WebKit-https.git
6 years agoEnsure that AudioNode deletion is synchronized with a stable state of the rendering...
crogers@google.com [Thu, 1 Nov 2012 23:23:51 +0000 (23:23 +0000)]
Ensure that AudioNode deletion is synchronized with a stable state of the rendering graph
https://bugs.webkit.org/show_bug.cgi?id=100994

Reviewed by Kenneth Russell.

In some rare cases it has been observed that nodes are getting deleted in the main thread
during an audio rendering quantum where the dirty inputs and outputs have not yet been cleaned
via calls to handleDirtyAudioSummingJunctions() and handleDirtyAudioNodeOutputs().
This was possible because nodes marked for deletion with markForDeletion() could be picked
up in a subsequent call to deleteMarkedNodes() before the render quantum has finished and
handlePostRenderTasks() has had a chance to reconcile these marked nodes and clean the dirty state.
The solution is to manage the marked nodes in a separate vector which only gets copied to another
vector truly eligible for deletion which is synchronized in handlePostRenderTasks().

* Modules/webaudio/AudioContext.cpp:
(WebCore::AudioContext::markForDeletion):
(WebCore::AudioContext::scheduleNodeDeletion):
(WebCore::AudioContext::deleteMarkedNodes):
* Modules/webaudio/AudioContext.h:
(AudioContext):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@133239 268f45cc-cd09-0410-ab3c-d52691b4dbfc

6 years agoBuild fix after r133224 as suggested by Enrica.
rniwa@webkit.org [Thu, 1 Nov 2012 22:58:30 +0000 (22:58 +0000)]
Build fix after r133224 as suggested by Enrica.
Confirmed the test failures go away on Chromium Mac locally.

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

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@133238 268f45cc-cd09-0410-ab3c-d52691b4dbfc

6 years agoUnreviewed. Skip fast/xmlhttprequest/xmlhttprequest-recursive-sync-event.html.
roger_fong@apple.com [Thu, 1 Nov 2012 22:58:04 +0000 (22:58 +0000)]
Unreviewed. Skip fast/xmlhttprequest/xmlhttprequest-recursive-sync-event.html.
It is inherently flaky: https://bugs.webkit.org/show_bug.cgi?id=100117

* platform/win/TestExpectations:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@133237 268f45cc-cd09-0410-ab3c-d52691b4dbfc

6 years agoAdd Python-LLDB support for WTF::Vector types.
jer.noble@apple.com [Thu, 1 Nov 2012 22:33:10 +0000 (22:33 +0000)]
Add Python-LLDB support for WTF::Vector types.
https://bugs.webkit.org/show_bug.cgi?id=100980

Reviewed by Anders Carlsson.

Add a synthetic child provider for WTF::Vector types.

* lldb/lldb_webkit.py:
(__lldb_init_module): Add both the summary and synthetic providers for WTF::Vector
(WTFVector_SummaryProvider): Wrapper around WTFVectorProvider
(WTFVectorProvider):
(WTFVectorProvider.__init__):
(WTFVectorProvider.num_children): Return m_size+2 children to account for synthesized children.
(WTFVectorProvider.get_child_index): Synthesize child indexes for each entry in the Vector.
(WTFVectorProvider.get_child_at_index): Synthesize children for each entry in the Vector.
(WTFVectorProvider.update): Retrieve state from the underlying Vector.
(WTFVectorProvider.has_children): Vectors always have children.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@133234 268f45cc-cd09-0410-ab3c-d52691b4dbfc

6 years ago Rename HAVE(NETWORK_CFDATA_ARRAY_CALLBACK) to USE(NETWORK_CFDATA_ARRAY_CALLBACK)
ap@apple.com [Thu, 1 Nov 2012 22:29:23 +0000 (22:29 +0000)]
    Rename HAVE(NETWORK_CFDATA_ARRAY_CALLBACK) to USE(NETWORK_CFDATA_ARRAY_CALLBACK)

        Rubber-stamped by Joe Pecoraro.

        All CFNetwork based platforms have this now, but not all use it at the moment.

        * loader/ResourceLoader.h:
        * loader/SubresourceLoader.h:
        * loader/cf/SubresourceLoaderCF.cpp:
        * loader/mac/ResourceLoaderMac.mm:
        * platform/SharedBuffer.cpp:
        (WebCore::SharedBuffer::clear):
        (WebCore::SharedBuffer::buffer):
        (WebCore::SharedBuffer::getSomeData):
        * platform/SharedBuffer.h:
        * platform/cf/SharedBufferCF.cpp:
        * platform/network/ResourceHandle.h:
        * platform/network/ResourceHandleClient.h:
        (ResourceHandleClient):
        * platform/network/cf/ResourceHandleCFNet.cpp:
        (WebCore::willCacheResponse):
        (WebCore::ResourceHandle::createCFURLConnection):
        * platform/network/mac/ResourceHandleMac.mm:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@133233 268f45cc-cd09-0410-ab3c-d52691b4dbfc

6 years ago Fix HAVE(NETWORK_CFDATA_ARRAY_CALLBACK) build
ap@apple.com [Thu, 1 Nov 2012 22:26:55 +0000 (22:26 +0000)]
    Fix HAVE(NETWORK_CFDATA_ARRAY_CALLBACK) build
        https://bugs.webkit.org/show_bug.cgi?id=100979

        Reviewed by Joseph Pecoraro.

        * loader/ResourceBuffer.cpp: (WebCore::ResourceBuffer::append):
        * loader/ResourceBuffer.h:
        Added a version of append() that takes CFDataRef, passing it directly to SharedBuffer.

        * loader/mac/ResourceLoaderMac.mm: (WebCore::ResourceLoader::didReceiveDataArray):
        InspectorInstrumentation now takes a length. Unfortunately, we don't have encodedLength
        here, meaning that the results will likely be incorrect.

        * platform/network/cf/ResourceHandleCFNet.cpp:
        (WebCore::willCacheResponse): Removed a version that needlessly used deprecated API.
        (WebCore::ResourceHandle::createCFURLConnection): This code assumed that
        HAVE(NETWORK_CFDATA_ARRAY_CALLBACK) and USE(PROTECTION_SPACE_AUTH_CALLBACK) always
        come together. Changed to check for these separately.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@133232 268f45cc-cd09-0410-ab3c-d52691b4dbfc

6 years agoRemove PageLoad directory entry from PerformanceTests pattern_map
zoltan@webkit.org [Thu, 1 Nov 2012 21:59:16 +0000 (21:59 +0000)]
Remove PageLoad directory entry from PerformanceTests pattern_map
https://bugs.webkit.org/show_bug.cgi?id=100989

Reviewed by Ryosuke Niwa.

Since we removed PageLoad directory from PerformanceTests we don't need this entry in pattern_map.

* Scripts/webkitpy/performance_tests/perftest.py:
(PerfTestFactory):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@133229 268f45cc-cd09-0410-ab3c-d52691b4dbfc

6 years ago[BlackBerry] Add more form validation strings
rwlbuis@webkit.org [Thu, 1 Nov 2012 21:59:12 +0000 (21:59 +0000)]
[BlackBerry] Add more form validation strings
https://bugs.webkit.org/show_bug.cgi?id=100978

Reviewed by Yong Li.

PR 210072

Source/WebCore:

Implement these two methods properly.

* platform/blackberry/LocalizedStringsBlackBerry.cpp:
(WebCore::validationMessageRangeUnderflowText):
(WebCore::validationMessageRangeOverflowText):

LayoutTests:

Add the result we expect after adding translations for range over and underflow.

* platform/blackberry/fast/forms/number/number-validation-message-expected.txt: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@133228 268f45cc-cd09-0410-ab3c-d52691b4dbfc

6 years ago[CSS Shaders] Move MeshBoxType out of CustomFilterOperation
achicu@adobe.com [Thu, 1 Nov 2012 21:48:11 +0000 (21:48 +0000)]
[CSS Shaders] Move MeshBoxType out of CustomFilterOperation
https://bugs.webkit.org/show_bug.cgi?id=100782

Reviewed by Dean Jackson.

Source/WebCore:

Moved CustomFilter::MeshBoxType enum to the CustomFilterConstants.h file and renamed it to CustomFilterMeshBoxType.
This way we can reuse it without including the CustomFilterOperation file.

No new tests, just refactoring code.

* css/CSSPrimitiveValueMappings.h:
(WebCore::CSSPrimitiveValue::CSSPrimitiveValue):
(WebCore::CSSPrimitiveValue::operator CustomFilterMeshBoxType):
* css/StyleResolver.cpp:
(WebCore::StyleResolver::createCustomFilterOperation):
* platform/graphics/filters/CustomFilterArrayParameter.h: Adding AnimationUtilities.h for WebCore::blend, looks like
CustomFilterOperation already included AnimationsUtilities.h indirectly.
* platform/graphics/filters/CustomFilterConstants.h: Added CustomFilterMeshBoxType.
* platform/graphics/filters/CustomFilterNumberParameter.h:
* platform/graphics/filters/CustomFilterOperation.cpp:
(WebCore::CustomFilterOperation::CustomFilterOperation):
* platform/graphics/filters/CustomFilterOperation.h:
(WebCore::CustomFilterOperation::create):
(WebCore::CustomFilterOperation::meshBoxType):
(CustomFilterOperation):
* platform/graphics/filters/CustomFilterRenderer.cpp:
(WebCore::CustomFilterRenderer::create):
(WebCore::CustomFilterRenderer::CustomFilterRenderer):
* platform/graphics/filters/CustomFilterRenderer.h: Also removed the need to include CustomFilterOperation.h. Some other
headers were needed to replace the files that were automatically incldued because of that file.
(WebCore):
(CustomFilterRenderer):
* platform/graphics/filters/FECustomFilter.cpp:
(WebCore::FECustomFilter::FECustomFilter):
(WebCore::FECustomFilter::create):
* platform/graphics/filters/FECustomFilter.h:
(FECustomFilter):

Source/WebKit2:

Renamed the CustomFilterOperation::MeshBoxType to CustomFilterMeshBoxType.

* Shared/CoordinatedGraphics/CoordinatedGraphicsArgumentCoders.cpp:
(CoreIPC::::decode):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@133227 268f45cc-cd09-0410-ab3c-d52691b4dbfc

6 years ago[chromium] Unreviewed; marking another inspector test as flaky.
senorblanco@chromium.org [Thu, 1 Nov 2012 20:57:51 +0000 (20:57 +0000)]
[chromium] Unreviewed; marking another inspector test as flaky.

* platform/chromium/TestExpectations:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@133226 268f45cc-cd09-0410-ab3c-d52691b4dbfc

6 years ago[WK1] Fixed-position elements jiggle up and down slightly during scrolling on a Retin...
aroben@webkit.org [Thu, 1 Nov 2012 20:57:46 +0000 (20:57 +0000)]
[WK1] Fixed-position elements jiggle up and down slightly during scrolling on a Retina display
https://bugs.webkit.org/show_bug.cgi?id=100957

Reviewed by Simon Fraser.

WebCore doesn't yet support subpixel scrolling. WebKit2 forces
scrolling to always be integral. Now WebKit1 forces this as well.

I'm not sure how to write a test for this.

* WebView/WebDynamicScrollBarsView.mm:
(shouldRoundScrollOrigin): Returns YES if there are any position:fixed
or position:sticky elements in the page.
(-[WebDynamicScrollBarsView scrollClipView:toPoint:]): Round the
scroll origin to the nearest pixel if needed.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@133225 268f45cc-cd09-0410-ab3c-d52691b4dbfc

6 years agoPart2 of: Extend -webkit-user-select with new value "all"
enrica@apple.com [Thu, 1 Nov 2012 20:54:20 +0000 (20:54 +0000)]
Part2 of: Extend -webkit-user-select with new value "all"
<rdar://problem/10161404>
https://bugs.webkit.org/show_bug.cgi?id=91912

Reviewed by Ryosuke Niwa.

Source/WebCore:

The new value "all" for -webkit-user-select property gives content none-or-all selection option.
The patch was originally prepared by Alice Cheng but the approach has been changed.
The idea is to treat these elements like non editable, meaning that we should skip over them entirely
when moving the cursor and a deletion should delete the element and all its descentants at once.
The key change is in Node::rendererIsEditable where we now return false if the element style is
userSelect: all. The other change is in the way we create the selection on mouse click and dragging
over the element. In both cases we force the selection to extend over the entire element with
the user-select: all attribute.
This is currently enabled only for the Mac port.

Test: editing/selection/user-select-all-selection.html

* dom/Node.cpp: Added a parameter to isContentEditable to behave differently
when called from JavaScript. Internally isContentEditable returns false on
nodes with user-select: all style.
(WebCore::Node::isContentEditable):
(WebCore::Node::isContentRichlyEditable):
(WebCore::Node::rendererIsEditable):
(WebCore::Node::shouldUseInputMethod):
(WebCore::Node::willRespondToMouseClickEvents):
* dom/Node.h:
(WebCore::Node::rendererIsEditable):
(WebCore::Node::rendererIsRichlyEditable):
* dom/Position.cpp:
(WebCore::Position::nodeIsUserSelectAll): Added.
(WebCore::Position::rootUserSelectAllForNode): Added.
* dom/Position.h: Added static functions described above.
* editing/ApplyStyleCommand.cpp:
(WebCore::ApplyStyleCommand::removeInlineStyleFromElement): Added parameter to
isContentEditable() call.
(WebCore::ApplyStyleCommand::surroundNodeRangeWithElement): Added parameter to
isContentEditable() call.
* editing/DeleteFromTextNodeCommand.cpp:
(WebCore::DeleteFromTextNodeCommand::doApply): Added parameter to
isContentEditable() call.
* editing/FrameSelection.cpp:
(WebCore::adjustForwardPositionForUserSelectAll): New helper function.
(WebCore::adjustBackwardPositionForUserSelectAll): New helper function.
(WebCore::FrameSelection::modifyExtendingRight):
(WebCore::FrameSelection::modifyExtendingForward):
(WebCore::FrameSelection::modifyExtendingLeft):
(WebCore::FrameSelection::modifyExtendingBackward):
(WebCore::FrameSelection::modify):
(WebCore::CaretBase::invalidateCaretRect): Added parameter to
isContentEditable() call.
* editing/InsertNodeBeforeCommand.cpp:
(WebCore::InsertNodeBeforeCommand::doApply): Ditto.
(WebCore::InsertNodeBeforeCommand::doUnapply): Ditto.
* editing/RemoveNodeCommand.cpp:
(WebCore::RemoveNodeCommand::doApply): Ditto.
* editing/visible_units.cpp:
(WebCore::startOfParagraph): We should not consider a paragraph break and element
with user-select: all style, like we do at the border of editability.
(WebCore::endOfParagraph): Ditto.
* page/EventHandler.cpp:
(WebCore::EventHandler::updateSelectionForMouseDownDispatchingSelectStart): Create a selection
around the element whose style is user-select: all.
(WebCore::EventHandler::updateSelectionForMouseDrag): Ditto.
* rendering/RootInlineBox.cpp:
(WebCore::RootInlineBox::selectionState): Fixed a bug uncovered during this work.
If the selection starts in one of the leaf boxes and after we encounter one with SelectionNone,
we should return the selection state as SelectionBoth, assuming we went past the end selection.
This avoids doing an incorrect gap filling for the selection highlighting.

LayoutTests:

Testing moving and extending selections with the new CSS propertyvalue, with mouse movements
and with the keyboard.
Updated test expectations for all the platforms, since this is only enabled for the Mac port.

* editing/selection/user-select-all-selection-expected.txt: Added.
* editing/selection/user-select-all-selection.html: Added.
* platform/chromium/TestExpectations:
* platform/qt/TestExpectations:
* platform/gtk/TestExpectations:
* platform/win/TestExpectations:
* platform/efl/TestExpectations:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@133224 268f45cc-cd09-0410-ab3c-d52691b4dbfc

6 years agoIndexedDB: Fix Windows build by re-adding a #include
senorblanco@chromium.org [Thu, 1 Nov 2012 20:50:43 +0000 (20:50 +0000)]
IndexedDB: Fix Windows build by re-adding a #include
https://bugs.webkit.org/show_bug.cgi?id=100986

Patch by Alec Flett <alecflett@chromium.org> on 2012-11-01
Reviewed by Stephen White.

