WebKit-https.git
9 years ago<http://webkit.org/b/56381> Objective-C classes should be typedef-ed as structs ...
ddkilzer@apple.com [Tue, 15 Mar 2011 18:22:50 +0000 (18:22 +0000)]
<webkit.org/b/56381> Objective-C classes should be typedef-ed as structs (not void*) in C++

Reviewed by Simon Fraser.

Typedef-ing Objective-C classes as void* for pure C++ makes it
easier for bugs to creep in because compilers can't do any type
checking for void pointers.

* platform/graphics/GraphicsContext3D.h: Changed typedef
declarations for CALayer and WebGLLayer from void* to structs.
(WebCore::GraphicsContext3D::platformLayer): Changed
static_cast<CALayer*> to reinterpret_cast<CALayer*> now that
CALayer and WebGLLayer are not void pointers.
* platform/graphics/GraphicsLayer.h: Changed typedef declaration
for PlatformLayer from void* to struct CALayer.
* platform/graphics/ca/PlatformCAAnimation.h: Changed typedef
declaration for CAPropertyAnimation from void* to a struct.
Extracted typdef for PlatformAnimationRef.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@81156 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2011-03-15 Ilya Sherman <isherman@chromium.org>
commit-queue@webkit.org [Tue, 15 Mar 2011 18:02:55 +0000 (18:02 +0000)]
2011-03-15  Ilya Sherman  <isherman@chromium.org>

        Reviewed by Tony Chang.

        Autofilled form elements are assigned fixed background color but not text color
        https://bugs.webkit.org/show_bug.cgi?id=48382

        * fast/forms/input-autofilled-expected.txt: Added.
        * fast/forms/input-autofilled.html: Added.
        * platform/win/fast/forms/input-autofilled-expected.txt: Added.
2011-03-15  Ilya Sherman  <isherman@chromium.org>

        Reviewed by Tony Chang.

        Autofilled form elements are assigned fixed background color but not text color
        https://bugs.webkit.org/show_bug.cgi?id=48382

        Test: fast/forms/input-autofilled.html

        * css/html.css:
        (input:-webkit-autofill): Added foreground color: #000000
        * css/wml.css:
        (input:-webkit-autofill): Added foreground color: #000000
2011-03-15  Ilya Sherman  <isherman@chromium.org>

        Reviewed by Tony Chang.

        Autofilled form elements are assigned fixed background color but not text color
        https://bugs.webkit.org/show_bug.cgi?id=48382

        * WebCoreSupport/DumpRenderTreeSupportGtk.cpp:
        (DumpRenderTreeSupportGtk::setAutofilled): Added.
        * WebCoreSupport/DumpRenderTreeSupportGtk.h:
2011-03-15  Ilya Sherman  <isherman@chromium.org>

        Reviewed by Tony Chang.

        Autofilled form elements are assigned fixed background color but not text color
        https://bugs.webkit.org/show_bug.cgi?id=48382

        * DOM/WebDOMOperations.mm:
        (-[DOMHTMLInputElement _setAutofilled:]): Added (wrapper).
        * DOM/WebDOMOperationsPrivate.h:
2011-03-15  Ilya Sherman  <isherman@chromium.org>

        Reviewed by Tony Chang.

        Autofilled form elements are assigned fixed background color but not text color
        https://bugs.webkit.org/show_bug.cgi?id=48382

        * DumpRenderTree/LayoutTestController.cpp:
        (setAutofilledCallback): Added.
        (LayoutTestController::staticFunctions):
        * DumpRenderTree/LayoutTestController.h:
        * DumpRenderTree/chromium/LayoutTestController.cpp:
        (LayoutTestController::LayoutTestController):
        (LayoutTestController::setAutofilled): Added.
        * DumpRenderTree/chromium/LayoutTestController.h:
        * DumpRenderTree/gtk/LayoutTestControllerGtk.cpp:
        (LayoutTestController::setAutofilled): Added.
        * DumpRenderTree/mac/LayoutTestControllerMac.mm:
        (LayoutTestController::setAutofilled): Added.
        (LayoutTestController::setValueForUser): Style tweak.
        * DumpRenderTree/win/LayoutTestControllerWin.cpp:
        (LayoutTestController::setAutofilled): Added (stub).
        * DumpRenderTree/wx/LayoutTestControllerWx.cpp:
        (LayoutTestController::setAutofilled): Added (stub).

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@81155 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoFix broken assert in new code.
barraclough@apple.com [Tue, 15 Mar 2011 18:02:13 +0000 (18:02 +0000)]
Fix broken assert in new code.

Reviewed by Sam Weinig.

* dfg/DFGAliasTracker.h:
(JSC::DFG::AliasTracker::recordPutByVal):
    - recordPutByVal is called for both PutByVal & PutByValAlias.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@81154 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoRemoved redundant code from BytecodeGenerator.
barraclough@apple.com [Tue, 15 Mar 2011 17:57:31 +0000 (17:57 +0000)]
Removed redundant code from BytecodeGenerator.

Rubber stamped by Sam Weinig.

* bytecompiler/BytecodeGenerator.cpp:
* bytecompiler/BytecodeGenerator.h:
    - delete uncalled code missed when reparsing was removed.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@81153 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2011-03-15 Tony Chang <tony@chromium.org>
tony@chromium.org [Tue, 15 Mar 2011 17:56:48 +0000 (17:56 +0000)]
2011-03-15  Tony Chang  <tony@chromium.org>

        Revert r81147 and r81149, broke the chromium win build.

        * WebKit.gyp:
        * scripts/generate_devtools_grd.py: Removed.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@81152 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2011-03-15 Adam Roben <aroben@apple.com>
aroben@apple.com [Tue, 15 Mar 2011 17:51:06 +0000 (17:51 +0000)]
2011-03-15  Adam Roben  <aroben@apple.com>

        Fix a leak in WebStorageTrackerClient::dispatchDidModifyOrigin

        Fixes <http://webkit.org/b/56385> SecurityOrigin leaks seen in
        WebStorageTrackerClient::dispatchDidModifyOrigin

        Reviewed by Antti Koivisto.

        * Storage/WebStorageTrackerClient.mm:
        (WebStorageTrackerClient::dispatchDidModifyOriginOnMainThread): Adopt the reference passed
        to us by dispatchDidModifyOrigin.
        (WebStorageTrackerClient::dispatchDidModifyOrigin): Added a comment and removed an
        unnecessary cast.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@81150 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2011-03-15 Tony Chang <tony@chromium.org>
tony@chromium.org [Tue, 15 Mar 2011 17:49:50 +0000 (17:49 +0000)]
2011-03-15  Tony Chang  <tony@chromium.org>

        Fix chromium build using python 2.5 (with statement).

        * scripts/generate_devtools_grd.py:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@81149 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2011-03-15 Tony Chang <tony@chromium.org>
tony@chromium.org [Tue, 15 Mar 2011 17:40:44 +0000 (17:40 +0000)]
2011-03-15  Tony Chang  <tony@chromium.org>

        Reviewed by Adam Barth.

        [chromium] add a build target to create a grd file with devtool resources
        https://bugs.webkit.org/show_bug.cgi?id=56335

        This is the webkit side of the fix for
        http://code.google.com/p/chromium/issues/detail?id=35793

        * WebKit.gyp:
        * scripts/generate_devtools_grd.py: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@81147 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoUpdate WebKit2 Skipped file for some new tests that rely on unimplemented features
aroben@apple.com [Tue, 15 Mar 2011 17:35:36 +0000 (17:35 +0000)]
Update WebKit2 Skipped file for some new tests that rely on unimplemented features

* platform/mac-wk2/Skipped:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@81146 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2011-03-15 Alexis Menard <alexis.menard@openbossa.org>
commit-queue@webkit.org [Tue, 15 Mar 2011 17:32:24 +0000 (17:32 +0000)]
2011-03-15  Alexis Menard  <alexis.menard@openbossa.org>

        Reviewed by Eric Carlson.

        HTMLMediaElement::mediaPlayerPlaybackStateChanged should not change the "public" state of the element
        if it's an internal pause for example.
        https://bugs.webkit.org/show_bug.cgi?id=56374

        In case of an internal pause, the callback from the mediaplayer should be ignored to avoid reflecting the
        change into the DOM.

        No new tests: Verified manually.

        * html/HTMLMediaElement.cpp:
        (WebCore::HTMLMediaElement::mediaPlayerPlaybackStateChanged):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@81145 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2011-03-15 Bill Budge <bbudge@chromium.org>
commit-queue@webkit.org [Tue, 15 Mar 2011 17:24:32 +0000 (17:24 +0000)]
2011-03-15  Bill Budge  <bbudge@chromium.org>

        Reviewed by David Levin.

        AssociatedURLLoader does not support Cross Origin Requests
        https://bugs.webkit.org/show_bug.cgi?id=53925

        No new tests. No new functionality exposed.

        * public/WebURLLoader.h:
        (WebKit::WebURLLoaderOptions::WebURLLoaderOptions):
        * src/AssociatedURLLoader.cpp:
        (WebKit::AssociatedURLLoader::ClientAdapter::clearClient):
        (WebKit::AssociatedURLLoader::ClientAdapter::create):
        (WebKit::AssociatedURLLoader::ClientAdapter::ClientAdapter):
        (WebKit::AssociatedURLLoader::ClientAdapter::willSendRequest):
        (WebKit::AssociatedURLLoader::ClientAdapter::didSendData):
        (WebKit::AssociatedURLLoader::ClientAdapter::didReceiveResponse):
        (WebKit::AssociatedURLLoader::ClientAdapter::didReceiveData):
        (WebKit::AssociatedURLLoader::ClientAdapter::didReceiveCachedMetadata):
        (WebKit::AssociatedURLLoader::ClientAdapter::didFinishLoading):
        (WebKit::AssociatedURLLoader::ClientAdapter::didFail):
        (WebKit::AssociatedURLLoader::AssociatedURLLoader):
        (WebKit::AssociatedURLLoader::~AssociatedURLLoader):
        (WebKit::AssociatedURLLoader::loadSynchronously):
        (WebKit::AssociatedURLLoader::loadAsynchronously):
        (WebKit::AssociatedURLLoader::cancel):
        (WebKit::AssociatedURLLoader::setDefersLoading):
        * src/AssociatedURLLoader.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@81144 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2011-03-15 Alok priyadarshi <alokp@chromium.org>
commit-queue@webkit.org [Tue, 15 Mar 2011 17:14:22 +0000 (17:14 +0000)]
2011-03-15  Alok priyadarshi  <alokp@chromium.org>

        Reviewed by James Robinson.

        Remove unnecessary function WebPlugin::getBackingTextureId
        https://bugs.webkit.org/show_bug.cgi?id=56122

        It will be tested by Pepper2D and Pepper3D layout/smoke tests
        on the Chromium side.

        * public/WebPlugin.h:
        * src/WebPluginContainerImpl.cpp:
        (WebKit::WebPluginContainerImpl::platformLayer):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@81143 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoBuild fix after r81135.
mitz@apple.com [Tue, 15 Mar 2011 16:51:43 +0000 (16:51 +0000)]
Build fix after r81135.

* TestWebKitAPI/Configurations/InjectedBundle.xcconfig:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@81142 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2011-03-15 Mikhail Naganov <mnaganov@chromium.org>
mnaganov@chromium.org [Tue, 15 Mar 2011 16:33:49 +0000 (16:33 +0000)]
2011-03-15  Mikhail Naganov  <mnaganov@chromium.org>

        Unreviewed. Update chromium deps to pick up the latest V8.

        * DEPS:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@81141 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoBuild fix after r81135.
mitz@apple.com [Tue, 15 Mar 2011 16:30:51 +0000 (16:30 +0000)]
Build fix after r81135.

* WebKitTestRunner/Configurations/InjectedBundle.xcconfig:
* WebKitTestRunner/mac/TestControllerMac.mm:
* WebKitTestRunner/mac/TestInvocationMac.mm:
* WebKitTestRunner/mac/main.mm:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@81140 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2011-03-15 Balazs Kelemen <kbalazs@webkit.org>
kbalazs@webkit.org [Tue, 15 Mar 2011 16:01:22 +0000 (16:01 +0000)]
2011-03-15  Balazs Kelemen  <kbalazs@webkit.org>

        Reviewed by Adam Barth.

        REGRESSION(80977): breaks Qt ARMv7 Linux Release bot
        https://bugs.webkit.org/show_bug.cgi?id=56349

        * QtTestBrowser/QtTestBrowser.pro: Link against fontconfig
        on platforms where we use it. Copied from DumpRenderTree.pro.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@81139 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoBuild fix after r81135.
mitz@apple.com [Tue, 15 Mar 2011 15:51:44 +0000 (15:51 +0000)]
Build fix after r81135.

* DumpRenderTree/cf/WebArchiveDumpSupport.cpp:
* DumpRenderTree/mac/WebArchiveDumpSupportMac.mm:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@81138 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2011-03-15 Martin Robinson <mrobinson@igalia.com>
mrobinson@webkit.org [Tue, 15 Mar 2011 15:44:32 +0000 (15:44 +0000)]
2011-03-15  Martin Robinson  <mrobinson@igalia.com>

        Reviewed by Adam Barth.

        [GTK] [WebKit2] The UIProcess never changes the mouse cursor
        https://bugs.webkit.org/show_bug.cgi?id=56333

        Add an implementation for WebView::addCursor for WebKit2 GTK+. This allows
        the cursor to change as the user mouses around the page. There is currently
        no test infrastructure to track cursor changes.

