Timeline
Apr 26, 2013:
- 11:43 PM Changeset in webkit [149236] by
-
- 16 edits2 adds in trunk
Source/JavaScriptCore: Re-landing <http://trac.webkit.org/changeset/148999>
Filled out more cases of branch folding in bytecode when emitting
expressions into a branching context
https://bugs.webkit.org/show_bug.cgi?id=115057
Reviewed by Phil Pizlo.
We can't fold the number == 1 case to boolean because all non-zero numbers
down-cast to true, but only 1 is == to true.
Source/WTF: Filled out more cases of branch folding in bytecode when emitting expressions into a branching context
https://bugs.webkit.org/show_bug.cgi?id=115057
Reviewed by Filip Pizlo.
Added a helper constructor for TriState so clients can make one without
branching or making assumptions about the integer values of TriStates.
- wtf/TriState.h:
(WTF::triState):
(WTF):
LayoutTests: Re-landing <http://trac.webkit.org/changeset/148999>
Filled out more cases of branch folding in bytecode when emitting
expressions into a branching context
https://bugs.webkit.org/show_bug.cgi?id=115057
Reviewed by Phil Pizlo.
Added a more exhaustive correctness test.
- fast/js/branch-fold-correctness-expected.txt: Added.
- fast/js/branch-fold-correctness.html: Added.
- 11:33 PM Changeset in webkit [149235] by
-
- 2 edits in branches/dfgFourthTier/Source/JavaScriptCore
Fix indentation of SymbolTable.h
Rubber stamped by Mark Hahnenberg.
- runtime/SymbolTable.h:
- 11:28 PM Changeset in webkit [149234] by
-
- 2 edits in trunk/Source/JavaScriptCore
Correct indentation of SymbolTable.h
Rubber stamped by Mark Hahnenberg.
- runtime/SymbolTable.h:
- 10:44 PM Changeset in webkit [149233] by
-
- 16 edits in branches/dfgFourthTier/Source/JavaScriptCore
fourthTier: CFA should defend against results seeming inconsistent due to a watchpoint firing during compilation
https://bugs.webkit.org/show_bug.cgi?id=115083
Reviewed by Geoffrey Garen.
This ruggedizes our racyness with respect to watchpoints. We want to be able to assert,
in some places, that a watchpoint-based optimization has only occurred if the
watchpoint set was still valid. But currently we *can* soundly do watchpoint-based
optimizations even for invalid watchpoints, so long as we recorded in the IR that we
had done so; this will then lead to the code being insta-jettisoned after compilation
completes. Obviously, we don't want this to happen often - but we do want to allow it
precisely in the case of watchpoint races.
This adds the ability to assert that we hadn't over-watchpointed ourselves, with and
exemption for races.
- dfg/DFGAbstractState.cpp:
(JSC::DFG::AbstractState::executeEffects):
- dfg/DFGAbstractValue.cpp:
(JSC::DFG::AbstractValue::setFuturePossibleStructure):
(JSC::DFG::AbstractValue::filterFuturePossibleStructure):
- dfg/DFGByteCodeParser.cpp:
(JSC::DFG::ByteCodeParser::addStructureTransitionCheck):
(JSC::DFG::ByteCodeParser::parseResolveOperations):
(JSC::DFG::ByteCodeParser::parseBlock):
- dfg/DFGConstantFoldingPhase.cpp:
(JSC::DFG::ConstantFoldingPhase::addStructureTransitionCheck):
- dfg/DFGDesiredWatchpoints.h:
(GenericDesiredWatchpoints):
(JSC::DFG::GenericDesiredWatchpoints::isStillValid):
(JSC::DFG::GenericDesiredWatchpoints::shouldAssumeMixedState):
(JSC::DFG::GenericDesiredWatchpoints::isValidOrMixed):
(JSC::DFG::DesiredWatchpoints::isStillValid):
(JSC::DFG::DesiredWatchpoints::shouldAssumeMixedState):
(JSC::DFG::DesiredWatchpoints::isValidOrMixed):
(DesiredWatchpoints):
- dfg/DFGFixupPhase.cpp:
(JSC::DFG::FixupPhase::canOptimizeStringObjectAccess):
- dfg/DFGGraph.h:
(JSC::DFG::Graph::masqueradesAsUndefinedWatchpointIsStillValid):
(Graph):
- dfg/DFGJITCompiler.cpp:
(JSC::DFG::JITCompiler::link):
(JSC::DFG::JITCompiler::compile):
(JSC::DFG::JITCompiler::compileFunction):
- dfg/DFGJITCompiler.h:
(JSC::DFG::JITCompiler::addLazily):
(JITCompiler):
- dfg/DFGSpeculativeJIT.cpp:
(JSC::DFG::SpeculativeJIT::compilePeepHoleObjectEquality):
- dfg/DFGSpeculativeJIT.h:
(SpeculativeJIT):
(JSC::DFG::SpeculativeJIT::masqueradesAsUndefinedWatchpointIsStillValid):
(JSC::DFG::SpeculativeJIT::speculationWatchpointForMasqueradesAsUndefined):
(JSC::DFG::SpeculativeJIT::speculateStringObjectForStructure):
- dfg/DFGSpeculativeJIT32_64.cpp:
(JSC::DFG::SpeculativeJIT::nonSpeculativeNonPeepholeCompareNull):
(JSC::DFG::SpeculativeJIT::nonSpeculativePeepholeBranchNull):
(JSC::DFG::SpeculativeJIT::compileObjectEquality):
(JSC::DFG::SpeculativeJIT::compileObjectToObjectOrOtherEquality):
(JSC::DFG::SpeculativeJIT::compilePeepHoleObjectToObjectOrOtherEquality):
(JSC::DFG::SpeculativeJIT::compileObjectOrOtherLogicalNot):
(JSC::DFG::SpeculativeJIT::emitObjectOrOtherBranch):
(JSC::DFG::SpeculativeJIT::compile):
- dfg/DFGSpeculativeJIT64.cpp:
(JSC::DFG::SpeculativeJIT::nonSpeculativeNonPeepholeCompareNull):
(JSC::DFG::SpeculativeJIT::nonSpeculativePeepholeBranchNull):
(JSC::DFG::SpeculativeJIT::compileObjectEquality):
(JSC::DFG::SpeculativeJIT::compileObjectToObjectOrOtherEquality):
(JSC::DFG::SpeculativeJIT::compilePeepHoleObjectToObjectOrOtherEquality):
(JSC::DFG::SpeculativeJIT::compileObjectOrOtherLogicalNot):
(JSC::DFG::SpeculativeJIT::emitObjectOrOtherBranch):
(JSC::DFG::SpeculativeJIT::compile):
- ftl/FTLCompile.cpp:
(JSC::FTL::compile):
- ftl/FTLState.h:
(State):
- 10:30 PM Changeset in webkit [149232] by
-
- 5 edits3 adds in trunk
Video playback has corruption on the edges of the video
https://bugs.webkit.org/show_bug.cgi?id=115216
Reviewed by Simon Fraser.
Source/WebCore:
Test: media/video-poster-background.html
RenderImage correctly answers foregroundIsKnownToBeOpaqueInRect() method
when a poster image is displayed, but once the video begins playing and
the poster image is no longer displayed, RenderImage will continue to
answer 'true' even when the video does not occupy the entire render box.
Override foregroundIsKnownToBeOpaqueInRect() in order to more correctly
answer the question for the video layer.
- rendering/RenderImage.h: Make foregroundIsKnownToBeOpaqueInRect() protected
(vs. private).
- rendering/RenderVideo.cpp:
(WebCore::RenderVideo::foregroundIsKnownToBeOpaqueInRect): Added. Return
true if the rect parameter is entirely contained by the video box.
- rendering/RenderVideo.h:
LayoutTests:
- media/content/test.jpg: Added.
- media/video-poster-background-expected.html: Added.
- media/video-poster-background.html: Added.
- 9:41 PM Changeset in webkit [149231] by
-
- 2 edits in trunk/Tools
[EFL] Build break using jhbuild on ubuntu 13.04
https://bugs.webkit.org/show_bug.cgi?id=115225
Reviewed by Gyuyoung Kim.
libxslt in Ubuntu 13.04 requires libxml 2.9.0 or higher and
libxml 2.9.0 has a bug to make regression which r138422 mentioned.
So this patch bumped libxml version to 2.9.1.
- efl/jhbuild.modules:
- 6:31 PM Changeset in webkit [149230] by
-
- 3 edits in trunk/Source/WebKit2
[WK2][EFL] Build break after r149212
https://bugs.webkit.org/show_bug.cgi?id=115294
Patch by Yael Aharon <yael.aharon@intel.com> on 2013-04-26
Reviewed by Simon Fraser.
r144672 bumped up the revision of WKContextClient API, and introduced a version array
APIClientTraits<WKContextClient>::interfaceSizesByVersion .
In r149212, the version was bumped back down and the version array was emptied but not removed.
This causes a warning "array subscript is below array bounds", and in EFL the warning is
treated as error.
- Shared/APIClientTraits.cpp:
(WebKit):
- Shared/APIClientTraits.h:
- 5:51 PM Changeset in webkit [149229] by
-
- 2 edits in trunk/LayoutTests
Unreviewed. Skipping some more tests that still need rebaselining after disabling subpixel layout on mac.
http://bugs.webkit.org/show_bug.cgi?id=115292
- 5:51 PM Changeset in webkit [149228] by
-
- 2 edits in trunk/Tools
delete-stale-build-files does not clear .d files
http://bugs.webkit.org/show_bug.cgi?id=115293
Reviewed by Ryosuke Niwa.
- BuildSlaveSupport/delete-stale-build-files:
(main): Add .d to the file type list
- 5:38 PM Changeset in webkit [149227] by
-
- 2 edits in trunk/WebKitLibraries
Disable sub-pixel-layout on Apple Windows port, VS2005 edition.
- win/tools/vsprops/FeatureDefines.vsprops:
- 5:35 PM Changeset in webkit [149226] by
-
- 12 edits in trunk/Source/WebCore
[CSS Exclusions] ExclusionShape bounding box methods should return LayoutRects
https://bugs.webkit.org/show_bug.cgi?id=115117
Reviewed by Dirk Schulze.
Redefined the ExclusionShape API in terms of LayoutUnits, instead of floats: all of the
ExclusionShape methods now have LayoutUnit parameters and return LayoutUnit values.
This is more natural, since the callers work exclusively in LayoutUnits.
This is strictly a refactoring, no new tests were needed.
- rendering/ExclusionPolygon.cpp:
(WebCore::ExclusionPolygon::getExcludedIntervals):
(WebCore::ExclusionPolygon::getIncludedIntervals):
(WebCore::ExclusionPolygon::firstIncludedIntervalLogicalTop):
- rendering/ExclusionPolygon.h:
- rendering/ExclusionRectangle.cpp:
(WebCore::ExclusionRectangle::getExcludedIntervals):
(WebCore::ExclusionRectangle::getIncludedIntervals):
(WebCore::ExclusionRectangle::firstIncludedIntervalLogicalTop):
- rendering/ExclusionRectangle.h:
- rendering/ExclusionShape.cpp:
(WebCore::ExclusionShape::createExclusionShape):
- rendering/ExclusionShape.h:
(LineSegment):
(ExclusionShape):
- rendering/ExclusionShapeInfo.cpp:
(WebCore):
(WebCore::::computedShape):
- rendering/ExclusionShapeInfo.h:
(WebCore):
(WebCore::ExclusionShapeInfo::shapeLogicalTop):
(WebCore::ExclusionShapeInfo::shapeLogicalBottom):
(ExclusionShapeInfo):
- rendering/ExclusionShapeInsideInfo.cpp:
(WebCore::ExclusionShapeInsideInfo::adjustLogicalLineTop):
- rendering/ExclusionShapeInsideInfo.h:
- rendering/ExclusionShapeOutsideInfo.h:
- 5:17 PM Changeset in webkit [149225] by
-
- 2 edits in trunk/Source/WebCore
Nil-check the results of -[AVPlayerItemVideoOutput copyPixelBufferForItemTime:itemTimeForDisplay:].
https://bugs.webkit.org/show_bug.cgi?id=115265
Reviewed by Eric Carlson.
copyPixelBufferForItemTime:itemTimeForDisplay: can return nil, even
when hasNewPixelBufferForItemTime: returns YES. Check the results
before passing the buffer on to VTPixelTransferSessionTransferImage()
which does not NULL-check its parameters.
- platform/graphics/avfoundation/objc/MediaPlayerPrivateAVFoundationObjC.mm:
(WebCore::MediaPlayerPrivateAVFoundationObjC::createPixelBuffer):
- 5:09 PM Changeset in webkit [149224] by
-
- 2 edits in trunk/LayoutTests
Unreviewed. Skipping some tests that still need rebaselining after disabling subpixel layout on mac.
http://bugs.webkit.org/show_bug.cgi?id=115292
- 4:45 PM Changeset in webkit [149223] by
-
- 2 edits in trunk/Source/WebCore
Uninflate caret rect.
http://bugs.webkit.org/show_bug.cgi?id=114997.
<rdar://problem/12629007>.
Reviewed by Timothy Horton.
Inflation was causing painting errors with focus rings.
It is no longer needed after disabling subpixel layout.
- editing/FrameSelection.cpp:
(WebCore::repaintCaretForLocalRect):
- 4:40 PM Changeset in webkit [149222] by
-
- 3 edits66 adds in trunk/LayoutTests
Unreviewed. More rebaselines for mac after disabling sub-pixel layout.
- 4:40 PM Changeset in webkit [149221] by
-
- 2 edits in trunk/WebKitLibraries
Disable sub-pixel-layout on Apple Windows port.
- win/tools/vsprops/FeatureDefines.props:
- 4:17 PM Changeset in webkit [149220] by
-
- 216 edits1 delete in trunk
Make Apple Windows VS2010 build results into and get dependencies from 32 suffixed folders.
Make the DebugSuffix configuration use _debug dependencies.
- 4:06 PM Changeset in webkit [149219] by
-
- 5 edits in trunk/Source/WebKit2
[GTK] Add methods to add a user style sheet to the WebKit2 GTK+ API
https://bugs.webkit.org/show_bug.cgi?id=99081
Reviewed by Carlos Garcia Campos, Gustavo Noronha Silva, and Benjamin Poulain.
Add methods to WebKitWebViewGroup to add and remove user style sheets.
This allows clients to inject style sheets into pages with a set of
rules for when those style sheets apply.
- UIProcess/API/gtk/WebKitWebViewGroup.cpp:
(toImmutableArray): Added this helper which converts the GList* parameters into
ImmutableArrays for use with the WebKit2 internal API.
(webkit_web_view_group_add_user_style_sheet): Added new API for adding a style sheet.
(webkit_web_view_group_remove_all_user_style_sheets): Add new API for clearing out all style sheets.
- UIProcess/API/gtk/WebKitWebViewGroup.h: Added new method declarations.
- UIProcess/API/gtk/docs/webkit2gtk-sections.txt: Added new API to the documentation.
- UIProcess/API/gtk/tests/TestWebKitWebViewGroup.cpp: Added a test for the new API.
(isStyleSheetInjectedForURLAtPath): Function to check whether the style sheet has been injected for a given URL.
(fillURLListFromPaths): Helper which converts paths passed via varargs into a whitelist or blacklist.
(removeOldInjectedStyleSheetsAndResetLists): Function to start afresh.
(testWebViewGroupInjectedStyleSheet): The actual test.
(serverCallback): Server callback for use with the test. We cannot use loadHTML or
loadAlternateHTML, because that checks the whitelist and blacklist against about:blank.
(beforeAll): Initialize the server and new test.
(afterAll): Clean up the server.
- 3:29 PM Changeset in webkit [149218] by
-
- 2 edits in trunk/Source/WebCore
Unreviewed build fix.
- platform/LayoutUnit.h:
(WebCore::LayoutUnit::LayoutUnit):
- 3:22 PM Changeset in webkit [149217] by
-
- 8 edits in trunk/Source/WebKit2
2013-04-26 Simon Cooper <scooper@apple.com>
Permit ~/Library or ~/Library/Preferences to be symlinks
https://bugs.webkit.org/show_bug.cgi?id=115142
<rdar://problem/12953603>
Reviewed by Alexey Proskuryakov.
Pass in two new parameters which are the resolved paths to ~/Library
and ~/Library/Preferences. Added new expansion functions and
substituted the new functions for home-regex, home-subpath and
home-literal for places referencing resources in "~/Library" or
"~/Library/Preferences".
- Resources/PlugInSandboxProfiles/com.apple.QuickTime Plugin.plugin.sb:
- Resources/PlugInSandboxProfiles/com.apple.WebKit.plugin-common.sb:
- Resources/PlugInSandboxProfiles/com.apple.ist.ds.appleconnect.webplugin.sb:
- Resources/PlugInSandboxProfiles/com.macromedia.Flash Player.plugin.sb:
- Resources/PlugInSandboxProfiles/com.microsoft.SilverlightPlugin.sb:
- Resources/PlugInSandboxProfiles/com.oracle.java.JavaAppletPlugin.sb:
- Shared/mac/ChildProcessMac.mm: (WebKit::ChildProcess::initializeSandbox):
- 3:20 PM Changeset in webkit [149216] by
-
- 4 edits in trunk/Source/WebCore
WebCore ObjC bridge is missing support for bool type
https://bugs.webkit.org/show_bug.cgi?id=115276
Reviewed by Geoffrey Garen.
Added code to handle conversion between ObjC and JS booleans.
