WebKit GIT trees - WebKit-https.git/log

BUILD FIX (r144727): Change type of columnRange and rowRange to pair<unsigned, unsigned>
<http://webkit.org/b/110050>

Fixes the following build failures:

    Source/WebCore/accessibility/ios/WebAccessibilityObjectWrapperIOS.mm:741:30: error: non-const lvalue reference to type 'pair<unsigned int, unsigned int>' cannot bind to a value of unrelated type 'pair<int, int>'
        tableCell->rowIndexRange(rowRange);
                                 ^~~~~~~~
    In file included from Source/WebCore/accessibility/ios/WebAccessibilityObjectWrapperIOS.mm:33:
    Source/WebCore/accessibility/AccessibilityTableCell.h:47:58: note: passing argument to parameter 'rowRange' here
        virtual void rowIndexRange(pair<unsigned, unsigned>& rowRange);
                                                             ^
    Source/WebCore/accessibility/ios/WebAccessibilityObjectWrapperIOS.mm:742:33: error: non-const lvalue reference to type 'pair<unsigned int, unsigned int>' cannot bind to a value of unrelated type 'pair<int, int>'
        tableCell->columnIndexRange(columnRange);
                                    ^~~~~~~~~~~
    In file included from Source/WebCore/accessibility/ios/WebAccessibilityObjectWrapperIOS.mm:33:
    Source/WebCore/accessibility/AccessibilityTableCell.h:49:61: note: passing argument to parameter 'columnRange' here
        virtual void columnIndexRange(pair<unsigned, unsigned>& columnRange);
                                                                ^
    Source/WebCore/accessibility/ios/WebAccessibilityObjectWrapperIOS.mm:804:30: error: non-const lvalue reference to type 'pair<unsigned int, unsigned int>' cannot bind to a value of unrelated type 'pair<int, int>'
        tableCell->rowIndexRange(rowRange);
                                 ^~~~~~~~
    In file included from Source/WebCore/accessibility/ios/WebAccessibilityObjectWrapperIOS.mm:33:
    Source/WebCore/accessibility/AccessibilityTableCell.h:47:58: note: passing argument to parameter 'rowRange' here
        virtual void rowIndexRange(pair<unsigned, unsigned>& rowRange);
                                                             ^
    Source/WebCore/accessibility/ios/WebAccessibilityObjectWrapperIOS.mm:818:33: error: non-const lvalue reference to type 'pair<unsigned int, unsigned int>' cannot bind to a value of unrelated type 'pair<int, int>'
        tableCell->columnIndexRange(columnRange);
                                    ^~~~~~~~~~~
    In file included from Source/WebCore/accessibility/ios/WebAccessibilityObjectWrapperIOS.mm:33:
    Source/WebCore/accessibility/AccessibilityTableCell.h:49:61: note: passing argument to parameter 'columnRange' here
        virtual void columnIndexRange(pair<unsigned, unsigned>& columnRange);
                                                    ^
    4 errors generated.

* accessibility/ios/WebAccessibilityObjectWrapperIOS.mm:
(-[WebAccessibilityObjectWrapper accessibilityHeaderElements]):
(-[WebAccessibilityObjectWrapper accessibilityRowRange]):
(-[WebAccessibilityObjectWrapper accessibilityColumnRange]):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@144751 268f45cc-cd09-0410-ab3c-d52691b4dbfc

[BlackBerry] Clean up animated scroll/zoom code in WebPagePrivate
https://bugs.webkit.org/show_bug.cgi?id=111373

Patch by Andrew Lo <anlo@rim.com> on 2013-03-05
Reviewed by Rob Buis.

Internally reviewed by Jakob Petsovits.
Internal PR 285468.

Have WebPageClient pass the final scale and position to WebPage when a zoom/scroll
animation finishes, rather than keeping those values as instance variables in
WebPagePrivate.

Rename some functions & variables to reflect that they are used for both block zoom
and generic zoom/scroll animations.

* Api/WebPage.cpp:
(BlackBerry::WebKit::WebPagePrivate::WebPagePrivate):
(BlackBerry::WebKit::WebPagePrivate::zoomAnimationFinished):
(BlackBerry::WebKit::WebPage::zoomAnimationFinished):
(BlackBerry::WebKit::WebPagePrivate::resetBlockZoom):
(BlackBerry::WebKit::WebPage::blockZoom):
(BlackBerry::WebKit::WebPagePrivate::animateToScaleAndDocumentScrollPosition):
* Api/WebPage.h:
* Api/WebPageClient.h:
* Api/WebPage_p.h:
(WebPagePrivate):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@144750 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Web Inspector: [refactoring] set frame identifiers in timeline records early, do not keep them in event stack
https://bugs.webkit.org/show_bug.cgi?id=111345

Reviewed by Pavel Feldman.

* inspector/InspectorTimelineAgent.cpp:
(WebCore::InspectorTimelineAgent::pushGCEventRecords):
(WebCore::InspectorTimelineAgent::addRecordToTimeline):
(WebCore::InspectorTimelineAgent::innerAddRecordToTimeline):
(WebCore::InspectorTimelineAgent::setFrameIdentifier):
(WebCore):
(WebCore::InspectorTimelineAgent::didCompleteCurrentRecord):
(WebCore::InspectorTimelineAgent::appendRecord):
(WebCore::InspectorTimelineAgent::pushCurrentRecord):
(WebCore::InspectorTimelineAgent::commitFrameRecord):
* inspector/InspectorTimelineAgent.h:
(WebCore::InspectorTimelineAgent::TimelineRecordEntry::TimelineRecordEntry):
(TimelineRecordEntry):
(InspectorTimelineAgent):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@144749 268f45cc-cd09-0410-ab3c-d52691b4dbfc

MediaStream API: Add the getStreamById method on RTCPeerConnection
https://bugs.webkit.org/show_bug.cgi?id=111311

Reviewed by Adam Barth.

Source/Platform:

Added an initial initialize method which doesn't take an id;
the id will be generated for you.

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

Source/WebCore:

Adding the missing method on RTCPeerConnection defined here:
http://dev.w3.org/2011/webrtc/editor/webrtc.html#widl-RTCPeerConnection-getStreamById-MediaStream-DOMString-streamId

Existing tests have been extended to cover patch.

* Modules/mediastream/RTCPeerConnection.cpp:
(WebCore::RTCPeerConnection::getStreamById):
(WebCore):
* Modules/mediastream/RTCPeerConnection.h:
(RTCPeerConnection):
* Modules/mediastream/RTCPeerConnection.idl:
* platform/chromium/support/WebMediaStream.cpp:
(WebKit::WebMediaStream::initialize):
(WebKit):
* platform/mediastream/MediaStreamSource.cpp:
(WebCore::MediaStreamSource::setReadyState):

Tools:

Changing mocks to be more realistic.

* DumpRenderTree/chromium/TestRunner/src/MockWebRTCPeerConnectionHandler.cpp:
(WebTestRunner::MockWebRTCPeerConnectionHandler::addStream):
(WebTestRunner::MockWebRTCPeerConnectionHandler::removeStream):
* DumpRenderTree/chromium/TestRunner/src/WebUserMediaClientMock.cpp:
(WebTestRunner::WebUserMediaClientMock::requestUserMedia):

LayoutTests:

* fast/mediastream/MediaStream-add-remove-tracks.html:
* fast/mediastream/RTCPeerConnection-AddRemoveStream-expected.txt:
* fast/mediastream/RTCPeerConnection-AddRemoveStream.html:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@144748 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Web Inspector: allow retrieval of composited layers in a given DOM subtree
https://bugs.webkit.org/show_bug.cgi?id=111312

Source/WebCore:

Introduce the LayerTreeAgent.layersForNode(node) method allowing the front-end
to gather all composited layers associated with nodes in the subtree of which
the provided node is the root.

In order to gather the layers in the subtree, we first traverse the node's
renderer's RenderObject hierarchy and whenever we encounter a RenderObject
that has a RenderLayer, we then traverse that renderLayer's RenderLayer
hierarchy. This allows for a quick path through the relevant objects we're
gathering.

Layers gathered will push the node to which they're associated, allowing a
nodeId for this layer to be listed in the Layer object sent to the front-end.
It is crucial to be able to provide a nodeId as well as a layerId for a Layer
object in order to be able to correctly assess mutations in the layer tree.
For instance, it is expected that a node's layer be replaced by a new layer
to represent a slightly different rendering of its content, but the front-end
should be able to represent this layer as an object for which only certain
attributes have changed (like the "paintCount" property).

Layer objects also indicate whether they're associated to a node hosted in a
shadow tree (the optional "isInShadowTree" property) in order for the front-end
to be able to only show this layer if the option to show nodes hosted in shadow
tree is enabled.

Finally, since we're only gathering composited layers, we're removing the
"isLayerComposited" property and removing the optional flag on the "paintCount",
"memory" and "compositedBounds" properties.

Reviewed by Simon Fraser.

Test: inspector-protocol/layers/layers-for-node.html

* inspector/Inspector.json:
* inspector/InspectorLayerTreeAgent.cpp:
(WebCore::InspectorLayerTreeAgent::layersForNode):
(WebCore):
(WebCore::InspectorLayerTreeAgent::gatherLayersUsingRenderObjectHierarchy):
(WebCore::InspectorLayerTreeAgent::gatherLayersUsingRenderLayerHierarchy):
(WebCore::InspectorLayerTreeAgent::buildObjectForLayer):
(WebCore::InspectorLayerTreeAgent::idForNode):
* inspector/InspectorLayerTreeAgent.h:
(InspectorLayerTreeAgent):

LayoutTests:

New test for the LayerTree.layersForNode() method.

Reviewed by Simon Fraser.

* inspector-protocol/layers/layers-for-node-expected.txt: Added.
* inspector-protocol/layers/layers-for-node.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@144747 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Web Inspector: use timestamps consistent with rest of Timeline in WebSocket events.
https://bugs.webkit.org/show_bug.cgi?id=111344

- create web socket related records in a manner consistent with the rest of timeline.

Reviewed by Pavel Feldman.

* inspector/InspectorTimelineAgent.cpp:
(WebCore::InspectorTimelineAgent::didCreateWebSocket):
(WebCore::InspectorTimelineAgent::willSendWebSocketHandshakeRequest):
(WebCore::InspectorTimelineAgent::didReceiveWebSocketHandshakeResponse):
(WebCore::InspectorTimelineAgent::didDestroyWebSocket):
* inspector/InspectorTimelineAgent.h:
(InspectorTimelineAgent):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@144746 268f45cc-cd09-0410-ab3c-d52691b4dbfc

[WK2][EFL] Add requestCheckingOfString implementation.
https://bugs.webkit.org/show_bug.cgi?id=108172

Reviewed by Hajime Morrita.

Provide implementation of 'requestCheckingOfString' for WK2-EFL.
It uses the unified text checker feature.

* UIProcess/efl/TextCheckerEfl.cpp:
(WebKit::TextChecker::requestCheckingOfString):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@144745 268f45cc-cd09-0410-ab3c-d52691b4dbfc

[CSS Regions] Crash when MathML used in CSS Regions
https://bugs.webkit.org/show_bug.cgi?id=110686

Reviewed by David Hyatt.

Source/WebCore:

The MathML blocks trigger a children layout when computing the preferred widths. This happens to determine the preferred
logical heights of the children. When the layout reaches the line boxes computation the preferred width of the containing block
is requested. Because it wasn't computed, the layout of the children is started again to determine the preferred logical heights.
This causes an infinite recursion and triggers a stack overflow.

The patch introduces a new RAII utility class that disables fragmentation when the constructor is called and restores it
on the destructor. This class is used when computing the preferred height for the children of a MathML block.

Test: fast/regions/mathml-crash.html

* rendering/RenderObject.cpp:
(WebCore::RenderObject::setFlowThreadStateIncludingDescendants): Do not cross RenderFlowThread boundaries when updating the flow thread
state flag. The innermost flow threads need to manage their descendants flag values.
* rendering/RenderView.cpp:
(WebCore::FragmentationDisabler::FragmentationDisabler):
(WebCore):
(WebCore::FragmentationDisabler::~FragmentationDisabler):
* rendering/RenderView.h:
(FragmentationDisabler):
(WebCore):
* rendering/mathml/RenderMathMLBlock.cpp:
(WebCore::RenderMathMLBlock::computeChildrenPreferredLogicalHeights):

LayoutTests:

Add a test to verify regions and MathML do not crash.

* fast/regions/mathml-crash-expected.txt: Added.
* fast/regions/mathml-crash.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@144744 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Cleanup: Move HitTestLocation and HitTestResult into separate files.
https://bugs.webkit.org/show_bug.cgi?id=111294

Reviewed by Allan Sandfeld Jensen.

HitTestLocation and HitTestResult currently both live in
HitTestResult.h. That makes things more difficult to understand than
they should be, and violates the one-class-one-file principle that
WebKit generally follows. Happily, it's fairly trivial to fix.

This should have no visible change; it's strictly an internal
refactoring. It doesn't even touch any implementation code: we're
just moving bits from one place to another.

