WebCore::XPath::Filter::Filter): Move in the arguments instead of copying them.
(WebCore::XPath::Filter::evaluate): Updated for name and type changes.
(WebCore::XPath::LocationPath::LocationPath): Ditto.
(WebCore::XPath::LocationPath::evaluate): Ditto. Also use std::move instead of
Value::adopt and instead of swap.
(WebCore::XPath::LocationPath::appendStep): Changed to take ownership of a
unique_ptr.
(WebCore::XPath::LocationPath::prependStep): Ditto. Renamed from insertFirstStep.
(WebCore::XPath::Path::Path): Move in the arguments.

  • xml/XPathPath.h: Removed unneeded includes. Changed arugument types to use

std::unique_ptr to pass ownership in. Added override to all virtual functions.
Changed data members to use std::unique_ptr.

  • xml/XPathPredicate.cpp:

(WebCore::XPath::StringExpression::StringExpression): Use move.
(WebCore::XPath::Negative::Negative): Added.
(WebCore::XPath::Negative::evaluate): Updated for name changes.
(WebCore::XPath::NumericOp::NumericOp): Use move.
(WebCore::XPath::NumericOp::evaluate): Tweak formatting.
(WebCore::XPath::EqTestOp::EqTestOp): Use move.
(WebCore::XPath::EqTestOp::evaluate): Updated for name changes.
(WebCore::XPath::LogicalOp::LogicalOp): Use move.
(WebCore::XPath::LogicalOp::shortCircuitOn): Made shorter.
(WebCore::XPath::LogicalOp::evaluate): Updated for name changes.
(WebCore::XPath::Union::Union): Added.
(WebCore::XPath::Union::evaluate): Updated for name changes.
(WebCore::XPath::evaluatePredicate): Updated for name changes, to use
ASCIILiteral, and to be a free function that takes an expression.
(WebCore::XPath::predicateIsContextPositionSensitive): Added.
Replaces the Predicate::isContextPositionSensitive function.

  • xml/XPathPredicate.h: Made all the classes FINAL and added a lot of OVERRIDE.

Added a constructor for Negative and Union. Got rid of the Predicate class and
instead provided two functions that operate on an Expression.

  • xml/XPathStep.cpp:

(WebCore::XPath::Step::Step): Use move instea dof copying.
(WebCore::XPath::Step::~Step): Removed calls to deleteAllValues.
(WebCore::XPath::Step::optimize): Use move instead of swap and copy. Also
operate directly on the data members of the node test instead of using functions
that just return references to those data members.
(WebCore::XPath::optimizeStepPair): Use references instead of pointers, move
instead of swap, and early return instead of nested if statements.
(WebCore::XPath::Step::predicatesAreContextListInsensitive): Use references.
(WebCore::XPath::Step::evaluate): Use references instead of pointers and move
instead of swap.
(WebCore::XPath::nodeMatchesBasicTest): Use references instead of pointers and
more specific types when possible.
(WebCore::XPath::nodeMatches): Ditto.
(WebCore::XPath::Step::nodesInAxis): Use references instead of pointers. Added
braces to match style. Use words instead of letters for local variable names.

  • xml/XPathStep.h: Make almost everything in NodeTest private since callers

just set these up and Step is what actually uses them. Changed the predicate
vectors to be vectors of Predicate instead of Predicate*, since a Predicate
is just a wrapper around a single std::unique_ptr<Expression>. Changed to use
move instead of copy when making a Step and to use references instead of pointers.

  • xml/XPathValue.cpp: Use std::numeric_limits explicitly instead of using using.

Got rid of Value::adopt.
(WebCore::XPath::Value::toNodeSet): Use NeverDestroyed instead of DEFINE_STATE_LOCAL.
Updated for name changes.
(WebCore::XPath::Value::modifiableNodeSet): Ditto.
(WebCore::XPath::Value::toBoolean): Ditto.
(WebCore::XPath::Value::toNumber): Ditto.
(WebCore::XPath::Value::toString): Ditto. Use ASCIILiteral.

  • xml/XPathValue.h: Moved ValueData class inside the Value class and renamed

it Value::Data. Removed fancy trick that avoids turning pointers into bool, and
just take the risk. Made many, but not all, of the Value constructors explicit
since we normally are explicit at call sites. Removed unneeded unsigned long
overload of the constructor. Changed the NodeSet version of the constructor to
use std::move and use that instead of AdoptTag.

6:59 PM Changeset in webkit [157204] by rniwa@webkit.org
  • 2 edits in trunk/Source/WebCore

Build fix. Add back the missing call to EventRetargeter::adjustForTouchEvent.

  • dom/EventDispatcher.cpp:

(WebCore::EventDispatcher::dispatch):

6:44 PM Changeset in webkit [157203] by rniwa@webkit.org
  • 20 edits
    2 deletes in trunk/Source/WebCore

EventDispatchMediator is goner
​https://bugs.webkit.org/show_bug.cgi?id=122581

Reviewed by Antti Koivisto.

Killed the death.

  • CMakeLists.txt:
  • GNUmakefile.list.am:
  • WebCore.vcxproj/WebCore.vcxproj:
  • WebCore.vcxproj/WebCore.vcxproj.filters:
  • WebCore.xcodeproj/project.pbxproj:
  • dom/DOMAllInOne.cpp:
  • dom/Element.cpp:

(WebCore::Element::dispatchWheelEvent):
(WebCore::Element::dispatchKeyEvent):
(WebCore::Element::dispatchFocusEvent):
(WebCore::Element::dispatchBlurEvent):

  • dom/EventDispatchMediator.cpp: Removed.
  • dom/EventDispatchMediator.h: Removed.
  • dom/EventDispatcher.cpp:

(WebCore::EventDispatcher::dispatchEvent):
(WebCore::EventDispatcher::dispatchScopedEvent):

  • dom/EventDispatcher.h:
  • dom/KeyboardEvent.h:
  • dom/MouseEvent.h:
  • dom/Node.cpp:

(WebCore::Node::dispatchScopedEvent):
(WebCore::Node::dispatchEvent):
(WebCore::Node::dispatchGestureEvent):
(WebCore::Node::dispatchTouchEvent):
(WebCore::Node::dispatchUIRequestEvent):

  • dom/Node.h:
  • dom/ScopedEventQueue.cpp:

(WebCore::ScopedEventQueue::~ScopedEventQueue):
(WebCore::ScopedEventQueue::enqueueEvent):
(WebCore::ScopedEventQueue::dispatchAllEvents):
(WebCore::ScopedEventQueue::dispatchEvent):

  • dom/ScopedEventQueue.h:
  • dom/TouchEvent.cpp:
  • dom/TouchEvent.h:
  • dom/UIEvent.h:
  • dom/WheelEvent.h:
6:37 PM Changeset in webkit [157202] by akling@apple.com
  • 2 edits in trunk/Source/WebCore

Kill some RenderObject::firstChildSlow/lastChildSlow usage.
<​https://webkit.org/b/122582>

Reviewed by Antti Koivisto.

Use tighter typing in some random places to flush out unnecessary use
of virtual first/last child getters on RenderObject.

There are more of these opportunities, but I left them untouched to
avoid snowballing this patch.

6:23 PM Changeset in webkit [157201] by commit-queue@webkit.org
  • 2 edits in trunk/Tools

webkit-patch failure-reason fails with UnicodeDecodeError when name has non-ascii characters
​https://bugs.webkit.org/show_bug.cgi?id=122552

Patch by Gustavo Noronha Silva <​gustavo.noronha@collabora.com> on 2013-10-09
Reviewed by Darin Adler.

  • Scripts/webkitpy/common/checkout/commitinfo.py:

(CommitInfo.blame_string): ensure objects are made into unicode strings, not regular ones that
will cause UnicodeDecodeError for non-ascii characters.

6:23 PM Changeset in webkit [157200] by Simon Fraser
  • 2 edits in trunk/Source/WebCore

Fix some bad paths in the Xcode project
​https://bugs.webkit.org/show_bug.cgi?id=122583

Reviewed by Mark Rowe.

Fix paths for some storage quota-related files, and remove
reference to mediaControlsQuickTime.css which was removed in r156546.

  • WebCore.xcodeproj/project.pbxproj:
6:17 PM Changeset in webkit [157199] by Alexandru Chiculita
  • 7 edits
    16 adds in trunk

Web Inspector: [CSS Regions] Display the correct fragment boxes for content inside flow threads
​https://bugs.webkit.org/show_bug.cgi?id=122291

Reviewed by Timothy Hatcher.

Source/WebCore:

Tests: inspector/elements/highlight-regions-flow-horiz-bt.html

inspector/elements/highlight-regions-flow-transform-horiz-bt.html
inspector/elements/highlight-regions-flow-transform-vert-lr.html
inspector/elements/highlight-regions-flow-transform-vert-rl.html
inspector/elements/highlight-regions-flow-transform.html
inspector/elements/highlight-regions-flow-vert-lr.html
inspector/elements/highlight-regions-flow-vert-rl.html
inspector/elements/highlight-regions-flow.html

Nodes inside a RenderFlowThread can be split across multiple regions. This patch adds the required
code to compute "fragments" for each region. Each fragment will display its own label.

There are a couple of changes:

  1. The node highlight object now has a "scroll" property to hold the main view's scroll offset.
  2. There's a "fragment" array to get a list of quads for each renderer fragment. Renderers outside

RenderFlowThreads will just have one fragment.

  1. ElementInfo now has two more properties: "regionFlowInfo" (renamed from former flowInfo) and "contentFlowInfo".

If the renderer is a RenderRegion it will have a regionFlowInfo. If the renderer is inside a RenderFlowThread
then it will have a "contentFlowInfo".

  • inspector/InspectorOverlay.cpp:

(WebCore::InspectorOverlay::getHighlight): It should now take a "region" parameter to customize the output for a specific
containing region.
(WebCore::buildObjectForHighlight):
(WebCore::buildObjectForCSSRegionContentClip): We need to know how to clip the fragments inside regions.
(WebCore::buildObjectForRendererFragments): Extracted code from buildObjectForHighlight and added
the new regions lookup code.
(WebCore::buildObjectForElementInfo): Extracted from buildObjectForHighlight.
(WebCore::InspectorOverlay::buildObjectForHighlightedNode):
(WebCore::InspectorOverlay::drawQuadHighlight):

  • inspector/InspectorOverlayPage.css: There can be multiple labels on screen at once, so I've

created the #element-title-template element that we clone for each label. .element-title-container is now
holding all the .element-title elements, so that we can remove them all at once.
(.element-title-container):
(.element-title):
(#element-title-template):
(.tag-name):
(.node-id):
(.class-name):

  • inspector/InspectorOverlayPage.html:
  • inspector/InspectorOverlayPage.js:

(reset):
(_toggleOptionalValue):
(_createElementTitle): Creates an element populated with data read from elementInfo.
(_drawElementTitle): Added code to create more labels for multiple fragments.
(_drawRegionsHighlight):
(_drawFragmentHighlight):
(drawNodeHighlight): Consolidated the scroll property into a canvas.transform.

LayoutTests:

Added tests to check the result of the InspectorOverlay when it inspects nodes inside RenderFlowThreads.

  • http/tests/inspector/elements-test.js:
  • inspector/elements/highlight-regions-flow-expected.txt: Added.
  • inspector/elements/highlight-regions-flow-horiz-bt-expected.txt: Added.
  • inspector/elements/highlight-regions-flow-horiz-bt.html: Added.
  • inspector/elements/highlight-regions-flow-transform-expected.txt: Added.
  • inspector/elements/highlight-regions-flow-transform-horiz-bt-expected.txt: Added.
  • inspector/elements/highlight-regions-flow-transform-horiz-bt.html: Added.
  • inspector/elements/highlight-regions-flow-transform-vert-lr-expected.txt: Added.
  • inspector/elements/highlight-regions-flow-transform-vert-lr.html: Added.
  • inspector/elements/highlight-regions-flow-transform-vert-rl-expected.txt: Added.
  • inspector/elements/highlight-regions-flow-transform-vert-rl.html: Added.
  • inspector/elements/highlight-regions-flow-transform.html: Added.
  • inspector/elements/highlight-regions-flow-vert-lr-expected.txt: Added.
  • inspector/elements/highlight-regions-flow-vert-lr.html: Added.
  • inspector/elements/highlight-regions-flow-vert-rl-expected.txt: Added.
  • inspector/elements/highlight-regions-flow-vert-rl.html: Added.
  • inspector/elements/highlight-regions-flow.html: Added.
5:53 PM Changeset in webkit [157198] by Simon Fraser
  • 2 edits in trunk/Source/WebCore

Layout assertion failure in WebCore::RenderView::paint
​https://bugs.webkit.org/show_bug.cgi?id=121925

Reviewed by Tim Horton.

This fixes the assertion seen when running WK2 API tests.

If RenderLayerCompositor::flushPendingLayerChanges() was called before
the root layer was attached, it would set m_shouldFlushOnReattach
to do the flush later, after attaching (bypassing the runloop observer
that updated layout if necessary).

This could cause us to get to a CA commit without having updated
layout, hence the assertion.

Fix by having attachRootLayer() schedule a layer flush, rather than doing
one right away.

  • rendering/RenderLayerCompositor.cpp:

(WebCore::RenderLayerCompositor::attachRootLayer):

5:35 PM Changeset in webkit [157197] by Bem Jones-Bey
  • 9 edits in trunk/Source/WebCore

Move float logical location/dimension methods to RenderBlockFlow
​https://bugs.webkit.org/show_bug.cgi?id=122282

Reviewed by David Hyatt.

I recently moved the logical location and dimension methods from
RenderBlock to FloatingObject. This made it so that the methods needed
to take in the horizontal writing mode of the FloatingObject's
container in order to do their thing. When looking into how to fix
this, I came to understand why RenderBlock has methods to determine
the logical location and dimensions of its children, when the children
themselves have those methods: the methods on RenderBlock are for
determining the logical value in the writing mode of that RenderBlock,
and the method on the child is for determining the logical value in
the writing mode of the child.

Since the floating object list has moved from RenderBlock to
RenderBlockFlow, this change moves the logical dimension methods from
FloatingObject to RenderBlockFlow.

No new tests, no new behavior.

  • rendering/FloatingObjects.cpp:

(WebCore::ComputeFloatOffsetAdapter::ComputeFloatOffsetAdapter):
(WebCore::FloatingObjects::FloatingObjects):
(WebCore::::updateOffsetIfNeeded):
(WebCore::::collectIfNeeded):
(WebCore::::getHeightRemaining):

  • rendering/FloatingObjects.h:
  • rendering/LineWidth.cpp:

(WebCore::LineWidth::shrinkAvailableWidthForNewFloatIfNeeded):

  • rendering/RenderBlockFlow.cpp:

(WebCore::RenderBlockFlow::clearFloats):
(WebCore::RenderBlockFlow::repaintOverhangingFloats):
(WebCore::RenderBlockFlow::insertFloatingObject):
(WebCore::RenderBlockFlow::removeFloatingObject):
(WebCore::RenderBlockFlow::removeFloatingObjectsBelow):
(WebCore::RenderBlockFlow::computeLogicalLocationForFloat):
(WebCore::RenderBlockFlow::positionNewFloats):
(WebCore::RenderBlockFlow::nextFloatLogicalBottomBelow):
(WebCore::RenderBlockFlow::lowestFloatLogicalBottom):
(WebCore::RenderBlockFlow::addOverhangingFloats):
(WebCore::RenderBlockFlow::hasOverhangingFloat):
(WebCore::RenderBlockFlow::addIntrudingFloats):

  • rendering/RenderBlockFlow.h:

(WebCore::RenderBlockFlow::logicalTopForFloat):
(WebCore::RenderBlockFlow::logicalBottomForFloat):
(WebCore::RenderBlockFlow::logicalLeftForFloat):
(WebCore::RenderBlockFlow::logicalRightForFloat):
(WebCore::RenderBlockFlow::logicalWidthForFloat):
(WebCore::RenderBlockFlow::logicalHeightForFloat):
(WebCore::RenderBlockFlow::logicalSizeForFloat):
(WebCore::RenderBlockFlow::pixelSnappedLogicalTopForFloat):
(WebCore::RenderBlockFlow::pixelSnappedLogicalBottomForFloat):
(WebCore::RenderBlockFlow::pixelSnappedLogicalLeftForFloat):
(WebCore::RenderBlockFlow::pixelSnappedLogicalRightForFloat):
(WebCore::RenderBlockFlow::setLogicalTopForFloat):
(WebCore::RenderBlockFlow::setLogicalLeftForFloat):
(WebCore::RenderBlockFlow::setLogicalHeightForFloat):
(WebCore::RenderBlockFlow::setLogicalWidthForFloat):

  • rendering/RenderBlockLineLayout.cpp:

(WebCore::RenderBlockFlow::adjustLogicalLineTopAndLogicalHeightIfNeeded):
(WebCore::RenderBlockFlow::checkPaginationAndFloatsAtEndLine):
(WebCore::updateSegmentsForShapes):
(WebCore::LineBreaker::nextSegmentBreak):
(WebCore::RenderBlockFlow::positionNewFloatOnLine):

  • rendering/shapes/ShapeOutsideInfo.cpp:

(WebCore::ShapeOutsideInfo::updateDeltasForContainingBlockLine):

  • rendering/shapes/ShapeOutsideInfo.h:
5:29 PM Changeset in webkit [157196] by rniwa@webkit.org
  • 5 edits in trunk/Source/WebCore

Remove MouseEventDispatchMediator
​https://bugs.webkit.org/show_bug.cgi?id=122580

Reviewed by Antti Koivisto.