- bridge/objc/objc_instance.mm:
(ObjcInstance::invokeObjcMethod):
- bridge/objc/objc_utility.h:
- bridge/objc/objc_utility.mm:
(JSC::Bindings::convertValueToObjcValue):
(JSC::Bindings::convertObjcValueToValue):
(JSC::Bindings::objcValueTypeForType):
- 3:15 PM Changeset in webkit [149215] by
-
- 1 edit in tags/Safari-537.39.1/Source/WebKit/mac/WebCoreSupport/WebEditorClient.mm
Build fix.
- 3:15 PM Changeset in webkit [149214] by
-
- 4 edits in tags/Safari-537.39.1/Source
Versioning.
- 3:09 PM Changeset in webkit [149213] by
-
- 1 copy in tags/Safari-537.39.1
New tag.
- 2:54 PM Changeset in webkit [149212] by
-
- 4 edits in trunk/Source/WebKit2
[WK2] WKContextClient doesn't need to be versioned yet
https://bugs.webkit.org/show_bug.cgi?id=115279
Reviewed by Anders Carlsson.
- Shared/APIClientTraits.cpp:
- Shared/APIClientTraits.h:
- UIProcess/API/C/WKContext.h:
- 2:39 PM Changeset in webkit [149211] by
-
- 8 edits in trunk/Source/WebKit2
Downloads need to be converted from the NSURLConnection, not canceled and restarted, when using the NetworkProcess
https://bugs.webkit.org/show_bug.cgi?id=115277
<rdar://problem/12890184>
Reviewed by Sam Weinig.
- NetworkProcess/NetworkConnectionToWebProcess.cpp:
(WebKit::NetworkConnectionToWebProcess::convertMainResourceLoadToDownload):
Ask the download manager to convert the main resource load to a download.
- NetworkProcess/NetworkConnectionToWebProcess.messages.in:
Add new message.
- NetworkProcess/NetworkResourceLoader.h:
(WebKit::NetworkResourceLoader::handle):
Add getter.
- Shared/Downloads/mac/DownloadMac.mm:
(dispatchOnMainThread):
Helper function that ensures that a given block is called on the main thread.
(-[WKDownloadAsDelegate downloadDidBegin:]):
(-[WKDownloadAsDelegate download:didReceiveAuthenticationChallenge:]):
(-[WKDownloadAsDelegate download:didReceiveResponse:]):
(-[WKDownloadAsDelegate download:didReceiveDataOfLength:]):
(-[WKDownloadAsDelegate download:shouldDecodeSourceDataOfMIMEType:]):
(-[WKDownloadAsDelegate download:decideDestinationWithSuggestedFilename:]):
(-[WKDownloadAsDelegate download:didCreateDestination:]):
(-[WKDownloadAsDelegate downloadDidFinish:]):
(-[WKDownloadAsDelegate download:didFailWithError:]):
Use dispatchOnMainThread.
- UIProcess/WebPageProxy.cpp:
(WebKit::WebPageProxy::receivedPolicyDecision):
Remove code that cancels the current load and starts a new download.
- WebProcess/WebPage/WebFrame.cpp:
(WebKit::WebFrame::convertMainResourceLoadToDownload):
Send the ConvertMainResourceLoadToDownload message to the network process.
- 2:25 PM Changeset in webkit [149210] by
-
- 865 edits359 adds262 deletes in trunk/LayoutTests
Unreviewed. Rebaseline a bunch of tests and run optimize-baselines after disabling sub-pixel layout on mac.
https://bugs.webkit.org/show_bug.cgi?id=114999.
- 2:19 PM Changeset in webkit [149209] by
-
- 9 edits in trunk/Source
Disable sub-pixel layout on mac.
https://bugs.webkit.org/show_bug.cgi?id=114999.
Reviewed by Simon Fraser.
- Configurations/FeatureDefines.xcconfig:
- WebCore.exp.in:
Export symbol needed to be added for IntRect(const LayoutRect&) constructor,
which is now being used in pixelSnappedIntRect(const LayoutRect&) in LayoutRect.h after disabling sub-pixel layout.
- 2:13 PM Changeset in webkit [149208] by
-
- 2 edits in trunk/Source/WebKit2
Inspector window is often blank when opened
https://bugs.webkit.org/show_bug.cgi?id=115220
<rdar://problem/13735496>
Reviewed by Tim Horton.
After disabling tiled scrolling and accelerated drawing in r149006, the
Inspector window was often missing its content after opening. This only
happened when the user has a preference on disk to use accelerated drawing
in the Inspector.
The missing content was caused by a failure to connect the correct contextID
for remote layer hosting. On window creation, we hook up a context using
WindowServer layer hosting (the default), but then switch back to app-hosted
layers for the Inspector window. This switching happens in didUpdateBackingStoreState(),
but that can be re-entered via its call to sendUpdateBackingStoreState(). In
that cause, on unwinding the stack it would call enterAcceleratedCompositingMode()
with a stale layerTreeContext.
Fix by moving the code that exits/enters accelerated compositing mode to before
the call to sendUpdateBackingStoreState() that can cause re-entrancy. That avoids
use of a possibly stale layerTreeContext.
- UIProcess/DrawingAreaProxyImpl.cpp:
(WebKit::DrawingAreaProxyImpl::didUpdateBackingStoreState):
- 2:13 PM Changeset in webkit [149207] by
-
- 4 edits in trunk/Source
Layer-backed WebViews don't repaint content outside the visible area
https://bugs.webkit.org/show_bug.cgi?id=115275
Reviewed by Beth Dakin.
Source/WebCore:
Export ScrollView::setPaintsEntireContents(bool).
- WebCore.exp.in:
Source/WebKit/mac:
When a client of WebView makes the WebView be layer-backed, then
AppKit propagaes the layer-backing down to the WebHTMLView (but
no further).
In this situation, we had a bug where content that was not
in the visible content rect would not get repainted, because
WebCore clips repaints to the visible rect.
Fix by calling ScrollView::setPaintsEntireContents(bool) from
-[WebHTMLView setLayer:], saying that we need
to paint the entire contents if we have a layer.
- WebView/WebHTMLView.mm:
(-[WebHTMLView setLayer:]):
- 1:34 PM Changeset in webkit [149206] by
-
- 2 edits in trunk/Source/WebKit/blackberry
[BlackBerry] Check for valid element in elementTouched
https://bugs.webkit.org/show_bug.cgi?id=115205
Patch by Nima Ghanavatian <nghanavatian@blackberry.com> on 2013-04-26
Reviewed by Rob Buis.
Internally reviewed by Genevieve Mak.
PR 331546
We might receive a null ptr from nodeAsElementIfApplicable which
is passed in here. Check to make sure it's valid before using.
- WebKitSupport/InputHandler.cpp:
(BlackBerry::WebKit::InputHandler::elementTouched):
- 1:12 PM Changeset in webkit [149205] by
-
- 2 edits in trunk/Source/JavaScriptCore
Make stack tracing more robust
https://bugs.webkit.org/show_bug.cgi?id=115272
Reviewed by Geoffrey Garen.
CallFrame already handles stack walking confusion robustly,
so we should make sure that the actual walk handles that as well.
- interpreter/Interpreter.cpp:
(JSC::getCallerInfo):
- 12:42 PM Changeset in webkit [149204] by
-
- 2 edits in trunk/Source/WebKit2
Crashes under ResourceHandle::continueDidReceiveResponse when loading blobs
https://bugs.webkit.org/show_bug.cgi?id=115273
Reviewed by Anders Carlsson.
Covered by multiple existing tests (when using NetworkProcess), including:
- http/tests/fileapi/blob-url-in-subframe.html
- http/tests/fileapi/create-blob-url-from-data-url.html
- NetworkProcess/NetworkResourceLoader.cpp: (WebKit::NetworkResourceLoader::continueDidReceiveResponse): Work around BlobResourceHandle badness.
- 12:09 PM Changeset in webkit [149203] by
-
- 2 edits in trunk/Source/WebCore
Change RenderMeter::valueRatio() visibility to public
https://bugs.webkit.org/show_bug.cgi?id=115266
This function is currently unused.
However, it is necessary to implement a RenderTheme capable of rendering Meter elements.
Patch by Daker Fernandes Pinheiro <daker.pinheiro@openbossa.org> on 2013-04-26
Reviewed by Alexis Menard.
No new tests, because no behaviour is changed.
- rendering/RenderMeter.h:
(RenderMeter):
- 12:08 PM Changeset in webkit [149202] by
-
- 6 edits1 add in trunk/Source
Add Runtime.parse to the Inspector protocol.
This will be used to parse console expressions for errors
before evaluating them fully.
Reviewed by Oliver Hunt.
- ForwardingHeaders/parser/ParserError.h: Added.
- inspector/Inspector.json:
- inspector/InspectorRuntimeAgent.cpp:
(WebCore::InspectorRuntimeAgent::parse):
- inspector/InspectorRuntimeAgent.h:
(InspectorRuntimeAgent):
- 12:04 PM Changeset in webkit [149201] by
-
- 2 edits in trunk/Source/WebCore
Fix the copyright years after r149057
Patch by Benjamin Poulain <bpoulain@apple.com> on 2013-04-26
- page/DiagnosticLoggingKeys.cpp: I accidentally removed a year from
the copyright in r149057.
- 11:57 AM Changeset in webkit [149200] by
-
- 3 edits2 adds in trunk
Web Inspector: Crash due to null items from getDOMStorageItems
https://bugs.webkit.org/show_bug.cgi?id=115176
Patch by Konrad Piascik <kpiascik@blackberry.com> on 2013-04-26
Reviewed by Joseph Pecoraro.
Source/WebCore:
findStorageArea was returning a null storageArea causing the items
input paramater to not be set. This was happening without any error
being set at all. Set an error to prevent a crash when we try to
convert the result to a JSON string.
Added tests to check if session and local storage are empty that
they are still functional.
- inspector/InspectorDOMStorageAgent.cpp:
(WebCore::InspectorDOMStorageAgent::getDOMStorageItems):
LayoutTests:
Tests to check that if local and session storage are empty
that they still work corerctly.
- inspector/storage-panel-dom-storage-empty-expected.txt: Added.
- inspector/storage-panel-dom-storage-empty.html: Added.
- 11:40 AM Changeset in webkit [149199] by
-
- 3 edits in trunk/Source/WebKit2
[WK2] Remove NetworkResourceLoader::cleanupOnMainThread
https://bugs.webkit.org/show_bug.cgi?id=115268
Reviewed by Brady Eidson.
All cleanup() callers are now on main thread, and this indirection only complicates debugging.
- NetworkProcess/NetworkResourceLoader.cpp: (WebKit::NetworkResourceLoader::connectionToWebProcessDidClose): (WebKit::NetworkResourceLoader::abortInProgressLoad): (WebKit::NetworkResourceLoader::didFinishLoading): (WebKit::NetworkResourceLoader::didFail):
- NetworkProcess/NetworkResourceLoader.h:
- 11:24 AM Changeset in webkit [149198] by
-
- 3 edits in trunk/Source/WebKit2
[WebKit2] CustomProtocolManager's m_registeredSchemes HashSet can be accessed by multiple threads
https://bugs.webkit.org/show_bug.cgi?id=115267
Reviewed by Alexey Proskuryakov.
- Shared/Network/CustomProtocols/CustomProtocolManager.h: Added a Mutex to protect m_registeredSchemes.
- Shared/Network/CustomProtocols/mac/CustomProtocolManagerMac.mm:
(WebKit::CustomProtocolManager::registerScheme): Acquired the mutex before accessing m_registeredSchemes.
(WebKit::CustomProtocolManager::unregisterScheme): Ditto.
(WebKit::CustomProtocolManager::supportsScheme): Ditto.
- 11:10 AM Changeset in webkit [149197] by
-
- 2 edits in trunk/LayoutTests
[Mac] media/track/track-mode.html sometimes trigger InvalidStateError
https://bugs.webkit.org/show_bug.cgi?id=115045
Reviewed by Alexey Proskuryakov.
- media/track/track-mode.html: Don't assume that the video element is ready for seeking
after a 100ms timeout.
- 10:51 AM Changeset in webkit [149196] by
-
- 2 edits in trunk/Source/WebKit2
Blobs are not associated with connection when there are no sandbox extensions
https://bugs.webkit.org/show_bug.cgi?id=115264
Reviewed by Brady Eidson.
Covered by multiple fast/files tests when using NetworkProcess.
- NetworkProcess/FileAPI/NetworkBlobRegistry.cpp: (WebKit::NetworkBlobRegistry::registerBlobURL): Fixed a logic error.
- 10:41 AM Changeset in webkit [149195] by
-
- 2 edits in trunk/Tools
[GTK] Clean up the TestWebKitAPI GNUmakefile.am
https://bugs.webkit.org/show_bug.cgi?id=115233
Reviewed by Martin Robinson.
Specify unconditional, non-installable program targets in one declaration. Only conditionally specify
WebKit2-specific targets, not the complete target setups (i.e. targets' sources, cppflags listings etc.).
Add missing unit test source files to the TestWTF program, namely HashSet.cpp and MetaAllocator.cpp.
Specify the TestWTF program's CXXFLAGS to be the same as the global cxxflags as the source files need to
be compiled with the -fno-rtti flag that's present in the global_cxxflags variable.
- TestWebKitAPI/GNUmakefile.am:
- 9:55 AM Changeset in webkit [149194] by
-
- 11 edits3 copies2 adds in trunk
[WebKit2] Loading a resource from a custom protocol in a synchronous XHR times out
https://bugs.webkit.org/show_bug.cgi?id=115223
Reviewed by Darin Adler.
Source/WebKit2:
When WebKit performs a synchronous load on the Mac, it spins a nested
run loop in a mode that only accepts input from the NSURLConnection
performing the load. When the load is for a custom protocol in WebKit2,
we proxy it to the UI process via CoreIPC messages, but the response
messages from the UI process are never processed as long as the run
loop is in a non-common mode (and we wouldn't want to process messages
that could re-enter WebCore in the middle of loading, anyway). Since
these messages never make it back to the NSURLConnection handling the
request, the connection eventually times out.
Fix this by using a work queue to handle custom protocol messages in
the networking process. The work queue can process incoming custom
protocol messages while the main thread is blocked.
- NetworkProcess/NetworkProcess.cpp:
(WebKit::NetworkProcess::initializeConnection): Called
initializeConnection() on all the NetworkProcess's supplements.
- Shared/ChildProcessSupplement.h: Added a base class for
NetworkProcessSupplement and WebProcessSupplement which defines
initializeConnection and provides an empty default implementation.
(WebKit::ChildProcessSupplement::~ChildProcessSupplement):
(WebKit::ChildProcessSupplement::initializeConnection):
- Shared/Network/CustomProtocols/CustomProtocolManager.h:
- Shared/Network/CustomProtocols/mac/CustomProtocolManagerMac.mm:
(WebKit::CustomProtocolManager::CustomProtocolManager): Instantiated a
work queue for message processing.
(WebKit::CustomProtocolManager::initializeConnection): Added the work
queue as a message receiver on the CoreIPC connection.
(WebKit::CustomProtocolManager::initialize): If we're in the web
process and the network process is being used, unregister and destroy
the work queue we previously created. It'd be better to not create it
in the first place, but we have to register our work queue with the
CoreIPC connection before it is established, which is before the UI
process has told us whether the network process is in use.
- Shared/Network/NetworkProcessSupplement.h: Inherited from
ChildProcessSupplement.
- WebKit2.xcodeproj/project.pbxproj: Added ChildProcessSupplement.h.
- WebProcess/WebProcess.cpp:
(WebKit::WebProcess::initializeConnection): Called
initializeConnection() on all the WebProcess's supplements.
- WebProcess/WebProcessSupplement.h: Inherited from
ChildProcessSupplement.
Tools:
Added an API test.
- TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj: Added new files.
- TestWebKitAPI/Tests/CustomProtocolsSyncXHRTest.mm: Added.
(TestWebKitAPI::TEST): Tested that a synchronous XHR does not time out
when it loads a request with a custom protocol.
- TestWebKitAPI/Tests/WebKit2/custom-protocol-sync-xhr.html: Added.
- TestWebKitAPI/Tests/WebKit2ObjC/CustomProtocolsTest.mm: Moved the
NSURLProtocol subclass to TestProtocol.{h, mm} and did some
miscellaneous cleanup.
- TestWebKitAPI/mac/TestProtocol.h: Copied from Source/WebKit2/WebProcess/WebProcessSupplement.h.
- TestWebKitAPI/mac/TestProtocol.mm: Copied from Tools/TestWebKitAPI/Tests/WebKit2ObjC/CustomProtocolsTest.mm.
(+[TestProtocol canInitWithRequest:]):
(+[TestProtocol canonicalRequestForRequest:]):
(+[TestProtocol requestIsCacheEquivalent:toRequest:]):
(+[TestProtocol scheme]):
(-[TestProtocol startLoading]):
(-[TestProtocol stopLoading]):
- 9:24 AM Changeset in webkit [149193] by
-
- 49 edits in trunk
Remove the remaining Skia #ifdefs
https://bugs.webkit.org/show_bug.cgi?id=114886
Reviewed by Benjamin Poulain.
Source/WebCore:
- html/HTMLCanvasElement.cpp: Remove Skia #ifdef references.
- platform/graphics/BitmapImage.cpp: Ditto.
- platform/graphics/FloatPoint.h: Ditto.
- platform/graphics/FloatRect.h: Ditto.
- platform/graphics/FontCache.h: Ditto.
- platform/graphics/Gradient.cpp: Ditto.