        * UIProcess/gtk/WebView.cpp:
        (WebKit::WebView::setCursor): Ported implementation from WebKit1.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@81137 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2011-03-12 Pavel Podivilov <podivilov@chromium.org>
podivilov@chromium.org [Tue, 15 Mar 2011 15:12:14 +0000 (15:12 +0000)]
2011-03-12  Pavel Podivilov  <podivilov@chromium.org>

        Reviewed by Yury Semikhatsky.

        Web Inspector: re-implement xhr breakpoints.
        https://bugs.webkit.org/show_bug.cgi?id=56252

        * inspector/debugger/xhr-breakpoints-expected.txt: Added.
        * inspector/debugger/xhr-breakpoints.html: Added.
        * platform/gtk/Skipped:
        * platform/mac/Skipped:
        * platform/qt/Skipped:
        * platform/win/Skipped:
2011-03-12  Pavel Podivilov  <podivilov@chromium.org>

        Reviewed by Yury Semikhatsky.

        Web Inspector: re-implement xhr breakpoints.
        https://bugs.webkit.org/show_bug.cgi?id=56252

        - restore xhr breakpoints one by one instead of using setAllBrowserBreakpoints
        - store xhr breakpoints in a separate setting
        - move presentation-related code from BreakpointManager to XHRBreakpointsSidebarPane

        Test: inspector/debugger/xhr-breakpoints.html

        * inspector/InspectorBrowserDebuggerAgent.cpp:
        (WebCore::InspectorBrowserDebuggerAgent::InspectorBrowserDebuggerAgent):
        (WebCore::InspectorBrowserDebuggerAgent::inspectedURLChanged):
        (WebCore::InspectorBrowserDebuggerAgent::restoreStickyBreakpoint):
        (WebCore::InspectorBrowserDebuggerAgent::setXHRBreakpoint):
        (WebCore::InspectorBrowserDebuggerAgent::removeXHRBreakpoint):
        (WebCore::InspectorBrowserDebuggerAgent::willSendXMLHttpRequest):
        (WebCore::InspectorBrowserDebuggerAgent::clear):
        * inspector/InspectorBrowserDebuggerAgent.h:
        * inspector/front-end/BreakpointManager.js:
        (WebInspector.BreakpointManager.prototype.setXHRBreakpoint):
        (WebInspector.BreakpointManager.prototype.removeXHRBreakpoint):
        (WebInspector.BreakpointManager.prototype.breakpointViewForEventData):
        (WebInspector.BreakpointManager.prototype._projectChanged):
        (WebInspector.BreakpointManager.prototype._validateBreakpoints):
        (WebInspector.BreakpointManager.prototype._createEventListenerBreakpointId):
        * inspector/front-end/BreakpointsSidebarPane.js:
        (WebInspector.XHRBreakpointsSidebarPane):
        (WebInspector.XHRBreakpointsSidebarPane.prototype._addButtonClicked.finishEditing):
        (WebInspector.XHRBreakpointsSidebarPane.prototype._addButtonClicked):
        (WebInspector.XHRBreakpointsSidebarPane.prototype._setBreakpoint):
        (WebInspector.XHRBreakpointsSidebarPane.prototype._removeBreakpoint):
        (WebInspector.XHRBreakpointsSidebarPane.prototype._contextMenu.removeBreakpoint):
        (WebInspector.XHRBreakpointsSidebarPane.prototype._contextMenu):
        (WebInspector.XHRBreakpointsSidebarPane.prototype._checkboxClicked):
        (WebInspector.XHRBreakpointsSidebarPane.prototype._labelClicked.finishEditing):
        (WebInspector.XHRBreakpointsSidebarPane.prototype._labelClicked):
        (WebInspector.XHRBreakpointsSidebarPane.prototype.highlightBreakpoint):
        (WebInspector.XHRBreakpointsSidebarPane.prototype.clearBreakpointHighlight):
        (WebInspector.XHRBreakpointsSidebarPane.prototype._saveBreakpoints):
        (WebInspector.XHRBreakpointsSidebarPane.prototype._restoreBreakpoints):
        (WebInspector.XHRBreakpointsSidebarPane.prototype._projectChanged):
        * inspector/front-end/CallStackSidebarPane.js:
        (WebInspector.CallStackSidebarPane.prototype.update):
        (WebInspector.CallStackSidebarPane.prototype._xhrBreakpointHit):
        * inspector/front-end/ScriptsPanel.js:
        (WebInspector.ScriptsPanel):
        (WebInspector.ScriptsPanel.prototype._debuggerPaused):
        (WebInspector.ScriptsPanel.prototype._clearInterface):
        * inspector/front-end/Settings.js:
        (WebInspector.Settings):
        * inspector/front-end/inspector.js:
        (WebInspector.resetFocusElement):
        (WebInspector.set attached):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@81136 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoReviewed by Darin Adler.
kevino@webkit.org [Tue, 15 Mar 2011 14:59:20 +0000 (14:59 +0000)]
Reviewed by Darin Adler.

Introduce WTF_USE_EXPORT_MACROS, which will allow us to put shared library import/export
info into the headers rather than in export symbol definition files, but disable it on
all platforms initially so we can deal with port build issues one port at a time.

https://bugs.webkit.org/show_bug.cgi?id=27551

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@81135 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2011-03-15 Ilya Tikhonovsky <loislo@chromium.org>
loislo@chromium.org [Tue, 15 Mar 2011 14:41:54 +0000 (14:41 +0000)]
2011-03-15  Ilya Tikhonovsky  <loislo@chromium.org>

        Unreviewed build fix.

        Chromium: shared lib linux build are failing.

        Two exclude rules for LocalizedNumberNone and TextEncodingDetectorNone were added to the wrong library.
        It was webcore_remaining instead of webcore_platform.

        * WebCore.gyp/WebCore.gyp:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@81134 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2011-03-15 Yury Semikhatsky <yurys@chromium.org>
yurys@chromium.org [Tue, 15 Mar 2011 14:38:53 +0000 (14:38 +0000)]
2011-03-15  Yury Semikhatsky  <yurys@chromium.org>

        Reviewed by Pavel Feldman.

        Web Inspector: expanding/collapsing object shouldn&apos;t affect outer console.group expansion state
        https://bugs.webkit.org/show_bug.cgi?id=56373

        * inspector/front-end/Section.js:
        (WebInspector.Section):
        (WebInspector.Section.prototype.toggleExpanded):
        (WebInspector.Section.prototype.handleClick): stop click even propagation if it was handled by this section.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@81133 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2011-03-15 Pavel Feldman <pfeldman@chromium.org>
pfeldman@chromium.org [Tue, 15 Mar 2011 14:13:03 +0000 (14:13 +0000)]
2011-03-15  Pavel Feldman  <pfeldman@chromium.org>

        Reviewed by Yury Semikhatsky.

        Web Inspector: crash upon "//html//@id" search in elements panel.
        https://bugs.webkit.org/show_bug.cgi?id=56334

        * inspector/elements/elements-panel-search-expected.txt:
        * inspector/elements/elements-panel-search.html:
2011-03-15  Pavel Feldman  <pfeldman@chromium.org>

        Reviewed by Yury Semikhatsky.

        Web Inspector: crash upon "//html//@id" search in elements panel.
        https://bugs.webkit.org/show_bug.cgi?id=56334

        * inspector/InspectorDOMAgent.cpp:
        * inspector/front-end/ElementsTreeOutline.js:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@81132 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2011-03-15 Andrey Kosyakov <caseq@chromium.org>
caseq@chromium.org [Tue, 15 Mar 2011 14:02:56 +0000 (14:02 +0000)]
2011-03-15  Andrey Kosyakov  <caseq@chromium.org>

        Unreviewed. Swap two calls to setMonitoringXHREnabled to get rid of side effects on other tests.

        * inspector/protocol/console-agent-expected.txt:
        * inspector/protocol/console-agent.html:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@81131 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2011-03-15 Andrey Kosyakov <caseq@chromium.org>
caseq@chromium.org [Tue, 15 Mar 2011 13:47:37 +0000 (13:47 +0000)]
2011-03-15  Andrey Kosyakov  <caseq@chromium.org>

        Reviewed by Yury Semikhatsky.

        Web Inspector: further extension API cleanup (removed inspectedPage, add experimental prefix)
        https://bugs.webkit.org/show_bug.cgi?id=56327

        * inspector/front-end/ExtensionAPI.js:
        (WebInspector.injectedExtensionAPI):
        * inspector/front-end/ExtensionAPISchema.json:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@81130 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2011-03-15 Yury Semikhatsky <yurys@chromium.org>
yurys@chromium.org [Tue, 15 Mar 2011 13:08:54 +0000 (13:08 +0000)]
2011-03-15  Yury Semikhatsky  <yurys@chromium.org>

        Reviewed by Pavel Feldman.

        Web Inspector: when console.groupEnd calls twice
        https://bugs.webkit.org/show_bug.cgi?id=56114

        * inspector/console/console-nested-group-expected.txt: Added.
        * inspector/console/console-nested-group.html: Added.
2011-03-15  Yury Semikhatsky  <yurys@chromium.org>

        Reviewed by Pavel Feldman.

        Web Inspector: when console.groupEnd calls twice
        https://bugs.webkit.org/show_bug.cgi?id=56114

        Test: inspector/console/console-nested-group.html

        * inspector/ConsoleMessage.h:
        (WebCore::ConsoleMessage::type):
        * inspector/InspectorConsoleAgent.cpp:
        (WebCore::InspectorConsoleAgent::addConsoleMessage): do not coalesce adjacent EndGroup messages.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@81129 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2011-03-15 Chris Mumford <chris.mumford@palm.com>
commit-queue@webkit.org [Tue, 15 Mar 2011 10:47:29 +0000 (10:47 +0000)]
2011-03-15  Chris Mumford  <chris.mumford@palm.com>

        Reviewed by Adam Barth.

        Initializing several member variables that were not initialized in
        their constructors. These values were all read prior to initialization
        as reported by Valgrind.

        No new tests: No feature additions/removals.

        * accessibility/AccessibilityImageMapLink.cpp:
        (WebCore::AccessibilityImageMapLink::AccessibilityImageMapLink):
        * editing/DeleteSelectionCommand.cpp:
        (WebCore::DeleteSelectionCommand::DeleteSelectionCommand):
        * html/HTMLCanvasElement.cpp:
        (WebCore::HTMLCanvasElement::HTMLCanvasElement):
        * xml/XPathResult.cpp:
        (WebCore::XPathResult::XPathResult):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@81128 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2011-03-15 Hayato Ito <hayato@chromium.org>
hayato@chromium.org [Tue, 15 Mar 2011 10:46:21 +0000 (10:46 +0000)]
2011-03-15  Hayato Ito  <hayato@chromium.org>

        Reviewed by Shinichiro Hamaji.

        [NRWT] Add support for reftests to new-run-webkit-tests.

        https://bugs.webkit.org/show_bug.cgi?id=55457

        * Scripts/webkitpy/layout_tests/layout_package/single_test_runner.py:
        * Scripts/webkitpy/layout_tests/layout_package/test_failures.py:
        * Scripts/webkitpy/layout_tests/layout_package/test_result_writer.py:
        * Scripts/webkitpy/layout_tests/port/base.py:
        * Scripts/webkitpy/layout_tests/port/dryrun.py:
        * Scripts/webkitpy/layout_tests/port/test.py:
        * Scripts/webkitpy/layout_tests/port/test_files.py:
        * Scripts/webkitpy/layout_tests/run_webkit_tests_unittest.py:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@81127 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2011-03-15 Ben Taylor <bentaylor.solx86@gmail.com>
commit-queue@webkit.org [Tue, 15 Mar 2011 10:33:47 +0000 (10:33 +0000)]
2011-03-15  Ben Taylor  <bentaylor.solx86@gmail.com>

        Reviewed by Adam Barth.

        https://bugs.webkit.org/show_bug.cgi?id=56255
        Fix build on Solaris 10/Sun Studio 12 C++

        No new tests. This is to fix compilation on Solaris 10 with Sun Studio 12 C++

        * bridge/runtime_array.h:
        (JSC::RuntimeArray::getConcreteArray):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@81126 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2011-03-15 Philippe Normand <pnormand@igalia.com>
philn@webkit.org [Tue, 15 Mar 2011 10:32:13 +0000 (10:32 +0000)]
2011-03-15  Philippe Normand  <pnormand@igalia.com>

        Unreviewed, skip failing media test on GTK.

        * platform/gtk/Skipped:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@81125 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2011-03-15 Alexander Pavlov <apavlov@chromium.org>
apavlov@chromium.org [Tue, 15 Mar 2011 10:19:45 +0000 (10:19 +0000)]
2011-03-15  Alexander Pavlov  <apavlov@chromium.org>

        [Chromium] Unreviewed, update test expectations

        * platform/chromium/test_expectations.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@81124 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2011-03-15 Alejandro G. Castro <alex@igalia.com>
alex@webkit.org [Tue, 15 Mar 2011 10:10:53 +0000 (10:10 +0000)]
2011-03-15  Alejandro G. Castro  <alex@igalia.com>

        Reviewed by Martin Robinson.

