Always set the current NSGraphicsContext before calling drawWithFrame
https://bugs.webkit.org/show_bug.cgi?id=42542

Reviewed by Dan Bernstein.

* rendering/RenderThemeMac.mm:
(WebCore::RenderThemeMac::paintMeter):
(WebCore::RenderThemeMac::paintSearchField):
(WebCore::RenderThemeMac::paintSearchFieldCancelButton):
(WebCore::RenderThemeMac::paintSearchFieldResultsDecoration):
(WebCore::RenderThemeMac::paintSearchFieldResultsButton):

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@63642 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2010-07-18  Andreas Kling  <andreas.kling@nokia.com>

        Reviewed by Antonio Gomes.

        [Qt] tst_QWebPage fails cursorMovements and textSelection
        https://bugs.webkit.org/show_bug.cgi?id=42531

        Correct the HTML used by these tests.

        * tests/qwebpage/tst_qwebpage.cpp:
        (tst_QWebPage::cursorMovements):
        (tst_QWebPage::textSelection):
        (tst_QWebPage::textEditing):

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@63641 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Another attempt at fixing the Windows build.

* WebProcess/Plugins/NPJSObjectMap.h:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@63640 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Try to fix Windows build.

* win/WebKit2.vcproj:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@63639 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Move PluginWidget to WebKit
https://bugs.webkit.org/show_bug.cgi?id=42530

Reviewed by Sam Weinig.

WebCore:

Rename the PluginWidget class to PluginViewBase and make it an abstract base class.

This is a stopgap measure until we have a single PluginView class that we can use everywhere.

* WebCore.exp.in:
Remove PluginWidget symbols.

* WebCore.xcodeproj/project.pbxproj:
Update.

* platform/Widget.h:
(WebCore::Widget::isPluginViewBase):
Return false.

* plugins/PluginViewBase.h: Added.
(WebCore::PluginViewBase::platformLayer):
Always return 0 here now.

(WebCore::PluginViewBase::isPluginViewBase):
Return true.

* plugins/PluginWidget.h: Removed.
* plugins/mac/PluginWidgetMac.mm: Removed.
* rendering/RenderEmbeddedObject.cpp:
(WebCore::RenderEmbeddedObject::allowsAcceleratedCompositing):
Cast to PluginWidgetBase instead.

* rendering/RenderLayerBacking.cpp:
(WebCore::RenderLayerBacking::updateGraphicsLayerConfiguration):
Cast to PluginWidgetBase instead.

WebKit/mac:

* Plugins/WebBaseNetscapePluginView.h:
* Plugins/WebBaseNetscapePluginView.mm:
(-[WebBaseNetscapePluginView pluginLayer]):
Add default pluginLayer implementation.

* WebCoreSupport/WebFrameLoaderClient.mm:
(PluginWidget::PluginWidget):
(PluginWidget::invalidateRect):
Move the guts of PluginWidget here from WebCore.

(NetscapePluginWidget::platformLayer):
Get the layer from the plug-in view.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@63638 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2010-07-18  Andreas Kling  <andreas.kling@nokia.com>

        Reviewed by Antonio Gomes.

        [Qt] tst_QWebElement fails firstChildNextSibling and lastChildPreviousSibling
        https://bugs.webkit.org/show_bug.cgi?id=42527

        Correct the HTML used by these tests.

        * tests/qwebelement/tst_qwebelement.cpp:
        (tst_QWebElement::firstChildNextSibling):
        (tst_QWebElement::lastChildPreviousSibling):

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@63637 268f45cc-cd09-0410-ab3c-d52691b4dbfc

More NPRuntime work
https://bugs.webkit.org/show_bug.cgi?id=42526

Reviewed by Sam Weinig.

* WebProcess/Plugins/NPJSObjectMap.cpp:
(WebKit::identifierFromIdentifierRep):
(WebKit::NPJSObject::hasProperty):
Check if the JSObject has the given property.

(WebKit::NPJSObject::getProperty):
Add stubbed out function.

(WebKit::NPJSObject::npClass):
Add NP_HasProperty and NP_GetProperty.

(WebKit::NPJSObject::NP_HasProperty):
Call NPJSObject::hasProperty.

(WebKit::NPJSObject::NP_GetProperty):
Call NPJSObject::getProperty.

* WebProcess/Plugins/NPRuntimeUtilities.cpp:
(WebKit::releaseNPVariantValue):
Release the given NPVariant.

* WebProcess/Plugins/NPRuntimeUtilities.h:
* WebProcess/Plugins/Netscape/NetscapeBrowserFuncs.cpp:
(WebKit::NPN_GetProperty):
Call the NPClass GetProperty function.

(WebKit::NPN_HasProperty):
Call the NPClass HasProperty function.

(WebKit::NPN_ReleaseVariantValue):
Call releaseNPVariantValue.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@63636 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2010-07-18  Dean Jackson  <dino@apple.com>

        Unreviewed (although this was discussed with SimonF in
        original review).

        https://bugs.webkit.org/show_bug.cgi?id=41259

        Unfortunately the test in the previous commit is platform
        specific (I was hoping it would not be) and is failing
        the bots. This update exercises the same thing, but doesn't
        check for exact values.

        * fast/transforms/scrollIntoView-transformed-expected.txt:
        * fast/transforms/scrollIntoView-transformed.html:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@63635 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2010-07-18  Dean Jackson  <dino@apple.com>

        Unreviewed.

        Remove the unwanted extra line that Xcode
        added in my last commit.

        * WebCore.xcodeproj/project.pbxproj:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@63634 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2010-07-18  Dean Jackson  <dino@apple.com>

        Reviewed by Simon Fraser.

        https://bugs.webkit.org/show_bug.cgi?id=41259
        Interacting with a <select> element within a transformed and clipped
        container scrolls the container

        The Node::getRect and ContainerNode::getRect functions were not
        transform-aware. This fixes both, and has a test to make sure
        we're not breaking any existing scrollToView code. This means
        that a <select> popup will appear in the correct place if it
        is within a transformed and scrolled container.

        The test makes sure that existing scrollToView
        code doesn't break, but also checks that a <select> popup will appear
        in the correct place.

        Test: fast/transforms/scrollIntoView-transformed.html

        * dom/ContainerNode.cpp:
        (WebCore::ContainerNode::getUpperLeftCorner):
        (WebCore::ContainerNode::getLowerRightCorner):
        - make sure we call localToAbsolute in the right order
        (after we've done a local move) and pass in the flags to
        indicate it should look for transforms.
        * dom/Node.cpp:
        (WebCore::Node::getRect):
        - make sure localToAbsolute gets told to look for transforms.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@63633 268f45cc-cd09-0410-ab3c-d52691b4dbfc

<webkit.org/b/42522> Add missing (id) return type to Obj-C methods

Reviewed by Dan Bernstein.

* Carbon/CarbonWindowAdapter.mm:
(-[NSWindow(HIWebFrameView) _initContent:styleMask:backing:defer:contentView:]):
(-[CarbonWindowAdapter _destroyRealWindow:]):
(-[CarbonWindowAdapter _clearModalWindowLevel]):
* Misc/WebIconDatabase.mm:
(-[WebIconDatabase init]):
* Misc/WebNSPasteboardExtras.mm:
(-[NSFilePromiseDragSource initWithSource:]):
* WebView/WebHTMLRepresentation.mm:
(-[WebHTMLRepresentation init]):
* WebView/WebPreferences.mm:
(-[WebPreferences init]):
* WebView/WebView.mm:
(-[WebView UIDelegate]):
(-[WebView resourceLoadDelegate]):
(-[WebView downloadDelegate]):
(-[WebView policyDelegate]):
(-[WebView frameLoadDelegate]):

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@63632 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Add NPJSObjectMap class
https://bugs.webkit.org/show_bug.cgi?id=42524

Reviewed by Sam Weinig.

WebCore:

Export ScriptController functions.

* WebCore.exp.in:

WebKit2:

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

* WebProcess/Plugins/NPJSObjectMap.cpp: Added.
* WebProcess/Plugins/NPJSObjectMap.h: Added.
Add NPJSObjectMap, a map which contains NPObjects that wrap JavaScript objects.

* WebProcess/Plugins/PluginView.cpp:
(WebKit::PluginView::PluginView):
Initialize the map.

(WebKit::PluginView::~PluginView):
Invalidate the map.

(WebKit::PluginView::frame):
Add frame getter.

(WebKit::PluginView::windowScriptNPObject):
Wrap the window object.

(WebKit::PluginView::pluginElementNPObject):
Wrap the plug-in element object.

* WebProcess/Plugins/PluginView.h:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@63631 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Implement more NPRuntime related NPN_ functions
https://bugs.webkit.org/show_bug.cgi?id=42520

Reviewed by Sam Weinig.

* WebKit2.xcodeproj/project.pbxproj:
* WebProcess/Plugins/NPRuntimeUtilities.cpp: Added.
* WebProcess/Plugins/NPRuntimeUtilities.h: Added.
Add new file with NPRuntime related utility functions.

* WebProcess/Plugins/Netscape/NetscapeBrowserFuncs.cpp:
(WebKit::NPN_CreateObject):
(WebKit::NPN_RetainObject):
(WebKit::NPN_ReleaseObject):
Call the corresponding NPRuntimeUtilities functions.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@63630 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Begin work on NPRuntime support
https://bugs.webkit.org/show_bug.cgi?id=42519

Reviewed by Dan Bernstein.

* WebProcess/Plugins/Netscape/NetscapeBrowserFuncs.cpp:
(WebKit::NPN_GetValue):
Handle NPNVWindowNPObject and NPNVPluginElementNPObject.

* WebProcess/Plugins/Netscape/NetscapePlugin.cpp:
(WebKit::NetscapePlugin::windowScriptNPObject):
(WebKit::NetscapePlugin::pluginElementNPObject):
Call the plug-in controller.

* WebProcess/Plugins/PluginController.h:
Add new windowScriptNPObject and pluginElementNPObject functions.

* WebProcess/Plugins/PluginView.cpp:
(WebKit::PluginView::windowScriptNPObject):
(WebKit::PluginView::pluginElementNPObject):
Add stubbed out functions.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@63629 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Implement some NPRuntime related NPN_ functions
https://bugs.webkit.org/show_bug.cgi?id=42518

Reviewed by Dan Bernstein.

WebCore:

* WebCore.exp.in:
Export IdentifierRep functions.

WebKit2:

* WebProcess/Plugins/Netscape/NetscapeBrowserFuncs.cpp:
(WebKit::NPN_GetStringIdentifier):
(WebKit::NPN_GetStringIdentifiers):
(WebKit::NPN_GetIntIdentifier):
(WebKit::NPN_IdentifierIsString):
(WebKit::NPN_UTF8FromIdentifier):
(WebKit::NPN_IntFromIdentifier):
(WebKit::NPN_CreateObject):

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@63628 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Add dumping of statusbar text to WebKitTestRunner
https://bugs.webkit.org/show_bug.cgi?id=42516

Reviewed by Dan Bernstein.

