WebKit.git
11 years ago2010-01-22 Maciej Stachowiak <mjs@apple.com>
mjs@apple.com [Sat, 23 Jan 2010 00:43:37 +0000 (00:43 +0000)]
2010-01-22  Maciej Stachowiak  <mjs@apple.com>

        Reviewed by Adam Roben.

        Use stale cache data when going back and forward but not using WebCore’s page cache
        https://bugs.webkit.org/show_bug.cgi?id=33993
        <rdar://problem/7383392>

        No tests since this is a performance change.

        * loader/CachePolicy.h:
        (WebCore::): Add CachePolicyAllowStale.
        * loader/DocLoader.cpp:
        (WebCore::DocLoader::checkForReload): Never reload in the case of
        CachePolicyAllowStale.
        * loader/FrameLoader.cpp:
        (WebCore::FrameLoader::subresourceCachePolicy): Return CachePolicyAllowStale
        if the main resource is being loaded to allow stale data.
        (WebCore::FrameLoader::addExtraFieldsToRequest): Use ReturnCacheDataElseLoad
        on back/forward loads - needed when going back of forward to a page with frames.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@53737 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-01-22 Kelly Norton <knorton@google.com>
knorton@google.com [Sat, 23 Jan 2010 00:31:07 +0000 (00:31 +0000)]
2010-01-22  Kelly Norton  <knorton@google.com>

        Reviewed by Pavel Feldman.

        Web Inspector: JavaScript Error in DOMAgent.js:375 (_attributesUpdated)
        https://bugs.webkit.org/show_bug.cgi?id=33915

        Errors were caused by the fact that Element::attributes calls Element::setAttribute to
        synchronized the styleAttr. The fix is to simply check the synchronizing style attribute
        flag.

        * dom/Element.cpp:
        (WebCore::Element::setAttribute): Checked for case where styleAttr is being synchronized.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@53736 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-01-22 Adam Barth <abarth@webkit.org>
abarth@webkit.org [Sat, 23 Jan 2010 00:27:11 +0000 (00:27 +0000)]
2010-01-22  Adam Barth  <abarth@webkit.org>

        Unreviewed.  Arg.  Someone renamed limit to output_limit on me.

        * Scripts/webkitpy/commands/queues.py:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@53735 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-01-22 Adam Barth <abarth@webkit.org>
abarth@webkit.org [Sat, 23 Jan 2010 00:21:53 +0000 (00:21 +0000)]
2010-01-22  Adam Barth  <abarth@webkit.org>

        Rubber stamped by Eric Seidel.

        Make __init__.py and buildbot.py pass pep8 style checker.

        * Scripts/webkitpy/__init__.py:
        * Scripts/webkitpy/buildbot.py:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@53734 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-01-21 Joe Mason <jmason@rim.com>
abarth@webkit.org [Sat, 23 Jan 2010 00:12:12 +0000 (00:12 +0000)]
2010-01-21  Joe Mason  <jmason@rim.com>

        Reviewed by Adam Barth.

        webkit-patch should retry on invalid password
        https://bugs.webkit.org/show_bug.cgi?id=33955

        Ask for bugs.webkit.org authentication in a loop.

        * Scripts/webkitpy/bugzilla.py:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@53733 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoNo review. Updated Qt layout tests due to r53722.
jianli@chromium.org [Sat, 23 Jan 2010 00:10:47 +0000 (00:10 +0000)]
No review. Updated Qt layout tests due to r53722.

* platform/qt/fast/dom/Window/window-properties-expected.txt:
* platform/qt/fast/dom/Window/window-property-descriptors-expected.txt:
* platform/qt/fast/dom/prototype-inheritance-expected.txt:
* platform/qt/fast/js/global-constructors-expected.txt:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@53732 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-01-22 Adam Barth <abarth@webkit.org>
abarth@webkit.org [Sat, 23 Jan 2010 00:06:53 +0000 (00:06 +0000)]
2010-01-22  Adam Barth  <abarth@webkit.org>

        Unreviewed.  Don't call seek on a NoneType.

        * Scripts/webkitpy/statusserver.py:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@53731 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-01-22 Petri Latvala <petri.latvala@nomovok.com>
abarth@webkit.org [Fri, 22 Jan 2010 23:58:44 +0000 (23:58 +0000)]
2010-01-22  Petri Latvala  <petri.latvala@nomovok.com>

        Reviewed by David Levin.

        check-webkit-style breaks on files with unknown types
        https://bugs.webkit.org/show_bug.cgi?id=34001

        For files with type FileType.NONE, dispatch_processor returns None.

        * Scripts/webkitpy/style/checker.py: Don't call process_file with a processor of value None.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@53730 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-01-22 Adam Barth <abarth@webkit.org>
abarth@webkit.org [Fri, 22 Jan 2010 23:58:03 +0000 (23:58 +0000)]
2010-01-22  Adam Barth  <abarth@webkit.org>

        Reviewed by Eric Seidel.

        Make bugzilla.py and webkitport.py conform to pep8
        https://bugs.webkit.org/show_bug.cgi?id=34015

        This patch makes webkitport.py and bugzilla.py mostly conform to PEP8
        style as enforced by pep8.py.  I wasn't able to get rid of all the
        errors because I'm not sure how to wrap some lines properly.  Also,
        there are a few deprication errors that I couldn't resolve easily.
        However, this is a massive improvement in compliance.

        * Scripts/webkitpy/bugzilla.py:
        * Scripts/webkitpy/webkitport.py:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@53729 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-01-22 Adam Barth <abarth@webkit.org>
abarth@webkit.org [Fri, 22 Jan 2010 23:57:30 +0000 (23:57 +0000)]
2010-01-22  Adam Barth  <abarth@webkit.org>

        Reviewed by Darin Adler.

        Limit length of EWS results to 5MB
        https://bugs.webkit.org/show_bug.cgi?id=34016

        Hopefully this will fix the bug where the results link doesn't appear.
        Our current theory is that the results blob is too big and the server
        is rejecting the request with a 500 error.  That causes us to re-try
        the post, but when we re-try the StringIO buffer has its seek pointer
        at the end.

        * Scripts/webkitpy/commands/queues.py:
        * Scripts/webkitpy/statusserver.py:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@53728 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoNo review. Build fix.
jianli@chromium.org [Fri, 22 Jan 2010 23:54:27 +0000 (23:54 +0000)]
No review. Build fix.
Updated the newly added test to count the change from r53722.

* fast/dom/Window/window-property-descriptors-expected.txt:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@53727 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago[Qt] Platform dependent expected files added for tests introduced in r53656,
ossy@webkit.org [Fri, 22 Jan 2010 23:50:36 +0000 (23:50 +0000)]
[Qt] Platform dependent expected files added for tests introduced in r53656,
because these tests pass. Compared to png files.

Reviewed by Nikolas Zimmermann.

* platform/qt/svg/custom/relative-sized-shadow-tree-content-with-symbol-expected.txt: Added.
* platform/qt/svg/custom/relative-sized-use-on-symbol-expected.txt: Added.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@53726 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-01-22 Eric Carlson <eric.carlson@apple.com>
eric.carlson@apple.com [Fri, 22 Jan 2010 23:49:39 +0000 (23:49 +0000)]
2010-01-22  Eric Carlson  <eric.carlson@apple.com>

        Reviewed by Simon Fraser.

        Update r53711 for GraphicsLayer method rename.

        * platform/graphics/win/MediaPlayerPrivateQuickTimeWin.cpp:
        (WebCore::MediaPlayerPrivate::createLayerForMovie): Rename from change setContentsToVideo.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@53725 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoNot reviewed, test fix.
pkasting@chromium.org [Fri, 22 Jan 2010 23:41:17 +0000 (23:41 +0000)]
Not reviewed, test fix.

Update a layout test for change in page scroll size.
https://bugs.webkit.org/show_bug.cgi?id=32595

* fast/events/scrollbar-double-click-expected.txt:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@53724 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-01-22 Nikolas Zimmermann <nzimmermann@rim.com>
zimmermann@webkit.org [Fri, 22 Jan 2010 23:36:57 +0000 (23:36 +0000)]
2010-01-22  Nikolas Zimmermann  <nzimmermann@rim.com>

        Not reviewed. Fix comment in the a SVG <use> testcase, it should say "green rectangles" instead of "blue rectangles", as noticed by Csaba Osztrogonac.

        * platform/mac/svg/custom/relative-sized-shadow-tree-content-with-symbol-expected.checksum:
        * platform/mac/svg/custom/relative-sized-shadow-tree-content-with-symbol-expected.png:
        * platform/mac/svg/custom/relative-sized-shadow-tree-content-with-symbol-expected.txt:
        * svg/custom/relative-sized-shadow-tree-content-with-symbol.xhtml:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@53723 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoWebCore: Add BlobConstructor to DOMWindow.
jianli@chromium.org [Fri, 22 Jan 2010 23:03:27 +0000 (23:03 +0000)]
WebCore: Add BlobConstructor to DOMWindow.
https://bugs.webkit.org/show_bug.cgi?id=33982

Reviewed by Darin Adler.

* page/DOMWindow.idl:

LayoutTests: Updated the affected tests due to adding BlobConstructor to DOMWindow.
https://bugs.webkit.org/show_bug.cgi?id=33982

Reviewed by Darin Adler.

* fast/dom/Window/window-properties-expected.txt:
* fast/dom/prototype-inheritance-2-expected.txt:
* fast/dom/prototype-inheritance-expected.txt:
* fast/js/global-constructors-expected.txt:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@53722 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago[WebFrame loadRequest:] does not null check the WebCore::Frame
beidson@apple.com [Fri, 22 Jan 2010 23:03:01 +0000 (23:03 +0000)]
[WebFrame loadRequest:] does not null check the WebCore::Frame
<rdar://problem/7416286> and https://bugs.webkit.org/show_bug.cgi?id=34021

Reviewed by Adele Peterson.

* WebView/WebFrame.mm:
(-[WebFrame loadRequest:]): Null check the WebCore::Frame as it can disappear
  out of underneath a WebFrame.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@53721 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoAdd test for new attributes introduced in File and Blob interfaces.
jianli@chromium.org [Fri, 22 Jan 2010 22:59:12 +0000 (22:59 +0000)]
Add test for new attributes introduced in File and Blob interfaces.
https://bugs.webkit.org/show_bug.cgi?id=33980

Reviewed by Darin Adler.

* editing/pasteboard/file-input-files-access-expected.txt:
* editing/pasteboard/script-tests/file-input-files-access.js:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@53720 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoRubber-stamped by Adam Roben.
bweinstein@apple.com [Fri, 22 Jan 2010 22:18:59 +0000 (22:18 +0000)]
Rubber-stamped by Adam Roben.

Commit files that were supposed to be part of r53696, but weren't
landed.

* English.lproj/localizedStrings.js:
* manual-tests/debugger-caught-uncaught-exceptions.html: Added.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@53719 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoWhen scrolling by page, hold back 1/8th of the visible size instead of
pkasting@chromium.org [Fri, 22 Jan 2010 22:17:17 +0000 (22:17 +0000)]
When scrolling by page, hold back 1/8th of the visible size instead of
40 px.
https://bugs.webkit.org/show_bug.cgi?id=32595

Reviewed by David Hyatt.

WebCore:

* editing/EditorCommand.cpp:
(WebCore::verticalScrollDistance):
* platform/ScrollView.cpp:
(WebCore::ScrollView::updateScrollbars):
(WebCore::ScrollView::wheelEvent):
* platform/Scrollbar.h:
* platform/wx/ScrollViewWx.cpp:
(WebCore::ScrollView::ScrollViewPrivate::OnScrollWinEvents):
* rendering/RenderLayer.cpp:
(WebCore::RenderLayer::updateScrollInfoAfterLayout):

WebKit/mac:

* WebView/WebFrameView.mm:
(-[WebFrameView _verticalPageScrollDistance]):
(-[WebFrameView initWithFrame:]):
(-[WebFrameView _horizontalPageScrollDistance]):

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@53718 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoNot reviewed, attempted Snow Leopard build fix.
dimich@chromium.org [Fri, 22 Jan 2010 22:13:23 +0000 (22:13 +0000)]
Not reviewed, attempted Snow Leopard build fix.