Removed. Will remove EventDispatchMediator next.

  • dom/Element.cpp:

(WebCore::Element::dispatchMouseEvent):

  • dom/MouseEvent.cpp:
  • dom/MouseEvent.h:
  • dom/Node.cpp:

(WebCore::Node::dispatchEvent):
(WebCore::Node::dispatchTouchEvent):

4:49 PM Changeset in webkit [157195] by rniwa@webkit.org
  • 11 edits in trunk/Source/WebCore

Remove all subclasses of EventDispatchMediator except MouseEventDispatchMediator
​https://bugs.webkit.org/show_bug.cgi?id=122579

Reviewed by Antti Koivisto.

Removed UIRequestEventDispatchMediator, GestureEventDispatchMediator, KeyboardEventDispatchMediator,
and WheelEventDispatchMediator. Moved the code in the respective mediateAndDispatchEvent functions
to their respective dispatch*Event functions in Node.cpp and Element.cpp.

  • Modules/indieui/UIRequestEvent.cpp:

(WebCore::UIRequestEvent::eventInterface):

  • Modules/indieui/UIRequestEvent.h:
  • dom/Element.cpp:

(WebCore::deltaMode):
(WebCore::Element::dispatchWheelEvent):
(WebCore::Element::dispatchKeyEvent):

  • dom/GestureEvent.cpp:
  • dom/GestureEvent.h:
  • dom/KeyboardEvent.cpp:
  • dom/KeyboardEvent.h:
  • dom/Node.cpp:

(WebCore::Node::dispatchGestureEvent):
(WebCore::Node::dispatchUIRequestEvent):

  • dom/WheelEvent.cpp:
  • dom/WheelEvent.h:
4:03 PM Changeset in webkit [157194] by ap@apple.com
  • 2 edits in trunk/Tools

Would like a link to dashboard from build.webkit.org main page
​https://bugs.webkit.org/show_bug.cgi?id=122577

Reviewed by Tim Hatcher.

  • BuildSlaveSupport/build.webkit.org-config/templates/root.html: Added.
3:57 PM Changeset in webkit [157193] by fpizlo@apple.com
  • 2 edits in trunk/Source/JavaScriptCore

Minor clean-ups in the JSC Xcode project.

Rubber stamped by Mark Rowe.

  • When we copy the jsc binary into the framework, $(BUILT_PRODUCTS_DIR)/JavaScriptCore.framework/Resources/jsc is the *output* file not the input file. The input file is $(BUILT_PRODUCTS_DIR)/jsc.


  • Correct capitalization of "JavaScriptcore.framework" in a comment in a shell script in the project.
3:54 PM Changeset in webkit [157192] by Bem Jones-Bey
  • 7 edits in trunk

[CSS Shapes] Lines that don't intersect shape-outside should ignore both left and right margins
​https://bugs.webkit.org/show_bug.cgi?id=122570

Reviewed by Alexandru Chiculita.

Source/WebCore:

Due to an oversight, the left margin on left floats and the right
margin on right floats was having an effect on lines that did not
intersect the shape (but did intersect the margin box of the float).
By the spec, lines that do not intersect the shape should act as if
the float isn't there, so those margins should have no effect on
those lines. This patch fixes this.

No new tests, it is covered by the existing tests now that they have
been fixed to have the proper behavior.

  • rendering/shapes/ShapeOutsideInfo.cpp:

(WebCore::ShapeOutsideInfo::updateDeltasForContainingBlockLine):

LayoutTests:

Update tests to work with the proper behavior for margins. Yes, the
tests were not correct before.

  • csswg/contributors/adobe/submitted/shapes/shape-outside/resources/w3c-import.log:
  • csswg/contributors/adobe/submitted/shapes/shape-outside/shape-outside-floats-margin-000.html:
  • csswg/contributors/adobe/submitted/shapes/shape-outside/shape-outside-floats-margin-007.html:
  • csswg/contributors/adobe/submitted/shapes/shape-outside/shape-outside-floats-margin-009.html:
  • csswg/contributors/adobe/submitted/shapes/shape-outside/w3c-import.log:
  • fast/shapes/shape-outside-floats/shape-outside-floats-different-writing-modes.html:
3:39 PM Changeset in webkit [157191] by Brent Fulgham
  • 2 edits in trunk/Tools

[Win] Check for MathML fonts and warn user if they are missing.
​https://bugs.webkit.org/show_bug.cgi?id=122568

Reviewed by Tim Horton.

  • Scripts/webkitdirs.pm:

(checkRequiredSystemConfig): Use common gperf/bison/flex tests, rather than
Windows-specific code I added previously.
(fontExists): Added
(checkInstalledTools): Add check for MathML fonts.

3:32 PM Changeset in webkit [157190] by ap@apple.com
  • 3 edits in trunk/Tools

​http://build.webkit.org/dashboard/ doesn't link to webkitpy failures
​https://bugs.webkit.org/show_bug.cgi?id=122471

Reviewed by Tim Hatcher.

Added links to webkitpy, webkitperl and bindings-tests results.

  • BuildSlaveSupport/build.webkit.org-config/public_html/dashboard/Scripts/Buildbot.js:

(Buildbot.prototype.apiTestResultsURLForIteration):
(Buildbot.prototype.webkitpyTestResultsURLForIteration):
(Buildbot.prototype.webkitperlTestResultsURLForIteration):
(Buildbot.prototype.bindingsTestResultsURLForIteration):

  • BuildSlaveSupport/build.webkit.org-config/public_html/dashboard/Scripts/BuildbotTesterQueueView.js:

(BuildbotTesterQueueView.prototype.update.appendBuilderQueueStatus):

3:03 PM Changeset in webkit [157189] by ap@apple.com
  • 3 edits in trunk/Tools

Move some functions from WebKitBuildbot to Buildbot class.

Suggested by Tim Hatcher.

  • BuildSlaveSupport/build.webkit.org-config/public_html/dashboard/Scripts/Buildbot.js:

(Buildbot.prototype.javascriptTestResultsURLForIteration):
(Buildbot.prototype.apiTestResultsURLForIteration):

  • BuildSlaveSupport/build.webkit.org-config/public_html/dashboard/Scripts/WebKitBuildbot.js:

(WebKitBuildbot.prototype.layoutTestResultsURLForIteration):

2:58 PM Changeset in webkit [157188] by ap@apple.com
  • 3 edits in trunk/Tools

Add API tests to the Dashbord page.

​https://bugs.webkit.org/show_bug.cgi?id=122283

Patch by Timothy Hatcher <​timothy@apple.com> on 2013-10-09
Reviewed by Timothy Hatcher.

  • BuildSlaveSupport/build.webkit.org-config/public_html/dashboard/Scripts/BuildbotTesterQueueView.js:

(BuildbotTesterQueueView.prototype.update.appendBuilderQueueStatus):

  • BuildSlaveSupport/build.webkit.org-config/public_html/dashboard/Scripts/WebKitBuildbot.js:

(WebKitBuildbot.prototype.apiTestResultsURLForIteration):

2:56 PM Changeset in webkit [157187] by timothy@apple.com
  • 3 edits in trunk/Tools

Add API tests to the Dashbord page.

​https://bugs.webkit.org/show_bug.cgi?id=122283

Reviewed by Timothy Hatcher.

  • BuildSlaveSupport/build.webkit.org-config/public_html/dashboard/Scripts/BuildbotTesterQueueView.js:

(BuildbotTesterQueueView.prototype.update.appendBuilderQueueStatus):

  • BuildSlaveSupport/build.webkit.org-config/public_html/dashboard/Scripts/WebKitBuildbot.js:

(WebKitBuildbot.prototype.apiTestResultsURLForIteration):

2:45 PM Changeset in webkit [157186] by Bem Jones-Bey
  • 5 edits
    8 adds in trunk

[CSS Shapes] Clip shape-outside to the bottom of the margin box
​https://bugs.webkit.org/show_bug.cgi?id=122562

Reviewed by Alexandru Chiculita.

Source/WebCore:

The spec says that a shape-outside shape must be clipped to the
margin box. This implements clipping the bottom of the shape to the
bottom of the margin box.

Test: csswg/contributors/adobe/submitted/shapes/shape-outside/shape-outside-floats-clipped-000.html

csswg/contributors/adobe/submitted/shapes/shape-outside/shape-outside-floats-clipped-001.html
csswg/contributors/adobe/submitted/shapes/shape-outside/shape-outside-floats-clipped-002.html
csswg/contributors/adobe/submitted/shapes/shape-outside/shape-outside-floats-clipped-003.html

  • rendering/RenderBlockFlow.cpp:

(WebCore::RenderBlockFlow::nextFloatLogicalBottomBelow):

LayoutTests:

Test that the shape is properly clipped to the bottom of the float.

  • csswg/contributors/adobe/submitted/shapes/shape-outside/resources/w3c-import.log:
  • csswg/contributors/adobe/submitted/shapes/shape-outside/shape-outside-floats-clipped-000-expected.html: Added.
  • csswg/contributors/adobe/submitted/shapes/shape-outside/shape-outside-floats-clipped-000.html: Added.
  • csswg/contributors/adobe/submitted/shapes/shape-outside/shape-outside-floats-clipped-001-expected.html: Added.
  • csswg/contributors/adobe/submitted/shapes/shape-outside/shape-outside-floats-clipped-001.html: Added.
  • csswg/contributors/adobe/submitted/shapes/shape-outside/shape-outside-floats-clipped-002-expected.html: Added.
  • csswg/contributors/adobe/submitted/shapes/shape-outside/shape-outside-floats-clipped-002.html: Added.
  • csswg/contributors/adobe/submitted/shapes/shape-outside/shape-outside-floats-clipped-003-expected.html: Added.
  • csswg/contributors/adobe/submitted/shapes/shape-outside/shape-outside-floats-clipped-003.html: Added.
  • csswg/contributors/adobe/submitted/shapes/shape-outside/w3c-import.log:
2:34 PM Changeset in webkit [157185] by timothy@apple.com
  • 2 edits in trunk/Tools

Attempt to fix faulty Dashboard iteration updating by always rebuilding test result objects.

​https://bugs.webkit.org/show_bug.cgi?id=122515

Reviewed by Alexey Proskuryakov.

  • BuildSlaveSupport/build.webkit.org-config/public_html/dashboard/Scripts/BuildbotIteration.js:

(BuildbotIteration.prototype.update):

2:01 PM Changeset in webkit [157184] by commit-queue@webkit.org
  • 4 edits in trunk/Source/WebCore

[CoordinatedGraphics] ASSERTION FAILED: !m_flushingLayers (after r156291)
​https://bugs.webkit.org/show_bug.cgi?id=122016

Patch by Sergio Correia <Sergio Correia> on 2013-10-09
Reviewed by Noam Rosenthal.

To prevent calling scheduleLayerFlush() inside flushCompositingState(),
we now only call m_client->notifyFlushRequired() - which will trigger
scheduleLayerFlush() - if we are not already flushing layer changes.

No new tests, covered by existing ones.

  • platform/graphics/texmap/coordinated/CompositingCoordinator.cpp:

(WebCore::CompositingCoordinator::notifyFlushRequired): Check if we
are already flushing layer changes before calling notifyFlushRequired.
(WebCore::CompositingCoordinator::sizeDidChange): Use notifyFlushRequired()
from CompositingCoordinator instead of calling the one from m_client
directly.
(WebCore::CompositingCoordinator::createGraphicsLayer): Ditto.
(WebCore::CompositingCoordinator::detachLayer): Ditto.

  • platform/graphics/texmap/coordinated/CoordinatedGraphicsLayer.cpp:

(WebCore::CoordinatedGraphicsLayer::notifyFlushRequired): Added helper
to check whether we are flushing layer changes before calling client()'s
notifyFlushRequired(). It returns a boolean indicating whether it
called the client's method.
(WebCore::CoordinatedGraphicsLayer::didChangeLayerState): Use the
notifyFlushRequired() helper.
(WebCore::CoordinatedGraphicsLayer::didChangeAnimations): Ditto.
(WebCore::CoordinatedGraphicsLayer::didChangeChildren): Ditto.
(WebCore::CoordinatedGraphicsLayer::didChangeFilters): Ditto.
(WebCore::CoordinatedGraphicsLayer::didChangeImageBacking): Ditto.
(WebCore::CoordinatedGraphicsLayer::setContentsNeedsDisplay): Ditto.
(WebCore::CoordinatedGraphicsLayer::setContentsToCanvas): Ditto.
(WebCore::CoordinatedGraphicsLayer::flushCompositingState): Ditto.
(WebCore::CoordinatedGraphicsLayer::tiledBackingStoreHasPendingTileCreation):Ditto.

  • platform/graphics/texmap/coordinated/CoordinatedGraphicsLayer.h:

Declaration of the notifyFlushRequired() helper.

2:00 PM Changeset in webkit [157183] by Chris Fleizach
  • 4 edits
    3 adds in trunk

AX: VoiceOver double speaks all items in <li> nodes
​https://bugs.webkit.org/show_bug.cgi?id=122564

Reviewed by Brent Fulgham.

Source/WebCore:

Test: platform/mac/accessibility/native-list-item-should-not-expose-label.html

  • accessibility/AccessibilityNodeObject.cpp:

(WebCore::AccessibilityNodeObject::visibleText):

LayoutTests:

Make a mac specific expectation for listitem-title on Mac.
Make a more directed test for Mac regarding listitem labeling.

  • platform/mac/accessibility/lists-expected.txt:
  • platform/mac/accessibility/listitem-title-expected.txt: Added.
  • platform/mac/accessibility/native-list-item-should-not-expose-label-expected.txt: Added.
  • platform/mac/accessibility/native-list-item-should-not-expose-label.html: Added.
1:44 PM Changeset in webkit [157182] by kov@webkit.org
  • 2 edits in trunk/LayoutTests

Unreviewed gardening. Many inspector protocol tests are timing out on the bots, unreproducible locally.

  • platform/gtk/TestExpectations:
1:31 PM Changeset in webkit [157181] by kov@webkit.org
  • 2 edits in trunk/LayoutTests

Unreviewed gardening. A couple of paste tests began timing out after
​http://trac.webkit.org/changeset/156678.

  • platform/gtk/TestExpectations:
1:26 PM Changeset in webkit [157180] by commit-queue@webkit.org
  • 3 edits
    2 adds in trunk

CoordinatedGraphics: Fix integer rounding when computing pixel alignment
​https://bugs.webkit.org/show_bug.cgi?id=122565

Patch by Sergio Correia <Sergio Correia> on 2013-10-09
Reviewed by Noam Rosenthal.

Source/WebCore:

Replace roundedIntRect call in computePixelAlignment() with enclosingIntRect,
so that very small rectangles (after content scaling) don't appear to be empty,
creating a crash later.

Test: transforms/3d/general/rotate-45-degrees-long-inner-text-in-dt.html

  • platform/graphics/texmap/coordinated/CoordinatedGraphicsLayer.cpp:

(WebCore::CoordinatedGraphicsLayer::computePixelAlignment):

LayoutTests:

Added test that rotates 45 degrees a very long text in a dt tag that
might crash the browser depending on how it handles the integer
rounding when computing pixel alignment.

  • transforms/3d/general/rotate-45-degrees-long-inner-text-in-dt-expected.txt: Added.
  • transforms/3d/general/rotate-45-degrees-long-inner-text-in-dt.html: Added.
1:22 PM Changeset in webkit [157179] by ap@apple.com
  • 2 edits in trunk/Tools

​http://build.webkit.org/dashboard/ links to a wrong jscore-test results page
​https://bugs.webkit.org/show_bug.cgi?id=122569

Reviewed by Tim Hatcher.

  • BuildSlaveSupport/build.webkit.org-config/public_html/dashboard/Scripts/WebKitBuildbot.js:

(WebKitBuildbot.prototype.javascriptTestResultsURLForIteration): "actual.html" used
to be produced by the old Mozilla test harness that we don't use any more.

1:15 PM Changeset in webkit [157178] by commit-queue@webkit.org
  • 2 edits in trunk/Source/WebKit2

Set view to active in ViewClientEfl after WebProcess has relaunched
​https://bugs.webkit.org/show_bug.cgi?id=118650

Patch by Sergio Correia <Sergio Correia> on 2013-10-09
Reviewed by Noam Rosenthal.

Once WebProcess relaunches, the underlying scene from the view is recreated
and is not active by default, which means it will be in `background'. In
practice, from that point on we are going to get a blank screen, unless we
set it to active.

In the normal flow, the view is set to active during its initialization; this
patch sets it to active also in ViewClientEfl, in the webProcessDidRelaunch
callback.