        [GTK] Fix compilation warning after r80892
        https://bugs.webkit.org/show_bug.cgi?id=56302

        * DumpRenderTree/gtk/LayoutTestControllerGtk.cpp:
        (LayoutTestController::originsWithLocalStorage):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@81123 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2011-03-15 Robert Hogan <robert@webkit.org>
robert@webkit.org [Tue, 15 Mar 2011 09:41:22 +0000 (09:41 +0000)]
2011-03-15  Robert Hogan  <robert@webkit.org>

        Reviewed by Adam Barth.

        Document the qmakearg argument, and hint how to build webkit2 on Qt
        https://bugs.webkit.org/show_bug.cgi?id=56326

        * Scripts/build-webkit: add --qmakearg to build-webkit --help

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@81122 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2011-03-15 Ilya Sherman <isherman@chromium.org>
commit-queue@webkit.org [Tue, 15 Mar 2011 09:09:31 +0000 (09:09 +0000)]
2011-03-15  Ilya Sherman  <isherman@chromium.org>

        Reviewed by David Levin.

        Update Chromium test expectations for fast/forms/onchange-setvalueforuser
        https://bugs.webkit.org/show_bug.cgi?id=56365

        * platform/chromium/fast/forms/onchange-setvalueforuser-expected.txt: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@81121 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoMake Windows 7 Release (Tests) a core builder
aroben@apple.com [Tue, 15 Mar 2011 07:28:19 +0000 (07:28 +0000)]
Make Windows 7 Release (Tests) a core builder

We can probably make Windows XP Debug (Tests) a core builder, too, but we should probably
give it a few more days of greenness first.

Fixes <http://webkit.org/b/55665> Windows 7 Release (Tests) should be a core builder

Reviewed by Sam Weinig.

* Scripts/webkitpy/common/net/buildbot/buildbot.py:
(BuildBot): Added a regexp to match Windows 7 Release (Tests).

* Scripts/webkitpy/common/net/buildbot/buildbot_unittest.py:
(BuildBotTest): Updated the list of builders to match the current list on build.webkit.org.
Updated the core regexps to match the buildbot module. Updated the expected results to match
the current set of core builders.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@81120 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2011-03-14 Ryosuke Niwa <rniwa@webkit.org>
rniwa@webkit.org [Tue, 15 Mar 2011 06:58:45 +0000 (06:58 +0000)]
2011-03-14  Ryosuke Niwa  <rniwa@webkit.org>

        Another Chromium rebaseline for r81095.

        * platform/chromium-win/editing/pasteboard/smart-paste-008-expected.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@81119 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2011-03-14 Laszlo Gombos <laszlo.1.gombos@nokia.com>
laszlo.1.gombos@nokia.com [Tue, 15 Mar 2011 06:50:36 +0000 (06:50 +0000)]
2011-03-14  Laszlo Gombos  <laszlo.1.gombos@nokia.com>

        Unreviewed build fix.

        Buildfix when JIT is not enabled after r81079
        https://bugs.webkit.org/show_bug.cgi?id=56361

        * runtime/Executable.cpp:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@81118 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoAdd forgotten file from r81116
aroben@apple.com [Tue, 15 Mar 2011 06:34:16 +0000 (06:34 +0000)]
Add forgotten file from r81116

* TestWebKitAPI/Configurations/TestWebKitAPICFNetwork.vsprops: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@81117 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoabout:blank fake responses don't get serialized when sent the UIProcess
weinig@apple.com [Tue, 15 Mar 2011 06:21:41 +0000 (06:21 +0000)]
about:blank fake responses don't get serialized when sent the UIProcess
<rdar://problem/9108119>
https://bugs.webkit.org/show_bug.cgi?id=56357

Source/WebCore:

Reviewed by Adam Roben

Test: AboutBlankLoad

* platform/network/cf/ResourceResponse.h:
* platform/network/cf/ResourceResponseCFNet.cpp:
(WebCore::ResourceResponse::cfURLResponse):
Create a CFURLResponseRef if one does not exist yet as we do for
NSURLResponses on the mac.

Tools:

Reviewed by Adam Roben.

* TestWebKitAPI/Tests/WebKit2/AboutBlankLoad.cpp: Added.
(TestWebKitAPI::decidePolicyForResponse):
(TestWebKitAPI::TEST):
Add test that loads of about:blank have a response with a MIMEType of text/html.

* TestWebKitAPI/PlatformUtilities.h:
* TestWebKitAPI/mac/PlatformUtilitiesMac.mm:
(TestWebKitAPI::Util::MIMETypeForWKURLResponse):
* TestWebKitAPI/win/PlatformUtilitiesWin.cpp:
(TestWebKitAPI::Util::MIMETypeForWKURLResponse):
Add helper to get the MIMEType from a WKURLResponse.

* TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj:
* TestWebKitAPI/win/TestWebKitAPI.vcproj:
Add new files.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@81116 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2011-03-14 Alexey Proskuryakov <ap@apple.com>
ap@apple.com [Tue, 15 Mar 2011 06:11:59 +0000 (06:11 +0000)]
2011-03-14  Alexey Proskuryakov  <ap@apple.com>

        Reviewed by Darin Adler.

        REGRESSION: Print preview is blank when selecting multiple pages per sheet
        https://bugs.webkit.org/show_bug.cgi?id=56341
        <rdar://problem/8991382>

        * UIProcess/API/mac/WKPrintingView.mm: (pageDidDrawToPDF): Don't reset latest expected
        callback number after receiving a different one. We only update preview after receiving
        the latest expected page data (and the assumption is that AppKit will ask for pages in
        natural order for N-up).

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@81115 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2011-03-14 Alexis Menard <alexis.menard@openbossa.org>
commit-queue@webkit.org [Tue, 15 Mar 2011 05:59:15 +0000 (05:59 +0000)]
2011-03-14  Alexis Menard  <alexis.menard@openbossa.org>

        Reviewed by Kenneth Rohde Christiansen.

        [Qt] Seeking videos using the timeline bar does not work properly and stop the video playback.
        https://bugs.webkit.org/show_bug.cgi?id=56145

        We do not need seekTimeout and queuedSeekTimeout anymore. setPosition on QMediaPlayer is good enough.
        positionChanged() will be emitted when the data is buffered. On Linux the signal was not emitted because
        of a bug in QtMultimedia.

        * platform/graphics/qt/MediaPlayerPrivateQt.cpp:
        (WebCore::MediaPlayerPrivateQt::MediaPlayerPrivateQt):
        (WebCore::MediaPlayerPrivateQt::seek):
        (WebCore::MediaPlayerPrivateQt::stateChanged):
        (WebCore::MediaPlayerPrivateQt::positionChanged):
        * platform/graphics/qt/MediaPlayerPrivateQt.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@81114 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2011-03-14 Alexey Marinichev <amarinichev@chromium.org>
commit-queue@webkit.org [Tue, 15 Mar 2011 05:56:32 +0000 (05:56 +0000)]
2011-03-14  Alexey Marinichev  <amarinichev@chromium.org>

        Reviewed by Kenneth Russell.

        Graphics Context is not properly recovered if the GPU process dies.
        https://bugs.webkit.org/show_bug.cgi?id=56148

        Split context recreation into two steps.

        * src/WebViewImpl.cpp:
        (WebKit::WebViewImpl::composite):
        (WebKit::WebViewImpl::doComposite):
        (WebKit::WebViewImpl::reallocateRenderer):
        * src/WebViewImpl.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@81113 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2011-03-14 MORITA Hajime <morrita@google.com>
morrita@google.com [Tue, 15 Mar 2011 05:50:57 +0000 (05:50 +0000)]
2011-03-14  MORITA Hajime  <morrita@google.com>

        Unreviewed, skipped tests that uses an unsupported DRT API.

        * platform/qt/Skipped:
        * platform/win/Skipped:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@81112 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2011-03-14 Alexey Proskuryakov <ap@apple.com>
ap@apple.com [Tue, 15 Mar 2011 05:48:07 +0000 (05:48 +0000)]
2011-03-14  Alexey Proskuryakov  <ap@apple.com>

        Reviewed by Adam Roben.

        https://bugs.webkit.org/show_bug.cgi?id=44138
        Crashes beneath SocketStreamHandle::readStreamCallback when running WebSocket tests.

        * platform/win/Skipped: Unskipped tests that should be fixed now. I couldn't verify that
        locally, becasue most WebSocket tests time out on my machine.
2011-03-14  Alexey Proskuryakov  <ap@apple.com>

        Reviewed by Adam Roben.

        https://bugs.webkit.org/show_bug.cgi?id=44138
        Crash beneath SocketStreamHandle::readStreamCallback when running websocket/tests/workers/worker-handshake-challenge-randomness.html

        https://bugs.webkit.org/show_bug.cgi?id=55375
        http/tests/websocket/tests/reload-crash.html sometimes crashes beneath SocketStreamHandle::readStreamCallback on Windows

        https://bugs.webkit.org/show_bug.cgi?id=56185
        http/tests/websocket/tests/url-with-credential.html sometimes crashes beneath SocketStreamHandle::readStreamCallback on Windows

        * platform/network/cf/SocketStreamHandle.h: Made SocketStreamHandle ThreadSafeShared, so that
        a pointer can be passed across threads when wrapped in a RefPtr.

        * platform/network/cf/SocketStreamHandleCFNet.cpp: Make sure that an object still exists
        when executing a method on main thread by using RefPtr.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@81111 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2011-03-14 Ryosuke Niwa <rniwa@webkit.org>
rniwa@webkit.org [Tue, 15 Mar 2011 05:44:15 +0000 (05:44 +0000)]
2011-03-14  Ryosuke Niwa  <rniwa@webkit.org>

        Chromium rebaselines for r81095.

        * platform/chromium-linux/editing/deleting/smart-delete-001-expected.txt:
        * platform/chromium-linux/editing/deleting/smart-delete-002-expected.txt:
        * platform/chromium-linux/editing/pasteboard/smart-drag-drop-expected.txt:
        * platform/chromium-linux/editing/pasteboard/smart-paste-003-expected.txt:
        * platform/chromium-linux/editing/pasteboard/smart-paste-004-expected.txt:
        * platform/chromium-linux/editing/selection/fake-drag-expected.txt: Removed.
        * platform/chromium-linux/editing/selection/select-from-textfield-outwards-expected.txt:
        * platform/chromium-win/editing/deleting/smart-delete-001-expected.txt:
        * platform/chromium-win/editing/deleting/smart-delete-002-expected.txt:
        * platform/chromium-win/editing/pasteboard/smart-drag-drop-expected.txt:
        * platform/chromium-win/editing/pasteboard/smart-paste-001-expected.txt:
        * platform/chromium-win/editing/pasteboard/smart-paste-003-expected.txt:
        * platform/chromium-win/editing/pasteboard/smart-paste-004-expected.txt:
        * platform/chromium-win/editing/pasteboard/smart-paste-007-expected.txt:
        * platform/chromium-win/editing/selection/after-line-break-expected.txt:
        * platform/chromium-win/editing/selection/expanding-selections-expected.txt:
        * platform/chromium-win/editing/selection/expanding-selections2-expected.txt:
        * platform/chromium-win/editing/selection/fake-drag-expected.txt:
        * platform/chromium-win/editing/selection/select-from-textfield-outwards-expected.txt:
        * platform/chromium-win/editing/selection/triple-click-in-pre-expected.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@81110 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2011-03-14 Ryosuke Niwa <rniwa@webkit.org>
rniwa@webkit.org [Tue, 15 Mar 2011 05:31:29 +0000 (05:31 +0000)]
2011-03-14  Ryosuke Niwa  <rniwa@webkit.org>

        GTK and Qt rebaselines for r81095.

        * platform/gtk/editing/deleting/smart-delete-001-expected.txt:
        * platform/gtk/editing/deleting/smart-delete-002-expected.txt:
        * platform/gtk/editing/pasteboard/smart-paste-001-expected.txt:
        * platform/gtk/editing/pasteboard/smart-paste-003-expected.txt:
        * platform/gtk/editing/pasteboard/smart-paste-004-expected.txt:
        * platform/gtk/editing/pasteboard/smart-paste-007-expected.txt:
        * platform/gtk/editing/pasteboard/smart-paste-008-expected.txt:
        * platform/gtk/editing/selection/expanding-selections-expected.txt:
        * platform/gtk/editing/selection/expanding-selections2-expected.txt:
        * platform/gtk/editing/selection/fake-drag-expected.txt:
        * platform/gtk/editing/selection/select-from-textfield-outwards-expected.txt:
        * platform/gtk/editing/selection/triple-click-in-pre-expected.txt:
        * platform/qt/editing/deleting/smart-delete-001-expected.txt:
        * platform/qt/editing/pasteboard/smart-paste-001-expected.txt:
        * platform/qt/editing/pasteboard/smart-paste-003-expected.txt:
        * platform/qt/editing/pasteboard/smart-paste-004-expected.txt:
        * platform/qt/editing/pasteboard/smart-paste-007-expected.txt:
        * platform/qt/editing/selection/expanding-selections-expected.txt:
        * platform/qt/editing/selection/expanding-selections2-expected.txt:
        * platform/qt/editing/selection/triple-click-in-pre-expected.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@81109 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2011-03-14 Jarkko Sakkinen <jarkko.j.sakkinen@gmail.com>
commit-queue@webkit.org [Tue, 15 Mar 2011 05:24:34 +0000 (05:24 +0000)]
2011-03-14  Jarkko Sakkinen  <jarkko.j.sakkinen@gmail.com>

        Reviewed by Kenneth Rohde Christiansen.