* wtf/ThreadingPthreads.cpp: Add a forward declaration of a function which is not 'static'.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@53717 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoNot reviewed, backout.
pkasting@chromium.org [Fri, 22 Jan 2010 22:05:40 +0000 (22:05 +0000)]
Not reviewed, backout.

Back out r52673, which caused several regressions.
https://bugs.webkit.org/show_bug.cgi?id=32533

WebCore:

* platform/PopupMenuClient.h:
* platform/chromium/PopupMenuChromium.cpp:
(WebCore::PopupContainer::hidePopup):
(WebCore::PopupListBox::handleKeyEvent):
(WebCore::PopupListBox::abandon):
(WebCore::PopupListBox::acceptIndex):
(WebCore::PopupListBox::selectIndex):
(WebCore::PopupListBox::clearSelection):
(WebCore::PopupListBox::hidePopup):
* platform/gtk/PopupMenuGtk.cpp:
(WebCore::PopupMenu::menuUnmapped):
* platform/mac/PopupMenuMac.mm:
(WebCore::PopupMenu::show):
* platform/qt/QtAbstractWebPopup.cpp:
(WebCore::QtAbstractWebPopup::popupDidHide):
* platform/qt/QtAbstractWebPopup.h:
* platform/win/PopupMenuWin.cpp:
(WebCore::PopupMenu::hide):
* platform/wx/PopupMenuWx.cpp:
(WebCore::PopupMenu::OnMenuItemSelected):
* rendering/RenderMenuList.cpp:
(WebCore::RenderMenuList::popupDidHide):
* rendering/RenderMenuList.h:
* rendering/RenderTextControlSingleLine.cpp:
(WebCore::RenderTextControlSingleLine::popupDidHide):
* rendering/RenderTextControlSingleLine.h:

WebKit/chromium:

* src/AutocompletePopupMenuClient.cpp:
(WebKit::AutocompletePopupMenuClient::initialize):
(WebKit::AutocompletePopupMenuClient::popupDidHide):
(WebKit::AutocompletePopupMenuClient::setTextFromItem):
* src/AutocompletePopupMenuClient.h:

WebKit/qt:

* WebCoreSupport/QtFallbackWebPopup.cpp:
(WebCore::QtFallbackWebPopupCombo::hidePopup):

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@53716 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago33791: Proposed patch 1 (amended with preferred license)
cjerdonek@webkit.org [Fri, 22 Jan 2010 22:04:51 +0000 (22:04 +0000)]
33791: Proposed patch 1 (amended with preferred license)

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@53715 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoFix the leak of ThreadIdentifiers in threadMap across threads.
dimich@chromium.org [Fri, 22 Jan 2010 22:00:37 +0000 (22:00 +0000)]
Fix the leak of ThreadIdentifiers in threadMap across threads.
https://bugs.webkit.org/show_bug.cgi?id=32689

Reviewed by Maciej Stachowiak.

JavaScriptCore:

Test is added to DumpRenderTree.mm.

* Android.mk: Added file ThreadIdentifierDataPthreads.(h|cpp) to build.
* Android.v8.wtf.mk: Ditto.
* GNUmakefile.am: Ditto.
* JavaScriptCore.gyp/JavaScriptCore.gyp: Ditto.
* JavaScriptCore.gypi: Ditto.
* JavaScriptCore.xcodeproj/project.pbxproj: Ditto.

* wtf/ThreadIdentifierDataPthreads.cpp: Added. Contains custom implementation of thread-specific data that uses custom destructor.
(WTF::ThreadIdentifierData::~ThreadIdentifierData): Removes the ThreadIdentifier from the threadMap.
(WTF::ThreadIdentifierData::identifier):
(WTF::ThreadIdentifierData::initialize):
(WTF::ThreadIdentifierData::destruct): Custom thread-specific destructor. Resets the value for the key again to cause second invoke.
(WTF::ThreadIdentifierData::initializeKeyOnceHelper):
(WTF::ThreadIdentifierData::initializeKeyOnce): Need to use pthread_once since initialization may come on any thread(s).
* wtf/ThreadIdentifierDataPthreads.h: Added.
(WTF::ThreadIdentifierData::ThreadIdentifierData):

* wtf/Threading.cpp:
(WTF::threadEntryPoint): Move initializeCurrentThreadInternal to after the lock to make
                         sure it is invoked when ThreadIdentifier is already established.

* wtf/Threading.h: Rename setThreadNameInternal -> initializeCurrentThreadInternal since it does more then only set the name now.
* wtf/ThreadingNone.cpp:
(WTF::initializeCurrentThreadInternal): Ditto.
* wtf/ThreadingWin.cpp:
(WTF::initializeCurrentThreadInternal): Ditto.
(WTF::initializeThreading): Ditto.
* wtf/gtk/ThreadingGtk.cpp:
(WTF::initializeCurrentThreadInternal): Ditto.
* wtf/qt/ThreadingQt.cpp:
(WTF::initializeCurrentThreadInternal): Ditto.

* wtf/ThreadingPthreads.cpp:
(WTF::establishIdentifierForPthreadHandle):
(WTF::clearPthreadHandleForIdentifier): Make it not 'static' so the ~ThreadIdentifierData() in another file can call it.
(WTF::initializeCurrentThreadInternal): Set the thread-specific data. The ThreadIdentifier is already established by creating thread.
(WTF::waitForThreadCompletion): Remove call to clearPthreadHandleForIdentifier(threadID) since it is now done in ~ThreadIdentifierData().
(WTF::detachThread): Ditto.
(WTF::currentThread): Use the thread-specific data to get the ThreadIdentifier. It's many times faster then Mutex-protected iteration through the map.
                      Also, set the thread-specific data if called first time on the thread.

WebKitTools:

Add a new test to verify the ThreadIdentifiers are not reused across threads.
The test runs in the beginning of DumpRenderTree and spawns 2 non-WTF treads sequentially,
waiting for the previous thread to terminate before starting the next.
The treads use WTF::currentThread() in their thread function. Without a fix, this
causes both threads to have the same ThreadIdentifier which triggers ASSERT in thread function.
It also starts another thread using WTF. Without the fix, this finds pthread handle from previous
threads in the WTF threadMap and asserts in WTF::establishIdentifierForPthreadHandle().
The test practically does not affect the DRT run time because the threads end immediately.

* DumpRenderTree/mac/DumpRenderTree.mm:
(runThread): Test thread function.
(testThreadIdentifierMap):
(dumpRenderTree):

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@53714 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoRubber-stamped by Darin Adler.
mitz@apple.com [Fri, 22 Jan 2010 21:52:40 +0000 (21:52 +0000)]
Rubber-stamped by Darin Adler.

Revert r53693 because it broke scrolling of pages with fixed elements on
Mac OS X.

* page/FrameView.cpp:
* page/FrameView.h:
* platform/ScrollView.cpp:
(WebCore::ScrollView::scrollContents):
* platform/ScrollView.h:
* rendering/RenderObject.cpp:
(WebCore::RenderObject::styleWillChange):
(WebCore::RenderObject::destroy):

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@53713 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago Reviewed by Darin Adler.
ap@apple.com [Fri, 22 Jan 2010 20:59:13 +0000 (20:59 +0000)]
    Reviewed by Darin Adler.

        https://bugs.webkit.org/show_bug.cgi?id=34008
        Assertion failure in KURL::setProtocol when running DOM Fuzzer

        Test: fast/dom/Window/invalid-protocol.html

        * bindings/js/JSLocationCustom.cpp: (WebCore::JSLocation::setProtocol): Raise an exception
        if KURL::setProtocol fails.

        * html/HTMLAnchorElement.cpp: (WebCore::HTMLAnchorElement::setProtocol): Move argument
        tweaking logic into KURL. This way, the ':' trick applies to both JSLocation and
        HTMLAnchorElement, matching IE (but not Firefox). IE behavior is more permissive, and even
        more logical in my opinion.

        * loader/FrameLoader.cpp: (WebCore::FrameLoader::iconURL): Assert that setting protocol
        succeeded.

        * platform/KURL.cpp: (WebCore::KURL::setProtocol): Remove everything past ':', if present.
        Return false if the protocol to set is not valid.
        (WebCore::isValidProtocol): Made this work correctly for empty strings.

        * platform/KURL.h: isValidProtocol() is now static in KURL.cpp, it's only used in setProtocol().

        * platform/KURLGoogle.cpp:
        (WebCore::KURL::setProtocol): Always return true. This should hopefully prevent Chromium build
        breakage, alhough tests will likely fail.
        (WebCore::isValidProtocol): Removed, as this isn't used at the moment.

        * websockets/WebSocketHandshake.cpp: (WebCore::WebSocketHandshake::httpURLForAuthenticationAndCookies):
        Assert that setting protocol succeeded.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@53712 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-01-22 Eric Carlson <eric.carlson@apple.com>
eric.carlson@apple.com [Fri, 22 Jan 2010 20:40:13 +0000 (20:40 +0000)]
2010-01-22  Eric Carlson  <eric.carlson@apple.com>

        Reviewed by Adam Roben.

        <rdar://problem/7373568>
        https://bugs.webkit.org/show_bug.cgi?id=34007
        Implement hardware acceleration of video compositing on Windows

        Internally we still have QuickTime render to a bitmap, but frames are then composited
        with the page using an accelerated renderer.

        * platform/graphics/win/MediaPlayerPrivateQuickTimeWin.cpp:
        (WebCore::MediaPlayerPrivate::MediaPlayerPrivate): Initialize m_newFrameAvailable.
        (WebCore::MediaPlayerPrivate::~MediaPlayerPrivate): Call tearDownVideoRendering.
        (WebCore::MediaPlayerPrivate::cancelLoad): Call tearDownVideoRendering.
        (WebCore::MediaPlayerPrivate::updateStates): Call setUpVideoRendering once the movie
            is ready to draw.
        (WebCore::MediaPlayerPrivate::isReadyForRendering): New, utility function returns true
            only when movie is ready to render.
        (WebCore::MediaPlayerPrivate::setSize): Remember the size as we need it when rendering
            in accelerated mode.
        (WebCore::MediaPlayerPrivate::setVisible): Call setUpVideoRendering when made visible.
        (WebCore::MediaPlayerPrivate::paint): Do nothing when rendering to a layer. Move frame
            rate drawing code to paintCompleted.
        (WebCore::MediaPlayerPrivate::paintCompleted): New, clear m_newFrameAvailable and, when
            built with DRAW_FRAME_RATE, draw framerate.
        (WebCore::MediaPlayerPrivate::movieNewImageAvailable): Set m_newFrameAvailable, trigger
            repaint differently when drawing to a layer.
        (WebCore::MediaPlayerPrivate::currentRenderingMode): New, return the current rendering mode.
        (WebCore::MediaPlayerPrivate::preferredRenderingMode): New, return preferred rendering mode
            (render to a layer whenever possible).
        (WebCore::MediaPlayerPrivate::setUpVideoRendering): New, tear down current rendering mode
            and set up new mode when necessary.
        (WebCore::MediaPlayerPrivate::tearDownVideoRendering): New, noop unless rendering to a layer.
        (WebCore::MediaPlayerPrivate::hasSetUpVideoRendering): New.
        (WebCore::MediaPlayerPrivate::paintContents): New, callback from compositing layer when it
            is time to draw the current frame.
        (WebCore::MediaPlayerPrivate::createLayerForMovie): New, allocate a layer for the movie.
        (WebCore::MediaPlayerPrivate::destroyLayerForMovie): New, delete movie layer.
        (WebCore::MediaPlayerPrivate::supportsAcceleratedRendering): New, return true unless the
            movie is not ready to render.
        (WebCore::MediaPlayerPrivate::acceleratedRenderingStateChanged): New, deal with change in
            accelerated rendering state.
        (WebCore::MediaPlayerPrivate::notifySyncRequired): New, callback from compositing layer
            when it wants to schedule a redraw.
        * platform/graphics/win/MediaPlayerPrivateQuickTimeWin.h:
        (WebCore::MediaPlayerPrivate::notifyAnimationStarted):
        (WebCore::MediaPlayerPrivate::showDebugBorders):
        (WebCore::MediaPlayerPrivate::showRepaintCounter):

        * platform/graphics/win/QTMovieWin.cpp:
        (QTMovieWin::getCurrentFrameInfo):
            New, return information about the offscreen render buffer.
        (QTMovieWin::initializeQuickTime):
        * platform/graphics/win/QTMovieWin.h:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@53711 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-01-21 Jeremy Orlow <jorlow@chromium.org>
jorlow@chromium.org [Fri, 22 Jan 2010 20:33:41 +0000 (20:33 +0000)]
2010-01-21  Jeremy Orlow  <jorlow@chromium.org>

        Reviewed by Darin Fisher.

