WebKit-https.git
8 years agoUse xcrun to find nm and ranlib on OS X builds
ddkilzer@apple.com [Wed, 28 Mar 2012 20:13:27 +0000 (20:13 +0000)]
Use xcrun to find nm and ranlib on OS X builds
<http://webkit.org/b/82466>

Reviewed by Dan Bernstein.

* Scripts/build-webkit: Use xcrun to find ranlib.
* Scripts/webkitdirs.pm:
(determineNmPath): Added.  Use xcrun to find nm on OS X.
(nmPath): Added.
* Scripts/webkitperl/features.pm:
(libraryContainsSymbol): Use nmPath().

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@112431 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoMore flaky tests.
dpranke@chromium.org [Wed, 28 Mar 2012 19:56:07 +0000 (19:56 +0000)]
More flaky tests.

Unreviewed, expectations update.

* platform/chromium/test_expectations.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@112429 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoWeb Inspector: CodeGeneratorInspector.py: switch Runtime, Network and DOM domains...
vsevik@chromium.org [Wed, 28 Mar 2012 19:44:57 +0000 (19:44 +0000)]
Web Inspector: CodeGeneratorInspector.py: switch Runtime, Network and DOM domains to typed API
https://bugs.webkit.org/show_bug.cgi?id=81558

Patch by Peter Rybin <peter.rybin@gmail.com> on 2012-03-28
Reviewed by Pavel Feldman.

Client code is switched to typed API (all InspectorObject and InspectorArray types are
replaced with generated types from TypeBuilder according to Inspector.json).

Missing array of int specialization is added. Code generator now has a check against
misspelled (non-existing) domain names.

* inspector/CodeGeneratorInspector.py:
(Generator.go): check against misspelled domain names added.
* inspector/InjectedScript.cpp:
(WebCore::InjectedScript::evaluate):
(WebCore::InjectedScript::callFunctionOn):
(WebCore::InjectedScript::evaluateOnCallFrame):
(WebCore::InjectedScript::getProperties):
(WebCore::InjectedScript::wrapObject):
(WebCore::InjectedScript::wrapNode):
(WebCore::InjectedScript::wrapSerializedObject):
(WebCore::InjectedScript::makeEvalCall):
* inspector/InjectedScript.h:
(InjectedScript):
* inspector/InspectorDOMAgent.cpp:
(WebCore::InspectorDOMAgent::getDocument):
(WebCore::InspectorDOMAgent::pushChildNodesToFrontend):
(WebCore::InspectorDOMAgent::querySelectorAll):
(WebCore::InspectorDOMAgent::pushNodePathToFrontend):
(WebCore::InspectorDOMAgent::getEventListenersForNode):
(WebCore::InspectorDOMAgent::getSearchResults):
(WebCore::InspectorDOMAgent::resolveNode):
(WebCore::InspectorDOMAgent::getAttributes):
(WebCore::InspectorDOMAgent::buildObjectForNode):
(WebCore::InspectorDOMAgent::buildArrayForElementAttributes):
(WebCore::InspectorDOMAgent::buildArrayForContainerChildren):
(WebCore::InspectorDOMAgent::loadEventFired):
(WebCore::InspectorDOMAgent::didInsertDOMNode):
(WebCore::InspectorDOMAgent::styleAttributeInvalidated):
(WebCore::InspectorDOMAgent::pushNodeByPathToFrontend):
* inspector/InspectorDOMAgent.h:
(InspectorDOMAgent):
* inspector/InspectorPageAgent.cpp:
(WebCore::InspectorPageAgent::resourceTypeJson):
(WebCore::InspectorPageAgent::cachedResourceTypeJson):
(WebCore::InspectorPageAgent::buildObjectForFrameTree):
* inspector/InspectorPageAgent.h:
* inspector/InspectorResourceAgent.cpp:
(WebCore::buildObjectForResourceRequest):
(WebCore::buildObjectForResourceResponse):
(WebCore::buildObjectForCachedResource):
(WebCore::InspectorResourceAgent::willSendRequest):
(WebCore::InspectorResourceAgent::didReceiveResponse):
(WebCore::InspectorResourceAgent::didLoadResourceFromMemoryCache):
(WebCore::InspectorResourceAgent::buildInitiatorObject):
(WebCore::InspectorResourceAgent::willSendWebSocketHandshakeRequest):
(WebCore::InspectorResourceAgent::didReceiveWebSocketHandshakeResponse):
* inspector/InspectorResourceAgent.h:
(InspectorResourceAgent):
* inspector/InspectorRuntimeAgent.cpp:
(WebCore::InspectorRuntimeAgent::evaluate):
(WebCore::InspectorRuntimeAgent::callFunctionOn):
(WebCore::InspectorRuntimeAgent::getProperties):
* inspector/InspectorRuntimeAgent.h:
(InspectorRuntimeAgent):
* inspector/InspectorValues.h:
(InspectorArray):
(WebCore::InspectorArray::pushInt):
(WebCore):
* inspector/ScriptCallFrame.cpp:
(WebCore::ScriptCallFrame::buildInspectorObject):
* inspector/ScriptCallFrame.h:
(ScriptCallFrame):
* inspector/ScriptCallStack.cpp:
(WebCore::ScriptCallStack::buildInspectorArray):
* inspector/ScriptCallStack.h:
(ScriptCallStack):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@112428 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agosetNeedsLayout(true, false) is super confusing to read and should use an enum instead
eric@webkit.org [Wed, 28 Mar 2012 19:25:02 +0000 (19:25 +0000)]
setNeedsLayout(true, false) is super confusing to read and should use an enum instead
https://bugs.webkit.org/show_bug.cgi?id=82369

Reviewed by Julien Chaffraix.

I replaced all uses of the markParents bool with a MarkingBehavior
enum (which has two values: MarkContainingBlockChain and MarkOnlyThis).
I'm not sure the naming is perfect (as it's not clear to me if markParents is
used to mean the same thing in these 3 functions), but hopefully this code is more
clear.  I welcome further suggested adjustment from layout experts.

* page/FrameView.cpp:
(WebCore::FrameView::scheduleRelayout):
* rendering/RenderBlock.cpp:
(WebCore::RenderBlock::layoutBlock):
(WebCore::RenderBlock::adjustPositionedBlock):
(WebCore::RenderBlock::layoutBlockChildren):
(WebCore::RenderBlock::layoutBlockChild):
(WebCore::RenderBlock::layoutPositionedObjects):
(WebCore::RenderBlock::markForPaginationRelayoutIfNeeded):
(WebCore::RenderBlock::removePositionedObjects):
(WebCore::RenderBlock::insertFloatingObject):
(WebCore::RenderBlock::positionNewFloats):
(WebCore::RenderBlock::markAllDescendantsWithFloatsForLayout):
(WebCore::RenderBlock::adjustBlockChildForPagination):
* rendering/RenderBlockLineLayout.cpp:
(WebCore::RenderBlock::layoutRunsAndFloats):
(WebCore::RenderBlock::layoutInlineChildren):
(WebCore::RenderBlock::positionNewFloatOnLine):
* rendering/RenderBox.cpp:
(WebCore::RenderBox::positionLineBox):
* rendering/RenderDeprecatedFlexibleBox.cpp:
(WebCore::RenderDeprecatedFlexibleBox::layoutBlock):
(WebCore::gatherFlexChildrenInfo):
(WebCore::RenderDeprecatedFlexibleBox::layoutHorizontalBox):
(WebCore::RenderDeprecatedFlexibleBox::layoutVerticalBox):
(WebCore::RenderDeprecatedFlexibleBox::applyLineClamp):
* rendering/RenderFlexibleBox.cpp:
(WebCore::RenderFlexibleBox::prepareChildForPositionedLayout):
* rendering/RenderLayer.cpp:
(WebCore::RenderLayer::updateScrollInfoAfterLayout):
* rendering/RenderMedia.cpp:
(WebCore::RenderMedia::layout):
* rendering/RenderObject.cpp:
(WebCore::RenderObject::setPreferredLogicalWidthsDirty):
* rendering/RenderObject.h:
(RenderObject):
(WebCore::RenderObject::setNeedsLayout):
(WebCore::RenderObject::setChildNeedsLayout):
* rendering/RenderRubyRun.cpp:
(WebCore::RenderRubyRun::layoutSpecialExcludedChild):
* rendering/RenderTable.cpp:
(WebCore::RenderTable::layout):
* rendering/RenderTableRow.cpp:
(WebCore::RenderTableRow::layout):
* rendering/RenderTableSection.cpp:
(WebCore::RenderTableSection::calcRowLogicalHeight):
(WebCore::RenderTableSection::layoutRows):
* rendering/RenderTextControl.cpp:
(WebCore::RenderTextControl::layoutSpecialExcludedChild):
* rendering/RenderTextControlSingleLine.cpp:
(WebCore::RenderTextControlSingleLine::layout):
* rendering/RenderView.cpp:
(WebCore::RenderView::RenderView):
(WebCore::RenderView::layout):
* rendering/mathml/RenderMathMLRoot.cpp:
(WebCore::RenderMathMLRoot::layout):
* rendering/mathml/RenderMathMLRow.cpp:
(WebCore::RenderMathMLRow::layout):
* rendering/mathml/RenderMathMLSubSup.cpp:
(WebCore::RenderMathMLSubSup::layout):
* rendering/svg/SVGRenderSupport.cpp:
(WebCore::SVGRenderSupport::layoutChildren):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@112425 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years ago<http://webkit.org/b/82419> Web Inspector: Create -[DOMNode inspect] from IDL
commit-queue@webkit.org [Wed, 28 Mar 2012 19:22:07 +0000 (19:22 +0000)]
<webkit.org/b/82419> Web Inspector: Create -[DOMNode inspect] from IDL

Source/WebCore:

Instead of manually creating the method in DOM.mm we can specify it in
Node.idl and autogenerate part of the interface.

Patch by Joseph Pecoraro <pecoraro@apple.com> on 2012-03-28
Reviewed by Timothy Hatcher.

* bindings/objc/DOM.mm:
* bindings/objc/DOMPrivate.h:
Move the call to InspectorController::inspect from here ...

* dom/Node.h:
* dom/Node.idl: IDL only in an ObjC block.
* dom/Node.cpp:
(WebCore::Node::inspect):
... to here.

* WebCore.xcodeproj/project.pbxproj:
Add the new generated file so we can Copy it in the build phase.

Source/WebKit/mac:

Patch by Joseph Pecoraro <pecoraro@apple.com> on 2012-03-28
Reviewed by Timothy Hatcher.

* MigrateHeaders.make: Copy DOMNodePrivate.h to PrivateHeaders.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@112424 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoAdd history navigation to garden-o-matic
ojan@chromium.org [Wed, 28 Mar 2012 19:19:36 +0000 (19:19 +0000)]
Add history navigation to garden-o-matic
https://bugs.webkit.org/show_bug.cgi?id=82495

Reviewed by Dimitri Glazkov.

Also, maintain scroll offsets when returning to a tab. This makes addressing
expected failures considerably easier when dealing with below-the-fold tests.

* BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui.js:
-Store the scrollTop before switching to a new tab and restore the appropriate scrollTop after
switching.
-Modify window.location with the tabName in the hash. Can't use pushState because this is served
from a file URL and Chrome puts each different file URL in it's own origin.

* BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui/results.js:
Remove the href so that clicking on the accordion item does not modify window.location.hash.

* BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui_unittests.js:
* BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/styles/results.css:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@112423 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years ago[chromium] MediaStream API (JSEP): Introducing WebPeerConnection00Handler
tommyw@google.com [Wed, 28 Mar 2012 18:44:06 +0000 (18:44 +0000)]
[chromium] MediaStream API (JSEP): Introducing WebPeerConnection00Handler
https://bugs.webkit.org/show_bug.cgi?id=82450

Reviewed by Adam Barth.

Source/WebCore:

This change removes the default PeerConnection00Handler.cpp from the Chromium build.

A major overhaul of the existing layout tests is forthcoming.

* WebCore.gypi:
* platform/mediastream/PeerConnection00Handler.h:
(WebCore):
(PeerConnection00Handler):

Source/WebKit/chromium:

Last patch to enable the PeerConnection00 functionality in Chromium introduces
WebPeerConnection00Handler and WebPeerConnection00HandlerClient, together with the necessary plumbing.