  • UIProcess/efl/ViewClientEfl.cpp:

(WebKit::ViewClientEfl::webProcessDidRelaunch): Added WKViewSetIsActive call.

1:05 PM Changeset in webkit [157177] by rniwa@webkit.org
  • 14 edits in trunk/Source/WebCore

Rename EventRetargeter::adjustForRelatedTarget to EventPath::setRelatedTarget
​https://bugs.webkit.org/show_bug.cgi?id=122539

Reviewed by Darin Adler.

Renamed EventRetargeter::adjustForRelatedTarget to EventPath::setRelatedTarget and got rid of
EventDispatchMediators for focus events. We set the related target inside EventDispatcher's
dispatch via Event::relatedTarget, which is now virtual and overridden by mouse/focus events.

  • dom/Element.cpp: The following functions no longer use FocusEventDispatchMediators.

(WebCore::Element::dispatchFocusInEvent):
(WebCore::Element::dispatchFocusOutEvent):
(WebCore::Element::dispatchFocusEvent):
(WebCore::Element::dispatchBlurEvent):

  • dom/Event.h:

(WebCore::Event::relatedTarget): Added as a virtual function.

  • dom/EventDispatcher.cpp:

(WebCore::EventDispatcher::dispatchEvent): Cache Event*.
(WebCore::EventDispatcher::dispatch): Call setRelatedTargetToPath at the beginning to update
related targets in the path.

  • dom/EventDispatcher.h:
  • dom/EventRetargeter.cpp:

(WebCore::EventPath::EventPath): Initialize m_origin, used in EventPath::setRelatedTarget.
(WebCore::EventRetargeter::adjustTouchList):
(WebCore::EventPath::setRelatedTarget): Renamed from EventRetargeter::adjustForRelatedTarget.
Delegated the nullity checks to callers by taking references instead of pointers.
(WebCore::EventRetargeter::calculateAdjustedNodes):

  • dom/EventRetargeter.h: Removed AdjustedNodes typedef. It's just a Vector of RefPtr<Node>.
  • dom/FocusEvent.cpp: Removed FocusEventDispatchMediators for focus events.
  • dom/FocusEvent.h:
  • dom/MouseEvent.cpp:

(WebCore::MouseEvent::toElement): Cache the related target now that relatedTarget() is virtual.
(WebCore::MouseEvent::fromElement): Ditto.
(WebCore::MouseEventDispatchMediator::mediateAndDispatchEvent): No longer sets the related
targets since that's done in EventDispatcher::dispatch now.

  • dom/MouseEvent.h:
  • dom/Node.cpp: Removed obnoxious Node::dispatchScopedEventDispatchMediator.

(WebCore::Node::dispatchScopedEvent):

  • dom/Node.h:
  • html/shadow/MediaControlElements.cpp:

(WebCore::MediaControlVolumeSliderContainerElement::defaultEventHandler): Cache the related
target in a local variable now that relatedTarget() is virtual.

11:22 AM Changeset in webkit [157176] by msaboff@apple.com
  • 2 edits in trunk/Source/WTF

Change to std::unique_ptr breaks wtf/DataLog.cpp when DATA_LOG_TO_FILE set to 1
​https://bugs.webkit.org/show_bug.cgi?id=122557

Reviewed by Filip Pizlo.

Changed the leakPtr() to release().

  • wtf/DataLog.cpp:

(WTF::initializeLogFileOnce):

10:07 AM Changeset in webkit [157175] by commit-queue@webkit.org
  • 3 edits
    1 add in trunk/Tools

Add GTK 64 bits release to the buildbot dashboard
​https://bugs.webkit.org/show_bug.cgi?id=122548

Patch by Gustavo Noronha Silva <​gustavo.noronha@collabora.com> on 2013-10-09
Reviewed by Darin Adler.

  • BuildSlaveSupport/build.webkit.org-config/public_html/dashboard/Images/GTK.png: Added.
  • BuildSlaveSupport/build.webkit.org-config/public_html/dashboard/Scripts/WebKitBuildbot.js:

(WebKitBuildbot): add 64 bits release wk2 and wk1 testers.

  • BuildSlaveSupport/build.webkit.org-config/public_html/dashboard/Styles/Main.css:

(table.queue-grid tr.platform.linux-gtk img.logo): use the GTK+ logo in its dashboard line.

10:05 AM Changeset in webkit [157174] by commit-queue@webkit.org
  • 2 edits in trunk/Source/WebKit/win

[Win] BeginPaint should only be called in response to WM_PAINT.
​https://bugs.webkit.org/show_bug.cgi?id=122549

Patch by ​peavo@outlook.com <​peavo@outlook.com> on 2013-10-09
Reviewed by Brent Fulgham.

According to the documentation, the Win32 api function BeginPaint should only be called in response to a WM_PAINT message.
Currently, the WebView window procedure might call this function (via WebView::paint(0, 0)) for all types of messages.
Also, there is potential for double drawing. If some message is dispatched before WM_PAINT, WebView::paint(0, 0) will be called.
This call does not validate the area drawn, so later the WM_PAINT will arrive, drawing the same area.
This can be fixed by calling the Win32 function UpdateWindow(), which will do a synchronous window update, and validate the area.

  • WebView.cpp:

(WebView::WebViewWndProc): Use Win32 function UpdateWindow() to draw invalid area.

10:02 AM Changeset in webkit [157173] by commit-queue@webkit.org
  • 2 edits in trunk/Source/JavaScriptCore

[arm] Inverted src and dest FP registers in DFG speculative JIT when using hardfp.
​https://bugs.webkit.org/show_bug.cgi?id=122555

Patch by Julien Brianceau <​jbriance@cisco.com> on 2013-10-09
Reviewed by Michael Saboff.

  • dfg/DFGSpeculativeJIT.h:

(JSC::DFG::SpeculativeJIT::appendCallWithExceptionCheckSetResult):
(JSC::DFG::SpeculativeJIT::appendCallSetResult):

9:47 AM Changeset in webkit [157172] by commit-queue@webkit.org
  • 3 edits in trunk/LayoutTests

[EFL] Some accessibility tests are failing
​https://bugs.webkit.org/show_bug.cgi?id=122553

Unreviewed EFL gardening

Patch by Krzysztof Czech <​k.czech@samsung.com> on 2013-10-09

  • platform/efl-wk2/TestExpectations:
  • platform/efl/TestExpectations:
9:16 AM Changeset in webkit [157171] by Chris Fleizach
  • 2 edits in trunk/Source/WebCore

AX: VoiceOver does not speak aria-expanded state on iOS
​https://bugs.webkit.org/show_bug.cgi?id=122540

Expose aria-pressed state for iOS.

Reviewed by Mario Sanchez Prada.

  • accessibility/ios/WebAccessibilityObjectWrapperIOS.mm:

(-[WebAccessibilityObjectWrapper accessibilityIsExpanded]):

8:33 AM Changeset in webkit [157170] by Carlos Garcia Campos
  • 2 edits in trunk/Source/WebCore

Unreviewed. Update GObject DOM symbols file after r157163.

  • bindings/gobject/webkitdom.symbols:
8:24 AM Changeset in webkit [157169] by commit-queue@webkit.org
  • 2 edits in trunk/Websites/planet.webkit.org

Add Frédéric Wang's blog to the planet
​https://bugs.webkit.org/show_bug.cgi?id=122541

Patch by Frédéric Wang <​fred.wang@free.fr> on 2013-10-09
Reviewed by Martin Robinson.

  • config.ini: Add my blog to the planet.
8:24 AM Changeset in webkit [157168] by andersca@apple.com
  • 6 edits
    1 delete in trunk/Source/WebCore

Remove RunLoopMac.mm
​https://bugs.webkit.org/show_bug.cgi?id=122527

Reviewed by Antti Koivisto.

RunLoop::setUseApplicationRunLoopOnMainRunLoop() is no longer used, so we can just use the run() and stop()
implementations from RunLoopCF.

  • WebCore.exp.in:
  • WebCore.xcodeproj/project.pbxproj:
  • platform/RunLoop.cpp:
  • platform/RunLoop.h:
  • platform/cf/RunLoopCF.cpp:

(WebCore::RunLoop::stop):

  • platform/mac/RunLoopMac.mm: Removed.
7:54 AM Changeset in webkit [157167] by kov@webkit.org
  • 2 edits in trunk/LayoutTests

Unreviewed gardening. Add timeout as expected result for fast/history/go-back-to-iframe-with-plugin.html,
since it times out sometimes.

  • platform/gtk/TestExpectations:
7:48 AM Changeset in webkit [157166] by kov@webkit.org
  • 2 edits in trunk/LayoutTests

Unreviewed gardening. Add timeout as expected result for mathml/very-large-stretchy-operators.html,
since it times out on the bots.

  • platform/gtk/TestExpectations:
7:16 AM Changeset in webkit [157165] by mario@webkit.org
  • 6 edits
    1 delete in trunk

[ATK] Implement new API in AtkText: atk_text_get_string_at_offset()
​https://bugs.webkit.org/show_bug.cgi?id=120638

Reviewed by Gustavo Noronha Silva.

Source/WebCore:

Implemented new atk_text_get_string_at_offset() API, introduced in
ATK 2.9.4 to simplify how text at a given offset is retrieved.

  • accessibility/atk/WebKitAccessibleInterfaceText.cpp:

(webkitAccessibleTextGetStringAtOffset): New function implementing
the new API, which basically translates calls to the new API to
calls to the old API using 'at' positions and START boundaries.
(webkitAccessibleTextInterfaceInit): Hook the new function.

Source/WebKit/gtk:

Update ATK unit test to check the new API as well if the version
of ATK is new enough.

  • tests/testatk.c:

(testGetStringFunction): New helper function, to check the new API.
(runGetStringTests): Ditto.
(testWebkitAtkGetStringAtOffset): New unit test, similar to the
one already present to check the old API but focused in the new one.
(testWebkitAtkGetStringAtOffsetNewlines): Ditto.
(testWebkitAtkGetStringAtOffsetTextarea): Ditto.
(testWebkitAtkGetStringAtOffsetTextInput): Ditto.
(testWebkitAtkGetStringAtOffsetWithPreformattedText): Ditto.
(testWebkitAtkGetStringAtOffsetWithSpecialCharacters): Ditto.
(testWebkitAtkGetStringAtOffsetWithWrappedLines): Ditto.
(testWebkitAtkGetStringAtOffsetWithEmbeddedObjects): Ditto.
(testWebkitAtkGetExtents): Add checks for the new API, if possible.
(testWebkitAtkLinksWithInlineImages): Ditto.
(main): Add the new unit tests to check the new API, if possible.

Tools:

  • gtk/jhbuild.modules: Raised ATK and AT-SPI versions to 2.10.0, so we

can check both the old and new APIs to retrieve text at a given offset.

6:29 AM Changeset in webkit [157164] by msaboff@apple.com
  • 13 edits in trunk/Source/JavaScriptCore

Transition call and construct JITStubs to CCallHelper functions
​https://bugs.webkit.org/show_bug.cgi?id=122453

Reviewed by Geoffrey Garen.

Transitioned cti_op_call_eval to operationCallEval. Migrated baseline JIT to use the same
call thunks as the DFG. Eliminated all of the "oldStyle" thunks and related functions.

  • bytecode/CallLinkInfo.cpp:

(JSC::CallLinkInfo::unlink):

  • jit/JIT.cpp:

(JSC::JIT::linkFor):
(JSC::JIT::linkSlowCall):

  • jit/JIT.h:
  • jit/JITCall.cpp:

(JSC::JIT::compileCallEval):
(JSC::JIT::compileCallEvalSlowCase):
(JSC::JIT::compileOpCallSlowCase):
(JSC::JIT::privateCompileClosureCall):

  • jit/JITCall32_64.cpp:

(JSC::JIT::compileCallEval):
(JSC::JIT::compileCallEvalSlowCase):
(JSC::JIT::compileOpCallSlowCase):
(JSC::JIT::privateCompileClosureCall):

  • jit/JITInlines.h:

(JSC::JIT::callOperationWithCallFrameRollbackOnException):

  • jit/JITOperations.cpp:
  • jit/JITOperations.h:
  • jit/JITStubs.cpp:
  • jit/JITStubs.h:
  • jit/ThunkGenerators.cpp:
  • jit/ThunkGenerators.h:
6:07 AM Changeset in webkit [157163] by Carlos Garcia Campos
  • 3 edits in trunk/Source/WebCore

[GTK] mouse wheel events are not considered to be mouse events
​https://bugs.webkit.org/show_bug.cgi?id=120648

Reviewed by Gustavo Noronha Silva.

Expose WheelEvent in the GObject DOM bindings API.

  • bindings/gobject/GNUmakefile.am: Add new files to compilation.
  • bindings/gobject/WebKitDOMPrivate.cpp:

(WebKit::wrap): Create a WebKitDOMWheelEvent when the given Event
is a WheelEvent.

5:50 AM Changeset in webkit [157162] by commit-queue@webkit.org
  • 3 edits in trunk/Source/WebKit2

Unreviewed, rolling out r157157.
​http://trac.webkit.org/changeset/157157
​https://bugs.webkit.org/show_bug.cgi?id=122547

Caused crashes on several bots including gtk, mac-wk2 lion and
mountain lion (Requested by kov on #webkit).

  • Shared/Plugins/NPRemoteObjectMap.cpp:

(WebKit::NPRemoteObjectMap::registerNPObject):
(WebKit::NPRemoteObjectMap::pluginDestroyed):

  • Shared/Plugins/NPRemoteObjectMap.h:
4:21 AM Changeset in webkit [157161] by commit-queue@webkit.org
  • 3 edits in trunk/Source/JavaScriptCore

[sh4] Fix lots of unused parameter warnings.
​https://bugs.webkit.org/show_bug.cgi?id=122545

Patch by Julien Brianceau <​jbriance@cisco.com> on 2013-10-09
Reviewed by Csaba Osztrogonác.

  • assembler/MacroAssemblerSH4.h:

(JSC::MacroAssemblerSH4::revertJumpReplacementToPatchableBranchPtrWithPatch):

  • assembler/SH4Assembler.h:

(JSC::SH4Assembler::andlImm8r):
(JSC::SH4Assembler::orlImm8r):
(JSC::SH4Assembler::xorlImm8r):
(JSC::SH4Assembler::cmpEqImmR0):
(JSC::SH4Assembler::testlImm8r):
(JSC::SH4Assembler::movwPCReg):
(JSC::SH4Assembler::movwMemReg):
(JSC::SH4Assembler::movbMemReg):
(JSC::SH4Assembler::printInstr):
(JSC::SH4Assembler::printBlockInstr):

4:06 AM Changeset in webkit [157160] by commit-queue@webkit.org
  • 3 edits in trunk/LayoutTests

[EFL] accessibility/table-cells.html is failing
​https://bugs.webkit.org/show_bug.cgi?id=122544

Unreviewed EFL gardening

Changed expectations so that they show proper name of the parent (the table)

Patch by Krzysztof Czech <​k.czech@samsung.com> on 2013-10-09

  • platform/efl-wk1/accessibility/table-cells-expected.txt:
  • platform/efl-wk2/accessibility/table-cells-expected.txt:
2:16 AM Changeset in webkit [157159] by commit-queue@webkit.org
  • 4 edits in trunk

[sh4] Add sh4 support when building with CMake.
​https://bugs.webkit.org/show_bug.cgi?id=122542

Patch by Julien Brianceau <​jbriance@cisco.com> on 2013-10-09
Reviewed by Csaba Osztrogonác.

.:

  • CMakeLists.txt:

Source/JavaScriptCore:

  • CMakeLists.txt:

Oct 8, 2013:

11:23 PM Changeset in webkit [157158] by gyuyoung.kim@samsung.com
  • 2 edits in trunk/Source/WebCore

Unreviewed. Fix EFL buildbreak since r157152.

  • dom/EventRetargeter.cpp:

(WebCore::EventRetargeter::adjustForTouchEvent):

10:26 PM Changeset in webkit [157157] by Darin Adler
  • 3 edits in trunk/Source/WebKit2

Remove use of deleteAllValues in NPRemoteObjectMap::pluginDestroyed
​https://bugs.webkit.org/show_bug.cgi?id=122496

Reviewed by Andreas Kling.

  • Shared/Plugins/NPRemoteObjectMap.cpp:

(WebKit::NPRemoteObjectMap::registerNPObject): Don't call release when puttin objects
into the map.
(WebKit::NPRemoteObjectMap::pluginDestroyed): Instead of making an array of receivers
and deleting them all, remove all of the receivers from the map, which will take care
of deleting them. Also change the other loop to use a similar coding style.

  • Shared/Plugins/NPRemoteObjectMap.h: Change the value type of m_registeredNPObjects

to be unique_ptr instead of a raw pointer.

7:36 PM Changeset in webkit [157156] by kov@webkit.org
  • 2 edits in trunk/LayoutTests

Fix expectation for geolocation test that has been timing out.

  • platform/gtk-wk2/TestExpectations:
7:07 PM Changeset in webkit [157155] by kov@webkit.org
  • 3 edits in trunk/LayoutTests

Unreviewed gardening. Move http/tests/loading/307-after-303-after-post.html failure
expectation to the global gtk expectations file, since it's also true for gtk-wk2.