        The Chromium WebKit API needs to expose storage event related data
        https://bugs.webkit.org/show_bug.cgi?id=33985

        setItem and removeItem on WebStorageArea need to expose what the previous
        value was for the key being modified.  Clear needs to return whether it
        actually cleared anything.

        * public/WebStorageArea.h:
        (WebKit::WebStorageArea::setItem):
        (WebKit::WebStorageArea::removeItem):
        (WebKit::WebStorageArea::clear):
        * src/StorageAreaProxy.cpp:
        (WebCore::StorageAreaProxy::StorageAreaProxy):
        (WebCore::StorageAreaProxy::setItem):
        (WebCore::StorageAreaProxy::removeItem):
        (WebCore::StorageAreaProxy::clear):
        (WebCore::StorageAreaProxy::storageEvent):
        * src/StorageAreaProxy.h:
        * src/StorageNamespaceProxy.cpp:
        (WebCore::StorageNamespace::localStorageNamespace):
        (WebCore::StorageNamespace::sessionStorageNamespace):
        (WebCore::StorageNamespaceProxy::StorageNamespaceProxy):
        (WebCore::StorageNamespaceProxy::copy):
        (WebCore::StorageNamespaceProxy::storageArea):
        * src/StorageNamespaceProxy.h:
        * src/WebStorageAreaImpl.cpp:
        (WebKit::WebStorageAreaImpl::setItem):
        (WebKit::WebStorageAreaImpl::removeItem):
        (WebKit::WebStorageAreaImpl::clear):
        * src/WebStorageAreaImpl.h:
2010-01-21  Jeremy Orlow  <jorlow@chromium.org>

        Reviewed by Darin Fisher.

        The Chromium WebKit API needs to expose storage event related data
        https://bugs.webkit.org/show_bug.cgi?id=33985

        This change is not visible to layoutTests/web pages.

        * storage/StorageArea.h:
        * storage/StorageAreaImpl.cpp:
        (WebCore::StorageAreaImpl::setItem): return the old value
        (WebCore::StorageAreaImpl::removeItem): return the old value
        (WebCore::StorageAreaImpl::clear): return whether there was anything to clear
        * storage/StorageAreaImpl.h:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@53710 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoAn element that doesn't validate and an invalid element shouldn't share style
adele@apple.com [Fri, 22 Jan 2010 20:06:46 +0000 (20:06 +0000)]
An element that doesn't validate and an invalid element shouldn't share style
https://bugs.webkit.org/show_bug.cgi?id=34010

Reviewed by Dan Bernstein.

Test: fast/css/pseudo-invalid-novalidate-001.html

* css/CSSStyleSelector.cpp: (WebCore::CSSStyleSelector::canShareStyleWithElement):
  Separate the check for willValidate from the check for isValidFormControlElement.
* html/HTMLFormControlElement.h: The validity method doesn't need to be virtual.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@53709 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoCrash in fast/dom/Window/window-properties if Geolocation enabled
sfalken@apple.com [Fri, 22 Jan 2010 19:49:56 +0000 (19:49 +0000)]
Crash in fast/dom/Window/window-properties if Geolocation enabled
https://bugs.webkit.org/show_bug.cgi?id=34013

Reviewed by Dan Bernstein.

Add null check to fix Geolocation crash in fast/dom/Window/window-properties.

* page/Geolocation.cpp:
(WebCore::createGeoposition):

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@53708 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoFix for the failing test on qt buildbot.
enrica@apple.com [Fri, 22 Jan 2010 19:13:10 +0000 (19:13 +0000)]
Fix for the failing test on qt buildbot.
This tests had been added as part of the patch for https://bugs.webkit.org/show_bug.cgi?id=33970

Reviewed by Adele Peterson.

* editing/pasteboard/paste-noscript-svg-expected.txt:
* editing/resources/svgcontent.xhtml:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@53707 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-01-22 Kent Hansen <kent.hansen@nokia.com>
eric@webkit.org [Fri, 22 Jan 2010 19:09:47 +0000 (19:09 +0000)]
2010-01-22  Kent Hansen  <kent.hansen@nokia.com>

        Reviewed by Darin Adler.

        Object.getOwnPropertyDescriptor(window) returns descriptors for properties in the prototype chain
        https://bugs.webkit.org/show_bug.cgi?id=33948

        Even though prototype properties are proxied by JSDOMWindow::getOwnPropertySlot(),
        that proxying should not be performed by JSDOMWindow::getOwnPropertyDescriptor().

        This makes getOwnPropertyDescriptor() consistent with getOwnPropertyNames().

        Test: fast/dom/Window/window-property-descriptors.html

        * bindings/js/JSDOMWindowCustom.cpp:
        (WebCore::JSDOMWindow::getOwnPropertyDescriptor):

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@53706 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-01-21 Darin Fisher <darin@chromium.org>
darin@chromium.org [Fri, 22 Jan 2010 18:44:51 +0000 (18:44 +0000)]
2010-01-21  Darin Fisher  <darin@chromium.org>

        Reviewed by David Levin.

        [Chromium] spurious WebViewClient::didStopLoading if changing
        location.hash while a subframe is still loading

        https://bugs.webkit.org/show_bug.cgi?id=33884

        This is a refinement of http://trac.webkit.org/changeset/51548

        * src/FrameLoaderClientImpl.cpp:
        (WebKit::FrameLoaderClientImpl::dispatchDidChangeLocationWithinPage):
        (WebKit::FrameLoaderClientImpl::postProgressStartedNotification):
        (WebKit::FrameLoaderClientImpl::postProgressFinishedNotification):
        * src/WebViewImpl.cpp:
        (WebKit::WebViewImpl::WebViewImpl):
        (WebKit::WebViewImpl::didStartLoading):
        (WebKit::WebViewImpl::didStopLoading):
        * src/WebViewImpl.h:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@53705 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoCheck in expected failure results for text-zoom.xhtml on Windows
aroben@apple.com [Fri, 22 Jan 2010 18:31:24 +0000 (18:31 +0000)]
Check in expected failure results for text-zoom.xhtml on Windows

Fixing this test is covered by <http://webkit.org/b/34006>.

Rubber-stamped by Dan Bernstein.

* platform/win/svg/custom/text-zoom-expected.txt: Added.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@53704 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago[Qt] Save the QWebPageClient instead of the ownerWidget in QtAbstractWebPopup
girish@forwardbias.in [Fri, 22 Jan 2010 17:01:57 +0000 (17:01 +0000)]
[Qt] Save the QWebPageClient instead of the ownerWidget in QtAbstractWebPopup

Reviewed by Simon Hausmann.

The QWebPageClient is required for the QtFallbackWebPopup. QtFallbackWebPopup will
need it to create a QGraphicsProxyWidget (in a future commit) for the
QGraphicsWebView's web popup.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@53703 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago[Qt] http/tests/misc/slow-loading-mask.html timed out.
ossy@webkit.org [Fri, 22 Jan 2010 16:11:58 +0000 (16:11 +0000)]
[Qt] http/tests/misc/slow-loading-mask.html timed out.
https://bugs.webkit.org/show_bug.cgi?id=34004

* platform/qt/Skipped: http/tests/misc/slow-loading-mask.html skipped.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@53702 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-01-22 Shinichiro Hamaji <hamaji@chromium.org>
hamaji@chromium.org [Fri, 22 Jan 2010 14:28:56 +0000 (14:28 +0000)]
2010-01-22  Shinichiro Hamaji  <hamaji@chromium.org>

        Reviewed by Darin Adler.

        incorrect results for css2.1 layout tests that use setTimeout in body onload event
        https://bugs.webkit.org/show_bug.cgi?id=17557

        * css2.1/README.txt: Added.
        * css2.1/t1204-increment-00-c-o.html-disabled: Renamed from LayoutTests/css2.1/t1204-increment-00-c-o.html.
        * css2.1/t1204-increment-01-c-o.html-disabled: Renamed from LayoutTests/css2.1/t1204-increment-01-c-o.html.
        * css2.1/t1204-increment-02-c-o.html-disabled: Renamed from LayoutTests/css2.1/t1204-increment-02-c-o.html.
        * css2.1/t1204-reset-00-c-o.html-disabled: Renamed from LayoutTests/css2.1/t1204-reset-00-c-o.html.
        * css2.1/t1204-reset-01-c-o.html-disabled: Renamed from LayoutTests/css2.1/t1204-reset-01-c-o.html.
        * css2.1/t1204-reset-02-c-o.html-disabled: Renamed from LayoutTests/css2.1/t1204-reset-02-c-o.html.
        * fast/css/counters/counter-increment-000-expected.txt: Renamed from LayoutTests/fast/css/counters/t1204-increment-00-c-o-expected.txt.
        * fast/css/counters/counter-increment-000.html: Renamed from LayoutTests/fast/css/counters/t1204-increment-00-c-o.html.
        * fast/css/counters/counter-increment-001-expected.txt: Renamed from LayoutTests/fast/css/counters/t1204-increment-01-c-o-expected.txt.
        * fast/css/counters/counter-increment-001.html: Renamed from LayoutTests/fast/css/counters/t1204-increment-01-c-o.html.
        * fast/css/counters/counter-reset-001-expected.txt: Added.
        * fast/css/counters/counter-reset-001.html: Added.
        * platform/mac/css2.1/t1204-increment-00-c-o-expected.checksum: Removed.
        * platform/mac/css2.1/t1204-increment-00-c-o-expected.png: Removed.
        * platform/mac/css2.1/t1204-increment-00-c-o-expected.txt: Removed.
        * platform/mac/css2.1/t1204-increment-01-c-o-expected.checksum: Removed.
        * platform/mac/css2.1/t1204-increment-01-c-o-expected.png: Removed.
        * platform/mac/css2.1/t1204-increment-01-c-o-expected.txt: Removed.
        * platform/mac/css2.1/t1204-increment-02-c-o-expected.checksum: Removed.
        * platform/mac/css2.1/t1204-increment-02-c-o-expected.png: Removed.
        * platform/mac/css2.1/t1204-increment-02-c-o-expected.txt: Removed.
        * platform/mac/css2.1/t1204-reset-00-c-o-expected.checksum: Removed.
        * platform/mac/css2.1/t1204-reset-00-c-o-expected.png: Removed.
        * platform/mac/css2.1/t1204-reset-00-c-o-expected.txt: Removed.
        * platform/mac/css2.1/t1204-reset-01-c-o-expected.checksum: Removed.
        * platform/mac/css2.1/t1204-reset-01-c-o-expected.png: Removed.
        * platform/mac/css2.1/t1204-reset-01-c-o-expected.txt: Removed.
        * platform/mac/css2.1/t1204-reset-02-c-o-expected.checksum: Removed.
        * platform/mac/css2.1/t1204-reset-02-c-o-expected.png: Removed.
        * platform/mac/css2.1/t1204-reset-02-c-o-expected.txt: Removed.
        * platform/qt/css2.1/t1204-increment-00-c-o-expected.txt: Removed.
        * platform/qt/css2.1/t1204-increment-01-c-o-expected.txt: Removed.
        * platform/qt/css2.1/t1204-increment-02-c-o-expected.txt: Removed.
        * platform/qt/css2.1/t1204-reset-00-c-o-expected.txt: Removed.
        * platform/qt/css2.1/t1204-reset-01-c-o-expected.txt: Removed.
        * platform/qt/css2.1/t1204-reset-02-c-o-expected.txt: Removed.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@53701 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-01-22 Janusz Lewandowski <lew21@xtreeme.org>
hamaji@chromium.org [Fri, 22 Jan 2010 14:13:27 +0000 (14:13 +0000)]
2010-01-22  Janusz Lewandowski  <lew21@xtreeme.org>

        Reviewed by Maciej Stachowiak.

        run-webkit-tests fails on Ruby 1.9
        https://bugs.webkit.org/show_bug.cgi?id=33554
        inject does not work with strings any more, so BugsSite/PrettyPatch/PrettyPatch.rb prints lots of errors.