WebKit2:

* WebProcess/InjectedBundle/API/c/WKBundlePage.h:
Add setStatusbarText callback to WKBundlePageUIClient.

* WebProcess/InjectedBundle/InjectedBundlePageUIClient.cpp:
(WebKit::InjectedBundlePageUIClient::setStatusbarText):
Call setStatusbarText.

* WebProcess/InjectedBundle/InjectedBundlePageUIClient.h:
* WebProcess/WebCoreSupport/WebChromeClient.cpp:
(WebKit::WebChromeClient::setStatusbarText):
Call the bundle page UI client.

WebKitTools:

* WebKitTestRunner/InjectedBundle/InjectedBundlePage.cpp:
(WTR::InjectedBundlePage::InjectedBundlePage):
(WTR::InjectedBundlePage::_setStatusbarText):
(WTR::InjectedBundlePage::setStatusbarText):
Dump the statusbar text.

* WebKitTestRunner/InjectedBundle/InjectedBundlePage.h:
* WebKitTestRunner/InjectedBundle/LayoutTestController.cpp:
(WTR::LayoutTestController::LayoutTestController):
Initialize m_dumpStatusCallbacks to false.

(WTR::dumpStatusCallbacksCallback):
Implement JSC callback.

(WTR::LayoutTestController::staticFunctions):
Add dumpStatusCallbacks.

* WebKitTestRunner/InjectedBundle/LayoutTestController.h:
(WTR::LayoutTestController::dumpStatusCallbacks):
(WTR::LayoutTestController::setDumpStatusCallbacks):
Add setter and getter for m_dumpStatusCallbacks.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@63627 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2010-07-16  Pavel Podivilov  <podivilov@chromium.org>

        Reviewed by Pavel Feldman.

        [V8] Web Inspector: get actual breakpoint location from v8 and pass it
        to ScripDebugServer.
        https://bugs.webkit.org/show_bug.cgi?id=42452

        * src/js/DebuggerScript.js:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@63626 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2010-07-18  Pavel Feldman  <pfeldman@chromium.org>

        Not reviewed. Chromium: fixed one more expectation to
        reflect changes in r63616.

        * platform/chromium/test_expectations.txt:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@63625 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2010-07-18  Pavel Feldman  <pfeldman@chromium.org>

        Not reviewed. Chromium: fixed test expectations (reflected changes
        in r63616).

        * platform/chromium/test_expectations.txt:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@63624 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2010-07-17  Andreas Kling  <andreas.kling@nokia.com>

        Rubber-stamped by Simon Hausmann.

        Import new version of 2d.pattern.image.string.html from Philip Taylor.
        The original test was buggy and we now pass it.

        * canvas/philip/tests/2d.pattern.image.string-expected.txt:
        * canvas/philip/tests/2d.pattern.image.string.html:
        * platform/mac/Skipped:
        * platform/qt/Skipped:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@63623 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2010-07-17  TJ Lee  <tjlee0909@gmail.com>

        Reviewed by Timothy Hatcher.

        HTMLLinkElement ignores dnsPrefetchingEnabled setting
        https://bugs.webkit.org/show_bug.cgi?id=42500

        Changed the HTML Link tag to check that the browser
        has DNS-prefetching enabled before calling ResourceHandle::prepareForURL.

        There are no test cases for this patch because it was unclear how to test
        this using a layout test. A possible test case would be to
        clear the DNS cache on the client's machine before loading a page with
        <link rel="dns-prefetch" href="SomeSiteThatsNotTheCurrentOne.com"> and
        then check the number of DNS cache entries.

        * html/HTMLLinkElement.cpp:
        (WebCore::HTMLLinkElement::process):

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@63622 268f45cc-cd09-0410-ab3c-d52691b4dbfc

WebKit2: WebKitTestRunner should load the test plug-in
https://bugs.webkit.org/show_bug.cgi?id=42509

Reviewed by Maciej Stachowiak.

* UIProcess/API/C/WKContext.cpp:
(_WKContextSetAdditionalPluginPath):
Add a private function for setting a single additional plug-in path. the WebKit1 SPI that does the
same thing takes an array of paths, but this is good enough for now.

* UIProcess/Plugins/PluginInfoStore.cpp:
(WebKit::PluginInfoStore::setAdditionalPluginPaths):
Set the additional plug-in paths vector and refresh the database.

(WebKit::PluginInfoStore::loadPluginsIfNecessary):
First try to load plug-ins in the additional plug-in paths.

* UIProcess/WebContext.cpp:
(WebKit::WebContext::setAdditionalPluginPath):
Call PluginInfoStore::setAdditionalPluginPaths.

* UIProcess/WebContext.h:
(WebKit::WebContext::pluginInfoStore):
Make the plug-in info store per context instead of having a single shared info store.

* UIProcess/WebProcessProxy.cpp:
(WebKit::WebProcessProxy::getPlugins):
(WebKit::WebProcessProxy::getPluginHostConnection):
* UIProcess/WebProcessProxy.h:
Get the plug-in info store from the context.

WebKitTools: * WebKitTestRunner/TestController.cpp:
(WTR::TestController::initialize):
Get the test plug-in path and set it on the context.

Reviewed by Maciej Stachowiak.

* WebKitTestRunner/TestController.h:
* WebKitTestRunner/mac/TestControllerMac.mm:
(WTR::TestController::initializeInjectedBundlePath):
Simplify this code by using NSBundle.

(WTR::TestController::testPluginPath):
Return the test plug-in path.

* WebKitTestRunner/win/TestControllerWin.cpp:
(WTR::TestController::testPluginPath):
Ditto.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@63621 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2010-07-16  Maciej Stachowiak  <mjs@apple.com>

        Reviewed by Sam Weinig.

        Asynchronous policy checks make FrameLoader think it is done loading prematurely
        https://bugs.webkit.org/show_bug.cgi?id=42489

        This caused many (~100) layout tsts to fail under WebKit2.

        * loader/SubframeLoader.cpp:
        (WebCore::SubframeLoader::loadSubframe): Right after loading a new subframe,
        if m_complete is true, do not consider it done if it has a provisional loader.
        This will happen in the case where the policy check is asynchronous.
        * loader/FrameLoader.cpp:
        (WebCore::FrameLoader::subframeIsLoading): For similar reasons,
        consider a subframe to be loading if it has a policy decision pending.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@63620 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2010-07-17  Andreas Kling  <andreas.kling@nokia.com>

        [Qt] Reskip 2d.path.quadraticCurveTo.scaled.html

        This test fails with the Qt version running on the bot.

        * platform/qt/Skipped:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@63619 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Another attempt at fixing this test.

* plugins/npruntime/browser-object-identity.html:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@63618 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2010-07-17  Andreas Kling  <andreas.kling@nokia.com>

        Rubber-stamped by Simon Hausmann.

        [Qt] Unskipped two canvas tests that are passing

        - canvas/philip/tests/2d.path.arc.twopie.3.html
        - canvas/philip/tests/2d.path.quadraticCurveTo.scaled.html

        * platform/qt/Skipped:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@63617 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Move platform specific plug-ins to platform/
https://bugs.webkit.org/show_bug.cgi?id=42506

Reviewed by Sam Weinig.

* platform/gtk/Skipped:
* platform/mac-leopard/plugins/qt-qwidget-plugin-expected.checksum: Removed.
* platform/mac-leopard/plugins/qt-qwidget-plugin-expected.png: Removed.
* platform/mac/plugins/call-as-function-test-expected.txt: Renamed from LayoutTests/plugins/call-as-function-test-expected.txt.
* platform/mac/plugins/call-as-function-test.html: Renamed from LayoutTests/plugins/call-as-function-test.html.
* platform/mac/plugins/jsobjc-dom-wrappers-expected.txt: Renamed from LayoutTests/plugins/jsobjc-dom-wrappers-expected.txt.
* platform/mac/plugins/jsobjc-dom-wrappers.html: Renamed from LayoutTests/plugins/jsobjc-dom-wrappers.html.
* platform/mac/plugins/jsobjc-simple-expected.txt: Renamed from LayoutTests/plugins/jsobjc-simple-expected.txt.
* platform/mac/plugins/jsobjc-simple.html: Renamed from LayoutTests/plugins/jsobjc-simple.html.
* platform/mac/plugins/qt-qwidget-plugin-expected.checksum: Removed.
* platform/mac/plugins/qt-qwidget-plugin-expected.png: Removed.
* platform/mac/plugins/qt-qwidget-plugin-expected.txt: Removed.
* platform/mac/plugins/root-object-premature-delete-crash-expected.txt: Renamed from LayoutTests/plugins/root-object-premature-delete-crash-expected.txt.
* platform/mac/plugins/root-object-premature-delete-crash.html: Renamed from LayoutTests/plugins/root-object-premature-delete-crash.html.
* platform/mac/plugins/throw-on-dealloc-expected.txt: Renamed from LayoutTests/plugins/throw-on-dealloc-expected.txt.
* platform/mac/plugins/throw-on-dealloc.html: Renamed from LayoutTests/plugins/throw-on-dealloc.html.
* platform/mac/plugins/undefined-property-crash-expected.txt: Renamed from LayoutTests/plugins/undefined-property-crash-expected.txt.
* platform/mac/plugins/undefined-property-crash.html: Renamed from LayoutTests/plugins/undefined-property-crash.html.
* platform/qt/Skipped:
* platform/qt/plugins/qt-qwidget-plugin.html: Renamed from LayoutTests/plugins/qt-qwidget-plugin.html.
* platform/win/Skipped:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@63616 268f45cc-cd09-0410-ab3c-d52691b4dbfc

[EFL] Unreviewed build system cleanup.

Patch by Leandro Pereira <leandro@profusion.mobi> on 2010-07-16
Remove reference to "mediaControlsGtk.css".

* cmake/WebKitGenerators.cmake:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@63615 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2010-07-15  Qi Zhang  <qi.2.zhang@nokia.com>

        Reviewed by Kenneth Rohde Christiansen.

        [Qt] Failure on http://philip.html5.org/tests/canvas/suite/tests/2d.shadow.alpha.5.html
        https://bugs.webkit.org/show_bug.cgi?id=38400

        FillRect with shadow need take alpha information from fillstyle

        * platform/graphics/qt/GraphicsContextQt.cpp:
        (WebCore::GraphicsContext::fillRect):
2010-07-15  Qi Zhang  <qi.2.zhang@nokia.com>

        Reviewed by Kenneth Rohde Christiansen.

        [Qt] Failure on http://philip.html5.org/tests/canvas/suite/tests/2d.shadow.alpha.5.html
        https://bugs.webkit.org/show_bug.cgi?id=38400

        Remove the following test case from Skipped:
        canvas/philip/tests/2d.shadow.alpha.5.html
        canvas/philip/tests/2d.shadow.gradient.alpha.html
        canvas/philip/tests/2d.shadow.gradient.transparent.1.html
        canvas/philip/tests/2d.shadow.image.scale.html
        canvas/philip/tests/2d.shadow.pattern.alpha.html
        canvas/philip/tests/2d.shadow.pattern.transparent.1.html

        * platform/qt/Skipped:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@63614 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Move NPRuntime related tests to plugins/npruntime
https://bugs.webkit.org/show_bug.cgi?id=42505

Reviewed by Dan Bernstein.