Fix #include that broke windows.

* Modules/indexeddb/IDBCallbacks.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@133223 268f45cc-cd09-0410-ab3c-d52691b4dbfc

6 years agoREGRESSION (r128837): mathml/presentation/subsup.xhtml became flaky
dbarton@mathscribe.com [Thu, 1 Nov 2012 20:42:29 +0000 (20:42 +0000)]
REGRESSION (r128837): mathml/presentation/subsup.xhtml became flaky
https://bugs.webkit.org/show_bug.cgi?id=97390

Reviewed by Ojan Vafai.

Source/WebCore:

In a RenderMathMLSubSup, we need to set the base's wrapper's style so that baseHeight
in layout() will be the base's true height, without any flexbox stretching.

Tested by existing tests.

* rendering/mathml/RenderMathMLSubSup.cpp:
(WebCore::RenderMathMLSubSup::fixAnonymousStyles):
(WebCore::RenderMathMLSubSup::addChild):
(WebCore::RenderMathMLSubSup::styleDidChange):
* rendering/mathml/RenderMathMLSubSup.h:
(RenderMathMLSubSup):

LayoutTests:

* platform/chromium/TestExpectations:
* platform/efl/TestExpectations:
* platform/gtk/TestExpectations:
* platform/mac/TestExpectations:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@133221 268f45cc-cd09-0410-ab3c-d52691b4dbfc

6 years ago[chromium] Unreviewed gardening. Mark an inspector test as flaky.
senorblanco@chromium.org [Thu, 1 Nov 2012 20:30:04 +0000 (20:30 +0000)]
[chromium] Unreviewed gardening.  Mark an inspector test as flaky.

* platform/chromium/TestExpectations:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@133218 268f45cc-cd09-0410-ab3c-d52691b4dbfc

6 years ago[chromium] Unreviewed gardening.
senorblanco@chromium.org [Thu, 1 Nov 2012 20:25:31 +0000 (20:25 +0000)]
[chromium] Unreviewed gardening.

* platform/chromium-linux-x86/compositing/reflections/nested-reflection-anchor-point-expected.png: Removed.
* platform/chromium-mac/compositing/reflections/nested-reflection-anchor-point-expected.png:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@133217 268f45cc-cd09-0410-ab3c-d52691b4dbfc

6 years ago[chromium] Unreviewed gardening. Move baselines for a couple of
senorblanco@chromium.org [Thu, 1 Nov 2012 20:23:43 +0000 (20:23 +0000)]
[chromium] Unreviewed gardening.  Move baselines for a couple of
compositing tests to platform-mac, so MountainLion can use them too.

* platform/chromium-mac-lion/compositing/geometry/vertical-scroll-composited-expected.png: Removed.
* platform/chromium-mac-lion/platform/chromium/compositing/huge-layer-rotated-expected.png: Removed.
* platform/chromium-mac/compositing/geometry/vertical-scroll-composited-expected.png:
* platform/chromium-mac/platform/chromium/compositing/huge-layer-rotated-expected.png:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@133216 268f45cc-cd09-0410-ab3c-d52691b4dbfc

6 years agoREGRESSION (r132941): attribute modification 10% performance regression
antti@apple.com [Thu, 1 Nov 2012 19:41:08 +0000 (19:41 +0000)]
REGRESSION (r132941): attribute modification 10% performance regression
https://bugs.webkit.org/show_bug.cgi?id=100873

Reviewed by Ojan Vafai.

Don't do the class change finding by mutating SpaceSplitString. It is slow. Instead use a bit vector
to mark the unchanged classes

* css/StyleResolver.cpp:
(WebCore):
* css/StyleResolver.h:
(WebCore::StyleResolver::hasSelectorForAttribute):
(WebCore):
(WebCore::StyleResolver::hasSelectorForClass):
(WebCore::StyleResolver::hasSelectorForId):

    Inlined these and moved value validity testing to clients.

* dom/Element.cpp:
(WebCore::checkNeedsStyleInvalidationForIdChange):
(WebCore):
(WebCore::Element::attributeChanged):

    Clean up id testing too.

(WebCore::checkNeedsStyleInvalidationForClassChange):

    Use bit vector for marking seen values. Avoids allocations and reffing.

(WebCore::Element::classAttributeChanged):

    Don't test if style is already invalid.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@133214 268f45cc-cd09-0410-ab3c-d52691b4dbfc

6 years ago[chromium] Unreviewed gardening. Mark yet another test as flaky.
senorblanco@chromium.org [Thu, 1 Nov 2012 19:31:48 +0000 (19:31 +0000)]
[chromium] Unreviewed gardening.  Mark yet another test as flaky.

* platform/chromium/TestExpectations:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@133213 268f45cc-cd09-0410-ab3c-d52691b4dbfc

6 years ago[chromium] Unreviewed gardening. Mark a test as flaky.
senorblanco@chromium.org [Thu, 1 Nov 2012 19:30:12 +0000 (19:30 +0000)]
[chromium] Unreviewed gardening.  Mark a test as flaky.

* platform/chromium/TestExpectations:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@133212 268f45cc-cd09-0410-ab3c-d52691b4dbfc

6 years agoAdd LLDB-Python scripts to support WebKit data types.
jer.noble@apple.com [Thu, 1 Nov 2012 19:16:08 +0000 (19:16 +0000)]
Add LLDB-Python scripts to support WebKit data types.
https://bugs.webkit.org/show_bug.cgi?id=100898

Reviewed by Anders Carlsson.

Add Summary providers for common WebKit string types.

* lldb/lldb_webkit.py: Added.
(__lldb_init_module): Register these providers with the debugger.
(WTFString_SummaryProvider): Wrapper around WTFStringProvider.
(WTFStringImpl_SummaryProvider): Wrapper around WTFStringImplProvider.
(WTFAtomicString_SummaryProvider): Ditto.
(guess_string_length): Walk through the string looking for a null-byte(s).
(ustring_to_string): Walk through the string, extracting two-byte values, and concatenate them.
(lstring_to_string): Walk through the string, extracting byte values, and concatenate them.
(WTFStringImplProvider):
(WTFStringImplProvider.__init__):
(WTFStringImplProvider.get_length): Extract the m_length variable.
(WTFStringImplProvider.get_data8): Extract the m_data8 variable.
(WTFStringImplProvider.get_data16): Extract the m_data16 variable.
(WTFStringImplProvider.to_string): Pass the appropriate data variable to {u,l}string_to_string.
(WTFStringImplProvider.is_8bit): Check the m_hashAndFlags variable.
(WTFStringProvider):
(WTFStringProvider.__init__):
(WTFStringProvider.stringimpl): Extract the m_impl variable.
(WTFStringProvider.get_length): Pass through to WTFStringImplProvider.
(WTFStringProvider.to_string): Ditto.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@133211 268f45cc-cd09-0410-ab3c-d52691b4dbfc

6 years ago[Mac] Crash in Range::editingStartPosition
rniwa@webkit.org [Thu, 1 Nov 2012 19:02:16 +0000 (19:02 +0000)]
[Mac] Crash in Range::editingStartPosition
https://bugs.webkit.org/show_bug.cgi?id=100972

Reviewed by Enrica Casucci.

Add a missing null pointer check. toNormalizedRange() can return null here, in which case the call to
pastLastNode() will crash.

No new tests since determining the exact condition under which we go through the said code path and
toNormalRange returns null is hard.

* editing/mac/EditorMac.mm:
(WebCore::Editor::fontForSelection):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@133207 268f45cc-cd09-0410-ab3c-d52691b4dbfc

6 years agoUnreviewed, remove failing expectations after rebaseline.
danakj@chromium.org [Thu, 1 Nov 2012 19:01:11 +0000 (19:01 +0000)]
Unreviewed, remove failing expectations after rebaseline.

* platform/chromium/TestExpectations:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@133206 268f45cc-cd09-0410-ab3c-d52691b4dbfc

6 years ago.: https://bugs.webkit.org/show_bug.cgi?id=100917
bdakin@apple.com [Thu, 1 Nov 2012 18:57:39 +0000 (18:57 +0000)]
.: https://bugs.webkit.org/show_bug.cgi?id=100917
There should be a way to dump the scrolling tree from the layout tests

Reviewed by Simon Fraser.

* Source/autotools/symbols.filter:

Source/WebCore: https://bugs.webkit.org/show_bug.cgi?id=100917
There should be a way to dump the scrolling tree from the layout tests

Reviewed by Simon Fraser.

New Internals call to dump the scrolling state tree as text. Calls through to the
ScrollingCoordinator.
* testing/Internals.cpp:
(WebCore::Internals::scrollingStateTreeAsText):
(WebCore):
* testing/Internals.h:
* testing/Internals.idl:
* WebCore.exp.in:
* page/Page.cpp:
(WebCore::Page::scrollingStateTreeAsText):
(WebCore):
* page/Page.h:
(Page):

ScrollingCoordinator tells the rootStateNode to dump the tree as text.
* page/scrolling/mac/ScrollingCoordinatorMac.h:
(ScrollingCoordinatorMac):
* page/scrolling/mac/ScrollingCoordinatorMac.mm:
(WebCore::ScrollingCoordinatorMac::scrollingStateTreeAsText):
(WebCore):
* page/scrolling/ScrollingCoordinator.cpp:
(WebCore::ScrollingCoordinator::scrollingStateTreeAsText):
(WebCore):
* page/scrolling/ScrollingCoordinator.h:
(ScrollingCoordinator):

dumpNode() dumps generic stuff for each node, and dumpProperties will dump the
properties that are specific to different types of nodes.
* page/scrolling/ScrollingStateNode.cpp:
(WebCore::ScrollingStateNode::writeIndent):
(WebCore):
(WebCore::ScrollingStateNode::dumpNode):
(WebCore::ScrollingStateNode::scrollingStateTreeAsText):
* page/scrolling/ScrollingStateNode.h:
(WebCore):
(ScrollingStateNode):
* page/scrolling/ScrollingStateScrollingNode.cpp:
(WebCore::ScrollingStateScrollingNode::dumpProperties):
(WebCore):
* page/scrolling/ScrollingStateScrollingNode.h:
(ScrollingStateScrollingNode):

Source/WebKit2: https://bugs.webkit.org/show_bug.cgi?id=100917
There should be a way to dump the scrolling tree from the layout tests

Reviewed by Simon Fraser.

* win/WebKit2.def:
* win/WebKit2CFLite.def:

LayoutTests: https://bugs.webkit.org/show_bug.cgi?id=100917
There should be a way to dump the scrolling tree from the layout tests

Reviewed by Simon Fraser.

* platform/mac/tiled-drawing/scrolling-tree-after-scroll-expected.txt: Added.
* platform/mac/tiled-drawing/scrolling-tree-after-scroll.html: Added.
* platform/mac/tiled-drawing/scrolling-tree-slow-scrolling-expected.txt: Added.
* platform/mac/tiled-drawing/scrolling-tree-slow-scrolling.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@133205 268f45cc-cd09-0410-ab3c-d52691b4dbfc

6 years agoRebaseline after compositor migration to gfx types.
danakj@chromium.org [Thu, 1 Nov 2012 18:56:11 +0000 (18:56 +0000)]
Rebaseline after compositor migration to gfx types.
http://crbug.com/147395

Unreviewed rebaselines.

* platform/chromium-linux/compositing/geometry/fixed-position-transform-composited-page-scale-down-expected.png:
* platform/chromium-linux/compositing/geometry/vertical-scroll-composited-expected.png:
* platform/chromium-linux/compositing/reflections/nested-reflection-anchor-point-expected.png:
* platform/chromium-linux/platform/chromium/compositing/huge-layer-rotated-expected.png:
* platform/chromium-linux/transforms/3d/point-mapping/3d-point-mapping-3-expected.png:
* platform/chromium-linux/transforms/3d/point-mapping/3d-point-mapping-preserve-3d-expected.png:
* platform/chromium-mac-lion/compositing/geometry/vertical-scroll-composited-expected.png: Added.
* platform/chromium-mac-lion/compositing/reflections/nested-reflection-anchor-point-expected.png:
* platform/chromium-mac-lion/platform/chromium/compositing/huge-layer-rotated-expected.png: Added.
* platform/chromium-mac-lion/transforms/3d/point-mapping/3d-point-mapping-3-expected.png: Added.
* platform/chromium-mac-lion/transforms/3d/point-mapping/3d-point-mapping-origins-expected.png: Added.
* platform/chromium-mac-lion/transforms/3d/point-mapping/3d-point-mapping-preserve-3d-expected.png: Added.
* platform/chromium-mac-snowleopard/compositing/reflections/nested-reflection-anchor-point-expected.png:
* platform/chromium-mac-snowleopard/transforms/3d/point-mapping/3d-point-mapping-3-expected.png:
* platform/chromium-mac-snowleopard/transforms/3d/point-mapping/3d-point-mapping-origins-expected.png:
* platform/chromium-mac-snowleopard/transforms/3d/point-mapping/3d-point-mapping-preserve-3d-expected.png:
* platform/chromium-win-xp/compositing/reflections/nested-reflection-anchor-point-expected.png: Removed.
* platform/efl-wk1/compositing/geometry/fixed-position-transform-composited-page-scale-down-expected.png: Renamed from LayoutTests/platform/efl/compositing/geometry/fixed-position-transform-composited-page-scale-down-expected.png.
* platform/efl-wk1/compositing/reflections/nested-reflection-anchor-point-expected.png: Renamed from LayoutTests/platform/efl/compositing/reflections/nested-reflection-anchor-point-expected.png.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@133204 268f45cc-cd09-0410-ab3c-d52691b4dbfc

6 years ago[chromium] Provide WebKit API interface for platform gesture curves
rjkroege@chromium.org [Thu, 1 Nov 2012 18:56:07 +0000 (18:56 +0000)]
[chromium] Provide WebKit API interface for platform gesture curves
https://bugs.webkit.org/show_bug.cgi?id=100675

Reviewed by James Robinson.

Define interfaces to permit the Chromium embedder to provide its own
implementation of a gesture animation curve and add an entry point
to Platform.h that permits WebKit to obtain an instance of an embedder-
provided gesture curve.

No tests needed in this CL: only interfaces are being added.

* Platform.gypi:
* chromium/public/Platform.h:
(WebKit):
(Platform):
(WebKit::Platform::createFlingAnimationCurve):
* chromium/public/WebGestureCurve.h: Added.
(WebKit):
(WebGestureCurve):
(WebKit::WebGestureCurve::~WebGestureCurve):
* chromium/public/WebGestureCurveTarget.h: Added.
(WebKit):
(WebGestureCurveTarget):
(WebKit::WebGestureCurveTarget::~WebGestureCurveTarget):
* chromium/src/WebActiveGestureAnimation.cpp: Added.
(WebKit):
(WebKit::WebActiveGestureAnimation::createAtAnimationStart):
(WebKit::WebActiveGestureAnimation::createWithTimeOffset):
(WebKit::WebActiveGestureAnimation::~WebActiveGestureAnimation):
(WebKit::WebActiveGestureAnimation::WebActiveGestureAnimation):
(WebKit::WebActiveGestureAnimation::animate):
* chromium/src/WebActiveGestureAnimation.h: Added.
(WebKit):
(WebActiveGestureAnimation):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@133203 268f45cc-cd09-0410-ab3c-d52691b4dbfc

6 years agoAdd RGB to supported destination formats of ImageBuffer::copyToPlatformTexture
kbr@google.com [Thu, 1 Nov 2012 18:53:40 +0000 (18:53 +0000)]
Add RGB to supported destination formats of ImageBuffer::copyToPlatformTexture
https://bugs.webkit.org/show_bug.cgi?id=100971

Reviewed by Stephen White.

