WebKit-https.git
11 years ago2009-10-19 Marshall Culpepper <mculpepper@appcelerator.com>
eric@webkit.org [Mon, 19 Oct 2009 22:28:41 +0000 (22:28 +0000)]
2009-10-19  Marshall Culpepper  <mculpepper@appcelerator.com>

        Reviewed by Eric Seidel.

        added cairo include and lib directories to debug_wincairo.vsprops
        https://bugs.webkit.org/show_bug.cgi?id=29831

        * win/tools/vsprops/debug_wincairo.vsprops:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@49817 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-10-19 Andrew Scherkus <scherkus@chromium.org>
eric@webkit.org [Mon, 19 Oct 2009 22:10:38 +0000 (22:10 +0000)]
2009-10-19  Andrew Scherkus  <scherkus@chromium.org>

        Reviewed by Eric Seidel.

        Replace boolean operator indentation example with an if statement.

        https://bugs.webkit.org/show_bug.cgi?id=30538

        * coding/coding-style.html: Replace boolean indenting example to use "if"

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@49816 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-10-17 Pavel Feldman <pfeldman@chromium.org>
pfeldman@chromium.org [Mon, 19 Oct 2009 22:00:40 +0000 (22:00 +0000)]
2009-10-17  Pavel Feldman  <pfeldman@chromium.org>

        Reviewed by Timothy Hatcher.

        Web Inspector: Add more stop characters into console completions.

        https://bugs.webkit.org/show_bug.cgi?id=30477

        * inspector/front-end/ConsoleView.js:
        (WebInspector.ConsoleView.prototype.completions):
        (WebInspector.ConsoleView.prototype._reportCompletions):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@49815 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoFix crashes/assertions when calling WebLocalizedString from multiple threads concurrently
aroben@apple.com [Mon, 19 Oct 2009 22:00:12 +0000 (22:00 +0000)]
Fix crashes/assertions when calling WebLocalizedString from multiple threads concurrently

Fixes <http://webkit.org/b/30534> WebLocalizedString asserts if called
from multiple threads concurrently

Reviewed by John Sullivan.

* WebLocalizableStrings.cpp:
(mainBundleLocStringsMutex):
(frameworkLocStringsMutex):
Added these new getters.

(findCachedString):
(cacheString):
Lock the relevant mutex before accessing each string map. Otherwise
bad things could happen if two threads end up here at the same time.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@49814 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoGet rid of a few static initializers/exit-time destructors in WebLocalizableStrings
aroben@apple.com [Mon, 19 Oct 2009 22:00:02 +0000 (22:00 +0000)]
Get rid of a few static initializers/exit-time destructors in WebLocalizableStrings

Reviewed by Dan Bernstein.

* WebLocalizableStrings.cpp:
(mainBundleLocStrings):
(frameworkLocStrings):
Added these functions instead of file-scope variables.

(findCachedString):
(cacheString):
Updated to use the new functions.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@49813 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-10-19 Nate Chapin <japhet@chromium.org>
japhet@chromium.org [Mon, 19 Oct 2009 21:41:07 +0000 (21:41 +0000)]
2009-10-19  Nate Chapin  <japhet@chromium.org>

        Unreviewed, build fix.

        Update call to FrameLoader::loadFrameRequest() in qt and
        FrameLoader::urlSelected() in win.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@49812 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-10-19 Daniel Bates <dbates@webkit.org>
dbates@webkit.org [Mon, 19 Oct 2009 21:39:57 +0000 (21:39 +0000)]
2009-10-19  Daniel Bates  <dbates@webkit.org>

        Reviewed by Eric Seidel.

        https://bugs.webkit.org/show_bug.cgi?id=30266

        Added the test case for bug #30266 to the Windows skip list because
        event.dataTransfer.types does not behave correctly. See bug #30527.

        * platform/win/Skipped: Added test case to Windows
        skipped list.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@49811 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-10-19 Marshall Culpepper <mculpepper@appcelerator.com>
eric@webkit.org [Mon, 19 Oct 2009 21:31:25 +0000 (21:31 +0000)]
2009-10-19  Marshall Culpepper  <mculpepper@appcelerator.com>

        Reviewed by Adam Roben.

        implemented ClipboardWin::files()
        https://bugs.webkit.org/show_bug.cgi?id=29666

        * platform/win/ClipboardWin.cpp:
        (WebCore::ClipboardWin::files):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@49810 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-10-19 Nate Chapin <japhet@chromium.org>
japhet@chromium.org [Mon, 19 Oct 2009 21:15:06 +0000 (21:15 +0000)]
2009-10-19  Nate Chapin  <japhet@chromium.org>

        Reviewed by Alexey Proskuryakov.

        Add support for link relations in HTMLAnchorElement and implement rel="noreferrer".

        https://bugs.webkit.org/show_bug.cgi?id=28986

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@49809 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoDocumentation fix for -[NSObject(WebResourceLoadDelegate) webView:resource:willSendRe...
ddkilzer@apple.com [Mon, 19 Oct 2009 21:08:24 +0000 (21:08 +0000)]
Documentation fix for -[NSObject(WebResourceLoadDelegate) webView:resource:willSendRequest:redirectResponse:fromDataSource:]

Reviewed by Dan Bernstein.

* WebView/WebResourceLoadDelegate.h: Prepended "webView:" to the
@method declaration of
-webView:resource:willSendRequest:redirectResponse:fromDataSource:
which was absent before r3953 and was missed during the great
documentation clean up of r4269.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@49808 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoFix build.
andersca@apple.com [Mon, 19 Oct 2009 21:01:00 +0000 (21:01 +0000)]
Fix build.

* Plugins/WebNetscapePluginView.mm:
(-[WebNetscapePluginView saveAndSetNewPortStateForUpdate:]):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@49807 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoSplit assertions of conjunctions into separate statements or remove
mitz@apple.com [Mon, 19 Oct 2009 20:30:27 +0000 (20:30 +0000)]
Split assertions of conjunctions into separate statements or remove
redundant assertions.

Reviewed by Anders Carlsson.

* Plugins/Hosted/NetscapePluginHostProxy.mm:
(WebKit::NetscapePluginHostProxy::deadNameNotificationCallback):
* Plugins/Hosted/WebHostedNetscapePluginView.mm:
(-[WebHostedNetscapePluginView setAttributeKeys:andValues:]):
(-[WebHostedNetscapePluginView _webPluginContainerCancelCheckIfAllowedToLoadRequest:]):
* Plugins/WebBaseNetscapePluginView.mm:
(-[WebBaseNetscapePluginView halt]):
(-[WebBaseNetscapePluginView resumeFromHalt]):
* Plugins/WebNetscapePluginView.mm:
(-[WebNetscapePluginView saveAndSetNewPortStateForUpdate:]):
(-[WebNetscapePluginView _webPluginContainerCancelCheckIfAllowedToLoadRequest:]):
* WebCoreSupport/WebContextMenuClient.mm:
(WebContextMenuClient::getCustomMenuFromDefaultItems):
* WebCoreSupport/WebEditorClient.mm:
(WebEditorClient::checkGrammarOfString):
(WebEditorClient::checkTextOfParagraph):
* WebView/WebTextCompletionController.mm:
(-[WebTextCompletionController _reflectSelection]):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@49806 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-10-19 Nate Chapin <japhet@chromium.org>
japhet@chromium.org [Mon, 19 Oct 2009 20:25:51 +0000 (20:25 +0000)]
2009-10-19  Nate Chapin  <japhet@chromium.org>

        Reviewed by Geoffrey Garen.

        Add plugins/netscape-invoke-failures.html to Snow Leopard skipped list.

        https://bugs.webkit.org/show_bug.cgi?id=30528

        * platform/mac-snowleopard/Skipped: Add plugins/netscape-invoke-failures.html.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@49805 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoRefactor out some of the code in paint() into paintPanScrollIcon()
kenneth@webkit.org [Mon, 19 Oct 2009 20:25:25 +0000 (20:25 +0000)]
Refactor out some of the code in paint() into paintPanScrollIcon()
and paintScrollbars() in preparation of some Qt API changes.

Patch by Kenneth Rohde Christiansen <kenneth@webkit.org> on 2009-10-19
Reviewed by David Hyatt.

* platform/ScrollView.cpp:
(WebCore::ScrollView::paint):
* platform/ScrollView.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@49804 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago[v8] typeof(HTMLMediaElement) should return undefined if media
hclam@chromium.org [Mon, 19 Oct 2009 20:18:18 +0000 (20:18 +0000)]
[v8] typeof(HTMLMediaElement) should return undefined if media
engine is not available
https://bugs.webkit.org/show_bug.cgi?id=30343

Patch by Alpha Lam <hclam@chromium.org> on 2009-10-19
Reviewed by Dimitri Glazkov (dglazkov@chromium.org).

Check for availability of the media engine to disable
HTMLMediaElement, HTMLAudioElement, HTMLVideoElement, MediaError
in runtime.

Try runs and review in Chromium:
http://codereview.chromium.org/276011

Test: manual-tests/chromium/media-player-not-available.html

The above test can work work in Chromium. Testing procedures:
1. Remove all media support libraries in Chromium (e.g. ffmpeg libraries)
2. Open Chromium with above test page
3. Verify the test results according to the test page

* bindings/v8/custom/V8CustomBinding.h:
Declare enabler methods.
* bindings/v8/custom/V8DOMWindowCustom.cpp:
Report enabled only if media player is available.
* manual-tests/chromium/media-player-not-available.html: Added.
Manual test, following instruction in this test for procedures.
* page/DOMWindow.idl:
Mark HTMLMediaElement, HTMLAudioElement, HTMLVideElement, MediaError
as enabled at runtime.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@49803 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoAdded a private API for getting a global context from a context, for
ggaren@apple.com [Mon, 19 Oct 2009 19:59:32 +0000 (19:59 +0000)]
Added a private API for getting a global context from a context, for
clients who want to preserve a context for a later callback.

Patch by Geoffrey Garen <ggaren@apple.com> on 2009-10-19
Reviewed by Sam Weinig.

