WebKit-https.git
7 years ago[v8] Convert string conversion calls to one byte versions
commit-queue@webkit.org [Tue, 22 Jan 2013 11:59:19 +0000 (11:59 +0000)]
[v8] Convert string conversion calls to one byte versions
https://bugs.webkit.org/show_bug.cgi?id=107524

Patch by Dan Carney <dcarney@google.com> on 2013-01-22
Reviewed by Kentaro Hara.

No new tests. No change in functionality.

* bindings/v8/SerializedScriptValue.cpp:
* bindings/v8/V8StringResource.cpp:
(WebCore::WebCoreStringResourceBase::toWebCoreStringResourceBase):
(StringTraits):
(WebCore::true):
(WebCore::v8StringToWebCoreString):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@140409 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoWeb Inspector: [Network] CookiesTable refactoring.
commit-queue@webkit.org [Tue, 22 Jan 2013 11:57:31 +0000 (11:57 +0000)]
Web Inspector: [Network] CookiesTable refactoring.
https://bugs.webkit.org/show_bug.cgi?id=107511

Patch by Eugene Klyuchnikov <eustas@chromium.org> on 2013-01-22
Reviewed by Pavel Feldman.

CookiesTable uses integer indexes as column identifiers.
This leads to cryptic code.

Descriptive column identifiers will make code clearer.
Also "addCookiesFolder" replaced with "setCookieFolders" to avoid
repetitive "_rebuildTable" invocation.

* inspector/front-end/CookiesTable.js:
(WebInspector.CookiesTable): Use descriptive column identifiers.
(WebInspector.CookiesTable.prototype.setCookieFolders):
Accept multiple folders.
(WebInspector.CookiesTable.prototype._rebuildTable):
Use descriptive column identifiers.
(WebInspector.CookiesTable.prototype._sortCookies):
Ditto.
(WebInspector.CookiesTable.prototype._createGridNode):
Ditto.
* inspector/front-end/RequestCookiesView.js:
(WebInspector.RequestCookiesView.prototype._buildCookiesTable):
Use "setCookieFolders".

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@140408 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years ago[GTK] Disable support for unprefixed CSS animations, transitions, transforms in relea...
zandobersek@gmail.com [Tue, 22 Jan 2013 11:41:46 +0000 (11:41 +0000)]
[GTK] Disable support for unprefixed CSS animations, transitions, transforms in release builds
https://bugs.webkit.org/show_bug.cgi?id=107436

Reviewed by Alexis Menard.

Disable the unprefixed CSS animations, transitions and transforms support
in the release builds for now. The work in this area is not yet complete
and should not ship in such state.

No new tests - no new functionality.

* GNUmakefile.features.am.in:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@140407 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoUnreviewed GTK gardening.
zandobersek@gmail.com [Tue, 22 Jan 2013 11:31:36 +0000 (11:31 +0000)]
Unreviewed GTK gardening.

Adding back an expectation that was errornously removed in r140404.

* platform/gtk/TestExpectations:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@140406 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoWeb Inspector: [REGRESSION] SASSSourceMapping broken: _bindUISourceCode method is...
apavlov@chromium.org [Tue, 22 Jan 2013 11:13:13 +0000 (11:13 +0000)]
Web Inspector: [REGRESSION] SASSSourceMapping broken: _bindUISourceCode method is absent
https://bugs.webkit.org/show_bug.cgi?id=107529

Reviewed by Pavel Feldman.

* inspector/front-end/SASSSourceMapping.js:
(WebInspector.SASSSourceMapping.prototype._fileSaveFinished):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@140405 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoUnreviewed GTK gardening.
zandobersek@gmail.com [Tue, 22 Jan 2013 10:34:29 +0000 (10:34 +0000)]
Unreviewed GTK gardening.

Adding timeout expectations for two tests from the JSRegress suite that
were added in r140276. The timeouts are GTK-specific.
Expanding the main failure expectation for all Encrypted Media tests to
expect either failures or timeouts.

* platform/gtk/TestExpectations:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@140404 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoAtomicMarkupTokenBase must die
abarth@webkit.org [Tue, 22 Jan 2013 10:26:20 +0000 (10:26 +0000)]
AtomicMarkupTokenBase must die
https://bugs.webkit.org/show_bug.cgi?id=107513

Reviewed by Eric Seidel.

Now that NEW_XML has been removed from the tree, there is no reason to
have AtomicMarkupTokenBase exist as a templated base class for
AtomicHTMLToken. This patch just merges the code back into
AtomicHTMLToken.

* html/parser/HTMLToken.h:
(AtomicHTMLToken):
(WebCore::AtomicHTMLToken::type):
(WebCore::AtomicHTMLToken::name):
(WebCore::AtomicHTMLToken::setName):
(WebCore::AtomicHTMLToken::selfClosing):
(WebCore::AtomicHTMLToken::getAttributeItem):
(WebCore::AtomicHTMLToken::attributes):
(WebCore::AtomicHTMLToken::characters):
(WebCore::AtomicHTMLToken::charactersLength):
(WebCore::AtomicHTMLToken::isAll8BitData):
(WebCore::AtomicHTMLToken::comment):
(WebCore::AtomicHTMLToken::publicIdentifier):
(WebCore::AtomicHTMLToken::systemIdentifier):
(WebCore::AtomicHTMLToken::clearExternalCharacters):
(WebCore::AtomicHTMLToken::AtomicHTMLToken):
(WebCore::AtomicMarkupTokenBase::initializeAttributes):
(WebCore):
* xml/parser/MarkupTokenBase.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@140403 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoTurn on ENABLE_THREADED_HTML_PARSER for Chromium (it's still disabled at runtime)
eric@webkit.org [Tue, 22 Jan 2013 10:24:30 +0000 (10:24 +0000)]
Turn on ENABLE_THREADED_HTML_PARSER for Chromium (it's still disabled at runtime)
https://bugs.webkit.org/show_bug.cgi?id=107519

Reviewed by Adam Barth.

Source/WebCore:

Add some ASSERTs to be double-dog-sure that we're never using
any of the threading code when threading is not enabled.

I also removed an if-branch in the !ENABLE_THREADED_HTML_PARSER case
as it's not needed.

* html/parser/HTMLDocumentParser.cpp:
(WebCore::HTMLDocumentParser::prepareToStopParsing):
(WebCore::HTMLDocumentParser::startBackgroundParser):
(WebCore::HTMLDocumentParser::stopBackgroundParser):

Source/WebKit/chromium:

This makes our development lives easier, and makes it possible for the bots
to run threaded-parser-only tests by toggling the runtime enable
via window.internals.settings.

* features.gypi:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@140402 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years ago[V8] Make an Isolate parameter mandatory in toV8()
haraken@chromium.org [Tue, 22 Jan 2013 09:48:22 +0000 (09:48 +0000)]
[V8] Make an Isolate parameter mandatory in toV8()
https://bugs.webkit.org/show_bug.cgi?id=107520

Reviewed by Adam Barth.

Now it's safe to remove an optional Isolate parameter.

No tests. No change in behavior.

Source/WebCore:

* bindings/scripts/CodeGeneratorV8.pm:
(GenerateHeader):
* bindings/scripts/test/V8/V8Float64Array.h:
(WebCore::toV8):
* bindings/scripts/test/V8/V8TestActiveDOMObject.h:
(WebCore::toV8):
* bindings/scripts/test/V8/V8TestCustomNamedGetter.h:
(WebCore::toV8):
* bindings/scripts/test/V8/V8TestEventConstructor.h:
(WebCore::toV8):
* bindings/scripts/test/V8/V8TestEventTarget.h:
(WebCore::toV8):
* bindings/scripts/test/V8/V8TestException.h:
(WebCore::toV8):
* bindings/scripts/test/V8/V8TestInterface.h:
(WebCore::toV8):
* bindings/scripts/test/V8/V8TestMediaQueryListListener.h:
(WebCore::toV8):
* bindings/scripts/test/V8/V8TestNamedConstructor.h:
(WebCore::toV8):
* bindings/scripts/test/V8/V8TestNode.h:
(WebCore::toV8):
* bindings/scripts/test/V8/V8TestObj.h:
(WebCore::toV8):
* bindings/scripts/test/V8/V8TestOverloadedConstructors.h:
(WebCore::toV8):
* bindings/scripts/test/V8/V8TestSerializedScriptValueInterface.h:
(WebCore::toV8):
* css/MediaQueryListListener.cpp:
(WebCore::MediaQueryListListener::queryChanged):
* testing/v8/WebCoreTestSupport.cpp:
(WebCoreTestSupport::injectInternalsObject):

Source/WebKit/chromium:

* src/InspectorFrontendClientImpl.cpp:
(WebKit::InspectorFrontendClientImpl::windowObjectCleared):
* src/WebArrayBuffer.cpp:
(WebKit::WebArrayBuffer::toV8Value):
* src/WebBlob.cpp:
(WebKit::WebBlob::toV8Value):
* src/WebFrameImpl.cpp:
(WebKit::WebFrameImpl::createFileSystem):
(WebKit::WebFrameImpl::createSerializableFileSystem):
(WebKit::WebFrameImpl::createFileEntry):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@140401 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years ago[CSS Regions] min-max height will not trigger a relayout when set on a region with...
mihnea@adobe.com [Tue, 22 Jan 2013 09:43:45 +0000 (09:43 +0000)]
[CSS Regions] min-max height will not trigger a relayout when set on a region with auto-height
https://bugs.webkit.org/show_bug.cgi?id=103738

Reviewed by David Hyatt.

Source/WebCore:

When the style of an auto-height region changes with different min/max-height, we have to make sure
that the 2-pass layout algorithm is run so that the region height is properly computed. When doing a 2-pass layout
for auto-height regions, we have to make sure that in the first pass, all auto-height regions start without a previously
computed overrideLogicalContentHeight, otherwise we may end-up with incorrect results, illustrated by tests
autoheight-two-pass-layout-complex-001.html and autoheight-two-pass-layout-complex-002.html.

If we do not have auto height regions, a single pass layout in enough to correctly flow content inside regions.
When we have auto height regions, the algorithm to lay out content should be:
1. If the flow thread content does not need lay out, we do a single pass layout.
If during the layout, an auto-height region needs layout or a normal (not auto height) region changes its box dimensions,
we need to perform a new 2-pass layout after this layout completes. If not, we bail out and skip step 2.
2. Do a full two pass layout and make sure that all auto height regions start the 2-pass layout without a previously computed
override logical content height.

Tests: fast/regions/autoheight-region-decrease-maxheight.html
       fast/regions/autoheight-region-decrease-minheight.html
       fast/regions/autoheight-region-decrease-width.html
       fast/regions/autoheight-region-increase-maxheight.html
       fast/regions/autoheight-region-increase-minheight.html
       fast/regions/autoheight-region-increase-width.html
       fast/regions/autoheight-region-new-maxheight-vrl.html
       fast/regions/autoheight-region-new-maxheight.html
       fast/regions/autoheight-region-new-minheight-vlr.html
       fast/regions/autoheight-region-new-minheight.html
       fast/regions/autoheight-region-remove-maxheight.html
       fast/regions/autoheight-region-remove-minheight.html
       fast/regions/autoheight-two-pass-layout-complex-001.html
       fast/regions/autoheight-two-pass-layout-complex-002.html
       fast/regions/region-height-auto-to-defined.html
       fast/regions/region-height-defined-to-auto.html

* rendering/FlowThreadController.cpp: Added a flag that is needed to mark whether we need a full 2-pass layout.
(WebCore::FlowThreadController::FlowThreadController):
* rendering/FlowThreadController.h:
(WebCore::FlowThreadController::needsTwoPassLayoutForAutoHeightRegions):
(WebCore::FlowThreadController::setNeedsTwoPassLayoutForAutoHeightRegions):
* rendering/RenderFlowThread.cpp:
(WebCore::RenderFlowThread::resetRegionsOverrideLogicalContentHeight): Use invalidateRegions because we need also to mark the flow thread as needing layout.
(WebCore::RenderFlowThread::markAutoLogicalHeightRegionsForLayout):
* rendering/RenderRegion.cpp:
(WebCore::RenderRegion::layout): Make sure we do a 2-pass layout if needed.
* rendering/RenderView.cpp:
(WebCore::RenderView::layoutContentInAutoLogicalHeightRegions): Added a new function that models the layout algorithm for the case when we have auto height regions
and flow threads.
(WebCore::RenderView::layout):
(WebCore::RenderView::checkTwoPassLayoutForAutoHeightRegions):
* rendering/RenderView.h:

LayoutTests:

Added tests for dynamically change the min/max-height for auto-height regions, including also vertical writing modes.
Added tests for dynamically changing the width of auto-height regions to make sure we are correctly computing their height.
Added two tests for complex auto-height regions layout which require a full two-pass layout after some regions from region
chain are laid out.

* fast/regions/autoheight-region-decrease-maxheight-expected.txt: Added.
* fast/regions/autoheight-region-decrease-maxheight.html: Added.
* fast/regions/autoheight-region-decrease-minheight-expected.txt: Added.
* fast/regions/autoheight-region-decrease-minheight.html: Added.
* fast/regions/autoheight-region-decrease-width-expected.txt: Added.
* fast/regions/autoheight-region-decrease-width.html: Added.
* fast/regions/autoheight-region-increase-maxheight-expected.txt: Added.
* fast/regions/autoheight-region-increase-maxheight.html: Added.
* fast/regions/autoheight-region-increase-minheight-expected.txt: Added.
* fast/regions/autoheight-region-increase-minheight.html: Added.
* fast/regions/autoheight-region-increase-width-expected.txt: Added.
* fast/regions/autoheight-region-increase-width.html: Added.
* fast/regions/autoheight-region-new-maxheight-expected.txt: Added.
* fast/regions/autoheight-region-new-maxheight-vrl-expected.txt: Added.
* fast/regions/autoheight-region-new-maxheight-vrl.html: Added.
* fast/regions/autoheight-region-new-maxheight.html: Added.
* fast/regions/autoheight-region-new-minheight-expected.txt: Added.
* fast/regions/autoheight-region-new-minheight-vlr-expected.txt: Added.
* fast/regions/autoheight-region-new-minheight-vlr.html: Added.
* fast/regions/autoheight-region-new-minheight.html: Added.
* fast/regions/autoheight-region-remove-maxheight-expected.txt: Added.
* fast/regions/autoheight-region-remove-maxheight.html: Added.
* fast/regions/autoheight-region-remove-minheight-expected.txt: Added.
* fast/regions/autoheight-region-remove-minheight.html: Added.
* fast/regions/autoheight-two-pass-layout-complex-001-expected.txt: Added.
* fast/regions/autoheight-two-pass-layout-complex-001.html: Added.
* fast/regions/autoheight-two-pass-layout-complex-002-expected.txt: Added.
* fast/regions/autoheight-two-pass-layout-complex-002.html: Added.
* fast/regions/region-height-auto-to-defined-expected.txt: Added.
* fast/regions/region-height-auto-to-defined.html: Added.
* fast/regions/region-height-defined-to-auto-expected.txt: Added.
* fast/regions/region-height-defined-to-auto.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@140400 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoENABLE(NEW_XML) isn't used by anyone and no one is actively working on it
abarth@webkit.org [Tue, 22 Jan 2013 09:16:16 +0000 (09:16 +0000)]
ENABLE(NEW_XML) isn't used by anyone and no one is actively working on it
https://bugs.webkit.org/show_bug.cgi?id=100710

Reviewed by Eric Seidel.