- platform/graphics/Gradient.h: Ditto.
- platform/graphics/GraphicsContext.cpp: Ditto.
- platform/graphics/GraphicsContext.h: Ditto.
- platform/graphics/GraphicsContext3D.h: Ditto.
- platform/graphics/ImageBuffer.cpp: Ditto.
- platform/graphics/ImageBuffer.h: Ditto.
- platform/graphics/ImageBufferData.h: Ditto.
- platform/graphics/IntPoint.h: Ditto.
- platform/graphics/IntRect.h: Ditto.
- platform/graphics/NativeImagePtr.h: Ditto.
- platform/graphics/Path.h: Ditto.
- platform/graphics/Pattern.cpp: Ditto.
- platform/graphics/Pattern.h: Ditto.
- platform/graphics/filters/FEBlend.h: Ditto.
- platform/graphics/filters/FEColorMatrix.h: Ditto.
- platform/graphics/filters/FEComponentTransfer.h: Ditto.
- platform/graphics/filters/FEComposite.h: Ditto.
- platform/graphics/filters/FEConvolveMatrix.h: Ditto.
- platform/graphics/filters/FEDisplacementMap.h: Ditto.
- platform/graphics/filters/FEGaussianBlur.h: Ditto.
- platform/graphics/filters/FELighting.h: Ditto.
- platform/graphics/filters/FEMorphology.h: Ditto.
- platform/graphics/filters/FEOffset.h: Ditto.
- platform/graphics/filters/FilterEffect.cpp: Ditto.
- platform/graphics/filters/FilterEffect.h: Ditto.
- platform/graphics/transforms/AffineTransform.h: Ditto.
- platform/graphics/transforms/TransformationMatrix.h: Ditto.
- platform/image-decoders/ImageDecoder.cpp: Ditto.
- platform/image-decoders/ImageDecoder.h: Ditto.
- platform/image-decoders/jpeg/JPEGImageDecoder.cpp: Ditto.
- platform/image-decoders/webp/WEBPImageDecoder.cpp: Ditto.
- platform/mac/LocalCurrentGraphicsContext.h: Ditto.
- platform/mac/LocalCurrentGraphicsContext.mm: Ditto.
- rendering/svg/RenderSVGResourceSolidColor.cpp: Ditto.
- svg/graphics/SVGImage.cpp: Ditto.
- svg/graphics/SVGImage.h: Ditto.
Source/WebKit/blackberry:
- Api/WebPage.cpp: Remove Skia #ifdef references.
- WebCoreSupport/AboutDataUseFeatures.in: Ditto.
Source/WTF:
- wtf/Platform.h:
Tools:
- DumpRenderTree/blackberry/PixelDumpSupportBlackBerry.cpp:
(createBitmapContextFromWebView): Remove Skia #ifdef references.
- 9:16 AM Changeset in webkit [149192] by
-
- 2 edits in trunk/Source/WebKit2
[GTK] Compilation of ProcessLauncherGtk.cpp fails due to unresolved symbols
https://bugs.webkit.org/show_bug.cgi?id=115240
Patch by Eduardo Lima Mitev <elima@igalia.com> on 2013-04-26
Reviewed by Martin Robinson.
- UIProcess/Launcher/gtk/ProcessLauncherGtk.cpp: Includes sys/socket.h if OS is Linux
- 9:15 AM Changeset in webkit [149191] by
-
- 2 edits in trunk/LayoutTests
Sort jsc-test-list
$ sort -f -o LayoutTests/fast/js/jsc-test-list LayoutTests/fast/js/jsc-test-list
- fast/js/jsc-test-list: Sort case-insensitively.
- 9:15 AM Changeset in webkit [149190] by
-
- 2 edits in trunk/LayoutTests
Upstream iOS changes to jsc-test-list
Reviewed by Michael Saboff.
- fast/js/jsc-test-list: Add tests.
- 9:12 AM Changeset in webkit [149189] by
-
- 2 edits in trunk/LayoutTests
[Qt] Unreviewed gardening. Skip two failing tests.
- platform/qt/TestExpectations:
- 9:08 AM Changeset in webkit [149188] by
-
- 2 edits in trunk/Source/WebCore
Assert in JSC::Heap::unprotect when closing facebook.com web site
https://bugs.webkit.org/show_bug.cgi?id=115058
Reviewed by Darin Adler.
Grab a JSLock before calling RootObject::invalidate().
- bindings/js/ScriptController.cpp:
(WebCore::ScriptController::~ScriptController):
- 8:17 AM Changeset in webkit [149187] by
-
- 6 edits in trunk/Source/WebCore
[Mac] in-band cues sometimes have incorrect duration
https://bugs.webkit.org/show_bug.cgi?id=115200
Reviewed by Jer Noble.
No new tests, this is not possible to test in DRT.
- html/track/InbandTextTrack.cpp:
(WebCore::InbandTextTrack::addGenericCue): Don't add completed cues to the map.
(WebCore::InbandTextTrack::removeGenericCue): Log when a cue is removed from the track.
- platform/graphics/avfoundation/InbandTextTrackPrivateAVF.cpp:
(WebCore::InbandTextTrackPrivateAVF::InbandTextTrackPrivateAVF): Initialize m_pendingCueStatus.
(WebCore::InbandTextTrackPrivateAVF::processCue): Never call update() on a cue that is delivered
while seeking.
(WebCore::InbandTextTrackPrivateAVF::beginSeeking): Flush all incomplete cues, remember that
we are seeking.
(WebCore::InbandTextTrackPrivateAVF::resetCueValues):
- platform/graphics/avfoundation/InbandTextTrackPrivateAVF.h:
- platform/graphics/avfoundation/MediaPlayerPrivateAVFoundation.cpp:
(WebCore::MediaPlayerPrivateAVFoundation::seek): Call track->beginSeeking() instead of resetCueValues().
(WebCore::MediaPlayerPrivateAVFoundation::seekCompleted): Call track->endSeeking().
- platform/graphics/avfoundation/objc/InbandTextTrackPrivateAVFObjC.mm:
(WebCore::InbandTextTrackPrivateAVFObjC::kind): Include class name in Kind enum values to
avoid compile error.
- 8:14 AM Changeset in webkit [149186] by
-
- 21 edits in trunk/Source/WebCore
Remove wxWebKit from WebCore.
<http://webkit.org/b/115255>
Reviewed by Anders Carlsson.
- DerivedSources.make:
- WebCorePrefix.h:
- bindings/js/JSInspectorFrontendHostCustom.cpp:
(WebCore::JSInspectorFrontendHost::port):
- config.h:
- html/HTMLFormElement.cpp:
- platform/ContextMenu.h:
(ContextMenu):
- platform/ContextMenuItem.h:
- platform/Cursor.h:
- platform/DragData.h:
- platform/DragImage.h:
- platform/FileSystem.h:
- platform/PlatformExportMacros.h:
- platform/PlatformKeyboardEvent.h:
(PlatformKeyboardEvent):
- platform/PlatformMenuDescription.h:
- platform/PlatformMouseEvent.h:
(PlatformMouseEvent):
- platform/PlatformWheelEvent.h:
(PlatformWheelEvent):
- platform/ScrollView.cpp:
- platform/ScrollView.h:
(ScrollView):
- platform/Widget.h:
- platform/win/SharedTimerWin.cpp:
- 8:13 AM Changeset in webkit [149185] by
-
- 6 edits3 adds in trunk
use-after-free removing a frame from its parent in a beforeload event of an OBJECT element
https://bugs.webkit.org/show_bug.cgi?id=113964
Source/WebCore:
Object elements have the tendecny to modify or even fully remove
the containing Document inside beforeload callback. While Document is removed,
RenderArena gets destroyed. Retained RenderWidgets fails to function with NULL arena.
Protect RendereArena from getting wiped out, when Document is removed
during FrameView::updateWidget().
Patch by Zalan Bujtas <Alan Bujtas> on 2013-04-26
Reviewed by Antti Koivisto.
Test: fast/frames/crash-remove-iframe-during-object-beforeload.html
- dom/Document.cpp:
(WebCore::Document::attach):
- dom/Document.h:
(Document):
- page/FrameView.cpp:
(WebCore::FrameView::updateWidgets):
- rendering/RenderArena.h:
(RenderArena):
(WebCore::RenderArena::create):
LayoutTests:
Patch by Zalan Bujtas <Alan Bujtas> on 2013-04-26
Reviewed by Antti Koivisto.
- fast/frames/crash-remove-iframe-during-object-beforeload-expected.txt: Added.
- fast/frames/crash-remove-iframe-during-object-beforeload.html: Added.
- fast/frames/resources/remove-this-during-object-beforeload.html: Added.
- 7:56 AM Changeset in webkit [149184] by
-
- 2 edits in trunk/Source/WTF
Add move semantics to RefPtr
https://bugs.webkit.org/show_bug.cgi?id=115033
Reviewed by Anders Carlsson.
Add move constructors and move assignment operators to RefPtr when
COMPILER_SUPPORTS(CXX_RVALUE_REFERENCES).
This obviates unnecessary reffing/ureffing when RefPtr is created
or assigned from rvalue references.
- wtf/RefPtr.h:
(RefPtr):
(WTF::RefPtr::RefPtr):
(WTF::RefPtr::operator=):
- 7:32 AM Changeset in webkit [149183] by
-
- 2 edits in trunk/Source/JavaScriptCore
REGRESSION(r149165): It made many tests crash on 32 bit
https://bugs.webkit.org/show_bug.cgi?id=115227
Reviewed by Csaba Osztrogonác.
m_reservation is uninitialized when ENABLE(SUPER_REGION) is false.
- heap/SuperRegion.cpp:
(JSC::SuperRegion::~SuperRegion):
- 6:59 AM Changeset in webkit [149182] by
-
- 2 edits in trunk/Source/WebKit/blackberry
[BlackBerry] Should check if it's cached resource before download
https://bugs.webkit.org/show_bug.cgi?id=115101
Reviewed by Rob Buis.
Since main resource maybe cached, if user want to save the resource, we first check
if it's cached. If yes, don't need to initiate a fresh load again, but get the
cached resource data out to save.
RIM bug# 324003, internally reviewed by Charles Wei.
- WebCoreSupport/FrameLoaderClientBlackBerry.cpp:
(WebCore::FrameLoaderClientBlackBerry::convertMainResourceLoadToDownload):
- 6:20 AM Changeset in webkit [149181] by
-
- 3 edits20 adds in trunk/LayoutTests
[GTK] Enable Web Audio layout tests on WK2, provide platform-specific expected output
https://bugs.webkit.org/show_bug.cgi?id=114513
Reviewed by Philippe Normand.
Unskip the Web Audio layout tests for GTK WK2 and generate platform-specific baselines for the tests.
The audio baselines are not audibly different from the generic baselines, but it can be expected that
different backends won't produce the same binary output.
The remaining failures are marked as such in the TestExpectations, each failure having the appropriate
bug handle covering the issue. The tests are still skipped for GTK WK1 as DumpRenderTree is still missing
infrastructure support for running these tests.
- platform/gtk-wk1/TestExpectations:
- platform/gtk/TestExpectations:
- platform/gtk/webaudio/audiobuffersource-loop-points-expected.wav: Added.
- platform/gtk/webaudio/audiobuffersource-playbackrate-expected.wav: Added.
- platform/gtk/webaudio/codec-tests/aac/vbr-128kbps-44khz-expected.wav: Added.
- platform/gtk/webaudio/codec-tests/mp3/128kbps-44khz-expected.wav: Added.
- platform/gtk/webaudio/codec-tests/vorbis/vbr-128kbps-44khz-expected.wav: Added.
- platform/gtk/webaudio/codec-tests/vorbis/vbr-70kbps-44khz-expected.wav: Added.
- platform/gtk/webaudio/codec-tests/vorbis/vbr-96kbps-44khz-expected.wav: Added.
- platform/gtk/webaudio/codec-tests/wav/24bit-22khz-resample-expected.wav: Added.
- platform/gtk/webaudio/codec-tests/wav/24bit-44khz-expected.wav: Added.
- platform/gtk/webaudio/oscillator-custom-expected.wav: Added.
- platform/gtk/webaudio/oscillator-sawtooth-expected.wav: Added.
- platform/gtk/webaudio/oscillator-sine-expected.wav: Added.
- platform/gtk/webaudio/oscillator-square-expected.wav: Added.
- platform/gtk/webaudio/oscillator-triangle-expected.wav: Added.
- 6:16 AM Changeset in webkit [149180] by
-
- 5 edits in trunk
[BlackBerry] Clean up load interface in WebPage
https://bugs.webkit.org/show_bug.cgi?id=113267
Reviewed by Rob Buis.
Source/WebKit/blackberry:
Remove unused loadExtended(), combine load() and download() api in WebPage.
RIM Bug# 315535, internally reviewed by Joe Mason.
- Api/WebPage.cpp:
(BlackBerry::WebKit::WebPage::load):
- Api/WebPage.h:
Tools:
Update DumpRenderTree with change of load() interface in WebPage.
RIM Bug# 315535, internally reviewed by Joe Mason.
- DumpRenderTree/blackberry/DumpRenderTree.cpp:
(BlackBerry::WebKit::DumpRenderTree::runTest):
- 6:09 AM Changeset in webkit [149179] by
-
- 2 edits in trunk/LayoutTests
[GTK] Unreviewed gardnering. Unskip passing test.
- platform/gtk/TestExpectations: Unskip http/tests/inspector/compiler-source-mapping-debug.html
- 6:07 AM Changeset in webkit [149178] by
-
- 8 edits3 deletes in trunk/Source/WTF
Remove wxWebKit from WTF.
<http://webkit.org/b/115249>
Reviewed by Antti Koivisto.
- wscript: Removed.
- wtf/ArrayBufferView.h:
(ArrayBufferView):
- wtf/CurrentTime.cpp:
- wtf/DisallowCType.h:
- wtf/ExportMacros.h:
- wtf/FeatureDefines.h:
- wtf/Platform.h:
- wtf/text/WTFString.h:
(String):
- wtf/wx/MainThreadWx.cpp: Removed.
- wtf/wx/StringWx.cpp: Removed.
- 6:03 AM Changeset in webkit [149177] by
-
- 2 edits in trunk/Source/WebCore
Optimize function and interface object length computation in bindings generator
https://bugs.webkit.org/show_bug.cgi?id=115247
Reviewed by Kentaro Hara.
Introduce new GetFunctionLength() function that efficiently compute the length
of a function (i.e. its number of mandatory parameters).
We now call GetFunctionLength() instead of GenerateFunctionParametersCheck()
whenever we care only interested in the function length and not the actual
expression for checking the parameters. This is much more efficient as
GenerateFunctionParametersCheck() does a lot more processing than we need in
this case.
No new tests, no behavior change.
- bindings/scripts/CodeGeneratorJS.pm:
(GetFunctionLength):
(GenerateImplementation):
(GenerateConstructorHelperMethods):
- 6:02 AM Changeset in webkit [149176] by
-
- 1 edit1 delete in trunk/Tools
Remove the HeapGraphSerializer unit test
https://bugs.webkit.org/show_bug.cgi?id=115231
Reviewed by Andreas Kling.
Remove the HeapGraphSerializer unit test. The code it was testing was removed in r148921,
specifically the Source/WebCore/inspector/HeapGraphSerializer.(cpp|h) files.
- TestWebKitAPI/Tests/WebCore/HeapGraphSerializerTest.cpp: Removed.
- 5:09 AM FeatureFlags edited by
- (diff)
- 4:48 AM Changeset in webkit [149175] by
-
- 5 edits in trunk/Source/WebCore
Web Audio: Remove reduplicative addInput() in AnalyserNode.
<http://webkit.org/b/115244>
From Blink r149155 by <james.wei@intel.com>:
The parent class BasicInspectorNode already called addInput() and addOutput().
So it is reduplicative to call them in AnalyserNode.
- Modules/webaudio/AnalyserNode.cpp:
(WebCore::AnalyserNode::AnalyserNode):
- Modules/webaudio/AudioBasicInspectorNode.cpp:
(WebCore::AudioBasicInspectorNode::AudioBasicInspectorNode):
- Modules/webaudio/AudioBasicInspectorNode.h:
(AudioBasicInspectorNode):
- Modules/webaudio/MediaStreamAudioDestinationNode.cpp:
(WebCore::MediaStreamAudioDestinationNode::MediaStreamAudioDestinationNode):
- 3:30 AM Changeset in webkit [149174] by
-
- 2 edits in trunk/Source/WebKit/efl
[EFL] Add unit test cases for ewk_settings APIs
https://bugs.webkit.org/show_bug.cgi?id=114897
Patch by Jose Lejin PJ <jose.lejin@gmail.com> on 2013-04-26
Reviewed by Gyuyoung Kim.
Added unit tests for ewk_settings APIs.
- tests/test_ewk_setting.cpp:
(TEST_F):
- 2:42 AM Changeset in webkit [149173] by
-
- 20 edits4 adds in trunk
Mouseenter and mouseleave events not supported
https://bugs.webkit.org/show_bug.cgi?id=18930
Reviewed by David Hyatt.
Source/WebCore:
Implements mouseenter and mouseleave events from W3C DOM Level 3 Events.
These event are already supported by all other major browsers.
To avoid performance regressions the new events are only dispatched when
there are event listeners for them.