* API/APICast.h:
(toGlobalRef): Added an ASSERT, since this function is used more often
than before.

* API/JSContextRef.cpp:
* API/JSContextRefPrivate.h: Added. The new API.

* API/tests/testapi.c:
(print_callAsFunction):
(main): Test the new API.

* JavaScriptCore.exp:
* JavaScriptCore.xcodeproj/project.pbxproj: Build and export the new API.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@49802 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-10-19 Jan Michael Alonzo <jmalonzo@webkit.org>
jmalonzo@webkit.org [Mon, 19 Oct 2009 19:38:36 +0000 (19:38 +0000)]
2009-10-19  Jan Michael Alonzo  <jmalonzo@webkit.org>

        Reviewed by Alexey Proskuryakov.

        [Gtk] ApplicationCacheStorage error while compiling WebKit/gtk/webkit/webkitapplicationcache.cpp
        https://bugs.webkit.org/show_bug.cgi?id=30358

        Guard calls to cacheStorage() with ENABLE(OFFLINE_WEB_APPLICATIONS).

        * webkit/webkitapplicationcache.cpp:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@49801 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoMake the local static panScrollIcon into a Image* instead
kenneth@webkit.org [Mon, 19 Oct 2009 19:24:34 +0000 (19:24 +0000)]
Make the local static panScrollIcon into a Image* instead
of a RefPtr<Image>.

Patch by Kenneth Rohde Christiansen <kenneth@webkit.org> on 2009-10-19
Reviewed by Adam Roben.

* platform/ScrollView.cpp:
(WebCore::ScrollView::wheelEvent):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@49800 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoForcefully set antialiasing for Chromium media controls.
ajwong@chromium.org [Mon, 19 Oct 2009 19:19:10 +0000 (19:19 +0000)]
Forcefully set antialiasing for Chromium media controls.

Patch by Andrew Scherkus <scherkus@chromium.org> on 2009-10-19
Reviewed by NOBODY (OOPS!).

https://bugs.webkit.org/show_bug.cgi?id=30521

Before it was nondeterministiacally being enabled based on the previous state of GraphicsContext.

Covered by existing layout tests.

* rendering/RenderMediaControlsChromium.cpp:
(WebCore::paintMediaSlider): Added setShouldAntialias(true).

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@49799 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-10-19 Jens Alfke <jens@mooseyard.com>
eric@webkit.org [Mon, 19 Oct 2009 19:11:37 +0000 (19:11 +0000)]
2009-10-19  Jens Alfke  <jens@mooseyard.com>

        Reviewed by Darin Adler.

        Optimize string upper/lowercasing
        https://bugs.webkit.org/show_bug.cgi?id=30261

        - Added AtomicString::upper() and lower()
        - Further optimized StringImpl::lower()
        - Removed StringImpl::isLower()
        - Added QualifiedName::localNameUpper(), which is cached, thereby saving
          thousands of upper() calls and string allocations.

         * dom/Element.cpp:
         (WebCore::Element::setAttribute): Call AtomicString::lower()
         * dom/QualifiedName.cpp:
         (WebCore::QualifiedName::localNameUpper): New method
         * dom/QualifiedName.h: Added localNameUpper() method
         * dom/StyledElement.cpp:
         (WebCore::StyledElement::parseMappedAttribute):  Call AtomicString::lower()
         * html/HTMLDocument.cpp:
         (WebCore::HTMLDocument::createElement): Call AtomicString::lower()
         * html/HTMLElement.cpp:
         (WebCore::HTMLElement::nodeName): Call localNameUpper()
         * platform/text/AtomicString.cpp:
         (WebCore::AtomicString::lower): New method
         (WebCore::AtomicString::upper): New method
         * platform/text/AtomicString.h: Added lower() and upper()
         * platform/text/StringImpl.cpp: Removed isLower()
         (WebCore::StringImpl::lower): Further optimization of initial loop
         * platform/text/StringImpl.h: Removed isLower()

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@49798 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-10-19 Zan Dobersek <zandobersek@gmail.com>
eric@webkit.org [Mon, 19 Oct 2009 18:59:27 +0000 (18:59 +0000)]
2009-10-19  Zan Dobersek  <zandobersek@gmail.com>

        Reviewed by Gustavo Noronha.

        Update test results and Skipped list after changes in r49791.

        Some test results were accidentally committed in r49192. At that time
        results were incorrect, but testing still succeeded because of
        unimplemented features. After r49791 tests, though generating semantically
        equal output to Mac's render trees, started to fail and need to have
        generated results updated. This patch does just that.

        * platform/gtk/Skipped: Skip a test that does not generate semantically
        equal render trees.
        * platform/gtk/editing/inserting/insert-3907422-fix-expected.txt:
        * platform/gtk/editing/pasteboard/4076267-2-expected.txt:
        * platform/gtk/editing/pasteboard/4076267-3-expected.txt:
        * platform/gtk/editing/pasteboard/4076267-expected.txt:
        * platform/gtk/editing/pasteboard/paste-table-001-expected.txt:
        * platform/gtk/editing/pasteboard/paste-text-001-expected.txt:
        * platform/gtk/editing/pasteboard/paste-text-002-expected.txt:
        * platform/gtk/editing/pasteboard/paste-text-003-expected.txt:
        * platform/gtk/editing/pasteboard/paste-text-004-expected.txt:
        * platform/gtk/editing/pasteboard/paste-text-005-expected.txt:
        * platform/gtk/editing/pasteboard/paste-text-006-expected.txt:
        * platform/gtk/editing/pasteboard/paste-text-007-expected.txt:
        * platform/gtk/editing/pasteboard/paste-text-008-expected.txt:
        * platform/gtk/editing/pasteboard/paste-text-009-expected.txt:
        * platform/gtk/editing/pasteboard/paste-text-010-expected.txt:
        * platform/gtk/editing/pasteboard/paste-text-015-expected.txt:
        * platform/gtk/editing/pasteboard/paste-xml-expected.txt:
        * platform/gtk/editing/pasteboard/select-element-1-expected.txt: Removed.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@49797 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-10-19 Nate Chapin <japhet@chromium.org>
eric@webkit.org [Mon, 19 Oct 2009 18:50:26 +0000 (18:50 +0000)]
2009-10-19  Nate Chapin  <japhet@chromium.org>

        Reviewed by Adam Barth.

        Test that the proper exception is thrown into JS when a plugin
        invoke call returns false.

        https://bugs.webkit.org/show_bug.cgi?id=30239

        * plugins/netscape-invoke-failure-expected.txt: Added.
        * plugins/netscape-invoke-failure.html: Added.
2009-10-19  Nate Chapin  <japhet@chromium.org>

        Reviewed by Adam Barth.

        If a call to a plugin's invokeMethod, invokeDefault or construct
        returns false, throw an exception into JS.

        https://bugs.webkit.org/show_bug.cgi?id=30239

        Test: plugins/netscape-invoke-failure.html

        * bindings/v8/V8NPObject.cpp: Check return values of invokeMethod, invokeDefault and construct.
        (npObjectInvokeImpl):
        * bridge/c/c_instance.cpp:
        (JSC::Bindings::CInstance::invokeMethod): Check return value
        (JSC::Bindings::CInstance::invokeDefaultMethod): Check return value
        (JSC::Bindings::CInstance::invokeConstruct): Check return value
2009-10-19  Nate Chapin  <japhet@chromium.org>

        Reviewed by Adam Barth.

        Add testFail() to test plugin so we can test our handling of a
        plugin invoke call returning false.

        https://bugs.webkit.org/show_bug.cgi?id=30239

        * DumpRenderTree/TestNetscapePlugIn.subproj/PluginObject.cpp: Add testFail().
        (pluginInvoke):
        (testIdentifierToString): Always return true, since returning false will now cause an exception to be thrown.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@49796 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoTightened up some instanceof code generation.
ggaren@apple.com [Mon, 19 Oct 2009 18:47:40 +0000 (18:47 +0000)]
Tightened up some instanceof code generation.
https://bugs.webkit.org/show_bug.cgi?id=30488

Patch by Geoffrey Garen <ggaren@apple.com> on 2009-10-17
Reviewed by Oliver Hunt.

* jit/JITOpcodes.cpp:
(JSC::JIT::emit_op_instanceof):
(JSC::JIT::emitSlow_op_instanceof): No need to do object type checks -
cell type checks and ImplementsDefaultHasIntance checks implicitly
supersede object type checks.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@49795 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoWebCore: Fix for chromium/skia canvas arc start/end positions, when the arc is
senorblanco@chromium.org [Mon, 19 Oct 2009 18:38:23 +0000 (18:38 +0000)]
WebCore: Fix for chromium/skia canvas arc start/end positions, when the arc is
>= 360 degrees.
https://bugs.webkit.org/show_bug.cgi?id=30449

Reviewed by Adam Barth.

Covered by new test:
LayoutTests/fast/canvas/arc360.html

* platform/graphics/skia/PathSkia.cpp:
(WebCore::Path::addArc):

LayoutTests: New test for 360 degree arc endpoints, and expected results.
http://bugs.webkit.org/show_bug.cgi?id=30449

Reviewed by Adam Barth.

* fast/canvas/arc360.html: Added.
* platform/mac/fast/canvas/arc360-expected.checksum: Added.
* platform/mac/fast/canvas/arc360-expected.png: Added.
* platform/mac/fast/canvas/arc360-expected.txt: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@49794 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoRubber-stamped by Adam Roben.
bweinstein@apple.com [Mon, 19 Oct 2009 18:14:55 +0000 (18:14 +0000)]
Rubber-stamped by Adam Roben.

Add tests that have never passed on Windows to the Skipped list, and reopen
bugs.

* platform/win/Skipped:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@49793 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-10-19 Kent Tamura <tkent@chromium.org>
eric@webkit.org [Mon, 19 Oct 2009 17:35:59 +0000 (17:35 +0000)]
2009-10-19  Kent Tamura  <tkent@chromium.org>

        Reviewed by Alexey Proskuryakov.

