WebKit-https.git
10 years agoRemove some dead Autofill code
isherman@chromium.org [Wed, 24 Aug 2011 23:01:55 +0000 (23:01 +0000)]
Remove some dead Autofill code
https://bugs.webkit.org/show_bug.cgi?id=66899
http://code.google.com/p/chromium/issues/detail?id=51644

Reviewed by James Robinson.

* WebKit.gyp:
* public/WebAutofillClient.h:
* public/WebFrame.h:
* public/WebPasswordAutocompleteListener.h: Removed.
* src/EditorClientImpl.cpp:
(WebKit::EditorClientImpl::EditorClientImpl):
(WebKit::EditorClientImpl::handleKeyboardEvent):
(WebKit::EditorClientImpl::textFieldDidBeginEditing):
(WebKit::EditorClientImpl::textFieldDidEndEditing):
(WebKit::EditorClientImpl::textDidChangeInTextField):
(WebKit::EditorClientImpl::doTextFieldCommandFromEvent):
* src/EditorClientImpl.h:
* src/FrameLoaderClientImpl.cpp:
(WebKit::FrameLoaderClientImpl::dispatchDidFinishDocumentLoad):
* src/WebFrameImpl.cpp:
(WebKit::frameContentAsPlainText):
(WebKit::WebFrameImpl::scopeStringMatches):
(WebKit::WebFrameImpl::~WebFrameImpl):
* src/WebFrameImpl.h:
* src/WebViewImpl.cpp:
(WebKit::WebViewImpl::mouseDown):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@93747 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago[EFL] Fix build with ENABLE_GEOLOCATION.
paroga@webkit.org [Wed, 24 Aug 2011 22:44:46 +0000 (22:44 +0000)]
[EFL] Fix build with ENABLE_GEOLOCATION.
https://bugs.webkit.org/show_bug.cgi?id=66881

Use adoptPtr() to create a new GeolocationService instance, otherwise
the build will fail when creating the PassOwnPtr object.

Patch by Raphael Kubo da Costa <kubo@profusion.mobi> on 2011-08-24

* platform/efl/GeolocationServiceEfl.cpp:
(WebCore::GeolocationServiceEfl::create):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@93746 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoXSS filter bypass via document.write(location.href) and fragments
dbates@webkit.org [Wed, 24 Aug 2011 22:35:52 +0000 (22:35 +0000)]
XSS filter bypass via document.write(location.href) and fragments
https://bugs.webkit.org/show_bug.cgi?id=66585

Reviewed by Darin Adler.

Add a test case for a variation of a DOM-based XSS attack using an anchor URL.

* http/tests/security/xssAuditor/anchor-url-dom-write-location2-expected.txt: Added.
* http/tests/security/xssAuditor/anchor-url-dom-write-location2.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@93745 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago[CMake] Fix build with ENABLE_BLOB or ENABLE_FILE_SYSTEM.
paroga@webkit.org [Wed, 24 Aug 2011 22:28:50 +0000 (22:28 +0000)]
[CMake] Fix build with ENABLE_BLOB or ENABLE_FILE_SYSTEM.
https://bugs.webkit.org/show_bug.cgi?id=66880

Generate and build some additional files required by these options.

No new tests, as this is a build fix for some build options.

Patch by Raphael Kubo da Costa <kubo@profusion.mobi> on 2011-08-24

* CMakeLists.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@93744 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago[chromium] Don't call glDeleteTexture(0) in TextureManager
commit-queue@webkit.org [Wed, 24 Aug 2011 22:09:57 +0000 (22:09 +0000)]
[chromium] Don't call glDeleteTexture(0) in TextureManager
https://bugs.webkit.org/show_bug.cgi?id=66862

Delete(0) is allowed in standard OpenGL, but not in Chrome.
See http://code.google.com/p/chromium/issues/detail?id=85268

Patch by Iain Merrick <husky@google.com> on 2011-08-24
Reviewed by James Robinson.

* platform/graphics/chromium/TextureManager.cpp:
(WebCore::TextureManager::deleteEvictedTextures):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@93743 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago[chromium] Update expectations for tests that no longer crash.
jamesr@google.com [Wed, 24 Aug 2011 22:07:35 +0000 (22:07 +0000)]
[chromium] Update expectations for tests that no longer crash.

* platform/chromium/test_expectations.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@93742 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoRemove empty directory.
abarth@webkit.org [Wed, 24 Aug 2011 22:05:34 +0000 (22:05 +0000)]
Remove empty directory.

* BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/ui: Removed.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@93741 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago[chromium] Properly initialize CCTiledLayerImpl::m_skipsDraw
enne@google.com [Wed, 24 Aug 2011 21:56:11 +0000 (21:56 +0000)]
[chromium] Properly initialize CCTiledLayerImpl::m_skipsDraw
https://bugs.webkit.org/show_bug.cgi?id=66889

Reviewed by James Robinson.

* platform/graphics/chromium/cc/CCTiledLayerImpl.cpp:
(WebCore::CCTiledLayerImpl::CCTiledLayerImpl):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@93740 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoFinish writing up details-view in garden-o-matic
abarth@webkit.org [Wed, 24 Aug 2011 21:33:38 +0000 (21:33 +0000)]
Finish writing up details-view in garden-o-matic
https://bugs.webkit.org/show_bug.cgi?id=66891

Reviewed by Dimitri Glazkov.

This patch mostly just copies a bunch of CSS from main.css to
summary.css.  Presumably we'll delete main.css soon.

* BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/controllers.js:
* BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui/actions.js:
* BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui/actions_unittests.js:
* BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/styles/summary.css:
* BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/summary.html:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@93738 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoSome compositing LayoutTests flakily crashing on Mac 10.6
jamesr@google.com [Wed, 24 Aug 2011 21:27:25 +0000 (21:27 +0000)]
Some compositing LayoutTests flakily crashing on Mac 10.6
https://bugs.webkit.org/show_bug.cgi?id=66740

Reviewed by Adam Barth.

Tools:

Null-check obj->pluginTest before dereferencing it. This can be null in chromium if the plugin load fails, for
example if the test attempts to load a PDF plugin the load will fail since we don't have such a thing in the
chromium configuration.

* DumpRenderTree/TestNetscapePlugIn/main.cpp:
(NPP_Destroy):

LayoutTests:

Remove = CRASH expectations for plugin tests that should no longer crash.

* platform/chromium/test_expectations.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@93737 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoNo review necessary.
adauria@apple.com [Wed, 24 Aug 2011 21:20:13 +0000 (21:20 +0000)]
No review necessary.

Adding myself to list of committers.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@93735 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoSource/WebCore: chromium WebFrameImpl - don't load javascript URLs against chrome...
commit-queue@webkit.org [Wed, 24 Aug 2011 21:06:43 +0000 (21:06 +0000)]
Source/WebCore: chromium WebFrameImpl - don't load javascript URLs against chrome internal
pages.  Adds a mechanism for registering sensitive schemes which should
not be subject to manipulation by javascript urls typed into a location bar
or, more importantly, their bookmarklet equivalents.
https://bugs.webkit.org/show_bug.cgi?id=66720

Patch by Tom Sepez <tsepez@chromium.org> on 2011-08-24
Reviewed by Adam Barth.

Test: Chromium WebKit API unit test in chromium specific directory.

* platform/SchemeRegistry.cpp:
(WebCore::notAllowingJavascriptURLsSchemes):
(WebCore::SchemeRegistry::registerURLSchemeAsNotAllowingJavascriptURLs):
(WebCore::SchemeRegistry::shouldTreatURLSchemeAsNotAllowingJavascriptURLs):
* platform/SchemeRegistry.h:

Source/WebKit/chromium: chromium WebFrameImpl - don't load javascript URLs against chrome internal pages.
https://bugs.webkit.org/show_bug.cgi?id=66720

Patch by Tom Sepez <tsepez@chromium.org> on 2011-08-24
Reviewed by Adam Barth.

* public/WebSecurityPolicy.h:
* src/WebFrameImpl.cpp:
(WebKit::WebFrameImpl::loadJavaScriptURL):
* src/WebSecurityPolicy.cpp:
(WebKit::WebSecurityPolicy::registerURLSchemeAsNotAllowingJavascriptURLs):
* tests/WebFrameTest.cpp:
(WebKit::WebFrameTest::WebFrameTest):
(WebKit::WebFrameTest::registerMockedHttpURLLoad):
(WebKit::WebFrameTest::registerMockedChromeURLLoad):
(WebKit::WebFrameTest::serveRequests):
(WebKit::WebFrameTest::loadHttpFrame):
(WebKit::WebFrameTest::loadChromeFrame):
(WebKit::WebFrameTest::registerMockedURLLoad):
(WebKit::WebFrameTest::loadFrame):
(WebKit::TEST_F):
* tests/data/history.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@93734 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago[chromium] More test expectation updates.
pkasting@chromium.org [Wed, 24 Aug 2011 20:39:20 +0000 (20:39 +0000)]
[chromium] More test expectation updates.

Unreviewed, test expectations update.

* platform/chromium/test_expectations.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@93733 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago Plug-in tests are flaky on Mac
ap@apple.com [Wed, 24 Aug 2011 20:36:15 +0000 (20:36 +0000)]
    Plug-in tests are flaky on Mac
        https://bugs.webkit.org/show_bug.cgi?id=66822

        Reviewed by Anders Carlsson.

        * Plugins/Hosted/NetscapePluginInstanceProxy.mm:
        (WebKit::NetscapePluginInstanceProxy::evaluateJavaScript):
        Protect objects that can be removed by a script.

        * WebView/WebFrame.mm:
        (-[WebFrame _stringByEvaluatingJavaScriptFromString:forceUserGesture:]): Ditto.
        (-[WebFrame _stringByEvaluatingJavaScriptFromString:withGlobalObject:inScriptWorld:]): Ditto.
        Also fixed some nonsense code.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@93732 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago[CSSRegions] Fix Element::getBoundingClientRect and Element::getClientRects for conte...
commit-queue@webkit.org [Wed, 24 Aug 2011 20:20:57 +0000 (20:20 +0000)]
[CSSRegions] Fix Element::getBoundingClientRect and Element::getClientRects for content flow
https://bugs.webkit.org/show_bug.cgi?id=66641

Original author of the patch is Mihnea Ovidenie <mihnea@adobe.com>.

Source/WebCore:

Redirected the mapLocalToContainer method to the appropriate RenderRegion.

Note: RenderObject::absoluteQuads method needs some refactor to support multiple regions,
will do that in a separate bug.

Patch by Alexandru Chiculita <achicu@adobe.com> on 2011-08-24
Reviewed by David Hyatt.

Tests: fast/regions/flowed-content-bounding-client-rect-horizontal-bt.html
       fast/regions/flowed-content-bounding-client-rect-vertical-rl.html
       fast/regions/flowed-content-bounding-client-rect-vertical.html
       fast/regions/flowed-content-bounding-client-rect.html
       fast/regions/flowed-content-transform-bounding-client-rect.html

* rendering/RenderBox.cpp:
(WebCore::RenderBox::mapLocalToContainer):
* rendering/RenderFlowThread.cpp:
(WebCore::RenderFlowThread::renderRegionForLine):
(WebCore::RenderFlowThread::mapFromFlowToRegion):
* rendering/RenderFlowThread.h:

LayoutTests:

Patch by Alexandru Chiculita <achicu@adobe.com> on 2011-08-24
Reviewed by David Hyatt.

* fast/regions/flowed-content-bounding-client-rect-expected.txt: Added.
* fast/regions/flowed-content-bounding-client-rect-horizontal-bt-expected.txt: Added.
* fast/regions/flowed-content-bounding-client-rect-horizontal-bt.html: Added.
* fast/regions/flowed-content-bounding-client-rect-vertical-expected.txt: Added.
* fast/regions/flowed-content-bounding-client-rect-vertical-rl-expected.txt: Added.
* fast/regions/flowed-content-bounding-client-rect-vertical-rl.html: Added.
* fast/regions/flowed-content-bounding-client-rect-vertical.html: Added.
* fast/regions/flowed-content-bounding-client-rect.html: Added.
* fast/regions/flowed-content-transform-bounding-client-rect-expected.txt: Added.
* fast/regions/flowed-content-transform-bounding-client-rect.html: Added.
* fast/regions/resources/helper.js: Added.
(rectToArray):
(areEqualNumbers):
(areEqualRects):
(assertEqualRects):
(testBoundingRects):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@93728 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago[chromium] Root layer is not updated when only a portion of tile is updated
alokp@chromium.org [Wed, 24 Aug 2011 20:17:43 +0000 (20:17 +0000)]
[chromium] Root layer is not updated when only a portion of tile is updated
https://bugs.webkit.org/show_bug.cgi?id=66506

This only happened in the accelerated path which changes stencil state.
Incorrect stencil state was used to draw the root layer.

Reviewed by James Robinson.

