WebKit-https.git
9 years ago2010-10-28 Peter Varga <pvarga@inf.u-szeged.hu>
commit-queue@webkit.org [Thu, 28 Oct 2010 11:17:18 +0000 (11:17 +0000)]
2010-10-28  Peter Varga  <pvarga@inf.u-szeged.hu>

        Reviewed by Csaba Osztrogonác.

        resetAssertionMatches() is an unused function in YARR Interpreter
        https://bugs.webkit.org/show_bug.cgi?id=48503

        The resetAssertionMatches() function is removed from YARR Interpreter
        because it's never called.

        * yarr/RegexInterpreter.cpp:
        (JSC::Yarr::Interpreter::resetMatches):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@70764 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2010-10-28 MORITA Hajime <morrita@google.com>
morrita@google.com [Thu, 28 Oct 2010 10:41:58 +0000 (10:41 +0000)]
2010-10-28  MORITA Hajime  <morrita@google.com>

        Reviewed by Kent Tamura.

        WebKit needs to expose an Element::spellcheck attribute to javascript.
        https://bugs.webkit.org/show_bug.cgi?id=25536

        * editing/spelling/script-tests/spelling-attribute-change.js: Added.
        * editing/spelling/spelling-attribute-change-expected.txt: Added.
        * editing/spelling/spelling-attribute-change.html: Added.
        * fast/dom/HTMLElement/script-tests/spellcheck.js: Added.
        * fast/dom/HTMLElement/spellcheck-expected.txt: Added.
        * fast/dom/HTMLElement/spellcheck.html: Added.
        * platform/chromium/test_expectations.txt:
        * platform/gtk/Skipped:
        * platform/qt/Skipped:
2010-10-27  MORITA Hajime  <morrita@google.com>

        Reviewed by Kent Tamura.

        WebKit needs to expose an Element::spellcheck attribute to javascript.
        https://bugs.webkit.org/show_bug.cgi?id=25536

        Published Element.spellcheck property on IDL, implemented it.

        Tests: editing/spelling/spelling-attribute-change.html
               fast/dom/HTMLElement/spellcheck.html

        * html/HTMLElement.cpp:
        (WebCore::HTMLElement::spellcheck):
        (WebCore::HTMLElement::setSpellcheck):
        * html/HTMLElement.h:
        * html/HTMLElement.idl:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@70763 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2010-10-27 Pavel Feldman <pfeldman@chromium.org>
pfeldman@chromium.org [Thu, 28 Oct 2010 10:22:51 +0000 (10:22 +0000)]
2010-10-27  Pavel Feldman  <pfeldman@chromium.org>

        Reviewed by Timothy Hatcher.

        Web Inspector: make debugger work against new resource manager.
        https://bugs.webkit.org/show_bug.cgi?id=48376

        * inspector/Inspector.idl:
        * inspector/InspectorController.cpp:
        (WebCore::InspectorController::identifierForInitialRequest):
        * inspector/InspectorResourceAgent.cpp:
        (WebCore::InspectorResourceAgent::identifierForInitialRequest):
        * inspector/InspectorResourceAgent.h:
        * inspector/front-end/ExtensionServer.js:
        (WebInspector.ExtensionServer.prototype._onGetResourceContent):
        * inspector/front-end/Resource.js:
        (WebInspector.Resource.prototype.set content):
        * inspector/front-end/ResourceManager.js:
        (WebInspector.ResourceManager):
        (WebInspector.ResourceManager.prototype.identifierForInitialRequest):
        (WebInspector.ResourceManager.prototype._createResource):
        (WebInspector.ResourceManager.prototype.willSendRequest):
        (WebInspector.ResourceManager.prototype._updateResourceWithRequest):
        (WebInspector.ResourceManager.prototype.didFinishLoading):
        (WebInspector.ResourceManager.prototype.didFailLoading):
        (WebInspector.ResourceManager.prototype.setOverrideContent):
        (WebInspector.ResourceManager.prototype._processCachedResources):
        (WebInspector.ResourceManager.prototype._addFramesRecursively):
        (WebInspector.ResourceManager.prototype.resourceForURL):
        (WebInspector.ResourceManager.prototype.addConsoleMessage):
        (WebInspector.ResourceManager.prototype.clearConsoleMessages):
        (WebInspector.ResourceManager.prototype.forAllResources):
        (WebInspector.ResourceManager.prototype._bindResourceURL):
        (WebInspector.ResourceManager.prototype._unbindResourceURL):
        (WebInspector.ResourceManager.getContent):
        (WebInspector.ResourceTreeModel):
        (WebInspector.ResourceTreeModel.prototype.didCommitLoadForFrame):
        (WebInspector.ResourceTreeModel.prototype.addResourceToFrame):
        (WebInspector.ResourceTreeModel.prototype._clearResources):
        * inspector/front-end/StoragePanel.js:
        (WebInspector.StoragePanel.prototype.addOrUpdateFrame):
        * inspector/front-end/networkPanel.css:
        (.network.panel .resources-dividers):

2010-10-27  Pavel Feldman  <pfeldman@chromium.org>

        Reviewed by Timothy Hatcher.

        Web Inspector: make debugger work against new resource manager.
        https://bugs.webkit.org/show_bug.cgi?id=48376

        * http/tests/inspector/inspector-test2.js:
        (initialize_InspectorTest.InspectorTest.reloadPageIfNeeded):
        (initialize_InspectorTest.InspectorTest.pageReloaded):
        * http/tests/inspector/resource-har-conversion.html:
        * http/tests/inspector/resource-parameters.html:
        * inspector/extensions-resources.html:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@70762 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2010-10-28 Kimmo Kinnunen <kimmo.t.kinnunen@nokia.com>
commit-queue@webkit.org [Thu, 28 Oct 2010 09:02:30 +0000 (09:02 +0000)]
2010-10-28  Kimmo Kinnunen  <kimmo.t.kinnunen@nokia.com>

         Reviewed by Kenneth Rohde Christiansen.

         [Qt] Web process cannot be debugged if it crashes in initialization
         https://bugs.webkit.org/show_bug.cgi?id=47399

         Wait 3 seconds for the debugger on debug mode if
         WEBKIT2_PAUSE_WEB_PROCESS_ON_LAUNCH environment variable is set.

         The debugging can be initiated for example in following way:
          export WEBKIT2_PAUSE_WEB_PROCESS_ON_LAUNCH=1
          ./MiniBrowser & while [ -z `pidof QtWebProcess` ]; do sleep 1; done; \
           gdb --pid=`pidof QtWebProcess`

         * WebProcess/qt/WebProcessMainQt.cpp:
         (WebKit::WebProcessMainQt):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@70760 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoOne more Qt fix due to r70754.
aestes@apple.com [Thu, 28 Oct 2010 08:33:16 +0000 (08:33 +0000)]
One more Qt fix due to r70754.

* html/HTMLObjectElement.cpp:
(WebCore::HTMLObjectElement::hasValidClassId):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@70759 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2010-10-28 Pascal Massimino <pascal.massimino@gmail.com>
commit-queue@webkit.org [Thu, 28 Oct 2010 08:21:48 +0000 (08:21 +0000)]
2010-10-28  Pascal Massimino  <pascal.massimino@gmail.com>

        Reviewed by Adam Barth.

        fix setSize() call flow: it was only called the first
        time (when onlySize is true)
        Bug URL: https://bugs.webkit.org/show_bug.cgi?id=48494

        Testing: difficult to exercise the bug, as it requires
        some manual flow (change tab, scroll,...)

        * platform/image-decoders/webp/WEBPImageDecoder.cpp:
        (WebCore::WEBPImageDecoder::decode):
           fix: call setSize() even when onlySize is false

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@70758 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoFix the Qt build after r70754.
aestes@apple.com [Thu, 28 Oct 2010 08:16:33 +0000 (08:16 +0000)]
Fix the Qt build after r70754.

* html/HTMLObjectElement.cpp:
(WebCore::HTMLObjectElement::hasValidClassId):
(WebCore::HTMLObjectElement::updateWidget):
* html/HTMLObjectElement.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@70757 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago32-bit build fix.
mrowe@apple.com [Thu, 28 Oct 2010 08:08:09 +0000 (08:08 +0000)]
32-bit build fix.

Qualify references to Rect to avoid ambiguity between WebCore::Rect and Carbon's Rect.

* WebProcess/Plugins/Netscape/mac/NetscapePluginMac.mm:
(WebKit::NetscapePlugin::platformPostInitialize):
(WebKit::NetscapePlugin::windowFrameChanged):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@70756 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2010-10-28 Adam Barth <abarth@webkit.org>
abarth@webkit.org [Thu, 28 Oct 2010 08:02:53 +0000 (08:02 +0000)]
2010-10-28  Adam Barth  <abarth@webkit.org>

        Reviewed by Alexey Proskuryakov.

        Fix test description
        https://bugs.webkit.org/show_bug.cgi?id=48501

        I mistakenly copy/pasted this text from another test.

        * fast/dom/allowed-children-expected.txt:
        * fast/dom/allowed-children.html:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@70755 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2010-10-28 Andy Estes <aestes@apple.com>
aestes@apple.com [Thu, 28 Oct 2010 07:47:56 +0000 (07:47 +0000)]
2010-10-28  Andy Estes  <aestes@apple.com>

        Reviewed by Adam Barth.

        Fix a test failure introduced in r70748 by supporting Qt's non-standard
        use of classid.
        https://bugs.webkit.org/show_bug.cgi?id=45679

        * html/HTMLObjectElement.cpp:
        (WebCore::objectHasSupportedClassId): Return true if the object's
        serviceType is 'application/x-qt-plugin'.
        (WebCore::HTMLObjectElement::updateWidget): Do not render fallback
        content if a non-empty classid is specified for a Qt plugin object.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@70754 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoStop allowing deprecated methods to be used in NetscapePluginHostProxy.mm now
mrowe@apple.com [Thu, 28 Oct 2010 07:47:06 +0000 (07:47 +0000)]
Stop allowing deprecated methods to be used in NetscapePluginHostProxy.mm now
that deprecated methods are no longer used.

* WebKit.xcodeproj/project.pbxproj:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@70753 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoWrong instruction form for BKPT
zherczeg@webkit.org [Thu, 28 Oct 2010 07:22:38 +0000 (07:22 +0000)]
Wrong instruction form for BKPT
https://bugs.webkit.org/show_bug.cgi?id=48427

Reviewed by Andreas Kling.

One '0' is missing from BKPT instruction.
Thanks for Jacob Bramley for reporting this error.

* assembler/ARMAssembler.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@70752 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2010-10-28 Ivan Krstić <ike@apple.com>
commit-queue@webkit.org [Thu, 28 Oct 2010 07:14:41 +0000 (07:14 +0000)]
2010-10-28  Ivan Krstić  <ike@apple.com>

        Reviewed by Mark Rowe.

        Remove unused experimental proxied panel interface.
        <rdar://problem/7237059>

        * WebCore.exp.in:
        * platform/mac/WebCoreSystemInterface.h:
        * platform/mac/WebCoreSystemInterface.mm:
2010-10-28  Ivan Krstić  <ike@apple.com>

        Reviewed by Mark Rowe.

        Remove unused experimental proxied panel interface.
        <rdar://problem/7237059>

        * Plugins/Hosted/NetscapePluginHostManager.h:
        * Plugins/Hosted/NetscapePluginHostManager.mm:
        (WebKit::NetscapePluginHostManager::hostForPlugin):
        (WebKit::NetscapePluginHostManager::spawnPluginHost):
        (WebKit::NetscapePluginHostManager::instantiatePlugin):
        * Plugins/Hosted/NetscapePluginHostProxy.mm:
        * Plugins/Hosted/WebKitPluginClient.defs:
        * Plugins/Hosted/WebKitPluginHost.defs:
        * WebCoreSupport/WebSystemInterface.mm:
        (InitWebCoreSystemInterface):
        * WebView/WebPreferenceKeysPrivate.h:
        * WebView/WebPreferences.mm:
        (+[WebPreferences initialize]):
        * WebView/WebPreferencesPrivate.h:
2010-10-28  Ivan Krstić  <ike@apple.com>

        Reviewed by Mark Rowe.

        Remove unused experimental proxied panel interface.
        <rdar://problem/7237059>

        * WebProcess/WebCoreSupport/mac/WebSystemInterface.mm:
        (InitWebCoreSystemInterface):
2010-10-28  Ivan Krstić  <ike@apple.com>

        Reviewed by Mark Rowe.

        Remove unused experimental proxied panel interface.
        <rdar://problem/7237059>

        * WebKitSystemInterface.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@70751 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2010-10-28 Xan Lopez <xlopez@igalia.com>
xan@webkit.org [Thu, 28 Oct 2010 07:11:01 +0000 (07:11 +0000)]
2010-10-28  Xan Lopez  <xlopez@igalia.com>

        Try to fix Snow Leopard build.

        * jit/JITPropertyAccess.cpp:
        (JSC::JIT::testPrototype):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@70750 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2010-10-28 Xan Lopez <xlopez@igalia.com>
xan@webkit.org [Thu, 28 Oct 2010 07:00:37 +0000 (07:00 +0000)]
2010-10-28  Xan Lopez  <xlopez@igalia.com>

        Reviewed by Oliver Hunt.

        Do not have two different asCell APIs in JSValue
        https://bugs.webkit.org/show_bug.cgi?id=47979

        Remove JSCell* asCell(JSValue) in favor of only using
        JSValue::asCell().

        * API/APICast.h:
        (toRef):
        * jit/JITPropertyAccess32_64.cpp:
        (JSC::JIT::testPrototype):
        * jit/JITStubs.cpp:
        (JSC::JITThunks::tryCachePutByID):
        (JSC::JITThunks::tryCacheGetByID):
        (JSC::DEFINE_STUB_FUNCTION):
        * runtime/GetterSetter.h:
        (JSC::asGetterSetter):
        * runtime/JSByteArray.h:
        (JSC::asByteArray):
        * runtime/JSCell.h:
        (JSC::JSCell::getCallData):
        (JSC::JSCell::getConstructData):
        * runtime/JSString.h:
        (JSC::RopeBuilder::appendValueInConstructAndIncrementLength):
        (JSC::asString):
        * runtime/JSValue.h:
        * runtime/Operations.cpp:
        (JSC::jsIsObjectType):
        * runtime/Operations.h:
        (JSC::normalizePrototypeChain):
        * runtime/Protect.h:
        (JSC::gcProtect):
        (JSC::gcUnprotect):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@70749 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2010-10-27 Andy Estes <aestes@apple.com>
aestes@apple.com [Thu, 28 Oct 2010 06:15:09 +0000 (06:15 +0000)]
2010-10-27  Andy Estes  <aestes@apple.com>

        Reviewed by Adam Barth.

        r66156 broke AtlasCT library, formerly affected http://map.d.co.il/
        https://bugs.webkit.org/show_bug.cgi?id=45679

        This patch removes WebKit's map of classid values to MIME types. It also
        changes the behavior of object elements to render fallback content when
        a non-empty classid attribute is specified, which is the behavior HTML5
        specifies when a UA encounters a classid it doesn't understand.