        * PrettyPatch/PrettyPatch.rb:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@53700 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoUnreviewed Qt "build" fix.
hausmann@webkit.org [Fri, 22 Jan 2010 13:42:35 +0000 (13:42 +0000)]
Unreviewed Qt "build" fix.

* WebCore.pro: Fix warning about missing header file that isn't present anymore.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@53699 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoQState::polished() was renamed to QState::propertiesAssigned() when
girish@forwardbias.in [Fri, 22 Jan 2010 13:34:46 +0000 (13:34 +0000)]
QState::polished() was renamed to QState::propertiesAssigned() when
Qt 4.6.0 was released.

Reviewed by Kenneth Rohde Christiansen.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@53698 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago[Qt] Skip test introduced in r53684 because of missing eventSender.textZoomIn()
ossy@webkit.org [Fri, 22 Jan 2010 12:56:31 +0000 (12:56 +0000)]
[Qt] Skip test introduced in r53684 because of missing eventSender.textZoomIn()

* platform/qt/Skipped: svg/custom/text-zoom.xhtml skipped.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@53697 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-01-22 Brian Weinstein <bweinstein@apple.com>
yurys@chromium.org [Fri, 22 Jan 2010 12:52:52 +0000 (12:52 +0000)]
2010-01-22  Brian Weinstein  <bweinstein@apple.com>

        Reviewed by Timothy Hatcher.

        Web Inspector: Caught exceptions still pause the debugger.
        <https://bugs.webkit.org/show_bug.cgi?id=28622>.

        Turn the pauseOnExceptions variable from a bool to an int, because we now have
        3 states, in an enum: DontPauseOnExceptions, PauseOnAllExceptions, and
        PauseOnUncaughtExceptions. The status button for pausing on exceptions is now
        a tri-state button, which cycles from Don't pause (no background) to Pause on All
        (blue background) to Pause on Uncaught (purple background).

        Also added the ability for a status button to have more than 2 states, and added
        style rules for a CSS three state button.

        Added a manual test, manual-tests/debugger-caught-uncaught-exceptions.html, that
        generates caught and uncaught exceptions.

        * English.lproj/localizedStrings.js: Added updated Action / Click to Action text.
        * inspector/InspectorBackend.cpp:
        (WebCore::InspectorBackend::pauseOnExceptions): Changed from a bool to an int.
        (WebCore::InspectorBackend::setPauseOnExceptions): Ditto.
        * inspector/InspectorBackend.h: Ditto.
        * inspector/InspectorBackend.idl: Ditto.
        * inspector/JavaScriptDebugServer.cpp:
        (WebCore::JavaScriptDebugServer::JavaScriptDebugServer): Changed from bool to PauseOnExceptionsState.
        (WebCore::JavaScriptDebugServer::setPauseOnExceptions): Ditto.
        (WebCore::JavaScriptDebugServer::exception): Updated logic to see if we are breaking
            on uncaught exceptions, and check if we have a handler.
        * inspector/JavaScriptDebugServer.h:
        (WebCore::JavaScriptDebugServer::): Changed from bool to PauseOnExceptionsState.
        (WebCore::JavaScriptDebugServer::pauseOnExceptions): Ditto.
        * inspector/front-end/InspectorBackendStub.js:
        (.WebInspector.InspectorBackendStub.prototype.pauseOnExceptionsState): Changed from returning false to returning 0 (stub function).
        * inspector/front-end/ScriptsPanel.js:
        (WebInspector.ScriptsPanel):
        (WebInspector.ScriptsPanel.prototype._updatePauseOnExceptionsButton): Updated to new text/new cycling of variables.
        (WebInspector.ScriptsPanel.prototype._togglePauseOnExceptions): Updated to new text/new cycling of variables.
        * inspector/front-end/StatusBarButton.js:
        (WebInspector.StatusBarButton): Added the option for a different number of states.
        (WebInspector.StatusBarButton.prototype.set toggled): Added rules to toggle on/off with multiple states.
        * inspector/front-end/inspector.css: Added rules for toggled-1 and toggled-2.
        * manual-tests/debugger-caught-uncaught-exceptions.html: Added.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@53696 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-01-22 Ben Murdoch <benm@google.com>
eric@webkit.org [Fri, 22 Jan 2010 12:30:23 +0000 (12:30 +0000)]
2010-01-22  Ben Murdoch  <benm@google.com>

        Reviewed by Eric Seidel.

        [Android] Android platform specific PlatformTouchEvent and PlatformTouchPoint implementations are not upstream.
        https://bugs.webkit.org/show_bug.cgi?id=33719

        Add the Android specific files and update the Android makefile.

        No new tests as this is all Android-specific code.

        * Android.mk: Add Touch event related files to the makefile.
        * platform/PlatformTouchEvent.h: Add Android constructor.
        * platform/PlatformTouchPoint.h: Ditto.
        * platform/android/PlatformTouchEventAndroid.cpp: Added.
        (WebCore::PlatformTouchEvent::PlatformTouchEvent): Android implementation.
        * platform/android/PlatformTouchPointAndroid.cpp: Added.
        (WebCore::PlatformTouchPoint::PlatformTouchPoint): Android implementation.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@53695 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoUnreviewed trivial fix.
ossy@webkit.org [Fri, 22 Jan 2010 12:04:35 +0000 (12:04 +0000)]
Unreviewed trivial fix.

[Qt] css1_forward_compatible_parsing.html moved from css2.1 to fast/css by r53664.

* platform/qt/Skipped: path fixed.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@53694 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-01-22 Benjamin Poulain <benjamin.poulain@nokia.com>
eric@webkit.org [Fri, 22 Jan 2010 12:01:28 +0000 (12:01 +0000)]
2010-01-22  Benjamin Poulain  <benjamin.poulain@nokia.com>

        Reviewed by Kenneth Rohde Christiansen.

        Do not render the full frame when there is some elements with fixed positioning
        https://bugs.webkit.org/show_bug.cgi?id=33150

        The RenderObjects with fixed position register to the view
        to be taken into account while scrolling.
        The scrolling code update specifically those elements after blitting the pixels.

        * page/FrameView.cpp:
        (WebCore::FrameView::registerFixedPositionedObject):
        (WebCore::FrameView::unregisterFixedPositionedObject):
        (WebCore::FrameView::scrollContentsFastPath):
        * page/FrameView.h:
        * platform/ScrollView.cpp:
        (WebCore::ScrollView::scrollContents):
        (WebCore::ScrollView::scrollContentsFastPath):
        * platform/ScrollView.h:
        * rendering/RenderObject.cpp:
        (WebCore::RenderObject::styleWillChange):
        (WebCore::RenderObject::destroy):

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@53693 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-01-22 Fumitoshi Ukai <ukai@chromium.org>
ukai@chromium.org [Fri, 22 Jan 2010 11:57:14 +0000 (11:57 +0000)]
2010-01-22  Fumitoshi Ukai  <ukai@chromium.org>

        Reviewed by Maciej Stachowiak.

        Split websocket/tests/bad-sub-protocol.html
        https://bugs.webkit.org/show_bug.cgi?id=33646

        * websocket/tests/bad-sub-protocol-control-chars-expected.txt: Added.
        * websocket/tests/bad-sub-protocol-control-chars.html: Added.
        * websocket/tests/bad-sub-protocol-empty-expected.txt: Added.
        * websocket/tests/bad-sub-protocol-empty.html: Added.
        * websocket/tests/bad-sub-protocol-expected.txt: Removed.
        * websocket/tests/bad-sub-protocol-non-ascii-expected.txt: Added.
        * websocket/tests/bad-sub-protocol-non-ascii.html: Added.
        * websocket/tests/bad-sub-protocol.html: Removed.
        * websocket/tests/script-tests/bad-sub-protocol-control-chars.js: Added.
        * websocket/tests/script-tests/bad-sub-protocol-empty.js: Added.
        * websocket/tests/script-tests/bad-sub-protocol-non-ascii.js: Added.
        * websocket/tests/script-tests/bad-sub-protocol.js: Removed.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@53692 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-01-22 Kent Tamura <tkent@chromium.org>
eric@webkit.org [Fri, 22 Jan 2010 10:48:05 +0000 (10:48 +0000)]
2010-01-22  Kent Tamura  <tkent@chromium.org>

        Reviewed by Maciej Stachowiak.

        Improve portability of listbox-deselect-scroll.html
        https://bugs.webkit.org/show_bug.cgi?id=33989

        - Convert the test to dumpAsTest
        - Use eventSender.mouseDown(0, ['addSelectionKey'])
        - Avoid hard-coded line-heights.

        * fast/forms/listbox-deselect-scroll-expected.txt: Added.
        * fast/forms/listbox-deselect-scroll.html:
        * fast/forms/script-tests/listbox-deselect-scroll.js: Added.
        * platform/mac/fast/forms/listbox-deselect-scroll-expected.checksum: Removed.
        * platform/mac/fast/forms/listbox-deselect-scroll-expected.png: Removed.
        * platform/mac/fast/forms/listbox-deselect-scroll-expected.txt: Removed.
2010-01-22  Kent Tamura  <tkent@chromium.org>

        Reviewed by Maciej Stachowiak.

        Fix a bug that mouseDown:withModifiers: is never called.
        https://bugs.webkit.org/show_bug.cgi?id=33989

        * DumpRenderTree/mac/EventSendingController.mm:
        (+[EventSendingController isSelectorExcludedFromWebScript:]):
        (+[EventSendingController webScriptNameForSelector:]):

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@53691 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-01-22 Anton Muhin <antonm@chromium.org>
eric@webkit.org [Fri, 22 Jan 2010 10:26:03 +0000 (10:26 +0000)]
2010-01-22  Anton Muhin  <antonm@chromium.org>

        Reviewed by Eric Seidel.

        Adding ScriptWrappable into WebCore.
        This allows to associate custom information when Node
        gets wrapped into JavaScript wrapper.

        [v8, dom] Add a pointer field to C++ Node class
        https://bugs.webkit.org/show_bug.cgi?id=32430

        * WebCore.gypi:
        * WebCore.pro:
        * WebCore.vcproj/WebCore.vcproj:
        * WebCore.xcodeproj/project.pbxproj:
        * bindings/js/ScriptWrappable.h: Added.
        (WebCore::ScriptWrappable::ScriptWrappable):
        * bindings/v8/ScriptWrappable.h: Added.
        (WebCore::ScriptWrappable::ScriptWrappable):
        (WebCore::ScriptWrappable::wrapper):
        (WebCore::ScriptWrappable::setWrapper):
        (WebCore::ScriptWrappable::clearWrapper):
        * dom/Node.h:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@53690 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoemikhatsky <yurys@chromium.org>
yurys@chromium.org [Fri, 22 Jan 2010 09:42:09 +0000 (09:42 +0000)]
emikhatsky  <yurys@chromium.org>

        Reviewed by Pavel Feldman.

        Add stub implementation of dispatchOnInjectedScript method so that
        Chromium source can compile against it and the main fix for the bug
        28622 can be landed without breaking Chromium build.

        https://bugs.webkit.org/show_bug.cgi?id=28622

        * inspector/InspectorBackend.h:
        (WebCore::InspectorBackend::dispatchOnInjectedScript):

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@53689 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-01-22 Kwang Yul Seo <skyul@company100.net>
eric@webkit.org [Fri, 22 Jan 2010 09:19:41 +0000 (09:19 +0000)]
2010-01-22  Kwang Yul Seo  <skyul@company100.net>

        Reviewed by Dmitry Titov.

        ENABLE(DATABASE) guard is missing in ScriptExecutionContext::~ScriptExecutionContext
        https://bugs.webkit.org/show_bug.cgi?id=33990

        Add missing ENABLE(DATABASE) guard.