Test: compositing/overflow/overflow-scroll.html (existing)

* platform/graphics/chromium/LayerRendererChromium.cpp:
(WebCore::LayerRendererChromium::drawLayersInternal):
* platform/graphics/chromium/LayerTextureUpdaterCanvas.cpp:
(WebCore::LayerTextureUpdaterSkPicture::updateTextureRect):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@93727 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2011-08-24 Nate Chapin <japhet@chromium.org>
japhet@chromium.org [Wed, 24 Aug 2011 20:16:45 +0000 (20:16 +0000)]
2011-08-24  Nate Chapin  <japhet@chromium.org>

        Fix an intermittent crash in EventSource, when a
        reconnect timer triggers after a navigation
        away from the current page begins.
        https://bugs.webkit.org/show_bug.cgi?id=45202

        Reviewed by Alexey Proskuryakov.

        Test: http/tests/eventsource/eventsource-reconnect-during-navigate-crash.html

        * page/EventSource.cpp:
        (WebCore::EventSource::connect): Don't set m_requestInFlight
            to true if ThreadableLoader::create() failed.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@93726 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoChange clip routine to promote the clipbounds up to (local) floats and then perform...
reed@google.com [Wed, 24 Aug 2011 20:08:51 +0000 (20:08 +0000)]
Change clip routine to promote the clipbounds up to (local) floats and then perform the intersection,
rather than rounding the srcRect down to integers.
https://bugs.webkit.org/show_bug.cgi?id=66810

Reviewed by Kenneth Russell.

No new tests. Existing tests that resize images will exercise this code path

* platform/graphics/skia/SkiaUtils.cpp:
(WebCore::ClipRectToCanvas):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@93725 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoCrash when restoring NSGraphicsContext when running the Chromium GPU layout tests...
commit-queue@webkit.org [Wed, 24 Aug 2011 20:04:44 +0000 (20:04 +0000)]
Crash when restoring NSGraphicsContext when running the Chromium GPU layout tests on Mac
https://bugs.webkit.org/show_bug.cgi?id=66875

Patch by Avi Drissman <avi@google.com> on 2011-08-24
Reviewed by James Robinson.

* DumpRenderTree/chromium/WebThemeEngineDRTMac.mm:
(WebThemeEngineDRTMac::paintNSScrollerScrollbarThumb):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@93724 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoRename -[WKTextInputWindowController keyboardInputSourceChanged] to unmarkText to...
andersca@apple.com [Wed, 24 Aug 2011 20:02:37 +0000 (20:02 +0000)]
Rename -[WKTextInputWindowController keyboardInputSourceChanged] to unmarkText to indicate what the method does
https://bugs.webkit.org/show_bug.cgi?id=66883

Reviewed by Beth Dakin.

* UIProcess/API/mac/WKTextInputWindowController.h:
* UIProcess/API/mac/WKTextInputWindowController.mm:
(-[WKTextInputPanel _unmarkText]):
(-[WKTextInputWindowController unmarkText]):
* UIProcess/API/mac/WKView.mm:
(-[WKView _pluginFocusOrWindowFocusChanged:pluginComplexTextInputIdentifier:]):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@93723 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago[chromium] Test expectation updates.
pkasting@chromium.org [Wed, 24 Aug 2011 19:56:41 +0000 (19:56 +0000)]
[chromium] Test expectation updates.

Unreviewed, test expectations update.

* platform/chromium/test_expectations.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@93722 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago[CSSRegions] RenderRegion is not used if there's another renderer after it
commit-queue@webkit.org [Wed, 24 Aug 2011 19:53:32 +0000 (19:53 +0000)]
[CSSRegions] RenderRegion is not used if there's another renderer after it
https://bugs.webkit.org/show_bug.cgi?id=66844

Source/WebCore:

There was no call to attachRegion in RenderObjectChildList::insertChildNode, so
the region was not added to the RenderFlowThread.

Patch by Alexandru Chiculita <achicu@adobe.com> on 2011-08-24
Reviewed by David Hyatt.

Test: fast/regions/content-flowed-into-regions-dynamically-inserted.html

* rendering/RenderObjectChildList.cpp:
(WebCore::RenderObjectChildList::insertChildNode):

LayoutTests:

Patch by Alexandru Chiculita <achicu@adobe.com> on 2011-08-24
Reviewed by David Hyatt.

* fast/regions/content-flowed-into-regions-dynamically-inserted-expected.png: Added.
* fast/regions/content-flowed-into-regions-dynamically-inserted-expected.txt: Added.
* fast/regions/content-flowed-into-regions-dynamically-inserted.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@93721 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoJSNPObject and JSNPMethod create their structure in their constructors
oliver@apple.com [Wed, 24 Aug 2011 19:34:03 +0000 (19:34 +0000)]
JSNPObject and JSNPMethod create their structure in their constructors
https://bugs.webkit.org/show_bug.cgi?id=66879

Reviewed by Anders Carlsson.

It's not safe to create the Structure for an object inside its constructor
so we hoist construction out into their ::create methods and move the methods
into the cpp file.

* WebProcess/Plugins/Netscape/JSNPMethod.cpp:
(WebKit::JSNPMethod::JSNPMethod):
(WebKit::JSNPMethod::create):
* WebProcess/Plugins/Netscape/JSNPMethod.h:
(WebKit::JSNPMethod::create):
* WebProcess/Plugins/Netscape/JSNPObject.cpp:
(WebKit::JSNPObject::JSNPObject):
(WebKit::JSNPObject::create):
* WebProcess/Plugins/Netscape/JSNPObject.h:
(WebKit::JSNPObject::create):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@93720 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoUpdate SVN properties to help out the Windows EWS bots.
lforschler@apple.com [Wed, 24 Aug 2011 19:32:01 +0000 (19:32 +0000)]
Update SVN properties to help out the Windows EWS bots.

Reviewed by Adam Roben.

* win: Modified properties svn:ignore.
* win/include: Modified property svn:ignore.
* win/lib: Modified property svn:ignore.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@93719 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago[chromium] Organize SLOW section of expectations.
pkasting@chromium.org [Wed, 24 Aug 2011 19:06:10 +0000 (19:06 +0000)]
[chromium] Organize SLOW section of expectations.

Unreviewed, test expectations update.

* platform/chromium/test_expectations.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@93718 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoJSHTMLImageElement (and associated Node) is abandoned when image load is canceled
weinig@apple.com [Wed, 24 Aug 2011 18:28:12 +0000 (18:28 +0000)]
JSHTMLImageElement (and associated Node) is abandoned when image load is canceled
<rdar://problem/9925630>
https://bugs.webkit.org/show_bug.cgi?id=66864

Reviewed by Anders Carlsson.

In the JSC bindings, we use HTMLImageElement::hasPendingActivity() (which in turn calls
HTMLImageLoader::haveFiredLoadEvent()) to reason about the liveness of JSHTMLImageElements.
In the case that an image load is canceled, the haveFiredLoadEvent never true, so the
JSHTMLImageElement is kept alive forever (and since it references the global object, it
keeps the entire graph alive as well).

* loader/ImageLoader.cpp:
(WebCore::ImageLoader::notifyFinished):
Set m_firedLoad to true in the case of a canceled load, to reset the state back to
its initial values.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@93717 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2011-08-24 Andrew Scherkus <scherkus@chromium.org>
scherkus@chromium.org [Wed, 24 Aug 2011 18:23:42 +0000 (18:23 +0000)]
2011-08-24  Andrew Scherkus  <scherkus@chromium.org>

        Unreviewed, rebaselining media/video-currentTime-set.html due to
        FFmpeg udpate.

        * platform/chromium-cg-mac/media/video-currentTime-set-expected.txt:
        * platform/chromium-win/media/video-currentTime-set-expected.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@93715 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoExpose fullscreen API on WebElement/WebDocument
commit-queue@webkit.org [Wed, 24 Aug 2011 18:09:08 +0000 (18:09 +0000)]
Expose fullscreen API on WebElement/WebDocument
https://bugs.webkit.org/show_bug.cgi?id=66746

Patch by Jeremy Apthorp <jeremya@google.com> on 2011-08-24
Reviewed by Darin Fisher.

* public/WebPluginContainer.h:
* src/WebPluginContainerImpl.cpp:
(WebKit::WebPluginContainerImpl::setFullscreen):
* src/WebPluginContainerImpl.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@93714 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoMediaStream API: add createObjectURL functionality
commit-queue@webkit.org [Wed, 24 Aug 2011 18:07:43 +0000 (18:07 +0000)]
MediaStream API: add createObjectURL functionality
https://bugs.webkit.org/show_bug.cgi?id=65925

Source/WebCore:

This patch introduces the functionality for generating the url associated with a MediaStream,
which can then be assigned to the src attribute of the <video> tag for example.

Patch by Tommy Widenflycht <tommyw@google.com> on 2011-08-24
Reviewed by Adam Barth.

* CMakeLists.txt:
* GNUmakefile.list.am:
* WebCore.gypi:
* WebCore.pro:
* WebCore.vcproj/WebCore.vcproj:
* WebCore.xcodeproj/project.pbxproj:
* dom/ScriptExecutionContext.cpp:
(WebCore::ScriptExecutionContext::~ScriptExecutionContext):
(WebCore::ScriptExecutionContext::createPublicBlobURL):
(WebCore::ScriptExecutionContext::revokePublicBlobURL):
* dom/ScriptExecutionContext.h:
* html/DOMURL.cpp:
(WebCore::DOMURL::createObjectURL):
* html/DOMURL.h:
* html/DOMURL.idl:
* platform/MediaStreamRegistry.cpp: Added.
(WebCore::MediaStreamRegistry::registry):
(WebCore::MediaStreamRegistry::registerMediaStreamURL):
(WebCore::MediaStreamRegistry::unregisterMediaStreamURL):
(WebCore::MediaStreamRegistry::mediaStream):
* platform/MediaStreamRegistry.h: Copied from Source/WebCore/html/DOMURL.h.

Tests for the Media Stream API will be provided by the bug 56587, pending enough landed code.

LayoutTests:

Patch by Tommy Widenflycht <tommyw@google.com> on 2011-08-24
Reviewed by Adam Barth.

* fast/files/create-blob-url-crash-expected.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@93713 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoMore plug-in complex text input scaffolding
andersca@apple.com [Wed, 24 Aug 2011 18:00:31 +0000 (18:00 +0000)]
More plug-in complex text input scaffolding
https://bugs.webkit.org/show_bug.cgi?id=66865

Reviewed by Sam Weinig.

Make -[WKView _doneWithKeyEvent:eventWasHandled:] take an NSEvent * directly. Also
add methods for handling plug-in text input key-down events.

* UIProcess/API/mac/PageClientImpl.mm:
(WebKit::PageClientImpl::doneWithKeyEvent):
* UIProcess/API/mac/WKView.mm:
(-[WKView _tryHandlePluginComplexTextInputKeyDown:]):
(-[WKView keyDown:]):
(-[WKView _tryPostProcessPluginComplexTextInputKeyDown:]):
(-[WKView _doneWithKeyEvent:eventWasHandled:]):
* UIProcess/API/mac/WKViewInternal.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@93712 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago[GTK] WebProcess crash on hitting assertion m_isWaitingForDidUpdate in DrawingAreaImpl
commit-queue@webkit.org [Wed, 24 Aug 2011 17:55:44 +0000 (17:55 +0000)]
[GTK] WebProcess crash on hitting assertion m_isWaitingForDidUpdate in DrawingAreaImpl
https://bugs.webkit.org/show_bug.cgi?id=66640

Patch by Amruth Raj <amruthraj@motorola.com> on 2011-08-24
Reviewed by Martin Robinson.

The crash happens when a new timer is created in an existing timer callback. Since
TimerBase has a single GSource* it gets overridden and the old timer can never get
stopped. To handle this, keep a copy of the old timer and use it for destruction.

* Platform/RunLoop.h:
* Platform/gtk/RunLoopGtk.cpp:
(RunLoop::TimerBase::timerFiredCallback):
Store the timerSource so that even if it is modified by the timer callback,
it can be restored for freeing it.
(RunLoop::TimerBase::start):
Remove destroyNotifyCallback. Destruction is now handled in timerFiredCallback itself.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@93711 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoErrorInstance::create sometimes has two heap object constructions
fpizlo@apple.com [Wed, 24 Aug 2011 17:44:35 +0000 (17:44 +0000)]
ErrorInstance::create sometimes has two heap object constructions
in flight at once
https://bugs.webkit.org/show_bug.cgi?id=66845

Reviewed by Darin Adler.

The fix is simple since there is already a second create() method
that takes a UString.

* runtime/ErrorInstance.cpp:
(JSC::ErrorInstance::create):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@93710 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoMessageEvent.ports shouldn't ever be null.
levin@chromium.org [Wed, 24 Aug 2011 17:42:20 +0000 (17:42 +0000)]
MessageEvent.ports shouldn't ever be null.
https://bugs.webkit.org/show_bug.cgi?id=66789

Reviewed by Darin Adler.

Source/WebCore:

* bindings/js/JSMessageEventCustom.cpp:
(WebCore::JSMessageEvent::ports): Never return null.
* bindings/v8/custom/V8MessageEventCustom.cpp:
(WebCore::V8MessageEvent::portsAccessorGetter): Ditto.

LayoutTests:

Fix all tests which verified that MessageEvents.ports was null in various cases.