As discussed on webkit-dev
<http://lists.webkit.org/pipermail/webkit-dev/2012-August/022103.html>,
ENABLE(NEW_XML) isn't used by anyone and no one is actively working on
it. We should remove it from the code base for the time being. If folks
want to work on it again, we can restore it from SVN history.

* GNUmakefile.list.am:
* Target.pri:
* WebCore.gypi:
* WebCore.vcproj/WebCore.vcproj:
* WebCore.xcodeproj/project.pbxproj:
* dom/Document.cpp:
(WebCore::Document::createParser):
* dom/DocumentFragment.cpp:
(WebCore::DocumentFragment::parseXML):
* xml/parser/NewXMLDocumentParser.cpp: Removed.
* xml/parser/NewXMLDocumentParser.h: Removed.
* xml/parser/XMLCharacterReferenceParser.cpp: Removed.
* xml/parser/XMLCharacterReferenceParser.h: Removed.
* xml/parser/XMLToken.h: Removed.
* xml/parser/XMLTokenizer.cpp: Removed.
* xml/parser/XMLTokenizer.h: Removed.
* xml/parser/XMLTreeBuilder.cpp: Removed.
* xml/parser/XMLTreeBuilder.h: Removed.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@140399 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoSource/WebKit/gtk: [GTK] Add listener for direction-changed signal in WebKitWebView
commit-queue@webkit.org [Tue, 22 Jan 2013 09:15:07 +0000 (09:15 +0000)]
Source/WebKit/gtk: [GTK] Add listener for direction-changed signal in WebKitWebView
https://bugs.webkit.org/show_bug.cgi?id=107131

Patch by Manuel Rego Casasnovas <rego@igalia.com> on 2013-01-22
Reviewed by Philippe Normand.

* webkit/webkitwebview.cpp:
(webkit_web_view_init): Add listener for direction-changed signal.
(webkitWebViewDirectionChanged): Implement listener using
Editor::setBaseWritingDirection().

Tools: [GTK] Implement TestRunner::setTextDirection
https://bugs.webkit.org/show_bug.cgi?id=107131

Patch by Manuel Rego Casasnovas <rego@igalia.com> on 2013-01-22
Reviewed by Philippe Normand.

* DumpRenderTree/gtk/DumpRenderTree.cpp:
(resetDefaultsToConsistentValues): Reset direction to default value.
* DumpRenderTree/gtk/TestRunnerGtk.cpp:
(TestRunner::setTextDirection): Implement method using
gtk_widget_set_direction.

LayoutTests: [GTK] Implement testRunner::setTextDirection
https://bugs.webkit.org/show_bug.cgi?id=107131

Patch by Manuel Rego Casasnovas <rego@igalia.com> on 2013-01-22
Reviewed by Philippe Normand.

* platform/gtk-wk2/TestExpectations: Unflag
fast/html/set-text-direction.html as it was already passing in WK2.
* platform/gtk/TestExpectations: Remove
fast/html/set-text-direction.html.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@140398 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years ago[V8] Pass an Isolate to toV8() of event listeners
haraken@chromium.org [Tue, 22 Jan 2013 09:06:22 +0000 (09:06 +0000)]
[V8] Pass an Isolate to toV8() of event listeners
https://bugs.webkit.org/show_bug.cgi?id=107517

Reviewed by Adam Barth.

By using Context::GetIsolate(), this patch passes an Isolate to toV8().
I think this would be the last patch to pass an Isolate to toV8().
In the next patch, I will make an Isolate parameter mandatory for toV8().

No tests. No change in behavior.

* bindings/v8/V8AbstractEventListener.cpp:
(WebCore::V8AbstractEventListener::getReceiverObject):
* bindings/v8/V8AbstractEventListener.h:
(V8AbstractEventListener):
* bindings/v8/V8EventListener.cpp:
(WebCore::V8EventListener::callListenerFunction):
* bindings/v8/V8LazyEventListener.cpp:
(WebCore::toObjectWrapper):
(WebCore::V8LazyEventListener::callListenerFunction):
(WebCore::V8LazyEventListener::prepareListenerObject):
* bindings/v8/V8WorkerContextEventListener.cpp:
(WebCore::V8WorkerContextEventListener::getReceiverObject):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@140397 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years ago[chromium] move spell checking mocks to TestRunner library
jochen@chromium.org [Tue, 22 Jan 2013 09:00:37 +0000 (09:00 +0000)]
[chromium] move spell checking mocks to TestRunner library
https://bugs.webkit.org/show_bug.cgi?id=107455

Reviewed by Kent Tamura.

The content module doesn't provide its own WebSpellCheckClient.

* DumpRenderTree/DumpRenderTree.gypi:
* DumpRenderTree/chromium/TestRunner/public/WebTestDelegate.h:
(WebTestRunner::WebTestDelegate::fillSpellingSuggestionList):
* DumpRenderTree/chromium/TestRunner/public/WebTestProxy.h:
(WebKit):
(WebTestRunner):
(WebTestProxyBase):
* DumpRenderTree/chromium/TestRunner/src/EventSender.cpp:
(WebTestRunner):
(WebTestRunner::makeMenuItemStringsFor):
* DumpRenderTree/chromium/TestRunner/src/MockGrammarCheck.cpp: Renamed from Tools/DumpRenderTree/chromium/MockGrammarCheck.cpp.
(MockGrammarCheck::checkGrammarOfString):
* DumpRenderTree/chromium/TestRunner/src/MockGrammarCheck.h: Renamed from Tools/DumpRenderTree/chromium/MockGrammarCheck.h.
(WebKit):
(MockGrammarCheck):
* DumpRenderTree/chromium/TestRunner/src/MockSpellCheck.cpp: Renamed from Tools/DumpRenderTree/chromium/MockSpellCheck.cpp.
(MockSpellCheck::MockSpellCheck):
(MockSpellCheck::~MockSpellCheck):
(isNotASCIIAlpha):
(MockSpellCheck::spellCheckWord):
(MockSpellCheck::fillSuggestionList):
(MockSpellCheck::initializeIfNeeded):
* DumpRenderTree/chromium/TestRunner/src/MockSpellCheck.h: Renamed from Tools/DumpRenderTree/chromium/MockSpellCheck.h.
(MockSpellCheck):
* DumpRenderTree/chromium/TestRunner/src/SpellCheckClient.cpp: Added.
(WebTestRunner::SpellCheckClient::SpellCheckClient):
(WebTestRunner):
(WebTestRunner::SpellCheckClient::~SpellCheckClient):
(WebTestRunner::SpellCheckClient::setDelegate):
(WebTestRunner::SpellCheckClient::spellCheck):
(WebTestRunner::SpellCheckClient::checkTextOfParagraph):
(WebTestRunner::SpellCheckClient::requestCheckingOfText):
(WebTestRunner::SpellCheckClient::finishLastTextCheck):
(WebTestRunner::SpellCheckClient::autoCorrectWord):
* DumpRenderTree/chromium/TestRunner/src/SpellCheckClient.h: Copied from Tools/DumpRenderTree/chromium/TestRunner/src/TestDelegate.h.
(WebTestRunner):
(SpellCheckClient):
(WebTestRunner::SpellCheckClient::taskList):
(WebTestRunner::SpellCheckClient::mockSpellCheck):
* DumpRenderTree/chromium/TestRunner/src/TestDelegate.h:
(TestDelegate):
* DumpRenderTree/chromium/TestRunner/src/WebTestInterfaces.cpp:
(WebTestInterfaces::Internal):
* DumpRenderTree/chromium/TestRunner/src/WebTestProxy.cpp:
(WebTestRunner::WebTestProxyBase::WebTestProxyBase):
(WebTestRunner::WebTestProxyBase::~WebTestProxyBase):
(WebTestRunner::WebTestProxyBase::setDelegate):
(WebTestRunner::WebTestProxyBase::spellCheckClient):
(WebTestRunner):
* DumpRenderTree/chromium/WebViewHost.cpp:
(WebViewHost::WebViewHost):
(WebViewHost::setWebWidget):
* DumpRenderTree/chromium/WebViewHost.h:
(WebViewHost):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@140396 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoUnreviewed EFL gardening.
christophe.dumez@intel.com [Tue, 22 Jan 2013 08:58:38 +0000 (08:58 +0000)]
Unreviewed EFL gardening.

Unskip 2 tests that were sometimes asserting on the bots as
the problems seems to have been fixed in r140166.

* platform/efl/TestExpectations:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@140395 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoPDFPlugin: Build PDFPlugin everywhere, enable at runtime
timothy_horton@apple.com [Tue, 22 Jan 2013 08:38:29 +0000 (08:38 +0000)]
PDFPlugin: Build PDFPlugin everywhere, enable at runtime
https://bugs.webkit.org/show_bug.cgi?id=107117

Reviewed by Alexey Proskuryakov.

Since PDFLayerController SPI is all forward-declared, the plugin should build
on all Mac platforms, and can be enabled at runtime.

* Configurations/FeatureDefines.xcconfig:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@140394 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoAX: Should assert if we try to create two AXObjects that point to the same renderer...
dmazzoni@google.com [Tue, 22 Jan 2013 08:33:45 +0000 (08:33 +0000)]
AX: Should assert if we try to create two AXObjects that point to the same renderer or node
https://bugs.webkit.org/show_bug.cgi?id=107504

Reviewed by Chris Fleizach.

If two accessibility objects get created that point to the
same widget, renderer, or node, that can lead to crashes or
memory corruption later; make it assert instead of crashing.

* accessibility/AXObjectCache.cpp:
(WebCore::AXObjectCache::getOrCreate):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@140393 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years ago[V8] Pass an Isolate to toV8()
haraken@chromium.org [Tue, 22 Jan 2013 08:32:01 +0000 (08:32 +0000)]
[V8] Pass an Isolate to toV8()
https://bugs.webkit.org/show_bug.cgi?id=107512

Reviewed by Adam Barth.

By using Context::GetIsolate(), this patch passes an Isolate to toV8().

No tests. No change in behavior.

* bindings/v8/IDBBindingUtilities.cpp:
(WebCore::injectIDBKeyIntoScriptValue):
(WebCore::idbKeyToScriptValue):
* bindings/v8/ScriptController.cpp:
(WebCore::createScriptObject):
(WebCore::ScriptController::createScriptObjectForPluginElement):
* bindings/v8/ScriptDebugServer.cpp:
(WebCore::ScriptDebugServer::currentCallFrame):
* bindings/v8/ScriptObject.cpp:
(WebCore::ScriptGlobalObject::set):
* bindings/v8/V8AbstractEventListener.cpp:
(WebCore::V8AbstractEventListener::handleEvent):
* bindings/v8/V8DOMWindowShell.cpp:
(WebCore::V8DOMWindowShell::updateDocumentProperty):
* bindings/v8/V8MutationCallback.cpp:
(WebCore::V8MutationCallback::handleEvent):
* bindings/v8/V8NodeFilterCondition.cpp:
(WebCore::V8NodeFilterCondition::acceptNode):
* bindings/v8/V8WorkerContextEventListener.cpp:
(WebCore::V8WorkerContextEventListener::handleEvent):
* bindings/v8/custom/V8CustomSQLStatementErrorCallback.cpp:
(WebCore::V8SQLStatementErrorCallback::handleEvent):
* bindings/v8/custom/V8InjectedScriptHostCustom.cpp:
(WebCore::InjectedScriptHost::nodeAsScriptValue):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@140392 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoWeb Inspector: Show requests in `curl` syntax in DevTools → Network → Headers
commit-queue@webkit.org [Tue, 22 Jan 2013 08:13:08 +0000 (08:13 +0000)]
Web Inspector: Show requests in `curl` syntax in DevTools → Network → Headers
https://bugs.webkit.org/show_bug.cgi?id=107276

Patch by Sergey Ryazanov <serya@chromium.org> on 2013-01-22
Reviewed by Pavel Feldman.

Source/WebCore:

Test: inspector/curl-command.html

* inspector/front-end/NetworkPanel.js:
(WebInspector.NetworkLogView.prototype._generateCurlCommand):

LayoutTests:

* inspector/curl-command-expected.txt: Added.
* inspector/curl-command.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@140391 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoWeb Inspector: reuse JS heap profiler UI for native heap graph representaion
yurys@chromium.org [Tue, 22 Jan 2013 08:10:10 +0000 (08:10 +0000)]
Web Inspector: reuse JS heap profiler UI for native heap graph representaion
https://bugs.webkit.org/show_bug.cgi?id=107452

Reviewed by Pavel Feldman.

NativeHeapSnapshot is introduced to leverage JS heap snapshot processing algorithms
for native heap graph. New snapshot type ("Native Heap Snapshot") is added to the
profiles panel. It will represent native heap snapshots using the same UI as we use
for JS heap snapshots.

