WebKit-https.git
3 years agoUnreviewed rollout r200924. Caused js/regress/string-replace-generic.html to fail.
joepeck@webkit.org [Mon, 16 May 2016 22:31:13 +0000 (22:31 +0000)]
Unreviewed rollout r200924. Caused js/regress/string-replace-generic.html to fail.

.:

* ManualTests/inspector/profiler-test-call.html: Added.
* ManualTests/inspector/profiler-test-many-calls-in-the-same-scope.html: Added.

Source/JavaScriptCore:

* API/JSProfilerPrivate.cpp: Copied from Source/JavaScriptCore/profiler/ProfilerJettisonReason.h.
(JSStartProfiling):
(JSEndProfiling):
* API/JSProfilerPrivate.h: Copied from Source/JavaScriptCore/profiler/ProfilerJettisonReason.h.
* CMakeLists.txt:
* JavaScriptCore.xcodeproj/project.pbxproj:
* bytecode/BytecodeList.json:
* bytecode/BytecodeUseDef.h:
(JSC::computeUsesForBytecodeOffset):
(JSC::computeDefsForBytecodeOffset):
* bytecode/CodeBlock.cpp:
(JSC::CodeBlock::dumpBytecode):
* bytecode/UnlinkedFunctionExecutable.cpp:
(JSC::generateUnlinkedFunctionCodeBlock):
(JSC::UnlinkedFunctionExecutable::unlinkedCodeBlockFor):
* bytecode/UnlinkedFunctionExecutable.h:
* bytecompiler/BytecodeGenerator.cpp:
(JSC::BytecodeGenerator::BytecodeGenerator):
(JSC::BytecodeGenerator::emitCall):
(JSC::BytecodeGenerator::emitCallVarargs):
(JSC::BytecodeGenerator::emitCallVarargsInTailPosition):
(JSC::BytecodeGenerator::emitConstructVarargs):
(JSC::BytecodeGenerator::emitConstruct):
* bytecompiler/BytecodeGenerator.h:
(JSC::CallArguments::profileHookRegister):
(JSC::BytecodeGenerator::shouldEmitProfileHooks):
* bytecompiler/NodesCodegen.cpp:
(JSC::CallArguments::CallArguments):
(JSC::CallFunctionCallDotNode::emitBytecode):
(JSC::ApplyFunctionCallDotNode::emitBytecode):
* dfg/DFGAbstractInterpreterInlines.h:
(JSC::DFG::AbstractInterpreter<AbstractStateType>::executeEffects):
* dfg/DFGByteCodeParser.cpp:
(JSC::DFG::ByteCodeParser::parseBlock):
* dfg/DFGCapabilities.cpp:
(JSC::DFG::capabilityLevel):
* dfg/DFGClobberize.h:
(JSC::DFG::clobberize):
* dfg/DFGDoesGC.cpp:
(JSC::DFG::doesGC):
* dfg/DFGFixupPhase.cpp:
(JSC::DFG::FixupPhase::fixupNode):
* dfg/DFGNodeType.h:
* dfg/DFGPredictionPropagationPhase.cpp:
* dfg/DFGSafeToExecute.h:
(JSC::DFG::safeToExecute):
* dfg/DFGSpeculativeJIT32_64.cpp:
(JSC::DFG::SpeculativeJIT::compile):
* dfg/DFGSpeculativeJIT64.cpp:
(JSC::DFG::SpeculativeJIT::compile):
* inspector/InjectedScriptBase.cpp:
(Inspector::InjectedScriptBase::callFunctionWithEvalEnabled):
* inspector/protocol/Timeline.json:
* interpreter/Interpreter.cpp:
(JSC::UnwindFunctor::operator()):
(JSC::Interpreter::execute):
(JSC::Interpreter::executeCall):
(JSC::Interpreter::executeConstruct):
* jit/JIT.cpp:
(JSC::JIT::privateCompileMainPass):
* jit/JIT.h:
* jit/JITOpcodes.cpp:
(JSC::JIT::emit_op_profile_will_call):
(JSC::JIT::emit_op_profile_did_call):
* jit/JITOpcodes32_64.cpp:
(JSC::JIT::emit_op_profile_will_call):
(JSC::JIT::emit_op_profile_did_call):
* jit/JITOperations.cpp:
* jit/JITOperations.h:
* jsc.cpp:
* llint/LLIntSlowPaths.cpp:
(JSC::LLInt::LLINT_SLOW_PATH_DECL):
* llint/LLIntSlowPaths.h:
* llint/LowLevelInterpreter.asm:
* parser/ParserModes.h:
* profiler/CallIdentifier.h: Added.
(JSC::CallIdentifier::CallIdentifier):
(JSC::CallIdentifier::functionName):
(JSC::CallIdentifier::url):
(JSC::CallIdentifier::lineNumber):
(JSC::CallIdentifier::columnNumber):
(JSC::CallIdentifier::operator==):
(JSC::CallIdentifier::operator!=):
(JSC::CallIdentifier::Hash::hash):
(JSC::CallIdentifier::Hash::equal):
(JSC::CallIdentifier::hash):
(JSC::CallIdentifier::operator const char*):
(JSC::CallIdentifier::c_str):
(WTF::HashTraits<JSC::CallIdentifier>::constructDeletedValue):
(WTF::HashTraits<JSC::CallIdentifier>::isDeletedValue):
* profiler/LegacyProfiler.cpp: Added.
(JSC::LegacyProfiler::profiler):
(JSC::LegacyProfiler::startProfiling):
(JSC::LegacyProfiler::stopProfiling):
(JSC::callFunctionForProfilesWithGroup):
(JSC::LegacyProfiler::suspendProfiling):
(JSC::LegacyProfiler::unsuspendProfiling):
(JSC::LegacyProfiler::willExecute):
(JSC::LegacyProfiler::didExecute):
(JSC::LegacyProfiler::exceptionUnwind):
(JSC::LegacyProfiler::createCallIdentifier):
(JSC::createCallIdentifierFromFunctionImp):
* profiler/LegacyProfiler.h: Added.
(JSC::LegacyProfiler::currentProfiles):
* profiler/Profile.cpp: Added.
(JSC::Profile::create):
(JSC::Profile::Profile):
(JSC::Profile::~Profile):
(JSC::Profile::debugPrint):
(JSC::functionNameCountPairComparator):
(JSC::Profile::debugPrintSampleStyle):
* profiler/Profile.h: Copied from Source/JavaScriptCore/profiler/ProfilerJettisonReason.h.
* profiler/ProfileGenerator.cpp: Added.
(JSC::ProfileGenerator::create):
(JSC::ProfileGenerator::ProfileGenerator):
(JSC::AddParentForConsoleStartFunctor::AddParentForConsoleStartFunctor):
(JSC::AddParentForConsoleStartFunctor::foundParent):
(JSC::AddParentForConsoleStartFunctor::operator()):
(JSC::ProfileGenerator::addParentForConsoleStart):
(JSC::ProfileGenerator::title):
(JSC::ProfileGenerator::beginCallEntry):
(JSC::ProfileGenerator::endCallEntry):
(JSC::ProfileGenerator::willExecute):
(JSC::ProfileGenerator::didExecute):
(JSC::ProfileGenerator::exceptionUnwind):
(JSC::ProfileGenerator::stopProfiling):
(JSC::ProfileGenerator::removeProfileStart):
(JSC::ProfileGenerator::removeProfileEnd):
* profiler/ProfileGenerator.h: Added.
(JSC::ProfileGenerator::profile):
(JSC::ProfileGenerator::origin):
(JSC::ProfileGenerator::profileGroup):
(JSC::ProfileGenerator::setIsSuspended):
* profiler/ProfileNode.cpp: Added.
(JSC::ProfileNode::ProfileNode):
(JSC::ProfileNode::addChild):
(JSC::ProfileNode::removeChild):
(JSC::ProfileNode::spliceNode):
(JSC::ProfileNode::traverseNextNodePostOrder):
(JSC::ProfileNode::debugPrint):
(JSC::ProfileNode::debugPrintSampleStyle):
(JSC::ProfileNode::debugPrintRecursively):
(JSC::ProfileNode::debugPrintSampleStyleRecursively):
* profiler/ProfileNode.h: Added.
(JSC::ProfileNode::create):
(JSC::ProfileNode::Call::Call):
(JSC::ProfileNode::Call::startTime):
(JSC::ProfileNode::Call::setStartTime):
(JSC::ProfileNode::Call::elapsedTime):
(JSC::ProfileNode::Call::setElapsedTime):
(JSC::ProfileNode::operator==):
(JSC::ProfileNode::callerCallFrame):
(JSC::ProfileNode::callIdentifier):
(JSC::ProfileNode::id):
(JSC::ProfileNode::functionName):
(JSC::ProfileNode::url):
(JSC::ProfileNode::lineNumber):
(JSC::ProfileNode::columnNumber):
(JSC::ProfileNode::parent):
(JSC::ProfileNode::setParent):
(JSC::ProfileNode::calls):
(JSC::ProfileNode::lastCall):
(JSC::ProfileNode::appendCall):
(JSC::ProfileNode::children):
(JSC::ProfileNode::firstChild):
(JSC::ProfileNode::lastChild):
(JSC::ProfileNode::nextSibling):
(JSC::ProfileNode::setNextSibling):
(JSC::ProfileNode::forEachNodePostorder):
(JSC::CalculateProfileSubtreeDataFunctor::operator()):
(JSC::CalculateProfileSubtreeDataFunctor::returnValue):
* profiler/ProfilerJettisonReason.cpp:
(WTF::printInternal):
* profiler/ProfilerJettisonReason.h:
* runtime/CodeCache.cpp:
(JSC::CodeCache::getGlobalCodeBlock):
(JSC::CodeCache::getProgramCodeBlock):
(JSC::CodeCache::getEvalCodeBlock):
(JSC::CodeCache::getModuleProgramCodeBlock):
* runtime/CodeCache.h:
* runtime/Executable.cpp:
(JSC::ScriptExecutable::newCodeBlockFor):
* runtime/JSGlobalObject.cpp:
(JSC::JSGlobalObject::~JSGlobalObject):
(JSC::JSGlobalObject::hasLegacyProfiler):
(JSC::JSGlobalObject::createProgramCodeBlock):
(JSC::JSGlobalObject::createEvalCodeBlock):
(JSC::JSGlobalObject::createModuleProgramCodeBlock):
* runtime/JSGlobalObject.h:
(JSC::JSGlobalObject::supportsLegacyProfiling):
* runtime/Options.h:
* runtime/VM.cpp:
(JSC::VM::VM):
(JSC::SetEnabledProfilerFunctor::operator()):
(JSC::VM::setEnabledProfiler):
* runtime/VM.h:
(JSC::VM::enabledProfiler):
(JSC::VM::enabledProfilerAddress):

Source/WebCore:

Tests: fast/profiler/*

* CMakeLists.txt:
* DerivedSources.cpp:
* DerivedSources.make:
* ForwardingHeaders/profiler/LegacyProfiler.h: Added.
* ForwardingHeaders/profiler/Profile.h: Added.
* ForwardingHeaders/profiler/ProfileNode.h: Added.
* WebCore.xcodeproj/project.pbxproj:
* bindings/js/JSCustomXPathNSResolver.cpp:
* bindings/js/JSDOMWindowBase.cpp:
(WebCore::JSDOMWindowBase::supportsLegacyProfiling):
(WebCore::JSDOMWindowBase::supportsRichSourceInfo):
* bindings/js/JSDOMWindowBase.h:
* bindings/js/JSWorkerGlobalScopeBase.cpp:
(WebCore::JSWorkerGlobalScopeBase::supportsLegacyProfiling):
* bindings/js/JSWorkerGlobalScopeBase.h:
* bindings/js/ScriptCachedFrameData.cpp:
* bindings/js/ScriptController.cpp:
(WebCore::ScriptController::clearWindowShell):
* bindings/js/ScriptProfile.h: Copied from Source/JavaScriptCore/profiler/ProfilerJettisonReason.h.
* bindings/js/ScriptProfileNode.h: Copied from Source/JavaScriptCore/profiler/ProfilerJettisonReason.h.
* bindings/scripts/CodeGeneratorJS.pm:
(AddClassForwardIfNeeded):
* bindings/scripts/test/JS/JSTestObj.cpp:
(WebCore::jsTestObjPrototypeFunctionMethodWithSequenceArg):
(WebCore::jsTestObjPrototypeFunctionMethodReturningSequence):
* bindings/scripts/test/TestObj.idl:
* css/CSSParser.cpp:
* dom/Document.cpp:
* inspector/InspectorConsoleInstrumentation.h:
(WebCore::InspectorInstrumentation::stopProfiling):
* inspector/InspectorController.cpp:
(WebCore::InspectorController::InspectorController):
(WebCore::InspectorController::legacyProfilerEnabled):
(WebCore::InspectorController::setLegacyProfilerEnabled):
* inspector/InspectorController.h:
* inspector/InspectorInstrumentation.cpp:
(WebCore::InspectorInstrumentation::stopProfilingImpl):
* inspector/InspectorInstrumentation.h:
* inspector/InspectorTimelineAgent.cpp:
(WebCore::InspectorTimelineAgent::startFromConsole):
(WebCore::InspectorTimelineAgent::stopFromConsole):
* inspector/InspectorTimelineAgent.h:
* inspector/PageDebuggerAgent.cpp:
* inspector/PageRuntimeAgent.cpp:
* inspector/ScriptProfile.idl: Copied from Source/JavaScriptCore/profiler/ProfilerJettisonReason.h.
* inspector/ScriptProfileNode.idl: Copied from Source/JavaScriptCore/profiler/ProfilerJettisonReason.h.
* inspector/TimelineRecordFactory.cpp:
(WebCore::buildAggregateCallInfoInspectorObject):
(WebCore::buildInspectorObject):
(WebCore::buildProfileInspectorObject):
(WebCore::TimelineRecordFactory::appendProfile):
* inspector/TimelineRecordFactory.h:
* page/DOMWindow.cpp:
* page/Page.cpp:
* page/PageConsoleClient.cpp:
(WebCore::PageConsoleClient::profileEnd):
(WebCore::PageConsoleClient::clearProfiles):
* page/PageConsoleClient.h:
* testing/Internals.cpp:
(WebCore::Internals::resetToConsistentState):
(WebCore::Internals::consoleProfiles):
(WebCore::Internals::setLegacyJavaScriptProfilingEnabled):
* testing/Internals.h:
* testing/Internals.idl:
* testing/js/WebCoreTestSupport.cpp:
* xml/XSLStyleSheetLibxslt.cpp:
* xml/XSLTProcessorLibxslt.cpp:

Source/WebKit/win:

* Interfaces/IWebInspector.idl:
* WebCoreStatistics.cpp:
* WebInspector.cpp:
(WebInspector::isJavaScriptProfilingEnabled):
(WebInspector::setJavaScriptProfilingEnabled):
* WebInspector.h:
* WebView.cpp:

LayoutTests:

* fast/profiler/anonymous-event-handler-expected.txt: Added.
* fast/profiler/anonymous-event-handler.html: Added.
* fast/profiler/anonymous-function-called-from-different-contexts-expected.txt: Added.
* fast/profiler/anonymous-function-called-from-different-contexts.html: Added.
* fast/profiler/anonymous-function-calls-built-in-functions-expected.txt: Added.
* fast/profiler/anonymous-function-calls-built-in-functions.html: Added.
* fast/profiler/anonymous-function-calls-eval-expected.txt: Added.
* fast/profiler/anonymous-function-calls-eval.html: Added.
* fast/profiler/anonymous-functions-with-display-names-expected.txt: Added.
* fast/profiler/anonymous-functions-with-display-names.html: Added.
* fast/profiler/apply-expected.txt: Added.
* fast/profiler/apply.html: Added.
* fast/profiler/built-in-function-calls-anonymous-expected.txt: Added.
* fast/profiler/built-in-function-calls-anonymous.html: Added.
* fast/profiler/built-in-function-calls-user-defined-function-expected.txt: Added.
* fast/profiler/built-in-function-calls-user-defined-function.html: Added.
* fast/profiler/call-expected.txt: Added.
* fast/profiler/call-register-leak-expected.txt: Added.
* fast/profiler/call-register-leak.html: Added.
* fast/profiler/call.html: Added.
* fast/profiler/calling-the-function-that-started-the-profiler-from-another-scope-expected.txt: Added.
* fast/profiler/calling-the-function-that-started-the-profiler-from-another-scope.html: Added.
* fast/profiler/compare-multiple-profiles-expected.txt: Added.
* fast/profiler/compare-multiple-profiles.html: Added.
* fast/profiler/constructor-expected.txt: Added.
* fast/profiler/constructor.html: Added.
* fast/profiler/dead-time-expected.txt: Added.
* fast/profiler/dead-time.html: Added.
* fast/profiler/document-dot-write-expected.txt: Added.
* fast/profiler/document-dot-write.html: Added.
* fast/profiler/event-handler-expected.txt: Added.
* fast/profiler/event-handler.html: Added.
* fast/profiler/execution-context-and-eval-on-same-line-expected.txt: Added.
* fast/profiler/execution-context-and-eval-on-same-line.html: Added.
* fast/profiler/inline-event-handler-expected.txt: Added.
* fast/profiler/inline-event-handler.html: Added.
* fast/profiler/many-calls-in-the-same-scope-expected.txt: Added.
* fast/profiler/many-calls-in-the-same-scope.html: Added.
* fast/profiler/multiple-and-different-scoped-anonymous-function-calls-expected.txt: Added.
* fast/profiler/multiple-and-different-scoped-anonymous-function-calls.html: Added.
* fast/profiler/multiple-and-different-scoped-function-calls-expected.txt: Added.
* fast/profiler/multiple-and-different-scoped-function-calls.html: Added.
* fast/profiler/multiple-anonymous-functions-called-from-the-same-function-expected.txt: Added.
* fast/profiler/multiple-anonymous-functions-called-from-the-same-function.html: Added.
* fast/profiler/multiple-frames-expected.txt: Added.
* fast/profiler/multiple-frames.html: Added.
* fast/profiler/named-functions-with-display-names-expected.txt: Added.
* fast/profiler/named-functions-with-display-names.html: Added.
* fast/profiler/nested-anonymous-functon-expected.txt: Added.
* fast/profiler/nested-anonymous-functon.html: Added.
* fast/profiler/nested-start-and-stop-profiler-expected.txt: Added.
* fast/profiler/nested-start-and-stop-profiler.html: Added.
* fast/profiler/no-execution-context-expected.txt: Added.
* fast/profiler/no-execution-context.html: Added.
* fast/profiler/one-execution-context-expected.txt: Added.
* fast/profiler/one-execution-context.html: Added.
* fast/profiler/profile-calls-in-included-file-expected.txt: Added.
* fast/profiler/profile-calls-in-included-file.html: Added.
* fast/profiler/profile-with-no-title-expected.txt: Added.
* fast/profiler/profile-with-no-title.html: Added.
* fast/profiler/profiling-from-a-nested-location-but-stop-profiling-outside-the-nesting-expected.txt: Added.
* fast/profiler/profiling-from-a-nested-location-but-stop-profiling-outside-the-nesting.html: Added.
* fast/profiler/profiling-from-a-nested-location-expected.txt: Added.
* fast/profiler/profiling-from-a-nested-location.html: Added.
* fast/profiler/resources/other-frame.html: Added.
* fast/profiler/resources/other-window.html: Added.
* fast/profiler/resources/profiler-test-JS-resources.js: Added.
(endTest):
(insertGivenText):
(insertNewText):
(arrayOperatorFunction):
(anonymousFunction):
(anotherAnonymousFunction):
(intermediaryFunction):
(isEqualToFive):
(startProfile):
(printHeavyProfilesDataWithoutTime):
(printProfilesDataWithoutTime):
(printProfileNodeWithoutTime):
* fast/profiler/simple-event-call-expected.txt: Added.
* fast/profiler/simple-event-call.html: Added.
* fast/profiler/simple-no-level-change-expected.txt: Added.
* fast/profiler/simple-no-level-change.html: Added.
* fast/profiler/start-and-stop-profiler-multiple-times-expected.txt: Added.
* fast/profiler/start-and-stop-profiler-multiple-times.html: Added.
* fast/profiler/start-and-stop-profiling-in-the-same-function-expected.txt: Added.
* fast/profiler/start-and-stop-profiling-in-the-same-function.html: Added.
* fast/profiler/start-but-dont-stop-profiling-expected.txt: Added.
* fast/profiler/start-but-dont-stop-profiling.html: Added.
* fast/profiler/stop-profiling-after-setTimeout-expected.txt: Added.
* fast/profiler/stop-profiling-after-setTimeout.html: Added.
* fast/profiler/stop-then-function-call-expected.txt: Added.
* fast/profiler/stop-then-function-call.html: Added.
* fast/profiler/throw-exception-from-eval-expected.txt: Added.
* fast/profiler/throw-exception-from-eval.html-disabled: Added.
* fast/profiler/two-execution-contexts-expected.txt: Added.
* fast/profiler/two-execution-contexts.html: Added.
* fast/profiler/user-defined-function-calls-built-in-functions-expected.txt: Added.
* fast/profiler/user-defined-function-calls-built-in-functions.html: Added.
* fast/profiler/window-dot-eval-expected.txt: Added.
* fast/profiler/window-dot-eval.html: Added.
* platform/efl/TestExpectations:
* platform/gtk/TestExpectations:
* platform/ios-simulator/TestExpectations:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@200973 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoText selection is basically impossible on plain text pages.
enrica@apple.com [Mon, 16 May 2016 22:20:16 +0000 (22:20 +0000)]
Text selection is basically impossible on plain text pages.
https://bugs.webkit.org/show_bug.cgi?id=157681
rdar://problem/26065660

Reviewed by Darin Adler.

When dealing with a plain text file, the rules for deciding whether
a position is selectable should be different and we should never
switch to block selection.

* WebProcess/WebPage/ios/WebPageIOS.mm:
(WebKit::WebPage::shouldSwitchToBlockModeForHandle):
(WebKit::rectIsTooBigForSelection): Added helper function.
(WebKit::WebPage::selectTextWithGranularityAtPoint):
(WebKit::WebPage::getPositionInformation):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@200972 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoRenderLayer::hitTestList could mutate the list of candidate layers.
zalan@apple.com [Mon, 16 May 2016 22:00:30 +0000 (22:00 +0000)]
RenderLayer::hitTestList could mutate the list of candidate layers.
https://bugs.webkit.org/show_bug.cgi?id=157718
<rdar://problem/22556046>

Reviewed by Simon Fraser.

This patch ensures that we always start hittesting a clean render tree at EventHandler::hitTestResultAtPoint.

Speculative fix.

* page/EventHandler.cpp:
(WebCore::EventHandler::hitTestResultAtPoint):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@200971 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoUnreviewed, fixed typo in a comment.
annulen@yandex.ru [Mon, 16 May 2016 21:53:44 +0000 (21:53 +0000)]
Unreviewed, fixed typo in a comment.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@200970 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoRemove unused initializer for WebEvent on iOS.
enrica@apple.com [Mon, 16 May 2016 21:42:07 +0000 (21:42 +0000)]
Remove unused initializer for WebEvent on iOS.
https://bugs.webkit.org/show_bug.cgi?id=157689

Reviewed by Anders Carlsson.

Removing use of characterSet property.

* WebView/WebHTMLView.mm:
(-[WebHTMLView _handleEditingKeyEvent:]):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@200969 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoRemove unused initializer for WebEvent on iOS.
enrica@apple.com [Mon, 16 May 2016 21:39:04 +0000 (21:39 +0000)]
Remove unused initializer for WebEvent on iOS.
https://bugs.webkit.org/show_bug.cgi?id=157689

Reviewed by Anders Carlsson.