        Fix uninitialized variable reference in Element::removeAttribute().
        https://bugs.webkit.org/show_bug.cgi?id=30502

        * dom/Element.cpp:
        (WebCore::Element::removeAttribute):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@49792 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-10-19 Zan Dobersek <zandobersek@gmail.com>
eric@webkit.org [Mon, 19 Oct 2009 17:25:37 +0000 (17:25 +0000)]
2009-10-19  Zan Dobersek  <zandobersek@gmail.com>

        Reviewed by Gustavo Noronha.

        Enable tests that pass when DOM pasting is allowed.

        * platform/gtk/Skipped:
2009-10-19  Zan Dobersek  <zandobersek@gmail.com>

        Reviewed by Gustavo Noronha.

        Add a settings to allow DOM pastes to be performed.

        * webkit/webkitwebsettings.cpp:
        (webkit_web_settings_class_init):
        (webkit_web_settings_set_property):
        (webkit_web_settings_get_property):
        (webkit_web_settings_copy):
        * webkit/webkitwebview.cpp:
        (webkit_web_view_update_settings):
        (webkit_web_view_settings_notify):
2009-10-19  Zan Dobersek  <zandobersek@gmail.com>

        Reviewed by Gustavo Noronha.

        Enable DOM pasting when running layout tests.

        * DumpRenderTree/gtk/DumpRenderTree.cpp:
        (resetDefaultsToConsistentValues):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@49791 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-10-19 Eric Carlson <eric.carlson@apple.com>
eric.carlson@apple.com [Mon, 19 Oct 2009 17:13:08 +0000 (17:13 +0000)]
2009-10-19  Eric Carlson  <eric.carlson@apple.com>

        Reviewed by Simon Fraser.

        https://bugs.webkit.org/show_bug.cgi?id=30463
        Remove HTML5 media element 'loadend' event

        * dom/EventNames.h:
        * html/HTMLMediaElement.cpp:
        (WebCore::HTMLMediaElement::parseMappedAttribute): Remove 'loadend'.
        (WebCore::HTMLMediaElement::loadInternal): Ditto.
        (WebCore::HTMLMediaElement::noneSupported): Ditto.
        (WebCore::HTMLMediaElement::mediaEngineError): Ditto.
        (WebCore::HTMLMediaElement::setNetworkState): Ditto.
        (WebCore::HTMLMediaElement::userCancelledLoad): Ditto.

2009-10-19  Eric Carlson  <eric.carlson@apple.com>

        Reviewed by Simon Fraser.

        https://bugs.webkit.org/show_bug.cgi?id=30463
        Remove HTML5 media element 'loadend' event

        * media/event-attributes-expected.txt:
        * media/event-attributes.html: Switch from 'loadend' to 'canplaythrough'.
        * media/media-load-event-expected.txt:
        * media/media-load-event.html: Ditto.
        * media/progress-event-at-least-one-expected.txt:
        * media/progress-event-at-least-one.html: Ditto.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@49790 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-10-19 Jocelyn Turcotte <jocelyn.turcotte@nokia.com>
eric@webkit.org [Mon, 19 Oct 2009 16:26:02 +0000 (16:26 +0000)]
2009-10-19  Jocelyn Turcotte  <jocelyn.turcotte@nokia.com>

        Reviewed by Adam Barth.

        [Qt] Remove deletion of outgoing FormData object when reply is finished.
        The buffers will be destroyed a bit later, when the QNetworkReply is deleted.
        Bug: https://bugs.webkit.org/show_bug.cgi?id=29551

        * platform/network/qt/QNetworkReplyHandler.cpp:
        * platform/network/qt/QNetworkReplyHandler.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@49789 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-10-19 Dimitri Glazkov <dglazkov@chromium.org>
eric@webkit.org [Mon, 19 Oct 2009 16:18:19 +0000 (16:18 +0000)]
2009-10-19  Dimitri Glazkov  <dglazkov@chromium.org>

        Reviewed by Darin Adler.

        Fix hard-to-reproduce crash in HTMLTokenizer by avoiding a rare
        fastRealloc edge case.
        https://bugs.webkit.org/show_bug.cgi?id=29313

        No test, the crash shows up occasionally in crash dumps, we weren't able
        to reproduce it locally.

        * html/HTMLTokenizer.cpp:
        (WebCore::HTMLTokenizer::enlargeScriptBuffer): Added an early exit to
            avoid calling fastRealloc with the size of 0.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@49788 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-10-19 Andrew Scherkus <scherkus@chromium.org>
eric@webkit.org [Mon, 19 Oct 2009 16:08:22 +0000 (16:08 +0000)]
2009-10-19  Andrew Scherkus  <scherkus@chromium.org>

        Reviewed by Eric Seidel.

        Use fillRect() instead of drawRect() to fix Chromium media controls rendering.

        https://bugs.webkit.org/show_bug.cgi?id=30371

        Chromium Mac layout tests will need to be rebaselined.  Existing layout tests for Linux/Win still pass.

        * rendering/RenderMediaControlsChromium.cpp:
        (WebCore::paintMediaSlider): Use fillRect() instead of drawRect().

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@49787 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-10-19 Dmitry Titov <dimich@chromium.org>
eric@webkit.org [Mon, 19 Oct 2009 16:00:18 +0000 (16:00 +0000)]
2009-10-19  Dmitry Titov  <dimich@chromium.org>

        Reviewed by Adam Barth.

        Crash in DOMWindow::clearTimeout etc when DOMWindow is not connected to a frame.
        https://bugs.webkit.org/show_bug.cgi?id=29832

        Need to make sure the script caches are reset when frame gets disconnected from still-alive DOMWindow.
        This will prevent JS from calling DOMWindow methods that can not be completed w/o the frame.

        I am not sure it's possible to test this since the only file that repros the problem need ~10 seconds to cause crash.

        * page/Frame.cpp:
        (WebCore::Frame::~Frame): Right after frame disconnects from DOMWindow, clear WindowShell.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@49786 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-10-19 Viatcheslav Ostapenko <ostapenko.viatcheslav@nokia.com>
laszlo.1.gombos@nokia.com [Mon, 19 Oct 2009 15:53:36 +0000 (15:53 +0000)]
2009-10-19  Viatcheslav Ostapenko  <ostapenko.viatcheslav@nokia.com>

        Reviewed by Ariya Hidayat.

        Add QWebElement::render API which allows rendering of single
        element.

        * Api/qwebelement.cpp:
        (QWebElement::render):
        * Api/qwebelement.h:
        * tests/qwebelement/tst_qwebelement.cpp:
        (tst_QWebElement::render):
        * tests/qwebelement/qwebelement.qrc:
        * tests/qwebelement/image.png: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@49785 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoQWebPage: Doc: setNetworkAccessManager should only be called once.
ariya@webkit.org [Mon, 19 Oct 2009 15:16:52 +0000 (15:16 +0000)]
QWebPage: Doc: setNetworkAccessManager should only be called once.

Patch by Markus Goetz <Markus.Goetz@nokia.com> on 2009-10-19
Reviewed by Ariya Hidayat.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@49784 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-10-19 Csaba Osztrogonác <ossy@webkit.org>
ossy@webkit.org [Mon, 19 Oct 2009 15:02:33 +0000 (15:02 +0000)]
2009-10-19  Csaba Osztrogonác  <ossy@webkit.org>

        Unreviewed build warning fix.
        [Qt] Apply changes to WebCore.pro introduced in r49778

        * WebCore.pro: Removed svg/graphics/filters/SVGFEGaussianBlur.h from HEADERS

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@49783 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoWrong ifdef combination in QGraphicsWebView's event method.
tonikitoo@webkit.org [Mon, 19 Oct 2009 15:01:26 +0000 (15:01 +0000)]
Wrong ifdef combination in QGraphicsWebView's event method.

Patch by Antonio Gomes <tonikitoo@webkit.org> on 2009-10-19
Reviewed by Tor Arne.

* Api/qgraphicswebview.cpp:
(QGraphicsWebView::event):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@49782 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago[Qt] Skip failing tests for now and add test-specific information about the cause...
ossy@webkit.org [Mon, 19 Oct 2009 14:42:56 +0000 (14:42 +0000)]
[Qt] Skip failing tests for now and add test-specific information about the cause to the Skipped list.

Patch by Andras Becsi <becsi.andras@stud.u-szeged.hu> on 2009-10-19
Reviewed by Tor Arne Vestbø.

* platform/qt/Skipped:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@49781 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoMake WebKitHitTestResult actually appear in the index
kov@webkit.org [Mon, 19 Oct 2009 14:16:50 +0000 (14:16 +0000)]
Make WebKitHitTestResult actually appear in the index

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@49780 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago[Qt] Add platform expected file for canvas test.
ossy@webkit.org [Mon, 19 Oct 2009 14:00:23 +0000 (14:00 +0000)]
[Qt] Add platform expected file for canvas test.

Patch by Andras Becsi <becsi.andras@stud.u-szeged.hu> on 2009-10-19
Reviewed by Tor Arne Vestbø.

* platform/qt/fast/canvas/canvas-empty-image-pattern-expected.txt: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@49779 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-10-19 Dirk Schulze <krit@webkit.org>
krit@webkit.org [Mon, 19 Oct 2009 12:57:43 +0000 (12:57 +0000)]
2009-10-19  Dirk Schulze  <krit@webkit.org>

        Reviewed by Darin Adler.

        Move feGaussianBlur from WebCore/svg to WebCore/platform
        [https://bugs.webkit.org/show_bug.cgi?id=30495]

        This patch moves SVGFRGaussianBlur from svg/graphics/filters
        to platform/graphics/filters/FEGaussianBlur. This is needed
        for shadow support on Qt and Cairo.
        No change in functionality, therfore no new test case needed.

        * GNUmakefile.am:
        * WebCore.gypi:
        * WebCore.pro:
        * WebCore.vcproj/WebCore.vcproj:
        * WebCore.xcodeproj/project.pbxproj:
        * platform/graphics/filters/FEGaussianBlur.cpp: Added.
        * platform/graphics/filters/FEGaussianBlur.h: Added.
        * svg/SVGFEGaussianBlurElement.h:
        * svg/graphics/filters/SVGFEGaussianBlur.cpp: Removed.
        * svg/graphics/filters/SVGFEGaussianBlur.h: Removed.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@49778 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoUnreviewed.
yael.aharon@nokia.com [Mon, 19 Oct 2009 12:49:55 +0000 (12:49 +0000)]
Unreviewed.