        * dom/ScriptExecutionContext.cpp:
        (WebCore::ScriptExecutionContext::~ScriptExecutionContext):

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@53688 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago[Qt] Separate out the WebPage class into it's own
kenneth@webkit.org [Fri, 22 Jan 2010 09:17:41 +0000 (09:17 +0000)]
[Qt] Separate out the WebPage class into it's own
cpp/header files. Also, removed the assumption that
the view is a QWebView, in preparation of a merger
of the two Qt WebKit launchers.

Rubberstamped by Antti Koivisto.

* QtLauncher/QtLauncher.pro:
* QtLauncher/main.cpp:
(WebView::mousePressEvent):
* QtLauncher/webpage.cpp: Added.
(WebPage::supportsExtension):
(WebPage::extension):
(WebPage::acceptNavigationRequest):
(WebPage::openUrlInDefaultBrowser):
* QtLauncher/webpage.h: Added.
(WebPage::WebPage):

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@53687 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2010-01-22 Xan Lopez <xlopez@igalia.com>
xan@webkit.org [Fri, 22 Jan 2010 08:42:52 +0000 (08:42 +0000)]
2010-01-22  Xan Lopez  <xlopez@igalia.com>

        Reviewed by Gustavo Noronha.

        [GTK] GObject DOM bindings
        https://bugs.webkit.org/show_bug.cgi?id=33590

        * bindings/scripts/generate-bindings.pl:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@53686 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2010-01-22 Xan Lopez <xlopez@igalia.com>
xan@webkit.org [Fri, 22 Jan 2010 08:42:35 +0000 (08:42 +0000)]
2010-01-22  Xan Lopez  <xlopez@igalia.com>

        Reviewed by Gustavo Noronha.

        [GTK] GObject DOM bindings
        https://bugs.webkit.org/show_bug.cgi?id=33590

        Rename FEATURE_DEFINES_JAVASCRIPT to FEATURE_DEFINES and remove
        the hardcoded LANGUAGE_JAVASCRIPT=1, in preparation for its use by
        the GObject DOM bindings.

        * GNUmakefile.am:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@53685 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago<rdar://problem/7568696> Zoom applied to embedded SVG incorrectly
mitz@apple.com [Fri, 22 Jan 2010 08:20:22 +0000 (08:20 +0000)]
<rdar://problem/7568696> Zoom applied to embedded SVG incorrectly
https://bugs.webkit.org/show_bug.cgi?id=33988

Reviewed by Darin Adler.

WebCore:

Test: svg/custom/text-zoom.xhtml

* svg/SVGSVGElement.cpp:
(WebCore::SVGSVGElement::SVGSVGElement): Initialize m_scale.
(WebCore::SVGSVGElement::currentScale): If this is the document element,
return the frame’s zoom factor. Otherwise, return m_scale.
(WebCore::SVGSVGElement::setCurrentScale): If this is the document element,
set the frame’s zoom factor. Otherwise, set m_scale and mark for layout.
* svg/SVGSVGElement.h: Added m_scale member.

LayoutTests:

* svg/custom/text-zoom-expected.checksum: Added.
* svg/custom/text-zoom-expected.png: Added.
* svg/custom/text-zoom-expected.txt: Added.
* svg/custom/text-zoom.xhtml: Added.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@53684 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2010-01-21 Adam Barth <abarth@webkit.org>
abarth@webkit.org [Fri, 22 Jan 2010 07:48:53 +0000 (07:48 +0000)]
2010-01-21  Adam Barth  <abarth@webkit.org>

        Unreviewed.  Added a blank line requested by the great an powerful
        fishd.

        * public/WebFrameClient.h:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@53683 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2010-01-21 Kwang Yul Seo <skyul@company100.net>
eric@webkit.org [Fri, 22 Jan 2010 07:36:29 +0000 (07:36 +0000)]
2010-01-21  Kwang Yul Seo  <skyul@company100.net>

        Reviewed by Alexey Proskuryakov.

        Add ThreadSpecific for ENABLE(SINGLE_THREADED)
        https://bugs.webkit.org/show_bug.cgi?id=33878

        Implement ThreadSpecific with a simple getter/setter
        when ENABLE(SINGLE_THREADED) is true.

        Due to the change in https://bugs.webkit.org/show_bug.cgi?id=33236,
        an implementation of ThreadSpecific must be available to build WebKit.
        This causes a build failure for platforms without a proper
        ThreadSpecific implementation.

        * wtf/ThreadSpecific.h:
        (WTF::::ThreadSpecific):
        (WTF::::~ThreadSpecific):
        (WTF::::get):
        (WTF::::set):
        (WTF::::destroy):

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@53682 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2010-01-21 Adam Barth <abarth@webkit.org>
eric@webkit.org [Fri, 22 Jan 2010 07:11:24 +0000 (07:11 +0000)]
2010-01-21  Adam Barth  <abarth@webkit.org>

        Reviewed by Eric Seidel.

        [chromium] Add allowPlugins callback to allow per-site enabling of plugins
        https://bugs.webkit.org/show_bug.cgi?id=33974

        Add a callback into the FrameLoaderClient to let the embedder enable or
        disable plugins on a per-site basis.

        * loader/FrameLoader.cpp:
        (WebCore::FrameLoader::requestObject):
        * loader/FrameLoaderClient.h:
        (WebCore::FrameLoaderClient::allowPlugins):
2010-01-21  Adam Barth  <abarth@webkit.org>

        Reviewed by Eric Seidel.

        [chromium] Add allowPlugins callback to allow per-site enabling of plugins
        https://bugs.webkit.org/show_bug.cgi?id=33974

        Forward the new allowPlugins callback to WebKitClient so that it can
        make the policy decision.

        * public/WebFrameClient.h:
        (WebKit::WebFrameClient::allowPlugins):
        * src/FrameLoaderClientImpl.cpp:
        (WebKit::FrameLoaderClientImpl::allowJavaScript):
        (WebKit::FrameLoaderClientImpl::allowPlugins):
        * src/FrameLoaderClientImpl.h:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@53681 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2010-01-21 Tony Chang <tony@chromium.org>
eric@webkit.org [Fri, 22 Jan 2010 06:53:14 +0000 (06:53 +0000)]
2010-01-21  Tony Chang  <tony@chromium.org>

        Reviewed by Darin Adler.

        https://bugs.webkit.org/show_bug.cgi?id=25501
        Remove empty font tags when applying a style to a content editable
        region.

        * editing/execCommand/change-font.html: Added.
        * editing/execCommand/change-font-expected.txt: Added.
2010-01-21  Tony Chang  <tony@chromium.org>

        Reviewed by Darin Adler.

        https://bugs.webkit.org/show_bug.cgi?id=25501
        Remove empty font tags when applying a style to a content editable
        region.

        Test: editing/execCommand/change-font.html

        * editing/ApplyStyleCommand.cpp:
        (WebCore::isEmptyFontTag):

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@53680 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years agoAdded screenshots and clearer instructions on how to submit
cjerdonek@webkit.org [Fri, 22 Jan 2010 06:37:57 +0000 (06:37 +0000)]
Added screenshots and clearer instructions on how to submit
a patch for review.

Reviewed by David Levin.

https://bugs.webkit.org/show_bug.cgi?id=32542

Also added that a bug report should be selected or created
prior to submitting a patch.

* coding/contributing.html:
* coding/images: Added.
* coding/images/contribute_add_attachment.png: Added.
* coding/images/contribute_mark_review.png: Added.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@53679 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2010-01-21 James Robinson <jamesr@chromium.org>
eric@webkit.org [Fri, 22 Jan 2010 06:21:55 +0000 (06:21 +0000)]
2010-01-21  James Robinson  <jamesr@chromium.org>

        Reviewed by Darin Fisher.

        Bumps the refcount when WebNode::toElement/toConstElement are called.
        https://bugs.webkit.org/show_bug.cgi?id=33977

        Ensures that when a wrapper around a WebNode is create via the
        toElement() or toConstElement() calls that the refcount on the WebNode
        is increased, by forcing a call to WebNode::assign().  This is needed
        so that the following code:
        {
          WebNode n = someNode();
          WebElement e = n.toElement<WebElement>();
        }
        does not decrease the overall refcount on the Node that someNode()
        returns.

        This caused crashes in the 'Save page as..' feature.

        * public/WebNode.h:
        (WebKit::WebNode::toElement):
        (WebKit::WebNode::toConstElement):

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@53678 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2010-01-21 Kwang Yul Seo <skyul@company100.net>
eric@webkit.org [Fri, 22 Jan 2010 06:05:46 +0000 (06:05 +0000)]
2010-01-21  Kwang Yul Seo  <skyul@company100.net>

        Reviewed by Maciej Stachowiak.

        Add fastStrDup to FastMalloc
        https://bugs.webkit.org/show_bug.cgi?id=33937

        The new string returned by fastStrDup is obtained with fastMalloc,
        and can be freed with fastFree. This makes the memory management
        more consistent because we don't need to keep strdup allocated pointers
        and free them with free(). Instead we can use fastFree everywhere.

        * wtf/FastMalloc.cpp:
        (WTF::fastStrDup):
        * wtf/FastMalloc.h:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@53677 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2010-01-21 Fumitoshi Ukai <ukai@chromium.org>
ukai@chromium.org [Fri, 22 Jan 2010 06:01:45 +0000 (06:01 +0000)]
2010-01-21  Fumitoshi Ukai  <ukai@chromium.org>

        Reviewed by Shinichiro Hamaji.

        Move fast/websockets into websocket/tests
        https://bugs.webkit.org/show_bug.cgi?id=32258

        * fast/websockets: Removed.
        * websocket/tests/script-tests/websocket-event-target.js: Copied from LayoutTests/fast/websockets/script-tests/websocket-event-target.js.
        * websocket/tests/websocket-event-target-expected.txt: Copied from LayoutTests/fast/websockets/websocket-event-target-expected.txt.
        * websocket/tests/websocket-event-target.html: Copied from LayoutTests/fast/websockets/websocket-event-target.html.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@53676 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2010-01-21 Chris Jerdonek <cjerdonek@webkit.org>
eric@webkit.org [Fri, 22 Jan 2010 05:39:24 +0000 (05:39 +0000)]
2010-01-21  Chris Jerdonek  <cjerdonek@webkit.org>

        Reviewed by Shinichiro Hamaji.

        Refactored to move file name and file-reading related code
        from cpp_style.py and text_style.py to checker.py.

        https://bugs.webkit.org/show_bug.cgi?id=33775

        * Scripts/check-webkit-style:
          - Updates caused by changes to checker.py.

        * Scripts/webkitpy/style/checker.py:
          - Added SKIPPED_FILES_WITH_WARNING list.
          - Added SKIPPED_FILES_WITHOUT_WARNING list.
          - Added FileType class.
          - Added ProcessorDispatcher class.
          - In StyleChecker class:
            - Renamed process_patch() to check_patch().
            - Renamed process_file() to check_file().
            - Added _process_file().
            - Related refactoring.
            - Addressed check_patch() FIXME to share code with process_file().

        * Scripts/webkitpy/style/checker_unittest.py:
          - Added ProcessorDispatcherSkipTest class.
          - Added ProcessorDispatcherDispatchTest class.
          - Added StyleCheckerCheckFileTest class.

        * Scripts/webkitpy/style/cpp_style.py:
          - Renamed process_file_data() to _process_lines.
          - Removed process_file() (moved logic to checker.py).
          - Removed can_handle() (moved logic to checker.py).
          - Added CppProcessor class.
          - Removed is_exempt() (moved logic to checker.py).
          - Added process_file_data() back as a wrapper function.

        * Scripts/webkitpy/style/cpp_style_unittest.py:
          - Removed test_can_handle().
          - Removed test_is_exempt().
          - Added CppProcessorTest class.

        * Scripts/webkitpy/style/text_style.py:
          - Added TextProcessor class.
          - Removed process_file().
          - Removed can_handle().

        * Scripts/webkitpy/style/text_style_unittest.py:
          - Removed test_can_handle().
          - Added TextProcessorTest class.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@53675 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2010-01-21 Joe Mason <jmason@rim.com>
eric@webkit.org [Fri, 22 Jan 2010 05:23:12 +0000 (05:23 +0000)]
2010-01-21  Joe Mason  <jmason@rim.com>

        Reviewed by Darin Adler.