* WebCore.gypi:
* WebCore.vcproj/WebCore.vcproj:
* inspector/compile-front-end.py:
* inspector/front-end/HeapSnapshot.js:
(WebInspector.HeapSnapshot):
(WebInspector.HeapSnapshot.prototype._init):
* inspector/front-end/HeapSnapshotProxy.js:
(WebInspector.HeapSnapshotWorker):
* inspector/front-end/HeapSnapshotView.js:
(WebInspector.HeapProfileHeader.prototype.load):
(WebInspector.HeapProfileHeader.prototype.startSnapshotTransfer): extracted a few methods
to support overrides for native heap snapshots.
(WebInspector.HeapProfileHeader.prototype.snapshotConstructorName):
(WebInspector.HeapProfileHeader.prototype._setupWorker):
* inspector/front-end/NativeHeapSnapshot.js: Added.
(WebInspector.NativeHeapSnapshot):
(WebInspector.NativeHeapSnapshot.prototype.createNode):
(WebInspector.NativeHeapSnapshot.prototype.createEdge):
(WebInspector.NativeHeapSnapshot.prototype.createRetainingEdge):
(WebInspector.NativeHeapSnapshot.prototype._markInvisibleEdges):
(WebInspector.NativeHeapSnapshot.prototype._calculateFlags):
(WebInspector.NativeHeapSnapshot.prototype.canHaveDistanceOne):
(WebInspector.NativeHeapSnapshot.prototype.userObjectsMapAndFlag):
(WebInspector.NativeHeapSnapshotNode):
(WebInspector.NativeHeapSnapshotNode.prototype.canBeQueried):
(WebInspector.NativeHeapSnapshotNode.prototype.isUserObject):
(WebInspector.NativeHeapSnapshotNode.prototype.className):
(WebInspector.NativeHeapSnapshotNode.prototype.classIndex):
(WebInspector.NativeHeapSnapshotNode.prototype.id):
(WebInspector.NativeHeapSnapshotNode.prototype.name):
(WebInspector.NativeHeapSnapshotNode.prototype.isHidden):
(WebInspector.NativeHeapSnapshotNode.prototype.isSynthetic):
(WebInspector.NativeHeapSnapshotNode.prototype.isWindow):
(WebInspector.NativeHeapSnapshotNode.prototype.isDetachedDOMTreesRoot):
(WebInspector.NativeHeapSnapshotNode.prototype.isDetachedDOMTree):
(WebInspector.NativeHeapSnapshotEdge):
(WebInspector.NativeHeapSnapshotEdge.prototype.clone):
(WebInspector.NativeHeapSnapshotEdge.prototype.hasStringName):
(WebInspector.NativeHeapSnapshotEdge.prototype.isElement):
(WebInspector.NativeHeapSnapshotEdge.prototype.isHidden):
(WebInspector.NativeHeapSnapshotEdge.prototype.isWeak):
(WebInspector.NativeHeapSnapshotEdge.prototype.isInternal):
(WebInspector.NativeHeapSnapshotEdge.prototype.isInvisible):
(WebInspector.NativeHeapSnapshotEdge.prototype.isShortcut):
(WebInspector.NativeHeapSnapshotEdge.prototype.name):
(WebInspector.NativeHeapSnapshotEdge.prototype.toString):
(WebInspector.NativeHeapSnapshotEdge.prototype._nameOrIndex):
(WebInspector.NativeHeapSnapshotRetainerEdge):
(WebInspector.NativeHeapSnapshotRetainerEdge.prototype.clone):
(WebInspector.NativeHeapSnapshotRetainerEdge.prototype.isElement):
(WebInspector.NativeHeapSnapshotRetainerEdge.prototype.isHidden):
(WebInspector.NativeHeapSnapshotRetainerEdge.prototype.isInternal):
(WebInspector.NativeHeapSnapshotRetainerEdge.prototype.isInvisible):
(WebInspector.NativeHeapSnapshotRetainerEdge.prototype.isShortcut):
(WebInspector.NativeHeapSnapshotRetainerEdge.prototype.isWeak):
* inspector/front-end/NativeMemorySnapshotView.js:
(WebInspector.NativeSnapshotProfileType):
(WebInspector.NativeSnapshotProfileType.prototype.get buttonTooltip):
(WebInspector.NativeSnapshotProfileType.prototype.buttonClicked.didReceiveMemorySnapshot):
(WebInspector.NativeSnapshotProfileType.prototype.buttonClicked):
(WebInspector.NativeSnapshotProfileType.prototype.get treeItemTitle):
(WebInspector.NativeSnapshotProfileType.prototype.get description):
(WebInspector.NativeSnapshotProfileType.prototype.createTemporaryProfile):
(WebInspector.NativeSnapshotProfileType.prototype.createProfile):
(WebInspector.NativeSnapshotProfileHeader):
(WebInspector.NativeSnapshotProfileHeader.prototype.createView):
(WebInspector.NativeSnapshotProfileHeader.prototype.startSnapshotTransfer):
(WebInspector.NativeSnapshotProfileHeader.prototype.snapshotConstructorName):
(WebInspector.NativeSnapshotProfileHeader.prototype._takeNativeSnapshot): build a datastructure
resembling one used for JS heap snapshots but populated with the data from the native snaphot.
(WebInspector.NativeHeapSnapshotView):
(WebInspector.NativeHeapSnapshotView.prototype.get profile):
(WebInspector.NativeMemoryProfileType):
(WebInspector.NativeMemoryProfileType.prototype.get buttonTooltip):
* inspector/front-end/ProfilesPanel.js:
(WebInspector.ProfilesPanel):
* inspector/front-end/WebKit.qrc:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@140390 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoSource/WebKit: [EFL] Adds Accessibility support to wk1
commit-queue@webkit.org [Tue, 22 Jan 2013 08:02:36 +0000 (08:02 +0000)]
Source/WebKit: [EFL] Adds Accessibility support to wk1
https://bugs.webkit.org/show_bug.cgi?id=107440

Patch by Krzysztof Czech <k.czech@samsung.com> on 2013-01-22
Reviewed by Gyuyoung Kim.

* PlatformEfl.cmake: Adds ATK headers and libraries.

Source/WebKit/efl: [EFL] Adds Accessibility support to wk1.
https://bugs.webkit.org/show_bug.cgi?id=107440

Patch by Krzysztof Czech <k.czech@samsung.com> on 2013-01-22
Reviewed by Gyuyoung Kim.

Adds possibility of focusing and retrieving accessible object.

* WebCoreSupport/DumpRenderTreeSupportEfl.cpp:
(DumpRenderTreeSupportEfl::rootAccessibleElement):
(DumpRenderTreeSupportEfl::focusedAccessibleElement):
* WebCoreSupport/DumpRenderTreeSupportEfl.h:

Tools: [EFL] Adds Accessibility support to wk1
https://bugs.webkit.org/show_bug.cgi?id=107440

Patch by Krzysztof Czech <k.czech@samsung.com> on 2013-01-22
Reviewed by Gyuyoung Kim.

* DumpRenderTree/efl/CMakeLists.txt: Adds ATK headers and libraries.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@140389 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years ago[Mac] WK1 MiniBrowser should clear delegates before releasing webview
commit-queue@webkit.org [Tue, 22 Jan 2013 07:39:52 +0000 (07:39 +0000)]
[Mac] WK1 MiniBrowser should clear delegates before releasing webview
https://bugs.webkit.org/show_bug.cgi?id=107506

Patch by Joseph Pecoraro <pecoraro@apple.com> on 2013-01-21
Reviewed by Dan Bernstein.

* MiniBrowser/mac/WK1BrowserWindowController.m:
(-[WK1BrowserWindowController dealloc]):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@140388 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years ago[V8] ScriptState::isolate() should use Context::GetIsolate()
haraken@chromium.org [Tue, 22 Jan 2013 07:33:16 +0000 (07:33 +0000)]
[V8] ScriptState::isolate() should use Context::GetIsolate()
https://bugs.webkit.org/show_bug.cgi?id=107490

Reviewed by Adam Barth.

Now V8 provides Context::GetIsolate(). We should use it
in ScriptState::isolate().

* bindings/v8/ScriptState.h:
(WebCore::ScriptState::isolate):
(ScriptState):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@140387 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years ago[EFL] Missing context menus strings
commit-queue@webkit.org [Tue, 22 Jan 2013 06:42:23 +0000 (06:42 +0000)]
[EFL] Missing context menus strings
https://bugs.webkit.org/show_bug.cgi?id=107104

Patch by Michał Pakuła vel Rutka <m.pakula@samsung.com> on 2013-01-21
Reviewed by Kenneth Rohde Christiansen.

Add 5 missing localization strings for text direction context menus.

* platform/efl/LocalizedStringsEfl.cpp:
(WebCore::contextMenuItemTagRightToLeft):
(WebCore::contextMenuItemTagLeftToRight):
(WebCore::contextMenuItemTagWritingDirectionMenu):
(WebCore::contextMenuItemTagTextDirectionMenu):
(WebCore::contextMenuItemTagDefaultDirection):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@140386 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoDate selection from calendar picker should dispatch 'input' event in addition to...
tkent@chromium.org [Tue, 22 Jan 2013 06:17:21 +0000 (06:17 +0000)]
Date selection from calendar picker should dispatch 'input' event in addition to 'change' event
https://bugs.webkit.org/show_bug.cgi?id=107427

Reviewed by Kentaro Hara.

Source/WebCore:

According to the specification and Opera's behavior, we should dispatch
not only 'change' event but also 'input' event when a user chooses a
date from the calender picker.

http://www.whatwg.org/specs/web-apps/current-work/multipage/common-input-element-attributes.html#common-event-behaviors
> When the user agent changes the element's value on behalf of the user
> (e.g. as part of a form prefilling feature), the user agent must follow
> these steps:
> 1. If the input event applies, queue a task to fire a simple event
> that bubbles named input at the input element.
> 2. If the change event applies, queue a task to fire a simple event
> that bubbles named change at the input element.

Tests: platform/chromium/fast/forms/calendar-picker/date-picker-events.html
       platform/chromium/fast/forms/calendar-picker/datetimelocal-picker-events.html

* html/InputType.cpp:
(WebCore::InputType::setValue): Add DispatchInputAndChangeEvent support.
* html/BaseChooserOnlyDateAndTimeInputType.cpp:
(WebCore::BaseChooserOnlyDateAndTimeInputType::didChooseValue):
Use DispatchInputAndChangeEvent, not DispatchChangeEvent.
* html/BaseMultipleFieldsDateAndTimeInputType.cpp:
(WebCore::BaseMultipleFieldsDateAndTimeInputType::pickerIndicatorChooseValue):
Ditto.

LayoutTests:

* platform/chromium/fast/forms/calendar-picker/date-picker-events-expected.txt: Added.
* platform/chromium/fast/forms/calendar-picker/date-picker-events.html: Added.
* platform/chromium/fast/forms/calendar-picker/datetimelocal-picker-events-expected.txt: Added.
* platform/chromium/fast/forms/calendar-picker/datetimelocal-picker-events.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@140385 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years ago[CHROMIUM] Suppress c4267 build warnings for Win64 targets
jschuh@chromium.org [Tue, 22 Jan 2013 06:13:26 +0000 (06:13 +0000)]
[CHROMIUM] Suppress c4267 build warnings for Win64 targets
https://bugs.webkit.org/show_bug.cgi?id=107499

Reviewed by Abhishek Arya.

Source/JavaScriptCore:

* JavaScriptCore.gyp/JavaScriptCore.gyp:

Source/Platform:

* Platform.gyp/Platform.gyp:

Source/WebCore:

No new tests. No code changes.

* WebCore.gyp/WebCore.gyp:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@140384 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoCSSSelector::value() should assert that m_match != Tag.
akling@apple.com [Tue, 22 Jan 2013 06:02:03 +0000 (06:02 +0000)]
CSSSelector::value() should assert that m_match != Tag.
<http://webkit.org/b/107500>

Reviewed by Anders Carlsson.

Add an assertion to catch anyone trying to extract a value() from a Tag selector.

* css/CSSSelector.h:
(CSSSelector):
(WebCore::CSSSelector::value):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@140383 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoMove HTML Attribute case-sensitivity logic out of SelectorChecker to HTMLDocument.
dglazkov@chromium.org [Tue, 22 Jan 2013 05:07:16 +0000 (05:07 +0000)]
Move HTML Attribute case-sensitivity logic out of SelectorChecker to HTMLDocument.
https://bugs.webkit.org/show_bug.cgi?id=107460

The case-sensitivity of HTML attributes is defined in HTML spec and is only applicable in HTML documents,
so having this logic in HTMLDocument seems like a good thing.

Reviewed by Eric Seidel.

Refactoring, covered by existing tests.

* css/SelectorChecker.cpp:
(WebCore): Moved code out of here.
(WebCore::isFastCheckableMatch): Changed the callsite to use new location.
(WebCore::SelectorChecker::checkOne): Ditto.
* html/HTMLDocument.cpp:
(WebCore::addLocalNameToSet): Moved the code into here.
(WebCore): Ditto.
(WebCore::createHtmlCaseInsensitiveAttributesSet): Ditto.
(WebCore::HTMLDocument::isCaseSensitiveAttribute): Ditto.
* html/HTMLDocument.h:
(HTMLDocument): Added decl for the newly moved function.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@140382 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoFrameSelection should use shadowHost instead of shadowAncestorNode
shinyak@chromium.org [Tue, 22 Jan 2013 05:04:29 +0000 (05:04 +0000)]
FrameSelection should use shadowHost instead of shadowAncestorNode
https://bugs.webkit.org/show_bug.cgi?id=107220

Reviewed by Kent Tamura.

shadowAncestorNode() is deprecated and we want to use shadowHost.
Here, all objects calling shadowAncestorNode() is in ShadowDOM, we can safely convert
the call to shadowHost().

No new tests, simple refactoring.

* editing/FrameSelection.cpp:
(WebCore::FrameSelection::selectAll):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@140381 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoUnreviewed. Rolled Chromium DEPS to r177962. Requested by
commit-queue@webkit.org [Tue, 22 Jan 2013 03:29:42 +0000 (03:29 +0000)]
Unreviewed.  Rolled Chromium DEPS to r177962.  Requested by
"Yoshifumi Inoue" <yosin@chromium.org> via sheriffbot.

Patch by Sheriff Bot <webkit.review.bot@gmail.com> on 2013-01-21

* DEPS:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@140380 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years ago[chromium] REGRESSION(r139347) roll chromium deps broke webkit-unit-tests
noel.gordon@gmail.com [Tue, 22 Jan 2013 02:41:42 +0000 (02:41 +0000)]
[chromium] REGRESSION(r139347) roll chromium deps broke webkit-unit-tests
https://bugs.webkit.org/show_bug.cgi?id=106631

Reviewed by Kent Tamura.

Fixed in http://crrev.com/177931 and rolled. Renable these tests.

* tests/WebImageTest.cpp:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@140379 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years ago[chromium] Update webaudio/realtimeanalyser-fft-sizing.html expectation on Win
noel.gordon@gmail.com [Tue, 22 Jan 2013 02:37:16 +0000 (02:37 +0000)]
[chromium] Update webaudio/realtimeanalyser-fft-sizing.html expectation on Win

Unreviewed gardening.

* platform/chromium-win-xp/webaudio/realtimeanalyser-fft-sizing-expected.txt: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@140378 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoAdd build flag for Canvas's Path object (disabled by default)
commit-queue@webkit.org [Tue, 22 Jan 2013 02:26:20 +0000 (02:26 +0000)]
Add build flag for Canvas's Path object (disabled by default)
https://bugs.webkit.org/show_bug.cgi?id=107473

Patch by Dirk Schulze <dschulze@adobe.com> on 2013-01-21
Reviewed by Dean Jackson.

.:

Add CANVAS_PATH build flag to build systems.

* Source/cmake/WebKitFeatures.cmake:
* Source/cmakeconfig.h.cmake:
* configure.ac:

Source/JavaScriptCore:

Add CANVAS_PATH build flag to build systems.

* Configurations/FeatureDefines.xcconfig:

Source/WebCore:

Add CANVAS_PATH build flag to build systems.

* Configurations/FeatureDefines.xcconfig:
* GNUmakefile.am:

Source/WebKit/chromium:

Add CANVAS_PATH build flag to build systems.

* features.gypi:

Source/WebKit/mac:

* Configurations/FeatureDefines.xcconfig:

Source/WebKit2:

Add CANVAS_PATH build flag to build systems.

* Configurations/FeatureDefines.xcconfig:

Tools:

Add CANVAS_PATH build flag to build systems.

* Scripts/webkitperl/FeatureList.pm:

WebKitLibraries:

Add CANVAS_PATH build flag to build systems.

* win/tools/vsprops/FeatureDefines.vsprops:
* win/tools/vsprops/FeatureDefinesCairo.vsprops:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@140377 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years ago[EFL][WK2] Add APIs to set/get view source mode
commit-queue@webkit.org [Tue, 22 Jan 2013 01:52:48 +0000 (01:52 +0000)]
[EFL][WK2] Add APIs to set/get view source mode
https://bugs.webkit.org/show_bug.cgi?id=106633

Patch by Jinwoo Song <jinwoo7.song@samsung.com> on 2013-01-21
Reviewed by Gyuyoung Kim.

Source/WebKit2:

Added APIs to set/get view source mode for enabling to load
the source code of the web page.

* UIProcess/API/efl/ewk_view.cpp:
(ewk_view_source_mode_set):
(ewk_view_source_mode_get):
* UIProcess/API/efl/ewk_view.h:
* UIProcess/API/efl/tests/test_ewk2_view.cpp:
(TEST_F):
* UIProcess/WebPageProxy.cpp:
(WebKit::WebPageProxy::WebPageProxy):
(WebKit::WebPageProxy::setMainFrameInViewSourceMode):
* UIProcess/WebPageProxy.h:
(WebKit::WebPageProxy::mainFrameInViewSourceMode):
(WebPageProxy):

Tools:

Added a shotcut 'F8' in Minibrowser to display a source code
of the web page in a new window.