The ImageBuffer::copyToPlatformTexture restriction added in
r132965 was too restrictive. Allow RGB textures to prevent
performance regressions.

* html/canvas/WebGLRenderingContext.cpp:
(WebCore):
(WebCore::WebGLRenderingContext::texImage2D):
* platform/graphics/ImageBuffer.h:
(ImageBuffer):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@133202 268f45cc-cd09-0410-ab3c-d52691b4dbfc

6 years agoUnreviewed, marking tests as failing preemptively.
danakj@chromium.org [Thu, 1 Nov 2012 18:41:06 +0000 (18:41 +0000)]
Unreviewed, marking tests as failing preemptively.
http://crbug.com/147395

* platform/chromium/TestExpectations:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@133201 268f45cc-cd09-0410-ab3c-d52691b4dbfc

6 years ago[Chromium-Android] Apache doesn't properly clean up ipc semaphores after a layout...
peter@chromium.org [Thu, 1 Nov 2012 18:40:29 +0000 (18:40 +0000)]
[Chromium-Android] Apache doesn't properly clean up ipc semaphores after a layout test run
https://bugs.webkit.org/show_bug.cgi?id=100950

Reviewed by Dirk Pranke.

When a test run would fail to complete due to an exception in one of
the workers, the HTTP server wouldn't get a chance to gracefully shut
down. This caused too much IPC semaphores to be left on the server,
causing Apache to fail to start in subsequent runs.

By unifying the Android-specific code with other ports, we no longer
fail to call the ChromiumPort/Base setup_test_run() and clean_up_test_run()
methods either. Furthermore, the number_of_servers argument for starting
the HTTP server is now available as well.

Because not all tests require an HTTP server, it's not guaranteed that
it will be started. Android depends on this, so add a new method to Port
and override it for Android: requires_http_server().

* Scripts/webkitpy/layout_tests/controllers/layout_test_runner.py:
(LayoutTestRunner.run_tests):
* Scripts/webkitpy/layout_tests/controllers/manager.py:
(Manager._run_tests):
* Scripts/webkitpy/layout_tests/port/base.py:
(Port.to.requires_http_server):
* Scripts/webkitpy/layout_tests/port/base_unittest.py:
(PortTest.test_dont_require_http_server):
* Scripts/webkitpy/layout_tests/port/chromium_android.py:
(ChromiumAndroidPort.requires_http_server):
(ChromiumAndroidPort.start_http_server):
* Scripts/webkitpy/layout_tests/port/chromium_android_unittest.py:
(ChromiumAndroidPortTest.test_must_require_http_server):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@133200 268f45cc-cd09-0410-ab3c-d52691b4dbfc

6 years ago[BlackBerry] Web page view state should be preserved for pages loaded from page cache
commit-queue@webkit.org [Thu, 1 Nov 2012 18:37:23 +0000 (18:37 +0000)]
[BlackBerry] Web page view state should be preserved for pages loaded from page cache
https://bugs.webkit.org/show_bug.cgi?id=100694
Internal PR: 220488

Internally reviewed by Lianghui Chen, Joe Mason
Patch by Michael Matovsky <mmatovsky@rim.com> on 2012-11-01
Reviewed by Rob Buis.

The web page view specific metadata should be preserved for web pages loaded from page cache (by using back/forward).
This requires web page view state to be saved in and restored from the corresponding page cache history item.

Source/WebCore:

No new tests for platform specific internal change.

* history/blackberry/HistoryItemViewState.h:
(HistoryItemViewState):

Source/WebKit/blackberry:

* Api/WebPageClient.h:
* WebCoreSupport/FrameLoaderClientBlackBerry.cpp:
(WebCore::FrameLoaderClientBlackBerry::saveViewStateToItem):
(WebCore::FrameLoaderClientBlackBerry::restoreViewState):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@133199 268f45cc-cd09-0410-ab3c-d52691b4dbfc

6 years ago[chromium] Allow implicit conversion between gfx::Vector2d and WebSize
danakj@chromium.org [Thu, 1 Nov 2012 18:36:32 +0000 (18:36 +0000)]
[chromium] Allow implicit conversion between gfx::Vector2d and WebSize
https://bugs.webkit.org/show_bug.cgi?id=100553

Reviewed by James Robinson.

The gfx::Vector2d class is used in chromium for defining distances. The
IntSize class is used equivalently inside WebCore. This lets us convert
between vectors and sizes along the API boundary.

* chromium/public/WebSize.h:
(WebKit::WebSize::WebSize):
(WebSize):
(WebKit::WebSize::operator=):
(WebKit::WebSize::operator gfx::Vector2d):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@133198 268f45cc-cd09-0410-ab3c-d52691b4dbfc

6 years ago[chromium] Unreviewed gardening. Mark a test as flaky.
senorblanco@chromium.org [Thu, 1 Nov 2012 18:28:56 +0000 (18:28 +0000)]
[chromium] Unreviewed gardening.  Mark a test as flaky.

* platform/chromium/TestExpectations:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@133197 268f45cc-cd09-0410-ab3c-d52691b4dbfc

6 years agoUnreviewed. Skip fast/workers/worker-domurl.html on Windows. ENABLE_BLOB not enabled.
roger_fong@apple.com [Thu, 1 Nov 2012 18:12:01 +0000 (18:12 +0000)]
Unreviewed. Skip fast/workers/worker-domurl.html on Windows. ENABLE_BLOB not enabled.

* platform/win/TestExpectations:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@133196 268f45cc-cd09-0410-ab3c-d52691b4dbfc

6 years agoIndexedDB: add methods to support id-based backend APIs
alecflett@chromium.org [Thu, 1 Nov 2012 18:07:22 +0000 (18:07 +0000)]
IndexedDB: add methods to support id-based backend APIs
https://bugs.webkit.org/show_bug.cgi?id=100425

Reviewed by Tony Chang.

Source/WebCore:

First half of refactor involves adding a number of methods that
are int64_t-based rather than String-based. As a part of this, the
IDB*Metadata structs and the backend objectStore/index maps all
use int64_t as keys, rather than String.

In addition, there were a number of cleanups that came out of the
refactor:

- The list of object stores active in a transaction is now
  maintained by the frontend IDBTransaction rather than the backend
  IDBTransactionBackendImpl. This also had a simplifying rippling
  effect through other call signatures.

- I was able to remove an apparently old FIXME from
  IDBTransactionBackendImpl::objectStore and replace it with an ASSERT.

- IDBObjectStoreBackendImpl's IndexWriter class got a little
  simpler since the id is now easily available in the
  IDBIndexMetadata.

- A number of methods got simpler in their int64_t versions,
  specifically dropping a number of ExceptionCodes.

There is also some glue code
(getIndexId/getIndexIds/getObjectStoreId) that will go away with
the 2nd half of this: https://bugs.webkit.org/show_bug.cgi?id=100425

No new tests, no new functionality as this is just a refactor.

* Modules/indexeddb/IDBDatabase.cpp:
(WebCore::IDBDatabase::objectStoreNames):
(WebCore::IDBDatabase::createObjectStore):
(WebCore::IDBDatabase::deleteObjectStore):
(WebCore::IDBDatabase::transaction):
* Modules/indexeddb/IDBDatabaseBackendImpl.cpp:
(WebCore::IDBDatabaseBackendImpl::metadata):
(WebCore::IDBDatabaseBackendImpl::createObjectStore):
(WebCore::IDBDatabaseBackendImpl::objectStore):
(WebCore::IDBDatabaseBackendImpl::getObjectStoreId):
(WebCore):
(WebCore::IDBDatabaseBackendImpl::deleteObjectStore):
(WebCore::IDBDatabaseBackendImpl::transaction):
(WebCore::IDBDatabaseBackendImpl::loadObjectStores):
(WebCore::IDBDatabaseBackendImpl::removeObjectStoreFromMap):
(WebCore::IDBDatabaseBackendImpl::addObjectStoreToMap):
* Modules/indexeddb/IDBDatabaseBackendImpl.h:
(WebCore::IDBDatabaseBackendImpl::deleteObjectStore):
(IDBDatabaseBackendImpl):
* Modules/indexeddb/IDBDatabaseBackendInterface.h:
(IDBDatabaseBackendInterface):
* Modules/indexeddb/IDBIndex.h:
(WebCore::IDBIndex::id):
(WebCore::IDBIndex::openKeyCursor):
* Modules/indexeddb/IDBIndexBackendInterface.h:
* Modules/indexeddb/IDBMetadata.h:
(WebCore::IDBIndexMetadata::IDBIndexMetadata):
(IDBIndexMetadata):
(IDBObjectStoreMetadata):
(WebCore::IDBObjectStoreMetadata::containsIndex):
(IDBDatabaseMetadata):
(WebCore::IDBDatabaseMetadata::IDBDatabaseMetadata):
(WebCore::IDBDatabaseMetadata::containsObjectStore):
* Modules/indexeddb/IDBObjectStore.cpp:
(WebCore::IDBObjectStore::indexNames):
(WebCore::IDBObjectStore::put):
(WebCore::IDBObjectStore::createIndex):
(WebCore::IDBObjectStore::index):
(WebCore::IDBObjectStore::deleteIndex):
* Modules/indexeddb/IDBObjectStoreBackendImpl.cpp:
(WebCore::IDBObjectStoreBackendImpl::put):
(WebCore::IDBObjectStoreBackendImpl::putWithIndexKeys):
(WebCore):
(WebCore::makeIndexWriters):
(WebCore::IDBObjectStoreBackendImpl::setIndexKeys):
(WebCore::IDBObjectStoreBackendImpl::setIndexesReady):
(WebCore::IDBObjectStoreBackendImpl::setIndexesReadyInternal):
(WebCore::IDBObjectStoreBackendImpl::putInternal):
(WebCore::IDBObjectStoreBackendImpl::deleteInternal):
(WebCore::IDBObjectStoreBackendImpl::createIndex):
(WebCore::IDBObjectStoreBackendImpl::index):
(WebCore::IDBObjectStoreBackendImpl::getIndexId):
(WebCore::IDBObjectStoreBackendImpl::getIndexIds):
(WebCore::IDBObjectStoreBackendImpl::deleteIndex):
(WebCore::IDBObjectStoreBackendImpl::loadIndexes):
(WebCore::IDBObjectStoreBackendImpl::removeIndexFromMap):
(WebCore::IDBObjectStoreBackendImpl::addIndexToMap):
* Modules/indexeddb/IDBObjectStoreBackendImpl.h:
(IDBObjectStoreBackendImpl):
* Modules/indexeddb/IDBObjectStoreBackendInterface.h:
* Modules/indexeddb/IDBOpenDBRequest.cpp:
(WebCore::IDBOpenDBRequest::onUpgradeNeeded):
* Modules/indexeddb/IDBRequest.cpp:
(WebCore::IDBRequest::onSuccess):
(WebCore::IDBRequest::uncaughtExceptionInEventHandler):
* Modules/indexeddb/IDBTransaction.cpp:
(WebCore::IDBTransaction::create):
(WebCore::IDBTransaction::IDBTransaction):
(WebCore::IDBTransaction::objectStore):
* Modules/indexeddb/IDBTransaction.h:
(IDBTransaction):
* Modules/indexeddb/IDBTransactionBackendImpl.cpp:
(WebCore::IDBTransactionBackendImpl::create):
(WebCore::IDBTransactionBackendImpl::IDBTransactionBackendImpl):
(WebCore::IDBTransactionBackendImpl::objectStore):
(WebCore):
(WebCore::IDBTransactionBackendImpl::scheduleTask):
* Modules/indexeddb/IDBTransactionBackendImpl.h:
(IDBTransactionBackendImpl):
* Modules/indexeddb/IDBTransactionBackendInterface.h:

Source/WebKit/chromium:

Chromium Proxy and WebIDB* support for new int64_t-based methods.

* public/WebIDBDatabase.h:
(WebKit::WebIDBDatabase::deleteObjectStore):
(WebKit::WebIDBDatabase::transaction):
* public/WebIDBObjectStore.h:
(WebKit::WebIDBObjectStore::put):
(WebKit::WebIDBObjectStore::setIndexKeys):
(WebKit::WebIDBObjectStore::setIndexesReady):
(WebKit::WebIDBObjectStore::index):
(WebKit::WebIDBObjectStore::deleteIndex):
* public/WebIDBTransaction.h:
(WebKit::WebIDBTransaction::objectStore):
* src/IDBDatabaseBackendProxy.cpp:
(WebKit::IDBDatabaseBackendProxy::deleteObjectStore):
(WebKit):
(WebKit::IDBDatabaseBackendProxy::transaction):
* src/IDBDatabaseBackendProxy.h:
(IDBDatabaseBackendProxy):
* src/IDBObjectStoreBackendProxy.cpp:
(WebKit::IDBObjectStoreBackendProxy::put):
(WebKit):
(WebKit::IDBObjectStoreBackendProxy::setIndexKeys):
(WebKit::IDBObjectStoreBackendProxy::setIndexesReady):
(WebKit::IDBObjectStoreBackendProxy::index):
(WebKit::IDBObjectStoreBackendProxy::deleteIndex):
* src/IDBObjectStoreBackendProxy.h:
(IDBObjectStoreBackendProxy):
* src/IDBTransactionBackendProxy.cpp:
(WebKit::IDBTransactionBackendProxy::objectStore):
* src/IDBTransactionBackendProxy.h:
(IDBTransactionBackendProxy):
* src/WebIDBMetadata.cpp:
(WebKit::WebIDBMetadata::WebIDBMetadata):
(WebKit::WebIDBMetadata::operator IDBDatabaseMetadata):
* src/WebIDBTransactionImpl.cpp:
(WebKit::WebIDBTransactionImpl::objectStore):
(WebKit):
* src/WebIDBTransactionImpl.h:
* tests/IDBDatabaseBackendTest.cpp:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@133195 268f45cc-cd09-0410-ab3c-d52691b4dbfc

6 years ago[V8] The DOMWrapperVisitor abstraction is no longer needed
abarth@webkit.org [Thu, 1 Nov 2012 17:48:32 +0000 (17:48 +0000)]
[V8] The DOMWrapperVisitor abstraction is no longer needed
https://bugs.webkit.org/show_bug.cgi?id=100965

Reviewed by Kentaro Hara.

This patch removes the DOMWrapperVisitor interface because it is no
longer needed. As a consequence, DOMWrapperMaps no longer need to
support enumeration, and we can move more DOM objects to use the faster
intrusive wrappers.

There was one remaining user of DOMWrapperVisitor in the
ScriptProfiler, which I've moved over to enumerating objects directly
from V8, similar to a function above it in the same file.

* bindings/v8/DOMWrapperMap.h:
(WebCore):
(DOMWrapperMap):
* bindings/v8/IntrusiveDOMWrapperMap.h:
* bindings/v8/ScriptProfiler.cpp:
(WebCore::ScriptProfiler::visitNodeWrappers):
(WebCore::ScriptProfiler::visitExternalArrays):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@133194 268f45cc-cd09-0410-ab3c-d52691b4dbfc

6 years agoCSP 1.0: Warn when old-style directives encountered.
mkwst@chromium.org [Thu, 1 Nov 2012 17:46:28 +0000 (17:46 +0000)]
CSP 1.0: Warn when old-style directives encountered.
https://bugs.webkit.org/show_bug.cgi?id=100883

Reviewed by Adam Barth.

Source/WebCore:

In Mozilla's pre-W3C-spec implementation, a few directives are
implemented that were either renamed, reworked, or removed from CSP 1.0.
This patch adds special warning messages for three of those directives
to set developer expectations correctly.

Test: http/tests/security/contentSecurityPolicy/source-list-parsing-deprecated.html

* page/ContentSecurityPolicy.cpp:
(WebCore::CSPDirectiveList::parseDirective):
(WebCore::CSPDirectiveList::addDirective):
(WebCore::ContentSecurityPolicy::reportUnsupportedDirective):
* page/ContentSecurityPolicy.h:
    Rename 'reportUnrecognizedDirectives' to
    'reportUnsupportedDirectives', and teach it to give more descriptive
    error messages when encountering 'allow', 'options', and
    'policy-uri'.