* WebKit.gyp:
* bridge/PeerConnection00Handler.cpp: Added.
(WebCore):
(WebCore::PeerConnection00Handler::create):
(WebCore::PeerConnection00Handler::PeerConnection00Handler):
(WebCore::PeerConnection00Handler::~PeerConnection00Handler):
(WebCore::PeerConnection00Handler::createOffer):
(WebCore::PeerConnection00Handler::createAnswer):
(WebCore::PeerConnection00Handler::setLocalDescription):
(WebCore::PeerConnection00Handler::setRemoteDescription):
(WebCore::PeerConnection00Handler::localDescription):
(WebCore::PeerConnection00Handler::remoteDescription):
(WebCore::PeerConnection00Handler::startIce):
(WebCore::PeerConnection00Handler::processIceMessage):
(WebCore::PeerConnection00Handler::addStream):
(WebCore::PeerConnection00Handler::removeStream):
(WebCore::PeerConnection00Handler::stop):
* bridge/PeerConnection00HandlerInternal.cpp: Added.
(WebCore):
(WebCore::PeerConnection00HandlerInternal::PeerConnection00HandlerInternal):
(WebCore::PeerConnection00HandlerInternal::~PeerConnection00HandlerInternal):
(WebCore::PeerConnection00HandlerInternal::createOffer):
(WebCore::PeerConnection00HandlerInternal::createAnswer):
(WebCore::PeerConnection00HandlerInternal::setLocalDescription):
(WebCore::PeerConnection00HandlerInternal::setRemoteDescription):
(WebCore::PeerConnection00HandlerInternal::localDescription):
(WebCore::PeerConnection00HandlerInternal::remoteDescription):
(WebCore::PeerConnection00HandlerInternal::startIce):
(WebCore::PeerConnection00HandlerInternal::processIceMessage):
(WebCore::PeerConnection00HandlerInternal::addStream):
(WebCore::PeerConnection00HandlerInternal::removeStream):
(WebCore::PeerConnection00HandlerInternal::stop):
(WebCore::PeerConnection00HandlerInternal::didGenerateICECandidate):
(WebCore::PeerConnection00HandlerInternal::didChangeReadyState):
(WebCore::PeerConnection00HandlerInternal::didChangeICEState):
(WebCore::PeerConnection00HandlerInternal::didAddRemoteStream):
(WebCore::PeerConnection00HandlerInternal::didRemoveRemoteStream):
* bridge/PeerConnection00HandlerInternal.h: Copied from Source/WebCore/platform/mediastream/PeerConnection00Handler.h.
(WebKit):
(WebCore):
(PeerConnection00HandlerInternal):
* public/platform/WebKitPlatformSupport.h:
(WebKit):
(WebKit::WebKitPlatformSupport::createPeerConnection00Handler):
* public/platform/WebPeerConnection00Handler.h: Added.
(WebKit):
(WebPeerConnection00Handler):
(WebKit::WebPeerConnection00Handler::~WebPeerConnection00Handler):
* public/platform/WebPeerConnection00HandlerClient.h: Added.
(WebKit):
(WebPeerConnection00HandlerClient):
(WebKit::WebPeerConnection00HandlerClient::~WebPeerConnection00HandlerClient):
* src/AssertMatchingEnums.cpp:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@112421 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years ago[REGRESSION] Web Inspector: column caption delimiters are misaligned in DataGrid
robert@webkit.org [Wed, 28 Mar 2012 18:32:28 +0000 (18:32 +0000)]
[REGRESSION] Web Inspector: column caption delimiters are misaligned in DataGrid
https://bugs.webkit.org/show_bug.cgi?id=82193

Reviewed by Yury Semikhatsky.

Since r111742 column width in fixed layout tables is set to the width of the cell plus its left padding plus its right padding
plus half its left border plus half its right border in the collapsing border model. The layout of the panel
headers in the inspector depended on pre-r111742 behaviour, so ended up misaligned with the content below. To cater for the
new method of calculating column width enforce a fixed width for the corner column by removing padding and borders (except collapsed
borders from adjacent cells) and specify the expected width (15px) minus the expected width of the collapsed borders (1px).

* inspector/front-end/dataGrid.css:
(.data-grid th.corner):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@112419 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoIncrease code sharing between CSSProperty and CSSPropertyLonghand.
alexis.menard@openbossa.org [Wed, 28 Mar 2012 18:29:54 +0000 (18:29 +0000)]
Increase code sharing between CSSProperty and CSSPropertyLonghand.
https://bugs.webkit.org/show_bug.cgi?id=82479

Reviewed by Antti Koivisto.

Use longhands declarations from CSSPropertyLonghand in CSSProperty to avoid
code duplication.

No new tests : refactoring only, we shouldn't have any behavior difference.

* css/CSSProperty.cpp:
(WebCore::resolveToPhysicalProperty):
(WebCore::borderDirections):
(WebCore):
(WebCore::CSSProperty::resolveDirectionAwareProperty):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@112418 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years ago[chromium] Compositor visibility setting must be updated even if not actively compositing
commit-queue@webkit.org [Wed, 28 Mar 2012 18:24:01 +0000 (18:24 +0000)]
[chromium] Compositor visibility setting must be updated even if not actively compositing
https://bugs.webkit.org/show_bug.cgi?id=82406

Patch by James Robinson <jamesr@chromium.org> on 2012-03-28
Reviewed by Adrienne Walker.

Propagate the visibility bit to the WebLayerTreeView even when compositing is inactive.

* src/WebViewImpl.cpp:
(WebKit::WebViewImpl::setVisibilityState):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@112417 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoWeb Inspector: Implement snippets renaming in scripts navigator.
vsevik@chromium.org [Wed, 28 Mar 2012 18:23:00 +0000 (18:23 +0000)]
Web Inspector: Implement snippets renaming in scripts navigator.
https://bugs.webkit.org/show_bug.cgi?id=82477

Reviewed by Pavel Feldman.

Snippets renaming is implemented as editing of snippet name in scripts navigator.

* inspector/front-end/ScriptsNavigator.js:
(WebInspector.ScriptsNavigator.prototype._handleRenameSnippet):
(WebInspector.ScriptsNavigator.prototype._fileRenamed):
(WebInspector.ScriptsNavigator.prototype.rename.commitHandler):
(WebInspector.ScriptsNavigator.prototype.rename.cancelHandler):
(WebInspector.ScriptsNavigator.prototype.rename.afterEditing):
(WebInspector.BaseNavigatorTreeElement.prototype.set titleText):
* inspector/front-end/scriptsPanel.css:
(#scripts-navigator-tabbed-pane .navigator .base-navigator-tree-element-title.editing):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@112416 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoREGRESSION (r109826): Can't type into Flash text fields
andersca@apple.com [Wed, 28 Mar 2012 18:20:59 +0000 (18:20 +0000)]
REGRESSION (r109826): Can't type into Flash text fields
https://bugs.webkit.org/show_bug.cgi?id=82488
<rdar://problem/11022004>

Reviewed by Sam Weinig.

This was caused by the fix in r109826. Revert that change and fix the original bug by notifying the UI process
that the plug-in lost focus when it's destroyed. This will correctly reset the text input state (merely setting
the text input state to PluginComplexTextInputDisabled doesn't reset the state correctly).

* UIProcess/API/mac/WKView.mm:
(-[WKView _setPluginComplexTextInputState:]):
(-[WKView _handlePluginComplexTextInputKeyDown:]):
* WebProcess/Plugins/PluginView.cpp:
(WebKit::PluginView::~PluginView):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@112415 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoWeb Inspector: Add context menu for snippets control.
vsevik@chromium.org [Wed, 28 Mar 2012 18:13:47 +0000 (18:13 +0000)]
Web Inspector: Add context menu for snippets control.
https://bugs.webkit.org/show_bug.cgi?id=82475

Reviewed by Pavel Feldman.

This is a preliminary implementation of snippets control using context menu.
We might want to add a more discoverable one before taking snippets out of experiments.

* English.lproj/localizedStrings.js:
* inspector/front-end/ScriptsNavigator.js:
(WebInspector.ScriptsNavigator.prototype._showScriptFoldersSettingChanged):
(WebInspector.ScriptsNavigator.prototype._createSnippetsTree):
(WebInspector.ScriptsNavigator.prototype._handleSnippetContextMenuEvent):
(WebInspector.ScriptsNavigator.prototype._showSnippetContextMenu):
(WebInspector.ScriptsNavigator.prototype._handleEvaluateSnippet):
(WebInspector.ScriptsNavigator.prototype._handleRenameSnippet):
(WebInspector.ScriptsNavigator.prototype._handleRemoveSnippet):
(WebInspector.ScriptsNavigator.prototype._handleCreateSnippet):
(WebInspector.NavigatorScriptTreeElement.prototype.get navigator):
(WebInspector.NavigatorScriptTreeElement.prototype.onattach):
(WebInspector.NavigatorScriptTreeElement.prototype.onenter):
(WebInspector.NavigatorScriptTreeElement.prototype._handleContextMenuEvent):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@112414 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoWeb Inspector: ScriptsNavigator should save scroll position when switching tabs.
vsevik@chromium.org [Wed, 28 Mar 2012 18:08:04 +0000 (18:08 +0000)]
Web Inspector: ScriptsNavigator should save scroll position when switching tabs.
https://bugs.webkit.org/show_bug.cgi?id=82472

Reviewed by Pavel Feldman.

WebInspector.View provides save/restore scroll position capabilities.
This patch makes them used in ScriptsNavigator.

* inspector/front-end/ScriptsNavigator.js:
(WebInspector.ScriptsNavigator):
(WebInspector.ScriptsNavigator.prototype._createSnippetsTree):
* inspector/front-end/scriptsPanel.css:
(#scripts-navigator-tabbed-pane .tabbed-pane-content):
(#scripts-navigator-tabbed-pane .navigator-container):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@112413 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoUnreviewed, skipping 3 tests timing out after the libsoup upgrade
sergio@webkit.org [Wed, 28 Mar 2012 18:00:57 +0000 (18:00 +0000)]
Unreviewed, skipping 3 tests timing out after the libsoup upgrade
to 2.37.92.

* platform/gtk/Skipped:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@112412 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years ago[Chromium] Enable Viewport define by default
fsamuel@chromium.org [Wed, 28 Mar 2012 17:55:04 +0000 (17:55 +0000)]
[Chromium] Enable Viewport define by default
https://bugs.webkit.org/show_bug.cgi?id=79514

Reviewed by Darin Fisher.

* features.gypi:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@112411 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoWeb Inspector: only update Timeline overview when really needed
caseq@chromium.org [Wed, 28 Mar 2012 17:17:02 +0000 (17:17 +0000)]
Web Inspector: only update Timeline overview when really needed
https://bugs.webkit.org/show_bug.cgi?id=81909

Reviewed by Pavel Feldman.

Switch timeline overview to using raw records, thus removing dependency on instance members of
TimelinePresentationModel. This requires moving record styles/categories to static getters.

* inspector/front-end/TimelineModel.js:
(WebInspector.TimelineModel):
(WebInspector.TimelineModel.startTime):
(WebInspector.TimelineModel.endTime):
(WebInspector.TimelineModel.duration):
(WebInspector.TimelineModel.prototype._addRecord):
(WebInspector.TimelineModel.prototype.reset):
(WebInspector.TimelineModel.prototype.minimumRecordTime):
(WebInspector.TimelineModel.prototype.maximumRecordTime):
(WebInspector.TimelineModel.prototype._updateBoundaries):
* inspector/front-end/TimelineOverviewPane.js:
(WebInspector.TimelineOverviewPane):
(WebInspector.TimelineOverviewPane.prototype._showTimelines):
(WebInspector.TimelineOverviewPane.prototype._showMemoryGraph):
(WebInspector.TimelineOverviewPane.prototype._setVerticalOverview):
(WebInspector.TimelineOverviewPane.prototype._onCategoryVisibilityChanged):
(WebInspector.TimelineOverviewPane.prototype._update):
(WebInspector.TimelineOverviewPane.prototype._updateCategoryStrips.markPercentagesForRecord):
(WebInspector.TimelineOverviewPane.prototype._updateCategoryStrips):
(WebInspector.TimelineOverviewPane.prototype.windowStartTime):
(WebInspector.TimelineOverviewPane.prototype.windowEndTime):
(WebInspector.TimelineOverviewPane.prototype._onWindowChanged):
(WebInspector.TimelineOverviewPane.prototype.setShowShortEvents):
(WebInspector.TimelineOverviewPane.prototype._scheduleRefresh):
(WebInspector.TimelineOverviewCalculator.prototype.computeBarGraphPercentages):
(WebInspector.TimelineOverviewCalculator.prototype.setWindow):
(WebInspector.TimelineOverviewCalculator.prototype.reset):
(WebInspector.HeapGraph):
(WebInspector.HeapGraph.prototype.update):
(WebInspector.TimelineVerticalOverview):
(WebInspector.TimelineVerticalOverview.prototype.reset):
(WebInspector.TimelineVerticalOverview.prototype.update):
(WebInspector.TimelineVerticalOverview.prototype.wasShown):
(WebInspector.TimelineVerticalOverview.prototype.willHide):
(WebInspector.TimelineVerticalOverview.prototype._aggregateFrameStatistics):
(WebInspector.TimelineVerticalOverview.prototype._aggregateRecords):
(WebInspector.TimelineVerticalOverview.prototype.getWindowTimes):
(WebInspector.TimelineVerticalOverview.prototype._onRecordAdded):
(WebInspector.TimelineAggregatedRecord):
(WebInspector.TimelineAggregatedRecord.prototype._aggregateStatistics):
* inspector/front-end/TimelinePanel.js:
(WebInspector.TimelinePanel):
(WebInspector.TimelinePanel.prototype._createStatusbarButtons):
(WebInspector.TimelinePanel.prototype._toggleFilterButtonClicked):
(WebInspector.TimelinePanel.prototype._resetPanel):
(WebInspector.TimelinePanel.prototype._refresh):
(WebInspector.TimelineCalculator):
(WebInspector.TimelineCalculator.prototype.formatTime):
* inspector/front-end/TimelinePresentationModel.js:
(WebInspector.TimelinePresentationModel):
(WebInspector.TimelinePresentationModel.categories):
(WebInspector.TimelinePresentationModel.recordStyle):
(WebInspector.TimelinePresentationModel.categoryForRecord):
(WebInspector.TimelinePresentationModel.prototype.reset):
(WebInspector.TimelinePresentationModel.prototype.addRecord):
(WebInspector.TimelinePresentationModel.Record):
(WebInspector.TimelinePresentationModel.Record.prototype.generatePopupContent):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@112409 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years ago[GTK] generate-gtk-doc doesn't cope with custom build directory
philn@webkit.org [Wed, 28 Mar 2012 16:33:03 +0000 (16:33 +0000)]
[GTK] generate-gtk-doc doesn't cope with custom build directory
https://bugs.webkit.org/show_bug.cgi?id=82448

Reviewed by Martin Robinson.