       * MiniBrowser/efl/main.c:
(on_key_down):
(on_window_create):
(window_create):
(elm_main):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@140376 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years ago[chromium] Don't archive generated source files.
thakis@chromium.org [Tue, 22 Jan 2013 01:44:42 +0000 (01:44 +0000)]
[chromium] Don't archive generated source files.
https://bugs.webkit.org/show_bug.cgi?id=106736

Reviewed by Eric Seidel.

They shouldn't be needed to run tests. They don't need a lot of space,
but add many lines to the "extracting" stdout. This attempts to reduce
that noise a bit.

* BuildSlaveSupport/built-product-archive:
(archiveBuiltProduct):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@140375 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoUnreviewed gardening. Updating expectations for Mac platforms after r140370.
leviw@chromium.org [Tue, 22 Jan 2013 01:39:17 +0000 (01:39 +0000)]
Unreviewed gardening. Updating expectations for Mac platforms after r140370.

* platform/chromium-mac-snowleopard/platform/chromium/fast/events: Added.
* platform/chromium-mac-snowleopard/platform/chromium/fast/events/touch: Added.
* platform/chromium-mac-snowleopard/platform/chromium/fast/events/touch/touch-hit-rects-in-iframe-expected.txt: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@140374 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years ago[chromium] Unreviewed, update baselines after r140358. See also bug 107476.
thakis@chromium.org [Tue, 22 Jan 2013 01:04:37 +0000 (01:04 +0000)]
[chromium] Unreviewed, update baselines after r140358. See also bug 107476.
https://bugs.webkit.org/show_bug.cgi?id=107483

* platform/chromium-mac/fast/block/float/024-expected.txt:
* platform/chromium-mac/fast/block/margin-collapse/empty-clear-blocks-expected.txt:
* platform/chromium/fast/block/float/024-expected.txt: Renamed from LayoutTests/platform/chromium-win/fast/block/float/024-expected.txt.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@140373 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years ago[EFL][WK2] Implement WebInspector::localizedStringsURL() on EFL
commit-queue@webkit.org [Tue, 22 Jan 2013 01:04:34 +0000 (01:04 +0000)]
[EFL][WK2] Implement WebInspector::localizedStringsURL() on EFL
https://bugs.webkit.org/show_bug.cgi?id=107363

Patch by Sudarsana Nagineni <sudarsana.nagineni@intel.com> on 2013-01-21
Reviewed by Benjamin Poulain.

Implement WebInspector::localizedStringsURL() method to return the
file URL of the localizedStrings.js.

This prevents printing warnings about "Localized string not found" in
the console and fixes Web inspector related failing tests on the bots.

* WebProcess/WebPage/efl/WebInspectorEfl.cpp:
(WebKit::WebInspector::localizedStringsURL):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@140372 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoCSS: Make tag sub-selectors standalone CSSSelectors.
akling@apple.com [Tue, 22 Jan 2013 00:45:14 +0000 (00:45 +0000)]
CSS: Make tag sub-selectors standalone CSSSelectors.
<http://webkit.org/b/107111>

Reviewed by Antti Koivisto.

TL;DR: Instead of storing a QualifiedName with every CSSSelector, special-case tag selectors
       by promoting them to stand-alone CSSSelectors.
       33% reduction in CSS selector memory usage, 4.77 MB progression on Membuster3.
       Fixed some bugs in Shadow DOM that were exposed by these changes.

A selector like this:

    div.tripp.trapp#trull { }

Would previously be represented by a chain of 3 CSSSelector objects like so:

    [ Tag: "div",  Type: Class,  Value: "tripp" ]
    [ Tag:     *,  Type: Class,  Value: "trapp" ]
    [ Tag:     *,  Type: ID,     Value: "trull" ]

After this change, the memory layout becomes:

    [ Type: Tag,    Value: "div"   ]
    [ Type: Class,  Value: "tripp" ]
    [ Type: Class,  Value: "trapp" ]
    [ Type: ID,     Value: "trull" ]

This is a huge net memory win since the majority of selectors don't even have a tag name
and those that do now have a flat cost of one more CSSSelector.
Traversal is also slightly cleaner since any tag name will now be in a predictable place.

* css/SelectorChecker.cpp:
(WebCore::isFastCheckableMatch):
* html/shadow/HTMLContentElement.cpp:
(WebCore::validateSubSelector):

    Renamed CSSSelector::None to CSSSelector::Tag.

* css/SelectorChecker.h:
(WebCore::SelectorChecker::tagMatches):
* css/StyleResolver.cpp:
(WebCore::StyleResolver::ruleMatches):

    Changed SelectorChecker::tagMatches() to take a QualifiedName instead of a CSSSelector.

* dom/QualifiedName.cpp:
(WebCore::QualifiedName::deref):
(WebCore::QualifiedName::QualifiedNameImpl::~QualifiedNameImpl):
* dom/QualifiedName.h:
(QualifiedNameImpl):

    Give QualifiedNameImpl a destructor so it can remove itself from the global cache instead of
    having QualifiedName do it. This makes it possible to participate in ownership of QualifiedNames
    via a QualifiedNameImpl pointer, as used by the union member in CSSSelector.

* css/CSSGrammar.y.in:
* css/CSSParser.cpp:
(WebCore::CSSParser::createFloatingSelectorWithTagName):
(WebCore::CSSParser::updateSpecifiersWithNamespaceIfNeeded):
(WebCore::CSSParser::updateSpecifiersWithElementName):
* css/CSSParser.h:
* css/CSSParserValues.h:
(CSSParserSelector):
* css/CSSParserValues.cpp:
(WebCore::CSSParserSelector::CSSParserSelector):
(WebCore::CSSParserSelector::prependTagSelector):

    Make the CSS parser slap a CSSSelector with m_match=Tag at the head of all selectors that match
    one or more of these criteria:

    - The selector starts with a tag, e.g "div.foo" or just plain "span"
    - There is a @namespace rule in effect, and the override namespace needs to be stored with the selector.
      These will have CSSSelectors::m_isTagForNamespaceRule set, this is only so that selector serialization
      can avoid outputting a '*' tag where we previously didn't.

(WebCore::CSSParserSelector::isSimple):

    Merged the CSSSelector::isSimple() logic into CSSParserSelector, since it's only needed during parsing
    to figure out if a selector is allowed within :not().

* css/CSSSelector.h:
(CSSSelector):
(WebCore::CSSSelector::setValue):
(WebCore::CSSSelector::CSSSelector):
(WebCore::CSSSelector::~CSSSelector):
(WebCore::CSSSelector::tagQName):
* css/CSSSelector.cpp:
(WebCore::CSSSelector::createRareData):
(WebCore::CSSSelector::operator==):

    Add a QualifiedNameImpl* m_tagQName member to the CSSSelector data union. This union pointer is used
    if m_match == Tag. tagQName() is used to retrieve the tag (renamed from tag().)

(WebCore::CSSSelector::selectorText):

    Only serialize Tag selector components that aren't namespace placeholders. This behavior is web-facing
    so we make an effort to stay consistent.

(WebCore::CSSSelector::specificityForOneSelector):
(WebCore::CSSSelector::specificityForPage):
* css/CSSSelectorList.cpp:
(WebCore::SelectorNeedsNamespaceResolutionFunctor::operator()):
* css/RuleSet.cpp:
(WebCore::isSelectorMatchingHTMLBasedOnRuleHash):
* css/SelectorFilter.cpp:
(WebCore::collectDescendantSelectorIdentifierHashes):
(WebCore::SelectorFilter::collectIdentifierHashes):

    Adapt algorithms to having Tag selectors.

(WebCore::selectorListContainsUncommonAttributeSelector):

    Loop through all selector components when looking for uncommon attributes.

(WebCore::determinePropertyWhitelistType):

    Loop through all selector components when looking for ::cue().

(WebCore::RuleSet::addRule):
(WebCore::RuleSet::findBestRuleSetAndAdd):

    Break addRule into two methods to be able to peek ahead if the first selector is a Tag.
    Otherwise we'd end up sticking most selectors in m_tagRules, breaking the class/ID/etc optimizations.

* css/SelectorChecker.cpp:
(WebCore::SelectorChecker::fastCheck):
(WebCore::SelectorChecker::fastCheckSingleSelector):

    It's no longer necessary to check the tagQName for every selector component, so don't.
    Also simplified the signature of the template argument function.

(WebCore::SelectorChecker::fastCheckRightmostSelector):

    Updated for Tag selectors.

(WebCore::SelectorChecker::checkOne):
(WebCore::SelectorChecker::matches):
(WebCore::SelectorChecker::match):

    Have match() take a SiblingTraversalStrategy so it can forward that to checkOne().
    This is necessary for shadow DOM code that was incorrectly calling checkOne() instead of match().

* css/StyleResolver.cpp:
(WebCore::StyleResolver::ruleMatches):

    The meaning of "single-part selector" changes a bit with this patch, and no longer includes "div.foo"
    as that is now a Tag, followed by a Class. Given that, we can't assume the tag check is unnecessary
    just because the rightmost descendant was found in one of the hashes.

(WebCore::StyleResolver::matchPageRulesForList):

    Loop through all selector components when matching @page since pseudo types may not always be in
    the first subselector now.

* html/shadow/ContentDistributor.cpp:
(WebCore::ContentDistributor::collectSelectFeatureSetFrom):

    Collect feature information from subselectors to make sure nothing is missed. (This bug was exposed
    by offsetting the subselectors.)

* html/shadow/ContentSelectorQuery.cpp:
(WebCore::ContentSelectorChecker::checkContentSelector):

    Call SelectorChecker::match() instead of checkOne() to make subselector traversal work properly.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@140371 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoEvent target rects on the top level document shouldn't be clipped.
leviw@chromium.org [Tue, 22 Jan 2013 00:31:57 +0000 (00:31 +0000)]
Event target rects on the top level document shouldn't be clipped.
https://bugs.webkit.org/show_bug.cgi?id=107339

Reviewed by James Robinson.

Source/WebCore:

clippedOverflowRectForRepaint clips the top-level RenderView to the viewport, which
is wrong for generating event target rects, as the result will not extend to the bounds
of the document on pages that scroll. Changing the top-level view to use documentRect
instead.

Tests updated to cover bug: platform/chromium/fast/events/touch/touch-hit-rects-in-iframe.html
                            platform/chromium/fast/events/touch/compositor-touch-hit-rects.html

* page/scrolling/ScrollingCoordinator.cpp:
(WebCore::accumulateRendererTouchEventTargetRects): Use converToRootView instead of
a loop around convertToContaining view. This is not a change in behavior.
(WebCore::accumulateDocumentEventTargetRects): Switch to use documentRect instead of
clippedOverflowRectForRepaint for the top-level Document, and use converToRootView
to put rects in the coordinates of the top-level document.

LayoutTests:

Updating existing tests to cover this issue.

* platform/chromium-linux/platform/chromium/fast/events/touch/compositor-touch-hit-rects-expected.txt:
* platform/chromium/fast/events/touch/compositor-touch-hit-rects.html:
* platform/chromium/fast/events/touch/touch-hit-rects-in-iframe-expected.txt:
* platform/chromium/fast/events/touch/touch-hit-rects-in-iframe.html:
* platform/chromium/fast/events/touch/resources/frame-with-document-touch-handler.html: Added.
* platform/chromium/TestExpectations:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@140370 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoUnreviewed. Rolled Chromium DEPS to r177956. Requested by
commit-queue@webkit.org [Tue, 22 Jan 2013 00:27:39 +0000 (00:27 +0000)]
Unreviewed.  Rolled Chromium DEPS to r177956.  Requested by
thakis_ via sheriffbot.

Patch by Sheriff Bot <webkit.review.bot@gmail.com> on 2013-01-21

* DEPS:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@140369 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years ago[chromium] Update expectations.
thakis@chromium.org [Tue, 22 Jan 2013 00:03:30 +0000 (00:03 +0000)]
[chromium] Update expectations.
https://bugs.webkit.org/show_bug.cgi?id=107481

Unreviewed, expectations update.

After https://src.chromium.org/viewvc/chrome?view=rev&revision=177952,
these tests no longer fail on linux.

* platform/chromium/TestExpectations:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@140368 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoEnable process suppression by default on Mac
commit-queue@webkit.org [Mon, 21 Jan 2013 22:45:29 +0000 (22:45 +0000)]
Enable process suppression by default on Mac
https://bugs.webkit.org/show_bug.cgi?id=106804

Patch by Kiran Muppala <cmuppala@apple.com> on 2013-01-21
Reviewed by Alexey Proskuryakov.

* UIProcess/WebContext.cpp:
(WebKit::WebContext::WebContext): Change initialized value of
m_processSuppressionEnabled from false to true.
* UIProcess/mac/WebContextMac.mm:
(WebKit::WebContext::platformInitialize): Since process suppression
is no longer disabled on construction, do not disable it for global
child processes either.  Instead enable occlusion notifications as
they are needed to trigger process suppression.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@140367 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoAdd a USE() macro for content filtering code
aestes@apple.com [Mon, 21 Jan 2013 22:43:56 +0000 (22:43 +0000)]
Add a USE() macro for content filtering code
https://bugs.webkit.org/show_bug.cgi?id=107098

Reviewed by Mark Rowe.

Source/WebCore:

* WebCore.exp.in: Use USE(CONTENT_FILTERING).
* loader/MainResourceLoader.cpp:
(WebCore::MainResourceLoader::MainResourceLoader): Ditto.
(WebCore::MainResourceLoader::~MainResourceLoader): Ditto.
(WebCore::MainResourceLoader::cancel): Ditto.
(WebCore::MainResourceLoader::responseReceived): Ditto.
(WebCore::MainResourceLoader::dataReceived): Ditto.
(WebCore::MainResourceLoader::didFinishLoading): Ditto.
(WebCore::MainResourceLoader::notifyFinished): Ditto.
* loader/MainResourceLoader.h:
(MainResourceLoader): Ditto.
* platform/mac/WebCoreSystemInterface.h: Ditto.
* platform/mac/WebCoreSystemInterface.mm: Ditto.

Source/WebKit/mac:

* WebCoreSupport/WebSystemInterface.mm:
(InitWebCoreSystemInterface): Use USE(CONTENT_FILTERING).

Source/WebKit2:

* WebProcess/WebCoreSupport/mac/WebSystemInterface.mm:
(InitWebCoreSystemInterface): Use USE(CONTENT_FILTERING).

Source/WTF:

* wtf/Platform.h: Define WTF_USE_CONTENT_FILTERING on Mac platforms
more recent than Lion.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@140366 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years ago[CSS Exclusions] shape-outside on floats for rectangle shapes positioning
commit-queue@webkit.org [Mon, 21 Jan 2013 22:35:28 +0000 (22:35 +0000)]
[CSS Exclusions] shape-outside on floats for rectangle shapes positioning
https://bugs.webkit.org/show_bug.cgi?id=100399

Patch by Bem Jones-Bey <bjonesbe@adobe.com> on 2013-01-21
Reviewed by David Hyatt.

Source/WebCore:

When a float has a shape-outside with non-zero x or y, the float's
exclusion behavior can happen at a completely different location than
where the float's content is painted. This patch brings initial
support for this behavior.