LayoutTests:

* http/tests/security/contentSecurityPolicy/directive-parsing-01-expected.txt:
* http/tests/security/contentSecurityPolicy/directive-parsing-multiple-headers-expected.txt:
* http/tests/security/contentSecurityPolicy/inline-script-blocked-goofy-expected.txt:
* http/tests/security/contentSecurityPolicy/inline-script-blocked-javascript-url-expected.txt:
* http/tests/security/contentSecurityPolicy/javascript-url-allowed-expected.txt:
* http/tests/security/contentSecurityPolicy/javascript-url-blocked-expected.txt:
* http/tests/security/contentSecurityPolicy/script-src-in-iframe-expected.txt:
* http/tests/security/contentSecurityPolicy/script-src-none-expected.txt:
* http/tests/security/contentSecurityPolicy/source-list-parsing-deprecated-expected.txt: Added.
* http/tests/security/contentSecurityPolicy/source-list-parsing-deprecated.html: Added.
* platform/chromium/http/tests/security/contentSecurityPolicy/inline-script-blocked-javascript-url-expected.txt:
* platform/chromium/http/tests/security/contentSecurityPolicy/javascript-url-blocked-expected.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@133193 268f45cc-cd09-0410-ab3c-d52691b4dbfc

6 years agoUnreviewed. Rolled DEPS.
commit-queue@webkit.org [Thu, 1 Nov 2012 17:45:01 +0000 (17:45 +0000)]
Unreviewed.  Rolled DEPS.

Patch by Sheriff Bot <webkit.review.bot@gmail.com> on 2012-11-01

* DEPS:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@133192 268f45cc-cd09-0410-ab3c-d52691b4dbfc

6 years agoWeb Inspector: expose Inspector.detached event factory to the embedder.
pfeldman@chromium.org [Thu, 1 Nov 2012 17:42:42 +0000 (17:42 +0000)]
Web Inspector: expose Inspector.detached event factory to the embedder.
https://bugs.webkit.org/show_bug.cgi?id=100967

Reviewed by Vsevolod Vlasov.

That way embedder will be able to send messages on behalf of the backend being destructed.

* public/WebDevToolsAgent.h:
(WebDevToolsAgent):
* src/WebDevToolsAgentImpl.cpp:
(SerializingFrontendChannel):
(WebKit::SerializingFrontendChannel::sendMessageToFrontend):
(WebKit::WebDevToolsAgent::inspectorDetachedEvent):
(WebKit):
(WebKit::WebDevToolsAgent::workerDisconnectedFromWorkerEvent):
(WebKit::WebDevToolsAgent::disconnectEventAsText):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@133191 268f45cc-cd09-0410-ab3c-d52691b4dbfc

6 years ago[Chromium] Unreviewed, marking http/tests/xmlhttprequest/xmlhttprequest-test-send...
tony@chromium.org [Thu, 1 Nov 2012 17:31:01 +0000 (17:31 +0000)]
[Chromium] Unreviewed, marking http/tests/xmlhttprequest/xmlhttprequest-test-send-flag.html
as slow on Linux Debug based on the flakiness dashboard.

* platform/chromium/TestExpectations:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@133190 268f45cc-cd09-0410-ab3c-d52691b4dbfc

6 years agoWeb Inspector: Update CodeMirror to v3
pfeldman@chromium.org [Thu, 1 Nov 2012 17:05:21 +0000 (17:05 +0000)]
Web Inspector: Update CodeMirror to v3
https://bugs.webkit.org/show_bug.cgi?id=99319

Reviewed by Vsevolod Vlasov.

Updated to ToT v3.

* inspector/front-end/CodeMirrorTextEditor.js:
(WebInspector.CodeMirrorTextEditor):
(WebInspector.CodeMirrorTextEditor.prototype._gutterClick):
(WebInspector.CodeMirrorTextEditor.prototype.addBreakpoint):
(WebInspector.CodeMirrorTextEditor.prototype.removeBreakpoint):
(WebInspector.CodeMirrorTextEditor.prototype.setExecutionLine):
(WebInspector.CodeMirrorTextEditor.prototype.clearExecutionLine):
(WebInspector.CodeMirrorTextEditor.prototype.highlightLine):
(WebInspector.CodeMirrorTextEditor.prototype.clearLineHighlight):
(WebInspector.CodeMirrorTextEditor.prototype._change):
* inspector/front-end/cm/cmdevtools.css:
(.CodeMirror):
(.CodeMirror-linenumber):
(.cm-breakpoint):
* inspector/front-end/cm/codemirror.css:
(.CodeMirror):
(.CodeMirror-scroll):
(.CodeMirror-lines):
(.CodeMirror pre):
(.CodeMirror-scrollbar-filler):
(.CodeMirror-gutters):
(.CodeMirror-linenumbers):
(.CodeMirror-linenumber):
(.CodeMirror pre.CodeMirror-cursor):
(.CodeMirror pre.CodeMirror-secondarycursor):
(.cm-keymap-fat-cursor pre.CodeMirror-cursor):
(.cm-keymap-fat-cursor pre.CodeMirror-cursor:not(#nonsense_id)):
(.CodeMirror pre.CodeMirror-cursor.CodeMirror-overwrite):
(.cm-s-default .cm-keyword):
(.cm-s-default .cm-atom):
(.cm-s-default .cm-number):
(.cm-s-default .cm-def):
(.cm-s-default .cm-variable):
(.cm-s-default .cm-variable-2):
(.cm-s-default .cm-variable-3):
(.cm-s-default .cm-property):
(.cm-s-default .cm-operator):
(.cm-s-default .cm-comment):
(.cm-s-default .cm-string):
(.cm-s-default .cm-string-2):
(.cm-s-default .cm-meta):
(.cm-s-default .cm-error):
(.cm-s-default .cm-qualifier):
(.cm-s-default .cm-builtin):
(.cm-s-default .cm-bracket):
(.cm-s-default .cm-tag):
(.cm-s-default .cm-attribute):
(.cm-s-default .cm-header):
(.cm-s-default .cm-quote):
(.cm-s-default .cm-hr):
(.cm-s-default .cm-link):
(.cm-header, .cm-strong):
(.cm-em):
(.cm-emstrong):
(.cm-link):
(.cm-invalidchar):
(div.CodeMirror span.CodeMirror-matchingbracket):
(div.CodeMirror span.CodeMirror-nonmatchingbracket):
(.CodeMirror-sizer):
(.CodeMirror-vscrollbar, .CodeMirror-hscrollbar, .CodeMirror-scrollbar-filler):
(.CodeMirror-vscrollbar):
(.CodeMirror-hscrollbar):
(.CodeMirror-gutter):
(.CodeMirror-gutter-elt):
(.CodeMirror-linebackground):
(.CodeMirror-linewidget):
(.CodeMirror-measure):
(.CodeMirror-measure pre):
(.CodeMirror-selected):
(.CodeMirror-focused .CodeMirror-selected):
(.CodeMirror span):
* inspector/front-end/cm/codemirror.js:
(window.CodeMirror.):
(window.CodeMirror):
* inspector/front-end/utilities.js:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@133189 268f45cc-cd09-0410-ab3c-d52691b4dbfc

6 years agoSource/WebCore: [BlackBerry] Update BB10 form theme.
commit-queue@webkit.org [Thu, 1 Nov 2012 16:51:17 +0000 (16:51 +0000)]
Source/WebCore: [BlackBerry] Update BB10 form theme.
https://bugs.webkit.org/show_bug.cgi?id=100760

Patch by Tiancheng Jiang <tijiang@rim.com> on 2012-11-01
Reviewed by Rob Buis.

RIM PR 235194.

Check img pointer is null, if so, do early return.

* platform/blackberry/RenderThemeBlackBerry.cpp:
(WebCore::drawControl):
(WebCore::drawThreeSlice):
(WebCore::drawNineSlice):

Source/WebKit/chromium: [Chromium] debug builds: Use after free under ~PageOverlay()
https://bugs.webkit.org/show_bug.cgi?id=100930

Patch by Vsevolod Vlasov <vsevik@chromium.org> on 2012-11-01
Reviewed by Pavel Feldman.

Changed members order in PageOverlay so that GraphicsLayer is destroyed before GraphicsLayerClient.

* src/PageOverlay.h:
(PageOverlay):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@133188 268f45cc-cd09-0410-ab3c-d52691b4dbfc

6 years ago[V8] The V8DOMMap visitors are no longer needed
abarth@webkit.org [Thu, 1 Nov 2012 16:47:53 +0000 (16:47 +0000)]
[V8] The V8DOMMap visitors are no longer needed
https://bugs.webkit.org/show_bug.cgi?id=100963

Reviewed by Kentaro Hara.

This patch inlines visitAllDOMNodes into its one caller (and removes
one layer of visitor adaptor abstraction).

* bindings/v8/ScriptProfiler.cpp:
(WebCore::ScriptProfiler::visitNodeWrappers):
* bindings/v8/V8DOMMap.cpp:
* bindings/v8/V8DOMMap.h:
(WebCore):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@133187 268f45cc-cd09-0410-ab3c-d52691b4dbfc

6 years ago[BlackBerry] Update BB10 form theme.
commit-queue@webkit.org [Thu, 1 Nov 2012 16:46:10 +0000 (16:46 +0000)]
[BlackBerry] Update BB10 form theme.
https://bugs.webkit.org/show_bug.cgi?id=100760

Patch by Tiancheng Jiang <tijiang@rim.com> on 2012-11-01
Reviewed by Rob Buis.

RIM PR 235194.

Check img pointer is null, if so, do early return.

* platform/blackberry/RenderThemeBlackBerry.cpp:
(WebCore::drawControl):
(WebCore::drawThreeSlice):
(WebCore::drawNineSlice):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@133186 268f45cc-cd09-0410-ab3c-d52691b4dbfc

6 years ago[chromium] Make WebLayerTreeView::adjustEventPointForPinchZoom pure virtual.
wjmaclean@chromium.org [Thu, 1 Nov 2012 16:45:41 +0000 (16:45 +0000)]
[chromium] Make WebLayerTreeView::adjustEventPointForPinchZoom pure virtual.
https://bugs.webkit.org/show_bug.cgi?id=100875

Reviewed by James Robinson.

This CL removes the default implementation for adjustEventPointForPinchZoom(), which
was necessary for landing the original patch in https://bugs.webkit.org/show_bug.cgi?id=100542.

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

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@133185 268f45cc-cd09-0410-ab3c-d52691b4dbfc

6 years ago[BlackBerry] Use clipped background rect in backing store
leoyang@rim.com [Thu, 1 Nov 2012 16:24:31 +0000 (16:24 +0000)]
[BlackBerry] Use clipped background rect in backing store
https://bugs.webkit.org/show_bug.cgi?id=100960

Reviewed by Yong Li.
Reviewed internally by Arvid Nilsson.

We calculated clipped background rect but didn't really use it when painting
the background. Use the correct rect to avoid ASSERT failure in platform code.

* Api/BackingStore.cpp:
(BlackBerry::WebKit::BackingStorePrivate::paintDefaultBackground):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@133184 268f45cc-cd09-0410-ab3c-d52691b4dbfc

6 years agoUnreviewed chromium windows rebaseline.
eae@chromium.org [Thu, 1 Nov 2012 16:11:56 +0000 (16:11 +0000)]
Unreviewed chromium windows rebaseline.

* platform/chromium-linux-x86/svg/zoom: Removed.
* platform/chromium-linux/svg/zoom/page/zoom-svg-through-object-with-percentage-size-expected.txt: Removed.
* platform/chromium-win/compositing/overflow/overflow-compositing-descendant-expected.png:
* platform/chromium-win/compositing/overflow/scroll-ancestor-update-expected.png:
* platform/chromium-win/svg/zoom/page/zoom-svg-through-object-with-absolute-size-2-expected.png:
* platform/chromium-win/svg/zoom/page/zoom-svg-through-object-with-absolute-size-2-expected.txt:
* platform/chromium-win/svg/zoom/page/zoom-svg-through-object-with-absolute-size-expected.txt:
* platform/chromium-win/svg/zoom/page/zoom-svg-through-object-with-auto-size-expected.png:
* platform/chromium-win/svg/zoom/page/zoom-svg-through-object-with-percentage-size-expected.png:
* platform/chromium-win/svg/zoom/page/zoom-svg-through-object-with-percentage-size-expected.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@133183 268f45cc-cd09-0410-ab3c-d52691b4dbfc

6 years agoCoordinated Graphics: Clean up headers of Coordinated Graphics.
commit-queue@webkit.org [Thu, 1 Nov 2012 15:55:53 +0000 (15:55 +0000)]
Coordinated Graphics: Clean up headers of Coordinated Graphics.
https://bugs.webkit.org/show_bug.cgi?id=100907

Patch by Huang Dongsung <luxtella@company100.net> on 2012-11-01
Reviewed by Noam Rosenthal.

This patch performs:
1. Remove unused methods.
2. Change public methods to private if needed.
3. Put virtual, OVERRIDE and explicit keywords if needed.

* UIProcess/CoordinatedGraphics/LayerTreeCoordinatorProxy.h:
(LayerTreeCoordinatorProxy):
* UIProcess/CoordinatedGraphics/LayerTreeRenderer.h:
(LayerTreeRenderer):
* WebProcess/WebPage/CoordinatedGraphics/CoordinatedGraphicsLayer.cpp:
* WebProcess/WebPage/CoordinatedGraphics/CoordinatedGraphicsLayer.h:
(CoordinatedGraphicsLayer):
(WebCore::CoordinatedGraphicsLayer::fixedToViewport):
(WebCore::CoordinatedGraphicsLayer::setMaskTarget):
* WebProcess/WebPage/CoordinatedGraphics/LayerTreeCoordinator.h:
(LayerTreeCoordinator):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@133182 268f45cc-cd09-0410-ab3c-d52691b4dbfc

6 years agoUnreviewed chromium rebaseline for r133172.
eae@chromium.org [Thu, 1 Nov 2012 15:51:20 +0000 (15:51 +0000)]
Unreviewed chromium rebaseline for r133172.

* platform/chromium-linux-x86/svg/zoom/page: Removed.
* platform/chromium-linux-x86/svg/zoom/page/zoom-svg-as-relative-image-expected.txt: Removed.
* platform/chromium-mac-lion/compositing/overflow/overflow-compositing-descendant-expected.png:
* platform/chromium-mac-lion/compositing/overflow/scroll-ancestor-update-expected.png:
* platform/chromium-mac-lion/svg/zoom/page/zoom-svg-through-object-with-absolute-size-2-expected.png: Added.
* platform/chromium-mac-lion/svg/zoom/page/zoom-svg-through-object-with-percentage-size-expected.png:
* platform/chromium-mac-snowleopard/compositing/overflow/overflow-compositing-descendant-expected.png:
* platform/chromium-mac-snowleopard/compositing/overflow/scroll-ancestor-update-expected.png:
* platform/chromium-mac-snowleopard/svg/zoom/page/zoom-svg-through-object-with-absolute-size-2-expected.png:
* platform/chromium-mac-snowleopard/svg/zoom/page/zoom-svg-through-object-with-percentage-size-expected.png:
* platform/chromium-mac/compositing/overflow/overflow-compositing-descendant-expected.png:
* platform/chromium-mac/compositing/overflow/scroll-ancestor-update-expected.png:
* platform/chromium-mac/svg/as-image/svg-non-integer-scaled-image-expected.png:
* platform/chromium-mac/svg/zoom/page/zoom-svg-through-object-with-absolute-size-2-expected.png:
* platform/chromium-mac/svg/zoom/page/zoom-svg-through-object-with-absolute-size-2-expected.txt:
* platform/chromium-mac/svg/zoom/page/zoom-svg-through-object-with-absolute-size-expected.txt:
* platform/chromium-mac/svg/zoom/page/zoom-svg-through-object-with-auto-size-expected.png:
* platform/chromium-mac/svg/zoom/page/zoom-svg-through-object-with-auto-size-expected.txt:
* platform/chromium-mac/svg/zoom/page/zoom-svg-through-object-with-percentage-size-expected.png:
* platform/chromium-mac/svg/zoom/page/zoom-svg-through-object-with-percentage-size-expected.txt:
* platform/efl/compositing/overflow/overflow-compositing-descendant-expected.png: Removed.
* platform/efl/compositing/overflow/scroll-ancestor-update-expected.png: Removed.
* platform/efl/svg/zoom/page/zoom-svg-through-object-with-absolute-size-2-expected.png: Removed.
* platform/efl/svg/zoom/page/zoom-svg-through-object-with-auto-size-expected.png: Removed.
* platform/efl/svg/zoom/page/zoom-svg-through-object-with-percentage-size-expected.png: Removed.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@133181 268f45cc-cd09-0410-ab3c-d52691b4dbfc

6 years agoAdd suppressions for images for new skia roll
reed@google.com [Thu, 1 Nov 2012 15:41:31 +0000 (15:41 +0000)]
Add suppressions for images for new skia roll
https://bugs.webkit.org/show_bug.cgi?id=100961

Reviewed by NOBODY.