  • platform/gtk-wk1/TestExpectations:
  • platform/gtk/TestExpectations:
6:16 PM Changeset in webkit [157154] by commit-queue@webkit.org
  • 8 edits
    1 delete in trunk

Improved WinLauncher.
​https://bugs.webkit.org/show_bug.cgi?id=122536

Patch by Alex Christensen <​achristensen@webkit.org> on 2013-10-08
Reviewed by Brent Fulgham.

Source/WebCore:

  • platform/network/curl/ResourceHandleManager.cpp:

(WebCore::cookieJarPath):
Put cookies.dat into the AppData directory on Windows.

Tools:

  • WinLauncher/PrintWebUIDelegate.cpp:

(PrintWebUIDelegate::runJavaScriptAlertPanelWithMessage): Added.
(PrintWebUIDelegate::runJavaScriptConfirmPanelWithMessage): Added.

  • WinLauncher/PrintWebUIDelegate.h: Removed stubs.
  • WinLauncher/WinLauncher.cpp:

(createCrashReport): Use the executable name instead of "WinLauncher".

  • WinLauncher/WinLauncher.vcxproj/WinLauncher.exe.manifest: Removed.
  • WinLauncher/WinLauncher.vcxproj/WinLauncherLibCommon.props: Link to Wininet.lib.
  • win/DLLLauncher/DLLLauncherMain.cpp: Remove VS2005 dependency for WinCairo.
5:45 PM Changeset in webkit [157153] by rniwa@webkit.org
  • 2 edits in trunk/Source/WebCore

Remove the code erroneously in the previous commit.

  • dom/EventDispatcher.h:

(WebCore::EventPath::contextAt):

5:41 PM Changeset in webkit [157152] by rniwa@webkit.org
  • 8 edits in trunk/Source/WebCore

Turn EventPath into a real class
​https://bugs.webkit.org/show_bug.cgi?id=122533

Reviewed by Antti Koivisto.

Turned EventPath into a real class and made EventRetargeter::calculateEventPath its constructor.
We can turn a lot of functions in EventRetargeter.cpp member functions of EventPath since many
of them manipulate EventPath in one way or another.

  • dom/EventContext.h:

(WebCore::toMouseOrFocusEventContext):
(WebCore::toTouchEventContext):

  • dom/EventDispatcher.cpp:

(WebCore::EventDispatcher::EventDispatcher):
(WebCore::EventDispatcher::dispatchEventAtCapturing):
(WebCore::EventDispatcher::dispatchEventAtTarget):
(WebCore::EventDispatcher::dispatchEventAtBubbling):
(WebCore::EventDispatcher::dispatchEventPostProcess):
(WebCore::EventDispatcher::topEventContext):
(WebCore::EventPath::hasEventListeners):

  • dom/EventDispatcher.h:

(WebCore::EventPath::isEmpty):
(WebCore::EventPath::size):
(WebCore::EventPath::item):
(WebCore::EventPath::targetRespectingTargetRules):
(WebCore::EventPath::shrink):
(WebCore::EventPath::lastContextIfExists):

  • dom/EventRetargeter.cpp:

(WebCore::EventPath::EventPath):
(WebCore::EventRetargeter::adjustForTouchEvent):
(WebCore::EventRetargeter::adjustForRelatedTarget):
(WebCore::EventRetargeter::calculateAdjustedNodes):

  • dom/EventRetargeter.h:
  • inspector/InspectorInstrumentation.cpp:

(WebCore::eventHasListeners):

  • inspector/InspectorInstrumentation.h:
5:38 PM Changeset in webkit [157151] by kov@webkit.org
  • 2 edits in trunk/LayoutTests

Unreviewed gardening. The 64 bits release bot has been made to use 8 bpp depth, so
fast/dom/Window/window-screen-properties.html won't pass there.

  • platform/gtk/TestExpectations:
5:24 PM Changeset in webkit [157150] by oliver@apple.com
  • 12 edits in trunk

Convert for-of iteration to in-band signalling so we can trivially avoid unnecessary object allocation
​https://bugs.webkit.org/show_bug.cgi?id=122532

Reviewed by Michael Saboff.

Source/JavaScriptCore:

Switch for-of enumeration to use in band signalling to determine the end
of iteration. This allows us to trivially remove an otherwise unnecessary
object allocation, and paves the way for optimised thunks in future.

We can re-add explicit .next() functions in future that would marshall
the true iteration functions, but for now we'll ignore them.

This results in a huge improvement in the performance of for-of (in the order
of 2x) but there's still a long way to go in order to get the performance to
a satisfactory level.

  • bytecompiler/NodesCodegen.cpp:

(JSC::ForOfNode::emitBytecode):

  • runtime/ArrayIteratorPrototype.cpp:

(JSC::ArrayIteratorPrototype::finishCreation):
(JSC::createIteratorResult):

  • runtime/CommonIdentifiers.cpp:

(JSC::CommonIdentifiers::CommonIdentifiers):

  • runtime/CommonIdentifiers.h:
  • runtime/Identifier.cpp:

(JSC::Identifier::addSlowCase):

  • runtime/JSObject.h:

(JSC::JSFinalObject::create):

  • runtime/VM.cpp:

(JSC::VM::VM):

  • runtime/VM.h:

LayoutTests:

Update tests to reflect our non-exposure of .next()

  • js/array-iterators-expected.txt:
  • js/script-tests/array-iterators.js:
5:21 PM Changeset in webkit [157149] by kov@webkit.org
  • 1 edit
    1 move
    1 add in trunk/LayoutTests

Unreviewed gardening. Make platform-specific expectation file apply only to gtk-wk1, since
gtk-wk2 gets the output right.

  • platform/gtk-wk1/fast/loader/show-only-one-beforeunload-dialog-expected.txt: Renamed from LayoutTests/platform/gtk/fast/loader/show-only-one-beforeunload-dialog-expected.txt.
5:08 PM Changeset in webkit [157148] by kov@webkit.org
  • 2 edits in trunk/LayoutTests

Unreviewed gardening. Update gtk-wk2-specific expectations for a couple of
fast/dom/Window tests:

  • fast/dom/Window/window-postmessage-clone.html (times out)
  • fast/dom/Window/open-window-min-size.html (fails)
  • platform/gtk-wk2/TestExpectations:
5:05 PM Changeset in webkit [157147] by commit-queue@webkit.org
  • 9 edits in trunk/Source

Source/JavaScriptCore: Fixed compile errors while compiling without the JIT enabled.
​https://bugs.webkit.org/show_bug.cgi?id=122530

Patch by Alex Christensen <​achristensen@webkit.org> on 2013-10-08
Reviewed by Brent Fulgham.

  • jit/JITOperations.cpp:

Protected with #if ENABLE(JIT) like the rest of the JIT source.

Source/ThirdParty/ANGLE: Generate pdb files for debugging ANGLE DLLs.
​https://bugs.webkit.org/show_bug.cgi?id=122530

Patch by Alex Christensen <​achristensen@webkit.org> on 2013-10-08
Reviewed by Brent Fulgham.

  • ANGLE.vcxproj/libEGLCommon.props:
  • ANGLE.vcxproj/libGLESv2Common.props:

Generate debug information.

Source/WebCore: Enabled all necessary files for Win64 build.
​https://bugs.webkit.org/show_bug.cgi?id=122530

Patch by Alex Christensen <​achristensen@webkit.org> on 2013-10-08
Reviewed by Brent Fulgham.

  • WebCore.vcxproj/WebCore.vcxproj:

Enabled MediaPlayerPrivateFullscreenWindow and
RenderLayerModelObject to be built in Win64 build.

Source/WebKit: Fixed Win64 linking.
​https://bugs.webkit.org/show_bug.cgi?id=122530

Patch by Alex Christensen <​achristensen@webkit.org> on 2013-10-08
Reviewed by Brent Fulgham.

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

Added 64-bit linker symbols.

4:49 PM Changeset in webkit [157146] by weinig@apple.com
  • 14 edits in trunk/Source

Consolidate findString functions
​https://bugs.webkit.org/show_bug.cgi?id=122480

Reviewed by Darin Adler.

../WebCore:

  • editing/Editor.cpp:
  • editing/Editor.h:
  • editing/EditorCommand.cpp:

(WebCore::executeFindString):

  • page/DOMWindow.cpp:

(WebCore::DOMWindow::find):

  • page/Page.cpp:
  • page/Page.h:

Fix a long standing FIXME.

../WebKit/efl:

  • ewk/ewk_frame.cpp:

(ewk_frame_text_search):

  • ewk/ewk_view.cpp:

(ewk_view_text_search):

../WebKit/gtk:

  • webkit/webkitwebview.cpp:

(webkit_web_view_search_text):

../WebKit/win:

  • WebView.cpp:

(WebView::searchFor):

4:25 PM Changeset in webkit [157145] by kov@webkit.org
  • 2 edits in trunk/LayoutTests

Upgrade accessibility/title-ui-element-correctness.html from crash (wk2-specific)
to fail (gtk-wide) expectation.

  • platform/gtk-wk2/TestExpectations:
4:20 PM Changeset in webkit [157144] by Bem Jones-Bey
  • 13 edits in trunk/Source/WebCore

Move m_floatingObjects to RenderBlockFlow from RenderBlock
​https://bugs.webkit.org/show_bug.cgi?id=122512

Reviewed by David Hyatt.

Floats are part of block flow, not every RenderBlock can contain
floars. This change makes it so that RenderBlock no longer can contain
floats, only RenderBlockFlow can.

This also moves a bunch of line layout stuff over to RenderBlockFlow,
but makes no effort to move all of line layout, as all of that is
destined to move out of both RenderBlockFlow and RenderBlock into its
own class.

No new tests, no behavior change.

  • rendering/InlineIterator.h:

(WebCore::InlineBidiResolver::appendRun):

  • rendering/LineWidth.cpp:

(WebCore::LineWidth::LineWidth):

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

(WebCore::RenderBlock::styleWillChange):
(WebCore::RenderBlock::styleDidChange):
(WebCore::RenderBlock::splitFlow):
(WebCore::RenderBlock::deleteLineBoxTree):
(WebCore::RenderBlock::removeChild):
(WebCore::RenderBlock::computeOverflow):
(WebCore::RenderBlock::selectionGaps):
(WebCore::RenderBlock::adjustForBorderFit):

  • rendering/RenderBlock.h:

(WebCore::RenderBlock::containsFloats):
(WebCore::RenderBlock::shouldSkipCreatingRunsForObject):
(WebCore::RenderBlock::addOverflowFromFloats):
(WebCore::RenderBlock::logicalRightFloatOffsetForLine):
(WebCore::RenderBlock::logicalLeftFloatOffsetForLine):
(WebCore::RenderBlock::moveAllChildrenOnRemovalTo):
(WebCore::RenderBlock::paintFloats):
(WebCore::RenderBlock::hitTestFloats):
(WebCore::RenderBlock::clipOutFloatingObjects):

  • rendering/RenderBlockFlow.cpp:

(WebCore::RenderBlockFlow::clearFloats):
(WebCore::RenderBlockFlow::layoutBlock):
(WebCore::RenderBlockFlow::layoutBlockChild):
(WebCore::RenderBlockFlow::collapseMargins):
(WebCore::RenderBlockFlow::clearFloatsIfNeeded):
(WebCore::RenderBlockFlow::adjustBlockChildForPagination):
(WebCore::RenderBlockFlow::containsFloat):
(WebCore::RenderBlockFlow::styleDidChange):
(WebCore::RenderBlockFlow::styleWillChange):
(WebCore::RenderBlockFlow::deleteLineBoxTree):
(WebCore::RenderBlockFlow::moveAllChildrenOnRemovalTo):
(WebCore::RenderBlockFlow::addOverflowFromFloats):
(WebCore::RenderBlockFlow::computeOverflow):
(WebCore::RenderBlockFlow::repaintOverhangingFloats):
(WebCore::RenderBlockFlow::paintFloats):
(WebCore::RenderBlockFlow::clipOutFloatingObjects):
(WebCore::RenderBlockFlow::createFloatingObjects):
(WebCore::RenderBlockFlow::removeFloatingObjects):
(WebCore::RenderBlockFlow::insertFloatingObject):
(WebCore::RenderBlockFlow::removeFloatingObject):
(WebCore::RenderBlockFlow::removeFloatingObjectsBelow):
(WebCore::RenderBlockFlow::computeLogicalLocationForFloat):
(WebCore::RenderBlockFlow::positionNewFloats):
(WebCore::RenderBlockFlow::newLine):
(WebCore::RenderBlockFlow::logicalLeftFloatOffsetForLine):
(WebCore::RenderBlockFlow::logicalRightFloatOffsetForLine):
(WebCore::RenderBlockFlow::nextFloatLogicalBottomBelow):
(WebCore::RenderBlockFlow::lowestFloatLogicalBottom):
(WebCore::RenderBlockFlow::addOverhangingFloats):
(WebCore::RenderBlockFlow::hasOverhangingFloat):
(WebCore::RenderBlockFlow::addIntrudingFloats):
(WebCore::RenderBlockFlow::markAllDescendantsWithFloatsForLayout):
(WebCore::RenderBlockFlow::markSiblingsWithFloatsForLayout):
(WebCore::RenderBlockFlow::getClearDelta):
(WebCore::RenderBlockFlow::hitTestFloats):
(WebCore::RenderBlockFlow::adjustForBorderFit):

  • rendering/RenderBlockFlow.h:

(WebCore::RenderBlockFlow::hasOverhangingFloats):

  • rendering/RenderBlockLineLayout.cpp:

(WebCore::LineBreaker::LineBreaker):
(WebCore::RenderBlockFlow::appendRunsForObject):
(WebCore::RenderBlockFlow::createLineBoxes):
(WebCore::RenderBlockFlow::constructLine):
(WebCore::RenderBlockFlow::setMarginsForRubyRun):
(WebCore::updateLogicalInlinePositions):
(WebCore::RenderBlockFlow::computeInlineDirectionPositionsForLine):
(WebCore::RenderBlockFlow::computeInlineDirectionPositionsForSegment):
(WebCore::RenderBlockFlow::computeBlockDirectionPositionsForLine):
(WebCore::setStaticPositions):
(WebCore::RenderBlockFlow::handleTrailingSpaces):
(WebCore::RenderBlockFlow::appendFloatingObjectToLastLine):
(WebCore::constructBidiRunsForLine):
(WebCore::RenderBlockFlow::createLineBoxesFromBidiRuns):
(WebCore::RenderBlockFlow::layoutRunsAndFloats):
(WebCore::RenderBlockFlow::restartLayoutRunsAndFloatsInRange):
(WebCore::pushShapeContentOverflowBelowTheContentBox):
(WebCore::RenderBlockFlow::updateShapeAndSegmentsForCurrentLine):
(WebCore::RenderBlockFlow::updateShapeAndSegmentsForCurrentLineInFlowThread):
(WebCore::RenderBlockFlow::adjustLogicalLineTopAndLogicalHeightIfNeeded):
(WebCore::RenderBlockFlow::layoutRunsAndFloatsInRange):
(WebCore::RenderBlockFlow::linkToEndLineIfNeeded):
(WebCore::RenderBlockFlow::repaintDirtyFloats):
(WebCore::RenderBlockFlow::checkFloatsInCleanLine):
(WebCore::RenderBlockFlow::determineStartPosition):
(WebCore::RenderBlockFlow::determineEndPosition):
(WebCore::RenderBlockFlow::checkPaginationAndFloatsAtEndLine):
(WebCore::RenderBlockFlow::matchedEndLine):
(WebCore::shouldSkipWhitespaceAfterStartObject):
(WebCore::updateSegmentsForShapes):
(WebCore::RenderBlockFlow::deleteEllipsisLineBoxes):
(WebCore::RenderBlockFlow::checkLinesForTextOverflow):
(WebCore::RenderBlockFlow::positionNewFloatOnLine):

  • rendering/RenderBox.cpp:

(WebCore::RenderBox::outermostBlockContainingFloatingObject):
(WebCore::RenderBox::removeFloatingOrPositionedChildFromBlockLists):

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

(WebCore::RenderBoxModelObject::moveChildrenTo):

  • rendering/RenderInline.cpp:

(WebCore::RenderInline::splitFlow):

4:05 PM Changeset in webkit [157143] by timothy_horton@apple.com
  • 2 edits in trunk/Source/WTF

Turn threaded scrolling on for all Mac ports.
Rubber-stamped by Simon Fraser and Sam Weinig.

  • wtf/FeatureDefines.h:
4:01 PM Changeset in webkit [157142] by mihnea@adobe.com
  • 3 edits
    130 moves in trunk/LayoutTests

[CSSRegions] Move autoheight* tests into fast/regions/auto-size
​https://bugs.webkit.org/show_bug.cgi?id=122514

Reviewed by Sam Weinig.

Move tests, fix include paths, change TestExpectations when necessary.