Test: fast/exclusions/shape-outside-floats/shape-outside-floats-positioning.html

* rendering/RenderBlock.cpp:
(WebCore::RenderBlock::flipFloatForWritingModeForChild): If called
    during painting, make sure to pass that state down to the float
    positioning so that it can use the proper position.
(WebCore::RenderBlock::paintFloats): Since this is concerned with
    painting, use the position of the float, not the shape.
(WebCore::RenderBlock::positionNewFloats): Set the correct location
    for the float, relative to the location of the shape.
(WebCore::RenderBlock::addOverhangingFloats): Add FIXME.
* rendering/RenderBlock.h:
(WebCore::RenderBlock::xPositionForFloatIncludingMargin):
(WebCore::RenderBlock::yPositionForFloatIncludingMargin): Since the
    position of the shape (FloatingObject) can be different from the
    position of the float's box (RenderBox), the position needs to be
    computed differently for painting than for determining layout.
(WebCore::RenderBlock::flipFloatForWritingModeForChild): Change
    signature to take painting flag.

LayoutTests:

* fast/exclusions/shape-outside-floats/shape-outside-floats-positioning-expected.html: Added.
* fast/exclusions/shape-outside-floats/shape-outside-floats-positioning.html: Added.
* fast/exclusions/shape-outside-floats/shape-outside-floats-simple-rectangle-horizontal-multiple.html:
    This test had incorrect coordinates for some of the shapes, causing them to break now that x and y are supported for the shapes.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@140365 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoUnreviewed, suppress tests requiring rebaseline after r140358
robert@webkit.org [Mon, 21 Jan 2013 22:27:29 +0000 (22:27 +0000)]
Unreviewed, suppress tests requiring rebaseline after r140358
* platform/efl/TestExpectations:
* platform/gtk/TestExpectations:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@140364 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoUnreviewed LayoutTests gardening.
jer.noble@apple.com [Mon, 21 Jan 2013 22:21:42 +0000 (22:21 +0000)]
Unreviewed LayoutTests gardening.

Rebaseline fast/block/margin-collapse/empty-clear-blocks.html
and fast/block/float/024.html after r140358.

* platform/mac/fast/block/float/024-expected.txt:
* platform/mac/fast/block/margin-collapse/empty-clear-blocks-expected.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@140363 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoUnreviewed, suppress tests requiring rebaseline after r140358
robert@webkit.org [Mon, 21 Jan 2013 22:20:00 +0000 (22:20 +0000)]
Unreviewed, suppress tests requiring rebaseline after r140358
* platform/qt/TestExpectations:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@140362 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoMac DumpRenderTree's installSignalHandlers() is unreachable code
benjamin@webkit.org [Mon, 21 Jan 2013 22:12:56 +0000 (22:12 +0000)]
Mac DumpRenderTree's installSignalHandlers() is unreachable code
https://bugs.webkit.org/show_bug.cgi?id=107383

Reviewed by Sam Weinig.

The CrashHandler of DumpRenderTree were originally used for restoring
the color profile of the main display even if DRT crashes.

That feature was later limited to pixel tests only in r29999.

In r104351, the DRT was changed to use device color space. Making the crash
handlers useless, but they were left in the code since the error reporting
code was still reachable.

Later, r124581 refactored the code and dumpPixelsForCurrentTest
is being used before being assigned its value.
This did cause any regression because the code had become useless with the previous change.

* DumpRenderTree/mac/DumpRenderTree.mm:
Remove crashHandler() and installSignalHandlers() as they have become useless and
have been unreachable code for 6 months.
(dumpRenderTree):
Remove the code that was unreachable due to dumpPixelsForCurrentTest being always false
at those points.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@140361 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoMake ninja the default build system for build-webkit --chromium on mac
thakis@chromium.org [Mon, 21 Jan 2013 22:04:19 +0000 (22:04 +0000)]
Make ninja the default build system for build-webkit --chromium on mac
https://bugs.webkit.org/show_bug.cgi?id=106737

Reviewed by Eric Seidel.

* Scripts/update-webkit-chromium:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@140360 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoSupport non-WebVTT cues from in-band text tracks
eric.carlson@apple.com [Mon, 21 Jan 2013 22:02:19 +0000 (22:02 +0000)]
Support non-WebVTT cues from in-band text tracks
https://bugs.webkit.org/show_bug.cgi?id=107046

Reviewed by Sam Weinig.

* CMakeLists.txt: Add new files.
* GNUmakefile.list.am: Ditto.
* Target.pri: Ditto.
* WebCore.gypi: Ditto.
* WebCore.vcproj/WebCore.vcproj: Ditto.
* WebCore.xcodeproj/project.pbxproj: Ditto.

* html/shadow/MediaControlElements.cpp:
(WebCore::MediaControlTextTrackContainerElement::updateDisplay): Pass video size to getDisplayTree.

* html/track/InbandTextTrack.cpp:
(WebCore::InbandTextTrack::addGenericCue): New, create a generic cue if it doesn't already exist.
(WebCore::InbandTextTrack::addWebVTTCue): Renamed from addCue.
* html/track/InbandTextTrack.h:

* html/track/TextTrack.cpp:
(WebCore::TextTrack::hasCue): Moved from InbandTextTrack.cpp, use operator == instead of inline
    comparison so derived TextTrackCues can be compared.
* html/track/TextTrack.h:

* html/track/TextTrackCue.cpp:
(WebCore::TextTrackCueBox::applyCSSProperties): Interface change.
(WebCore::TextTrackCue::TextTrackCue): Don't create the display tree in the constructor, it may
    never be needed.
(WebCore::TextTrackCue::createDisplayTree): New, create the display tree.
(WebCore::TextTrackCue::displayTreeInternal): Display tree accessor, allows it to be created lazily.
(WebCore::TextTrackCue::setAlign): Alignment -> CueAlignment.
(WebCore::TextTrackCue::setIsActive): Use displayTreeInternal().
(WebCore::TextTrackCue::getDisplayTree): Use displayTreeInternal(), pass video size to applyCSSProperties.
(WebCore::TextTrackCue::removeDisplayTree): Use displayTreeInternal().
(WebCore::TextTrackCue::operator==): New.
* html/track/TextTrackCue.h:
(WebCore::TextTrackCue::getAlignment): Alignment -> CueAlignment.
(WebCore::TextTrackCue::operator!=): New.
(WebCore::TextTrackCue::cueType): New, cue type identified needed by operator==.
(WebCore::TextTrackCue::ownerDocument): New.

* html/track/TextTrackCueGeneric.cpp: Added.
(WebCore::TextTrackCueGenericBox::create):
(WebCore::TextTrackCueGenericBox::TextTrackCueGenericBox):
(WebCore::TextTrackCueGenericBox::applyCSSProperties):
(WebCore:::TextTrackCue):
(WebCore::TextTrackCueGeneric::createDisplayTree):
(WebCore::TextTrackCueGeneric::operator==):
* html/track/TextTrackCueGeneric.h: Added.
(WebCore::TextTrackCueGeneric::create):
(WebCore::TextTrackCueGeneric::~TextTrackCueGeneric):
(WebCore::TextTrackCueGeneric::baseFontSizeRelativeToVideoHeight):
(WebCore::TextTrackCueGeneric::setBaseFontSizeRelativeToVideoHeight):
(WebCore::TextTrackCueGeneric::fontSizeMultiplier):
(WebCore::TextTrackCueGeneric::setFontSizeMultiplier):
(WebCore::TextTrackCueGeneric::fontName):
(WebCore::TextTrackCueGeneric::setFontName):
(WebCore::TextTrackCueGeneric::operator!=):
(WebCore::TextTrackCueGeneric::cueType):

* platform/graphics/InbandTextTrackPrivateClient.h:
* platform/graphics/avfoundation/InbandTextTrackPrivateAVF.cpp:
(WebCore::InbandTextTrackPrivateAVF::processCueAttributes): Copy cue attributes into GenericCueData
    struct instead of trying to convert to WebVTT settings string. Process font size,font base
    size, font name, and vertical layout attributes,
(WebCore::InbandTextTrackPrivateAVF::processCue): Create separate cues for each attributed
    string in the array because each one can be at a different screen location.
(WebCore::InbandTextTrackPrivateAVF::resetCueValues):
* platform/graphics/avfoundation/InbandTextTrackPrivateAVF.h:
(InbandTextTrackPrivateAVF):

* platform/graphics/avfoundation/MediaPlayerPrivateAVFoundation.cpp:
(WebCore::MediaPlayerPrivateAVFoundation::seekCompleted): Reset any partially accumulated cues
    delivered since the seek was started.
(WebCore::MediaPlayerPrivateAVFoundation::addGenericCue): Renamed from flushCurrentCue.
* platform/graphics/avfoundation/MediaPlayerPrivateAVFoundation.h:

* platform/graphics/avfoundation/objc/MediaPlayerPrivateAVFoundationObjC.mm:
(WebCore::MediaPlayerPrivateAVFoundationObjC::createAVPlayer): Whitespace cleanup.
(WebCore::MediaPlayerPrivateAVFoundationObjC::createAVPlayerItem): Ditto.
(WebCore::MediaPlayerPrivateAVFoundationObjC::tracksChanged): Tell legible output to deliver cues
    without any additional styling needed by the internal renderer.
(WebCore::MediaPlayerPrivateAVFoundationObjC::processTextTracks):
(WebCore::MediaPlayerPrivateAVFoundationObjC::setCurrentTrack): Do nothing if asked to change
    to the track that is already current.
(-[WebCoreAVFMovieObserver legibleOutput:didOutputAttributedStrings:nativeSampleBuffers:forItemTime:]):
    Check to make sure that the client callback isn't NULL.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@140359 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoMisaligned logo on www.nzherald.co.nz possibly due to negative margin-top
robert@webkit.org [Mon, 21 Jan 2013 21:58:23 +0000 (21:58 +0000)]
Misaligned logo on nzherald.co.nz possibly due to negative margin-top
https://bugs.webkit.org/show_bug.cgi?id=14664

Reviewed by David Hyatt.

Source/WebCore:

Test: fast/block/float/clear-negative-margin-top.html

* rendering/RenderBlock.cpp:
(WebCore::RenderBlock::collapseMargins): If a negative margin pulls a block up so that floats from
siblings overhang, then ensure addOverHangingFloats() uses the revised logical top position when
deciding whether a float does indeed overhang into the block after margin-collapsing.
(WebCore::RenderBlock::clearFloatsIfNeeded):

LayoutTests:

* fast/block/float/clear-negative-margin-top-expected.html: Added.
* fast/block/float/clear-negative-margin-top.html: Added.
* platform/chromium-win/fast/block/float/024-expected.txt:
* platform/chromium-win/fast/block/margin-collapse/empty-clear-blocks-expected.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@140358 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoUnreviewed GTK gardening.
zandobersek@gmail.com [Mon, 21 Jan 2013 21:26:58 +0000 (21:26 +0000)]
Unreviewed GTK gardening.

Rebaselining a few media tests after r140339.

* platform/gtk/fast/hidpi/video-controls-in-hidpi-expected.txt:
* platform/gtk/fast/layers/video-layer-expected.txt:
* platform/gtk/media/media-controls-clone-expected.txt:
* platform/gtk/media/video-empty-source-expected.txt:
* platform/gtk/media/video-no-audio-expected.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@140355 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoUnreviewed. Rolled Chromium DEPS to r177953. Requested by
commit-queue@webkit.org [Mon, 21 Jan 2013 21:07:07 +0000 (21:07 +0000)]
Unreviewed.  Rolled Chromium DEPS to r177953.  Requested by
thakis_ via sheriffbot.

Patch by Sheriff Bot <webkit.review.bot@gmail.com> on 2013-01-21

* DEPS:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@140353 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoProposal: Add support for even-odd fill and clip to Canvas
commit-queue@webkit.org [Mon, 21 Jan 2013 21:05:00 +0000 (21:05 +0000)]
Proposal: Add support for even-odd fill and clip to Canvas
https://bugs.webkit.org/show_bug.cgi?id=106188

Patch by Rik Cabanier <cabanier@adobe.com> on 2013-01-21
Reviewed by Dirk Schulze.

Source/WebCore:

This patch adds support for an optional winding rule to fill, clip
and isPointInPath in canvas 2d. By default, the fill is non-zero but
you can now pass in an optional winding rule string ("nonzero" or "evenodd")
to make an explicit choice.

Tests: fast/canvas/canvas-clip-rule.html
       fast/canvas/canvas-fill-rule.html
       fast/canvas/canvas-isPointInPath-winding.html

* html/canvas/CanvasRenderingContext2D.cpp: Change implementation so it sets the windig rule in the context.
(WebCore::parseWinding):
(WebCore):
(WebCore::CanvasRenderingContext2D::fill):
(WebCore::CanvasRenderingContext2D::clip):
(WebCore::CanvasRenderingContext2D::isPointInPath):
* html/canvas/CanvasRenderingContext2D.h:
(CanvasRenderingContext2D):
* html/canvas/CanvasRenderingContext2D.idl: Change the idl so the new API is exposed to JavaScript.
* platform/graphics/cg/GraphicsContextCG.cpp: Fixes a bug where the wrong winding rule was set for clipping in Core Graphics.
(WebCore::GraphicsContext::clip):

LayoutTests:

Add tests to verify that the winding rule work as expected with clip, fill and isPointInPath

* fast/canvas/canvas-clip-rule-expected.txt: Added.
* fast/canvas/canvas-clip-rule.html: Added.
* fast/canvas/canvas-fill-rule-expected.txt: Added.
* fast/canvas/canvas-fill-rule.html: Added.
* fast/canvas/canvas-isPointInPath-winding-expected.txt: Added.
* fast/canvas/canvas-isPointInPath-winding.html: Added.
* fast/canvas/script-tests/canvas-clip-rule.js: Added.
(pixelDataAtPoint):
(checkResult):
(prepareTestScenario):
* fast/canvas/script-tests/canvas-fill-rule.js: Added.
(pixelDataAtPoint):
(checkResult):
(prepareTestScenario):
* fast/canvas/script-tests/canvas-isPointInPath-winding.js: Added.
(prepareTestScenario):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@140352 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoDon't null out the IPC connection in ChildProcess:terminate
andersca@apple.com [Mon, 21 Jan 2013 20:51:19 +0000 (20:51 +0000)]
Don't null out the IPC connection in ChildProcess:terminate
https://bugs.webkit.org/show_bug.cgi?id=107469
<rdar://problem/13015294>

Reviewed by Dan Bernstein.

We already invalidate the connection, and Connection::sendMessage checks if a connection
is valid and bails if it isn't, so the only thing that nulling out the connection here does
is that it would force us to add random connection null checks everywhere.

* Shared/ChildProcess.cpp:
(WebKit::ChildProcess::terminate):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@140351 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years ago[GTK][AC] setOpacity doesn't work for GraphicsLayerClutter
commit-queue@webkit.org [Mon, 21 Jan 2013 20:22:47 +0000 (20:22 +0000)]
[GTK][AC] setOpacity doesn't work for GraphicsLayerClutter
https://bugs.webkit.org/show_bug.cgi?id=105436

Patch by ChangSeok Oh <shivamidow@gmail.com> on 2013-01-21
Reviewed by Gustavo Noronha Silva.

Implement GraphicsLayerClutter::setOpacity based on Mac's GraphicsLayer implementation.

No new tests since this patch can be covered by already existing gtk ac tests.