* platform/gtk/Skipped:
* platform/mac-snowleopard/Skipped:
* platform/win/Skipped:
Update skipped lists.

* plugins/netscape-identifier-conversion.html: Removed.
* plugins/npruntime/bindings-test-expected.txt: Renamed from LayoutTests/plugins/bindings-test-expected.txt.
* plugins/npruntime/bindings-test.html: Renamed from LayoutTests/plugins/bindings-test.html.
* plugins/npruntime/browser-object-identity-expected.txt: Renamed from LayoutTests/plugins/netscape-browser-object-identity-expected.txt.
* plugins/npruntime/browser-object-identity.html: Renamed from LayoutTests/plugins/netscape-browser-object-identity.html.
* plugins/npruntime/construct-expected.txt: Renamed from LayoutTests/plugins/netscape-construct-expected.txt.
* plugins/npruntime/construct.html: Renamed from LayoutTests/plugins/netscape-construct.html.
* plugins/npruntime/embed-property-expected.txt: Renamed from LayoutTests/plugins/embed-property-expected.txt.
* plugins/npruntime/embed-property.html: Renamed from LayoutTests/plugins/embed-property.html.
* plugins/npruntime/enumerate-expected.txt: Renamed from LayoutTests/plugins/netscape-enumerate-expected.txt.
* plugins/npruntime/enumerate.html: Renamed from LayoutTests/plugins/netscape-enumerate.html.
* plugins/npruntime/get-int-identifier-special-values-expected.txt: Renamed from LayoutTests/plugins/getintidentifier-special-values-expected.txt.
* plugins/npruntime/get-int-identifier-special-values.html: Renamed from LayoutTests/plugins/getintidentifier-special-values.html.
* plugins/npruntime/get-property-return-value-expected.txt: Renamed from LayoutTests/plugins/netscape-get-property-return-value-expected.txt.
* plugins/npruntime/get-property-return-value.html: Renamed from LayoutTests/plugins/netscape-get-property-return-value.html.
* plugins/npruntime/identifier-conversion-expected.txt: Renamed from LayoutTests/plugins/netscape-identifier-conversion-expected.txt.
* plugins/npruntime/identifier-conversion.html: Renamed from LayoutTests/plugins/netscape-identifier-conversion.html.
* plugins/npruntime/invoke-browserfuncs-expected.txt: Renamed from LayoutTests/plugins/netscape-invoke-browserfuncs-expected.txt.
* plugins/npruntime/invoke-browserfuncs.html: Renamed from LayoutTests/plugins/netscape-invoke-browserfuncs.html.
* plugins/npruntime/invoke-default-expected.txt: Renamed from LayoutTests/plugins/netscape-invoke-default-expected.txt.
* plugins/npruntime/invoke-default.html: Renamed from LayoutTests/plugins/netscape-invoke-default.html.
* plugins/npruntime/invoke-expected.txt: Renamed from LayoutTests/plugins/invoke-expected.txt.
* plugins/npruntime/invoke-failure-expected.txt: Renamed from LayoutTests/plugins/netscape-invoke-failure-expected.txt.
* plugins/npruntime/invoke-failure.html: Renamed from LayoutTests/plugins/netscape-invoke-failure.html.
* plugins/npruntime/invoke.html: Renamed from LayoutTests/plugins/invoke.html.
* plugins/npruntime/npruntime-expected.txt: Renamed from LayoutTests/plugins/npruntime-expected.txt.
* plugins/npruntime/npruntime.html: Renamed from LayoutTests/plugins/npruntime.html.
* plugins/npruntime/round-trip-npobject-expected.txt: Renamed from LayoutTests/plugins/round-trip-npobject-expected.txt.
* plugins/npruntime/round-trip-npobject.html: Renamed from LayoutTests/plugins/round-trip-npobject.html.
* plugins/npruntime/script-tests/TEMPLATE.html: Renamed from LayoutTests/plugins/script-tests/TEMPLATE.html.
* plugins/npruntime/script-tests/browser-object-identity.js: Renamed from LayoutTests/plugins/script-tests/netscape-browser-object-identity.js.
(description.gc):
* plugins/npruntime/script-tests/identifier-conversion.js: Renamed from LayoutTests/plugins/script-tests/netscape-identifier-conversion.js.
* plugins/npruntime/set-property-expected.txt: Renamed from LayoutTests/plugins/setProperty-expected.txt.
* plugins/npruntime/set-property.html: Renamed from LayoutTests/plugins/setProperty.html.
* plugins/npruntime/throw-exception-expected.txt: Renamed from LayoutTests/plugins/netscape-throw-exception-expected.txt.
* plugins/npruntime/throw-exception.html: Renamed from LayoutTests/plugins/netscape-throw-exception.html.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@63613 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Stop all NPStreams before destroying a plug-in
https://bugs.webkit.org/show_bug.cgi?id=42504

Reviewed by Dan Bernstein.

* WebProcess/Plugins/Netscape/NetscapePlugin.cpp:
(WebKit::NetscapePlugin::stopAllStreams):
Go through all streams and stop them.

(WebKit::NetscapePlugin::destroy):
Call stopAllStreams.

* WebProcess/Plugins/Netscape/NetscapePlugin.h:
Add stopAllStreams.

* WebProcess/Plugins/Netscape/NetscapePluginStream.h:
Make stop public.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@63612 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Open streams should not keep a plug-in view alive
https://bugs.webkit.org/show_bug.cgi?id=42503

Reviewed by Dan Bernstein.

PluginView::Stream now has a weak reference to its PluginView.

* WebProcess/Plugins/PluginView.cpp:
(WebKit::PluginView::Stream::~Stream):
Assert that the plug-in view is null.

(WebKit::PluginView::Stream::didFail):
After calling removeStream, set the plug-in view member variable to 0. This is OK to do
since we keep a reference to the Stream, so we're sure that the call to removeStream does not
destroy the stream.

(WebKit::PluginView::Stream::didFinishLoading):
Ditto .

(WebKit::PluginView::~PluginView):
Cancel all streams.

(WebKit::PluginView::cancelAllStreams):
Cancel all streams.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@63611 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2010-07-17  John Gregg  <johnnyg@google.com>

        Unreviewed, updating test expectations.

        [chromium] Updating the expecations for input-file-directory-upload.html
        https://bugs.webkit.org/show_bug.cgi?id=42419

        * platform/chromium/test_expectations.txt:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@63610 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2010-07-17  John Gregg  <johnnyg@google.com>

        Reviewed by Kent Tamura.

        [chromium] directory upload: pass the attribute up to the chrome client
        https://bugs.webkit.org/show_bug.cgi?id=42475

        This allows chromium to display the native system directory picker when
        the directory attribute is present on the input element.

        * public/WebFileChooserParams.h:
        * src/ChromeClientImpl.cpp:
        (WebKit::ChromeClientImpl::runOpenPanel):

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@63609 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2010-07-17  Nikita Vasilyev  <me@elv1s.ru>

        Reviewed by Pavel Feldman.

        Web Inspector: [REGRESSION] Edit long CSS attributes works incorrect
        https://bugs.webkit.org/show_bug.cgi?id=42476

        * inspector/front-end/inspector.css:
        (.editing):

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@63608 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2010-07-16  Andreas Kling  <andreas.kling@nokia.com>

        Reviewed by Sam Weinig.

        Failing 2d.path.arcTo.ensuresubpath.* philip canvas tests
        https://bugs.webkit.org/show_bug.cgi?id=42186

        Unskip tests:
        - canvas/philip/tests/2d.path.arcTo.ensuresubpath.1.html
        - canvas/philip/tests/2d.path.arcTo.ensuresubpath.2.html

        * platform/mac/Skipped:
2010-07-16  Andreas Kling  <andreas.kling@nokia.com>

        Reviewed by Sam Weinig.

        Failing 2d.path.arcTo.ensuresubpath.* philip canvas tests
        https://bugs.webkit.org/show_bug.cgi?id=42186

        Move code from Qt's Path::addArcTo() up to CanvasRenderingContext2D.

        * html/canvas/CanvasRenderingContext2D.cpp:
        (WebCore::CanvasRenderingContext2D::arcTo): Behave as moveTo(x1,y1)
        if the current path is empty.
        * platform/graphics/qt/PathQt.cpp:
        (WebCore::Path::addArcTo): Remove now-redundant code.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@63607 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2010-07-16  Andreas Kling  <andreas.kling@nokia.com>

        Reviewed by Oliver Hunt.

        QtWebkit creates an unnecessary deep copy of images when canvas drawing is done
        A https://bugs.webkit.org/show_bug.cgi?id=32530

        Solve this by adding ImageBuffer::imageForRendering() which returns an image
        that can be used for rendering now, but isn't a copy to be kept around.

        * platform/graphics/ImageBuffer.h:
        (WebCore::ImageBuffer::imageForRendering):
        * platform/graphics/qt/ImageBufferQt.cpp:
        (WebCore::ImageBuffer::imageForRendering): Added to provide an image that can
        be used for rendering now, but may change in the future.
        * platform/graphics/qt/StillImageQt.cpp:
        (WebCore::StillImage::StillImage):
        (WebCore::StillImage::~StillImage):
        (WebCore::StillImage::size):
        (WebCore::StillImage::nativeImageForCurrentFrame):
        (WebCore::StillImage::draw):
        * platform/graphics/qt/StillImageQt.h:
        (WebCore::StillImage::createForRendering): Added for use in
        ImageBuffer::imageForRendering(), provides a thin wrapper around a QPixmap*.
        * html/HTMLCanvasElement.cpp:
        (WebCore::HTMLCanvasElement::paint): Paint with ImageBuffer::imageForRendering()

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@63606 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2010-07-16  Andreas Kling  <andreas.kling@nokia.com>

        Reviewed by Oliver Hunt.

        [Qt] Remove redundant logic in Path::addArcTo()
        https://bugs.webkit.org/show_bug.cgi?id=42494

        Bounds checking for arcTo() is now done in cross-platform code
        thanks to <http://trac.webkit.org/changeset/63599>

        * platform/graphics/qt/PathQt.cpp:
        (WebCore::Path::addArcTo):

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@63605 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2010-07-16  Adam Barth  <abarth@webkit.org>

        Reviewed by Eric Seidel.

        Teach webkitpy about queues.webkit.org
        https://bugs.webkit.org/show_bug.cgi?id=42492

        * Scripts/webkitpy/common/net/statusserver.py:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@63604 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2010-07-16  Nico Weber  <thakis@chromium.org>

        Reviewed by Ojan Vafai.

        Add chromium/win and linux expectations for fast/canvas/toDataUrl-alpha.