        [Qt] Command-line options for QGLWidget and WebGL to QtTestBrowser
        https://bugs.webkit.org/show_bug.cgi?id=56006

        * QtTestBrowser/launcherwindow.cpp:
        (LauncherWindow::initializeView):
        * QtTestBrowser/main.cpp:
        (LauncherApplication::handleUserOptions):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@81108 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoMac build fix. Part 1 of N.
weinig@apple.com [Tue, 15 Mar 2011 04:58:08 +0000 (04:58 +0000)]
Mac build fix. Part 1 of N.

* WebCore.exp.in:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@81107 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2011-03-14 James Kozianski <koz@chromium.org>
commit-queue@webkit.org [Tue, 15 Mar 2011 04:49:34 +0000 (04:49 +0000)]
2011-03-14  James Kozianski  <koz@chromium.org>

        Reviewed by Ojan Vafai.

        Add method to make BuildBot return test outputs
        https://bugs.webkit.org/show_bug.cgi?id=54374

        The Builder.latest_cached_build will be used by the new rebaseline
        script to determine the build to use baselines from.

        * Scripts/webkitpy/common/net/buildbot/buildbot.py:
        * Scripts/webkitpy/common/net/buildbot/buildbot_unittest.py:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@81106 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoMake WKBundlePageCanHandleRequest return true for empty document URLs
aroben@apple.com [Tue, 15 Mar 2011 04:30:44 +0000 (04:30 +0000)]
Make WKBundlePageCanHandleRequest return true for empty document URLs

Reviewed by Sam Weinig.

Source/WebKit2:

* WebProcess/WebPage/WebPage.cpp:
(WebKit::WebPage::canHandleRequest): Return true for any URL schemes that are handled as
empty documents, and defer to the platform for everything else.

* WebProcess/WebPage/WebPage.h: Added platformCanHandleRequest.

* WebProcess/WebPage/mac/WebPageMac.mm:
(WebKit::WebPage::platformCanHandleRequest):
* WebProcess/WebPage/qt/WebPageQt.cpp:
(WebKit::WebPage::platformCanHandleRequest):
* WebProcess/WebPage/win/WebPageWin.cpp:
(WebKit::WebPage::platformCanHandleRequest):
Renamed from canHandleRequest.

Tools:

Test that WKBundlePageCanHandleRequest returns true for empty document URLs

* TestWebKitAPI/Tests/WebKit2/CanHandleRequest.cpp: Added.
(TestWebKitAPI::didReceiveMessageFromInjectedBundle): Store the result of the test.
(TestWebKitAPI::setInjectedBundleClient): Hook up our callback.
(TestWebKitAPI::TEST): Register "emptyscheme" as an empty document scheme, load a page to
ensure the web process is initialized, then ask the bundle to run the test and assert that
it succeeded.

* TestWebKitAPI/Tests/WebKit2/CanHandleRequest_Bundle.cpp: Added.
(TestWebKitAPI::CanHandleRequestTest::CanHandleRequestTest): Just call up to the base class.
(TestWebKitAPI::canHandleURL): Helper function to test whether WebKit2 claims to be able to
handle a given URL.
(TestWebKitAPI::runTest): Check that empty document URLs can be handled (and that an unknown
URL type cannot be handled).
(TestWebKitAPI::CanHandleRequestTest::didReceiveMessage): Run the test and return the
result.

* TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj:
* TestWebKitAPI/win/TestWebKitAPI.vcproj:
* TestWebKitAPI/win/TestWebKitAPIInjectedBundle.vcproj:
Added new files to the project.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@81105 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2011-03-14 Joseph Pecoraro <joepeck@webkit.org>
joepeck@webkit.org [Tue, 15 Mar 2011 04:11:30 +0000 (04:11 +0000)]
2011-03-14  Joseph Pecoraro  <joepeck@webkit.org>

        Reviewed by Eric Carlson.

        Stalled media elements don't stop delaying the load event
        https://bugs.webkit.org/show_bug.cgi?id=56316

        * http/tests/media/video-play-stall-before-meta-data-expected.txt: Added.
        * http/tests/media/video-play-stall-before-meta-data.html: Added.
2011-03-14  Joseph Pecoraro  <joepeck@webkit.org>

        Reviewed by Eric Carlson.

        Stalled media elements don't stop delaying the load event
        https://bugs.webkit.org/show_bug.cgi?id=56316

        We should stop delaying the load event when the load has
        stalled naturally, or if we require a user gesture to
        continue the load.

        Test: http/tests/media/video-play-stall-before-meta-data.html

        * html/HTMLMediaElement.cpp:
        (WebCore::HTMLMediaElement::setNetworkState): when suspending, stop delaying.
        (WebCore::HTMLMediaElement::progressEventTimerFired): when stalling, stop delaying.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@81104 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoRoll out r81097. It broke the 32-bit Mac build and caused assertion failures on...
mrowe@apple.com [Tue, 15 Mar 2011 03:56:29 +0000 (03:56 +0000)]
Roll out r81097.  It broke the 32-bit Mac build and caused assertion failures on the Windows Debug bots.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@81103 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2011-03-14 Ryuan Choi <ryuan.choi@samsung.com>
commit-queue@webkit.org [Tue, 15 Mar 2011 03:50:35 +0000 (03:50 +0000)]
2011-03-14  Ryuan Choi  <ryuan.choi@samsung.com>

        Reviewed by Martin Robinson.

        [GTK] Possible leaks after splitting TextCheckerClientEnchant.
        https://bugs.webkit.org/show_bug.cgi?id=55989

        Convert gchar* to GOwnPtr<gchar> and refactor the code.

        * WebCoreSupport/TextCheckerClientEnchant.cpp:
        (WebKit::TextCheckerClientEnchant::checkSpellingOfString):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@81102 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2011-03-13 MORITA Hajime <morrita@google.com>
morrita@google.com [Tue, 15 Mar 2011 02:38:01 +0000 (02:38 +0000)]
2011-03-13  MORITA Hajime  <morrita@google.com>

        Reviewed by Tony Chang.

        Crash when dragging and dropping in a document with an invalid XHTML header
        https://bugs.webkit.org/show_bug.cgi?id=48799

        * editing/pasteboard/drop-file-svg.html: Added.
        * editing/pasteboard/resources/drop-file-svg.svg: Added.
2011-03-13  MORITA Hajime  <morrita@google.com>

        Reviewed by Tony Chang.

        Crash when dragging and dropping in a document with an invalid XHTML header
        https://bugs.webkit.org/show_bug.cgi?id=48799

        DragController tried to dispatch textInput event even when the
        drag destination is not the editable area.
        This change skips the event dispatching on that case.

        Test: editing/pasteboard/drop-file-svg.html

        * page/DragController.cpp:
        (WebCore::DragController::dispatchTextInputEventFor):
        (WebCore::DragController::concludeEditDrag):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@81101 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2011-03-14 Geoffrey Garen <ggaren@apple.com>
ggaren@apple.com [Tue, 15 Mar 2011 02:28:03 +0000 (02:28 +0000)]
2011-03-14  Geoffrey Garen  <ggaren@apple.com>

        Reviewed by Oliver Hunt.

        Made the global object moving-GC-safe
        https://bugs.webkit.org/show_bug.cgi?id=56348

        SunSpider reports no change.

        * runtime/JSGlobalObject.cpp:
        (JSC::JSGlobalObject::markChildren): Removed a dubious comment that
        suggested we do not need to visit all our references during GC, since
        that is not true in a moving GC.

        Re-sorted data members by type, removed one duplicate, and added back
        the one missing mark I found.

        * runtime/JSGlobalObject.h: Re-sorted data members by type.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@81100 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoUnreviewed, rolling out r81094.
levin@chromium.org [Tue, 15 Mar 2011 02:27:48 +0000 (02:27 +0000)]
Unreviewed, rolling out r81094.
http://trac.webkit.org/changeset/81094
https://bugs.webkit.org/show_bug.cgi?id=56355

Patch by Sheriff Bot <webkit.review.bot@gmail.com> on 2011-03-14
Broke the chromium DRT related build. (Requested by dave_levin
on #webkit).

Source/WebCore:

* WebCore.exp.in:
* WebCore.order:
* page/Frame.cpp:
(WebCore::Frame::layerTreeAsText):
* page/Frame.h:
* rendering/RenderLayerCompositor.cpp:
(WebCore::RenderLayerCompositor::layerTreeAsText):
* rendering/RenderLayerCompositor.h:

Source/WebKit/chromium:

* public/WebFrame.h:
* src/WebFrameImpl.cpp:
(WebKit::WebFrameImpl::renderTreeAsText):
(WebKit::WebFrameImpl::layerTreeAsText):
* src/WebFrameImpl.h:

Tools:

* DumpRenderTree/chromium/DumpRenderTree.cpp:
(main):
* DumpRenderTree/chromium/LayoutTestController.cpp:
(LayoutTestController::LayoutTestController):
(LayoutTestController::layerTreeAsText):
* DumpRenderTree/chromium/LayoutTestController.h:
* DumpRenderTree/chromium/TestShell.cpp:
(TestShell::runFileTest):
(TestShell::dump):
* DumpRenderTree/chromium/TestShell.h:
(TestParams::TestParams):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@81099 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2011-03-14 Adam Barth <abarth@webkit.org>
abarth@webkit.org [Tue, 15 Mar 2011 02:24:37 +0000 (02:24 +0000)]
2011-03-14  Adam Barth  <abarth@webkit.org>

        Apparently we need to exclude DefaultSharedWorkerRepository.cpp from
        the Chromium build, otherwise the objects visible in the global scope
        change.

        * WebCore.gyp/WebCore.gyp:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@81098 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2011-03-14 Rik Cabanier <cabanier@gmail.com>
commit-queue@webkit.org [Tue, 15 Mar 2011 02:17:53 +0000 (02:17 +0000)]
2011-03-14  Rik Cabanier  <cabanier@gmail.com>

        Reviewed by David Hyatt.

        bug 56052: percentages are incorrectly rounded in WebKit
        https://bugs.webkit.org/show_bug.cgi?id=56052

        * platform/mac/fast/css/percentage-non-integer-expected.txt:
2011-03-14  Rik Cabanier  <cabanier@adobe.com>

        Reviewed by David Hyatt.

        bug 56052: percentages are incorrectly rounded in WebKit
        https://bugs.webkit.org/show_bug.cgi?id=56052

        test: fast/css/percentage-non-integer.html

        * css/CSSStyleSelector.cpp:
        (WebCore::CSSStyleSelector::applyProperty):
        * page/PrintContext.cpp:
        (WebCore::PrintContext::pageProperty):
        * platform/Length.h:
        (WebCore::Length::Length):
        (WebCore::Length::operator*=):
        (WebCore::Length::value):
        (WebCore::Length::percent):
        (WebCore::Length::setValue):
        (WebCore::Length::calcValue):
        (WebCore::Length::calcMinValue):
        (WebCore::Length::isUndefined):
        (WebCore::Length::isZero):
        (WebCore::Length::isPositive):
        (WebCore::Length::isNegative):
        * rendering/AutoTableLayout.cpp:
        (WebCore::AutoTableLayout::recalcColumn):
        (WebCore::AutoTableLayout::computePreferredLogicalWidths):
        (WebCore::AutoTableLayout::calcEffectiveLogicalWidth):
        (WebCore::AutoTableLayout::layout):
        * rendering/FixedTableLayout.cpp:
        (WebCore::FixedTableLayout::calcWidthArray):
        (WebCore::FixedTableLayout::layout):
        * rendering/RenderTableSection.cpp:
        (WebCore::RenderTableSection::addCell):
        (WebCore::RenderTableSection::layoutRows):
        * rendering/style/BorderData.h:
        (WebCore::BorderData::hasBorderRadius):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@81097 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2011-03-14 Adam Barth <abarth@webkit.org>
abarth@webkit.org [Tue, 15 Mar 2011 02:09:43 +0000 (02:09 +0000)]
2011-03-14  Adam Barth  <abarth@webkit.org>

        Attempted Chromium build fix.  Exclude AllInOne harder.

        * WebCore.gyp/WebCore.gyp:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@81096 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2011-03-14 Ryosuke Niwa <rniwa@webkit.org>
rniwa@webkit.org [Tue, 15 Mar 2011 02:07:09 +0000 (02:07 +0000)]
2011-03-14  Ryosuke Niwa  <rniwa@webkit.org>

        Reviewed by Darin Adler.

        EventHandler calls shouldChangeSelection needlessly
        https://bugs.webkit.org/show_bug.cgi?id=56324

        Extracted setSelectionIfNeeded and setNonDirectionalSelectionIfNeeded and
        avoided calling shouldChangeSelection and setSelection when the existing
        selection is identical to that of new selection.

        * page/EventHandler.cpp:
        (WebCore::setSelectionIfNeeded): Extracted.
        (WebCore::setNonDirectionalSelectionIfNeeded): Extracted.
        (WebCore::EventHandler::selectClosestWordFromMouseEvent): Calls a helper function above.
        (WebCore::EventHandler::selectClosestWordOrLinkFromMouseEvent): Ditto.
        (WebCore::EventHandler::handleMousePressEventTripleClick): Ditto.
        (WebCore::EventHandler::handleMousePressEventSingleClick): Ditto.
        (WebCore::EventHandler::updateSelectionForMouseDrag): Ditto.
        (WebCore::EventHandler::handleMouseReleaseEvent): Ditto.
2011-03-14  Anton Muhin  <antonm@chromium.org>

        Reviewed by Adam Barth.