* CMakeLists.txt:
* GNUmakefile.list.am:
* Target.pri:
* WebCore.gypi:
* WebCore.vcproj/WebCore.vcproj:
* WebCore.xcodeproj/project.pbxproj:
    Hey, look: a new file! Let's add it to six different build systems! :)
* rendering/HitTestLocation.cpp: Added.
(WebCore):
(WebCore::HitTestLocation::HitTestLocation):
(WebCore::HitTestLocation::~HitTestLocation):
(WebCore::HitTestLocation::operator=):
(WebCore::HitTestLocation::move):
(WebCore::HitTestLocation::intersectsRect):
(WebCore::HitTestLocation::intersects):
(WebCore::HitTestLocation::rectForPoint):
* rendering/HitTestLocation.h: Added.
(WebCore):
(HitTestLocation):
(WebCore::HitTestLocation::point):
(WebCore::HitTestLocation::roundedPoint):
(WebCore::HitTestLocation::region):
(WebCore::HitTestLocation::isRectBasedTest):
(WebCore::HitTestLocation::isRectilinear):
(WebCore::HitTestLocation::boundingBox):
(WebCore::HitTestLocation::topPadding):
(WebCore::HitTestLocation::rightPadding):
(WebCore::HitTestLocation::bottomPadding):
(WebCore::HitTestLocation::leftPadding):
(WebCore::HitTestLocation::transformedPoint):
(WebCore::HitTestLocation::transformedRect):
* rendering/HitTestResult.cpp:
* rendering/HitTestResult.h:
    Move all this code from HitTestResult.* to HitTestLocation.*.
* rendering/RenderBlock.cpp:
    Include HitTestLocation.h.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@144743 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Refactor UserGestureIndicator::Token to be a top-level class
https://bugs.webkit.org/show_bug.cgi?id=111412

Reviewed by Nico Weber.

This will allow for forward declaring the class in a public WebKit
API header. The goal is to enable the embedder to hold on to a
UserGestureToken.

* dom/UserGestureIndicator.cpp:
(WebCore::UserGestureIndicator::UserGestureIndicator):
(WebCore::UserGestureIndicator::currentToken):
* dom/UserGestureIndicator.h:
(UserGestureToken):
(WebCore::UserGestureToken::~UserGestureToken):
(WebCore):
(UserGestureIndicator):
* page/DOMTimer.h:
(DOMTimer):
* page/EventHandler.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@144742 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Remove redundant code in Document::updateHoverActiveState.
https://bugs.webkit.org/show_bug.cgi?id=111303

Reviewed by Darin Adler.

Document::updateHoverActiveState currently looks for touchrelease events
and explictly clears out the hover state for all nodes between the
currently hovered node and the top of the hover chain. This is actually
redundant with the logic in the rest of the function; we can simplify by
setting the Element* we're working with to 0, which causes the later
loops to do the necessary work.

There should be no web-visible effect from this change; it should simply
make this function less complex.

Extracted from Allan Sandfeld Jensen's original patch to wkbug.com/98168

* dom/Document.cpp:
(WebCore::Document::updateHoverActiveState):
    Set innerElementInDocument to 0 rather than walking the hover chain
    and clearing it when we see a touchrelease event. The rest of the
    code in this function will have the same effect.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@144741 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Add content shell bots to garden-o-matic for general failurse
https://bugs.webkit.org/show_bug.cgi?id=111168

Reviewed by Adam Barth.

All but the WebKit (Content Shell) Android bot is stable enough. While
they're not green yet, there must not be any non-layout test failures on
them.

* BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/config.js:
(.):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@144739 268f45cc-cd09-0410-ab3c-d52691b4dbfc

[V8] Remove unused custom indexedPropertyGetter
https://bugs.webkit.org/show_bug.cgi?id=111409

Patch by Koji Hara <kojih@chromium.org> on 2013-03-05
Reviewed by Kentaro Hara.

Remove unused custom indexedPropertyGetter.
There are 9 custom indexedPropertyGetter, including an unused one(NamedNodeMap).
This commit assures all existing custom indexedPropertyGetter(for V8) are used.

No new tests. Simple refactoring.

* bindings/v8/custom/V8NamedNodeMapCustom.cpp: Remove indexedPropertyGetter

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@144738 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Unreviewed, EFL gardening.

Skip failing tests on EFL bots.

* platform/efl/TestExpectations:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@144737 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Unreviewed; updating test expectations.

fast/regions/seamless-iframe-flowed-into-regions.html is failing not only with debug build.

* platform/chromium/TestExpectations:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@144736 268f45cc-cd09-0410-ab3c-d52691b4dbfc

ShadowRoot needs guardRef() and guardDeref()
https://bugs.webkit.org/show_bug.cgi?id=109777

Reviewed by Dimitri Glazkov.

This change moves m_guardRefCount from Document to TreeScope,
which allows ShadowRoot to be guarded by guardRef() mechanism as
Document. After r137524, Node referes TreeScope instead of
Document. This is natural consequence of the change: It no longer
makes sense to guardRef() Document pointer from Node.

Detail:

- Document::m_guardRefCount and related funcdtions are moved to TreeScope
- Document::removedLastRef is factored out into TreeScope::removedLastRefToScope(),
  TreeScope::dispose() and Docuent::dispose(). ShadowRoot also got its own dispose() implementation.
- Moved guardRef() and guardDeref() calls to TreeScope and Node.
  Note that there are two "guarded" TreeScope references. One is
  Node::m_treeScope and another is TreeScope::m_parentTreeScope.
  The guarded-ref management is now encapsulated in these two classes.

No new tests. Covered by existing tests.

* WebCore.exp.in:
* dom/Document.cpp:
(WebCore::Document::Document):
(WebCore::Document::~Document):
(WebCore::Document::dispose): Extracted from removedLastRef()
* dom/Document.h:
(WebCore::Node::isTreeScope):
(WebCore::Node::Node):
* dom/DocumentFragment.cpp:
(WebCore::DocumentFragment::DocumentFragment): Remove ASSERT() and move it to ...
(WebCore::DocumentFragment::create): ... here, to allow NULL document from ShadowRoot.
* dom/Node.cpp:
(WebCore::Node::~Node):
(WebCore::Node::removedLastRef):
* dom/Node.h:
(WebCore::Node::setTreeScope):
* dom/Element.cpp:
(WebCore::Element::ensureAttr): This has been wrong and is fixed in this revision since the incorrectness is unveiled by this change.
* dom/ShadowRoot.cpp:
(WebCore::ShadowRoot::ShadowRoot): Passed NULL document to superclass. This aligns what Document is doing.
(WebCore::ShadowRoot::dispose): Added.
* dom/ShadowRoot.h:
(ShadowRoot):
* dom/TreeScope.cpp:
(SameSizeAsTreeScope):
(WebCore::TreeScope::TreeScope):
(WebCore::TreeScope::~TreeScope):
(WebCore::TreeScope::dispose): Added.
(WebCore::TreeScope::setParentTreeScope):
(WebCore::TreeScope::deletionHasBegun):
(WebCore::TreeScope::beginDeletion):
(WebCore::TreeScope::refCount): Added.
* dom/TreeScope.h: Turned m_rootNode to Node* from ContainerNode* for Node::isTreeScope to be inlined.
(WebCore::TreeScope::guardRef): Pulled up from Document.
(WebCore::TreeScope::guardDeref): Ditto.
(WebCore::TreeScope::hasGuardRefCount): Added to hide m_guardRefCount.
(WebCore::TreeScope::deletionHasBegun): Added.
(WebCore::TreeScope::beginDeletion): Added.
(WebCore::TreeScope::removedLastRefToScope): Pulled up from Document.
* dom/TreeScopeAdopter.cpp:
(WebCore::TreeScopeAdopter::moveTreeToNewScope):
(WebCore::TreeScopeAdopter::moveNodeToNewDocument):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@144735 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2013-03-04  Kinuko Yasuda  <kinuko@chromium.org>

        Unreviewed, adding crashing test to the expectations.

        * platform/chromium/TestExpectations:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@144734 268f45cc-cd09-0410-ab3c-d52691b4dbfc

[chromium] remove deprecated WebTestRunner::setFocus API
https://bugs.webkit.org/show_bug.cgi?id=111354

Reviewed by James Robinson.

* DumpRenderTree/chromium/TestRunner/public/WebTestDelegate.h:
(WebTestDelegate):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@144733 268f45cc-cd09-0410-ab3c-d52691b4dbfc

getPropertyValue for -webkit-text-stroke returns null, should compute the shorthand value
https://bugs.webkit.org/show_bug.cgi?id=111018

Patch by Uday Kiran <udaykiran@motorola.com> on 2013-03-04
Reviewed by Alexander Pavlov.

Source/WebCore:

Test: fast/css/getPropertyValue-webkit-text-stroke.html

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

LayoutTests:

Added test for getPropertyValue("-webkit-text-stroke").

* fast/css/getPropertyValue-webkit-text-stroke-expected.txt: Added.
* fast/css/getPropertyValue-webkit-text-stroke.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@144732 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Run the bindings generation tests on EWS and CQ bots
https://bugs.webkit.org/show_bug.cgi?id=111358

Reviewed by Adam Barth.

The tests are super fast and not flaky.

* Scripts/webkitpy/common/config/ports.py:
(DeprecatedPort.run_bindings_tests_command):
* Scripts/webkitpy/common/config/ports_mock.py:
(MockPort.run_webkit_tests_command):
(MockPort):
(MockPort.run_bindings_tests_command):
* Scripts/webkitpy/tool/commands/download_unittest.py:
* Scripts/webkitpy/tool/steps/runtests.py:
(RunTests.run):
* Scripts/webkitpy/tool/steps/runtests_unittest.py:
(RunTestsTest.test_webkit_run_unit_tests):
* Scripts/webkitpy/tool/steps/steps_unittest.py:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@144731 268f45cc-cd09-0410-ab3c-d52691b4dbfc

getPropertyValue for -webkit-columns returns null, should compute the shorthand value
https://bugs.webkit.org/show_bug.cgi?id=111011

Patch by Uday Kiran <udaykiran@motorola.com> on 2013-03-04
Reviewed by Alexander Pavlov.

Source/WebCore:

Test: fast/css/getPropertyValue-columns.html

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

LayoutTests:

Added test for getPropertyValue("-webkit-columns").

* fast/css/getPropertyValue-columns-expected.txt: Added.
* fast/css/getPropertyValue-columns.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@144730 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Web Inspector: Do not accumulate file system files, add them to workspace immediately
https://bugs.webkit.org/show_bug.cgi?id=111325

Reviewed by Pavel Feldman.

Source/WebCore:

* inspector/front-end/FileSystemProjectDelegate.js:
(WebInspector.FileSystemProjectDelegate.prototype.populate.fileLoaded):
(WebInspector.FileSystemProjectDelegate.prototype.populate):
* inspector/front-end/IsolatedFileSystem.js:
(WebInspector.IsolatedFileSystem.prototype.innerCallback):
(WebInspector.IsolatedFileSystem.prototype.requestFilesRecursive):

LayoutTests:

* inspector/file-system-project.html:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@144729 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Unreviewed Windows build fix attempt.

floor method is ambigious so explicitly specify the type.

* html/canvas/WebGLRenderingContext.cpp:
(WebCore::WebGLRenderingContext::validateCompressedTexFuncData):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@144728 268f45cc-cd09-0410-ab3c-d52691b4dbfc

AX: cellForColumnAndRow fails for tables with hidden table cells
https://bugs.webkit.org/show_bug.cgi?id=110050

Reviewed by Tim Horton.

Source/WebCore:

If a table had hidden cells, then accessibility code was being confused in a few ways.
1) The cellForColumnAndRow method would return the wrong information since that was
using the RenderTableSection to retrieve a cell, which did not have the same data as the AXTable
2) The way we were adding children made it impossible to determine column and row range because we
would skip rows that had hidden children
3) AccessibilityARIAGrid and AccessibilityTable were using different methods for cellForColumnAndRow

The fix does a few things to make things right:
1) Always add an accessibility row, even if there are no visible cells in that row.
2) Have one method for AXTable and AXARIAGrid for cellForColumnAndRow.
3) Change cellForColumnAndRow to query the accessibility children rather than the RenderTableSection in determining the row, col info.
4) cellForColumnAndRow should use unsigned values instead of int values.

Test: accessibility/table-with-hidden-head-section.html

* accessibility/AccessibilityARIAGrid.cpp:
(WebCore):
* accessibility/AccessibilityARIAGrid.h:
(AccessibilityARIAGrid):
* accessibility/AccessibilityARIAGridCell.cpp:
(WebCore::AccessibilityARIAGridCell::rowIndexRange):
(WebCore::AccessibilityARIAGridCell::columnIndexRange):
* accessibility/AccessibilityARIAGridCell.h:
(AccessibilityARIAGridCell):
* accessibility/AccessibilityTable.cpp:
(WebCore::AccessibilityTable::addChildren):
(WebCore::AccessibilityTable::cellForColumnAndRow):
* accessibility/AccessibilityTable.h:
(WebCore):
(AccessibilityTable):
* accessibility/AccessibilityTableCell.cpp:
(WebCore::AccessibilityTableCell::rowIndexRange):
(WebCore::AccessibilityTableCell::columnIndexRange):
* accessibility/AccessibilityTableCell.h:
(AccessibilityTableCell):
* accessibility/atk/WebKitAccessibleInterfaceTable.cpp:
(webkitAccessibleTableGetColumnAtIndex):
(webkitAccessibleTableGetRowAtIndex):
(webkitAccessibleTableGetColumnExtentAt):
(webkitAccessibleTableGetRowExtentAt):
(webkitAccessibleTableGetColumnHeader):
(webkitAccessibleTableGetRowHeader):
* accessibility/mac/WebAccessibilityObjectWrapperMac.mm:
(-[WebAccessibilityObjectWrapper accessibilityAttributeValue:]):
* rendering/RenderTableSection.h:
(RenderTableSection):
(WebCore::RenderTableSection::rowRendererAt):

Source/WebKit/chromium:

* src/WebAccessibilityObject.cpp:
(WebKit::WebAccessibilityObject::cellColumnIndex):
(WebKit::WebAccessibilityObject::cellColumnSpan):
(WebKit::WebAccessibilityObject::cellRowIndex):
(WebKit::WebAccessibilityObject::cellRowSpan):

LayoutTests:

* accessibility/table-with-hidden-head-section-expected.txt: Added.
* accessibility/table-with-hidden-head-section.html: Added.
* platform/chromium/TestExpectations:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@144727 268f45cc-cd09-0410-ab3c-d52691b4dbfc

getAttribute does not behave correctly for mixed-case attributes on HTML elements
https://bugs.webkit.org/show_bug.cgi?id=105713

Patch by Arpita Bahuguna <a.bah@samsung.com> on 2013-03-04
Reviewed by Andreas Kling.