        * platform/chromium-linux/fast/canvas/toDataURL-alpha-expected.checksum: Added.
        * platform/chromium-linux/fast/canvas/toDataURL-alpha-expected.png: Added.
        * platform/chromium-win/fast/canvas/toDataURL-alpha-expected.checksum: Added.
        * platform/chromium-win/fast/canvas/toDataURL-alpha-expected.png: Added.
        * platform/chromium-win/fast/canvas/toDataURL-alpha-expected.txt: Added.
        * platform/chromium/test_expectations.txt:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@63603 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2010-07-16  Zhe Su  <suzhe@chromium.org>

        Reviewed by Darin Adler.

        REGRESSION(r61484): Broke focus behaviour on Qt and probably other platforms
        https://bugs.webkit.org/show_bug.cgi?id=42253

        This patch just reverts the change to WebCore/page/FocusController.cpp
        made in changeset 61484, and add a new method named
        willSetInputMethodState in EditorClient interface, which gets called
        in FocusController just before changing the focused node.

        No new tests, because no new functionality.

        * loader/EmptyClients.h:
        (WebCore::EmptyEditorClient::willSetInputMethodState):
        * page/EditorClient.h:
        * page/FocusController.cpp:
        (WebCore::FocusController::setFocusedNode):
2010-07-16  Zhe Su  <suzhe@chromium.org>

        Reviewed by Darin Adler.

        REGRESSION(r61484): Broke focus behaviour on Qt and probably other platforms
        https://bugs.webkit.org/show_bug.cgi?id=42253

        Dummy implementation of EditorClient::willSetInputMethodState.

        * efl/WebCoreSupport/EditorClientEfl.cpp:
        (WebCore::EditorClientEfl::willSetInputMethodState):
        * efl/WebCoreSupport/EditorClientEfl.h:
2010-07-16  Zhe Su  <suzhe@chromium.org>

        Reviewed by Darin Adler.

        REGRESSION(r61484): Broke focus behaviour on Qt and probably other platforms
        https://bugs.webkit.org/show_bug.cgi?id=42253

        Implement EditorClient::willSetInputMethodState which replaces
        setInputMethodState method.

        * src/EditorClientImpl.cpp:
        (WebKit::EditorClientImpl::willSetInputMethodState):
        (WebKit::EditorClientImpl::setInputMethodState):
        * src/EditorClientImpl.h:
2010-07-16  Zhe Su  <suzhe@chromium.org>

        Reviewed by Darin Adler.

        REGRESSION(r61484): Broke focus behaviour on Qt and probably other platforms
        https://bugs.webkit.org/show_bug.cgi?id=42253

        Dummy implementation of EditorClient::willSetInputMethodState.

        * WebCoreSupport/EditorClientGtk.cpp:
        (WebKit::EditorClient::willSetInputMethodState):
        * WebCoreSupport/EditorClientGtk.h:
2010-07-16  Zhe Su  <suzhe@chromium.org>

        Reviewed by Darin Adler.

        REGRESSION(r61484): Broke focus behaviour on Qt and probably other platforms
        https://bugs.webkit.org/show_bug.cgi?id=42253

        Dummy implementation of EditorClient::willSetInputMethodState.

        * WebCoreSupport/EditorClientHaiku.cpp:
        (WebCore::EditorClientHaiku::willSetInputMethodState):
        * WebCoreSupport/EditorClientHaiku.h:
2010-07-16  Zhe Su  <suzhe@chromium.org>

        Reviewed by Darin Adler.

        REGRESSION(r61484): Broke focus behaviour on Qt and probably other platforms
        https://bugs.webkit.org/show_bug.cgi?id=42253

        Dummy implementation of EditorClient::willSetInputMethodState.

        * WebCoreSupport/WebEditorClient.h:
        * WebCoreSupport/WebEditorClient.mm:
        (WebEditorClient::willSetInputMethodState):
2010-07-16  Zhe Su  <suzhe@chromium.org>

        Reviewed by Darin Adler.

        REGRESSION(r61484): Broke focus behaviour on Qt and probably other platforms
        https://bugs.webkit.org/show_bug.cgi?id=42253

        Dummy implementation of EditorClient::willSetInputMethodState.

        * WebCoreSupport/EditorClientQt.cpp:
        (WebCore::EditorClientQt::willSetInputMethodState):
        * WebCoreSupport/EditorClientQt.h:
2010-07-16  Zhe Su  <suzhe@chromium.org>

        Reviewed by Darin Adler.

        REGRESSION(r61484): Broke focus behaviour on Qt and probably other platforms
        https://bugs.webkit.org/show_bug.cgi?id=42253

        Dummy implementation of EditorClient::willSetInputMethodState.

        * WebCoreSupport/WebEditorClient.cpp:
        (WebEditorClient::willSetInputMethodState):
        * WebCoreSupport/WebEditorClient.h:
2010-07-16  Zhe Su  <suzhe@chromium.org>

        Reviewed by Darin Adler.

        REGRESSION(r61484): Broke focus behaviour on Qt and probably other platforms
        https://bugs.webkit.org/show_bug.cgi?id=42253

        Dummy implementation of EditorClient::willSetInputMethodState.

        * WebKitSupport/EditorClientWx.cpp:
        (WebCore::EditorClientWx::willSetInputMethodState):
        * WebKitSupport/EditorClientWx.h:
2010-07-16  Zhe Su  <suzhe@chromium.org>

        Reviewed by Darin Adler.

        REGRESSION(r61484): Broke focus behaviour on Qt and probably other platforms
        https://bugs.webkit.org/show_bug.cgi?id=42253

        Dummy implementation of EditorClient::willSetInputMethodState.

        * WebProcess/WebCoreSupport/WebEditorClient.cpp:
        (WebKit::WebEditorClient::willSetInputMethodState):
        * WebProcess/WebCoreSupport/WebEditorClient.h:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@63602 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2010-07-16  Tony Chang  <tony@chromium.org>

        Not reviewed, updating test expectations.

        [chromium] clean up test expectations
        https://bugs.webkit.org/show_bug.cgi?id=42491

        * platform/chromium/test_expectations.txt: fast/dynamic/flash-replacement-test.html doesn't exist
          and svg/filters/filter-empty-g.svg is a duplicate on Linux.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@63601 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Build fix, not reviewed.

Reverted http://trac.webkit.org/changeset/63585 because getopt isn't
available in the OpenSource support libraries

* WebKit2.sln:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@63600 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2010-07-16  Matthew Delaney  <mdelaney@apple.com>

        Reviewed by Sam Weinig.

        Failing 2d.path.stroke.prune.arc philip canvas test
        https://bugs.webkit.org/show_bug.cgi?id=42188

        * platform/mac/Skipped: Unskipped now passing tests.
2010-07-16  Matthew Delaney  <mdelaney@apple.com>

        Reviewed by Sam Weinig.

        Failing 2d.path.stroke.prune.arc philip canvas test
        https://bugs.webkit.org/show_bug.cgi?id=42188

        * html/canvas/CanvasRenderingContext2D.cpp:
        Note, updated parameter names to match spec.
        (WebCore::CanvasRenderingContext2D::lineTo): Pulled bound checking code out of lower code to have checks for all platforms.
        (WebCore::CanvasRenderingContext2D::arcTo): Bound checking per the spec for arcTo parameters. Updated parameter names to match spec.
        * platform/graphics/Path.h: Added in new method to expose the current position.
        * platform/graphics/cairo/PathCairo.cpp:
        (WebCore::Path::currentPoint): Added in likely correct implementation for this call with a FIXME just in case.
        * platform/graphics/cg/PathCG.cpp:
        (WebCore::Path::currentPoint): Added in support for this call.
        * platform/graphics/haiku/PathHaiku.cpp:
        (WebCore::Path::currentPoint): Added in stub for this call.
        * platform/graphics/openvg/PathOpenVG.cpp:
        (WebCore::Path::currentPoint): Added in likely implementation for this call with a FIXME just in case.
        * platform/graphics/qt/PathQt.cpp:
        (WebCore::Path::currentPoint): Added in implementation for this call courtesy of Andrea Kling.
        * platform/graphics/skia/PathSkia.cpp:
        (WebCore::Path::currentPoint): Added in stub for this call.
        * platform/graphics/wince/PathWince.cpp:
        (WebCore::Path::currentPoint): Added in stub for this call.
        * platform/graphics/wx/PathWx.cpp:
        (WebCore::Path::currentPoint): Added in stub for this call.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@63599 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2010-07-16  Sheriff Bot  <webkit.review.bot@gmail.com>

        Unreviewed, rolling out r63593.
        http://trac.webkit.org/changeset/63593
        https://bugs.webkit.org/show_bug.cgi?id=42487

        Broke a few chromium pixel tests (Requested by tony^work on
        #webkit).

        * platform/chromium-linux/svg/stroke/path-zero-strokewidth-test-expected.checksum: Removed.
        * platform/chromium-linux/svg/stroke/path-zero-strokewidth-test-expected.png: Removed.
        * platform/chromium-linux/svg/stroke/path-zero-strokewidth-test-expected.txt: Removed.
        * platform/chromium-mac/svg/stroke/path-zero-strokewidth-test-expected.checksum: Removed.
        * platform/chromium-mac/svg/stroke/path-zero-strokewidth-test-expected.png: Removed.
        * platform/chromium-mac/svg/stroke/path-zero-strokewidth-test-expected.txt: Removed.
        * platform/mac/svg/stroke/path-zero-strokewidth-test-expected.checksum: Removed.
        * platform/mac/svg/stroke/path-zero-strokewidth-test-expected.png: Removed.
        * platform/mac/svg/stroke/path-zero-strokewidth-test-expected.txt: Removed.
        * svg/stroke/path-zero-strokewidth-test.svg: Removed.
2010-07-16  Sheriff Bot  <webkit.review.bot@gmail.com>

        Unreviewed, rolling out r63593.
        http://trac.webkit.org/changeset/63593
        https://bugs.webkit.org/show_bug.cgi?id=42487

        Broke a few chromium pixel tests (Requested by tony^work on
        #webkit).

        * platform/graphics/skia/GraphicsContextSkia.cpp:
        (WebCore::GraphicsContext::drawConvexPolygon):
        (WebCore::GraphicsContext::drawEllipse):
        (WebCore::GraphicsContext::drawLine):
        (WebCore::GraphicsContext::strokeArc):
        (WebCore::GraphicsContext::strokePath):
        (WebCore::GraphicsContext::strokeRect):

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@63598 268f45cc-cd09-0410-ab3c-d52691b4dbfc

    Reviewed by Alice Liu.

        Fix Windows build.

        * WebKitTestRunner/InjectedBundle/win/InjectedBundle.vcproj:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@63597 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2010-07-16  Ojan Vafai  <ojan@chromium.org>

        Unreviewed. Update expectations with new failures.