Pass the product directory in generate-gtkdoc arguments from
build-webkit.

* Scripts/webkitdirs.pm:
(buildAutotoolsProject):
* gtk/common.py:
(get_build_path):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@112408 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years ago[BlackBerry] Plumb through GraphicsLayer::contentsOpaque() to LayerTiler
commit-queue@webkit.org [Wed, 28 Mar 2012 16:03:29 +0000 (16:03 +0000)]
[BlackBerry] Plumb through GraphicsLayer::contentsOpaque() to LayerTiler
https://bugs.webkit.org/show_bug.cgi?id=82457

Patch by Arvid Nilsson <anilsson@rim.com> on 2012-03-28
Reviewed by Rob Buis.

The LayerTiler already knows not to turn on GL_BLEND for opaque layers.
However, it only ever sets the opaque flag for image layers and color
layers, never for content layers.

This was no big deal, because contentsOpaque() is currently false for
all layers except the root layer, which we always drew using the
BlackBerry::WebKit::BackingStore anyway.

When we start using RenderLayerBacking::m_usingTiledCacheLayer=true on
the root layer in situations where the BackingStore is unavailable, we
can speed up rendering of the root layer by honouring the opaque flag.

Fixed by plumbing through the GraphicsLayer::contentsOpaque() flag all
the way to LayerTiler and on to Texture.

* platform/graphics/blackberry/LayerData.h:
(WebCore::LayerData::LayerData):
(WebCore::LayerData::isOpaque):
(LayerData):
* platform/graphics/blackberry/LayerTile.cpp:
(WebCore::LayerTile::updateContents):
* platform/graphics/blackberry/LayerTile.h:
(LayerTile):
* platform/graphics/blackberry/LayerTiler.cpp:
(WebCore::LayerTiler::updateTextureContentsIfNeeded):
(WebCore::LayerTiler::performTileJob):
* platform/graphics/blackberry/LayerTiler.h:
(WebCore::LayerTiler::TextureJob::TextureJob):
(WebCore::LayerTiler::TextureJob::updateContents):
* platform/graphics/blackberry/LayerWebKitThread.h:
(WebCore::LayerWebKitThread::setOpaque):
* platform/graphics/blackberry/Texture.h:
(Texture):
* platform/graphics/blackberry/TextureCacheCompositingThread.cpp:
(WebCore::TextureCacheCompositingThread::updateContents):
* platform/graphics/blackberry/TextureCacheCompositingThread.h:
(TextureCacheCompositingThread):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@112407 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years ago[GTK] Add --skipped command line option to Tools/Scripts/run-gtk-tests
carlosgc@webkit.org [Wed, 28 Mar 2012 15:53:47 +0000 (15:53 +0000)]
[GTK] Add --skipped command line option to Tools/Scripts/run-gtk-tests
https://bugs.webkit.org/show_bug.cgi?id=82341

Reviewed by Martin Robinson.

Add --skipped=skip|ignore|only command line option.

* Scripts/run-gtk-tests:
(TestRunner._test_cases_to_skip): Return the list of test cases
to skip for the given test depending on the skipped action.
(TestRunner._should_run_test): Decide whether to run the test or
not depending on the skipped action.
(TestRunner.run_tests): Only show the list of skipped tests when
skipped action is 'skip'.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@112406 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoNot reviewed: updated chromium test expectations.
pfeldman@chromium.org [Wed, 28 Mar 2012 15:49:13 +0000 (15:49 +0000)]
Not reviewed: updated chromium test expectations.

* platform/chromium/test_expectations.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@112405 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years ago[GTK] Use gtester -s to skip individual test cases instead of unit tests as a whole
carlosgc@webkit.org [Wed, 28 Mar 2012 15:47:15 +0000 (15:47 +0000)]
[GTK] Use gtester -s to skip individual test cases instead of unit tests as a whole
https://bugs.webkit.org/show_bug.cgi?id=82333

Reviewed by Martin Robinson.

* Scripts/run-gtk-tests:
(TestRunner.SkippedTest): Add SkippedTest class to store more
information about skipped tests.
(TestRunner._setup_testing_environment): Use SkippedTest to add
TestWebKitAccessibility to the skipped list.
(TestRunner._find_skipped_test): Return the SkippedTest object for
a given test path.
(TestRunner._test_cases_to_skip): Returns the list of test cases
that should be skipped for a given test path.
(TestRunner._should_run_test): Helper funtion to decide whether
the given test path should be run or not. Tests in the skipped
list that don't have a list of failing test cases are not run at
all. Tests in the skipped list that contain a list of test cases
are run with gtester -s to skip the individual test cases.
(TestRunner._run_test): Helper funtion to run a given test.
(TestRunner.run_tests): Build the list of tests to run based on
_should_run_test() and use helper function _run_test() to run
every test.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@112404 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoWeb Inspector: nuke HeapGraph.setSize(), do it automagically on update
caseq@chromium.org [Wed, 28 Mar 2012 15:36:03 +0000 (15:36 +0000)]
Web Inspector: nuke HeapGraph.setSize(), do it automagically on update
https://bugs.webkit.org/show_bug.cgi?id=82471

Reviewed by Yury Semikhatsky.

* inspector/front-end/TimelineOverviewPane.js:
(WebInspector.TimelineOverviewPane.prototype.update):
(WebInspector.HeapGraph.prototype.update):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@112403 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoWeb Inspector: dispatch console-message-added and console-messages-cleared events...
podivilov@chromium.org [Wed, 28 Mar 2012 15:20:06 +0000 (15:20 +0000)]
Web Inspector: dispatch console-message-added and console-messages-cleared events on UISourceCode.
https://bugs.webkit.org/show_bug.cgi?id=82463

Reviewed by Vsevolod Vlasov.

Source/WebCore:

Those events are related to specific UISourceCode. See bug 82224 for more details.

* inspector/front-end/DebuggerPresentationModel.js:
(WebInspector.DebuggerPresentationModel):
(WebInspector.DebuggerPresentationModel.prototype._addConsoleMessageToScript):
(WebInspector.DebuggerPresentationModel.prototype._consoleCleared):
(WebInspector.DebuggerPresentationModel.prototype._debuggerReset):
(WebInspector.UISourceCodeImpl):
(WebInspector.UISourceCodeImpl.prototype.consoleMessages):
(WebInspector.UISourceCodeImpl.prototype.consoleMessageAdded):
(WebInspector.UISourceCodeImpl.prototype.consoleMessagesCleared):
* inspector/front-end/JavaScriptSourceFrame.js:
(WebInspector.JavaScriptSourceFrame):
(WebInspector.JavaScriptSourceFrame.prototype.beforeTextChanged):
(WebInspector.JavaScriptSourceFrame.prototype._consoleMessageAdded):
(WebInspector.JavaScriptSourceFrame.prototype._consoleMessagesCleared):
(WebInspector.JavaScriptSourceFrame.prototype._onTextViewerContentLoaded):
* inspector/front-end/ScriptsPanel.js:
(WebInspector.ScriptsPanel.prototype._sourceFrameLoaded):
* inspector/front-end/UISourceCode.js:
(WebInspector.UISourceCode.prototype.breakpoints):
(WebInspector.UISourceCode.prototype.consoleMessages):

LayoutTests:

* inspector/debugger/scripts-panel.html:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@112402 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years ago[GTK] Implement PlatformStrategies
carlosgc@webkit.org [Wed, 28 Mar 2012 15:13:51 +0000 (15:13 +0000)]
[GTK] Implement PlatformStrategies
https://bugs.webkit.org/show_bug.cgi?id=82454

Reviewed by Xan Lopez.

Source/WebCore:

* GNUmakefile.list.am: Remove PluginDataGtk.cpp. The same
functionality is now implemented using platform strategies.
* plugins/gtk/PluginDataGtk.cpp: Removed.

Source/WebKit/gtk:

* GNUmakefile.am: Add new files to compilation.
* WebCoreSupport/PlatformStrategiesGtk.cpp: Added.
(PlatformStrategiesGtk::initialize): Initialize platform
strategies.
(PlatformStrategiesGtk::PlatformStrategiesGtk):
(PlatformStrategiesGtk::createCookiesStrategy): Return this.
(PlatformStrategiesGtk::createPluginStrategy): Ditto.
(PlatformStrategiesGtk::createVisitedLinkStrategy): Ditto.
(PlatformStrategiesGtk::createPasteboardStrategy): Return 0, since
PasteboardStrategy is only used by mac code for now.
(PlatformStrategiesGtk::notifyCookiesChanged):
(PlatformStrategiesGtk::refreshPlugins): Refresh the plugin
database.
(PlatformStrategiesGtk::getPluginInfo): Initialize plugins and
populate the given vector with plugins information.
(PlatformStrategiesGtk::isLinkVisited): Return whether the given
hash is a visited link of the page group.
(PlatformStrategiesGtk::addVisitedLink): Add the given hash to the
page group visited links.
* WebCoreSupport/PlatformStrategiesGtk.h: Added.
* webkit/webkitglobals.cpp:
(webkitInit): Initialize PlatformStrategiesGtk.

Source/WTF:

* wtf/Platform.h: Define WTF_USE_PLATFORM_STRATEGIES for GTK+
platform too.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@112401 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoWeb Inspector: [Regression] TabIndex is not correctly restored after editing.
vsevik@chromium.org [Wed, 28 Mar 2012 15:07:04 +0000 (15:07 +0000)]
Web Inspector: [Regression] TabIndex is not correctly restored after editing.
https://bugs.webkit.org/show_bug.cgi?id=82464

Reviewed by Pavel Feldman.

This fixes incorrect usage of isNaN that returns false for null value.
Drive-by javascript compilation fix.

* inspector/front-end/CompilerScriptMapping.js:
(WebInspector.CompilerScriptMapping.prototype.addScript): compilation fix.
* inspector/front-end/ResourceScriptMapping.js:
(WebInspector.ResourceScriptMapping.prototype._uiSourceCodeChanged): compilation fix.
* inspector/front-end/SnippetsModel.js:
(WebInspector.SnippetsScriptMapping.prototype._createUISourceCodeForScript): compilation fix.
* inspector/front-end/UIUtils.js:
(WebInspector.startEditing.cleanUpAfterEditing): incorrect usage of isNaN fix.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@112400 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoAn extra line break is inserted when pasting into a font element.
yi.4.shen@nokia.com [Wed, 28 Mar 2012 15:00:42 +0000 (15:00 +0000)]
An extra line break is inserted when pasting into a font element.
https://bugs.webkit.org/show_bug.cgi?id=71207

Reviewed by Ryosuke Niwa.

Fix an editing bug where inserting text into a font element would
create an extra div element in the dom tree. The
WebCore::positionAvoidingPrecedingNodes() tries to set the correct
destination position by checking the next visible position, however,
it causes the position moves into the child element in somecase.
Instead, we should only check the position in parent after node.

Source/WebCore:

Test: editing/inserting/insert-text-into-font.html

* editing/ReplaceSelectionCommand.cpp:
(WebCore::positionAvoidingPrecedingNodes):

LayoutTests:

* editing/inserting/insert-text-into-font-expected.txt: Added.
* editing/inserting/insert-text-into-font.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@112399 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years ago[CSS Regions] Add region styling tests for interaction between region styling and...
commit-queue@webkit.org [Wed, 28 Mar 2012 14:56:51 +0000 (14:56 +0000)]
[CSS Regions] Add region styling tests for interaction between region styling and scoped styles
https://bugs.webkit.org/show_bug.cgi?id=82340

Patch by Mihai Balan <mibalan@adobe.com> on 2012-03-28
Reviewed by Antti Koivisto.

Adding tests for style interaction between region styling snd scoped style elements. Currently skipped,
since scoped style elements are not enabled by default.

* fast/regions/style-scoped-in-flow-expected.html: Added.
* fast/regions/style-scoped-in-flow-override-container-style-expected.html: Added.
* fast/regions/style-scoped-in-flow-override-container-style.html: Added.
* fast/regions/style-scoped-in-flow-override-region-styling-expected.html: Added.
* fast/regions/style-scoped-in-flow-override-region-styling-multiple-regions-expected.html: Added.
* fast/regions/style-scoped-in-flow-override-region-styling-multiple-regions.html: Added.
* fast/regions/style-scoped-in-flow-override-region-styling.html: Added.
* fast/regions/style-scoped-in-flow.html: Added.
* platform/efl/Skipped:
* platform/gtk/Skipped:
* platform/mac/Skipped:
* platform/qt/Skipped:
* platform/win/Skipped:
* platform/wincairo/Skipped:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@112398 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years ago[Qt] Couple of tests have different results on 64 bit and/or in debug mode compared...
ossy@webkit.org [Wed, 28 Mar 2012 14:54:03 +0000 (14:54 +0000)]
[Qt] Couple of tests have different results on 64 bit and/or in debug mode compared to 32 bit and/or release mode
https://bugs.webkit.org/show_bug.cgi?id=52810

Reviewed by Simon Hausmann.

Use SSE2 floating point math on 32 bit instead of the default
387 to make layout test result same on 32 and on 64 bit builds.