* fast/dom/Window/window-postmessage-args-expected.txt:
* fast/events/message-port-multi-expected.txt:
* fast/events/resources/message-port-multi.js:
(channel.port2.onmessage):
* fast/workers/resources/worker-context-multi-port.js:
(worker.onmessage):
* fast/workers/resources/worker-messageport.js:
(onmessage):
* fast/workers/resources/worker-thread-multi-port.js:
(onmessage):
* fast/workers/worker-context-multi-port-expected.txt:
* fast/workers/worker-messageport-expected.txt:
* fast/workers/worker-multi-port-expected.txt:
* platform/chromium-win/fast/events/message-port-multi-expected.txt:
* platform/chromium-win/fast/workers/worker-context-multi-port-expected.txt:
* platform/chromium-win/fast/workers/worker-multi-port-expected.txt:
* platform/chromium/fast/dom/Window/window-postmessage-args-expected.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@93709 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agohttps://bugs.webkit.org/show_bug.cgi?id=66799
commit-queue@webkit.org [Wed, 24 Aug 2011 17:30:39 +0000 (17:30 +0000)]
https://bugs.webkit.org/show_bug.cgi?id=66799
RenderFlowThread creates a list of child elements, so that it can search
the insertion point of a new element without searching the whole DOM.
The issue is that it used addChild/removeChild, but that wasn't taking care of
elements added in anonymous blocks created inside the RenderFlowThread.

Patch by Alexandru Chiculita <achicu@adobe.com> on 2011-08-24
Reviewed by David Hyatt.

Source/WebCore:

Test: fast/regions/flow-anonymous-children.html

* rendering/RenderFlowThread.cpp:
(WebCore::RenderFlowThread::nextRendererForNode):
(WebCore::RenderFlowThread::previousRendererForNode):
(WebCore::RenderFlowThread::addFlowChild):
(WebCore::RenderFlowThread::removeFlowChild):
* rendering/RenderFlowThread.h:
* rendering/RenderObjectChildList.cpp:
(WebCore::renderFlowThreadContainer):
(WebCore::RenderObjectChildList::removeChildNode):
(WebCore::RenderObjectChildList::appendChildNode):
(WebCore::RenderObjectChildList::insertChildNode):

LayoutTests:

* fast/regions/flow-anonymous-children-expected.txt: Added.
* fast/regions/flow-anonymous-children.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@93708 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago[SKIA] Allow WebKitGraphicsContext3D implementation to provide a compatible GrGLInterface
commit-queue@webkit.org [Wed, 24 Aug 2011 15:04:32 +0000 (15:04 +0000)]
[SKIA] Allow WebKitGraphicsContext3D implementation to provide a compatible GrGLInterface
https://bugs.webkit.org/show_bug.cgi?id=66809

Patch by Brian Salomon <bsalomon@google.com> on 2011-08-24
Reviewed by Stephen White.

* public/WebGraphicsContext3D.h:
(WebKit::WebGraphicsContext3D::grGLInterface):
* src/GraphicsContext3DChromium.cpp:
(WebCore::GraphicsContext3DInternal::grContext):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@93707 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoWeb Inspector: Add 9 more javascript files to the compilation.
pfeldman@chromium.org [Wed, 24 Aug 2011 13:35:07 +0000 (13:35 +0000)]
Web Inspector: Add 9 more javascript files to the compilation.
https://bugs.webkit.org/show_bug.cgi?id=66847

Reviewed by Yury Semikhatsky.

* WebCore.gypi:
* WebCore.vcproj/WebCore.vcproj:
* inspector/compile-front-end.sh:
* inspector/front-end/BreakpointManager.js:
* inspector/front-end/CSSStyleModel.js:
* inspector/front-end/Checkbox.js:
* inspector/front-end/ConsoleMessage.js: Added.
(WebInspector.ConsoleStackFrame):
(WebInspector.ConsoleMessage):
(WebInspector.ConsoleMessage.createTextMessage):
(WebInspector.ConsoleMessage.prototype.get stackTrace):
* inspector/front-end/ConsoleView.js:
(WebInspector.ConsoleView.prototype._registerConsoleDomainDispatcher.dispatcher.messageAdded):
(WebInspector.ConsoleView.prototype._registerConsoleDomainDispatcher.dispatcher.messageRepeatCountUpdated):
* inspector/front-end/CookieParser.js:
* inspector/front-end/DOMAgent.js:
* inspector/front-end/DOMStorage.js:
* inspector/front-end/Database.js:
* inspector/front-end/DebuggerModel.js:
* inspector/front-end/DebuggerPresentationModel.js:
* inspector/front-end/InspectorFrontendHostStub.js:
(WebInspector.isMac):
* inspector/front-end/NetworkManager.js:
* inspector/front-end/RemoteObject.js:
(WebInspector.RemoteObject.prototype.setPropertyValue):
* inspector/front-end/Resource.js:
(WebInspector.Resource):
(WebInspector.Resource.Type.Document.0.Stylesheet.1.Image.2.Font.3.Script.4.XHR.5.WebSocket.7.Other.8.isTextType):
(WebInspector.Resource.Type.toUIString):
(WebInspector.Resource.Type.toString):
(WebInspector.Resource.prototype._checkWarnings):
(WebInspector.ResourceDomainModelBinding):
(WebInspector.ResourceDomainModelBinding.prototype.canSetContent):
(WebInspector.ResourceDomainModelBinding.prototype.setContent):
* inspector/front-end/ResourceCategory.js:
(WebInspector.ResourceCategory):
(WebInspector.ResourceCategory.prototype.toString):
(WebInspector.ResourceCategory.prototype.get title):
* inspector/front-end/ResourceTreeModel.js:
* inspector/front-end/ScriptFormatter.js:
(WebInspector.ScriptFormatter.prototype._didFormatContent):
(WebInspector.FormatterMappingPayload):
* inspector/front-end/TimelineManager.js:
* inspector/front-end/WebKit.qrc:
* inspector/front-end/externs.js:
(console.assert):
(WebInspector.linkifyURLAsNode):
(WebInspector.linkifyStringAsFragment):
(WebInspector.UIString):
(Element.prototype.scrollIntoViewIfNeeded):
* inspector/front-end/inspector.html:
* inspector/front-end/inspector.js:
(WebInspector.UIString):
* inspector/front-end/treeoutline.js:
* inspector/front-end/utilities.js:
* inspector/generate-protocol-externs:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@93706 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoWeb Inspector: extract ConsoleMessage into a separate file.
pfeldman@chromium.org [Wed, 24 Aug 2011 12:52:41 +0000 (12:52 +0000)]
Web Inspector: extract ConsoleMessage into a separate file.
https://bugs.webkit.org/show_bug.cgi?id=66853

Reviewed by Yury Semikhatsky.

* WebCore.gypi:
* WebCore.vcproj/WebCore.vcproj:
* inspector/front-end/ConsoleMessage.js: Added.
(WebInspector.ConsoleMessage):
(WebInspector.ConsoleMessage.createTextMessage):
(WebInspector.ConsoleMessage.prototype._formatMessage):
(WebInspector.ConsoleMessage.prototype._linkifyLocation):
(WebInspector.ConsoleMessage.prototype._linkifyCallFrame):
(WebInspector.ConsoleMessage.prototype.isErrorOrWarning):
(WebInspector.ConsoleMessage.prototype._format):
(WebInspector.ConsoleMessage.prototype._formatWithSubstitutionString.valueFormatter):
(WebInspector.ConsoleMessage.prototype._formatWithSubstitutionString.append):
(WebInspector.ConsoleMessage.prototype.clearHighlight):
(WebInspector.ConsoleMessage.prototype.highlightSearchResults):
(WebInspector.ConsoleMessage.prototype.matchesRegex):
(WebInspector.ConsoleMessage.prototype.toMessageElement):
(WebInspector.ConsoleMessage.prototype._populateStackTraceTreeElement):
(WebInspector.ConsoleMessage.prototype._updateRepeatCount):
(WebInspector.ConsoleMessage.prototype.toString):
(WebInspector.ConsoleMessage.prototype.isEqual):
(WebInspector.ConsoleMessage.prototype.get stackTrace):
* inspector/front-end/ConsoleView.js:
* inspector/front-end/WebKit.qrc:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@93705 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2011-08-24 Pavel Podivilov <podivilov@chromium.org>
podivilov@chromium.org [Wed, 24 Aug 2011 11:40:57 +0000 (11:40 +0000)]
2011-08-24  Pavel Podivilov  <podivilov@chromium.org>

         Unreviewed, chromium rebaseline for gradient-after-transparent-border.

        * platform/chromium-cg-mac-leopard/fast/gradients/gradient-after-transparent-border-expected.png:
        * platform/chromium-cg-mac/fast/gradients/gradient-after-transparent-border-expected.png:
        * platform/chromium/test_expectations.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@93704 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2011-08-24 Pavel Podivilov <podivilov@chromium.org>
podivilov@chromium.org [Wed, 24 Aug 2011 10:52:54 +0000 (10:52 +0000)]
2011-08-24  Pavel Podivilov  <podivilov@chromium.org>

        Unreviewed, update test expectations for svg/custom/stroke-opacity-update.

        * platform/chromium-linux/svg/custom/stroke-opacity-update-expected.png: Added.
        * platform/chromium-win/svg/custom/stroke-opacity-update-expected.png: Added.
        * platform/chromium/test_expectations.txt:
        * platform/mac/svg/custom/stroke-opacity-update-expected.txt: Removed.
        * platform/qt/svg/custom/stroke-opacity-update-expected.txt: Removed.
        * svg/custom/stroke-opacity-update-expected.txt: Renamed from LayoutTests/platform/gtk/svg/custom/stroke-opacity-update-expected.txt.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@93703 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoSource/WebCore: Paint SliderTrack and SliderThumb independently.
commit-queue@webkit.org [Wed, 24 Aug 2011 10:51:22 +0000 (10:51 +0000)]
Source/WebCore: Paint SliderTrack and SliderThumb independently.
https://bugs.webkit.org/show_bug.cgi?id=50453
Also fixes https://bugs.webkit.org/show_bug.cgi?id=65456 crash.

Patch by Tatiana Meshkova <tatiana.meshkova@nokia.com> on 2011-08-24
Reviewed by Kenneth Rohde Christiansen.

RenderSlider works only with <input type=range>. Nowadays RenderTheme
supports painting of SliderTrack and SliderThumb without range input.
In order to be in sync with other platforms this patch gets rid of
upcasting in RenderThemeQt::paintSliderTrack and allows to paint
SliderTrack and SliderThumb independently.

Test: platform/qt/fast/forms/range/slider-crash.html

* platform/qt/RenderThemeQt.cpp:
(WebCore::RenderThemeQt::paintSliderTrack): Draw SC_SliderGroove.
Allow paintSliderThumb() to care about SC_SliderHandle. We don't need
to setup value and position anymore, since handle is drawn separately.
Draw focus explicitly, since QStyleOptionSlider doesn't allow to focus
on CC_Slider without handle subcontrol.
(WebCore::RenderThemeQt::paintSliderThumb): Draw SC_SliderHandle.

LayoutTests: Pixel crash test for RenderThemeQt::paintSliderTrack.
https://bugs.webkit.org/show_bug.cgi?id=50453

Patch by Tatiana Meshkova <tatiana.meshkova@nokia.com> on 2011-08-24
Reviewed by Kenneth Rohde Christiansen.

Support -webkit-appearance: slider-horizontal/vertical
painting without range input.

* platform/qt/fast/forms/range/slider-crash-expected.png: Added.
* platform/qt/fast/forms/range/slider-crash-expected.txt: Added.
* platform/qt/fast/forms/range/slider-crash.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@93702 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoWeb Inspector: annotate and prepare for compilation first 14 JavaScript files within...
pfeldman@chromium.org [Wed, 24 Aug 2011 10:18:59 +0000 (10:18 +0000)]
Web Inspector: annotate and prepare for compilation first 14 JavaScript files within 4 modules.
https://bugs.webkit.org/show_bug.cgi?id=66776

Reviewed by Yury Semikhatsky.