  • fast/regions/auto-size/abspos-autoheight-nested-region-expected.txt: Renamed from LayoutTests/fast/regions/abspos-autoheight-nested-region-expected.txt.
  • fast/regions/auto-size/abspos-autoheight-nested-region.html: Renamed from LayoutTests/fast/regions/abspos-autoheight-nested-region.html.
  • fast/regions/auto-size/autoheight-abspos-bottom-align-expected.txt: Renamed from LayoutTests/fast/regions/autoheight-abspos-bottom-align-expected.txt.
  • fast/regions/auto-size/autoheight-abspos-bottom-align.html: Renamed from LayoutTests/fast/regions/autoheight-abspos-bottom-align.html.
  • fast/regions/auto-size/autoheight-abspos-region-expected.html: Renamed from LayoutTests/fast/regions/autoheight-abspos-region-expected.html.
  • fast/regions/auto-size/autoheight-abspos-region.html: Renamed from LayoutTests/fast/regions/autoheight-abspos-region.html.
  • fast/regions/auto-size/autoheight-allregions-expected.html: Renamed from LayoutTests/fast/regions/autoheight-allregions-expected.html.
  • fast/regions/auto-size/autoheight-allregions-nobreaks-expected.html: Renamed from LayoutTests/fast/regions/autoheight-allregions-nobreaks-expected.html.
  • fast/regions/auto-size/autoheight-allregions-nobreaks.html: Renamed from LayoutTests/fast/regions/autoheight-allregions-nobreaks.html.
  • fast/regions/auto-size/autoheight-allregions.html: Renamed from LayoutTests/fast/regions/autoheight-allregions.html.
  • fast/regions/auto-size/autoheight-break-after-expected.html: Renamed from LayoutTests/fast/regions/autoheight-break-after-expected.html.
  • fast/regions/auto-size/autoheight-break-after.html: Renamed from LayoutTests/fast/regions/autoheight-break-after.html.
  • fast/regions/auto-size/autoheight-breakafteralways-maxheight-expected.html: Renamed from LayoutTests/fast/regions/autoheight-breakafteralways-maxheight-expected.html.
  • fast/regions/auto-size/autoheight-breakafteralways-maxheight.html: Renamed from LayoutTests/fast/regions/autoheight-breakafteralways-maxheight.html.
  • fast/regions/auto-size/autoheight-breakbefore-wrongheight-expected.txt: Renamed from LayoutTests/fast/regions/autoheight-breakbefore-wrongheight-expected.txt.
  • fast/regions/auto-size/autoheight-breakbefore-wrongheight.html: Renamed from LayoutTests/fast/regions/autoheight-breakbefore-wrongheight.html.
  • fast/regions/auto-size/autoheight-breakbeforealways-expected.html: Renamed from LayoutTests/fast/regions/autoheight-breakbeforealways-expected.html.
  • fast/regions/auto-size/autoheight-breakbeforealways.html: Renamed from LayoutTests/fast/regions/autoheight-breakbeforealways.html.
  • fast/regions/auto-size/autoheight-correct-region-for-lines-2-expected.html: Renamed from LayoutTests/fast/regions/autoheight-correct-region-for-lines-2-expected.html.
  • fast/regions/auto-size/autoheight-correct-region-for-lines-2.html: Renamed from LayoutTests/fast/regions/autoheight-correct-region-for-lines-2.html.
  • fast/regions/auto-size/autoheight-correct-region-for-lines-expected.html: Renamed from LayoutTests/fast/regions/autoheight-correct-region-for-lines-expected.html.
  • fast/regions/auto-size/autoheight-correct-region-for-lines.html: Renamed from LayoutTests/fast/regions/autoheight-correct-region-for-lines.html.
  • fast/regions/auto-size/autoheight-definedheight-changenotdetected-expected.html: Renamed from LayoutTests/fast/regions/autoheight-definedheight-changenotdetected-expected.html.
  • fast/regions/auto-size/autoheight-definedheight-changenotdetected.html: Renamed from LayoutTests/fast/regions/autoheight-definedheight-changenotdetected.html.
  • fast/regions/auto-size/autoheight-dynamic-update-expected.html: Renamed from LayoutTests/fast/regions/autoheight-dynamic-update-expected.html.
  • fast/regions/auto-size/autoheight-dynamic-update.html: Renamed from LayoutTests/fast/regions/autoheight-dynamic-update.html.
  • fast/regions/auto-size/autoheight-firstregion-breakalways-expected.html: Renamed from LayoutTests/fast/regions/autoheight-firstregion-breakalways-expected.html.
  • fast/regions/auto-size/autoheight-firstregion-breakalways.html: Renamed from LayoutTests/fast/regions/autoheight-firstregion-breakalways.html.
  • fast/regions/auto-size/autoheight-flexbox-expected.html: Renamed from LayoutTests/fast/regions/autoheight-flexbox-expected.html.
  • fast/regions/auto-size/autoheight-flexbox.html: Renamed from LayoutTests/fast/regions/autoheight-flexbox.html.
  • fast/regions/auto-size/autoheight-floatright-region-expected.html: Renamed from LayoutTests/fast/regions/autoheight-floatright-region-expected.html.
  • fast/regions/auto-size/autoheight-floatright-region.html: Renamed from LayoutTests/fast/regions/autoheight-floatright-region.html.
  • fast/regions/auto-size/autoheight-horizontal-bt-expected.html: Renamed from LayoutTests/fast/regions/autoheight-horizontal-bt-expected.html.
  • fast/regions/auto-size/autoheight-horizontal-bt.html: Renamed from LayoutTests/fast/regions/autoheight-horizontal-bt.html.
  • fast/regions/auto-size/autoheight-inlineblock-normalflow-region-expected.html: Renamed from LayoutTests/fast/regions/autoheight-inlineblock-normalflow-region-expected.html.
  • fast/regions/auto-size/autoheight-inlineblock-normalflow-region.html: Renamed from LayoutTests/fast/regions/autoheight-inlineblock-normalflow-region.html.
  • fast/regions/auto-size/autoheight-lastregion-overflowauto-breaksignored-expected.html: Renamed from LayoutTests/fast/regions/autoheight-lastregion-overflowauto-breaksignored-expected.html.
  • fast/regions/auto-size/autoheight-lastregion-overflowauto-breaksignored.html: Renamed from LayoutTests/fast/regions/autoheight-lastregion-overflowauto-breaksignored.html.
  • fast/regions/auto-size/autoheight-lastregion-overflowauto-expected.html: Renamed from LayoutTests/fast/regions/autoheight-lastregion-overflowauto-expected.html.
  • fast/regions/auto-size/autoheight-lastregion-overflowauto.html: Renamed from LayoutTests/fast/regions/autoheight-lastregion-overflowauto.html.
  • fast/regions/auto-size/autoheight-maxheight-mixed-break-expected.txt: Renamed from LayoutTests/fast/regions/autoheight-maxheight-mixed-break-expected.txt.
  • fast/regions/auto-size/autoheight-maxheight-mixed-break.html: Renamed from LayoutTests/fast/regions/autoheight-maxheight-mixed-break.html.
  • fast/regions/auto-size/autoheight-maxheight-region-expected.html: Renamed from LayoutTests/fast/regions/autoheight-maxheight-region-expected.html.
  • fast/regions/auto-size/autoheight-maxheight-region.html: Renamed from LayoutTests/fast/regions/autoheight-maxheight-region.html.
  • fast/regions/auto-size/autoheight-maxheight-simple-break-expected.txt: Renamed from LayoutTests/fast/regions/autoheight-maxheight-simple-break-expected.txt.
  • fast/regions/auto-size/autoheight-maxheight-simple-break.html: Renamed from LayoutTests/fast/regions/autoheight-maxheight-simple-break.html.
  • fast/regions/auto-size/autoheight-maxheight-simple-nobreak-expected.txt: Renamed from LayoutTests/fast/regions/autoheight-maxheight-simple-nobreak-expected.txt.
  • fast/regions/auto-size/autoheight-maxheight-simple-nobreak.html: Renamed from LayoutTests/fast/regions/autoheight-maxheight-simple-nobreak.html.
  • fast/regions/auto-size/autoheight-middleregion-expected.html: Renamed from LayoutTests/fast/regions/autoheight-middleregion-expected.html.
  • fast/regions/auto-size/autoheight-middleregion.html: Renamed from LayoutTests/fast/regions/autoheight-middleregion.html.
  • fast/regions/auto-size/autoheight-minheight-region-expected.html: Renamed from LayoutTests/fast/regions/autoheight-minheight-region-expected.html.
  • fast/regions/auto-size/autoheight-minheight-region.html: Renamed from LayoutTests/fast/regions/autoheight-minheight-region.html.
  • fast/regions/auto-size/autoheight-minmaxheight-mixed-break-expected.txt: Renamed from LayoutTests/fast/regions/autoheight-minmaxheight-mixed-break-expected.txt.
  • fast/regions/auto-size/autoheight-minmaxheight-mixed-break-hbt-expected.txt: Renamed from LayoutTests/fast/regions/autoheight-minmaxheight-mixed-break-hbt-expected.txt.
  • fast/regions/auto-size/autoheight-minmaxheight-mixed-break-hbt.html: Renamed from LayoutTests/fast/regions/autoheight-minmaxheight-mixed-break-hbt.html.
  • fast/regions/auto-size/autoheight-minmaxheight-mixed-break-vlr-expected.txt: Renamed from LayoutTests/fast/regions/autoheight-minmaxheight-mixed-break-vlr-expected.txt.
  • fast/regions/auto-size/autoheight-minmaxheight-mixed-break-vlr.html: Renamed from LayoutTests/fast/regions/autoheight-minmaxheight-mixed-break-vlr.html.
  • fast/regions/auto-size/autoheight-minmaxheight-mixed-break-vrl-expected.txt: Renamed from LayoutTests/fast/regions/autoheight-minmaxheight-mixed-break-vrl-expected.txt.
  • fast/regions/auto-size/autoheight-minmaxheight-mixed-break-vrl.html: Renamed from LayoutTests/fast/regions/autoheight-minmaxheight-mixed-break-vrl.html.
  • fast/regions/auto-size/autoheight-minmaxheight-mixed-break.html: Renamed from LayoutTests/fast/regions/autoheight-minmaxheight-mixed-break.html.
  • fast/regions/auto-size/autoheight-minmaxheight-simple-break-expected.txt: Renamed from LayoutTests/fast/regions/autoheight-minmaxheight-simple-break-expected.txt.
  • fast/regions/auto-size/autoheight-minmaxheight-simple-break.html: Renamed from LayoutTests/fast/regions/autoheight-minmaxheight-simple-break.html.
  • fast/regions/auto-size/autoheight-minmaxheight-simple-nobreak-expected.txt: Renamed from LayoutTests/fast/regions/autoheight-minmaxheight-simple-nobreak-expected.txt.
  • fast/regions/auto-size/autoheight-minmaxheight-simple-nobreak.html: Renamed from LayoutTests/fast/regions/autoheight-minmaxheight-simple-nobreak.html.
  • fast/regions/auto-size/autoheight-mixed-nested-complex-regions-expected.txt: Renamed from LayoutTests/fast/regions/autoheight-mixed-nested-complex-regions-expected.txt.
  • fast/regions/auto-size/autoheight-mixed-nested-complex-regions.html: Renamed from LayoutTests/fast/regions/autoheight-mixed-nested-complex-regions.html.
  • fast/regions/auto-size/autoheight-mixed-nested-regions-expected.txt: Renamed from LayoutTests/fast/regions/autoheight-mixed-nested-regions-expected.txt.
  • fast/regions/auto-size/autoheight-mixed-nested-regions.html: Renamed from LayoutTests/fast/regions/autoheight-mixed-nested-regions.html.
  • fast/regions/auto-size/autoheight-mixed-parallel-regions-expected.txt: Renamed from LayoutTests/fast/regions/autoheight-mixed-parallel-regions-expected.txt.
  • fast/regions/auto-size/autoheight-mixed-parallel-regions.html: Renamed from LayoutTests/fast/regions/autoheight-mixed-parallel-regions.html.
  • fast/regions/auto-size/autoheight-nested-regions-expected.txt: Renamed from LayoutTests/fast/regions/autoheight-nested-regions-expected.txt.
  • fast/regions/auto-size/autoheight-nested-regions.html: Renamed from LayoutTests/fast/regions/autoheight-nested-regions.html.
  • fast/regions/auto-size/autoheight-normalflow-region-expected.html: Renamed from LayoutTests/fast/regions/autoheight-normalflow-region-expected.html.
  • fast/regions/auto-size/autoheight-normalflow-region.html: Renamed from LayoutTests/fast/regions/autoheight-normalflow-region.html.
  • fast/regions/auto-size/autoheight-region-decrease-maxheight-expected.txt: Renamed from LayoutTests/fast/regions/autoheight-region-decrease-maxheight-expected.txt.
  • fast/regions/auto-size/autoheight-region-decrease-maxheight.html: Renamed from LayoutTests/fast/regions/autoheight-region-decrease-maxheight.html.
  • fast/regions/auto-size/autoheight-region-decrease-minheight-expected.txt: Renamed from LayoutTests/fast/regions/autoheight-region-decrease-minheight-expected.txt.
  • fast/regions/auto-size/autoheight-region-decrease-minheight.html: Renamed from LayoutTests/fast/regions/autoheight-region-decrease-minheight.html.
  • fast/regions/auto-size/autoheight-region-decrease-width-expected.txt: Renamed from LayoutTests/fast/regions/autoheight-region-decrease-width-expected.txt.
  • fast/regions/auto-size/autoheight-region-decrease-width.html: Renamed from LayoutTests/fast/regions/autoheight-region-decrease-width.html.
  • fast/regions/auto-size/autoheight-region-increase-maxheight-expected.txt: Renamed from LayoutTests/fast/regions/autoheight-region-increase-maxheight-expected.txt.
  • fast/regions/auto-size/autoheight-region-increase-maxheight.html: Renamed from LayoutTests/fast/regions/autoheight-region-increase-maxheight.html.
  • fast/regions/auto-size/autoheight-region-increase-minheight-expected.txt: Renamed from LayoutTests/fast/regions/autoheight-region-increase-minheight-expected.txt.
  • fast/regions/auto-size/autoheight-region-increase-minheight.html: Renamed from LayoutTests/fast/regions/autoheight-region-increase-minheight.html.
  • fast/regions/auto-size/autoheight-region-increase-width-expected.txt: Renamed from LayoutTests/fast/regions/autoheight-region-increase-width-expected.txt.
  • fast/regions/auto-size/autoheight-region-increase-width.html: Renamed from LayoutTests/fast/regions/autoheight-region-increase-width.html.
  • fast/regions/auto-size/autoheight-region-new-maxheight-expected.txt: Renamed from LayoutTests/fast/regions/autoheight-region-new-maxheight-expected.txt.
  • fast/regions/auto-size/autoheight-region-new-maxheight-vrl-expected.txt: Renamed from LayoutTests/fast/regions/autoheight-region-new-maxheight-vrl-expected.txt.
  • fast/regions/auto-size/autoheight-region-new-maxheight-vrl.html: Renamed from LayoutTests/fast/regions/autoheight-region-new-maxheight-vrl.html.
  • fast/regions/auto-size/autoheight-region-new-maxheight.html: Renamed from LayoutTests/fast/regions/autoheight-region-new-maxheight.html.
  • fast/regions/auto-size/autoheight-region-new-minheight-expected.txt: Renamed from LayoutTests/fast/regions/autoheight-region-new-minheight-expected.txt.
  • fast/regions/auto-size/autoheight-region-new-minheight-vlr-expected.txt: Renamed from LayoutTests/fast/regions/autoheight-region-new-minheight-vlr-expected.txt.
  • fast/regions/auto-size/autoheight-region-new-minheight-vlr.html: Renamed from LayoutTests/fast/regions/autoheight-region-new-minheight-vlr.html.
  • fast/regions/auto-size/autoheight-region-new-minheight.html: Renamed from LayoutTests/fast/regions/autoheight-region-new-minheight.html.
  • fast/regions/auto-size/autoheight-region-remove-maxheight-expected.txt: Renamed from LayoutTests/fast/regions/autoheight-region-remove-maxheight-expected.txt.
  • fast/regions/auto-size/autoheight-region-remove-maxheight.html: Renamed from LayoutTests/fast/regions/autoheight-region-remove-maxheight.html.
  • fast/regions/auto-size/autoheight-region-remove-minheight-expected.txt: Renamed from LayoutTests/fast/regions/autoheight-region-remove-minheight-expected.txt.
  • fast/regions/auto-size/autoheight-region-remove-minheight.html: Renamed from LayoutTests/fast/regions/autoheight-region-remove-minheight.html.
  • fast/regions/auto-size/autoheight-regionlayoutupdate-multipletimes-expected.txt: Renamed from LayoutTests/fast/regions/autoheight-regionlayoutupdate-multipletimes-expected.txt.
  • fast/regions/auto-size/autoheight-regionlayoutupdate-multipletimes.html: Renamed from LayoutTests/fast/regions/autoheight-regionlayoutupdate-multipletimes.html.
  • fast/regions/auto-size/autoheight-regions-mark-expected.txt: Renamed from LayoutTests/fast/regions/autoheight-regions-mark-expected.txt.
  • fast/regions/auto-size/autoheight-regions-mark.html: Renamed from LayoutTests/fast/regions/autoheight-regions-mark.html.
  • fast/regions/auto-size/autoheight-secondregion-breakoutside-expected.html: Renamed from LayoutTests/fast/regions/autoheight-secondregion-breakoutside-expected.html.
  • fast/regions/auto-size/autoheight-secondregion-breakoutside.html: Renamed from LayoutTests/fast/regions/autoheight-secondregion-breakoutside.html.
  • fast/regions/auto-size/autoheight-secondregion-expected.html: Renamed from LayoutTests/fast/regions/autoheight-secondregion-expected.html.
  • fast/regions/auto-size/autoheight-secondregion.html: Renamed from LayoutTests/fast/regions/autoheight-secondregion.html.
  • fast/regions/auto-size/autoheight-singleregion-breakafteralways-expected.html: Renamed from LayoutTests/fast/regions/autoheight-singleregion-breakafteralways-expected.html.
  • fast/regions/auto-size/autoheight-singleregion-breakafteralways-maxheight-expected.html: Renamed from LayoutTests/fast/regions/autoheight-singleregion-breakafteralways-maxheight-expected.html.
  • fast/regions/auto-size/autoheight-singleregion-breakafteralways-maxheight.html: Renamed from LayoutTests/fast/regions/autoheight-singleregion-breakafteralways-maxheight.html.
  • fast/regions/auto-size/autoheight-singleregion-breakafteralways.html: Renamed from LayoutTests/fast/regions/autoheight-singleregion-breakafteralways.html.
  • fast/regions/auto-size/autoheight-singleregion-breakaftermargin-expected.html: Renamed from LayoutTests/fast/regions/autoheight-singleregion-breakaftermargin-expected.html.
  • fast/regions/auto-size/autoheight-singleregion-breakaftermargin.html: Renamed from LayoutTests/fast/regions/autoheight-singleregion-breakaftermargin.html.
  • fast/regions/auto-size/autoheight-singleregion-breakbeforealways-expected.html: Renamed from LayoutTests/fast/regions/autoheight-singleregion-breakbeforealways-expected.html.
  • fast/regions/auto-size/autoheight-singleregion-breakbeforealways-minheight-expected.html: Renamed from LayoutTests/fast/regions/autoheight-singleregion-breakbeforealways-minheight-expected.html.
  • fast/regions/auto-size/autoheight-singleregion-breakbeforealways-minheight.html: Renamed from LayoutTests/fast/regions/autoheight-singleregion-breakbeforealways-minheight.html.
  • fast/regions/auto-size/autoheight-singleregion-breakbeforealways.html: Renamed from LayoutTests/fast/regions/autoheight-singleregion-breakbeforealways.html.
  • fast/regions/auto-size/autoheight-singleregion-multiplebreaks-expected.html: Renamed from LayoutTests/fast/regions/autoheight-singleregion-multiplebreaks-expected.html.
  • fast/regions/auto-size/autoheight-singleregion-multiplebreaks.html: Renamed from LayoutTests/fast/regions/autoheight-singleregion-multiplebreaks.html.
  • fast/regions/auto-size/autoheight-singleregion-overflowauto-breaksignored-expected.html: Renamed from LayoutTests/fast/regions/autoheight-singleregion-overflowauto-breaksignored-expected.html.
  • fast/regions/auto-size/autoheight-singleregion-overflowauto-breaksignored.html: Renamed from LayoutTests/fast/regions/autoheight-singleregion-overflowauto-breaksignored.html.
  • fast/regions/auto-size/autoheight-singleregion-overflowauto-expected.html: Renamed from LayoutTests/fast/regions/autoheight-singleregion-overflowauto-expected.html.
  • fast/regions/auto-size/autoheight-singleregion-overflowauto.html: Renamed from LayoutTests/fast/regions/autoheight-singleregion-overflowauto.html.
  • fast/regions/auto-size/autoheight-two-pass-layout-complex-001-expected.txt: Renamed from LayoutTests/fast/regions/autoheight-two-pass-layout-complex-001-expected.txt.
  • fast/regions/auto-size/autoheight-two-pass-layout-complex-001.html: Renamed from LayoutTests/fast/regions/autoheight-two-pass-layout-complex-001.html.
  • fast/regions/auto-size/autoheight-two-pass-layout-complex-002-expected.txt: Renamed from LayoutTests/fast/regions/autoheight-two-pass-layout-complex-002-expected.txt.
  • fast/regions/auto-size/autoheight-two-pass-layout-complex-002.html: Renamed from LayoutTests/fast/regions/autoheight-two-pass-layout-complex-002.html.
  • fast/regions/auto-size/autoheight-vertical-lr-expected.html: Renamed from LayoutTests/fast/regions/autoheight-vertical-lr-expected.html.
  • fast/regions/auto-size/autoheight-vertical-lr.html: Renamed from LayoutTests/fast/regions/autoheight-vertical-lr.html.
  • fast/regions/auto-size/autoheight-vertical-rl-expected.html: Renamed from LayoutTests/fast/regions/autoheight-vertical-rl-expected.html.
  • fast/regions/auto-size/autoheight-vertical-rl.html: Renamed from LayoutTests/fast/regions/autoheight-vertical-rl.html.
  • platform/efl/TestExpectations:
  • platform/gtk/TestExpectations:
3:40 PM Changeset in webkit [157141] by ggaren@apple.com
  • 4 edits in trunk/Websites/webkit.org