* qmake/mkspecs/features/unix/default_post.prf:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@112397 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years ago[Soup] DNS prefetching spams resolver, shoots self in the foot
sergio@webkit.org [Wed, 28 Mar 2012 14:45:45 +0000 (14:45 +0000)]
[Soup] DNS prefetching spams resolver, shoots self in the foot
https://bugs.webkit.org/show_bug.cgi?id=41630

Reviewed by Martin Robinson.

.:

Bump libsoup and glib dependencies.

* Source/cmake/OptionsEfl.cmake:
* configure.ac:

Source/WebCore:

Added generic DNSResolveQueue class to throttle DNS
prefetches. It's an abstract refactoring of CFNET's
DNSResolveQueue. Platform specific methods implemented for soup
and CFNET backends.

No new tests required as we're just refactoring existing code to
be used by two different ports.

* CMakeLists.txt: added new file.
* GNUmakefile.list.am: ditto.
* WebCore.vcproj/WebCore.vcproj: ditto.
* WebCore.xcodeproj/project.pbxproj: ditto.
* platform/network/DNSResolveQueue.cpp: Added.
(WebCore):
(WebCore::DNSResolveQueue::add): adds a new host to be prefetched.
(WebCore::DNSResolveQueue::fired): by using a delay we coalesce
several prefetch requests and try to resolve them all here.
* platform/network/DNSResolveQueue.h: Added.
(WebCore):
(DNSResolveQueue): class that implements DNS prefetch
throttling using a template pattern.
(WebCore::DNSResolveQueue::shared):
(WebCore::DNSResolveQueue::decrementRequestCount):
* platform/network/cf/DNSCFNet.cpp:
(WebCore::DNSResolveQueue::platformProxyIsEnabledInSystemPreferences):
(WebCore::DNSResolveQueue::platformResolve):
* platform/network/soup/DNSSoup.cpp:
(WebCore):
(WebCore::DNSResolveQueue::platformProxyIsEnabledInSystemPreferences):
(WebCore::resolvedCallback):
(WebCore::DNSResolveQueue::platformResolve):
(WebCore::prefetchDNS):

Tools:

Bump libsoup and glib dependencies.

* efl/jhbuild.modules:
* gtk/jhbuild.modules:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@112396 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agowindow.scrollBy() scrolls incorrectly when zoomed in/out
commit-queue@webkit.org [Wed, 28 Mar 2012 14:37:56 +0000 (14:37 +0000)]
window.scrollBy() scrolls incorrectly when zoomed in/out
https://bugs.webkit.org/show_bug.cgi?id=45228

Patch by Eugene Girard <girard@chromium.org> on 2012-03-28
Reviewed by Simon Fraser.

scrollBy now uses layout (not CSS) units. (As scrollTo does.)

Source/WebCore:

Test: fast/dom/zoom-scroll-page-test.html

* page/DOMWindow.cpp:
(WebCore::DOMWindow::scrollBy):

LayoutTests:

* fast/dom/zoom-scroll-page-test-expected.txt: Added.
* fast/dom/zoom-scroll-page-test.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@112395 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoNull dereference in SVGTextElement::animatedLocalTransform()
schenney@chromium.org [Wed, 28 Mar 2012 14:28:02 +0000 (14:28 +0000)]
Null dereference in SVGTextElement::animatedLocalTransform()
https://bugs.webkit.org/show_bug.cgi?id=82375

Reviewed by Nikolas Zimmermann.

Source/WebCore:

A poorly structured SVG file with a text element inside an
animatedTransform element leaves the text element placed outside of
the SVG root and it has no renderer(). Later attempts to use the
renderer in animatedLocalTransform fail.

Test: svg/custom/get-text-element-transform-crash.html

* svg/SVGTextElement.cpp:
(WebCore::SVGTextElement::animatedLocalTransform): Added a check for null renderer, and do not check
for style transform if there is no renderer.

LayoutTests:

* svg/custom/get-text-element-transform-crash-expected.txt: Added.
* svg/custom/get-text-element-transform-crash.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@112394 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years ago[GTK] Respect NOCONFIGURE option in autogen.sh
xan@webkit.org [Wed, 28 Mar 2012 14:19:14 +0000 (14:19 +0000)]
[GTK] Respect NOCONFIGURE option in autogen.sh
https://bugs.webkit.org/show_bug.cgi?id=82447

Reviewed by Philippe Normand.

Make autogen.sh follow the new build-api for GNOME, see
http://people.gnome.org/~walters/docs/build-api.txt

* autogen.sh: respect the NOCONFIGURE environment flag, per the
new GNOME build-api.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@112393 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years ago[GTK] Disable introspection for jhbuild modules
kov@webkit.org [Wed, 28 Mar 2012 14:11:49 +0000 (14:11 +0000)]
[GTK] Disable introspection for jhbuild modules
https://bugs.webkit.org/show_bug.cgi?id=82395

Reviewed by Philippe Normand.

* efl/jhbuild.modules: add --disable-introspection to all modules
that support it
* gtk/jhbuild.modules: ditto.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@112392 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoEnable animVal support for SVGAnimatedPreserveAspectRatio
zimmermann@webkit.org [Wed, 28 Mar 2012 13:14:20 +0000 (13:14 +0000)]
Enable animVal support for SVGAnimatedPreserveAspectRatio
https://bugs.webkit.org/show_bug.cgi?id=82326

Reviewed by Zoltan Herczeg.

Source/WebCore:

Enable animVal support for SVGAnimatedPreserveAspectRatio. Very simple now that everything is prepared.
All we have to do is add startAnimValAnimation/etc. methods to SVGAnimatedPreserveAspectRatioAnimator.

Modernize SVGPreserveAspectRatio, align parsing this object from String, with the other SVG DOM primitives,
who supply a "void parse(const String&)" method, for this task.

Extended existing tests to cover this.

* svg/SVGAnimatedPreserveAspectRatio.cpp:
(WebCore::SVGAnimatedPreserveAspectRatioAnimator::constructFromString):
(WebCore::SVGAnimatedPreserveAspectRatioAnimator::startAnimValAnimation):
(WebCore::SVGAnimatedPreserveAspectRatioAnimator::stopAnimValAnimation):
(WebCore::SVGAnimatedPreserveAspectRatioAnimator::resetAnimValToBaseVal):
(WebCore::SVGAnimatedPreserveAspectRatioAnimator::animValWillChange):
(WebCore::SVGAnimatedPreserveAspectRatioAnimator::animValDidChange):
* svg/SVGAnimatedPreserveAspectRatio.h:
(SVGAnimatedPreserveAspectRatioAnimator):
* svg/SVGAnimatedType.cpp:
(WebCore::SVGAnimatedType::valueAsString):
(WebCore::SVGAnimatedType::setValueAsString):
(WebCore::SVGAnimatedType::supportsAnimVal):
* svg/SVGFEImageElement.cpp:
(WebCore::SVGFEImageElement::parseAttribute):
* svg/SVGFitToViewBox.cpp:
(WebCore::SVGFitToViewBox::parseAttribute):
* svg/SVGImageElement.cpp:
(WebCore::SVGImageElement::parseAttribute):
* svg/SVGPreserveAspectRatio.cpp:
(WebCore::SVGPreserveAspectRatio::parse):
(WebCore):
* svg/SVGPreserveAspectRatio.h:
(SVGPreserveAspectRatio):
* svg/SVGViewSpec.cpp:
(WebCore::SVGViewSpec::setPreserveAspectRatioString):
(WebCore::SVGViewSpec::parseViewSpec):

LayoutTests:

Extend SVGAnimatedPreserveAspectRatio tests to cover animVal support.

* svg/animations/script-tests/svgPreserveAspectRatio-animation-1.js:
(sample1):
(sample2):
(sample3):
* svg/animations/svgPreserveAspectRatio-animation-1-expected.txt:
* svg/dom/preserve-aspect-ratio-parser-expected.txt:
* svg/dom/preserve-aspect-ratio-parser.html: Parsing errors now yield to 'unknown' pAR values, as desired.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@112391 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoWeb Inspector: [v8] inspection of CanvasPixelArray is slow.
pfeldman@chromium.org [Wed, 28 Mar 2012 13:05:59 +0000 (13:05 +0000)]
Web Inspector: [v8] inspection of CanvasPixelArray is slow.
https://bugs.webkit.org/show_bug.cgi?id=82455

Reviewed by Vsevolod Vlasov.

Added CanvasPixelArray to the list of classes that we consider
arrays.

* bindings/v8/custom/V8InjectedScriptHostCustom.cpp:
(WebCore::V8InjectedScriptHost::typeCallback):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@112390 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoSelectively retrieve text content around a given position.
leandrogracia@chromium.org [Wed, 28 Mar 2012 13:02:55 +0000 (13:02 +0000)]
Selectively retrieve text content around a given position.
https://bugs.webkit.org/show_bug.cgi?id=78361

Reviewed by Ryosuke Niwa.

Tests: a new layout test will be introduced once the required WebKit API code is ready.

Introduce the DOMTextContentWalker class to selectively retrieve content
around a given VisiblePosition into a string in such a way that a text
Range can be extracted later from a pair of the string's positions.

A new boolean called m_shouldStop is also introduced in the TextIterator to force
it advancing under certain conditions. Although this is usually done by setting
m_positionNode to zero, it can't be applied in this case since by doing so the
iterator's range is lost.

* WebCore.gypi:
* editing/SurroundingText.cpp: Added.
(WebCore):
(WebCore::SurroundingText::SurroundingText):
(WebCore::SurroundingText::rangeFromContentOffsets):
(WebCore::SurroundingText::content):
(WebCore::SurroundingText::positionOffsetInContent):
* editing/SurroundingText.h: Added.
(WebCore):
(SurroundingText):
* editing/TextIterator.cpp:
(WebCore::TextIterator::TextIterator):
(WebCore::TextIterator::advance):
(WebCore::SimplifiedBackwardsTextIterator::SimplifiedBackwardsTextIterator):
(WebCore::SimplifiedBackwardsTextIterator::advance):
* editing/TextIterator.h:
(WebCore::TextIterator::atEnd):
(TextIterator):
(WebCore::SimplifiedBackwardsTextIterator::atEnd):
(SimplifiedBackwardsTextIterator):
* html/HTMLFormControlElement.cpp:
(WebCore::HTMLFormControlElement::enclosingFormControlElement):
(WebCore):
* html/HTMLFormControlElement.h:
(HTMLFormControlElement):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@112389 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years ago[Qt] Unreviewed, fix API tests after adding full platforms. (r111658)
ossy@webkit.org [Wed, 28 Mar 2012 13:02:08 +0000 (13:02 +0000)]
[Qt] Unreviewed, fix API tests after adding full platforms. (r111658)

* BuildSlaveSupport/build.webkit.org-config/master.cfg:
(BuildAndTestFactory.__init__):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@112388 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoSpeed up updates of existing CSS properties from JS.
alexis.menard@openbossa.org [Wed, 28 Mar 2012 12:40:31 +0000 (12:40 +0000)]
Speed up updates of existing CSS properties from JS.
https://bugs.webkit.org/show_bug.cgi?id=82235

Reviewed by Andreas Kling.

Source/WebCore:

Improve the way we handle updating an existing CSS property by replacing its value
by the new one rather than removing the old value and then adding the new one. This
speed up by 35% PerformanceTests/CSS/CSSPropertyUpdateValue.html.

No new tests : Updating existing tests that were relying on the order of the properties
after modification. Each updated property is not appended to the list of properties of the
style therefore the cssText value is slightly different. It matches Firefox behavior but not
Opera.

* css/StylePropertySet.cpp:
(WebCore::StylePropertySet::addParsedProperty):

LayoutTests:

Updated expectations after behavior change on how we handle updates of
CSS properties from JS. We match now Firefox where the cssText keeps the
order of the properties (but modify its value) whereas before we were appending
the new value at the end of the list of properties.

* editing/pasteboard/data-transfer-items-expected.txt:
* editing/pasteboard/onpaste-text-html-expected.txt:
* fast/dom/css-set-property-exception-expected.txt:
* fast/events/ondrop-text-html-expected.txt:
* fast/mutation/observe-attributes-expected.txt:
* fast/mutation/observe-attributes.html:
* inspector/styles/styles-new-API-expected.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@112387 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years ago[Qt] Unreviewed typo fix after r112383.
ossy@webkit.org [Wed, 28 Mar 2012 12:39:40 +0000 (12:39 +0000)]
[Qt] Unreviewed typo fix after r112383.

* Scripts/run-qtwebkit-tests:
(Main.check_crash_occurences):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@112386 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years ago[BlackBerry] Use WebCore::PlatformMouseEvent::globalPosition() instead of WebCore...
leo.yang@torchmobile.com.cn [Wed, 28 Mar 2012 12:31:20 +0000 (12:31 +0000)]
[BlackBerry] Use WebCore::PlatformMouseEvent::globalPosition() instead of WebCore::PlatformMouseEvent::globalPos() in WebPage.cpp
https://bugs.webkit.org/show_bug.cgi?id=82420

Reviewed by Rob Buis.

In PlatformMouseEvent.h the blackberry porting added globalPos() (but not upstreamed yet)
for PlatformMouseEvent, but this method is same as globalPosition() that is available for
every porting. We should use globalPosition().

* Api/WebPage.cpp:
(BlackBerry::WebKit::WebPagePrivate::updateCursor):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@112385 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years ago[BlackBerry] Remove PlatformMouseEvent::setClickCount for BlackBerry
leo.yang@torchmobile.com.cn [Wed, 28 Mar 2012 12:26:03 +0000 (12:26 +0000)]
[BlackBerry] Remove PlatformMouseEvent::setClickCount for BlackBerry
https://bugs.webkit.org/show_bug.cgi?id=82422

Reviewed by Rob Buis.