        Test: fast/replaced/object-with-non-empty-classid-triggers-fallback.html

        * html/HTMLObjectElement.cpp: Remove serviceTypeForClassId(),
        createClassIdToTypeMap(), and the ClassIdToTypeMap typedef.
        (WebCore::HTMLObjectElement::updateWidget): Do not call
        serviceTypeForClassId() when there is no type attribute, and render
        fallback content if the classid attribute is non-empty.
2010-10-27  Andy Estes  <aestes@apple.com>

        Reviewed by Adam Barth.

        r66156 broke AtlasCT library, formerly affected http://map.d.co.il/
        https://bugs.webkit.org/show_bug.cgi?id=45679

        * fast/dom/object-embed-plugin-scripting.html: Changed classid
        attributes to type attributes.
        * fast/dom/object-plugin-hides-properties.html: Ditto.
        * fast/replaced/object-with-non-empty-classid-triggers-fallback-expected.txt: Added.
        * fast/replaced/object-with-non-empty-classid-triggers-fallback.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@70748 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2010-10-27 Eric Seidel <eric@webkit.org>
eric@webkit.org [Thu, 28 Oct 2010 05:12:39 +0000 (05:12 +0000)]
2010-10-27  Eric Seidel  <eric@webkit.org>

        Reviewed by Adam Barth.

        webkit-patch suggest-reviewers -g 260550a6e30b7bf34f16bdb4a5396cf26264fc1c is very slow
        https://bugs.webkit.org/show_bug.cgi?id=48500

        This doesn't fix the problem, but it makes things slightly better.
        Each git svn find-rev call takes about .25 seconds on my desktop.
        This patch uses a new memoized class to avoid those calls when possible.

        The real slowness is still git log on some files, like:
        git log --pretty=format:%H -5 -- /Projects/WebKit/WebCore/platform/wx/SearchPopupMenuWx.h
        I'm not yet sure how to make the pathological git logs better.

        * Scripts/webkitpy/common/checkout/scm.py:
        * Scripts/webkitpy/common/memoized.py: Added.
        * Scripts/webkitpy/common/memoized_unittest.py: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@70747 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2010-10-27 Kinuko Yasuda <kinuko@chromium.org>
kinuko@chromium.org [Thu, 28 Oct 2010 04:00:14 +0000 (04:00 +0000)]
2010-10-27  Kinuko Yasuda  <kinuko@chromium.org>

        Unreviewed, updating drt_expectations.txt.

        * platform/chromium/drt_expectations.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@70746 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2010-10-27 Eric Uhrhane <ericu@chromium.org>
commit-queue@webkit.org [Thu, 28 Oct 2010 03:57:58 +0000 (03:57 +0000)]
2010-10-27  Eric Uhrhane  <ericu@chromium.org>

        Reviewed by David Levin.

        [Chromium] Metadata queries should return full file info, not just modification time
        https://bugs.webkit.org/show_bug.cgi?id=48098

        LayoutTests that use this functionality via WorkerAsyncFileWriter are in review.

        Build file changes to add FileMetadata.
        * GNUmakefile.am:
        * WebCore.gypi:
        * WebCore.pro:
        * WebCore.vcproj/WebCore.vcproj:
        * WebCore.xcodeproj/project.pbxproj:

        The rest of the changes here just pass the new FileMetadata object through calls that used to take a modification time.
        * platform/FileMetadata.h: Added.
        (WebCore::FileMetadata::FileMetadata):
        * fileapi/FileSystemCallbacks.cpp:
        (WebCore::FileSystemCallbacksBase::didReadMetadata):
        (WebCore::MetadataCallbacks::didReadMetadata):
        * fileapi/FileSystemCallbacks.h:
        * platform/AsyncFileSystemCallbacks.h:
2010-10-27  Eric Uhrhane  <ericu@chromium.org>

        Reviewed by David Levin.

        [Chromium] Metadata queries should return full file info, not just modification time
        https://bugs.webkit.org/show_bug.cgi?id=48098

        * src/AssertMatchingEnums.cpp: Ensure WebFileInfo and FileMetadata use the same values for file/directory indicators.
        * src/WebFileSystemCallbacksImpl.cpp:
        (WebKit::WebFileSystemCallbacksImpl::didReadMetadata): Pass through the new information.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@70745 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2010-10-27 Kinuko Yasuda <kinuko@chromium.org>
kinuko@chromium.org [Thu, 28 Oct 2010 03:48:24 +0000 (03:48 +0000)]
2010-10-27  Kinuko Yasuda  <kinuko@chromium.org>

        Unreviewed, update drt_expectations.txt.

        * platform/chromium/drt_expectations.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@70744 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2010-10-27 Chris Rogers <crogers@google.com>
crogers@google.com [Thu, 28 Oct 2010 03:38:08 +0000 (03:38 +0000)]
2010-10-27  Chris Rogers  <crogers@google.com>

        Reviewed by Kenneth Russell.

        Add VectorMath files
        https://bugs.webkit.org/show_bug.cgi?id=48467

        No new tests since audio API is not yet implemented.

        * platform/audio/VectorMath.cpp: Added.
        (WebCore::VectorMath::vsmul):
        (WebCore::VectorMath::vadd):
        * platform/audio/VectorMath.h: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@70743 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2010-10-27 Chris Rogers <crogers@google.com>
crogers@google.com [Thu, 28 Oct 2010 03:21:56 +0000 (03:21 +0000)]
2010-10-27  Chris Rogers  <crogers@google.com>

        Reviewed by Kenneth Russell.

        Fixup files affected by VectorMath and related API changes
        https://bugs.webkit.org/show_bug.cgi?id=48481

        No new tests since audio API is not yet implemented.

        * platform/audio/AudioBus.cpp:
        * platform/audio/AudioChannel.cpp:
        * platform/audio/Biquad.cpp:
        * platform/audio/FFTConvolver.cpp:
        * platform/audio/FFTConvolver.h:
        * platform/audio/Panner.cpp:
        (WebCore::Panner::create):
        * platform/audio/Reverb.cpp:
        (WebCore::calculateNormalizationScale):
        (WebCore::Reverb::initialize):
        (WebCore::Reverb::process):
        * platform/audio/ReverbAccumulationBuffer.cpp:
        * platform/audio/ReverbAccumulationBuffer.h:
        * platform/audio/ReverbConvolver.cpp:
        (WebCore::ReverbConvolver::ReverbConvolver):
        (WebCore::ReverbConvolver::process):
        * platform/audio/ReverbConvolver.h:
        * platform/audio/ReverbConvolverStage.cpp:
        (WebCore::ReverbConvolverStage::ReverbConvolverStage):
        * platform/audio/ReverbConvolverStage.h:
        * platform/audio/ReverbInputBuffer.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@70742 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2010-10-27 Kinuko Yasuda <kinuko@chromium.org>
kinuko@chromium.org [Thu, 28 Oct 2010 03:18:10 +0000 (03:18 +0000)]
2010-10-27  Kinuko Yasuda  <kinuko@chromium.org>

        Reviewed by Dumitru Daniliuc.

        DOMFileSystem's reference should be kept while there're any active Entries/callbacks
        https://bugs.webkit.org/show_bug.cgi?id=47535

        Changed to keep DOMFileSystem's reference in each EntryBase, DirectoryReaderBase, EntryCallbacks and EntriesCallbacks.

        Test: fast/filesystem/filesystem-reference.html

        * fileapi/DOMFileSystemBase.cpp:
        (WebCore::DOMFileSystemBase::readDirectory): Changed to pass/take
        PassRefPtr but not a raw pointer.
        * fileapi/DOMFileSystemBase.h:
        * fileapi/DirectoryEntry.cpp:
        (WebCore::DirectoryEntry::DirectoryEntry): Ditto.
        (WebCore::DirectoryEntry::createReader):
        * fileapi/DirectoryEntry.h:
        (WebCore::DirectoryEntry::create): Ditto.
        * fileapi/DirectoryEntrySync.cpp:
        (WebCore::DirectoryEntrySync::DirectoryEntrySync): Ditto.
        * fileapi/DirectoryEntrySync.h:
        (WebCore::DirectoryEntrySync::create): Ditto.
        * fileapi/DirectoryReader.h:
        (WebCore::DirectoryReader::filesystem):
        * fileapi/DirectoryReaderBase.h:
        (WebCore::DirectoryReaderBase::filesystem):
        * fileapi/Entry.cpp:
        (WebCore::Entry::Entry): Ditto.
        * fileapi/Entry.h:
        (WebCore::Entry::filesystem):
        * fileapi/EntryBase.h:
        (WebCore::EntryBase::filesystem):
        (WebCore::EntryBase::EntryBase): Ditto.
        * fileapi/EntrySync.cpp:
        (WebCore::EntrySync::create): Ditto.
        (WebCore::EntrySync::getParent):
        (WebCore::EntrySync::EntrySync): Ditto.
        * fileapi/EntrySync.h:
        (WebCore::EntrySync::filesystem):
        * fileapi/FileEntry.cpp:
        (WebCore::FileEntry::FileEntry): Ditto.
        * fileapi/FileEntry.h:
        (WebCore::FileEntry::create): Ditto.
        * fileapi/FileEntrySync.cpp:
        (WebCore::FileEntrySync::FileEntrySync): Ditto.
        * fileapi/FileEntrySync.h:
        (WebCore::FileEntrySync::create): Ditto.
        * fileapi/FileSystemCallbacks.cpp:
        (WebCore::EntriesCallbacks::create): Ditto.
        (WebCore::EntriesCallbacks::EntriesCallbacks): Ditto.
        * fileapi/FileSystemCallbacks.h:
2010-10-27  Kinuko Yasuda  <kinuko@chromium.org>

        Reviewed by Dumitru Daniliuc.

        DOMFileSystem's reference should be kept while there're any active Entries/callbacks
        https://bugs.webkit.org/show_bug.cgi?id=47535

        * fast/filesystem/filesystem-reference-expected.txt: Added.
        * fast/filesystem/filesystem-reference.html: Added.
        * fast/filesystem/script-tests/filesystem-reference.js: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@70741 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2010-10-27 Yuzo Fujishima <yuzo@google.com>
yuzo@google.com [Thu, 28 Oct 2010 03:11:33 +0000 (03:11 +0000)]
2010-10-27  Yuzo Fujishima  <yuzo@google.com>

        Reviewed by Eric Seidel.

        Fix for Bug 48310 - SVG font-face-name without name attribute causes a crash
        https://bugs.webkit.org/show_bug.cgi?id=48310

        * svg/custom/font-face-name-without-name-attr-expected.txt: Added.
        * svg/custom/font-face-name-without-name-attr.svg: Added.
2010-10-27  Yuzo Fujishima  <yuzo@google.com>

        Reviewed by Eric Seidel.

        Fix for Bug 48310 - SVG font-face-name without name attribute causes a crash
        https://bugs.webkit.org/show_bug.cgi?id=48310

        Test: svg/custom/font-face-name-without-name-attr.svg

        * svg/SVGFontFaceElement.cpp:
        (WebCore::SVGFontFaceElement::rebuildFontFace): Don't create src
        property if the src value list is empty.
        * svg/SVGFontFaceSrcElement.cpp:
        (WebCore::SVGFontFaceSrcElement::srcValue): Don't add empty src
        values to the src value list.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@70740 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2010-10-27 Chris Rogers <crogers@google.com>
crogers@google.com [Thu, 28 Oct 2010 03:02:37 +0000 (03:02 +0000)]
2010-10-27  Chris Rogers  <crogers@google.com>

        Reviewed by Kenneth Russell.

        Allow AudioDSPKernel to function without an AudioDSPKernelProcessor
        https://bugs.webkit.org/show_bug.cgi?id=48472

        No new tests since audio API is not yet implemented.

        * platform/audio/AudioDSPKernel.h:
        (WebCore::AudioDSPKernel::AudioDSPKernel):
        (WebCore::AudioDSPKernel::sampleRate):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@70739 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoBuild fix.
mrowe@apple.com [Thu, 28 Oct 2010 02:37:00 +0000 (02:37 +0000)]
Build fix.

* UIProcess/API/mac/PageClientImpl.mm: Reference the file using the correct case.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@70738 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoFind indicators do not bounce
andersca@apple.com [Thu, 28 Oct 2010 01:44:58 +0000 (01:44 +0000)]
Find indicators do not bounce
https://bugs.webkit.org/show_bug.cgi?id=48490
<rdar://problem/8564276>

Reviewed by Sam Weinig.

WebKit2:

* UIProcess/API/mac/FindIndicatorWindow.h:
* UIProcess/API/mac/FindIndicatorWindow.mm:
(-[WebFindIndicatorWindowAnimation _initWithFindIndicatorWindow:WebKit::animationDuration:animationProgressCallback:WebKit::FindIndicatorWindow::animationDidEndCallback:WebKit::FindIndicatorWindow::]):
Add an animationDuration parameter.

(WebKit::FindIndicatorWindow::FindIndicatorWindow):
Initialize m_bounceAnimationContext.

(WebKit::FindIndicatorWindow::setFindIndicator):
Create a bounce animation and start it.

(WebKit::FindIndicatorWindow::closeWindow):
Stop the bounce animation and destroy the bounce animation context.

(WebKit::FindIndicatorWindow::startFadeOutTimerFired):
pass the fade out duration.

(WebKit::FindIndicatorWindow::bounceAnimationCallback):
Set the bounce animation progress.

(WebKit::FindIndicatorWindow::bounceAnimationDidEnd):
Destroy the bounce animation context.

WebKitLibraries:

Add bounce animation context functions.

* WebKitSystemInterface.h:
* libWebKitSystemInterfaceLeopard.a:
* libWebKitSystemInterfaceSnowLeopard.a:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@70737 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2010-10-27 Chao-ying Fu <fu@mips.com>
commit-queue@webkit.org [Thu, 28 Oct 2010 01:30:09 +0000 (01:30 +0000)]
2010-10-27  Chao-ying Fu  <fu@mips.com>

        Reviewed by Oliver Hunt.

        Support emit_op_mod() for MIPS on JSVALUE32_64
        https://bugs.webkit.org/show_bug.cgi?id=46511

        This patch uses MIPS div instructions for op_mod to improve performance.

        * jit/JITArithmetic32_64.cpp:
        (JSC::JIT::emit_op_mod):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@70736 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2010-10-27 Stephen White <senorblanco@chromium.org>
senorblanco@chromium.org [Thu, 28 Oct 2010 01:07:03 +0000 (01:07 +0000)]
2010-10-27  Stephen White  <senorblanco@chromium.org>

        Reviewed by Kenneth Russell.

        [CHROMIUM] Unskip the failing tests in chromium-gpu, and adjust the
        expectations to match reality.
        https://bugs.webkit.org/show_bug.cgi?id=48473

        Some bugs were already fixed and the test lines could be removed, some
        tests have started newly-failing.  Rebaselined one test that had a bad
        checksum.