New skia roll improves clipped antialias hairlines, and this means a few images need to be rebaselined.
crbug.com/148638

* platform/chromium/TestExpectations:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@133180 268f45cc-cd09-0410-ab3c-d52691b4dbfc

6 years agoCrash beneath WKRelease after failed load in MiniBrowser
aroben@webkit.org [Thu, 1 Nov 2012 15:40:31 +0000 (15:40 +0000)]
Crash beneath WKRelease after failed load in MiniBrowser
https://bugs.webkit.org/show_bug.cgi?id=100954

Reviewed by Sam Weinig.

There were just some simple retain/release bugs.

* MiniBrowser/mac/WK2BrowserWindowController.m:
(-[WK2BrowserWindowController updateTextFieldFromURL:]):
(-[WK2BrowserWindowController updateProvisionalURLForFrame:]):
Don't try to release null URLs.

(-[WK2BrowserWindowController updateCommittedURLForFrame:]): Don't
release a URL for which we don't have ownership.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@133179 268f45cc-cd09-0410-ab3c-d52691b4dbfc

6 years agoUnreviewed. Build fix.
abarth@webkit.org [Thu, 1 Nov 2012 15:33:29 +0000 (15:33 +0000)]
Unreviewed. Build fix.

* bindings/v8/V8GCController.cpp:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@133178 268f45cc-cd09-0410-ab3c-d52691b4dbfc

6 years ago[EFL][WK2] Add Ewk_Auth_Request API
commit-queue@webkit.org [Thu, 1 Nov 2012 15:32:42 +0000 (15:32 +0000)]
[EFL][WK2] Add Ewk_Auth_Request API
https://bugs.webkit.org/show_bug.cgi?id=100858

Patch by Christophe Dumez <christophe.dumez@intel.com> on 2012-11-01
Reviewed by Kenneth Rohde Christiansen.

Source/WebCore:

Provide an AuthenticationClient for EFL port in WebCore so
that we can handle authentication similarly to other ports
of WebKit.

No new tests, no behavior change for layout tests.

* PlatformEfl.cmake: Add AuthenticationChallengeSoup.cpp to CMake.
* loader/ResourceLoader.cpp:
(WebCore::ResourceLoader::didReceiveAuthenticationChallenge):
* platform/network/ResourceHandle.h:
(ResourceHandle): Have ResourceHandle subclass AuthenticationClient like
most of the other ports.
* platform/network/soup/AuthenticationChallengeSoup.cpp: Fix initialization
of previousFailureCount member. It should be 1 if we are retrying
authentication and 0 otherwise, not the opposite.
(WebCore::AuthenticationChallenge::AuthenticationChallenge):
* platform/network/soup/ResourceHandleSoup.cpp: Provide implementation for EFL
port of AuthenticationClient methods.
(WebCore::WebCoreSynchronousLoader::didReceiveAuthenticationChallenge):
(WebCoreSynchronousLoader):
(WebCore):
(WebCore::ResourceHandle::didReceiveAuthenticationChallenge):
(WebCore::ResourceHandle::receivedRequestToContinueWithoutCredential):
(WebCore::ResourceHandle::receivedCredential):
(WebCore::ResourceHandle::receivedCancellation):
(WebCore::authenticateCallback):
(WebCore::ResourceHandle::defaultSession):

Source/WebKit2:

Add Ewk_Auth_Request API for the client to handle
HTTP authentication.

* PlatformEfl.cmake:
* UIProcess/API/efl/EWebKit2.h:
* UIProcess/API/efl/EwkViewCallbacks.h:
(EwkViewCallbacks):
* UIProcess/API/efl/ewk_auth_request.cpp: Added.
(Ewk_Auth_Request::Ewk_Auth_Request):
(Ewk_Auth_Request::~Ewk_Auth_Request):
(Ewk_Auth_Request::suggestedUsername):
(Ewk_Auth_Request::realm):
(Ewk_Auth_Request::host):
(Ewk_Auth_Request::cancel):
(Ewk_Auth_Request::authenticate):
(Ewk_Auth_Request::isRetrying):
(ewk_auth_request_ref):
(ewk_auth_request_unref):
(ewk_auth_request_suggested_username_get):
(ewk_auth_request_cancel):
(ewk_auth_request_authenticate):
(ewk_auth_request_retrying_get):
(ewk_auth_request_realm_get):
(ewk_auth_request_host_get):
* UIProcess/API/efl/ewk_auth_request.h: Added.
* UIProcess/API/efl/ewk_auth_request_private.h: Added.
(WebKit):
(Ewk_Auth_Request):
(Ewk_Auth_Request::create):
* UIProcess/API/efl/ewk_view.h:
* UIProcess/API/efl/tests/test_ewk2_auth_request.cpp: Added corresponding API tests.
(serverCallback):
(onAuthenticationRequest):
(TEST_F):
(onResourceLoadResponse):
* UIProcess/efl/PageLoadClientEfl.cpp:
(WebKit::PageLoadClientEfl::didReceiveAuthenticationChallengeInFrame):
(WebKit):
(WebKit::PageLoadClientEfl::PageLoadClientEfl):
* UIProcess/efl/PageLoadClientEfl.h:
(PageLoadClientEfl):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@133177 268f45cc-cd09-0410-ab3c-d52691b4dbfc

6 years agoFix the Mac breakage
jocelyn.turcotte@digia.com [Thu, 1 Nov 2012 15:19:04 +0000 (15:19 +0000)]
Fix the Mac breakage
https://bugs.webkit.org/show_bug.cgi?id=100383

Reviewed by Kenneth Rohde Christiansen.

acceleratedCompositingForFixedPositionEnabled needs to be enabled
for fixed layout to work efficiently but it's wrong to force it
to false if fixed layout isn't enabled.

Separate hard requirements for fixed layout from rendering optimizations
and guard the later with USE(COORDINATED_GRAPHICS).
Also add an early return to make sure that ports always having a false
useFixedLayout creation parameter won't get their settings adjusted.

* WebProcess/WebPage/WebPage.cpp:
(WebKit::WebPage::setUseFixedLayout):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@133176 268f45cc-cd09-0410-ab3c-d52691b4dbfc

6 years agoWebIconDatabase: Properly clean up on destruction
jocelyn.turcotte@digia.com [Thu, 1 Nov 2012 15:17:43 +0000 (15:17 +0000)]
WebIconDatabase: Properly clean up on destruction
https://bugs.webkit.org/show_bug.cgi?id=100237

Reviewed by Brady Eidson.

Source/WebCore:

It's now possible that the IconDatabase gets destroyed since WebKit2
isn't using it as a singleton.

Check that the database was properly closed rather than asserting that
it won't get deleted.

* loader/icon/IconDatabase.cpp:
(WebCore::IconDatabase::~IconDatabase):

Source/WebKit2:

* UIProcess/WebIconDatabase.cpp:
(WebKit::WebIconDatabase::invalidate):
Make sure that IconDatabaseBase::iconDatabase won't return a dangling pointer.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@133175 268f45cc-cd09-0410-ab3c-d52691b4dbfc

6 years agoWeb Inspector: make component subitems use parent color in native memory snapshots.
commit-queue@webkit.org [Thu, 1 Nov 2012 15:16:25 +0000 (15:16 +0000)]
Web Inspector: make component subitems use parent color in native memory snapshots.
https://bugs.webkit.org/show_bug.cgi?id=100876

A component subitems use the color of the component itself if the color
is not explicitly specified for the subitem.

Patch by Alexei Filippov <alph@chromium.org> on 2012-11-01
Reviewed by Yury Semikhatsky.

* inspector/front-end/NativeMemorySnapshotView.js:
(WebInspector.NativeSnapshotNode.prototype._createSizeCell):
(WebInspector.MemoryBlockViewProperties._initialize):
(WebInspector.MemoryBlockViewProperties._forMemoryBlock):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@133174 268f45cc-cd09-0410-ab3c-d52691b4dbfc

6 years ago[Chromium] Unreviewed gardening. Remove some no-longer-flaky tests
senorblanco@chromium.org [Thu, 1 Nov 2012 15:06:41 +0000 (15:06 +0000)]
[Chromium] Unreviewed gardening.  Remove some no-longer-flaky tests
from TestExpectations.

* platform/chromium/TestExpectations:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@133173 268f45cc-cd09-0410-ab3c-d52691b4dbfc

6 years ago[subpixel] Change intrinsicSize to LayoutUnit
eae@chromium.org [Thu, 1 Nov 2012 15:00:26 +0000 (15:00 +0000)]
[subpixel] Change intrinsicSize to LayoutUnit
https://bugs.webkit.org/show_bug.cgi?id=99104

Reviewed by Levi Weintraub.

Source/WebCore:

Change RenderReplaced and intrinsicSize to LayoutUnit to avoid
rounding problems when zooming/scaling. Also change imageSize to
LayoutUnit as it can return a scaled size.

Test: fast/sub-pixel/tiled-canvas-elements.html

* html/ImageDocument.cpp:
(WebCore::ImageDocumentParser::finish):
* loader/cache/CachedImage.cpp:
(WebCore::CachedImage::imageSizeForRenderer):
* loader/cache/CachedImage.h:
(CachedImage):
* platform/graphics/FractionalLayoutSize.h:
(FractionalLayoutSize):
(WebCore::FractionalLayoutSize::scale):
(WebCore::FractionalLayoutSize::clampToMinimumSize):
* rendering/RenderBox.h:
(WebCore::RenderBox::intrinsicSize):
* rendering/RenderHTMLCanvas.cpp:
(WebCore::RenderHTMLCanvas::canvasSizeChanged):
* rendering/RenderImage.cpp:
(WebCore::RenderImage::imageChanged):
(WebCore::RenderImage::updateIntrinsicSizeIfNeeded):
(WebCore::RenderImage::paintReplaced):
(WebCore::RenderImage::minimumReplacedHeight):
* rendering/RenderImage.h:
(RenderImage):
* rendering/RenderImageResource.h:
(WebCore::RenderImageResource::imageSize):
* rendering/RenderImageResourceStyleImage.h:
* rendering/RenderReplaced.cpp:
(WebCore::RenderReplaced::RenderReplaced):
(WebCore::RenderReplaced::computeAspectRatioInformationForRenderBox):
(WebCore::RenderReplaced::computeReplacedLogicalWidth):
(WebCore::RenderReplaced::computeReplacedLogicalHeight):
* rendering/RenderReplaced.h:
(RenderReplaced):
(WebCore::RenderReplaced::minimumReplacedHeight):
(WebCore::RenderReplaced::setIntrinsicSize):
* rendering/RenderVideo.cpp:
(WebCore::RenderVideo::updateIntrinsicSize):
(WebCore::RenderVideo::calculateIntrinsicSize):
(WebCore::RenderVideo::videoBox):
(WebCore::RenderVideo::minimumReplacedHeight):
* rendering/RenderVideo.h:
(RenderVideo):
* rendering/style/StyleCachedImage.cpp:
(WebCore::StyleCachedImage::imageSize):
* rendering/style/StyleCachedImage.h:
(StyleCachedImage):
* rendering/style/StyleCachedImageSet.cpp:
(WebCore::StyleCachedImageSet::imageSize):
* rendering/style/StyleCachedImageSet.h:
(StyleCachedImageSet):
* rendering/style/StyleGeneratedImage.cpp:
(WebCore::StyleGeneratedImage::imageSize):
(WebCore::StyleGeneratedImage::computeIntrinsicDimensions):
* rendering/style/StyleGeneratedImage.h:
(StyleGeneratedImage):
* rendering/style/StyleImage.h:
(StyleImage):
* rendering/style/StylePendingImage.h:

LayoutTests:

Add test for tiled replaced elements and update existing expectations as needed.

* fast/sub-pixel/tiled-canvas-elements-expected.txt: Added.
* fast/sub-pixel/tiled-canvas-elements.html: Added.
* platform/chromium-linux/compositing/overflow/overflow-compositing-descendant-expected.png:
* platform/chromium-linux/compositing/overflow/scroll-ancestor-update-expected.png:
* platform/chromium-linux/svg/zoom/page/zoom-svg-through-object-with-absolute-size-2-expected.png:
* platform/chromium-linux/svg/zoom/page/zoom-svg-through-object-with-absolute-size-2-expected.txt:
* platform/chromium-linux/svg/zoom/page/zoom-svg-through-object-with-absolute-size-expected.txt:
* platform/chromium-linux/svg/zoom/page/zoom-svg-through-object-with-auto-size-expected.png:
* platform/chromium-linux/svg/zoom/page/zoom-svg-through-object-with-percentage-size-expected.png:
* platform/chromium-linux/svg/zoom/page/zoom-svg-through-object-with-percentage-size-expected.txt:
* platform/chromium-win/svg/as-image/svg-non-integer-scaled-image-expected.png:
* platform/chromium-win/svg/zoom/page/zoom-svg-through-object-with-auto-size-expected.txt:
* platform/chromium/fast/reflections/reflection-with-zoom-expected.txt:
* platform/chromium/svg/as-image/svg-non-integer-scaled-image-expected.txt:
* platform/chromium/svg/zoom/page/zoom-svg-as-image-expected.txt:
* platform/chromium/svg/zoom/page/zoom-svg-as-relative-image-expected.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@133172 268f45cc-cd09-0410-ab3c-d52691b4dbfc

6 years ago[Chromium] Unreviewed gardening. New reftest from
senorblanco@chromium.org [Thu, 1 Nov 2012 14:56:35 +0000 (14:56 +0000)]
[Chromium] Unreviewed gardening.  New reftest from
http://trac.webkit.org/changeset/133146 doesn't seem to work on
MacOS 10.8.

* platform/chromium/TestExpectations:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@133171 268f45cc-cd09-0410-ab3c-d52691b4dbfc

6 years ago[EFL] Fix event mapping for WebKit2 with coordinated graphics
kenneth@webkit.org [Thu, 1 Nov 2012 14:56:16 +0000 (14:56 +0000)]
[EFL] Fix event mapping for WebKit2 with coordinated graphics
https://bugs.webkit.org/show_bug.cgi?id=100956

Reviewed by Alexis Menard.

Apply scale after translation.

* UIProcess/API/efl/ewk_view.cpp:
(toWebContentTransform):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@133170 268f45cc-cd09-0410-ab3c-d52691b4dbfc

6 years ago[V8] Unify the V8GCController visitors
abarth@webkit.org [Thu, 1 Nov 2012 14:52:07 +0000 (14:52 +0000)]
[V8] Unify the V8GCController visitors
https://bugs.webkit.org/show_bug.cgi?id=100897

Reviewed by Eric Seidel.

After this patch, we use a single visitor for all DOM wrappers,
regardless of type. We also visit all the wrappers in one pass by
calling v8::V8::VisitHandlesWithClassIds directly rather than via
visitAllDOMNodes.