Source/WebCore:

getAttribute() and getAttributeNode() APIs do not convert the
passed attribute name to lowercase before comparing against the
existing attributes.
The specification however states that the passed name should
be converted to ASCII lowercase before checking for the existence
of the given attribute. [www.w3.org/TR/domcore/#dom-element-getattribute]

Test: fast/dom/Element/getAttribute-case-insensitivity.html

* dom/Element.h:
(WebCore::ElementData::getAttributeItemIndex):
getAttributeItemIndex() accepts a bool param 'shouldIgnoreAttributeCase'
which specifies whether the attribute's case should be ignored
before comparison. But inspite of this param's value we still carry
out a case sensitive search.
Have modified the same to convert the passed attribute's name to
lowercase if 'shouldIgnoreAttributeCase' is true.

LayoutTests:

* fast/dom/Element/getAttribute-case-insensitivity-expected.txt: Added.
* fast/dom/Element/getAttribute-case-insensitivity.html: Added.
Layout test added for verifying that getAttribute() and getAttributeNode()
APIs convert the passed attribute name to lowercase before comparing
against the existing attributes.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@144726 268f45cc-cd09-0410-ab3c-d52691b4dbfc

[Chromium] Add runtime flag for font load events
https://bugs.webkit.org/show_bug.cgi?id=111296

Patch by Kunihiko Sakamoto <ksakamoto@chromium.org> on 2013-03-04
Reviewed by Kent Tamura.

Add a runtime feature flag for FontLoader interface (disabled by default),
and enable ENABLE_FONT_LOAD_EVENTS build flag for chromium.

Source/WebCore:

* bindings/generic/RuntimeEnabledFeatures.cpp:
(WebCore):
* bindings/generic/RuntimeEnabledFeatures.h:
(RuntimeEnabledFeatures):
(WebCore::RuntimeEnabledFeatures::setFontLoadEventsEnabled):
(WebCore::RuntimeEnabledFeatures::fontLoadEventsEnabled):

Source/WebKit/chromium:

* features.gypi:
* public/WebRuntimeFeatures.h:
(WebRuntimeFeatures):
* src/WebRuntimeFeatures.cpp:
(WebKit::WebRuntimeFeatures::enableFontLoadEvents):
(WebKit):
(WebKit::WebRuntimeFeatures::isFontLoadEventsEnabled):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@144725 268f45cc-cd09-0410-ab3c-d52691b4dbfc

[V8] Add LayoutTests for SerializedScriptValue Int32/Uint32 cases
https://bugs.webkit.org/show_bug.cgi?id=111364

Reviewed by Kentaro Hara.

A lack of layout tests for SerializedScriptValue special cases let
crbug.com/179280 slip in and out un-noticed. Rectify the situation.

* platform/chromium/fast/storage/serialized-script-value-expected.txt:
* platform/chromium/fast/storage/serialized-script-value.html:
* storage/indexeddb/structured-clone-expected.txt:
* storage/indexeddb/structured-clone.html:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@144724 268f45cc-cd09-0410-ab3c-d52691b4dbfc

WEBGL_compressed_texture_atc implementation
https://bugs.webkit.org/show_bug.cgi?id=110496

Patch by Brandon Jones <bajones@google.com> on 2013-03-04
Reviewed by Kenneth Russell.

Source/WebCore:

Test: webgl/conformance/extensions/webgl-compressed-texture-atc.html

Exposes the WEBGL_compressed_texture_atc extension, but is unverified since no desktop hardware supports
the format that I am aware of. Should enable mobile ports to expose the format, however.

* CMakeLists.txt:
* DerivedSources.make:
* DerivedSources.pri:
* GNUmakefile.list.am:
* Target.pri:
* WebCore.gypi:
* WebCore.xcodeproj/project.pbxproj:
* bindings/js/JSWebGLRenderingContextCustom.cpp:
(WebCore::toJS):
* bindings/v8/custom/V8WebGLRenderingContextCustom.cpp:
(WebCore::toV8Object):
* html/canvas/WebGLCompressedTextureATC.cpp: Copied from Source/WebCore/html/canvas/WebGLExtension.h.
(WebCore):
(WebCore::WebGLCompressedTextureATC::WebGLCompressedTextureATC):
(WebCore::WebGLCompressedTextureATC::~WebGLCompressedTextureATC):
(WebCore::WebGLCompressedTextureATC::getName):
(WebCore::WebGLCompressedTextureATC::create):
(WebCore::WebGLCompressedTextureATC::supported):
* html/canvas/WebGLCompressedTextureATC.h: Copied from Source/WebCore/html/canvas/WebGLExtension.h.
(WebCore):
(WebGLCompressedTextureATC):
* html/canvas/WebGLCompressedTextureATC.idl: Copied from Source/WebCore/html/canvas/WebGLExtension.h.
* html/canvas/WebGLExtension.h:
* html/canvas/WebGLRenderingContext.cpp:
(WebCore):
(WebCore::WebGLRenderingContext::paintRenderingResultsToCanvas):
(WebCore::WebGLRenderingContext::getExtension):
(WebCore::WebGLRenderingContext::getSupportedExtensions):
(WebCore::WebGLRenderingContext::validateCompressedTexFuncData):
* html/canvas/WebGLRenderingContext.h:
(WebCore):
(WebGLRenderingContext):
* platform/graphics/Extensions3D.h:

LayoutTests:

* webgl/conformance/extensions/webgl-compressed-texture-atc-expected.txt: Added.
* webgl/conformance/extensions/webgl-compressed-texture-atc.html: Added.
* webgl/resources/webgl_test_files/conformance/extensions/webgl-compressed-texture-atc.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@144723 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Web Inspector: Exceptions when WebInspector run in hosted mode.
https://bugs.webkit.org/show_bug.cgi?id=111326

Reviewed by Pavel Feldman.

Source/WebCore:

There is bug in commands code generation in InspectorBackend.js
As a consequence all protocol enums have wrong names and cause
exception when refererenced.

Currently protocol enums are referenced only from NetworkManager.

Test: inspector/inspector-backend-commands-generation.html

* inspector/front-end/InspectorBackend.js:
(InspectorBackendClass.prototype.loadFromJSONIfNeeded):
Use static method to generate code.
(InspectorBackendClass._generateCommands):
Extracted from loadFromJSONIfNeeded. Typo fixed.

LayoutTests:

Test runtime commands code generator output.

* inspector/inspector-backend-commands-generation-expected.txt: Added.
* inspector/inspector-backend-commands-generation.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@144722 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Source/Platform: Add audio hardware support for multi-channel by implementing AudioDestinationNode.maxChannelCount
https://bugs.webkit.org/show_bug.cgi?id=111385

Reviewed by James Robinson.

* chromium/public/Platform.h:
(WebKit::Platform::audioHardwareOutputChannels):

Source/WebCore: Add audio hardware support for multi-channel output by implementing AudioDestinationNode.maxChannelCount
https://bugs.webkit.org/show_bug.cgi?id=111385

Reviewed by James Robinson.

See specification details for .maxChannelCount:
https://dvcs.w3.org/hg/audio/raw-file/tip/webaudio/specification.html#AudioDestinationNode

* Modules/webaudio/AudioDestinationNode.h:
(WebCore::AudioDestinationNode::maxChannelCount):
* Modules/webaudio/AudioDestinationNode.idl:
* Modules/webaudio/AudioNode.h:
Implement .maxChannelCount attribute.

(AudioNode):
* Modules/webaudio/DefaultAudioDestinationNode.h:
* Modules/webaudio/DefaultAudioDestinationNode.cpp:
(WebCore::DefaultAudioDestinationNode::initialize):
(WebCore::DefaultAudioDestinationNode::createDestination):
(WebCore::DefaultAudioDestinationNode::enableInput):
Call to createDestination() now takes no argument.
Instead we maintain a member variable for the input device ID.
(WebCore::DefaultAudioDestinationNode::maxChannelCount):
(WebCore::DefaultAudioDestinationNode::setChannelCount):
Check port-specific hardware support for maxChannelCount(),
and allow setting the channel count based on what the hardware
supports.

* platform/audio/AudioDestination.h:
(AudioDestination):
* platform/audio/chromium/AudioDestinationChromium.cpp:
(WebCore::AudioDestination::maxChannelCount):
(WebCore::AudioDestinationChromium::render):
Call into chromium WebKit API: audioHardwareOutputChannels().

* platform/audio/gstreamer/AudioDestinationGStreamer.cpp:
(WebCore::AudioDestination::maxChannelCount):
Add basic implementation for maxChannelCount() only supporting stereo output.

* platform/audio/mac/AudioDestinationMac.cpp:
(WebCore::AudioDestination::maxChannelCount):
Add basic implementation for maxChannelCount() only supporting stereo output.
Later, the mac port can support more than this with some small improvements to AudioDestinationMac.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@144720 268f45cc-cd09-0410-ab3c-d52691b4dbfc

run-webkit-tests should support --profile and --profiler= just like run-perf-tests does
https://bugs.webkit.org/show_bug.cgi?id=111384

Reviewed by Ryosuke Niwa.

Sometimes I want to know why various LayoutTests are slow
without having to convert them into PerformanceTests in order
to use our fancy profiling. The Driver already knows how to
do the right thing, we just need to expose the options to the user.

Handling of these options is already unittested in the perftests_unittest.

* Scripts/webkitpy/layout_tests/run_webkit_tests.py:
(parse_args):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@144719 268f45cc-cd09-0410-ab3c-d52691b4dbfc

BUILD FIX (r144662): Import JSBase.h in WebFrameLoadDelegate.h
<http://webkit.org/b/111269>

Fixes the following build failure:

    In file included from Source/WebKit/ios/DefaultDelegates/WebDefaultFrameLoadDelegate.m:11:
    In file included from Source/WebKit/mac/WebView/WebFrameLoadDelegatePrivate.h:29:
    Source/WebKit/mac/WebView/WebFrameLoadDelegate.h:41:5: error: 'JSC_OBJC_API_ENABLED' is not defined, evaluates to 0 [-Werror,-Wundef]
    #if JSC_OBJC_API_ENABLED
        ^
    Source/WebKit/mac/WebView/WebFrameLoadDelegate.h:207:5: error: 'JSC_OBJC_API_ENABLED' is not defined, evaluates to 0 [-Werror,-Wundef]
    #if JSC_OBJC_API_ENABLED
        ^
    2 errors generated.

* WebView/WebFrameLoadDelegate.h: Include JSBase.h.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@144717 268f45cc-cd09-0410-ab3c-d52691b4dbfc

check-webkit-style says "Have to enable auto props in the subversion config file"
https://bugs.webkit.org/show_bug.cgi?id=107724

Patch by Alan Cutter <alancutter@chromium.org> on 2013-03-04
Reviewed by Tony Chang.

Added a standard svn config file for WebKit developers to use.
Updated the style and commit queue bot build scripts to load this file.

* EWSTools/GoogleComputeEngine/build-commit-queue.sh:
* EWSTools/GoogleComputeEngine/build-feeder-style-sheriffbot.sh:
* EWSTools/configure-svn-config.sh: Copied from Tools/EWSTools/GoogleComputeEngine/build-commit-queue.sh.
* svn-config: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@144716 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Rebaseline 4 virtual/softwarecompositing tests.

Unreviewed rebaseline of test expectations.

* platform/chromium-linux/platform/chromium/virtual/softwarecompositing/geometry/layer-due-to-layer-children-deep-expected.png:
* platform/chromium-linux/platform/chromium/virtual/softwarecompositing/geometry/layer-due-to-layer-children-expected.png:
* platform/chromium-linux/platform/chromium/virtual/softwarecompositing/geometry/vertical-scroll-composited-expected.png:
* platform/chromium-linux/platform/chromium/virtual/softwarecompositing/reflections/animation-inside-reflection-expected.png:
* platform/chromium-mac-lion/platform/chromium/virtual/softwarecompositing/geometry/layer-due-to-layer-children-deep-expected.png:
* platform/chromium-mac-lion/platform/chromium/virtual/softwarecompositing/geometry/layer-due-to-layer-children-expected.png:
* platform/chromium-mac-lion/platform/chromium/virtual/softwarecompositing/reflections/animation-inside-reflection-expected.png:
* platform/chromium-mac-snowleopard/platform/chromium/virtual/softwarecompositing/geometry/layer-due-to-layer-children-deep-expected.png:
* platform/chromium-mac-snowleopard/platform/chromium/virtual/softwarecompositing/geometry/layer-due-to-layer-children-expected.png:
* platform/chromium-mac-snowleopard/platform/chromium/virtual/softwarecompositing/reflections/animation-inside-reflection-expected.png:
* platform/chromium-mac/platform/chromium/virtual/softwarecompositing/geometry/layer-due-to-layer-children-deep-expected.png:
* platform/chromium-mac/platform/chromium/virtual/softwarecompositing/geometry/layer-due-to-layer-children-expected.png:
* platform/chromium-mac/platform/chromium/virtual/softwarecompositing/geometry/vertical-scroll-composited-expected.png:
* platform/chromium-mac/platform/chromium/virtual/softwarecompositing/reflections/animation-inside-reflection-expected.png:
* platform/chromium-win/platform/chromium/virtual/softwarecompositing/geometry/layer-due-to-layer-children-deep-expected.png:
* platform/chromium-win/platform/chromium/virtual/softwarecompositing/geometry/layer-due-to-layer-children-expected.png:
* platform/chromium-win/platform/chromium/virtual/softwarecompositing/geometry/vertical-scroll-composited-expected.png:
* platform/chromium-win/platform/chromium/virtual/softwarecompositing/reflections/animation-inside-reflection-expected.png:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@144715 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Background HTML parser can rewind the tokenizer after end-of-file
https://bugs.webkit.org/show_bug.cgi?id=111365

Reviewed by Eric Seidel.