        [v8] Rework object group building.
        https://bugs.webkit.org/show_bug.cgi?id=55399

        Instead of going top-down (from owner to owned elements), go up---from objects
        to their group ids.  That fits better to v8's object grouping model and guarantees
        that each wrapper belongs to the single group.

        Alas, this cannot be implemented for one kind of objects---CSSProperties.

        Part of core GC algorithm and tested extensively by exisiting layout tests.

        * platform/chromium/test_expectations.txt: Temporary suppress text differences.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@81095 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2011-03-14 Daniel Sievers <sievers@google.com>
commit-queue@webkit.org [Tue, 15 Mar 2011 01:56:37 +0000 (01:56 +0000)]
2011-03-14  Daniel Sievers  <sievers@google.com>

        Reviewed by Simon Fraser.

        [Chromium] Make RenderAsTextBehavior and LayerTreeAsTextBehavior tweakable from the DumpRenderTree commandline
        https://bugs.webkit.org/show_bug.cgi?id=56139

        * WebCore.exp.in:
        * WebCore.order:
        * page/Frame.cpp:
        (WebCore::Frame::layerTreeAsText):
        * page/Frame.h:
        * rendering/RenderLayerCompositor.cpp:
        (WebCore::RenderLayerCompositor::layerTreeAsText):
        * rendering/RenderLayerCompositor.h:
2011-03-14  Daniel Sievers  <sievers@google.com>

        Reviewed by Simon Fraser.

        [Chromium] Make RenderAsTextBehavior and LayerTreeAsTextBehavior tweakable from the DumpRenderTree commandline
        https://bugs.webkit.org/show_bug.cgi?id=56139

        * public/WebFrame.h:
        * src/WebFrameImpl.cpp:
        (WebKit::WebFrameImpl::renderTreeAsText):
        (WebKit::WebFrameImpl::layerTreeAsText):
        * src/WebFrameImpl.h:
2011-03-14  Daniel Sievers  <sievers@google.com>

        Reviewed by Simon Fraser.

        [Chromium] Make RenderAsTextBehavior and LayerTreeAsTextBehavior tweakable from the DumpRenderTree commandline
        https://bugs.webkit.org/show_bug.cgi?id=56139

        * DumpRenderTree/chromium/DumpRenderTree.cpp:
        (main):
        * DumpRenderTree/chromium/LayoutTestController.cpp:
        (LayoutTestController::LayoutTestController):
        (LayoutTestController::layerTreeAsText):
        * DumpRenderTree/chromium/LayoutTestController.h:
        (LayoutTestController::setShowDebugLayerTree):
        * DumpRenderTree/chromium/TestShell.cpp:
        (TestShell::runFileTest):
        (TestShell::dump):
        * DumpRenderTree/chromium/TestShell.h:
        (TestParams::TestParams):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@81094 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2011-03-14 Adam Barth <abarth@webkit.org>
abarth@webkit.org [Tue, 15 Mar 2011 01:42:59 +0000 (01:42 +0000)]
2011-03-14  Adam Barth  <abarth@webkit.org>

        Reviewed by Dimitri Glazkov.

        Add remaining files to WebCore.gypi
        https://bugs.webkit.org/show_bug.cgi?id=56351

        Adding the remaining files to WebCore.gypi required updating the
        include/exclude lists in WebCore.gyp.  These lists aren't overly
        elegant, but we can try to improve them in the future.

        * WebCore.gyp/WebCore.gyp:
        * WebCore.gypi:
        * bindings/v8/ScriptCachedFrameData.cpp:
            - Added ifdefs to this file to match the header.
        * platform/graphics/WOFFFileFormat.cpp:
            - Fixed build error when compiling without ENABLE(OPENTYPE_SANITIZER).

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@81093 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2011-03-14 Anton Muhin <antonm@chromium.org>
antonm@chromium.org [Tue, 15 Mar 2011 01:19:14 +0000 (01:19 +0000)]
2011-03-14  Anton Muhin  <antonm@chromium.org>

        Reviewed by Adam Barth.

        [v8] Rework object group building.
        https://bugs.webkit.org/show_bug.cgi?id=55399

        Instead of going top-down (from owner to owned elements), go up---from objects
        to their group ids.  That fits better to v8's object grouping model and guarantees
        that each wrapper belongs to the single group.

        Alas, this cannot be implemented for one kind of objects---CSSProperties.

        Part of core GC algorithm and tested extensively by exisiting layout tests.

        * platform/chromium/test_expectations.txt: Temporary suppress text differences.
2011-03-14  Anton Muhin  <antonm@chromium.org>

        Reviewed by Adam Barth.

        [v8] Rework object group building.
        https://bugs.webkit.org/show_bug.cgi?id=55399

        Instead of going top-down (from owner to owned elements), go up---from objects
        to their group ids.  That fits better to v8's object grouping model and guarantees
        that each wrapper belongs to the single group.

        Alas, this cannot be implemented for one kind of objects---CSSProperties.

        Part of core GC algorithm and tested extensively by exisiting layout tests.

        * bindings/scripts/CodeGeneratorV8.pm:
        * bindings/scripts/test/V8/V8TestInterface.cpp:
        * bindings/scripts/test/V8/V8TestMediaQueryListListener.cpp:
        * bindings/scripts/test/V8/V8TestObj.cpp:
        * bindings/v8/V8GCController.cpp:
        (WebCore::calculateGroupId):
        (WebCore::calculateRootStyleSheet):
        (WebCore::GrouperVisitor::visitDOMWrapper):
        (WebCore::GrouperVisitor::applyGrouping):
        (WebCore::V8GCController::gcPrologue):
        * bindings/v8/WrapperTypeInfo.h:
        (WebCore::WrapperTypeInfo::isSubclass):
        * css/CSSRuleList.h:
        (WebCore::CSSRuleList::styleList):
        * css/StyleSheetList.h:
        (WebCore::StyleSheetList::document):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@81092 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoSkipping some tests on Windows due to problems on mouse click handling.
luiz@webkit.org [Tue, 15 Mar 2011 01:15:47 +0000 (01:15 +0000)]
Skipping some tests on Windows due to problems on mouse click handling.
See https://bugs.webkit.org/show_bug.cgi?id=56347.

* platform/win/Skipped:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@81091 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2011-03-14 Ryosuke Niwa <rniwa@webkit.org>
rniwa@webkit.org [Tue, 15 Mar 2011 01:13:19 +0000 (01:13 +0000)]
2011-03-14  Ryosuke Niwa  <rniwa@webkit.org>

        Mac Leopard rebaselines for r81053 (Grabbed off of Chromium bots).

        * platform/mac-leopard/fast/text/atsui-rtl-override-selection-expected.checksum:
        * platform/mac-leopard/fast/text/atsui-rtl-override-selection-expected.png:
        * platform/mac-leopard/editing/selection/drag-select-1-expected.checksum:
        * platform/mac-leopard/editing/selection/drag-select-1-expected.png:
        * platform/mac-leopard/editing/selection/fake-drag-expected.checksum:
        * platform/mac-leopard/editing/selection/fake-drag-expected.png:
        * platform/mac-leopard/editing/selection/inline-closest-leaf-child-expected.checksum:
        * platform/mac-leopard/editing/selection/inline-closest-leaf-child-expected.png:
        * platform/mac-leopard/svg/custom/mouse-move-on-svg-container-expected.checksum:
        * platform/mac-leopard/svg/custom/mouse-move-on-svg-container-expected.png:
        * platform/mac-leopard/svg/custom/mouse-move-on-svg-container-standalone-expected.checksum:
        * platform/mac-leopard/svg/custom/mouse-move-on-svg-container-standalone-expected.png:
        * platform/mac-leopard/svg/custom/mouse-move-on-svg-root-expected.checksum:
        * platform/mac-leopard/svg/custom/mouse-move-on-svg-root-expected.png:
        * platform/mac-leopard/svg/custom/mouse-move-on-svg-root-standalone-expected.checksum:
        * platform/mac-leopard/svg/custom/mouse-move-on-svg-root-standalone-expected.png:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@81090 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoCrashTracer: 60 crashes in WebProcess at com.apple.WebKit2: -[AccessibilityWebPageObj...
cfleizach@apple.com [Tue, 15 Mar 2011 01:11:35 +0000 (01:11 +0000)]
CrashTracer: 60 crashes in WebProcess at com.apple.WebKit2: -[AccessibilityWebPageObject accessibilityHitTest:] + 248
https://bugs.webkit.org/show_bug.cgi?id=56336

Reviewed by Beth Dakin.

Crash trace indicates a nil pointer access in one of these m_page->mainFrame()->coreFrame()->view()
pointers.

* WebProcess/WebPage/mac/AccessibilityWebPageObject.mm:
(-[AccessibilityWebPageObject accessibilityHitTest:]):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@81089 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoAssertion failure by form validation message for <select required> with float:left
tkent@chromium.org [Tue, 15 Mar 2011 00:53:53 +0000 (00:53 +0000)]
Assertion failure by form validation message for <select required> with float:left
https://bugs.webkit.org/show_bug.cgi?id=55995

Reviewed by James Robinson.

Source/WebCore:

Test: fast/forms/interactive-validation-select-crash.html

* rendering/RenderBlock.cpp:
(WebCore::canMergeContiguousAnonymousBlocks):
  isAnonymousBlock() doesn't mean it is a RenderBlock. We need to check isRenderBlock().

LayoutTests:

* fast/forms/interactive-validation-select-crash-expected.txt: Added.
* fast/forms/interactive-validation-select-crash.html: Added.
* platform/gtk/Skipped:
* platform/mac/Skipped:
* platform/qt/Skipped:
* platform/win/Skipped:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@81088 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoProvide API for creating WKImage from CGImageRef
alice.liu@apple.com [Tue, 15 Mar 2011 00:48:45 +0000 (00:48 +0000)]
Provide API for creating WKImage from CGImageRef
https://bugs.webkit.org/show_bug.cgi?id=56159

Reviewed by Sam Weinig.

Add function for creating a WKImage from CGImage data.
* Shared/API/c/cg/WKImageCG.cpp:
(WKImageCreateFromCGImage): Added. Creates a graphics context for the bitmap-backed image and draws into it.
* Shared/API/c/cg/WKImageCG.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@81087 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2011-03-14 Geoffrey Garen <ggaren@apple.com>
ggaren@apple.com [Tue, 15 Mar 2011 00:39:56 +0000 (00:39 +0000)]
2011-03-14  Geoffrey Garen  <ggaren@apple.com>

        Reviewed by Oliver Hunt.

        Made JSWrapperObject and subclasses moving-GC-safe
        https://bugs.webkit.org/show_bug.cgi?id=56346

        SunSpider reports no change.

        * runtime/BooleanObject.cpp:
        (JSC::BooleanObject::BooleanObject):
        * runtime/DateInstance.cpp:
        (JSC::DateInstance::DateInstance): No more need for JSGlobalData, since
        we don't initialize the wrapped value in our constructor.

        * runtime/DateInstance.h: Don't set the OverridesMarkChildren flag because
        we do not in fact override markChildren.

        * runtime/DatePrototype.h: Declare an anonymous slot, since wrapper object
        no longer does so for us. Also added an ASSERT to catch a latent bug,
        where DatePrototype stomped on its base class's anonymous slot. Hard-coded
        anonymous slots are a plague on our code. This doesn't cause any problems
        in our existing code since the base class never reads the anonymous slot
        it declares, but it caused crashes when I tried to start using the slot
        in an initial version of this patch.

        * runtime/JSWrapperObject.h:
        (JSC::JSWrapperObject::JSWrapperObject):
        (JSC::JSWrapperObject::internalValue):
        (JSC::JSWrapperObject::setInternalValue): Resolved a problem where
        our internal value was stored in two places: an anonymous slot, and a
        data member which was not always visited during GC. Now, we only use the
        data member, and we always visit it. (Instead of relying on certain
        subclasses to set the OverridesMarkChildren bit, we set it ourselves.)

        * runtime/NumberObject.cpp:
        (JSC::NumberObject::NumberObject): No more need for JSGlobalData, since
        we don't initialize the wrapped value in our constructor.

        * runtime/NumberObject.h: Removed meaningless declaration.

        * runtime/StringObject.cpp:
        (JSC::StringObject::StringObject): No more need for JSGlobalData, since
        we don't initialize the wrapped value in our constructor.

        * runtime/StringObject.h: Don't set the OverridesMarkChildren flag because
        we do not in fact override markChildren.

        * runtime/StringPrototype.h: Declare an anonymous slot, since wrapper object
        no longer does so for us. Also added an ASSERT to catch a latent bug,
        where DatePrototype stomped on its base class's anonymous slot. Hard-coded
        anonymous slots are a plague on our code.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@81086 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2011-03-14 Michael Saboff <msaboff@apple.com>
msaboff@apple.com [Tue, 15 Mar 2011 00:36:23 +0000 (00:36 +0000)]
2011-03-14  Michael Saboff  <msaboff@apple.com>

        Reviewed by Gavin Barraclough.

        Look-ahead assertions with back references don’t work as expected
        https://bugs.webkit.org/show_bug.cgi?id=56082

        Changed parentheses assertion processing to temporarily back out the
        number of known characters after the assertion while processing the
        assertion.  This was done so that assertions don't fail due to
        checking the number of required characters as additional to the
        rest of the express since assertions don't "consume" input.
        Added a byte code to uncheck characters to support the change.