This patch also introduces a wrapper class ID for non-Node DOM objects.
Previously, only DOM nodes had a class ID.

* bindings/v8/IntrusiveDOMWrapperMap.h:
* bindings/v8/ScriptProfiler.cpp:
(WebCore::retainedDOMInfo):
(WebCore::ScriptProfiler::initialize):
* bindings/v8/V8DOMMap.cpp:
(WebCore::visitAllDOMNodes):
* bindings/v8/V8DOMWrapper.cpp:
(WebCore::V8DOMWrapper::setJSWrapperForDOMNode):
* bindings/v8/V8DOMWrapper.h:
(WebCore::V8DOMWrapper::setJSWrapperForDOMObject):
* bindings/v8/V8GCController.cpp:
(WebCore::GCHandleVisitor::notifyFinished):
(GCHandleVisitor):
(WebCore::V8GCController::majorGCPrologue):
* bindings/v8/WrapperTypeInfo.h:
(WebCore):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@133169 268f45cc-cd09-0410-ab3c-d52691b4dbfc

6 years ago[Chromium] Unreviewed gardening.
mkwst@chromium.org [Thu, 1 Nov 2012 14:44:52 +0000 (14:44 +0000)]
[Chromium] Unreviewed gardening.

Skipping http/tests/security/contentSecurityPolicy/object-src-none-blocked.html
as it's consistently leaking state into the next text (object-src-url-allowed.html).
Filed https://bugs.webkit.org/show_bug.cgi?id=100955 to follow up.

* platform/chromium/TestExpectations:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@133168 268f45cc-cd09-0410-ab3c-d52691b4dbfc

6 years agoUnreviewed, rolling out r133143.
senorblanco@chromium.org [Thu, 1 Nov 2012 14:26:09 +0000 (14:26 +0000)]
Unreviewed, rolling out r133143.
http://trac.webkit.org/changeset/133143
https://bugs.webkit.org/show_bug.cgi?id=96894

Causing content_browsertests failures

Source/WebCore:

* CMakeLists.txt:
* GNUmakefile.list.am:
* Target.pri:
* WebCore.gypi:
* WebCore.vcproj/WebCore.vcproj:
* WebCore.xcodeproj/project.pbxproj:
* dom/DeviceOrientationClient.h:
(DeviceOrientationClient):
* dom/DeviceOrientationController.cpp:
(WebCore::DeviceOrientationController::DeviceOrientationController):
(WebCore):
(WebCore::DeviceOrientationController::~DeviceOrientationController):
(WebCore::DeviceOrientationController::timerFired):
(WebCore::DeviceOrientationController::addListener):
(WebCore::DeviceOrientationController::removeListener):
(WebCore::DeviceOrientationController::removeAllListeners):
(WebCore::DeviceOrientationController::suspendEventsForAllListeners):
(WebCore::DeviceOrientationController::resumeEventsForAllListeners):
(WebCore::DeviceOrientationController::didChangeDeviceOrientation):
(WebCore::DeviceOrientationController::supplementName):
* dom/DeviceOrientationController.h:
(WebCore):
(DeviceOrientationController):
(WebCore::DeviceOrientationController::isActive):
(WebCore::DeviceOrientationController::client):
(WebCore::DeviceOrientationController::from):
* dom/Document.cpp:
(WebCore::Document::suspendActiveDOMObjects):
(WebCore::Document::resumeActiveDOMObjects):
* loader/EmptyClients.h:
(WebCore::EmptyDeviceOrientationClient::startUpdating):
(WebCore::EmptyDeviceOrientationClient::stopUpdating):
* page/DOMWindow.cpp:
(WebCore::DOMWindow::addEventListener):
(WebCore::DOMWindow::removeEventListener):
(WebCore::DOMWindow::removeAllEventListeners):
* page/DeviceClient.h: Removed.
* page/DeviceController.cpp: Removed.
* page/DeviceController.h: Removed.

Source/WebKit/qt:

* WebCoreSupport/DumpRenderTreeSupportQt.cpp:
(DumpRenderTreeSupportQt::setMockDeviceOrientation):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@133167 268f45cc-cd09-0410-ab3c-d52691b4dbfc

6 years ago[BlackBerry] Fix editing/text-iterator/findString-expected.html fail
rwlbuis@webkit.org [Thu, 1 Nov 2012 14:25:10 +0000 (14:25 +0000)]
[BlackBerry] Fix editing/text-iterator/findString-expected.html fail
https://bugs.webkit.org/show_bug.cgi?id=100902

Reviewed by Yong Li.

PR 210039

Fix the test my using the Page::findString method instead of our own in-page searcher.
This also syncs some of the changes we did regarding String usage.

* DumpRenderTree/blackberry/DumpRenderTree.cpp:
(BlackBerry::WebKit::DumpRenderTree::dumpFramesAsText):
(BlackBerry::WebKit::dumpHistoryItem):
(BlackBerry::WebKit::dumpBackForwardListForWebView):
(BlackBerry::WebKit::DumpRenderTree::dump):
(BlackBerry::WebKit::DumpRenderTree::didFinishDocumentLoadForFrame):
(BlackBerry::WebKit::DumpRenderTree::didReceiveTitleForFrame):
(BlackBerry::WebKit::DumpRenderTree::runJavaScriptAlert):
* DumpRenderTree/blackberry/DumpRenderTreeBlackBerry.h:
(DumpRenderTree):
* DumpRenderTree/blackberry/TestRunnerBlackBerry.cpp:
(TestRunner::setMockGeolocationError):
(TestRunner::layerTreeAsText):
(TestRunner::findString):
(TestRunner::setStorageDatabaseIdleInterval):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@133166 268f45cc-cd09-0410-ab3c-d52691b4dbfc

6 years agoRemove empty directory after r132043.
rakuco@webkit.org [Thu, 1 Nov 2012 14:24:34 +0000 (14:24 +0000)]
Remove empty directory after r132043.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@133165 268f45cc-cd09-0410-ab3c-d52691b4dbfc

6 years ago[chromium] Avoid attempting to defer commits when they are already deferred
commit-queue@webkit.org [Thu, 1 Nov 2012 14:22:21 +0000 (14:22 +0000)]
[chromium] Avoid attempting to defer commits when they are already deferred
https://bugs.webkit.org/show_bug.cgi?id=100913

Patch by James Robinson <jamesr@chromium.org> on 2012-11-01
Reviewed by Adrienne Walker.

In threaded compositing mode, it's possible to load and unload FrameViews without any
intervening invalidations. In this case, we enter setIsAcceleratedCompositingActive
with commits already deferred.  There's no need to defer again and doing so trips
an assertion in the implementation.

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

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@133164 268f45cc-cd09-0410-ab3c-d52691b4dbfc

6 years ago[EFL][WK2] Add File Chooser dialog support to MiniBrowser
commit-queue@webkit.org [Thu, 1 Nov 2012 14:17:40 +0000 (14:17 +0000)]
[EFL][WK2] Add File Chooser dialog support to MiniBrowser
https://bugs.webkit.org/show_bug.cgi?id=100822

Patch by Christophe Dumez <christophe.dumez@intel.com> on 2012-11-01
Reviewed by Kenneth Rohde Christiansen.

Add very basic file picker dialog to MiniBrowser so that
the user can select a file from the file system when
clicking an input element of type "file".

* MiniBrowser/efl/main.c:
(close_file_picker):
(on_filepicker_deletion):
(on_fileselector_done):
(on_file_chooser_request):
(window_create):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@133163 268f45cc-cd09-0410-ab3c-d52691b4dbfc

6 years ago[EFL][AC]Free GL resources allocated by GraphicsContext3DEfl.
commit-queue@webkit.org [Thu, 1 Nov 2012 14:17:01 +0000 (14:17 +0000)]
[EFL][AC]Free GL resources allocated by GraphicsContext3DEfl.
https://bugs.webkit.org/show_bug.cgi?id=100923.

Patch by Kondapally Kalyan <kalyan.kondapally@intel.com> on 2012-11-01
Reviewed by Kenneth Rohde Christiansen.

GraphicsContext3DEfl creates FBO's, textures and render buffer's, but doesn't free them.
This patch makes sure that GraphicsContext3dEfl frees all the GL resources allocated by it.

* platform/graphics/efl/GraphicsContext3DEfl.cpp:
(WebCore::GraphicsContext3D::~GraphicsContext3D):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@133162 268f45cc-cd09-0410-ab3c-d52691b4dbfc

6 years ago[EFL] Correct our use of the coordinated graphics
kenneth@webkit.org [Thu, 1 Nov 2012 14:14:06 +0000 (14:14 +0000)]
[EFL] Correct our use of the coordinated graphics
https://bugs.webkit.org/show_bug.cgi?id=100947

Reviewed by Alexis Menard.

Replace the display(...) call with a regular update() on the view.
In the coordinated graphics case this updates our view using
paintToCurrentGLContext with our viewmodel matrix and clips it to
the viewport. This avoid the need to manually joining update regions.

Unneeded update calls have also been removed and the resizing
logic has been improved.

* UIProcess/API/efl/EwkViewImpl.cpp:
(EwkViewImpl::displayTimerFired):
(EwkViewImpl::update):
* UIProcess/API/efl/EwkViewImpl.h:
(WebCore):
(EwkViewImpl):
(EwkViewImpl::clearEvasGLSurface):
* UIProcess/API/efl/ewk_view.cpp:
(_ewk_view_smart_calculate):
* UIProcess/efl/PageClientImpl.cpp:
(WebKit::PageClientImpl::setViewNeedsDisplay):
* UIProcess/efl/PageViewportControllerClientEfl.cpp:
(WebKit::PageViewportControllerClientEfl::setVisibleContentsRect):
(WebKit::PageViewportControllerClientEfl::didChangeContentsSize):
(WebKit::PageViewportControllerClientEfl::didChangeVisibleContents):
* UIProcess/efl/PageViewportControllerClientEfl.h:
(PageViewportControllerClientEfl):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@133161 268f45cc-cd09-0410-ab3c-d52691b4dbfc

6 years agoFix StylePropertySet/ElementAttributeData custom allocation in debug builds.
rakuco@webkit.org [Thu, 1 Nov 2012 14:10:37 +0000 (14:10 +0000)]
Fix StylePropertySet/ElementAttributeData custom allocation in debug builds.
<http://webkit.org/b/100753>

Unreviewed debug bot crash fix after r133138.

There's additional padding after StylePropertySet and ElementAttributeData
in 64-bit debug builds since there are additional members in RefCountedBase.
Use 'sizeof(ImmutableFoo) - sizeof(void*)' as the base size of ImmutableFoo.

Patch by Andreas Kling <kling@webkit.org> on 2012-11-01

* css/StylePropertySet.cpp:
(WebCore::sizeForImmutableStylePropertySetWithPropertyCount):
* dom/ElementAttributeData.cpp:
(WebCore::sizeForImmutableElementAttributeDataWithAttributeCount):
(WebCore::ElementAttributeData::createImmutable):
(WebCore::ElementAttributeData::reportMemoryUsage):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@133160 268f45cc-cd09-0410-ab3c-d52691b4dbfc

6 years agoUnreviewed. Rolled DEPS.
commit-queue@webkit.org [Thu, 1 Nov 2012 14:05:34 +0000 (14:05 +0000)]
Unreviewed.  Rolled DEPS.

Patch by Sheriff Bot <webkit.review.bot@gmail.com> on 2012-11-01

* DEPS:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@133159 268f45cc-cd09-0410-ab3c-d52691b4dbfc

6 years agoFEImage::m_document is never cleared. Why not?
schenney@chromium.org [Thu, 1 Nov 2012 13:54:49 +0000 (13:54 +0000)]
FEImage::m_document is never cleared. Why not?
https://bugs.webkit.org/show_bug.cgi?id=99243

Reviewed by Dirk Schulze.

Adding a comment to explain why the failure to clear m_document is not a problem.

No new tests because no code change at all.

* svg/graphics/filters/SVGFEImage.h:
(FEImage): Add a comment on the lifetime of m_document.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@133158 268f45cc-cd09-0410-ab3c-d52691b4dbfc

6 years agoWeb Inspector: Timeline: make "addRecord" unambiguous
commit-queue@webkit.org [Thu, 1 Nov 2012 13:50:30 +0000 (13:50 +0000)]
Web Inspector: Timeline: make "addRecord" unambiguous
https://bugs.webkit.org/show_bug.cgi?id=100761

Patch by Eugene Klyuchnikov <eustas.bug@gmail.com> on 2012-11-01
Reviewed by Pavel Feldman.

TimelinePresentationModel.addRecord accepts two parameters:
record and parentRecord. parentRecord is always root record.
Make this explicit by removing parentRecord parameter.

* inspector/front-end/TimelinePanel.js: Removed parameter.
* inspector/front-end/TimelinePresentationModel.js:
(WebInspector.TimelinePresentationModel.prototype.addRecord):
Used root recoed instead of parent record.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@133157 268f45cc-cd09-0410-ab3c-d52691b4dbfc

6 years ago[BlackBerry] Browser prematurely sends wrong credentials
charles.wei@torchmobile.com.cn [Thu, 1 Nov 2012 13:47:19 +0000 (13:47 +0000)]
[BlackBerry] Browser prematurely sends wrong credentials
https://bugs.webkit.org/show_bug.cgi?id=100585

Reviewed by George Staikos.

Don't send credentials to the server before been challenged.

* platform/network/blackberry/NetworkManager.cpp:
(WebCore::NetworkManager::startJob):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@133156 268f45cc-cd09-0410-ab3c-d52691b4dbfc

6 years agoSVG as an image may recreate the renderer on zoom
schenney@chromium.org [Thu, 1 Nov 2012 13:39:54 +0000 (13:39 +0000)]
SVG as an image may recreate the renderer on zoom
https://bugs.webkit.org/show_bug.cgi?id=99508

Reviewed by Abhishek Arya.

Source/WebCore:

The SVGImage code, when SVG is used in <img> tags, caches the renderer
at the start of the painting method and re-uses the pointer at the end
of the method. However, when the page is zoomed the renderer may be
detached mid-method, thus leaving a stray pointer. The fix is to
re-fetch the pointer after the zooms.

Test: svg/as-image/img-zoom-svg-stylesheet.html

* svg/graphics/SVGImage.cpp:
(WebCore::SVGImage::drawSVGToImageBuffer): Re-fetch the renderer after
the zoom operations.

LayoutTests:

Test and ref-test-result for zooming of SVG content when loaded as-image, and the svg is actually a css file.

* svg/as-image/img-zoom-svg-stylesheet-expected.html: Added.
* svg/as-image/img-zoom-svg-stylesheet.html: Added.
* svg/as-image/resources/svg-stylesheet-for-zoom.svg: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@133155 268f45cc-cd09-0410-ab3c-d52691b4dbfc

6 years agoUnreviewed. Change my email address.
yael@webkit.org [Thu, 1 Nov 2012 13:27:37 +0000 (13:27 +0000)]
Unreviewed. Change my email address.

* Scripts/webkitpy/common/config/committers.py:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@133154 268f45cc-cd09-0410-ab3c-d52691b4dbfc

6 years ago[Qt] Double clicks/taps aren't passed down to the page
jocelyn.turcotte@digia.com [Thu, 1 Nov 2012 13:18:13 +0000 (13:18 +0000)]
[Qt] Double clicks/taps aren't passed down to the page
https://bugs.webkit.org/show_bug.cgi?id=100949

Reviewed by Kenneth Rohde Christiansen.

Source/WebKit2:

Since we are passing double click events as press events and that
they are preceeded by a real press event, double clicks are passed
down to the page as triple clicks.

Ignore double click events and let WebCore::EventHandler handle it
through press and release events.

* Shared/qt/WebEventFactoryQt.cpp:
(WebKit::webEventTypeForEvent):
* UIProcess/API/qt/qquickwebview.cpp:
(QQuickWebViewPrivate::handleMouseEvent):
* UIProcess/qt/QtWebPageEventHandler.cpp:
(WebKit::QtWebPageEventHandler::handleInputEvent):

Tools:

Touch mocking shouldn't transform double click events as well,
this adds an extra touch update between clicks which prevents
double-tap detection from triggering on web pages.