* inspector/compile-front-end.sh:
* inspector/front-end/ContextMenu.js:
* inspector/front-end/DOMAgent.js:
(WebInspector.DOMAgent.prototype._loadNodeAttributes):
(WebInspector.DOMAgent.prototype._childNodeRemoved):
(WebInspector.DOMAgent.prototype._inspectElementRequested):
(WebInspector.DOMDispatcher.prototype.inspectElementRequested):
* inspector/front-end/DebuggerModel.js:
(WebInspector.DebuggerModel.Location):
* inspector/front-end/ElementsPanel.js:
(WebInspector.ElementsPanel):
(WebInspector.ElementsPanel.prototype._inspectElementRequested):
* inspector/front-end/InspectorFrontendHostStub.js:
* inspector/front-end/KeyboardShortcut.js:
* inspector/front-end/Object.js:
(WebInspector.Object.prototype.addEventListener):
(WebInspector.Object.prototype.removeEventListener):
(WebInspector.Object.prototype.hasEventListeners):
* inspector/front-end/Panel.js:
(WebInspector.Panel):
* inspector/front-end/Placard.js:
* inspector/front-end/Popover.js:
(WebInspector.Popover.prototype.show):
* inspector/front-end/ResourcesPanel.js:
(WebInspector.ResourcesPanel):
(WebInspector.StorageCategoryTreeElement):
* inspector/front-end/Script.js:
(WebInspector.Script.prototype.editSource):
* inspector/front-end/ScriptsPanel.js:
* inspector/front-end/Settings.js:
(WebInspector.Settings):
(WebInspector.Settings.prototype.createSetting):
* inspector/front-end/SoftContextMenu.js:
(.WebInspector.SoftContextMenu.prototype._menuItemMouseUp):
(.WebInspector.SoftContextMenu.prototype._triggerAction):
(.WebInspector.SoftContextMenu.prototype._menuKeyDown):
* inspector/front-end/TabbedPane.js:
* inspector/front-end/TextPrompt.js:
* inspector/front-end/UserMetrics.js:
* inspector/front-end/View.js:
* inspector/front-end/externs.js:
(JSON.parse):
(JSON.stringify):
* inspector/front-end/inspector.js:
():
* inspector/generate-protocol-externs:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@93701 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2011-08-24 Pavel Podivilov <podivilov@chromium.org>
podivilov@chromium.org [Wed, 24 Aug 2011 10:10:13 +0000 (10:10 +0000)]
2011-08-24  Pavel Podivilov  <podivilov@chromium.org>

        Unreviewed, fix chromium expectations for fast/preloader/script.html.

        * platform/chromium/fast/preloader/script-expected.txt: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@93700 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoSelectElementData: Reorder members to reduce memory usage.
andreas.kling@nokia.com [Wed, 24 Aug 2011 09:52:27 +0000 (09:52 +0000)]
SelectElementData: Reorder members to reduce memory usage.
https://bugs.webkit.org/show_bug.cgi?id=66787

Reviewed by Darin Adler.

Reduced the size of SelectElementData by 24 bytes on 64-bit by rearranging
its data members.

Also added missing initialization of m_userDrivenChange.

* dom/SelectElement.cpp:
(WebCore::SelectElementData::SelectElementData):
* dom/SelectElement.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@93699 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoThere is no facility for profiling how the write barrier is used
fpizlo@apple.com [Wed, 24 Aug 2011 09:50:40 +0000 (09:50 +0000)]
There is no facility for profiling how the write barrier is used
https://bugs.webkit.org/show_bug.cgi?id=66747

Reviewed by Geoffrey Garen.

Added facilities for the JIT to specify the kind of write barrier
being executed.  Added code for profiling the number of each kind
of barrier encountered.

* GNUmakefile.list.am:
* JavaScriptCore.exp:
* JavaScriptCore.pro:
* JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.vcproj:
* JavaScriptCore.xcodeproj/project.pbxproj:
* dfg/DFGJITCodeGenerator.cpp:
(JSC::DFG::JITCodeGenerator::writeBarrier):
(JSC::DFG::JITCodeGenerator::cachedPutById):
* dfg/DFGJITCodeGenerator.h:
* dfg/DFGJITCompiler.cpp:
(JSC::DFG::JITCompiler::emitCount):
* dfg/DFGJITCompiler.h:
(JSC::DFG::JITCompiler::emitCount):
* dfg/DFGNonSpeculativeJIT.cpp:
(JSC::DFG::NonSpeculativeJIT::compile):
* dfg/DFGRepatch.cpp:
(JSC::DFG::tryCachePutByID):
* dfg/DFGSpeculativeJIT.cpp:
(JSC::DFG::SpeculativeJIT::compile):
* heap/Heap.h:
(JSC::Heap::writeBarrier):
* heap/WriteBarrierSupport.cpp: Added.
(JSC::WriteBarrierCounters::initialize):
* heap/WriteBarrierSupport.h: Added.
(JSC::WriteBarrierCounters::WriteBarrierCounters):
(JSC::WriteBarrierCounters::jitCounterFor):
(JSC::WriteBarrierCounters::countWriteBarrier):
* jit/JIT.h:
* jit/JITPropertyAccess.cpp:
(JSC::JIT::emit_op_put_by_id):
(JSC::JIT::privateCompilePutByIdTransition):
(JSC::JIT::emit_op_put_scoped_var):
(JSC::JIT::emit_op_put_global_var):
(JSC::JIT::emitWriteBarrier):
* jit/JITPropertyAccess32_64.cpp:
(JSC::JIT::emit_op_put_by_val):
(JSC::JIT::emit_op_put_by_id):
(JSC::JIT::privateCompilePutByIdTransition):
(JSC::JIT::emit_op_put_scoped_var):
(JSC::JIT::emit_op_put_global_var):
(JSC::JIT::emitWriteBarrier):
* runtime/InitializeThreading.cpp:
(JSC::initializeThreadingOnce):
* runtime/WriteBarrier.h:
(JSC::WriteBarrierBase::setWithoutWriteBarrier):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@93698 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago[EFL] Add dummy IconDatabaseClientEfl.
commit-queue@webkit.org [Wed, 24 Aug 2011 09:48:47 +0000 (09:48 +0000)]
[EFL] Add dummy IconDatabaseClientEfl.
http://bugs.webkit.org/show_bug.cgi?id=65596

Implements a dummy IconDatabaseClientEfl class.
Adds two files, IconDatabaseClientEfl.h | .cpp.

Patch by Jaehun Lim <ljaehun.lim@samsung.com> on 2011-08-24
Reviewed by Kent Tamura.

* CMakeListsEfl.txt:
* WebCoreSupport/IconDatabaseClientEfl.cpp: Added.
(WebCore::IconDatabaseClientEfl::performImport):
(WebCore::IconDatabaseClientEfl::didRemoveAllIcons):
(WebCore::IconDatabaseClientEfl::didImportIconURLForPageURL):
(WebCore::IconDatabaseClientEfl::didImportIconDataForPageURL):
(WebCore::IconDatabaseClientEfl::didChangeIconForPageURL):
(WebCore::IconDatabaseClientEfl::didFinishURLImport):
* WebCoreSupport/IconDatabaseClientEfl.h: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@93697 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoWeb Inspector: clearPseudoState should retain Element while clearing
commit-queue@webkit.org [Wed, 24 Aug 2011 09:07:59 +0000 (09:07 +0000)]
Web Inspector: clearPseudoState should retain Element while clearing
its state.
https://bugs.webkit.org/show_bug.cgi?id=66790

Patch by Pavel Feldman <pfeldman@google.com> on 2011-08-24
Reviewed by Yury Semikhatsky.

* inspector/InspectorCSSAgent.cpp:
(WebCore::InspectorCSSAgent::clearPseudoState):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@93696 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoExtra baseline for bug 66036 (canvas fill composite modes)
koz@chromium.org [Wed, 24 Aug 2011 05:24:42 +0000 (05:24 +0000)]
Extra baseline for bug 66036 (canvas fill composite modes)
https://bugs.webkit.org/show_bug.cgi?id=66835

Unreviewed, new baseline for pixel tests.

Patch by Ben Wells <benwells@chromium.org> on 2011-08-23

* platform/chromium-gpu-linux/fast/canvas/canvas-composite-expected.png:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@93695 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago[EFL] Change parameter type for some APIs in ewk_tile_matrix.
commit-queue@webkit.org [Wed, 24 Aug 2011 04:37:44 +0000 (04:37 +0000)]
[EFL] Change parameter type for some APIs in ewk_tile_matrix.
https://bugs.webkit.org/show_bug.cgi?id=66639

As type of both col and row parameter for ewk_tile_matrix's some APIs listed by this patch
are different from each other and eina_matrixsparse associated with ewk_tile_matrix is now
using unsigned long type for both col and row parameter, row parameter for ewk_tile_matrix
is also better to be specified as unsigned long type.

Patch by KwangHyuk Kim <hyuki.kim@samsung.com> on 2011-08-23
Reviewed by Anders Carlsson.

* ewk/ewk_tiled_matrix.c:
(ewk_tile_matrix_tile_exact_get):
(ewk_tile_matrix_tile_exact_exists):
(ewk_tile_matrix_tile_new):
(ewk_tile_matrix_tile_update):
(ewk_tile_matrix_tile_update_full):
* ewk/ewk_tiled_matrix.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@93693 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoWebSearchableFormData crashes when given a detached HTMLFormElement
abarth@webkit.org [Wed, 24 Aug 2011 04:32:42 +0000 (04:32 +0000)]
WebSearchableFormData crashes when given a detached HTMLFormElement
https://bugs.webkit.org/show_bug.cgi?id=66831

Reviewed by Dimitri Glazkov.

We need to null-check the Frame.

* src/WebSearchableFormData.cpp:
(HTMLNames::GetFormEncoding):
* tests/WebFrameTest.cpp:
(WebKit::TEST_F):
* tests/data/form.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@93692 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoMake notifications appear in time order again.
dglazkov@chromium.org [Wed, 24 Aug 2011 04:24:08 +0000 (04:24 +0000)]
Make notifications appear in time order again.
https://bugs.webkit.org/show_bug.cgi?id=66815

Reviewed by Adam Barth.

* BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui/notifications.js: Added setting index of the notification.
* BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui/notifications_unittests.js: Updated the test.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@93691 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoExtra baseline for bug 66442 (skia webkit-transform breaks webkit-mask)
commit-queue@webkit.org [Wed, 24 Aug 2011 02:14:23 +0000 (02:14 +0000)]
Extra baseline for bug 66442 (skia webkit-transform breaks webkit-mask)
https://bugs.webkit.org/show_bug.cgi?id=66833

Patch by Ben Wells <benwells@chromium.org> on 2011-08-23
Reviewed by Kenneth Russell.

* platform/chromium-gpu-linux/compositing/masks/multiple-masks-expected.png: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@93690 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago[chromium] Make WebGL context current before querying for extensions
commit-queue@webkit.org [Wed, 24 Aug 2011 02:11:11 +0000 (02:11 +0000)]
[chromium] Make WebGL context current before querying for extensions
https://bugs.webkit.org/show_bug.cgi?id=66788

We're making changes on the Chromium side for threaded compositing that
will require makeContextCurrent to be called on the appropriate thread
before any other GL calls.

In line with other GraphicsContext3D implementations, we do this
implicitly. This is the minimum change that makes WebGL compliant.

Patch by Iain Merrick <husky@google.com> on 2011-08-23
Reviewed by Kenneth Russell.

* src/GraphicsContext3DChromium.cpp:
(WebCore::GraphicsContext3DInternal::initializeExtensions):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@93689 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoAdd checks to ensure allocation does not take place during initialization of GC-manag...
commit-queue@webkit.org [Wed, 24 Aug 2011 02:05:33 +0000 (02:05 +0000)]
Add checks to ensure allocation does not take place during initialization of GC-managed objects
https://bugs.webkit.org/show_bug.cgi?id=65288

Patch by Mark Hahnenberg <mhahnenberg@apple.com> on 2011-08-23
Reviewed by Darin Adler.

Source/JavaScriptCore:

Adding the new validation functionality.  In its current state, it will performs checks,
but they don't fail unless you do allocation in the arguments to the parent constructor in the
initialization list of a class.  The allocateCell() method turns on the global flag disallowing any new
allocations, and the constructorBody() method in JSCell turns it off.  This way, allocation is still
allowed in constructor bodies while other refactoring efforts continue.

* runtime/JSCell.h:
(JSC::JSCell::JSCell::constructorBody):
(JSC::JSCell::JSCell::JSCell):
(JSC::JSCell::allocateCell):
* runtime/JSGlobalData.cpp:
(JSC::JSGlobalData::JSGlobalData):
* runtime/JSGlobalData.h:
(JSC::JSGlobalData::isInitializingObject):
(JSC::JSGlobalData::setInitializingObject):
* runtime/StringObjectThatMasqueradesAsUndefined.h:
(JSC::StringObjectThatMasqueradesAsUndefined::create):

Source/WebCore:

No new tests.

Adding the new validation functionality.  In its current state, it will performs checks,
but they don't fail unless you do allocation in the arguments to the parent constructor in the
initialization list of a class.  The allocateCell() method turns on the global flag disallowing any new
allocations, and the constructorBody() method in JSCell turns it off.  This way, allocation is still
allowed in constructor bodies while other refactoring efforts continue.

* bindings/js/JSDOMWindowShell.cpp:
(WebCore::JSDOMWindowShell::operator new):
* bindings/js/JSDOMWindowShell.h:

Source/WebKit/mac:

Adding the new validation functionality.  In its current state, it will performs checks,
but they don't fail unless you do allocation in the arguments to the parent constructor in the
initialization list of a class.  The allocateCell() method turns on the global flag disallowing any new
allocations, and the constructorBody() method in JSCell turns it off.  This way, allocation is still
allowed in constructor bodies while other refactoring efforts continue.