        * platform/chromium/test_expectations.txt:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@63596 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Part of <rdar://problem/7233974> Deprecate +[WebView _setShouldUseFontSmoothing:]
https://bugs.webkit.org/show_bug.cgi?id=29355

Reviewed by Sam Weinig.

WebCore:

* WebCore.exp.in: Updated.
* platform/graphics/Font.cpp:
(WebCore::Font::Font): Added a font smoothing mode parameter to the constructor.
Set the font smoothing mode in the font description.
* platform/graphics/Font.h:

WebKit/mac:

* Misc/WebKitNSStringExtras.h:
* Misc/WebKitNSStringExtras.mm:
(-[NSString _web_drawAtPoint:font:textColor:]): Now calls through to
-_web_drawAtPoint:font:textColor:allowingFontSmoothing: passing YES for the last
parameter.
(-[NSString _web_drawAtPoint:font:textColor:allowingFontSmoothing:]): Added
allowingFontSmoothing:. If font smoothing is not allowed, sets the font smoothing
mode to antialiased. Otherwise, sets it to auto.
(-[NSString _web_drawDoubledAtPoint:withTopColor:bottomColor:font:]): Instead of
changing the font smoothing setting in the graphics context, call
-_web_drawAtPoint:font:textColor:allowingFontSmoothing: passing NO for the last
parameter.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@63595 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2010-07-16  Satish Sampath  <satish@chromium.org>

        Reviewed by Anders Carlsson.

        Add speech attribute to IDL for enabling access from JS.
        https://bugs.webkit.org/show_bug.cgi?id=42483

        No tests added, this change is a pre-requisite for future layout tests.

        * html/HTMLInputElement.cpp:
        (WebCore::HTMLInputElement::parseMappedAttribute): Update the renderer when speech attribute changes.
        * html/HTMLInputElement.idl: Added the speech attribute to IDL.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@63594 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2010-07-16  Fady Samuel  <fsamuel@chromium.org>

        Reviewed by David Levin.

        Added layout test to show correct zero-width stroke.

        SVG - stroke-width:0 bug with stroke other than "none"
        https://bugs.webkit.org/show_bug.cgi?id=42387

        * platform/chromium-linux/svg/stroke/path-zero-strokewidth-test-expected.checksum: Added.
        * platform/chromium-linux/svg/stroke/path-zero-strokewidth-test-expected.png: Added.
        * platform/chromium-linux/svg/stroke/path-zero-strokewidth-test-expected.txt: Added.
        * platform/chromium-mac/svg/stroke/path-zero-strokewidth-test-expected.checksum: Added.
        * platform/chromium-mac/svg/stroke/path-zero-strokewidth-test-expected.png: Added.
        * platform/chromium-mac/svg/stroke/path-zero-strokewidth-test-expected.txt: Added.
        * platform/mac/svg/stroke/path-zero-strokewidth-test-expected.checksum: Added.
        * platform/mac/svg/stroke/path-zero-strokewidth-test-expected.png: Added.
        * platform/mac/svg/stroke/path-zero-strokewidth-test-expected.txt: Added.
        * svg/stroke/path-zero-strokewidth-test.svg: Added.
2010-07-16  Fady Samuel  <fsamuel@chromium.org>

        Reviewed by David Levin.

        Avoids adding stroke when stroke-width is zero.

        SVG - stroke-width:0 bug with stroke other than "none"
        https://bugs.webkit.org/show_bug.cgi?id=42387

        Test: svg/stroke/path-zero-strokewidth-test.svg

        * platform/graphics/skia/GraphicsContextSkia.cpp:
        (WebCore::GraphicsContext::drawConvexPolygon):
        (WebCore::GraphicsContext::drawEllipse):
        (WebCore::GraphicsContext::drawLine):
        (WebCore::GraphicsContext::strokeArc):
        (WebCore::GraphicsContext::strokePath):
        (WebCore::GraphicsContext::strokeRect):

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@63593 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2010-07-16  Kent Tamura  <tkent@chromium.org>

        Unreviewed, build fix.

        * rendering/RenderThemeChromiumWin.cpp:
        (WebCore::RenderThemeChromiumWin::getThemeData): Fix a typo.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@63592 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Remove semicolons from Objective-C method declarations.

Reviewed by Sam Weinig.

* WebView/WebRenderLayer.mm:
(+[WebRenderLayer nameForLayer:]):
(+[WebRenderLayer compositingInfoForLayer:]):

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@63591 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Fix for https://bugs.webkit.org/show_bug.cgi?id=42482
<rdar://problem/8197701>
Add notification of when the BackForwardList changes
to aid invalidation of Back/Forward related UI elements.

Reviewed by Anders Carlsson.

WebKit2:

* UIProcess/API/C/WKPage.h:
Add didChangeBackForwardList to the WKPageLoaderClient. This
fires whenever an item is added or removed from the back forward
list or when the cursor changes position.

* UIProcess/WebBackForwardList.cpp:
(WebKit::WebBackForwardList::addItem):
(WebKit::WebBackForwardList::goToItem):
* UIProcess/WebLoaderClient.cpp:
(WebKit::WebLoaderClient::didChangeBackForwardList):
* UIProcess/WebLoaderClient.h:
* UIProcess/WebPageProxy.cpp:
(WebKit::WebPageProxy::didChangeBackForwardList):
* UIProcess/WebPageProxy.h:
Pipe changes to the WebBackForwardList up to the page load client.

WebKitTools:

* MiniBrowser/mac/BrowserWindowController.m:
(_didStartProvisionalLoadForFrame):
(_didCommitLoadForFrame):
(_didFailLoadWithErrorForFrame):
(_didChangeBackForwardList):
(-[BrowserWindowController awakeFromNib]):
* WebKitTestRunner/WebKitTestRunner.xcodeproj/project.pbxproj:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@63590 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2010-07-16  Kent Tamura  <tkent@chromium.org>

        Reviewed by Darin Fisher.

        [Chromium] <input type=number> UI implementation for Windows
        https://bugs.webkit.org/show_bug.cgi?id=42259

        No additional tests.  Existing tests cover this change and we'll
        update expectations.

        * platform/chromium/ChromiumBridge.h: Add paintSpinButton().
        * rendering/RenderThemeChromiumWin.cpp:
        (WebCore::RenderThemeChromiumWin::determineState):
         Add ControlSubPart parameter and add support for spin buttons.
        (WebCore::RenderThemeChromiumWin::determineClassicState): ditto.
        (WebCore::RenderThemeChromiumWin::getThemeData): ditto.
        (WebCore::RenderThemeChromiumWin::adjustInnerSpinButtonStyle): Added.
        (WebCore::RenderThemeChromiumWin::paintInnerSpinButton): Added.
        * rendering/RenderThemeChromiumWin.h:
          Declare ControlSubpart, and add it to some functions.
2010-07-16  Kent Tamura  <tkent@chromium.org>

        Reviewed by Darin Fisher.

        [Chromium] <input type=number> UI implementation for Windows
        https://bugs.webkit.org/show_bug.cgi?id=42259

        * public/WebThemeEngine.h:
        (WebKit::WebThemeEngine::paintSpinButton): Added.
        * src/ChromiumBridge.cpp:
        (WebCore::ChromiumBridge::paintSpinButton):
          Added. It calls WebThemeEngine::paintSpinButton().
2010-07-16  Kent Tamura  <tkent@chromium.org>

        Reviewed by Darin Fisher.

        [Chromium] <input type=number> UI implementation for Windows
        https://bugs.webkit.org/show_bug.cgi?id=42259

        * DumpRenderTree/chromium/WebThemeEngineDRT.cpp:
        (WebThemeEngineDRT::paintSpinButton):
         Added. Check state consistency and use existing arrow painting code.
        * DumpRenderTree/chromium/WebThemeEngineDRT.h:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@63589 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2010-07-16  Lucas De Marchi  <lucas.demarchi@profusion.mobi>

        Reviewed by Antonio Gomes.

        [EFL] Use function provided by EFL for system beep
        https://bugs.webkit.org/show_bug.cgi?id=42481

        EFL port does not support automated tests, yet.

        * platform/efl/SoundEfl.cpp:
        (WebCore::systemBeep):

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@63588 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2010-07-16  Sarah Strong  <sarah.e.strong@gmail.com>

        Reviewed by Gustavo Noronha Silva.

        [GTK] Clipboard data is lost on exit
        https://bugs.webkit.org/show_bug.cgi?id=27411

        No new tests. To manually test the bug fixed by this patch:
        1) Open an application that uses a webkit webview on a GNOME-based system
        2) Copy some text from that application
        3) Optional: paste it to another application. This should work properly with and without this patch.
        4) Exit the application completely.
        5) Paste:
                  Without this patch, you cannot because your clipboard is empty.
                  With this patch, pasting succeeds.
        I have not included an automated test because of the difficulty of testing behaviour after application exit.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@63587 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Keyboard operations for <input type=number>
https://bugs.webkit.org/show_bug.cgi?id=42076

Reviewed by Darin Fisher.

WebCore:

- The up arrow key works as stepUp().
- The down arrow key works as stepDown().
- Reject characters other than + - 0-9 . e E

Test: fast/forms/input-number-keyoperation.html

* html/HTMLInputElement.cpp:
(WebCore::HTMLInputElement::defaultEventHandler):
 Add up/down arrow keys support, and call handleBeforeTextInsertedEvent().
(WebCore::isNumberCharacter):
(WebCore::HTMLInputElement::handleBeforeTextInsertedEvent):
 For type=number, remove unacceptable characters.
* html/HTMLInputElement.h:

LayoutTests:

* fast/forms/input-number-keyoperation-expected.txt: Added.
* fast/forms/input-number-keyoperation.html: Added.
* fast/forms/script-tests/input-number-keyoperation.js: Added.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@63586 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Add WebKitTestRunner to the WebKit2 solution

Reviewed by Sam Weinig.

* WebKit2.sln: Add InjectedBundle and WebKitTestRunner projects.  Also change the build dependency order from
DumpRunderTree --> WebKitAPITest to
DumpRenderTree --> InjectedBundle --> WebKitTestRunner --> WebKitAPITest

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@63585 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Land updated expected results for fast/dom/prototype-inheritance-2-expected.txt on Windows.

* platform/win/fast/dom/prototype-inheritance-2-expected.txt:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@63584 268f45cc-cd09-0410-ab3c-d52691b4dbfc

https://bugs.webkit.org/show_bug.cgi?id=42430
Make WebKitTestRunner on Windows actually load and run a test

Reviewed by Sam Weinig.