PlatformMouseEvent::setClickCount() is never used by BlackBerry.
The declaration was added in PlatformMouseEvent.h internally for
BlackBerry porting, but it has not been upstreamed yet.

* platform/blackberry/PlatformMouseEventBlackBerry.cpp:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@112384 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years ago[Qt] run-qtwebkit-tests should report crashes
ossy@webkit.org [Wed, 28 Mar 2012 12:11:45 +0000 (12:11 +0000)]
[Qt] run-qtwebkit-tests should report crashes
https://bugs.webkit.org/show_bug.cgi?id=76905

Added checking of crashing tests. Now final summary includes the total number of crashing tests too.
Made unit test for this modification by János Badics <jbadics@inf.u-szeged.hu>.

Patch by Ádám Kallai <kadam@inf.u-szeged.hu> on 2012-03-28
Reviewed by Csaba Osztrogonác.

* BuildSlaveSupport/build.webkit.org-config/master.cfg:
(RunQtAPITests.commandComplete):
(RunQtAPITests.evaluateCommand):
* BuildSlaveSupport/build.webkit.org-config/mastercfg_unittest.py:
(RunQtAPITestsTest):
(test_failure):
(test_crash):
* Scripts/run-qtwebkit-tests:
(Main.announce_results_html):
(Main):
(Main.check_crash_occurences):
(Main.convert_to_stdout):
(Main.convert_to_html):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@112383 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoSeparate @import rules from other rules in CSSStyleSheet
antti@apple.com [Wed, 28 Mar 2012 10:41:18 +0000 (10:41 +0000)]
Separate @import rules from other rules in CSSStyleSheet
https://bugs.webkit.org/show_bug.cgi?id=82384

Reviewed by Andreas Kling.

Import rules always come before all other rules (except @charset). They currently live
in the generic child rule vector. They can be moved to a vector of their own for stronger
typing and more focused traversal. This will also make future refactoring easier.

- @import rules go to m_importRules
- the rest go to m_childRules

* css/CSSStyleSelector.cpp:
(WebCore::CSSStyleSelector::collectMatchingRulesForList):
* css/CSSStyleSheet.cpp:
(WebCore::CSSStyleSheet::parserAppendRule):
(WebCore::CSSStyleSheet::length):
(WebCore::CSSStyleSheet::item):
(WebCore::CSSStyleSheet::clearRules):
(WebCore::CSSStyleSheet::rules):
(WebCore::CSSStyleSheet::insertRule):
(WebCore::CSSStyleSheet::deleteRule):
(WebCore::CSSStyleSheet::isLoading):
(WebCore::CSSStyleSheet::addSubresourceStyleURLs):
* css/CSSStyleSheet.h:
(WebCore):
(CSSStyleSheet):
(WebCore::CSSStyleSheet::childRules):
(WebCore::CSSStyleSheet::importRules):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@112382 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoWeb Inspector: REGRESSION: Stack overflow on the page with > 100kloc
pfeldman@chromium.org [Wed, 28 Mar 2012 10:10:14 +0000 (10:10 +0000)]
Web Inspector: REGRESSION: Stack overflow on the page with > 100kloc
https://bugs.webkit.org/show_bug.cgi?id=82436

Reviewed by Yury Semikhatsky.

This change migrates to manual splice implementation that uses additional
information about the range being inserted to make it work faster / allocate
less memory.

* inspector/front-end/TextEditorModel.js:
(WebInspector.TextEditorModel.endsWithBracketRegex.):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@112381 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoAdd support for Intel compiler on Linux QT port.
commit-queue@webkit.org [Wed, 28 Mar 2012 10:09:34 +0000 (10:09 +0000)]
Add support for Intel compiler on Linux QT port.
https://bugs.webkit.org/show_bug.cgi?id=81077

Patch by Kwonjin Jeong <gram@company100.net> on 2012-03-28
Reviewed by Tor Arne Vestbø.

* qmake/mkspecs/features/default_post.prf:
Don't hide symbols when building with Intel compiler.
The symbols are needed when building QtWebKit tests.

* qmake/mkspecs/features/unix/icc.prf:
Added to ignore icc-specific warnings.
This feature is used with command-line arguments for build-webkit script.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@112380 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoWeb Inspector: breakpoints are not shown in sidebar pane after reload.
podivilov@chromium.org [Wed, 28 Mar 2012 09:58:07 +0000 (09:58 +0000)]
Web Inspector: breakpoints are not shown in sidebar pane after reload.
https://bugs.webkit.org/show_bug.cgi?id=82351

Reviewed by Pavel Feldman.

Source/WebCore:

When UISourceCode is added to ScriptsPanel, it could already have breakpoints.
We should iterate over existing breakpoints and add them to sidebar pane.

* inspector/front-end/ScriptsPanel.js:
(WebInspector.ScriptsPanel.prototype._uiSourceCodeAdded):

LayoutTests:

* inspector/debugger/set-breakpoint-expected.txt:
* inspector/debugger/set-breakpoint.html:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@112379 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years ago[Qt] REGRESSION(r111480): It made svg/as-image tests assert
kkristof@inf.u-szeged.hu [Wed, 28 Mar 2012 09:56:36 +0000 (09:56 +0000)]
[Qt] REGRESSION(r111480): It made svg/as-image tests assert
https://bugs.webkit.org/show_bug.cgi?id=82052

It also made svg/as-object tests assert

Reviewed by Csaba Osztrogonác.

* platform/qt/Skipped: Skip failing tests.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@112378 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years ago[WebSocket]The Sec-WebSocket-Protocol must not appear more than once in an HTTP response
commit-queue@webkit.org [Wed, 28 Mar 2012 09:12:00 +0000 (09:12 +0000)]
[WebSocket]The Sec-WebSocket-Protocol must not appear more than once in an HTTP response
https://bugs.webkit.org/show_bug.cgi?id=82432

Patch by Li Yin <li.yin@intel.com> on 2012-03-28
Reviewed by Kent Tamura.

Source/WebCore:

From RFC6455: http://tools.ietf.org/html/rfc6455#section-11.3.4
The |Sec-WebSocket-Protocol| header field must not appear
more than once in an HTTP response.

Test: http/tests/websocket/tests/hybi/handshake-fail-by-more-protocol-header.html

* Modules/websockets/WebSocketHandshake.cpp:
(WebCore::WebSocketHandshake::readHTTPHeaders):

LayoutTests:

From RFC6455: http://tools.ietf.org/html/rfc6455#section-11.3.4
The |Sec-WebSocket-Protocol| header field must not appear
more than once in an HTTP response.

* http/tests/websocket/tests/hybi/handshake-fail-by-more-protocol-header-expected.txt: Added.
* http/tests/websocket/tests/hybi/handshake-fail-by-more-protocol-header.html: Added.
* http/tests/websocket/tests/hybi/handshake-fail-by-more-protocol-header_wsh.py: Added.
(web_socket_do_extra_handshake):
(web_socket_transfer_data):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@112377 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoFix QWARN in test_loadIgnoreEmptyUrl API test.
kbalazs@webkit.org [Wed, 28 Mar 2012 08:59:43 +0000 (08:59 +0000)]
Fix QWARN in test_loadIgnoreEmptyUrl API test.

Rubber-stamped by Csaba Osztrogonác.

* UIProcess/API/qt/tests/qmltests/WebView/tst_loadUrl.qml:
There is no load() API of the view, we have to use the url property.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@112376 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoUnreviewed, trivial update-webkit-libs-jhbuild system call fix.
philn@webkit.org [Wed, 28 Mar 2012 08:54:38 +0000 (08:54 +0000)]
Unreviewed, trivial update-webkit-libs-jhbuild system call fix.

* Scripts/update-webkitgtk-libs: This script is written in
Perl. Not Python. So use perl vocabulary to refer to command line arguments.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@112375 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoUnreviewed, GTK rebaseline after r112254.
philn@webkit.org [Wed, 28 Mar 2012 08:40:24 +0000 (08:40 +0000)]
Unreviewed, GTK rebaseline after r112254.

* platform/gtk/editing/selection/select-across-readonly-input-5-expected.txt:
* platform/gtk/fast/forms/input-readonly-autoscroll-expected.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@112374 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agokill-old-processes doesn't do anything on linux
rniwa@webkit.org [Wed, 28 Mar 2012 08:26:37 +0000 (08:26 +0000)]
kill-old-processes doesn't do anything on linux
https://bugs.webkit.org/show_bug.cgi?id=82388

Reviewed by Tony Chang.

Kill cc1plus, ld, etc... on linux. Also ps aux + grep to kill old run-webkit-tests processes.
This should help stabilizing Chromium linux bots.

* BuildSlaveSupport/kill-old-processes:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@112373 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years ago[GTK] jhbuild wrongly tricked into clean-up
philn@webkit.org [Wed, 28 Mar 2012 08:22:33 +0000 (08:22 +0000)]
[GTK] jhbuild wrongly tricked into clean-up
https://bugs.webkit.org/show_bug.cgi?id=82216

Reviewed by Gustavo Noronha Silva.

Save jhbuild{rc,.modules}.md5sum in Dependencies instead of the
build tree. Also when checking if the jhbuild configuration
changed use absolute md5sum paths.

* Scripts/webkitdirs.pm:
(runAutogenForAutotoolsProjectIfNecessary):
(jhbuildConfigurationChanged):
(buildAutotoolsProject):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@112372 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agokill-old-processes doesn't do anything on linux
ossy@webkit.org [Wed, 28 Mar 2012 08:16:24 +0000 (08:16 +0000)]
kill-old-processes doesn't do anything on linux
https://bugs.webkit.org/show_bug.cgi?id=82388

One more attempt to disable kill-old-processes on the Qt bots not to kill each others.

The platform in Factory comes from config.json, accordingly platform is fullplatform,
for example: qt-4.8, qt-5.0-wk2. We have to check only the prefix of platform.

Reviewed by Ryosuke Niwa.

* BuildSlaveSupport/build.webkit.org-config/master.cfg:
(Factory.__init__):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@112371 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years ago[Qt] REGRESSION(r111480): It made svg/as-image tests assert
ossy@webkit.org [Wed, 28 Mar 2012 07:36:49 +0000 (07:36 +0000)]
[Qt] REGRESSION(r111480): It made svg/as-image tests assert
https://bugs.webkit.org/show_bug.cgi?id=82052

* platform/qt/Skipped: Skip failing tests.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@112370 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years ago[Qt] One more unreviewed gardening after r112301.
ossy@webkit.org [Wed, 28 Mar 2012 07:25:59 +0000 (07:25 +0000)]
[Qt] One more unreviewed gardening after r112301.

* platform/qt-5.0/fast/dom/Window/window-properties-expected.png: Renamed from LayoutTests/platform/qt-5.0-wk1/fast/dom/Window/window-properties-expected.png.
* platform/qt-5.0/fast/dom/Window/window-properties-expected.txt: Renamed from LayoutTests/platform/qt-5.0-wk1/fast/dom/Window/window-properties-expected.txt.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@112369 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoMake Chromium port for Android use hardware gpu path default.
commit-queue@webkit.org [Wed, 28 Mar 2012 07:05:07 +0000 (07:05 +0000)]
Make Chromium port for Android use hardware gpu path default.
https://bugs.webkit.org/show_bug.cgi?id=82033

Patch by Hao Zheng <zhenghao@chromium.org> on 2012-03-28
Reviewed by Dirk Pranke.

Skip canvas tests, which are run in virtual gpu test suites.

* Scripts/webkitpy/layout_tests/port/chromium_android.py:
(ChromiumAndroidPort.__init__):
(ChromiumAndroidPort.skipped_tests):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@112368 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agokill-old-processes doesn't do anything on linux
ossy@webkit.org [Wed, 28 Mar 2012 06:43:30 +0000 (06:43 +0000)]
kill-old-processes doesn't do anything on linux
https://bugs.webkit.org/show_bug.cgi?id=82388

Disable kill-old-processes on the Qt bots not to kill each others.

Reviewed by Ryosuke Niwa.

* BuildSlaveSupport/build.webkit.org-config/master.cfg:
(Factory.__init__):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@112367 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years ago[Qt] Unreviewed gardening after r112301.
ossy@webkit.org [Wed, 28 Mar 2012 06:40:46 +0000 (06:40 +0000)]
[Qt] Unreviewed gardening after r112301.

* platform/qt-4.8/fast/dom/Window/window-properties-expected.png:
* platform/qt-4.8/fast/dom/Window/window-properties-expected.txt:
* platform/qt-5.0-wk1/fast/dom/Window/window-properties-expected.png: Added.
* platform/qt-5.0-wk1/fast/dom/Window/window-properties-expected.txt: Renamed from LayoutTests/platform/qt/fast/dom/Window/window-properties-expected.txt.
* platform/qt-5.0-wk2/Skipped:
* platform/qt-5.0-wk2/fast/dom/Window/window-properties-expected.txt:
* platform/qt/fast/dom/Window/window-properties-expected.png: Removed.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@112366 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoNot reviewed: rebaselined chromium expectations.
pfeldman@chromium.org [Wed, 28 Mar 2012 06:12:21 +0000 (06:12 +0000)]
Not reviewed: rebaselined chromium expectations.