Refined power management in SunSpider 1.0.2
​https://bugs.webkit.org/show_bug.cgi?id=122525

Reviewed by Gavin Barraclough.

Sync'd sunspider-1.0.2 to PerformanceTests/SunSpider/hosted/sunspider-1.0.2.

  • perf/sunspider-1.0.2/sunspider-1.0.2/driver.html:
  • perf/sunspider-1.0.2/sunspider-1.0.2/sunspider-test-contents.js:
  • perf/sunspider-1.0.2/sunspider.css:

(#frameparent):
(#countdown):

3:40 PM Changeset in webkit [157140] by ggaren@apple.com
  • 4 edits in trunk/PerformanceTests/SunSpider

Refined power management in SunSpider 1.0.2
​https://bugs.webkit.org/show_bug.cgi?id=122525

Reviewed by Gavin Barraclough.

  • hosted/sunspider.css:

(#frameparent):
(#countdown):

  • resources/TEMPLATE.html: Provide an explicit countdown element in the

parent, instead of hacking it into each child frame.

  • resources/driver-TEMPLATE.html: Be more explicit about how we respond

to power management. Just moving setTimeout(0) inside the loop is a
little too subtle, and may not work across browsers.

Also, leave a little extra time after changing the countdown graphic,
to give the browser some time to paint.

2:57 PM Changeset in webkit [157139] by dino@apple.com
  • 2 edits in trunk/LayoutTests

Unreviewed followup to ​https://bugs.webkit.org/show_bug.cgi?id=122523
This test now passes, so updating expected file.

  • fast/canvas/webgl/webgl-unprefixed-context-id-expected.txt:
2:17 PM Changeset in webkit [157138] by dino@apple.com
  • 4 edits
    2 adds in trunk

Support unprefixed WebGL context creation
​https://bugs.webkit.org/show_bug.cgi?id=122523
<rdar://problem/15179463>

Reviewed by Simon Fraser.

Source/WebCore:

Add support for the unprefixed "webgl" canvas context type.

Test: fast/canvas/webgl/unprefixed-context.html

  • html/HTMLCanvasElement.cpp:

(WebCore::HTMLCanvasElement::is3dType):

  • inspector/InjectedScriptCanvasModuleSource.js:

LayoutTests:

Tests that we can create a context without the
"experimental-" prefix. Meanwhile, most of the other
WebGL tests should start to use the unprefixed
form automatically (the helper script attempts
"webgl" first).

  • fast/canvas/webgl/unprefixed-context-expected.txt: Added.
  • fast/canvas/webgl/unprefixed-context.html: Added.
2:05 PM Changeset in webkit [157137] by andersca@apple.com
  • 15 edits in trunk/Source/WebKit2

WebProcess crash on SAP WebCycle web app
​https://bugs.webkit.org/show_bug.cgi?id=122520
<rdar://problem/15030605>

Reviewed by Darin Adler.

Stop trying to use RunLoop to manage the top-level run loop and just have the child process and their
delegate subclasses start and stop the run loops. This fixes a bug with showModalDialog where we would
unintentionally call -[NSApp stop] when closing a modal dialog.

This also lets us move all knowledge of NSApplication from RunLoop. (Both the web process and plug-in process
need to use -[NSApp run] and -[NSApp stop:]).

  • PluginProcess/EntryPoint/mac/LegacyProcess/PluginProcessMain.mm:

(WebKit::PluginProcessMainDelegate::doPreInitializationWork):
Remove call to RunLoop::setUseApplicationRunLoopOnMainRunLoop. Add a startRunLoop override that calls
-[NSApp run].

  • PluginProcess/EntryPoint/mac/XPCService/PluginServiceEntryPoint.mm:

(PluginServiceInitializer):
Remove call to RunLoop::setUseApplicationRunLoopOnMainRunLoop.

  • PluginProcess/PluginProcess.h:

Add stopRunLoop() override on Mac.

  • PluginProcess/mac/PluginProcessMac.mm:

(WebKit::PluginProcess::stopRunLoop):
Call -[NSApp stop:] and tickle the event system.

  • Shared/ChildProcess.cpp:

(WebKit::ChildProcess::stopRunLoop):
Add default implementation that just calls RunLoop::main()->stop().

(WebKit::ChildProcess::terminate):
Call stopRunLoop().

  • Shared/ChildProcess.h:
  • Shared/EntryPointUtilities/mac/LegacyProcess/ChildProcessEntryPoint.h:

Add startRunLoop member function and call it instead of RunLoop::run().

  • Shared/EntryPointUtilities/mac/LegacyProcess/ChildProcessEntryPoint.mm:

(WebKit::ChildProcessMainDelegate::startRunLoop):
Call RunLoop::run().

  • WebProcess/EntryPoint/mac/LegacyProcess/WebContentProcessMain.mm:

(WebKit::WebContentProcessMainDelegate::doPreInitializationWork):
Remove call to RunLoop::setUseApplicationRunLoopOnMainRunLoop.
Add startRunLoop override that calls -[NSApp run].

  • WebProcess/EntryPoint/mac/XPCService/WebContentServiceEntryPoint.mm:

(WebContentServiceInitializer):
Remove call to RunLoop::setUseApplicationRunLoopOnMainRunLoop.

  • WebProcess/WebProcess.cpp:

(WebKit::WebProcess::didClose):
Call stopRunLoop().

  • WebProcess/WebProcess.h:

Add stopRunLoop override.

  • WebProcess/mac/WebProcessMac.mm:

(WebKit::WebProcess::stopRunLoop):
Call -[NSApp stop:] and tickle the event system.

1:58 PM Changeset in webkit [157136] by rniwa@webkit.org
  • 18 edits in trunk/Source/WebCore

Rename EventDispatchMediator::dispatchEvent to mediateAndDispatchEvent
​https://bugs.webkit.org/show_bug.cgi?id=122522

Reviewed by Darin Adler.

Renamed EventDispatchMediator::dispatchEvent to EventDispatchMediator::mediateAndDispatchEvent
to make the future refactoring easier.

  • Modules/indieui/UIRequestEvent.cpp:

(WebCore::UIRequestEventDispatchMediator::mediateAndDispatchEvent):

  • Modules/indieui/UIRequestEvent.h:
  • dom/EventDispatchMediator.cpp:

(WebCore::EventDispatchMediator::mediateAndDispatchEvent):

  • dom/EventDispatchMediator.h:
  • dom/EventDispatcher.cpp:

(WebCore::EventDispatcher::dispatchEvent):

  • dom/FocusEvent.cpp:

(WebCore::FocusEventDispatchMediator::mediateAndDispatchEvent):
(WebCore::BlurEventDispatchMediator::mediateAndDispatchEvent):
(WebCore::FocusInEventDispatchMediator::mediateAndDispatchEvent):
(WebCore::FocusOutEventDispatchMediator::mediateAndDispatchEvent):

  • dom/FocusEvent.h:
  • dom/GestureEvent.cpp:

(WebCore::GestureEventDispatchMediator::mediateAndDispatchEvent):

  • dom/GestureEvent.h:
  • dom/KeyboardEvent.cpp:

(WebCore::KeyboardEventDispatchMediator::mediateAndDispatchEvent):

  • dom/KeyboardEvent.h:
  • dom/MouseEvent.cpp:

(WebCore::MouseEventDispatchMediator::mediateAndDispatchEvent):

  • dom/MouseEvent.h:
  • dom/TouchEvent.cpp:

(WebCore::TouchEventDispatchMediator::mediateAndDispatchEvent):

  • dom/TouchEvent.h:
  • dom/WheelEvent.cpp:

(WebCore::WheelEventDispatchMediator::mediateAndDispatchEvent):

  • dom/WheelEvent.h:
1:42 PM Changeset in webkit [157135] by Brent Fulgham
  • 2 edits in trunk/Source/WebCore

[MathML] Use of floating point floor/ceil on LayoutUnits seems wrong
​https://bugs.webkit.org/show_bug.cgi?id=122367

Reviewed by Darin Adler.

  • rendering/mathml/RenderMathMLOperator.cpp:

(WebCore::RenderMathMLOperator::paintCharacter): Use member methods to compute
correct floor/ceil values for LayoutUnits.

1:38 PM Changeset in webkit [157134] by kov@webkit.org
  • 2 edits in trunk/LayoutTests

Skip a new WebGL test along with the others.

  • platform/gtk/TestExpectations:
1:34 PM Changeset in webkit [157133] by dino@apple.com
  • 3 edits
    2 adds in trunk

Video -> pixel buffer output should not manage color spaces
​https://bugs.webkit.org/show_bug.cgi?id=122486

Reviewed by Tim Horton.

Source/WebCore:

On Mac, when we draw a video into a canvas or WebGL, the
output of the video should not be color managed. Instead, leave it
as DeviceRGB so that it matches other colors in those elements.
We'll hopefully solve this properly soon - making everything sRGB.

Test: media/video-canvas-drawing-output.html

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

(WebCore::MediaPlayerPrivateAVFoundationObjC::createImageForTimeInRect): Pass on
a CGImageRef via CGImageCreateCopyWithColorSpace rather than the raw data.
(WebCore::MediaPlayerPrivateAVFoundationObjC::paintWithVideoOutput): Tell CoreImage
to allocate a pixel buffer that is unmanaged.

LayoutTests:

  • media/video-canvas-drawing-output-expected.txt: Added.
  • media/video-canvas-drawing-output.html: Added.
1:30 PM Changeset in webkit [157132] by lauro.neto@openbossa.org
  • 2 edits in trunk/Source/WebCore

[MediaStream] Fix segfault in initializeMockSources()
​https://bugs.webkit.org/show_bug.cgi?id=122517

Reviewed by Darin Adler.

Calling release() and a member function on the same RefPtr
as arguments to a function can lead to segfaults if the release
call is evaluated first.

  • platform/mock/MockMediaStreamCenter.cpp:

(WebCore::initializeMockSources):

1:26 PM Changeset in webkit [157131] by mark.lam@apple.com
  • 3 edits in trunk/Tools

Fix ruby LoadError when running run-javascriptcore-tests.
​https://bugs.webkit.org/show_bug.cgi?id=122519.

Reviewed by Mark Hahnenberg.

The scripts needed to explicitly require 'rubygems' for older versions of ruby
(versions before 1.9).

  • Scripts/jsc-stress-test-helpers/profiler-test-helper:
  • Scripts/run-jsc-stress-tests:
1:25 PM Changeset in webkit [157130] by kov@webkit.org
  • 2 edits in trunk/LayoutTests

Unreviewed gardening.

  • platform/gtk/TestExpectations: fast/backgrounds/background-opaque-images-over-color.html

failing.

1:03 PM Changeset in webkit [157129] by mihnea@adobe.com
  • 4 edits in trunk

[CSSRegions] Regions with overflow: hidden should paint over positioned sibling
​https://bugs.webkit.org/show_bug.cgi?id=122389

Reviewed by Alexandru Chiculita.

Source/WebCore:

As a follow-up from ​https://bugs.webkit.org/show_bug.cgi?id=122321, a region should not
be a normal flow layer when it has overflow clip. Change existing test by adding overflow:hidden
to the region element.

  • rendering/RenderLayer.cpp:

(WebCore::RenderLayer::shouldBeNormalFlowOnly):

LayoutTests:

Add overflow: hidden to the region to test that it does not move to the normal flow layers.

  • fast/regions/layers/region-normalflow-stacking-context.html:
12:58 PM Changeset in webkit [157128] by kov@webkit.org
  • 4 edits in trunk/LayoutTests

Unreviewed gardening, these tests had a couple pixels shift at some point,
needing a new baseline.

  • platform/gtk/fast/hidpi/clip-text-in-hidpi-expected.txt:
  • platform/gtk/fast/hidpi/image-set-background-dynamic-expected.txt:
  • platform/gtk/fast/hidpi/image-set-border-image-dynamic-expected.txt:
12:44 PM Changeset in webkit [157127] by rniwa@webkit.org
  • 2 edits in trunk/Source/WebCore

Simplify the loop in EventRetargeter::calculateEventPath
​https://bugs.webkit.org/show_bug.cgi?id=122500

Reviewed by Antti Koivisto.