Source/WebCore:

This is no longer used and can be removed.
The logic tied to isPopupVariant has been incorporated
in keyboard flags. The characterSet property is no longer needed too.

* platform/ios/WebEvent.h:
* platform/ios/WebEvent.mm:
(-[WebEvent initWithKeyEventType:timeStamp:characters:charactersIgnoringModifiers:modifiers:isRepeating:withFlags:keyCode:isTabKey:characterSet:]):
(-[WebEvent initWithKeyEventType:timeStamp:characters:charactersIgnoringModifiers:modifiers:isRepeating:isPopupVariant:keyCode:isTabKey:characterSet:]): Deleted.
(-[WebEvent _characterSetDescription]): Deleted.
(-[WebEvent isPopupVariant]): Deleted.
(-[WebEvent characterSet]): Deleted.

Tools:

Adopting different initializer for WebEvent to fix the build.

* DumpRenderTree/mac/EventSendingController.mm:
(-[EventSendingController keyDown:withModifiers:withLocation:]):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@200968 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoAdd a WebCore logging channel for images
simon.fraser@apple.com [Mon, 16 May 2016 21:36:15 +0000 (21:36 +0000)]
Add a WebCore logging channel for images
https://bugs.webkit.org/show_bug.cgi?id=157752

Reviewed by Zalan Bujtas.

Create an Images log channel, and log various things related to decoding and drawing
images.

* platform/Logging.h:
* platform/graphics/BitmapImage.cpp:
(WebCore::BitmapImage::destroyDecodedDataIfNecessary):
(WebCore::BitmapImage::cacheFrame):
(WebCore::BitmapImage::startAnimation):
* platform/graphics/cg/GraphicsContextCG.cpp:
(WebCore::GraphicsContext::drawNativeImage):
* platform/graphics/cg/ImageDecoderCG.cpp:
(WebCore::ImageDecoder::createFrameImageAtIndex):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@200967 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years ago[iOS] Remove dispatch_after in -[WKFormInputSession setSuggestions:]
commit-queue@webkit.org [Mon, 16 May 2016 20:52:33 +0000 (20:52 +0000)]
[iOS] Remove dispatch_after in -[WKFormInputSession setSuggestions:]
https://bugs.webkit.org/show_bug.cgi?id=157745

Patch by Chelsea Pugh <cpugh@apple.com> on 2016-05-16
Reviewed by Dan Bernstein.

* UIProcess/ios/WKContentViewInteraction.mm:
(-[WKFormInputSession setSuggestions:]): Remove the dispatch_after since
there is no longer a chance of keyboard suggestions replacing the
suggestions set here.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@200966 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoDon't execute JavaScript within HTMLMediaElement::stop()
adachan@apple.com [Mon, 16 May 2016 20:46:01 +0000 (20:46 +0000)]
Don't execute JavaScript within HTMLMediaElement::stop()
https://bugs.webkit.org/show_bug.cgi?id=157655

Reviewed by Chris Dumez.

* dom/ScriptExecutionContext.cpp:
(WebCore::ScriptExecutionContext::suspendActiveDOMObjects):
Move the setting of m_activeDOMObjectsAreSuspended to true earlier so we won't execute
any JS while suspending the objects.
* html/HTMLMediaElement.cpp:
(WebCore::HTMLMediaElement::updateMediaControlsAfterPresentationModeChange):
Bail early if the controls script hasn't been injected yet or the DOM object has been
stopped or suspended.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@200965 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoFocus ordering should respect slot elements
rniwa@webkit.org [Mon, 16 May 2016 20:26:40 +0000 (20:26 +0000)]
Focus ordering should respect slot elements
https://bugs.webkit.org/show_bug.cgi?id=151379

Reviewed by Antti Koivisto.

Source/WebCore:

Implemented the sequential focus navigation ordering as discussed on
https://github.com/w3c/webcomponents/issues/375

New behavior treats each shadow root and slot as a "focus scope". The focus navigation ordering
is defined within each "focus scope" using tabindex, treating any "focus scope owner"
(e.g. shadow host or a slot) as if it was having tabindex=0 if it wasn't itself focusable.

This patch modifies FocusNavigationScope to support a focus scope defined for a slot element in
addition to the one defined for a shadow tree and a document as previously supported.

Tests: fast/shadow-dom/focus-across-details-element.html
       fast/shadow-dom/focus-navigation-across-slots.html

* dom/Node.cpp:
(WebCore::parentShadowRoot): Extracted from assignedSlot.
(WebCore::Node::assignedSlot):
(WebCore::Node::assignedSlotForBindings): Added.
* dom/Node.h:
* dom/NonDocumentTypeChildNode.idl:
* html/HTMLDetailsElement.h:
(HTMLDetailsElement::hasCustomFocusLogic): Added. Don't treat details element as a "focus scope".
* html/HTMLSummaryElement.h:
(HTMLSummaryElement::hasCustomFocusLogic): Ditto for summary element.
* page/FocusController.cpp:
(WebCore::hasCustomFocusLogic): Moved.
(WebCore::isFocusScopeOwner): Added. Returns true on a shadow host without a custom focus logic or
on a slot inside a shadow tree whose shadow host doesn't have a custom focus logic.
(WebCore::FocusNavigationScope::firstChildInScope): Now takes a reference. Call isFocusScopeOwner
to check for both slots and shadow roots instead of just the latter. This fixes a subtle bug that
focus may never get out of textarea in some cases due to its failure to check hasCustomFocusLogic.
(WebCore::FocusNavigationScope::lastChildInScope): Ditto.
(WebCore::FocusNavigationScope::parentInScope): Made this a member function since it needs to check
against m_slotElement inside the focus scope of a slot.
(WebCore::FocusNavigationScope::nextSiblingInScope): Added. Finds the next assigned node in a slot
in the focus scope defined for a slot. Just calls nextSibling() in the focus scope for shadow tree
and document.
(WebCore::FocusNavigationScope::previousSiblingInScope): Ditto for finding the previous sibling.
(WebCore::FocusNavigationScope::firstNodeInScope): Added. This function replaces rootNode() which
doesn't exist for the focus scope of a slot element.
(WebCore::FocusNavigationScope::lastNodeInScope): Ditto for the last node.
(WebCore::FocusNavigationScope::nextInScope):
(WebCore::FocusNavigationScope::previousInScope):
(WebCore::FocusNavigationScope::FocusNavigationScope): Added a variant that takes HTMLSlotElement.
(WebCore::FocusNavigationScope::owner): Added the support for slot elements.
(WebCore::FocusNavigationScope::scopeOf): Ditto.
(WebCore::FocusNavigationScope::scopeOwnedByScopeOwner): Ditto.
(WebCore::isFocusableElementOrScopeOwner): Added the support for slot elements and renamed from
isFocusableOrHasShadowTreeWithoutCustomFocusLogic.
(WebCore::isNonFocusableScopeOwner): Ditto. Renamed from isNonFocusableShadowHost.
(WebCore::isFocusableScopeOwner): Ditto. Renamed from isFocusableShadowHost.
(WebCore::shadowAdjustedTabIndex): Added the support for slot elements.
(WebCore::FocusController::findFocusableElementAcrossFocusScope):
(WebCore::FocusController::nextFocusableElementWithinScope):
(WebCore::FocusController::previousFocusableElementWithinScope):
(WebCore::FocusController::findElementWithExactTabIndex):
(WebCore::nextElementWithGreaterTabIndex): Call firstNodeInScope() instead of rootNode() here since
there is no root node for the focus scope defined for a slot element.
(WebCore::previousElementWithLowerTabIndex): Ditto for scope.lastNodeInScope().
(WebCore::FocusController::nextFocusableElementOrScopeOwner):
(WebCore::FocusController::previousFocusableElementOrScopeOwner):
(WebCore::parentInScope): Deleted.
(WebCore::FocusNavigationScope::rootNode): Deleted.
(WebCore::FocusNavigationScope::scopeOwnedByShadowHost): Deleted.
(WebCore::isNonFocusableShadowHost): Deleted.
(WebCore::isFocusableShadowHost): Deleted.
(WebCore::isFocusableOrHasShadowTreeWithoutCustomFocusLogic): Deleted.

LayoutTests:

Added regression tests for moving focus by tab and shift+tab across
user-defined shadow trees with slots and details element.

* fast/shadow-dom/focus-across-details-element-expected.txt: Added.
* fast/shadow-dom/focus-across-details-element.html: Added.
* fast/shadow-dom/focus-navigation-across-slots-expected.txt: Added.
* fast/shadow-dom/focus-navigation-across-slots.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@200964 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoUse WTF::Optional for ScrollView's m_deferredScrollDelta / m_deferredScrollOffsets
cdumez@apple.com [Mon, 16 May 2016 20:24:52 +0000 (20:24 +0000)]
Use WTF::Optional for ScrollView's m_deferredScrollDelta / m_deferredScrollOffsets
https://bugs.webkit.org/show_bug.cgi?id=157747

Reviewed by Zalan Bujtas.

Use WTF::Optional for ScrollView's m_deferredScrollDelta / m_deferredScrollOffsets
instead of std::unique_ptr as it is more suited for this purpose.

* platform/ScrollView.cpp:
(WebCore::ScrollView::scrollOffsetChangedViaPlatformWidget):
(WebCore::ScrollView::handleDeferredScrollUpdateAfterContentSizeChange):
(WebCore::ScrollView::scrollTo):
* platform/ScrollView.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@200963 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoWeb Inspector: Contents of Duration column are covered by always on (legacy) scroll...
nvasilyev@apple.com [Mon, 16 May 2016 20:19:24 +0000 (20:19 +0000)]
Web Inspector: Contents of Duration column are covered by always on (legacy) scroll bars
https://bugs.webkit.org/show_bug.cgi?id=157590

Reviewed by Timothy Hatcher.

Set the right padding of the DataGrid header to match the scrollbar width.

* UserInterface/Views/DataGrid.css:
(.data-grid .data-container):
(.data-grid.inline .data-container):
Don't show scrollbars for inline data grids.

(.data-grid > .header-wrapper):
(.data-grid.no-header > .header-wrapper > table.header):
(.data-grid.no-header > table.header): Deleted.
(.data-grid th): Deleted.
* UserInterface/Views/DataGrid.js:
(WebInspector.DataGrid):
We can't add padding-rigth to a table. Wrap the table in div.header-wrapper.

(WebInspector.DataGrid.prototype._updateScrollbarPadding):
(WebInspector.DataGrid.prototype.layout):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@200962 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoCrash simulating keystrokes at WebKit::WebAutomationSession::platformSimulateKeyStroke
bburg@apple.com [Mon, 16 May 2016 20:01:03 +0000 (20:01 +0000)]
Crash simulating keystrokes at WebKit::WebAutomationSession::platformSimulateKeyStroke
https://bugs.webkit.org/show_bug.cgi?id=157737
<rdar://problem/26292946>

Reviewed by Timothy Hatcher.

* UIProcess/Cocoa/WebAutomationSessionCocoa.mm:
(WebKit::WebAutomationSession::platformSimulateKeyStroke):
AppKit expects characters passed to [NSEvent keyEventWithType:...]
to be non-nil even if there are no characters. Initialize characters
to the empty string @"". It will be overwritten if the keystroke should
produce any unicode characters.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@200961 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoRebaseline tests for ios-simulator
ryanhaddad@apple.com [Mon, 16 May 2016 19:57:55 +0000 (19:57 +0000)]
Rebaseline tests for ios-simulator

Unreviewed test gardening.

* fast/events/touch/document-create-touch-list-ios-expected.txt:
* platform/ios-simulator/ios/touch/construct-TouchList-expected.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@200960 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoWeb Automation: Automation.inspectBrowsingContext should automatically start page...
bburg@apple.com [Mon, 16 May 2016 19:53:00 +0000 (19:53 +0000)]
Web Automation: Automation.inspectBrowsingContext should automatically start page profiling
https://bugs.webkit.org/show_bug.cgi?id=157739

Reviewed by Timothy Hatcher.

* UIProcess/API/C/WKInspector.cpp:
(WKInspectorTogglePageProfiling):
Implicitly show the Web Inspector in the C API command to preserve existing behavior.

* UIProcess/Automation/WebAutomationSession.cpp:
(WebKit::WebAutomationSession::inspectorFrontendLoaded):
If the frontend loaded, it was either because the user opened Web Inspector (and
turning on page profiling is harmless), or it was loaded but not shown by the
inspectBrowsingContext command. For the latter, we want to start page profiling
before processing any additional commands so subsequent execution is captured.

* UIProcess/WebInspectorProxy.cpp:
(WebKit::WebInspectorProxy::togglePageProfiling):
Send the start/stop profiling messages directly to the WebInspectorUI process instead of
bouncing through the inspected page's process, which does an implicit show() we don't want.

* UIProcess/WebPageProxy.cpp:
(WebKit::WebPageProxy::inspector): Make it const.
* UIProcess/WebPageProxy.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@200959 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoFixupPhase should be more eager to demote bit math to untyped
fpizlo@apple.com [Mon, 16 May 2016 19:41:01 +0000 (19:41 +0000)]
FixupPhase should be more eager to demote bit math to untyped
https://bugs.webkit.org/show_bug.cgi?id=157746

Reviewed by Mark Lam.
Source/JavaScriptCore:

This just makes the logic for how we fixup bit math match the way we do it in other places.
This doesn't affect performance on any major benchmark but it's a big win on new
microbenchmarks added in this change.

Details:

object-and                                     11.1610+-0.7602     ^      4.8105+-0.1690        ^ definitely 2.3201x faster
object-or                                      11.0845+-0.2487     ^      4.7146+-0.0374        ^ definitely 2.3511x faster
object-xor                                     10.2946+-0.9946     ^      4.7278+-0.0814        ^ definitely 2.1775x faster
object-lshift                                  10.4896+-1.0867     ^      4.7699+-0.0721        ^ definitely 2.1991x faster
object-rshift                                  11.1239+-0.5010     ^      4.7194+-0.0445        ^ definitely 2.3570x faster
object-urshift                                 10.9745+-0.1315     ^      4.7848+-0.0479        ^ definitely 2.2936x faster

* dfg/DFGFixupPhase.cpp:
(JSC::DFG::FixupPhase::fixupNode):

LayoutTests:

* js/regress/object-and-expected.txt: Added.
* js/regress/object-and.html: Added.
* js/regress/object-int-and-array-expected.txt: Added.
* js/regress/object-int-and-array.html: Added.
* js/regress/object-lshift-expected.txt: Added.
* js/regress/object-lshift.html: Added.
* js/regress/object-or-expected.txt: Added.
* js/regress/object-or.html: Added.
* js/regress/object-rshift-expected.txt: Added.
* js/regress/object-rshift.html: Added.
* js/regress/object-urshift-expected.txt: Added.
* js/regress/object-urshift.html: Added.
* js/regress/object-xor-expected.txt: Added.
* js/regress/object-xor.html: Added.
* js/regress/script-tests/object-and.js: Added.
(o.valueOf):
* js/regress/script-tests/object-int-and-array.js: Added.
(i.o.valueOf):
* js/regress/script-tests/object-lshift.js: Added.
(o.valueOf):
* js/regress/script-tests/object-or.js: Added.
(o.valueOf):
* js/regress/script-tests/object-rshift.js: Added.
(o.valueOf):
* js/regress/script-tests/object-urshift.js: Added.
(o.valueOf):
* js/regress/script-tests/object-xor.js: Added.
(o.valueOf):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@200958 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoFix some deprecation warnings.
conrad_shultz@apple.com [Mon, 16 May 2016 19:20:40 +0000 (19:20 +0000)]
Fix some deprecation warnings.

* UIProcess/mac/WKPrintingView.mm:
(-[WKPrintingView _drawPDFDocument:page:atPoint:]):
* WebProcess/Plugins/PDF/DeprecatedPDFPlugin.mm:
(WebKit::PDFPlugin::setActiveAnnotation):
(WebKit::PDFPlugin::lookupTextAtLocation):
* WebProcess/Plugins/PDF/PDFPluginChoiceAnnotation.h:
* WebProcess/Plugins/PDF/PDFPluginChoiceAnnotation.mm:
(WebKit::PDFPluginChoiceAnnotation::createAnnotationElement):
* WebProcess/Plugins/PDF/PDFPluginTextAnnotation.h:
* WebProcess/Plugins/PDF/PDFPluginTextAnnotation.mm:
(WebKit::PDFPluginTextAnnotation::createAnnotationElement):
* WebProcess/WebPage/mac/WebPageMac.mm:
(WebKit::drawPDFPage):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@200957 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoAdd machine-readable results for JSC API tests to the Buildbot json log
commit-queue@webkit.org [Mon, 16 May 2016 18:57:28 +0000 (18:57 +0000)]
Add machine-readable results for JSC API tests to the Buildbot json log
https://bugs.webkit.org/show_bug.cgi?id=157642

Patch by Srinivasan Vijayaraghavan <svijayaraghavan@apple.com> on 2016-05-16
Reviewed by Geoffrey Garen.

* BuildSlaveSupport/build.webkit.org-config/master.cfg:
(RunJavaScriptCoreTests): Use --no-fail-fast option on run-javascriptcore-tests
* Scripts/run-javascriptcore-tests:
(runJSCStressTests): Abstracted out writing JSON to file (new function below)
(writeJsonDataIfApplicable): Added

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@200956 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoWeb Inspector: Background stripes don't span the width of Network sidebar
nvasilyev@apple.com [Mon, 16 May 2016 18:54:41 +0000 (18:54 +0000)]
Web Inspector: Background stripes don't span the width of Network sidebar
https://bugs.webkit.org/show_bug.cgi?id=157744

Reviewed by Timothy Hatcher.

* UserInterface/Views/NetworkSidebarPanel.css:
(.sidebar > .panel.navigation.network.network-grid-content-view-showing > .content):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@200955 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agocontainingBlockFor*Position functions should take the renderer instead of the parent.
zalan@apple.com [Mon, 16 May 2016 18:44:02 +0000 (18:44 +0000)]
containingBlockFor*Position functions should take the renderer instead of the parent.
https://bugs.webkit.org/show_bug.cgi?id=157659

Reviewed by Simon Fraser.

containingBlockForFixedPosition, containingBlockForAbsolutePosition and containingBlockForObjectInFlow functions
expect the renderer's parent to be passed in (unless it is a RenderInline!). It is rather misleading and highly error-prone.
We should call them with the renderer itself instead.

* dom/Element.cpp:
(WebCore::layoutOverflowRectContainsAllDescendants): This expects ancestor containing block.
* rendering/LogicalSelectionOffsetCaches.h:
(WebCore::LogicalSelectionOffsetCaches::LogicalSelectionOffsetCaches):
* rendering/RenderElement.cpp:
(WebCore::containingBlockForFixedPosition):
(WebCore::containingBlockForAbsolutePosition):
(WebCore::containingBlockForObjectInFlow):
* rendering/RenderElement.h:
* rendering/RenderInline.cpp:
(WebCore::RenderInline::styleWillChange):
* rendering/RenderLineBreak.cpp:
(WebCore::RenderLineBreak::collectSelectionRects): Not a behaviour change.
* rendering/RenderObject.cpp:
(WebCore::RenderObject::containingBlock): RenderScrollbarPart renderer now returns
the containing block based on its owning renderer's style.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@200953 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoWeb Inspector: Computed style shows both prefixed and unprefixed variants of properties
bburg@apple.com [Mon, 16 May 2016 18:26:32 +0000 (18:26 +0000)]
Web Inspector: Computed style shows both prefixed and unprefixed variants of properties
https://bugs.webkit.org/show_bug.cgi?id=157674
<rdar://problem/24339756>

Patch by Antoine Quint <graouts@apple.com> on 2016-05-16
Reviewed by Timothy Hatcher.

We make the CSSProperty `implicit` property read-write, and in the case where a computed style
has a property marked as explicit, we also check that it's found in matching rules to consider
it non-implicit. This correctly filters out variants of properties set explicitly.

* UserInterface/Models/CSSProperty.js:
(WebInspector.CSSProperty.prototype.set implicit):
* UserInterface/Models/DOMNodeStyles.js:
(WebInspector.DOMNodeStyles.prototype.refresh.fetchedComputedStyle):
(WebInspector.DOMNodeStyles.prototype.refresh):
(WebInspector.DOMNodeStyles.prototype._isPropertyFoundInMatchingRules):
(WebInspector.DOMNodeStyles):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@200952 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years ago[WK2] Add API to WKWebViewConfiguration to control autoplay policy.
jer.noble@apple.com [Mon, 16 May 2016 18:23:22 +0000 (18:23 +0000)]
[WK2] Add API to WKWebViewConfiguration to control autoplay policy.
https://bugs.webkit.org/show_bug.cgi?id=156312

Reviewed by Dan Bernstein.

Source/WebKit2:

Add a new API to WKWebViewConfiguration to allow fine-grained control over when media is allowed
to play without a user gesture, and simultaneously deprecate the existing, all-or-nothing API,
and deprecated the equivalent SPIs.

* UIProcess/API/Cocoa/WKWebView.mm:
(-[WKWebView _initializeWithConfiguration:]):
* UIProcess/API/Cocoa/WKWebViewConfiguration.h:
* UIProcess/API/Cocoa/WKWebViewConfiguration.mm:
(-[WKWebViewConfiguration init]):
(-[WKWebViewConfiguration encodeWithCoder:]):
(-[WKWebViewConfiguration initWithCoder:]):
(-[WKWebViewConfiguration copyWithZone:]):
(-[WKWebViewConfiguration _requiresUserActionForVideoPlayback]):
(-[WKWebViewConfiguration _setRequiresUserActionForVideoPlayback:]):
(-[WKWebViewConfiguration _requiresUserActionForAudioPlayback]):
(-[WKWebViewConfiguration _setRequiresUserActionForAudioPlayback:]):
(-[WKWebViewConfiguration requiresUserActionForMediaPlayback]):
(-[WKWebViewConfiguration setRequiresUserActionForMediaPlayback:]):
* UIProcess/API/Cocoa/WKWebViewConfigurationPrivate.h:

Tools:

* WebKitTestRunner/cocoa/TestControllerCocoa.mm:
(WTR::initializeWebViewConfiguration):
* TestWebKitAPI/Tests/WebKit2Cocoa/RequiresUserActionForPlayback.mm:
(TEST_F):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@200951 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoWeb Automation: WebAutomationSessionProxy.js gets injected more than once sometimes
bburg@apple.com [Mon, 16 May 2016 18:09:46 +0000 (18:09 +0000)]
Web Automation: WebAutomationSessionProxy.js gets injected more than once sometimes
https://bugs.webkit.org/show_bug.cgi?id=157716
<rdar://problem/26287306>

Reviewed by Timothy Hatcher.

Whenever a script was injected into a non-normal world by a WebKit client,
the WebProcess's WebAutomationSessionProxy singleton would get a
notification that the window was cleared from the relevant frame.
This notification happens when creating the world's window shell for
the first time.

This code should ignore such notifications that originate from non-main world
contexts. Web Inspector's instrumentation already ignored this, but
the automation session notification comes in via a different WebKit2 layer.

* WebProcess/WebCoreSupport/WebFrameLoaderClient.cpp:
(WebKit::WebFrameLoaderClient::dispatchDidClearWindowObjectInWorld):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@200950 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoWeb Inspector: DataGrid _updateVisibleRows dominates profiles of timeline recordings...
timothy@apple.com [Mon, 16 May 2016 18:04:25 +0000 (18:04 +0000)]
Web Inspector: DataGrid _updateVisibleRows dominates profiles of timeline recordings when data grid (Overview or TimelineDataGrids) is showing
https://bugs.webkit.org/show_bug.cgi?id=157664
rdar://problem/26262219

Reviewed by Joseph Pecoraro.