        * yarr/YarrInterpreter.cpp:
        (JSC::Yarr::Interpreter::matchDisjunction):
        (JSC::Yarr::ByteCompiler::uncheckInput):
        (JSC::Yarr::ByteCompiler::emitDisjunction):
        * yarr/YarrInterpreter.h:
        (JSC::Yarr::ByteTerm::UncheckInput):
2011-03-14  Michael Saboff  <msaboff@apple.com>

        Reviewed by Gavin Barraclough.

        Look-ahead assertions with back references don’t work as expected
        https://bugs.webkit.org/show_bug.cgi?id=56082

        New tests to verify the corresponding changes.

        * fast/regex/assertion-expected.txt: Added.
        * fast/regex/assertion.html: Added.
        * fast/regex/script-tests/assertion.js: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@81085 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2011-03-14 Balazs Kelemen <kbalazs@webkit.org>
kbalazs@webkit.org [Tue, 15 Mar 2011 00:32:53 +0000 (00:32 +0000)]
2011-03-14  Balazs Kelemen  <kbalazs@webkit.org>

        Reviewed by Adam Roben.

        [Qt][WK2]Unbreak InjectedBundle on Qt
        https://bugs.webkit.org/show_bug.cgi?id=54109

        No code changes so no new tests.

        Revert the changes that were needed to use KURL
        in WebKitTestRunner.

        * Configurations/WebCore.xcconfig:
        * WebCore.exp.in:
2011-03-14  Balazs Kelemen  <kbalazs@webkit.org>

        Reviewed by Adam Roben.

        [Qt][WK2]Unbreak InjectedBundle on Qt
        https://bugs.webkit.org/show_bug.cgi?id=54109

        Add API's to WKURL and WKString to satisfy
        WebKitTestRunner's needs.

        * Shared/API/c/WKString.cpp:
        (WKStringIsEqualToUTF8CStringIgnoringCase):
        * Shared/API/c/WKString.h:
        * Shared/API/c/WKURL.cpp:
        (WKURLCopyHostName):
        (WKURLCopyScheme):
        * Shared/API/c/WKURL.h:
        * Shared/WebString.h:
        (WebKit::WebString::equalToUTF8StringIgnoringCase):
        * Shared/WebURL.h: Added OwnPtr<KURL> member to be able
        to lazily parse the URL and store the result.
        (WebKit::WebURL::host):
        (WebKit::WebURL::protocol):
        (WebKit::WebURL::parseURLIfNecessary):
        * win/WebKit2.def: Revert the symbol exports that were
        needed to use KURL in WebKitTestRunner.
2011-03-14  Balazs Kelemen  <kbalazs@webkit.org>

        Reviewed by Adam Roben.

        [Qt][WK2]Unbreak InjectedBundle on Qt
        https://bugs.webkit.org/show_bug.cgi?id=54109

        * WebKitTestRunner/InjectedBundle/InjectedBundlePage.cpp:
        (WTR::InjectedBundlePage::willSendRequestForFrame): Use new WebKit2
        API's instead of KURL to analyze the URL.
        * WebKitTestRunner/StringFunctions.h:
        (WTR::adoptWK): Moved from InjectedBundlePage.cpp to make it available globally
        in the WebKitTestRunner project.
        (WTR::toWK): Use adoptWK.
        (WTR::operator<<): Added helper for printing a WKURL.
        * WebKitTestRunner/WebKitTestRunner.xcodeproj/project.pbxproj: Do not link
        against WebCore.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@81084 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoIt's the same old song (about the linux debug crash).
levin@chromium.org [Mon, 14 Mar 2011 23:59:38 +0000 (23:59 +0000)]
It's the same old song (about the linux debug crash).

* platform/chromium/test_expectations.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@81083 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2011-03-14 Jeff Miller <jeffm@apple.com>
jeffm@apple.com [Mon, 14 Mar 2011 23:52:53 +0000 (23:52 +0000)]
2011-03-14  Jeff Miller  <jeffm@apple.com>

        Reviewed by Adam Roben.

        Add WKViewSetScrollOffsetOnNextResize() to C API on Windows
        https://bugs.webkit.org/show_bug.cgi?id=56340

        * UIProcess/API/C/win/WKView.cpp:
        (WKViewSetScrollOffsetOnNextResize): Added.
        * UIProcess/API/C/win/WKView.h: Added WKViewSetScrollOffsetOnNextResize().
        * UIProcess/win/WebView.cpp:
        (WebKit::WebView::onSizeEvent): Apply any scroll offset when setting the drawing area size.
        (WebKit::WebView::setScrollOffsetOnNextResize): Added.
        * UIProcess/win/WebView.h: Added setScrollOffsetOnNextResize().

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@81082 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2011-03-14 Viatcheslav Ostapenko <ostapenko.viatcheslav@nokia.com>
commit-queue@webkit.org [Mon, 14 Mar 2011 23:47:07 +0000 (23:47 +0000)]
2011-03-14  Viatcheslav Ostapenko  <ostapenko.viatcheslav@nokia.com>

        Reviewed by Laszlo Gombos.

        [Qt] Warning that round/roundf functions are already defined when compiled with RVCT 4 on symbian.
        https://bugs.webkit.org/show_bug.cgi?id=56133

        Add condition to not compile webkit internal math round functions on RVCT compiler versions
        from 3.0.0 because they are already defined in compiler math library.

        * wtf/MathExtras.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@81081 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoFix two memory leaks within LayoutTestController's storage support.
mrowe@apple.com [Mon, 14 Mar 2011 23:40:39 +0000 (23:40 +0000)]
Fix two memory leaks within LayoutTestController's storage support.

Reviewed by Adam Roben.

* DumpRenderTree/mac/LayoutTestControllerMac.mm:
(LayoutTestController::originsWithLocalStorage): Don't leak the JSStringRef that is returned
by JSStringCreateWithCFString.
(LayoutTestController::deleteLocalStorageForOrigin): Don't leak the WebSecurityOrigin that we
allocate.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@81080 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoBug 56284 - Add a dataflow intermediate representation for use in JIT generation.
barraclough@apple.com [Mon, 14 Mar 2011 23:31:00 +0000 (23:31 +0000)]
Bug 56284 - Add a dataflow intermediate representation for use in JIT generation.

Reviewed by Geoffrey Garen & Oliver Hunt.

The JSC JIT presently generates code directly from the bytecode used by the interpreter.
This is not an optimal intermediate representation for JIT code generation, since it does
not capture liveness information of values, and provides little opportunity to perform
any static analysis for even primitive types. The JIT currently generates two code paths,
a fast path handling common cases, and a slower path handling less common operand types.
However the slow path jumps back into the fast path, meaning that information arising
from the earlier type checks cannot be propagated to later operations.

This patch adds:
    * a dataflow intermediate representation capable of describing a single basic block
      of operations,
    * a mechanism to convert a simple, single-block bytecode functions to the new IR,
    * and a JIT code generator capable of generating code from this representation.

The JIT generates two code paths, with the slower path not reentering the fast path
mid-block, allowing speculative optimizations to be made on the hot path, with type
information arising from these speculative decisions able to be propagated through the
dataflow. Code generation of both speculative and non-speculative paths exploits the type
and liveness information represented in the dataflow graph to attempt to avoid redundant
boxing and type-checking of values, and to remove unnecessary spills of temporary values
to the RegisterFile.

The dataflow JIT currently can only support a subset of bytecode operations, limited to
arithmetic, bit-ops, and basic property access. Functions that cannot be compiled by the
dataflow JIT will be run using the existing JIT. The coverage of the dataflow JIT will be
expanded to include, control-flow, function calls, and then the long-tail of remaining
bytecode instructions. The JIT presently only support JSVALUE64, and as a consequence of
this only supports x86-64.

The status of the dataflow JIT is currently work-in-progress. Limitations of the present
JIT code generation may cause performance regressions, particularly:
    * the policy to only generate arithmetic code on the speculative path using integer
      instructions, never using floating point.
    * the policy to only generate arithmetic code on the non-speculative path using
      floating point instructions, never using integer.
    * always generating JSValue adds on the non-speculative path as a call out to a
      C-function, never handling this in JIT code.
    * always assuming by-Value property accesses on the speculative path to be array
      accesses.
    * generating all by-Value property accesses from the non-speculative path as a call
      out to a C-function.
    * generating all by-Indentifer property accesses as a call out to a C-function.
Due to these regressions, the code is landed in a state where it is disabled in most
cases by the ENABLE_DFG_JIT_RESTRICTIONS guard in Platform.h. As these regressions are
addressed, the JIT will be allowed to trigger in more cases.

* JavaScriptCore.xcodeproj/project.pbxproj:
    - Added new files to Xcode project.
* dfg: Added.
    - Added directory for new code.
* dfg/DFGByteCodeParser.cpp: Added.
* dfg/DFGByteCodeParser.h: Added.
    - Contruct a DFG::Graph representation from a bytecode CodeBlock.
* dfg/DFGGenerationInfo.h: Added.
    - Track type & register information for VirtualRegisters during JIT code generation.
* dfg/DFGGraph.cpp: Added.
* dfg/DFGGraph.h: Added.
    - Dataflow graph intermediate representation for code generation.
* dfg/DFGJITCodeGenerator.cpp: Added.
* dfg/DFGJITCodeGenerator.h: Added.
    - Base class for SpeculativeJIT & NonSpeculativeJIT to share common functionality.
* dfg/DFGJITCompiler.cpp: Added.
* dfg/DFGJITCompiler.h: Added.
    - Class responsible for driving code generation of speculativeJIT & non-speculative
      code paths from the dataflow graph.
* dfg/DFGNonSpeculativeJIT.cpp: Added.
* dfg/DFGNonSpeculativeJIT.h: Added.
    - Used to generate the non-speculative code path, this make no assumptions
      about operand types.
* dfg/DFGOperations.cpp: Added.
* dfg/DFGOperations.h: Added.
    - Helper functions called from the JIT generated code.
* dfg/DFGRegisterBank.h: Added.
    - Used to track contents of physical registers during JIT code generation.
* dfg/DFGSpeculativeJIT.cpp: Added.
* dfg/DFGSpeculativeJIT.h: Added.
    - Used to generate the speculative code path, this make assumptions about
      operand types to enable optimization.
* runtime/Executable.cpp:
    - Add code to attempt to use the DFG JIT to compile a function, with fallback
      to the existing JIT.
* wtf/Platform.h:
    - Added compile guards to enable the DFG JIT.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@81079 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoAdd more linux debug crashers to the sad list.
levin@chromium.org [Mon, 14 Mar 2011 23:29:36 +0000 (23:29 +0000)]
Add more linux debug crashers to the sad list.

* platform/chromium/test_expectations.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@81078 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2011-03-14 Balazs Kelemen <kbalazs@webkit.org>
kbalazs@webkit.org [Mon, 14 Mar 2011 23:27:12 +0000 (23:27 +0000)]
2011-03-14  Balazs Kelemen  <kbalazs@webkit.org>

        Reviewed by Kenneth Rohde Christiansen.

        check-webkit-style falsely should not complain about WebKit2 API headers in the Shared directory
        https://bugs.webkit.org/show_bug.cgi?id=56321

        * Scripts/webkitpy/style/checker.py: Add the patch "Source/WebKit2/Shared/API/c"
        to the list of exceptions for WebKit2 C API headers.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@81077 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2011-03-14 Martin Robinson <mrobinson@igalia.com>
mrobinson@webkit.org [Mon, 14 Mar 2011 23:24:01 +0000 (23:24 +0000)]
2011-03-14  Martin Robinson  <mrobinson@igalia.com>

        Update GTK+ pixel results after r81053.

        * platform/gtk/editing/selection/drag-select-1-expected.checksum:
        * platform/gtk/editing/selection/drag-select-1-expected.png:
        * platform/gtk/editing/selection/fake-drag-expected.checksum: Added.
        * platform/gtk/editing/selection/fake-drag-expected.png: Added.
        * platform/gtk/editing/selection/inline-closest-leaf-child-expected.checksum: Added.
        * platform/gtk/editing/selection/inline-closest-leaf-child-expected.png: Added.
        * platform/gtk/svg/custom/mouse-move-on-svg-container-standalone-expected.checksum:
        * platform/gtk/svg/custom/mouse-move-on-svg-container-standalone-expected.png:
        * platform/gtk/svg/custom/mouse-move-on-svg-root-standalone-expected.checksum:
        * platform/gtk/svg/custom/mouse-move-on-svg-root-standalone-expected.png:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@81076 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoShow the list of recent leaky builds in reverse chronological order
aroben@apple.com [Mon, 14 Mar 2011 23:14:02 +0000 (23:14 +0000)]
Show the list of recent leaky builds in reverse chronological order

Previously they were being shown in a deterministic but surprising order based on
lexicographically sorting the build offsets (-1, -10, -2, -3...).

Fixes <http://webkit.org/b/56342> Leaks Viewer shows recent builds in a strange order

Reviewed by Darin Adler.