Replaced the convoluted loop by two nested loops to make semantics clear.

  • dom/EventRetargeter.cpp:

(WebCore::EventRetargeter::calculateEventPath):

12:27 PM Changeset in webkit [157126] by g.czajkowski@samsung.com
  • 2 edits in trunk/LayoutTests

[Mac] Unskip editing/spelling/spelling-changed-text.html
​https://bugs.webkit.org/show_bug.cgi?id=122502

Unreviewed gardening.

  • platform/mac/TestExpectations:

spelling-changed.html is passing after r156901.

12:13 PM Changeset in webkit [157125] by kov@webkit.org
  • 2 edits in trunk/LayoutTests

Unreviewed gardening. A couple of media-query tests rely on 3d transforms,
so skip them along with the 3d transforms tests.

  • platform/gtk/TestExpectations:
11:25 AM Changeset in webkit [157124] by ChangSeok Oh
  • 2 edits in trunk/Source/WebCore

[GTK] Leak: GdkEvent retured by gtk_get_current_event is not freed.
​https://bugs.webkit.org/show_bug.cgi?id=122506

Reviewed by Gustavo Noronha Silva.

Gdk event returned by gtk_get_current_event should be freed after using it.

No functionality changed.

  • platform/gtk/PopupMenuGtk.cpp:

(WebCore::PopupMenuGtk::show):

11:03 AM Changeset in webkit [157123] by rniwa@webkit.org
  • 6 edits in trunk/Source/WebCore

Use references in EventRetargeter::calculateEventPath and EventRetargeter::eventTargetRespectingTargetRules
​https://bugs.webkit.org/show_bug.cgi?id=122494

Reviewed by Antti Koivisto.

Use referenecs instead of pointers in ventRetargeter's calculateEventPath and eventTargetRespectingTargetRules.
Also replace Vector<EventTarget*, 32> targetStack by a single pointer since we only append an item to
the vector when it's empty and we always use the last item.

  • dom/EventDispatcher.cpp:

(WebCore::EventDispatcher::EventDispatcher):
(WebCore::EventDispatcher::dispatchScopedEvent):
(WebCore::EventDispatcher::dispatch):
(WebCore::EventDispatcher::dispatchEventPostProcess):

  • dom/EventDispatcher.h:
  • dom/EventRetargeter.cpp:

(WebCore::EventRetargeter::calculateEventPath):

  • dom/EventRetargeter.h:

(WebCore::EventRetargeter::eventTargetRespectingTargetRules):

  • dom/Node.cpp:

(WebCore::Node::dispatchScopedEventDispatchMediator):

10:59 AM Changeset in webkit [157122] by Christophe Dumez
  • 2 edits in trunk/Tools

Unreviewed, add my new email to contributors.json.

Patch by Christophe Dumez <Christophe Dumez> on 2013-10-08

  • Scripts/webkitpy/common/config/contributors.json:
10:33 AM Changeset in webkit [157121] by mihnea@adobe.com
  • 8 edits
    2 adds in trunk

[CSSRegions] Computed z-Index should return 0 instead of auto for a region
​https://bugs.webkit.org/show_bug.cgi?id=122405

Reviewed by David Hyatt.

Source/WebCore:

Test: fast/regions/layers/region-zIndex-computedStyle.html

Adjust the z-index value for a region. Also, since a box with a z-Index
that is non-auto receives a layer, i removed the requiresLayer override
in RenderRegion and RenderMultiColumnSet.

  • css/StyleResolver.cpp:

(WebCore::StyleResolver::adjustRenderStyle):

  • rendering/RenderMultiColumnSet.cpp:
  • rendering/RenderMultiColumnSet.h:
  • rendering/RenderRegion.cpp:
  • rendering/RenderRegion.h:
  • rendering/style/RenderStyle.h:

LayoutTests:

  • fast/regions/layers/region-zIndex-computedStyle-expected.txt: Added.
  • fast/regions/layers/region-zIndex-computedStyle.html: Added.
10:33 AM Changeset in webkit [157120] by abucur@adobe.com
  • 5 edits
    2 adds in trunk

[CSS Regions] Widows don't work if the first line in a region is aligned with the top of the region
​https://bugs.webkit.org/show_bug.cgi?id=122450

Reviewed by David Hyatt.

Source/WebCore:

The patch adds a new function updateRegionForLine that updates the containing region for a line and the
flag used to determine if it's the first line in that region. This is necessary because the code in
adjustLinePositionForPagination is not sufficient to determine when an unforced break occurs and the line
being positioned is the first one in the fragmentation container or not.

Test: fast/regions/regions-widows-float-top-aligned.html

  • rendering/RenderBlockFlow.cpp:

(WebCore::RenderBlockFlow::adjustLinePositionForPagination): Added FIXME comment.

  • rendering/RenderBlockFlow.h:
  • rendering/RenderBlockLineLayout.cpp:

(WebCore::RenderBlock::layoutRunsAndFloatsInRange):
(WebCore::RenderBlock::linkToEndLineIfNeeded):
(WebCore::RenderBlock::determineStartPosition):
(WebCore::RenderBlockFlow::updateRegionForLine): New function used to update the containing region and
the first line flag.

LayoutTests:

Simple test veryfing that widows are correctly computed when there is a float affecting the lines
and when they are vertically aligned with the regions height.

  • fast/regions/regions-widows-float-top-aligned-expected.html: Added.
  • fast/regions/regions-widows-float-top-aligned.html: Added.
10:25 AM Changeset in webkit [157119] by mhahnenberg@apple.com
  • 3 edits in trunk/Source/JavaScriptCore

JSManagedValue should be able to store non-object JSValues
​https://bugs.webkit.org/show_bug.cgi?id=122351

Reviewed by Oliver Hunt.

We decided not to support this because we thought it didn't make sense to have a
"weak" JSValue that wasn't an object.

Our general thought process was if you have a JSObject-ObjC object pair (i.e. an
Obj-C object that you exported to JavaScript-land), it makes more sense to store
a non-object JSValue on the JavaScript-land version of the object rather than as
an ivar in the Objective-C object.

In retrospect, this may not have been a good decision at least w.r.t. consistency
in client code. If you're storing a bag of JSValues off an Obj-C object, you'd
like to store all of them either in ObjC-land or JavaScript-land, but doing some
in one and some in the other doesn't sound too good. Also, what if the object you
want to hang these values off of doesn't have a corresponding object in JavaScript-
land in which to store them?

The solution is to fix JSManagedValue to be able to reference non-object JSValues.
Right now, all JSManagedValues contain a Weak<JSObject>. We'll change this so that
they can contain either a non-cell JSValue or a JSObject*, along with a weak
reference to the JSGlobalObject for reconstructing a JSValue later on.

  • API/JSManagedValue.mm:

(PrimitiveOrObject::PrimitiveOrObject):
(PrimitiveOrObject::~PrimitiveOrObject):
(PrimitiveOrObject::clear):
(PrimitiveOrObject::isClear):
(PrimitiveOrObject::isSet):
(PrimitiveOrObject::isPrimitive):
(PrimitiveOrObject::isObject):
(PrimitiveOrObject::setPrimitive):
(PrimitiveOrObject::setObject):
(PrimitiveOrObject::object):
(PrimitiveOrObject::primitive):
(-[JSManagedValue initWithValue:]):
(-[JSManagedValue value]):
(-[JSManagedValue disconnectValue]):

10:16 AM Changeset in webkit [157118] by kov@webkit.org
  • 2 edits in trunk/LayoutTests

Unreviewed gardening. Adding ImageOnlyFailure expectations to:

  • fast/regions/selection/selecting-text-ignoring-region-horiz-bt.html
  • fast/regions/selection/selecting-text-ignoring-region-vert-lr.html
  • fast/regions/selection/selecting-text-in-2-regions-horiz-bt.html
  • fast/regions/selection/selecting-text-in-2-regions-vert-lr.html
  • platform/gtk/TestExpectations:
10:12 AM Changeset in webkit [157117] by ap@apple.com
  • 2 edits in trunk/LayoutTests

Layout Test inspector/styles/add-new-rule-with-style-after-body.html is flaky
​https://bugs.webkit.org/show_bug.cgi?id=122510

  • platform/mac/TestExpectations: Marking as such.
10:05 AM Changeset in webkit [157116] by ap@apple.com
  • 2 edits in trunk/LayoutTests

Layout Test inspector/debugger/debugger-activation-crash2.html is flaky
​https://bugs.webkit.org/show_bug.cgi?id=122509

  • platform/mac/TestExpectations: Marking as such.
9:52 AM Changeset in webkit [157115] by ap@apple.com
  • 2 edits in trunk/LayoutTests

Layout Test compositing/video/video-reflection.html is flaky
​https://bugs.webkit.org/show_bug.cgi?id=122508

  • platform/mac/TestExpectations: Marking as such.
9:47 AM Changeset in webkit [157114] by ap@apple.com
  • 2 edits in trunk/LayoutTests

Layout Test media/track/track-cue-rendering.html is flaky
​https://bugs.webkit.org/show_bug.cgi?id=122507

  • platform/mac/TestExpectations: Marking as such.
9:43 AM Changeset in webkit [157113] by kov@webkit.org
  • 1 edit
    1 delete in trunk/LayoutTests

Unreviewed, rolling out r157080.
​http://trac.webkit.org/changeset/157080

Not a proper baseline

  • platform/gtk/fast/regions/selection/selecting-text-ignoring-region-horiz-bt-expected.png: Removed.
  • platform/gtk/fast/regions/selection/selecting-text-ignoring-region-vert-lr-expected.png: Removed.
  • platform/gtk/fast/regions/selection/selecting-text-in-2-regions-horiz-bt-expected.png: Removed.
  • platform/gtk/fast/regions/selection/selecting-text-in-2-regions-vert-lr-expected.png: Removed.
9:42 AM Changeset in webkit [157112] by Brent Fulgham
  • 2 edits in trunk/Tools

Unreviewed build correction.

  • Scripts/webkitpy/port/win.py:

(WinPort.write_registry_value): Use string representation of error message to avoid
format conversion problem.

9:38 AM Changeset in webkit [157111] by ap@apple.com
  • 2 edits in trunk/LayoutTests

Flaky Test: media/video-layer-crash.html
​https://bugs.webkit.org/show_bug.cgi?id=114744

  • platform/mac/TestExpectations: Marking as such.
9:36 AM Changeset in webkit [157110] by commit-queue@webkit.org
  • 3 edits in trunk/Source/WebCore

Get EFL and GTK compiling with media stream enabled
​https://bugs.webkit.org/show_bug.cgi?id=122505

Patch by Thiago de Barros Lacerda <​thiago.lacerda@openbossa.org> on 2013-10-08
Reviewed by Philippe Normand.

After r157068

No new tests needed.

  • GNUmakefile.list.am:
  • UseJSC.cmake:
9:31 AM Changeset in webkit [157109] by Darin Adler
  • 3 edits in trunk/Source/WebKit/mac

WebNetscapePluginView should no longer use deleteAllValues
​https://bugs.webkit.org/show_bug.cgi?id=122495

Reviewed by Anders Carlsson.

  • Plugins/WebNetscapePluginView.h: Use unique_ptr for both the pointer to

the timers hash table, and for the values in the hash table.

  • Plugins/WebNetscapePluginView.mm:

(-[WebNetscapePluginView stopTimers]): Rewrote loop so it works with unique_ptr.
(-[WebNetscapePluginView startTimers]): Ditto.
(-[WebNetscapePluginView fini]): Removed unneeded code to delete timers.
(-[WebNetscapePluginView scheduleTimerWithInterval:repeat:timerFunc:]): Make
the timers hash table with make-unique. Fixed the loop that chooses the timer
ID so it won't ever try to use a deleted value, although that never happens
in practice. Use HashMap::add instead of using both contains and set.
Rearranged code so we don't need to make a raw pointer for the timer.
(-[WebNetscapePluginView unscheduleTimer:]): Use remove instead of take/delete,
since remove takes care of the deletion automatically since we are using the
unique_ptr class.

8:56 AM Changeset in webkit [157108] by commit-queue@webkit.org
  • 3 edits
    2 adds in trunk

Composited layer that painted into composited ancestor is not repainted after moving
​https://bugs.webkit.org/show_bug.cgi?id=114655

Patch by Hurnjoo Lee <​hurnjoo.lee@samsung.com> on 2013-10-08
Reviewed by Simon Fraser.

Source/WebCore:

Test: compositing/repaint/absolute-painted-into-composited-ancestor.html

Composited layers are currently skipping repainting after layout,
if the repaint status is not NeedsFullRepaint. However, in case layers
were painted into composited ancestor, they need to repaint.

  • rendering/RenderLayer.cpp:

(WebCore::RenderLayer::shouldRepaintAfterLayout):

LayoutTests:

Add a text repaint test to check the repainting of composited layer
that painted into composited ancestor.

  • compositing/repaint/absolute-painted-into-composited-ancestor-expected.txt: Added.
  • compositing/repaint/absolute-painted-into-composited-ancestor.html: Added.
8:51 AM Changeset in webkit [157107] by Brent Fulgham
  • 3 edits in trunk/Tools

[Win] Add checks for proper build and test environment
​https://bugs.webkit.org/show_bug.cgi?id=122466

Reviewed by Darin Adler.

This patch makes two changes:

  1. Update webkitdirs.pm to check for some common misconfigurations and provide helpful warnings or error messages so that people know what to fix to get things working.
  2. Modify the win.py port to handle the 32-bit and 64-bit test environments, as well as making sure the various environment settings are properly configured for our test environment.
  • Scripts/webkitdirs.pm:

(checkInstalledTools): Add checks for some common Cygwin configuration problems that have
caused problems in the past.
(setupAppleWinEnv):

  • Scripts/webkitpy/port/win.py:

(WinPort):
(WinPort.read_registry_string): Parameterize for run architecture and registry root.
(WinPort.write_registry_value): Parameterize registry writing by data type (string versus
DWORD, etc.)
(WinPort.write_registry_string): Revise to call new generic registry write function.
(WinPort.setup_crash_log_saving): Revise for new API
(WinPort.restore_crash_log_saving): Ditto
(WinPort.prevent_error_dialogs): Add new methods to prevent Windows dialogs that block our test
servers from running.
(WinPort.allow_error_dialogs): Ditto.
(WinPort.setup_test_run): Call new API to turn off user input dialogs on program failure.
(WinPort.clean_up_test_run): Call new API to reactivatea user input dialogs after test run finished.

8:45 AM Changeset in webkit [157106] by Philippe Normand
  • 2 edits in trunk/Tools

[GStreamer] enable v4l2 plugin
​https://bugs.webkit.org/show_bug.cgi?id=122503

Reviewed by Martin Robinson.

  • gtk/jhbuild.modules: Enable the plugin, it now builds fine since

we updated our GStreamer version to 1.0.8.

8:28 AM Changeset in webkit [157105] by commit-queue@webkit.org
  • 3 edits in trunk/Source/WebKit2

Unreviewed, rolling out r157090.
​http://trac.webkit.org/changeset/157090
​https://bugs.webkit.org/show_bug.cgi?id=122504

Caused crashes in many plug-in tests (Requested by ap on
#webkit).

  • Shared/Plugins/NPRemoteObjectMap.cpp:

(WebKit::NPRemoteObjectMap::registerNPObject):
(WebKit::NPRemoteObjectMap::pluginDestroyed):

  • Shared/Plugins/NPRemoteObjectMap.h:
8:15 AM Changeset in webkit [157104] by commit-queue@webkit.org
  • 2 edits in trunk

[GTK] Re-enable MathML for release builds
​https://bugs.webkit.org/show_bug.cgi?id=122361

Patch by Martin Robinson <​mrobinson@igalia.com> on 2013-10-08
Reviewed by Darin Adler.

  • Source/autotools/SetupWebKitFeatures.m4: Enable MathML for release builds.
8:12 AM Changeset in webkit [157103] by Lucas Forschler
  • 5 edits in tags/Safari-538.1.3/Source

Versioning.

8:10 AM Changeset in webkit [157102] by Lucas Forschler
  • 1 copy in tags/Safari-538.1.3

New Tag.

7:56 AM Changeset in webkit [157101] by Darin Adler
  • 3 edits in trunk/Source/WebKit/mac

Eliminate use of deleteAllValues in ProxyInstance
​https://bugs.webkit.org/show_bug.cgi?id=122499

Reviewed by Anders Carlsson.

  • Plugins/Hosted/ProxyInstance.h: Use unique_ptr for map values in m_fields and m_methods.
  • Plugins/Hosted/ProxyInstance.mm:

(WebKit::ProxyInstance::~ProxyInstance): Removed calls to deleteAllValues.
(WebKit::ProxyInstance::methodNamed): Use make_unique and get to deal with map entries
that are unique_ptr. Also fixed a bug where this would do an expensive no-op if the map
already had an entry with a null in it. We would call the plug-in, but eventually
isNewEntry would be false so we would return null.
(WebKit::ProxyInstance::fieldNamed): Use make_unique and get to deal with map entries
that are unique_ptr.