* platform/graphics/clutter/GraphicsLayerClutter.cpp:
(WebCore::GraphicsLayerClutter::setOpacity):
(WebCore):
(WebCore::GraphicsLayerClutter::updateOpacityOnLayer):
(WebCore::GraphicsLayerClutter::commitLayerChangesBeforeSublayers):
* platform/graphics/clutter/GraphicsLayerClutter.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@140350 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years ago[BlackBerry] Skip spell checking on single character strings
nghanavatian@rim.com [Mon, 21 Jan 2013 19:46:48 +0000 (19:46 +0000)]
[BlackBerry] Skip spell checking on single character strings
https://bugs.webkit.org/show_bug.cgi?id=107463

Reviewed by Yong Li.

Check for a minimum string length when firing off spellcheck since we don't
want to mark single letter words.

Internally reviewed by Mike Fenton.

* WebKitSupport/InputHandler.cpp:
* WebKitSupport/SpellingHandler.cpp:
(BlackBerry::WebKit::SpellingHandler::createSpellCheckRequest):
(BlackBerry::WebKit::SpellingHandler::parseBlockForSpellChecking):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@140349 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years ago[EFL] API unit tests are running extremely slow on the bots
joone.hur@intel.com [Mon, 21 Jan 2013 19:31:56 +0000 (19:31 +0000)]
[EFL] API unit tests are running extremely slow on the bots
https://bugs.webkit.org/show_bug.cgi?id=104665

Reviewed by Gyuyoung Kim.

Clear HTTP cache files before running the unit tests, which prevents
performance degradation due to so many cache files.

* UIProcess/API/efl/tests/UnitTestUtils/EWK2UnitTestBase.cpp:
(EWK2UnitTest::EWK2UnitTestBase::SetUp):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@140348 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years ago[Texmap][GTK] The poster-circle doesn't appear.
commit-queue@webkit.org [Mon, 21 Jan 2013 19:16:29 +0000 (19:16 +0000)]
[Texmap][GTK] The poster-circle doesn't appear.
https://bugs.webkit.org/show_bug.cgi?id=106672

Patch by ChangSeok Oh <shivamidow@gmail.com> on 2013-01-21
Reviewed by Noam Rosenthal.

Source/WebCore:

We need to call setNeedsDisplay to redraw GraphicsLayer after we set drawsContent
a new value for the layer. Otherwise we can't get a chance for first drawing the layer.

Test: compositing/visibility/visibility-composited-animation.html

* platform/graphics/texmap/TextureMapperLayer.cpp:
(WebCore::TextureMapperLayer::flushCompositingStateForThisLayerOnly):

LayoutTests:

A new pixel test is added. It tests if animated layer is drawn for the first scene.

* compositing/visibility/visibility-composited-animation-expected.png: Added.
* compositing/visibility/visibility-composited-animation-expected.txt: Added.
* compositing/visibility/visibility-composited-animation.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@140347 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoAdd a test to ensure that content written into a seamless IFrame inherits style.
mkwst@chromium.org [Mon, 21 Jan 2013 19:03:19 +0000 (19:03 +0000)]
Add a test to ensure that content written into a seamless IFrame inherits style.
https://bugs.webkit.org/show_bug.cgi?id=99289

Reviewed by Eric Seidel.

This patch adds a single test to ensure that content written into a
seamless IFrame via 'contentDocument.write()' properly inherits style.
The root of the issue was fixed in wkbug.com/103539, so no code changes
are necesary, but that patch didn't test this particular edge case.

* fast/frames/seamless/seamless-document-write-expected.txt: Added.
* fast/frames/seamless/seamless-document-write.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@140346 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoRegression(r140262): Causes a lot of flakiness
christophe.dumez@intel.com [Mon, 21 Jan 2013 18:44:19 +0000 (18:44 +0000)]
Regression(r140262): Causes a lot of flakiness
https://bugs.webkit.org/show_bug.cgi?id=107454

Reviewed by Alexey Proskuryakov.

r140262 introduced a lot of flakiness due to fixed layout mode
not getting disabled on the view after being enabled for
specific test cases. This patch makes sure fixed layout mode
properly gets reset in WebKitTestRunner to avoid flakiness.

* WebKitTestRunner/TestInvocation.cpp:
(WTR::updateLayoutType):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@140344 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoConsolidate and run ARIA-list-and-listitem accessibility test
commit-queue@webkit.org [Mon, 21 Jan 2013 18:26:58 +0000 (18:26 +0000)]
Consolidate and run ARIA-list-and-listitem accessibility test
https://bugs.webkit.org/show_bug.cgi?id=107283

Patch by Brian Holt <brian.holt@samsung.com> on 2013-01-21
Reviewed by Chris Fleizach.

After revision r131674 for bug 99640, the output from
AccessibilityUIElement::role() is consistent across most platforms,
therefore platform specific expections can be consolidated into a
generic expectation with a Mac exception and enabled.

* accessibility/aria-list-and-listitem-expected.txt: Renamed from LayoutTests/platform/chromium/accessibility/aria-list-and-listitem-expected.txt.
* platform/gtk/TestExpectations: Removed passing test.
* platform/gtk/accessibility/aria-list-and-listitem-expected.txt: Removed.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@140343 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years ago[EFL][Qt][WebGl] Random crash in GraphicsContext3D::drawArrays
commit-queue@webkit.org [Mon, 21 Jan 2013 17:57:06 +0000 (17:57 +0000)]
[EFL][Qt][WebGl] Random crash in GraphicsContext3D::drawArrays
https://bugs.webkit.org/show_bug.cgi?id=107178

Patch by Viatcheslav Ostapenko <sl.ostapenko@samsung.com> on 2013-01-21
Reviewed by Noam Rosenthal.

Workaround for the problem in mesa when internal llvm pipe object is deleted
later than the screen object. Screen object is deleted because the corresponding
X server display connection closed.
Keep X server display connection open until program shutdown.
OffScreenRootWindow::display is now static, so there is no need to create
OffScreenRootWindow object on client side.

Fixes crash that appears during run of fast/canavs/webgl tests.

* platform/graphics/surfaces/glx/GraphicsSurfaceGLX.cpp:
(OffScreenRootWindow):
(WebCore::OffScreenRootWindow::DisplayConnection::DisplayConnection):
(DisplayConnection):
(WebCore::OffScreenRootWindow::DisplayConnection::~DisplayConnection):
(WebCore::OffScreenRootWindow::DisplayConnection::display):
(WebCore::OffScreenRootWindow::display):
(WebCore::OffScreenRootWindow::~OffScreenRootWindow):
(WebCore):
(WebCore::GraphicsSurfacePrivate::GraphicsSurfacePrivate):
(WebCore::GraphicsSurfacePrivate::createSurface):
(WebCore::GraphicsSurfacePrivate::createPixmap):
(WebCore::GraphicsSurfacePrivate::makeCurrent):
(WebCore::GraphicsSurfacePrivate::doneCurrent):
(WebCore::GraphicsSurfacePrivate::swapBuffers):
(WebCore::GraphicsSurfacePrivate::display):
(WebCore::GraphicsSurfacePrivate::size):
(WebCore::GraphicsSurfacePrivate::findFBConfigWithAlpha):
(WebCore::GraphicsSurfacePrivate::clear):
(GraphicsSurfacePrivate):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@140342 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoImplement the HTML <main> element.
commit-queue@webkit.org [Mon, 21 Jan 2013 17:10:34 +0000 (17:10 +0000)]
Implement the HTML <main> element.
https://bugs.webkit.org/show_bug.cgi?id=103172

Patch by Michael[tm] Smith <mike@w3.org> on 2013-01-21
Reviewed by Chris Fleizach.

Source/WebCore:

Authored by Steve Faulkner <faulkner.steve@gmail.com>

<main> should behave the same as <nav>, <section>, and <article>, and <aside>.
<main> should expose AXLandmarkMain correctly on each platform.

Tests: fast/html/main-element.html
       platform/chromium/accessibility/main-element.html
       platform/mac/accessibility/main-element.html

* accessibility/AccessibilityRenderObject.cpp:
(WebCore::AccessibilityRenderObject::determineAccessibilityRole):
* css/html.css:
(article, aside, footer, header, hgroup, main, nav, section):
* editing/FormatBlockCommand.cpp:
(WebCore::isElementForFormatBlock):
* html/HTMLTagNames.in:
* html/parser/HTMLStackItem.h:
(WebCore::HTMLStackItem::isSpecialNode):
* html/parser/HTMLTreeBuilder.cpp:
(WebCore::HTMLTreeBuilder::processStartTagForInBody):
(WebCore::HTMLTreeBuilder::processEndTagForInBody):

LayoutTests:

<main> should behave the same as <nav>, <section>, <article>, and <aside>.
<main> has no specific parsing rules.
<main> should expose AXLandmarkMain correctly for each platform.

* editing/execCommand/query-format-block-expected.txt:
* editing/execCommand/remove-format-elements-expected.txt:
* editing/execCommand/remove-format-elements.html:
* editing/execCommand/script-tests/query-format-block.js:
* fast/block/child-not-removed-from-parent-lineboxes-crash.html:
* fast/dom/click-method-on-html-element-expected.txt:
* fast/dom/click-method-on-html-element.html:
* fast/dom/wrapper-classes-expected.txt:
* fast/dom/wrapper-classes.html:
* fast/forms/access-key-for-all-elements-expected.txt:
* fast/forms/access-key-for-all-elements.html:
* fast/html/main-element-expected.txt: Added.
* fast/html/main-element.html: Added.
* fast/html/script-tests/main-element.js: Added.
(getWeight):
* platform/chromium/accessibility/main-element-expected.txt: Added.
* platform/chromium/accessibility/main-element.html: Added.
* platform/chromium/fast/dom/wrapper-classes-expected.txt:
* platform/mac/accessibility/main-element-expected.txt: Added.
* platform/mac/accessibility/main-element.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@140341 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoCrash in AccessibilityTableCell::parentTable()
commit-queue@webkit.org [Mon, 21 Jan 2013 16:57:38 +0000 (16:57 +0000)]
Crash in AccessibilityTableCell::parentTable()
https://bugs.webkit.org/show_bug.cgi?id=107261

Patch by Joanmarie Diggs <jdiggs@igalia.com> on 2013-01-21
Reviewed by Chris Fleizach.

Source/WebCore:

Test: accessibility/table-destroyed-crash.html

Getting the parent table in order to get the role value should not be
done when objects are being destroyed. Also, it does not seem safe to
assume we have an AXObjectCache.

Moving the logic from roleValue() to determineAccessibilityRole() has
the side effect of not being able to verify the cell is in an AXTable
when that AXTable has not yet been created. Therefore isTableCell()
should look to see if it is the descendant of an AXRow.

* accessibility/AccessibilityTableCell.cpp:
(WebCore::AccessibilityTableCell::parentTable):
(WebCore::AccessibilityTableCell::isTableCell):
(WebCore::AccessibilityTableCell::determineAccessibilityRole):
* accessibility/AccessibilityTableCell.h:
(AccessibilityTableCell):

LayoutTests:

Getting the parent table in order to get the role value should not be
done when objects are being destroyed. Also, it does not seem safe to
assume we have an AXObjectCache.

* accessibility/table-destroyed-crash-expected.txt: Added.
* accessibility/table-destroyed-crash.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@140340 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years ago[GTK] Volume button should not be shown for videos without audio
commit-queue@webkit.org [Mon, 21 Jan 2013 16:39:34 +0000 (16:39 +0000)]
[GTK] Volume button should not be shown for videos without audio
https://bugs.webkit.org/show_bug.cgi?id=106436

Patch by Halton Huo <halton.huo@intel.com> on 2013-01-21
Reviewed by Philippe Normand.

The volume control is no longer shown for videos with no audio.

No new tests, tested by media/video-no-audio.html.

* platform/gtk/RenderThemeGtk.cpp:
(WebCore::RenderThemeGtk::hasOwnDisabledStateHandlingFor):
(WebCore):
* platform/gtk/RenderThemeGtk.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@140339 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years ago[Soup] Work around a glib bug
danw@gnome.org [Mon, 21 Jan 2013 16:23:13 +0000 (16:23 +0000)]
[Soup] Work around a glib bug
https://bugs.webkit.org/show_bug.cgi?id=106789

Reviewed by Martin Robinson.

In glib <= 2.35.4, g_input_stream_skip_async() applied to a
libsoup response stream will do a synchronous skip() in another
thread, which libsoup isn't expecting and doesn't have proper
locking for. Work around this until the next time we bump the glib
requirement by using read_async() (and throwing away the result)
instead of skip_async().

No new tests; fixes a race condition in existing tests with older
glib

* platform/network/soup/ResourceHandleSoup.cpp:
(WebCore::redirectSkipCallback):
(WebCore::sendRequestCallback):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@140338 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoREGRESSION (r140285): GTK uses WKPageResourceLoadClient which was removed
kov@webkit.org [Mon, 21 Jan 2013 16:17:26 +0000 (16:17 +0000)]
REGRESSION (r140285): GTK uses WKPageResourceLoadClient which was removed
https://bugs.webkit.org/show_bug.cgi?id=107421

Source/WebKit2:

Unreviewed buildfix.

* GNUmakefile.list.am:
* UIProcess/API/gtk/WebKitResourceLoadClient.cpp: Removed.
* UIProcess/API/gtk/WebKitResourceLoadClient.h: Removed.
* UIProcess/API/gtk/WebKitWebView.cpp: No longer attach unexisting ResourceLoadClient.
(webkitWebViewConstructed):

Tools:

Unreviewed build fix.

* Scripts/run-gtk-tests:
(TestRunner): skip Resource tests.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@140337 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years ago[Gtk] Dispatching event list into input element.
commit-queue@webkit.org [Mon, 21 Jan 2013 16:14:34 +0000 (16:14 +0000)]
[Gtk] Dispatching event list into input element.
https://bugs.webkit.org/show_bug.cgi?id=107259

Patch by Oleg Smirnov <oleg.smirnov@lge.com> on 2013-01-21
Reviewed by Martin Robinson.

Clear pending command list after dispatching events.

* WebCoreSupport/EditorClientGtk.cpp:
(WebKit::EditorClient::handleKeyboardEvent):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@140336 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoWeb Inspector: data grid resize is slow
caseq@chromium.org [Mon, 21 Jan 2013 16:07:43 +0000 (16:07 +0000)]
Web Inspector: data grid resize is slow
https://bugs.webkit.org/show_bug.cgi?id=107291

Reviewed by Pavel Feldman.

- avoid (most of) unnecessary relayouts and style recalculation when dragging data grid resizer.

* inspector/front-end/DataGrid.js:
(WebInspector.DataGrid.prototype._positionResizers):
(WebInspector.DataGrid.prototype._resizerDragging):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@140335 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years ago[Qt] Unreviewed gardening. Skip failing test.
kadam@inf.u-szeged.hu [Mon, 21 Jan 2013 15:33:57 +0000 (15:33 +0000)]
[Qt] Unreviewed gardening. Skip failing test.
https://bugs.webkit.org/show_bug.cgi?id=105574.

* platform/qt/TestExpectations:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@140334 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years ago[Inspector] Layout Elements panel in a single column when docked right.
commit-queue@webkit.org [Mon, 21 Jan 2013 15:32:47 +0000 (15:32 +0000)]
[Inspector] Layout Elements panel in a single column when docked right.
https://bugs.webkit.org/show_bug.cgi?id=107129

Patch by Vladislav Kaznacheev <kaznacheev@chromium.org> on 2013-01-21
Reviewed by Pavel Feldman.