* WebKitTestRunner/InjectedBundle/InjectedBundleMain.cpp: Export WKBundleInitialize.
(WKBundleInitialize):
* WebKitTestRunner/InjectedBundle/win/InjectedBundle.vcproj: Added.  Output is a dll.
* WebKitTestRunner/TestInvocation.cpp: Create a Windows-style path on Windows
(WTR::createWKURL):
* WebKitTestRunner/WebKitTestRunnerPrefix.h: Change check for Windows platform to match
other existing checks, and avoid including Platform.h
* WebKitTestRunner/win/PlatformWebViewWin.cpp:
(WTR::registerWindowClass): Added.
(WTR::PlatformWebView::PlatformWebView): Implemented.
(WTR::PlatformWebView::~PlatformWebView): Implemented.
(WTR::PlatformWebView::page): Implemented.
* WebKitTestRunner/win/TestControllerWin.cpp:
(WTR::TestController::initializeInjectedBundlePath): Implemented. Provide build
configuration specific path to InjectedBundle.dll.
* WebKitTestRunner/win/TestInvocationWin.cpp:
(WTR::TestInvocation::runUntil): Implemented.
* WebKitTestRunner/win/WebKitTestRunner.sln: Added InjectedBundle project.
* WebKitTestRunner/win/WebKitTestRunner.vcproj: Cleaned up unnecessary dependencies.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@63583 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2010-07-16  Kent Tamura  <tkent@chromium.org>

        Reviewed by Darin Fisher.

        Improve hover state handling for spin buttons
        https://bugs.webkit.org/show_bug.cgi?id=42260

        Background:
        When we move the mouse cursor to a node from the outside of the node,
        the following steps are executed.
         1. setHovered(true) is called.
         2. The node is repainted for the hover state.
         3. 'mousemove' event is dispatched for the node.
        For a spin-button, RenderTheme::paint{Inner,Outer}SpinButton() is
        called before the event handler of the spin-button. So we can't
        detect which of the up part or the down part is hovered correctly.

        Solution:
        The hover state of a spin-button is one of three states;
        Indeterminate, Up, and Down. The state is Indeterminate since
        setHovered(true) is called and until 'mousemove' event is
        dispatched.

        No new tests because there are no implementation of spin-buttons
        with hovered state yet.

        * rendering/RenderTheme.cpp:
        (WebCore::RenderTheme::isSpinUpButtonPartPressed):
        (WebCore::RenderTheme::isHovered):
         Return false if the node is a spin-button and the state is Indeterminate.
        (WebCore::RenderTheme::isSpinUpButtonPartHovered):
        * rendering/TextControlInnerElements.cpp:
        (WebCore::SpinButtonElement::SpinButtonElement):
         Initialize m_upDownState.
        (WebCore::SpinButtonElement::defaultEventHandler):
        (WebCore::SpinButtonElement::setHovered):
         Set the state to Indeterminate.
        * rendering/TextControlInnerElements.h:
        (WebCore::SpinButtonElement::upDownState):

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@63582 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2010-07-16  Leandro Pereira  <leandro@profusion.mobi>

        [EFL] Unreviewed build system cleanup.

        Move ExecutableAllocator{FixedVMPool,Posix,Symbian,Win}.cpp from
        root CMakeLists.txt to the platform CMakeLists.txt.

        * CMakeLists.txt:
        * CMakeListsEfl.txt: Add ExecutableAllocatorPosix.cpp.
2010-07-16  Leandro Pereira  <leandro@profusion.mobi>

        [EFL] Unreviewed build system cleanup.

        Remove reference to "mediaControlsGtk.css".

        * cmake/WebKitGenerators.cmake:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@63581 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2010-07-16  James Robinson  <jamesr@chromium.org>

        Reviewed by Darin Fisher.

        [chromium] Avoid crashing if offscreen GLES2 context creation fails
        https://bugs.webkit.org/show_bug.cgi?id=42469

        If creating or initializing an offscreen GLES2Context fails we should return null
        rather than crashing or returning an invalid context.

        * src/WebViewImpl.cpp:
        (WebKit::WebViewImpl::getOffscreenGLES2Context):

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@63580 268f45cc-cd09-0410-ab3c-d52691b4dbfc

<rdar://problem/7527532> Crash beneath setSelection() during detach()
https://bugs.webkit.org/show_bug.cgi?id=42020

Reviewed by Simon Fraser.

No test because I am unable to reproduce the crash.

* rendering/RenderView.cpp:
(WebCore::RenderView::setSelection): In the clearSelection() case, where the repaint mode is
RepaintNewMinusOld, avoid making RenderBlockSelectionInfo instances, and thereby avoid calling
localToAbsolute() during detach().

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@63579 268f45cc-cd09-0410-ab3c-d52691b4dbfc

clang++ build fixes for JavaScriptCore and WebCore
https://bugs.webkit.org/show_bug.cgi?id=42478

Reviewed by Sam Weinig.

JavaScriptCore:

* runtime/RegExpKey.h:
(JSC::operator==):
Move the RegExpKey equals operator into the JSC namespace so it can be found by ADL.

WebCore:

* platform/network/Credential.cpp:
(WebCore::Credential::type):
* platform/network/Credential.h:
Remove const qualifier on Credential::type since it doesn't have an effect on the type.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@63578 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Fix release build.

* storage/Database.cpp:
(WebCore::DerefContextTask::performTask):

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@63577 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2010-07-16  Simon Fraser  <simon.fraser@apple.com>

        Reviewed by Sam Weinig.

        Safari pegs CPU and drops frames on http://neography.com/experiment/circles/solarsystem/ (CSS animations)
        https://bugs.webkit.org/show_bug.cgi?id=41409

        AnimationController::isAnimatingPropertyOnRenderer() really asked whether an accelerated animation
        or transition was running. This prevented us from falling into compositing layers for animation
        on platforms, like Windows, that don't have accelerated animations.

        Fix by making things more explicit: we now have two methods, isRunningAnimationOnRenderer()
        and isRunningAcceleratedAnimationOnRenderer().

        Changes are more extensive because I flipped the sense of 'm_fallbackAnimating', which is
        now 'm_isAccelerated', for clarity.

        Test: compositing/animation/animation-compositing.html

        * page/animation/AnimationBase.cpp:
        (WebCore::AnimationBase::AnimationBase): m_fallbackAnimating -> m_isAccelerated
        (WebCore::AnimationBase::blendProperties): Ditto.
        (WebCore::AnimationBase::updateStateMachine): Ditto.
        * page/animation/AnimationBase.h:
        (WebCore::AnimationBase::isAnimatingProperty): Takes new acceleratedOnly parameter
        which causes the method to only return true if the animation is accelerated.
        (WebCore::AnimationBase::isAccelerated): Changed from isFallbackAnimating.

        * page/animation/AnimationController.cpp:
        (WebCore::AnimationControllerPrivate::isRunningAnimationOnRenderer):
        (WebCore::AnimationControllerPrivate::isRunningAcceleratedAnimationOnRenderer):
        (WebCore::AnimationController::isRunningAnimationOnRenderer):
        (WebCore::AnimationController::isRunningAcceleratedAnimationOnRenderer):

        * page/animation/CompositeAnimation.h:
        * page/animation/AnimationController.h: Rename isAnimatingPropertyOnRenderer(), add
        isRunningAcceleratedAnimationOnRenderer().
        * page/animation/AnimationControllerPrivate.h: Ditto.
        * page/animation/CompositeAnimation.cpp:
        (WebCore::CompositeAnimation::updateTransitions): !isFallbackAnimating() -> isAccelerated().
        (WebCore::CompositeAnimation::isAnimatingProperty): Pass acceleratedOnly down.

        * page/animation/ImplicitAnimation.cpp:
        (WebCore::ImplicitAnimation::timeToNextService): !isFallbackAnimating() -> isAccelerated().

        * page/animation/KeyframeAnimation.cpp:
        (WebCore::KeyframeAnimation::timeToNextService): isFallbackAnimating() -> !isAccelerated().

        * rendering/RenderLayerBacking.cpp:
        (WebCore::RenderLayerBacking::updateGraphicsLayerGeometry): Avoid touching the transform or
        opacity if an accelerated animation is running.

        * rendering/RenderLayerCompositor.cpp:
        (WebCore::RenderLayerCompositor::requiresCompositingForAnimation): Make compositing layers
        if an animation of transform or opacity is running.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@63576 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2010-07-16  Ojan Vafai  <ojan@chromium.org>

        Unreviewed. Put expected result from r63568 in the right place.
        It should apply to windows and linux, not just linux.

        * platform/chromium-win/svg/custom/relative-sized-content-with-resources-expected.txt: Renamed from LayoutTests/platform/chromium-linux/svg/custom/relative-sized-content-with-resources-expected.txt.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@63575 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2010-07-16  Tony Gentilcore  <tonyg@chromium.org>

        Reviewed by Ojan Vafai.

        Fix fast/dom/prototype-inheritance.html for chromium
        https://bugs.webkit.org/show_bug.cgi?id=42477

        * fast/dom/script-tests/prototype-inheritance.js:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@63574 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2010-07-16  Simon Fraser  <simon.fraser@apple.com>

        Reviewed by Sam Weinig.

        Assertion when turning Accelerated Compositing off on a composited page
        https://bugs.webkit.org/show_bug.cgi?id=42408

        When accelerated compositing is turned off, check m_hasAcceleratedCompositing
        before saying that the root layer will be composited. Fixes an assertion.

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

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@63573 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Spatial navigation: do not consider outline for focusable element boundaries
https://bugs.webkit.org/show_bug.cgi?id=42474

Reviewed by Simon Fraser.
Patch by Antonio Gomes <tonikitoo@webkit.org>

WebCore:

Test: fast/events/spatial-navigation/snav-zero-margin-content.html

Currently in WebCore::renderRectRelativeToRootDocument function, we are calling
RenderObject::absoluteClippedOverflowRect to obtain the rect boundary of a given
renderer/element. This method deals with outline, which is out of elements boundary.
It makes spatial navigation to fail on common sites like google.gom: "Web, Images, Map, etc"
are inaccessible.

Patch replaces RenderObject::absoluteClippedOverflowRect by Node::getRect,
which returns only the absolute bounding box rect of the Element.

* page/SpatialNavigation.cpp:
(WebCore::renderRectRelativeToRootDocument):
(WebCore::checkNegativeCoordsForNode):

LayoutTests:

* fast/events/spatial-navigation/snav-zero-margin-content-expected.txt: Added.
* fast/events/spatial-navigation/snav-zero-margin-content.html: Added.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@63572 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2010-07-15  Antonio Gomes  <tonikitoo@webkit.org>

        Reviewed by Gustavo Noronha.

        [Qt] Remove unnecessary WebKit headers inclusion from WebCore files
        https://bugs.webkit.org/show_bug.cgi?id=42416

        There are some places in WebCore unnecessarily including WebKit headers.

        Cleaning up only, no testing needed.

        * platform/network/qt/ResourceHandleQt.cpp:
        * platform/qt/RenderThemeQt.c

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@63571 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Refinement of r63556: moved the tab width computation back to Font, but added a
SimpleFontData parameter.

Reviewed by Anders Carlsson.

* platform/graphics/Font.h:
(WebCore::Font::tabWidth):
* platform/graphics/WidthIterator.cpp:
(WebCore::WidthIterator::advance):
* platform/graphics/mac/ComplexTextController.cpp:
(WebCore::ComplexTextController::adjustGlyphsAndAdvances):

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@63570 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2010-07-16  Tony Gentilcore  <tonyg@chromium.org>

        Reviewed by Pavel Feldman.