Patch by Yael Aharon <yael.aharon@nokia.com> on 2009-10-19
Added reviewer name to r49574.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@49777 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-10-19 Andras Becsi <becsi.andras@stud.u-szeged.hu>
ossy@webkit.org [Mon, 19 Oct 2009 12:48:42 +0000 (12:48 +0000)]
2009-10-19  Andras Becsi  <becsi.andras@stud.u-szeged.hu>

        Reviewed by Tor Arne Vestbø.

        [Qt] Update outdated metrics for these passing tests.

        * platform/qt/css1/font_properties/font-expected.txt:
        * platform/qt/fast/xsl/document-function-expected.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@49776 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-10-19 Joanmarie Diggs <joanmarie.diggs@gmail.com>
eric@webkit.org [Mon, 19 Oct 2009 11:58:38 +0000 (11:58 +0000)]
2009-10-19  Joanmarie Diggs  <joanmarie.diggs@gmail.com>

        Reviewed by Xan Lopez.

        https://bugs.webkit.org/show_bug.cgi?id=30489
        [Gtk] The document frame/html container claims to be parentless

        Work around for the parentless claim made by the accessible associated
        with the embedded WebView. When we identify this condition, get the
        widget's parent, and then get the AtkObject from that. Admittedly an ugly
        hack, but we'll need it for Yelp and other Gtk+/GNOME applications which
        plan to switch over to WebKit.

        * accessibility/gtk/AccessibilityObjectWrapperAtk.cpp:
        (webkit_accessible_get_parent):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@49775 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-10-19 Benjamin Otte <otte@gnome.org>
jmalonzo@webkit.org [Mon, 19 Oct 2009 11:51:09 +0000 (11:51 +0000)]
2009-10-19  Benjamin Otte  <otte@gnome.org>

        Reviewed by Jan Alonzo.

        Handle duration queries properly
        https://bugs.webkit.org/show_bug.cgi?id=29999

        Previously duration queries failed to handle unknown duration
        and didn't treat the returned duration as an unsigned.

        * platform/graphics/gtk/MediaPlayerPrivateGStreamer.cpp:
        (WebCore::MediaPlayerPrivate::duration):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@49774 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-10-19 Daniel Macks <dmacks@netspace.org>
eric@webkit.org [Mon, 19 Oct 2009 11:50:55 +0000 (11:50 +0000)]
2009-10-19  Daniel Macks  <dmacks@netspace.org>

        Reviewed by Adam Barth.

        Inconsistent handling of gettext domain can cause compile failure
        https://bugs.webkit.org/show_bug.cgi?id=30292

        * GNUmakefile.am: Use DOMAIN variable instead of hard-coded value

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@49773 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-10-19 Jocelyn Turcotte <jocelyn.turcotte@nokia.com>
eric@webkit.org [Mon, 19 Oct 2009 11:28:01 +0000 (11:28 +0000)]
2009-10-19  Jocelyn Turcotte  <jocelyn.turcotte@nokia.com>

        Reviewed by Tor Arne Vestbø.

        [Qt] Fix build on Windows.

        * plugins/win/PluginViewWin.cpp:
        (windowHandleForPageClient):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@49772 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoWindowed Plugins: Don't crash when client is 0.
zecke@webkit.org [Mon, 19 Oct 2009 11:17:42 +0000 (11:17 +0000)]
Windowed Plugins: Don't crash when client is 0.

Client is 0 when we use QWebPage without a QWebView or QGraphicsWebView.
In addition, setFrameRect()/updatePluginWidget() is called even if the
plugin was not succesfully loaded. updatePluginWidget() updates the
window rect which is, in theory, useful to draw something that indicates
that we didn't load successfully.

So, a status check is added to setNPWindowIfNeeded.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@49771 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-10-19 Girish Ramakrishnan <girish@forwardbias.in>
eric@webkit.org [Mon, 19 Oct 2009 10:44:43 +0000 (10:44 +0000)]
2009-10-19  Girish Ramakrishnan  <girish@forwardbias.in>

        Reviewed by Holger Freyther.

        [Qt] Windowed Plugins: Fix crash when QWebPage is deleted after QWebView.

        Fixes various sources of crashes:
        1. The PluginContainer is a child of QWebView. When the view gets deleted,
        the PluginView is not notified about the deletion of PluginContainer.
        2. QWebView destructor does not set client to 0.
        3. Sometimes pending paint events are sent after the plugin has died, so add
        a check in PluginView::setNPWindowIfNeeded.

        https://bugs.webkit.org/show_bug.cgi?id=30354

        * plugins/qt/PluginContainerQt.cpp:
        (PluginContainerQt::~PluginContainerQt):
        * plugins/qt/PluginViewQt.cpp:
        (WebCore::PluginView::setNPWindowIfNeeded):
2009-10-19  Girish Ramakrishnan  <girish@forwardbias.in>

        Reviewed by Holger Freyther.

        [Qt] Windowed Plugins: Fix crash when QWebPage is deleted after QWebView.

        Fixes various sources of crashes:
        1. The PluginContainer is a child of QWebView. When the view gets deleted,
        the PluginView is not notified about the deletion of PluginContainer.
        2. QWebView destructor does not set client to 0.
        3. Sometimes pending paint events are sent after the plugin has died, so add
        a check in PluginView::setNPWindowIfNeeded.

        https://bugs.webkit.org/show_bug.cgi?id=30354

        * Api/qwebview.cpp:
        (QWebView::~QWebView):
        * tests/qwebview/qwebview.pro:
        * tests/qwebview/tst_qwebview.cpp:
        (tst_QWebView::reusePage_data):
        (tst_QWebView::reusePage):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@49770 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-10-19 Jakob Truelsen <antialize@gmail.com>
eric@webkit.org [Mon, 19 Oct 2009 07:17:40 +0000 (07:17 +0000)]
2009-10-19  Jakob Truelsen  <antialize@gmail.com>

        Reviewed by Adam Barth.

        https://bugs.webkit.org/show_bug.cgi?id=29042
        Allow one to customize the minimal and maximal shrink factors used when printing.

        * page/PrintContext.cpp:
        (WebCore::PrintContext::begin):
        * page/Settings.cpp:
        (WebCore::Settings::Settings):
        (WebCore::Settings::setPrintingMinimumShrinkFactor):
        (WebCore::Settings::setPrintingMaximumShrinkFactor):
        * page/Settings.h:
        (WebCore::Settings::printingMinimumShrinkFactor):
        (WebCore::Settings::printingMaximumShrinkFactor):
2009-10-19  Jakob Truelsen  <antialize@gmail.com>

        Reviewed by Adam Barth.

        https://bugs.webkit.org/show_bug.cgi?id=29042

        Allow one to costumize the minimal and maximal shrink factors,
        Added methods setPrintingMinimumShrinkFactor, printingMinimumShrinkFactor,
        setPrintingMaximumShrinkFactor, printingMaximumShrinkFactor to QWebSettings.

        * Api/qwebsettings.cpp:
        (QWebSettingsPrivate::apply):
        (QWebSettings::QWebSettings):
        (QWebSettings::setPrintingMinimumShrinkFactor):
        (QWebSettings::printingMinimumShrinkFactor):
        (QWebSettings::setPrintingMaximumShrinkFactor):
        (QWebSettings::printingMaximumShrinkFactor):
        * Api/qwebsettings.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@49769 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoFix ChangeLog spelling.
abarth@webkit.org [Mon, 19 Oct 2009 07:04:09 +0000 (07:04 +0000)]
Fix ChangeLog spelling.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@49768 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-10-18 Erik Arvidsson <arv@chromium.org>
eric@webkit.org [Mon, 19 Oct 2009 06:59:58 +0000 (06:59 +0000)]
2009-10-18  Erik Arvidsson  <arv@chromium.org>

        Reviewed by Adam Barth.

        Fixes issue where doubleclicking a word could select following adjacent newlines.
        https://bugs.webkit.org/show_bug.cgi?id=30234

        * platform/mac/editing/selection/doubleclick-should-not-expand-across-lines-expected.txt: Added.
        * platform/mac/editing/selection/doubleclick-should-not-expand-across-lines.html: Added.
        * platform/mac/editing/selection/script-tests/TEMPLATE.html: Added.
        * platform/mac/editing/selection/script-tests/doubleclick-should-not-expand-across-lines.js: Added.
        (description.doubleClickOnElement):
        (doubleClickAt):
        (selectionShouldBe):
        * platform/win/editing/selection/doubleclick-should-not-expand-across-lines-expected.txt: Added.
        * platform/win/editing/selection/doubleclick-should-not-expand-across-lines.html: Added.
        * platform/win/editing/selection/script-tests/TEMPLATE.html: Added.
        * platform/win/editing/selection/script-tests/doubleclick-should-not-expand-across-lines.js: Added.
        (description.doubleClickOnElement):
        (doubleClickAt):
        (selectionShouldBe):
2009-10-18  Erik Arvidsson  <arv@chromium.org>

        Reviewed by Adam Barth.

        Fixes issue where doubleclicking a word could select following adjacent newlines.
        https://bugs.webkit.org/show_bug.cgi?id=30234

        Tests: platform/mac/editing/selection/doubleclick-should-not-expand-across-lines.html
               platform/mac/editing/selection/script-tests/TEMPLATE.html
               platform/win/editing/selection/doubleclick-should-not-expand-across-lines.html
               platform/win/editing/selection/script-tests/TEMPLATE.html

        * editing/VisibleSelection.cpp:
        (WebCore::VisibleSelection::appendTrailingWhitespace):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@49767 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-10-18 Kenneth Rohde Christiansen <kenneth@webkit.org>
eric@webkit.org [Mon, 19 Oct 2009 06:51:24 +0000 (06:51 +0000)]
2009-10-18  Kenneth Rohde Christiansen  <kenneth@webkit.org>

        Reviewed by Simon Hausmann.