* UserInterface/Views/DataGrid.js:
(WebInspector.DataGrid): Added new members.
(WebInspector.DataGrid.prototype.layout): Reset _cachedScrollTop and _cachedScrollHeight on resize.
(WebInspector.DataGrid.prototype._noteScrollPositionChanged): Added.
(WebInspector.DataGrid.prototype._updateVisibleRows): Cache sizes and positions when possible.
(WebInspector.DataGridNode.prototype.set hidden): Added call to _noteRowsChanged.
(WebInspector.DataGridNode.prototype.collapse): Call _noteRowsChanged instead of needsLayout.
(WebInspector.DataGridNode.prototype.expand): Call _noteRowsChanged instead of needsLayout.
(WebInspector.DataGrid.prototype._updateFilter): Removed direct call to _updateVisibleRows, this is
better handled by DataGridNode's hidden setter.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@200949 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoWeb Inspector: Many DataGrid instances do not save/restore their scroll position
timothy@apple.com [Mon, 16 May 2016 17:41:41 +0000 (17:41 +0000)]
Web Inspector: Many DataGrid instances do not save/restore their scroll position
https://bugs.webkit.org/show_bug.cgi?id=157709
rdar://problem/26286090

Reviewed by Brian Burg.

* UserInterface/Models/BackForwardEntry.js:
(WebInspector.BackForwardEntry.prototype.makeCopy): Added.
* UserInterface/Views/ContentViewContainer.js:
(WebInspector.ContentViewContainer.prototype.showContentView): Copy the last entry for the view.
* UserInterface/Views/DOMStorageContentView.js:
(WebInspector.DOMStorageContentView.prototype.get scrollableElements): Added.
* UserInterface/Views/HeapSnapshotContentView.js:
(WebInspector.HeapSnapshotContentView.prototype.get scrollableElements): Added.
* UserInterface/Views/IndexedDatabaseObjectStoreContentView.js:
(WebInspector.IndexedDatabaseObjectStoreContentView.prototype.get scrollableElements): Added.
* UserInterface/Views/MemoryTimelineView.js:
(WebInspector.MemoryTimelineView.prototype.get scrollableElements): Added.
* UserInterface/Views/ProfileView.js:
(WebInspector.ProfileView.prototype.get scrollableElements): Added.
(WebInspector.ProfileView.prototype._repopulateDataGridFromTree): Removed unused skipRefresh argument.
* UserInterface/Views/ScriptClusterTimelineView.js:
(WebInspector.ScriptClusterTimelineView.prototype.get scrollableElements): Added.
* UserInterface/Views/ScriptProfileTimelineView.js:
(WebInspector.ScriptProfileTimelineView.prototype.get scrollableElements): Added.
* UserInterface/Views/TimelineView.js:
(WebInspector.TimelineView.prototype.get scrollableElements): Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@200947 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoRegExp /y flag incorrect handling of mixed-length alternation
msaboff@apple.com [Mon, 16 May 2016 17:40:15 +0000 (17:40 +0000)]
RegExp /y flag incorrect handling of mixed-length alternation
https://bugs.webkit.org/show_bug.cgi?id=157723

Reviewed by Filip Pizlo.

Source/JavaScriptCore:

Previously for sticky patterns, we were bailing out and exiting when backtracking
alternatives with dissimilar match lengths.  Deleted that code.  Instead, for
sticky patterns we need to process the backtracking except for advancing to the
next input index.

* yarr/YarrJIT.cpp:
(JSC::Yarr::YarrGenerator::backtrack):

LayoutTests:

Added tests for alternatives with shorter to longer lengths.

* js/regexp-sticky-expected.txt:
* js/script-tests/regexp-sticky.js:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@200946 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoREGRESSION (r192098): Content missing after copy and paste to Notes App on retina...
bfulgham@apple.com [Mon, 16 May 2016 17:35:30 +0000 (17:35 +0000)]
REGRESSION (r192098): Content missing after copy and paste to Notes App on retina displays
https://bugs.webkit.org/show_bug.cgi?id=157630
<rdar://problem/25277577>

Reviewed by Simon Fraser.

Source/WebCore:

Tested by http/tests/images/hidpi-srcset-copy.html

The code to create a WebArchive (or other representation) of a webpage was using the 'src' attribute of
the image element, which might be changed to reflect the optimal image dimensions for a specific
display. Instead, we should use the 'imageSourceURL' method of the Element class so we get the correct
'original' image URL in our cached resources.

* editing/cocoa/HTMLConverter.mm:
(HTMLConverter::_processElement): Use 'imageSourceURL', rather than getting the raw 'src' attribute.
* html/HTMLImageElement.cpp:
(WebCore::HTMLImageElement::addSubresourceAttributeURLs): Ditto.

Tools:

Update test infrastructure to support checking the number of images in the pasteboard.

* DumpRenderTree/TestRunner.cpp:
(imageCountInGeneralPasteboardCallback): Added.
(TestRunner::staticFunctions):
* DumpRenderTree/TestRunner.h:
* DumpRenderTree/mac/TestRunnerMac.mm:
(TestRunner::imageCountInGeneralPasteboard): Added.
* DumpRenderTree/win/TestRunnerWin.cpp:
(TestRunner::imageCountInGeneralPasteboard): Added stub method.
* WebKitTestRunner/InjectedBundle/Bindings/TestRunner.idl:
* WebKitTestRunner/InjectedBundle/InjectedBundle.cpp:
(WTR::InjectedBundle::imageCountInGeneralPasteboard): Added.
* WebKitTestRunner/InjectedBundle/InjectedBundle.h:
* WebKitTestRunner/InjectedBundle/TestRunner.cpp:
(WTR::TestRunner::imageCountInGeneralPasteboard): Added.
* WebKitTestRunner/InjectedBundle/TestRunner.h:
* WebKitTestRunner/TestController.cpp:
(WTR::TestController::imageCountInGeneralPasteboard): Added stub implementation.
* WebKitTestRunner/TestController.h:
* WebKitTestRunner/TestInvocation.cpp:
(WTR::TestInvocation::didReceiveSynchronousMessageFromInjectedBundle): Handle copy message.
* WebKitTestRunner/cocoa/TestControllerCocoa.mm:
(WTR::TestController::imageCountInGeneralPasteboard): Added.

LayoutTests:

* http/tests/images/hidpi-srcset-copy-expected.txt: Added.
* http/tests/images/hidpi-srcset-copy.html: Added.
* http/tests/resources/img-srcset: Added.
* http/tests/resources/img-srcset/image1-620x442.jpg: Added.
* http/tests/resources/img-srcset/image2-1024x731.jpg: Added.
* http/tests/resources/img-srcset/image2-1400x1000.jpg: Added.
* http/tests/resources/img-srcset/image2-300x214.jpg: Added.
* http/tests/resources/img-srcset/image2-620x442.jpg: Added.
* http/tests/resources/img-srcset/image2-840x600.jpg: Added.
* http/tests/resources/img-srcset/image3-250x178.jpg: Added.
* http/tests/resources/img-srcset/image3-615x438.jpg: Added.
* http/tests/resources/img-srcset/image3-620x440.jpg: Added.
* http/tests/resources/img-srcset/image3-700x498.jpg: Added.
* http/tests/resources/img-srcset/image3-840x598.jpg: Added.
* http/tests/resources/img-srcset/image3.jpg: Added.
* http/tests/resources/img-srcset/image4-1024x612.jpg: Added.
* http/tests/resources/img-srcset/image4-300x179.jpg: Added.
* http/tests/resources/img-srcset/image4-620x370.jpg: Added.
* http/tests/resources/img-srcset/image4-768x459.jpg: Added.
* http/tests/resources/img-srcset/image4-840x502.jpg: Added.
* http/tests/resources/img-srcset/image4-940x560.jpg: Added.
* platform/efl/TestExpectations: No pasteboard implementation. Skip.
* platform/gtk/TestExpectations: Ditto.
* platform/win/TestExpectations: Ditto.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@200945 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoRandom CachedFont.cpp cleanup.
beidson@apple.com [Mon, 16 May 2016 17:22:13 +0000 (17:22 +0000)]
Random CachedFont.cpp cleanup.
https://bugs.webkit.org/show_bug.cgi?id=157740

Reviewed by Myles C. Maxfield.

No new tests (Refactor/cleanup, no behavior change)

* loader/cache/CachedFont.cpp:
(WebCore::CachedFont::ensureCustomFontData): Remove unneeded local RefPtr<>
(WebCore::CachedFont::createCustomFontData): Refactor to not require unnecessary Ref-churn in the common case.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@200944 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoModernize Track classes' code
cdumez@apple.com [Mon, 16 May 2016 16:37:26 +0000 (16:37 +0000)]
Modernize Track classes' code
https://bugs.webkit.org/show_bug.cgi?id=157735

Reviewed by Eric Carlson.

Modernize Track classes' code a bit.

* Modules/mediasource/MediaSource.cpp:
(WebCore::MediaSource::removeSourceBuffer):
* Modules/mediasource/SourceBuffer.cpp:
(WebCore::SourceBuffer::sourceBufferPrivateDidReceiveInitializationSegment):
* bindings/js/JSTrackCustom.cpp:
(WebCore::toTrack):
(WebCore::toJS):
* html/HTMLMediaElement.cpp:
(WebCore::HTMLMediaElement::audioTrackEnabledChanged):
(WebCore::HTMLMediaElement::textTrackModeChanged):
(WebCore::HTMLMediaElement::videoTrackSelectedChanged):
(WebCore::HTMLMediaElement::textTrackRemoveCues):
(WebCore::HTMLMediaElement::mediaPlayerDidAddTextTrack):
(WebCore::HTMLMediaElement::addAudioTrack):
(WebCore::HTMLMediaElement::addTextTrack):
(WebCore::HTMLMediaElement::addVideoTrack):
(WebCore::HTMLMediaElement::removeAudioTrack):
(WebCore::HTMLMediaElement::removeTextTrack):
(WebCore::HTMLMediaElement::removeVideoTrack):
(WebCore::HTMLMediaElement::forgetResourceSpecificTracks):
(WebCore::HTMLMediaElement::audioTracks):
(WebCore::HTMLMediaElement::textTracks):
(WebCore::HTMLMediaElement::videoTracks):
(WebCore::HTMLMediaElement::didAddTextTrack):
(WebCore::HTMLMediaElement::didRemoveTextTrack):
(WebCore::HTMLMediaElement::setSelectedTextTrack):
(WebCore::HTMLMediaElement::textTrackAddCues): Deleted.
(WebCore::HTMLMediaElement::closeCaptionTracksChanged): Deleted.
* html/HTMLMediaElement.h:
* html/shadow/MediaControlElements.cpp:
(WebCore::MediaControlClosedCaptionsTrackListElement::updateDisplay):
(WebCore::MediaControlClosedCaptionsTrackListElement::rebuildTrackListMenu):
* html/track/AudioTrack.cpp:
(WebCore::AudioTrack::~AudioTrack):
(WebCore::AudioTrack::setPrivate):
(WebCore::AudioTrack::isValidKind):
(WebCore::AudioTrack::willRemove):
(WebCore::AudioTrack::setEnabled): Deleted.
(WebCore::AudioTrack::updateKindFromPrivate): Deleted.
* html/track/AudioTrack.h:
(isType):
* html/track/AudioTrackList.cpp:
(AudioTrackList::append):
(AudioTrackList::item):
(AudioTrackList::getTrackById):
(AudioTrackList::eventTargetInterface): Deleted.
* html/track/AudioTrackList.h:
* html/track/InbandTextTrack.cpp:
(WebCore::InbandTextTrack::willRemove):
* html/track/TextTrack.cpp:
(WebCore::TextTrack::setKind):
(WebCore::TextTrack::trackIndex):
(WebCore::TextTrack::trackIndexRelativeToRenderedTracks):
(WebCore::TextTrack::setLanguage):
* html/track/TextTrack.h:
(isType):
* html/track/TextTrackList.cpp:
(TextTrackList::getTrackIndex):
(TextTrackList::getTrackIndexRelativeToRenderedTracks):
(TextTrackList::item):
(TextTrackList::getTrackById):
(TextTrackList::invalidateTrackIndexesAfterTrack):
(TextTrackList::append):
(TextTrackList::remove):
(TextTrackList::contains):
* html/track/TextTrackList.h:
* html/track/TrackBase.h:
* html/track/TrackEvent.cpp:
(WebCore::TrackEvent::TrackEvent):
* html/track/TrackEvent.h:
* html/track/TrackListBase.cpp:
(TrackListBase::remove):
(TrackListBase::contains):
(TrackListBase::scheduleTrackEvent):
(TrackListBase::scheduleAddTrackEvent):
(TrackListBase::scheduleRemoveTrackEvent):
* html/track/TrackListBase.h:
* html/track/VideoTrack.cpp:
(WebCore::VideoTrack::willRemove):
(WebCore::VideoTrack::setKind):
(WebCore::VideoTrack::setLanguage):
* html/track/VideoTrack.h:
(isType):
* html/track/VideoTrackList.cpp:
(VideoTrackList::append):
(VideoTrackList::item):
(VideoTrackList::getTrackById):
(VideoTrackList::selectedIndex):
* html/track/VideoTrackList.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@200943 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoUnreviewed gardening
jdiggs@igalia.com [Mon, 16 May 2016 14:50:07 +0000 (14:50 +0000)]
Unreviewed gardening

Re-mark a failing test as failing. Update the expectations file so
that it fails rather than passes unexpectedly.

* accessibility/gtk/title-and-alt-expected.txt:
* platform/gtk/TestExpectations:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@200942 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoCSSParser.cpp triggers -Wunused-parameter
mcatanzaro@igalia.com [Mon, 16 May 2016 14:40:57 +0000 (14:40 +0000)]
CSSParser.cpp triggers -Wunused-parameter
https://bugs.webkit.org/show_bug.cgi?id=157734

Reviewed by Csaba Osztrogonác.

* css/CSSParser.cpp:
(WebCore::isValidKeywordPropertyAndValue):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@200941 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years ago[Cairo] GraphicsContext3D::ImageExtractor should use the correct size for copying...
zandobersek@gmail.com [Mon, 16 May 2016 10:28:19 +0000 (10:28 +0000)]
[Cairo] GraphicsContext3D::ImageExtractor should use the correct size for copying non-image surfaces
https://bugs.webkit.org/show_bug.cgi?id=157580

Reviewed by Darin Adler.

GraphicsContext3D::ImageExtractor::extractImage() shouldn't use m_imageWidth
and m_imageHeight members when copying the non-image-backed Cairo surface into
the image-based replacement simply because these two are not initialized until
later in this method.

Instead, the size of the to-be-copied image should be queried via the
cairoSurfaceSize() utility function which properly handles Cairo surfaces of
different types.

* platform/graphics/cairo/GraphicsContext3DCairo.cpp:
(WebCore::GraphicsContext3D::ImageExtractor::extractImage):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@200940 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoREGRESSION (r199821): Large animated GIFs with slow network do not animate till the...
commit-queue@webkit.org [Mon, 16 May 2016 08:05:13 +0000 (08:05 +0000)]
REGRESSION (r199821): Large animated GIFs with slow network do not animate till the last frame
https://bugs.webkit.org/show_bug.cgi?id=157500
Source/WebCore:

Patch by Said Abou-Hallawa <sabouhallawa@apple.com> on 2016-05-16
Reviewed by Darin Adler.

r199821 relies on the assumption that the image frameCount can be retrieved
only once when the image size is available and it is not going to change. It
turned out this assumption is wrong for some animated GIFs. The frameCount
can change every time a new data block is decoded.

The fix is to invalidate the image cached metadata every time a new data
block is decoded.

Test: http/tests/misc/slow-loading-animated-image.html

* platform/graphics/BitmapImage.cpp:
(WebCore::BitmapImage::BitmapImage): Move the "fromImage" BitmapImage constructor
to BitmapImage.cpp. The ImageSource needs to be initialized differently for
this case from the decoder image case. Also get rid of the template constructor.
Most of the flags of the BitmapImage will be deleted with the asynchronous image
decoding work.

(WebCore::BitmapImage::cacheFrame): Delete unused flag m_hasUniformFrameSize.
It was added in r35761 and was mainly used by BitmapImage::currentFrameSize().
But this function itself was deleted in r172348.

(WebCore::BitmapImage::dataChanged): Invalidate the image metadata cache.

* platform/graphics/BitmapImage.h:
(WebCore::BitmapImage::BitmapImage): Deleted.
Delete unused flag m_hasUniformFrameSize. Also define new static functions in
FrameData to calculate platform dependent properties of an image.

* platform/graphics/ImageSource.cpp:
(WebCore::ImageSource::ImageSource): Add a new constructor for the "fromImage"
BitmapImage case. The purpose of this constructor is to keep m_needsUpdateMetadata
and m_maximumSubsamplingLevel be initialized as they are in the class definition.
This constructor will be used with the asynchronous image decoding work where the
ImageSource needs to cache the FrameData.

(WebCore::ImageSource::updateMetadata): If needsUpdateMetaData() is true, cache new metadata.

(WebCore::ImageSource::subsamplingLevelForScale): Call updateMetadata() with its new name.
(WebCore::ImageSource::frameCount): Ditto
(WebCore::ImageSource::cacheMetadata): Deleted.

* platform/graphics/ImageSource.h:
(WebCore::ImageSource::setNeedsUpdateMetadata): Invalidate the image cached metadata.

* platform/graphics/cairo/BitmapImageCairo.cpp:
(WebCore::NativeImage::size):
(WebCore::NativeImage::hasAlpha):
(WebCore::BitmapImage::BitmapImage): Deleted.
Delete the "fromImage" BitmapImage constructor from this file since it is now platform
independent and has been moved to BitmapImage.cpp. Also add the new NativeImage platform
dependent functions which are used by the new "fromImage" BitmapImage constructor.

* platform/graphics/cg/BitmapImageCG.cpp:
(WebCore::NativeImage::size):
(WebCore::NativeImage::hasAlpha):
(WebCore::BitmapImage::BitmapImage): Deleted.
Ditto.

LayoutTests:

Patch by Said Abou-Hallawa <sabouhallawa@apple.com> on 2016-05-16
Reviewed by Darin Adler.

Ensure the large animated gif will get the correct frameCount even if it
is loaded through a slow network.

* http/tests/misc/resources/large-animated.gif: Added.
This is a large animated GIF with 48 frames. The frameCount keeps changing
every time a new data block is decoded. Each frame except the last one lasts
for 20 ms. The last one lasts for 10 seconds which is similar to setting no-
loop to true. I created all the images myself. All the images except the last
one are rectangles filled with gradient color so their sizes can be large and
get the bug to repro.

* http/tests/misc/resources/slow-image-load.pl: Added.
* http/tests/misc/resources/slow-png-load.pl: Removed.
I wanted to use the slow-png-load.pl but only changing the image mimetype.
So I changed its name and I added a new query paramater called 'mimetype'.

* http/tests/misc/slow-loading-animated-image-expected.html: Added.
* http/tests/misc/slow-loading-animated-image.html: Added.
This test loads the animated image in 1 second. It waits for another 1 second
to ensure that all the frames are displayed. All we need is to capture the
last frame which is distinctly filled with solid green color.

* http/tests/misc/slow-loading-image-in-pattern.html:
* http/tests/misc/slow-loading-mask.html:
Use slow-image-load.pl and pass the image mime type.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@200939 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoSmall removal of useless code for MathML token elements
fred.wang@free.fr [Mon, 16 May 2016 06:23:36 +0000 (06:23 +0000)]
Small removal of useless code for MathML token elements
https://bugs.webkit.org/show_bug.cgi?id=157568

Patch by Frederic Wang <fwang@igalia.com> on 2016-05-15
Reviewed by Darin Adler.

We remove two pieces of code for MathML token elements with unclear purposes and that do not
have any effect on math rendering.

No new tests, behavior is unchanged.

* css/mathml.css:
(mtext): Deleted. This was probably a hack that is no longer needed with the latest math font
ascent/descent management.
* mathml/MathMLTextElement.cpp:
(WebCore::MathMLTextElement::createElementRenderer): Deleted. Removing this does not break
the mentioned test.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@200938 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years ago[GTK] Fix some bogus expectations broken by the previous commit
mcatanzaro@igalia.com [Mon, 16 May 2016 00:47:26 +0000 (00:47 +0000)]
[GTK] Fix some bogus expectations broken by the previous commit

* platform/gtk/TestExpectations:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@200937 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years ago[GTK] More unreviewed test expectations gardening
mcatanzaro@igalia.com [Mon, 16 May 2016 00:41:47 +0000 (00:41 +0000)]
[GTK] More unreviewed test expectations gardening

* platform/gtk/TestExpectations:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@200936 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years ago[GTK] More Unreviewed test expectations gardening
mcatanzaro@igalia.com [Mon, 16 May 2016 00:10:28 +0000 (00:10 +0000)]
[GTK] More Unreviewed test expectations gardening

* platform/gtk/TestExpectations:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@200935 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoUse more references in JS wrappers related code
cdumez@apple.com [Sun, 15 May 2016 23:30:11 +0000 (23:30 +0000)]
Use more references in JS wrappers related code
https://bugs.webkit.org/show_bug.cgi?id=157721

Reviewed by Darin Adler.

Source/WebCore:

- Use more references in JS wrappers related code.
- Avoid some refcounting churn when using toJSNewlyCreated() and in
  HTML/SVG elements factories by moving the Ref<> around.
- Add toJS() / toJSNewlyCreated() implementations for Text, XMLDocument
  and HTMLDocument. Now that toJSNewlyCreated() takes a Ref<>&& /
  RefPtr<>&& in, the compiler is no longer able to implicitly use the
  toJSNewlyCreated() implementation of a parent class for a subclass
  type (complains about calls being ambiguous). I thought about updating
  the bindings generator to always generate a toJSNewlyCreated() for
  all wrapper types (that would call the one on the parent class if the
  subclass does not have a proper implementation). However this would
  add some complexity to the bindings generator and it did not seem
  worth it because [NewObject] and therefore toJSNewlyCreated() is only
  used for a very limited set of subtypes. I have therefore decided to
  add the toJS() / toJSNewlyCreated() implementations for the subtypes
  in question. This also has the benefit of avoiding calling the more
  complex implementations on Node / Document.