        Fix unit error in calculating timings from ResourceLoadTiming API
        https://bugs.webkit.org/show_bug.cgi?id=42473

        The start and end offsets are already in milliseconds so they should not be multiplied by 1000.0 like requestTime.

        No new tests because, for a yet unknown reason, the ResourceLoadTiming API is not populated in the chromium test shell like it is in full chromium. This means the test that would have caught this bug (fast/dom/webtiming.html) currently has its expectation set to FAIL.

        * page/Timing.cpp:
        (WebCore::Timing::domainLookupStart):
        (WebCore::Timing::domainLookupEnd):
        (WebCore::Timing::connectStart):
        (WebCore::Timing::connectEnd):
        (WebCore::Timing::requestStart):
        (WebCore::Timing::requestEnd):
        (WebCore::Timing::responseStart):

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@63569 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2010-07-16  Ojan Vafai  <ojan@chromium.org>

        Unreviewed. Add another expected chromium failure and the missing the text expectation in r63567.

        * platform/chromium-linux/svg/custom/relative-sized-content-with-resources-expected.txt: Added.
        * platform/chromium/test_expectations.txt:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@63568 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2010-07-16  Ojan Vafai  <ojan@chromium.org>

        Unreviewed. Add new chromium expectations and update expected failures. All updated expected
        failures have the relevant details in the bugs.webkit.org bug.

        Below are the revisions that caused needing new expectations.

        // http://trac.webkit.org/changeset/63531/
        svg/custom/relative-sized-content-with-resources.xhtml
        svg/custom/marker-child-changes-css.svg

        // http://trac.webkit.org/changeset/63542/
        svg/text/selection-background-color.xhtml
        svg/text/selection-styles.xhtml

        // http://trac.webkit.org/changeset/63517/
        fast/css/empty-generated-content.html
        css2.1/t1204-root-e.html

        // http://trac.webkit.org/changeset/63517/
        html5lib/runner.html

        * platform/chromium-linux/svg/custom/marker-child-changes-css-expected.checksum: Copied from LayoutTests/platform/chromium-linux/svg/custom/marker-child-changes-expected.checksum.
        * platform/chromium-linux/svg/custom/marker-child-changes-css-expected.png: Copied from LayoutTests/platform/chromium-linux/svg/custom/marker-child-changes-expected.png.
        * platform/chromium-linux/svg/custom/relative-sized-content-with-resources-expected.checksum: Added.
        * platform/chromium-linux/svg/custom/relative-sized-content-with-resources-expected.png: Added.
        * platform/chromium-linux/svg/text/selection-background-color-expected.checksum:
        * platform/chromium-linux/svg/text/selection-background-color-expected.png:
        * platform/chromium-linux/svg/text/selection-styles-expected.checksum:
        * platform/chromium-linux/svg/text/selection-styles-expected.png:
        * platform/chromium-mac/html5lib/runner-expected.txt:
        * platform/chromium-win/css2.1/t1204-root-e-expected.txt:
        * platform/chromium-win/fast/css/empty-generated-content-expected.txt:
        * platform/chromium-win/html5lib/runner-expected.txt:
        * platform/chromium-win/svg/custom/marker-child-changes-css-expected.checksum: Copied from LayoutTests/platform/chromium-win/svg/custom/marker-child-changes-expected.checksum.
        * platform/chromium-win/svg/custom/marker-child-changes-css-expected.png: Copied from LayoutTests/platform/chromium-win/svg/custom/marker-child-changes-expected.png.
        * platform/chromium-win/svg/custom/relative-sized-content-with-resources-expected.checksum: Added.
        * platform/chromium-win/svg/custom/relative-sized-content-with-resources-expected.png: Added.
        * platform/chromium-win/svg/text/selection-background-color-expected.checksum:
        * platform/chromium-win/svg/text/selection-background-color-expected.png:
        * platform/chromium-win/svg/text/selection-styles-expected.checksum:
        * platform/chromium-win/svg/text/selection-styles-expected.png:
        * platform/chromium/test_expectations.txt:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@63567 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2010-07-16  Oliver Hunt  <oliver@apple.com>

        Reviewed by Geoffrey Garen.

        ES5 allows use of reserved words as IdentifierName
        https://bugs.webkit.org/show_bug.cgi?id=42471

        Modify the lexer to allow us to avoid identifying reserved
        words in those contexts where they are valid identifiers, and
        we know it's safe.  Additionally tag the reserved word tokens
        so we can easily identify them in those cases where we can't
        guarantee that we've skipped reserved word identification.

        * parser/JSParser.cpp:
        (JSC::JSParser::next):
        (JSC::JSParser::parseProperty):
        (JSC::JSParser::parseMemberExpression):
        * parser/JSParser.h:
        (JSC::):
        * parser/Lexer.cpp:
        (JSC::Lexer::lex):
        * parser/Lexer.h:
        (JSC::Lexer::):
2010-07-16  Oliver Hunt  <oliver@apple.com>

        Reviewed by Geoffrey Garen.

        ES5 allows use of reserved words as IdentifierName
        https://bugs.webkit.org/show_bug.cgi?id=42471

        Add tests to check for correct handling of reserved words with
        the ES5 semantics.

        * fast/js/reserved-words-as-property-expected.txt: Added.
        * fast/js/reserved-words-as-property.html: Added.
        * fast/js/script-tests/reserved-words-as-property.js: Added.
        ():
        * fast/js/sputnik/Conformance/11_Expressions/11.1_Primary_Expressions/11.1.5_Object_Initializer/S11.1.5_A4.1-expected.txt:
        * fast/js/sputnik/Conformance/11_Expressions/11.1_Primary_Expressions/11.1.5_Object_Initializer/S11.1.5_A4.2-expected.txt:
          These tests are wrong, unsure how to get sputnik tests corrected.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@63566 268f45cc-cd09-0410-ab3c-d52691b4dbfc

LayoutTests/http/tests/cookies/multiple-cookies.html sometimes fails because the cookies are returned in a different order
https://bugs.webkit.org/show_bug.cgi?id=42095

Reviewed by Alexey Proskuryakov.

Normalize the cookie strings by splitting them into an array, sorting the array and then joining back.

* http/tests/cookies/multiple-cookies-expected.txt:
* http/tests/cookies/resources/cookies-test-pre.js:
(normalizeCookie):
(testCookies):
* websocket/tests/httponly-cookie-expected.txt:
* websocket/tests/httponly-cookie.pl:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@63565 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2010-07-16  Brady Eidson  <beidson@apple.com>

        Reviewed by Sam Weinig.

        Make revalidation of back/forward work a little better with page loads
        https://bugs.webkit.org/show_bug.cgi?id=42470

        * MiniBrowser/mac/BrowserWindow.xib:
        * MiniBrowser/mac/BrowserWindowController.h:
        * MiniBrowser/mac/BrowserWindowController.m:
        (-[BrowserWindowController validateToolbar]):
        (_didStartProvisionalLoadForFrame):
        (_didCommitLoadForFrame):
        (_didFailLoadWithErrorForFrame):

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@63564 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Remove fast/dynamic/flash-replacement-test.html
https://bugs.webkit.org/show_bug.cgi?id=42467

Reviewed by Simon Fraser.

This test doesn't really test anything since the flash file it points to doesn't exist, and
the flash replacement test requires a SWF file that triggers a style layout when instantiating the plug-in.

Furthermore, we don't even instantiate Netscape plug-ins from attach anymore which makes this test
even less accurate.

Also, it's failing on Mac if Flash 10.1, which uses the Core Animation drawing model, is installed.

* fast/dynamic/flash-replacement-test.html: Removed.
* platform/chromium-linux/fast/dynamic/flash-replacement-test-expected.checksum: Removed.
* platform/chromium-linux/fast/dynamic/flash-replacement-test-expected.png: Removed.
* platform/chromium-win/fast/dynamic/flash-replacement-test-expected.checksum: Removed.
* platform/chromium-win/fast/dynamic/flash-replacement-test-expected.png: Removed.
* platform/chromium-win/fast/dynamic/flash-replacement-test-expected.txt: Removed.
* platform/mac/fast/dynamic/flash-replacement-test-expected.checksum: Removed.
* platform/mac/fast/dynamic/flash-replacement-test-expected.png: Removed.
* platform/mac/fast/dynamic/flash-replacement-test-expected.txt: Removed.
* platform/qt/fast/dynamic/flash-replacement-test-expected.txt: Removed.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@63563 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Really add WARN_UNUSED_RESULT to leakRef
https://bugs.webkit.org/show_bug.cgi?id=42464

Reviewed by David Levin.

JavaScriptCore:

* wtf/PassRefPtr.h:
(WTF::PassRefPtr::):
(WTF::NonNullPassRefPtr::):
Put the WARN_UNUSED_RESULT attribute at the right place.

* wtf/RetainPtr.h:
(WTF::RetainPtr::releaseRef):
Remove WARN_UNUSED_RESULT here for now, it leads to two warnings that need
to be fixed first.

WebCore:

Get rid of a call to releaseRef here by passing the ScriptExecutionContext
reference through to the DerefContextTask.

* storage/Database.cpp:
(WebCore::DerefContextTask::create):
(WebCore::DerefContextTask::performTask):
(WebCore::DerefContextTask::DerefContextTask):
(WebCore::Database::~Database):

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@63562 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2010-07-16  Ilya Tikhonovsky  <loislo@chromium.org>

        Reviewed by Yury Semikhatsky.

        WebInspector: The current implementation of generator is not ready
        for generation Backend part of Inspector interface. The full patch
        with Backend is quite big and I've split it.
        https://bugs.webkit.org/show_bug.cgi?id=42462

        * inspector/CodeGeneratorInspector.pm:
        * inspector/Inspector.idl:
        * inspector/InspectorValues.cpp:
        (WebCore::InspectorArray::get):
        * inspector/InspectorValues.h:
        (WebCore::InspectorArray::length):

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@63561 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2010-07-15  Rob Buis  <rwlbuis@gmail.com>

        Reviewed by Darin Adler.

        An empty value for xml:lang isn't considered
        https://bugs.webkit.org/show_bug.cgi?id=42042

        Allow :lang selector to match empty values for xml:lang and
        lang attributes.

        Test: fast/css/lang-selector-empty-attribute.xhtml

        * css/CSSStyleSelector.cpp:
        (WebCore::CSSStyleSelector::SelectorChecker::checkOneSelector):

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@63560 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2010-07-15  Tony Gentilcore  <tonyg@chromium.org>

        Reviewed by Darin Fisher.

        Update expectations for Web Timing enabled in chromium port.
        https://bugs.webkit.org/show_bug.cgi?id=42313

        Default expectations are to PASS (Web Timing enabled). Since the gtk, qt, win, and mac platforms do not have WEB_TIMING enabled, they must skip these tests.