* BuildSlaveSupport/build.webkit.org-config/public_html/LeaksViewer/RecentBuildsLoader.js:
(RecentBuildsLoader.prototype.start): Sort the builds by revision before passing them to our
callback.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@81075 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoI checked in Intel-only version of these files in r80975. This fixes that by building all
psolanki@apple.com [Mon, 14 Mar 2011 23:13:19 +0000 (23:13 +0000)]
I checked in Intel-only version of these files in r80975. This fixes that by building all
the correct slices - i386/x86_64/ppc for SnowLeopard and i386/x86_64/ppc/ppc64 for Leopard.

Rubber-stamped by Mark Rowe.

* libWebKitSystemInterfaceLeopard.a:
* libWebKitSystemInterfaceSnowLeopard.a:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@81074 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoUpstream test expectations from the chromium repository to WebKit.
levin@chromium.org [Mon, 14 Mar 2011 22:47:43 +0000 (22:47 +0000)]
Upstream test expectations from the chromium repository to WebKit.

* platform/chromium/test_expectations.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@81073 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2011-03-14 Ryosuke Niwa <rniwa@webkit.org>
rniwa@webkit.org [Mon, 14 Mar 2011 22:38:10 +0000 (22:38 +0000)]
2011-03-14  Ryosuke Niwa  <rniwa@webkit.org>

        Mac pixel rebaselines for r81053.

        * platform/mac/editing/selection/drag-select-1-expected.checksum:
        * platform/mac/editing/selection/drag-select-1-expected.png:
        * platform/mac/editing/selection/fake-drag-expected.checksum:
        * platform/mac/editing/selection/fake-drag-expected.png:
        * platform/mac/fast/text/atsui-rtl-override-selection-expected.checksum:
        * platform/mac/fast/text/atsui-rtl-override-selection-expected.png:
        * platform/mac/svg/custom/mouse-move-on-svg-container-expected.checksum:
        * platform/mac/svg/custom/mouse-move-on-svg-container-expected.png:
        * platform/mac/svg/custom/mouse-move-on-svg-container-standalone-expected.checksum:
        * platform/mac/svg/custom/mouse-move-on-svg-container-standalone-expected.png:
        * platform/mac/svg/custom/mouse-move-on-svg-root-expected.checksum:
        * platform/mac/svg/custom/mouse-move-on-svg-root-expected.png:
        * platform/mac/svg/custom/mouse-move-on-svg-root-standalone-expected.checksum:
        * platform/mac/svg/custom/mouse-move-on-svg-root-standalone-expected.png:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@81072 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2011-03-14 Geoffrey Garen <ggaren@apple.com>
ggaren@apple.com [Mon, 14 Mar 2011 22:31:29 +0000 (22:31 +0000)]
2011-03-14  Geoffrey Garen  <ggaren@apple.com>

        Reviewed by Oliver Hunt.

        Removed more cases of DeprecatedPtr (exception, SmallStrings)
        https://bugs.webkit.org/show_bug.cgi?id=56332

        * runtime/Identifier.cpp:
        (JSC::Identifier::add):
        (JSC::Identifier::addSlowCase): Use a variable instead of a hard-coded
        constant, to make this code less brittle.

        * runtime/JSGlobalData.h: Use HeapRoot instead of DeprecatedPtr because
        this reference is owned and managed directly by the heap.

        * runtime/JSString.cpp:
        (JSC::JSString::substringFromRope):
        * runtime/JSString.h:
        (JSC::jsSingleCharacterString):
        (JSC::jsSingleCharacterSubstring):
        (JSC::jsString):
        (JSC::jsStringWithFinalizer):
        (JSC::jsSubstring):
        (JSC::jsOwnedString): Use a variable instead of a hard-coded
        constant, to make this code less brittle.

        * runtime/SmallStrings.cpp:
        (JSC::SmallStringsStorage::rep):
        (JSC::SmallStringsStorage::SmallStringsStorage):
        (JSC::SmallStrings::SmallStrings):
        (JSC::SmallStrings::markChildren):
        (JSC::SmallStrings::clear):
        (JSC::SmallStrings::count): Use a variable instead of a hard-coded
        constant, to make this code less brittle.

        * runtime/SmallStrings.h:
        (JSC::SmallStrings::singleCharacterString): Use HeapRoot instead of
        DeprecatedPtr because these references are owned and managed directly by
        the heap.

        Stop using FixedArray because we only want a very limited set
        of classes to be able to use HeapRoot. (Replaced with manual ASSERTs.)

        * runtime/WriteBarrier.h:
        (JSC::operator==):
        (JSC::WriteBarrier::WriteBarrier):
        (JSC::HeapRoot::HeapRoot):
        (JSC::HeapRoot::operator=): Added HeapRoot, which is allowed to set
        without write barrier because we assume all HeapRoots are scanned during
        all GC passes.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@81071 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoUpdate the chromium windows baselines for r81053.
levin@chromium.org [Mon, 14 Mar 2011 22:28:49 +0000 (22:28 +0000)]
Update the chromium windows baselines for r81053.

* platform/chromium-win/editing/selection/drag-select-1-expected.checksum:
* platform/chromium-win/editing/selection/drag-select-1-expected.png:
* platform/chromium-win/editing/selection/drag-select-1-expected.txt:
* platform/chromium-win/editing/selection/fake-drag-expected.txt:
* platform/chromium-win/svg/custom/mouse-move-on-svg-container-expected.checksum:
* platform/chromium-win/svg/custom/mouse-move-on-svg-container-expected.png:
* platform/chromium-win/svg/custom/mouse-move-on-svg-container-expected.txt:
* platform/chromium-win/svg/custom/mouse-move-on-svg-container-standalone-expected.checksum:
* platform/chromium-win/svg/custom/mouse-move-on-svg-container-standalone-expected.png:
* platform/chromium-win/svg/custom/mouse-move-on-svg-container-standalone-expected.txt:
* platform/chromium-win/svg/custom/mouse-move-on-svg-root-expected.checksum:
* platform/chromium-win/svg/custom/mouse-move-on-svg-root-expected.png:
* platform/chromium-win/svg/custom/mouse-move-on-svg-root-expected.txt:
* platform/chromium-win/svg/custom/mouse-move-on-svg-root-standalone-expected.checksum:
* platform/chromium-win/svg/custom/mouse-move-on-svg-root-standalone-expected.png:
* platform/chromium-win/svg/custom/mouse-move-on-svg-root-standalone-expected.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@81070 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2011-03-14 Jarkko Sakkinen <jarkko.j.sakkinen@gmail.com>
commit-queue@webkit.org [Mon, 14 Mar 2011 22:25:24 +0000 (22:25 +0000)]
2011-03-14  Jarkko Sakkinen  <jarkko.j.sakkinen@gmail.com>

        Reviewed by Kenneth Rohde Christiansen.

        [Qt] Compilation fails with --3d-canvas
        https://bugs.webkit.org/show_bug.cgi?id=55964

        * WebCore.pro:
        * platform/graphics/qt/Extensions3DQt.cpp:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@81069 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoShow a list of up to 10 recent leaky builds when Leaks Viewer loads
aroben@apple.com [Mon, 14 Mar 2011 22:19:55 +0000 (22:19 +0000)]
Show a list of up to 10 recent leaky builds when Leaks Viewer loads

Fixes <http://webkit.org/b/56043> Leaks Viewer should present a list of recent builds to
analyze

Reviewed by David Kilzer and Joe Pecoraro.

* BuildSlaveSupport/build.webkit.org-config/public_html/LeaksViewer/LeaksViewer.css:
(#url-prompt): Made this a bit taller to make room for the list of builds.

(#recent-builds-loading-indicator):
(#recent-builds-list):
Added styles for the build list and loading indicator.

* BuildSlaveSupport/build.webkit.org-config/public_html/LeaksViewer/LeaksViewer.js:
(LeaksViewer.loaded): Pass the URL from the query through decodeURIComponent before loading
it. This will unescape it.
(LeaksViewer.urlPromptButtonClicked): Moved code from here to _urlChosenFromPrompt.
(LeaksViewer._didLoadRecentBuilds): Added. Stores the builds and updates the prompt.
(LeaksViewer._displayURLPrompt): Start loading the recent leaky builds.
(LeaksViewer._updateURLPrompt): Create a list of the recent leaky builds and add them to the
prompt UI.
(LeaksViewer._urlChosenFromPrompt): Added. Code came from urlPromptButtonClicked. This is
now also used when a recent build is clicked.

* BuildSlaveSupport/build.webkit.org-config/public_html/LeaksViewer/RecentBuildsLoader.js: Added.
(RecentBuildsLoader): Store the callback.
(RecentBuildsLoader.prototype.start): Load the N most recent builds from the specified
builder. Find the revision number, leak count, and results URL for each one. Pass the
resulting info to our callback.

* BuildSlaveSupport/build.webkit.org-config/public_html/LeaksViewer/Utilities.js:
(range): Added this helper function to return an array containing the integers [0, n).

* BuildSlaveSupport/build.webkit.org-config/public_html/LeaksViewer/index.html: Added
RecentBuildsLoader.js and added text to the prompt UI to show the recent builds.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@81068 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoFixed typo in the change log entry for r81066.
rniwa@webkit.org [Mon, 14 Mar 2011 22:17:32 +0000 (22:17 +0000)]
Fixed typo in the change log entry for r81066.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@81067 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2011-03-14 Ryosuke Niwa <rniwa@webkit.org>
rniwa@webkit.org [Mon, 14 Mar 2011 22:15:47 +0000 (22:15 +0000)]
2011-03-14  Ryosuke Niwa  <rniwa@webkit.org>

        GTK rebaselines or r81053.

        * platform/gtk/editing/selection/drag-select-1-expected.txt:
        * platform/gtk/editing/selection/fake-drag-expected.txt:
        * platform/gtk/editing/selection/inline-closest-leaf-child-expected.txt:
        * platform/gtk/svg/custom/mouse-move-on-svg-container-expected.txt:
        * platform/gtk/svg/custom/mouse-move-on-svg-container-standalone-expected.txt:
        * platform/gtk/svg/custom/mouse-move-on-svg-root-expected.txt:
        * platform/gtk/svg/custom/mouse-move-on-svg-root-standalone-expected.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@81066 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoFileSystemWin.cpp needs listDirectory() implementation
bweinstein@apple.com [Mon, 14 Mar 2011 22:15:27 +0000 (22:15 +0000)]
FileSystemWin.cpp needs listDirectory() implementation
https://bugs.webkit.org/show_bug.cgi?id=56331
<rdar://problem/9126635>

Reviewed by Adam Roben and Gavin Barraclough.

Source/JavaScriptCore:

Give StringConcatenate the ability to deal with const UChar*'s as a String type to append.

* wtf/text/StringConcatenate.h:

Source/WebCore:

Move PathWalker from an inline class in WebKit2 to its own class in WebCore,
so it can be used from both WebCore and WebKit2.

Implement FileSystemWin::listDirectory using PathWalker to populate the Vector
of paths matching the passed in pattern.

* WebCore.vcproj/WebCore.vcproj:
* platform/win/FileSystemWin.cpp:
(WebCore::listDirectory): Call through to PathWalker.
* platform/win/PathWalker.cpp: Added.
(WebCore::PathWalker::PathWalker): Moved from WebKit2. Added a second argument
    for the pattern to pass to the Windows File APIs.
(WebCore::PathWalker::~PathWalker): Moved from WebKit2.
(WebCore::PathWalker::isValid): Ditto.
(WebCore::PathWalker::data): Ditto.
(WebCore::PathWalker::step): Ditto.
* platform/win/PathWalker.h: Added.

Source/WebKit2:

Move PathWalker down into WebCore, and call the PathWalker in WebCore
(that now has a second argument, the pattern to take).