Source/WebCore:

Prior to this patch, it was possible to call didFailSpeculation after
processing the end-of-file token because checkForSpeculationFailure
didn't zero out m_tokenizer in some control paths.

This patch renames checkForSpeculationFailure to validateSpeculations
and ensures that it always takes ownership of the main thread's
HTMLTokenizer.

This patch also adds a number of ASSERTs to make sure the parser state
machine stays in the correct configuration (e.g., that we don't have a
main thread tokenizer while we're supposed to be tokenizing on the
background thread).

Test: fast/parser/document-write-fighting-eof.html

* html/parser/BackgroundHTMLInputStream.cpp:
(WebCore::BackgroundHTMLInputStream::rewindTo):
* html/parser/BackgroundHTMLParser.cpp:
(WebCore::BackgroundHTMLParser::append):
* html/parser/HTMLDocumentParser.cpp:
(WebCore::HTMLDocumentParser::validateSpeculations):
(WebCore::HTMLDocumentParser::processParsedChunkFromBackgroundParser):
(WebCore::HTMLDocumentParser::pumpPendingSpeculations):
(WebCore::HTMLDocumentParser::resumeParsingAfterScriptExecution):
* html/parser/HTMLDocumentParser.h:
(HTMLDocumentParser):

LayoutTests:

* fast/parser/document-write-fighting-eof.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@144714 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Make sure that clearOwnerNode also clears StyleResolver references (via didMutate).
https://bugs.webkit.org/show_bug.cgi?id=109446

Patch by Tim 'mithro' Ansell <mithro@mithis.com> on 2013-03-04
Reviewed by Eric Seidel.

Source/WebCore:

Test: fast/css/stylesheet.innerHTML-calls-didmutate.html

* css/CSSStyleSheet.h:

LayoutTests:

* fast/css/stylesheet.innerHTML-calls-didmutate-expected.txt: Added.
* fast/css/stylesheet.innerHTML-calls-didmutate.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@144713 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Unreviewed. Rolled Chromium DEPS to r185990. Requested by
jamesr_ via sheriffbot.

Patch by Sheriff Bot <webkit.review.bot@gmail.com> on 2013-03-04

* DEPS:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@144712 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Unreviewd build fix after r144701.

* platform/EventTracer.cpp:
(WebCore::EventTracer::initialize):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@144711 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Fix some style issues in RenderLayerCompositor
https://bugs.webkit.org/show_bug.cgi?id=111369

Patch by Min Qin <qinmin@chromium.org> on 2013-03-04
Reviewed by Simon Fraser.

Just style fix, no test needed

* rendering/RenderLayerCompositor.cpp:
(WebCore::RenderLayerCompositor::frameViewDidScroll):
(WebCore::RenderLayerCompositor::requiresCompositingForPosition):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@144710 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Revert changes to RenderLayer::scrollByRecursively() that were added for gesture-scrolling
https://bugs.webkit.org/show_bug.cgi?id=111007

Reviewed by Simon Fraser.

Due to http://trac.webkit.org/changeset/144519 landing, the changes made to RenderLayer in
http://trac.webkit.org/changeset/142057,
http://trac.webkit.org/changeset/140177, and
http://trac.webkit.org/changeset/142195
are no longer necessary (as RenderLayer::scrollByRecursively() is not used for
gesture-scrolling) and should be reverted.

No change in behavior, so no new tests needed.

* rendering/RenderLayer.cpp:
(WebCore::RenderLayer::scrollByRecursively):
* rendering/RenderLayer.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@144709 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Unused Structure property tables waste 14MB on Membuster.
<http://webkit.org/b/110854>
<rdar://problem/13292104>

Reviewed by Geoffrey Garen.

Turn PropertyTable into a GC object and have Structure drop unpinned tables when marking.
14 MB progression on Membuster3.

* CMakeLists.txt:
* GNUmakefile.list.am:
* JavaScriptCore.gypi:
* JavaScriptCore.vcxproj/JavaScriptCore.vcxproj:
* JavaScriptCore.xcodeproj/project.pbxproj:
* Target.pri:

    Added PropertyTable.cpp.

* runtime/PropertyTable.cpp: Added.
(JSC::PropertyTable::create):
(JSC::PropertyTable::clone):
(JSC::PropertyTable::PropertyTable):
(JSC::PropertyTable::destroy):
(JSC::PropertyTable::~PropertyTable):
(JSC::PropertyTable::visitChildren):

    Moved marking of property table values here from Structure::visitChildren().

* runtime/WriteBarrier.h:
(JSC::WriteBarrierBase::get):

    Move m_cell to a local before using it multiple times. This avoids a multiple-access race when
    Structure::checkOffsetConsistency() is used in assertions on the main thread while a marking thread
    zaps the property table.

* runtime/Structure.h:
(JSC::Structure::materializePropertyMapIfNecessary):
(JSC::Structure::materializePropertyMapIfNecessaryForPinning):
* runtime/StructureInlines.h:
(JSC::Structure::propertyTable):

    Added a getter for the Structure's PropertyTable that ASSERTs GC currently isn't active.
    Because GC can zap an unpinned property table at any time, it's not entirely safe to access it.
    Renamed the variable itself to m_propertyTableUnsafe to force call sites into explaining themselves.

(JSC::Structure::putWillGrowOutOfLineStorage):
(JSC::Structure::checkOffsetConsistency):

    Moved these out of Structure.h to break header dependency cycle between Structure/PropertyTable.

* runtime/Structure.cpp:
(JSC::Structure::visitChildren):

    Null out m_propertyTable if the table is unpinned. This'll cause the table to get GC'd.

* runtime/JSGlobalData.h:
* runtime/JSGlobalData.cpp:
(JSC::JSGlobalData::JSGlobalData):

    Add a global propertyTableStructure.

* runtime/PropertyMapHashTable.h:
(PropertyTable):
(JSC::PropertyTable::createStructure):
(JSC::PropertyTable::copy):

    Make PropertyTable a GC object.

* runtime/Structure.cpp:
(JSC::Structure::dumpStatistics):
(JSC::Structure::materializePropertyMap):
(JSC::Structure::despecifyDictionaryFunction):
(JSC::Structure::addPropertyTransition):
(JSC::Structure::changePrototypeTransition):
(JSC::Structure::despecifyFunctionTransition):
(JSC::Structure::attributeChangeTransition):
(JSC::Structure::toDictionaryTransition):
(JSC::Structure::sealTransition):
(JSC::Structure::freezeTransition):
(JSC::Structure::preventExtensionsTransition):
(JSC::Structure::nonPropertyTransition):
(JSC::Structure::isSealed):
(JSC::Structure::isFrozen):
(JSC::Structure::flattenDictionaryStructure):
(JSC::Structure::pin):
(JSC::Structure::copyPropertyTable):
(JSC::Structure::copyPropertyTableForPinning):
(JSC::Structure::get):
(JSC::Structure::despecifyFunction):
(JSC::Structure::despecifyAllFunctions):
(JSC::Structure::putSpecificValue):
(JSC::Structure::remove):
(JSC::Structure::createPropertyMap):
(JSC::Structure::getPropertyNamesFromStructure):
(JSC::Structure::checkConsistency):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@144708 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Unreviewed build fix after r144701.

* platform/EventTracer.cpp:
(WebCore::EventTracer::initialize):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@144707 268f45cc-cd09-0410-ab3c-d52691b4dbfc

REGRESSION (r143643): Buttons containing floats render differently
https://bugs.webkit.org/show_bug.cgi?id=110933

Patch by Christian Biesinger <cbiesinger@chromium.org> on 2013-03-04
Reviewed by Ojan Vafai.

Source/WebCore:

Test: fast/forms/button-with-float.html

* rendering/RenderBlock.cpp:
(WebCore::RenderBlock::expandsToEncloseOverhangingFloats):
Also include new flexbox as a renderer that needs to enclose
overhanging floats. Flex boxes establish a block formatting context,
so they need to contain floats:
http://dev.w3.org/csswg/css3-flexbox/#flex-items

LayoutTests:

* css3/flexbox/float-inside-flexitem-expected.html: Added.
* css3/flexbox/float-inside-flexitem.html: Added.
* fast/forms/button-with-float-expected.html: Added.
* fast/forms/button-with-float.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@144706 268f45cc-cd09-0410-ab3c-d52691b4dbfc

[chromium] Switch to new webkit_support::CreateLayerTreeView interface
https://bugs.webkit.org/show_bug.cgi?id=111285

Patch by James Robinson <jamesr@chromium.org> on 2013-03-04
Reviewed by Adrienne Walker.

This removes the need for WebCompositorSupport::initialize and exposes
slightly fewer entry points across the repository boundary.

* DumpRenderTree/chromium/TestShell.cpp:
(TestShell::initialize):
* DumpRenderTree/chromium/TestShell.h:
(TestShell::webCompositorThread):
(TestShell):
* DumpRenderTree/chromium/WebViewHost.cpp:
(WebViewHost::initializeLayerTreeView):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@144705 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Add PerformanceTest for document startup via iframe append/remove
https://bugs.webkit.org/show_bug.cgi?id=111389

Reviewed by Kentaro Hara.

We'd like pages with lots of iframes to be fast.
This test shows that at least for V8/Chromium we spend
all our time booting v8.

On my 32-core linux machine I see about 70 runs/sec in
Release and 5 runs/sec in Debug.

* DOM/iframe-append-remove.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@144704 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Support WebSpeech - Speech Synthesis
https://bugs.webkit.org/show_bug.cgi?id=106742

Build fix.

* Modules/speech/SpeechSynthesis.cpp:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@144703 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Unreviewed, rolling out r144595.
http://trac.webkit.org/changeset/144595
https://bugs.webkit.org/show_bug.cgi?id=90341

Causing multiple crashes in fast/dom/Element/* tests (hits
assert in Attr.cp:215)

Source/WebCore:

* dom/Element.cpp:
(WebCore::Element::setAttributeNode):

LayoutTests:

* fast/dom/Element/setAttributeNode-case-insensitivity-expected.txt: Removed.
* fast/dom/Element/setAttributeNode-case-insensitivity.html: Removed.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@144702 268f45cc-cd09-0410-ab3c-d52691b4dbfc

[chromium] Implement TRACE_EVENT macros for sampling profiling
https://bugs.webkit.org/show_bug.cgi?id=110932

Reviewed by Adam Barth.

We are implementing TRACE_EVENT macros for sampling profiling.
It works in the following mechanism:

- Chromium defines global state variables for sampling profiling.
(i.e. g_trace_state0, g_trace_state1, g_trace_state2 in trace_event.h)

- WebKit gets the addresses of the global state variables at the
initialization step. (i.e. EventTracer::initialize())

- WebKit updates the global states by using TRACE_EVENT_SAMPLING_STATE()
macros every time WebKit changes its state. (e.g. DOM attribute
getters/setters/methods)

- A sampling thread running in Chrome reads the global states
periodically and visualizes the profiling results into about://tracing.

This patch implements (1) a WebKit API to get the addresses of
the global states and (2) the TRACE_EVENT_SAMPLING_STATE() macro.

Source/Platform:

* chromium/public/Platform.h:
(Platform):
(WebKit::Platform::getTraceSamplingState):

Source/WebCore:

No tests. The sampling profiler is not enabled in Chrome by default.
I manually inserted TRACE_EVENT_SAMPLING_STATE() macros and
confirmed that the sampling profiling is working as expected.

* platform/EventTracer.cpp:
(WebCore::EventTracer::initialize):
(WebCore::EventTracer::getTraceCategoryEnabledFlag):
(WebCore):
* platform/EventTracer.h:
(WebCore):
(EventTracer):
* platform/chromium/EventTracerChromium.cpp:
(WebCore):
(WebCore::EventTracer::initialize):
* platform/chromium/TraceEvent.h:

Source/WebKit/chromium:

* src/WebKit.cpp:
(WebKit::initializeWithoutV8):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@144701 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Unreviewed gardening.

* platform/chromium/TestExpectations: touch-gesture-noscroll-body-* are failing on mac.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@144700 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Unreviewed. Get rid of some AppleWin VS2010 build warnings.

* DumpRenderTree/DumpRenderTree.vcxproj/TestNetscapePlugin/TestNetscapePluginCommon.props:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@144699 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Support WebSpeech - Speech Synthesis
https://bugs.webkit.org/show_bug.cgi?id=106742

Reviewed by Simon Fraser.

Source/JavaScriptCore:

Enable speech synthesis for the Mac.

* Configurations/FeatureDefines.xcconfig:

Source/WebCore:

Enable speech synthesis for the Mac.

* Configurations/FeatureDefines.xcconfig:

Source/WebKit/mac:

Enable speech synthesis for the Mac.

* Configurations/FeatureDefines.xcconfig:

Source/WebKit2:

Enable speech synthesis for the Mac.

* Configurations/FeatureDefines.xcconfig:

LayoutTests:

Unskip speech synthesis tests.

* fast/dom/script-tests/constructed-objects-prototypes.js:
(constructorPropertiesOnWindow):
* platform/mac/TestExpectations:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@144698 268f45cc-cd09-0410-ab3c-d52691b4dbfc

BUILD FIX (r144657): Fix DumpRenderTree build failure on iOS

    Tools/DumpRenderTree/ios/AccessibilityControllerIOS.mm:95:29: error: class method '+stringWithJSStringRef:' not found (return type defaults to 'id') [-Werror,-Wobjc-method-access]
        NSString *idAttribute = [NSString stringWithJSStringRef:idAttributeRef];
                                ^         ~~~~~~~~~~~~~~~~~~~~~
    1 error generated.

* DumpRenderTree/ios/AccessibilityControllerIOS.mm: Import
AccessibilityCommonMac.h.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@144697 268f45cc-cd09-0410-ab3c-d52691b4dbfc

UI process storage areas should hold on to storage maps
https://bugs.webkit.org/show_bug.cgi?id=111374

Reviewed by Beth Dakin.