If the Inspector is docked right, the style panes in the Elements panel
are put below the DOM tree pane, not to the right. This behavior
is experimental (hidden behind a new flag "elementsPanelSingleColumn").

No new tests.

* inspector/front-end/DockController.js:
(WebInspector.DockController.prototype.dockSide):
(WebInspector.DockController.prototype._updateUI):
(WebInspector.DockController.prototype._toggleDockState):
* inspector/front-end/ElementsPanel.js:
(WebInspector.ElementsPanel.prototype.onResize):
(WebInspector.ElementsPanel.prototype._onDockStateChanged):
(WebInspector.ElementsPanel.prototype._sidebarPosition):
* inspector/front-end/Panel.js:
(WebInspector.Panel.prototype.createSidebarView):
* inspector/front-end/Settings.js:
(WebInspector.ExperimentsSettings):
* inspector/front-end/SidebarView.js:
(WebInspector.SidebarView):
(WebInspector.SidebarView.prototype.get mainElement):
(WebInspector.SidebarView.prototype.get sidebarElement):
(WebInspector.SidebarView.prototype._setSidebarElementStyle):
(WebInspector.SidebarView.prototype.setSidebarPosition):
(WebInspector.SidebarView.prototype._innerSetSidebarPosition):
(WebInspector.SidebarView.prototype.setMinimumSidebarHeight):
(WebInspector.SidebarView.prototype.setMinimumMainHeightPercent):
(WebInspector.SidebarView.prototype.applyConstraints):
(WebInspector.SidebarView.prototype.hideMainElement):
(WebInspector.SidebarView.prototype.hideSidebarElement):
* inspector/front-end/SplitView.js:
(WebInspector.SplitView):
* inspector/front-end/splitView.css:
(.split-view-sidebar-top):
(.split-view-sidebar-top.maximized):
(.split-view-sidebar-bottom):
(.split-view-sidebar-bottom.maximized):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@140333 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoWeb Inspector: do not serialize native heap graph when collecting information about...
yurys@chromium.org [Mon, 21 Jan 2013 15:30:26 +0000 (15:30 +0000)]
Web Inspector: do not serialize native heap graph when collecting information about memory distribution
https://bugs.webkit.org/show_bug.cgi?id=107450

Reviewed by Pavel Feldman.

Native heap graph is not reported when we need to show only high-level memory distribution.

* inspector/front-end/NativeMemorySnapshotView.js:
(WebInspector.NativeMemorySnapshotView):
(WebInspector.NativeMemoryProfileType.prototype.buttonClicked.didReceiveMemorySnapshot):
(WebInspector.NativeMemoryProfileType.prototype.buttonClicked):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@140332 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoWeb Inspector: [REGRESSION] Forced :visited pseudoclass has no effect on A elements
apavlov@chromium.org [Mon, 21 Jan 2013 14:57:19 +0000 (14:57 +0000)]
Web Inspector: [REGRESSION] Forced :visited pseudoclass has no effect on A elements
https://bugs.webkit.org/show_bug.cgi?id=107448

Reviewed by Antti Koivisto.

* css/StyleResolver.cpp:
(WebCore::StyleResolver::sortAndTransferMatchedRules): Remove totally bad code.
(WebCore::StyleResolver::styleForElement):
  Use InsideVisitedLink if PseudoVisited has been forced by Inspector.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@140331 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years ago[Qt][WK1] Permission request callbacks for non-legacy notifications
allan.jensen@digia.com [Mon, 21 Jan 2013 14:28:11 +0000 (14:28 +0000)]
[Qt][WK1] Permission request callbacks for non-legacy notifications
https://bugs.webkit.org/show_bug.cgi?id=107438

Reviewed by Jocelyn Turcotte.

Tracks the new callbacks the same way legacy callbacks are tracked.

No test affected yet as we are still lacking support for testing
feature permission callbacks.

* WebCoreSupport/NotificationPresenterClientQt.cpp:
(WebCore::NotificationPresenterClientQt::requestPermission):
(WebCore::NotificationPresenterClientQt::allowNotificationForFrame):
* WebCoreSupport/NotificationPresenterClientQt.h:
(NotificationPresenterClientQt):
(CallbacksInfo):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@140330 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoWeb Inspector: modify generate_protocol_externs.py to generate JSON typedef's for...
aandrey@chromium.org [Mon, 21 Jan 2013 14:16:07 +0000 (14:16 +0000)]
Web Inspector: modify generate_protocol_externs.py to generate JSON typedef's for @constructors
https://bugs.webkit.org/show_bug.cgi?id=107287

Reviewed by Pavel Feldman.

Source/WebCore:

Modify generate_protocol_externs.py to generate @typedefs instead of @constructors,

* inspector/InjectedScriptCanvasModuleSource.js:
(.):
* inspector/InjectedScriptSource.js:
* inspector/compile-front-end.py:
* inspector/generate_protocol_externs.py:
(param_type):

LayoutTests:

* inspector/console/command-line-api-expected.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@140329 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoUnreviewed, rolling out r139781.
commit-queue@webkit.org [Mon, 21 Jan 2013 14:11:14 +0000 (14:11 +0000)]
Unreviewed, rolling out r139781.
http://trac.webkit.org/changeset/139781
https://bugs.webkit.org/show_bug.cgi?id=107443

This patch seems to cause flakiness on Qt (Requested by chris-
qBT_work on #webkit).

Patch by Sheriff Bot <webkit.review.bot@gmail.com> on 2013-01-21

* WebProcess/WebPage/CoordinatedGraphics/CoordinatedLayerTreeHost.cpp:
(WebKit::CoordinatedLayerTreeHost::forceRepaintAsync):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@140328 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years ago[EFL][WKTR] platformRunUntil() should keep the main loop running
christophe.dumez@intel.com [Mon, 21 Jan 2013 13:55:36 +0000 (13:55 +0000)]
[EFL][WKTR] platformRunUntil() should keep the main loop running
https://bugs.webkit.org/show_bug.cgi?id=107442

Reviewed by Noam Rosenthal.

EFL's implementation of platformRunUntil() should keep the main loop
running and avoid calling sleep() between calls to
ecore_main_loop_iterate(). This call to sleep seems to be a source
of deadlocks (see Bug 106884).

* WebKitTestRunner/efl/TestControllerEfl.cpp:
(WTR::TestController::platformRunUntil):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@140327 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years ago[Qt] Update the documentation about the storage of inspector settings
jocelyn.turcotte@digia.com [Mon, 21 Jan 2013 13:40:06 +0000 (13:40 +0000)]
[Qt] Update the documentation about the storage of inspector settings
https://bugs.webkit.org/show_bug.cgi?id=106777

Reviewed by Allan Sandfeld Jensen.

Since r76770, QSettings isn't queried anymore to persist inspector
settings like whether or not the debugger is enabled by default.
It still seems to be used for the default attach height.

Update the documentation accordingly.

* WidgetApi/qwebinspector.cpp:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@140326 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years ago[Gtk] [Cairo] Memory leak when is WebView destroy.
commit-queue@webkit.org [Mon, 21 Jan 2013 13:34:11 +0000 (13:34 +0000)]
[Gtk] [Cairo] Memory leak when is WebView destroy.
https://bugs.webkit.org/show_bug.cgi?id=107112

Patch by Oleg Smirnov <oleg.smirnov@lge.com> on 2013-01-21
Reviewed by Martin Robinson.

Make createSurfaceForBackingStore return a smart pointer in all supported
build configurations to avoid a memory leak.

No new tests. There is memory leak only.

* platform/cairo/WidgetBackingStoreCairo.cpp:
(WebCore::createSurfaceForBackingStore):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@140325 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoINPUT_MULTIPLE_FIELDS_UI: should not dispatch 'input' events if the element value...
tkent@chromium.org [Mon, 21 Jan 2013 12:50:30 +0000 (12:50 +0000)]
INPUT_MULTIPLE_FIELDS_UI: should not dispatch 'input' events if the element value is not updated
https://bugs.webkit.org/show_bug.cgi?id=107429

Reviewed by Kentaro Hara.

Source/WebCore:

http://www.whatwg.org/specs/web-apps/current-work/multipage/common-input-element-attributes.html#common-event-behaviors
> any time the user causes the element's value to change, the user agent
> must queue a task to fire a simple event that bubbles named input at the
> input element.

Tests:
fast/forms/time-multiple-fields/time-multiple-fields-keyboard-event.html
is updated to cover this change.

* html/BaseMultipleFieldsDateAndTimeInputType.cpp:
(WebCore::BaseMultipleFieldsDateAndTimeInputType::editControlValueChanged):
If the new value is equivalent to the old value, don't dispatch events.
However we should recalculate validity and call notifyFormStateChanged
because input.validity.badInput state might be changed.

LayoutTests:

* fast/forms/time-multiple-fields/time-multiple-fields-keyboard-events-expected.txt:
* fast/forms/time-multiple-fields/time-multiple-fields-keyboard-events.html:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@140324 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoQtTestBrowser should provide way to clear selected elements
vivek.vg@samsung.com [Mon, 21 Jan 2013 12:45:24 +0000 (12:45 +0000)]
QtTestBrowser should provide way to clear selected elements
https://bugs.webkit.org/show_bug.cgi?id=107437

Reviewed by Simon Hausmann.

Provide a way to "Clear selection" in the "Develop" menu to clear the previous search highlight.
Invoke the clearSelection prior to highlighting new elements.

* QtTestBrowser/launcherwindow.cpp:
(ElementHighlight):
(LauncherWindow::createChrome):
(LauncherWindow::selectElements):
(LauncherWindow::clearSelection):
* QtTestBrowser/launcherwindow.h:
(LauncherWindow):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@140323 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years ago[Qt] Crash in gmail on enabling desktop notifications
allan.jensen@digia.com [Mon, 21 Jan 2013 12:33:40 +0000 (12:33 +0000)]
[Qt] Crash in gmail on enabling desktop notifications
https://bugs.webkit.org/show_bug.cgi?id=106699

Patch by David Rosca.
Reviewed by Jocelyn Turcotte.

Protect against callback that may be null. This also matches WebKit2 behaviour.

* WebCoreSupport/NotificationPresenterClientQt.cpp:
(WebCore::NotificationPresenterClientQt::allowNotificationForFrame):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@140322 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoWeb Inspector: Do not dispatch mousemove when emulating touch event and no touch...
apavlov@chromium.org [Mon, 21 Jan 2013 12:06:57 +0000 (12:06 +0000)]
Web Inspector: Do not dispatch mousemove when emulating touch event and no touch is active
https://bugs.webkit.org/show_bug.cgi?id=107419

Reviewed by Pavel Feldman.

Source/WebCore:

mousemove event should not be dispatched when emulating touch events and mouse button is not pressed

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

LayoutTests:

* fast/events/touch/emulate-touch-events.html: Test that mousemove is dispatched not more than once
  and reset the touch event emulation flag in window.internals.settings upon test completion.
* fast/events/touch/emulated-touch-iframe.html:
  Reset the touch event emulation flag in window.internals.settings upon test completion.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@140321 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoWeb Inspector: fix highlight bug in DTE.
commit-queue@webkit.org [Mon, 21 Jan 2013 11:54:00 +0000 (11:54 +0000)]
Web Inspector: fix highlight bug in DTE.
https://bugs.webkit.org/show_bug.cgi?id=107433

Patch by Andrey Lushnikov <lushnikov@chromium.org> on 2013-01-21
Reviewed by Pavel Feldman.

Source/WebCore:

Do not re-create |state.ranges| array when it already contains highlight.

Test: inpector/editor/highlighter-chunk-limit.txt

* inspector/front-end/TextEditorHighlighter.js:
(WebInspector.TextEditorHighlighter.prototype._highlightLines):

LayoutTests:

Add a layout test to cover highlight bug specific situation.

* inspector/editor/highlighter-chunk-limit-expected.txt: Added.
* inspector/editor/highlighter-chunk-limit.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@140320 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years ago[Qt] Unreviewed gardening. Skip some failing.
kadam@inf.u-szeged.hu [Mon, 21 Jan 2013 11:33:54 +0000 (11:33 +0000)]
[Qt] Unreviewed gardening. Skip some failing.

* platform/qt/TestExpectations:
* platform/qt/editing/input/caret-at-the-edge-of-contenteditable-expected.png: Update after r140202.
* platform/qt/editing/input/caret-at-the-edge-of-contenteditable-expected.txt: Update after r140202.
* platform/qt/editing/input/reveal-caret-of-multiline-contenteditable-expected.png: Update after r140202.
* platform/qt/editing/input/reveal-caret-of-multiline-contenteditable-expected.txt: Update after r140202.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@140319 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years ago[chromium] move methods interacting with willSendRequest to TestRunner
jochen@chromium.org [Mon, 21 Jan 2013 11:16:22 +0000 (11:16 +0000)]
[chromium] move methods interacting with willSendRequest to TestRunner
https://bugs.webkit.org/show_bug.cgi?id=107270

Reviewed by Kent Tamura.

* DumpRenderTree/chromium/DRTTestRunner.cpp:
(DRTTestRunner::DRTTestRunner):
(DRTTestRunner::waitForPolicyDelegate):
* DumpRenderTree/chromium/DRTTestRunner.h:
(DRTTestRunner):
* DumpRenderTree/chromium/TestRunner/public/WebTestRunner.h:
(WebTestRunner::WebTestRunner::shouldStayOnPageAfterHandlingBeforeUnload):
(WebTestRunner::WebTestRunner::httpHeadersToClear):
(WebTestRunner::WebTestRunner::shouldBlockRedirects):
(WebTestRunner::WebTestRunner::willSendRequestShouldReturnNull):
* DumpRenderTree/chromium/TestRunner/src/TestRunner.cpp:
(WebTestRunner::TestRunner::TestRunner):
(WebTestRunner::TestRunner::reset):
(WebTestRunner::TestRunner::httpHeadersToClear):
(WebTestRunner):
(WebTestRunner::TestRunner::shouldBlockRedirects):
(WebTestRunner::TestRunner::willSendRequestShouldReturnNull):
(WebTestRunner::TestRunner::setWillSendRequestClearHeader):
(WebTestRunner::TestRunner::setWillSendRequestReturnsNullOnRedirect):
(WebTestRunner::TestRunner::setWillSendRequestReturnsNull):
* DumpRenderTree/chromium/TestRunner/src/TestRunner.h:
(TestRunner):
* DumpRenderTree/chromium/TestRunner/src/WebTestProxy.cpp:
(WebTestRunner::WebTestProxyBase::willSendRequest):
* DumpRenderTree/chromium/WebViewHost.cpp:
(WebViewHost::willSendRequest):
(WebViewHost::reset):
* DumpRenderTree/chromium/WebViewHost.h:
(WebViewHost):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@140318 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years ago[GTK] Stop building WebKit2 on GTK EWSs
zandobersek@gmail.com [Mon, 21 Jan 2013 11:10:27 +0000 (11:10 +0000)]
[GTK] Stop building WebKit2 on GTK EWSs
https://bugs.webkit.org/show_bug.cgi?id=107435

Reviewed by Philippe Normand.

The WebKit2 build of the GTK is at the moment not efficiently maintainable,
so stop building it on GTK EWSs.