        * platform/chromium-gpu-linux/compositing/iframes/composited-iframe-alignment-expected.checksum: Added.
        * platform/chromium-gpu-linux/compositing/iframes/composited-iframe-alignment-expected.png: Added.
        * platform/chromium-gpu-linux/compositing/iframes/composited-iframe-alignment-expected.txt: Added.
        * platform/chromium-gpu/test_expectations.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@70735 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2010-10-27 Kenneth Rohde Christiansen <kenneth@webkit.org>
andreas.kling@nokia.com [Thu, 28 Oct 2010 00:55:29 +0000 (00:55 +0000)]
2010-10-27  Kenneth Rohde Christiansen  <kenneth@webkit.org>

        Reviewed by Andreas Kling.

        Remove the reference to the pageNamespace in the QWKPagePrivate.
        Access it though the WebPageProxy instead which holds a strong
        ref to it.

        * UIProcess/API/qt/qwkpage.cpp:
        (QWKPagePrivate::QWKPagePrivate):
        (QWKPage::preferences):
        * UIProcess/API/qt/qwkpage_p.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@70734 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoMore work on downloads
andersca@apple.com [Thu, 28 Oct 2010 00:48:57 +0000 (00:48 +0000)]
More work on downloads
https://bugs.webkit.org/show_bug.cgi?id=48480

Reviewed by Sam Weinig.

* WebProcess/Downloads/Download.cpp:
(WebKit::Download::~Download):
Call platformInvalidate().

(WebKit::Download::didBegin):
(WebKit::Download::didReceiveData):
(WebKit::Download::didFinish):
Add empty stubs.

* WebProcess/Downloads/DownloadManager.cpp:
(WebKit::DownloadManager::startDownload):
Create a Download object and insert it into the m_downloads map.

* WebProcess/Downloads/cf/DownloadCFNet.cpp:
(WebKit::Download::platformInvalidate):
Add empty stub.

* WebProcess/Downloads/mac/DownloadMac.mm:
(WebKit::Download::platformInvalidate):
Tell the WKDownloadAsDelegate object that we're being invalidated.

(-[WKDownloadAsDelegate downloadDidBegin:]):
Call Download::didBegin.

(-[WKDownloadAsDelegate download:didReceiveDataOfLength:]):
Call Download::didReceiveData.

(-[WKDownloadAsDelegate downloadDidFinish:]):
Call Download::didFinish.

* WebProcess/Downloads/qt/DownloadQt.cpp:
(WebKit::Download::platformInvalidate):
Add empty stub.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@70733 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2010-10-27 Pratik Solanki <psolanki@apple.com>
commit-queue@webkit.org [Thu, 28 Oct 2010 00:42:35 +0000 (00:42 +0000)]
2010-10-27  Pratik Solanki  <psolanki@apple.com>

        Reviewed by Darin Adler.

        Improve memSize calculation in [WebView _setCacheModel]
        https://bugs.webkit.org/show_bug.cgi?id=48484

        * WebView/WebView.mm:
        (roundUpToPowerOf2): Added. Utility function to calculate the nearest power of 2.
        (+[WebView _setCacheModel:]): Update memSize calculation to set it to the next
        bigger power of 2. Also update WebCore cache settings.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@70732 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2010-10-27 Eric Seidel <eric@webkit.org>
eric@webkit.org [Thu, 28 Oct 2010 00:05:07 +0000 (00:05 +0000)]
2010-10-27  Eric Seidel  <eric@webkit.org>

        Reviewed by Ojan Vafai.

        EWS bots should not use --quiet when running build-webkit
        https://bugs.webkit.org/show_bug.cgi?id=48482

        --quiet is only correct when the sub-process does the error reporting.
        In the case of _can_build() the parent process is reporting the error.
        We'd like the full build log at queues.webkit.org so someone can look
        at the log and understand why the EWS is failing to build trunk.

        * Scripts/webkitpy/layout_tests/port/chromium_win_unittest.py:
        * Scripts/webkitpy/tool/commands/earlywarningsystem.py:
        * Scripts/webkitpy/tool/commands/earlywarningsystem_unittest.py:
        * Scripts/webkitpy/tool/mocktool.py:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@70731 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoWebCore: Support Appkit key bindings and custom key bindings in WebKit2
enrica@apple.com [Thu, 28 Oct 2010 00:04:50 +0000 (00:04 +0000)]
WebCore: Support Appkit key bindings and custom key bindings in WebKit2
https://bugs.webkit.org/show_bug.cgi?id=48271
<rdar://problem/7660723>

Reviewed by Alexey Proskuryakov.

* dom/KeyboardEvent.h:
(WebCore::KeypressCommand::KeypressCommand): Added default constructor and modified assert for text
insertion command constructor to support WebKit2 scenarios.
The default constructor is needed to support the generated WebKit2 message processing code.

WebKit2: Support Appkit key bindings and custom key bindings in WebKit2
https://bugs.webkit.org/show_bug.cgi?id=48271
<rdar://problem/7660723>

Reviewed by Alexey Proskuryakov.

We need to support AppKit key bindings and custom key bindings in WebKit2.
Every keyboard event is first sent to the WebProcess and we inform AppKit that
we don't need further processing. When the event is processed by the WebProcess and no handler consumes it,
we send a synchronous message back to the UI process to let AppKit perform the key bindings.
This operation can result in one or more editing commands to execute or a noop.
The WebProcess then replies back to the UI process to inform whether further processing is required, in
which case the event is sent back to the application to be mapped.
There is a potential for a race condition: in case the WebProcess is very slow to process the key events, the UI
process could be in a completely different state when the event is being resent (for example it might not have the
keyboard focus anymore) and the command could be lost. We should look out for user experience impact.

* Scripts/webkit2/messages.py: Added header file in the generated file.
* Shared/WebCoreArgumentCoders.h: Added encoder/decoder for KeypressCommand.
* UIProcess/API/mac/PageClientImpl.h:
* UIProcess/API/mac/PageClientImpl.mm:
(WebKit::PageClientImpl::interceptKeyEvent): Pass-through call to WKView.
(WebKit::PageClientImpl::didNotHandleKeyEvent): Added logic to resend the event to the application.
* UIProcess/API/mac/WKView.mm:
(-[WKView validateUserInterfaceItem:]): Changed the default return value to YES, otherwise no menu shortcut is performed.
(-[WKView doCommandBySelector:]): Added.
(-[WKView insertText:]): Added.
(-[WKView _handleStyleKeyEquivalent:]): Added to handle command-B and command-I.
(-[WKView performKeyEquivalent:]): Added to intercept key binding sequences.
(-[WKView _setEventBeingResent:]):
(-[WKView _interceptKeyEvent:]):
* UIProcess/API/mac/WKViewInternal.h:
* UIProcess/PageClient.h:
* UIProcess/WebPageProxy.cpp:
(WebKit::WebPageProxy::interpretKeyEvent):
* UIProcess/WebPageProxy.h:
* UIProcess/WebPageProxy.messages.in:
* WebProcess/WebCoreSupport/WebEditorClient.cpp:
* WebProcess/WebCoreSupport/mac/WebEditorClientMac.mm:
(WebKit::WebEditorClient::handleKeyboardEvent):
(WebKit::WebEditorClient::handleInputMethodKeydown):
* WebProcess/WebPage/WebPage.cpp:
* WebProcess/WebPage/WebPage.h:
* WebProcess/WebPage/mac/WebPageMac.mm: Moved handleKeyboardEvent and handleInputMethodKeydown
to the Mac specific implementation.
(WebKit::WebPage::interceptEditingKeyboardEvent): Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@70730 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoWebKit2: More Windows build fixage. Copy another new header into WebKitOutputDir.
bweinstein@apple.com [Wed, 27 Oct 2010 23:53:06 +0000 (23:53 +0000)]
WebKit2: More Windows build fixage. Copy another new header into WebKitOutputDir.

* win/WebKit2Generated.make:

WebKitTools: More Windows build fixage. Rename a variable that was named string.

* WebKitTestRunner/InjectedBundle/InjectedBundlePage.cpp:
(WTR::hasPrefix):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@70728 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoUnreviewed build correction.
bfulgham@webkit.org [Wed, 27 Oct 2010 23:40:56 +0000 (23:40 +0000)]
Unreviewed build correction.

* wtf/Platform.h: Make sure ACCELERATED_COMPOSITING is
turned off in the WinCairo port.  This isn't supported (yet.)

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@70726 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoWindows build fix. Copy a new header into WebKitOutputDir.
bweinstein@apple.com [Wed, 27 Oct 2010 23:39:43 +0000 (23:39 +0000)]
Windows build fix. Copy a new header into WebKitOutputDir.

* win/WebKit2Generated.make:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@70725 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2010-10-27 Ojan Vafai <ojan@chromium.org>
ojan@chromium.org [Wed, 27 Oct 2010 23:33:26 +0000 (23:33 +0000)]
2010-10-27  Ojan Vafai  <ojan@chromium.org>

        Reviewed by Dimitri Glazkov.

        [chromium] Make the test results server store which master the bot is on
        https://bugs.webkit.org/show_bug.cgi?id=48478

        The chromium bots recently changed so that there are multiple slaves with
        the same name on different masters. Up till now, the test results server
        assumed slave names were unique. Adds a master field to the file in order
        to distinguish.

        Also, for files that currently lack a master or testtype, set them appropriately.

        * TestResultServer/handlers/testfilehandler.py:
        * TestResultServer/index.yaml:
        * TestResultServer/model/jsonresults.py:
        * TestResultServer/model/testfile.py:
        * TestResultServer/templates/showfilelist.html:
        * TestResultServer/templates/uploadform.html:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@70724 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2010-10-26 Darin Adler <darin@apple.com>
darin@apple.com [Wed, 27 Oct 2010 23:18:01 +0000 (23:18 +0000)]
2010-10-26  Darin Adler  <darin@apple.com>

        Reviewed by Sam Weinig.

        WebKitTestRunner needs to support layoutTestController.dumpBackForwardList
        https://bugs.webkit.org/show_bug.cgi?id=42322
        rdar://problem/8193631

        WebKitTestRunner needs to support layoutTestController.clearBackForwardList
        https://bugs.webkit.org/show_bug.cgi?id=42333
        rdar://problem/8193643

        * History/WebBackForwardList.mm:
        (core): Return BackForwardListImpl.
        (kit): Take BackForwardListImpl.
        (-[WebBackForwardList initWithBackForwardList:]): Use BackForwardListImpl.
        (-[WebBackForwardList dealloc]): Ditto.
        (-[WebBackForwardList finalize]): Ditto.
        (-[WebBackForwardList description]): Ditto.
        (-[WebBackForwardList setPageCacheSize:]): Ditto.
        (-[WebBackForwardList pageCacheSize]): Ditto.
        * History/WebBackForwardListInternal.h: Ditto.
        * WebView/WebFrameView.mm:
        (-[WebFrameView keyDown:]): Ditto.
        * WebView/WebView.mm:
        (-[WebView initWithCoder:]): Ditto.
        (-[WebView encodeWithCoder:]): Ditto.
        (-[WebView backForwardList]): Ditto.
        (-[WebView setMaintainsBackForwardList:]): Ditto.
2010-10-26  Darin Adler  <darin@apple.com>

        Reviewed by Sam Weinig.

        WebKitTestRunner needs to support layoutTestController.dumpBackForwardList
        https://bugs.webkit.org/show_bug.cgi?id=42322
        rdar://problem/8193631

        WebKitTestRunner needs to support layoutTestController.clearBackForwardList
        https://bugs.webkit.org/show_bug.cgi?id=42333
        rdar://problem/8193643

        * WebCore.exp.in: Updated exports.

        * history/BackForwardList.h: Added comments about future refactoring.
        Moved functions that are not called inside WebCore into a separate
        section, to be removed later. Added an isActive function to replace the
        use of enabled() and entries().

        * history/BackForwardListImpl.h: Removed isBackForwardListImpl.
        Marked virtual function overrides virtual explicitly, except for the
        ones that will be non-virtual once we make the changes above.
        Added the isActive function.

        * history/PageCache.cpp:
        (WebCore::logCanCachePageDecision): Use isActive.
        (WebCore::PageCache::canCache): Ditto.

        * loader/FrameLoader.cpp:
        (WebCore::FrameLoader::checkDidPerformFirstNavigation): Check
        using currentItem, backItem, and forwardItem instead of
        using entries.

        * page/Settings.cpp:
        (WebCore::Settings::setUsesPageCache): Loop through the back/forward
        list entries using backListCount and forwardListCount instead of
        using entries.
2010-10-26  Darin Adler  <darin@apple.com>

        Reviewed by Sam Weinig.

        WebKitTestRunner needs to support layoutTestController.dumpBackForwardList
        https://bugs.webkit.org/show_bug.cgi?id=42322
        rdar://problem/8193631

        WebKitTestRunner needs to support layoutTestController.clearBackForwardList
        https://bugs.webkit.org/show_bug.cgi?id=42333
        rdar://problem/8193643

        * platform/mac-wk2/Skipped: Resorted some test lists. Removed the section
        listing all the tests that were blocked by dumpBackForwardList.
2010-10-26  Darin Adler  <darin@apple.com>

        Reviewed by Sam Weinig.

        WebKitTestRunner needs to support layoutTestController.dumpBackForwardList
        https://bugs.webkit.org/show_bug.cgi?id=42322
        rdar://problem/8193631

        WebKitTestRunner needs to support layoutTestController.clearBackForwardList
        https://bugs.webkit.org/show_bug.cgi?id=42333
        rdar://problem/8193643

        * WebKitTestRunner/Configurations/InjectedBundle.xcconfig:
        Renamed the product to WebKitTestRunnerInjectedBundle to avoid
        name conflicts in the build directory.

        * WebKitTestRunner/InjectedBundle/Bindings/LayoutTestController.idl:
        Added dumpBackForwardList and clearBackForwardList.

        * WebKitTestRunner/InjectedBundle/InjectedBundle.cpp:
        (WTR::InjectedBundle::InjectedBundle): Removed initialization of m_mainPage.
        (WTR::InjectedBundle::didCreatePage): Removed unneeded underscore.
        (WTR::InjectedBundle::willDestroyPage): Ditto.
        (WTR::InjectedBundle::didReceiveMessage): Ditto.
        (WTR::InjectedBundle::initialize): Ditto.
        (WTR::InjectedBundle::didCreatePage): Changed code to use m_pages
        instead of m_mainPage and m_otherPages.
        (WTR::InjectedBundle::willDestroyPage): Ditto.
        (WTR::InjectedBundle::page): Ditto.
        (WTR::InjectedBundle::beginTesting): Ditto.
        (WTR::InjectedBundle::done): Ditto.
        (WTR::InjectedBundle::closeOtherPages): Ditto.
        (WTR::InjectedBundle::dumpBackForwardListsForAllPages): Added.