Pass the session storage quota size through to the storage area constructor and
create a StorageMap with the given quota size.

* UIProcess/Storage/StorageManager.cpp:
(StorageManager::StorageArea):
(WebKit::StorageManager::StorageArea::create):
(WebKit::StorageManager::StorageArea::StorageArea):
(StorageManager::SessionStorageNamespace):
(WebKit::StorageManager::SessionStorageNamespace::create):
(WebKit::StorageManager::SessionStorageNamespace::SessionStorageNamespace):
(WebKit::StorageManager::SessionStorageNamespace::getOrCreateStorageArea):
(WebKit::StorageManager::createSessionStorageNamespace):
(WebKit::StorageManager::createSessionStorageNamespaceInternal):
* UIProcess/Storage/StorageManager.h:
(StorageManager):
* UIProcess/WebPageProxy.cpp:
(WebKit::WebPageProxy::WebPageProxy):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@144695 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Cache partitioning does not affect iframe MainResources
https://bugs.webkit.org/show_bug.cgi?id=111022

Reviewed by David Kilzer.

Source/WebCore:

Set the partition of a main resource request if in a subframe.

Test: http/tests/cache/partitioned-cache-iframe.html

* loader/FrameLoader.cpp:
(WebCore::FrameLoader::loadURL):

LayoutTests:

These tests results are for the case in which the cache is not
partitioned, as no shipping ports have cache partitioning enabled

* http/tests/cache/partitioned-cache-iframe-expected.txt: Added.
* http/tests/cache/partitioned-cache-iframe.html: Added.
* http/tests/cache/resources/random-cached-iframe.cgi: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@144694 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Unreviewed build fix after r144675.

Work around picky/old C++ compilers by separating the '>>' in
nested templates with a space.

* UIProcess/Storage/StorageManager.cpp:
(StorageManager::StorageArea):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@144693 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Unreviewed gardening.

* platform/chromium/TestExpectations: Marking all html5lib/generated/ as crashing.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@144692 268f45cc-cd09-0410-ab3c-d52691b4dbfc

More work on UI side storage
https://bugs.webkit.org/show_bug.cgi?id=111370

Reviewed by Andreas Kling.

Add a DispatchStorageEvent message and begin hooking it up.

* UIProcess/Storage/StorageManager.cpp:
(StorageManager::StorageArea):
(WebKit::StorageManager::StorageArea::setItem):
(WebKit):
(WebKit::StorageManager::StorageArea::dispatchEvents):
(WebKit::StorageManager::setItem):
(WebKit::StorageManager::findStorageArea):
* UIProcess/Storage/StorageManager.h:
(StorageManager):
* UIProcess/Storage/StorageManager.messages.in:
* WebProcess/Storage/StorageAreaProxy.cpp:
(WebKit::StorageAreaProxy::setItem):
(WebKit::StorageAreaProxy::dispatchStorageEvent):
(WebKit):
* WebProcess/Storage/StorageAreaProxy.h:
(StorageAreaProxy):
* WebProcess/Storage/StorageAreaProxy.messages.in:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@144691 268f45cc-cd09-0410-ab3c-d52691b4dbfc

AX: user agent shadow roots for plugins should be 'accessible'
https://bugs.webkit.org/show_bug.cgi?id=111339

Reviewed by Tim Horton.

If a plugin element uses a user agent shadow element, we should make that shadow element
into a button with the title and subtitle put together, and use that as the label. This
will probably be the best accessibility experience.

* html/HTMLPlugInImageElement.cpp:
(WebCore::HTMLPlugInImageElement::didAddUserAgentShadowRoot):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@144690 268f45cc-cd09-0410-ab3c-d52691b4dbfc

CSS3 calc: identify divide by zero cases at parse time
https://bugs.webkit.org/show_bug.cgi?id=80411

Patch by Alan Cutter <alancutter@chromium.org> on 2013-03-04
Reviewed by Tony Chang.

Added tests to check whether divide by 0 cases are caught.

* css3/calc/catch-divide-by-0-expected.txt: Added.
* css3/calc/catch-divide-by-0.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@144689 268f45cc-cd09-0410-ab3c-d52691b4dbfc

[chromium] Form controls on composited layers are not painted properly
https://bugs.webkit.org/show_bug.cgi?id=111366

Reviewed by James Robinson.

Set PlatformContextSkia::m_drawingToImageBuffer which is used as a hint to properly blend GDI-painted bitmap.

* platform/graphics/chromium/OpaqueRectTrackingContentLayerDelegate.cpp:
(WebCore::OpaqueRectTrackingContentLayerDelegate::paintContents):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@144688 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Give ViewportConstraints classes copy constructors
https://bugs.webkit.org/show_bug.cgi?id=111367

Reviewed by Beth Dakin.

Downstream code requires that ViewportConstraints objects have
copy constructors, so add them.

Also clean up some redundant initialization of stack objects
in RenderLayerCompositor.

No behavior change.

* page/scrolling/ScrollingConstraints.h:
(WebCore::ViewportConstraints::ViewportConstraints):
(ViewportConstraints):
(WebCore::FixedPositionViewportConstraints::FixedPositionViewportConstraints):
(FixedPositionViewportConstraints):
(WebCore::StickyPositionViewportConstraints::StickyPositionViewportConstraints):
(StickyPositionViewportConstraints):
* rendering/RenderLayerCompositor.cpp:
(WebCore::RenderLayerCompositor::computeFixedViewportConstraints):
(WebCore::RenderLayerCompositor::computeStickyViewportConstraints):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@144687 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Unreviewed. AppleWin build fix.

* html/parser/XSSAuditor.cpp:
(WebCore::isSemicolonSeparatedAttribute):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@144683 268f45cc-cd09-0410-ab3c-d52691b4dbfc

CSSParser: Avoid temporary String when parsing hex colors.
<http://webkit.org/b/111341>

Reviewed by Anders Carlsson.

Templatize CSSParser::fastParseColor() so we can pass a CSSParserString as the color name.
This avoids creating a temporary WTF::String from the parser string.

* css/CSSParser.cpp:
(WebCore::CSSParser::fastParseColor):
* css/CSSParser.h:
* css/CSSParserValues.h:
(CSSParserString):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@144682 268f45cc-cd09-0410-ab3c-d52691b4dbfc

[CSS Grid Layout] Add parsing for grid-{start|before}
https://bugs.webkit.org/show_bug.cgi?id=110988

Reviewed by Tony Chang.

Source/WebCore:

Test: fast/css-grid-layout/grid-item-start-before-get-set.html

Refactored how we store internally grid-{row|column} to be based on grid-{before|start}.
This is required as grid-{row|column} are shorthands for resp. grid-{before|after} and grid-{start|end}
in the specification and will also be in the code. That means a lot of methods were renamed to be more
aligned with the new design.

* css/CSSComputedStyleDeclaration.cpp:
(WebCore::CSSComputedStyleDeclaration::getPropertyCSSValue):
* css/CSSParser.cpp:
(WebCore::CSSParser::parseValue):
* css/CSSProperty.cpp:
(WebCore::CSSProperty::isInheritedProperty):
* css/CSSPropertyNames.in:
* css/StyleResolver.cpp:
(WebCore::StyleResolver::applyProperty):
Added the code to parse, apply and get back the new properties. Also
some renaming after the internal naming change below.

* rendering/RenderGrid.cpp:
(WebCore::RenderGrid::maximumIndexInDirection):
(WebCore::RenderGrid::placeItemsOnGrid):
(WebCore::RenderGrid::autoPlacementMajorAxisPositionForChild):
(WebCore::RenderGrid::autoPlacementMinorAxisPositionForChild):
* rendering/style/RenderStyle.h:
* rendering/style/StyleGridItemData.cpp:
(WebCore::StyleGridItemData::StyleGridItemData):
* rendering/style/StyleGridItemData.h:
(WebCore::StyleGridItemData::operator==):
Updated the naming now that we manipulate grid-{before|start} instead of grid-{row|column}.

LayoutTests:

* fast/css-grid-layout/grid-item-column-row-get-set-expected.txt:
* fast/css-grid-layout/grid-item-column-row-get-set.html:
Extended the test to make sure that both grid-column (resp. grid-row) and grid-start (resp. grid-before) are
in sync.

* fast/css-grid-layout/grid-item-start-before-get-set-expected.txt: Added.
* fast/css-grid-layout/grid-item-start-before-get-set.html: Added
Add this test for grid-{start|before}, similar to LayoutTests/fast/css-grid-layout/grid-item-column-row-get-set.html.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@144681 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Unreviewed build fix: wrap references to PluginInfoStore with
guards for platforms which do not enable NETSCAPE_PLUGIN_AP.

* UIProcess/WebContext.cpp:
(WebKit::WebContext::WebContext):
(WebKit::WebContext::~WebContext):
* UIProcess/WebContext.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@144680 268f45cc-cd09-0410-ab3c-d52691b4dbfc

WebSpeech: change voiceURI to voice
https://bugs.webkit.org/show_bug.cgi?id=110773

Reviewed by Tim Horton.

Source/WebCore:

Change the voiceURI property to a voice property that takes a SpeechSynthesisVoice.
This was recently added to the errata of the WebSpeech spec.

Test: platform/mac/fast/speechsynthesis/speech-synthesis-utterance-uses-voice.html

* Modules/speech/SpeechSynthesisUtterance.cpp:
(WebCore):
(WebCore::SpeechSynthesisUtterance::voice):
(WebCore::SpeechSynthesisUtterance::setVoice):
* Modules/speech/SpeechSynthesisUtterance.h:
(SpeechSynthesisUtterance):
* Modules/speech/SpeechSynthesisUtterance.idl:
* Modules/speech/SpeechSynthesisVoice.h:
(WebCore::SpeechSynthesisVoice::platformVoice):
(SpeechSynthesisVoice):
* platform/PlatformSpeechSynthesisUtterance.cpp:
(WebCore::PlatformSpeechSynthesisUtterance::PlatformSpeechSynthesisUtterance):
* platform/PlatformSpeechSynthesisUtterance.h:
(WebCore::PlatformSpeechSynthesisUtterance::voice):
(WebCore::PlatformSpeechSynthesisUtterance::setVoice):
(PlatformSpeechSynthesisUtterance):
* platform/mac/PlatformSpeechSynthesizerMac.mm:
(-[WebSpeechSynthesisWrapper speakUtterance:WebCore::]):

LayoutTests:

* platform/mac/fast/speechsynthesis/speech-synthesis-utterance-uses-voice-expected.txt: Added.
* platform/mac/fast/speechsynthesis/speech-synthesis-utterance-uses-voice.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@144679 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Filename passed to FormData.append() takes precedence over filename in
File.
https://bugs.webkit.org/show_bug.cgi?id=111255

Patch by Victor Costan <costan@gmail.com> on 2013-03-04
Reviewed by Alexey Proskuryakov.

Source/WebCore:

Test: http/tests/local/formdata/send-form-data-with-filename.html

* platform/network/FormData.cpp:
(WebCore::FormData::appendKeyValuePairItems): changed to implement the
behavior in the XMLHttpRequest W3C spec.

LayoutTests:

(runTest):
* http/tests/local/formdata/send-form-data-with-filename-expected.txt: Added.
* http/tests/local/formdata/send-form-data-with-filename.html: Added.
* http/tests/local/formdata/send-form-data-with-sliced-file-expected.txt: Updated
to reflect the behavior specified in the XMLHttpRequest W3C spec.
* platform/wk2/TestExpectations: newly added test doesn't work in
WebKit2 because beginDragWithFiles is not implemented

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@144677 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Create and destroy storage areas in the UI process
https://bugs.webkit.org/show_bug.cgi?id=111361

Reviewed by Sam Weinig.

* UIProcess/Storage/StorageManager.cpp:
(StorageManager::StorageArea):
Keep track of listener connection and storage area ID pairs.

(WebKit::StorageManager::StorageArea::~StorageArea):
Assert that we don't have any listeners left.

(WebKit::StorageManager::StorageArea::addListener):
Add the pair to the set of listeners.

(WebKit::StorageManager::StorageArea::removeListener):
Remove the pair from the set of listeners.

(WebKit::StorageManager::SessionStorageNamespace::getOrCreateStorageArea):
Given an origin, look up or create the storage area.

(WebKit::StorageManager::createStorageArea):
Create the storage area if it doesn't already exist.

(WebKit::StorageManager::destroyStorageArea):
Look up the storage area and destroy it.

* UIProcess/Storage/StorageManager.h:
Add a map of open storage areas.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@144675 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Canvas and DOM go out of sync
https://bugs.webkit.org/show_bug.cgi?id=107701

Patch by Alexis Hetu <sugoi@chromium.org> on 2013-03-04
Reviewed by Simon Fraser.

Source/WebCore:

Test: fast/canvas/canvas-overflow-hidden-animation.html

Added an update of the current compositor from the layer's stacking
container when the layer itself isn't a stacking container.

* rendering/RenderLayerBacking.cpp:
(WebCore::RenderLayerBacking::updateAfterLayout):
Added an update of the stacking container of the owning layer of the
layer backing when the owning layer isn't a stacking container itself.

LayoutTests:

Added a new layout test for moving an object (a div) within a div that
has the "overflow:hidden" property.

* fast/canvas/canvas-overflow-hidden-animation-expected.html: Added.
* fast/canvas/canvas-overflow-hidden-animation.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@144674 268f45cc-cd09-0410-ab3c-d52691b4dbfc

[chromium] Roll chromium DEPS 185662 -> 185963

Roll roll roll the deps
Gently from downstream.
Merrily merrily merrily merrily
Green bots are but a dream.

* DEPS:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@144673 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Add API to allow WK2 clients to query the list of installed plug-ins.
https://bugs.webkit.org/show_bug.cgi?id=111245

Reviewed by Alexey Proskuryakov.