* CMakeLists.txt:
* WebCore.xcodeproj/project.pbxproj:
* bindings/js/JSAnimationTimelineCustom.cpp:
(WebCore::toJS):
* bindings/js/JSAudioContextCustom.cpp:
(WebCore::constructJSAudioContext):
* bindings/js/JSBindingsAllInOne.cpp:
* bindings/js/JSBlobCustom.cpp:
(WebCore::constructJSBlob):
* bindings/js/JSCSSRuleCustom.cpp:
(WebCore::toJS):
* bindings/js/JSCSSValueCustom.cpp:
(WebCore::toJS):
* bindings/js/JSDOMBinding.h:
(WebCore::getCachedWrapper):
(WebCore::createWrapper):
(WebCore::wrap):
(WebCore::getExistingWrapper):
(WebCore::createNewWrapper):
(WebCore::toJS):
(WebCore::finiteInt32Value): Deleted.
* bindings/js/JSDocumentCustom.cpp:
(WebCore::createNewDocumentWrapper):
(WebCore::toJS):
(WebCore::cachedDocumentWrapper):
(WebCore::reportMemoryForFramelessDocument):
(WebCore::toJSNewlyCreated):
* bindings/js/JSDocumentCustom.h:
* bindings/js/JSDocumentFragmentCustom.cpp:
(WebCore::createNewDocumentFragmentWrapper):
(WebCore::toJSNewlyCreated):
(WebCore::toJS):
* bindings/js/JSElementCustom.cpp:
(WebCore::toJSNewlyCreated):
* bindings/js/JSElementCustom.h:
(WebCore::toJSNewlyCreated):
* bindings/js/JSEventCustom.cpp:
(WebCore::createNewEventWrapper):
(WebCore::toJS):
(WebCore::toJSNewlyCreated):
* bindings/js/JSFileCustom.cpp:
(WebCore::constructJSFile):
* bindings/js/JSHTMLCollectionCustom.cpp:
(WebCore::toJS):
* bindings/js/JSHTMLDocumentCustom.cpp:
(WebCore::createNewHTMLDocumentWrapper):
(WebCore::toJS):
(WebCore::toJSNewlyCreated):
* bindings/js/JSHTMLTemplateElementCustom.cpp:
(WebCore::JSHTMLTemplateElement::content):
* bindings/js/JSImageDataCustom.cpp:
(WebCore::toJS):
* bindings/js/JSNodeCustom.cpp:
(WebCore::createWrapperInline):
(WebCore::createWrapper):
(WebCore::toJSNewlyCreated):
(WebCore::getOutOfLineCachedWrapper):
* bindings/js/JSNodeCustom.h:
(WebCore::toJS):
* bindings/js/JSNodeListCustom.cpp:
(WebCore::createWrapper):
(WebCore::toJSNewlyCreated):
* bindings/js/JSNodeListCustom.h:
(WebCore::toJS):
* bindings/js/JSRTCPeerConnectionCustom.cpp:
(WebCore::constructJSRTCPeerConnection):
* bindings/js/JSRTCSessionDescriptionCustom.cpp:
(WebCore::constructJSRTCSessionDescription):
* bindings/js/JSSVGPathSegCustom.cpp:
(WebCore::toJS):
* bindings/js/JSStyleSheetCustom.cpp:
(WebCore::toJS):
* bindings/js/JSTextCustom.cpp: Copied from Source/WebCore/dom/XMLDocument.idl.
(WebCore::createNewTextWrapper):
(WebCore::toJS):
(WebCore::toJSNewlyCreated):
* bindings/js/JSTextTrackCueCustom.cpp:
(WebCore::toJS):
* bindings/js/JSTrackCustom.cpp:
(WebCore::toJS):
* bindings/js/JSXMLDocumentCustom.cpp: Copied from Source/WebCore/bindings/js/JSDocumentFragmentCustom.cpp.
(WebCore::createNewXMLDocumentWrapper):
(WebCore::toJS):
(WebCore::toJSNewlyCreated):
* bindings/scripts/CodeGeneratorJS.pm:
(GenerateHeader):
(GenerateImplementation):
* bindings/scripts/test/JS/*: Rebaseline bindings tests.
* dom/Element.cpp:
(WebCore::Element::setInnerHTML):
* dom/Text.idl:
* dom/XMLDocument.idl:
* dom/make_names.pl:
(printWrapperFunctions):
(printWrapperFactoryCppFile):
(printWrapperFactoryHeaderFile):
* editing/MarkupAccumulator.cpp:
(WebCore::MarkupAccumulator::serializeNodesWithNamespaces):
* html/HTMLDocument.idl:
* html/HTMLTemplateElement.cpp:
(WebCore::HTMLTemplateElement::content):
(WebCore::HTMLTemplateElement::cloneNodeInternal):
* html/HTMLTemplateElement.h:
* html/parser/HTMLConstructionSite.cpp:
(WebCore::insert):
(WebCore::HTMLConstructionSite::insertTextNode):
(WebCore::HTMLConstructionSite::ownerDocumentForCurrentNode):
* inspector/InspectorDOMAgent.cpp:
(WebCore::InspectorDOMAgent::buildObjectForNode):
* xml/parser/XMLDocumentParserLibxml2.cpp:
(WebCore::XMLDocumentParser::startElementNs):

Source/WTF:

Add new static_reference_cast() overload that takes a Ref<U>&& in
in order to downcast Ref<> types without causing ref-counting
churn.

* wtf/Ref.h:
(WTF::static_reference_cast):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@200934 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoDFG::Plan shouldn't read from its VM once it's been cancelled
fpizlo@apple.com [Sun, 15 May 2016 23:08:21 +0000 (23:08 +0000)]
DFG::Plan shouldn't read from its VM once it's been cancelled
https://bugs.webkit.org/show_bug.cgi?id=157726

Reviewed by Saam Barati.

Plan::vm was a reference, not a pointer, and so wasn't nulled by Plan::cancel(). So, a
cancelled plan may have a dangling pointer to a VM: we could delete the VM after cancelling
the plan.

Prior to http://trac.webkit.org/changeset/200705, this was probably fine because nobody
would read Plan::vm if the plan was cancelled. But r200705 changed that. It was a hard
regression to spot because usually a cancelled plan will still refer to a valid VM.

This change fixes the regression and makes it a lot easier to spot the regression in the
future. Plan::vm is now a pointer and we null it in Plan::cancel(). Now if you make this
mistake, you will get a crash anytime the Plan is cancelled, not just anytime the plan is
cancelled and the VM gets deleted. Also, it's now very clear what to do when you want to
use Plan::vm on the cancel path: you can null-check vm; if it's null, assume the worst.

Because we null the VM of a cancelled plan, we cannot have Safepoint::vm() return the
plan's VM anymore. That's because when we cancel a plan that is at a safepoint, we use the
safepoint's VM to determine whether this is one of our safepoints *after* the plan is
already cancelled. So, Safepoint now has its own copy of m_vm, and that copy gets nulled
when the Safepoint is cancelled. The Safepoint's m_vm will be nulled moments after Plan's
vm gets nulled (see Worklist::removeDeadPlans(), which has a cancel path for Plans in one
loop and a cancel path for Safepoints in the loop after it).

* dfg/DFGJITFinalizer.cpp:
(JSC::DFG::JITFinalizer::finalizeCommon):
* dfg/DFGPlan.cpp:
(JSC::DFG::Plan::Plan):
(JSC::DFG::Plan::computeCompileTimes):
(JSC::DFG::Plan::reportCompileTimes):
(JSC::DFG::Plan::compileInThreadImpl):
(JSC::DFG::Plan::reallyAdd):
(JSC::DFG::Plan::notifyCompiling):
(JSC::DFG::Plan::finalizeWithoutNotifyingCallback):
(JSC::DFG::Plan::cancel):
* dfg/DFGPlan.h:
(JSC::DFG::Plan::canTierUpAndOSREnter):
* dfg/DFGSafepoint.cpp:
(JSC::DFG::Safepoint::cancel):
(JSC::DFG::Safepoint::vm):
* dfg/DFGSafepoint.h:
* dfg/DFGWorklist.cpp:
(JSC::DFG::Worklist::isActiveForVM):
(JSC::DFG::Worklist::waitUntilAllPlansForVMAreReady):
(JSC::DFG::Worklist::removeAllReadyPlansForVM):
(JSC::DFG::Worklist::rememberCodeBlocks):
(JSC::DFG::Worklist::visitWeakReferences):
(JSC::DFG::Worklist::removeDeadPlans):
(JSC::DFG::Worklist::runThread):
* ftl/FTLJITFinalizer.cpp:
(JSC::FTL::JITFinalizer::finalizeFunction):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@200933 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoSomething tiny left out of the last patch.
darin@apple.com [Sun, 15 May 2016 22:13:53 +0000 (22:13 +0000)]
Something tiny left out of the last patch.

* WebCore.xcodeproj/project.pbxproj: Allow Xcode to update this file.
* editing/ApplyStyleCommand.cpp:
(WebCore::ApplyStyleCommand::applyInlineStyleChange): Removed unneeded null check; all
code paths already have null checks.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@200932 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoMore CTTE and other cleanups for HTML editing header
darin@apple.com [Sun, 15 May 2016 22:08:52 +0000 (22:08 +0000)]
More CTTE and other cleanups for HTML editing header
https://bugs.webkit.org/show_bug.cgi?id=157722

Reviewed by Chris Dumez.

Source/WebCore:

These changes are mostly things Chris asked for in his review of my last patch
or other related style and small efficiency improvements.

- using nullptr instead of 0
- using is<Text> instead of isTextNode and is<> instead of hasTagName
- changing argument type of functions like editingIgnoresContent to a reference
- using { } instead of constructor calls for classes like Position
- using words like position intead of abbreviations like pos
- using auto rather than explicit types
- rename functions that return true for HTMLElement to express that in the function names

* dom/Position.cpp:
(WebCore::nextRenderedEditable): nullptr
(WebCore::previousRenderedEditable): nullptr
(WebCore::Position::Position): is<>, reference
(WebCore::Position::moveToPosition): reference
(WebCore::Position::offsetForPositionAfterAnchor): Added a null assertion and
null check that replaces the one that used to be in lastOffsetForEditing.
(WebCore::Position::parentAnchoredEquivalent): { }, reference
(WebCore::Position::computeNodeBeforePosition): nullptr
(WebCore::Position::computeNodeAfterPosition): nullptr
(WebCore::Position::anchorTypeForLegacyEditingPosition): reference
(WebCore::Position::previous): reference
(WebCore::Position::next): reference
(WebCore::Position::parentEditingBoundary): nullptr
(WebCore::Position::previousCharacterPosition): { }, pos -> position
(WebCore::Position::nextCharacterPosition): { }, pos -> position
(WebCore::endsOfNodeAreVisuallyDistinctPositions): reference
(WebCore::Position::upstream): { }, pos -> position
(WebCore::Position::downstream): { }, pos -> position
(WebCore::Position::rootUserSelectAllForNode): nullptr
(WebCore::Position::isCandidate): auto, reference, is<>
(WebCore::Position::rendersInDifferentPosition): pos -> position, auto, is<>
(WebCore::Position::leadingWhitespacePosition): { }, is<>
(WebCore::Position::trailingWhitespacePosition): { }
(WebCore::Position::getInlineBoxAndOffset): reference
(WebCore::Position::equals): reference, is<>

* dom/PositionIterator.cpp:
(WebCore::PositionIterator::operator Position): reference
(WebCore::PositionIterator::isCandidate): reference

* editing/ApplyStyleCommand.cpp:
(WebCore::ApplyStyleCommand::applyRelativeFontStyleChange): is<>
(WebCore::ApplyStyleCommand::applyInlineStyleToNodeRange): is<>, reference
(WebCore::ApplyStyleCommand::removeConflictingInlineStyleFromRun): reference
(WebCore::ApplyStyleCommand::removeInlineStyle): auto, is<>, reference
(WebCore::ApplyStyleCommand::nodeFullySelected): is<>
(WebCore::ApplyStyleCommand::nodeFullyUnselected): is<>
(WebCore::ApplyStyleCommand::splitTextAtStart): is<>
(WebCore::ApplyStyleCommand::splitTextElementAtStart): is<>
(WebCore::ApplyStyleCommand::splitTextElementAtEnd): is<>
(WebCore::ApplyStyleCommand::isValidCaretPositionInTextNode): is<>
(WebCore::ApplyStyleCommand::surroundNodeRangeWithElement): Removed unneeded
is<Element> calls before calling areIdenticalElements.
(WebCore::ApplyStyleCommand::positionToComputeInlineStyleChange): Tightened
up logic for the loop to avoid redundant checks.
(WebCore::ApplyStyleCommand::applyInlineStyleChange): auto

* editing/CompositeEditCommand.cpp:
(WebCore::CompositeEditCommand::insertNodeAt): reference
(WebCore::CompositeEditCommand::appendNode): reference
(WebCore::CompositeEditCommand::moveParagraphs): isListHTMLElement

* editing/DeleteSelectionCommand.cpp:
(WebCore::DeleteSelectionCommand::handleGeneralDelete): reference

* editing/FormatBlockCommand.cpp: Fixed order of includes.
(WebCore::enclosingBlockToSplitTreeTo): isListHTMLElement

* editing/InsertParagraphSeparatorCommand.cpp:
(WebCore::InsertParagraphSeparatorCommand::doApply): reference

* editing/ModifySelectionListLevel.cpp:
(WebCore::getStartEndListChildren): isListHTMLElement
(WebCore::IncreaseSelectionListLevelCommand::doApply):isListHTMLElement
(WebCore::canDecreaseListLevel):isListHTMLElement

* editing/RemoveNodeCommand.cpp: Removed unneeded include.

* editing/ReplaceNodeWithSpanCommand.cpp:
(WebCore::ReplaceNodeWithSpanCommand::doApply): Use HTMLSpanElement::create.

* editing/ReplaceSelectionCommand.cpp:
(WebCore::isInterchangeNewlineNode): is<>, downcast<>
(WebCore::isInterchangeConvertedSpaceSpan): is<>, downcast<>
(WebCore::positionAvoidingPrecedingNodes): pos -> position, reference
(WebCore::ReplaceSelectionCommand::doApply): isListHTMLElement
(WebCore::ReplaceSelectionCommand::insertAsListItems): isListHTMLElement

* editing/SpellChecker.cpp: Removed unneeded include.

* editing/TextIterator.cpp:
(WebCore::SimplifiedBackwardsTextIterator::advance): reference

* editing/VisibleSelection.cpp:
(WebCore::VisibleSelection::selectionFromContentsOfNode): reference

* editing/VisibleUnits.cpp:
(WebCore::previousLinePosition): reference
(WebCore::nextLinePosition): reference
(WebCore::findStartOfParagraph): reference
(WebCore::findEndOfParagraph): reference

* editing/htmlediting.cpp:
(WebCore::canHaveChildrenForEditing): Moved here from header file so we don't
have to include "Text.h". Not critical to inline this.
(WebCore::isAtomicNode): reference
(WebCore::lastOffsetForEditing): reference
(WebCore::stringWithRebalancedWhitespace): Changed to use StringBuilder instead
of Vector<UChar> so this won't turn 8-bit strings into 16-bit strings. Changed
to not use the StringBuilder at all until we encounter a character that needs
to be changed.
(WebCore::isSpecialHTMLElement): Renamed to make it clear that it only can return
true for a Node that is an HTMLElement.
(WebCore::firstInSpecialElement): isSpecialHTMLElement
(WebCore::lastInSpecialElement): isSpecialHTMLElement
(WebCore::isListHTMLElement): Renamed to make it clear that it can only return
true for a Node that is an HTMLElement.
(WebCore::isListItem): isListHTMLElement, is<>
(WebCore::enclosingNodeOfType): p -> position, nullptr
(WebCore::highestEnclosingNodeOfType): p -> position
(WebCore::enclosingListChild): isListHTMLElement, is<>
(WebCore::embeddedSublist): isListHTMLElement
(WebCore::appendedSublist):isListHTMLElement
(WebCore::caretMaxOffset): is<>
(WebCore::caretRendersInsideNode): changed to take a reference
(WebCore::rendererForCaretPainting): reference

* editing/htmlediting.h: Updated for name changes and type changes above.

* editing/markup.cpp:
(WebCore::StyledMarkupAccumulator::traverseNodesForSerialization): reference
(WebCore::highestAncestorToWrapMarkup): auto, reference, isListHTMLElement

* html/HTMLSpanElement.cpp:
(WebCore::HTMLSpanElement::create): Added overload that does not require a tag name.
* html/HTMLSpanElement.h: Ditto.

* page/DOMSelection.cpp: Removed unneeded includes.

* rendering/RenderBox.cpp:
(WebCore::RenderBox::localCaretRect): reference

Source/WebKit/ios:

* WebCoreSupport/WebFrameIOS.mm:
(-[WebFrame previousUnperturbedDictationResultBoundaryFromPosition:]): auto
(-[WebFrame nextUnperturbedDictationResultBoundaryFromPosition:]): auto

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@200931 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years ago[GTK] Unreviewed test expectations gardening
mcatanzaro@igalia.com [Sun, 15 May 2016 22:05:56 +0000 (22:05 +0000)]
[GTK] Unreviewed test expectations gardening

* TestExpectations:
* platform/gtk/TestExpectations:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@200930 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years ago[GTK] Unreviewed test expectations gardening
mcatanzaro@igalia.com [Sun, 15 May 2016 22:05:05 +0000 (22:05 +0000)]
[GTK] Unreviewed test expectations gardening

* TestExpectations:
* platform/gtk/TestExpectations:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@200929 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoModernize Intl constructors; using InternalFunction::createSubclassStructure
utatane.tea@gmail.com [Sun, 15 May 2016 21:11:27 +0000 (21:11 +0000)]
Modernize Intl constructors; using InternalFunction::createSubclassStructure
https://bugs.webkit.org/show_bug.cgi?id=157082

Reviewed by Darin Adler.

Previously, Intl constructors retrieve "prototype" to inherit the "new.target".
At that time, this mis-assumed that getDirect() always returns meaningful JS value.
Actually, it returns an empty value if a property does not exist.

Instead of fixing this assertion, we now use InternalFunction::createSubclassStructure
in Intl constructors. It is modern and preferable way since it can cache the derived
structures in InternalFunction.

This patch also cleans up the workaround in Intl.NumberFormat and Intl.DateTimeFormat.
Those code are largely duplicate. This is now extracted into
constructIntlInstanceWithWorkaroundForLegacyIntlConstructor. This clean up does not
have any behavior changes. They are already tested in LayoutTests/js/intl-datetimeformat
and LayoutTests/js/intl-numberformat.

* JavaScriptCore.xcodeproj/project.pbxproj:
* runtime/IntlCollator.cpp:
(JSC::IntlCollator::create):
* runtime/IntlCollator.h:
* runtime/IntlCollatorConstructor.cpp:
(JSC::constructIntlCollator):
(JSC::callIntlCollator):
* runtime/IntlDateTimeFormat.cpp:
(JSC::IntlDateTimeFormat::create):
* runtime/IntlDateTimeFormat.h:
* runtime/IntlDateTimeFormatConstructor.cpp:
(JSC::constructIntlDateTimeFormat):
(JSC::callIntlDateTimeFormat):
* runtime/IntlDateTimeFormatPrototype.cpp:
(JSC::IntlDateTimeFormatPrototypeGetterFormat):
(JSC::IntlDateTimeFormatPrototypeFuncResolvedOptions):
* runtime/IntlNumberFormat.cpp:
(JSC::IntlNumberFormat::create):
* runtime/IntlNumberFormat.h:
* runtime/IntlNumberFormatConstructor.cpp:
(JSC::constructIntlNumberFormat):
(JSC::callIntlNumberFormat):
* runtime/IntlNumberFormatPrototype.cpp:
(JSC::IntlNumberFormatPrototypeGetterFormat):
(JSC::IntlNumberFormatPrototypeFuncResolvedOptions):
* runtime/IntlObjectInlines.h: Added.
(JSC::constructIntlInstanceWithWorkaroundForLegacyIntlConstructor):
* tests/stress/intl-constructors-with-proxy.js: Added.
(shouldBe):
(throw.new.Error.Empty):
(throw.new.Error):
(shouldBe.Empty):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@200928 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoUnreviewed, rolling out r200925.
commit-queue@webkit.org [Sun, 15 May 2016 15:36:57 +0000 (15:36 +0000)]
Unreviewed, rolling out r200925.
https://bugs.webkit.org/show_bug.cgi?id=157725

Does not build if MEDIA_SESSION is enabled (Requested by
cdumez_ on #webkit).

Reverted changeset:

"Use new Web IDL dictionary support for
MediaSession.setMetadata()"
https://bugs.webkit.org/show_bug.cgi?id=157711
http://trac.webkit.org/changeset/200925

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@200927 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoRemove Frédéric Wang's blog from the planet
fred.wang@free.fr [Sun, 15 May 2016 10:34:51 +0000 (10:34 +0000)]
Remove Frédéric Wang's blog from the planet
https://bugs.webkit.org/show_bug.cgi?id=157724

Reviewed by Philippe Normand.

* config.ini: Remove my feed url since my homepage changed and it is included twice after r169566.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@200926 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoUse new Web IDL dictionary support for MediaSession.setMetadata()
cdumez@apple.com [Sat, 14 May 2016 23:52:01 +0000 (23:52 +0000)]
Use new Web IDL dictionary support for MediaSession.setMetadata()
https://bugs.webkit.org/show_bug.cgi?id=157711

Reviewed by Eric Carlson.

Use new Web IDL dictionary support for MediaSession.setMetadata().

No new tests, no intended Web-exposed behavior change.

* Modules/mediasession/MediaSession.cpp:
(WebCore::MediaSession::setMetadata):
(WebCore::MediaSession::deactivate): Deleted.
* Modules/mediasession/MediaSession.h:
* Modules/mediasession/MediaSession.idl:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@200925 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoRemove LegacyProfiler
commit-queue@webkit.org [Sat, 14 May 2016 22:17:13 +0000 (22:17 +0000)]
Remove LegacyProfiler
https://bugs.webkit.org/show_bug.cgi?id=153565

Patch by Joseph Pecoraro <pecoraro@apple.com> on 2016-05-14
Reviewed by Mark Lam.

.:

* ManualTests/inspector/profiler-test-call.html: Removed.
* ManualTests/inspector/profiler-test-many-calls-in-the-same-scope.html: Removed.

Source/JavaScriptCore:

JavaScriptCore now provides a sampling profiler and it is enabled
by all ports. Web Inspector switched months ago to using the
sampling profiler and displaying its data. Remove the legacy
profiler, as it is no longer being used by anything other then
console.profile and tests. We will update console.profile's
behavior soon to have new behavior and use the sampling data.