* platform/chromium-mac-leopard/fast/text/unicode-variation-selector-expected.png:
* platform/chromium-mac-leopard/fast/text/unicode-variation-selector-expected.txt: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@112365 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years ago[chromium] Transfer wheel fling via WebCompositorInputHandlerClient
commit-queue@webkit.org [Wed, 28 Mar 2012 06:08:16 +0000 (06:08 +0000)]
[chromium] Transfer wheel fling via WebCompositorInputHandlerClient
https://bugs.webkit.org/show_bug.cgi?id=81740

Patch by James Robinson <jamesr@chromium.org> on 2012-03-27
Reviewed by Adrienne Walker.

Source/WebCore:

Adds the ability to construct an in-progress PlatformGestureAnimation.

* platform/ActivePlatformGestureAnimation.cpp:
(WebCore::ActivePlatformGestureAnimation::create):
(WebCore):
(WebCore::ActivePlatformGestureAnimation::ActivePlatformGestureAnimation):
* platform/ActivePlatformGestureAnimation.h:
(ActivePlatformGestureAnimation):
* platform/TouchpadFlingPlatformGestureCurve.cpp:
(WebCore::TouchpadFlingPlatformGestureCurve::create):
(WebCore::TouchpadFlingPlatformGestureCurve::TouchpadFlingPlatformGestureCurve):
* platform/TouchpadFlingPlatformGestureCurve.h:
(TouchpadFlingPlatformGestureCurve):

Source/WebKit/chromium:

Adds a path for transfering an active wheel fling animation out to the embedder from the compositor and back in
to a WebViewImpl via the embedder. This is used when we start a wheel fling animation on the compositor thread
but then hit a condition that we can't handle from the compositor, such as registered wheel event listeners or a
scrollable area we can't handle.

New tests added to WebCompositorInputHandlerTest for the transfering logic.

* public/WebActiveWheelFlingParameters.h: Copied from Source/WebKit/chromium/public/WebCompositorInputHandlerClient.h.
(WebKit):
(WebActiveWheelFlingParameters):
(WebKit::WebActiveWheelFlingParameters::WebActiveWheelFlingParameters):
* public/WebCompositorInputHandlerClient.h:
(WebKit):
(WebCompositorInputHandlerClient):
(WebKit::WebCompositorInputHandlerClient::transferActiveWheelFlingAnimation):
* public/WebView.h:
(WebKit):
(WebView):
* src/WebCompositorInputHandlerImpl.cpp:
(WebKit::WebCompositorInputHandlerImpl::handleGestureFling):
(WebKit::WebCompositorInputHandlerImpl::animate):
(WebKit::WebCompositorInputHandlerImpl::cancelCurrentFling):
(WebKit::WebCompositorInputHandlerImpl::scrollBy):
* src/WebCompositorInputHandlerImpl.h:
(WebCore):
* src/WebViewImpl.cpp:
(WebKit::WebViewImpl::transferActiveWheelFlingAnimation):
(WebKit):
* src/WebViewImpl.h:
(WebViewImpl):
* tests/WebCompositorInputHandlerImplTest.cpp:
(WebKit::MockWebCompositorInputHandlerClient::MockWebCompositorInputHandlerClient):
(MockWebCompositorInputHandlerClient):
(WebKit::TEST):
(WebKit::WebCompositorInputHandlerImplTest::WebCompositorInputHandlerImplTest):
(WebKit::WebCompositorInputHandlerImplTest::~WebCompositorInputHandlerImplTest):
(WebCompositorInputHandlerImplTest):
(WebKit::TEST_F):
(WebKit):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@112364 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years ago[chromium] Send wheel events to main thread even if we think nothing is scrollable
commit-queue@webkit.org [Wed, 28 Mar 2012 04:27:18 +0000 (04:27 +0000)]
[chromium] Send wheel events to main thread even if we think nothing is scrollable
https://bugs.webkit.org/show_bug.cgi?id=82408

Patch by James Robinson <jamesr@chromium.org> on 2012-03-27
Reviewed by Adrienne Walker.

* src/WebCompositorInputHandlerImpl.cpp:
(WebKit::WebCompositorInputHandlerImpl::handleInputEventInternal):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@112363 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoPerf-o-matic build fix.
rniwa@webkit.org [Wed, 28 Mar 2012 04:19:05 +0000 (04:19 +0000)]
Perf-o-matic build fix.

* Websites/webkit-perf.appspot.com/controller.py:
(RunsChartHandler.post):
* Websites/webkit-perf.appspot.com/models.py:
(Runs.chart_params):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@112362 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years ago[BlackBerry] Text selection - selection gets broken in test.com/individuals.htm
commit-queue@webkit.org [Wed, 28 Mar 2012 03:47:00 +0000 (03:47 +0000)]
[BlackBerry] Text selection - selection gets broken in test.com/individuals.htm
https://bugs.webkit.org/show_bug.cgi?id=82292

Change to check and avoid text selection across frames.

Internal reviewed by Mike Fenton

Patch by Sean Wang <Xuewen.Wang@torchmobile.com.cn> on 2012-03-27
Reviewed by Rob Buis.

* WebKitSupport/SelectionHandler.cpp:
(BlackBerry::WebKit::visiblePositionForPointIgnoringClipping):
    support selection across frames, so check if the *framePoint* is in
    the *frame*.
(BlackBerry::WebKit::SelectionHandler::setSelection):
    function returns a null VisablePosition, it stands for a invalid position
    or a position in the different frames, therefor we don't execute setting
    handle's position.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@112361 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years ago[chromium] Route monotonic clock up from compositor
nduca@chromium.org [Wed, 28 Mar 2012 03:15:57 +0000 (03:15 +0000)]
[chromium] Route monotonic clock up from compositor
https://bugs.webkit.org/show_bug.cgi?id=82154

Reviewed by James Robinson.

Source/Platform:

* chromium/public/WebLayerTreeViewClient.h:
(WebLayerTreeViewClient):

Source/WebCore:

* platform/graphics/chromium/cc/CCLayerTreeHost.cpp:
(WebCore::CCLayerTreeHost::updateAnimations):
* platform/graphics/chromium/cc/CCLayerTreeHost.h:
(CCLayerTreeHost):
* platform/graphics/chromium/cc/CCThreadProxy.cpp:
(WebCore::CCThreadProxy::scheduledActionBeginFrame):
(WebCore::CCThreadProxy::beginFrame):
* platform/graphics/chromium/cc/CCThreadProxy.h:
(WebCore::CCThreadProxy::BeginFrameAndCommitState::BeginFrameAndCommitState):
(BeginFrameAndCommitState):

Source/WebKit/chromium:

* public/WebWidget.h:
(WebKit::WebWidget::animate):
* src/WebLayerTreeViewImpl.cpp:
(WebKit::WebLayerTreeViewImpl::updateAnimations):
* src/WebLayerTreeViewImpl.h:
(WebLayerTreeViewImpl):
* src/WebViewImpl.cpp:
(WebKit::WebViewImpl::animate):
(WebKit::WebViewImpl::updateAnimations):
* src/WebViewImpl.h:
(WebViewImpl):
* tests/FakeCCLayerTreeHostClient.h:
(WebCore::FakeCCLayerTreeHostClient::updateAnimations):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@112360 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoHandle DRT crash caused by Android OOM.
commit-queue@webkit.org [Wed, 28 Mar 2012 03:05:31 +0000 (03:05 +0000)]
Handle DRT crash caused by Android OOM.
https://bugs.webkit.org/show_bug.cgi?id=82310

Patch by Hao Zheng <zhenghao@chromium.org> on 2012-03-27
Reviewed by Tony Chang.

When Android is OOM, it sends a SIGKILL (137) signal to DRT. DRT
is stopped silently and regarded as crashed. Re-run the test for
such crash.

Add some debug log to NRWT.

* Scripts/webkitpy/layout_tests/port/chromium.py:
(ChromiumDriver.run_test):
* Scripts/webkitpy/layout_tests/port/chromium_android.py:
(ChromiumAndroidPort.__init__):
(ChromiumAndroidPort.get_last_stacktrace):
(ChromiumAndroidDriver.__init__):
(ChromiumAndroidDriver._start):
(ChromiumAndroidDriver.run_test):
(ChromiumAndroidDriver._get_drt_return_value):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@112359 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years ago[BlackBerry] Upstream LayerAnimation.{cpp, h}
commit-queue@webkit.org [Wed, 28 Mar 2012 02:56:42 +0000 (02:56 +0000)]
[BlackBerry] Upstream LayerAnimation.{cpp, h}
https://bugs.webkit.org/show_bug.cgi?id=80123

Patch by Robin Cao <robin.cao@torchmobile.com.cn> on 2012-03-27
Reviewed by Rob Buis.

Initial upstream, no new tests.

* platform/graphics/blackberry/LayerAnimation.cpp: Added.
(WebCore):
(WebCore::solveEpsilon):
(WebCore::solveCubicBezierFunction):
(WebCore::solveStepsFunction):
(WebCore::timingFunctionForAnimationValue):
(WebCore::progress):
(WebCore::fetchIntervalEndpoints):
(WebCore::LayerAnimation::apply):
(WebCore::LayerAnimation::blendTransform):
(WebCore::LayerAnimation::blendOpacity):
(WebCore::LayerAnimation::validateTransformLists):
* platform/graphics/blackberry/LayerAnimation.h: Added.
(WebCore):
(LayerAnimation):
(WebCore::LayerAnimation::create):
(WebCore::LayerAnimation::clone):
(WebCore::LayerAnimation::~LayerAnimation):
(WebCore::LayerAnimation::name):
(WebCore::LayerAnimation::setStartTime):
(WebCore::LayerAnimation::idFromAnimation):
(WebCore::LayerAnimation::isEqualToAnimation):
(WebCore::LayerAnimation::id):
(WebCore::LayerAnimation::property):
(WebCore::LayerAnimation::boxSize):
(WebCore::LayerAnimation::timeOffset):
(WebCore::LayerAnimation::startTime):
(WebCore::LayerAnimation::valueCount):
(WebCore::LayerAnimation::timingFunction):
(WebCore::LayerAnimation::duration):
(WebCore::LayerAnimation::iterationCount):
(WebCore::LayerAnimation::direction):
(WebCore::LayerAnimation::valueAt):
(WebCore::LayerAnimation::LayerAnimation):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@112358 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years ago[Chromium] Uninitialized access in SimpleFontDataSkia::platformInit
bashi@chromium.org [Wed, 28 Mar 2012 02:42:05 +0000 (02:42 +0000)]
[Chromium] Uninitialized access in SimpleFontDataSkia::platformInit
https://bugs.webkit.org/show_bug.cgi?id=82411

Reviewed by Kent Tamura.

SimpleFontDataSkia::platformInit() could call widthForGlyph(), which
accesses m_zeroWidthSpaceGlyph. This causes a valgrind memcheck error
because m_zeroWidthSpaceGlyph isn't initialized at this point.
Initialize m_zeroWidthSpaceGlyph with zero so that widthForGlyph() can
return appropriate value(The value zero here means "unknown glyph").

No new tests. I manually confirmed this change fixes the valgrind
memcheck error.

* platform/graphics/skia/SimpleFontDataSkia.cpp:
(WebCore::SimpleFontData::platformInit):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@112357 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agowebkit_unit_tests crashing on chromium lion in CCLayerTreeHostTestCompositeAndReadbac...
commit-queue@webkit.org [Wed, 28 Mar 2012 02:29:21 +0000 (02:29 +0000)]
webkit_unit_tests crashing on chromium lion in CCLayerTreeHostTestCompositeAndReadbackWhileInvisible
https://bugs.webkit.org/show_bug.cgi?id=82382

Patch by James Robinson <jamesr@chromium.org> on 2012-03-27
Reviewed by Adrienne Walker.

Use UnsignedWithZeroKeyHashTraits trait for the HashSet<WebGLId> m_usedTextures in
CompositorFakeWebGraphicsContext3DWithTextureTracking. This set is used to track all textures that are bound in
a given context during a test. During ..CompositeAndReadback.. we bind texture id 0. By default, HashSet<T>
where T is an unsigned does not allow 0 as a value. WebGLId is typedef'd to unsigned.

* tests/CCLayerTreeHostTest.cpp:
(CompositorFakeWebGraphicsContext3DWithTextureTracking):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@112356 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoFix expected results for some unittest failures.
ojan@chromium.org [Wed, 28 Mar 2012 02:08:49 +0000 (02:08 +0000)]
Fix expected results for some unittest failures.

* BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui/failures_unittests.js:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@112355 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years ago[chromium] Fix --enabled-threaded-compositing flag in DRT
enne@google.com [Wed, 28 Mar 2012 02:06:54 +0000 (02:06 +0000)]
[chromium] Fix --enabled-threaded-compositing flag in DRT
https://bugs.webkit.org/show_bug.cgi?id=82405

Reviewed by James Robinson.

--enable-threaded-compositing wasn't working because there was no way
to set the value before it was used in the TestShell constructor.

To avoid adding yet another boolean parameter to TestShell's
constructor, add an explicit initialize function that can operate on
any boolean settings that are set post-constructor.

* DumpRenderTree/chromium/DumpRenderTree.cpp:
(main):
* DumpRenderTree/chromium/TestShell.cpp:
(TestShell::TestShell):
(TestShell::initialize):
* DumpRenderTree/chromium/TestShell.h:
(TestShell):
(TestShell::setTestShellMode):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@112354 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years ago[EFL][WK2] Add RunLoopEfl and WorkQueueEfl
commit-queue@webkit.org [Wed, 28 Mar 2012 02:02:18 +0000 (02:02 +0000)]
[EFL][WK2] Add RunLoopEfl and WorkQueueEfl
https://bugs.webkit.org/show_bug.cgi?id=62777

Patch by YoungTaeck Song <youngtaeck.song@samsung.com> on 2012-03-27
Reviewed by Hajime Morita.