Add API in WKContext, implemented by WebContext, that allows callers to register a callback
retrieve an array containing installed plugin information. This API is exposed through WebContext
because it owns the PluginInfoStore used to fulfill the request.

Bump the WKContextClient API by 1:
* Shared/APIClientTraits.cpp:
* Shared/APIClientTraits.h:
* UIProcess/API/C/WKContext.h:

Add a client protocol to PluginInfoStore to notify the client when plugIn information has been
successfully loaded:
* UIProcess/Plugins/PluginInfoStore.cpp:
(WebKit::PluginInfoStore::PluginInfoStore): Initialize m_client to 0.
(WebKit::PluginInfoStore::loadPluginsIfNecessary): If the client is present, notify after loading the plugin store.
* UIProcess/Plugins/PluginInfoStore.h:
(WebKit::PluginInfoStoreClient::~PluginInfoStoreClient): Default destructor.
(WebKit::PluginInfoStoreClient::PluginInfoStoreClient): Default constructor.
(WebKit::PluginInfoStore::setClient): Simple setter.
(WebKit::PluginInfoStore::client): Simple getter.

* UIProcess/WebContext.cpp:
(WebKit::WebContext::WebContext): Set self as the client of PluginInfoStore.
(WebKit::WebContext::~WebContext): Clear the client of PluginInfoStore.
(WebKit::WebContext::pluginInfoStoreDidLoadPlugins): Pass the callback to own client.
* UIProcess/WebContext.h:
* UIProcess/WebContextClient.cpp:
(WebKit::WebContextClient::plugInInformationBecameAvailable): Pass the callback to the registered WK callback, if present.
* UIProcess/WebContextClient.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@144672 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Small BlobResourceHandle cleanup
https://bugs.webkit.org/show_bug.cgi?id=111352

Reviewed by Sam Weinig.

1. Don't use PassRefPtr<BlobStorageData> when ownership is not being passed.
2. Added OVERRIDE to some functions.
3. Added const to ResourceHandle::firstRequest().

* platform/network/BlobRegistryImpl.cpp:
(WebCore::loadResourceSynchronously):
(WebCore::BlobRegistryImpl::createResourceHandle):
(WebCore::BlobRegistryImpl::getBlobDataFromURL):
* platform/network/BlobRegistryImpl.h:
(BlobRegistryImpl):
* platform/network/BlobResourceHandle.cpp:
(WebCore::BlobResourceHandle::createAsync):
(WebCore::BlobResourceHandle::loadResourceSynchronously):
* platform/network/BlobResourceHandle.h:
(BlobResourceHandle):
* platform/network/FormData.cpp:
(WebCore::appendBlobResolved):
* platform/network/ResourceHandle.cpp:
(WebCore::ResourceHandle::firstRequest):
* platform/network/ResourceHandle.h:
(ResourceHandle):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@144671 268f45cc-cd09-0410-ab3c-d52691b4dbfc

PDFPlugin: Hook up Services
https://bugs.webkit.org/show_bug.cgi?id=111246
<rdar://problem/13062672>

Reviewed by Alexey Proskuryakov.

Make the Application->Services menu work when the focused selection is a PDFPlugin.

* Shared/EditorState.cpp:
(WebKit::EditorState::encode):
(WebKit::EditorState::decode):
* Shared/EditorState.h:
(WebKit::EditorState::EditorState):
(EditorState):
Add isInPlugin property to EditorState.

* UIProcess/API/mac/WKView.mm:
(-[WKView validRequestorForSendType:returnType:]):
If the selection is currently in a Plugin, we only currently have the ability
to retrieve the selection as a plain-text string, so restrict sendType to NSStringPboardType.

* WebProcess/Plugins/PDF/SimplePDFPlugin.h:
* WebProcess/Plugins/Plugin.h:
* WebProcess/Plugins/PluginView.h:
* WebProcess/Plugins/PluginProxy.h:
* WebProcess/Plugins/Netscape/NetscapePlugin.h:
Add getStringSelection.

* WebProcess/Plugins/PDF/PDFLayerControllerDetails.h:
Add pdfLayerController:didChangeSelection: delegate method.

* WebProcess/Plugins/PDF/PDFPlugin.h:
(PDFPlugin):
Add getStringSelection and notifySelectionChanged.

* WebProcess/Plugins/PDF/PDFPlugin.mm:
(-[WKPDFLayerControllerDelegate pdfLayerController:didChangeSelection:]):
Forward selection change notifications to our PDFPlugin.

(WebKit::PDFPlugin::notifySelectionChanged):
Forward selection change notifications to our WebPage.

(WebKit::PDFPlugin::getStringSelection):
Retrieve PDFLayerController's selection as a plain text string.

* WebProcess/Plugins/PluginView.cpp:
(WebKit::PluginView::countFindMatches):
(WebKit::PluginView::findString):
Drive-bys, check for existance and initialization of the plugin before using it.

(WebKit::PluginView::getStringSelection): Added.

* WebProcess/WebCoreSupport/WebEditorClient.cpp:
(WebKit::WebEditorClient::respondToChangedSelection):
Use didChangeSelection instead of sending the EditorStateChanged message directly.

* WebProcess/WebPage/WebPage.cpp:
(WebKit::WebPage::editorState):
If a plugin is focused and has a selection, return a EditorState that reflects that.
Since PDFs aren't editable (when editing annotations, the plugin doesn't have focus),
I'm not adding support for the editing-related properties when a plugin has focus.

(WebKit::WebPage::focusedPluginViewForFrame):
(WebKit::WebPage::pluginViewForFrame):
Make these class methods instead of static functions so we can use them from
WebPageMac too, instead of duplicating code there.

(WebKit::WebPage::didChangeSelection): Added.

* WebProcess/WebPage/WebPage.h:
Add didChangeSelection and [focused]PluginViewForFrame.

* WebProcess/WebPage/mac/WebPageMac.mm:
(WebKit::WebPage::getStringSelectionForPasteboard):
Defer to the focused plugin (if it exists) when retrieving the plain-text selection.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@144670 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Web Inspector: Introduce WebInspector.EventTarget interface for interfaces that dispatch events.
https://bugs.webkit.org/show_bug.cgi?id=111320

Reviewed by Pavel Feldman.

* inspector/front-end/Object.js:
(WebInspector.EventTarget):
(WebInspector.EventTarget.prototype.addEventListener):
(WebInspector.EventTarget.prototype.removeEventListener):
(WebInspector.EventTarget.prototype.removeAllListeners):
(WebInspector.EventTarget.prototype.hasEventListeners):
(WebInspector.EventTarget.prototype.dispatchEventToListeners):
* inspector/front-end/ResourceScriptMapping.js:
* inspector/front-end/UISourceCode.js:
* inspector/front-end/Workspace.js:
(WebInspector.ProjectDelegate.prototype.searchInFileContent):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@144669 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Remove contextInternalContext from JSContextInternal.h
https://bugs.webkit.org/show_bug.cgi?id=111356

Reviewed by Geoffrey Garen.

We don't need it any more since we have globalContextRef in JSContext.

Source/JavaScriptCore:

* API/JSContext.mm:
* API/JSContextInternal.h:
* API/JSValue.mm:
(+[JSValue valueWithBool:inContext:]):
(+[JSValue valueWithDouble:inContext:]):
(+[JSValue valueWithInt32:inContext:]):
(+[JSValue valueWithUInt32:inContext:]):
(+[JSValue valueWithNewObjectInContext:]):
(+[JSValue valueWithNewArrayInContext:]):
(+[JSValue valueWithNewRegularExpressionFromPattern:flags:inContext:]):
(+[JSValue valueWithNewErrorFromMessage:inContext:]):
(+[JSValue valueWithNullInContext:]):
(+[JSValue valueWithUndefinedInContext:]):
(-[JSValue toBool]):
(-[JSValue toDouble]):
(-[JSValue toNumber]):
(-[JSValue toString]):
(-[JSValue toDate]):
(-[JSValue toArray]):
(-[JSValue toDictionary]):
(-[JSValue valueForProperty:]):
(-[JSValue setValue:forProperty:]):
(-[JSValue deleteProperty:]):
(-[JSValue hasProperty:]):
(-[JSValue valueAtIndex:]):
(-[JSValue setValue:atIndex:]):
(-[JSValue isUndefined]):
(-[JSValue isNull]):
(-[JSValue isBoolean]):
(-[JSValue isNumber]):
(-[JSValue isString]):
(-[JSValue isObject]):
(-[JSValue isEqualToObject:]):
(-[JSValue isEqualWithTypeCoercionToObject:]):
(-[JSValue isInstanceOf:]):
(-[JSValue callWithArguments:]):
(-[JSValue constructWithArguments:]):
(-[JSValue invokeMethod:withArguments:]):
(valueToObject):
(objectToValueWithoutCopy):
(objectToValue):
(-[JSValue initWithValue:inContext:]):
(-[JSValue dealloc]):
(-[JSValue description]):
* API/JSWrapperMap.mm:
(createObjectWithCustomBrand):
(-[JSObjCClassInfo allocateConstructorAndPrototypeWithSuperClassInfo:]):
(-[JSObjCClassInfo wrapperForObject:]):
(-[JSWrapperMap jsWrapperForObject:]):
* API/ObjCCallbackFunction.mm:
(ObjCCallbackFunction::call):
(objCCallbackFunctionForInvocation):

Source/WebKit/mac:

* WebView/WebScriptWorld.mm:
(+[WebScriptWorld scriptWorldForJavaScriptContext:]):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@144668 268f45cc-cd09-0410-ab3c-d52691b4dbfc

XSSAuditor bypass with --> comment syntax.
https://bugs.webkit.org/show_bug.cgi?id=111349

Reviewed by Adam Barth.

Source/WebCore:

Test: http/tests/security/xssAuditor/open-iframe-src-03.html

* html/parser/XSSAuditor.cpp:
(WebCore::isTerminatingCharacter):
Adds ">" to the list of terminating characters, so that comments of
the form --> end the snippet we seek to match.

LayoutTests:

* http/tests/security/xssAuditor/open-iframe-src-03-expected.txt: Added.
* http/tests/security/xssAuditor/open-iframe-src-03.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@144667 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Web Inspector: Modal dialogs sometimes lose focus.
https://bugs.webkit.org/show_bug.cgi?id=111317

Reviewed by Pavel Feldman.

setCurrentFocusElement doesn't give focus to an element anymore
if modal glass pane is shown and element is not inside it.

* inspector/front-end/Dialog.js:
(WebInspector.Dialog):
(WebInspector.Dialog.prototype._hide):
* inspector/front-end/InspectorView.js:
(WebInspector.InspectorView.prototype.defaultFocusedElement):
* inspector/front-end/ScriptsPanel.js:
(WebInspector.ScriptsPanel.prototype.defaultFocusedElement):
* inspector/front-end/UIUtils.js:
(WebInspector.GlassPane):
(WebInspector.GlassPane.prototype.dispose):
(WebInspector.setCurrentFocusElement):
* inspector/front-end/dialog.css:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@144666 268f45cc-cd09-0410-ab3c-d52691b4dbfc

[WK2][EFL] Add callbacks to the WKViewClient to handle Web Process crash and relaunch
https://bugs.webkit.org/show_bug.cgi?id=109828

Reviewed by Kenneth Rohde Christiansen.

Source/WebKit2:

Providing WKViewClient with Web Process crash and Web Process relaunch
callbacks brings better design as WebView should not be aware of
EFL-specific code handling the corresponding events.

The implementation of the mentioned Web Process callbacks was also added.

* UIProcess/API/C/efl/WKView.cpp:
(WKViewSetThemePath):
* UIProcess/API/C/efl/WKView.h:
* UIProcess/efl/ViewClientEfl.cpp:
(WebKit::ViewClientEfl::webProcessCrashed):
(WebKit):
(WebKit::ViewClientEfl::webProcessDidRelaunch):
(WebKit::ViewClientEfl::ViewClientEfl):
* UIProcess/efl/ViewClientEfl.h:
(ViewClientEfl):
* UIProcess/efl/WebView.cpp:
(WebKit::WebView::setThemePath):

    Accepts WTF::String instead of WKStringRef as it is
    more appropriate for C++ API implementation class.

(WebKit::WebView::processDidCrash):
(WebKit::WebView::didRelaunchProcess):
* UIProcess/efl/WebView.h:
(WebView):
* UIProcess/efl/WebViewClient.cpp:
(WebKit::WebViewClient::webProcessCrashed):
(WebKit):
(WebKit::WebViewClient::webProcessDidRelaunch):
* UIProcess/efl/WebViewClient.h:

Tools:

Added API test for newly added Web Process crash and Web Process relaunch WKViewClient
callbacks.

* TestWebKitAPI/CMakeLists.txt:

    Tests located in 'TestWebKitAPI/Tests/WebKit2' subdirectories are also
    considered.

* TestWebKitAPI/PlatformEfl.cmake:
* TestWebKitAPI/Tests/WebKit2/efl/WKViewClientWebProcessCallbacks.cpp: Added.
(TestWebKitAPI):
(TestWebKitAPI::didFinishLoadForFrame):
(TestWebKitAPI::setPageLoaderClient):
(TestWebKitAPI::webProcessCrashed):
(TestWebKitAPI::webProcessDidRelaunch):
(TestWebKitAPI::setViewClient):
(TestWebKitAPI::TEST):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@144664 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Unreviewed. AppleWin build fix.

* WebKit.vcxproj/FeatureDefines.props:
* WebKit.vcxproj/WebKitExportGenerator/WebKitExports.def.in:
* WebCore.vcxproj/WebCore.vcxproj:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@144663 268f45cc-cd09-0410-ab3c-d52691b4dbfc

JSC_OBJC_API_ENABLED should be removed from FeatureDefines.h
https://bugs.webkit.org/show_bug.cgi?id=111269

Reviewed by Mark Hahnenberg.