        Rename fixedContentsSize property to preferredContentsSize as
        agreed upon with Siman Hausmann and Matthias Ettrich.

        * Api/qwebpage.cpp:
        (QWebPage::preferredContentsSize):
        (QWebPage::setPreferredContentsSize):
        * Api/qwebpage.h:
        * WebCoreSupport/FrameLoaderClientQt.cpp:
        (WebCore::FrameLoaderClientQt::transitionToCommittedForNewPage):
2009-10-18  Kenneth Rohde Christiansen  <kenneth@webkit.org>

        Reviewed by Simon Hausmann.

        Use the setPreferredContentsSize method instead
        of setFixedContentsSize, as the method has been renamed.

        * DumpRenderTree/qt/DumpRenderTree.cpp:
        (WebCore::DumpRenderTree::open):
        * DumpRenderTree/qt/LayoutTestControllerQt.cpp:
        (LayoutTestController::setFixedContentsSize):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@49766 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-10-18 Kwang Yul Seo <skyul@company100.net>
eric@webkit.org [Mon, 19 Oct 2009 06:43:44 +0000 (06:43 +0000)]
2009-10-18  Kwang Yul Seo  <skyul@company100.net>

        Reviewed by Darin Adler.

        Use _stricmp and _strnicmp instead of deprecated stricmp and strnicmp.
        https://bugs.webkit.org/show_bug.cgi?id=30474

        stricmp and strnicmp are deprecated beginning in Visual
        C++ 2005. Use _stricmp and _strnicmp instead in StringExtras.h.

        * wtf/StringExtras.h:
        (strncasecmp):
        (strcasecmp):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@49765 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoCURL build fix, use proper header name.
kevino@webkit.org [Mon, 19 Oct 2009 03:53:02 +0000 (03:53 +0000)]
CURL build fix, use proper header name.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@49764 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago<rdar://problem/7309062> Crashes in WebCore::DocumentLoader::isStopping below WebKit...
mrowe@apple.com [Mon, 19 Oct 2009 02:38:59 +0000 (02:38 +0000)]
<rdar://problem/7309062> Crashes in WebCore::DocumentLoader::isStopping below WebKit::NetscapePluginInstanceProxy::loadRequest.

Reviewed by Dan Bernstein.

* Plugins/Hosted/NetscapePluginInstanceProxy.mm:
(WebKit::NetscapePluginInstanceProxy::loadRequest): Null-check the DocumentLoader before deferencing it.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@49763 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-10-18 Pavel Feldman <pfeldman@chromium.org>
pfeldman@chromium.org [Mon, 19 Oct 2009 02:09:14 +0000 (02:09 +0000)]
2009-10-18  Pavel Feldman  <pfeldman@chromium.org>

        Reviewed by Timothy Hatcher.

        Web Inspector: Properties on Arrays and NodeLists are not logged correctly.

        https://bugs.webkit.org/show_bug.cgi?id=30485

WebCore:
        * inspector/front-end/ConsoleView.js:
        (WebInspector.ConsoleView.prototype._formatarray):
        (WebInspector.ConsoleView.prototype._printArray):

LayoutTests:
        * inspector/console-format-expected.txt:
        * inspector/console-format.html:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@49762 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-10-18 Pavel Feldman <pfeldman@chromium.org>
pfeldman@chromium.org [Mon, 19 Oct 2009 01:20:14 +0000 (01:20 +0000)]
2009-10-18  Pavel Feldman  <pfeldman@chromium.org>

        Not reviewed: touching WebCore so that it
        initiated inspector frontend deployment on Windows.

        Web Inspector: frontend files are not deployed in Windows
        incremental build.

        * inspector/InspectorDOMAgent.cpp:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@49761 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-10-18 Pavel Feldman <pfeldman@chromium.org>
pfeldman@chromium.org [Mon, 19 Oct 2009 00:20:43 +0000 (00:20 +0000)]
2009-10-18  Pavel Feldman  <pfeldman@chromium.org>

        Reviewed by Timothy Hatcher.

        Web Inspector: make console tests more granular.

        https://bugs.webkit.org/show_bug.cgi?id=30496

        * inspector/console-dir-expected.txt: Added.
        * inspector/console-dir.html: Added.
        * inspector/console-dirxml-expected.txt: Added.
        * inspector/console-dirxml.html: Added.
        * inspector/console-format-expected.txt: Added.
        * inspector/console-format.html: Added.
        * inspector/console-tests-expected.txt:
        * inspector/console-tests.html:
        * platform/gtk/Skipped:
        * platform/qt/Skipped:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@49760 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-10-18 Pavel Feldman <pfeldman@chromium.org>
pfeldman@chromium.org [Sun, 18 Oct 2009 23:17:37 +0000 (23:17 +0000)]
2009-10-18  Pavel Feldman  <pfeldman@chromium.org>

        Reviewed by Timothy Hatcher.

        Web Inspector: Title of dir(["test", "test2"]) should be Array

        https://bugs.webkit.org/show_bug.cgi?id=30486

        Test: inspector/console-tests.html

        * inspector/front-end/InjectedScript.js:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@49758 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-10-18 Dirk Schulze <krit@webkit.org>
krit@webkit.org [Sun, 18 Oct 2009 21:15:59 +0000 (21:15 +0000)]
2009-10-18  Dirk Schulze  <krit@webkit.org>

        Reviewed by Nikolas Zimmermann.

        REGRESSION: SVG Mask doesn't work for maskContentUnits="objectBoundingBox"
        [https://bugs.webkit.org/show_bug.cgi?id=30480]

        Patch for SVGMaskElement landed in r49598 should correct the location of
        the mask image graphics context. It only corrects the location for userSpaceOnUse
        mode in maskContentUnits but breaks objectBoundingBoxMode.
        The maskDestRect shouldn't be moved. It is not responsible for the correct context
        postioin but for the correct position of the mask image.
        This patch calculates the context postion independently from the maskDestRect.
        It also uses lineareRGB color space for masking on CG now. This is the default
        color space for masking operations.
        We already have tests for both maskContentUnits modes.

            *svg/W3C-SVG-1.1/masking-intro-01-f.svg
            *svg/custom/mask-with-default-value.svg

        Some tests needed new pixel tests because of the new color space.

        * svg/SVGMaskElement.cpp:
        (WebCore::SVGMaskElement::drawMaskerContent):

        The color space of masking operations in SVG changed to linearSVG.
        This needs an update of some masking pixel tests.

        * platform/mac/svg/W3C-SVG-1.1/masking-intro-01-f-expected.checksum:
        * platform/mac/svg/W3C-SVG-1.1/masking-intro-01-f-expected.png:
        * platform/mac/svg/batik/masking/maskRegions-expected.checksum:
        * platform/mac/svg/batik/masking/maskRegions-expected.png:
        * platform/mac/svg/custom/grayscale-gradient-mask-expected.checksum:
        * platform/mac/svg/custom/grayscale-gradient-mask-expected.png:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@49757 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-10-18 Anton Muhin <antonm@chromium.org>
eric@webkit.org [Sun, 18 Oct 2009 19:55:49 +0000 (19:55 +0000)]
2009-10-18  Anton Muhin  <antonm@chromium.org>

        Reviewed by Adam Barth.

        Use v8::Integer::NewFromUnsigned when converting unsigneds into V8's numbers.
        That is notably faster for small numbers (most common case).
        https://bugs.webkit.org/show_bug.cgi?id=30493

        * bindings/scripts/CodeGeneratorV8.pm:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@49756 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-10-18 Jan Michael Alonzo <jmalonzo@webkit.org>
jmalonzo@webkit.org [Sun, 18 Oct 2009 19:29:49 +0000 (19:29 +0000)]
2009-10-18  Jan Michael Alonzo  <jmalonzo@webkit.org>

        Reviewed by Holger Freyther.

        [GTK] Add MathML to the build system
        https://bugs.webkit.org/show_bug.cgi?id=30487

        Add --enable-mathml to configure.

        * configure.ac:

        Add MathML sources if MathML support is enabled.

        * GNUmakefile.am:

        Include layout test baseline for Gtk.

        * platform/gtk/mathml/presentation/sub-expected.txt: Added.
        * platform/gtk/mathml/presentation/sup-expected.txt: Added.
        * platform/gtk/mathml/presentation/tables-expected.txt: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@49755 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoNon-PCH build fix. Added missing header.
kevino@webkit.org [Sun, 18 Oct 2009 18:21:08 +0000 (18:21 +0000)]
Non-PCH build fix. Added missing header.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@49754 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-10-18 Simon Fraser <simon.fraser@apple.com>
simon.fraser@apple.com [Sun, 18 Oct 2009 17:17:43 +0000 (17:17 +0000)]
2009-10-18  Simon Fraser  <simon.fraser@apple.com>

        Reviewed by Dan Bernstein.

        Compositing layers not correctly updated after partial layout
        https://bugs.webkit.org/show_bug.cgi?id=30425

        When a partial layout happened, for example as the result of a postion-change-only
        layout, then some compositing layer positions were not correctly updated. To fix
        this, updateLayerPositions() now carries along a flag that is set at the rootmost
        layer being updated, and used to determine when we hit the first compositing layer
        in this update. RenderLayerBacking::updateAfterLayout() makes use of this information
        to do a full geometry update on that layer, which is thus the rootmost compositing
        layer that is being updated.

        Test: compositing/geometry/partial-layout-update.html

        * page/FrameView.cpp:
        (WebCore::FrameView::layout):
        * rendering/RenderLayer.cpp:
        (WebCore::RenderLayer::updateLayerPositions):
        (WebCore::RenderLayer::scrollToOffset):
        * rendering/RenderLayer.h:
        (WebCore::RenderLayer::):
        * rendering/RenderLayerBacking.cpp:
        (WebCore::RenderLayerBacking::updateAfterLayout):
        * rendering/RenderLayerBacking.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@49753 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoReviewed by Kevin Ollivier.
kevino@webkit.org [Sun, 18 Oct 2009 17:02:58 +0000 (17:02 +0000)]
Reviewed by Kevin Ollivier.