Source/WebCore:

Add initial version RunLoopEfl for WebKit2 Efl.

* platform/RunLoop.h:
(TimerBase):
(RunLoop):
* platform/efl/RunLoopEfl.cpp:
(WebCore::RunLoop::RunLoop):
(WebCore::RunLoop::~RunLoop):
(WebCore):
(WebCore::RunLoop::run):
(WebCore::RunLoop::stop):
(WebCore::RunLoop::wakeUpEvent):
(WebCore::RunLoop::wakeUp):
(WebCore::RunLoop::TimerBase::TimerBase):
(WebCore::RunLoop::TimerBase::~TimerBase):
(WebCore::RunLoop::TimerBase::timerFired):
(WebCore::RunLoop::TimerBase::start):
(WebCore::RunLoop::TimerBase::stop):
(WebCore::RunLoop::TimerBase::isActive):

Source/WebKit2:

Add initial version WorkQueueEfl for WebKit2 Efl.

* Platform/CoreIPC/Connection.h:
* Platform/CoreIPC/unix/ConnectionUnix.cpp:
(CoreIPC::Connection::platformInvalidate):
(CoreIPC::Connection::open):
* Platform/PlatformProcessIdentifier.h:
(WebKit):
* Platform/WorkQueue.h:
(WorkQueue):
* Platform/efl/WorkQueueEfl.cpp: Added.
(TimerWorkItem):
(TimerWorkItem::TimerWorkItem):
(TimerWorkItem::~TimerWorkItem):
(TimerWorkItem::function):
(TimerWorkItem::queue):
(TimerWorkItem::timerID):
(WorkQueue::platformInitialize):
(WorkQueue::platformInvalidate):
(WorkQueue::performWork):
(WorkQueue::performFdWork):
(WorkQueue::sendMessageToThread):
(WorkQueue::workQueueThread):
(WorkQueue::registerSocketEventHandler):
(WorkQueue::unregisterSocketEventHandler):
(WorkQueue::dispatch):
(WorkQueue::timerFired):
(WorkQueue::dispatchAfterDelay):
* PlatformEfl.cmake:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@112353 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoMore flaky tests!
dpranke@chromium.org [Wed, 28 Mar 2012 01:58:32 +0000 (01:58 +0000)]
More flaky tests!

Unreviewed, expectations changes.

* platform/chromium/test_expectations.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@112352 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agowebkit_unit_tests crashing on chromium lion in ScrollbarLayerChromiumTest.resolveScro...
commit-queue@webkit.org [Wed, 28 Mar 2012 01:56:30 +0000 (01:56 +0000)]
webkit_unit_tests crashing on chromium lion in ScrollbarLayerChromiumTest.resolveScrollLayerPointer
https://bugs.webkit.org/show_bug.cgi?id=82381

Patch by James Robinson <jamesr@chromium.org> on 2012-03-27
Reviewed by Adrienne Walker.

Use the mock scrollbar theme for this unit test. The real ScrollbarThemeMac isn't compatible with the mocks this
test uses and the test isn't designed to cover the theming code, only code inside the compositor.

* tests/ScrollbarLayerChromiumTest.cpp:
(WebCore::TEST):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@112351 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoRestore a couple of leopard/snowleopard expectations I overzealously deleted.
dpranke@chromium.org [Wed, 28 Mar 2012 01:16:21 +0000 (01:16 +0000)]
Restore a couple of leopard/snowleopard expectations I overzealously deleted.

Unreviewed, expectations change.

* platform/chromium/test_expectations.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@112350 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoRemove move tests passing on win and linux.
dpranke@chromium.org [Wed, 28 Mar 2012 01:10:37 +0000 (01:10 +0000)]
Remove move tests passing on win and linux.

Unreviewed, expectations change.

* platform/chromium/test_expectations.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@112349 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoRemove more tests passing on lion
dpranke@chromium.org [Wed, 28 Mar 2012 01:01:40 +0000 (01:01 +0000)]
Remove more tests passing on lion

Unreviewed, expectations change.

* platform/chromium/test_expectations.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@112348 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoReinforce Geolocation to prevent accidental leak of the user position
benjamin@webkit.org [Wed, 28 Mar 2012 00:58:20 +0000 (00:58 +0000)]
Reinforce Geolocation to prevent accidental leak of the user position
https://bugs.webkit.org/show_bug.cgi?id=82396

Patch by Benjamin Poulain <bpoulain@apple.com> on 2012-03-27
Reviewed by Adam Barth.

It is very important not to provide the position of the user to a page
unless the user authorize it.

The code used to make it easy to cause such problems, because any part
of the Geolocation object could invoke the success callback directly.

This patch add encapsulation for all the attributes of GeoNotifier,
and add extra guards for the two callbacks.

In the case of the success callback, we do one extra check before sending
the value to the bindings.

* Modules/geolocation/Geolocation.cpp:
(WebCore::Geolocation::GeoNotifier::runSuccessCallback):
(WebCore::Geolocation::GeoNotifier::runErrorCallback):
(WebCore):
(WebCore::Geolocation::GeoNotifier::stopTimer):
(WebCore::Geolocation::GeoNotifier::timerFired):
(WebCore::Geolocation::startRequest):
(WebCore::Geolocation::sendError):
(WebCore::Geolocation::sendPosition):
(WebCore::Geolocation::stopTimer):
(WebCore::Geolocation::extractNotifiersWithCachedPosition):
(WebCore::Geolocation::startUpdating):
* Modules/geolocation/Geolocation.h:
(WebCore::Geolocation::isAllowed):
(Geolocation):
(GeoNotifier):
(WebCore::Geolocation::GeoNotifier::options):
(WebCore::Geolocation::GeoNotifier::useCachedPosition):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@112347 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoUnreviewed, rolling out r112339.
dpranke@chromium.org [Wed, 28 Mar 2012 00:52:40 +0000 (00:52 +0000)]
Unreviewed, rolling out r112339.
http://trac.webkit.org/changeset/112339
https://bugs.webkit.org/show_bug.cgi?id=82354

webkit_unit_tests crashing on chromium mac bots

* src/AssociatedURLLoader.cpp:
(AssociatedURLLoader::ClientAdapter):
* tests/AssociatedURLLoaderTest.cpp:
(WebKit::AssociatedURLLoaderTest::TearDown):
(WebKit):
(WebKit::TEST_F):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@112346 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoAdd suppression for flaky plugins/netscape-dom-access-and-reload.html on mac and...
dpranke@chromium.org [Wed, 28 Mar 2012 00:47:43 +0000 (00:47 +0000)]
Add suppression for flaky plugins/netscape-dom-access-and-reload.html on mac and linux.

Unreviewed, expectations change.

* platform/chromium/test_expectations.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@112344 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years ago[Coverity] Address some uninitialized constructor values.
commit-queue@webkit.org [Wed, 28 Mar 2012 00:37:41 +0000 (00:37 +0000)]
[Coverity] Address some uninitialized constructor values.
https://bugs.webkit.org/show_bug.cgi?id=82376

Patch by Kausalya Madhusudhanan <kmadhusu@chromium.org> on 2012-03-27
Reviewed by James Robinson.

Source/WebCore:

New tests are not required since I did not modify any code behavior.

* html/shadow/MediaControlRootElementChromium.cpp:
(WebCore::MediaControlRootElementChromium::MediaControlRootElementChromium):
* inspector/InspectorIndexedDBAgent.h:
(InspectorIndexedDBAgent):
* inspector/InspectorTimelineAgent.cpp:
(WebCore::InspectorTimelineAgent::InspectorTimelineAgent):
* page/scrolling/ScrollingCoordinator.cpp:
(WebCore::ScrollingCoordinator::ScrollingCoordinator):
* platform/ScrollAnimatorNone.cpp:
(WebCore::ScrollAnimatorNone::ScrollAnimatorNone):
* platform/chromium/DataTransferItemChromium.cpp:
(WebCore::DataTransferItemChromium::DataTransferItemChromium):
* platform/graphics/skia/ImageBufferSkia.cpp:
(WebCore::ImageBuffer::ImageBuffer):
* storage/StorageTask.cpp:
(WebCore::StorageTask::StorageTask):

Source/WebKit/chromium:

New tests are not required since I did not modify any code behavior. I just initialized the class member variables in the constructor.

* src/WebCompositorInputHandlerImpl.cpp:
(WebCore::PlatformGestureToCCGestureAdapter::PlatformGestureToCCGestureAdapter):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@112343 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoRoll chromium DEPS to r129300.
dpranke@chromium.org [Wed, 28 Mar 2012 00:28:07 +0000 (00:28 +0000)]
Roll chromium DEPS to r129300.

Unreviewed, deps change.

* DEPS:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@112342 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoSimplify CSSPropertyBorderSpacing parsing.
alexis.menard@openbossa.org [Wed, 28 Mar 2012 00:27:27 +0000 (00:27 +0000)]
Simplify CSSPropertyBorderSpacing parsing.
https://bugs.webkit.org/show_bug.cgi?id=82397

Reviewed by Benjamin Poulain.

Remove the local array of longhand properties as it doesn't really
bring much in this simple parsing algorithm.

No new tests, no functionality change intended.

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

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@112341 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoFix race condition when initializing the scrolling thread
andersca@apple.com [Wed, 28 Mar 2012 00:21:13 +0000 (00:21 +0000)]
Fix race condition when initializing the scrolling thread
https://bugs.webkit.org/show_bug.cgi?id=82398
<rdar://problem/11002166>

Reviewed by Sam Weinig.

Lock m_initializeRunLoopConditionMutex when assigning m_threadIdentifier since we're
asserting that it's not null in the scrolling thread.

* page/scrolling/ScrollingThread.cpp:
(WebCore::ScrollingThread::createThreadIfNeeded):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@112340 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoAssociatedURLLoader does not support Cross Origin Redirects when using
bbudge@chromium.org [Wed, 28 Mar 2012 00:19:47 +0000 (00:19 +0000)]
AssociatedURLLoader does not support Cross Origin Redirects when using
Access Control.
https://bugs.webkit.org/show_bug.cgi?id=82354

AssociatedURLLoader's internal adapter now overrides didFailRedirectCheck,
which cancels the load, causing didFail to notify the client that the
load failed. AssociatedURLLoaderTest adds test cases for CORS requests
that receive redirects and pass or fail the redirect access check.

Reviewed by Adam Barth.

* src/AssociatedURLLoader.cpp:
(AssociatedURLLoader::ClientAdapter):
(WebKit::AssociatedURLLoader::ClientAdapter::didFailRedirectCheck):
(WebKit):
* tests/AssociatedURLLoaderTest.cpp:
(WebKit::TEST_F):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@112339 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoPlug-ins using the Core Animation drawing model should work when hosting the layer...
andersca@apple.com [Wed, 28 Mar 2012 00:16:52 +0000 (00:16 +0000)]
Plug-ins using the Core Animation drawing model should work when hosting the layer tree in the window server
https://bugs.webkit.org/show_bug.cgi?id=82387
<rdar://problem/11031942>

Reviewed by Sam Weinig.

* PluginProcess/PluginControllerProxy.cpp:
(WebKit::PluginControllerProxy::PluginControllerProxy):
(WebKit::PluginControllerProxy::initialize):
Remove m_pluginCreationParameters; it was used by the old NPRuntime short-circuit code.
Make platformInitialize take creation parameters.

* PluginProcess/PluginControllerProxy.messages.in:
Add SetLayerHostingMode message.

* PluginProcess/gtk/PluginControllerProxyGtk.cpp:
(WebKit::PluginControllerProxy::platformInitialize):
Make platformInitialize take creation parameters.

* PluginProcess/mac/PluginControllerProxyMac.mm:
(WebKit::PluginControllerProxy::platformInitialize):
Call updateLayerHostingContext.

(WebKit::PluginControllerProxy::setLayerHostingMode):
Call updateLayerHostingContext and send back the new context ID.

(WebKit::PluginControllerProxy::updateLayerHostingContext):
Create a new LayerHostingContext given the layer hosting mode.

* PluginProcess/qt/PluginControllerProxyQt.cpp:
(WebKit::PluginControllerProxy::platformInitialize):
Make platformInitialize take creation parameters.

* WebProcess/Plugins/Netscape/NetscapePlugin.h:
* WebProcess/Plugins/Netscape/mac/NetscapePluginMac.mm:
(WebKit::NetscapePlugin::setLayerHostingMode):
Add stub.

* WebProcess/Plugins/Netscape/mac/PluginProxyMac.mm:
(WebKit::makeRenderLayer):
Add helper function for creating a new render layer.

(WebKit::PluginProxy::pluginLayer):
Call makeRenderLayer.

(WebKit::PluginProxy::setLayerHostingMode):
Send a message to the plug-in process.

(WebKit::PluginProxy::setLayerHostingContextID):
Update the context ID and make a new render layer.

* WebProcess/Plugins/PDF/BuiltInPDFView.cpp:
(WebKit::BuiltInPDFView::setLayerHostingMode):
Add stub.
* WebProcess/Plugins/PDF/BuiltInPDFView.h:

* WebProcess/Plugins/Plugin.cpp:
(WebKit::Plugin::Parameters::encode):
(WebKit::Plugin::Parameters::decode):
* WebProcess/Plugins/Plugin.h:
Add the layer hosting mode as a parameter. Add a new setLayerHostingMode pure virtual member function.