        Promote pow argument to double to resolve ambiguous overload (compile fix for RVCT 4.0)
        https://bugs.webkit.org/show_bug.cgi?id=33952

        * platform/graphics/filters/FEComponentTransfer.cpp:
        (WebCore::gamma):

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@53674 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2010-01-21 Andrei Popescu <andreip@google.com>
eric@webkit.org [Fri, 22 Jan 2010 05:06:41 +0000 (05:06 +0000)]
2010-01-21  Andrei Popescu  <andreip@google.com>

        Reviewed by David Levin.

        [Android] bindings/v8/ScriptController.cpp needs to include
        PlatformBridge.h instead of ChromiumBridge.h so that it can
        be used on both Chromium and Android.
        https://bugs.webkit.org/show_bug.cgi?id=33673

        Add "static NPObject* pluginScriptableObject(Widget*);" method
        to PlatformBridge.h
        Include PlatformBridge.h from ScriptController.cpp.

        No new tests, just platform code.

        * bindings/v8/ScriptController.cpp:
        (WebCore::ScriptController::createScriptInstanceForWidget):
        * platform/android/PlatformBridge.h:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@53673 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years agohistory.back() for same-document history traversals isn't synchronous as the specific...
beidson@apple.com [Fri, 22 Jan 2010 04:22:02 +0000 (04:22 +0000)]
history.back() for same-document history traversals isn't synchronous as the specification states.
<rdar://problem/7535011> and https://bugs.webkit.org/show_bug.cgi?id=33538

Reviewed by Maciej Stachowiak.

JavaScriptCore:

* wtf/Platform.h: Add a "HISTORY_ALWAYS_ASYNC" enable and turn it on for Chromium.

WebCore:

In resolving https://bugs.webkit.org/show_bug.cgi?id=25570, all history.back()/forward()/go() navigations
were made asynchronous. That doesn't agree with the HTML5 spec and might have been overreaching for fixing
that particular bug. In working with the new history state APIs I noticed some things that should've been
possible were not possible because of this change.

The change in loading behavior is well covered via modifications to previous tests.

* loader/RedirectScheduler.cpp:
(WebCore::RedirectScheduler::scheduleHistoryNavigation): Determine beforehand if the traversal is
  a same-document navigation. If it is, perform the load directly instead of scheduling it.

LayoutTests:

The change in loading behavior is well covered via modifications to these previous tests:

* fast/loader/stateobjects/document-destroyed-navigate-back-with-fragment-scroll.html:
* fast/loader/stateobjects/pushstate-with-fragment-urls-and-hashchange-expected.txt:
* fast/loader/stateobjects/pushstate-with-fragment-urls-and-hashchange.html:
* fast/loader/stateobjects/state-api-on-detached-frame-crash-expected.txt:
* fast/loader/stateobjects/state-api-on-detached-frame-crash.html:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@53672 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2010-01-21 Michael Nordman <michaeln@google.com>
eric@webkit.org [Fri, 22 Jan 2010 04:21:21 +0000 (04:21 +0000)]
2010-01-21  Michael Nordman  <michaeln@google.com>

        Reviewed by Darin Fisher.

        Adds WebKit::WebDataSource::applicationCacheHost() to the API.
        https://bugs.webkit.org/show_bug.cgi?id=33880

        * WebKit.gyp:
        * public/WebDataSource.h:
        * src/ApplicationCacheHost.cpp:
        * src/ApplicationCacheHostInternal.h: Copied from WebKit/chromium/src/ApplicationCacheHost.cpp.
        (WebCore::ApplicationCacheHostInternal::notifyEventListener):
        (WebCore::ApplicationCacheHostInternal::toWebApplicationCacheHost):
        * src/WebDataSourceImpl.cpp:
        (WebKit::WebDataSourceImpl::applicationCacheHost):
        * src/WebDataSourceImpl.h:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@53671 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2010-01-21 Jakub Wieczorek <faw217@gmail.com>
eric@webkit.org [Fri, 22 Jan 2010 04:04:51 +0000 (04:04 +0000)]
2010-01-21  Jakub Wieczorek  <faw217@gmail.com>

        Reviewed by Tor Arne Vestbø.

        [Qt] Don't allow Phonon's invisible video widget to keep the app running.
        https://bugs.webkit.org/show_bug.cgi?id=33842

        The change made in r38223 should not be limited to Qt < 4.5 only.

        * platform/graphics/qt/MediaPlayerPrivatePhonon.cpp:
        (WebCore::MediaPlayerPrivate::MediaPlayerPrivate):

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@53670 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2010-01-21 Kent Tamura <tkent@chromium.org>
tkent@chromium.org [Fri, 22 Jan 2010 03:29:30 +0000 (03:29 +0000)]
2010-01-21  Kent Tamura  <tkent@chromium.org>

        Reviewed by Darin Adler.

        HTMLInputElement::valueAsDate setter support for type=datetime.
        https://bugs.webkit.org/show_bug.cgi?id=33939

        Add setter tests to input-valueasdate-datetime.js, and update the
        expectation.

        Note: the expectation file contains some FAIL lines. They are
        intentional because they test a unimplemented feature.

        * fast/forms/input-valueasdate-datetime-expected.txt:
        * fast/forms/script-tests/input-valueasdate-datetime.js:

2010-01-21  Kent Tamura  <tkent@chromium.org>

        Reviewed by Darin Adler.

        HTMLInputElement::valueAsDate setter support for type=datetime.
        https://bugs.webkit.org/show_bug.cgi?id=33939

        Introduce ISODateTime::setMillisecondsSinceEpochForDateTime() and add
        DateTime type support to ISODateTime::toString().

        * html/HTMLInputElement.cpp:
        (WebCore::HTMLInputElement::setValueAsDate):
        * html/ISODateTime.cpp:
        (WebCore::ISODateTime::setMillisecondsSinceEpochForDate):
         Set Invalid to m_type.
        (WebCore::ISODateTime::setMillisecondsSinceEpochForDateTime):
        (WebCore::ISODateTime::setMillisecondsSinceEpochForMonth):
         Set Invalid to m_type.
        (WebCore::ISODateTime::setMillisecondsSinceMidnight):
         Set Invalid to m_type.
        (WebCore::ISODateTime::toString):
         Support DateTime type. This always produces UTC representation.
        * html/ISODateTime.h:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@53669 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2010-01-21 Geoffrey Garen <ggaren@apple.com>
ggaren@apple.com [Fri, 22 Jan 2010 03:25:16 +0000 (03:25 +0000)]
2010-01-21  Geoffrey Garen  <ggaren@apple.com>

        Reviewed by Sam Weinig.

        Fixed <rdar://problem/7562574> ASSERT in WebCore::removeWrapper() at the
        end of run-webkit-tests

        This was an ASSERT-only bug, introduced by isolated worlds, which
        created the novelty of a wrapper that might outlive its wrapper cache.

        When a wrapper outlived its wrapper cache, both the wrapper's destructor
        and the wrapper cache's destructor would claim to have uncached the wrapper,
        causing an ASSERT to fire.

        The solution is to distinguish between operations that logically add and
        remove cache entries, and operations that delete whole caches. We track
        when a cache entry is logically added, and when it's logically removed,
        independent of whether the actual cache still exists.

        * bindings/js/JSDOMBinding.cpp:
        (WebCore::willCacheWrapper):
        (WebCore::didUncacheWrapper): New names for these functions to help
        explain what they track.

        (WebCore::DOMWrapperWorld::~DOMWrapperWorld): Don't claim to uncache
        all the wrappers in the world; we're deleting the cache, not managing its
        entries.

        (WebCore::cacheDOMObjectWrapper):
        (WebCore::forgetDOMObject):
        (WebCore::forgetDOMNode):
        (WebCore::cacheDOMNodeWrapper):
        (WebCore::forgetAllDOMNodesForDocument):
        (WebCore::forgetWorldOfDOMNodesForDocument):
        (WebCore::takeWrappers):
        (WebCore::updateDOMNodeDocument): Updated for renames.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@53668 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years agoCreate a unit-tested subroutine to parse patch files created
cjerdonek@webkit.org [Fri, 22 Jan 2010 02:35:12 +0000 (02:35 +0000)]
Create a unit-tested subroutine to parse patch files created
by svn-create-patch.

Reviewed by David Kilzer.

https://bugs.webkit.org/show_bug.cgi?id=33475

* Scripts/VCSUtils.pm:
  - Added parseDiff().
  - Added parsePatch().

* Scripts/webkitperl/VCSUtils_unittest/parseDiff.pl: Added.
  - Added unit tests for parseDiff().

* Scripts/webkitperl/VCSUtils_unittest/parseDiffHeader.pl:
  - Total number of tests now computed dynamically.

* Scripts/webkitperl/VCSUtils_unittest/parsePatch.pl: Added.
  - Added unit tests for parsePatch().

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@53667 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2010-01-21 Nikolas Zimmermann <nzimmermann@rim.com>
zimmermann@webkit.org [Fri, 22 Jan 2010 02:23:34 +0000 (02:23 +0000)]
2010-01-21  Nikolas Zimmermann  <nzimmermann@rim.com>

        Reviewed by Sam Weinig.

        SVG JS bindings "context" pointer needs to move onto binding impls
        https://bugs.webkit.org/show_bug.cgi?id=27243

        Rewrite SVG DOM JSC bindings to use a global DOMObject <-> SVGElement context map, similar to V8's approach.
        This allows us to remove DOMObjectWithSVGContext and make all SVG JS objects use DOMObjectWithGlobalPointer.
        We're fitting again in JS_CELL_SIZE, and there's no need to special case SVG anymore.

        Not adding a new test, as we have yet to identify how to test that we're passing around correct global objects.
        For now this is just a code cleanup which lets me continue making progress in that area of the code.

        * GNUmakefile.am: Add new DOMObjectWithSVGContext.h to build.
        * WebCore.gypi: Ditto.
        * WebCore.vcproj/WebCore.vcproj: Ditto.
        * WebCore.xcodeproj/project.pbxproj: Ditto.
        * bindings/js/DOMObjectWithSVGContext.h: Removed.
        * bindings/js/JSDOMBinding.cpp:
        (WebCore::setDOMException): Add comment, that passing 0 context is intentional.
        * bindings/js/JSDOMBinding.h:
        (WebCore::createDOMObjectWrapper): Add to JSSVGContextCache, when creating wrappers.
        (WebCore::getDOMObjectWrapper): Add assertion guarding the JSSVGContextCache is in-sync with the passed context.
        * bindings/js/JSSVGContextCache.h: Added. Maps DOMObjects to SVGElements - just like V8 approachs to this problem.
        (WebCore::JSSVGContextCache::wrapperMap):
        (WebCore::JSSVGContextCache::addWrapper):
        (WebCore::JSSVGContextCache::forgetWrapper):
        (WebCore::JSSVGContextCache::propagateSVGDOMChange):
        (WebCore::JSSVGContextCache::svgContextForDOMObject):
        * bindings/js/JSSVGLengthCustom.cpp: Retrieve context from cache, instead of taking it from the JS* class.
        (WebCore::JSSVGLength::value):
        (WebCore::JSSVGLength::convertToSpecifiedUnits):
        * bindings/js/JSSVGMatrixCustom.cpp Ditto.:
        (WebCore::JSSVGMatrix::multiply):
        (WebCore::JSSVGMatrix::inverse):
        (WebCore::JSSVGMatrix::rotateFromVector):
        * bindings/js/JSSVGPODListCustom.h: Ditto.
        (WebCore::JSSVGPODListCustom::finishGetter):
        (WebCore::JSSVGPODListCustom::finishSetter):
        (WebCore::JSSVGPODListCustom::finishSetterReadOnlyResult):
        (WebCore::JSSVGPODListCustom::clear):
        (WebCore::JSSVGPODListCustom::initialize):
        * bindings/js/JSSVGPODTypeWrapper.h: Ditto.
        (WebCore::JSSVGDynamicPODTypeWrapper::commitChange):
        (WebCore::JSSVGStaticPODTypeWrapper::commitChange):
        (WebCore::JSSVGStaticPODTypeWrapperWithPODTypeParent::commitChange):
        (WebCore::JSSVGStaticPODTypeWrapperWithParent::commitChange):
        (WebCore::JSSVGPODTypeWrapperCreatorForList::create):
        (WebCore::JSSVGPODTypeWrapperCreatorForList::commitChange):
        * bindings/js/JSSVGPathSegCustom.cpp: Ditto.
        (WebCore::toJS):
        * bindings/js/JSSVGPathSegListCustom.cpp: Ditto.
        (WebCore::JSSVGPathSegList::clear):
        (WebCore::JSSVGPathSegList::initialize):
        (WebCore::JSSVGPathSegList::getItem):
        (WebCore::JSSVGPathSegList::insertItemBefore):
        (WebCore::JSSVGPathSegList::replaceItem):
        (WebCore::JSSVGPathSegList::removeItem):
        (WebCore::JSSVGPathSegList::appendItem):
        * bindings/scripts/CodeGeneratorJS.pm:
        * svg/SVGAngle.h: Remove unneeded associatedAttributeName() method.
        * svg/SVGPreserveAspectRatio.h: Ditto.
        * svg/SVGTransform.h: Ditto.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@53666 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years agowx build fixes, add a missing header and add a new dir to the build system.
kevino@webkit.org [Fri, 22 Jan 2010 02:07:34 +0000 (02:07 +0000)]
wx build fixes, add a missing header and add a new dir to the build system.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@53665 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2010-01-21 Shinichiro Hamaji <hamaji@chromium.org>
hamaji@chromium.org [Fri, 22 Jan 2010 01:40:58 +0000 (01:40 +0000)]
2010-01-21  Shinichiro Hamaji  <hamaji@chromium.org>

        Reviewed by Darin Adler.