* Plugins/Hosted/ProxyRuntimeObject.h:
(WebKit::ProxyRuntimeObject::create):
* Plugins/Hosted/ProxyRuntimeObject.mm:
(WebKit::ProxyRuntimeObject::ProxyRuntimeObject):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@93688 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoRebaselines for bug 66036 (fills with various canvas compositing modes)
commit-queue@webkit.org [Wed, 24 Aug 2011 02:00:10 +0000 (02:00 +0000)]
Rebaselines for bug 66036 (fills with various canvas compositing modes)
https://bugs.webkit.org/show_bug.cgi?id=66824

Patch by Ben Wells <benwells@chromium.org> on 2011-08-23
Reviewed by James Robinson.

* platform/chromium-gpu-win/fast/canvas/canvas-composite-expected.png: Added.
* platform/chromium-gpu-win/fast/canvas/canvas-composite-transformclip-expected.png: Added.
* platform/chromium/test_expectations.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@93687 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago[chromium] Add more flaky snowleopard GPU tests.
pkasting@chromium.org [Wed, 24 Aug 2011 01:57:33 +0000 (01:57 +0000)]
[chromium] Add more flaky snowleopard GPU tests.

Unreviewed, test expectations update.

* platform/chromium/test_expectations.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@93686 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago[EFL] Get the default value for `scripts_window_open' from the right method.
commit-queue@webkit.org [Wed, 24 Aug 2011 01:55:25 +0000 (01:55 +0000)]
[EFL] Get the default value for `scripts_window_open' from the right method.
https://bugs.webkit.org/show_bug.cgi?id=66775

Patch by Raphael Kubo da Costa <kubo@profusion.mobi> on 2011-08-23
Reviewed by Kent Tamura.

For some unknown reason, the default value for this setting was being
obtained from Settings::allowScriptsToCloseWindows() instead of
Settings::javaScriptCanOpenWindowsAutomatically(), which makes no
sense at all.

* ewk/ewk_view.cpp:
(_ewk_view_priv_new):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@93685 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago[chromium] Fix erroneous test expectation.
pkasting@chromium.org [Wed, 24 Aug 2011 01:39:41 +0000 (01:39 +0000)]
[chromium] Fix erroneous test expectation.

Unreviewed, test expectations update.

* platform/chromium/test_expectations.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@93684 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago[EFL] Change the tiled backing store to use constant size of tile.
commit-queue@webkit.org [Wed, 24 Aug 2011 01:39:30 +0000 (01:39 +0000)]
[EFL] Change the tiled backing store to use constant size of tile.
https://bugs.webkit.org/show_bug.cgi?id=65848

As current tiled backing store in Webkit EFL is using variable tile size according to the zoom level,
the number of tiles in viewport is not same always and the speed of panning is also not consistent
whenever zoom level is changed.
Therefore, tile size should be constant size in order to make sure the consistent panning speed.

Patch by KwangHyuk Kim <hyuki.kim@samsung.com> on 2011-08-23
Reviewed by Anders Carlsson.

* ewk/ewk_tiled_backing_store.c:
(_ewk_tiled_backing_store_model_matrix_create):
(_ewk_tiled_backing_store_smart_add):
(_ewk_tiled_backing_store_smart_calculate):
(_ewk_tiled_backing_store_zoom_set_internal):
(ewk_tiled_backing_store_zoom_weak_set):
(ewk_tiled_backing_store_flush):
(ewk_tiled_backing_store_pre_render_region):
(ewk_tiled_backing_store_pre_render_relative_radius):
* ewk/ewk_tiled_backing_store.h:
* ewk/ewk_tiled_matrix.c:
(_ewk_tile_matrix_cell_free):
(_ewk_tile_matrix_tile_free):
(ewk_tile_matrix_new):
(ewk_tile_matrix_tile_exact_get):
(ewk_tile_matrix_tile_exact_exists):
(ewk_tile_matrix_tile_new):
(ewk_tile_matrix_tile_update):
(ewk_tile_matrix_tile_update_full):
(_ewk_tile_matrix_slicer_setup):
(ewk_tile_matrix_update):
(ewk_tile_matrix_dbg):
* ewk/ewk_tiled_matrix.h:
* ewk/ewk_tiled_model.c:
(ewk_tile_new):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@93683 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoScrollAnimatorNone coasting implementation
commit-queue@webkit.org [Wed, 24 Aug 2011 01:36:50 +0000 (01:36 +0000)]
ScrollAnimatorNone coasting implementation
https://bugs.webkit.org/show_bug.cgi?id=66258

Source/WebCore:

Increase the animation time based on the velocity of the scroll, up to a
maximum amount. Fix bugs found by additional unit testing.

Patch by Scott Byer <scottbyer@chromium.org> on 2011-08-23
Reviewed by James Robinson.

Additional tests in ScrollAnimatorNoneTest: CurveMathQuartic, CurveMathCoast,
ScrollOnceQuartic, ScrollTwiceCubic, ScrollLotsCubic, ScrollLotsCubicSmoothed,
ScrollUpToBumperCoast, ScrollDownToBumperCoast, VaryingInputsEquivalency,
VaryingInputsEquivalencyCoast, VaryingInputsEquivalencyCoastLarge,
VaryingInputsEquivalencyCoastSteep

* platform/ScrollAnimatorNone.cpp:
(WebCore::ScrollAnimatorNone::Parameters::Parameters):
(WebCore::ScrollAnimatorNone::PerAxisData::curveAt):
(WebCore::ScrollAnimatorNone::PerAxisData::coastCurve):
(WebCore::ScrollAnimatorNone::PerAxisData::curveIntegralAt):
(WebCore::ScrollAnimatorNone::PerAxisData::attackArea):
(WebCore::ScrollAnimatorNone::PerAxisData::releaseArea):
(WebCore::ScrollAnimatorNone::PerAxisData::PerAxisData):
(WebCore::ScrollAnimatorNone::PerAxisData::updateDataFromParameters):
(WebCore::ScrollAnimatorNone::PerAxisData::updateVisibleLength):
(WebCore::ScrollAnimatorNone::ScrollAnimatorNone):
(WebCore::ScrollAnimatorNone::scroll):
(WebCore::ScrollAnimatorNone::willEndLiveResize):
(WebCore::ScrollAnimatorNone::didAddVerticalScrollbar):
(WebCore::ScrollAnimatorNone::didAddHorizontalScrollbar):
(WebCore::ScrollAnimatorNone::updateVisibleLengths):
* platform/ScrollAnimatorNone.h:

Source/WebKit/chromium:

Patch by Scott Byer <scottbyer@chromium.org> on 2011-08-23
Reviewed by James Robinson.

* tests/ScrollAnimatorNoneTest.cpp:
(MockScrollableArea::visibleHeight):
(MockScrollableArea::visibleWidth):
(ScrollAnimatorNoneTest::SavePerAxisData::SavePerAxisData):
(ScrollAnimatorNoneTest::SavePerAxisData::operator==):
(ScrollAnimatorNoneTest::SetUp):
(ScrollAnimatorNoneTest::updateDataFromParameters):
(ScrollAnimatorNoneTest::animateScroll):
(ScrollAnimatorNoneTest::attackArea):
(ScrollAnimatorNoneTest::releaseArea):
(ScrollAnimatorNoneTest::coastCurve):
(ScrollAnimatorNoneTest::curveTestInner):
(TEST_F):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@93682 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago[chromium] Need a way to test lost compositor context recovery
jamesr@google.com [Wed, 24 Aug 2011 01:33:12 +0000 (01:33 +0000)]
[chromium] Need a way to test lost compositor context recovery
https://bugs.webkit.org/show_bug.cgi?id=66820

Reviewed by Kenneth Russell.

Source/WebCore:

Adds support for recreating a context as if it was lost, and fixes
compositeAndReadback() to attempt context recovery.

Test: platform/chromium/compositing/lost-compositor-context.html

* platform/graphics/chromium/cc/CCLayerTreeHost.cpp:
(WebCore::CCLayerTreeHost::compositeAndReadback):
(WebCore::CCLayerTreeHost::loseCompositorContext):
* platform/graphics/chromium/cc/CCLayerTreeHost.h:

Source/WebKit/chromium:

Adds a testing-only API to simulate a lost compositor context.

* public/WebView.h:
* src/WebViewImpl.cpp:
(WebKit::WebViewImpl::loseCompositorContext):
* src/WebViewImpl.h:

Tools:

Exposes a LayoutTestController interface in chromium to simulate a
lost compositor context.

* DumpRenderTree/chromium/LayoutTestController.cpp:
(LayoutTestController::LayoutTestController):
(LayoutTestController::loseCompositorContext):
* DumpRenderTree/chromium/LayoutTestController.h:

LayoutTests:

Adds a simple test for losing the compositor context in a recoverable
way on a composited page. Tests that we recover without crashing and
render the updated content on the next frame.

* platform/chromium/compositing/lost-compositor-context.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@93681 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago[chromium] Implement CCThread in terms of WebThread
nduca@chromium.org [Wed, 24 Aug 2011 01:22:22 +0000 (01:22 +0000)]
[chromium] Implement CCThread in terms of WebThread
https://bugs.webkit.org/show_bug.cgi?id=66610

Reviewed by Darin Fisher.

Source/WebCore:

* WebCore.gypi:
* platform/graphics/chromium/cc/CCCompletionEvent.h:
* platform/graphics/chromium/cc/CCLayerTreeHost.h:
(WebCore::CCLayerTreeHost::client):
* platform/graphics/chromium/cc/CCLayerTreeHostImplProxy.cpp:
(WebCore::CCLayerTreeHostImplProxy::CCLayerTreeHostImplProxy):
* platform/graphics/chromium/cc/CCThread.h:
(WebCore::CCThread::~CCThread):

Source/WebKit/chromium:

* WebKit.gyp:
* public/WebThread.h:
(WebKit::WebThread::~WebThread):
* src/CCThreadImpl.cpp: Added.
(WebKit::GetThreadIDTask::GetThreadIDTask):
(WebKit::GetThreadIDTask::~GetThreadIDTask):
(WebKit::GetThreadIDTask::run):
(WebKit::CCThreadTaskAdapter::CCThreadTaskAdapter):
(WebKit::CCThreadTaskAdapter::~CCThreadTaskAdapter):
(WebKit::CCThreadTaskAdapter::run):
(WebKit::CCThreadImpl::create):
(WebKit::CCThreadImpl::~CCThreadImpl):
(WebKit::CCThreadImpl::postTask):
(WebKit::CCThreadImpl::threadID):
(WebKit::CCThreadImpl::CCThreadImpl):
* src/CCThreadImpl.h: Renamed from Source/WebCore/platform/graphics/chromium/cc/CCThread.cpp.
* src/WebViewImpl.cpp:
(WebKit::WebViewImpl::createCompositorThread):
* src/WebViewImpl.h:
* tests/CCThreadTest.cpp:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@93680 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago[chromium] Renderer crashes when about:gpucrash is loaded
commit-queue@webkit.org [Wed, 24 Aug 2011 01:18:00 +0000 (01:18 +0000)]
[chromium] Renderer crashes when about:gpucrash is loaded
https://bugs.webkit.org/show_bug.cgi?id=66814

Source/WebCore:

WebViewImpl::didRecreateGraphicsContext() was calling
setRootLayer() on the CC layer tree, but the root layer
doesn't actually change and the code wasn't defensive
against this. This CL adds some asserts to catch tree
corruption and removes the unnecessary calls.

Patch by Iain Merrick <husky@google.com> on 2011-08-23
Reviewed by James Robinson.

* platform/graphics/GraphicsLayer.cpp:
(WebCore::GraphicsLayer::setParent):
* platform/graphics/GraphicsLayer.h:
* platform/graphics/chromium/LayerChromium.cpp:
(WebCore::LayerChromium::setParent):
(WebCore::LayerChromium::hasAncestor):
* platform/graphics/chromium/LayerChromium.h:

Source/WebKit/chromium:

WebViewImpl::didRecreateGraphicsContext() was calling
setRootLayer() on the CC layer tree, but the root layer
doesn't actually change and the code wasn't defensive
against this. This CL adds some asserts to catch tree
corruption and removes the unnecessary calls.

Patch by Iain Merrick <husky@google.com> on 2011-08-23
Reviewed by James Robinson.

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

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@93679 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoUnreviewed, rolling out r93645.
commit-queue@webkit.org [Wed, 24 Aug 2011 01:00:07 +0000 (01:00 +0000)]
Unreviewed, rolling out r93645.
http://trac.webkit.org/changeset/93645
https://bugs.webkit.org/show_bug.cgi?id=66828

Broke webkit_unit_tests on chromium mac (Requested by jamesr
on #webkit).

Patch by Sheriff Bot <webkit.review.bot@gmail.com> on 2011-08-23

* platform/chromium/ScrollbarThemeChromiumMac.mm:
(WebCore::ScrollbarThemeChromiumMac::ScrollbarThemeChromiumMac):
(WebCore::ScrollbarThemeChromiumMac::paintOverhangAreas):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@93676 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoPrepare WKTextInputWindowController for the new text input model
andersca@apple.com [Wed, 24 Aug 2011 00:29:47 +0000 (00:29 +0000)]
Prepare WKTextInputWindowController for the new text input model
https://bugs.webkit.org/show_bug.cgi?id=66826

Reviewed by Sam Weinig.

Add -[WKTextInputWindowController hasMarkedText] and make interpretKeyEvent take an
additional usingLegacyCocoaTextInput flag. No functionality change.