* WebProcess/Plugins/PluginProxy.messages.in:
Add a SetLayerHostingContextID message.

* WebProcess/Plugins/PluginView.cpp:
(WebKit::PluginView::setLayerHostingMode):
Call down to the plug-in.

* WebProcess/WebCoreSupport/WebFrameLoaderClient.cpp:
(WebKit::WebFrameLoaderClient::createPlugin):
Set the layer hosting mode of the page.

* WebProcess/WebPage/WebPage.cpp:
(WebKit::WebPage::WebPage):
Initialize m_layerHostingMode to false.

* WebProcess/WebPage/WebPage.h:
(WebKit::WebPage::layerHostingMode):
Add m_layerHostingMode and a getter.

* WebProcess/WebPage/mac/TiledCoreAnimationDrawingArea.mm:
(WebKit::TiledCoreAnimationDrawingArea::setLayerHostingMode):
Call WebPage::setLayerHostingMode.

* WebProcess/WebPage/mac/WebPageMac.mm:
(WebKit::WebPage::setLayerHostingMode):
Tell all plug-ins that the layer hosting mode changed.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@112338 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agonew-run-webkit-tests: suppress debug output from run-safari on results file
dpranke@chromium.org [Wed, 28 Mar 2012 00:12:57 +0000 (00:12 +0000)]
new-run-webkit-tests: suppress debug output from run-safari on results file
https://bugs.webkit.org/show_bug.cgi?id=82400

Reviewed by Ryosuke Niwa.

Send the run-safari output to /dev/null rather than the console.

* Scripts/webkitpy/layout_tests/port/mac.py:
(MacPort.show_results_html_file):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@112337 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoAdd assertions to valueForLength() and RenderBox::computeLogicalWidthInRegionUsing...
macpherson@chromium.org [Wed, 28 Mar 2012 00:12:30 +0000 (00:12 +0000)]
Add assertions to valueForLength() and RenderBox::computeLogicalWidthInRegionUsing() to help with debugging.
https://bugs.webkit.org/show_bug.cgi?id=82393

Reviewed by Eric Seidel.

No new tests / adding assetions only.

* css/LengthFunctions.cpp:
(WebCore::valueForLength):
* rendering/RenderBox.cpp:
(WebCore::RenderBox::computeLogicalWidthInRegionUsing):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@112336 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoRemove more passing tests, tighten overly broad expectations.
dpranke@chromium.org [Wed, 28 Mar 2012 00:02:34 +0000 (00:02 +0000)]
Remove more passing tests, tighten overly broad expectations.

Unreviewed, expectations change.

* platform/chromium/test_expectations.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@112334 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agouse the correct size when computing flex-pack space
tony@chromium.org [Tue, 27 Mar 2012 23:56:11 +0000 (23:56 +0000)]
use the correct size when computing flex-pack space
https://bugs.webkit.org/show_bug.cgi?id=82378

Reviewed by Ojan Vafai.

Source/WebCore:

Fix 2 bugs:
- We weren't properly updating available space before computing
  packing space. If a min/max is not hit, we still need to adjust
  the available free space.
- For flex-pack:end, we need to put the overflow in the start edge.

New test cases in css3/flexbox/flex-pack.html

* rendering/RenderFlexibleBox.cpp:
(WebCore::RenderFlexibleBox::resolveFlexibleLengths):
(WebCore::initialPackingOffset):

LayoutTests:

* css3/flexbox/flex-pack-expected.txt:
* css3/flexbox/flex-pack.html:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@112333 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years ago[chromium] Unknown transforms should be treated as non-axis aligned on main thread
commit-queue@webkit.org [Tue, 27 Mar 2012 23:49:04 +0000 (23:49 +0000)]
[chromium] Unknown transforms should be treated as non-axis aligned on main thread
https://bugs.webkit.org/show_bug.cgi?id=82370

Patch by Dana Jansens <danakj@chromium.org> on 2012-03-27
Reviewed by Adrienne Walker.

Source/WebCore:

On main thread, animating transforms have "unknown" values as they are changing
out of sync on the impl thread. So treat them as non-axis-aligned since they
may be, when deciding to create a render surface.

In addition, since surfaces are cheap on main thread, create one for all layers
with animating transforms and a drawing descendant, as this allows paint culling
within the layer's subtree (the animated transform won't affect drawTransforms
inside the subtree).

Also renamed the layerIsInAnimatingSubtreeFor* to animatingTransformTo*.
The old name made me pause and think what it meant and I'm the one who
created it. Hopefully this is more clear.

Unit test: CCLayerTreeHostCommonTest.verifyAnimationsForRenderSurfaceHierarchy

* platform/graphics/chromium/cc/CCLayerTreeHostCommon.cpp:
(WebCore::transformToParentIsKnown):
(WebCore):
(WebCore::subtreeShouldRenderToSeparateSurface):
(WebCore::calculateDrawTransformsAndVisibilityInternal):

Source/WebKit/chromium:

* tests/CCLayerTreeHostCommonTest.cpp:
(WebKitTests::TEST):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@112332 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoRe-land r112277; reverting it doesn't seem to have fixed anything.
dpranke@chromium.org [Tue, 27 Mar 2012 23:21:14 +0000 (23:21 +0000)]
Re-land r112277; reverting it doesn't seem to have fixed anything.

Unreviewed, build fix.

* svg/SVGUseElement.cpp:
(WebCore::SVGUseElement::insertedIntoDocument):
(WebCore::SVGUseElement::svgAttributeChanged):
(WebCore::SVGUseElement::willRecalcStyle):
(WebCore::SVGUseElement::finishParsingChildren):
* xml/XMLErrors.cpp:
(WebCore::XMLErrors::insertErrorMessageBlock):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@112329 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years ago[chromium] Rename opaqueContentsRegion() to visibleContentOpaqueRegion()
commit-queue@webkit.org [Tue, 27 Mar 2012 22:46:36 +0000 (22:46 +0000)]
[chromium] Rename opaqueContentsRegion() to visibleContentOpaqueRegion()
https://bugs.webkit.org/show_bug.cgi?id=81689

Patch by Dana Jansens <danakj@chromium.org> on 2012-03-27
Reviewed by Adrienne Walker.

Source/WebCore:

The return value from this function is a region of opaque pixels in the
layer's content space that intersect with its visible rect. Rather than
adding a comment to this effect, renaming the method to make it clear.

* platform/graphics/chromium/LayerChromium.h:
(WebCore::LayerChromium::visibleContentOpaqueRegion):
* platform/graphics/chromium/TiledLayerChromium.cpp:
(WebCore::TiledLayerChromium::visibleContentOpaqueRegion):
* platform/graphics/chromium/TiledLayerChromium.h:
* platform/graphics/chromium/cc/CCLayerImpl.h:
(WebCore::CCLayerImpl::visibleContentOpaqueRegion):
* platform/graphics/chromium/cc/CCOcclusionTracker.cpp:
(WebCore::computeOcclusionBehindLayer):
* platform/graphics/chromium/cc/CCTiledLayerImpl.cpp:
(WebCore::CCTiledLayerImpl::visibleContentOpaqueRegion):
* platform/graphics/chromium/cc/CCTiledLayerImpl.h:
(CCTiledLayerImpl):

Source/WebKit/chromium:

* tests/CCLayerTreeHostTest.cpp:
(WTF::TestLayerChromium::opaqueVisibleContents):
* tests/CCOcclusionTrackerTest.cpp:
(WebCore::TestContentLayerChromium::opaqueVisibleContents):
(WebCore::TestContentLayerImpl::opaqueVisibleContents):
* tests/TiledLayerChromiumTest.cpp:
(WTF::TEST):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@112327 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years ago[chromium] A unit test to verify clipping is honored when moving occlusion to parent...
commit-queue@webkit.org [Tue, 27 Mar 2012 22:29:21 +0000 (22:29 +0000)]
[chromium] A unit test to verify clipping is honored when moving occlusion to parent surface
https://bugs.webkit.org/show_bug.cgi?id=81783

Patch by Dana Jansens <danakj@chromium.org> on 2012-03-27
Reviewed by Adrienne Walker.

* tests/CCOcclusionTrackerTest.cpp:
(WebCore::CCOcclusionTrackerTestSurfaceOcclusionTranslatesToParent::runMyTest):
(WebCore):
(CCOcclusionTrackerTestSurfaceOcclusionTranslatesWithClipping):
(WebCore::CCOcclusionTrackerTestSurfaceOcclusionTranslatesWithClipping::runMyTest):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@112326 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoScrollable plugins not registered properly in ScrollingCoordinator
commit-queue@webkit.org [Tue, 27 Mar 2012 22:24:36 +0000 (22:24 +0000)]
Scrollable plugins not registered properly in ScrollingCoordinator
https://bugs.webkit.org/show_bug.cgi?id=82163

Patch by James Robinson <jamesr@chromium.org> on 2012-03-27
Reviewed by Anders Carlsson.

Source/WebCore:

Whenever a ScrollableArea is added or removed from a FrameView's ScrollableAreaSet, we have to recalculate the
nonFastScrollableRegion. This can happen for certain types of plugins that are scrollable.

This also reverts 112142 which was a not quite right way to handle these plugins.

* page/FrameView.cpp:
(WebCore::FrameView::addScrollableArea):
(WebCore::FrameView::removeScrollableArea):
* page/scrolling/ScrollingCoordinator.cpp:
(WebCore::computeNonFastScrollableRegion):
(WebCore::ScrollingCoordinator::frameViewScrollableAreasDidChange):
(WebCore):
* page/scrolling/ScrollingCoordinator.h:
(ScrollingCoordinator):
* plugins/PluginViewBase.h:

Source/WebKit/chromium:

Since ScrollbarGroups are ScrollableAreas, they need to be able to report their bounds for the
ScrollingCoordinator's calculateNonFastScrollableRegion. This also changes ScrollbarGroups to only be registered
as ScrollableAreas on the FrameView's set when they actually have Scrollbars.

* src/ScrollbarGroup.cpp:
(WebKit::ScrollbarGroup::ScrollbarGroup):
(WebKit::ScrollbarGroup::~ScrollbarGroup):
(WebKit::ScrollbarGroup::scrollbarCreated):
(WebKit::ScrollbarGroup::scrollbarDestroyed):
(WebKit::ScrollbarGroup::setFrameRect):
(WebKit):
(WebKit::ScrollbarGroup::scrollableAreaBoundingBox):
* src/ScrollbarGroup.h:
(ScrollbarGroup):
* src/WebPluginContainerImpl.cpp:
(WebKit::WebPluginContainerImpl::reportGeometry):
(WebKit):
(WebKit::WebPluginContainerImpl::scrollbarGroup):
* src/WebPluginContainerImpl.h:
(WebPluginContainerImpl):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@112325 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoRemove a bunch of lines that are no longer failing.
dpranke@chromium.org [Tue, 27 Mar 2012 22:21:11 +0000 (22:21 +0000)]
Remove a bunch of lines that are no longer failing.

Unreviewed, expectations change.

* platform/chromium/test_expectations.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@112324 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoHold a reference to refChild in insertBefore before calling collectChildrenAndRemoveF...
adamk@chromium.org [Tue, 27 Mar 2012 22:13:06 +0000 (22:13 +0000)]
Hold a reference to refChild in insertBefore before calling collectChildrenAndRemoveFromOldParent
https://bugs.webkit.org/show_bug.cgi?id=82377

Reviewed by Ryosuke Niwa.

Source/WebCore:

This fixes a regression from r111925.

Test: fast/dom/insertBefore-refChild-crash.html

* dom/ContainerNode.cpp:
(WebCore::ContainerNode::insertBefore): Move the 'next' RefPtr above the call to
collectChildrenAndRemoveFromOldParent and rename refChildPreviousSibling
to 'prev' (matching appendChild and replaceChild).

LayoutTests:

* fast/dom/insertBefore-refChild-crash-expected.txt: Added.
* fast/dom/insertBefore-refChild-crash.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@112323 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoImprove error reporting in run-javascriptcore-tests
simon.fraser@apple.com [Tue, 27 Mar 2012 22:07:18 +0000 (22:07 +0000)]
Improve error reporting in run-javascriptcore-tests
https://bugs.webkit.org/show_bug.cgi?id=82379

Reviewed by Jessie Berlin.

When the script dies, have it print out the reason for dying.

* Scripts/run-javascriptcore-tests:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@112322 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agocssText should not generate literal 'initial' in shorthand properties
rniwa@webkit.org [Tue, 27 Mar 2012 22:01:03 +0000 (22:01 +0000)]
cssText should not generate literal 'initial' in shorthand properties
https://bugs.webkit.org/show_bug.cgi?id=82364

Reviewed by Antti Koivisto.

Source/WebCore:

Fixed the bug by treating initial value as if the value is not set.
While this is incorrect for properties that inherits by default,
it's strictly better than generating unparsable value as we do today.

The proper fix is for CSSInitialValue::cssText to fetch the respective
default value from what's currently in CSSStyleSelector code but that requires
a considerable amount of refactoring and work.

* css/StylePropertySet.cpp:
(WebCore::StylePropertySet::getShorthandValue):
(WebCore::StylePropertySet::getCommonValue):

LayoutTests:

* fast/css/cssText-shorthand-expected.txt: Rebaselined.
* fast/css/cssText-shorthand.html:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@112321 268f45cc-cd09-0410-ab3c-d52691b4dbfc