Tests: fast/events/mouseenter-mouseleave-capture.html
fast/events/mouseenter-mouseleave.html
- bindings/scripts/CodeGenerator.pm:
- dom/Document.cpp:
(WebCore::Document::prepareMouseEvent):
(WebCore::Document::updateHoverActiveState):
- dom/Document.h:
(Document):
- dom/Document.idl:
- dom/Element.h:
(Element):
- dom/Element.idl:
- dom/EventListenerMap.cpp:
(WebCore::EventListenerMap::containsCapturing):
- dom/EventListenerMap.h:
(EventListenerMap):
- dom/EventNames.h:
- dom/EventTarget.h:
(EventTarget):
(WebCore::EventTarget::hasCapturingEventListeners):
- dom/MouseEvent.cpp:
(WebCore::MouseEvent::create):
(WebCore::MouseEvent::toElement):
(WebCore::MouseEvent::fromElement):
- html/HTMLAttributeNames.in:
- html/HTMLElement.cpp:
(WebCore::HTMLElement::eventNameForAttributeName):
- page/DOMWindow.h:
(DOMWindow):
- page/DOMWindow.idl:
- svg/SVGElement.cpp:
(WebCore::SVGElement::parseAttribute):
- svg/SVGElementInstance.h:
(SVGElementInstance):
- svg/SVGElementInstance.idl:
LayoutTests:
To new tests that mouseenter and mouseleave works in both bubbling and capture phase.
- fast/events/mouseenter-mouseleave-capture-expected.txt: Added.
- fast/events/mouseenter-mouseleave-capture.html: Added.
- fast/events/mouseenter-mouseleave-expected.txt: Added.
- fast/events/mouseenter-mouseleave.html: Added.
- 2:36 AM Changeset in webkit [149172] by
-
- 2 edits in trunk/Source/JavaScriptCore
Fix SH4 build broken since r149159.
https://bugs.webkit.org/show_bug.cgi?id=115229
Add BranchTruncateType enum in SH4 port and handle it in branchTruncateDoubleToInt32.
Patch by Julien Brianceau <jbrianceau@nds.com> on 2013-04-26
Reviewed by Allan Sandfeld Jensen.
- assembler/MacroAssemblerSH4.h:
(JSC::MacroAssemblerSH4::branchTruncateDoubleToInt32):
- 2:16 AM Changeset in webkit [149171] by
-
- 5 edits in trunk/LayoutTests
[Qt] REGRESSION(r149001): It made two fast/dom/DeviceMotion tests fail.
https://bugs.webkit.org/show_bug.cgi?id=115235
Reviewed by Kentaro Hara.
Update window-property.html test cases for DeviceMotion and DeviceOrientation.
The global constructors are no longer enumerable after r149001.
- fast/dom/DeviceMotion/script-tests/window-property.js:
- fast/dom/DeviceMotion/window-property-expected.txt:
- fast/dom/DeviceOrientation/script-tests/window-property.js:
- fast/dom/DeviceOrientation/window-property-expected.txt:
- 12:38 AM WebKitIDL edited by
- Drop Supplemental in favor of partial interfaces due to r149170 (diff)
- 12:35 AM WebKitIDL edited by
- [Callback] only applies to parameters, not interfaces (diff)
- 12:29 AM Changeset in webkit [149170] by
-
- 33 edits in trunk/Source/WebCore
Add support for Web IDL partial interfaces to the bindings generator
https://bugs.webkit.org/show_bug.cgi?id=115228
Reviewed by Kentaro Hara.
Add support for Web IDL partial interfaces to the bindings generator:
http://dev.w3.org/2006/webapi/WebIDL/#dfn-partial-interface
Also drop support for the the WebKit-specific [Supplemental] extended
attribute and use partial interfaces instead in existing IDL files.
No new tests, no behavior change.
- Modules/battery/NavigatorBattery.idl:
- Modules/filesystem/DOMWindowFileSystem.idl:
- Modules/filesystem/DataTransferItemFileSystem.idl:
- Modules/filesystem/HTMLInputElementFileSystem.idl:
- Modules/filesystem/WorkerContextFileSystem.idl:
- Modules/gamepad/NavigatorGamepad.idl:
- Modules/geolocation/NavigatorGeolocation.idl:
- Modules/indexeddb/DOMWindowIndexedDatabase.idl:
- Modules/indexeddb/WorkerContextIndexedDatabase.idl:
- Modules/mediastream/DOMWindowMediaStream.idl:
- Modules/mediastream/NavigatorMediaStream.idl:
- Modules/navigatorcontentutils/NavigatorContentUtils.idl:
- Modules/networkinfo/NavigatorNetworkInfoConnection.idl:
- Modules/notifications/DOMWindowNotifications.idl:
- Modules/notifications/WorkerContextNotifications.idl:
- Modules/quota/DOMWindowQuota.idl:
- Modules/quota/NavigatorStorageQuota.idl:
- Modules/quota/WorkerNavigatorStorageQuota.idl:
- Modules/speech/DOMWindowSpeech.idl:
- Modules/speech/DOMWindowSpeechSynthesis.idl:
- Modules/vibration/NavigatorVibration.idl:
- Modules/webaudio/DOMWindowWebAudio.idl:
- Modules/webdatabase/DOMWindowWebDatabase.idl:
- Modules/webdatabase/WorkerContextWebDatabase.idl:
- Modules/websockets/DOMWindowWebSocket.idl:
- Modules/websockets/WorkerContextWebSocket.idl:
- bindings/scripts/IDLAttributes.txt:
- bindings/scripts/IDLParser.pm:
(parsePartialDefinition):
- bindings/scripts/generate-bindings.pl:
- bindings/scripts/preprocess-idls.pl:
(getPartialInterfaceNameFromIDLFile):
- bindings/scripts/test/TestSupplemental.idl:
- page/DOMWindowPagePopup.idl:
Apr 25, 2013:
- 11:50 PM Changeset in webkit [149169] by
-
- 3 edits in trunk/Source/WebKit2
[GTK] Plugin process broken due to a missing symbol
https://bugs.webkit.org/show_bug.cgi?id=114901
Reviewed by Gustavo Noronha Silva.
Work around the missing symbols in the WebKitPluginProcess by specifying the required libraries
multiple times in the program's LDADD list. Libtool fights fiercely to avoid duplicate static library
references among the linker arguments (even if using the --preserve-dup-deps flag) so the required libraries
are specified in various ways.
The unresolved symbols in object files are no longer ignored as this can lead to further trouble (for instance
this patch also adds a few missing source files to the list of WebKitPluginProcess sources). The Freetype and
gamepad dependencies' libraries are also added to the binary's LDADD list.
- GNUmakefile.am:
- GNUmakefile.list.am: Add a few missing files to the build.
- 11:23 PM Changeset in webkit [149168] by
-
- 4 edits2 adds in trunk
[CSS Regions] Hit testing is broken for absolutely positioned regions that have overflow: hidden
https://bugs.webkit.org/show_bug.cgi?id=113874
Reviewed by David Hyatt.
Source/WebCore:
Test: fast/regions/hit-test-abspos-overflow-region.html
When a region is an out-of-flow positioned object with an overflow clip, we need
to make sure that hit testing works also for cases other than foreground (content)
hit testing. This patch moves the previous hit testing code, that delegated foreground
hit testing to the region's flow thread hit testing, into the hitTestContent method,
now that RenderRegion is RenderBlock based.
- rendering/RenderRegion.cpp:
(WebCore::RenderRegion::hitTestContents):
- rendering/RenderRegion.h:
(RenderRegion):
LayoutTests:
Add test for hit-testing of border of an out-of-flow region with overflow: hidden.
- fast/regions/hit-test-abspos-overflow-region-expected.txt: Added.
- fast/regions/hit-test-abspos-overflow-region.html: Added.
- 9:51 PM Changeset in webkit [149167] by
-
- 4 edits2 adds in trunk
Copy and paste can strip !important CSS rules due to a bug in mergeStyleFromRules
https://bugs.webkit.org/show_bug.cgi?id=115217
Reviewed by Darin Adler.
Source/WebCore:
The bug was caused by mergeStyleFromRules overriding "important" style rules with "unimportant" inline styles.
Fixed the bug by using addParsedProperty, which respects !important, in MutableStylePropertySet's
mergeAndOverrideOnConflict, which was only used in editing code. Now that we've fixed this function, we can use
it in ViewportStyleResolver::addViewportRule as well.
Test: editing/pasteboard/copy-paste-with-important-rules.html
- css/StylePropertySet.cpp:
(WebCore::MutableStylePropertySet::mergeAndOverrideOnConflict): Fixed to respect !important.
- css/ViewportStyleResolver.cpp:
(WebCore::ViewportStyleResolver::addViewportRule): Use mergeAndOverrideOnConflict now that the code is identical.
LayoutTests:
Added a regression test.
- editing/pasteboard/copy-paste-with-important-rules-expected.txt: Added.
- editing/pasteboard/copy-paste-with-important-rules.html: Added.
- 9:50 PM Changeset in webkit [149166] by
-
- 3 edits in trunk/Tools
Remove support for Chromium from build/update scripts
https://bugs.webkit.org/show_bug.cgi?id=115218
Patch by Seokju Kwon <Seokju Kwon> on 2013-04-25
Reviewed by Darin Adler.
- Scripts/build-webkit:
(writeCongrats):
- Scripts/update-webkit:
- 7:47 PM Changeset in webkit [149165] by
-
- 4 edits in trunk/Source/JavaScriptCore
SuperRegion doesn't call deallocate() on its PageReservation
https://bugs.webkit.org/show_bug.cgi?id=115208
Reviewed by Geoffrey Garen.
It should. This doesn't cause us to leak physical memory, but it does cause us to leak virtual
address space (and probably mach ports), which is also bad :-( FixedVMPoolExecutableAllocator
also has this bug, but it doesn't matter much because there's only one instance of that class
throughout the entire lifetime of the process, whereas each VM has its own SuperRegion.
- heap/SuperRegion.cpp:
(JSC::SuperRegion::~SuperRegion):
- heap/SuperRegion.h:
(SuperRegion):
- jit/ExecutableAllocatorFixedVMPool.cpp:
(FixedVMPoolExecutableAllocator):
(JSC::FixedVMPoolExecutableAllocator::~FixedVMPoolExecutableAllocator):
- 7:30 PM Changeset in webkit [149164] by
-
- 3 edits in trunk/Tools
Remove support for Chromium from api-test
https://bugs.webkit.org/show_bug.cgi?id=115211
Patch by Seokju Kwon <Seokju Kwon> on 2013-04-25
Reviewed by Dirk Pranke.
- Scripts/build-api-tests:
- Scripts/run-api-tests:
(isSupportedPlatform):
- 6:38 PM Changeset in webkit [149163] by
-
- 2 edits in trunk/Tools
Get rid of Chromium in build-dumprendertree
https://bugs.webkit.org/show_bug.cgi?id=115212
Patch by Seokju Kwon <Seokju Kwon> on 2013-04-25
Reviewed by Dirk Pranke.
- Scripts/build-dumprendertree:
- 6:18 PM Changeset in webkit [149162] by
-
- 7 edits in trunk/Source/JavaScriptCore
DFG doesn't support to_jsnumber
https://bugs.webkit.org/show_bug.cgi?id=115129
Reviewed by Geoffrey Garen.
Based on Oliver's patch. Implements to_jsnumber as Identity(Number:@thingy), and then does
an optimization in Fixup to turn Identity(Number:) into Identity(Int32:) if the predictions
tell us to. Identity is later turned into Phantom.
Also fixed BackPropMask, which appeared to have NodeDoesNotExit included in it. That's
wrong; NodeDoesNotExit is not a backward propagation property.
Also fixed Identity to be marked as CanExit (i.e. not NodeDoesNotExit).
This more than doubles the FPS on ammo.
- dfg/DFGByteCodeParser.cpp:
(JSC::DFG::ByteCodeParser::parseBlock):
- dfg/DFGCapabilities.h:
(JSC::DFG::canCompileOpcode):
- dfg/DFGFixupPhase.cpp:
(JSC::DFG::FixupPhase::fixupNode):
(FixupPhase):
(JSC::DFG::FixupPhase::observeUseKindOnNode):
(JSC::DFG::FixupPhase::observeUseKindOnEdge):
- dfg/DFGNodeFlags.h:
(DFG):
- dfg/DFGNodeType.h:
(DFG):
- dfg/DFGPredictionPropagationPhase.cpp:
(JSC::DFG::PredictionPropagationPhase::propagate):
- 5:50 PM Changeset in webkit [149161] by
-
- 4 edits in trunk/Source
Versioning.
- 5:49 PM Changeset in webkit [149160] by
-
- 1 copy in tags/Safari-537.39
New tag.
- 5:41 PM Changeset in webkit [149159] by
-
- 21 edits12 adds in trunk
Add support for Math.imul
https://bugs.webkit.org/show_bug.cgi?id=115143
Reviewed by Filip Pizlo.
Source/JavaScriptCore:
Add support for Math.imul, a thunk generator for Math.imul,
and an intrinsic.
Fairly self explanatory set of changes, DFG intrinsics simply
leverages the existing ValueToInt32 nodes.
- create_hash_table:
- dfg/DFGAbstractState.cpp:
(JSC::DFG::AbstractState::executeEffects):
- dfg/DFGBackwardsPropagationPhase.cpp:
(JSC::DFG::BackwardsPropagationPhase::propagate):
- dfg/DFGByteCodeParser.cpp:
(JSC::DFG::ByteCodeParser::handleIntrinsic):
- dfg/DFGCSEPhase.cpp:
(JSC::DFG::CSEPhase::performNodeCSE):
- dfg/DFGFixupPhase.cpp:
(JSC::DFG::FixupPhase::fixupNode):
- dfg/DFGNodeType.h:
(DFG):
- dfg/DFGPredictionPropagationPhase.cpp:
(JSC::DFG::PredictionPropagationPhase::propagate):
- dfg/DFGSpeculativeJIT.cpp:
(JSC::DFG::SpeculativeJIT::compileArithIMul):
- dfg/DFGSpeculativeJIT.h:
(SpeculativeJIT):
- dfg/DFGSpeculativeJIT32_64.cpp:
(JSC::DFG::SpeculativeJIT::compile):
- dfg/DFGSpeculativeJIT64.cpp:
(JSC::DFG::SpeculativeJIT::compile):
- jit/ThunkGenerators.cpp:
(JSC::imulThunkGenerator):
(JSC):
- jit/ThunkGenerators.h:
(JSC):
- runtime/Intrinsic.h:
- runtime/MathObject.cpp:
(JSC):
(JSC::mathProtoFuncIMul):
- runtime/VM.cpp:
(JSC::thunkGeneratorForIntrinsic):
LayoutTests:
Add a bunch of tests for Math.imul
- fast/js/Object-getOwnPropertyNames-expected.txt:
- fast/js/imul-expected.txt: Added.
- fast/js/imul.html: Added.
- fast/js/regress/imul-double-only-expected.txt: Added.
- fast/js/regress/imul-double-only.html: Added.
- fast/js/regress/imul-int-only-expected.txt: Added.
- fast/js/regress/imul-int-only.html: Added.
- fast/js/regress/imul-mixed-expected.txt: Added.
- fast/js/regress/imul-mixed.html: Added.
- fast/js/regress/script-tests/imul-double-only.js: Added.
(f):
- fast/js/regress/script-tests/imul-int-only.js: Added.
(f):
- fast/js/regress/script-tests/imul-mixed.js: Added.
(f):
- fast/js/script-tests/Object-getOwnPropertyNames.js:
- fast/js/script-tests/imul.js: Added.
(testIMul):
- 5:40 PM Changeset in webkit [149158] by
-
- 15 edits in trunk/Source
Unreviewed, roll out http://trac.webkit.org/changeset/148999
It broke http://kripken.github.io/ammo.js/examples/new/ammo.html
Source/JavaScriptCore:
- JavaScriptCore.order:
- bytecompiler/BytecodeGenerator.cpp:
(JSC::BytecodeGenerator::emitNewArray):
(JSC::BytecodeGenerator::emitThrowReferenceError):
(JSC::BytecodeGenerator::emitReadOnlyExceptionIfNeeded):
- bytecompiler/BytecodeGenerator.h:
(JSC::BytecodeGenerator::shouldEmitProfileHooks):
(BytecodeGenerator):
- bytecompiler/NodesCodegen.cpp:
(JSC):
(JSC::NullNode::emitBytecode):
(JSC::BooleanNode::emitBytecode):
(JSC::NumberNode::emitBytecode):
(JSC::StringNode::emitBytecode):
(JSC::IfNode::emitBytecode):
(JSC::IfElseNode::emitBytecode):
- parser/ASTBuilder.h:
(JSC::ASTBuilder::createIfStatement):
(ASTBuilder):
- parser/NodeConstructors.h:
(JSC):
(JSC::NullNode::NullNode):
(JSC::BooleanNode::BooleanNode):
(JSC::NumberNode::NumberNode):
(JSC::StringNode::StringNode):
(JSC::IfNode::IfNode):
(JSC::IfElseNode::IfElseNode):
- parser/Nodes.h:
(JSC::ExpressionNode::isPure):
(JSC::ExpressionNode::isSubtract):
(StatementNode):
(NullNode):
(JSC::NullNode::isNull):
(BooleanNode):
(JSC::BooleanNode::isPure):
(NumberNode):
(JSC::NumberNode::value):
(JSC::NumberNode::isPure):
(StringNode):
(JSC::StringNode::isPure):
(JSC::StringNode::isString):
(BinaryOpNode):
(IfNode):
(JSC):
(IfElseNode):
(ContinueNode):
(BreakNode):
- parser/Parser.cpp:
(JSC::::parseIfStatement):
- parser/ResultType.h:
(ResultType):
- runtime/JSCJSValueInlines.h:
(JSC::JSValue::pureToBoolean):
- runtime/JSCell.h:
(JSCell):
- runtime/JSCellInlines.h:
(JSC):
Source/WTF:
- wtf/TriState.h:
- 5:33 PM Changeset in webkit [149157] by
-
- 4 edits in trunk/Source/WebCore
StylePropertySet::getPropertyShorthand() should return a String.