* UIProcess/Plugins/win/PluginInfoStoreWin.cpp:
(WebKit::PluginInfoStore::pluginPathsInDirectory):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@81065 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoAdd rebaselines due to two changes:
levin@chromium.org [Mon, 14 Mar 2011 22:14:29 +0000 (22:14 +0000)]
Add rebaselines due to two changes:
For r81053, we have
   editing/selection/drag-select-1.html, editing/selection/fake-drag.html, svg/custom/mouse-move-on-svg-container-standalone.svg,
   svg/custom/mouse-move-on-svg-container.xhtml, svg/custom/mouse-move-on-svg-root-standalone.svg, svg/custom/mouse-move-on-svg-root.xhtml

For r81049, we have
   fast/table/overflowHidden.html, fast/table/relative-position-containment.html, fast/table/relative-position-offsets.html, fast/table/relative-position-stacking.html

* platform/chromium-linux/editing/selection/drag-select-1-expected.checksum:
* platform/chromium-linux/editing/selection/drag-select-1-expected.png:
* platform/chromium-linux/editing/selection/drag-select-1-expected.txt: Added.
* platform/chromium-linux/editing/selection/fake-drag-expected.txt: Added.
* platform/chromium-linux/fast/table/relative-position-containment-expected.checksum: Added.
* platform/chromium-linux/fast/table/relative-position-containment-expected.png: Added.
* platform/chromium-linux/fast/table/relative-position-offsets-expected.checksum: Added.
* platform/chromium-linux/fast/table/relative-position-offsets-expected.png: Added.
* platform/chromium-linux/fast/table/relative-position-stacking-expected.checksum: Added.
* platform/chromium-linux/fast/table/relative-position-stacking-expected.png: Added.
* platform/chromium-linux/svg/custom/mouse-move-on-svg-container-expected.checksum:
* platform/chromium-linux/svg/custom/mouse-move-on-svg-container-expected.png:
* platform/chromium-linux/svg/custom/mouse-move-on-svg-container-expected.txt: Added.
* platform/chromium-linux/svg/custom/mouse-move-on-svg-container-standalone-expected.checksum:
* platform/chromium-linux/svg/custom/mouse-move-on-svg-container-standalone-expected.png:
* platform/chromium-linux/svg/custom/mouse-move-on-svg-container-standalone-expected.txt: Added.
* platform/chromium-linux/svg/custom/mouse-move-on-svg-root-expected.checksum:
* platform/chromium-linux/svg/custom/mouse-move-on-svg-root-expected.png:
* platform/chromium-linux/svg/custom/mouse-move-on-svg-root-expected.txt: Added.
* platform/chromium-linux/svg/custom/mouse-move-on-svg-root-standalone-expected.checksum:
* platform/chromium-linux/svg/custom/mouse-move-on-svg-root-standalone-expected.png:
* platform/chromium-linux/svg/custom/mouse-move-on-svg-root-standalone-expected.txt: Added.
* platform/chromium-win/fast/table/overflowHidden-expected.txt:
* platform/chromium-win/fast/table/relative-position-containment-expected.checksum: Added.
* platform/chromium-win/fast/table/relative-position-containment-expected.png: Added.
* platform/chromium-win/fast/table/relative-position-containment-expected.txt: Added.
* platform/chromium-win/fast/table/relative-position-offsets-expected.checksum: Added.
* platform/chromium-win/fast/table/relative-position-offsets-expected.png: Added.
* platform/chromium-win/fast/table/relative-position-offsets-expected.txt: Added.
* platform/chromium-win/fast/table/relative-position-stacking-expected.checksum: Added.
* platform/chromium-win/fast/table/relative-position-stacking-expected.png: Added.
* platform/chromium-win/fast/table/relative-position-stacking-expected.txt: Added.
* platform/chromium/test_expectations.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@81064 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2011-03-14 Ryosuke Niwa <rniwa@webkit.org>
rniwa@webkit.org [Mon, 14 Mar 2011 21:55:48 +0000 (21:55 +0000)]
2011-03-14  Ryosuke Niwa  <rniwa@webkit.org>

        Removed redundant baselines from Windows port so that rebaselines done in r81059 take effect.

        * platform/win/svg/custom/mouse-move-on-svg-container-expected.txt: Removed.
        * platform/win/svg/custom/mouse-move-on-svg-container-standalone-expected.txt: Removed.
        * platform/win/svg/custom/mouse-move-on-svg-root-expected.txt: Removed.
        * platform/win/svg/custom/mouse-move-on-svg-root-standalone-expected.txt: Removed.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@81063 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoAdjusting Snow Leopard expected results after r81035.
luiz@webkit.org [Mon, 14 Mar 2011 21:48:20 +0000 (21:48 +0000)]
Adjusting Snow Leopard expected results after r81035.

* platform/mac/fast/html/details-add-summary-1-and-click-expected.txt: Added.
* platform/mac/fast/html/details-add-summary-1-expected.txt: Added.
* platform/mac/fast/html/details-add-summary-10-and-click-expected.txt: Added.
* platform/mac/fast/html/details-add-summary-10-expected.txt: Added.
* platform/mac/fast/html/details-add-summary-2-and-click-expected.txt: Added.
* platform/mac/fast/html/details-add-summary-2-expected.txt: Added.
* platform/mac/fast/html/details-add-summary-3-and-click-expected.txt: Added.
* platform/mac/fast/html/details-add-summary-3-expected.txt: Added.
* platform/mac/fast/html/details-add-summary-4-and-click-expected.txt: Added.
* platform/mac/fast/html/details-add-summary-4-expected.txt: Added.
* platform/mac/fast/html/details-add-summary-5-and-click-expected.txt: Added.
* platform/mac/fast/html/details-add-summary-5-expected.txt: Added.
* platform/mac/fast/html/details-add-summary-6-and-click-expected.txt: Added.
* platform/mac/fast/html/details-add-summary-6-expected.txt: Added.
* platform/mac/fast/html/details-add-summary-7-and-click-expected.txt: Added.
* platform/mac/fast/html/details-add-summary-7-expected.txt: Added.
* platform/mac/fast/html/details-add-summary-8-and-click-expected.txt: Added.
* platform/mac/fast/html/details-add-summary-8-expected.txt: Added.
* platform/mac/fast/html/details-add-summary-9-and-click-expected.txt: Added.
* platform/mac/fast/html/details-add-summary-9-expected.txt: Added.
* platform/mac/fast/html/details-no-summary1-expected.txt: Added.
* platform/mac/fast/html/details-no-summary2-expected.txt: Added.
* platform/mac/fast/html/details-no-summary3-expected.txt: Added.
* platform/mac/fast/html/details-no-summary4-expected.txt: Added.
* platform/mac/fast/html/details-open-javascript-expected.txt: Added.
* platform/mac/fast/html/details-open1-expected.txt: Added.
* platform/mac/fast/html/details-open2-expected.txt: Added.
* platform/mac/fast/html/details-open3-expected.txt: Added.
* platform/mac/fast/html/details-open4-expected.txt: Added.
* platform/mac/fast/html/details-open5-expected.txt: Added.
* platform/mac/fast/html/details-open6-expected.txt: Added.
* platform/mac/fast/html/details-position-expected.txt: Added.
* platform/mac/fast/html/details-remove-summary-1-and-click-expected.txt: Added.
* platform/mac/fast/html/details-remove-summary-1-expected.txt: Added.
* platform/mac/fast/html/details-remove-summary-2-and-click-expected.txt: Added.
* platform/mac/fast/html/details-remove-summary-2-expected.txt: Added.
* platform/mac/fast/html/details-remove-summary-3-and-click-expected.txt: Added.
* platform/mac/fast/html/details-remove-summary-3-expected.txt: Added.
* platform/mac/fast/html/details-remove-summary-4-and-click-expected.txt: Added.
* platform/mac/fast/html/details-remove-summary-4-expected.txt: Added.
* platform/mac/fast/html/details-remove-summary-5-and-click-expected.txt: Added.
* platform/mac/fast/html/details-remove-summary-5-expected.txt: Added.
* platform/mac/fast/html/details-remove-summary-6-and-click-expected.txt: Added.
* platform/mac/fast/html/details-remove-summary-6-expected.txt: Added.
* platform/mac/fast/html/details-writing-mode-expected.txt: Added.
* platform/mac/fast/multicol/span/clone-anonymous-block-non-inline-child-crash-expected.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@81062 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2011-03-14 Ryosuke Niwa <rniwa@webkit.org>
rniwa@webkit.org [Mon, 14 Mar 2011 21:45:48 +0000 (21:45 +0000)]
2011-03-14  Ryosuke Niwa  <rniwa@webkit.org>

        Qt rebaselines for r81053.

        * platform/qt/editing/selection/drag-select-1-expected.txt:
        * platform/qt/svg/custom/mouse-move-on-svg-container-expected.txt:
        * platform/qt/svg/custom/mouse-move-on-svg-container-standalone-expected.txt:
        * platform/qt/svg/custom/mouse-move-on-svg-root-expected.txt:
        * platform/qt/svg/custom/mouse-move-on-svg-root-standalone-expected.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@81061 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoTry to fix the EFL build-bustage that their EWS bot did not see, but their core-build...
beidson@apple.com [Mon, 14 Mar 2011 21:44:00 +0000 (21:44 +0000)]
Try to fix the EFL build-bustage that their EWS bot did not see, but their core-builder did.  (sigh)

* ewk/ewk_history.cpp:
(ewk_history_item_icon_surface_get):
(ewk_history_item_icon_object_add):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@81060 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2011-03-14 Ryosuke Niwa <rniwa@webkit.org>
rniwa@webkit.org [Mon, 14 Mar 2011 21:35:42 +0000 (21:35 +0000)]
2011-03-14  Ryosuke Niwa  <rniwa@webkit.org>

        Mac rebaselines for r81053.

        * platform/mac/fast/text/atsui-rtl-override-selection-expected.txt:
        * platform/mac/svg/custom/mouse-move-on-svg-container-expected.txt:
        * platform/mac/svg/custom/mouse-move-on-svg-container-standalone-expected.txt:
        * platform/mac/svg/custom/mouse-move-on-svg-root-expected.txt:
        * platform/mac/svg/custom/mouse-move-on-svg-root-standalone-expected.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@81059 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agohttps://bugs.webkit.org/show_bug.cgi?id=56320
beidson@apple.com [Mon, 14 Mar 2011 21:25:51 +0000 (21:25 +0000)]
https://bugs.webkit.org/show_bug.cgi?id=56320
Remove HistoryItem::icon() and the WebCore dependency on "IconDatabaseBase::defaultIcon()"

Reviewed by Anders Carlsson.

../WebCore:

Remove HistoryItem::icon():
* history/HistoryItem.cpp:
* history/HistoryItem.h:
* WebCore.exp.in:

* loader/icon/IconDatabaseBase.h:
(WebCore::IconDatabaseBase::defaultIcon):

../WebKit/efl:

* ewk/ewk_history.cpp:
(ewk_history_item_icon_surface_get): Use IconDatabase directly.

../WebKit/mac:

* History/WebHistoryItem.mm:
(-[WebHistoryItem icon]): This far-future FIXME is now irrelevant.

../WebKit/qt:

* Api/qwebhistory.cpp:
(QWebHistoryItem::icon): Use IconDatabase directly.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@81058 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2011-03-13 Andy Estes <aestes@apple.com>
aestes@apple.com [Mon, 14 Mar 2011 21:22:32 +0000 (21:22 +0000)]
2011-03-13  Andy Estes  <aestes@apple.com>

        Reviewed by Darin Adler.

        Timer-based events should inherit the user gesture state of their
        originating event in certain cases.
        https://bugs.webkit.org/show_bug.cgi?id=55104

        * fast/events/popup-blocking-timers-expected.txt: Added.
        * fast/events/popup-blocking-timers.html: Added.
2011-03-14  Andy Estes  <aestes@apple.com>

        Reviewed by Darin Adler.

        Timer-based events should inherit the user gesture state of their
        originating event in certain cases.
        https://bugs.webkit.org/show_bug.cgi?id=55104

        If a timer is installed by a gesture-originated event and will fire
        within one second, the timer-initiated event should behave as if it
        were also initiated by a user gesture. Multi-shot timers should only
        get this behavior on their first execution. Nested timers should not
        get this behavior. This makes us compatible with Gecko when handling
        popups and file chooser dialogs created from timer events.

        Test: fast/events/popup-blocking-timers.html

        * page/DOMTimer.cpp:
        (WebCore::timeoutId): Create a helper function so that m_timeoutId can
        be initialized in the data member initialization list.
        (WebCore::shouldForwardUserGesture): Ditto, but for
        m_shouldForwardUserGesture.
        (WebCore::DOMTimer::DOMTimer): Move initialization of data members from
        the ctor body to the data member initialization list. Also rename the
        argument 'timeout' to 'interval'.
        (WebCore::DOMTimer::fired): Create a UserGestureIndicator and set its
        state based on the value of m_shouldForwardUserGesture.
        (WebCore::DOMTimer::adjustMinimumTimerInterval): m_originalTimeout was
        renamed to m_originalInterval.
        * page/DOMTimer.h: Add m_shouldForwardUserGesture and rename
        m_originalTimeout to m_originalInterval.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@81057 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2011-03-09 Levi Weintraub <leviw@chromium.org>
leviw@chromium.org [Mon, 14 Mar 2011 21:19:29 +0000 (21:19 +0000)]
2011-03-09  Levi Weintraub  <leviw@chromium.org>

        Reviewed by Ryosuke Niwa.

        Deleting content directly following a button inserts an unnecessary placeholder
        https://bugs.webkit.org/show_bug.cgi?id=56053

        Test to ensure unnecessary placeholders aren't inserted after buttons when delete is pressed.

        * editing/deleting/delete-inserts-br-after-button-expected.txt: Added.
        * editing/deleting/delete-inserts-br-after-button.html: Added.
2011-03-09  Levi Weintraub  <leviw@chromium.org>

        Reviewed by Ryosuke Niwa.

        Deleting content directly following a button inserts an unnecessary placeholder
        https://bugs.webkit.org/show_bug.cgi?id=56053

        Fixing a use of Node's enclosingBlockFlowElement with enclosingBlock htmlediting's
        enclosingBlock, as enclosingBlockFlowElement would return inline-block elements despite
        DeleteSelectionCommand treating them as blockflow.

        Test: editing/deleting/delete-inserts-br-after-button.html

        * editing/DeleteSelectionCommand.cpp:
        (WebCore::DeleteSelectionCommand::mergeParagraphs):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@81056 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agohttps://bugs.webkit.org/show_bug.cgi?id=45164
hyatt@apple.com [Mon, 14 Mar 2011 20:56:18 +0000 (20:56 +0000)]
https://bugs.webkit.org/show_bug.cgi?id=45164

Reviewed by Dan Bernstein.

REGRESSION: <a><img align=top></a> Clickable area too large

Make sure to clamp hit testing of quirky inline flow boxes the same way we already clamped
painting.

Source/WebCore:

* rendering/InlineFlowBox.cpp:
(WebCore::InlineFlowBox::nodeAtPoint):

LayoutTests:

* fast/inline/inline-position-top-align-expected.txt: Added.
* fast/inline/inline-position-top-align.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@81055 268f45cc-cd09-0410-ab3c-d52691b4dbfc