Source/WebKit/mac:

Include JavaScriptCore/JSBase.h for the definition of JSC_OBJC_API_ENABLED.

* WebView/WebDelegateImplementationCaching.h:

Source/WTF:

JSC_OBJC_API_ENABLED flag was added to FeatureDefines.h (r143637)
so that it could be used in WebCore.exp.in to determine whether
or not we should be exporting certain symbols based on whether
the WebKit-related portion of the JSC ObjC API was enabled.

It turned out conditionally including certain symbols didn't
work when building fat binaries, so it's not even necessary any more.

* wtf/FeatureDefines.h: Remove JSC_OBJC_API_ENABLED.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@144662 268f45cc-cd09-0410-ab3c-d52691b4dbfc

[v8] add isolate parameter to heap statistics calls
https://bugs.webkit.org/show_bug.cgi?id=111308

Patch by Dan Carney <dcarney@google.com> on 2013-03-04
Reviewed by Jochen Eisinger.

No new tests.

* bindings/v8/ScriptGCEvent.cpp:
(WebCore::ScriptGCEvent::getHeapSize):
(WebCore::ScriptGCEvent::getUsedHeapSize):
* bindings/v8/V8AdaptorFunction.cpp:
(WebCore::V8AdaptorFunction::getTemplate):
* bindings/v8/V8GCController.cpp:
(WebCore::V8GCController::checkMemoryUsage):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@144661 268f45cc-cd09-0410-ab3c-d52691b4dbfc

[chromium] remove deprecated hyphenation methods from Platform
https://bugs.webkit.org/show_bug.cgi?id=111101

Reviewed by Adam Barth.

Source/Platform:

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

Source/WebCore:

* platform/text/chromium/Hyphenation.cpp:
(WebCore::canHyphenate):
(WebCore::lastHyphenLocation):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@144660 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Complete the plug-in URL string before sending it to the UI process
https://bugs.webkit.org/show_bug.cgi?id=111355
<rdar://problem/13326713>

Reviewed by Andreas Kling.

If the URL is not absolute, we won't be able to find it by path extension in the plug-in info store.

* WebProcess/WebCoreSupport/WebChromeClient.cpp:
(WebKit::WebChromeClient::unavailablePluginButtonClicked):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@144659 268f45cc-cd09-0410-ab3c-d52691b4dbfc

AX: Upstream iOS Accessibility DumpRenderTree changes
https://bugs.webkit.org/show_bug.cgi?id=111220

Reviewed by David Kilzer.

Upstream the iOS changes for accessibility in DumpRenderTree.

* DumpRenderTree/AccessibilityUIElement.cpp:
(headerElementAtIndexCallback):
(linkedElementCallback):
(elementsForRangeCallback):
(increaseTextSelectionCallback):
(decreaseTextSelectionCallback):
(assistiveTechnologySimulatedFocusCallback):
(stringForSelectionCallback):
(getIPhoneLabelCallback):
(getIPhoneHintCallback):
(getIPhoneValueCallback):
(getIPhoneIdentifierCallback):
(getIPhoneTraitsCallback):
(getIPhoneIsElementCallback):
(getIPhoneElementTextPositionCallback):
(getIPhoneElementTextLengthCallback):
(AccessibilityUIElement::getJSClass):
* DumpRenderTree/AccessibilityUIElement.h:
(AccessibilityUIElement):
* DumpRenderTree/DumpRenderTree.xcodeproj/project.pbxproj:
* DumpRenderTree/ios: Added.
* DumpRenderTree/ios/AccessibilityControllerIOS.mm: Added.
(AccessibilityController::AccessibilityController):
(AccessibilityController::~AccessibilityController):
(AccessibilityController::elementAtPoint):
(AccessibilityController::focusedElement):
(AccessibilityController::rootElement):
(findAccessibleObjectById):
(AccessibilityController::accessibleElementById):
(AccessibilityController::setLogFocusEvents):
(AccessibilityController::setLogScrollingStartEvents):
(AccessibilityController::setLogValueChangeEvents):
(AccessibilityController::setLogAccessibilityEvents):
(AccessibilityController::addNotificationListener):
(AccessibilityController::removeNotificationListener):
* DumpRenderTree/ios/AccessibilityUIElementIOS.mm: Added.
(AccessibilityUIElement::AccessibilityUIElement):
(AccessibilityUIElement::~AccessibilityUIElement):
(concatenateAttributeAndValue):
(AccessibilityUIElement::iphoneLabel):
(AccessibilityUIElement::iphoneHint):
(AccessibilityUIElement::iphoneValue):
(AccessibilityUIElement::iphoneIdentifier):
(AccessibilityUIElement::iphoneTraits):
(AccessibilityUIElement::iphoneIsElement):
(AccessibilityUIElement::iphoneElementTextPosition):
(AccessibilityUIElement::iphoneElementTextLength):
(AccessibilityUIElement::url):
(AccessibilityUIElement::x):
(AccessibilityUIElement::y):
(AccessibilityUIElement::width):
(AccessibilityUIElement::height):
(AccessibilityUIElement::clickPointX):
(AccessibilityUIElement::clickPointY):
(AccessibilityUIElement::getChildren):
(AccessibilityUIElement::getChildrenWithRange):
(AccessibilityUIElement::childrenCount):
(AccessibilityUIElement::elementAtPoint):
(AccessibilityUIElement::indexOfChild):
(AccessibilityUIElement::getChildAtIndex):
(AccessibilityUIElement::headerElementAtIndex):
(AccessibilityUIElement::linkedElement):
(AccessibilityUIElement::linkedUIElementAtIndex):
(AccessibilityUIElement::ariaOwnsElementAtIndex):
(AccessibilityUIElement::ariaFlowToElementAtIndex):
(AccessibilityUIElement::disclosedRowAtIndex):
(AccessibilityUIElement::selectedRowAtIndex):
(AccessibilityUIElement::rowAtIndex):
(AccessibilityUIElement::titleUIElement):
(AccessibilityUIElement::parentElement):
(AccessibilityUIElement::disclosedByRow):
(AccessibilityUIElement::increaseTextSelection):
(AccessibilityUIElement::decreaseTextSelection):
(AccessibilityUIElement::stringForSelection):
(AccessibilityUIElement::stringForRange):
(AccessibilityUIElement::attributedStringForRange):
(AccessibilityUIElement::attributedStringRangeIsMisspelled):
(AccessibilityUIElement::elementsForRange):
(AccessibilityUIElement::getLinkedUIElements):
(AccessibilityUIElement::getDocumentLinks):
(AccessibilityUIElement::attributesOfLinkedUIElements):
(AccessibilityUIElement::attributesOfDocumentLinks):
(AccessibilityUIElement::attributesOfChildren):
(AccessibilityUIElement::allAttributes):
(AccessibilityUIElement::stringAttributeValue):
(AccessibilityUIElement::isPressActionSupported):
(AccessibilityUIElement::isIncrementActionSupported):
(AccessibilityUIElement::isDecrementActionSupported):
(AccessibilityUIElement::boolAttributeValue):
(AccessibilityUIElement::isAttributeSettable):
(AccessibilityUIElement::isAttributeSupported):
(AccessibilityUIElement::parameterizedAttributeNames):
(AccessibilityUIElement::role):
(AccessibilityUIElement::subrole):
(AccessibilityUIElement::roleDescription):
(AccessibilityUIElement::title):
(AccessibilityUIElement::description):
(AccessibilityUIElement::orientation):
(AccessibilityUIElement::stringValue):
(AccessibilityUIElement::language):
(AccessibilityUIElement::helpText):
(AccessibilityUIElement::intValue):
(AccessibilityUIElement::minValue):
(AccessibilityUIElement::maxValue):
(AccessibilityUIElement::valueDescription):
(AccessibilityUIElement::insertionPointLineNumber):
(AccessibilityUIElement::isEnabled):
(AccessibilityUIElement::isRequired):
(AccessibilityUIElement::isFocused):
(AccessibilityUIElement::isSelected):
(AccessibilityUIElement::isExpanded):
(AccessibilityUIElement::isChecked):
(AccessibilityUIElement::hierarchicalLevel):
(AccessibilityUIElement::ariaIsGrabbed):
(AccessibilityUIElement::ariaDropEffects):
(AccessibilityUIElement::lineForIndex):
(AccessibilityUIElement::boundsForRange):
(AccessibilityUIElement::attributesOfColumnHeaders):
(AccessibilityUIElement::attributesOfRowHeaders):
(AccessibilityUIElement::attributesOfColumns):
(AccessibilityUIElement::attributesOfRows):
(AccessibilityUIElement::attributesOfVisibleCells):
(AccessibilityUIElement::attributesOfHeader):
(AccessibilityUIElement::rowCount):
(AccessibilityUIElement::columnCount):
(AccessibilityUIElement::indexInTable):
(AccessibilityUIElement::rowIndexRange):
(AccessibilityUIElement::columnIndexRange):
(AccessibilityUIElement::cellForColumnAndRow):
(AccessibilityUIElement::selectedTextRange):
(AccessibilityUIElement::assistiveTechnologySimulatedFocus):
(AccessibilityUIElement::setSelectedTextRange):
(AccessibilityUIElement::increment):
(AccessibilityUIElement::decrement):
(AccessibilityUIElement::showMenu):
(AccessibilityUIElement::press):
(AccessibilityUIElement::accessibilityValue):
(AccessibilityUIElement::documentEncoding):
(AccessibilityUIElement::documentURI):
(_accessibilityNotificationCallback):
(AccessibilityUIElement::addNotificationListener):
(AccessibilityUIElement::removeNotificationListener):
(AccessibilityUIElement::isFocusable):
(AccessibilityUIElement::isSelectable):
(AccessibilityUIElement::isMultiSelectable):
(AccessibilityUIElement::isSelectedOptionActive):
(AccessibilityUIElement::isVisible):
(AccessibilityUIElement::isOffScreen):
(AccessibilityUIElement::isCollapsed):
(AccessibilityUIElement::isIgnored):
(AccessibilityUIElement::hasPopup):
(AccessibilityUIElement::takeFocus):
(AccessibilityUIElement::takeSelection):
(AccessibilityUIElement::addSelection):
(AccessibilityUIElement::removeSelection):
(AccessibilityUIElement::uiElementForSearchPredicate):
(AccessibilityUIElement::numberAttributeValue):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@144657 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Use initialization from literal for MediaFeatureNames
https://bugs.webkit.org/show_bug.cgi?id=111283

Reviewed by Andreas Kling.

* css/MediaFeatureNames.cpp:
(WebCore::MediaFeatureNames::init):
Initialization from literal is faster and use less memory. The names are
defined statically in MediaFeatureNames.h.

Also use the NotNull placement new to avoid a branch on silly allocators.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@144656 268f45cc-cd09-0410-ab3c-d52691b4dbfc

(tests) flexbox/resources.css should use unprefixed properties, for firefox
https://bugs.webkit.org/show_bug.cgi?id=111010

Patch by Christian Biesinger <cbiesinger@chromium.org> on 2013-03-04
Reviewed by Tony Chang.

Search & replace to remove -moz- prefixes, as per
https://developer.mozilla.org/en-US/docs/CSS/Using_CSS_flexible_boxes:
"Gecko implementation is unprefixed (and is no longer behind a
preference as of Firefox 20)"

* css3/flexbox/resources/flexbox.css:
(.flexbox):
(.inline-flexbox):
(.flex-none):
(.flex-auto):
(.flex-one):
(.row):
(.row-reverse):
(.column):
(.column-reverse):
(.wrap):
(.wrap-reverse):
(.align-content-flex-start):
(.align-content-flex-end):
(.align-content-center):
(.align-content-space-between):
(.align-content-space-around):
(.align-content-stretch):
(.align-items-flex-start):
(.align-items-flex-end):
(.align-items-center):
(.align-items-baseline):
(.align-items-stretch):
(.align-self-auto):
(.align-self-flex-start):
(.align-self-flex-end):
(.align-self-center):
(.align-self-baseline):
(.align-self-stretch):
(.justify-content-flex-start):
(.justify-content-flex-end):
(.justify-content-center):
(.justify-content-space-between):
(.justify-content-space-around):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@144655 268f45cc-cd09-0410-ab3c-d52691b4dbfc

[chromium] Add ability to wait on async upload completion
https://bugs.webkit.org/show_bug.cgi?id=110987

Reviewed by James Robinson.

Add a graphics context entry point for waiting on asynchronous upload
completion. The wait is performed server-side so the call is
non-blocking.

See https://codereview.chromium.org/12210129/

* chromium/public/WebGraphicsContext3D.h:
(WebKit::WebGraphicsContext3D::waitAsyncTexImage2DCHROMIUM):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@144654 268f45cc-cd09-0410-ab3c-d52691b4dbfc

[GTK] Some WebKit2 GTK+ unit tests are failing in 32 bits bot
https://bugs.webkit.org/show_bug.cgi?id=111346

Patch by Carlos Garcia Campos <cgarcia@igalia.com> on 2013-03-04
Reviewed by Martin Robinson.

The problem is the use of a temporary CString object in the macro
g_assert_cmpstr. It's a common mistake because we usually forget
that g_assert_cmpstr is not a function but a macro, that expands
to multiple lines. This patch adds a new macro ASSERT_CMP_CSTRING
with the same implementation that g_assert_cmpstr, but using
CStrings instead of const char*. It fixes all the cases where a
temporary CString was used in g_assert_cmpstr, and uses the new
macro also for the cases where we were caching the CString just
for g_assert_cmpstr.