Add the ability to specify a proxy for wxWebKit.

https://bugs.webkit.org/show_bug.cgi?id=30446

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@49752 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-10-18 Joanmarie Diggs <joanmarie.diggs@gmail.com>
eric@webkit.org [Sun, 18 Oct 2009 16:23:15 +0000 (16:23 +0000)]
2009-10-18  Joanmarie Diggs  <joanmarie.diggs@gmail.com>

        Reviewed by Xan Lopez

        https://bugs.webkit.org/show_bug.cgi?id=25901
        Use ATK_ROLE_SECTION for divTag and ATK_ROLE_LABEL for labelTag

        Replaces the use of ATK_ROLE_PANEL with the expected accessible roles.

        * accessibility/gtk/AccessibilityObjectWrapperAtk.cpp:
        (webkit_accessible_get_role):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@49751 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoReviewed by Kevin Ollivier.
kevino@webkit.org [Sun, 18 Oct 2009 16:08:41 +0000 (16:08 +0000)]
Reviewed by Kevin Ollivier.

Add basic database support to wx API

https://bugs.webkit.org/show_bug.cgi?id=30445

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@49750 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-10-18 Eric Carlson <eric.carlson@apple.com>
eric.carlson@apple.com [Sun, 18 Oct 2009 16:00:40 +0000 (16:00 +0000)]
2009-10-18  Eric Carlson  <eric.carlson@apple.com>

        Skip media/video-empty-source.html until <rdar://problem/6563117> has been fixed.

        * platform/mac/Skipped:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@49749 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-10-17 Yuta Kitamura <yutak@chromium.org>
eric@webkit.org [Sun, 18 Oct 2009 06:27:17 +0000 (06:27 +0000)]
2009-10-17  Yuta Kitamura  <yutak@chromium.org>

        Reviewed by Darin Adler.

        Fix inefficient string concatenation.

        LayoutTests/fast/css/large-list-of-rules-crash.html contains code that
        concatenate strings in an inefficient way. Concatenation of strings should be
        done with Array.join method. This patch fixes this issue, and also fixes typos
        in the test.

        fast/css/large-list-of-rules-crash.html concatenates strings in an inefficient way
        https://bugs.webkit.org/show_bug.cgi?id=30436

        * fast/css/large-list-of-rules-crash.html:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@49748 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoVersioning.
mrowe@apple.com [Sat, 17 Oct 2009 23:07:42 +0000 (23:07 +0000)]
Versioning.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@49747 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-10-17 Nikolas Zimmermann <nzimmermann@rim.com>
zimmermann@webkit.org [Sat, 17 Oct 2009 22:52:54 +0000 (22:52 +0000)]
2009-10-17  Nikolas Zimmermann  <nzimmermann@rim.com>

        Reviewed by George Staikos.

        Cleanup SVGElement code, preparing for animVal support
        https://bugs.webkit.org/show_bug.cgi?id=30466

        Add new SynchronizablePropertyController and move the code for SVG property <-> XML attribute synchronization
        from SVGElement. This is a further preparation for animVal support, where SynchronizablePropertyController will
        be used to control the start/end state of an animated property.

        We're currently tracking animated properties that need synchronization, generalize this concept and use
        it to track their creation (bound to certain SVG*Element classes) as well as a flag determing the need
        of synchronization (no SVG DOM object wrappers, no synchronization needed).

        No change in functionality, thus no tests.

        * GNUmakefile.am: Add 'SynchronizablePropertyController.cpp/h' to build
        * WebCore.gypi: Ditto.
        * WebCore.pro: Ditto.
        * WebCore.vcproj/WebCore.vcproj: Ditto.
        * WebCore.xcodeproj/project.pbxproj: Ditto.
        * svg/SVGAllInOne.cpp: Ditto.
        * svg/SVGAnimatedProperty.h: Add registerProperty() function
        (WebCore::::SVGAnimatedProperty): Call registerProperty() function
        (WebCore::::registerProperty): Add new registerProperty() function, announcing new properties to SynchronizablePropertyController.
        * svg/SVGAnimatedTemplate.h: Use propertyController() method, to access the new functions moved from SVGElement.
        (WebCore::lookupOrCreateWrapper):
        * svg/SVGElement.cpp: Adapt to code moving to SynchronizablePropertyController.
        (WebCore::SVGElement::updateAnimatedSVGAttribute):
        * svg/SVGElement.h: Move handling SVG property synchronization to SynchronizablePropertyController.
        (WebCore::SVGElement::propertyController): Expose reference to SynchronizablePropertyController object.
        * svg/SVGViewSpec.cpp: Reorder initialization order of contextElement, leading to possible crashes.
        (WebCore::SVGViewSpec::SVGViewSpec):
        (WebCore::SVGViewSpec::viewTarget):
        * svg/SVGViewSpec.h: Ditto.
        (WebCore::SVGViewSpec::contextElement):
        * svg/SynchronizablePropertyController.cpp: Added.
        (WebCore::SynchronizablePropertyController::SynchronizablePropertyController):
        (WebCore::SynchronizablePropertyController::registerProperty):
        (WebCore::SynchronizablePropertyController::setPropertyNeedsSynchronization):
        (WebCore::SynchronizablePropertyController::synchronizeProperty):
        (WebCore::SynchronizablePropertyController::synchronizeAllProperties):
        (WebCore::SynchronizablePropertyController::startAnimation):
        (WebCore::SynchronizablePropertyController::stopAnimation):
        * svg/SynchronizablePropertyController.h: Added.
        (WebCore::SynchronizableProperty::SynchronizableProperty):
        (WebCore::SynchronizableProperty::isHashTableDeletedValue):
        (WebCore::SynchronizableProperty::operator==):
        (WebCore::SynchronizablePropertyHash::hash):
        (WebCore::SynchronizablePropertyHash::equal):
        (WebCore::SynchronizablePropertyHashTraits::constructDeletedValue):
        (WebCore::SynchronizablePropertyHashTraits::isDeletedValue):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@49745 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago Reviewed by Eric Carlson.
ap@apple.com [Sat, 17 Oct 2009 22:38:24 +0000 (22:38 +0000)]
    Reviewed by Eric Carlson.

        https://bugs.webkit.org/show_bug.cgi?id=30470
        Make marking a bug as duplicate easier

        * js/field.js: (setResolutionToDuplicate): Focus the edit field, too.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@49744 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-10-17 Joanmarie Diggs <joanmarie.diggs@gmail.com>
eric@webkit.org [Sat, 17 Oct 2009 20:00:33 +0000 (20:00 +0000)]
2009-10-17  Joanmarie Diggs  <joanmarie.diggs@gmail.com>

        Reviewed by Xan Lopez.

        https://bugs.webkit.org/show_bug.cgi?id=25413
        [GTK] Please expose the level of headings

        Exposes the heading level as an attribute of the AtkObject.

        * accessibility/gtk/AccessibilityObjectWrapperAtk.cpp:
        (addAttributeToSet):
        (webkit_accessible_get_attributes):
        (webkit_accessible_class_init):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@49743 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-10-17 Kwang Yul Seo <skyul@company100.net>
eric@webkit.org [Sat, 17 Oct 2009 15:54:18 +0000 (15:54 +0000)]
2009-10-17  Kwang Yul Seo  <skyul@company100.net>

        Reviewed by Eric Seidel.

        EventHandler::eventLoopHandleMouseDragged needs ENABLE(DRAG_SUPPORT) guards
        https://bugs.webkit.org/show_bug.cgi?id=30472

        Put ENABLE(DRAG_SUPPORT) guards around EventHandler::eventLoopHandleMouseDragged
        in EventHandler.cpp. MSVC fails to compile when DRAG_SUPPORT not enabled.

        * page/EventHandler.cpp:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@49742 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoReviewed by Kevin Ollivier.
kevino@webkit.org [Sat, 17 Oct 2009 15:49:02 +0000 (15:49 +0000)]
Reviewed by Kevin Ollivier.

Optionally allow the user to zoom text using the mouse wheel.

https://bugs.webkit.org/show_bug.cgi?id=30444

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@49741 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-10-17 Alpha Lam <hclam@chromium.org>
eric@webkit.org [Sat, 17 Oct 2009 15:45:32 +0000 (15:45 +0000)]
2009-10-17  Alpha Lam  <hclam@chromium.org>

        Reviewed by Eric Seidel.

        [chromium] Video controls not zoomed / transformed correctly
        https://bugs.webkit.org/show_bug.cgi?id=30461

        Draw the images in the controls scaled.

        No new tests. With this change Chromium will pass the following tests:
        LayoutTests/media/video-controls-zoomed.html
        LayoutTests/media/video-controls-transformed.html

        * rendering/RenderMediaControlsChromium.cpp:
        (WebCore::paintMediaButton):
        Pass along the target rect directly.
        (WebCore::RenderMediaControlsChromium::adjustMediaSliderThumbSize):
        Adjust the thumb slide applied with the scale factor.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@49740 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-10-17 Laszlo Gombos <laszlo.1.gombos@nokia.com>
laszlo.1.gombos@nokia.com [Sat, 17 Oct 2009 07:53:44 +0000 (07:53 +0000)]
2009-10-17  Laszlo Gombos  <laszlo.1.gombos@nokia.com>

        Unreviewed.

        [Gtk] Build fix for WebKitGtk after r49723.

        Add bindings/js/JSExceptionBase.* to the build.

        * GNUmakefile.am:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@49739 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-10-17 Laszlo Gombos <laszlo.1.gombos@nokia.com>
laszlo.1.gombos@nokia.com [Sat, 17 Oct 2009 07:46:12 +0000 (07:46 +0000)]
2009-10-17  Laszlo Gombos  <laszlo.1.gombos@nokia.com>

        Unreviewed.

        [Qt] Build fix for QtWebKit after r49723.

        Add bindings/js/JSExceptionBase.cpp to the build.