        * WebKitTestRunner/InjectedBundle/InjectedBundle.h: Changed
        page function to longer be inline, and pageCount function
        to use m_pages. Added dumpBackForwardListsForAllPages, and
        removed some underscores. Replaced m_mainPage and m_otherPageas
        with m_pages.

        * WebKitTestRunner/InjectedBundle/InjectedBundlePage.cpp:
        (WTR::adoptWK): Added.
        (WTR::hasPrefix): Added.
        (WTR::InjectedBundlePage::reset): Added code to set up
        m_previousTestBackForwardListItem.
        (WTR::InjectedBundlePage::dump): Added code to call
        dumpBackForwardListsForAllPages.
        (WTR::compareByTargetName): Added.
        (WTR::dumpBackForwardListItem): Added.
        (WTR::InjectedBundlePage::dumpBackForwardList): Added.

        * WebKitTestRunner/InjectedBundle/InjectedBundlePage.h: Added
        dumpBackForwardList and m_previousTestBackForwardListItem.

        * WebKitTestRunner/InjectedBundle/LayoutTestController.cpp:
        (WTR::LayoutTestController::LayoutTestController): Initialize
        m_shouldDumpBackForwardListsForAllWindows to false.
        (WTR::LayoutTestController::clearBackForwardList): Added.

        * WebKitTestRunner/InjectedBundle/LayoutTestController.h:
        Added dumpBackForwardList, clearBackForwardList,
        shouldDumpBackForwardListsForAllWindows, and
        m_shouldDumpBackForwardListsForAllWindows.

        * WebKitTestRunner/TestController.cpp:
        (WTR::TestController::initialize): Set up
        didReceiveSynchronousMessageFromInjectedBundle.
        (WTR::TestController::didReceiveSynchronousMessageFromInjectedBundle):
        Added.
        * WebKitTestRunner/TestController.h: Ditto.

        * WebKitTestRunner/TestInvocation.cpp:
        (WTR::TestInvocation::didReceiveSynchronousMessageFromInjectedBundle):
        Added.
        * WebKitTestRunner/TestInvocation.h: Ditto.

        * WebKitTestRunner/WebKitTestRunner.xcodeproj/project.pbxproj:
        Renamed bundle to WebKitTestRunnerInjectedBundle (see above).
        * WebKitTestRunner/mac/TestControllerMac.mm:
        (WTR::TestController::initializeInjectedBundlePath): Ditto.
2010-10-26  Darin Adler  <darin@apple.com>

        Reviewed by Sam Weinig.

        WebKitTestRunner needs to support layoutTestController.dumpBackForwardList
        https://bugs.webkit.org/show_bug.cgi?id=42322
        rdar://problem/8193631

        WebKitTestRunner needs to support layoutTestController.clearBackForwardList
        https://bugs.webkit.org/show_bug.cgi?id=42333
        rdar://problem/8193643

        * Shared/API/c/WKBase.h: Added WKBundleBackForwardListRef and
        WKBundleBackForwardListItemRef.

        * Shared/API/c/WKURL.cpp:
        (WKURLCopyString): Added.
        * Shared/API/c/WKURL.h: Added WKURLCopyString.

        * Shared/API/c/cf/WKURLCF.h: Fixed incorrect argument name.

        * Shared/APIObject.h: Added TypeBundleBackForwardList and
        TypeBundleBackForwardListItem.

        * Shared/ImmutableArray.h: Marked destructor virtual explicitly.
        * Shared/ImmutableDictionary.h: Ditto.
        * UIProcess/WebBackForwardListItem.h: Ditto.
        * UIProcess/WebContext.h: Ditto.
        * UIProcess/WebFrameProxy.h: Ditto.
        * UIProcess/WebNavigationData.h: Ditto.
        * UIProcess/WebPageNamespace.h: Ditto.
        * UIProcess/WebPreferences.h: Ditto.
        * WebProcess/InjectedBundle/InjectedBundleScriptWorld.h: Ditto.
        * WebProcess/InjectedBundle/DOM/InjectedBundleNodeHandle.h: Ditto.
        * WebProcess/InjectedBundle/DOM/InjectedBundleRangeHandle.h: Ditto.

        * UIProcess/WebBackForwardList.cpp: Removed unneeded backListWithLimit
        and forwardListWithLimit functions.
        (WebKit::WebBackForwardList::clear): Added.
        * UIProcess/WebBackForwardList.h: Marked destructor virtual explicitly.
        Added clear, removed backListWithLimit and forwardListWithLimit functions.

        * UIProcess/WebPageProxy.cpp: Removed unneeded backForwardBackItem,
        backForwardCurrentItem, and backForwardForwardItem functions.
        (WebKit::WebPageProxy::backForwardClear): Added.
        * UIProcess/WebPageProxy.h: Ditto.
        * UIProcess/WebPageProxy.messages.in: Ditto.

        * WebKit2.pro: Added new files.
        * WebKit2.xcodeproj/project.pbxproj: Ditto.
        * win/WebKit2.vcproj: Ditto.

        * WebProcess/InjectedBundle/API/c/WKBundleAPICast.h: Added
        WKBundleBackForwardListRef and WKBundleBackForwardListItemRef.

        * WebProcess/InjectedBundle/API/c/WKBundleBackForwardList.cpp:
        Started with a copy of from WebKit2/UIProcess/API/C/WKBackForwardList.cpp.
        * WebProcess/InjectedBundle/API/c/WKBundleBackForwardList.h:
        Started with a copy of WebKit2/UIProcess/API/C/WKBackForwardList.h.

        * WebProcess/InjectedBundle/API/c/WKBundleBackForwardListItem.cpp:
        Started with a copy of WebKit2/UIProcess/API/C/WKBackForwardListItem.cpp.
        * WebProcess/InjectedBundle/API/c/WKBundleBackForwardListItem.h:
        Started with a copy of WebKit2/UIProcess/API/C/WKBackForwardListItem.h.

        * WebProcess/InjectedBundle/API/c/WKBundlePage.cpp:
        (WKBundlePageGetBackForwardList): Added.
        * WebProcess/InjectedBundle/API/c/WKBundlePage.h: Ditto.

        * WebProcess/InjectedBundle/InjectedBundleBackForwardList.cpp:
        Started with a copy of WebKit2/UIProcess/WebBackForwardList.cpp.
        * WebProcess/InjectedBundle/InjectedBundleBackForwardList.h:
        Started with a copy of WebKit2/UIProcess/WebBackForwardList.h.

        * WebProcess/InjectedBundle/InjectedBundleBackForwardListItem.cpp:
        Started with a copy of WebKit2/UIProcess/WebBackForwardListItem.cpp.
        * WebProcess/InjectedBundle/InjectedBundleBackForwardListItem.h:
        Started with a copy of WebKit2/UIProcess/WebBackForwardListItem.h.

        * WebProcess/WebPage/WebBackForwardListProxy.cpp: Removed
        many unneeded functions.
        (WebKit::WebBackForwardListProxy::WebBackForwardListProxy):
        Removed unneeded initialization of m_closed.
        (WebKit::WebBackForwardListProxy::addItem): Added a check for
        the case of a capacity of 0.
        (WebKit::WebBackForwardListProxy::goToItem): Added a check for
        when the proxy outlives its page.
        (WebKit::WebBackForwardListProxy::itemAtIndex): Ditto.
        (WebKit::WebBackForwardListProxy::backListCount): Ditto.
        (WebKit::WebBackForwardListProxy::forwardListCount): Ditto.
        (WebKit::WebBackForwardListProxy::close): Removed unneeded code
        setting m_closed.
        (WebKit::WebBackForwardListProxy::isActive): Added.
        (WebKit::WebBackForwardListProxy::clear): Added.

        * WebProcess/WebPage/WebBackForwardListProxy.h: Added clear,
        isActive, and detach functions. Removed many other unneeded
        function and data members.

        * WebProcess/WebPage/WebPage.cpp:
        (WebKit::WebPage::~WebPage): Added call to detach on the
        back/forward list.
        (WebKit::WebPage::backForwardList): Added.
        * WebProcess/WebPage/WebPage.h: Added backForwardList function
        and m_backForwardList data member.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@70723 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2010-10-27 Chris Rogers <crogers@google.com>
crogers@google.com [Wed, 27 Oct 2010 22:44:15 +0000 (22:44 +0000)]
2010-10-27  Chris Rogers  <crogers@google.com>

        Reviewed by Chris Marrin.

        Add ENABLE_WEB_AUDIO feature enable flag (initially disabled) to build-webkit
        https://bugs.webkit.org/show_bug.cgi?id=48279

        * Configurations/FeatureDefines.xcconfig:
2010-10-27  Chris Rogers  <crogers@google.com>

        Reviewed by Chris Marrin.

        Add ENABLE_WEB_AUDIO feature enable flag (initially disabled) to build-webkit
        https://bugs.webkit.org/show_bug.cgi?id=48279

        No new tests since flag is disabled.

        * Configurations/FeatureDefines.xcconfig:
2010-10-27  Chris Rogers  <crogers@google.com>

        Reviewed by Chris Marrin.

        Add ENABLE_WEB_AUDIO feature enable flag (initially disabled) to build-webkit
        https://bugs.webkit.org/show_bug.cgi?id=48279

        * Configurations/FeatureDefines.xcconfig:
2010-10-27  Chris Rogers  <crogers@google.com>

        Reviewed by Chris Marrin.

        Add ENABLE_WEB_AUDIO feature enable flag (initially disabled) to build-webkit
        https://bugs.webkit.org/show_bug.cgi?id=48279

        * Scripts/build-webkit:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@70722 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2010-10-27 Martin Robinson <mrobinson@igalia.com>
mrobinson@webkit.org [Wed, 27 Oct 2010 22:38:31 +0000 (22:38 +0000)]
2010-10-27  Martin Robinson  <mrobinson@igalia.com>

        Rebaseline some GTK+ tests after r70688.

        * platform/gtk/fast/css/font-face-opentype-expected.txt:
        * platform/gtk/fast/forms/select-visual-hebrew-expected.txt:
        * platform/gtk/fast/forms/select-writing-direction-natural-expected.txt:
        * platform/gtk/fast/forms/visual-hebrew-text-field-expected.txt:
        * platform/gtk/fast/text/backslash-to-yen-sign-euc-expected.txt:
        * platform/gtk/fast/text/backslash-to-yen-sign-expected.txt:
        * platform/gtk/fast/text/international/plane2-expected.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@70721 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2010-10-27 Eric Seidel <eric@webkit.org>
eric@webkit.org [Wed, 27 Oct 2010 22:37:43 +0000 (22:37 +0000)]
2010-10-27  Eric Seidel  <eric@webkit.org>

        Unreviewed.

        build-webkit should collect Visual Studio Express logs and display them
        https://bugs.webkit.org/show_bug.cgi?id=39199

        It turns out my previous patch wasn't actually working on the win-ews
        machine.  So I've fixed my mistakes from before.

        * Scripts/build-webkit:
         - Windows VSE builds change the CWD while building.  Why?  Who knows.
        * Scripts/print-vse-failure-logs:
         - windows VSE builds don't use Debug/Release as I expected, so I've
           moved off of --configuration to --top-level and added the /obj
           optimization while I was there.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@70720 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2010-10-27 Chris Rogers <crogers@google.com>
crogers@google.com [Wed, 27 Oct 2010 22:28:57 +0000 (22:28 +0000)]
2010-10-27  Chris Rogers  <crogers@google.com>

        Reviewed by Kenneth Russell.

        Add AudioResamplerKernel files
        https://bugs.webkit.org/show_bug.cgi?id=47624

        No new tests since audio API is not yet implemented.

        * platform/audio/AudioResamplerKernel.cpp: Added.
        (WebCore::AudioResamplerKernel::AudioResamplerKernel):
        (WebCore::AudioResamplerKernel::getSourcePointer):
        (WebCore::AudioResamplerKernel::process):
        (WebCore::AudioResamplerKernel::reset):
        (WebCore::AudioResamplerKernel::rate):
        * platform/audio/AudioResamplerKernel.h: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@70719 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2010-10-27 Adam Barth <abarth@webkit.org>
abarth@webkit.org [Wed, 27 Oct 2010 21:57:24 +0000 (21:57 +0000)]
2010-10-27  Adam Barth  <abarth@webkit.org>

        Reviewed by Ojan Vafai.

        REGRESSION(66511): CORDYS Process Factory doesn't work
        https://bugs.webkit.org/show_bug.cgi?id=48453

        Test some appendChild cases that didn't have coverage before.

        * fast/dom/allowed-children-expected.txt: Added.
        * fast/dom/allowed-children.html: Added.
2010-10-27  Adam Barth  <abarth@webkit.org>

        Reviewed by Ojan Vafai.

        REGRESSION(66511): CORDYS Process Factory doesn't work
        https://bugs.webkit.org/show_bug.cgi?id=48453

        When Ojan refactored this function in 66511, he missed an "else",
        which causes us to generated HIERARCHY_REQUEST_ERR in some cases where
        we're not supposed to.  This patch restores the "else" behavior (using
        an early return) and adds a test for the cases that caused problems in
        the field.

        Test: fast/dom/allowed-children.html

        * dom/Node.cpp:
        (WebCore::isChildTypeAllowed):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@70718 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2010-10-27 Kenneth Russell <kbr@google.com>
kbr@google.com [Wed, 27 Oct 2010 21:53:07 +0000 (21:53 +0000)]
2010-10-27  Kenneth Russell  <kbr@google.com>

        Reviewed by Dimitri Glazkov.

        [chromium] WebGL does not work with the compositor in test_shell
        https://bugs.webkit.org/show_bug.cgi?id=48470

        * src/WebGraphicsContext3DDefaultImpl.cpp:
        (WebKit::WebGraphicsContext3DDefaultImpl::resolveMultisampledFramebuffer):
         - Properly restore the draw framebuffer after resolving.
        (WebKit::WebGraphicsContext3DDefaultImpl::prepareTexture):
         - Make the context current before calling resolveMultisampledFramebuffer.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@70717 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoFix for https://bugs.webkit.org/show_bug.cgi?id=48385 Add WebKit
bdakin@apple.com [Wed, 27 Oct 2010 21:46:36 +0000 (21:46 +0000)]
Fix for https://bugs.webkit.org/show_bug.cgi?id=48385 Add WebKit
SPI to scale a WebView
-and corresponding-
<rdar://problem/8107667>

Reviewed by Darin Adler.

WebCore:

This patch adds SPI to Mac WebKit that scales the page by the given
scale factor. The scale factor is factored into the RenderStyle's
transform.

Set the current pageScale for the document.
* css/CSSStyleSelector.cpp:
(WebCore::CSSStyleSelector::styleForDocument):

Frame keeps track of the page scale factor. When
scalePage() is called, it modifies the RenderView's style to
include a transform of the appropriate scale and forces a style
recalc.
* page/Frame.cpp:
(WebCore::Frame::Frame):
(WebCore::Frame::scalePage):
* page/Frame.h:
(WebCore::Frame::pageScaleFactor):
* WebCore.exp.in:

Add a scale to the transformation matrix.
* rendering/style/RenderStyle.cpp:
(WebCore::RenderStyle::setPageScaleTransform):
* rendering/style/RenderStyle.h:

WebKit/mac:

This patch adds SPI to Mac WebKit that scales the page by the given
scale factor.