* UIProcess/API/mac/WKTextInputWindowController.h:
* UIProcess/API/mac/WKTextInputWindowController.mm:
(-[WKTextInputPanel _interpretKeyEvent:usingLegacyCocoaTextInput:string:]):
(-[WKTextInputPanel _hasMarkedText]):
(-[WKTextInputWindowController hasMarkedText]):
(-[WKTextInputWindowController interpretKeyEvent:usingLegacyCocoaTextInput:string:]):
* UIProcess/API/mac/WKView.mm:
(-[WKView keyDown:]):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@93670 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agohttps://bugs.webkit.org/show_bug.cgi?id=66244
bdakin@apple.com [Wed, 24 Aug 2011 00:29:27 +0000 (00:29 +0000)]
https://bugs.webkit.org/show_bug.cgi?id=66244
Cached pages don't fully update when going back after changing the display scale
factor
-and corresponding-
<rdar://problem/9955656>

Reviewed by Darin Adler.

This patch adds a generalized concept of needing a full style recalc to the
BackForwardController. So when the display scale factor is changed, the
BackForwardController can be informed that all pages will need a full style recalc
when they come out of the cache. This same mechanism is also used to fix a long-
standing bug with full-page/text zoom.

Iterate through the HistoryItems and mark all CachedPages as needing a full style
recalc.
* history/BackForwardController.cpp:
(WebCore::BackForwardController::markPagesForFullStyleRecalc):
* history/BackForwardController.h:

ChachedPage has a new bool -- m_needsFullStyleRecalc -- to track whether a full
style recalc is needed when the CachedPage is restored.
* history/CachedPage.cpp:
(WebCore::CachedPage::CachedPage):
(WebCore::CachedPage::restore):
(WebCore::CachedPage::clear):
* history/CachedPage.h:
(WebCore::CachedPage::markForFullStyleRecalc):

HistoryItem actually takes care of calling into CachedPage.
* history/HistoryItem.cpp:
(WebCore::HistoryItem::markForFullStyleRecalc):
* history/HistoryItem.h:

Fix style recalc issues for full-page/text zoom by calling our new function on
PageCache.
* page/Frame.cpp:
(WebCore::Frame::setPageAndTextZoomFactors):

Fix style recalc issues for display scale factor changes by calling our new
function on PageCache.
* page/Page.cpp:
(WebCore::Page::setDeviceScaleFactor):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@93669 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoFix baselines for fast/block/float/fit_line_below_floats.html. I forgot Windows...
abarth@webkit.org [Wed, 24 Aug 2011 00:13:51 +0000 (00:13 +0000)]
Fix baselines for fast/block/float/fit_line_below_floats.html.  I forgot Windows in my previous patch.

* platform/chromium-linux/fast/block/float/fit_line_below_floats-expected.txt: Removed.
* platform/chromium-win/fast/block/float/fit_line_below_floats-expected.png: Added.
* platform/chromium-win/fast/block/float/fit_line_below_floats-expected.txt: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@93663 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago[chromium] Remove comments formerly associated with now-rebaselined tests.
pkasting@chromium.org [Tue, 23 Aug 2011 23:59:33 +0000 (23:59 +0000)]
[chromium] Remove comments formerly associated with now-rebaselined tests.

Unreviewed, test expectations update.

* platform/chromium/test_expectations.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@93662 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoChange garden-o-matic frontend builder array to be a configuration map.
dglazkov@chromium.org [Tue, 23 Aug 2011 23:33:57 +0000 (23:33 +0000)]
Change garden-o-matic frontend builder array to be a configuration map.
https://bugs.webkit.org/show_bug.cgi?id=66816

Reviewed by Adam Barth.

* BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/builders.js: Made kBuilders a map, with value being a configuration object.
* BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/config.js: Adjusted callsites to expect a map.
* BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/main.js: Ditto.
* BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/model.js: Ditto.
* BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui.js: Ditto.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@93658 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoFix build.
andersca@apple.com [Tue, 23 Aug 2011 23:32:34 +0000 (23:32 +0000)]
Fix build.

* inspector/InstrumentingAgents.cpp:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@93657 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoAdded support for momentarily revealing last typed character in password input.
commit-queue@webkit.org [Tue, 23 Aug 2011 23:31:44 +0000 (23:31 +0000)]
Added support for momentarily revealing last typed character in password input.
Code change was partially based on Apple's iOS code and Samuel Nevala's work.
https://bugs.webkit.org/show_bug.cgi?id=32509

Patch by Chang Shu <cshu@webkit.org> on 2011-08-23
Reviewed by Alexey Proskuryakov.

Source/WebCore:

* editing/InsertIntoTextNodeCommand.cpp:
(WebCore::InsertIntoTextNodeCommand::doApply):
* rendering/RenderText.cpp:
(WebCore::SecureTextTimer::SecureTextTimer):
(WebCore::SecureTextTimer::restartWithNewText):
(WebCore::SecureTextTimer::invalidate):
(WebCore::SecureTextTimer::lastTypedCharacterOffset):
(WebCore::SecureTextTimer::fired):
(WebCore::RenderText::willBeDestroyed):
(WebCore::RenderText::setTextInternal):
(WebCore::RenderText::secureText):
(WebCore::RenderText::momentarilyRevealLastTypedCharacter):
* rendering/RenderText.h:
(WebCore::RenderText::isSecure):
* testing/Internals.cpp:
(WebCore::Internals::setPasswordEchoEnabled): Fixed some silly coding in Internals.
(WebCore::Internals::setPasswordEchoDurationInSeconds):
(WebCore::Internals::reset):

LayoutTests:

Updated failed expected files.

* editing/input/password-echo-passnode-expected.txt:
* editing/input/password-echo-passnode2-expected.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@93656 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoRebaselines for bug 66442 (skia webkit-transform breaks webkit-mask)
commit-queue@webkit.org [Tue, 23 Aug 2011 23:24:44 +0000 (23:24 +0000)]
Rebaselines for bug 66442 (skia webkit-transform breaks webkit-mask)
https://bugs.webkit.org/show_bug.cgi?id=66757

Patch by Ben Wells <benwells@chromium.org> on 2011-08-23
Reviewed by Kenneth Russell.

* platform/chromium-gpu-win/compositing/masks/multiple-masks-expected.png: Added.
* platform/chromium-gpu-win/compositing/masks/multiple-masks-expected.txt: Added.
* platform/chromium-linux/fast/backgrounds/mask-composite-expected.png: Added.
* platform/chromium-win/fast/backgrounds/mask-composite-expected.png: Added.
* platform/chromium-win/fast/backgrounds/mask-composite-expected.txt: Added.
* platform/chromium-win/fast/css/transformed-mask-expected.png: Added.
* platform/chromium-win/fast/css/transformed-mask-expected.txt: Renamed from LayoutTests/platform/chromium-linux/fast/css/transformed-mask-expected.txt.
* platform/chromium/test_expectations.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@93655 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agohttps://bugs.webkit.org/show_bug.cgi?id=66751
commit-queue@webkit.org [Tue, 23 Aug 2011 23:22:19 +0000 (23:22 +0000)]
https://bugs.webkit.org/show_bug.cgi?id=66751
[Chromium]WebWorkerClientImpl should always invoke
InspectorIntrumentation on main thread.
The fix moves calls to InspectorInstrumentation from
WebWorkerClientImpl to unedrlying WorkerMessagingProxy.

Patch by Dmitry Lomov <Dmitry Lomov (dslomov@google.com)> on 2011-08-23
Reviewed by Pavel Feldman.

Source/WebCore:

Covered by existing tests.

* inspector/InstrumentingAgents.cpp:
(WebCore::instrumentationForPage):
* workers/WorkerMessagingProxy.cpp:
(WebCore::WorkerMessagingProxy::workerContextDestroyedInternal):
(WebCore::WorkerMessagingProxy::terminateWorkerContext):

Source/WebKit/chromium:

* src/WebWorkerClientImpl.cpp:
(WebKit::WebWorkerClientImpl::terminateWorkerContext):
(WebKit::WebWorkerClientImpl::workerContextDestroyed):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@93654 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoTurn WebCanvas into CGContext before drawing scrollbar (DRT using Skia on Mac)
caryclark@google.com [Tue, 23 Aug 2011 23:18:34 +0000 (23:18 +0000)]
Turn WebCanvas into CGContext before drawing scrollbar (DRT using Skia on Mac)
https://bugs.webkit.org/show_bug.cgi?id=66803

This fixes the WebKit Canaries, in particular avoids a crash on tests
like css1/basic/comments.html.

Reviewed by James Robinson.

* DumpRenderTree/chromium/WebThemeEngineDRTMac.mm:
(WebThemeEngineDRTMac::paintNSScrollerScrollbarThumb):
The NSGraphicsContext call graphicsContextWithGraphicsPort takes a void*
parameter, so it took a WebCanvas* (mapped to SkCanvas* in Skia-land)
and treated it as if it were a CGContextRef.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@93653 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago[Qt] Button text is invisible when mobile theme is in use and application palette...
commit-queue@webkit.org [Tue, 23 Aug 2011 23:12:27 +0000 (23:12 +0000)]
[Qt] Button text is invisible when mobile theme is in use and application palette set to use light text color.
https://bugs.webkit.org/show_bug.cgi?id=66635

When mobile theme is in use and application palette is using light
color to display text button text becomes invisible. The reason for
that is that in RenderThemeQt::systemFont() we are using application
pallete while when are painting buttons lightGrayPalette is used.
Calling setPaletteFromPageClientIfExists() takes care of the issue.

Patch by Misha Tyutyunik <michael.tyutyunik@nokia.com> on 2011-08-23
Reviewed by Andreas Kling.

Covered by existing tests.

* platform/qt/RenderThemeQt.cpp:
(WebCore::RenderThemeQt::systemColor):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@93652 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoAdd handling of mix-width and max-width for flexitems
tony@chromium.org [Tue, 23 Aug 2011 23:10:45 +0000 (23:10 +0000)]
Add handling of mix-width and max-width for flexitems
https://bugs.webkit.org/show_bug.cgi?id=66723

Reviewed by David Hyatt.

Source/WebCore:

If we flex past a min/max width value, we need to mark the flexitem as
a fixed width and re-start the flexing algorithm.  We use a HashMap to
keep track of fixed width items.

This patch also split out the size computation from the actual layout
to avoid unnecessary layouts caused by restarting the flexing algorithm.

Test: css3/flexbox/003.html

* rendering/RenderFlexibleBox.cpp:
(WebCore::RenderFlexibleBox::layoutHorizontalBlock):
(WebCore::RenderFlexibleBox::runFreeSpaceAllocationAlgorithm):
* rendering/RenderFlexibleBox.h:

LayoutTests:

* css3/flexbox/003-expected.txt: Added.
* css3/flexbox/003.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@93651 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoUpdate expectations for pkasting. The rebaseline script is having
abarth@webkit.org [Tue, 23 Aug 2011 23:08:31 +0000 (23:08 +0000)]
Update expectations for pkasting.  The rebaseline script is having
trouble on Cygwin, so I ran the script as instructed by Peter.

* platform/chromium-cg-mac-leopard/fast/block/float/fit_line_below_floats-expected.png: Added.
* platform/chromium-cg-mac-leopard/fast/block/float/floats-and-text-indent-expected.png: Added.
* platform/chromium-cg-mac-leopard/fast/block/float/floats-and-text-indent-rl-expected.png: Added.
* platform/chromium-cg-mac-leopard/fast/multicol/span/span-as-nested-inline-block-child-expected.png: Added.
* platform/chromium-cg-mac/fast/multicol/span/span-as-nested-inline-block-child-expected.png: Added.
* platform/chromium-linux-x86/fast/css: Removed.
* platform/chromium-linux-x86/svg/dynamic-updates: Removed.
* platform/chromium-linux/fast/block/float/fit_line_below_floats-expected.png: Added.
* platform/chromium-linux/fast/block/float/fit_line_below_floats-expected.txt: Added.
* platform/chromium-linux/fast/block/float/floats-and-text-indent-expected.png: Added.
* platform/chromium-linux/fast/block/float/floats-and-text-indent-rl-expected.png: Added.
* platform/chromium-linux/fast/block/positioning/058-expected.png:
* platform/chromium-linux/fast/multicol/span/span-as-nested-inline-block-child-expected.png: Added.
* platform/chromium-linux/svg/custom/image-small-width-height-expected.png:
* platform/chromium-linux/svg/dynamic-updates/SVGFEImageElement-dom-preserveAspectRatio-attr-expected.png:
* platform/chromium-linux/svg/dynamic-updates/SVGFEImageElement-svgdom-preserveAspectRatio-prop-expected.png:
* platform/chromium-win-vista/svg/dynamic-updates: Removed.
* platform/chromium-win/fast/block/float/floats-and-text-indent-expected.png: Added.
* platform/chromium-win/fast/block/float/floats-and-text-indent-expected.txt: Added.
* platform/chromium-win/fast/block/float/floats-and-text-indent-rl-expected.png: Added.
* platform/chromium-win/fast/block/float/floats-and-text-indent-rl-expected.txt: Added.
* platform/chromium-win/fast/block/positioning/058-expected.png:
* platform/chromium-win/fast/block/positioning/058-expected.txt:
* platform/chromium-win/fast/css/border-radius-non-negative-expected.png: Added.
* platform/chromium-win/fast/multicol/span/span-as-nested-inline-block-child-expected.png: Added.
* platform/chromium-win/fast/multicol/span/span-as-nested-inline-block-child-expected.txt: Added.
* platform/chromium-win/svg/custom/image-small-width-height-expected.png:
* platform/chromium-win/svg/dynamic-updates/SVGFEImageElement-dom-preserveAspectRatio-attr-expected.png:
* platform/chromium-win/svg/dynamic-updates/SVGFEImageElement-svgdom-preserveAspectRatio-prop-expected.png:
* platform/chromium/test_expectations.txt:
* platform/gtk/svg/custom/image-small-width-height-expected.txt: Removed.
* platform/qt/svg/custom/image-small-width-height-expected.txt: Removed.
* svg/custom/image-small-width-height-expected.txt: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@93650 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoFix mergeChangeLogs test on native windows
paroga@webkit.org [Tue, 23 Aug 2011 23:07:20 +0000 (23:07 +0000)]
Fix mergeChangeLogs test on native windows
https://bugs.webkit.org/show_bug.cgi?id=66797

Reviewed by Adam Roben.