        * WebCore.pro:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@49738 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-10-17 Pavel Feldman <pfeldman@chromium.org>
pfeldman@chromium.org [Sat, 17 Oct 2009 07:32:57 +0000 (07:32 +0000)]
2009-10-17  Pavel Feldman  <pfeldman@chromium.org>

        Not reviewed: reduce inspector console test size - it times out.

        * inspector/console-tests-expected.txt:
        * inspector/console-tests.html:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@49737 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago Reviewed by Mark Rowe.
ap@apple.com [Sat, 17 Oct 2009 06:59:38 +0000 (06:59 +0000)]
    Reviewed by Mark Rowe.

        https://bugs.webkit.org/show_bug.cgi?id=30470
        Make marking a bug as duplicate easier

        * js/field.js: (setResolutionToDuplicate): Show edit field right away, don't make the user
        click again.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@49736 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agohttps://bugs.webkit.org/show_bug.cgi?id=30456
sfalken@apple.com [Sat, 17 Oct 2009 06:29:49 +0000 (06:29 +0000)]
https://bugs.webkit.org/show_bug.cgi?id=30456
Fixes for new Debug_All Windows build configuration.

Reviewed by Dan Bernstein.

WebCore:

* platform/network/cf/ResourceRequestCFNet.cpp:
(WebCore::findCFNetworkModule): Ask for the correct library instead of guessing.
* platform/win/ScrollbarThemeSafari.cpp: Use new DEBUG_ALL preprocessor define for library naming.
* rendering/RenderMediaControls.cpp: Use new DEBUG_ALL preprocessor define for library naming.
* rendering/RenderThemeSafari.cpp: Use new DEBUG_ALL preprocessor define for library naming.

WebKit/win:

* WebKitClassFactory.cpp: Use new DEBUG_ALL preprocessor define for library naming.

WebKitLibraries:

* win/tools/vsprops/debug_all.vsprops:
Define DEBUG_ALL in Debug_All configuration.
Continue to define USE_DEBUG_SAFARI_THEME for open source SafariTheme header usage.

WebKitTools:

* DumpRenderTree/win/DumpRenderTree.cpp:
(sharedCFURLCache): Use new DEBUG_ALL preprocessor define for library naming.
* DumpRenderTree/win/DumpRenderTree.vcproj: Remove unused DEBUG_WEBKIT_HAS_SUFFIX.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@49735 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoRolled back in r49717 with the build maybe working now?
ggaren@apple.com [Sat, 17 Oct 2009 05:52:20 +0000 (05:52 +0000)]
Rolled back in r49717 with the build maybe working now?

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@49734 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago Reviewed by Jon Honeycutt.
kdecker@apple.com [Sat, 17 Oct 2009 05:15:57 +0000 (05:15 +0000)]
    Reviewed by Jon Honeycutt.

        <rdar://problem/7304575>

        * Plugins/WebBaseNetscapePluginView.h: Added new boolean ivar and -hasBeenHalted instance method.
        * Plugins/WebBaseNetscapePluginView.mm:
        (-[WebBaseNetscapePluginView halt]):Update _hasBeenHalted.
        (-[WebBaseNetscapePluginView hasBeenHalted]): Added.
        * WebView/WebView.mm:
        (+[WebView _hasPluginForNodeBeenHalted:]): Checks to see if a plug-in node's been halted.
        * WebView/WebViewPrivate.h: Added -_hasPluginForNodeBeenHalted method.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@49733 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoAdd SPI to determine whether a plug-in has ever been halted.
jhoneycutt@apple.com [Sat, 17 Oct 2009 04:38:38 +0000 (04:38 +0000)]
Add SPI to determine whether a plug-in has ever been halted.

Part of <rdar://problem/7312158>.

Reviewed by Dan Bernstein.

WebCore:

* plugins/PluginView.cpp:
(WebCore::PluginView::PluginView):
Initialize m_hasBeenHalted.

* plugins/PluginView.h:
(WebCore::PluginView::hasBeenHalted):
Return m_hasBeenHalted.

* plugins/win/PluginViewWin.cpp:
(WebCore::PluginView::halt):
Set m_hasBeenHalted.

WebKit/win:

* Interfaces/IWebViewPrivate.idl:
Add hasPluginForNodeBeenHalted().

* WebView.cpp:
(pluginViewForNode):
Code factored out of isNodeHaltedPlugin().
(WebView::isNodeHaltedPlugin):
Remove STDMETHODCALLTYPE, as it has no use at the definition. Use
pluginViewForNode().
(WebView::restartHaltedPluginForNode):
Ditto.
(WebView::hasPluginForNodeBeenHalted):
Get the PluginView for the given node. Return the result of
PluginView::hasBeenHalted().

* WebView.h:
Declare hasPluginForNodeBeenHalted().

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@49732 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-10-16 Simon Fraser <simon.fraser@apple.com>
simon.fraser@apple.com [Sat, 17 Oct 2009 04:23:57 +0000 (04:23 +0000)]
2009-10-16  Simon Fraser  <simon.fraser@apple.com>

        Skip this test:
            compositing/animation/animated-composited-inside-hidden.html
        because the fix for https://bugs.webkit.org/show_bug.cgi?id=26770 was backed out.

        * platform/mac/Skipped:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@49731 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago[Qt] QGLauncher leaks WebPage object
tonikitoo@webkit.org [Sat, 17 Oct 2009 03:54:56 +0000 (03:54 +0000)]
[Qt] QGLauncher leaks WebPage object
https://bugs.webkit.org/show_bug.cgi?id=30465

Patch by Antonio Gomes <tonikitoo@webkit.org> on 2009-10-16
Reviewed by Simon Hausmann.

Make 'SharedScene' to own 'WebPage' reference and delete it at its destructor.

* Api/qwebpage.cpp:
(QWebPage::view):
* Api/qwebpage_p.h:
* QGVLauncher/main.cpp:
(SharedScene::SharedScene):
(SharedScene::~SharedScene):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@49730 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago[Qt] "dangling" pointer to qwebpage's view object can leads QGLauncher to crash
tonikitoo@webkit.org [Sat, 17 Oct 2009 03:51:08 +0000 (03:51 +0000)]
[Qt] "dangling" pointer to qwebpage's view object can leads QGLauncher to crash
https://bugs.webkit.org/show_bug.cgi?id=30459

Patch by Antonio Gomes <tonikitoo@webkit.org> on 2009-10-16
Reviewed by Simon Hausmann.

Remove all setView(ev->widget()) calls in QWebPage and QGWV event handling methods,
since QWebPageClient would do the trick.

* Api/qgraphicswebview.cpp:
(QGraphicsWebView::hoverMoveEvent):
* Api/qwebpage.cpp:
(QWebPagePrivate::mouseMoveEvent):
(QWebPagePrivate::mousePressEvent):
(QWebPagePrivate::mouseDoubleClickEvent):
(QWebPagePrivate::mouseReleaseEvent):
(QWebPagePrivate::wheelEvent):
(QWebPagePrivate::dragEnterEvent):
(QWebPagePrivate::dragLeaveEvent):
(QWebPagePrivate::dragMoveEvent):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@49729 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoUpdated expected results of tests that give CONSOLE messages for
bweinstein@apple.com [Sat, 17 Oct 2009 03:35:44 +0000 (03:35 +0000)]
Updated expected results of tests that give CONSOLE messages for
new verbose output.

* fast/dom/NamedNodeMap-setNamedItem-crash-expected.txt:
* fast/dom/timer-clear-interval-in-handler-and-generate-error-expected.txt:
* platform/mac/fast/dynamic/015-expected.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@49728 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-10-16 Adam Barth <abarth@webkit.org>
abarth@webkit.org [Sat, 17 Oct 2009 03:10:59 +0000 (03:10 +0000)]
2009-10-16  Adam Barth  <abarth@webkit.org>

        Reviewed by Alexey Proskuryakov.

        @charset rule after the first byte causes the rest of css to be ignored
        https://bugs.webkit.org/show_bug.cgi?id=18265

        Apparently we're supposed to tolerate whitespace before the charset
        rule so folks can write code like this:

        <style>
        @charset "utf-8"
        ...
        </style>

        I'm told this is one of the top compatability problems in China.
        Tests: fast/css/comment-before-charset-external.html
               fast/css/comment-before-charset.html
               fast/css/many-spaces-before-charset.html
               fast/css/space-before-charset-external.html
               fast/css/space-before-charset.html

        * css/CSSGrammar.y:
2009-10-16  Adam Barth  <abarth@webkit.org>

        Reviewed by Alexey Proskuryakov.

        @charset rule after the first byte causes the rest of css to be ignored
        https://bugs.webkit.org/show_bug.cgi?id=18265

        Test that we allow whitespace and comments before the charset rule.
        These test results all match Firefox.

        * fast/css/comment-before-charset-expected.txt: Added.
        * fast/css/comment-before-charset-external-expected.txt: Added.
        * fast/css/comment-before-charset-external.html: Added.
        * fast/css/comment-before-charset.html: Added.
        * fast/css/many-spaces-before-charset-expected.txt: Added.
        * fast/css/many-spaces-before-charset.html: Added.
        * fast/css/resources/comment-before-charset.css: Added.
        * fast/css/resources/space-before-charset.css: Added.
        * fast/css/space-before-charset-expected.txt: Added.
        * fast/css/space-before-charset-external-expected.txt: Added.
        * fast/css/space-before-charset-external.html: Added.
        * fast/css/space-before-charset.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@49727 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoRoll out r49717 as it broke the build.
mrowe@apple.com [Sat, 17 Oct 2009 02:31:42 +0000 (02:31 +0000)]
Roll out r49717 as it broke the build.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@49726 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoRubber-stamped by Adam Roben.
bweinstein@apple.com [Sat, 17 Oct 2009 01:23:43 +0000 (01:23 +0000)]
Rubber-stamped by Adam Roben.

Remove inspector tests from Windows Skipped list, because they pass locally
and the front-end changes are picked up by the bots.

* platform/win/Skipped:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@49725 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoBuild fix: apparently we shouldn't export those symbols?
ggaren@apple.com [Sat, 17 Oct 2009 01:20:12 +0000 (01:20 +0000)]
Build fix: apparently we shouldn't export those symbols?