        3 tests for CSS are misplaced
        https://bugs.webkit.org/show_bug.cgi?id=33942

        * css2.1/atrule_longest_match-expected.txt: Removed.
        * css2.1/atrule_longest_match.html: Removed.
        * css2.1/bogus-color-span.html: Removed.
        * css2.1/css1_forward_compatible_parsing.html: Removed.
        * fast/css/atrule_longest_match-expected.txt: Copied from LayoutTests/css2.1/atrule_longest_match-expected.txt.
        * fast/css/atrule_longest_match.html: Copied from LayoutTests/css2.1/atrule_longest_match.html.
        * fast/css/bogus-color-span.html: Copied from LayoutTests/css2.1/bogus-color-span.html.
        * fast/css/css1_forward_compatible_parsing.html: Copied from LayoutTests/css2.1/css1_forward_compatible_parsing.html.
        * platform/mac/css2.1/bogus-color-span-expected.checksum: Removed.
        * platform/mac/css2.1/bogus-color-span-expected.png: Removed.
        * platform/mac/css2.1/bogus-color-span-expected.txt: Removed.
        * platform/mac/css2.1/css1_forward_compatible_parsing-expected.checksum: Removed.
        * platform/mac/css2.1/css1_forward_compatible_parsing-expected.png: Removed.
        * platform/mac/css2.1/css1_forward_compatible_parsing-expected.txt: Removed.
        * platform/mac/fast/css/bogus-color-span-expected.checksum: Copied from LayoutTests/platform/mac/css2.1/bogus-color-span-expected.checksum.
        * platform/mac/fast/css/bogus-color-span-expected.png: Copied from LayoutTests/platform/mac/css2.1/bogus-color-span-expected.png.
        * platform/mac/fast/css/bogus-color-span-expected.txt: Copied from LayoutTests/platform/mac/css2.1/bogus-color-span-expected.txt.
        * platform/mac/fast/css/css1_forward_compatible_parsing-expected.checksum: Copied from LayoutTests/platform/mac/css2.1/css1_forward_compatible_parsing-expected.checksum.
        * platform/mac/fast/css/css1_forward_compatible_parsing-expected.png: Copied from LayoutTests/platform/mac/css2.1/css1_forward_compatible_parsing-expected.png.
        * platform/mac/fast/css/css1_forward_compatible_parsing-expected.txt: Copied from LayoutTests/platform/mac/css2.1/css1_forward_compatible_parsing-expected.txt.
        * platform/qt/css2.1/bogus-color-span-expected.txt: Removed.
        * platform/qt/css2.1/css1_forward_compatible_parsing-expected.txt: Removed.
        * platform/qt/fast/css/bogus-color-span-expected.txt: Copied from LayoutTests/platform/qt/css2.1/bogus-color-span-expected.txt.
        * platform/qt/fast/css/css1_forward_compatible_parsing-expected.txt: Copied from LayoutTests/platform/qt/css2.1/css1_forward_compatible_parsing-expected.txt.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@53664 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2010-01-21 Simon Fraser <simon.fraser@apple.com>
simon.fraser@apple.com [Fri, 22 Jan 2010 01:28:07 +0000 (01:28 +0000)]
2010-01-21  Simon Fraser  <simon.fraser@apple.com>

        Reviewed by Dan Bernstein.

        -webkit-mask-box-image draws a box while loading
        https://bugs.webkit.org/show_bug.cgi?id=33979
        <rdar://problem/7378662>

        Don't render masked elements until the mask images are fully loaded.

        Test: http/tests/misc/slow-loading-mask.html

        * rendering/RenderBox.cpp:
        (WebCore::RenderBox::paintMaskImages): Check that all mask images (both the single mask-box-image,
        and the possibly multiple mask-image) are fully loaded before allowing the masked content to render.
        If they are not fully loaded, use a fully transparent transparency layer.

        * rendering/style/FillLayer.h:
        * rendering/style/FillLayer.cpp:
        (WebCore::FillLayer::imagesAreLoaded): New utility function that returns true if the image
        in every FillLayer is loaded.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@53663 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years agoRubber-stamped by Eric Seidel.
bdakin@apple.com [Fri, 22 Jan 2010 01:20:55 +0000 (01:20 +0000)]
Rubber-stamped by Eric Seidel.

Skipping this until issues mentioned in https://bugs.webkit.org/show_bug.cgi?id=33923 are resolved.

* platform/mac/Skipped:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@53662 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2010-01-21 Adam Barth <abarth@webkit.org>
abarth@webkit.org [Fri, 22 Jan 2010 01:07:46 +0000 (01:07 +0000)]
2010-01-21  Adam Barth  <abarth@webkit.org>

        Reviewed by Eric Seidel.

        Make the EWS transactional
        https://bugs.webkit.org/show_bug.cgi?id=33978

        Now if the EWS gets interrupted in the middle of processing a patch,
        the bots will re-process the patch.

        * Scripts/test-webkitpy:
        * Scripts/webkitpy/commands/queues.py:
        * Scripts/webkitpy/commands/queues_unittest.py:
        * Scripts/webkitpy/patchcollection.py:
        * Scripts/webkitpy/patchcollection_unittest.py: Added.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@53661 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years agoREGRESSION (r52795): New message created in response to mailto: URL has random charac...
darin@apple.com [Fri, 22 Jan 2010 00:48:45 +0000 (00:48 +0000)]
REGRESSION (r52795): New message created in response to mailto: URL has random character in message body
rdar://problem/7565902

Reviewed by Mark Rowe.

This regression affects the Mac OS X Mail application.
I was not able to find a simple way to create a regression test.

* platform/cf/SharedBufferCF.cpp:
(WebCore::SharedBuffer::SharedBuffer): Initialize m_size to 0 as in the other constructors.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@53660 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years agoScript tags are copied and pasted, making cross-domain attacks possible.
enrica@apple.com [Fri, 22 Jan 2010 00:40:23 +0000 (00:40 +0000)]
Script tags are copied and pasted, making cross-domain attacks possible.
https://bugs.webkit.org/show_bug.cgi?id=33970

Reviewed by Darin Adler.

WebCore:

Tests: editing/pasteboard/paste-noscript-svg.html
       editing/pasteboard/paste-visible-script.html

We remove the content and the attributes of every script tag before
pasting into the destination.

* dom/Element.cpp:
(WebCore::Element::setAttributeMap): Now we are removing xlink:href
if it contains javascript protocol.
* html/HTMLParser.cpp:
(WebCore::HTMLParser::parseToken): We strip all the script tag attributes
we are parsing to create a fragment to paste.
* html/HTMLTokenizer.cpp:
(WebCore::HTMLTokenizer::HTMLTokenizer):
(WebCore::HTMLTokenizer::scriptHandler): Removes the script text after it was parsed.
* html/HTMLTokenizer.h:

LayoutTests:

* editing/pasteboard/paste-noscript-svg-expected.txt: Added.
* editing/pasteboard/paste-noscript-svg.html: Added.
* editing/pasteboard/paste-visible-script-expected.txt: Added.
* editing/pasteboard/paste-visible-script.html: Added.
* editing/resources/svgcontent.xhtml: Added.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@53659 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2010-01-21 Adam Barth <abarth@webkit.org>
abarth@webkit.org [Fri, 22 Jan 2010 00:34:49 +0000 (00:34 +0000)]
2010-01-21  Adam Barth  <abarth@webkit.org>

        Unreviewed.  Add missing "ago" for style in the status bubble.

        * QueueStatusServer/templates/statusbubble.html:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@53658 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2010-01-21 Geoffrey Garen <ggaren@apple.com>
ggaren@apple.com [Fri, 22 Jan 2010 00:06:00 +0000 (00:06 +0000)]
2010-01-21  Geoffrey Garen  <ggaren@apple.com>

        Reviewed by Oliver Hunt.

        Always create a prototype for automatically managed classes.

        This fixes some errors where prototype chains were not correctly hooked
        up, and also ensures that API classes work correctly with features like
        instanceof.

        * API/JSClassRef.cpp:
        (OpaqueJSClass::create): Cleaned up some of this code. Also changed it
        to always create a prototype class.

        * API/tests/testapi.c:
        (Derived2_class):
        (main): Fixed a null value crash in the exception checking code.
        * API/tests/testapi.js: Added some tests for the case where a prototype
        chain would not be hooked up correctly.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@53657 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2010-01-21 Nikolas Zimmermann <nzimmermann@rim.com>
zimmermann@webkit.org [Thu, 21 Jan 2010 23:22:56 +0000 (23:22 +0000)]
2010-01-21  Nikolas Zimmermann  <nzimmermann@rim.com>

        Reviewed by Dirk Schulze.

        <use> with % lengths does not update on window resize
        https://bugs.webkit.org/show_bug.cgi?id=14639

        appendChild broken for symbol-use
        https://bugs.webkit.org/show_bug.cgi?id=24802

        Test: svg/custom/relative-sized-shadow-tree-content-with-symbol.xhtml
              svg/custom/svg/custom/relative-sized-use-on-symbol.xhtml

        Fix reaction on window size changes for relative sized elements within <symbol> elements.
        Don't evaluate SVGUseElement width/height attributes in order to propagate them to the shadow tree,
        instead just propagate the attribute values itself so "100%" remains "100%" instead of being evaluated
        to an absolute pixel value in the parent <svg> user-space.

        Needs another fix in SVGStyledElement::childrenChanged(), which is a regression from my last <use>
        patch, that hasn't been detected, as the use-dynamic-append.svg testcase was broken.

        The combination of all the last <use> patches finally fixes the peepo.co.uk website, which was broken by several bugs for years.

        * svg/SVGStyledElement.cpp:
        (WebCore::SVGStyledElement::childrenChanged): Need to mark shadow trees needing recalcs, even when parsing.
        * svg/SVGUseElement.cpp:
        (WebCore::SVGUseElement::svgAttributeChanged): Don't reclone for width/height attribute changes, just propagate them down the render tree
        (WebCore::updateContainerSize): Added helper function, updating the width/height attributes in the shadow tree.
        (WebCore::SVGUseElement::updateContainerSizes):
        (WebCore::SVGUseElement::buildShadowAndInstanceTree): Update container width/height after shadow tree creation, instead of doing it inbetween.
        (WebCore::SVGUseElement::buildShadowTree): Don't alter width/height attributes here anymore.
        (WebCore::SVGUseElement::expandUseElementsInShadowTree): Ditto.
        (WebCore::SVGUseElement::expandSymbolElementsInShadowTree): Ditto.
        * svg/SVGUseElement.h:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@53656 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago Reviewed by Darin Adler.
ap@apple.com [Thu, 21 Jan 2010 23:19:57 +0000 (23:19 +0000)]
    Reviewed by Darin Adler.

        https://bugs.webkit.org/show_bug.cgi?id=33972
        Assertion failure in FrameLoader::checkLoadComplete()

        I couldn't make a test for this, after trying rather hard. I'm not too worried, because such
        a test has to depend on details of current implementation so intimately that it would become
        ineffective very quickly anyway.