<http://webkit.org/b/115213>
Reviewed by Anders Carlsson.
Return a String directly from getPropertyShorthand() instead of forcing clients
to call getPropertyNameString().
- css/PropertySetCSSStyleDeclaration.cpp:
(WebCore::PropertySetCSSStyleDeclaration::getPropertyShorthand):
- css/StylePropertySet.cpp:
(WebCore::StylePropertySet::getPropertyShorthand):
- css/StylePropertySet.h:
(StylePropertySet):
- 5:27 PM Changeset in webkit [149156] by
-
- 13 edits in trunk
WEB SPEECH: language support does not work as expected
https://bugs.webkit.org/show_bug.cgi?id=115119
Reviewed by Alexey Proskuryakov.
Source/WebCore:
Make the Mac platform synthesizer have access to all the voices installed on the system.
- WebCore.exp.in:
- platform/mac/PlatformSpeechSynthesizerMac.mm:
(WebCore::PlatformSpeechSynthesizer::initializeVoiceList):
- platform/mac/WebCoreSystemInterface.h:
- platform/mac/WebCoreSystemInterface.mm:
Source/WebKit/mac:
- WebCoreSupport/WebSystemInterface.mm:
(InitWebCoreSystemInterface):
Source/WebKit2:
- WebProcess/WebCoreSupport/mac/WebSystemInterface.mm:
(InitWebCoreSystemInterface):
WebKitLibraries:
- WebKitSystemInterface.h:
- 4:33 PM Changeset in webkit [149155] by
-
- 11 edits2 adds in trunk
Source/WebCore: <meter> element not exposed to accessibility
https://bugs.webkit.org/show_bug.cgi?id=109023
rdar://problem/13658964
Reviewed by Tim Horton.
Makes the <meter> element appear in the AX tree by reusing the
AccessibilityProgressIndicator element to handle either progress or meter
elements.
Test: accessibility/meter-element.html
- accessibility/AXObjectCache.cpp:
(WebCore::createFromRenderer):
(WebCore::AXObjectCache::getOrCreate):
- accessibility/AccessibilityNodeObject.cpp:
(WebCore::AccessibilityNodeObject::canHaveChildren):
(WebCore::AccessibilityNodeObject::visibleText):
- accessibility/AccessibilityProgressIndicator.cpp:
(WebCore::AccessibilityProgressIndicator::AccessibilityProgressIndicator):
(WebCore::AccessibilityProgressIndicator::create):
(WebCore::AccessibilityProgressIndicator::valueForRange):
(WebCore::AccessibilityProgressIndicator::maxValueForRange):
(WebCore::AccessibilityProgressIndicator::minValueForRange):
(WebCore::AccessibilityProgressIndicator::progressElement):
(WebCore::AccessibilityProgressIndicator::meterElement):
- accessibility/AccessibilityProgressIndicator.h:
(AccessibilityProgressIndicator):
- accessibility/AccessibilityRenderObject.cpp:
(WebCore::AccessibilityRenderObject::firstChild):
(WebCore::AccessibilityRenderObject::lastChild):
These methods are modified to handle when there is no rendered children, but there are Node
children. The node children are desired when calculating text within a node, for example.
LayoutTests: <meter> element not exposed to accessibility
https://bugs.webkit.org/show_bug.cgi?id=109023
Reviewed by Tim Horton.
- accessibility/meter-element-expected.txt: Added.
- accessibility/meter-element.html: Added.
- platform/mac/accessibility/role-subrole-roledescription-expected.txt:
- platform/mac/accessibility/role-subrole-roledescription.html:
- 4:10 PM Changeset in webkit [149154] by
-
- 3 edits in trunk/Source/JavaScriptCore
PreciseJumpTargets should treat loop_hint as a jump target
https://bugs.webkit.org/show_bug.cgi?id=115209
Reviewed by Mark Hahnenberg.
I didn't add a test but I turned this into a release assertion. Running Octane is enough
to trigger it.
- bytecode/PreciseJumpTargets.cpp:
(JSC::computePreciseJumpTargets):
- dfg/DFGByteCodeParser.cpp:
(JSC::DFG::ByteCodeParser::parseBlock):
- 4:06 PM Changeset in webkit [149153] by
-
- 5 edits in trunk/Source/WebCore
[CSS Shaders] Remove the meshType from the CustomFilterOperation
https://bugs.webkit.org/show_bug.cgi?id=102529
Patch by Qiankun Miao <qiankun.miao@intel.com> on 2013-04-25
Reviewed by Dean Jackson.
No new tests, no new functionality.
MeshType can be accessed from program, so remove the m_meshType
related code in CustomFilterOperation.
- css/StyleResolver.cpp:
(WebCore::StyleResolver::createCustomFilterOperationWithInlineSyntax):
- platform/graphics/filters/CustomFilterOperation.cpp:
(WebCore::CustomFilterOperation::CustomFilterOperation):
(WebCore::CustomFilterOperation::blend):
- platform/graphics/filters/CustomFilterOperation.h:
(WebCore::CustomFilterOperation::create):
(WebCore::CustomFilterOperation::meshType):
(CustomFilterOperation):
(WebCore::CustomFilterOperation::operator==):
- platform/graphics/texmap/coordinated/CoordinatedCustomFilterOperation.h:
(WebCore::CoordinatedCustomFilterOperation::CoordinatedCustomFilterOperation):
- 3:52 PM Changeset in webkit [149152] by
-
- 10 edits9 adds in trunk
Source/JavaScriptCore: Fix problems with processing negative zero on DFG.
https://bugs.webkit.org/show_bug.cgi?id=113862
Patch by Roman Zhuykov <zhroma@ispras.ru> on 2013-04-25
Reviewed by Filip Pizlo.
Fix NodeNeedsNegZero flag propagation in BackwardPropagationPhase.
Function arithNodeFlags should not mask NodeNeedsNegZero flag for ArithNegate and DoubleAsInt32
nodes and this flag should be always used to decide where we need to generate nezative-zero checks.
Remove unnecessary negative-zero checks from integer ArithDiv on ARM.
Also remove such checks from integer ArithMod on ARM and X86, and make them always to
check not only "modulo_result == 0" but also "dividend < 0".
Generate faster code for case when ArithMod operation divisor is constant power of 2 on ARMv7
in the same way as on ARMv7s, and add negative-zero checks into this code when needed.
Change speculationCheck ExitKind from Overflow to NegativeZero where applicable.
This shows 30% speedup of math-spectral-norm, and 5% speedup
on SunSpider overall on ARMv7 Linux.
- assembler/MacroAssemblerARM.h:
(JSC::MacroAssemblerARM::branchConvertDoubleToInt32):
- assembler/MacroAssemblerARMv7.h:
(JSC::MacroAssemblerARMv7::branchConvertDoubleToInt32):
- assembler/MacroAssemblerMIPS.h:
(JSC::MacroAssemblerMIPS::branchConvertDoubleToInt32):
- assembler/MacroAssemblerSH4.h:
(JSC::MacroAssemblerSH4::branchConvertDoubleToInt32):
- assembler/MacroAssemblerX86Common.h:
(JSC::MacroAssemblerX86Common::branchConvertDoubleToInt32):
- dfg/DFGBackwardsPropagationPhase.cpp:
(JSC::DFG::BackwardsPropagationPhase::isNotNegZero):
(JSC::DFG::BackwardsPropagationPhase::isNotPosZero):
(JSC::DFG::BackwardsPropagationPhase::propagate):
- dfg/DFGNode.h:
(JSC::DFG::Node::arithNodeFlags):
- dfg/DFGSpeculativeJIT.cpp:
(JSC::DFG::SpeculativeJIT::compileDoubleAsInt32):
(JSC::DFG::SpeculativeJIT::compileSoftModulo):
(JSC::DFG::SpeculativeJIT::compileArithNegate):
LayoutTests: Arithmetic operations with negative zero should be optimized correclty.
https://bugs.webkit.org/show_bug.cgi?id=113862
Patch by Roman Zhuykov <zhroma@ispras.ru> on 2013-04-25
Reviewed by Filip Pizlo.
- fast/js/regress/negative-zero-divide-expected.txt: Added.
- fast/js/regress/negative-zero-divide.html: Added.
- fast/js/regress/negative-zero-modulo-expected.txt: Added.
- fast/js/regress/negative-zero-modulo.html: Added.
- fast/js/regress/negative-zero-negate-expected.txt: Added.
- fast/js/regress/negative-zero-negate.html: Added.
- fast/js/regress/script-tests/negative-zero-divide.js: Added.
(foo):
- fast/js/regress/script-tests/negative-zero-modulo.js: Added.
(foo):
- fast/js/regress/script-tests/negative-zero-negate.js: Added.
(foo):
- 3:51 PM Changeset in webkit [149151] by
-
- 7 edits in trunk
Ignore invalid regular expressions for input[pattern].
https://bugs.webkit.org/show_bug.cgi?id=115204
Reviewed by Darin Adler.
Source/WebCore:
According to the specification, we should not proceed regular expression
matching if a pattern attribute value is an invalid regular
expression. We had a bug that invalid expressions such as
pattern=")foo(" made RegularExpression objects successfully.
If an input element has a pattern attribute specified, and the
attribute's value, when compiled as a JavaScript regular expression with
the global, ignoreCase, and multiline flags disabled (see ECMA262
Edition 5, sections 15.10.7.2 through 15.10.7.4), compiles successfully,
then the resulting regular expression is the element's compiled pattern
regular expression. If the element has no such attribute, or if the
value doesn't compile successfully, then the element has no compiled
pattern regular expression.
This imports a part of Blink r148951.
Tests: Update fast/forms/ValidityState-patternMismatch.html
- html/BaseTextInputType.cpp:
(WebCore::BaseTextInputType::patternMismatch):
Check correctness of pattern attribute value before wrapping with parentheses.
- platform/text/RegularExpression.cpp:
(WebCore::RegularExpression::isValid): Added.
- platform/text/RegularExpression.h:
(RegularExpression): Declare isValid.
LayoutTests:
- fast/forms/ValidityState-patternMismatch-expected.txt:
- fast/forms/ValidityState-patternMismatch.html:
- 3:50 PM Changeset in webkit [149150] by
-
- 2 edits in trunk/Source/WTF
Fix 32bit build
- 3:37 PM Changeset in webkit [149149] by
-
- 2 edits in trunk/Source/WebKit/mac
Build fix for smart compliers.
Reviewed by Tim Horton.
[NSDictionary dictionaryWithObjectsAndKeys:] returns an NSDictionary,
not a NSMutableDictionary.
- WebCoreSupport/WebEditorClient.mm:
(WebEditorClient::updateSpellingUIWithGrammarString):
- 3:25 PM Changeset in webkit [149148] by
-
- 8 edits in trunk/Tools
Pass relatedPage when creating a page in WebKitTestRunner
https://bugs.webkit.org/show_bug.cgi?id=115206
Reviewed by Geoffrey Garen.
To make sure that window.open() opens in the same process even when using multiple processes.
- WebKitTestRunner/PlatformWebView.h:
- WebKitTestRunner/TestController.cpp: (WTR::TestController::createOtherPage): (WTR::TestController::createWebViewWithOptions):
- WebKitTestRunner/efl/PlatformWebViewEfl.cpp: (WTR::PlatformWebView::PlatformWebView):
- WebKitTestRunner/gtk/PlatformWebViewGtk.cpp: (WTR::PlatformWebView::PlatformWebView):
- WebKitTestRunner/mac/PlatformWebViewMac.mm: (-[TestRunnerWKView initWithFrame:contextRef:pageGroupRef:relatedToPage:useTiledDrawing:]): (WTR::PlatformWebView::PlatformWebView):
- WebKitTestRunner/qt/PlatformWebViewQt.cpp: (WTR::PlatformWebView::PlatformWebView):
- WebKitTestRunner/win/PlatformWebViewWin.cpp: (WTR::PlatformWebView::PlatformWebView):
- 3:23 PM Changeset in webkit [149147] by
-
- 5 edits in trunk/Source/WebKit2
Rename DecidePolicyForResponse message to DecidePolicyForResponseSync
https://bugs.webkit.org/show_bug.cgi?id=115207
Reviewed by Beth Dakin.
In preparation for adding a new async DecidePolicyForResponse message,
rename the current one to DecidePolicyForResponseSync. Also split out the async part
into a separate WebPageProxy::decidePolicyForResponse member function.
- UIProcess/WebPageProxy.cpp:
(WebKit::WebPageProxy::WebPageProxy):
(WebKit::WebPageProxy::receivedPolicyDecision):
(WebKit::WebPageProxy::decidePolicyForResponse):
(WebKit::WebPageProxy::decidePolicyForResponseSync):
- UIProcess/WebPageProxy.h:
(WebPageProxy):
- UIProcess/WebPageProxy.messages.in:
- WebProcess/WebCoreSupport/WebFrameLoaderClient.cpp:
(WebKit::WebFrameLoaderClient::dispatchDecidePolicyForResponse):
- 3:17 PM Changeset in webkit [149146] by
-
- 4 edits in trunk/Source
Stack guards are too conservative
https://bugs.webkit.org/show_bug.cgi?id=115147
Reviewed by Mark Hahnenberg.
Source/JavaScriptCore:
Increase stack guard to closer to old size.
- interpreter/Interpreter.cpp:
(JSC::Interpreter::StackPolicy::StackPolicy):
Source/WTF:
Use getrlimit on darwin to get the stack size for the main thread.
- wtf/StackBounds.cpp:
(WTF::StackBounds::initialize):
- 2:39 PM Changeset in webkit [149145] by
-
- 2 edits in trunk/Source/WebKit/blackberry
[BlackBerry] Selection overlay on non-composited iframes are incorrectly positioned.
https://bugs.webkit.org/show_bug.cgi?id=115197
Patch by Andrew Lo <anlo@rim.com> on 2013-04-25
Reviewed by Rob Buis.
When drawing the selection overlay, the rects to
paint when selecting text on non-composited sub-frames
need to be adjusted by the frame position.
- WebKitSupport/SelectionOverlay.cpp:
(BlackBerry::WebKit::SelectionOverlay::paintContents):
- 2:29 PM Changeset in webkit [149144] by
-
- 2 edits in trunk/Source/WebCore
Glyphs may fail to render when using SVG font
https://bugs.webkit.org/show_bug.cgi?id=115193
Reviewed by Simon Fraser.
Calling SimpleFontData::applyTransforms() when the font used is
an SVG font makes little sense since Core Text doesn’t know or
understand SVG fonts and would be passed some other, unrelated
platform font.
- platform/graphics/SimpleFontData.h:
(WebCore::SimpleFontData::applyTransforms):
- 2:21 PM Changeset in webkit [149143] by
-
- 2 edits in trunk/Tools
Fix autocompletion for Benjamin. The Bugzilla email address must come first.
- Scripts/webkitpy/common/config/contributors.json:
- 2:02 PM Changeset in webkit [149142] by
-
- 3 edits in trunk/Source/WebCore
Media elements shouldn't resume playback when a page is restored from the back/forward cache if the WKView isn't in a window
https://bugs.webkit.org/show_bug.cgi?id=115191
Reviewed by Eric Carlson.
If a page is suspended, then resumed when its WebView or WKView has been removed from a window,
the page's media elements will unpause. Check whether media is allowed to start during resume()
via Page::canMediaStart() and if not, register for mediaCanStart notifications. Then, in
mediaCanStart() if the media is force-paused, unpause it.
- html/HTMLMediaElement.cpp:
(WebCore::HTMLMediaElement::suspend):
(WebCore::HTMLMediaElement::resume):
(WebCore::HTMLMediaElement::mediaCanStart):
- html/HTMLMediaElement.h:
(WebCore::HTMLMediaElement::pageConsentRequiredForResume):
- 2:01 PM Changeset in webkit [149141] by
-
- 2 edits in trunk/LayoutTests
Update a comment, mentioning a newly filed bug.
- platform/wk2/TestExpectations:
- 1:37 PM Changeset in webkit [149140] by
-
- 7 edits in trunk/Source
Remove ENABLE(PARSED_STYLE_SHEET_CACHING) and make it always-on.
Rubber-stamped by Anders Koivisto.
Source/WebCore:
- PlatformWinCE.cmake:
- accessibility/AccessibilityObject.h:
- html/HTMLLinkElement.cpp:
(WebCore::HTMLLinkElement::setCSSStyleSheet):
Source/WebKit:
- PlatformWinCE.cmake:
Source/WebKit/blackberry:
- WebCoreSupport/AboutDataEnableFeatures.in:
Source/WTF:
- wtf/FeatureDefines.h:
- 1:26 PM Changeset in webkit [149139] by
-
- 5 edits in trunk/Source
Remove OS(WINCE) from AccessibilityObject.h
https://bugs.webkit.org/show_bug.cgi?id=115192
Reviewed by Andreas Kling.
Use the same files like the the other PLATFORM(WIN) ports use.
Source/WebCore:
- PlatformWinCE.cmake:
- accessibility/AccessibilityObject.h:
(AccessibilityObject):
Source/WebKit:
- PlatformWinCE.cmake:
- 1:11 PM Changeset in webkit [149138] by
-
- 2 edits in trunk/Source/WebCore
Add definition of GlyphBufferGlyph for BlackBerry
https://bugs.webkit.org/show_bug.cgi?id=115183
Patch by Alberto Garcia <agarcia@igalia.com> on 2013-04-25
Reviewed by Xan Lopez.