* Scripts/webkitpy/common/config/ports.py:
(GtkPort.build_webkit_command):
* Scripts/webkitpy/common/config/ports_unittest.py:
(DeprecatedPortTest.test_gtk_port):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@140317 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years ago[Refactoring] rename StyleRuleBlock -> StyleRuleGroup
tasak@google.com [Mon, 21 Jan 2013 10:33:19 +0000 (10:33 +0000)]
[Refactoring] rename StyleRuleBlock -> StyleRuleGroup
https://bugs.webkit.org/show_bug.cgi?id=107415

Reviewed by Antti Koivisto.

StyleRuleBlock should be renamed StyleRuleGroup, because
CSSGroupingRule (not CSSBlockRule) is an interface for an at-rule that
contains other rules nested inside itself:
http://www.w3.org/TR/2012/WD-css3-conditional-20121213/#the-cssgroupingrule-interface
To match the name: CSSGroupingRule, StyleRuleGroup looks better.

No new tests, because just refactoring.

* css/CSSGroupingRule.cpp:
(WebCore::CSSGroupingRule::CSSGroupingRule):
(WebCore::CSSGroupingRule::reattach):
* css/CSSGroupingRule.h:
(CSSGroupingRule):
* css/StyleRule.cpp:
(WebCore::StyleRuleGroup::StyleRuleGroup):
(WebCore::StyleRuleGroup::wrapperInsertRule):
(WebCore::StyleRuleGroup::wrapperRemoveRule):
(WebCore::StyleRuleGroup::reportDescendantMemoryUsage):
(WebCore::StyleRuleMedia::StyleRuleMedia):
(WebCore::StyleRuleSupports::StyleRuleSupports):
(WebCore::StyleRuleRegion::StyleRuleRegion):
* css/StyleRule.h:
(StyleRuleGroup):
(WebCore::StyleRuleHost::StyleRuleHost):
(WebCore::toStyleRuleMedia):
(WebCore::toStyleRuleSupports):
(WebCore::toStyleRuleRegion):
Just replaced StyleRuleBlock with StyleRuleGroup.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@140316 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoImplement MediaStreamEvent constructor
haraken@chromium.org [Mon, 21 Jan 2013 10:12:03 +0000 (10:12 +0000)]
Implement MediaStreamEvent constructor
https://bugs.webkit.org/show_bug.cgi?id=107411

Reviewed by Hajime Morita.

Fixed wrong test cases.

* fast/events/constructors/media-stream-event-constructor-expected.txt:
* fast/events/constructors/media-stream-event-constructor.html:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@140315 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoWeb Inspector: refactoring to use string as argument for highlightRegex API
commit-queue@webkit.org [Mon, 21 Jan 2013 09:53:19 +0000 (09:53 +0000)]
Web Inspector: refactoring to use string as argument for highlightRegex API
https://bugs.webkit.org/show_bug.cgi?id=107243

Patch by Andrey Lushnikov <lushnikov@chromium.org> on 2013-01-21
Reviewed by Pavel Feldman.

Source/WebCore:

Use string as argument for highlightRegex API instead of using regex object.

No new tests: no change in behaviour.

* inspector/front-end/DefaultTextEditor.js:
(WebInspector.TextEditorMainPanel.prototype.highlightRegex):
* inspector/front-end/TextEditor.js:

LayoutTests:

Correct test to correspond to refactoring changes.

* inspector/editor/text-editor-highlight-regexp.html:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@140314 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoUnreviewed, rolling out r140311.
commit-queue@webkit.org [Mon, 21 Jan 2013 09:33:05 +0000 (09:33 +0000)]
Unreviewed, rolling out r140311.
http://trac.webkit.org/changeset/140311
https://bugs.webkit.org/show_bug.cgi?id=107431

Broke selection-related tests (Requested by apavlov on
#webkit).

Patch by Sheriff Bot <webkit.review.bot@gmail.com> on 2013-01-21

Source/WebCore:

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

LayoutTests:

* fast/events/touch/emulate-touch-events.html:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@140313 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoUnreviewed. Build fix.
haraken@chromium.org [Mon, 21 Jan 2013 09:11:58 +0000 (09:11 +0000)]
Unreviewed. Build fix.

* bindings/v8/V8NPObject.cpp:
(WebCore::createV8ObjectForNPObject):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@140312 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoWeb Inspector: Do not dispatch mousemove when emulating touch event and no touch...
apavlov@chromium.org [Mon, 21 Jan 2013 08:49:54 +0000 (08:49 +0000)]
Web Inspector: Do not dispatch mousemove when emulating touch event and no touch is active
https://bugs.webkit.org/show_bug.cgi?id=107419

Reviewed by Pavel Feldman.

Source/WebCore:

mousemove event should not be dispatched when emulating touch events and mouse button is not pressed

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

LayoutTests:

* fast/events/touch/emulate-touch-events.html: Test that mousemove is dispatched not more than once.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@140311 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoMediaStream API: Update the RTCPeerConnection states to match the latest specification
tommyw@google.com [Mon, 21 Jan 2013 08:32:32 +0000 (08:32 +0000)]
MediaStream API: Update the RTCPeerConnection states to match the latest specification
https://bugs.webkit.org/show_bug.cgi?id=107120

Reviewed by Adam Barth.

Source/Platform:

Updating the chromium WebKit API for the new states.

* chromium/public/WebRTCPeerConnectionHandlerClient.h:
(WebKit::WebRTCPeerConnectionHandlerClient::~WebRTCPeerConnectionHandlerClient):
(WebRTCPeerConnectionHandlerClient):
(WebKit::WebRTCPeerConnectionHandlerClient::didChangeSignalingState):
(WebKit::WebRTCPeerConnectionHandlerClient::didChangeICEGatheringState):
(WebKit::WebRTCPeerConnectionHandlerClient::didChangeICEConnectionState):
(WebKit::WebRTCPeerConnectionHandlerClient::didAddRemoteDataChannel):
(WebKit::WebRTCPeerConnectionHandlerClient::didChangeReadyState):
(WebKit::WebRTCPeerConnectionHandlerClient::didChangeICEState):

Source/WebCore:

http://dev.w3.org/2011/webrtc/editor/webrtc.html#interface-definition

RTCPeerConnection::readyState and iceState are history and instead signalingState,
iceConnectionState and iceGatheringState have been introduced.

Existing tests updated to cover this patch.

* Modules/mediastream/RTCPeerConnection.cpp:
(WebCore::RTCPeerConnection::RTCPeerConnection):
(WebCore::RTCPeerConnection::createOffer):
(WebCore::RTCPeerConnection::createAnswer):
(WebCore::RTCPeerConnection::setLocalDescription):
(WebCore::RTCPeerConnection::localDescription):
(WebCore::RTCPeerConnection::setRemoteDescription):
(WebCore::RTCPeerConnection::remoteDescription):
(WebCore::RTCPeerConnection::updateIce):
(WebCore::RTCPeerConnection::addIceCandidate):
(WebCore::RTCPeerConnection::readyState):
(WebCore):
(WebCore::RTCPeerConnection::signalingState):
(WebCore::RTCPeerConnection::iceConnectionState):
(WebCore::RTCPeerConnection::addStream):
(WebCore::RTCPeerConnection::removeStream):
(WebCore::RTCPeerConnection::createDataChannel):
(WebCore::RTCPeerConnection::close):
(WebCore::RTCPeerConnection::didChangeSignalingState):
(WebCore::RTCPeerConnection::didChangeIceGatheringState):
(WebCore::RTCPeerConnection::didChangeIceConnectionState):
(WebCore::RTCPeerConnection::didAddRemoteStream):
(WebCore::RTCPeerConnection::didRemoveRemoteStream):
(WebCore::RTCPeerConnection::didAddRemoteDataChannel):
(WebCore::RTCPeerConnection::stop):
(WebCore::RTCPeerConnection::changeSignalingState):
(WebCore::RTCPeerConnection::changeIceGatheringState):
(WebCore::RTCPeerConnection::changeIceConnectionState):
* Modules/mediastream/RTCPeerConnection.h:
(RTCPeerConnection):
* Modules/mediastream/RTCPeerConnection.idl:
* platform/mediastream/RTCPeerConnectionHandlerClient.h:
(RTCPeerConnectionHandlerClient):
* platform/mediastream/chromium/RTCPeerConnectionHandlerChromium.cpp:
(WebCore::RTCPeerConnectionHandlerChromium::didChangeSignalingState):
(WebCore::RTCPeerConnectionHandlerChromium::didChangeICEConnectionState):
* platform/mediastream/chromium/RTCPeerConnectionHandlerChromium.h:
(RTCPeerConnectionHandlerChromium):

Source/WebKit/chromium:

* src/AssertMatchingEnums.cpp:

Tools:

Updating mock to use the new states.

* DumpRenderTree/chromium/MockWebRTCPeerConnectionHandler.cpp:
(RTCPeerConnectionStateTask::RTCPeerConnectionStateTask):
(MockWebRTCPeerConnectionHandler::initialize):
(MockWebRTCPeerConnectionHandler::updateICE):

LayoutTests:

* fast/mediastream/RTCPeerConnection-datachannel-expected.txt:
* fast/mediastream/RTCPeerConnection-datachannel.html:
* fast/mediastream/RTCPeerConnection-ice-expected.txt:
* fast/mediastream/RTCPeerConnection-ice.html:
* fast/mediastream/RTCPeerConnection-state-expected.txt:
* fast/mediastream/RTCPeerConnection-state.html:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@140310 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoREGRESSION (r140285): EFL uses WKPageResourceLoadClient which was removed
thiago.santos@intel.com [Mon, 21 Jan 2013 08:18:18 +0000 (08:18 +0000)]
REGRESSION (r140285): EFL uses WKPageResourceLoadClient which was removed
https://bugs.webkit.org/show_bug.cgi?id=107417

Reviewed by Gyuyoung Kim.

We had no choice but remove the ResourceLoadClientEfl after r140285
and thus our public API signals depending on it. The test_ewk2_auth_request
unit test was skipped because it depends these signals and will be fixed in a
different patch.

* PlatformEfl.cmake:
* UIProcess/API/efl/EWebKit2.h:
* UIProcess/API/efl/EwkViewCallbacks.h:
* UIProcess/API/efl/EwkViewImpl.cpp:
(EwkViewImpl::EwkViewImpl):
* UIProcess/API/efl/EwkViewImpl.h:
(WebKit):
(EwkViewImpl):
* UIProcess/API/efl/ewk_resource.cpp: Removed.
* UIProcess/API/efl/ewk_resource.h: Removed.
* UIProcess/API/efl/ewk_resource_private.h: Removed.
* UIProcess/API/efl/ewk_view.cpp:
* UIProcess/API/efl/ewk_view.h:
* UIProcess/efl/ResourceLoadClientEfl.cpp: Removed.
* UIProcess/efl/ResourceLoadClientEfl.h: Removed.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@140309 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoUse ISO8601 date/time formats as fallbacks for date/time input types
tkent@chromium.org [Mon, 21 Jan 2013 08:09:21 +0000 (08:09 +0000)]
Use ISO8601 date/time formats as fallbacks for date/time input types
https://bugs.webkit.org/show_bug.cgi?id=107418

Reviewed by Kentaro Hara.

We obtain date/time formats from OSes to build date/time input type
UIs. If something wrong happens in the code, fallback formats are
used. Such fallback formats should not be English formats in order that
we can find defects easily.

No new tests. These formats should not be used unless the current code
has defects.

* html/DateTimeInputType.cpp:
(WebCore::DateTimeInputType::setupLayoutParameters):
Use an ISO8601 format which is same as a format used in HTML5.
* html/DateTimeLocalInputType.cpp:
(WebCore::DateTimeLocalInputType::setupLayoutParameters): Ditto.
* html/MonthInputType.cpp:
(WebCore::MonthInputType::setupLayoutParameters): Ditto.
* html/WeekInputType.cpp:
(WebCore::WeekInputType::setupLayoutParameters): Ditto.
* platform/text/LocaleICU.cpp:
(WebCore::LocaleICU::dateFormat): Ditto.
* platform/text/LocaleNone.cpp:
(WebCore::LocaleNone::dateFormat): Ditto.
(WebCore::LocaleNone::dateTimeFormatWithSeconds): Ditto.
(WebCore::LocaleNone::dateTimeFormatWithoutSeconds): Ditto.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@140308 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoElements must be reattached when inserted/removed from top layer
falken@chromium.org [Mon, 21 Jan 2013 07:59:39 +0000 (07:59 +0000)]
Elements must be reattached when inserted/removed from top layer
https://bugs.webkit.org/show_bug.cgi?id=105489

Relanding r139402 as rollout due to suspected perf regression did not help (bug 106726).

Reviewed by Julien Chaffraix.

Source/WebCore:

Ensure a reattach occurs when an element is inserted/removed from top layer, so its renderer can be inserted correctly:
as a child of RenderView in top layer sibling order if it's in the top layer, and in the usual place otherwise.

We previously relied on style recalc to catch when an element is inserted/removed from the top layer, because it
only happens on dialog.show/close which toggle display: none. But that is incorrect because, for example, close()
followed immediately by show() results in no style change.

Tests: fast/dom/HTMLDialogElement/removed-element-is-removed-from-top-layer.html
       fast/dom/HTMLDialogElement/top-layer-stacking-correct-order-remove-readd.html

* dom/Element.cpp:
(WebCore::Element::removedFrom): Call Document::removeFromTopLayer to let the element be removed from the top layer vector.
removeFromTopLayer calls Element::setIsInTopLayer(false) itself if needed.
(WebCore::Element::setIsInTopLayer): Ensure a reattach occurs if the element is already attached.

LayoutTests:

* fast/dom/HTMLDialogElement/removed-element-is-removed-from-top-layer-expected.html: Added.
* fast/dom/HTMLDialogElement/removed-element-is-removed-from-top-layer.html: Added.
This tests that a top layer element removed from the document does not reappear in the top layer if readded.
This test actually would pass before this patch, but just by good fortune (see bug).
* fast/dom/HTMLDialogElement/top-layer-stacking-correct-order-remove-readd-expected.html: Added.
* fast/dom/HTMLDialogElement/top-layer-stacking-correct-order-remove-readd.html: Added.
This tests that top layer ordering is correct after removing and readding an element to the top layer.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@140307 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoWeb Inspector: change HeapSnapshotLoader to allow loading native heap snapshots
yurys@chromium.org [Mon, 21 Jan 2013 07:50:51 +0000 (07:50 +0000)]
Web Inspector: change HeapSnapshotLoader to allow loading native heap snapshots
https://bugs.webkit.org/show_bug.cgi?id=107282

Reviewed by Pavel Feldman.

Conctete heap snapshot constructor name is passed as a parameter to HeapSnapshotLoaderProxy and
propagated to HeapSnapshotLoader.buildSnapshot to support different types of heap snapshots.

Source/WebCore:

* inspector/front-end/HeapSnapshotLoader.js:
(WebInspector.HeapSnapshotLoader.prototype.buildSnapshot):
* inspector/front-end/HeapSnapshotProxy.js:
(WebInspector.HeapSnapshotWorker.prototype.createLoader):
(WebInspector.HeapSnapshotLoaderProxy):
(WebInspector.HeapSnapshotLoaderProxy.prototype.close):
* inspector/front-end/HeapSnapshotView.js:
(WebInspector.HeapProfileHeader.prototype._setupWorker):

LayoutTests:

* inspector/profiler/heap-snapshot-loader.html:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@140306 268f45cc-cd09-0410-ab3c-d52691b4dbfc