* WebView/WebView.mm:
(-[WebView _scaleWebView:]):
(-[WebView _viewScaleFactor]):
* WebView/WebViewPrivate.h:

WebKit2:

This patch adds SPI to Mac WebKit that scales the page by the given
scale factor.

Implement scaleWebView(), which tells WebCore::Frame to scale, and
viewScaleFactor() which returns the current scale factor.
* Shared/API/c/WKSharedAPICast.h:
* UIProcess/API/C/WKPage.cpp:
(WKPageScaleWebView):
(WKPageGetViewScaleFactor):
* UIProcess/API/C/WKPage.h:
* UIProcess/WebPageProxy.cpp:
(WebKit::WebPageProxy::WebPageProxy):
(WebKit::WebPageProxy::scaleWebView):
* UIProcess/WebPageProxy.h:
(WebKit::WebPageProxy::viewScaleFactor):
* WebProcess/WebPage/WebPage.cpp:
(WebKit::WebPage::scaleWebView):
(WebKit::WebPage::viewScaleFactor):
* WebProcess/WebPage/WebPage.h:
* WebProcess/WebPage/WebPage.messages.in:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@70716 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoAdd Windows-specific results for test added in r70692
aroben@apple.com [Wed, 27 Oct 2010 21:45:55 +0000 (21:45 +0000)]
Add Windows-specific results for test added in r70692

This test isn't rendering correctly. That's covered by
<http://webkit.org/b/48459>.

* platform/win/fast/blockflow/border-vertical-lr-expected.checksum: Added.
* platform/win/fast/blockflow/border-vertical-lr-expected.png: Added.
* platform/win/fast/blockflow/border-vertical-lr-expected.txt: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@70715 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoREGRESSION (r70335): Incorrect article layout in Safari Reader
mitz@apple.com [Wed, 27 Oct 2010 21:38:46 +0000 (21:38 +0000)]
REGRESSION (r70335): Incorrect article layout in Safari Reader
https://bugs.webkit.org/show_bug.cgi?id=48436

Reviewed by Darin Adler.

WebCore:

* page/DOMWindow.cpp:
(WebCore::DOMWindow::getMatchedCSSRules): Return rules from all origins if the cross-origin
check is disabled.
* page/Settings.cpp:
(WebCore::Settings::Settings): Initialize m_crossOriginCheckInGetMatchedCSSRulesDisabled.
* page/Settings.h:
(WebCore::Settings::setCrossOriginCheckInGetMatchedCSSRulesDisabled): Added this accessor.
(WebCore::Settings::crossOriginCheckInGetMatchedCSSRulesDisabled): Ditto.

WebKit/mac:

* Misc/WebKitVersionChecks.h: Added WEBKIT_FIRST_VERSION_WITH_GET_MATCHED_CSS_RULES_RESTRICTIONS.
* WebView/WebView.mm:
(-[WebView _needsUnrestrictedGetMatchedCSSRules]): Added. Returns YES for Safari linked against
versions of WebKit that did not have the cross-origin restrictions in getMatchedCSSRules().
(-[WebView _preferencesChangedNotification:]): Call Settings::setCrossOriginCheckInGetMatchedCSSRulesDisabled()
based on the above.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@70714 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2010-10-27 Stuart Morgan <stuartmorgan@chromium.org>
commit-queue@webkit.org [Wed, 27 Oct 2010 21:38:41 +0000 (21:38 +0000)]
2010-10-27  Stuart Morgan  <stuartmorgan@chromium.org>

        Reviewed by David Levin.

        Include caps lock state when converting NSEvent modifiers to
        WebInputEvent

        https://bugs.webkit.org/show_bug.cgi?id=47917

        * src/mac/WebInputEventFactory.mm:
        (WebKit::modifiersFromEvent):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@70713 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2010-10-27 Chris Rogers <crogers@google.com>
crogers@google.com [Wed, 27 Oct 2010 21:35:42 +0000 (21:35 +0000)]
2010-10-27  Chris Rogers  <crogers@google.com>

        Reviewed by James Robinson.

        Add ENABLE_WEB_AUDIO feature enable flag (initially disabled) for Chromium
        https://bugs.webkit.org/show_bug.cgi?id=48465

        * features.gypi:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@70712 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoWindows build fix.
bweinstein@apple.com [Wed, 27 Oct 2010 21:33:58 +0000 (21:33 +0000)]
Windows build fix.

* jit/JITStubs.cpp:
(JSC::jitThrow):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@70711 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoUnreviewed. Add Qt specific expected results after r70657.
ossy@webkit.org [Wed, 27 Oct 2010 21:27:17 +0000 (21:27 +0000)]
Unreviewed. Add Qt specific expected results after r70657.

* platform/qt/animations/missing-values-first-keyframe-expected.txt: Added.
* platform/qt/animations/missing-values-last-keyframe-expected.txt: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@70710 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2010-10-27 Dimitri Glazkov <dglazkov@chromium.org>
dglazkov@chromium.org [Wed, 27 Oct 2010 21:22:48 +0000 (21:22 +0000)]
2010-10-27  Dimitri Glazkov  <dglazkov@chromium.org>

        [Chromium] Add expectations after r70692.

        * platform/chromium-linux/fast/borders/border-radius-inline-flow-expected.checksum: Added.
        * platform/chromium-linux/fast/borders/border-radius-inline-flow-expected.png: Added.
        * platform/chromium-mac/fast/borders/border-radius-inline-flow-expected.checksum: Added.
        * platform/chromium-win/fast/borders/border-radius-inline-flow-expected.checksum: Added.
        * platform/chromium-win/fast/borders/border-radius-inline-flow-expected.png: Added.
        * platform/chromium-win/fast/borders/border-radius-inline-flow-expected.txt: Added.
        * platform/chromium/test_expectations.txt: Tweaked.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@70709 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2010-10-18 Jer Noble <jer.noble@apple.com>
jer.noble@apple.com [Wed, 27 Oct 2010 21:21:02 +0000 (21:21 +0000)]
2010-10-18  Jer Noble  <jer.noble@apple.com>

        Reviewed by Eric Carlson.

        Safari fullscreen media element tickles when it should not.

        https://bugs.webkit.org/show_bug.cgi?id=47861

        * WebView/WebVideoFullscreenController.mm:
        (-[WebVideoFullscreenController updatePowerAssertions]):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@70708 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoRemove som unnecessary Windows-specific results
aroben@apple.com [Wed, 27 Oct 2010 21:16:21 +0000 (21:16 +0000)]
Remove som unnecessary Windows-specific results

These should have been removed back in r70630.

* platform/win/fast/blockflow/english-lr-text-expected.checksum: Removed.
* platform/win/fast/blockflow/english-lr-text-expected.png: Removed.
* platform/win/fast/blockflow/english-lr-text-expected.txt: Removed.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@70707 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2010-10-27 Zhenyao Mo <zmo@google.com>
zmo@google.com [Wed, 27 Oct 2010 21:09:18 +0000 (21:09 +0000)]
2010-10-27  Zhenyao Mo  <zmo@google.com>

        Reviewed by Kenneth Russell.

        refactor the nested large switch statements in GraphicsContext3DCG.cpp:getImageData()
        https://bugs.webkit.org/show_bug.cgi?id=47027

        * platform/graphics/GraphicsContext3D.cpp:
        (WebCore::doPacking): ASSERT false if undefined format is passed in.
        * platform/graphics/GraphicsContext3D.h: Add kSourceFormatUndefined enum.
        * platform/graphics/cg/GraphicsContext3DCG.cpp:
        (WebCore::getSourceDataFormat): Decide source data format from componentsPerPixel, alpha format, bitsPerComponet, etc.
        (WebCore::GraphicsContext3D::getImageData): Refactor the code to use getSourceDataFormat and remove nested switches.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@70706 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2010-10-27 Patrick Gansterer <paroga@webkit.org>
paroga@webkit.org [Wed, 27 Oct 2010 21:05:31 +0000 (21:05 +0000)]
2010-10-27  Patrick Gansterer  <paroga@webkit.org>

        Reviewed by Adam Barth.

        Use WTF::StringHasher::createBlobHash where possible
        https://bugs.webkit.org/show_bug.cgi?id=48228

        Replace StringImpl::computeHash with StringHasher::createBlobHash.
        This removes many duplicated code.

        * bindings/js/JSSVGPODTypeWrapper.h:
        (WebCore::PODTypeWrapperCacheInfoHash::hash):
        * page/SecurityOriginHash.h:
        (WebCore::SecurityOriginHash::hash):
        * platform/cf/SchedulePair.h:
        (WebCore::SchedulePairHash::hash):
        * platform/graphics/FontCache.cpp:
        (WebCore::computeHash):
        * platform/graphics/chromium/FontPlatformDataChromiumWin.h:
        (WebCore::FontPlatformData::RefCountedHFONT::hash):
        * platform/graphics/cocoa/FontPlatformData.h:
        (WebCore::FontPlatformData::hash):
        * platform/graphics/gtk/FontPlatformDataPango.h:
        (WebCore::FontPlatformData::hash):
        * platform/graphics/wince/FontPlatformData.cpp:
        (WebCore::FixedSizeFontDataKeyHash::hash):
        * platform/graphics/wx/FontPlatformDataWx.cpp:
        (WebCore::FontPlatformData::FontPlatformData):
        (WebCore::FontPlatformData::computeHash):
        * plugins/PluginPackage.cpp:
        (WebCore::PluginPackage::hash):
        * plugins/win/PluginPackageWin.cpp:
        (WebCore::PluginPackage::hash):
        * svg/DeprecatedSVGAnimatedTemplate.h:
        (WebCore::DeprecatedSVGAnimatedTypeWrapperKeyHash::hash):
        * svg/properties/SVGAnimatedPropertyDescription.h:
        (WebCore::SVGAnimatedPropertyDescriptionHash::hash):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@70705 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoBug 48365 - Remove output parameters from JITStackFrame
barraclough@apple.com [Wed, 27 Oct 2010 20:46:09 +0000 (20:46 +0000)]
Bug 48365 - Remove output parameters from JITStackFrame

Reviewed by Oliver Hunt.

The JIT stub functions presently use the stackframe to provide a couple of additional return values.
  * In the case of uncaught exceptions the exception value is returned on the stackframe.exception property.
  * In the case of caught exceptions the updated value for the callFrame register is returned on the stackframe.callFrame property.

Change exception returns such that exceptions are always returned on JSGlobalData::exception.
Change op_catch such that the new CallFrame value is returned from op_throw / vm_throw in regT0.

* JavaScriptCore.xcodeproj/project.pbxproj:
* debugger/Debugger.cpp:
(JSC::evaluateInGlobalCallFrame):
* debugger/DebuggerCallFrame.cpp:
(JSC::DebuggerCallFrame::evaluate):
* interpreter/CachedCall.h:
(JSC::CachedCall::CachedCall):
(JSC::CachedCall::call):
* interpreter/CallFrame.h:
(JSC::ExecState::exception):
* interpreter/Interpreter.cpp:
(JSC::Interpreter::callEval):
(JSC::Interpreter::Interpreter):
(JSC::Interpreter::execute):
(JSC::Interpreter::executeCall):
(JSC::Interpreter::executeConstruct):
(JSC::Interpreter::prepareForRepeatCall):
(JSC::Interpreter::privateExecute):
* interpreter/Interpreter.h:
* jit/JITCode.h:
(JSC::JITCode::execute):
* jit/JITOpcodes.cpp:
(JSC::JIT::emit_op_catch):
* jit/JITOpcodes32_64.cpp:
(JSC::JIT::emit_op_catch):
* jit/JITStubs.cpp:
(JSC::ctiTrampoline):
(JSC::jitThrow):
(JSC::DEFINE_STUB_FUNCTION):
* jit/JITStubs.h:
* runtime/ArrayPrototype.cpp:
(JSC::arrayProtoFuncFilter):
(JSC::arrayProtoFuncMap):
(JSC::arrayProtoFuncEvery):
(JSC::arrayProtoFuncForEach):
(JSC::arrayProtoFuncSome):
(JSC::arrayProtoFuncReduce):
(JSC::arrayProtoFuncReduceRight):
* runtime/CallData.cpp:
(JSC::call):
* runtime/Completion.cpp:
(JSC::evaluate):
* runtime/ConstructData.cpp:
(JSC::construct):
* runtime/ExceptionHelpers.cpp:
(JSC::createErrorForInvalidGlobalAssignment):
(JSC::throwOutOfMemoryError):
(JSC::throwStackOverflowError):
* runtime/ExceptionHelpers.h:
* runtime/JSArray.cpp:
(JSC::JSArray::sort):
* runtime/JSGlobalObjectFunctions.cpp:
(JSC::globalFuncEval):
* runtime/StringPrototype.cpp:
(JSC::stringProtoFuncReplace):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@70703 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoFix the WebKit2 API tests.
jberlin@webkit.org [Wed, 27 Oct 2010 20:38:36 +0000 (20:38 +0000)]
Fix the WebKit2 API tests.
https://bugs.webkit.org/show_bug.cgi?id=48461

Reviewed by Sam Weinig.

The API tests were failing because the tests expect to be able to send messages before the
WebProcess finishes launching.

Instead of dispatching the pending messages in processDidFinishLaunching, dispatch them in
ensureWebProcess but do not make messages pending if the process is launching.

* UIProcess/WebContext.cpp:
(WebKit::WebContext::processDidFinishLaunching):
Move dispatching the pending messages back from here ...
(WebKit::WebContext::ensureWebProcess):
... to here.
(WebKit::WebContext::postMessageToInjectedBundle):
Check whether the process can send messages in order to determine if a message needs to be
sent later.

* UIProcess/WebProcessProxy.h:
(WebKit::WebProcessProxy::canSendMessage):
The WebProcessProxy can send a message if it is valid or if it is launching.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@70702 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2010-10-27 Tony Chang <tony@chromium.org>
tony@chromium.org [Wed, 27 Oct 2010 20:34:33 +0000 (20:34 +0000)]
2010-10-27  Tony Chang  <tony@chromium.org>

        Unreviewed, build fix after r70692.

        * rendering/RenderBoxModelObject.cpp:
        (WebCore::RenderBoxModelObject::paintBorder):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@70701 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2010-10-27 Simon Fraser <simon.fraser@apple.com>
simon.fraser@apple.com [Wed, 27 Oct 2010 20:32:51 +0000 (20:32 +0000)]
2010-10-27  Simon Fraser  <simon.fraser@apple.com>

        Remove Leopard-specific results that are no longer required.