* API/JSProfilerPrivate.cpp: Removed.
* API/JSProfilerPrivate.h: Removed.
* CMakeLists.txt:
* JavaScriptCore.xcodeproj/project.pbxproj:
* bytecode/BytecodeList.json:
* bytecode/BytecodeUseDef.h:
(JSC::computeUsesForBytecodeOffset): Deleted.
(JSC::computeDefsForBytecodeOffset): Deleted.
* bytecode/CodeBlock.cpp:
(JSC::CodeBlock::dumpBytecode): Deleted.
* bytecode/UnlinkedFunctionExecutable.cpp:
(JSC::generateUnlinkedFunctionCodeBlock):
(JSC::UnlinkedFunctionExecutable::unlinkedCodeBlockFor):
* bytecode/UnlinkedFunctionExecutable.h:
* bytecompiler/BytecodeGenerator.cpp:
(JSC::BytecodeGenerator::BytecodeGenerator):
(JSC::BytecodeGenerator::emitCall):
(JSC::BytecodeGenerator::emitCallVarargs):
(JSC::BytecodeGenerator::emitCallVarargsInTailPosition):
(JSC::BytecodeGenerator::emitConstructVarargs):
(JSC::BytecodeGenerator::emitConstruct):
* bytecompiler/BytecodeGenerator.h:
(JSC::CallArguments::profileHookRegister): Deleted.
(JSC::BytecodeGenerator::shouldEmitProfileHooks): Deleted.
* bytecompiler/NodesCodegen.cpp:
(JSC::CallFunctionCallDotNode::emitBytecode):
(JSC::ApplyFunctionCallDotNode::emitBytecode):
(JSC::CallArguments::CallArguments): Deleted.
* dfg/DFGAbstractInterpreterInlines.h:
(JSC::DFG::AbstractInterpreter<AbstractStateType>::executeEffects): Deleted.
* dfg/DFGByteCodeParser.cpp:
(JSC::DFG::ByteCodeParser::parseBlock): Deleted.
* dfg/DFGCapabilities.cpp:
(JSC::DFG::capabilityLevel): Deleted.
* dfg/DFGClobberize.h:
(JSC::DFG::clobberize): Deleted.
* dfg/DFGDoesGC.cpp:
(JSC::DFG::doesGC): Deleted.
* dfg/DFGFixupPhase.cpp:
(JSC::DFG::FixupPhase::fixupNode): Deleted.
* dfg/DFGNodeType.h:
* dfg/DFGPredictionPropagationPhase.cpp:
* dfg/DFGSafeToExecute.h:
(JSC::DFG::safeToExecute): Deleted.
* dfg/DFGSpeculativeJIT32_64.cpp:
(JSC::DFG::SpeculativeJIT::compile): Deleted.
* dfg/DFGSpeculativeJIT64.cpp:
(JSC::DFG::SpeculativeJIT::compile): Deleted.
* inspector/InjectedScriptBase.cpp:
(Inspector::InjectedScriptBase::callFunctionWithEvalEnabled):
* inspector/protocol/Timeline.json:
* interpreter/Interpreter.cpp:
(JSC::UnwindFunctor::operator()): Deleted.
(JSC::Interpreter::execute): Deleted.
(JSC::Interpreter::executeCall): Deleted.
(JSC::Interpreter::executeConstruct): Deleted.
* jit/JIT.cpp:
(JSC::JIT::privateCompileMainPass): Deleted.
* jit/JIT.h:
* jit/JITOpcodes.cpp:
(JSC::JIT::emit_op_profile_will_call): Deleted.
(JSC::JIT::emit_op_profile_did_call): Deleted.
* jit/JITOpcodes32_64.cpp:
(JSC::JIT::emit_op_profile_will_call): Deleted.
(JSC::JIT::emit_op_profile_did_call): Deleted.
* jit/JITOperations.cpp:
* jit/JITOperations.h:
* jsc.cpp:
* llint/LLIntSlowPaths.cpp:
(JSC::LLInt::LLINT_SLOW_PATH_DECL): Deleted.
* llint/LLIntSlowPaths.h:
* llint/LowLevelInterpreter.asm:
* parser/ParserModes.h:
* profiler/CallIdentifier.h: Removed.
* profiler/LegacyProfiler.cpp: Removed.
* profiler/LegacyProfiler.h: Removed.
* profiler/Profile.cpp: Removed.
* profiler/Profile.h: Removed.
* profiler/ProfileGenerator.cpp: Removed.
* profiler/ProfileGenerator.h: Removed.
* profiler/ProfileNode.cpp: Removed.
* profiler/ProfileNode.h: Removed.
* profiler/ProfilerJettisonReason.cpp:
(WTF::printInternal): Deleted.
* profiler/ProfilerJettisonReason.h:
* runtime/CodeCache.cpp:
(JSC::CodeCache::getGlobalCodeBlock):
(JSC::CodeCache::getProgramCodeBlock):
(JSC::CodeCache::getEvalCodeBlock):
(JSC::CodeCache::getModuleProgramCodeBlock):
* runtime/CodeCache.h:
* runtime/Executable.cpp:
(JSC::ScriptExecutable::newCodeBlockFor):
* runtime/JSGlobalObject.cpp:
(JSC::JSGlobalObject::createProgramCodeBlock):
(JSC::JSGlobalObject::createEvalCodeBlock):
(JSC::JSGlobalObject::createModuleProgramCodeBlock):
(JSC::JSGlobalObject::~JSGlobalObject): Deleted.
(JSC::JSGlobalObject::hasLegacyProfiler): Deleted.
* runtime/JSGlobalObject.h:
(JSC::JSGlobalObject::supportsLegacyProfiling): Deleted.
* runtime/Options.h:
* runtime/VM.cpp:
(JSC::VM::VM): Deleted.
(JSC::SetEnabledProfilerFunctor::operator()): Deleted.
(JSC::VM::setEnabledProfiler): Deleted.
* runtime/VM.h:
(JSC::VM::enabledProfiler): Deleted.
(JSC::VM::enabledProfilerAddress): Deleted.

Source/WebCore:

* CMakeLists.txt:
* DerivedSources.cpp:
* DerivedSources.make:
* ForwardingHeaders/profiler/LegacyProfiler.h: Removed.
* ForwardingHeaders/profiler/Profile.h: Removed.
* ForwardingHeaders/profiler/ProfileNode.h: Removed.
* WebCore.xcodeproj/project.pbxproj:
* bindings/js/JSCustomXPathNSResolver.cpp:
* bindings/js/JSDOMWindowBase.cpp:
(WebCore::JSDOMWindowBase::supportsLegacyProfiling): Deleted.
(WebCore::JSDOMWindowBase::supportsRichSourceInfo): Deleted.
* bindings/js/JSDOMWindowBase.h:
* bindings/js/JSWorkerGlobalScopeBase.cpp:
(WebCore::JSWorkerGlobalScopeBase::supportsLegacyProfiling): Deleted.
* bindings/js/JSWorkerGlobalScopeBase.h:
* bindings/js/ScriptCachedFrameData.cpp:
* bindings/js/ScriptController.cpp:
(WebCore::ScriptController::clearWindowShell): Deleted.
* bindings/js/ScriptProfile.h: Removed.
* bindings/js/ScriptProfileNode.h: Removed.
* bindings/scripts/CodeGeneratorJS.pm:
(AddClassForwardIfNeeded): Deleted.
* bindings/scripts/test/JS/JSTestObj.cpp:
* bindings/scripts/test/TestObj.idl:
* css/CSSParser.cpp:
* dom/Document.cpp:
* inspector/InspectorConsoleInstrumentation.h:
(WebCore::InspectorInstrumentation::stopProfiling):
* inspector/InspectorController.cpp:
(WebCore::InspectorController::InspectorController):
(WebCore::InspectorController::legacyProfilerEnabled): Deleted.
(WebCore::InspectorController::setLegacyProfilerEnabled): Deleted.
* inspector/InspectorController.h:
* inspector/InspectorInstrumentation.cpp:
(WebCore::InspectorInstrumentation::stopProfilingImpl):
* inspector/InspectorInstrumentation.h:
* inspector/InspectorTimelineAgent.cpp:
(WebCore::InspectorTimelineAgent::startFromConsole):
(WebCore::InspectorTimelineAgent::stopFromConsole):
* inspector/InspectorTimelineAgent.h:
* inspector/PageDebuggerAgent.cpp:
* inspector/PageRuntimeAgent.cpp:
* inspector/ScriptProfile.idl: Removed.
* inspector/ScriptProfileNode.idl: Removed.
* inspector/TimelineRecordFactory.cpp:
(WebCore::buildAggregateCallInfoInspectorObject): Deleted.
(WebCore::buildInspectorObject): Deleted.
(WebCore::buildProfileInspectorObject): Deleted.
(WebCore::TimelineRecordFactory::appendProfile): Deleted.
* page/DOMWindow.cpp:
* page/Page.cpp:
* page/PageConsoleClient.cpp:
(WebCore::PageConsoleClient::profileEnd):
(WebCore::PageConsoleClient::clearProfiles): Deleted.
* page/PageConsoleClient.h:
* testing/Internals.cpp:
(WebCore::Internals::resetToConsistentState): Deleted.
(WebCore::Internals::consoleProfiles): Deleted.
(WebCore::Internals::setLegacyJavaScriptProfilingEnabled): Deleted.
* testing/Internals.h:
* testing/Internals.idl:
* testing/js/WebCoreTestSupport.cpp:
* xml/XSLStyleSheetLibxslt.cpp:
* xml/XSLTProcessorLibxslt.cpp:

Source/WebKit/win:

* Interfaces/IWebInspector.idl:
* WebView.cpp:
* WebInspector.cpp:
(WebInspector::isJavaScriptProfilingEnabled): Deleted.
(WebInspector::setJavaScriptProfilingEnabled): Deleted.
* WebInspector.h:

LayoutTests:

* fast/profiler/anonymous-event-handler-expected.txt: Removed.
* fast/profiler/anonymous-event-handler.html: Removed.
* fast/profiler/anonymous-function-called-from-different-contexts-expected.txt: Removed.
* fast/profiler/anonymous-function-called-from-different-contexts.html: Removed.
* fast/profiler/anonymous-function-calls-built-in-functions-expected.txt: Removed.
* fast/profiler/anonymous-function-calls-built-in-functions.html: Removed.
* fast/profiler/anonymous-function-calls-eval-expected.txt: Removed.
* fast/profiler/anonymous-function-calls-eval.html: Removed.
* fast/profiler/anonymous-functions-with-display-names-expected.txt: Removed.
* fast/profiler/anonymous-functions-with-display-names.html: Removed.
* fast/profiler/apply-expected.txt: Removed.
* fast/profiler/apply.html: Removed.
* fast/profiler/built-in-function-calls-anonymous-expected.txt: Removed.
* fast/profiler/built-in-function-calls-anonymous.html: Removed.
* fast/profiler/built-in-function-calls-user-defined-function-expected.txt: Removed.
* fast/profiler/built-in-function-calls-user-defined-function.html: Removed.
* fast/profiler/call-expected.txt: Removed.
* fast/profiler/call-register-leak-expected.txt: Removed.
* fast/profiler/call-register-leak.html: Removed.
* fast/profiler/call.html: Removed.
* fast/profiler/calling-the-function-that-started-the-profiler-from-another-scope-expected.txt: Removed.
* fast/profiler/calling-the-function-that-started-the-profiler-from-another-scope.html: Removed.
* fast/profiler/compare-multiple-profiles-expected.txt: Removed.
* fast/profiler/compare-multiple-profiles.html: Removed.
* fast/profiler/constructor-expected.txt: Removed.
* fast/profiler/constructor.html: Removed.
* fast/profiler/dead-time-expected.txt: Removed.
* fast/profiler/dead-time.html: Removed.
* fast/profiler/document-dot-write-expected.txt: Removed.
* fast/profiler/document-dot-write.html: Removed.
* fast/profiler/event-handler-expected.txt: Removed.
* fast/profiler/event-handler.html: Removed.
* fast/profiler/execution-context-and-eval-on-same-line-expected.txt: Removed.
* fast/profiler/execution-context-and-eval-on-same-line.html: Removed.
* fast/profiler/inline-event-handler-expected.txt: Removed.
* fast/profiler/inline-event-handler.html: Removed.
* fast/profiler/many-calls-in-the-same-scope-expected.txt: Removed.
* fast/profiler/many-calls-in-the-same-scope.html: Removed.
* fast/profiler/multiple-and-different-scoped-anonymous-function-calls-expected.txt: Removed.
* fast/profiler/multiple-and-different-scoped-anonymous-function-calls.html: Removed.
* fast/profiler/multiple-and-different-scoped-function-calls-expected.txt: Removed.
* fast/profiler/multiple-and-different-scoped-function-calls.html: Removed.
* fast/profiler/multiple-anonymous-functions-called-from-the-same-function-expected.txt: Removed.
* fast/profiler/multiple-anonymous-functions-called-from-the-same-function.html: Removed.
* fast/profiler/multiple-frames-expected.txt: Removed.
* fast/profiler/multiple-frames.html: Removed.
* fast/profiler/named-functions-with-display-names-expected.txt: Removed.
* fast/profiler/named-functions-with-display-names.html: Removed.
* fast/profiler/nested-anonymous-functon-expected.txt: Removed.
* fast/profiler/nested-anonymous-functon.html: Removed.
* fast/profiler/nested-start-and-stop-profiler-expected.txt: Removed.
* fast/profiler/nested-start-and-stop-profiler.html: Removed.
* fast/profiler/no-execution-context-expected.txt: Removed.
* fast/profiler/no-execution-context.html: Removed.
* fast/profiler/one-execution-context-expected.txt: Removed.
* fast/profiler/one-execution-context.html: Removed.
* fast/profiler/profile-calls-in-included-file-expected.txt: Removed.
* fast/profiler/profile-calls-in-included-file.html: Removed.
* fast/profiler/profile-with-no-title-expected.txt: Removed.
* fast/profiler/profile-with-no-title.html: Removed.
* fast/profiler/profiling-from-a-nested-location-but-stop-profiling-outside-the-nesting-expected.txt: Removed.
* fast/profiler/profiling-from-a-nested-location-but-stop-profiling-outside-the-nesting.html: Removed.
* fast/profiler/profiling-from-a-nested-location-expected.txt: Removed.
* fast/profiler/profiling-from-a-nested-location.html: Removed.
* fast/profiler/resources/other-frame.html: Removed.
* fast/profiler/resources/other-window.html: Removed.
* fast/profiler/resources/profiler-test-JS-resources.js: Removed.
* fast/profiler/simple-event-call-expected.txt: Removed.
* fast/profiler/simple-event-call.html: Removed.
* fast/profiler/simple-no-level-change-expected.txt: Removed.
* fast/profiler/simple-no-level-change.html: Removed.
* fast/profiler/start-and-stop-profiler-multiple-times-expected.txt: Removed.
* fast/profiler/start-and-stop-profiler-multiple-times.html: Removed.
* fast/profiler/start-and-stop-profiling-in-the-same-function-expected.txt: Removed.
* fast/profiler/start-and-stop-profiling-in-the-same-function.html: Removed.
* fast/profiler/start-but-dont-stop-profiling-expected.txt: Removed.
* fast/profiler/start-but-dont-stop-profiling.html: Removed.
* fast/profiler/stop-profiling-after-setTimeout-expected.txt: Removed.
* fast/profiler/stop-profiling-after-setTimeout.html: Removed.
* fast/profiler/stop-then-function-call-expected.txt: Removed.
* fast/profiler/stop-then-function-call.html: Removed.
* fast/profiler/throw-exception-from-eval-expected.txt: Removed.
* fast/profiler/throw-exception-from-eval.html-disabled: Removed.
* fast/profiler/two-execution-contexts-expected.txt: Removed.
* fast/profiler/two-execution-contexts.html: Removed.
* fast/profiler/user-defined-function-calls-built-in-functions-expected.txt: Removed.
* fast/profiler/user-defined-function-calls-built-in-functions.html: Removed.
* fast/profiler/window-dot-eval-expected.txt: Removed.
* fast/profiler/window-dot-eval.html: Removed.
* platform/efl/TestExpectations:
* platform/gtk/TestExpectations:
* platform/ios-simulator/TestExpectations:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@200924 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoToT WebKit doesn't show tooltip on perf dashboard's summary page
rniwa@webkit.org [Sat, 14 May 2016 21:07:28 +0000 (21:07 +0000)]
ToT WebKit doesn't show tooltip on perf dashboard's summary page
https://bugs.webkit.org/show_bug.cgi?id=157705

Reviewed by Darin Adler.

Source/WebCore:

The bug was caused by WebKit doesn't look for the title attribute across shadow boundaries.
Fixed it by using a newly added Node::parentNodeInComposedTree in HitTestResult::title.

Test: fast/shadow-dom/tooltip-on-composed-tree.html

* dom/Node.cpp:
(WebCore::Node::parentInComposedTree): Added.
* dom/Node.h:
* rendering/HitTestResult.cpp:
(WebCore::HitTestResult::title): Fixed the bug.
(WebCore::HitTestResult::innerTextIfTruncated): Fixed a related bug when ShowsToolTipOverTruncatedText
is enabled. Unfortunately, there is no machinery to test this feature yet.

LayoutTests:

Added a regression test for finding a tooltip across shadow boundaries.

* fast/shadow-dom/tooltip-on-composed-tree-expected.txt: Added.
* fast/shadow-dom/tooltip-on-composed-tree.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@200923 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoCTTE for the HTML editing header
darin@apple.com [Sat, 14 May 2016 20:09:50 +0000 (20:09 +0000)]
CTTE for the HTML editing header
https://bugs.webkit.org/show_bug.cgi?id=157676

Reviewed by Chris Dumez.

Source/WebCore:

* accessibility/AXObjectCache.cpp:
(WebCore::AccessibilityReplacedText::postTextStateChangeNotification): Use auto so we have a
change to compile more efficient code using the more specific types returned, rather than
explicitly using a less specific type like Node.
(WebCore::AXObjectCache::startCharacterOffsetOfParagraph): Ditto.
* accessibility/atk/WebKitAccessibleInterfaceText.cpp: Update for deprecatedIsEditingWhitespace name.
* dom/Position.cpp:
(WebCore::Position::offsetForPositionAfterAnchor): Update to use a reference instead of a pointer.
(WebCore::Position::next): Ditto.
(WebCore::Position::atFirstEditingPositionForNode): Ditto.
(WebCore::Position::atLastEditingPositionForNode): Ditto.
(WebCore::Position::atStartOfTree): Ditto.
(WebCore::Position::atEndOfTree): Ditto.
(WebCore::Position::upstream): Ditto.
(WebCore::Position::downstream): Ditto.
(WebCore::Position::rendersInDifferentPosition): Ditto.
* dom/PositionIterator.cpp:
(WebCore::PositionIterator::increment): Ditto.
(WebCore::PositionIterator::decrement): Ditto.
(WebCore::PositionIterator::atEnd): Ditto.
(WebCore::PositionIterator::atEndOfNode): Ditto.
* editing/AlternativeTextController.cpp:
(WebCore::AlternativeTextController::markPrecedingWhitespaceForDeletedAutocorrectionAfterCommand):
Use new name, deprecatedIsEditingWhitespace.
* editing/ApplyBlockElementCommand.cpp:
(WebCore::ApplyBlockElementCommand::formatSelection): Use auto (see rationale above).
* editing/ApplyStyleCommand.cpp:
(WebCore::ApplyStyleCommand::applyBlockStyle): More of the same.
(WebCore::ApplyStyleCommand::applyRelativeFontStyleChange): Ditto.
(WebCore::ApplyStyleCommand::fixRangeAndApplyInlineStyle): Ditto.
(WebCore::ApplyStyleCommand::isValidCaretPositionInTextNode): Ditto.
(WebCore::ApplyStyleCommand::mergeStartWithPreviousIfIdentical): Ditto.
(WebCore::ApplyStyleCommand::mergeEndWithNextIfIdentical): Ditto.
(WebCore::ApplyStyleCommand::surroundNodeRangeWithElement): Ditto.
* editing/BreakBlockquoteCommand.cpp:
(WebCore::BreakBlockquoteCommand::doApply): Ditto.
* editing/CompositeEditCommand.cpp:
(WebCore::postTextStateChangeNotification): Ditto.
(WebCore::CompositeEditCommand::insertNodeAt): Ditto.
(WebCore::CompositeEditCommand::positionOutsideTabSpan): Ditto.
(WebCore::containsOnlyDeprecatedEditingWhitespace): Ditto.
(WebCore::CompositeEditCommand::shouldRebalanceLeadingWhitespaceFor): Ditto.
(WebCore::CompositeEditCommand::rebalanceWhitespaceAt): Ditto.
(WebCore::CompositeEditCommand::rebalanceWhitespaceOnTextSubstring): Ditto.
(WebCore::CompositeEditCommand::insertNewDefaultParagraphElementAt): Ditto.
(WebCore::CompositeEditCommand::moveParagraphWithClones): Ditto.
(WebCore::CompositeEditCommand::moveParagraphs): Ditto.
* editing/DeleteSelectionCommand.cpp:
(WebCore::DeleteSelectionCommand::initializeStartEnd): Ditto.
(WebCore::DeleteSelectionCommand::handleGeneralDelete): Ditto.
(WebCore::DeleteSelectionCommand::mergeParagraphs): Ditto.
(WebCore::DeleteSelectionCommand::doApply): Ditto.
* editing/EditCommand.cpp:
(WebCore::EditCommand::postTextStateChangeNotification): Ditto.
* editing/EditingStyle.cpp: Ditto.
* editing/Editor.cpp:
(WebCore::Editor::advanceToNextMisspelling): Ditto.
* editing/FormatBlockCommand.cpp:
(WebCore::FormatBlockCommand::formatRange): Ditto.
* editing/IndentOutdentCommand.cpp:
(WebCore::IndentOutdentCommand::outdentParagraph): Ditto.
* editing/InsertLineBreakCommand.cpp:
(WebCore::InsertLineBreakCommand::doApply): Ditto.
* editing/InsertListCommand.cpp:
(WebCore::InsertListCommand::fixOrphanedListChild): Ditto.
(WebCore::InsertListCommand::doApplyForSingleParagraph): Ditto.
(WebCore::InsertListCommand::unlistifyParagraph): Ditto.
(WebCore::InsertListCommand::listifyParagraph): Ditto.
* editing/InsertParagraphSeparatorCommand.cpp:
(WebCore::InsertParagraphSeparatorCommand::doApply): Ditto.
* editing/InsertTextCommand.cpp:
(WebCore::InsertTextCommand::insertTab): Ditto.
* editing/ModifySelectionListLevel.cpp:
(WebCore::IncreaseSelectionListLevelCommand::doApply): Ditto.
* editing/ReplaceSelectionCommand.cpp:
(WebCore::ReplacementFragment::removeUnrenderedNodes): Ditto.
(WebCore::ReplaceSelectionCommand::shouldMerge): Ditto.
(WebCore::ReplaceSelectionCommand::removeRedundantStylesAndKeepStyleSpanInline): Ditto.
(WebCore::ReplaceSelectionCommand::mergeEndIfNeeded): Ditto.
(WebCore::enclosingInline): Ditto.
(WebCore::ReplaceSelectionCommand::doApply): Ditto.
* editing/TextIterator.cpp:
(WebCore::maxOffsetIncludingCollapsedSpaces): Ditto.
* editing/TypingCommand.cpp:
(WebCore::TypingCommand::forwardDeleteKeyPressed): Ditto.
* editing/VisiblePosition.cpp:
(WebCore::VisiblePosition::honorEditingBoundaryAtOrBefore): Ditto.
(WebCore::VisiblePosition::honorEditingBoundaryAtOrAfter): Ditto.
* editing/VisibleSelection.cpp:
(WebCore::VisibleSelection::setStartAndEndFromBaseAndExtentRespectingGranularity): Ditto.
(WebCore::VisibleSelection::adjustSelectionToAvoidCrossingEditingBoundaries): Ditto.
* editing/VisibleUnits.cpp:
(WebCore::previousRootInlineBoxCandidatePosition): Ditto.
(WebCore::nextRootInlineBoxCandidatePosition): Ditto.
(WebCore::startOfParagraph): Ditto.
(WebCore::endOfParagraph): Ditto.
(WebCore::startOfEditableContent): Ditto.
(WebCore::endOfEditableContent): Ditto.