* UIProcess/API/gtk/tests/TestDownloads.cpp:
(testDownloadRemoteFile):
* UIProcess/API/gtk/tests/TestInspectorServer.cpp:
(testInspectorServerPageList):
* UIProcess/API/gtk/tests/TestLoaderClient.cpp:
* UIProcess/API/gtk/tests/TestMain.h:
* UIProcess/API/gtk/tests/TestResources.cpp:
* UIProcess/API/gtk/tests/TestWebKitFaviconDatabase.cpp:
(testGetFaviconURI):
* UIProcess/API/gtk/tests/TestWebKitSettings.cpp:
(assertThatUserAgentIsSentInHeaders):
(testWebKitSettingsUserAgent):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@144652 268f45cc-cd09-0410-ab3c-d52691b4dbfc

[V8] Unreviewed. Rebaselined run-bindings-tests following r144617

* bindings/scripts/test/V8/V8Float64Array.cpp:
(WebCore::ConfigureV8Float64ArrayTemplate):
(WebCore::V8Float64Array::GetTemplate):
* bindings/scripts/test/V8/V8Float64Array.h:
(V8Float64Array):
* bindings/scripts/test/V8/V8TestActiveDOMObject.cpp:
(WebCore::TestActiveDOMObjectV8Internal::postMessageAttrGetter):
(WebCore::TestActiveDOMObjectV8Internal::TestActiveDOMObjectDomainSafeFunctionSetter):
(WebCore::ConfigureV8TestActiveDOMObjectTemplate):
(WebCore::V8TestActiveDOMObject::GetTemplate):
* bindings/scripts/test/V8/V8TestActiveDOMObject.h:
(V8TestActiveDOMObject):
* bindings/scripts/test/V8/V8TestCustomNamedGetter.cpp:
(WebCore::ConfigureV8TestCustomNamedGetterTemplate):
(WebCore::V8TestCustomNamedGetter::GetTemplate):
* bindings/scripts/test/V8/V8TestCustomNamedGetter.h:
(V8TestCustomNamedGetter):
* bindings/scripts/test/V8/V8TestEventConstructor.cpp:
(WebCore::ConfigureV8TestEventConstructorTemplate):
(WebCore::V8TestEventConstructor::GetTemplate):
* bindings/scripts/test/V8/V8TestEventConstructor.h:
(V8TestEventConstructor):
* bindings/scripts/test/V8/V8TestEventTarget.cpp:
(WebCore::ConfigureV8TestEventTargetTemplate):
(WebCore::V8TestEventTarget::GetTemplate):
* bindings/scripts/test/V8/V8TestEventTarget.h:
(V8TestEventTarget):
* bindings/scripts/test/V8/V8TestException.cpp:
(WebCore::ConfigureV8TestExceptionTemplate):
(WebCore::V8TestException::GetTemplate):
* bindings/scripts/test/V8/V8TestException.h:
(V8TestException):
* bindings/scripts/test/V8/V8TestInterface.cpp:
(WebCore::ConfigureV8TestInterfaceTemplate):
(WebCore::V8TestInterface::GetTemplate):
* bindings/scripts/test/V8/V8TestInterface.h:
(V8TestInterface):
* bindings/scripts/test/V8/V8TestMediaQueryListListener.cpp:
(WebCore::ConfigureV8TestMediaQueryListListenerTemplate):
(WebCore::V8TestMediaQueryListListener::GetTemplate):
* bindings/scripts/test/V8/V8TestMediaQueryListListener.h:
(V8TestMediaQueryListListener):
* bindings/scripts/test/V8/V8TestNamedConstructor.cpp:
(WebCore::V8TestNamedConstructorConstructor::GetTemplate):
(WebCore::ConfigureV8TestNamedConstructorTemplate):
(WebCore::V8TestNamedConstructor::GetTemplate):
* bindings/scripts/test/V8/V8TestNamedConstructor.h:
(V8TestNamedConstructorConstructor):
(V8TestNamedConstructor):
* bindings/scripts/test/V8/V8TestNode.cpp:
(WebCore::ConfigureV8TestNodeTemplate):
(WebCore::V8TestNode::GetTemplate):
* bindings/scripts/test/V8/V8TestNode.h:
(V8TestNode):
* bindings/scripts/test/V8/V8TestObj.cpp:
(WebCore::ConfigureV8TestObjTemplate):
(WebCore::V8TestObj::GetTemplate):
(WebCore::V8TestObj::installPerContextPrototypeProperties):
* bindings/scripts/test/V8/V8TestObj.h:
(V8TestObj):
* bindings/scripts/test/V8/V8TestOverloadedConstructors.cpp:
(WebCore::ConfigureV8TestOverloadedConstructorsTemplate):
(WebCore::V8TestOverloadedConstructors::GetTemplate):
* bindings/scripts/test/V8/V8TestOverloadedConstructors.h:
(V8TestOverloadedConstructors):
* bindings/scripts/test/V8/V8TestSerializedScriptValueInterface.cpp:
(WebCore::ConfigureV8TestSerializedScriptValueInterfaceTemplate):
(WebCore::V8TestSerializedScriptValueInterface::GetTemplate):
* bindings/scripts/test/V8/V8TestSerializedScriptValueInterface.h:
(V8TestSerializedScriptValueInterface):
* bindings/scripts/test/V8/V8TestTypedefs.cpp:
(WebCore::ConfigureV8TestTypedefsTemplate):
(WebCore::V8TestTypedefs::GetTemplate):
* bindings/scripts/test/V8/V8TestTypedefs.h:
(V8TestTypedefs):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@144651 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Unreviewed, rolling out r144562.
http://trac.webkit.org/changeset/144562
https://bugs.webkit.org/show_bug.cgi?id=85642

Caused KeyPadInsert faluires in webkit_unittests

Source/WebCore:

* platform/chromium/KeyCodeConversionGtk.cpp:
(WebCore::windowsKeyCodeForKeyEvent):

Source/WebKit/chromium:

* WebKit.gypi:
* tests/KeyCodeConversionTestGtk.cpp: Removed.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@144649 268f45cc-cd09-0410-ab3c-d52691b4dbfc

getComputedStyle not implemented for -webkit-columns shorthand
https://bugs.webkit.org/show_bug.cgi?id=111236

Patch by Uday Kiran <udaykiran@motorola.com> on 2013-03-04
Reviewed by Alexis Menard.

Source/WebCore:

Implement getComputedStyle for -webkit-columns shorthand.

Test: fast/css/getComputedStyle/getComputedStyle-webkit-columns-shorthand.html

* css/CSSComputedStyleDeclaration.cpp:
(WebCore::CSSComputedStyleDeclaration::getPropertyCSSValue):

LayoutTests:

Add test to check getComputedStyle for -webkit-columns.

* fast/css/getComputedStyle/getComputedStyle-webkit-columns-shorthand-expected.txt: Added.
* fast/css/getComputedStyle/getComputedStyle-webkit-columns-shorthand.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@144647 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Add support for 8 bit TextRuns on Chromium Linux & Mac
https://bugs.webkit.org/show_bug.cgi?id=99393

Reviewed by Eric Seidel.

Source/WebCore:

Adding support for 8 bit TextRuns for Mac and Linux Chromium. To accomplish this,
8 bit text runs are upconverted to 16 bit in the complex text path during string
normalization, as HarfBuzz operates on UChars.

Windows has platfom assumptions that TextRuns are 16 bit that need to be addressed
before enabling this optimization.

No new tests. No change in behavior.

(WebCore::HarfBuzzShaperBase::setNormalizedBuffer):
* platform/graphics/harfbuzz/HarfBuzzShaperBase.h:
* platform/graphics/harfbuzz/HarfBuzzShaper.cpp:
(WebCore::normalizeCharacters):
(WebCore::HarfBuzzShaper::HarfBuzzShaper):

Source/WebKit/chromium:

Enabling 8 bit text runs for Linux and Mac platforms.

* features.gypi:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@144646 268f45cc-cd09-0410-ab3c-d52691b4dbfc

transition-property property and transition shorthand property doesn't accept "all, all".
https://bugs.webkit.org/show_bug.cgi?id=111201

Reviewed by Simon Fraser.

Source/WebCore:

Relax the parsing of transition-property to allow all, all as a value.
It is not very useful per say but it is possible by the spec and we
align with Firefox, IE and Opera.

Test: LayoutTests/transitions/transitions-parsing.html

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

LayoutTests:

Updated the tests to cover the bug.

* transitions/transitions-parsing-expected.txt:
* transitions/transitions-parsing.html:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@144644 268f45cc-cd09-0410-ab3c-d52691b4dbfc

[Chromium] Add a new dependency on jsr-305 for Android
https://bugs.webkit.org/show_bug.cgi?id=111340

Unreviewed build fix; new Android dependency. This is
currently breaking gyp generation on the Android bots.

* DEPS:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@144643 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Don't apply page scale to clipRect if applyPageScaleFactorInCompositor is set to true
https://bugs.webkit.org/show_bug.cgi?id=111229

Patch by Min Qin <qinmin@chromium.org> on 2013-03-04
Reviewed by Simon Fraser.

If applyPageScaleFactorInCompositor is set to true, webcore should use css coordinates.
As a result, we shouldn't apply page scale to clipRect in RenderLayerCompositor.
Layout tests set the flag to false, so we don't have anything to test against.

* rendering/RenderLayerCompositor.cpp:
(WebCore::RenderLayerCompositor::addToOverlapMap):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@144642 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Add simple vector traits for JSC::Identifier.
<http://webkit.org/b/111323>

Reviewed by Geoffrey Garen.

Identifiers are really just Strings, giving them simple vector traits makes
Vector move them with memcpy() instead of churning the refcounts.

* runtime/Identifier.h:
(WTF):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@144641 268f45cc-cd09-0410-ab3c-d52691b4dbfc

[BlackBerry] Upstream selection start animation.
https://bugs.webkit.org/show_bug.cgi?id=111335

Patch by Iris Wu <shuwu@rim.com> on 2013-03-04
Reviewed by Rob Buis.

PR 219960

To indicate selection starts, we draw an overlay rect larger than
the selected word at the start of selection. Then shrink it down
to the actual size.

Reviewed Internally by Genevieve Mak and Andrew Lo and Mike Fenton.

* Api/WebAnimation.cpp:
(BlackBerry::WebKit::WebAnimation::shrinkAnimation):
(WebKit):
* Api/WebAnimation.h:
* Api/WebPage.cpp:
(BlackBerry::WebKit::WebPagePrivate::init):
(BlackBerry::WebKit::WebPage::selectionHighlight):
(WebKit):
* Api/WebPage.h:
* Api/WebPage_p.h:
(WebPagePrivate):
* Api/WebTapHighlight.h:
* WebKitSupport/DefaultTapHighlight.cpp:
(WebKit):
(BlackBerry::WebKit::shrinkAnimationName):
(BlackBerry::WebKit::DefaultTapHighlight::draw):
* WebKitSupport/DefaultTapHighlight.h:
(DefaultTapHighlight):
* WebKitSupport/SelectionHandler.cpp:
(BlackBerry::WebKit::SelectionHandler::cancelSelection):
(BlackBerry::WebKit::textGranularityFromSelectionExpansionType):
(WebKit):
(BlackBerry::WebKit::SelectionHandler::selectNodeIfFatFingersResultIsLink):
(BlackBerry::WebKit::SelectionHandler::selectAtPoint):
* WebKitSupport/SelectionHandler.h:
(SelectionHandler):
* WebKitSupport/TouchEventHandler.cpp:
(BlackBerry::WebKit::TouchEventHandler::handleTouchPoint):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@144640 268f45cc-cd09-0410-ab3c-d52691b4dbfc

ASSERTION FAILED: m_clipRectsCache->m_respectingOverflowClip[clipRectsType] == (clipRectsContext.respectOverflowClip == RespectOverflowClip) in RenderLayer.
https://bugs.webkit.org/show_bug.cgi?id=108257

Reviewed by David Hyatt.

Source/WebCore:

With composited scrolling we paint both with and without respecting
overflow clip. To prevent collisions in the clip cache, and to prevent
throwing away cached clips unnecessarily, we keep two copies of the
clip cache -- one for when overflow clip is respected, and one for
when it isn't.

No new tests. Covered by existing tests (in debug):
  compositing/overflow/automatically-opt-into-composited-scrolling.html
  compositing/overflow/composited-scrolling-creates-a-stacking-container.html

* rendering/RenderLayer.cpp:
(WebCore::RenderLayer::updateClipRects):
  No longer asserts that our 'respect overflow clip' status is
  consistent. It also gets and sets the clip cache using the clip rect
  context rather than just the clip rect type.
(WebCore::RenderLayer::calculateClipRects):
(WebCore::RenderLayer::parentClipRects):
(WebCore::RenderLayer::clearClipRects):
(WebCore::ClipRectsCache::ClipRectsCache):
(WebCore::ClipRectsCache::getClipRects):
(WebCore::ClipRectsCache::setClipRects):
(WebCore::ClipRectsCache::getIndex):
  Get and set the cached clip rects using the context rather than
  type.
(WebCore::RenderLayer::clipRects):
  We now cache twice as many clip rects.
* rendering/RenderLayerBacking.cpp:
(WebCore::RenderLayerBacking::updateGraphicsLayerGeometry):
  ShouldRespectOverflowClip was moved out of RenderLayer.

* rendering/RenderLayer.cpp:
(WebCore::RenderLayer::updateClipRects):
(WebCore::RenderLayer::calculateClipRects):
(WebCore::RenderLayer::parentClipRects):
(WebCore::RenderLayer::clearClipRects):
* rendering/RenderLayer.h:
(WebCore::ClipRectsCache::ClipRectsCache):
(WebCore::ClipRectsCache::getClipRects):
(WebCore::ClipRectsCache::setClipRects):
(ClipRectsCache):
(WebCore::ClipRectsCache::getIndex):
(WebCore::RenderLayer::clipRects):
* rendering/RenderLayerBacking.cpp:
(WebCore::RenderLayerBacking::updateGraphicsLayerGeometry):

LayoutTests:

* platform/efl-wk2/TestExpectations:
* platform/mac/TestExpectations:
* platform/qt-5.0-wk2/TestExpectations:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@144639 268f45cc-cd09-0410-ab3c-d52691b4dbfc