        * platform/mac-leopard/fast/text/international/vertical-text-glyph-test-expected.txt: Removed.
        * platform/mac-leopard/fast/text/international/vertical-text-metrics-test-expected.txt: Removed.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@70700 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2010-10-27 Gabor Loki <loki@webkit.org>
loki@webkit.org [Wed, 27 Oct 2010 20:32:21 +0000 (20:32 +0000)]
2010-10-27  Gabor Loki  <loki@webkit.org>

        Reviewed by Oliver Hunt.

        https://bugs.webkit.org/show_bug.cgi?id=48060
        Speed up op_jeq_null and op_jneq_null.

        For both opcodes the NullTag and UndefinedTag are checked to control the
        jump. These values can be simply checked by AboveOrEqual or Below
        condition if they are the two highest unsigned integers from JSValue's
        Tag field.

        * jit/JITOpcodes32_64.cpp:
        (JSC::JIT::emit_op_jeq_null):
        (JSC::JIT::emit_op_jneq_null):
        * runtime/JSValue.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@70699 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoAdd Windows-specific results for tests added in r70664
aroben@apple.com [Wed, 27 Oct 2010 20:16:10 +0000 (20:16 +0000)]
Add Windows-specific results for tests added in r70664

These tests aren't rendering correctly. That's covered by
<http://webkit.org/b/48459>.

* platform/win/fast/blockflow/japanese-lr-text-expected.checksum: Added.
* platform/win/fast/blockflow/japanese-lr-text-expected.png: Added.
* platform/win/fast/blockflow/japanese-lr-text-expected.txt: Added.
* platform/win/fast/blockflow/japanese-rl-text-expected.checksum: Added.
* platform/win/fast/blockflow/japanese-rl-text-expected.png: Added.
* platform/win/fast/blockflow/japanese-rl-text-expected.txt: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@70698 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2010-10-27 Leandro Pereira <leandro@profusion.mobi>
leandro@webkit.org [Wed, 27 Oct 2010 20:15:44 +0000 (20:15 +0000)]
2010-10-27  Leandro Pereira  <leandro@profusion.mobi>

        Unreviewed. Build fix after r70651.

        * CMakeListsEfl.txt: Add SoupURIUtils.cpp.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@70697 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoWebPageProxy::isValid should return false if the page has been explicitly closed
andersca@apple.com [Wed, 27 Oct 2010 20:10:20 +0000 (20:10 +0000)]
WebPageProxy::isValid should return false if the page has been explicitly closed
https://bugs.webkit.org/show_bug.cgi?id=48458

Reviewed by Sam Weinig.

* UIProcess/WebPageProxy.cpp:
(WebKit::WebPageProxy::WebPageProxy):
(WebKit::WebPageProxy::isValid):
(WebKit::WebPageProxy::relaunch):
(WebKit::WebPageProxy::close):
(WebKit::WebPageProxy::processDidCrash):
* UIProcess/WebPageProxy.h:
(WebKit::WebPageProxy::isClosed):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@70696 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2010-10-27 Crystal Zhang <haizhang@rim.com>
commit-queue@webkit.org [Wed, 27 Oct 2010 19:53:07 +0000 (19:53 +0000)]
2010-10-27  Crystal Zhang  <haizhang@rim.com>

        Reviewed by Eric Carlson.

        https://bugs.webkit.org/show_bug.cgi?id=48446

        Fix an issue where we were passing the wrong parameter
        to MIMETypeRegistry::getMIMETypeForExtension().

        * platform/MIMETypeRegistry.cpp:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@70695 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agohttps://bugs.webkit.org/show_bug.cgi?id=48449
hyatt@apple.com [Wed, 27 Oct 2010 19:48:16 +0000 (19:48 +0000)]
https://bugs.webkit.org/show_bug.cgi?id=48449

Reviewed by Beth Dakin.

Make border drawing work correctly with vertical inline flows.  The top and bottom edge have to be omitted
rather than the left and right edges.

Also discovered that border-radius is broken with the new path-based-drawing code for all inline flows.  This
is a regression from the old drawing code.  I patched the code to be correct rather than clipping out stuff
that it shouldn't and applying border-radii on lines that don't even have them.

Added fast/blockflow/border-vertical-lr.html and fast/borders/border-radius-inline-flow.html

WebCore:

* rendering/RenderBoxModelObject.cpp:
(WebCore::RenderBoxModelObject::paintBorder):
(WebCore::RenderBoxModelObject::clipBorderSidePolygon):
* rendering/RenderBoxModelObject.h:

LayoutTests:

* fast/blockflow/border-vertical-lr.html: Added.
* fast/borders/border-radius-inline-flow.html: Added.
* platform/mac/fast/blockflow/border-vertical-lr-expected.checksum: Added.
* platform/mac/fast/blockflow/border-vertical-lr-expected.png: Added.
* platform/mac/fast/blockflow/border-vertical-lr-expected.txt: Added.
* platform/mac/fast/borders/border-radius-inline-flow-expected.checksum: Added.
* platform/mac/fast/borders/border-radius-inline-flow-expected.png: Added.
* platform/mac/fast/borders/border-radius-inline-flow-expected.txt: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@70692 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoSkip some failing tests on Windows
aroben@apple.com [Wed, 27 Oct 2010 19:43:05 +0000 (19:43 +0000)]
Skip some failing tests on Windows

The tests don't fail in a reliable way, so skipping them is our best
option for now.

* platform/win/Skipped:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@70691 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2010-10-27 Jocelyn Turcotte <jocelyn.turcotte@nokia.com>
jocelyn.turcotte@nokia.com [Wed, 27 Oct 2010 19:39:42 +0000 (19:39 +0000)]
2010-10-27  Jocelyn Turcotte  <jocelyn.turcotte@nokia.com>

        Reviewed by Kenneth Rohde Christiansen.

        [Qt] Fix tst_QWebFrame::popupFocus().
        https://bugs.webkit.org/show_bug.cgi?id=48432

        - Call setFocus() before show() to work around a but in Qt.
        - Remove the check on combo at the end since hidePopup() leads
          indirectly to its destruction and segfault.

        * tests/qwebframe/tst_qwebframe.cpp:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@70690 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2010-10-27 Martin Robinson <mrobinson@igalia.com>
mrobinson@webkit.org [Wed, 27 Oct 2010 19:35:45 +0000 (19:35 +0000)]
2010-10-27  Martin Robinson  <mrobinson@igalia.com>

        Add GTK+-specific baselines for a couple tests.

        * platform/gtk/animations/missing-values-first-keyframe-expected.txt: Added.
        * platform/gtk/animations/missing-values-last-keyframe-expected.txt: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@70689 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2010-10-27 Martin Robinson <mrobinson@igalia.com>
mrobinson@webkit.org [Wed, 27 Oct 2010 19:31:47 +0000 (19:31 +0000)]
2010-10-27  Martin Robinson  <mrobinson@igalia.com>

        Reviewed by Gustavo Noronha Silva.

        [Cairo] Font fallback determination is very ineffecient
        https://bugs.webkit.org/show_bug.cgi?id=42052

        Added a test which verifies the correct result rendering with a custom
        font that is missing many glyphs.

        * platform/gtk/fonts/custom-font-missing-glyphs-expected.checksum: Added.
        * platform/gtk/fonts/custom-font-missing-glyphs-expected.png: Added.
        * platform/gtk/fonts/custom-font-missing-glyphs-expected.txt: Added.
        * platform/gtk/fonts/custom-font-missing-glyphs.html: Added.
        * platform/gtk/fonts/resources/mostly-empty-font.ttf: Added. A font that I
        created in fotnforge which is composed of empty glyphs, except for 'A' which
        has a lonely polyhedron for a glyph.
2010-10-27  Martin Robinson  <mrobinson@igalia.com>

        Reviewed by Gustavo Noronha Silva.

        [Cairo] Font fallback determination is very ineffecient
        https://bugs.webkit.org/show_bug.cgi?id=42052

        When choosing a fallback font in FontCache::getFontDataForCharacters, use FontConfig's
        mechanism for searching the fallback list directly, instead of scanning the list manually.
        If the font is a custom font or any other font with no FontConfig pattern associated with
        it, as FontConfig for a font that can supply the missing glyphs.

        Test: platform/gtk/fonts/custom-font-missing-glyphs.html

        * platform/graphics/cairo/FontCacheFreeType.cpp:
        (WebCore::createFontConfigPatternForCharacters): Added this helper which turns
        a collection of unicode characters into a FontConfig pattern.
        (WebCore::findBestFontGivenFallbacks): Using a FontConfig pattern created with th
        above helper and a FcFontSet of fallback fonts, this helper will find a font in
        the set which fulfills the requirments.
        (WebCore::FontCache::getFontDataForCharacters): Given an existing font with missing
        glyphs, either find an appropriate fallback font from the list of FontConfig fallbacks
        or ask FontConfig directly for an appropriate font.
        * platform/graphics/cairo/FontPlatformDataFreeType.h: Made a member mutable, as it is
        initialized lazily.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@70688 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoBegin stubbing out the Download class
andersca@apple.com [Wed, 27 Oct 2010 19:21:41 +0000 (19:21 +0000)]
Begin stubbing out the Download class
https://bugs.webkit.org/show_bug.cgi?id=48447

Reviewed by Adam Roben.

* WebKit2.pro:
* WebKit2.xcodeproj/project.pbxproj:
Add new files.

* WebKit2Prefix.h:
Include WebCore/EmptyProtocolDefinitions.h

* WebProcess/Downloads/Download.cpp: Added.
* WebProcess/Downloads/Download.h: Added.

* WebProcess/Downloads/DownloadManager.cpp:
(WebKit::DownloadManager::startDownload):
Create a Download object and start it.

* WebProcess/Downloads/mac/DownloadMac.mm: Added.
(WebKit::Download::start):
Create an NSURLDownload.

* WebProcess/WebPage/WebFrame.cpp:
(WebKit::WebFrame::startDownload):
Ask the download manager to start downloading.

* win/WebKit2.vcproj:
* win/WebKit2Common.vsprops:
Add new files.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@70687 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2010-10-27 Renata Hodovan <reni@inf.u-szeged.hu>
commit-queue@webkit.org [Wed, 27 Oct 2010 19:09:21 +0000 (19:09 +0000)]
2010-10-27  Renata Hodovan  <reni@inf.u-szeged.hu>

        Reviewed by Dirk Schulze.

        preserveAlpha attributes were misspeled in SVGFEConcolveMatrixElement tests.
        https://bugs.webkit.org/show_bug.cgi?id=48403

        Since expecteds and pixel tests didn't change after correcting this attribute has no effect so it is removed.
        For checking preserveAlpha attribute we have already a dom test which is not affected by this patch.

        * svg/dynamic-updates/script-tests/SVGFEConvolveMatrixElement-dom-bias-attr.js:
        * svg/dynamic-updates/script-tests/SVGFEConvolveMatrixElement-dom-divisor-attr.js:
        * svg/dynamic-updates/script-tests/SVGFEConvolveMatrixElement-dom-in-attr.js:
        * svg/dynamic-updates/script-tests/SVGFEConvolveMatrixElement-dom-kernelMatrix-attr.js:
        * svg/dynamic-updates/script-tests/SVGFEConvolveMatrixElement-dom-kernelUnitLength-attr.js:
        * svg/dynamic-updates/script-tests/SVGFEConvolveMatrixElement-dom-order-attr.js:
        * svg/dynamic-updates/script-tests/SVGFEConvolveMatrixElement-dom-targetX-attr.js:
        * svg/dynamic-updates/script-tests/SVGFEConvolveMatrixElement-dom-targetY-attr.js:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@70686 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoDon't allow setting NetscapePlugin::m_isWindowed after NPP_New has completed
aroben@apple.com [Wed, 27 Oct 2010 19:05:54 +0000 (19:05 +0000)]
Don't allow setting NetscapePlugin::m_isWindowed after NPP_New has completed

In Firefox, Chrome, and WebKit1 trying to set this value after NPP_New
does not actually affect whether the plugin is windowed.

Fixes <http://webkit.org/b/46673> <rdar://problem/8484211> Assertion
failure in NetscapePlugin::platformDestroy when running
plugins/mouse-events.html in WebKit2 on Windows

Reviewed by Anders Carlsson.

* WebProcess/Plugins/Netscape/NetscapePlugin.cpp:
(WebKit::NetscapePlugin::setIsWindowed): Moved here from the header.
Bail out if m_isStarted is true, indicating that NPP_New has already
finished.

* WebProcess/Plugins/Netscape/NetscapePlugin.h: Removed the
implementation of setIsWindowed.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@70685 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2010-10-27 Dimitri Glazkov <dglazkov@chromium.org>
dglazkov@chromium.org [Wed, 27 Oct 2010 19:02:17 +0000 (19:02 +0000)]
2010-10-27  Dimitri Glazkov  <dglazkov@chromium.org>

        [Chromium] Update expectations after r70664.

        * platform/chromium-mac/fast/blockflow/japanese-lr-text-expected.checksum: Added.
        * platform/chromium-mac/fast/blockflow/japanese-rl-text-expected.checksum: Added.
        * platform/chromium-win/fast/blockflow/japanese-lr-text-expected.checksum: Added.
        * platform/chromium-win/fast/blockflow/japanese-lr-text-expected.png: Added.
        * platform/chromium-win/fast/blockflow/japanese-lr-text-expected.txt: Added.
        * platform/chromium-win/fast/blockflow/japanese-rl-text-expected.checksum: Added.
        * platform/chromium-win/fast/blockflow/japanese-rl-text-expected.png: Added.
        * platform/chromium-win/fast/blockflow/japanese-rl-text-expected.txt: Added.
        * platform/chromium/test_expectations.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@70684 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2010-10-27 Dimitri Glazkov <dglazkov@chromium.org>
dglazkov@chromium.org [Wed, 27 Oct 2010 18:58:18 +0000 (18:58 +0000)]
2010-10-27  Dimitri Glazkov  <dglazkov@chromium.org>

        Unreviewed, rolling out r70674.
        http://trac.webkit.org/changeset/70674
        https://bugs.webkit.org/show_bug.cgi?id=48053

        Broke Chromium Windows build.

        * Scripts/webkitpy/layout_tests/layout_package/dump_render_tree_thread.py:
        * Scripts/webkitpy/layout_tests/run_webkit_tests.py:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@70683 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2010-10-27 Alexander Pavlov <apavlov@chromium.org>
apavlov@chromium.org [Wed, 27 Oct 2010 18:53:55 +0000 (18:53 +0000)]
2010-10-27  Alexander Pavlov  <apavlov@chromium.org>

        Reviewed by Pavel Feldman.

        Web Inspector: "Reload Page and Audit on Load" fails to run audits when resource tracking is disabled
        https://bugs.webkit.org/show_bug.cgi?id=48431

        * http/tests/inspector/inspector-test2.js:
        (initialize_InspectorTest.InspectorTest.enableResourceTracking):
        (initialize_InspectorTest.InspectorTest.disableResourceTracking):
        * http/tests/inspector/resource-parameters.html:
        * inspector/audits-panel-functional.html:
        * inspector/styles-source-offsets.html:
2010-10-27  Alexander Pavlov  <apavlov@chromium.org>

        Reviewed by Pavel Feldman.