* editing/htmlediting.cpp:
(WebCore::highestEditableRoot): Changed return type to ContainerNode. Maybe later could
change it to Element.
(WebCore::lowestEditableAncestor): Changed return type to Element.
(WebCore::isEditableToAccessibility): Use auto.
(WebCore::isRichlyEditablePosition): Ditto.
(WebCore::editableRootForPosition): Ditto.
(WebCore::unsplittableElementForPosition):Ditto.
(WebCore::nextCandidate): Streamline.
(WebCore::nextVisuallyDistinctCandidate): Ditto.
(WebCore::previousCandidate): Ditto.
(WebCore::previousVisuallyDistinctCandidate): Ditto.
(WebCore::firstEditablePositionAfterPositionInRoot): Changed arugment type to
ContainerNode.
(WebCore::lastEditablePositionBeforePositionInRoot): Ditto.
(WebCore::lastOffsetForEditing): Changed argument type to reference.
(WebCore::stringWithRebalancedWhitespace): Optimized the case where the function does
not need to change the string, so it just returns the passed string rather than a copy.
(WebCore::isTableStructureNode): Use auto.
(WebCore::isSpecialElement): Made this function private to the file. Use auto.
(WebCore::firstInSpecialElement): Return an HTMLElement.
(WebCore::lastInSpecialElement): Ditto.
(WebCore::isFirstVisiblePositionInSpecialElement): Made this function private to the file.
(WebCore::positionBeforeContainingSpecialElement): Changed out argument to HTMLElement.
(WebCore::isLastVisiblePositionInSpecialElement): Ditto.
(WebCore::positionAfterContainingSpecialElement): Ditto.
(WebCore::positionOutsideContainingSpecialElement): Ditto.
(WebCore::isFirstPositionAfterTable): Changed return type to Element.
(WebCore::isLastPositionBeforeTable): Ditto.
(WebCore::visiblePositionBeforeNode): Changed argument type to reference.
(WebCore::visiblePositionAfterNode): Ditto.
(WebCore::enclosingElementWithTag): Use auto. Removed special case for null since the
algorithm already handles null correctly.
(WebCore::enclosingNodeOfType): Use auto.
(WebCore::highestEnclosingNodeOfType): Ditto.
(WebCore::highestNodeToRemoveInPruning): Ditto.
(WebCore::enclosingTableCell): Changed return type to Element.
(WebCore::enclosingAnchorElement): Removed special case for null since the algorithm
already handles null correctly.
(WebCore::enclosingList): Use auto.
(WebCore::enclosingListChild): Use auto.
(WebCore::enclosingEmptyListItem): Use auto.
(WebCore::outermostEnclosingList): Use auto.
(WebCore::canMergeLists): Use references and more specific type. Since both elements are
known to be HTMLElement, compare local names instead of using hasTagName.
(WebCore::isRenderedTable): Use auto.
(WebCore::isTableCell): Ditto.
(WebCore::isEmptyTableCell): Ditto.
(WebCore::createBreakElement): Deleted. Callers can use HTMLBRElement directly.
(WebCore::createOrderedListElement): Deleted. Callers can use HTMLOListElement directly.
(WebCore::createUnorderedListElement): Deleted. Callers can use HTMLUListElement directly.
(WebCore::createListItemElement): Deleted. Callers can use HTMLLIElement directly.
(WebCore::isTabSpanNode): Remove redundant checks and did more specific typecast.
(WebCore::isTabSpanTextNode): Removed redundant null checks.
(WebCore::tabSpanNode): Changed return type to HTMLSpanElement.
(WebCore::positionOutsideTabSpan): Deleted. Unused function.
(WebCore::createTabSpanElement): Made one of the overloads private to this file. Use auto.
Changed argument type since we don't need to support null. Moved createEditingTextNode code,
specific to the version without a string into that function rather than the helper function.
(WebCore::isNodeRendered): Changed argument type to a reference.
(WebCore::numEnclosingMailBlockquotes): Streamlined.
(WebCore::isMailBlockquote): Use downcast instead of static_cast.
(WebCore::caretMinOffset): Take a reference instead of a pointer.
(WebCore::caretMaxOffset): Ditto.
(WebCore::selectionForParagraphIteration): Use auto.
(WebCore::indexForVisiblePosition): Ditto.
(WebCore::visiblePositionForIndex): Ditto.
(WebCore::visiblePositionForIndexUsingCharacterIterator): Ditto.
(WebCore::isVisiblyAdjacent): Made this private to this file.
(WebCore::isNodeVisiblyContainedWithin): Changed argument type to a reference.
(WebCore::areIdenticalElements): Changed argument types to references.
(WebCore::adjustedSelectionStartForStyleComputation): Use auto.
(WebCore::isBlockFlowElement): Changed argument type to a reference. Use auto.
(WebCore::deprecatedEnclosingBlockFlowElement): Updated for reference changes.
(WebCore::rendererForCaretPainting): Use auto.

* editing/htmlediting.h: Use pragma once. Removed unneeded includes and forward declarations.
Removed various non-helpful comments (the sections still don't make sense; not really good to
group functions by their return types). Moved inline functoin bodies to the bottom of the file.
Renamed isWhitespace to deprecatedIsEditingWhitespace.

* editing/markup.cpp:
(WebCore::createFragmentFromText): Use HTMLBRElement.
* page/DOMSelection.cpp:
(WebCore::DOMSelection::extend): Updated since caretMaxOffset takes a reference.

Source/WebKit/ios:

* WebCoreSupport/WebFrameIOS.mm:
(-[WebFrame previousUnperturbedDictationResultBoundaryFromPosition:]): Update since
lastOffsetForEditing takes a reference now. Seems like this code should be in WebCore.
(-[WebFrame nextUnperturbedDictationResultBoundaryFromPosition:]): Ditto.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@200922 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoSupport ArrayBufferViews in the CSS Font Loading API
mmaxfield@apple.com [Sat, 14 May 2016 19:18:34 +0000 (19:18 +0000)]
Support ArrayBufferViews in the CSS Font Loading API
https://bugs.webkit.org/show_bug.cgi?id=157694
<rdar://problem/25554267>

Source/WebCore:

This patch adds a new mode to CSSFontFaceSource for immediate (ArrayBuffer) data.
Then, FontFace can simply be hooked up to this new mode.

Reviewed by Darin Adler.

Test: fast/text/css-font-loading-arraybuffer.html

* css/CSSFontFaceSource.cpp:
(WebCore::CSSFontFaceSource::CSSFontFaceSource):
(WebCore::CSSFontFaceSource::font):
* css/CSSFontFaceSource.h:
* css/FontFace.cpp:
(WebCore::FontFace::create):
* loader/cache/CachedFont.cpp:
(WebCore::CachedFont::ensureCustomFontData):
(WebCore::CachedFont::createCustomFontData):
(WebCore::CachedFont::createFont):
(WebCore::CachedFont::platformDataFromCustomData):
* loader/cache/CachedFont.h:

LayoutTests:

Reviewed by Darin Adler.

* fast/text/css-font-loading-arraybuffer-expected.txt: Added.
* fast/text/css-font-loading-arraybuffer.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@200921 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years ago[WebIDL] Add support for dictionary members of integer types
cdumez@apple.com [Sat, 14 May 2016 17:56:01 +0000 (17:56 +0000)]
[WebIDL] Add support for dictionary members of integer types
https://bugs.webkit.org/show_bug.cgi?id=157703

Reviewed by Darin Adler.

Add support for dictionary members of integer types, including support
for the [Clamp] and [EnforceRange] IDL extended attributes on such
members.

* bindings/js/JSDOMConvert.h:
(WebCore::convert):
(WebCore::convertOptional):
- Add the needed template specializations so we can use convertOptional()
  with integral types.
- Use std::enable_if so the template specializations meants to be used
  for floating point types or integral types can only be instantiated
  for such types.

* bindings/scripts/CodeGeneratorJS.pm:
(GenerateConversionRuleWithLeadingComma):
(GetIntegerConversionConfiguration):
(JSValueToNative):
Generate the right IntegerConversionConfiguration parameter for convert()
and convertOptional() when converting dictionary members of integral
types

* bindings/scripts/IDLParser.pm:
(parseDictionaryMember):
Fix bug in the IDL parser where we weren't initializing the extended
attributes for dictionary members.

* bindings/scripts/test/JS/JSTestObj.cpp:
* bindings/scripts/test/TestObj.idl:
Add bindings tests coverage.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@200920 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoAllocate MacGlyphToPathTranslator / CairoGlyphToPathTranslator on the stack
cdumez@apple.com [Sat, 14 May 2016 16:54:07 +0000 (16:54 +0000)]
Allocate MacGlyphToPathTranslator / CairoGlyphToPathTranslator on the stack
https://bugs.webkit.org/show_bug.cgi?id=157690

Reviewed by Myles C. Maxfield.

Allocate MacGlyphToPathTranslator / CairoGlyphToPathTranslator on the
stack. We also now use the subclass type for the local variable so it
will bypass the vtable for various virtual function calls (if the compiler
was not already smart enough to figure this out).

* platform/graphics/cairo/FontCairo.cpp:
(WebCore::FontCascade::dashesForIntersectionsWithRect):
* platform/graphics/cocoa/FontCascadeCocoa.mm:
(WebCore::FontCascade::dashesForIntersectionsWithRect):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@200919 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoWeb Inspector: Empty line shows arrow when hovering over Animation Frame Requested...
nvasilyev@apple.com [Sat, 14 May 2016 14:59:55 +0000 (14:59 +0000)]
Web Inspector: Empty line shows arrow when hovering over Animation Frame Requested in timeline recording
https://bugs.webkit.org/show_bug.cgi?id=157707

Reviewed by Timothy Hatcher.

* UserInterface/Views/DataGrid.css:
(.data-grid:not(.variable-height-rows) > .data-container > table.data > tbody > tr > td > div):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@200918 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoUnreviewed attempt to fix the iOS build after the protector variables renaming.
cdumez@apple.com [Sat, 14 May 2016 06:31:51 +0000 (06:31 +0000)]
Unreviewed attempt to fix the iOS build after the protector variables renaming.

* platform/network/cf/ResourceHandleCFURLConnectionDelegateWithOperationQueue.cpp:
(WebCore::ResourceHandleCFURLConnectionDelegateWithOperationQueue::willSendRequest):
(WebCore::ResourceHandleCFURLConnectionDelegateWithOperationQueue::didReceiveResponse):
(WebCore::ResourceHandleCFURLConnectionDelegateWithOperationQueue::didReceiveData):
(WebCore::ResourceHandleCFURLConnectionDelegateWithOperationQueue::didFinishLoading):
(WebCore::ResourceHandleCFURLConnectionDelegateWithOperationQueue::didFail):
(WebCore::ResourceHandleCFURLConnectionDelegateWithOperationQueue::willCacheResponse):
(WebCore::ResourceHandleCFURLConnectionDelegateWithOperationQueue::didReceiveChallenge):
(WebCore::ResourceHandleCFURLConnectionDelegateWithOperationQueue::didSendBodyData):
(WebCore::ResourceHandleCFURLConnectionDelegateWithOperationQueue::canRespondToProtectionSpace):
(WebCore::ResourceHandleCFURLConnectionDelegateWithOperationQueue::didReceiveDataArray):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@200917 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agojsc: samplingProfilerStackTraces() without starting sampling should not cause jsc...
commit-queue@webkit.org [Sat, 14 May 2016 05:51:38 +0000 (05:51 +0000)]
jsc: samplingProfilerStackTraces() without starting sampling should not cause jsc to crash
https://bugs.webkit.org/show_bug.cgi?id=157704

Patch by Joseph Pecoraro <pecoraro@apple.com> on 2016-05-13
Reviewed by Saam Barati.

* jsc.cpp:
(functionStartSamplingProfiler):
(functionSamplingProfilerStackTraces):
Throw an exception instead of crashing if we haven't started sampling.

* inspector/agents/InspectorScriptProfilerAgent.cpp:
(Inspector::InspectorScriptProfilerAgent::startTracking):
* runtime/VM.h:
* runtime/VM.cpp:
(JSC::VM::ensureSamplingProfiler):
Switch ensure to returning a reference, like most other ensures.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@200916 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoAll scrolling height/width values should be integral rounded.
zalan@apple.com [Sat, 14 May 2016 05:14:08 +0000 (05:14 +0000)]
All scrolling height/width values should be integral rounded.
https://bugs.webkit.org/show_bug.cgi?id=157687

Reviewed by Simon Fraser.

Both clientHeight(Width) and scrollHeight(width) need to be explicitly integral
rounded in order to ensure that scrollable content is computed properly.
This will evolve into subpixel snapping once we apply subpixel arithmetics on scrolling.

Source/WebCore:

Not testable.

* rendering/RenderBox.cpp:
(WebCore::RenderBox::scrollWidth):
* rendering/RenderLayer.cpp:
(WebCore::RenderLayer::scrollTo):
(WebCore::RenderLayer::horizontalScrollbarStart):
(WebCore::RenderLayer::scrollWidth):
(WebCore::RenderLayer::scrollHeight):
(WebCore::RenderLayer::computeScrollDimensions):
(WebCore::RenderLayer::updateScrollbarsAfterLayout):
* rendering/RenderLayer.h:

LayoutTests:

* platform/mac/fast/css/text-overflow-ellipsis-text-align-center-expected.txt:
* platform/mac/fast/css/text-overflow-ellipsis-text-align-justify-expected.txt:
* platform/mac/fast/css/text-overflow-ellipsis-text-align-left-expected.txt:
* platform/mac/fast/css/text-overflow-ellipsis-text-align-right-expected.txt:
* platform/mac/fast/css/text-overflow-input-expected.txt:
* platform/mac/fast/css/vertical-text-overflow-ellipsis-text-align-center-expected.txt:
* platform/mac/fast/css/vertical-text-overflow-ellipsis-text-align-justify-expected.txt:
* platform/mac/fast/css/vertical-text-overflow-ellipsis-text-align-left-expected.txt:
* platform/mac/fast/css/vertical-text-overflow-ellipsis-text-align-right-expected.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@200915 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoFix JSBench which causes math domain error in run-benchmark script.
dewei_zhu@apple.com [Sat, 14 May 2016 04:58:41 +0000 (04:58 +0000)]
Fix JSBench which causes math domain error in run-benchmark script.
https://bugs.webkit.org/show_bug.cgi?id=157697
<rdar://problem/26279292>

Reviewed by Ryosuke Niwa.

Some tests of JSBench give time usage to be 0 ms which will causes error while calculating geometric mean.
Also fix a round-off error while calculating stdev.

* Scripts/webkitpy/benchmark_runner/benchmark_results.py:
(BenchmarkResults._format_values):
* Scripts/webkitpy/benchmark_runner/data/patches/JSBench.patch:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@200914 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoExtend protector naming code style guideline to cover operator= assignment.
beidson@apple.com [Sat, 14 May 2016 04:45:42 +0000 (04:45 +0000)]
Extend protector naming code style guideline to cover operator= assignment.
https://bugs.webkit.org/show_bug.cgi?id=157693

Reviewed by Darin Adler.

Tools:

* Scripts/webkitpy/style/checkers/cpp.py:
(check_identifier_name_in_declaration):
* Scripts/webkitpy/style/checkers/cpp_unittest.py:
(WebKitStyleTest.test_names):

Websites/webkit.org:

* code-style.md:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@200913 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoFix iOS Build.
dino@apple.com [Sat, 14 May 2016 02:40:57 +0000 (02:40 +0000)]
Fix iOS Build.

* Shared/AssistedNodeInformation.h:
(WebKit::OptionItem::OptionItem):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@200912 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoREGRESSION (r200463): Autofill Credit Card expected but got Autofill Contact
weinig@apple.com [Sat, 14 May 2016 02:29:04 +0000 (02:29 +0000)]
REGRESSION (r200463): Autofill Credit Card expected but got Autofill Contact
<rdar://problem/26251538>
https://bugs.webkit.org/show_bug.cgi?id=157701

Reviewed by Dan Bernstein.

* WebProcess/InjectedBundle/API/Cocoa/WKWebProcessPlugInFrame.mm:
(-[WKWebProcessPlugInFrame _certificateChain]):
We need to do a retain-autorelease here now because the CertificateInfo's that contain the
cached certificate chain, are being destroyed, and therefore destroying the chain, before this
function returns. This used to work when the chain wasn't just a cache, because there was still
a strong reference to it being held by the ResourceResponse's CertificateInfo copy, but now
there is none.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@200911 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years ago[iOS] <select> elements popover should render right-aligned when in RTL mode
dino@apple.com [Sat, 14 May 2016 02:27:50 +0000 (02:27 +0000)]
[iOS] <select> elements popover should render right-aligned when in RTL mode
https://bugs.webkit.org/show_bug.cgi?id=157699
<rdar://problem/26282319>

Reviewed by Simon Fraser.

Some follow-up changes as suggested by Darin in review.

* Shared/AssistedNodeInformation.h: Use initializers so we can
remove the constructors.
(WebKit::OptionItem::OptionItem): Deleted.
(WebKit::AssistedNodeInformation::AssistedNodeInformation): Deleted.
* UIProcess/ios/forms/WKFormSelectPopover.mm: Add a comment to explain
why we are explicitly setting the language.
(-[WKSelectTableViewController initWithView:hasGroups:]):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@200910 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years ago[NetworkCache] Avoid having to re-parse URLs after deserializing them
cdumez@apple.com [Sat, 14 May 2016 02:26:23 +0000 (02:26 +0000)]
[NetworkCache] Avoid having to re-parse URLs after deserializing them
https://bugs.webkit.org/show_bug.cgi?id=157660

Reviewed by Darin Adler.

Source/WebCore:

Avoid having to re-parse URLs after deserializing them in the WebKit2
network cache storage implementation.

We previously serialized URLs as Strings, which meant that we had the
re-parse them upon deserialization. We now serialize all of the URL
data members to avoid having to parse the String again.

* platform/URL.h:
(WebCore::URL::encode):
(WebCore::URL::decode):
* platform/network/ResourceRequestBase.h:
(WebCore::ResourceRequestBase::encodeWithoutPlatformData):
(WebCore::ResourceRequestBase::decodeWithoutPlatformData):
* platform/network/ResourceResponseBase.h:
(WebCore::ResourceResponseBase::encode):
(WebCore::ResourceResponseBase::decode):

Source/WebKit2:

* NetworkProcess/cache/NetworkCacheCoders.cpp:
(WebKit::NetworkCache::Coder<WebCore::CertificateInfo>::encode): Deleted.
(WebKit::NetworkCache::Coder<WebCore::CertificateInfo>::decode): Deleted.
* NetworkProcess/cache/NetworkCacheCoders.h:
Drop template specializations used by our network cache for encoding
/ decoding URLs as Strings. This causes us to now rely on
URL::encode() / URL::decode() instead.

* Shared/WebCoreArgumentCoders.cpp:
(IPC::ArgumentCoder<BlobPart>::decode):
Fix bug in BlobPart encoding / decoding. It was encoding the url member
as a URL but decoding it as a String.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@200909 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoScriptController::processingUserGesture should propagate across postMessage boundaries
weinig@apple.com [Sat, 14 May 2016 02:17:31 +0000 (02:17 +0000)]
ScriptController::processingUserGesture should propagate across postMessage boundaries
<rdar://problem/26273173>
https://bugs.webkit.org/show_bug.cgi?id=157686

Reviewed by Anders Carlsson.

Source/WebCore:

Test: fast/dom/Window/post-message-user-action.html

* dom/UserGestureIndicator.h:
Export processingUserGesture() for testing.

* page/DOMWindow.cpp:
(WebCore::PostMessageTimer::PostMessageTimer):
(WebCore::PostMessageTimer::event):
Maintain the processingUserGesture bit through postMessage.

* testing/Internals.cpp:
(WebCore::Internals::setResourceTimingSupport):
(WebCore::Internals::isProcessingUserGesture):
* testing/Internals.h:
* testing/Internals.idl:
Add new internals function to get the current state of the UseGestureIndicator for testing.

LayoutTests:

* fast/dom/Window/post-message-user-action-expected.txt: Added.
* fast/dom/Window/post-message-user-action.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@200908 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoAlign window.scroll() / scrollTo() / scrollBy() with the CSSOM specification
cdumez@apple.com [Sat, 14 May 2016 02:10:27 +0000 (02:10 +0000)]
Align window.scroll() / scrollTo() / scrollBy() with the CSSOM specification
https://bugs.webkit.org/show_bug.cgi?id=157666

Reviewed by Darin Adler.

Source/WebCore:

Align window.scroll() / scrollTo() / scrollBy() with Firefox and the CSSOM
specification:
https://drafts.csswg.org/cssom-view/#extensions-to-the-window-interface

In particular, the following changes were made:
1. Make parameters to scroll() / scrollTo() / scrollBy() mandatory.
2. Add overloads for scroll() / scrollTo() / scrollBy() that take an
   optional ScrollToOptions dictionary.
3. Update API to use "unrestricted double" typing for x/y instead of
   "long". This matches the specification but it does not really change
   our behavior at this point because the values are still casted to
   int in our implementation.

Web-Exposed behavior changes:
1. JS can now pass a dictionary to scroll() / scrollTo() / scrollBy().
   This a new feature that Firefox already supports (Chrome does not).
2. Passing only 1 parameter to scroll() / scrollTo() / scrollBy() that
   is not a dictionary will now throw a TypeError. The compatibility
   risky should be low because Firefox and Chrome already throw in this
   case (Chrome has been throwing for 2 years and a half).
3. Calling scrollTo() / scroll() without any parameter no longer
   scrolls to 0. Instead we use the current viewport's x/y which means
   we don't scroll at all. The new behavior matches Firefox, Chrome and
   IE 11. This fixes scrolling on the following Website:
   https://members.chosun.com/cms_subscribe/application/index.jsp

No new tests, extended existing testing.

* bindings/js/JSDOMConvert.h:
(WebCore::convert):
(WebCore::convertOptional):
* bindings/scripts/CodeGeneratorJS.pm:
(ShouldAllowNonFiniteForFloatingPointType):
(GenerateConversionRuleWithLeadingComma):
(GenerateDictionaryImplementationContent):
(JSValueToNative):
* bindings/scripts/test/JS/JSTestObj.cpp:
(WebCore::convert<TestObj::Dictionary>):
* bindings/scripts/test/TestObj.idl:
* page/DOMWindow.cpp:
(WebCore::DOMWindow::scrollBy):
(WebCore::DOMWindow::scrollTo):
* page/DOMWindow.h:
* page/DOMWindow.idl:

LayoutTests:

Update / improve testing coverage for the API.

* fast/dom/Window/window-scroll-arguments-expected.txt:
* fast/dom/Window/window-scroll-arguments.html:
* fast/dom/non-numeric-values-numeric-parameters-expected.txt:
* fast/dom/script-tests/non-numeric-values-numeric-parameters.js:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@200907 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoDFG/FTL have a few bugs in their reasoning about the scope
sbarati@apple.com [Sat, 14 May 2016 02:03:10 +0000 (02:03 +0000)]
DFG/FTL have a few bugs in their reasoning about the scope
https://bugs.webkit.org/show_bug.cgi?id=157696

Reviewed by Benjamin Poulain.

1. When the debugger is enabled, it is easier for the DFG to reason
about the scope register by simply claiming all nodes read the scope
register. This prevents us from ever entering the runtime where we
may take a stack trace but there isn't a scope on the stack.

2. This patch fixes a bug where the FTL compilation wasn't properly
setting the CodeBlock register. It was only doing this when there
was inline data, but when the debugger is enabled, we never inline.
So this code just needed to be removed from that loop. It was never
right for it to be inside the loop.

* dfg/DFGClobberize.h:
(JSC::DFG::clobberize):
* ftl/FTLCompile.cpp:
(JSC::FTL::compile):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@200906 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoUnreviewed, rolling out r200894.
ryanhaddad@apple.com [Sat, 14 May 2016 00:03:43 +0000 (00:03 +0000)]
Unreviewed, rolling out r200894.
https://bugs.webkit.org/show_bug.cgi?id=157695

This change broke the iOS build (Requested by ryanhaddad on
#webkit).

Reverted changeset:

"Remove unused initializer for WebEvent on iOS."
https://bugs.webkit.org/show_bug.cgi?id=157689
http://trac.webkit.org/changeset/200894

Patch by Commit Queue <commit-queue@webkit.org> on 2016-05-13

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@200901 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agov3 UI shows full git hash instead of the first 8 characters for a blame range
rniwa@webkit.org [Sat, 14 May 2016 00:01:21 +0000 (00:01 +0000)]
v3 UI shows full git hash instead of the first 8 characters for a blame range
https://bugs.webkit.org/show_bug.cgi?id=157691

Reviewed by Stephanie Lewis.

Fixed the bug that v3 UI shows the full 40 character git hash instead of the first 8 character as done in v2 UI.

* public/v3/models/commit-log.js:
(CommitLog.prototype.diff): Fixed the bug.
* tools/run-tests.py:
(main): Add the support for running a subset of tests as mocha does.
* unit-tests/commit-log-tests.js: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@200900 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years ago[JSC] SetLocal without exit do not need phantoms
benjamin@webkit.org [Fri, 13 May 2016 23:45:44 +0000 (23:45 +0000)]
[JSC] SetLocal without exit do not need phantoms
https://bugs.webkit.org/show_bug.cgi?id=157653

Patch by Benjamin Poulain <bpoulain@apple.com> on 2016-05-13
Reviewed by Filip Pizlo.

I made a mistake in r200498.

If a SetLocal cannot possibly exit, we were not clearing
the source of the operand. As a result, we sometime kept
a value alive up to the end of the block.

That's uncommon because SetLocal typically appear
toward the end of blocks. That's probably why there was
no perf impact with that fix.