        * loader/FrameLoader.cpp: (WebCore::FrameLoader::stopAllLoaders): Stop the check timer, we
        don't need it after aborting load.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@53655 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years agoRubber-stamped by Adam Roben.
bweinstein@apple.com [Thu, 21 Jan 2010 23:06:01 +0000 (23:06 +0000)]
Rubber-stamped by Adam Roben.

Added platform specific (incorrect) results for plugins/mouse-events* for Windows,
and removed mouse-events from the Windows Skipped list. The landing of correct results
for these tests is being tracked by <http://webkit.org/b/33973>.

* platform/win/Skipped: Removed plugins/mouse-events from the Skipped List.
* platform/win/plugins: Added.
* platform/win/plugins/mouse-events-expected.txt: Added.
* platform/win/plugins/mouse-events-fixedpos-expected.txt: Added.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@53654 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2010-01-21 Simon Fraser <simon.fraser@apple.com>
simon.fraser@apple.com [Thu, 21 Jan 2010 23:04:34 +0000 (23:04 +0000)]
2010-01-21  Simon Fraser  <simon.fraser@apple.com>

        No review

        Skip plugins/mouse-events-fixedpos.html on gtk, since it requires a special test plugin.

        * platform/gtk/Skipped:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@53653 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years agoBuildfix for r53646.
ossy@webkit.org [Thu, 21 Jan 2010 22:53:23 +0000 (22:53 +0000)]
Buildfix for r53646.

Reviewed by Dave Hyatt.

* bindings/js/SerializedScriptValue.cpp: path of JSLock.h fixed.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@53652 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years agoUnreviewed buildfix after r53644.
ossy@webkit.org [Thu, 21 Jan 2010 22:31:15 +0000 (22:31 +0000)]
Unreviewed buildfix after r53644.

* platform/Scrollbar.cpp:
(WebCore::Scrollbar::scroll): missing HAVE(ACCESSIBILITY) guard added.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@53651 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2010-01-21 Darin Fisher <darin@chromium.org>
darin@chromium.org [Thu, 21 Jan 2010 22:24:55 +0000 (22:24 +0000)]
2010-01-21  Darin Fisher  <darin@chromium.org>

        Reviewed by David Levin.

        Unnecessary call to HistoryItem::targetItem in HistoryController::pushState
        https://bugs.webkit.org/show_bug.cgi?id=33969

        * loader/HistoryController.cpp:
        (WebCore::HistoryController::pushState): Remove unnecessary code.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@53650 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years agoNot reviewed, Chromium build fix. Add AccessibilityScrollbar.h and .cpp to WebCore...
dimich@chromium.org [Thu, 21 Jan 2010 22:17:47 +0000 (22:17 +0000)]
Not reviewed, Chromium build fix. Add AccessibilityScrollbar.h and .cpp to WebCore.gypi

* WebCore.gypi:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@53649 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago[chromium] Unreviewed build fix.
ajwong@chromium.org [Thu, 21 Jan 2010 22:10:44 +0000 (22:10 +0000)]
[chromium] Unreviewed build fix.

Fix possible use of initialized variable warning in release mode.

* bindings/v8/SerializedScriptValue.cpp:
(WebCore::ZigZag::Deserializer::doDeserialize):

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@53648 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago[Qt] Unreviewed buildfix after r53644.
ossy@webkit.org [Thu, 21 Jan 2010 22:02:15 +0000 (22:02 +0000)]
[Qt] Unreviewed buildfix after r53644.

* WebCore.pro: accessibility/AccessibilityScrollbar.cpp and accessibility/AccessibilityScrollbar.h added.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@53647 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years agoMake sure to do a JSLock before deserialization of script values, since they can...
hyatt@apple.com [Thu, 21 Jan 2010 21:57:18 +0000 (21:57 +0000)]
Make sure to do a JSLock before deserialization of script values, since they can heap allocate.

Reviewed by Oliver Hunt.

* bindings/js/SerializedScriptValue.cpp:
(WebCore::SerializedScriptValueData::deserialize):

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@53646 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years agoBuild fix, add new file to AccessibilityAllInOne
bweinstein@apple.com [Thu, 21 Jan 2010 21:54:46 +0000 (21:54 +0000)]
Build fix, add new file to AccessibilityAllInOne

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@53645 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years agoWebCore: Fix for https://bugs.webkit.org/show_bug.cgi?id=33923 REGRESSION
bdakin@apple.com [Thu, 21 Jan 2010 21:31:39 +0000 (21:31 +0000)]
WebCore: Fix for https://bugs.webkit.org/show_bug.cgi?id=33923 REGRESSION
(Safari 4): AXValueChanged no longer sent for text area scrollbars
-and corresponding-
<rdar://problem/6942686>

Reviewed by Oliver Hunt.

New class AccessibilityScrollbar inherits directly from
AccessibilityObject.
* accessibility/AccessibilityScrollbar.cpp: Added.
(WebCore::AccessibilityScrollbar::AccessibilityScrollbar):
(WebCore::AccessibilityScrollbar::create):
(WebCore::AccessibilityScrollbar::valueForRange):
* accessibility/AccessibilityScrollbar.h: Added.
(WebCore::AccessibilityScrollbar::setScrollbar):
(WebCore::AccessibilityScrollbar::roleValue):
(WebCore::AccessibilityScrollbar::accessibilityIsIgnored):
(WebCore::AccessibilityScrollbar::size):
(WebCore::AccessibilityScrollbar::elementRect):
(WebCore::AccessibilityScrollbar::parentObject):

AXObjectCache::getOrCreate() now has a case for ScrollBarRole,
which will create a new AccessibilityScrollbar. I also added a new
version of postNotification() that does not require a renderer. The
old postNotification() calls the new one.
* accessibility/AXObjectCache.cpp:
(WebCore::AXObjectCache::getOrCreate):
(WebCore::AXObjectCache::postNotification):
* accessibility/AXObjectCache.h:

When AX is enabled, getOrCreate an AccessibilityScrollbar and send
a notification.
* platform/Scrollbar.cpp:
(WebCore::Scrollbar::scroll):

Fix project files.
* GNUmakefile.am:
* WebCore.vcproj/WebCore.vcproj:
* WebCore.xcodeproj/project.pbxproj:
* WebCoreSources.bkl:

LayoutTests: Test for https://bugs.webkit.org/show_bug.cgi?id=33923 REGRESSION
(Safari 4): AXValueChanged no longer sent for text area scrollbars
-and corresponding-
<rdar://problem/6942686>

Reviewed by Oliver Hunt.

* platform/mac/accessibility/change-notification-on-scroll-expected.txt: Added.
* platform/mac/accessibility/change-notification-on-scroll.html: Added.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@53644 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2010-01-21 Simon Fraser <simon.fraser@apple.com>
simon.fraser@apple.com [Thu, 21 Jan 2010 21:08:24 +0000 (21:08 +0000)]
2010-01-21  Simon Fraser  <simon.fraser@apple.com>

        No review

        Skip plugins/mouse-events-fixedpos.html on Qt, since it requires a special test plugin.

        * platform/qt/Skipped:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@53643 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years agoForce JSC to create a prototype chain for API classes with a
oliver@apple.com [Thu, 21 Jan 2010 21:04:34 +0000 (21:04 +0000)]
Force JSC to create a prototype chain for API classes with a
parent class but no static functions.

Reviewed by Geoff Garen.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@53642 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years agoAdd changes missing from r53595, without which Chromium has no database.
dimich@chromium.org [Thu, 21 Jan 2010 21:02:15 +0000 (21:02 +0000)]
Add changes missing from r53595, without which Chromium has no database.

Patch by Eric Uhrhane <ericu@chromium.org> on 2010-01-21
Reviewed by Dmitry Titov.

https://bugs.webkit.org/show_bug.cgi?id=33966

No new tests - fixes database layout tests in Chromium.

WebCore:

* bindings/v8/RuntimeEnabledFeatures.cpp:
* bindings/v8/RuntimeEnabledFeatures.h:

WebKit/chromium:

* src/WebRuntimeFeatures.cpp:
(WebKit::WebRuntimeFeatures::enableDatabase):
(WebKit::WebRuntimeFeatures::isDatabaseEnabled):

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@53641 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2010-01-21 Kenneth Russell <kbr@google.com>
eric@webkit.org [Thu, 21 Jan 2010 21:01:38 +0000 (21:01 +0000)]
2010-01-21  Kenneth Russell  <kbr@google.com>

        Reviewed by Oliver Hunt.

        [Chromium] Implement texSubImage2D taking WebGLArray
        https://bugs.webkit.org/show_bug.cgi?id=33932

        * fast/canvas/webgl/tex-sub-image-2d-expected.txt: Added.
        * fast/canvas/webgl/tex-sub-image-2d.html: Added.
2010-01-21  Kenneth Russell  <kbr@google.com>

        Reviewed by Oliver Hunt.

        [Chromium] Implement texSubImage2D taking WebGLArray
        https://bugs.webkit.org/show_bug.cgi?id=33932

        * src/GraphicsContext3D.cpp:
        (WebCore::GraphicsContext3D::texSubImage2D):

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@53640 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2010-01-21 Jakob Petsovits <jpetsovits@rim.com>
jpetsovits@rim.com [Thu, 21 Jan 2010 20:56:25 +0000 (20:56 +0000)]
2010-01-21  Jakob Petsovits  <jpetsovits@rim.com>

        Reviewed by Adam Treat.

        Don't call vgGetError()/eglGetError() repeatedly
        https://bugs.webkit.org/show_bug.cgi?id=33959

        For vgGetError(), I missed the part of the spec where
        it says that calling that function clears the error
        and subsequent calls will return VG_NO_ERROR again.

        For eglGetError(), the specification doesn't mention
        that kind of behavior, and interpretations seem to
        differ between EGL implementations (even within
        Khronos: the OpenVG reference implementation doesn't
        reset the error code - and even mentions the difference
        to vgGetError() in a comment - whereas the online
        OpenGL ES API document explicitly specifies clearing
        the error code).

        It thus makes sense not to call either of the two
        error functions more than once for checking a single
        EGL/OpenVG call. This patch adapts assertions to
        accommodate for this behavior, and also needs to
        change surface creation methods as they previously
        relied on multiple calls of eglGetError().

        * platform/graphics/openvg/EGLDisplayOpenVG.cpp:
        (WebCore::EGLDisplayOpenVG::sharedPlatformSurface):
        (WebCore::EGLDisplayOpenVG::createPbufferSurface):
        * platform/graphics/openvg/EGLDisplayOpenVG.h:
        * platform/graphics/openvg/EGLUtils.h:
        * platform/graphics/openvg/SurfaceOpenVG.cpp:
        (WebCore::SurfaceOpenVG::SurfaceOpenVG):
        * platform/graphics/openvg/SurfaceOpenVG.h:
        * platform/graphics/openvg/VGUtils.h:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@53639 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2010-01-21 Kent Hansen <kent.hansen@nokia.com>
eric@webkit.org [Thu, 21 Jan 2010 20:46:03 +0000 (20:46 +0000)]
2010-01-21  Kent Hansen  <kent.hansen@nokia.com>

        Reviewed by Geoffrey Garen.

        Object.getOwnPropertyDescriptor always returns undefined for JS API objects
        https://bugs.webkit.org/show_bug.cgi?id=33946

        Ideally the getOwnPropertyDescriptor() reimplementation should return an
        access descriptor that wraps the property getter and setter callbacks, but
        that approach is much more involved than returning a value descriptor.
        Keep it simple for now.

        * API/JSCallbackObject.h:
        * API/JSCallbackObjectFunctions.h:
        (JSC::::getOwnPropertyDescriptor):
        * API/tests/testapi.js:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@53638 268f45cc-cd09-0410-ab3c-d52691b4dbfc