Tested on the mobile version of Google maps.

* MiniBrowser/qt/MiniBrowserApplication.cpp:
(MiniBrowserApplication::notify):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@133153 268f45cc-cd09-0410-ab3c-d52691b4dbfc

6 years agoWeb Inspector: introduce Inspector.detached message containing termination cause.
pfeldman@chromium.org [Thu, 1 Nov 2012 13:16:51 +0000 (13:16 +0000)]
Web Inspector: introduce Inspector.detached message containing termination cause.
https://bugs.webkit.org/show_bug.cgi?id=100948

Reviewed by Yury Semikhatsky.

Now protocol clients have more information to process connection termination.

* English.lproj/localizedStrings.js:
* inspector/Inspector.json:
* inspector/front-end/InspectorFrontendHostStub.js:
(WebInspector.RemoteDebuggingTerminatedScreen):
* inspector/front-end/inspector.js:
(WebInspector.loaded.WebInspector.socket.onopen):
(WebInspector.loaded):
(WebInspector.detached):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@133152 268f45cc-cd09-0410-ab3c-d52691b4dbfc

6 years agoWeb Inspector: Restoring breakpoints for formatted UISourceCode should be triggered...
vsevik@chromium.org [Thu, 1 Nov 2012 12:27:33 +0000 (12:27 +0000)]
Web Inspector: Restoring breakpoints for formatted UISourceCode should be triggered by BreakpointManager, not ScriptsPanel.
https://bugs.webkit.org/show_bug.cgi?id=100593

Reviewed by Pavel Feldman.

Source/WebCore:

BreakpointManager now listen for FormattedChanged event and restores formatted breakpoints on it.
Removed now redundant callback from UISourceCode.setFormatted().

* inspector/front-end/BreakpointManager.js:
(WebInspector.BreakpointManager.prototype._restoreBreakpoints):
(WebInspector.BreakpointManager.prototype._uiSourceCodeAdded):
(WebInspector.BreakpointManager.prototype._uiSourceCodeFormatted):
(WebInspector.BreakpointManager.prototype._uiSourceCodeRemoved):
* inspector/front-end/ScriptsPanel.js:
(WebInspector.ScriptsPanel.prototype._addUISourceCode):
(WebInspector.ScriptsPanel.prototype._revealExecutionLine):
(WebInspector.ScriptsPanel.prototype._toggleFormatSource):
* inspector/front-end/UISourceCode.js:
(WebInspector.UISourceCode.prototype._fireContentAvailable):
(WebInspector.UISourceCode.prototype.setFormatted.formattedChanged):
(WebInspector.UISourceCode.prototype.setFormatted.didGetContent):
(WebInspector.UISourceCode.prototype.setFormatted):

LayoutTests:

* inspector/debugger/script-formatter-breakpoints.html:
* inspector/debugger/script-formatter-console.html:
* inspector/debugger/script-formatter-search-expected.txt:
* inspector/debugger/script-formatter-search.html:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@133151 268f45cc-cd09-0410-ab3c-d52691b4dbfc

6 years agoWeb Inspector: Output code evaluated in the console the same as console.log
yurys@chromium.org [Thu, 1 Nov 2012 12:17:00 +0000 (12:17 +0000)]
Web Inspector: Output code evaluated in the console the same as console.log
https://bugs.webkit.org/show_bug.cgi?id=100695

Reviewed by Pavel Feldman.

Source/WebCore:

Added an option to return object preview from evaluation commands in the protocol.
The option is used to generate preview for console eval results.

* inspector/InjectedScript.cpp:
(WebCore::InjectedScript::evaluate):
(WebCore::InjectedScript::callFunctionOn):
(WebCore::InjectedScript::evaluateOnCallFrame):
* inspector/InjectedScript.h:
(InjectedScript):
* inspector/InjectedScriptSource.js:
(.):
* inspector/Inspector.json:
* inspector/InspectorDebuggerAgent.cpp:
(WebCore::InspectorDebuggerAgent::evaluateOnCallFrame):
* inspector/InspectorDebuggerAgent.h:
(InspectorDebuggerAgent):
* inspector/InspectorRuntimeAgent.cpp:
(WebCore::InspectorRuntimeAgent::evaluate):
(WebCore::InspectorRuntimeAgent::callFunctionOn):
* inspector/InspectorRuntimeAgent.h:
(InspectorRuntimeAgent):
* inspector/front-end/ConsoleMessage.js:
(WebInspector.ConsoleMessageImpl.prototype.useArrayPreviewInFormatter):
(WebInspector.ConsoleMessageImpl.prototype._formatParameterAsArray):
* inspector/front-end/ConsoleView.js:
(WebInspector.ConsoleCommandResult.prototype.useArrayPreviewInFormatter):
* inspector/front-end/DebuggerModel.js:
(WebInspector.DebuggerModel.prototype.evaluateOnSelectedCallFrame):
(WebInspector.DebuggerModel.CallFrame.prototype.evaluate):
* inspector/front-end/ExtensionServer.js:
(WebInspector.ExtensionServer.prototype.evaluate):
* inspector/front-end/JavaScriptSourceFrame.js:
(WebInspector.JavaScriptSourceFrame.prototype._resolveObjectForPopover):
* inspector/front-end/RemoteObject.js:
(WebInspector.RemoteObject.prototype.):
(WebInspector.RemoteObject.prototype.callFunction):
(WebInspector.RemoteObject.prototype.callFunctionJSON):
* inspector/front-end/RuntimeModel.js:
(WebInspector.RuntimeModel.prototype.evaluate):
(WebInspector.RuntimeModel.prototype.evaluated):
* inspector/front-end/WatchExpressionsSidebarPane.js:
(WebInspector.WatchExpressionsSection.prototype.update):

LayoutTests:

Rebaselined some tests after object preview is started beeing used for
console eval results.

* inspector/console/command-line-api-expected.txt:
* inspector/console/command-line-api-inspect-expected.txt:
* inspector/console/console-eval-global-expected.txt:
* inspector/console/console-eval-undefined-override-expected.txt:
* platform/chromium/inspector/console/console-eval-undefined-override-expected.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@133150 268f45cc-cd09-0410-ab3c-d52691b4dbfc

6 years agoWeb Inspector: Remove obsolete code from JavaScriptSourceFrame
vsevik@chromium.org [Thu, 1 Nov 2012 11:35:52 +0000 (11:35 +0000)]
Web Inspector: Remove obsolete code from JavaScriptSourceFrame
https://bugs.webkit.org/show_bug.cgi?id=100594

Reviewed by Yury Semikhatsky.

Removed obsolete WorkingCopyCommitted listener.
Removed redundant code that removes breakpoints from _innerSetContent.

* inspector/front-end/JavaScriptSourceFrame.js:
(WebInspector.JavaScriptSourceFrame):
(WebInspector.JavaScriptSourceFrame.prototype._innerSetContent):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@133149 268f45cc-cd09-0410-ab3c-d52691b4dbfc

6 years agoUpdate average StylePropertySet size estimation.
kling@webkit.org [Thu, 1 Nov 2012 11:32:14 +0000 (11:32 +0000)]
Update average StylePropertySet size estimation.
<http://webkit.org/b/100940>

Reviewed by Antti Koivisto.

Use sizeForImmutableStylePropertySetWithPropertyCount(2) as the average StylePropertySet
size to keep it in sync with the changed object memory layout.

* css/StylePropertySet.cpp:
(WebCore::sizeForImmutableStylePropertySetWithPropertyCount):
(WebCore::StylePropertySet::createImmutable):
(WebCore::StylePropertySet::averageSizeInBytes):
(WebCore::StylePropertySet::reportMemoryUsage):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@133148 268f45cc-cd09-0410-ab3c-d52691b4dbfc

6 years ago[EFL][Qt][AC] Remove an unnecessary connection to X-Server.
commit-queue@webkit.org [Thu, 1 Nov 2012 11:22:01 +0000 (11:22 +0000)]
[EFL][Qt][AC] Remove an unnecessary connection to X-Server.
https://bugs.webkit.org/show_bug.cgi?id=100628.

Patch by Kondapally Kalyan <kalyan.kondapally@intel.com> on 2012-11-01
Reviewed by Kenneth Rohde Christiansen.

With Changset: https://bugs.webkit.org/show_bug.cgi?id=100523
GraphicsSurfacePrivate should always use the Display returned by offscreenwindow.
However, in GraphicsSurfacePrivate constructor we call XOpenDisplay before asking
display from offscreenwindow.
This patch removes unnecessary call to XOpenDisplay made in GraphicsSurfacePrivate.

* platform/graphics/surfaces/qt/GraphicsSurfaceGLX.cpp:
(WebCore::GraphicsSurfacePrivate::GraphicsSurfacePrivate):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@133147 268f45cc-cd09-0410-ab3c-d52691b4dbfc

6 years ago[CSSRegions]Former auto-height regions should not ignore their defined height
mihnea@adobe.com [Thu, 1 Nov 2012 11:17:49 +0000 (11:17 +0000)]
[CSSRegions]Former auto-height regions should not ignore their defined height
https://bugs.webkit.org/show_bug.cgi?id=100749

Reviewed by Julien Chaffraix.

Source/WebCore:

When a region with height auto has its height defined, we should also clear the override logical content height.
Otherwise, the region will use the wrong height when laying out content from the associated render flow thread.

Test: fast/regions/autoheight-definedheight-changenotdetected.html

* rendering/RenderRegion.cpp:
(WebCore::RenderRegion::updateRegionHasAutoLogicalHeightFlag):

LayoutTests:

Add a test in which we flow content into a region chain with 2 regions, the first one having height auto.
After a first layout, the region with height auto has an override logical content height, which should be
cleared when the first region has its height changed from auto to defined.

* fast/regions/autoheight-definedheight-changenotdetected-expected.html: Added.
* fast/regions/autoheight-definedheight-changenotdetected.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@133146 268f45cc-cd09-0410-ab3c-d52691b4dbfc

6 years agoMemory instrumentation: do not report memory occupied by RenderObjects referenced...
yurys@chromium.org [Thu, 1 Nov 2012 11:06:45 +0000 (11:06 +0000)]
Memory instrumentation: do not report memory occupied by RenderObjects referenced from CSSImageGeneratorValue
https://bugs.webkit.org/show_bug.cgi?id=100934

Reviewed by Alexander Pavlov.

* css/CSSImageGeneratorValue.cpp:
(WTF): Skip rederences to RenderObjects from CSSImageGeneratorValue when collecting
memory usage data.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@133145 268f45cc-cd09-0410-ab3c-d52691b4dbfc

6 years ago[AC][EFL][WK2] Rename PageViewportControllerClientEfl::m_pageViewportController to...
commit-queue@webkit.org [Thu, 1 Nov 2012 11:04:42 +0000 (11:04 +0000)]
[AC][EFL][WK2] Rename PageViewportControllerClientEfl::m_pageViewportController to m_controller
https://bugs.webkit.org/show_bug.cgi?id=100933

Patch by Mikhail Pozdnyakov <mikhail.pozdnyakov@intel.com> on 2012-11-01
Reviewed by Kenneth Rohde Christiansen.

PageViewportControllerClientEfl::m_pageViewportController has to be renamed to m_controller
to keep consistency with other ports. Assertinons were also added.

* UIProcess/efl/PageViewportControllerClientEfl.cpp:
(WebKit::PageViewportControllerClientEfl::PageViewportControllerClientEfl):
(WebKit::PageViewportControllerClientEfl::updateViewportSize):
(WebKit::PageViewportControllerClientEfl::setVisibleContentsRect):
(WebKit::PageViewportControllerClientEfl::didResumeContent):
(WebKit::PageViewportControllerClientEfl::setController):
* UIProcess/efl/PageViewportControllerClientEfl.h:
(PageViewportControllerClientEfl):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@133144 268f45cc-cd09-0410-ab3c-d52691b4dbfc

6 years agoAdd DeviceController base-class to remove duplication of DeviceXXXControler
kihong.kwon@samsung.com [Thu, 1 Nov 2012 10:38:04 +0000 (10:38 +0000)]
Add DeviceController base-class to remove duplication of DeviceXXXControler
https://bugs.webkit.org/show_bug.cgi?id=96894

Reviewed by Hajime Morita.

Source/WebCore:

Add DeviceController which is extracted from DeviceOrientationController to remove duplication.
And soon-to-be-added DeviceMotionController and ProximityController.

Covered by existing tests.

* CMakeLists.txt:
* GNUmakefile.list.am:
* Target.pri:
* WebCore.gypi:
* WebCore.vcproj/WebCore.vcproj:
* WebCore.xcodeproj/project.pbxproj:
* dom/DeviceOrientationClient.h:
* dom/DeviceOrientationController.cpp:
Remove member functions to move to DeviceController.
- addListener(), removeListener(), removeAllListeners(), isActive()
(WebCore::DeviceOrientationController::DeviceOrientationController):
(WebCore::DeviceOrientationController::didChangeDeviceOrientation):
(WebCore::DeviceOrientationController::client):
(WebCore::DeviceOrientationController::hasLastData):
(WebCore::DeviceOrientationController::getLastEvent):
(WebCore::DeviceOrientationController::from):
(WebCore):
* dom/DeviceOrientationController.h:
(WebCore):
(WebCore::DeviceOrientationController::~DeviceOrientationController):
(DeviceOrientationController):
* dom/Document.cpp:
Remove suspendEventsForAllListeners() and resumeEventsForAllListeners() function calls.
These calls can be made by checking activeDOMObjectsAreSuspended() and activeDOMObjectsAreStopped() before dispatchEvent.
(WebCore::Document::suspendActiveDOMObjects):
(WebCore::Document::resumeActiveDOMObjects):
* loader/EmptyClients.h:
(EmptyDeviceClient):
(WebCore::EmptyDeviceClient::startUpdating):
(WebCore::EmptyDeviceClient::stopUpdating):
(WebCore):
* page/DOMWindow.cpp:
(WebCore::DOMWindow::addEventListener):
(WebCore::DOMWindow::removeEventListener):
(WebCore::DOMWindow::removeAllEventListeners):
* page/DeviceClient.h: Added.
(WebCore):
(DeviceClient):
(WebCore::DeviceClient::~DeviceClient):
* page/DeviceController.cpp: Added.
DeviceController has extracted functions from DeviceOrientationController and DeviceMotionController.
- addDeviceEventListener(), removeDeviceEventlistener(), removeAllDeviceEventListeners(), dispatchDeviceEvent()
All kind of device event controller which has DeviceClient can be inherited from DeviceController.
(WebCore):
(WebCore::DeviceController::DeviceController):
(WebCore::DeviceController::addDeviceEventListener):
(WebCore::DeviceController::removeDeviceEventListener):
(WebCore::DeviceController::removeAllDeviceEventListeners):
(WebCore::DeviceController::dispatchDeviceEvent):
(WebCore::DeviceController::fireDeviceEvent):
* page/DeviceController.h: Added.
(WebCore):
(DeviceController):
(WebCore::DeviceController::~DeviceController):
(WebCore::DeviceController::isActive):
(WebCore::DeviceController::client):
(WebCore::DeviceController::hasLastData):
(WebCore::DeviceController::getLastEvent):

Source/WebKit/qt:

Change client() to deviceOrientationClient() to get DeviceOrientationClient*.

* WebCoreSupport/DumpRenderTreeSupportQt.cpp:
(DumpRenderTreeSupportQt::setMockDeviceOrientation):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@133143 268f45cc-cd09-0410-ab3c-d52691b4dbfc

6 years ago[chromium] Add new baseline for exif-orientation-css.html on linux.
podivilov@chromium.org [Thu, 1 Nov 2012 10:17:32 +0000 (10:17 +0000)]
[chromium] Add new baseline for exif-orientation-css.html on linux.