Windows has no /tmp/ directory and the "TMPDIR" environment variable isn't set in the default setup.
Add the "TEMP" environment variable as additional source for the temporary directory used in writeTempFile.

* Scripts/webkitperl/VCSUtils_unittest/mergeChangeLogs.pl:
(writeTempFile):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@93649 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoResourceRequest::setStorageSession should update NSURLRequest as well
psolanki@apple.com [Tue, 23 Aug 2011 23:05:10 +0000 (23:05 +0000)]
ResourceRequest::setStorageSession should update NSURLRequest as well
https://bugs.webkit.org/show_bug.cgi?id=66811

Reviewed by Darin Adler.

* platform/network/cf/ResourceRequestCFNet.cpp:
(WebCore::ResourceRequest::setStorageSession): Update the m_nsRequest field as well since we
have a new CFURLRequestRef.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@93648 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago[chromium] Remove obsolete child context list from LayerRendererChromium
commit-queue@webkit.org [Tue, 23 Aug 2011 23:01:08 +0000 (23:01 +0000)]
[chromium] Remove obsolete child context list from LayerRendererChromium
https://bugs.webkit.org/show_bug.cgi?id=66743

Patch by James Robinson <jamesr@chromium.org> on 2011-08-23
Reviewed by Kenneth Russell.

Before we had proper semantics for flush() in the command buffer, we
used a scheme of latches across contexts to ensure consistent frames
were presented for WebGL and canvas. We no longer have latches and
flush is a proper synchronization primative, so the child context list
no longer does anything. Canvas2DLayerChromium and
WebGLLayerChromium's implementations of updateCompositorResources() do
flushes on the appropriate context already - canvas layers via
DrawingBuffer::publishToPlatformLayer(), and WebGL layers via the call
to prepareTexture().

No new tests since the code being removed doesn't do anything.

* platform/graphics/chromium/Canvas2DLayerChromium.cpp:
(WebCore::Canvas2DLayerChromium::~Canvas2DLayerChromium):
(WebCore::Canvas2DLayerChromium::setDrawingBuffer):
* platform/graphics/chromium/Canvas2DLayerChromium.h:
* platform/graphics/chromium/LayerRendererChromium.cpp:
(WebCore::LayerRendererChromium::drawLayers):
* platform/graphics/chromium/LayerRendererChromium.h:
* platform/graphics/chromium/WebGLLayerChromium.cpp:
(WebCore::WebGLLayerChromium::~WebGLLayerChromium):
(WebCore::WebGLLayerChromium::setContext):
* platform/graphics/chromium/WebGLLayerChromium.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@93647 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoChromium Mac: Use a custom pattern image for rubber banding overhang area
commit-queue@webkit.org [Tue, 23 Aug 2011 22:59:34 +0000 (22:59 +0000)]
Chromium Mac: Use a custom pattern image for rubber banding overhang area
https://bugs.webkit.org/show_bug.cgi?id=66707

Patch by Alexei Svitkine <asvitkine@chromium.org> on 2011-08-23
Reviewed by Dimitri Glazkov.

No new tests since its just changing the Chromium-specific overhang pattern.

* platform/chromium/ScrollbarThemeChromiumMac.mm:
(WebCore::ScrollbarThemeChromiumMac::ScrollbarThemeChromiumMac):
(WebCore::ScrollbarThemeChromiumMac::paintOverhangAreas):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@93645 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoSource/WebCore: Prepare to paint slider thumb separately.
commit-queue@webkit.org [Tue, 23 Aug 2011 22:57:49 +0000 (22:57 +0000)]
Source/WebCore: Prepare to paint slider thumb separately.
Provide proper size values for vertical SliderThumb.
https://bugs.webkit.org/show_bug.cgi?id=66732

Patch by Tatiana Meshkova <tatiana.meshkova@nokia.com> on 2011-08-23
Reviewed by Kenneth Rohde Christiansen.

* platform/qt/RenderThemeQt.cpp:
(WebCore::RenderThemeQt::adjustSliderThumbSize):
Swap length and thickness for sliderthumb-vertical.

LayoutTests: Swap width and height for sliderthumb-vertical.
https://bugs.webkit.org/show_bug.cgi?id=66732

Patch by Tatiana Meshkova <tatiana.meshkova@nokia.com> on 2011-08-23
Reviewed by Kenneth Rohde Christiansen.

* platform/qt/fast/forms/range/thumbslider-no-parent-slider-expected.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@93644 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago[chromium] Typo fix.
pkasting@chromium.org [Tue, 23 Aug 2011 22:35:00 +0000 (22:35 +0000)]
[chromium] Typo fix.

Unreviewed, test expectations update.

* platform/chromium/test_expectations.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@93643 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago[chromium] Revert unintentional changes to test expectations in r93638.
pkasting@chromium.org [Tue, 23 Aug 2011 22:22:24 +0000 (22:22 +0000)]
[chromium] Revert unintentional changes to test expectations in r93638.

Unreviewed, test expectations update.

* platform/chromium/test_expectations.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@93642 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agogarden-o-matic relative time is wrong and updates too frequently.
dglazkov@chromium.org [Tue, 23 Aug 2011 22:10:26 +0000 (22:10 +0000)]
garden-o-matic relative time is wrong and updates too frequently.
https://bugs.webkit.org/show_bug.cgi?id=66802

Reviewed by Adam Barth.

* BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/base.js: Changed to account for rounding correctly.
* BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/base_unittests.js: Updated tests.
* BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/config.js: Upped the interval to 1 minute.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@93641 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago[chromium] Fix potential nullptr deref in TiledLayerChromium::pushPropertiesTo
enne@google.com [Tue, 23 Aug 2011 22:09:23 +0000 (22:09 +0000)]
[chromium] Fix potential nullptr deref in TiledLayerChromium::pushPropertiesTo
https://bugs.webkit.org/show_bug.cgi?id=66808

Reviewed by James Robinson.

* platform/graphics/chromium/TiledLayerChromium.cpp:
(WebCore::TiledLayerChromium::pushPropertiesTo):
* platform/graphics/chromium/cc/CCTiledLayerImpl.cpp:
(WebCore::CCTiledLayerImpl::draw):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@93640 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoUnreviewed, rolling out r93452.
pkasting@chromium.org [Tue, 23 Aug 2011 21:51:57 +0000 (21:51 +0000)]
Unreviewed, rolling out r93452.
http://trac.webkit.org/changeset/93452
https://bugs.webkit.org/show_bug.cgi?id=66423

Broke Leopard, Snowleopard, and Chromium bots

Source/WebCore:

* xml/parser/XMLDocumentParserLibxml2.cpp:
(WebCore::XMLDocumentParser::XMLDocumentParser):

LayoutTests:

* fast/parser/innerhtml-with-prefixed-elements.xhtml: Removed.
* platform/chromium/test_expectations.txt:
* platform/mac/fast/parser/innerhtml-with-prefixed-elements-expected.png: Removed.
* platform/mac/fast/parser/innerhtml-with-prefixed-elements-expected.txt: Removed.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@93638 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoSource/WebKit: Patch by Jonas M. Gastal <jgastal@profusion.mobi> on 2011-08-23
demarchi@webkit.org [Tue, 23 Aug 2011 21:01:52 +0000 (21:01 +0000)]
Source/WebKit: Patch by Jonas M. Gastal <jgastal@profusion.mobi> on 2011-08-23
Reviewed by Antonio Gomes.

[EFL] Exporting JavaScript objects for EFL port
https://bugs.webkit.org/show_bug.cgi?id=43446

This patch allows to export objects to JavaScript API,
just like qt does, but it uses the netscape api
to do the convertion to javascript object.

* CMakeLists.txt:
Adding necessary files for the patch.

Source/WebKit/efl: [EFL] Exporting JavaScript objects for EFL port
https://bugs.webkit.org/show_bug.cgi?id=43446

Patch by Jonas M. Gastal <jgastal@profusion.mobi> on 2011-08-23
Reviewed by Antonio Gomes.

This patch allows to export objects to JavaScript API,
just like qt does, but it uses the netscape api
to do the convertion to javascript object.

* CMakeLists.txt:
* WebCoreSupport/FrameLoaderClientEfl.cpp:
(WebCore::FrameLoaderClientEfl::dispatchDidClearWindowObjectInWorld):
* ewk/EWebKit.h:
* ewk/ewk_js.cpp: Added.
(ewk_js_variant_to_npvariant):
(ewk_js_npvariant_to_variant):
(ewk_js_property_has):
(ewk_js_property_get):
(ewk_js_property_set):
(ewk_js_property_remove):
(ewk_js_properties_enumerate):
(ewk_js_method_has):
(ewk_js_method_invoke):
(ewk_js_class_new):
(ewk_js_class_free):
(ewk_js_object_new):
(ewk_js_object_free):
(ewk_js_object_properties_get):
(ewk_js_object_name_get):
(ewk_js_variant_free):
* ewk/ewk_js.h: Added.
* ewk/ewk_main.cpp:
(ewk_init):
* ewk/ewk_private.h:
* ewk/ewk_view.cpp:
(ewk_view_js_window_object_clear):
(ewk_view_js_object_add):
* ewk/ewk_view.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@93637 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago[Qt] Unnecessary folders in WebKit2.
andreas.kling@nokia.com [Tue, 23 Aug 2011 20:48:48 +0000 (20:48 +0000)]
[Qt] Unnecessary folders in WebKit2.
https://bugs.webkit.org/show_bug.cgi?id=66583

Patch by Gopal Raghavan <gopal.1.raghavan@nokia.com> on 2011-08-23
Reviewed by Benjamin Poulain.

Remove empty autotest folders that were left over from QGraphicsWKView times.

* UIProcess/API/qt/tests/qgraphicswkview: Removed.
* UIProcess/API/qt/tests/qwkhistory: Removed.
* UIProcess/API/qt/tests/qwkpage: Removed.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@93636 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago[chromium] Move result images into correct subdirectory.
pkasting@chromium.org [Tue, 23 Aug 2011 20:40:10 +0000 (20:40 +0000)]
[chromium] Move result images into correct subdirectory.
https://bugs.webkit.org/show_bug.cgi?id=66618

Unreviewed, file misplacement fix.

* platform/chromium-cg-mac/fast/repaint/scroll-absolute-layer-with-reflection-expected.png: Copied from platform/chromium-cg-mac/scroll-absolute-layer-with-reflection-expected.png.
* platform/chromium-cg-mac/fast/repaint/scroll-fixed-layer-with-reflection-expected.png: Copied from platform/chromium-cg-mac/scroll-fixed-layer-with-reflection-expected.png.
* platform/chromium-cg-mac/fast/repaint/scroll-fixed-layer-with-transformed-parent-layer-expected.png: Copied from platform/chromium-cg-mac/scroll-fixed-layer-with-transformed-parent-layer-expected.png.
* platform/chromium-cg-mac/fast/repaint/scroll-fixed-reflected-layer-expected.png: Copied from platform/chromium-cg-mac/scroll-fixed-reflected-layer-expected.png.
* platform/chromium-cg-mac/fast/repaint/scroll-in-clipped-layer-expected.png: Copied from platform/chromium-cg-mac/scroll-in-clipped-layer-expected.png.
* platform/chromium-cg-mac/fast/repaint/scroll-in-fixed-layer-expected.png: Copied from platform/chromium-cg-mac/scroll-in-fixed-layer-expected.png.
* platform/chromium-cg-mac/fast/repaint/scroll-in-transformed-layer-expected.png: Copied from platform/chromium-cg-mac/scroll-in-transformed-layer-expected.png.
* platform/chromium-cg-mac/fast/repaint/scroll-with-transformed-parent-layer-expected.png: Copied from platform/chromium-cg-mac/scroll-with-transformed-parent-layer-expected.png.
* platform/chromium-cg-mac/scroll-absolute-layer-with-reflection-expected.png: Removed.
* platform/chromium-cg-mac/scroll-fixed-layer-with-reflection-expected.png: Removed.
* platform/chromium-cg-mac/scroll-fixed-layer-with-transformed-parent-layer-expected.png: Removed.
* platform/chromium-cg-mac/scroll-fixed-reflected-layer-expected.png: Removed.
* platform/chromium-cg-mac/scroll-in-clipped-layer-expected.png: Removed.
* platform/chromium-cg-mac/scroll-in-fixed-layer-expected.png: Removed.
* platform/chromium-cg-mac/scroll-in-transformed-layer-expected.png: Removed.
* platform/chromium-cg-mac/scroll-with-transformed-parent-layer-expected.png: Removed.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@93635 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago[chromium] Remove a few FIXME #defines in API headers
adamk@chromium.org [Tue, 23 Aug 2011 20:19:09 +0000 (20:19 +0000)]
[chromium] Remove a few FIXME #defines in API headers
https://bugs.webkit.org/show_bug.cgi?id=66791

Reviewed by James Robinson.