7:39 AM Changeset in webkit [157100] by Alan Bujtas
  • 5 edits
    2 deletes in trunk

Unreviewed. Rollout of r156536. ~5% regression in inline layout performance.

Source/WebCore:

  • rendering/LineWidth.cpp:
  • rendering/LineWidth.h:

(WebCore::LineWidth::addUncommittedWidth):

  • rendering/RenderBlockLineLayout.cpp:

(WebCore::LineBreaker::nextSegmentBreak):

LayoutTests:

  • fast/css/unexpected-word-wrapping-with-non-empty-spans-expected.html: Removed.
  • fast/css/unexpected-word-wrapping-with-non-empty-spans.html: Removed.
5:57 AM Changeset in webkit [157099] by Michał Pakuła vel Rutka
  • 3 edits in trunk/LayoutTests

Unreviewed EFL gardening

Unskip tests that are already passing.

  • platform/efl-wk1/TestExpectations: Mark unskipped test as failing for WK1.
  • platform/efl/TestExpectations:
5:07 AM Changeset in webkit [157098] by gyuyoung.kim@samsung.com
  • 12 edits in trunk/Source/WebCore

Use toCSSFooValue() instead of using static_cast<const CSSFooValue*>
​https://bugs.webkit.org/show_bug.cgi?id=122491

Reviewed by Andreas Kling.

static_cast<const CSSFooValue*> wasn't cleanup. So, we need to use toCSSFooValue() for it.
To use toCSSValue() will help to detect bad type-cast.

Besides static_cast<const WebKitCSSFooValue*> helper functions need to be added to support this use.

No new tests, no behavior change.

  • css/CSSComputedStyleDeclaration.cpp:

(WebCore::ComputedStyleExtractor::propertyMatches):

  • css/CSSImageGeneratorValue.cpp:

(WebCore::CSSImageGeneratorValue::isFixedSize):
(WebCore::CSSImageGeneratorValue::isPending):
(WebCore::CSSImageGeneratorValue::knownToBeOpaque):

  • css/CSSValue.cpp:

(WebCore::CSSValue::isImplicitInitialValue):
(WebCore::CSSValue::addSubresourceStyleURLs):
(WebCore::CSSValue::hasFailedOrCanceledSubresources):
(WebCore::CSSValue::equals):
(WebCore::CSSValue::cssText):
(WebCore::CSSValue::serializeResolvingVariables):
(WebCore::CSSValue::cloneForCSSOM):

  • css/StylePropertySet.cpp:

(WebCore::StylePropertySet::PropertyReference::cssName):

  • css/WebKitCSSSVGDocumentValue.h:

(WebCore::toWebKitCSSSVGDocumentValue):

  • css/WebKitCSSTransformValue.h:

(WebCore::toWebKitCSSTransformValue):

4:39 AM Changeset in webkit [157097] by commit-queue@webkit.org
  • 2 edits in trunk/Source/WebCore

Get MediaStreamCenterGStreamer compiling again
​https://bugs.webkit.org/show_bug.cgi?id=122482

Patch by Thiago de Barros Lacerda <​thiago.lacerda@openbossa.org> on 2013-10-08
Reviewed by Eric Carlson.

r157068 was including a nonexistent header

No new tests needed.

  • platform/mediastream/gstreamer/MediaStreamCenterGStreamer.cpp:
3:55 AM Changeset in webkit [157096] by mario@webkit.org
  • 5 edits in trunk

[ATK] Expose state of aria-invalid as ATK_STATE_INVALID_ENTRY
​https://bugs.webkit.org/show_bug.cgi?id=122104

Reviewed by Chris Fleizach.

Source/WebCore:

Map invalid state of an AccessibilityObject to the proper AtkState.

  • accessibility/atk/WebKitAccessibleWrapperAtk.cpp:

(setAtkStateSetFromCoreObject): Add the mapping.

Tools:

Updated stringAttributeValue to double check that the exposure of
aria-invalid as a state is consistent with its exposure as a text
attribute as well, when it applies (implementations of AtkText).

  • DumpRenderTree/atk/AccessibilityUIElementAtk.cpp:

(AccessibilityUIElement::stringAttributeValue): Add double check.

  • WebKitTestRunner/InjectedBundle/atk/AccessibilityUIElementAtk.cpp:

(WTR::AccessibilityUIElement::stringAttributeValue): Ditto.

3:45 AM Changeset in webkit [157095] by mario@webkit.org
  • 2 edits in trunk/LayoutTests

Unreviewed GTK gardening. Name for tables are now properly showing
up when exposing information for the parent of a table's cell.

  • platform/gtk/accessibility/table-cells-expected.txt: Updated.
2:56 AM Changeset in webkit [157094] by g.czajkowski@samsung.com
  • 12 edits in trunk/LayoutTests

context-menu-suggestions-multiword-selection.html and context-menu-suggestions-subword-selection.html are flaky after r156901
​https://bugs.webkit.org/show_bug.cgi?id=122414

Reviewed by Ryosuke Niwa.

On Mac, first call of 'initSpellTest' from spelling/resources/util.js
often fails while checking spelling marker.
Use shouldBecomeDifferent instead of calling timers in the tests.
It should have been done at r156137 when shouldBecomeEqual/Different
was used in asynchronous spell checking tests.

  • editing/spelling/resources/util.js:

Use shouldBecomeDifferent instead of timers.

  • editing/spelling/context-menu-suggestions-multiword-selection-expected.txt:
  • editing/spelling/context-menu-suggestions-subword-selection-expected.txt:
  • editing/spelling/spelling-changed-text-expected.txt:
  • editing/spelling/spelling-double-clicked-word-expected.txt:
  • editing/spelling/spelling-exactly-selected-multiple-words-expected.txt:
  • editing/spelling/spelling-exactly-selected-word-expected.txt:
  • editing/spelling/spelling-should-select-multiple-words-expected.txt:
  • editing/spelling/spelling-with-punctuation-selection-expected.txt:
  • editing/spelling/spelling-with-whitespace-selection-expected.txt:

Rebaseline the tests which include util.js due to
shouldBecomeDifferent outputs more logs.

  • platform/mac/TestExpectations:

Now context-menu-suggestions-multiword-selection.html and
context-menu-suggestions-subword-selection.html are passing for Mac.

2:41 AM Changeset in webkit [157093] by commit-queue@webkit.org
  • 2 edits in trunk/Source/JavaScriptCore

JavaScriptCore fails to build
​https://bugs.webkit.org/show_bug.cgi?id=122440

Patch by Robert Plociennik <​r.plociennik@samsung.com> on 2013-10-08
Reviewed by Darin Adler.

Compilation fails in debug due to 'comparison of unsigned expression >= 0 is
always true'.

  • debugger/DebuggerCallFrame.cpp:

(JSC::DebuggerCallFrame::positionForCallFrame): Removed the offending ASSERTS.

2:01 AM Changeset in webkit [157092] by mihnea@adobe.com
  • 1 edit
    32 moves
    1 add in trunk/LayoutTests

[CSSRegions] Move autosize and autowidth tests into fast/regions/auto-size
​https://bugs.webkit.org/show_bug.cgi?id=122493

Reviewed by Andreas Kling.

Simple move, changed paths in 2 tests.

  • fast/regions/auto-size/autosize-abspos-anchoredregion-expected.txt: Renamed from LayoutTests/fast/regions/autosize-abspos-anchoredregion-expected.txt.
  • fast/regions/auto-size/autosize-abspos-anchoredregion-vertlr-expected.txt: Renamed from LayoutTests/fast/regions/autosize-abspos-anchoredregion-vertlr-expected.txt.
  • fast/regions/auto-size/autosize-abspos-anchoredregion-vertlr.html: Renamed from LayoutTests/fast/regions/autosize-abspos-anchoredregion-vertlr.html.
  • fast/regions/auto-size/autosize-abspos-anchoredregion.html: Renamed from LayoutTests/fast/regions/autosize-abspos-anchoredregion.html.
  • fast/regions/auto-size/autosize-region-container-resize-expected.html: Renamed from LayoutTests/fast/regions/autosize-region-container-resize-expected.html.
  • fast/regions/auto-size/autosize-region-container-resize.html: Renamed from LayoutTests/fast/regions/autosize-region-container-resize.html.
  • fast/regions/auto-size/autosize-region-window-resize-expected.html: Renamed from LayoutTests/fast/regions/autosize-region-window-resize-expected.html.
  • fast/regions/auto-size/autosize-region-window-resize.html: Renamed from LayoutTests/fast/regions/autosize-region-window-resize.html.
  • fast/regions/auto-size/autowidth-abspos-expected.html: Renamed from LayoutTests/fast/regions/autowidth-abspos-expected.html.
  • fast/regions/auto-size/autowidth-abspos-regionchain-expected.html: Renamed from LayoutTests/fast/regions/autowidth-abspos-regionchain-expected.html.
  • fast/regions/auto-size/autowidth-abspos-regionchain.html: Renamed from LayoutTests/fast/regions/autowidth-abspos-regionchain.html.
  • fast/regions/auto-size/autowidth-abspos.html: Renamed from LayoutTests/fast/regions/autowidth-abspos.html.
  • fast/regions/auto-size/autowidth-attachedinvalidregion-expected.txt: Renamed from LayoutTests/fast/regions/autowidth-attachedinvalidregion-expected.txt.
  • fast/regions/auto-size/autowidth-attachedinvalidregion.html: Renamed from LayoutTests/fast/regions/autowidth-attachedinvalidregion.html.
  • fast/regions/auto-size/autowidth-float-expected.html: Renamed from LayoutTests/fast/regions/autowidth-float-expected.html.
  • fast/regions/auto-size/autowidth-float.html: Renamed from LayoutTests/fast/regions/autowidth-float.html.
  • fast/regions/auto-size/autowidth-inlineblock-expected.html: Renamed from LayoutTests/fast/regions/autowidth-inlineblock-expected.html.
  • fast/regions/auto-size/autowidth-inlineblock.html: Renamed from LayoutTests/fast/regions/autowidth-inlineblock.html.
  • fast/regions/auto-size/autowidth-nonreplaced-abspos-expected.html: Renamed from LayoutTests/fast/regions/autowidth-nonreplaced-abspos-expected.html.
  • fast/regions/auto-size/autowidth-nonreplaced-abspos.html: Renamed from LayoutTests/fast/regions/autowidth-nonreplaced-abspos.html.
  • fast/regions/auto-size/autowidth-nonreplacedblock-normalflow-expected.html: Renamed from LayoutTests/fast/regions/autowidth-nonreplacedblock-normalflow-expected.html.
  • fast/regions/auto-size/autowidth-nonreplacedblock-normalflow.html: Renamed from LayoutTests/fast/regions/autowidth-nonreplacedblock-normalflow.html.
  • fast/regions/auto-size/autowidth-normalflow-expected.html: Renamed from LayoutTests/fast/regions/autowidth-normalflow-expected.html.
  • fast/regions/auto-size/autowidth-normalflow-maxwidth-expected.html: Renamed from LayoutTests/fast/regions/autowidth-normalflow-maxwidth-expected.html.
  • fast/regions/auto-size/autowidth-normalflow-maxwidth.html: Renamed from LayoutTests/fast/regions/autowidth-normalflow-maxwidth.html.
  • fast/regions/auto-size/autowidth-normalflow-minmaxwidth-expected.html: Renamed from LayoutTests/fast/regions/autowidth-normalflow-minmaxwidth-expected.html.
  • fast/regions/auto-size/autowidth-normalflow-minmaxwidth.html: Renamed from LayoutTests/fast/regions/autowidth-normalflow-minmaxwidth.html.
  • fast/regions/auto-size/autowidth-normalflow-minwidth-expected.html: Renamed from LayoutTests/fast/regions/autowidth-normalflow-minwidth-expected.html.
  • fast/regions/auto-size/autowidth-normalflow-minwidth.html: Renamed from LayoutTests/fast/regions/autowidth-normalflow-minwidth.html.
  • fast/regions/auto-size/autowidth-normalflow-vertrl-expected.html: Renamed from LayoutTests/fast/regions/autowidth-normalflow-vertrl-expected.html.
  • fast/regions/auto-size/autowidth-normalflow-vertrl.html: Renamed from LayoutTests/fast/regions/autowidth-normalflow-vertrl.html.
  • fast/regions/auto-size/autowidth-normalflow.html: Renamed from LayoutTests/fast/regions/autowidth-normalflow.html.
1:07 AM Changeset in webkit [157091] by Carlos Garcia Campos
  • 4 edits in trunk

[GTK] media/video-src-blob.html
​https://bugs.webkit.org/show_bug.cgi?id=102586

Reviewed by Martin Robinson.

Source/WebCore:

Add support for loading videos using blob URLs.

Fixes media/video-src-blob.html.

  • platform/graphics/gstreamer/WebKitWebSourceGStreamer.cpp:

(webKitWebSrcStart): Use BufferData buffering policy for blob
URLs.
(webKitWebSrcGetProtocols): Add blob to the list of supported
protocols.
(webKitWebSrcSetUri): Consider blob URLs as valid.

LayoutTests:

  • platform/gtk/TestExpectations: Unskip

media/video-src-blob.html.

12:58 AM Changeset in webkit [157090] by Darin Adler
  • 3 edits in trunk/Source/WebKit2

Remove use of deleteAllValues in NPRemoteObjectMap::pluginDestroyed
​https://bugs.webkit.org/show_bug.cgi?id=122496

Reviewed by Andreas Kling.

  • Shared/Plugins/NPRemoteObjectMap.cpp:

(WebKit::NPRemoteObjectMap::registerNPObject): Don't call release when putting objects
into the map.
(WebKit::NPRemoteObjectMap::pluginDestroyed): Instead of making an array of receivers
and deleting them all, remove all of the receivers from the map, which will take care
of deleting them. Also change the other loop to use a similar coding style.

  • Shared/Plugins/NPRemoteObjectMap.h: Change the value type of m_registeredNPObjects

to be unique_ptr instead of a raw pointer.

12:45 AM Changeset in webkit [157089] by Darin Adler
  • 7 edits in trunk/Source/WebKit/mac

Change NetscapePluginInstanceProxy::m_replies to use unique_ptr instead of deleteAllValues
​https://bugs.webkit.org/show_bug.cgi?id=122492

Reviewed by Andreas Kling.

  • Plugins/Hosted/NetscapePluginHostManager.mm:

(WebKit::NetscapePluginHostManager::instantiatePlugin): Use auto so this works with
unique_ptr instead of auto_ptr.

  • Plugins/Hosted/NetscapePluginHostProxy.mm:

(WKPCGetScriptableNPObjectReply): Use make_unique instead of using new directly.
(WKPCBooleanReply): Ditto.
(WKPCBooleanAndDataReply): Ditto.
(WKPCInstantiatePluginReply): Ditto.

  • Plugins/Hosted/NetscapePluginInstanceProxy.h: Changed m_replies to map to unique_ptr

instead of to a raw pointer.
(WebKit::NetscapePluginInstanceProxy::setCurrentReply): Changed to take a unique_ptr
instead of a raw pointer since this takes ownership of the pointer.
(WebKit::NetscapePluginInstanceProxy::waitForReply): Changed to return a unique_ptr
instead of a auto_ptr because that's better. Use unique_ptr inside the function
too to avoid the need for an explicit delete.

  • Plugins/Hosted/NetscapePluginInstanceProxy.mm:

(WebKit::NetscapePluginInstanceProxy::~NetscapePluginInstanceProxy): Removed the call to
deleteAllValues.
(WebKit::NetscapePluginInstanceProxy::cancelStreamLoad): Removed unneeded initialization
of a local variable that was already set in all code paths.
(WebKit::NetscapePluginInstanceProxy::wheelEvent): Use auto so this works with unique_ptr
instead of auto_ptr.
(WebKit::NetscapePluginInstanceProxy::print): Ditto.
(WebKit::NetscapePluginInstanceProxy::snapshot): Ditto. Ditto.
(WebKit::NetscapePluginInstanceProxy::processRequestsAndWaitForReply): Changed to return
a unique_ptr. Also removed an unneeded assertion.
(WebKit::NetscapePluginInstanceProxy::createBindingsInstance): Use auto so this works
with unique_ptr instead of auto_ptr.

  • Plugins/Hosted/ProxyInstance.h: Changed return type of waitForReply to unique_ptr.
  • Plugins/Hosted/ProxyInstance.mm:

(WebKit::ProxyInstance::invoke): Use auto so this works with unique_ptr instead of auto_ptr.
(WebKit::ProxyInstance::supportsInvokeDefaultMethod): Ditto.
(WebKit::ProxyInstance::supportsConstruct): Ditto.
(WebKit::ProxyInstance::getPropertyNames): Ditto.
(WebKit::ProxyInstance::methodNamed): Ditto.
(WebKit::ProxyInstance::fieldNamed): Ditto.
(WebKit::ProxyInstance::fieldValue): Ditto.
(WebKit::ProxyInstance::setFieldValue): Ditto.

Note: See TracTimeline for information about the timeline view.