        Web Inspector: "Reload Page and Audit on Load" fails to run audits when resource tracking is disabled
        https://bugs.webkit.org/show_bug.cgi?id=48431

        * inspector/front-end/AuditsPanel.js:
        (WebInspector.AuditsPanel.prototype._reloadResources):
        * inspector/front-end/ResourcesPanel.js:
        (WebInspector.ResourcesPanel.prototype._createPanelEnabler):
        (WebInspector.ResourcesPanel.prototype._enableResourceTracking):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@70682 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2010-10-27 Abhishek Arya <inferno@chromium.org>
inferno@chromium.org [Wed, 27 Oct 2010 18:46:12 +0000 (18:46 +0000)]
2010-10-27  Abhishek Arya  <inferno@chromium.org>

        Reviewed by Dave Hyatt.

        Add a function to make sure child is allowed before adding to a
        render view.
        https://bugs.webkit.org/show_bug.cgi?id=48328

        Test: fast/inline/inline-child-height-width-calc-crash.html

        * rendering/RenderView.cpp:
        (WebCore::RenderView::isChildAllowed):
        * rendering/RenderView.h:
2010-10-27  Abhishek Arya  <inferno@chromium.org>

        Reviewed by Dave Hyatt.

        Tests that adding a br element to a new document does not result in crash.
        https://bugs.webkit.org/show_bug.cgi?id=48328

        * fast/inline/inline-child-height-width-calc-crash-expected.txt: Added.
        * fast/inline/inline-child-height-width-calc-crash.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@70681 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoAdd API for enabling the developer extras (Web Inspector).
timothy@apple.com [Wed, 27 Oct 2010 18:45:15 +0000 (18:45 +0000)]
Add API for enabling the developer extras (Web Inspector).

http://webkit.org/b/48439

Reviewed by John Sullivan.

* Shared/WebPreferencesStore.cpp:
(WebKit::WebPreferencesStore::WebPreferencesStore):
(WebKit::WebPreferencesStore::encode):
(WebKit::WebPreferencesStore::decode):
* Shared/WebPreferencesStore.h:
* UIProcess/API/C/WKPreferences.cpp:
(WKPreferencesSetDeveloperExtrasEnabled):
(WKPreferencesGetDeveloperExtrasEnabled):
* UIProcess/API/C/WKPreferences.h:
* UIProcess/WebPreferences.cpp:
(WebKit::WebPreferences::setDeveloperExtrasEnabled):
(WebKit::WebPreferences::developerExtrasEnabled):
* UIProcess/WebPreferences.h:
* WebProcess/WebPage/WebPage.cpp:
(WebKit::WebPage::updatePreferences):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@70680 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2010-10-25 Tony Chang <tony@chromium.org>
tony@chromium.org [Wed, 27 Oct 2010 18:42:30 +0000 (18:42 +0000)]
2010-10-25  Tony Chang  <tony@chromium.org>

        Reviewed by Anders Carlsson.

        compile TestNetscapePlugIn on chromium linux
        https://bugs.webkit.org/show_bug.cgi?id=48274

        * WebKit.gyp: Enable compilation of TestNetscapePlugIn on Linux.
2010-10-25  Tony Chang  <tony@chromium.org>

        Reviewed by Anders Carlsson.

        compile TestNetscapePlugIn on chromium linux
        https://bugs.webkit.org/show_bug.cgi?id=48274

        * DumpRenderTree/TestNetscapePlugIn/main.cpp:
        (NP_Initialize): On Linux, plugin funcs are set in initialize.
        (NPP_New): Mark the plugin as windowless.
        (NPP_GetValue): Handle mime type values.
        (NP_GetMIMEDescription):
        (NP_GetValue):
        * DumpRenderTree/chromium/TestNetscapePlugIn/ForwardingHeaders/WebKit/npapi.h: define TRUE and FALSE, which are in webkit's npapi.h.
        * DumpRenderTree/chromium/TestNetscapePlugIn/ForwardingHeaders/WebKit/npfunctions.h: Pull in npapi.h to get TRUE/FALSE (matches webkit's npfunctions.h)

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@70679 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2010-10-27 Dimitri Glazkov <dglazkov@chromium.org>
dglazkov@chromium.org [Wed, 27 Oct 2010 18:36:41 +0000 (18:36 +0000)]
2010-10-27  Dimitri Glazkov  <dglazkov@chromium.org>

        [Chromium] Update expectations after r70664 and r70657.

        * platform/chromium-linux/fast/blockflow/japanese-lr-text-expected.checksum: Added.
        * platform/chromium-linux/fast/blockflow/japanese-lr-text-expected.png: Added.
        * platform/chromium-linux/fast/blockflow/japanese-lr-text-expected.txt: Added.
        * platform/chromium-linux/fast/blockflow/japanese-rl-text-expected.checksum: Added.
        * platform/chromium-linux/fast/blockflow/japanese-rl-text-expected.png: Added.
        * platform/chromium-linux/fast/blockflow/japanese-rl-text-expected.txt: Added.
        * platform/chromium-mac/animations/missing-values-first-keyframe-expected.checksum: Added.
        * platform/chromium-mac/animations/missing-values-first-keyframe-expected.png: Added.
        * platform/chromium-mac/animations/missing-values-last-keyframe-expected.checksum: Added.
        * platform/chromium-mac/animations/missing-values-last-keyframe-expected.png: Added.
        * platform/chromium-win/animations/missing-values-first-keyframe-expected.checksum: Added.
        * platform/chromium-win/animations/missing-values-first-keyframe-expected.png: Added.
        * platform/chromium-win/animations/missing-values-first-keyframe-expected.txt: Added.
        * platform/chromium-win/animations/missing-values-last-keyframe-expected.checksum: Added.
        * platform/chromium-win/animations/missing-values-last-keyframe-expected.png: Added.
        * platform/chromium-win/animations/missing-values-last-keyframe-expected.txt: Added.
        * platform/chromium/test_expectations.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@70678 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2010-10-27 Martin Robinson <mrobinson@igalia.com>
mrobinson@webkit.org [Wed, 27 Oct 2010 18:36:22 +0000 (18:36 +0000)]
2010-10-27  Martin Robinson  <mrobinson@igalia.com>

        Reviewed by Gustavo Noronha Silva.

        [GTK] r69296 includes a cast from GdkWindow* to GtkWindow*
        https://bugs.webkit.org/show_bug.cgi?id=48445

        Pass the pageClient GtkWidget* to gtk_xtbin_new instead of the
        GdkWindow*. The pageClient is used to get the proper visual for the
        plugin container widget.

        No new tests. This patch should prevent plugins/null-npp-getvalue-pointer.html
        from crashing.

        * plugins/gtk/PluginViewGtk.cpp:
        (WebCore::PluginView::platformStart): Pass the widget instead of the window.
        * plugins/gtk/gtk2xtbin.c: Accept the widget instead of the window.
        (gtk_xtbin_new):
        * plugins/gtk/gtk2xtbin.h: Update the declaration.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@70677 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2010-10-27 Kenneth Rohde Christiansen <kenneth@webkit.org>
kenneth@webkit.org [Wed, 27 Oct 2010 18:19:41 +0000 (18:19 +0000)]
2010-10-27  Kenneth Rohde Christiansen  <kenneth@webkit.org>

        Reviewed by Anders Carlsson.

        Move contentsSizeChange from the UIClient into the PageClient and
        and make it a private Qt-only API in the process.

        Make WKPageContentsSizeChangedCallback be a private API
        https://bugs.webkit.org/show_bug.cgi?id=48409

        * UIProcess/API/C/WKPage.h:
        * UIProcess/API/qt/qwkpage.cpp:
        (QWKPagePrivate::didChangeContentsSize):
        (QWKPage::QWKPage):
        * UIProcess/API/qt/qwkpage_p.h:
        * UIProcess/PageClient.h:
        * UIProcess/WebPageProxy.cpp:
        (WebKit::WebPageProxy::didChangeContentsSize):
        * UIProcess/WebPageProxy.h:
        * UIProcess/WebPageProxy.messages.in:
        * UIProcess/WebUIClient.cpp:
        * WebProcess/WebCoreSupport/WebChromeClient.cpp:
        (WebKit::WebChromeClient::contentsSizeChanged):
2010-10-27  Kenneth Rohde Christiansen  <kenneth@webkit.org>

        Reviewed by Anders Carlsson.

        Remove contentSizeChanged callbacks as it is no longer
        part of the public UIClient.

        Make WKPageContentsSizeChangedCallback be a private API
        https://bugs.webkit.org/show_bug.cgi?id=48409

        * MiniBrowser/mac/BrowserWindowController.m:
        (-[BrowserWindowController awakeFromNib]):
        * MiniBrowser/win/BrowserView.cpp:
        (BrowserView::create):
        * WebKitTestRunner/TestController.cpp:
        (WTR::createOtherPage):
        (WTR::TestController::initialize):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@70676 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2010-10-27 Kwang Yul Seo <skyul@company100.net>
commit-queue@webkit.org [Wed, 27 Oct 2010 18:05:20 +0000 (18:05 +0000)]
2010-10-27  Kwang Yul Seo  <skyul@company100.net>

        Reviewed by James Robinson.

        [BREWMP] Port graphics backend
        https://bugs.webkit.org/show_bug.cgi?id=40275

        Add PLATFORM(BREWMP) guard to skia's FontCustomPlatformData
        to share OS(LINUX) implementation.

        * loader/CachedFont.cpp:
        * platform/graphics/skia/FontCustomPlatformData.cpp:
        (WebCore::FontCustomPlatformData::~FontCustomPlatformData):
        (WebCore::FontCustomPlatformData::fontPlatformData):
        (WebCore::createFontCustomPlatformData):
        * platform/graphics/skia/FontCustomPlatformData.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@70675 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoMake http locking default in NRWT.
ossy@webkit.org [Wed, 27 Oct 2010 18:02:25 +0000 (18:02 +0000)]
Make http locking default in NRWT.
https://bugs.webkit.org/show_bug.cgi?id=48053

Patch by Gabor Rapcsanyi <rgabor@inf.u-szeged.hu> on 2010-10-27
Reviewed by Ojan Vafai.

* Scripts/webkitpy/layout_tests/layout_package/dump_render_tree_thread.py:
* Scripts/webkitpy/layout_tests/run_webkit_tests.py:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@70674 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoJavaScriptCore: https://bugs.webkit.org/show_bug.cgi?id=41948
ggaren@apple.com [Wed, 27 Oct 2010 17:59:57 +0000 (17:59 +0000)]
JavaScriptCore: https://bugs.webkit.org/show_bug.cgi?id=41948
REGRESSION(r60392): Registerfile can be unwound too far following an exception

Reviewed by Oliver Hunt.

SunSpider reports no change.

* interpreter/Interpreter.cpp:
(JSC::Interpreter::throwException): Walk the stack to calculate the high
water mark currently in use. It's not safe to assume that the current
CallFrame's high water mark is the highest high water mark because
calls do not always set up at the end of a CallFrame. A large caller
CallFrame can encompass a small callee CallFrame.

* jit/JITOpcodes.cpp:
(JSC::JIT::privateCompileCTINativeCall):
* jit/JITOpcodes32_64.cpp:
(JSC::JIT::privateCompileCTINativeCall): Make sure to set a 0 CodeBlock
in the CallFrame of a host call, like the Interpreter does, instead of
leaving the CodeBlock field uninitialized. The backtracing code requires
a valid CodeBlock field in each CallFrame.

LayoutTests: Added a test for:

Reviewed by Oliver Hunt.

https://bugs.webkit.org/show_bug.cgi?id=41948
REGRESSION(r60392): Registerfile can be unwound too far following an exception

* fast/js/exception-registerfile-shrink-expected.txt: Added.
* fast/js/exception-registerfile-shrink.html: Added.
* fast/js/script-tests/exception-registerfile-shrink.js: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@70673 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2010-10-27 Gabor Rapcsanyi <rgabor@inf.u-szeged.hu>
commit-queue@webkit.org [Wed, 27 Oct 2010 17:51:26 +0000 (17:51 +0000)]
2010-10-27  Gabor Rapcsanyi  <rgabor@inf.u-szeged.hu>

        Reviewed by Tony Chang.

        [NRWT] Fix http lock on Windows platform
        https://bugs.webkit.org/show_bug.cgi?id=48321

        * Scripts/webkitpy/layout_tests/port/http_lock.py:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@70672 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2010-10-27 Ryosuke Niwa <rniwa@webkit.org>
rniwa@webkit.org [Wed, 27 Oct 2010 17:50:45 +0000 (17:50 +0000)]
2010-10-27  Ryosuke Niwa  <rniwa@webkit.org>

        Unreviewed; grouped editing test failures in one place.

        * platform/chromium/test_expectations.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@70671 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoUnreviewed. Update Qt specific expected files after r70546.
ossy@webkit.org [Wed, 27 Oct 2010 17:48:29 +0000 (17:48 +0000)]
Unreviewed. Update Qt specific expected files after r70546.

* platform/qt/fast/blockflow/floats-in-block-layout-expected.txt:
* platform/qt/fast/blockflow/self-collapsing-block-expected.txt:
* platform/qt/fast/multicol/nested-columns-expected.checksum:
* platform/qt/fast/multicol/nested-columns-expected.png:
* platform/qt/fast/multicol/nested-columns-expected.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@70670 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2010-10-27 Satish Sampath <satish@chromium.org>
satish@chromium.org [Wed, 27 Oct 2010 17:46:18 +0000 (17:46 +0000)]
2010-10-27  Satish Sampath  <satish@chromium.org>

        Unreviewed, rolling out r70665.
        http://trac.webkit.org/changeset/70665
        https://bugs.webkit.org/show_bug.cgi?id=47089

        Need to address Alexey's review comments.

        * fast/speech/input-text-language-tag-expected.txt: Removed.
        * fast/speech/input-text-language-tag.html: Removed.
        * fast/speech/input-text-speechbutton.html:
        * fast/speech/speech-button-ignore-generated-events.html:
        * platform/chromium/test_expectations.txt:
2010-10-27  Satish Sampath  <satish@chromium.org>

        Unreviewed, rolling out r70665.
        http://trac.webkit.org/changeset/70665
        https://bugs.webkit.org/show_bug.cgi?id=47089

        Need to address Alexey's review comments.

        * page/SpeechInput.cpp:
        (WebCore::SpeechInput::startRecognition):
        * page/SpeechInput.h:
        * page/SpeechInputClient.h:
        * platform/mock/SpeechInputClientMock.cpp:
        (WebCore::SpeechInputClientMock::startRecognition):
        (WebCore::SpeechInputClientMock::setRecognitionResult):
        (WebCore::SpeechInputClientMock::timerFired):
        * platform/mock/SpeechInputClientMock.h:
        * rendering/TextControlInnerElements.cpp:
        (WebCore::TextControlInnerElement::attachInnerElement):
        (WebCore::SearchFieldCancelButtonElement::detach):
        (WebCore::SpinButtonElement::defaultEventHandler):
        (WebCore::InputFieldSpeechButtonElement::defaultEventHandler):
        (WebCore::InputFieldSpeechButtonElement::detach):
2010-10-27  Satish Sampath  <satish@chromium.org>

        Unreviewed, rolling out r70665.
        http://trac.webkit.org/changeset/70665
        https://bugs.webkit.org/show_bug.cgi?id=47089

        Need to address Alexey's review comments.