BlackBerry uses an unsigned int.
- platform/graphics/GlyphBuffer.h:
(WebCore):
- 1:02 PM Changeset in webkit [149137] by
-
- 5 edits in trunk/Source/WebKit/blackberry
[BlackBerry] Make scroll position adjustment work with pages with fixed position elements.
https://bugs.webkit.org/show_bug.cgi?id=115178
Patch by Iris Wu <shuwu@blackberry.com> on 2013-04-25
Reviewed by Rob Buis.
PR 308796
Currently the position WebPage::adjustDocumentScrollPosition adjusts is the top
left point of the viewport.
On the page with fixed position elements, we want it to adjust the position beneath
the fixed elements so it can be always visible.
The basic idea is:
- Detect if there are fixed position elements before going through ProximityDetector.
- If the fixed element exists, calculate its the size and the actual visible position
beneath it.
- Pass the position to ProximityDetector. Then according to the new position we get,
calculate the top left position of the viewport (final scroll position).
- Api/WebPage.cpp:
(BlackBerry::WebKit::WebPage::fixedElementSizeDelta):
(WebKit):
- Api/WebPage.h:
- Api/WebPageCompositor.cpp:
(BlackBerry::WebKit::WebPageCompositorPrivate::findFixedElementRect):
(WebKit):
- Api/WebPageCompositor_p.h:
(WebPageCompositorPrivate):
- 1:00 PM Changeset in webkit [149136] by
-
- 2 edits in trunk/Source/JavaScriptCore
Stack guards are too conservative
https://bugs.webkit.org/show_bug.cgi?id=115147
Reviewed by Geoffrey Garen.
Reduce the limits and simplify the decision making.
- interpreter/Interpreter.cpp:
(JSC::Interpreter::StackPolicy::StackPolicy):
- 12:56 PM Changeset in webkit [149135] by
-
- 2 edits in trunk/Source/WebKit/blackberry
[BlackBerry] Enable balanced page group load deferrer behaviour.
https://bugs.webkit.org/show_bug.cgi?id=115189
Patch by Mike Lattanzio <mlattanzio@blackberry.com> on 2013-04-25
Reviewed by Rob Buis.
Prevent a possible deadlock by enabling balanced deferrers.
Internally reviewed by: Joe Mason
PR 329986
- Api/WebPage.cpp:
(BlackBerry::WebKit::WebPagePrivate::init):
- 12:54 PM Changeset in webkit [149134] by
-
- 2 edits in trunk/Source/JavaScriptCore
JSC: Fix interpreter misbehavior in builds with JIT disabled
https://bugs.webkit.org/show_bug.cgi?id=115190
Patch by Nick Diego Yamane <nick.yamane@openbossa.org> on 2013-04-25
Reviewed by Oliver Hunt.
Commit http://trac.webkit.org/changeset/147858 modified
some details on how JS stack traces are built. The method
"getLineNumberForCallFrame", renamed in that changeset to
"getBytecodeOffsetForCallFrame" is always returning `0' when
JIT is disabled
How to reproduce:
- Build webkit with JIT disabled
- Open MiniBrowser, for example, with http://google.com
- In a debug build, WebProcess will hit the following ASSERT: Source/JavaScriptCore/bytecode/UnlinkedCodeBlock.cpp:279 ASSERT(low);
- interpreter/Interpreter.cpp:
(JSC::getBytecodeOffsetForCallFrame):
- 12:50 PM Changeset in webkit [149133] by
-
- 2 edits in trunk/Tools
Build fix. Forgot to merge args.
- Scripts/webkitpy/tool/bot/irc_command.py:
(Hi.execute):
- 12:44 PM Changeset in webkit [149132] by
-
- 2 edits in trunk/Tools
webkitbot should recognize its own nickname in greetings.
https://bugs.webkit.org/show_bug.cgi?id=115196
Reviewed by Andreas Kling.
Make webkitbot recognize other forms of its nick. Also be tolerant of spaces between the nick and !.
- Scripts/webkitpy/tool/bot/irc_command.py:
(Hi.execute):
- 12:31 PM Changeset in webkit [149131] by
-
- 21 edits2 adds2 deletes in trunk
ScriptExecutionContext log exception should include a column number
https://bugs.webkit.org/show_bug.cgi?id=114315
Reviewed by Oliver Hunt.
Source/WebCore:
Test: inspector/console/console-exception-stack-traces.html
- bindings/js/ScriptCallStackFactory.h:
- bindings/js/ScriptCallStackFactory.cpp:
(WebCore::createScriptCallStackFromException):
Generate a ScriptCallStack from an exception. Use the vm.exceptionStack
if available, and fallback to the exception object where needed.
- bindings/js/JSDOMBinding.cpp:
(WebCore::reportException):
Always include a non-empty call stack with exceptions.
Where not provided, fallback to the exception object.
- dom/Document.h:
- dom/Document.cpp:
(WebCore::Document::logExceptionToConsole):
- dom/ScriptExecutionContext.h:
- dom/ScriptExecutionContext.cpp:
(WebCore::ScriptExecutionContext::PendingException::PendingException):
(ScriptExecutionContext::PendingException):
(WebCore::ScriptExecutionContext::reportException):
- workers/DefaultSharedWorkerRepository.cpp:
- workers/SharedWorkerContext.cpp:
(WebCore::SharedWorkerContext::logExceptionToConsole):
- workers/SharedWorkerContext.h:
- workers/WorkerContext.cpp:
(WebCore::WorkerContext::logExceptionToConsole):
- workers/WorkerContext.h:
- workers/WorkerMessagingProxy.cpp:
(WebCore::WorkerExceptionTask::performTask):
Plumb columnNumber through as needed.
LayoutTests:
- inspector/console/console-exception-stack-traces-expected.txt: Added.
- inspector/console/console-exception-stack-traces.html: Added.
Verifies that exceptions have stack traces and match a console.trace stack.
- http/tests/inspector-enabled/console-exception-while-no-inspector-expected.txt: Removed.
- http/tests/inspector-enabled/console-exception-while-no-inspector.html: Removed.
We now save exception backtraces even if the inspector is not open.
- fast/dom/javascript-url-exception-isolation-expected.txt
- fast/events/remove-target-with-shadow-in-drag-expected.txt
- fast/events/set-attribute-listener-window-onerror-crash-expected.txt
- inspector-protocol/media-query-listener-exception-expected.txt:
- inspector/console/console-uncaught-exception-expected.txt:
- inspector/console/console-uncaught-exception-in-eval-expected.txt:
Update results that now include backtraces or different data. Most better, some worse.
- 12:02 PM WebKitIDL edited by
- [Callback] only applies to parameters, not interfaces (diff)
- 11:59 AM Changeset in webkit [149130] by
-
- 4 edits in trunk/Source/JavaScriptCore
Make checkSyntax take a VM instead of an ExecState
RS=Tim
- 11:57 AM WebKitIDL edited by
- Update [Callback] extended attribute doc due to r149113 (diff)
- 11:50 AM Changeset in webkit [149129] by
-
- 5 edits in trunk/Source/WebCore
Fix build files so WebGLDebugXXX extensions can be used if enabled.
https://bugs.webkit.org/show_bug.cgi?id=113976
Reviewed by Dean Jackson.
No new tests as no new functionality.
- DerivedSources.make:
- GNUmakefile.list.am:
- WebCore.xcodeproj/project.pbxproj:
- bindings/js/JSWebGLRenderingContextCustom.cpp:
- 11:35 AM Changeset in webkit [149128] by
-
- 2 edits in trunk/Source/JavaScriptCore
32 Bit: Crash due to RegExpTest nodes not setting result type to Boolean
https://bugs.webkit.org/show_bug.cgi?id=115188
Reviewed by Geoffrey Garen.
Changed the RegExpTest node to set the AbstractValue to boolean, since that
what it is.
- dfg/DFGAbstractState.cpp:
(JSC::DFG::AbstractState::executeEffects):
- 11:33 AM Changeset in webkit [149127] by
-
- 2 edits in trunk/Source/WebCore
cloneChildNodes looks for deleteButtonController in each level of recursion
https://bugs.webkit.org/show_bug.cgi?id=115146
Reviewed by Andreas Kling.
Obtain the delete button controller upfront, and shallow copy descendents of each child
so that we don't look for the delete button controller inside cloneNode called on each child.
Performance Tests: DOM/CloneNodes.html
- dom/ContainerNode.cpp:
(WebCore::cloneChildNodesAvoidingDeleteButon): Extracted.
(WebCore::ContainerNode::cloneChildNodes):
- 11:32 AM Changeset in webkit [149126] by
-
- 8 edits in trunk
HTMLOptionsCollection's namedItem and name getter should return the first item
https://bugs.webkit.org/show_bug.cgi?id=115150
Reviewed by Andreas Kling.
Source/WebCore:
Following the resolution in http://lists.whatwg.org/pipermail/whatwg-whatwg.org/2012-December/038355.html,
the spefication has been updated to only return the first item when there are multiple items of the same name
in HTMLOptionsCollection; this new behavior matches that of Firefox and Opera (Presto).
Implement this new behavior to remove the custom binding code and use the fast path in namedItem and name
getter of HTMLOptionsCollection. (Obtaining all items for a given name is expensive!).
Tests: fast/dom/HTMLSelectElement/named-options.html
fast/dom/html-collections-named-getter.html
- bindings/js/JSHTMLOptionsCollectionCustom.cpp:
(WebCore): Removed the custom bindings for name getter and namedItem.
- html/HTMLOptionsCollection.idl:
LayoutTests:
Changed the expectations of the tests.
- fast/dom/HTMLSelectElement/named-options-expected.txt:
- fast/dom/HTMLSelectElement/script-tests/named-options.js:
- fast/dom/html-collections-named-getter-expected.txt:
- fast/dom/html-collections-named-getter.html:
- 11:31 AM Changeset in webkit [149125] by
-
- 59 edits1 move3 adds1 delete in trunk
Web Inspector: ConsoleMessage should include line and column number where possible
https://bugs.webkit.org/show_bug.cgi?id=114929
Source/WebCore:
- adds "m_column" to WebCore::ConsoleMessage
- adds "column" to Console.ConsoleMessage in the inspector protocol
- set the column number for console.* functions (Console.cpp)
- set the column number for XSLT errors (XSLTProcessor)
- plumb columnNumber everywhere else it is needed, set it to 0 and file bugs for all cases missing columnNumber that could provide it.
Reviewed by Timothy Hatcher.
Test: inspector/console/console-url-line-column.html
inspector/console/console-messages-stack-traces.html
- inspector/ConsoleMessage.h:
- inspector/ConsoleMessage.cpp:
(WebCore::ConsoleMessage::ConsoleMessage):
(WebCore::ConsoleMessage::autogenerateMetadata):
(WebCore::ConsoleMessage::addToFrontend):
(WebCore::ConsoleMessage::isEqual):
Add m_column and set it where appropriate.
- inspector/Inspector.json:
Add column property to Console.ConsoleMessage.
- page/Console.cpp:
(WebCore::internalAddMessage):
(WebCore::Console::profile):
Set columnNumber like lineNumber from the last stack frame.
(WebCore::Console::groupEnd):
Line and column are unused in this message type, set both to 0.
- xml/XSLTProcessorLibxslt.cpp:
(WebCore::XSLTProcessor::parseErrorFunc):
- xml/XSLTProcessorQt.cpp:
(WebCore::XSLTMessageHandler::handleMessage):
Add real column numbers, the XSLT handlers already had it available.
- bindings/js/JSCustomXPathNSResolver.cpp:
(WebCore::JSCustomXPathNSResolver::lookupNamespaceURI):
- css/CSSParser.cpp:
(WebCore::CSSParser::logError):
- dom/ScriptExecutionContext.h:
- dom/ScriptExecutionContext.cpp:
(WebCore::ScriptExecutionContext::addConsoleMessage):
- dom/Document.cpp:
(WebCore::Document::logExceptionToConsole):
(WebCore::Document::addMessage):
- dom/Document.h:
- inspector/InspectorConsoleAgent.cpp:
(WebCore::InspectorConsoleAgent::addMessageToConsole):
(WebCore::InspectorConsoleAgent::stopTiming):
(WebCore::InspectorConsoleAgent::didFinishXHRLoading):
(WebCore::InspectorConsoleAgent::didReceiveResponse):
(WebCore::InspectorConsoleAgent::didFailLoading):
- inspector/InspectorConsoleAgent.h:
- inspector/InspectorConsoleInstrumentation.h:
(WebCore::InspectorInstrumentation::addMessageToConsole):
(WebCore::InspectorInstrumentation::addStartProfilingMessageToConsole):
- inspector/InspectorInstrumentation.cpp:
(WebCore::InspectorInstrumentation::addMessageToConsoleImpl):
(WebCore::InspectorInstrumentation::addStartProfilingMessageToConsoleImpl):
(WebCore::InspectorInstrumentation::addProfileImpl):
- inspector/InspectorInstrumentation.h:
(InspectorInstrumentation):
- inspector/InspectorProfilerAgent.cpp:
(WebCore::InspectorProfilerAgent::addProfile):
(WebCore::InspectorProfilerAgent::addProfileFinishedMessageToConsole):
(WebCore::InspectorProfilerAgent::addStartProfilingMessageToConsole):
(WebCore::InspectorProfilerAgent::start):
(WebCore::InspectorProfilerAgent::stop):
- inspector/InspectorProfilerAgent.h:
(InspectorProfilerAgent):
- loader/EmptyClients.h:
(WebCore::EmptyChromeClient::addMessageToConsole):
- page/ChromeClient.h:
(WebCore::ChromeClient::addMessageToConsole):
- page/ContentSecurityPolicy.cpp:
(WebCore::ContentSecurityPolicy::logToConsole):
- page/PageConsole.cpp:
(WebCore::PageConsole::addMessage):
- page/PageConsole.h:
- workers/DefaultSharedWorkerRepository.cpp:
(SharedWorkerProxy):
(WebCore::postExceptionTask):
(WebCore::SharedWorkerProxy::postExceptionToWorkerObject):
(WebCore::postConsoleMessageTask):
(WebCore::SharedWorkerProxy::postConsoleMessageToWorkerObject):
- workers/SharedWorkerContext.cpp:
(WebCore::SharedWorkerContext::logExceptionToConsole):
- workers/WorkerContext.cpp:
(WebCore::WorkerContext::logExceptionToConsole):
(WebCore::WorkerContext::addConsoleMessage):
(WebCore::WorkerContext::addMessage):
(WebCore::WorkerContext::addMessageToWorkerConsole):
- workers/WorkerContext.h:
- workers/WorkerMessagingProxy.cpp:
(WebCore::WorkerExceptionTask::create):
(WebCore::WorkerExceptionTask::WorkerExceptionTask):
(WebCore::WorkerExceptionTask::performTask):
(WorkerExceptionTask):
(WebCore::WorkerMessagingProxy::postExceptionToWorkerObject):
(WebCore::postConsoleMessageTask):
(WebCore::WorkerMessagingProxy::postConsoleMessageToWorkerObject):
- workers/WorkerMessagingProxy.h:
(WorkerMessagingProxy):
- workers/WorkerReportingProxy.h:
(WorkerReportingProxy):
Plumb columnNumber through where appropriate. File bugs where missing.
Source/WebKit/blackberry:
Reviewed by Timothy Hatcher.
- Api/DumpRenderTreeClient.h:
- Api/WebPageClient.h:
- WebCoreSupport/ChromeClientBlackBerry.cpp:
(WebCore::ChromeClientBlackBerry::addMessageToConsole):
- WebCoreSupport/ChromeClientBlackBerry.h:
(ChromeClientBlackBerry):
Source/WebKit/efl:
Reviewed by Timothy Hatcher.
- WebCoreSupport/ChromeClientEfl.cpp:
(WebCore::ChromeClientEfl::addMessageToConsole):
- WebCoreSupport/ChromeClientEfl.h:
(ChromeClientEfl):
Source/WebKit/gtk:
Reviewed by Timothy Hatcher.
- WebCoreSupport/ChromeClientGtk.cpp:
(WebKit::ChromeClient::addMessageToConsole):
- WebCoreSupport/ChromeClientGtk.h:
(ChromeClient):
Source/WebKit/mac:
Reviewed by Timothy Hatcher.
- WebCoreSupport/WebChromeClient.h:
- WebCoreSupport/WebChromeClient.mm:
(WebChromeClient::addMessageToConsole):
Source/WebKit/qt:
Reviewed by Timothy Hatcher.
- WebCoreSupport/ChromeClientQt.cpp:
(WebCore::ChromeClientQt::addMessageToConsole):
- WebCoreSupport/ChromeClientQt.h:
Source/WebKit/win:
Reviewed by Timothy Hatcher.
- WebCoreSupport/WebChromeClient.cpp:
(WebChromeClient::addMessageToConsole):
- WebCoreSupport/WebChromeClient.h:
Source/WebKit/wince:
Reviewed by Timothy Hatcher.
- WebCoreSupport/ChromeClientWinCE.cpp:
(WebKit::ChromeClientWinCE::addMessageToConsole):
- WebCoreSupport/ChromeClientWinCE.h:
Source/WebKit2:
Reviewed by Timothy Hatcher.