* DEPS:
Roll chromium_rev to r97891.
* public/WebFileSystem.h:
#define made obsolete in http://crrev.com/97520.
* public/WebKitClient.h:
#define made obsolete in http://crrev.com/97524 and
http://crrev.com/97891.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@93632 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoFixed warnings produced by gcc-4.6.0.
commit-queue@webkit.org [Tue, 23 Aug 2011 20:10:46 +0000 (20:10 +0000)]
Fixed warnings produced by gcc-4.6.0.
https://bugs.webkit.org/show_bug.cgi?id=62168

Patch by Ahmad Sharif <asharif@chromium.org> on 2011-08-23
Reviewed by Adam Barth.

* dom/Element.cpp:
(WebCore::Element::recalcStyle):
* platform/ScrollAnimatorNone.cpp:
(WebCore::ScrollAnimatorNone::PerAxisData::curveAt):
(WebCore::ScrollAnimatorNone::PerAxisData::curveDerivativeAt):
* thirdparty: Copied from Source/WebCore/thirdparty.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@93631 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoAdd missing quotes around filenames in mergeChangeLogs
paroga@webkit.org [Tue, 23 Aug 2011 20:08:09 +0000 (20:08 +0000)]
Add missing quotes around filenames in mergeChangeLogs
https://bugs.webkit.org/show_bug.cgi?id=66794

Reviewed by Adam Roben.

* Scripts/VCSUtils.pm:
(mergeChangeLogs):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@93630 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago[chromium] Attempt to make test expectations more accurate.
pkasting@chromium.org [Tue, 23 Aug 2011 19:58:33 +0000 (19:58 +0000)]
[chromium] Attempt to make test expectations more accurate.

Unreviewed, test expectations update.

* platform/chromium/test_expectations.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@93629 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoText overlaps with the floats
commit-queue@webkit.org [Tue, 23 Aug 2011 19:54:28 +0000 (19:54 +0000)]
Text overlaps with the floats
https://bugs.webkit.org/show_bug.cgi?id=66671

Source/WebCore:

The problem was that LineWidth::fitBelowFloats is not updating the m_left and m_right fields.
Next time LineWidth::shrinkAvailableWidthForNewFloatIfNeeded was called the width was recalculated
using the old values, making the line bigger.

Patch by Alexandru Chiculita <achicu@adobe.com> on 2011-08-23
Reviewed by David Hyatt.

Test: fast/block/float/fit_line_below_floats.html

* rendering/RenderBlockLineLayout.cpp:
(WebCore::LineWidth::fitBelowFloats):

LayoutTests:

Patch by Alexandru Chiculita <achicu@adobe.com> on 2011-08-23
Reviewed by David Hyatt.

* fast/block/float/fit_line_below_floats.html: Added.
* platform/chromium/test_expectations.txt:
* platform/gtk/Skipped:
* platform/mac/fast/block/float/fit_line_below_floats-expected.png: Added.
* platform/mac/fast/block/float/fit_line_below_floats-expected.txt: Added.
* platform/qt/Skipped:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@93628 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago[CSSRegions] RenderFlowThread layout should use the attached region sizes
commit-queue@webkit.org [Tue, 23 Aug 2011 19:44:43 +0000 (19:44 +0000)]
[CSSRegions] RenderFlowThread layout should use the attached region sizes
https://bugs.webkit.org/show_bug.cgi?id=66143

Source/WebCore:

When RenderFlowThread is doing the layout it also sets the current RenderFlowThread on
the RenderView. RenderFlowThread cannot be nested inside a single RenderView, so there's no
need to put that on the LayoutState.

The right offset used to layout the inlines inside a RenderFlowThread
is now adjusted to fit inside the current region.

Patch by Alexandru Chiculita <achicu@adobe.com> on 2011-08-23
Reviewed by David Hyatt.

Tests: fast/regions/webkit-flow-floats-inside-regions-bounds.html
       fast/regions/webkit-flow-inlines-inside-regions-bounds-vertical-rl.html
       fast/regions/webkit-flow-inlines-inside-regions-bounds-vertical.html
       fast/regions/webkit-flow-inlines-inside-regions-bounds.html

* rendering/LayoutState.cpp:
(WebCore::LayoutState::LayoutState):
* rendering/RenderBlock.cpp:
(WebCore::RenderBlock::layoutBlock):
(WebCore::RenderBlock::adjustForRegionFittingIfNeeded):
(WebCore::RenderBlock::logicalRightOffsetForLine):
* rendering/RenderBlock.h:
* rendering/RenderFlowThread.cpp:
(WebCore::RenderFlowThread::RenderFlowThread):
(WebCore::CurrentRenderFlowThreadMaintainer::CurrentRenderFlowThreadMaintainer):
(WebCore::CurrentRenderFlowThreadMaintainer::~CurrentRenderFlowThreadMaintainer):
(WebCore::RenderFlowThread::layout):
(WebCore::RenderFlowThread::renderRegionForLine):
(WebCore::RenderFlowThread::regionLogicalWidthForLine):
* rendering/RenderFlowThread.h:
(WebCore::RegionFittingDisabler::RegionFittingDisabler):
(WebCore::RegionFittingDisabler::~RegionFittingDisabler):
* rendering/RenderRegion.cpp:
(WebCore::RenderRegion::layout):
* rendering/RenderRegion.h:
* rendering/RenderView.cpp:
(WebCore::RenderView::RenderView):
* rendering/RenderView.h:
(WebCore::RenderView::hasRenderFlowThread):
(WebCore::RenderView::currentRenderFlowThread):
(WebCore::RenderView::setCurrentRenderFlowThread):
(WebCore::RenderView::pushLayoutState):

LayoutTests:

Patch by Alexandru Chiculita <achicu@adobe.com> on 2011-08-23
Reviewed by David Hyatt.

* fast/regions/webkit-flow-floats-inside-regions-bounds-expected.txt: Added.
* fast/regions/webkit-flow-floats-inside-regions-bounds.html: Added.
* fast/regions/webkit-flow-inlines-inside-regions-bounds-expected.txt: Added.
* fast/regions/webkit-flow-inlines-inside-regions-bounds-vertical-expected.txt: Added.
* fast/regions/webkit-flow-inlines-inside-regions-bounds-vertical-rl-expected.txt: Added.
* fast/regions/webkit-flow-inlines-inside-regions-bounds-vertical-rl.html: Added.
* fast/regions/webkit-flow-inlines-inside-regions-bounds-vertical.html: Added.
* fast/regions/webkit-flow-inlines-inside-regions-bounds.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@93627 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoText-indent and floats push the text out of the container box
commit-queue@webkit.org [Tue, 23 Aug 2011 19:38:05 +0000 (19:38 +0000)]
Text-indent and floats push the text out of the container box
https://bugs.webkit.org/show_bug.cgi?id=66662

Source/WebCore:

Text indent was not propagated when a new float was pushed on the line.
It was only calculated when the logicalLeft/RightOffsetForLine was called.
Because of that RenderBlock::LineBreaker::nextLineBreak used a bigger width
than actually available and made the line render on top of the borders.

Patch by Alexandru Chiculita <achicu@adobe.com> on 2011-08-23
Reviewed by David Hyatt.

Tests: fast/block/float/floats-and-text-indent-rl.html
       fast/block/float/floats-and-text-indent.html

* rendering/RenderBlock.cpp:
(WebCore::RenderBlock::textIndentOffset):
(WebCore::RenderBlock::logicalLeftOffsetForLine):
(WebCore::RenderBlock::logicalRightOffsetForLine):
* rendering/RenderBlock.h:
* rendering/RenderBlockLineLayout.cpp:
(WebCore::LineWidth::shrinkAvailableWidthForNewFloatIfNeeded):

LayoutTests:

Patch by Alexandru Chiculita <achicu@adobe.com> on 2011-08-23
Reviewed by David Hyatt.

* fast/block/float/floats-and-text-indent-rl.html: Added.
* fast/block/float/floats-and-text-indent.html: Added.
* platform/chromium/test_expectations.txt:
* platform/gtk/Skipped:
* platform/mac/fast/block/float/floats-and-text-indent-expected.png: Added.
* platform/mac/fast/block/float/floats-and-text-indent-expected.txt: Added.
* platform/mac/fast/block/float/floats-and-text-indent-rl-expected.png: Added.
* platform/mac/fast/block/float/floats-and-text-indent-rl-expected.txt: Added.
* platform/qt/Skipped:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@93626 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoAdd shareResources flag to 3D graphics context attributes
commit-queue@webkit.org [Tue, 23 Aug 2011 19:30:52 +0000 (19:30 +0000)]
Add shareResources flag to 3D graphics context attributes
https://bugs.webkit.org/show_bug.cgi?id=66516

Patch by Al Patrick <apatrick@chromium.org> on 2011-08-23
Reviewed by James Robinson.

Source/WebCore:

* html/canvas/WebGLRenderingContext.cpp:
(WebCore::WebGLRenderingContext::create):
* platform/graphics/GraphicsContext3D.h:
(WebCore::GraphicsContext3D::Attributes::Attributes):
* platform/graphics/gpu/SharedGraphicsContext3D.cpp:
(WebCore::SharedGraphicsContext3D::create):

Source/WebKit/chromium:

* public/WebGraphicsContext3D.h:
(WebKit::WebGraphicsContext3D::Attributes::Attributes):
* src/GraphicsContext3DChromium.cpp:
(WebCore::GraphicsContext3DInternal::initialize):
* src/WebViewImpl.cpp:
(std::getCompositorContextAttributes):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@93625 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agohttps://bugs.webkit.org/show_bug.cgi?id=55347
barraclough@apple.com [Tue, 23 Aug 2011 19:28:23 +0000 (19:28 +0000)]
https://bugs.webkit.org/show_bug.cgi?id=55347
"name" and "message" enumerable on *Error.prototype

Reviewed by Sam Weinig.

Source/JavaScriptCore:

The default value of a NativeErrorPrototype's message
property is "", not the name of the error.

* runtime/NativeErrorConstructor.cpp:
(JSC::NativeErrorConstructor::NativeErrorConstructor):
* runtime/NativeErrorConstructor.h:
(JSC::NativeErrorConstructor::create):
(JSC::NativeErrorConstructor::constructorBody):
* runtime/NativeErrorPrototype.cpp:
(JSC::NativeErrorPrototype::NativeErrorPrototype):
(JSC::NativeErrorPrototype::constructorBody):
* runtime/NativeErrorPrototype.h:
(JSC::NativeErrorPrototype::create):
* runtime/StringPrototype.cpp:
(JSC::StringPrototype::StringPrototype):
* runtime/StringPrototype.h:
(JSC::StringPrototype::create):

LayoutTests:

* fast/js/exception-properties-expected.txt:
* fast/js/script-tests/exception-properties.js:
    - Add tests for value of RangeError.__proto__.{name|message}

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@93624 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agowebkit-perl tests fail on win32 Perl due to lack of list form of pipe open implementation
paroga@webkit.org [Tue, 23 Aug 2011 19:27:32 +0000 (19:27 +0000)]
webkit-perl tests fail on win32 Perl due to lack of list form of pipe open implementation
https://bugs.webkit.org/show_bug.cgi?id=49080

Reviewed by Adam Roben.

Use the string form of pipe open instead (like we do at all other places).

* Scripts/VCSUtils.pm:
(mergeChangeLogs):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@93623 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoNew MainFrame views should inherit fixed layout settings from existing MainFrame...
fsamuel@chromium.org [Tue, 23 Aug 2011 19:21:05 +0000 (19:21 +0000)]
New MainFrame views should inherit fixed layout settings from existing MainFrame view in Chromium.
https://bugs.webkit.org/show_bug.cgi?id=66780

Fixed layout mode is set from WebViewImpl, but new main FrameViews are created on loading new pages. The new page's FrameView should get its
fixed layout settings from the current page.

Reviewed by Darin Fisher.

* src/WebFrameImpl.cpp:
(WebKit::WebFrameImpl::createFrameView):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@93622 268f45cc-cd09-0410-ab3c-d52691b4dbfc