        * public/WebSpeechInputController.h:
        (WebKit::WebSpeechInputController::startRecognition):
        * public/WebSpeechInputControllerMock.h:
        * src/SpeechInputClientImpl.cpp:
        (WebKit::SpeechInputClientImpl::startRecognition):
        * src/SpeechInputClientImpl.h:
        * src/WebSpeechInputControllerMockImpl.cpp:
        (WebKit::WebSpeechInputControllerMockImpl::setMockRecognitionResult):
        (WebKit::WebSpeechInputControllerMockImpl::startRecognition):
        * src/WebSpeechInputControllerMockImpl.h:
        * src/WebViewImpl.cpp:
        (WebKit::WebViewImpl::WebViewImpl):
        * src/WebViewImpl.h:
2010-10-27  Satish Sampath  <satish@chromium.org>

        Unreviewed, rolling out r70665.
        http://trac.webkit.org/changeset/70665
        https://bugs.webkit.org/show_bug.cgi?id=47089

        Need to address Alexey's review comments.

        * DumpRenderTree/LayoutTestController.cpp:
        (setMockSpeechInputResultCallback):
        * DumpRenderTree/LayoutTestController.h:
        * DumpRenderTree/chromium/LayoutTestController.cpp:
        (LayoutTestController::setMockSpeechInputResult):
        * DumpRenderTree/gtk/LayoutTestControllerGtk.cpp:
        (LayoutTestController::setMockSpeechInputResult):
        * DumpRenderTree/mac/LayoutTestControllerMac.mm:
        (LayoutTestController::setMockSpeechInputResult):
        * DumpRenderTree/qt/LayoutTestControllerQt.cpp:
        (LayoutTestController::setMockSpeechInputResult):
        * DumpRenderTree/qt/LayoutTestControllerQt.h:
        * DumpRenderTree/win/LayoutTestControllerWin.cpp:
        (LayoutTestController::setMockSpeechInputResult):
        * DumpRenderTree/wx/LayoutTestControllerWx.cpp:
        (LayoutTestController::setMockSpeechInputResult):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@70669 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoChange the results to account for a typo in the Japanese text string.
hyatt@apple.com [Wed, 27 Oct 2010 17:26:34 +0000 (17:26 +0000)]
Change the results to account for a typo in the Japanese text string.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@70668 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2010-10-27 Dimitri Glazkov <dglazkov@chromium.org>
dglazkov@chromium.org [Wed, 27 Oct 2010 17:25:33 +0000 (17:25 +0000)]
2010-10-27  Dimitri Glazkov  <dglazkov@chromium.org>

        [Chromium] Update expectations after r70654.
        https://bugs.webkit.org/show_bug.cgi?id=48347

        * platform/chromium-*/svg/dynamic-updates/*: Updated.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@70667 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2010-10-27 ZHenyao Mo <zmo@google.com>
zmo@google.com [Wed, 27 Oct 2010 17:24:45 +0000 (17:24 +0000)]
2010-10-27  ZHenyao Mo  <zmo@google.com>

        Reviewed by Kenneth Russell.

        Must enable GL_POINT_SPRITE in GraphicsContext3D implementations
        https://bugs.webkit.org/show_bug.cgi?id=45908

        * src/WebGraphicsContext3DDefaultImpl.cpp: Enable for chromium --in-process-webgl port.
        (WebKit::WebGraphicsContext3DDefaultImpl::initialize):
2010-10-27  ZHenyao Mo  <zmo@google.com>

        Reviewed by Kenneth Russell.

        Must enable GL_POINT_SPRITE in GraphicsContext3D implementations
        https://bugs.webkit.org/show_bug.cgi?id=45908

        * platform/graphics/mac/GraphicsContext3DMac.mm: Enable for Mac port.
        (WebCore::GraphicsContext3D::GraphicsContext3D):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@70666 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2010-10-27 Leandro Gracia Gil <leandrogracia@google.com>
satish@chromium.org [Wed, 27 Oct 2010 17:19:42 +0000 (17:19 +0000)]
2010-10-27  Leandro Gracia Gil  <leandrogracia@google.com>

        Reviewed by Jeremy Orlow.

        Added a layout test for the language tag in speech input. Updated
        the speech button tests to include the new language parameter in the
        mock result.
        https://bugs.webkit.org/show_bug.cgi?id=47089

        * fast/speech/input-text-language-tag-expected.txt: Added.
        * fast/speech/input-text-language-tag.html: Added.
        * fast/speech/input-text-speechbutton.html:
        * fast/speech/speech-button-ignore-generated-events.html:
        * platform/chromium/test_expectations.txt:
2010-10-27  Leandro Gracia Gil  <leandrogracia@google.com>

        Reviewed by Jeremy Orlow.

        Patch the current speech input implementation to use the nearest
        language tag. The language is now passed to the startRecognition
        methods so that language-specific recognition could be used. Also added
        a second parameter to setMockSpeechInputResult for the language used in
        speech recognition.
        https://bugs.webkit.org/show_bug.cgi?id=47089

        This is the 2nd of a 4-sided patch in Chromium and WebKit. For more
        details see http://codereview.chromium.org/3615005/show and
        http://codereview.chromium.org/3595018/show. The last of the 4 patches
        depends also on the language tag validation provided by this patch:
        https://bugs.webkit.org/show_bug.cgi?id=48225.

        Test: fast/speech/input-text-language-tag.html

        * page/SpeechInput.cpp:
        (WebCore::SpeechInput::startRecognition):
        * page/SpeechInput.h:
        * page/SpeechInputClient.h:
        * platform/mock/SpeechInputClientMock.cpp:
        (WebCore::SpeechInputClientMock::startRecognition):
        (WebCore::SpeechInputClientMock::setRecognitionResult):
        (WebCore::SpeechInputClientMock::timerFired):
        * platform/mock/SpeechInputClientMock.h:
        * rendering/TextControlInnerElements.cpp:
        (WebCore::TextControlInnerElement::attachInnerElement):
        (WebCore::SearchFieldCancelButtonElement::detach):
        (WebCore::SpinButtonElement::defaultEventHandler):
        (WebCore::InputFieldSpeechButtonElement::defaultEventHandler):
        (WebCore::InputFieldSpeechButtonElement::detach):
2010-10-27  Leandro Gracia Gil  <leandrogracia@google.com>

        Reviewed by Jeremy Orlow.

        Patch the current speech input implementation to use the nearest
        language tag. The language is now passed to the startRecognition
        methods so that language-specific recognition could be used. Also added
        a second parameter to setMockSpeechInputResult for the language used in
        speech recognition.
        https://bugs.webkit.org/show_bug.cgi?id=47089

        This is the 2nd of a 4-sided patch in Chromium and WebKit. For more
        details see http://codereview.chromium.org/3615005/show and
        http://codereview.chromium.org/3595018/show. The last of the 4 patches
        depends also on the language tag validation provided by this patch:
        https://bugs.webkit.org/show_bug.cgi?id=48225.

        * public/WebSpeechInputController.h:
        (WebKit::WebSpeechInputController::startRecognition):
        * public/WebSpeechInputControllerMock.h:
        * src/SpeechInputClientImpl.cpp:
        (WebKit::SpeechInputClientImpl::create):
        (WebKit::SpeechInputClientImpl::startRecognition):
        * src/SpeechInputClientImpl.h:
        * src/WebSpeechInputControllerMockImpl.cpp:
        (WebKit::WebSpeechInputControllerMockImpl::setMockRecognitionResult):
        (WebKit::WebSpeechInputControllerMockImpl::startRecognition):
        * src/WebSpeechInputControllerMockImpl.h:
        * src/WebViewImpl.cpp:
        (WebKit::WebViewImpl::WebViewImpl):
        * src/WebViewImpl.h:
2010-10-27  Leandro Gracia Gil  <leandrogracia@google.com>

        Reviewed by Jeremy Orlow.

        Added a second parameter to setMockSpeechInputResult for
        the language used in speech input.
        https://bugs.webkit.org/show_bug.cgi?id=47089

        * DumpRenderTree/LayoutTestController.cpp:
        (setMockSpeechInputResultCallback):
        * DumpRenderTree/LayoutTestController.h:
        * DumpRenderTree/chromium/LayoutTestController.cpp:
        (LayoutTestController::setMockSpeechInputResult):
        * DumpRenderTree/gtk/LayoutTestControllerGtk.cpp:
        (LayoutTestController::setMockSpeechInputResult):
        * DumpRenderTree/mac/LayoutTestControllerMac.mm:
        (LayoutTestController::setMockSpeechInputResult):
        * DumpRenderTree/qt/LayoutTestControllerQt.cpp:
        (LayoutTestController::setMockSpeechInputResult):
        * DumpRenderTree/qt/LayoutTestControllerQt.h:
        * DumpRenderTree/win/LayoutTestControllerWin.cpp:
        (LayoutTestController::setMockSpeechInputResult):
        * DumpRenderTree/wx/LayoutTestControllerWx.cpp:
        (LayoutTestController::setMockSpeechInputResult):
        (LayoutTestController::pageSizeAndMarginsInPixels):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@70665 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agohttps://bugs.webkit.org/show_bug.cgi?id=48381
hyatt@apple.com [Wed, 27 Oct 2010 17:16:07 +0000 (17:16 +0000)]
https://bugs.webkit.org/show_bug.cgi?id=48381

Reviewed by Dan Bernstein.

Make basic layer positioning work with vertical-lr writing-mode.

WebCore:

* rendering/RenderBlock.cpp:
(WebCore::RenderBlock::paintChildren):
(WebCore::RenderBlock::paintFloats):
* rendering/RenderBox.cpp:
(WebCore::RenderBox::adjustForFlippedBlocksWritingMode):
(WebCore::RenderBox::locationOffsetIncludingFlipping):
* rendering/RenderBox.h:
* rendering/RenderLayer.cpp:
(WebCore::RenderLayer::updateLayerPosition):

LayoutTests:

* fast/blockflow/japanese-lr-text.html: Added.
* fast/blockflow/japanese-rl-text.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@70664 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2010-10-27 Zhenyao Mo <zmo@google.com>
zmo@google.com [Wed, 27 Oct 2010 17:14:29 +0000 (17:14 +0000)]
2010-10-27  Zhenyao Mo  <zmo@google.com>

        Reviewed by Kenneth Russell.

        object-deletion-behaviour.html fails with mesa backend
        https://bugs.webkit.org/show_bug.cgi?id=46667

        * html/canvas/WebGLRenderingContext.cpp:
        (WebCore::WebGLRenderingContext::getFramebufferAttachmentParameter): If object_type is 0, we return early because object_name should be always be 0.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@70663 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoReset TestNetscapePlugIn's NPP_GetValue pointer when NullNPPGetValuePointer finishes...
aroben@apple.com [Wed, 27 Oct 2010 17:13:35 +0000 (17:13 +0000)]
Reset TestNetscapePlugIn's NPP_GetValue pointer when NullNPPGetValuePointer finishes running

Fixes <http://webkit.org/b/48435> REGRESSION (r70655): Many plugins
tests are failing on Qt

Reviewed by Anders Carlsson.

* DumpRenderTree/TestNetscapePlugIn/PluginTest.cpp:
(PluginTest::NPP_Destroy):
* DumpRenderTree/TestNetscapePlugIn/PluginTest.h:
Added a do-nothing NPP_Destroy implementation.

* DumpRenderTree/TestNetscapePlugIn/Tests/NullNPPGetValuePointer.cpp:
(NullNPPGetValuePointer::NullNPPGetValuePointer): Save the original
NPP_GetValue pointer in m_originalNPPGetValuePointer so we can restore
it later.
(NullNPPGetValuePointer::NPP_Destroy): Added. Restores the original
NPP_GetValue pointer so it can be used in other tests.

* DumpRenderTree/TestNetscapePlugIn/main.cpp:
(NPP_Destroy):
* DumpRenderTree/unix/TestNetscapePlugin/TestNetscapePlugin.cpp:
(webkit_test_plugin_destroy_instance):
Call through to the PluginTest when NPP_Destroy is called.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@70662 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2010-10-27 Zhenyao Mo <zmo@google.com>
zmo@google.com [Wed, 27 Oct 2010 17:10:14 +0000 (17:10 +0000)]
2010-10-27  Zhenyao Mo  <zmo@google.com>

        Reviewed by Kenneth Russell.

        If WebGL is running on top of a strict version of OpenGL ES it should make sure attribs have buffers assigned at all times
        https://bugs.webkit.org/show_bug.cgi?id=47964

        * html/canvas/WebGLRenderingContext.cpp:
        (WebCore::WebGLRenderingContext::validateRenderingState): Minimum checking: if input <= 0, only check if each enabled vertex attribute is bound to a buffer.
        (WebCore::WebGLRenderingContext::drawArrays): If underlying GL performs bounds checking, we still need to do the minimum checking using validateRenderingState.
        (WebCore::WebGLRenderingContext::drawElements): Ditto.
        * html/canvas/WebGLRenderingContext.h: Add a comment for input <= 0 in validateRenderingState.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@70661 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2010-10-27 Dimitri Glazkov <dglazkov@chromium.org>
dglazkov@chromium.org [Wed, 27 Oct 2010 17:04:14 +0000 (17:04 +0000)]
2010-10-27  Dimitri Glazkov  <dglazkov@chromium.org>

        [Chromium] Update results after r70569.

        * platform/chromium-linux/fast/blockflow/english-lr-text-expected.checksum: Added.
        * platform/chromium-mac/fast/blockflow/english-lr-text-expected.checksum: Added.
        * platform/chromium-win/fast/blockflow/english-lr-text-expected.checksum: Added.
        * platform/chromium-win/fast/blockflow/english-lr-text-expected.png: Added.
        * platform/chromium-win/fast/blockflow/english-lr-text-expected.txt: Added.
        * platform/chromium/test_expectations.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@70660 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoAdd DownloadManager class
andersca@apple.com [Wed, 27 Oct 2010 17:00:52 +0000 (17:00 +0000)]
Add DownloadManager class
https://bugs.webkit.org/show_bug.cgi?id=48388

Reviewed by Sam Weinig.

* WebKit2.pro:
* WebKit2.xcodeproj/project.pbxproj:
Add new files.

* WebProcess/Downloads/DownloadManager.cpp: Added.
(WebKit::DownloadManager::shared):
(WebKit::DownloadManager::DownloadManager):
* WebProcess/Downloads/DownloadManager.h: Added.

* win/WebKit2.vcproj:
* win/WebKit2Common.vsprops:
Add new files.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@70659 268f45cc-cd09-0410-ab3c-d52691b4dbfc