Patch by Geoffrey Garen <ggaren@apple.com> on 2009-10-16
* JavaScriptCore.exp:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@49724 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoWebCore: Fixes part of <http://webkit.org/b/30412>.
bweinstein@apple.com [Sat, 17 Oct 2009 01:18:24 +0000 (01:18 +0000)]
WebCore: Fixes part of <webkit.org/b/30412>.
Web Inspector should get human readable DOM Exceptions.

Reviewed by Darin Adler.

Add a description field to ExceptionBase, and call it through
reportException which allows the user/developer to get a more detailed
and coherent error explanation through the Web Inspector.

This only applies to exceptions that come from a script on the page when
it is run. DOM Exceptions that come from the code run in the console, or DOM
exceptions that are caught and logged will come in a future patch.

Tests: inspector/uncaught-dom1-exception.html
       inspector/uncaught-dom3-exception.html
       inspector/uncaught-dom8-exception.html

* WebCore.gypi:
* WebCore.vcproj/WebCore.vcproj:
* WebCore.xcodeproj/project.pbxproj:
* bindings/js/JSDOMBinding.cpp:
(WebCore::reportException):
* bindings/js/JSExceptionBase.cpp: Added.
(WebCore::toExceptionBase):
* bindings/js/JSExceptionBase.h: Added.
* dom/ExceptionBase.cpp:
(WebCore::ExceptionBase::ExceptionBase):
* dom/ExceptionBase.h:
(WebCore::ExceptionBase::description):
* dom/ExceptionCode.cpp:
(WebCore::):
(WebCore::getExceptionCodeDescription):
* dom/ExceptionCode.h:

LayoutTests: Fixes part of <http://webkit.org/b/30412>.
Web Inspector should get human readable DOM Exceptions.

Reviewed by Darin Adler.

Adds tests for output of uncaught exceptions to the Web Inspector
Console. Tests DOM exceptions 1, 3, and 8 (INDEX_SIZE_ERR, NOT_FOUND_ERR, and
HIERARCHY_REQUEST_ERR).

* inspector/uncaught-dom1-exception-expected.txt: Added.
* inspector/uncaught-dom1-exception.html: Added.
* inspector/uncaught-dom3-exception-expected.txt: Added.
* inspector/uncaught-dom3-exception.html: Added.
* inspector/uncaught-dom8-exception-expected.txt: Added.
* inspector/uncaught-dom8-exception.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@49723 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoBuild fix: export some symbols.
ggaren@apple.com [Sat, 17 Oct 2009 01:09:35 +0000 (01:09 +0000)]
Build fix: export some symbols.

Patch by Geoffrey Garen <ggaren@apple.com> on 2009-10-16
* JavaScriptCore.exp:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@49722 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agostructure typeinfo flags should be inherited.
oliver@apple.com [Sat, 17 Oct 2009 01:06:40 +0000 (01:06 +0000)]
structure typeinfo flags should be inherited.
https://bugs.webkit.org/show_bug.cgi?id=30468

Reviewed by Gavin Barraclough.

Add StructureFlag constant to the various JSC classes and use
it for the TypeInfo construction.  This allows us to simply
accumulate flags by basing each classes StructureInfo on its parents.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@49721 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoBuild fix: forgot to check in this #include.
ggaren@apple.com [Sat, 17 Oct 2009 00:45:32 +0000 (00:45 +0000)]
Build fix: forgot to check in this #include.

Patch by Geoffrey Garen <ggaren@apple.com> on 2009-10-16
* bridge/runtime_root.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@49720 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago* coding/RefPtr.html: Minor revision, updating a link and tweaking
darin@apple.com [Sat, 17 Oct 2009 00:40:16 +0000 (00:40 +0000)]
* coding/RefPtr.html: Minor revision, updating a link and tweaking
wording a bit in few places.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@49719 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoFast for-in enumeration: Cache JSPropertyNameIterator; cache JSStrings
ggaren@apple.com [Sat, 17 Oct 2009 00:28:19 +0000 (00:28 +0000)]
Fast for-in enumeration: Cache JSPropertyNameIterator; cache JSStrings
in JSPropertyNameIterator; inline more code.

Patch by Geoffrey Garen <ggaren@apple.com> on 2009-10-16
Reviewed by Oliver Hunt.

1.024x as fast on SunSpider (fasta: 1.43x as fast).

* bytecode/CodeBlock.cpp:
(JSC::CodeBlock::dump):
* bytecode/Opcode.h:
* bytecompiler/BytecodeGenerator.cpp:
(JSC::BytecodeGenerator::emitGetPropertyNames):
(JSC::BytecodeGenerator::emitNextPropertyName):
* bytecompiler/BytecodeGenerator.h: Added a few extra operands to
op_get_pnames and op_next_pname so that we can track iteration state
in the register file instead of in the JSPropertyNameIterator. (To be
cacheable, the JSPropertyNameIterator must be stateless.)

* interpreter/Interpreter.cpp:
(JSC::Interpreter::tryCachePutByID):
(JSC::Interpreter::tryCacheGetByID): Updated for rename to
"normalizePrototypeChain" and removal of "isCacheable".

(JSC::Interpreter::privateExecute): Updated for in-RegisterFile
iteration state tracking.

* jit/JIT.cpp:
(JSC::JIT::privateCompileMainPass):
* jit/JIT.h:
* jit/JITOpcodes.cpp:
(JSC::JIT::emit_op_get_pnames): Updated for in-RegisterFile
iteration state tracking.

(JSC::JIT::emit_op_next_pname): Inlined code generation for op_next_pname.

* jit/JITStubs.cpp:
(JSC::JITThunks::tryCachePutByID):
(JSC::JITThunks::tryCacheGetByID): Updated for rename to
"normalizePrototypeChain" and removal of "isCacheable".

(JSC::DEFINE_STUB_FUNCTION):
* jit/JITStubs.h:
(JSC::): Added has_property and to_object stubs. Removed op_next_pname
stub, since has_property is all we need anymore.

* parser/Nodes.cpp:
(JSC::ForInNode::emitBytecode): Updated for in-RegisterFile
iteration state tracking.

* runtime/JSCell.h:
* runtime/JSObject.cpp:
(JSC::JSObject::getPropertyNames): Don't do caching at this layer
anymore, since we don't create a JSPropertyNameIterator at this layer.

* runtime/JSPropertyNameIterator.cpp:
(JSC::JSPropertyNameIterator::create): Do do caching at this layer.
(JSC::JSPropertyNameIterator::get):  Updated for in-RegisterFile
iteration state tracking.
(JSC::JSPropertyNameIterator::markChildren): Mark our JSStrings.

* runtime/JSPropertyNameIterator.h:
(JSC::JSPropertyNameIterator::size):
(JSC::JSPropertyNameIterator::setCachedStructure):
(JSC::JSPropertyNameIterator::cachedStructure):
(JSC::JSPropertyNameIterator::setCachedPrototypeChain):
(JSC::JSPropertyNameIterator::cachedPrototypeChain):
(JSC::JSPropertyNameIterator::JSPropertyNameIterator):
(JSC::Structure::setEnumerationCache): Don't store iteration state in
a JSPropertyNameIterator. Do cache a JSPropertyNameIterator in a
Structure.

* runtime/JSValue.h:
(JSC::asCell):
* runtime/MarkStack.h: Make those mischievous #include gods happy.

* runtime/ObjectConstructor.cpp:

* runtime/Operations.h:
(JSC::normalizePrototypeChain): Renamed countPrototypeChainEntriesAndCheckForProxies
to normalizePrototypeChain, since it changes dictionary prototypes to
non-dictionary objects.

* runtime/PropertyNameArray.cpp:
(JSC::PropertyNameArray::add):
* runtime/PropertyNameArray.h:
(JSC::PropertyNameArrayData::PropertyNameArrayData):
(JSC::PropertyNameArray::data):
(JSC::PropertyNameArray::size):
(JSC::PropertyNameArray::begin):
(JSC::PropertyNameArray::end): Simplified some code here to help with
current and future refactoring.

* runtime/Protect.h:
* runtime/Structure.cpp:
(JSC::Structure::~Structure):
(JSC::Structure::addPropertyWithoutTransition):
(JSC::Structure::removePropertyWithoutTransition): No need to clear
the enumeration cache with adding / removing properties without
transition. It is an error to add / remove properties without transition
once an object has been observed, and we can ASSERT to catch that.

* runtime/Structure.h:
(JSC::Structure::enumerationCache): Changed the enumeration cache to
hold a JSPropertyNameIterator.

* runtime/StructureChain.cpp:
* runtime/StructureChain.h:
(JSC::StructureChain::head): Removed StructureChain::isCacheable because
it was wrong-headed in two ways: (1) It gave up when a prototype was a
dictionary, but instead we want un-dictionary heavily accessed
prototypes; (2) It folded a test for hasDefaultGetPropertyNames() into
a generic test for "cacheable-ness", but hasDefaultGetPropertyNames()
is only releavant to for-in caching.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@49717 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-10-16 Simon Fraser <simon.fraser@apple.com>
simon.fraser@apple.com [Sat, 17 Oct 2009 00:20:45 +0000 (00:20 +0000)]
2009-10-16  Simon Fraser  <simon.fraser@apple.com>

        Reviewed by Dan Bernstein.

        Transform layer gets stuck with the wrong transform after an animation ends
        <rdar://problem/7311662>

        Work around a CoreAnimation bug which causes an animated transform layer to
        end up with a stale transform.

        Test: compositing/animation/state-at-end-event-transform-layer.html

        * platform/graphics/mac/GraphicsLayerCA.mm:
        (WebCore::bug7311367Workaround):
        (WebCore::GraphicsLayerCA::removeAnimationFromLayer):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@49713 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agowxMSW non-precomp headers build fix.
kevino@webkit.org [Sat, 17 Oct 2009 00:20:16 +0000 (00:20 +0000)]
wxMSW non-precomp headers build fix.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@49712 268f45cc-cd09-0410-ab3c-d52691b4dbfc