        * fast/dom/Window/window-properties-performance-expected.txt: Added.
        * fast/dom/Window/window-properties-performance.html: Added.
        * fast/dom/Window/window-properties.html: Skip webkitPerformance so that its platform-specific expectations can be managed in window-properties-performance.html.
        * fast/dom/navigation-type-back-forward-expected.txt: Update to PASS
        * fast/dom/navigation-type-navigate-expected.txt: Update to PASS
        * fast/dom/navigation-type-reload-expected.txt: Update to PASS
        * fast/dom/webtiming-expected.txt: Updated to mostly PASS. Currently has failures because chromium is not populating the ResourceLoadTiming API. This will be resolved when http://codereview.chromium.org/2909016 lands.
        * fast/dom/webtiming.html: Update test based on new backfill functionality from https://bugs.webkit.org/show_bug.cgi?id=42303.
        * http/tests/misc/webtiming-one-redirect-expected.txt: Update to PASS
        * http/tests/misc/webtiming-two-redirects-expected.txt: Update to PASS
        * platform/gtk/Skipped:
        * platform/mac/Skipped:
        * platform/qt/Skipped:
        * platform/win/Skipped:
2010-07-15  Tony Gentilcore  <tonyg@chromium.org>

        Reviewed by Darin Fisher.

        Enable window.webkitPerformance (Web Timing) for chromium
        https://bugs.webkit.org/show_bug.cgi?id=42313

        * features.gypi:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@63559 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2010-07-16  Alexander Pavlov  <apavlov@chromium.org>

        Reviewed by Pavel Feldman.

        Web Inspector: Shift-Enter does not do a reverse search. It searches forward.
        https://bugs.webkit.org/show_bug.cgi?id=42459

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

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@63558 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2010-07-16  Pavel Feldman  <pfeldman@chromium.org>

        Reviewed by Yury Semikhatsky.

        Web Inspector: do not include SSL time into Waiting time.

        https://bugs.webkit.org/show_bug.cgi?id=42458

        * inspector/front-end/ResourcesPanel.js:
        (WebInspector.ResourcesPanel.prototype._showPopover):

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@63557 268f45cc-cd09-0410-ab3c-d52691b4dbfc

<rdar://problem/8198266> white-space: pre text containing tabs is not laid out correctly when the font lacks a space glyph
https://bugs.webkit.org/show_bug.cgi?id=42437

Reviewed by Anders Carlsson.

No test because none of the fonts available to DumpRenderTree are missing a space glyph.

Changed the tab width computation to use the width of the space glyph from the font that has
a space glyph, which may be a fallback font if the primary font lacks a space glyph.

* platform/graphics/Font.h: Removed Font::tabWidth().
* platform/graphics/WidthIterator.cpp:
(WebCore::WidthIterator::advance):
* platform/graphics/mac/ComplexTextController.cpp:
(WebCore::ComplexTextController::adjustGlyphsAndAdvances):

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@63556 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2010-07-16  Sheriff Bot  <webkit.review.bot@gmail.com>

        Unreviewed, rolling out r63551.
        http://trac.webkit.org/changeset/63551
        https://bugs.webkit.org/show_bug.cgi?id=42460

        "Problems with gyp" (Requested by yurys on #webkit).

        * DEPS:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@63555 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2010-07-16  Pavel Feldman  <pfeldman@chromium.org>

        Reviewed by Yury Semikhatsky.

        Web Inspector: add SSL time label into the resources popover.

        https://bugs.webkit.org/show_bug.cgi?id=42458

        * English.lproj/localizedStrings.js:
        * inspector/InspectorResource.cpp:
        (WebCore::InspectorResource::buildObjectForTiming):
        * inspector/front-end/ResourcesPanel.js:
        (WebInspector.ResourcesPanel.prototype._showPopover):

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@63554 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2010-07-16  Yury Semikhatsky  <yurys@chromium.org>

        Unreviewed. Chromium build fix. Add missing include.

        * inspector/ConsoleMessage.h:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@63553 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2010-07-16  Andreas Kling  <andreas.kling@nokia.com>

        Reviewed by Antonio Gomes.

        [Qt] Path: Avoid creating a new GraphicsContext in strokeContains() and strokeBoundingRect()
        https://bugs.webkit.org/show_bug.cgi?id=42456

        * platform/graphics/qt/PathQt.cpp:
        (WebCore::scratchContext): Added, provides a scratch GraphicsContext.
        (WebCore::Path::strokeContains): Use the scratch context instead
        of creating a new GraphicsContext.
        (WebCore::Path::strokeBoundingRect): Ditto.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@63552 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2010-07-16  Yury Semikhatsky  <yurys@chromium.org>

        Unreviewed. Bump up Chromium revision.

        * DEPS:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@63551 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2010-07-16  Pavel Podivilov  <podivilov@chromium.org>

        Reviewed by Yury Semikhatsky.

        [V8] V8Proxy::retrieve may return null if javascript is disabled. Add a check
        to avoid crashes in inspected page.
        https://bugs.webkit.org/show_bug.cgi?id=42065

        * bindings/v8/ScriptDebugServer.cpp:
        (WebCore::ScriptDebugServer::addListener):

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@63550 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2010-07-16  Pavel Feldman  <pfeldman@chromium.org>

        Reviewed by Yury Semikhatsky.

        Web Inspector: disable AppCache in chromium.

        https://bugs.webkit.org/show_bug.cgi?id=41858

        * inspector/front-end/Settings.js:
        * inspector/front-end/StoragePanel.js:
        (WebInspector.StoragePanel):
        (WebInspector.StoragePanel.prototype.reset):
        (WebInspector.StoragePanel.prototype.addApplicationCache):2010-07-16  Pavel Feldman  <pfeldman@chromium.org>

        Reviewed by Yury Semikhatsky.

        Web Inspector: disable AppCache in chromium.

        https://bugs.webkit.org/show_bug.cgi?id=41858

        * src/js/DevTools.js:
        (WebInspector.loaded):

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@63549 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2010-07-16  Yury Semikhatsky  <yurys@chromium.org>

        Reviewed by Pavel Feldman.

        console.trace should show file and line number for each function in the stack
        https://bugs.webkit.org/show_bug.cgi?id=21180

        Test: inspector/console-trace.html

        * bindings/js/ScriptCallStack.cpp:
        (WebCore::ScriptCallStack::initialize):
        * bindings/v8/ScriptCallFrame.cpp:
        (WebCore::ScriptCallFrame::ScriptCallFrame):
        * bindings/v8/ScriptCallFrame.h:
        * bindings/v8/ScriptCallStack.cpp:
        (WebCore::getFrameLocation):
        (WebCore::toScriptCallFrame):
        (WebCore::ScriptCallStack::create):
        (WebCore::ScriptCallStack::ScriptCallStack):
        (WebCore::ScriptCallStack::at):
        (WebCore::ScriptCallStack::size):
        * bindings/v8/ScriptCallStack.h:
        * bindings/v8/custom/V8ConsoleCustom.cpp:
        (WebCore::V8Console::traceCallback):
        * inspector/ConsoleMessage.cpp:
        (WebCore::ConsoleMessage::CallFrame::CallFrame):
        (WebCore::ConsoleMessage::CallFrame::isEqual):
        (WebCore::ConsoleMessage::CallFrame::createFrontendObject):
        (WebCore::ConsoleMessage::ConsoleMessage):
        (WebCore::ConsoleMessage::addToFrontend):
        (WebCore::ConsoleMessage::isEqual):
        * inspector/ConsoleMessage.h:
        * inspector/front-end/ConsoleView.js:
        (WebInspector.ConsoleMessage.prototype._formatMessage):
        (WebInspector.ConsoleMessage.prototype._createStackTraceElement):
        (WebInspector.ConsoleMessage.prototype._createSourceUrlLink):
        * inspector/front-end/inspector.css:
        (.console-message.expandable > .console-message-text::before):
        (.console-message.expandable.collapsed > .console-message-text::before):
        (.console-message.expandable.collapsed > ol.stack-trace):
        (.console-message > ol.stack-trace):
        (.console-message.repeated-message > ol.stack-trace):
        (.console-message.repeated-message > ol.stack-trace.trace-message):
        * page/Console.idl:

2010-07-16  Yury Semikhatsky  <yurys@chromium.org>

        Reviewed by Pavel Feldman.

        console.trace should show file and line number for each function in the stack
        https://bugs.webkit.org/show_bug.cgi?id=21180

        * http/tests/inspector/console-tests.js:
        (frontend_dumpConsoleMessages):
        * inspector/console-trace-expected.txt: Added.
        * inspector/console-trace.html: Added.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@63548 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2010-07-16  Alexander Pavlov  <apavlov@chromium.org>

        Reviewed by Yury Semikhatsky.

        Web Inspector: [Chromium] A non-existent resource is referenced from devTools.css

        The erroneous rule has been removed.
        https://bugs.webkit.org/show_bug.cgi?id=42454

        * src/js/devTools.css:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@63547 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2010-07-16  Lucas De Marchi  <lucas.demarchi@profusion.mobi>

        Reviewed by Antonio Gomes.

        [EFL] Build with MathML enabled. Fix build when MathML is enabled and
        make it the default.
        https://bugs.webkit.org/show_bug.cgi?id=42453

        * cmake/OptionsEfl.cmake: build with MathML enabled as default.
2010-07-16  Lucas De Marchi  <lucas.demarchi@profusion.mobi>

        Reviewed by Antonio Gomes.

        [EFL] Build with MathML enabled. Fix build when MathML is enabled and
        make it the default.
        https://bugs.webkit.org/show_bug.cgi?id=42453

        EFL port does not support automated tests, yet.

        * CMakeLists.txt: add missing source file and sort.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@63546 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2010-07-16  Andrey Kosyakov  <caseq@chromium.org>

        Reviewed by Pavel Feldman.

        Check that end node of a range has a valid renderer in WebViewImpl::caretOrSelectionBounds()
        to avoid crash in Position::getInlineBoxAndOffset() when start node has renderer but end one doesn't.
        https://bugs.webkit.org/show_bug.cgi?id=42449

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

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@63545 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2010-07-16  Hans Wennborg  <hans@chromium.org>

        Reviewed by Steve Block.

        DeviceOrientationEvent.h should not forward-declare DeviceOrientation
        https://bugs.webkit.org/show_bug.cgi?id=42447

        When destructing m_orientation, DeviceOrientation cannot be an incomplete type.

        * dom/DeviceOrientationEvent.h:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@63544 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2010-07-16  Pavel Podivilov  <podivilov@chromium.org>

        Reviewed by Yury Semikhatsky.

        Remove wrong test expectations for clip-path-child-changes.svg from chromium-mac
        and mark it as flaky on mac.
        https://bugs.webkit.org/show_bug.cgi?id=42445

        * platform/chromium-mac/svg/custom/clip-path-child-changes-expected.txt: Removed.
        * platform/chromium/test_expectations.txt:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@63543 268f45cc-cd09-0410-ab3c-d52691b4dbfc