* dfg/DFGPhantomInsertionPhase.cpp:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@200898 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years ago[JSC] Move the CheckTierUp function calls out of the main path
benjamin@webkit.org [Fri, 13 May 2016 23:45:15 +0000 (23:45 +0000)]
[JSC] Move the CheckTierUp function calls out of the main path
https://bugs.webkit.org/show_bug.cgi?id=157668

Patch by Benjamin Poulain <bpoulain@apple.com> on 2016-05-13
Reviewed by Mark Lam.

If you have a tiny tiny loop (for example, Sunspider's bits-in-byte),
the size of CheckTierUp is a problem.

On multi-issue CPUs, the node is so big that we do not
get to run anything from the loop in the instruction fetch.

On x86, having a bigger loop also pushes us out of the LSD.

This is a 6% improvement on bits-in-byte. Other Sunspider tests
only improves marginally.

* dfg/DFGSpeculativeJIT.cpp:
(JSC::DFG::SpeculativeJIT::addSlowPathGenerator):
(JSC::DFG::SpeculativeJIT::runSlowPathGenerators):
* dfg/DFGSpeculativeJIT.h:
(JSC::DFG::SpeculativeJIT::silentSpill):
(JSC::DFG::SpeculativeJIT::silentFill):
* dfg/DFGSpeculativeJIT64.cpp:
(JSC::DFG::SpeculativeJIT::compile):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@200897 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years ago[JSC] Emit the loads of emitLoadWithStructureCheck() in the order they are used
benjamin@webkit.org [Fri, 13 May 2016 23:44:31 +0000 (23:44 +0000)]
[JSC] Emit the loads of emitLoadWithStructureCheck() in the order they are used
https://bugs.webkit.org/show_bug.cgi?id=157671

Patch by Benjamin Poulain <bpoulain@apple.com> on 2016-05-13
Reviewed by Mark Lam.

This improves the chances of having a value
when issuing the TEST.

* jit/JITPropertyAccess.cpp:
(JSC::JIT::emitLoadWithStructureCheck):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@200896 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoRename all protector variables that should be called `protectedThis`
beidson@apple.com [Fri, 13 May 2016 23:42:17 +0000 (23:42 +0000)]
Rename all protector variables that should be called `protectedThis`
https://bugs.webkit.org/show_bug.cgi?id=157610

Reviewed by Alex Christensen.

No new tests (Renaming, no behavior change).

WebCore has so many instances of these protectors as well as so much variance in naming them,
that I think it deserved a new style pass.

* Modules/fetch/FetchResponse.cpp:
* Modules/geolocation/GeoNotifier.cpp:
* Modules/geolocation/Geolocation.cpp:
* Modules/indexeddb/IDBTransaction.cpp:
* Modules/indexeddb/client/TransactionOperation.h:
* Modules/indexeddb/server/UniqueIDBDatabase.cpp:
* Modules/indexeddb/server/UniqueIDBDatabaseConnection.cpp:
* Modules/indexeddb/server/UniqueIDBDatabaseTransaction.cpp:
* Modules/indexeddb/shared/InProcessIDBServer.cpp:
* Modules/mediastream/UserMediaRequest.cpp:
* Modules/notifications/NotificationCenter.cpp:
* Modules/webaudio/AudioContext.cpp:
* Modules/webdatabase/Database.cpp:
* Modules/websockets/WebSocket.cpp:
* Modules/websockets/WebSocketChannel.cpp:
* Modules/websockets/WorkerThreadableWebSocketChannel.cpp:
* accessibility/AccessibilityObject.cpp:
* accessibility/AccessibilityRenderObject.cpp:
* bindings/gobject/GObjectEventListener.cpp:
* bindings/js/JSCustomElementInterface.cpp:
* bindings/js/JSCustomSQLStatementErrorCallback.cpp:
* bindings/js/JSCustomXPathNSResolver.cpp:
* bindings/js/JSDOMGlobalObjectTask.cpp:
* bindings/js/JSDOMWindowBase.cpp:
* bindings/js/JSErrorHandler.cpp:
* bindings/js/JSEventListener.cpp:
* bindings/js/JSMutationCallback.cpp:
* bindings/js/JSNodeFilterCustom.cpp:
* bindings/js/SerializedScriptValue.cpp:
* bindings/scripts/CodeGeneratorJS.pm:
* bindings/scripts/test/JS/JSTestCallback.cpp:
* bindings/scripts/test/JS/JSTestCallbackFunction.cpp:
* bridge/runtime_root.cpp:
* css/CSSFontSelector.cpp:
* css/StyleSheetContents.cpp:
* dom/CharacterData.cpp:
* dom/ContainerNode.cpp:
* dom/Document.cpp:
* dom/EventTarget.cpp:
* dom/Node.cpp:
* dom/ScriptExecutionContext.cpp:
* dom/ScriptedAnimationController.cpp:
* dom/StringCallback.cpp:
* html/HTMLEmbedElement.cpp:
* html/HTMLFormControlElement.cpp:
* html/HTMLFormElement.cpp:
* html/HTMLInputElement.cpp:
* html/HTMLLinkElement.cpp:
* html/HTMLMediaElement.cpp:
* html/HTMLObjectElement.cpp:
* html/HTMLOptionElement.cpp:
* html/HTMLScriptElement.cpp:
* html/HTMLTableElement.cpp:
* html/HTMLTextAreaElement.cpp:
* html/HTMLTitleElement.cpp:
* html/parser/HTMLDocumentParser.cpp:
* html/shadow/SliderThumbElement.cpp:
* html/shadow/SpinButtonElement.cpp:
* inspector/InspectorFrontendClientLocal.cpp:
* loader/DocumentLoader.cpp:
* loader/DocumentThreadableLoader.cpp:
* loader/MediaResourceLoader.cpp:
* loader/NetscapePlugInStreamLoader.cpp:
* loader/ResourceLoader.cpp:
* loader/SubresourceLoader.cpp:
* loader/cache/CachedRawResource.cpp:
* loader/cf/SubresourceLoaderCF.cpp:
* loader/mac/ResourceLoaderMac.mm:
* page/DOMWindow.cpp:
* page/DOMWindowExtension.cpp:
* page/FrameView.cpp:
* page/animation/AnimationBase.cpp:
* page/scrolling/ThreadedScrollingTree.cpp:
* platform/ScrollView.cpp:
* platform/graphics/DisplayRefreshMonitor.cpp:
* platform/graphics/ca/win/CACFLayerTreeHost.cpp:
* platform/graphics/gstreamer/InbandTextTrackPrivateGStreamer.cpp:
* platform/ios/WebVideoFullscreenControllerAVKit.mm:
* platform/ios/WebVideoFullscreenInterfaceAVKit.mm:
* platform/mac/WidgetMac.mm:
* platform/network/BlobResourceHandle.cpp:
* platform/network/cf/ResourceHandleCFNet.cpp:
* platform/network/cf/ResourceHandleCFURLConnectionDelegateWithOperationQueue.cpp:
* platform/network/curl/CurlDownload.cpp:
* platform/network/mac/ResourceHandleMac.mm:
* platform/network/soup/SocketStreamHandleSoup.cpp:
* platform/win/PopupMenuWin.cpp:
* workers/WorkerScriptLoader.cpp:
* xml/XMLHttpRequest.cpp:
* xml/parser/XMLDocumentParser.cpp:
* xml/parser/XMLDocumentParserLibxml2.cpp:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@200895 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoRemove unused initializer for WebEvent on iOS.
enrica@apple.com [Fri, 13 May 2016 23:39:00 +0000 (23:39 +0000)]
Remove unused initializer for WebEvent on iOS.
https://bugs.webkit.org/show_bug.cgi?id=157689

Reviewed by Anders Carlsson.

Source/WebCore:

This is no longer used and can be removed.
The logic tied to isPopupVariant has been incorporated
in keyboard flags. The characterSet property is no longer needed too.

* platform/ios/WebEvent.h:
* platform/ios/WebEvent.mm:
(-[WebEvent initWithKeyEventType:timeStamp:characters:charactersIgnoringModifiers:modifiers:isRepeating:withFlags:keyCode:isTabKey:characterSet:]):
(-[WebEvent initWithKeyEventType:timeStamp:characters:charactersIgnoringModifiers:modifiers:isRepeating:isPopupVariant:keyCode:isTabKey:characterSet:]): Deleted.
(-[WebEvent _characterSetDescription]): Deleted.
(-[WebEvent isPopupVariant]): Deleted.
(-[WebEvent characterSet]): Deleted.

Source/WebKit/mac:

Removing use of characterSet property.

* WebView/WebHTMLView.mm:
(-[WebHTMLView _handleEditingKeyEvent:]):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@200894 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoWeb Inspector: Inform augmenting client when inspector controller is destroyed
joepeck@webkit.org [Fri, 13 May 2016 23:29:13 +0000 (23:29 +0000)]
Web Inspector: Inform augmenting client when inspector controller is destroyed
https://bugs.webkit.org/show_bug.cgi?id=157688
<rdar://problem/25832724>

Reviewed by Timothy Hatcher.

* inspector/JSGlobalObjectInspectorController.cpp:
(Inspector::JSGlobalObjectInspectorController::~JSGlobalObjectInspectorController):
* inspector/augmentable/AugmentableInspectorControllerClient.h:
There is a weak relationship between the InspectorController and the
AugmentingClient. Let the augmenting client know when the controller
is destroyed so it doesn't try to use us anymore.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@200893 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoTry to fix some non-iOS builds.
mitz@apple.com [Fri, 13 May 2016 23:26:59 +0000 (23:26 +0000)]
Try to fix some non-iOS builds.

* mac/postprocess-framework-headers.sh:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@200891 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoUnreviewed. Added the missing executable bits.
rniwa@webkit.org [Fri, 13 May 2016 23:15:38 +0000 (23:15 +0000)]
Unreviewed. Added the missing executable bits.

* tools/bundle-v3-scripts.py: Added property svn:executable.
* tools/detect-changes.js: Added property svn:executable.
* tools/process-maintenance-backlog.py: Added property svn:executable.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@200890 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agocross-fade() rendering doesn't match expectation
simon.fraser@apple.com [Fri, 13 May 2016 22:40:57 +0000 (22:40 +0000)]
cross-fade() rendering doesn't match expectation
https://bugs.webkit.org/show_bug.cgi?id=157665
rdar://problem/17917708

Reviewed by Dean Jackson.

Source/WebCore:

Cross-fading two opaque images would result in a non-opaque result in between the endpoints,
probably because r157045 caused both images to be drawn with srcOver, since drawImage()
clobbers the composite operation in the context.

Fix by passing the composite operation to the drawImage() calls for the non-transparency layer
code path.

Converted css3/images/cross-fade-blending.html into a ref test to test this.

* platform/graphics/CrossfadeGeneratedImage.cpp:
(WebCore::drawCrossfadeSubimage):

LayoutTests:

Make css3/images/cross-fade-blending.html a ref test.

* css3/images/cross-fade-blending-expected.html: Added.
* css3/images/cross-fade-blending-expected.png: Removed.
* css3/images/cross-fade-blending-expected.txt: Removed.
* css3/images/cross-fade-blending.html:
* platform/mac/css3/images/cross-fade-blending-expected.png: Removed.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@200889 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoUnprefix -webkit-cross-fade()
simon.fraser@apple.com [Fri, 13 May 2016 22:40:52 +0000 (22:40 +0000)]
Unprefix -webkit-cross-fade()
https://bugs.webkit.org/show_bug.cgi?id=157632

Reviewed by Darin Adler.
Source/WebCore:

Support unprefixed cross-fade() in addition to the prefixed value when parsing,
and have computed style round-trip the correct one.

If computed style is queried during an image transition, return an unprefixed
cross-fade(). If blending cross-fade values, only return the prefixed function if
both inputs are prefixed.

Tested by modified and existing tests.

* css/CSSCrossfadeValue.cpp:
(WebCore::CSSCrossfadeValue::customCSSText):
(WebCore::CSSCrossfadeValue::blend):
* css/CSSCrossfadeValue.h:
(WebCore::CSSCrossfadeValue::create):
(WebCore::CSSCrossfadeValue::isPrefixed):
(WebCore::CSSCrossfadeValue::CSSCrossfadeValue):
* css/CSSParser.cpp:
(WebCore::CSSParser::isGeneratedImageValue):
(WebCore::CSSParser::parseGeneratedImage):
(WebCore::CSSParser::parseCrossfade):
* css/CSSParser.h:

Source/WebInspectorUI:

* UserInterface/Models/CSSKeywordCompletions.js:

LayoutTests:

* animations/cross-fade-background-image-expected.html:
* animations/cross-fade-background-image.html:
* animations/cross-fade-border-image-source.html:
* animations/cross-fade-list-style-image.html:
* animations/cross-fade-webkit-mask-box-image.html:
* animations/cross-fade-webkit-mask-image.html:
* animations/resources/animation-test-helpers.js:
(parseCSSImage):
(parseCrossFade):
(compareCSSImages):
* fast/css/getComputedStyle/computed-style-cross-fade-expected.txt:
* fast/css/getComputedStyle/computed-style-cross-fade.html:
* transitions/resources/transition-test-helpers.js:
(parseCrossFade):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@200888 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoResourceTiming entries for cached resources and XHR
yoav@yoav.ws [Fri, 13 May 2016 22:34:29 +0000 (22:34 +0000)]
ResourceTiming entries for cached resources and XHR
https://bugs.webkit.org/show_bug.cgi?id=157669

Reviewed by Alex Christensen.

Source/WebCore:

 - Moves the ResourceTiming storage and addition logic into its own class, so that it
   can be accessed by both CachedResourceLoader and DocumentThreadableLoader.
 - Using the above, adds ResourceTiming collection into DocumentThreadableLoader,
   in order to support ResourceTiming entries for XHR based requests.
 - Adds ResourceTiming entries for resources that are reused from the memory cache.

Test: http/tests/performance/performance-resource-timing-cached-entries.html

* CMakeLists.txt: Add ResourceTimingInformation.
* WebCore.xcodeproj/project.pbxproj: Add ResourceTimingInformation.
* loader/DocumentThreadableLoader.cpp:
(WebCore::DocumentThreadableLoader::didFinishLoading): Add a call to addResourceTiming.
(WebCore::DocumentThreadableLoader::loadRequest): Store the initiator information.
* loader/DocumentThreadableLoader.h:
* loader/ResourceTimingInformation.cpp: Added.
(WebCore::ResourceTimingInformation::addResourceTiming): Moved addResourceTiming logic from CachedResourceLoader.
(WebCore::ResourceTimingInformation::storeResourceTimingInitiatorInformation): Moved addResourceTiming logic from
CachedResourceLoader. Removed reliance on the committingFirstRealLoad bool when storing initiator info, as I don't
see why it is required, and it made no sense in the context of DocumentThreadableLoader.
* loader/ResourceTimingInformation.h: Added.
* loader/cache/CachedResourceLoader.cpp:
(WebCore::CachedResourceLoader::requestResource): Add a ResourceTiming entry when a resource is reused from MemoryCache.
(WebCore::CachedResourceLoader::revalidateResource): Use ResourceTimingInformation::storeResourceTimingInitiatorInformation.
(WebCore::CachedResourceLoader::loadResource): Use ResourceTimingInformation::storeResourceTimingInitiatorInformation.
(WebCore::CachedResourceLoader::loadDone): Use ResourceTimingInformation::addResourceTiming.
(WebCore::CachedResourceLoader::storeResourceTimingInitiatorInformation): Deleted.
* loader/cache/CachedResourceLoader.h:

LayoutTests:

These tests make sure that cacheable resources as well as XHR based resources
have ResourceTiming entries.

* http/tests/performance/performance-resource-timing-cached-entries-expected.txt: Added.
* http/tests/performance/performance-resource-timing-cached-entries.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@200887 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoWeb Inspector: Typing "layout" in filter for Layout and Rendering Timeline matches...
mattbaker@apple.com [Fri, 13 May 2016 22:22:45 +0000 (22:22 +0000)]
Web Inspector: Typing "layout" in filter for Layout and Rendering Timeline matches everything
https://bugs.webkit.org/show_bug.cgi?id=157684
<rdar://problem/26276588>

Reviewed by Joseph Pecoraro.

Hidden columns should not participate in built-in grid filtering
using DataGrid.filterText. Hidden columns can still be custom filtered,
which is how TimelineDataGrid filters scopebar columns.

* UserInterface/Views/DataGrid.js:
(WebInspector.DataGridNode.prototype.get filterableData):
Skip hidden columns.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@200886 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoRunaway malloc memory usage in this simple JSC program
ggaren@apple.com [Fri, 13 May 2016 21:24:34 +0000 (21:24 +0000)]
Runaway malloc memory usage in this simple JSC program
https://bugs.webkit.org/show_bug.cgi?id=157682

Reviewed by Mark Lam.

* heap/WeakSet.cpp:
(JSC::WeakSet::sweep): Whenever we might add a block to
m_logicallyEmptyWeakBlocks, be sure also to sweep a block in
m_logicallyEmptyWeakBlocks. Otherwise, additions might outpace removals
even when all memory is freed.

We do this whenever we *might* add a block and not just whenever we *do*
add a block because we'd like to sweep the entries in
m_logicallyEmptyWeakBlocks promptly even when it's not growing, and this
is a reasonably rate-limited opportunity to do so.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@200884 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoDashboard code restructuring
aakash_jain@apple.com [Fri, 13 May 2016 20:35:58 +0000 (20:35 +0000)]
Dashboard code restructuring
https://bugs.webkit.org/show_bug.cgi?id=157680
<rdar://problem/14736888>

Reviewed by Alexey Proskuryakov.

* BuildSlaveSupport/build.webkit.org-config/public_html/dashboard/Scripts/BubbleQueueView.js:
(BubbleQueueView.prototype.addLinkToRow): Moved to base class.
(BubbleQueueView.prototype.addTextToRow): Moved to base class.
(BubbleQueueView.prototype._addDividerToPopover): Moved to base class.
* BuildSlaveSupport/build.webkit.org-config/public_html/dashboard/Scripts/BuildbotBuilderQueueView.js:
(BuildbotBuilderQueueView.prototype.update): Passing additional argument to _appendPendingRevisionCount.
* BuildSlaveSupport/build.webkit.org-config/public_html/dashboard/Scripts/BuildbotCombinedQueueView.js:
(BuildbotCombinedQueueView.prototype.update): Passing additional argument to _appendPendingRevisionCount.
* BuildSlaveSupport/build.webkit.org-config/public_html/dashboard/Scripts/BuildbotTesterQueueView.js:
(BuildbotTesterQueueView.prototype.update): Passing additional argument to _appendPendingRevisionCount.
* BuildSlaveSupport/build.webkit.org-config/public_html/dashboard/Scripts/BuildbotQueueView.js:
(BuildbotQueueView.prototype._appendPendingRevisionCount): Moved to base class and modified to take
additional argument.
(BuildbotQueueView.prototype._popoverLinesForCommitRange): Moved to base class.
(BuildbotQueueView.prototype._presentPopoverForPendingCommits): Moved to base class.
(BuildbotQueueView.prototype._addDividerToPopover): Moved to base class.
(BuildbotQueueView.prototype._formatRevisionForDisplay): Moved to base class.
* BuildSlaveSupport/build.webkit.org-config/public_html/dashboard/Scripts/Main.js:
(documentReady): Display any customView if available.
* BuildSlaveSupport/build.webkit.org-config/public_html/dashboard/Scripts/QueueView.js:
(QueueView.prototype.addLinkToRow): Moved from child class, no changes.
(QueueView.prototype.addTextToRow): Moved from child class, no changes.
(QueueView.prototype._addDividerToPopover): Moved from child class, no changes.
(QueueView.prototype._appendPendingRevisionCount): Moved from child class, made it more generic
by taking latestIterationGetter method as an argument.
(QueueView.prototype._popoverLinesForCommitRange): Moved from child class, no changes.
(QueueView.prototype._presentPopoverForPendingCommits): Moved from child class, no changes.
(QueueView.prototype._formatRevisionForDisplay): Moved from child class, no changes.
* BuildSlaveSupport/build.webkit.org-config/public_html/dashboard/Styles/Main.css:
(a:hover:link): Underline only when there is a link.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@200880 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoWe should have one calleeSaveRegistersBuffer per VMEntryFrame, not one per VM.
mark.lam@apple.com [Fri, 13 May 2016 20:16:29 +0000 (20:16 +0000)]
We should have one calleeSaveRegistersBuffer per VMEntryFrame, not one per VM.
https://bugs.webkit.org/show_bug.cgi?id=157537
<rdar://problem/24794845>

Reviewed by Michael Saboff.

Source/JavaScriptCore:

The pre-existing code behaves this way:

1. When JS code throws an exception, it saves callee save registers in
   the VM calleeSaveRegistersBuffer.  These values are meant to be restored
   to the callee save registers later either at the catch handler or at the
   uncaught exception handler.

2. If the Inspector is enable, the VM will invoke inspector C++ code to inspect
   the exception.  That C++ code can change the values of the callee save
   registers.

   The inspector code in turn re-enters the VM to execute JS inspector code.

   The JS inspector code can run hot enough that we do an enterOptimizationCheck
   on it.  The enterOptimizationCheck first saves all callee save registers
   into the VM calleeSaveRegistersBuffer.

   This effectively overwrites the values in the VM calleeSaveRegistersBuffer
   from (1).

3. Eventually, execution returns to the catch handler or the uncaught exception
   handler which restores the overwritten values in the VM
   calleeSaveRegistersBuffer to the callee save registers.

   When execution returns to the C++ code that entered the VM before (1), the
   values in the callee registers are not what that code expects, and badness
   and/or crashes ensues.

This patch applies the following fix:

1. Allocate space in the VMEntryFrame for the calleeSaveRegistersBuffer.
   This ensures that each VM entry session has its own buffer to use, and will
   not corrupt the one from the previous VM entry session.

   Delete the VM calleeSaveRegistersBuffer.

2. Change all locations that uses the VM calleeSaveRegistersBuffer to use the
   calleeSaveRegistersBuffer in the current VMEntryFrame.

3. Renamed all uses of the term "VMCalleeSavesBuffer" to
   "VMEntryFrameCalleeSavesBuffer".

This fix has been tested on the following configurations:
1. JSC and layout tests on a debug ASan build for 64-bit x86_64.
2. JSC tests on a release ASan build for 32-bit x86.
3. JSC tests on a release normal (non-ASan) build for ARM64.
4. JSC tests on a release normal (non-ASan) build for ARMv7 and ARMv7s.
5. JSC tests on a release ASan CLOOP build for x86_64.

These test runs did not produce any new crashes.  The ASan CLOOP has some
pre-existing crashes which are not due to this patch.

This bug can be tested by running the inspector/debugger/regress-133182.html test
on an ASan build.