- WebProcess/WebCoreSupport/WebChromeClient.cpp:
(WebKit::WebChromeClient::addMessageToConsole):
- WebProcess/WebCoreSupport/WebChromeClient.h:
LayoutTests:
Update a test that was outputting the url and line number of console
method calls, to output the column number as well. Added a test that
verifies that console API messages have stack traces if appropriate.
Reviewed by Timothy Hatcher.
- inspector/console/console-url-and-line-expected.txt: Removed.
- inspector/console/console-url-line-column-expected.txt: Added.
- inspector/console/console-url-line-column.html: Renamed from LayoutTests/inspector/console/console-url-and-line.html.
- inspector/console/console-messages-stack-traces-expected.txt: Added.
- inspector/console/console-messages-stack-traces.html: Added.
- 11:29 AM Changeset in webkit [149124] by
-
- 2 edits in trunk/Source/WebKit/blackberry
[BlackBerry] Get rid of return in void method
https://bugs.webkit.org/show_bug.cgi?id=115186
Patch by Konrad Piascik <kpiascik@blackberry.com> on 2013-04-25
Reviewed by Rob Buis.
- Api/WebPage.cpp:
(BlackBerry::WebKit::WebPage::setForcedTextEncoding):
- 11:19 AM Changeset in webkit [149123] by
-
- 5 edits in trunk/Source
[Qt] Animated opacity does not trigger accelerated compositing
https://bugs.webkit.org/show_bug.cgi?id=115107
Reviewed by Simon Fraser.
Source/WebCore:
Add extra option to also trigger compositing on animated opacity.
- page/ChromeClient.h:
- rendering/RenderLayerCompositor.cpp:
(WebCore::RenderLayerCompositor::requiresCompositingForAnimation):
Source/WebKit/qt:
We should trigger compositing on animated opacity as it is always faster on Qt.
- WebCoreSupport/ChromeClientQt.cpp:
VideoTrigger removed because we have already disabled AC for video in qwebsettings.
- 11:09 AM Changeset in webkit [149122] by
-
- 1 edit in trunk/Tools/Scripts/webkitpy/common/config/contributors.json
Unreviewed. Adding myself to the reviwers list.
- 10:59 AM Changeset in webkit [149121] by
-
- 3 edits in trunk/Source/WebKit2
Thread safety issues in WKCustomProtocol.
<rdar://problem/13247304> and https://bugs.webkit.org/show_bug.cgi?id=115185
Reviewed by Alexey Proskuryakov.
- Shared/Network/CustomProtocols/CustomProtocolManager.h:
- Shared/Network/CustomProtocols/mac/CustomProtocolManagerMac.mm:
(WebKit::CustomProtocolManager::addCustomProtocol): Protect m_customProtocolMap with its mutex.
(WebKit::CustomProtocolManager::removeCustomProtocol): Ditto.
(WebKit::CustomProtocolManager::didFailWithError):
(WebKit::CustomProtocolManager::didLoadData):
(WebKit::CustomProtocolManager::didReceiveResponse):
(WebKit::CustomProtocolManager::didFinishLoading):
(WebKit::CustomProtocolManager::protocolForID): Protect m_customProtocolMap with its mutex, and return
a RetainPtr instead of a raw pointer.
- 10:48 AM Changeset in webkit [149120] by
-
- 4 edits3 adds in trunk
Sec-WebSocket-Extensions header field must not appear more than once in an HTTP response.
https://bugs.webkit.org/show_bug.cgi?id=115128
Patch by Lamarque V. Souza <Lamarque.Souza@basyskom.com> on 2013-04-25
Reviewed by Alexey Proskuryakov.
Source/WebCore:
According to WebSocket specification Sec-WebSocket-Extensions header field
must not appear more than once in an HTTP response. It is ok if it appears
more than once in client request. Also add a check for invalid UTF-8
characters when parsing Sec-WebSocket-Extensions quoted string.
Test: http/tests/websocket/tests/hybi/handshake-fail-by-more-extensions-header.html
- Modules/websockets/WebSocketExtensionParser.cpp:
(WebCore::WebSocketExtensionParser::consumeQuotedString): Return false if there is
invalid character in the quoted string being parsed.
- Modules/websockets/WebSocketHandshake.cpp:
(WebCore::WebSocketHandshake::readHTTPHeaders): Check if Sec-WebSocket-Extensions
appears more than once in header response. Abort handshake if it does.
LayoutTests:
- http/tests/websocket/tests/hybi/handshake-fail-by-more-extensions-header-expected.txt: Added.
- http/tests/websocket/tests/hybi/handshake-fail-by-more-extensions-header.html: Added.
- http/tests/websocket/tests/hybi/handshake-fail-by-more-extensions-header_wsh.py: Added.
- 10:31 AM Changeset in webkit [149119] by
-
- 3 edits in trunk/Source/WebKit2
Further build fixes: correct two more misnames of WebProcessShim.dyld.
- WebProcess/EntryPoint/mac/LegacyProcess/WebContentProcessMain.mm:
(WebKit::WebContentProcessMainDelegate::doPreInitializationWork):
- WebProcess/EntryPoint/mac/XPCService/WebContentService/Info.plist:
- 10:12 AM Changeset in webkit [149118] by
-
- 2 edits in trunk/Source/WebKit2
Make gobject-introspection warnings non-fatal
https://bugs.webkit.org/show_bug.cgi?id=115184
Reviewed by Carlos Garcia Campos.
- GNUmakefile.am: Make it so that gobject-introspection errors do not fail the buld
until we get to a point where gobject-introspection gives consistent warnings.
- 9:46 AM Changeset in webkit [149117] by
-
- 14 edits4 adds in trunk/Source/WebKit2
[GTK] Add WebKitWebViewGroup to WebKit2 GTK+ API
https://bugs.webkit.org/show_bug.cgi?id=111265
Reviewed by Anders Carlsson.
In the current API the settings are always shared by all web
views, because they are always created on the default page
group. This makes impossible to have different settings on
different web views. Expose the page group in the API,
so that users can create web views on a group different than the
default one to have their own settings.
- GNUmakefile.list.am: Add new files to compilation.
- UIProcess/API/gtk/WebKitSettings.cpp:
(webkitSettingsGetPreferences): Helper private method to get the
WebPreferences wrapped by a WebKitSettings object.
- UIProcess/API/gtk/WebKitSettingsPrivate.h:
- UIProcess/API/gtk/WebKitWebContext.cpp:
(_WebKitWebContextPrivate): Add a default WebKitWebViewGroup.
(webkitWebContextCreatePageForWebView): Create the WebPageProxy
with the WebPageGroup of the given WebKitWebViewGroup.
(webkitWebContextGetDefaultWebViewGroup): Helper private method to
get the default WebKitWebViewGroup.
- UIProcess/API/gtk/WebKitWebContextPrivate.h:
- UIProcess/API/gtk/WebKitWebView.cpp:
(_WebKitWebViewPrivate): Add a WebKitWebViewGroup in case the view
is created with a group.
(webkitWebViewUpdateSettings): Keep a reference of the group
settings to be able to disconnect the signals when the settings
object is replaced in the view group and connect to the signals
for the custom settings.
(webkitWebViewSettingsChanged): Disconnect the signals for the
previous settings and set the new settings.
(webkitWebViewDisconnectSettingsChangedSignalHandler): Disconnect
notify::settings signal of WebKitWebViewGroup.
(webkitWebViewConstructed): Pass the current WebKitWebViewGroup to
webkitWebContextCreatePageForWebView().
(webkitWebViewSetProperty): Add setter for group property.
(webkitWebViewGetProperty): Add getter for group property.
(webkitWebViewDispose): Call
webkitWebViewDisconnectSettingsChangedSignalHandler().
(webkit_web_view_class_init): Add WebKitWebView:group property.
(webkitWebViewHandleAuthenticationChallenge): Use
webkit_web_view_get_settings().
(webkit_web_view_new_with_group): Create a new web view with the
given WebKitWebViewGroup.
(webkit_web_view_get_group): Return the current WebKitWebViewGroup
or the default one if the view was not created with
webkit_web_view_new_with_group().
(webkit_web_view_set_settings): Set the settings of the current
WebKitWebViewGroup.
(webkit_web_view_get_settings): Get the settings of the current
WebKitWebViewGroup.
(webkit_web_view_set_zoom_level): Use
webkit_web_view_get_settings().
(webkit_web_view_get_zoom_level): Ditto.
- UIProcess/API/gtk/WebKitWebView.h:
- UIProcess/API/gtk/WebKitWebViewGroup.cpp: Added.
(_WebKitWebViewGroupPrivate):
(webkitWebViewGroupSetProperty):
(webkitWebViewGroupGetProperty):
(webkitWebViewGroupConstructed):
(webkit_web_view_group_class_init):
(webkitWebViewGroupAttachSettingsToPageGroup): Set the
WebPreferences of the current WebKitSettings to the WebPageGroup.
(webkitWebViewGroupCreate): Create a new WebKitWebViewGroup for
the given WebPageGroup.
(webkitWebViewGroupGetPageGroup): Helper private method to get the
WebPageGroup wrapped by the WebKitWebViewGroup.
(webkit_web_view_group_new): Create a new WebKitWebViewGroup.
(webkit_web_view_group_get_name): Return the name of a
WebKitWebViewGroup.
(webkit_web_view_group_get_settings): Get the settings of a
WebKitWebViewGroup.
(webkit_web_view_group_set_settings): Set new settings of a
WebKitWebViewGroup.
- UIProcess/API/gtk/WebKitWebViewGroup.h: Added.
(_WebKitWebViewGroup):
(_WebKitWebViewGroupClass):
- UIProcess/API/gtk/WebKitWebViewGroupPrivate.h: Added.
- UIProcess/API/gtk/docs/webkit2gtk-docs.sgml: Add
WebKitWebViewGroup section.
- UIProcess/API/gtk/docs/webkit2gtk-sections.txt: Add new symbols.
- UIProcess/API/gtk/docs/webkit2gtk.types: Add
webkit_web_view_group_get_type.
- UIProcess/API/gtk/tests/GNUmakefile.am: Add new files to
compilation.
- UIProcess/API/gtk/tests/TestWebKitWebView.cpp:
(testWebViewSettings): We can't watch settings in the default view
group anymore, because they are not released with the web view but
with the web context that is freed after the Test destructor.
- UIProcess/API/gtk/tests/TestWebKitWebViewGroup.cpp: Added.
(testWebViewGroupDefault):
(testWebViewGroupNewGroup):
(testWebViewNewWithGroup):
(testWebViewGroupSettings):
(beforeAll):
(afterAll):
- UIProcess/API/gtk/webkit2.h: Include WebKitWebViewGroup.h.
- 9:39 AM Changeset in webkit [149116] by
-
- 2 edits in trunk/Source/WTF
[Windows] Compile fix.
https://bugs.webkit.org/show_bug.cgi?id=115170
Patch by peavo@outlook.com <peavo@outlook.com> on 2013-04-25
Reviewed by Filip Pizlo.
Forward declare function, instead of including intrin.h.
- wtf/Atomics.h:
- 9:32 AM Changeset in webkit [149115] by
-
- 2 edits in trunk/Source/WebCore
REGRESSION(r148758): Remove WTFLogAlways call from IconLoader.
https://bugs.webkit.org/show_bug.cgi?id=115182
Reviewed by Dan Bernstein.
The call added in r148758 was apparently a leftover from some debugging
session, and it makes the Interactive/window-resize.html performance
test randomly fail on Qt and EFL due to WTFLogAlways writing to stderr.
- loader/icon/IconLoader.cpp:
(WebCore::IconLoader::notifyFinished):
- 9:18 AM Changeset in webkit [149114] by
-
- 2 edits in trunk/Source/JavaScriptCore
REGRESSION(r137994): Random crashes occur with SH4 JSC.
https://bugs.webkit.org/show_bug.cgi?id=115167.
Patch by Julien Brianceau <jbrianceau@nds.com> on 2013-04-25
Reviewed by Oliver Hunt.
Since r137994, uncommited pages could be inside the area of memory in
parameter of the cacheFlush function. That's why we have to flush each
page separately to avoid a fail of the whole flush, if an uncommited page
is in the area.
This patch is very similar to changeset 145194 made for ARMv7 architecture,
see https://bugs.webkit.org/show_bug.cgi?id=111441 for further information.
- assembler/SH4Assembler.h:
(JSC::SH4Assembler::cacheFlush):
- 8:55 AM Changeset in webkit [149113] by
-
- 32 edits in trunk/Source/WebCore
Add support for Web IDL callback interfaces to the bindings generator
https://bugs.webkit.org/show_bug.cgi?id=115179
Reviewed by Kentaro Hara.
Add support in the bindings generator for Web IDL callback interfaces:
http://dev.w3.org/2006/webapi/WebIDL/#dfn-callback-interface
Drop support for the WebKit-specific [Callback] extended attributes
on Web IDL interfaces and update existing IDL files to use proper
callback interfaces instead.
No new tests, no behavior change.
- Modules/filesystem/EntriesCallback.idl:
- Modules/filesystem/EntryCallback.idl:
- Modules/filesystem/ErrorCallback.idl:
- Modules/filesystem/FileCallback.idl:
- Modules/filesystem/FileSystemCallback.idl:
- Modules/filesystem/FileWriterCallback.idl:
- Modules/filesystem/MetadataCallback.idl:
- Modules/geolocation/PositionCallback.idl:
- Modules/geolocation/PositionErrorCallback.idl:
- Modules/mediastream/NavigatorUserMediaErrorCallback.idl:
- Modules/mediastream/NavigatorUserMediaSuccessCallback.idl:
- Modules/mediastream/RTCErrorCallback.idl:
- Modules/mediastream/RTCSessionDescriptionCallback.idl:
- Modules/mediastream/RTCStatsCallback.idl:
- Modules/notifications/NotificationPermissionCallback.idl:
- Modules/quota/StorageErrorCallback.idl:
- Modules/quota/StorageQuotaCallback.idl:
- Modules/quota/StorageUsageCallback.idl:
- Modules/webaudio/AudioBufferCallback.idl:
- Modules/webdatabase/DatabaseCallback.idl:
- Modules/webdatabase/SQLStatementCallback.idl:
- Modules/webdatabase/SQLStatementErrorCallback.idl:
- Modules/webdatabase/SQLTransactionCallback.idl:
- Modules/webdatabase/SQLTransactionErrorCallback.idl:
- Modules/webdatabase/SQLTransactionSyncCallback.idl:
- bindings/scripts/CodeGeneratorJS.pm:
(GenerateInterface):
- bindings/scripts/IDLParser.pm:
(parseCallbackRestOrInterface):
- bindings/scripts/test/TestCallback.idl:
- dom/RequestAnimationFrameCallback.idl:
- dom/StringCallback.idl:
- html/VoidCallback.idl:
- 8:17 AM Changeset in webkit [149112] by
-
- 2 edits in trunk/Source/WTF
Use std::move in AtomicString
https://bugs.webkit.org/show_bug.cgi?id=115102
Reviewed by Anders Carlsson.
Use std::move in AtomicString instead of direct casting to rvalue references.
- wtf/text/AtomicString.h:
(AtomicString):
(WTF::AtomicString::AtomicString):
(WTF::AtomicString::operator=):
- 7:52 AM Changeset in webkit [149111] by
-
- 8 edits1 copy in trunk
[GStreamer] Add audio/speex MIME type as supported
https://bugs.webkit.org/show_bug.cgi?id=115032
Patch by Adrian Perez de Castro <Adrian Perez de Castro> on 2013-04-25
Reviewed by Philippe Normand.
Source/WebCore:
GStreamer has support for the Speex codec (http://www.speex.org).
Speex streams may be embedded in a Ogg container, or standalone.
The case of the Ogg container is already covered by the "*/ogg"
MIME types declared as supported by the GStreamer media playing
code, but for standalone streams to work, "audio/speex" has to
be added.
With this, and the needed GStreamer plugin installed, the GTK+
launcher is able to correctly play Speex streams served with
"Content-type: audio/speex".
- platform/graphics/gstreamer/MediaPlayerPrivateGStreamer.cpp:
LayoutTests:
Speex (http://www.speex.org) streams can be embedded into Ogg
Ogg containers, so add a check for it in the test case.
- media/media-can-play-ogg-expected.txt:
- media/media-can-play-ogg.html:
- platform/blackberry/media/media-can-play-ogg-expected.txt:
- platform/mac/media/media-can-play-ogg-expected.txt:
- platform/qt-linux/media/media-can-play-ogg-expected.txt:
- platform/win/media/media-can-play-ogg-expected.txt:
- 7:33 AM Changeset in webkit [149110] by
-
- 9 edits in trunk/Source/WebCore
Add FINAL decorators to the InlineBox class hierarchy.
<http://webkit.org/b/115177>
Reviewed by Antti Koivisto.
From Blink r148628 by <cevans@chromium.org>:
FINAL is a macro in wtf/Compiler.h that does the correct thing if the compiler does not support "final")
The approach used is as simple as possible whilst being thorough.
So, leaf classes have FINAL applied to the whole class whereas intermediary classes have FINAL applied to relevant methods.
FINAL allows a compiler to devirtualize call sites and turn them into direct calls. As you might expect, this is perf positive:
(clang on Linux):
- line_layout.html goes from 120 runs/s -> 123 runs/2, +2.5%
- html5-full-render.html goes from 3176ms -> 3162ms, +0.4%
I have confidence that the former result is statistically significant (as the numbers are very very stable) but not the latter.