* platform/chromium-linux/platform/chromium/virtual/deferred/fast/images/exif-orientation-css-expected.png: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@133142 268f45cc-cd09-0410-ab3c-d52691b4dbfc

6 years agoMemory instrumentation: report memory occupied by ResourceRequest instead of its...
yurys@chromium.org [Thu, 1 Nov 2012 09:28:19 +0000 (09:28 +0000)]
Memory instrumentation: report memory occupied by ResourceRequest instead of its base ResourceRequestBase
https://bugs.webkit.org/show_bug.cgi?id=100497

Reviewed by Alexander Pavlov.

Source/WebCore:

Added memory reporting method to chromium implementation of ResourceRequest.

* platform/network/ResourceRequestBase.cpp:
(WebCore::ResourceRequestBase::reportMemoryUsageBase): Renamed reportMemoryUsage
on ResourceRequestBase to reportMemoryUsageBase and made it protected. I'd
rather make ResourceRequestBase::reportMemoryUsage virtual and override it
in the descendant but ResourceRequestBase doesn't have any virtual methods
and shouldn't be used directly(ResourceRequest should be used instead).
* platform/network/ResourceRequestBase.h:
(ResourceRequestBase):
* platform/network/chromium/ResourceRequest.cpp:
(WebCore::ResourceRequest::reportMemoryUsage):
(WebCore):
* platform/network/chromium/ResourceRequest.h:
(ResourceRequest):

Source/WebKit/chromium:

Updated MemoryInstrumentationTest.ImageObserver so that it doesn't fail
when instrumentation of objects referenced by CachedImage changes.

* tests/MemoryInstrumentationTest.cpp:

Source/WTF:

* wtf/MemoryInstrumentation.h:
(WTF::MemoryClassInfo::addPrivateBuffer): do not report memory usage for empty buffer.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@133141 268f45cc-cd09-0410-ab3c-d52691b4dbfc

6 years ago[EFL] Gardening after r133056
commit-queue@webkit.org [Thu, 1 Nov 2012 09:08:38 +0000 (09:08 +0000)]
[EFL] Gardening after r133056
https://bugs.webkit.org/show_bug.cgi?id=100929

Unreviewed gardening.

Newly added compositing/tiling tests are failing on EFL port due to missing AC support.

Patch by Mikhail Pozdnyakov <mikhail.pozdnyakov@intel.com> on 2012-11-01

* platform/efl/TestExpectations:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@133140 268f45cc-cd09-0410-ab3c-d52691b4dbfc

6 years ago[chromium] Unreviewed, two fast/images/exif-orientation-* tests now pass.
podivilov@chromium.org [Thu, 1 Nov 2012 09:00:37 +0000 (09:00 +0000)]
[chromium] Unreviewed, two fast/images/exif-orientation-* tests now pass.

* platform/chromium-win/platform/chromium/virtual/deferred/fast/images/exif-orientation-css-expected.png: Added.
* platform/chromium/TestExpectations:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@133139 268f45cc-cd09-0410-ab3c-d52691b4dbfc

6 years agoPack immutable StylePropertySets harder on 64-bit.
kling@webkit.org [Thu, 1 Nov 2012 08:55:58 +0000 (08:55 +0000)]
Pack immutable StylePropertySets harder on 64-bit.
<http://webkit.org/b/100753>
<rdar://problem/12599155>

Reviewed by Antti Koivisto.

Move away from using CSSProperty as internal storage for immutable StylePropertySets.
Instead use two arrays, one for property metadata (ID, shorthand ID, !important, ...)
and one for the CSSValue pointers. This saves 4 bytes per property on 64-bit.

Old object layout:

    Ref count                   (4 bytes)
    Metadata                    (4 bytes)
    CSSProperty [N]             (16 bytes each)

New object layout:

    Ref count                   (4 bytes)
    Metadata                    (4 bytes)
    CSSValue* [N]               (8 bytes each)
    StylePropertyMetadata [N]   (4 bytes each)

901kB progress on Membuster3 (22% overall reduction in StylePropertySet memory.)

The CSSProperty class sticks around for now, it's still used in mutable StylePropertySets
and by the StylePropertySet constructors.

* css/CSSProperty.cpp:
* css/CSSProperty.h:
(CSSProperty):
(WebCore::CSSProperty::CSSProperty):
(WebCore::CSSProperty::id):
(WebCore::CSSProperty::shorthandID):
(WebCore::CSSProperty::isImportant):
(WebCore::CSSProperty::metadata):

    Break the bitfield from CSSProperty out into a StylePropertyMetadata class (actually a union.)

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

    Updated size calculation for immutable StylePropertySets, 1/4 smaller!

(WebCore::ImmutableStylePropertySet::ImmutableStylePropertySet):
(WebCore::ImmutableStylePropertySet::~ImmutableStylePropertySet):
(WebCore::MutableStylePropertySet::MutableStylePropertySet):
(WebCore::StylePropertySet::mergeAndOverrideOnConflict):
(WebCore::StylePropertySet::reportMemoryUsage):
(WebCore::StylePropertySet::PropertyReference::cssName):
(WebCore::StylePropertySet::PropertyReference::cssText):
* css/StylePropertySet.h:
(WebCore::StylePropertySet::PropertyReference::id):
(WebCore::StylePropertySet::PropertyReference::shorthandID):
(WebCore::StylePropertySet::PropertyReference::isImportant):
(WebCore::StylePropertySet::PropertyReference::isInherited):
(WebCore::StylePropertySet::PropertyReference::isImplicit):
(PropertyReference):
(WebCore::StylePropertySet::PropertyReference::value):
(WebCore::StylePropertySet::PropertyReference::toCSSProperty):
(WebCore::StylePropertySet::PropertyReference::propertyMetadata):
(WebCore::StylePropertySet::PropertyReference::propertyValue):
(StylePropertySet):
(ImmutableStylePropertySet):
(WebCore::StylePropertySet::immutableValueArray):
(WebCore::StylePropertySet::immutableMetadataArray):

    Refactored internal storage for StylePropertySet.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@133138 268f45cc-cd09-0410-ab3c-d52691b4dbfc

6 years ago2012-11-01 Pavel Podivilov <podivilov@google.com>
podivilov@chromium.org [Thu, 1 Nov 2012 08:53:47 +0000 (08:53 +0000)]
2012-11-01  Pavel Podivilov  <podivilov@google.com>

        [chromium] Unreviewed, remove several tests that started to pass from TestExpectations.

        * platform/chromium/TestExpectations:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@133137 268f45cc-cd09-0410-ab3c-d52691b4dbfc

6 years agoUnreviewed, rolling out r133126.
commit-queue@webkit.org [Thu, 1 Nov 2012 08:04:37 +0000 (08:04 +0000)]
Unreviewed, rolling out r133126.
http://trac.webkit.org/changeset/133126
https://bugs.webkit.org/show_bug.cgi?id=100925

This made EFL API test break on EFL buildbots (Requested by
gyuyoung on #webkit).

Patch by Sheriff Bot <webkit.review.bot@gmail.com> on 2012-11-01

* PlatformEfl.cmake:
* UIProcess/API/efl/EWebKit2.h:
* UIProcess/API/efl/ewk_context.cpp:
(Ewk_Context::Ewk_Context):
* UIProcess/API/efl/ewk_context.h:
* UIProcess/API/efl/ewk_context_private.h:
(Ewk_Context):
* UIProcess/API/efl/ewk_security_origin.cpp: Removed.
* UIProcess/API/efl/ewk_security_origin.h: Removed.
* UIProcess/API/efl/ewk_security_origin_private.h: Removed.
* UIProcess/API/efl/ewk_storage_manager.cpp: Removed.
* UIProcess/API/efl/ewk_storage_manager.h: Removed.
* UIProcess/API/efl/ewk_storage_manager_private.h: Removed.
* UIProcess/API/efl/tests/test_ewk2_context.cpp:
* UIProcess/API/efl/tests/test_ewk2_storage_manager.cpp: Removed.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@133136 268f45cc-cd09-0410-ab3c-d52691b4dbfc

6 years agoDFG optimized string access code should be enabled
fpizlo@apple.com [Thu, 1 Nov 2012 07:41:43 +0000 (07:41 +0000)]
DFG optimized string access code should be enabled
https://bugs.webkit.org/show_bug.cgi?id=100825

Reviewed by Oliver Hunt.

- Removes prediction checks from the parser.

- Fixes the handling of array mode refinement for strings. I.e. we don't do
  any refinement - we already know it's going to be a string. We could
  revisit this in the future, but for now the DFG lacks the ability to
  handle any array modes other than Array::String for string intrinsics, so
  this is as good as it gets.

- Removes uses of isBlahSpeculation for checking if a mode is already
  checked. isBlahSpeculation implicitly checks if the SpeculatedType is not
  BOTTOM ("empty"), which breaks for checking if a mode is already checked
  since a mode may already be "checked" in the sense that we've proven that
  the code is unreachable.

~1% speed-up on V8v7, mostly from a speed-up on crypto, which uses string
intrinsics in one of the hot functions.

* bytecode/SpeculatedType.h:
(JSC::speculationChecked):
(JSC):
* dfg/DFGArrayMode.cpp:
(JSC::DFG::ArrayMode::alreadyChecked):
* dfg/DFGByteCodeParser.cpp:
(JSC::DFG::ByteCodeParser::handleIntrinsic):
* dfg/DFGFixupPhase.cpp:
(JSC::DFG::FixupPhase::fixupNode):
* dfg/DFGSpeculativeJIT.cpp:
(JSC::DFG::SpeculativeJIT::compileGetCharCodeAt):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@133135 268f45cc-cd09-0410-ab3c-d52691b4dbfc

6 years agoRemove unused Locale::parseDateTime
tkent@chromium.org [Thu, 1 Nov 2012 07:31:58 +0000 (07:31 +0000)]
Remove unused Locale::parseDateTime
https://bugs.webkit.org/show_bug.cgi?id=100910

Reviewed by Kentaro Hara.

Source/WebCore:

For date/time input types, InputType::convertFromVisibleValue is never
called. convertFromVisibleValue is called when an inner editable node is
updated. However input elements don't have such editable nodes if
ENABLE_INPUT_MULTIPLE_FIELDS_UI is enabled, and a user can't edit the
inner editable node otherwise because we open date/time pickers when a
user try to change the field value.

We had used convertFromVisibleValue for input[type=date] with an old UI.

No new tests because of no behavior changes.

* html/BaseDateAndTimeInputType.cpp:
(WebCore::BaseDateAndTimeInputType::convertFromVisibleValue):
Remove Locale::parseDateTiem and put ASSERT_NOT_REACHED.

* platform/text/PlatformLocale.h: Remove parseDateTime.

* platform/text/LocaleICU.h: Remove parseDateTime and a bogus comment.
* platform/text/LocaleICU.cpp: Remove parseDateTime.

* platform/text/LocaleNone.cpp: Remove parseDateTime.

* platform/text/mac/LocaleMac.h: Remove parseDateTime.
* platform/text/mac/LocaleMac.mm: Ditto.
* platform/text/win/LocaleWin.h:
- Remove parseDateTime, its helpers, and m_baseYear.
- Remove appendNumber, appendTwoDigitsNumber, appendFourDigitsNumber,
  and formatDate. They don't exist.
* platform/text/win/LocaleWin.cpp:
Remove the above functions.
(WebCore::LocaleWin::LocaleWin): Remove m_baseYear iniitalization.

Source/WebKit/chromium:

* tests/LocaleMacTest.cpp: Remove tests for parseDateTime.
* tests/LocaleWinTest.cpp: Ditto.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@133134 268f45cc-cd09-0410-ab3c-d52691b4dbfc

6 years ago[WK2] Add APIs to get/set encoding detector
commit-queue@webkit.org [Thu, 1 Nov 2012 07:26:00 +0000 (07:26 +0000)]
[WK2] Add APIs to get/set encoding detector
https://bugs.webkit.org/show_bug.cgi?id=100066

Patch by Yuni Jeong <yhnet.jung@samsung.com> on 2012-11-01
Reviewed by Gyuyoung Kim.

Add setting APIs for encoding dector and a unit test.

* Shared/WebPreferencesStore.h:
(WebKit):
* UIProcess/API/C/WKPreferences.cpp:
(WKPreferencesSetUsesEncodingDetector):
(WKPreferencesGetUsesEncodingDetector):
* UIProcess/API/C/WKPreferences.h:
* UIProcess/API/efl/ewk_settings.cpp:
(ewk_setting_uses_encoding_detector_set):
(ewk_setting_uses_encoding_detector_get):
* UIProcess/API/efl/ewk_settings.h:
* UIProcess/API/efl/tests/test_ewk2_settings.cpp:
(TEST_F):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@133133 268f45cc-cd09-0410-ab3c-d52691b4dbfc

6 years ago[Chromium] Page popup shows up at wrong locations when root view is scrolled
keishi@webkit.org [Thu, 1 Nov 2012 07:01:00 +0000 (07:01 +0000)]
[Chromium] Page popup shows up at wrong locations when root view is scrolled
https://bugs.webkit.org/show_bug.cgi?id=100830

Reviewed by Kent Tamura.

We need to account for the scroll position when calculating the root view rectangle in the screen.

* src/DateTimeChooserImpl.cpp:
(WebKit::DateTimeChooserImpl::writeDocument):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@133132 268f45cc-cd09-0410-ab3c-d52691b4dbfc

6 years agoSparse array size threshold should be increased to 100000
mark.lam@apple.com [Thu, 1 Nov 2012 06:55:40 +0000 (06:55 +0000)]
Sparse array size threshold should be increased to 100000
https://bugs.webkit.org/show_bug.cgi?id=100827

Patch by Filip Pizlo <fpizlo@apple.com> on 2012-10-31
Reviewed by Oliver Hunt.

This enables the use of contiguous arrays in programs that previously
couldn't use them. And I so far can't see any examples of this being
a downside. To the extent that there is a downside, it ought to be
addressed by GC: https://bugs.webkit.org/show_bug.cgi?id=100828

* runtime/ArrayConventions.h:
(JSC):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@133131 268f45cc-cd09-0410-ab3c-d52691b4dbfc

6 years agoRemove some CachedResource::Status's in favor of looking at CachedResource::m_error
japhet@chromium.org [Thu, 1 Nov 2012 06:20:00 +0000 (06:20 +0000)]
Remove some CachedResource::Status's in favor of looking at CachedResource::m_error
https://bugs.webkit.org/show_bug.cgi?id=100901

Reviewed by Adam Barth.

No new tests, refactor only.

* inspector/InspectorPageAgent.cpp:
(WebCore::InspectorPageAgent::buildObjectForFrameTree):
* loader/DocumentThreadableLoader.cpp:
(WebCore::DocumentThreadableLoader::notifyFinished):
* loader/FrameLoader.cpp:
(WebCore::FrameLoader::loadInSameDocument):
* loader/SubresourceLoader.cpp:
(WebCore::SubresourceLoader::didFail):
(WebCore::SubresourceLoader::willCancel):
* loader/cache/CachedResource.cpp:
(WebCore::CachedResource::stopLoading):
* loader/cache/CachedResource.h:
(WebCore::CachedResource::wasCanceled):
(WebCore::CachedResource::errorOccurred):
(WebCore::CachedResource::loadFailedOrCanceled):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@133130 268f45cc-cd09-0410-ab3c-d52691b4dbfc

6 years agoSparse array size threshold should be increased to 100000
fpizlo@apple.com [Thu, 1 Nov 2012 06:09:13 +0000 (06:09 +0000)]
Sparse array size threshold should be increased to 100000
https://bugs.webkit.org/show_bug.cgi?id=100827

Reviewed by Oliver Hunt.

This enables the use of contiguous arrays in programs that previously
couldn't use them. And I so far can't see any examples of this being
a downside. To the extent that there is a downside, it ought to be
addressed by GC: https://bugs.webkit.org/show_bug.cgi?id=100828

* runtime/ArrayConventions.h:
(JSC):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@133129 268f45cc-cd09-0410-ab3c-d52691b4dbfc