* bytecode/PolymorphicAccess.cpp:
(JSC::AccessGenerationState::emitExplicitExceptionHandler):
* dfg/DFGJITCompiler.cpp:
(JSC::DFG::JITCompiler::compileExceptionHandlers):
* dfg/DFGOSREntry.cpp:
(JSC::DFG::prepareOSREntry):
* dfg/DFGOSRExitCompiler.cpp:
* dfg/DFGOSRExitCompiler32_64.cpp:
(JSC::DFG::OSRExitCompiler::compileExit):
* dfg/DFGOSRExitCompiler64.cpp:
(JSC::DFG::OSRExitCompiler::compileExit):
* dfg/DFGThunks.cpp:
(JSC::DFG::osrEntryThunkGenerator):
* ftl/FTLCompile.cpp:
(JSC::FTL::compile):
* ftl/FTLLowerDFGToB3.cpp:
(JSC::FTL::DFG::LowerDFGToB3::lower):
* ftl/FTLOSRExitCompiler.cpp:
(JSC::FTL::compileStub):
* interpreter/Interpreter.cpp:
(JSC::UnwindFunctor::operator()):
(JSC::UnwindFunctor::copyCalleeSavesToVMEntryFrameCalleeSavesBuffer):
(JSC::UnwindFunctor::copyCalleeSavesToVMCalleeSavesBuffer): Deleted.
* interpreter/Interpreter.h:
(JSC::NativeCallFrameTracer::NativeCallFrameTracer):
* interpreter/VMEntryRecord.h:
(JSC::VMEntryRecord::calleeSaveRegistersBufferOffset):
(JSC::VMEntryRecord::prevTopCallFrame):
(JSC::VMEntryRecord::unsafePrevTopCallFrame):
(JSC::VMEntryFrame::vmEntryRecordOffset):
(JSC::VMEntryFrame::calleeSaveRegistersBufferOffset):
* jit/AssemblyHelpers.cpp:
(JSC::AssemblyHelpers::emitRandomThunk):
(JSC::AssemblyHelpers::restoreCalleeSavesFromVMEntryFrameCalleeSavesBuffer):
(JSC::AssemblyHelpers::restoreCalleeSavesFromVMCalleeSavesBuffer): Deleted.
* jit/AssemblyHelpers.h:
(JSC::AssemblyHelpers::emitRestoreSavedTagRegisters):
(JSC::AssemblyHelpers::copyCalleeSavesToVMEntryFrameCalleeSavesBuffer):
(JSC::AssemblyHelpers::copyCalleeSavesFromFrameOrRegisterToVMEntryFrameCalleeSavesBuffer):
(JSC::AssemblyHelpers::copyCalleeSavesToVMCalleeSavesBuffer): Deleted.
(JSC::AssemblyHelpers::copyCalleeSavesFromFrameOrRegisterToVMCalleeSavesBuffer): Deleted.
* jit/JIT.cpp:
(JSC::JIT::emitEnterOptimizationCheck):
(JSC::JIT::privateCompileExceptionHandlers):
* jit/JITOpcodes.cpp:
(JSC::JIT::emit_op_throw):
(JSC::JIT::emit_op_catch):
(JSC::JIT::emitSlow_op_loop_hint):
* jit/JITOpcodes32_64.cpp:
(JSC::JIT::emit_op_throw):
(JSC::JIT::emit_op_catch):
* jit/ThunkGenerators.cpp:
(JSC::throwExceptionFromCallSlowPathGenerator):
(JSC::nativeForGenerator):
* llint/LLIntThunks.cpp:
(JSC::vmEntryRecord):
* llint/LowLevelInterpreter.asm:
* llint/LowLevelInterpreter32_64.asm:
* llint/LowLevelInterpreter64.asm:
* runtime/VM.h:
(JSC::VM::getCTIStub):
(JSC::VM::calleeSaveRegistersBufferOffset): Deleted.
* wasm/WASMFunctionCompiler.h:
(JSC::WASMFunctionCompiler::endFunction):

LayoutTests:

* inspector/debugger/regress-133182-expected.txt:
- Rebased test results to update line numbers.
* platform/mac/TestExpectations:
- Unskip the test.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@200879 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoSummary page doesn't report some missing platforms
rniwa@webkit.org [Fri, 13 May 2016 19:42:38 +0000 (19:42 +0000)]
Summary page doesn't report some missing platforms
https://bugs.webkit.org/show_bug.cgi?id=157670

Reviewed by Darin Adler.

This patch improves the warning text for missing platforms and fixes the bug that platforms that don't have
any data reported for a given test would not be reported as missing.

* public/v3/pages/summary-page.js:
(SummaryPage.prototype.render): Added instrumentations.
(SummaryPage.prototype._constructRatioGraph): Always create both the ratio bar graph and the spinner icon.
(SummaryPage.prototype._renderCell): Extracted from _constructRatioGraph. Toggle the displayed-ness of the
spinner and the ratio bar graph in the cell by CSS for better performance.
(SummaryPage.prototype._warningTextForGroup): Extracted from _constructRatioGraph. Rephrased warning text
for clarity and adopted new API of SummaryPageConfigurationGroup.
(SummaryPage.prototype._warningTextForGroup.mapAndSortByName): Added.
(SummaryPage.prototype._warningTextForGroup.pluralizeIfNeeded): Added.
(SummaryPage.cssTemplate): Added rules to toggle the visibility of spinner icons and bar graphs.
(SummaryPageConfigurationGroup): Replaced this._warnings by more explicitly named this._missingPlatforms
and this._platformsWithoutBaseline. Also add a platform to this._missingPlatforms if it didn't appear in
any metrics. Note that adding a platform whenever it doesn't in any one metric would be incorrect since
some tests uses a different test name on different platforms: e.g. PLT-Mac and PLT-iPhone.
(SummaryPageConfigurationGroup.prototype.missingPlatforms): Added.
(SummaryPageConfigurationGroup.prototype.platformsWithoutBaseline): Added.
(SummaryPageConfigurationGroup.prototype._fetchAndComputeRatio):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@200878 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoAttempt to fix the iOS build after r200874
ryanhaddad@apple.com [Fri, 13 May 2016 19:42:22 +0000 (19:42 +0000)]
Attempt to fix the iOS build after r200874

Unreviewed build fix.

* UIProcess/API/Cocoa/WKWebViewConfiguration.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@200877 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoAX: Regressions in undo/redo accessibility from Bug 153361
d_russell@apple.com [Fri, 13 May 2016 19:34:48 +0000 (19:34 +0000)]
AX: Regressions in undo/redo accessibility from Bug 153361
https://bugs.webkit.org/show_bug.cgi?id=157652

Reviewed by Darin Adler.

Use indexForVisiblePosition & visiblePositionForIndex to store selection data
so that AccessibilityUndoReplacedText can build valid selection ranges and
retrieve strings for AT notifications.

Source/WebCore:

Tests: accessibility/mac/value-change/value-change-user-info-contenteditable-expected.txt:
       accessibility/mac/value-change/value-change-user-info-contenteditable.html:
       accessibility/mac/value-change/value-change-user-info-textarea-expected.txt:
       accessibility/mac/value-change/value-change-user-info-textarea.html:
       accessibility/mac/value-change/value-change-user-info-textfield-expected.txt:
       accessibility/mac/value-change/value-change-user-info-textfield.html:

* accessibility/AXObjectCache.cpp:
(WebCore::AccessibilityReplacedText::AccessibilityReplacedText):
(WebCore::AXObjectCache::postTextStateChangeNotification):
(WebCore::AXObjectCache::postTextReplacementNotification):
* editing/CompositeEditCommand.cpp:
(WebCore::AccessibilityUndoReplacedText::configureRangeDeletedByReapplyWithEndingSelection):
(WebCore::AccessibilityUndoReplacedText::configureRangeDeletedByReapplyWithStartingSelection):
(WebCore::AccessibilityUndoReplacedText::setRangeDeletedByUnapply):
(WebCore::AccessibilityUndoReplacedText::captureTextForUnapply):
(WebCore::AccessibilityUndoReplacedText::captureTextForReapply):
(WebCore::AccessibilityUndoReplacedText::textDeletedByUnapply):
(WebCore::AccessibilityUndoReplacedText::textDeletedByReapply):
(WebCore::AccessibilityUndoReplacedText::postTextStateChangeNotificationForUnapply):
(WebCore::AccessibilityUndoReplacedText::postTextStateChangeNotificationForReapply):
(WebCore::EditCommandComposition::EditCommandComposition):
(WebCore::EditCommandComposition::unapply):
(WebCore::EditCommandComposition::reapply):
(WebCore::EditCommandComposition::setStartingSelection):
(WebCore::EditCommandComposition::setEndingSelection):
(WebCore::EditCommandComposition::setRangeDeletedByUnapply):
(WebCore::AccessibilityUndoReplacedText::confgureTextToBeDeletedByUnapplyIndexesWithEditCommandEndingSelection): Deleted.
(WebCore::AccessibilityUndoReplacedText::confgureTextToBeDeletedByUnapplyStartIndexWithEditCommandStartingSelection): Deleted.
(WebCore::AccessibilityUndoReplacedText::setTextInsertedByUnapplyRange): Deleted.
(WebCore::AccessibilityUndoReplacedText::captureTextToBeDeletedByUnapply): Deleted.
(WebCore::AccessibilityUndoReplacedText::captureTextToBeDeletedByReapply): Deleted.
(WebCore::AccessibilityUndoReplacedText::textInsertedByUnapply): Deleted.
(WebCore::AccessibilityUndoReplacedText::textInsertedByReapply): Deleted.
(WebCore::EditCommandComposition::setTextInsertedByUnapplyRange): Deleted.
* editing/CompositeEditCommand.h:
* editing/Editor.cpp:
(WebCore::Editor::replaceSelectionWithFragment):
* editing/TypingCommand.cpp:
(WebCore::TypingCommand::postTextStateChangeNotificationForDeletion):
(WebCore::TypingCommand::insertTextAndNotifyAccessibility):
(WebCore::TypingCommand::insertLineBreakAndNotifyAccessibility):
(WebCore::TypingCommand::insertParagraphSeparatorAndNotifyAccessibility):
(WebCore::TypingCommand::insertParagraphSeparatorInQuotedContentAndNotifyAccessibility):

LayoutTests:

* accessibility/mac/value-change/value-change-user-info-contenteditable-expected.txt:
* accessibility/mac/value-change/value-change-user-info-contenteditable.html:
* accessibility/mac/value-change/value-change-user-info-textarea-expected.txt:
* accessibility/mac/value-change/value-change-user-info-textarea.html:
* accessibility/mac/value-change/value-change-user-info-textfield-expected.txt:
* accessibility/mac/value-change/value-change-user-info-textfield.html:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@200876 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoUnreviewed, rolling out r200837.
cdumez@apple.com [Fri, 13 May 2016 19:19:01 +0000 (19:19 +0000)]
Unreviewed, rolling out r200837.

Seems to have regressed Speedometer and JetStream on iOS

Reverted changeset:

"DOMPromise should only restrict the resolution type"
https://bugs.webkit.org/show_bug.cgi?id=157307
http://trac.webkit.org/changeset/200837

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@200875 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoRename WKDataDetectorTypeSpotlightSuggestion to WKDataDetectorTypeLookupSuggestion
weinig@apple.com [Fri, 13 May 2016 19:14:32 +0000 (19:14 +0000)]
Rename WKDataDetectorTypeSpotlightSuggestion to WKDataDetectorTypeLookupSuggestion
<rdar://problem/26269611>

Reviewed by Dan Bernstein and Enrica Casucci.

Source/WebCore:

* editing/cocoa/DataDetection.h:
* editing/cocoa/DataDetection.mm:
(WebCore::constructURLStringForResult):
(WebCore::DataDetection::detectContentInRange):

Source/WebKit2:

* Shared/API/Cocoa/WKFoundation.h:
* Shared/API/c/WKDeclarationSpecifiers.h:
* UIProcess/API/Cocoa/WKWebView.mm:
(fromWKDataDetectorTypes):
* UIProcess/API/Cocoa/WKWebViewConfiguration.h:
* mac/postprocess-framework-headers.sh:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@200874 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoWeb Inspector: Script ProfileViews should be searchable
mattbaker@apple.com [Fri, 13 May 2016 19:01:34 +0000 (19:01 +0000)]
Web Inspector: Script ProfileViews should be searchable
https://bugs.webkit.org/show_bug.cgi?id=157581
<rdar://problem/26228530>

Reviewed by Joseph Pecoraro.

* UserInterface/Views/DataGrid.js:
(WebInspector.DataGrid.prototype.get filterText):
Make filterText readable.

* UserInterface/Views/ProfileDataGridNode.js:
(WebInspector.ProfileDataGridNode.prototype.get callingContextTreeNode):
(WebInspector.ProfileDataGridNode.prototype.filterableDataForColumn):
Add filterable data for the "function" column.
(WebInspector.ProfileDataGridNode.prototype._updateChildrenForModifiers):
(WebInspector.ProfileDataGridNode.prototype.get node): Deleted.
Renamed callingContextTreeNode to be less ambiguous.

* UserInterface/Views/ProfileDataGridTree.js:
(WebInspector.ProfileDataGridTree.prototype._updateCurrentFocusDetails):

* UserInterface/Views/ProfileView.js:
(WebInspector.ProfileView.prototype.get dataGrid):
Expose data grid for use in parent view.

* UserInterface/Views/ScriptClusterTimelineView.js:
(WebInspector.ScriptClusterTimelineView.prototype.selectRecord):
Drive-by fix: forward property to current child TimelineView.

* UserInterface/Views/ScriptProfileTimelineView.js:
(WebInspector.ScriptProfileTimelineView):
(WebInspector.ScriptProfileTimelineView.prototype._scopeBarSelectionDidChange):
(WebInspector.ScriptProfileTimelineView.prototype._showProfileViewForOrientation):
Helper function to switch profile views. Persist filter text when
switching to the new profile view.

(WebInspector.ScriptProfileTimelineView.prototype.get showsFilterBar): Deleted.
Remove FIXME and show filter bar.

* UserInterface/Views/TimelineView.js:
(WebInspector.TimelineView.prototype.setupDataGrid):
Support switching to a new data grid.
(WebInspector.TimelineView.prototype.dataGridMatchNodeAgainstCustomFilters):
Hooking up filtering causes data grid nodes to be filtered based on the
ruler selection. Although ScriptProfileTimelineView performs its own
time-based filtering, this is necessary to prevent an assert.

(WebInspector.TimelineView.prototype._timelineDataGridSelectedNodeChanged):
(WebInspector.TimelineView.prototype._timelineDataGridNodeWasFiltered):
Converted arrow functions to member functions to allow unregistering
event listeners on outgoing data grid when swapping grids.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@200873 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoAX: Wrong CharacterOffset from an upstream VisiblePosition
n_wang@apple.com [Fri, 13 May 2016 18:57:38 +0000 (18:57 +0000)]
AX: Wrong CharacterOffset from an upstream VisiblePosition
https://bugs.webkit.org/show_bug.cgi?id=157644

Reviewed by Chris Fleizach.

Source/WebCore:

We should adjust the text marker's CharacterOffset if its corresponding
VisiblePosition has upstream affinity.

Test: accessibility/mac/character-offset-from-upstream-position.html

* accessibility/AXObjectCache.cpp:
(WebCore::AXObjectCache::characterOffsetForTextMarkerData):
(WebCore::AXObjectCache::traverseToOffsetInRange):
(WebCore::AXObjectCache::localCaretRectForCharacterOffset):

LayoutTests:

* accessibility/mac/character-offset-from-upstream-position-expected.txt: Added.
* accessibility/mac/character-offset-from-upstream-position.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@200872 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoBuild fix.
bdakin@apple.com [Fri, 13 May 2016 18:45:25 +0000 (18:45 +0000)]
Build fix.

* WebView/WebView.mm:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@200867 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoSource/JavaScriptCore:
bdakin@apple.com [Fri, 13 May 2016 18:29:52 +0000 (18:29 +0000)]
Source/JavaScriptCore:
Add dyldSPI.h for linked on or after checks, and add one for link preview
https://bugs.webkit.org/show_bug.cgi?id=157401
-and corresponding-
rdar://problem/26253396

Reviewed by Darin Adler.

Import #import <wtf/spi/darwin/dyldSPI.h> which now declares all of the
needed dyld code.
* API/JSWrapperMap.mm:

Source/WebCore:
Add dyldSPI.h for linked on or after checks, and add one for link preview
https://bugs.webkit.org/show_bug.cgi?id=157401
-and corresponding-
rdar://problem/26253396

Reviewed by Darin Adler.

This patch removes DynamicLinkerSPI.h and replaces it with
wtf/spi/darwin/dyldSPI.h

* WebCore.xcodeproj/project.pbxproj:
* html/HTMLObjectElement.cpp:
* platform/graphics/ios/FontServicesIOS.mm:
* platform/spi/cocoa/DynamicLinkerSPI.h: Removed.

Source/WebKit/ios:
Add dyldSPI.h for linked on or after checks, and add one for link preview
https://bugs.webkit.org/show_bug.cgi?id=157401
-and corresponding-
rdar://problem/26253396

Reviewed by Darin Adler.

This patch removes DynamicLinkerSPI.h and replaces it with
wtf/spi/darwin/dyldSPI.h

* Misc/WebUIKitSupport.mm:

Source/WebKit/mac:
Add dyldSPI.h for linked on or after checks, and add one for link preview
https://bugs.webkit.org/show_bug.cgi?id=157401
-and corresponding-
rdar://problem/26253396

Reviewed by Darin Adler.

This patch removes DynamicLinkerSPI.h and replaces it with
wtf/spi/darwin/dyldSPI.h

* WebView/WebView.mm:

Source/WebKit2:
Add dyldSPI.h for linked on or after checks, and add one for link preview
https://bugs.webkit.org/show_bug.cgi?id=157401

Reviewed by Darin Adler.

Set _allowsLinkPreview to YES for programs linked on or after
firstSDKVersionWithLinkPreviewEnabledByDefault.
* UIProcess/API/Cocoa/WKWebView.h:
* UIProcess/API/Cocoa/WKWebView.mm:
(-[WKWebView _initializeWithConfiguration:]):

Source/WTF:
Add dyldSPI.h
https://bugs.webkit.org/show_bug.cgi?id=157401
-and corresponding-
rdar://problem/26253396

Reviewed by Darin Adler.

New header for dyld spi.
* WTF.xcodeproj/project.pbxproj:
* wtf/spi/darwin/dyldSPI.h: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@200866 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoTurn on WEB_TIMING for all ports
yoav@yoav.ws [Fri, 13 May 2016 18:06:43 +0000 (18:06 +0000)]
Turn on WEB_TIMING for all ports
https://bugs.webkit.org/show_bug.cgi?id=157673

Reviewed by Alex Christensen.

.:

Turn on WEB_TIMING by default on the cmake Mac port, to match it with all other ports,
and make sure that the flag will be on by default for all ports.

* Source/cmake/OptionsGTK.cmake: Remove the specific private flag for WEB_TIMING.
* Source/cmake/OptionsMac.cmake: Remove the specific private flag for WEB_TIMING.
* Source/cmake/WebKitFeatures.cmake: Turn on WEB_TIMING for all ports by default.

Tools:

* Scripts/webkitperl/FeatureList.pm: Make sure that WEB_TIMING is on by default.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@200863 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoFullscreen view attempts to resize when in 2-up mode
jer.noble@apple.com [Fri, 13 May 2016 17:42:18 +0000 (17:42 +0000)]
Fullscreen view attempts to resize when in 2-up mode
https://bugs.webkit.org/show_bug.cgi?id=157558

Reviewed by Darin Adler.

No need to react to NSApplicationDidChangeScreenParametersNotification to resize the
fullscreen window when the screen parameters change; the NSWindow fullscreen machinery
will do that on our behalf.

* UIProcess/mac/WKFullScreenWindowController.mm:
(-[WKFullScreenWindowController windowDidLoad]): Deleted.
(-[WKFullScreenWindowController applicationDidChangeScreenParameters:]): Deleted.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@200862 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoGive CrossThreadCopier a default specialization for Vectors of objects.
beidson@apple.com [Fri, 13 May 2016 17:37:23 +0000 (17:37 +0000)]
Give CrossThreadCopier a default specialization for Vectors of objects.
https://bugs.webkit.org/show_bug.cgi?id=157675

Reviewed by Darin Adler.

No new tests (Refactor, no change in behavior).

* platform/CrossThreadCopier.cpp:
(WebCore::Vector<String>>::copy): Deleted.
* platform/CrossThreadCopier.h: Add a default specialization for Vector<T> where T is a type
  that is CrossThreadCopyable itself.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@200861 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoMarking fast/text-autosizing/ios/text-autosizing-after-back.html as a flaky timeout
ryanhaddad@apple.com [Fri, 13 May 2016 17:29:22 +0000 (17:29 +0000)]
Marking fast/text-autosizing/ios/text-autosizing-after-back.html as a flaky timeout
https://bugs.webkit.org/show_bug.cgi?id=157589

Unreviewed test gardening.

* TestExpectations:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@200860 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoProtector Ref/RefPtrs should have a specified naming style.
beidson@apple.com [Fri, 13 May 2016 16:16:51 +0000 (16:16 +0000)]
Protector Ref/RefPtrs should have a specified naming style.
https://bugs.webkit.org/show_bug.cgi?id=157591

Reviewed by Darin Adler.

Tools:

* Scripts/webkitpy/style/checkers/cpp.py:
(check_identifier_name_in_declaration):
(CppChecker):
* Scripts/webkitpy/style/checkers/cpp_unittest.py:
(WebKitStyleTest.test_names):

Websites/webkit.org:

* code-style.md:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@200859 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years ago[iOS] <select> elements popover should render right-aligned when in RTL mode
commit-queue@webkit.org [Fri, 13 May 2016 16:06:03 +0000 (16:06 +0000)]
[iOS] <select> elements popover should render right-aligned when in RTL mode
https://bugs.webkit.org/show_bug.cgi?id=157672
<rdar://problem/26193442>

Patch by Antoine Quint <graouts@apple.com> on 2016-05-13
Reviewed by Darin Adler.

Add a new isRTL field to the AssistedNodeInformation as specified by the assisted
node's render style and account for it when displaying the table view shown in the
popover attached to the assisted node using the UIView semanticContentAttribute
property.

* Shared/AssistedNodeInformation.cpp:
(WebKit::AssistedNodeInformation::encode):
(WebKit::AssistedNodeInformation::decode):
* Shared/AssistedNodeInformation.h:
(WebKit::AssistedNodeInformation::AssistedNodeInformation):
* UIProcess/ios/forms/WKFormSelectPopover.mm:
(-[WKSelectTableViewController initWithView:hasGroups:]):
(-[WKSelectTableViewController tableView:cellForRowAtIndexPath:]):
* WebProcess/WebPage/ios/WebPageIOS.mm:
(WebKit::WebPage::getAssistedNodeInformation):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@200858 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoHave the WorkerGlobalScope API return more references
cdumez@apple.com [Fri, 13 May 2016 14:58:53 +0000 (14:58 +0000)]
Have the WorkerGlobalScope API return more references
https://bugs.webkit.org/show_bug.cgi?id=157663

Reviewed by Alexey Proskuryakov.

Have the WorkerGlobalScope API return more references instead of
pointers.

* workers/WorkerGlobalScope.cpp:
(WebCore::WorkerGlobalScope::location):
(WebCore::WorkerGlobalScope::navigator):
* workers/WorkerGlobalScope.h:
(WebCore::WorkerGlobalScope::self):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@200857 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoAssertion failure for direct eval in non-class method
utatane.tea@gmail.com [Fri, 13 May 2016 14:28:30 +0000 (14:28 +0000)]
Assertion failure for direct eval in non-class method
https://bugs.webkit.org/show_bug.cgi?id=157138

Reviewed by Saam Barati.

This assertion was incorrect. In method definitions in object literals,
it can be sloppy mode, but its DerivedContextType may not be DerivedContextType::None.

* bytecode/EvalCodeCache.h:
(JSC::EvalCodeCache::CacheKey::CacheKey):
(JSC::EvalCodeCache::CacheKey::operator==):
(JSC::EvalCodeCache::CacheKey::Hash::equal):
(JSC::EvalCodeCache::tryGet):
(JSC::EvalCodeCache::getSlow):
* interpreter/Interpreter.cpp:
(JSC::eval):
* tests/stress/direct-eval-in-object-literal-methods.js: Added.
(shouldBe):
(throw.new.Error):
(shouldBe.Parent.prototype.l):
(shouldBe.Parent):
(shouldBe.Derived.prototype.m):
(shouldBe.Derived):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@200856 268f45cc-cd09-0410-ab3c-d52691b4dbfc