- rendering/EllipsisBox.h:
- rendering/InlineFlowBox.h:
- rendering/InlineTextBox.h:
- rendering/RootInlineBox.h:
- rendering/TrailingFloatsRootInlineBox.h:
- rendering/svg/SVGInlineFlowBox.h:
- rendering/svg/SVGInlineTextBox.h:
- rendering/svg/SVGRootInlineBox.h:
- 7:19 AM Changeset in webkit [149109] by
-
- 2 edits in trunk/Tools
[GTK] Tools/gtk/install-dependencies added ragel dependency
https://bugs.webkit.org/show_bug.cgi?id=115163
Patch by Xavier Castano <xcastanho@igalia.com> on 2013-04-25
Reviewed by Xan Lopez.
The install-dependencies script was missing a dependency on ragel
which is expected when building HarfBuzz.
- gtk/install-dependencies: Added a dependency on ragel for both
apt and yum.
- 7:05 AM Changeset in webkit [149108] by
-
- 2 edits in trunk/Tools
[GTK] Remove pango from optional jhbuild dependencies
https://bugs.webkit.org/show_bug.cgi?id=115162
Patch by Eduardo Lima Mitev <elima@igalia.com> on 2013-04-25
Reviewed by Martin Robinson.
Pango is not needed since revision r148293 that lowers the pango
dependency to 1.30, which is available in most distributions.
- gtk/jhbuild-optional.modules: Removes description of pango module.
- 6:46 AM Changeset in webkit [149107] by
-
- 5 edits in trunk/Source/WebKit2
Unreviewed build fix: Fix the name of DyldInterpose.h in
SecItemShimLibrary.mm, fix the name of WebProcessShim.dyld in
WebContentServiceEntryPoint.mm, and stop linking against libWTF.a
directly (which means removing ASSERTs from
CookieStorageShimLibrary.cpp).
- Shared/mac/CookieStorageShimLibrary.cpp:
(WebKit::shimCFHTTPCookieStorageCopyRequestHeaderFieldsForURL):
(WebKit::WebKitCookieStorageShimInitialize):
- WebKit2.xcodeproj/project.pbxproj:
- WebProcess/EntryPoint/mac/XPCService/WebContentServiceEntryPoint.mm:
(WebContentServiceInitializer):
- WebProcess/mac/SecItemShimLibrary.mm:
- 6:43 AM Changeset in webkit [149106] by
-
- 4 edits4 adds in trunk
CSS parser: Add error recovery while parsing @-webkit-keyframes key values.
<http://webkit.org/b/115175>
Source/WebCore:
From Blink r148714 by <apavlov@chromium.org>:
If not a percentage, "from", or "to" value is used in a key list, the rule is erroneous,
and due to the absense of recovery, the parser skips the following, valid CSS rule.
On a related note, keyframes, whose selectors contain invalid keys, should be discarded
altogether, according to <http://www.w3.org/TR/css3-animations/#keyframes>
Tests: animations/keyframes-invalid-keys.html
fast/css/webkit-keyframes-errors.html
- css/CSSGrammar.y.in:
- css/CSSParser.cpp:
(WebCore::CSSParser::rewriteSpecifiers):
LayoutTests:
From Blink r148714 by <apavlov@chromium.org>.
- animations/keyframes-invalid-keys-expected.txt: Added.
- animations/keyframes-invalid-keys.html: Added.
- fast/css/webkit-keyframes-errors-expected.html: Added.
- fast/css/webkit-keyframes-errors.html: Added.
- 6:23 AM Changeset in webkit [149105] by
-
- 7 edits in trunk/Source/WebCore
REGRESSION (r147797): Animations slideshows of images on www.thesuperficial.com are slow
https://bugs.webkit.org/show_bug.cgi?id=115172
Reviewed by Andreas Kling.
On this page ads dynamically loaded to subframes on slideshow navigation switch us to state where we throttle layer flushes.
Fix by ignoring any subframe-originated loads when determining throttling.
- loader/FrameLoader.cpp:
(WebCore::FrameLoader::loadProgressingStatusChanged):
- loader/ProgressTracker.cpp:
(WebCore::ProgressTracker::isMainLoadProgressing):
Rename, ignore subframe originated loads. Subframe loads that are initiated during the main load still count.
- loader/ProgressTracker.h:
(ProgressTracker):
- page/FrameView.cpp:
(WebCore::FrameView::updateLayerFlushThrottlingInAllFrames):
- page/FrameView.h:
(FrameView):
- rendering/RenderLayerBacking.cpp:
(WebCore::computeTileCoverage):
- 6:10 AM Changeset in webkit [149104] by
-
- 2 edits in trunk/LayoutTests
[Qt] Unreviewed gardening. Skip some failing tests after r148996.
- platform/qt/TestExpectations:
- 6:04 AM Changeset in webkit [149103] by
-
- 2 edits in trunk/Tools
Remove Chromium-specific cases from FeatureList.pm
https://bugs.webkit.org/show_bug.cgi?id=115169
Patch by Seokju Kwon <Seokju Kwon> on 2013-04-25
Reviewed by Andreas Kling.
- Scripts/webkitperl/FeatureList.pm:
- 5:56 AM Changeset in webkit [149102] by
-
- 4 edits2 adds in trunk
SVG: Fix viewBox animations on shapes with non-scaling-stroke.
<http://webkit.org/b/115173>
Source/WebCore:
From Blink r149058 by <pdr@chromium.org>:
Previously hasAttribute() to determine if a viewBox was set in the transform
code for non-scaling-strokes. hasAttribute() should not be used in this case,
as it will return false if the attribute is not set but the value is animating.
This patch switches to checking if the viewBox is empty instead of checking for
the presence of the attribute.
Test: svg/stroke/animated-non-scaling-stroke.html
- svg/SVGSVGElement.cpp:
(WebCore::SVGSVGElement::localCoordinateSpaceTransform):
- svg/SVGSVGElement.h:
(WebCore::SVGSVGElement::hasEmptyViewBox):
LayoutTests:
From Blink r149058 by <pdr@chromium.org>.
- svg/stroke/animated-non-scaling-stroke-expected.html: Added.
- svg/stroke/animated-non-scaling-stroke.html: Added.
- 5:32 AM Changeset in webkit [149101] by
-
- 2 edits in trunk/Source/WebCore
Remove isPluginElement hack in Document::setFocusedNode()
<http://webkit.org/b/115171>
From Blink r148800 by <tkent@chromium.org>:
This hack is no longer necessary because r147591 made isFocusable for plugin
elements work correctly.
- dom/Document.cpp:
(WebCore::Document::setFocusedNode):
- 5:13 AM Changeset in webkit [149100] by
-
- 4 edits in trunk/LayoutTests
[Qt] Unreviewed gardening. Unskip now passing tests.
- platform/qt-5.0-wk1/TestExpectations:
- platform/qt-5.0-wk2/TestExpectations:
- platform/qt/TestExpectations:
- 3:49 AM Changeset in webkit [149099] by
-
- 23 edits22 copies8 adds in trunk/LayoutTests
[Qt] Unreviewed gardening. Organize expected files after r149095.
- platform/qt-5.0-wk1/editing/deleting/delete-by-word-002-expected.txt: Copied from LayoutTests/platform/qt/editing/deleting/delete-by-word-002-expected.txt.
- platform/qt-5.0-wk1/editing/deleting/delete-cell-contents-expected.txt: Copied from LayoutTests/platform/qt/editing/deleting/delete-cell-contents-expected.txt.
- platform/qt-5.0-wk1/editing/execCommand/button-expected.txt: Copied from LayoutTests/platform/qt/editing/execCommand/button-expected.txt.
- platform/qt-5.0-wk1/editing/execCommand/createLink-expected.txt: Copied from LayoutTests/platform/qt/editing/execCommand/createLink-expected.txt.
- platform/qt-5.0-wk1/editing/execCommand/indent-pre-expected.txt: Copied from LayoutTests/platform/qt/editing/execCommand/indent-pre-expected.txt.
- platform/qt-5.0-wk1/editing/execCommand/unlink-expected.txt: Copied from LayoutTests/platform/qt/editing/execCommand/unlink-expected.txt.
- platform/qt-5.0-wk1/editing/input/emacs-ctrl-o-expected.txt: Copied from LayoutTests/platform/qt/editing/input/emacs-ctrl-o-expected.txt.
- platform/qt-5.0-wk1/editing/inserting/insert-before-link-1-expected.txt: Copied from LayoutTests/platform/qt/editing/inserting/insert-before-link-1-expected.txt.
- platform/qt-5.0-wk1/editing/pasteboard/display-block-on-spans-expected.txt: Copied from LayoutTests/platform/qt/editing/pasteboard/display-block-on-spans-expected.txt.
- platform/qt-5.0-wk1/editing/pasteboard/emacs-cntl-y-001-expected.txt: Copied from LayoutTests/platform/qt/editing/pasteboard/emacs-cntl-y-001-expected.txt.
- platform/qt-5.0-wk1/editing/pasteboard/merge-end-list-2-expected.txt: Copied from LayoutTests/platform/qt/editing/pasteboard/merge-end-list-2-expected.txt.
- platform/qt-5.0-wk1/editing/pasteboard/merge-end-table-2-expected.txt: Copied from LayoutTests/platform/qt/editing/pasteboard/merge-end-table-2-expected.txt.
- platform/qt-5.0-wk1/editing/pasteboard/page-zoom-expected.txt: Copied from LayoutTests/platform/qt/editing/pasteboard/page-zoom-expected.txt.
- platform/qt-5.0-wk1/editing/selection/extend-selection-word-expected.txt: Copied from LayoutTests/platform/qt/editing/selection/extend-selection-word-expected.txt.
- platform/qt-5.0-wk1/editing/style/apply-through-end-of-document-expected.txt: Copied from LayoutTests/platform/qt/editing/style/apply-through-end-of-document-expected.txt.
- platform/qt-5.0-wk1/editing/style/remove-underline-across-paragraph-expected.txt: Copied from LayoutTests/platform/qt/editing/style/remove-underline-across-paragraph-expected.txt.
- platform/qt-5.0-wk1/editing/style/remove-underline-across-paragraph-in-bold-expected.txt: Copied from LayoutTests/platform/qt/editing/style/remove-underline-across-paragraph-in-bold-expected.txt.
- platform/qt-5.0-wk1/editing/style/remove-underline-expected.txt: Copied from LayoutTests/platform/qt/editing/style/remove-underline-expected.txt.
- platform/qt-5.0-wk1/editing/style/remove-underline-from-stylesheet-expected.txt: Copied from LayoutTests/platform/qt/editing/style/remove-underline-from-stylesheet-expected.txt.
- platform/qt-5.0-wk1/editing/style/remove-underline-in-bold-expected.txt: Copied from LayoutTests/platform/qt/editing/style/remove-underline-in-bold-expected.txt.
- platform/qt-5.0-wk1/editing/style/unbold-in-bold-expected.txt: Copied from LayoutTests/platform/qt/editing/style/unbold-in-bold-expected.txt.
- platform/qt-5.0-wk1/editing/undo/replace-text-in-node-preserving-markers-crash-expected.txt: Copied from LayoutTests/platform/qt/editing/undo/replace-text-in-node-preserving-markers-crash-expected.txt.
- platform/qt/editing/deleting/delete-by-word-002-expected.txt:
- platform/qt/editing/deleting/delete-cell-contents-expected.txt:
- platform/qt/editing/execCommand/button-expected.txt:
- platform/qt/editing/execCommand/createLink-expected.txt:
- platform/qt/editing/execCommand/indent-pre-expected.txt:
- platform/qt/editing/execCommand/unlink-expected.txt:
- platform/qt/editing/input/emacs-ctrl-o-expected.txt:
- platform/qt/editing/inserting/insert-before-link-1-expected.txt:
- platform/qt/editing/pasteboard/display-block-on-spans-expected.txt:
- platform/qt/editing/pasteboard/emacs-cntl-y-001-expected.txt:
- platform/qt/editing/pasteboard/merge-end-list-2-expected.txt:
- platform/qt/editing/pasteboard/merge-end-table-2-expected.txt:
- platform/qt/editing/pasteboard/page-zoom-expected.txt:
- platform/qt/editing/selection/extend-selection-word-expected.txt:
- platform/qt/editing/style/apply-through-end-of-document-expected.txt:
- platform/qt/editing/style/remove-underline-across-paragraph-expected.txt:
- platform/qt/editing/style/remove-underline-across-paragraph-in-bold-expected.txt:
- platform/qt/editing/style/remove-underline-expected.txt:
- platform/qt/editing/style/remove-underline-from-stylesheet-expected.txt:
- platform/qt/editing/style/remove-underline-in-bold-expected.txt:
- platform/qt/editing/style/unbold-in-bold-expected.txt:
- platform/qt/editing/undo/replace-text-in-node-preserving-markers-crash-expected.txt:
- 2:51 AM Changeset in webkit [149098] by
-
- 4 edits2 adds in trunk
EventSource: Synchronous loader callback not handled properly
https://bugs.webkit.org/show_bug.cgi?id=115104
Source/WebCore:
When attempting a cross-origin request towards a non-HTTP URL, an early verification
of the protocol scheme will cause didFailAccessControlCheck to be called synchronously
before the loader has even finished being created. This special case was not handled
properly, since we tried to cancel a non-existing loader, which resulted in a crash.
In addition to checking whether a request is in flight before trying to cancel it,
this change also schedules the initial creation of the loader to happen asynchronously
when an EventSource is constructed, so that a script can register for the error event
before it is dispatched (as a result of passing a non-HTTP URL to the constructor).
Patch by Per-Erik Brodin <per-erik.brodin@ericsson.com> on 2013-04-25
Reviewed by Alexey Proskuryakov.
Test: http/tests/eventsource/eventsource-cors-non-http.html
- page/EventSource.cpp:
(WebCore::EventSource::EventSource):
(WebCore::EventSource::create):
(WebCore::EventSource::scheduleInitialConnect):
(WebCore):
(WebCore::EventSource::scheduleReconnect):
(WebCore::EventSource::connectTimerFired):
(WebCore::EventSource::close):
(WebCore::EventSource::abortConnectionAttempt):
- page/EventSource.h:
(EventSource):
LayoutTests:
Patch by Per-Erik Brodin <per-erik.brodin@ericsson.com> on 2013-04-25
Reviewed by Alexey Proskuryakov.
- http/tests/eventsource/eventsource-cors-non-http-expected.txt: Added.
- http/tests/eventsource/eventsource-cors-non-http.html: Added.
- 2:43 AM Changeset in webkit [149097] by
-
- 2 edits in trunk/Source/WTF
Unreviewed WinCE build fix after r148888.
- wtf/Atomics.h: WinCE version of intrin.h is called cmnintrin.h.
- 2:42 AM Changeset in webkit [149096] by
-
- 2 edits in trunk/Source/WebCore
Unreviewed WinCE build fix after r148545.
- page/wince/FrameWinCE.cpp:
- 1:01 AM Changeset in webkit [149095] by
-
- 7 edits21 adds in trunk/LayoutTests
[Qt] Unreviewed gardening. Added platform specific expected files after r148987.
- platform/qt/editing/deleting/delete-by-word-002-expected.txt: Added.
- platform/qt/editing/deleting/delete-cell-contents-expected.txt: Added.
- platform/qt/editing/execCommand/button-expected.txt: Added.
- platform/qt/editing/execCommand/createLink-expected.txt: Added.
- platform/qt/editing/execCommand/indent-pre-expected.txt: Added.
- platform/qt/editing/execCommand/unlink-expected.txt: Added.
- platform/qt/editing/input/emacs-ctrl-o-expected.png:
- platform/qt/editing/input/emacs-ctrl-o-expected.txt:
- platform/qt/editing/inserting/insert-before-link-1-expected.txt: Added.
- platform/qt/editing/pasteboard/display-block-on-spans-expected.txt: Added.
- platform/qt/editing/pasteboard/emacs-cntl-y-001-expected.png:
- platform/qt/editing/pasteboard/emacs-cntl-y-001-expected.txt:
- platform/qt/editing/pasteboard/merge-end-list-2-expected.txt: Added.
- platform/qt/editing/pasteboard/merge-end-table-2-expected.txt: Added.
- platform/qt/editing/pasteboard/page-zoom-expected.txt: Added.
- platform/qt/editing/selection/extend-selection-word-expected.txt: Added.
- platform/qt/editing/selection/selection-extend-should-not-move-across-caret-on-mac-expected.txt: Added.
- platform/qt/editing/style/apply-through-end-of-document-expected.txt: Added.
- platform/qt/editing/style/make-text-writing-direction-inline-win-expected.txt: Added.
- platform/qt/editing/style/remove-underline-across-paragraph-expected.txt: Added.
- platform/qt/editing/style/remove-underline-across-paragraph-in-bold-expected.txt: Added.
- platform/qt/editing/style/remove-underline-expected.txt: Added.
- platform/qt/editing/style/remove-underline-from-stylesheet-expected.txt: Added.
- platform/qt/editing/style/remove-underline-in-bold-expected.txt: Added.
- platform/qt/editing/style/unbold-in-bold-expected.png:
- platform/qt/editing/style/unbold-in-bold-expected.txt:
- platform/qt/editing/undo/replace-text-in-node-preserving-markers-crash-expected.txt: Added.
- 12:50 AM Changeset in webkit [149094] by
-
- 2 edits in trunk/Source/WebKit/efl
Unreviewed, rolling out r149032.
http://trac.webkit.org/changeset/149032
https://bugs.webkit.org/show_bug.cgi?id=115151
EFL WK1 API test has been broken after r149032 (Requested by
gyuyoung on #webkit).
- tests/test_ewk_setting.cpp:
(TEST_F):
