WebKit-https.git
11 years agoVersioning.k
mrowe@apple.com [Tue, 7 Jul 2009 20:16:39 +0000 (20:16 +0000)]
Versioning.k

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@45604 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-07-07 Nate Chapin <japhet@chromium.org>
japhet@chromium.org [Tue, 7 Jul 2009 20:02:10 +0000 (20:02 +0000)]
2009-07-07  Nate Chapin  <japhet@chromium.org>

        Reviewed by David Levin.

        Relanding r45559, which was rolled back at r45574.

        https://bugs.webkit.org/show_bug.cgi?id=26857

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@45603 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoBug 26995 - AX: Some webpages do not send AXLoadComplete
cfleizach@apple.com [Tue, 7 Jul 2009 19:05:45 +0000 (19:05 +0000)]
Bug 26995 - AX: Some webpages do not send AXLoadComplete
https://bugs.webkit.org/show_bug.cgi?id=26995

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@45602 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoWebCore:
pdherbemont@apple.com [Tue, 7 Jul 2009 18:02:52 +0000 (18:02 +0000)]
WebCore:

2009-07-06  Pierre d'Herbemont  <pdherbemont@apple.com>

        Reviewed by Simon Fraser.

        https://bugs.webkit.org/show_bug.cgi?id=27015

        Make sure that the CSS properties letter-spacing, word-spacing,
        line-height, text-transform, text-indent, text-shadow,
        text-decoration and color do not affect the media element controls,
        that display text.

        Controls that display text are only present in when the theme
        MediaControllerThemeQT is being used.

        Test: media/controls-styling.html

        * css/mediaControlsQT.css:
        * rendering/MediaControlElements.cpp:
        (WebCore::MediaControlElement::updateStyle): Special case for
        text-decoration. text-decoration can't be overriden from CSS, because
        text-decoration is additive for historical reasons.

LayoutTests:

2009-07-06  Pierre d'Herbemont  <pdherbemont@apple.com>

        Reviewed by Simon Fraser.

        Make sure overriding properties won't affect the media controller.

        * media/controls-styling.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@45601 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoBug 27025: Crashes and regression test failures related to regexps in 64-bit
mrowe@apple.com [Tue, 7 Jul 2009 17:54:04 +0000 (17:54 +0000)]
Bug 27025: Crashes and regression test failures related to regexps in 64-bit
Fix <https://bugs.webkit.org/show_bug.cgi?id=27025> / <rdar://problem/7033448>.

Reviewed by Darin Adler.

For x86_64 RegexGenerator uses rbx, a callee-save register, as a scratch register but
neglects to save and restore it.  The change in handling of the output vector in r45545
altered code generation so that the RegExp::match was now storing important data in rbx,
which caused crashes and bogus results when it was clobbered.

* yarr/RegexJIT.cpp:
(JSC::Yarr::RegexGenerator::generateEnter): Save rbx.
(JSC::Yarr::RegexGenerator::generateReturn): Restore rbx.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@45600 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-07-07 Albert Wong <ajwong@chromium.org>
dimich@chromium.org [Tue, 7 Jul 2009 17:30:09 +0000 (17:30 +0000)]
2009-07-07  Albert Wong  <ajwong@chromium.org>

        Not reviewed, Chromium build fix.

        Add in missing header + fix syntax issue that crept into last patch.
        https://bugs.webkit.org/show_bug.cgi?id=27027

        * rendering/RenderThemeChromiumMac.mm:
        (WebCore::RenderThemeChromiumMac::paintMediaFullscreenButton):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@45599 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-07-07 Simon Fraser <simon.fraser@apple.com>
simon.fraser@apple.com [Tue, 7 Jul 2009 17:15:16 +0000 (17:15 +0000)]
2009-07-07  Simon Fraser  <simon.fraser@apple.com>

        Reviewed by Dave Hyatt.

        Make use of geometry information to decide which layers become composited.

        <rdar://problem/7011947>
        <https://bugs.webkit.org/show_bug.cgi?id=27021>

        In addition to looking at painting order, also, optionally, take layer
        overlap into account when deciding which RenderLayers need to be composited.

        No testcase because DRT doesn't dump which layers are composited.

        * page/FrameView.h:
        * page/FrameView.cpp:
        (WebCore::FrameView::updateCompositingLayers):
        Removed the unused CompositingUpdate parameter, and just do an early return
        when there is no view.

        * rendering/RenderLayer.h:
        * rendering/RenderLayer.cpp:
        (WebCore::RenderLayer::rendererContentChanged):
        (WebCore::RenderLayer::dirtyZOrderLists):
        (WebCore::RenderLayer::dirtyNormalFlowList):
        (WebCore::RenderLayer::styleChanged):
        (WebCore::RenderLayer::mustOverlapCompositedLayers):
        (WebCore::RenderLayer::setMustOverlapCompositedLayers):
        Change "overlay" to "overlap" for consistency with the use of "overlap"
        elsewhere.

        * rendering/RenderLayerBacking.cpp:
        (WebCore::RenderLayerBacking::updateAfterLayout):
        compositingLayersNeedUpdate() renamed to compositingLayersNeedRebuild().

        (WebCore::RenderLayerBacking::startAnimation):
        (WebCore::RenderLayerBacking::startTransition):
        Tell the compositor that we started an accelerated animation or transition.

        * rendering/RenderLayerCompositor.h:
        (WebCore::RenderLayerCompositor::compositingLayersNeedRebuild):
        Renamed, since it's explicitly about changes in hierarchy now.

        (WebCore::RenderLayerCompositor::setCompositingConsultsOverlap):
        (WebCore::RenderLayerCompositor::compositingConsultsOverlap):
        New getter and setter for whether compositing should take overlap into account.

        * rendering/RenderLayerCompositor.cpp:
        (WebCore::RenderLayerCompositor::RenderLayerCompositor):
        Add m_compositingConsultsOverlap, re-order some member variables.

        (WebCore::RenderLayerCompositor::cacheAcceleratedCompositingEnabledFlag):
        Whitespace cleanup and method rename.

        (WebCore::RenderLayerCompositor::setCompositingLayersNeedRebuild):
        Method and variable renaming.

        (WebCore::RenderLayerCompositor::updateCompositingLayers):
        Now we have to run through the layer hierarchy every time if looking
        at overlap, because changes in layout can now alter compositing behavior.
        We minimize work by tracking whether we actually need to change the layer
        hierarchy via needLayerRebuild.

        (WebCore::RenderLayerCompositor::updateBacking):
        3D transforms turn off overlap mode.

        (WebCore::RenderLayerCompositor::layerWasAdded):
        (WebCore::RenderLayerCompositor::layerWillBeRemoved):
        Method rename.

        (WebCore::RenderLayerCompositor::addToOverlapMap):
        (WebCore::RenderLayerCompositor::overlapsCompositedLayers):
        Utility methods to add a layer to the overlapMap (computing the absolute
        bounding box only if we haven't already done so), and testing the map
        entries.

        (WebCore::RenderLayerCompositor::computeCompositingRequirements):
        If a layer would composite only because it comes after other compositing
        layers in the painting order, then consult the overlap map to determine whether
        it overlaps, and thus actually needs to composite.

        Add layers to the map when they must be composited.

        (WebCore::RenderLayerCompositor::rebuildCompositingLayerTree):
        Only do re-parenting work if the updateHierarchy flag is set.

        (WebCore::RenderLayerCompositor::didStartAcceleratedAnimation):
        When we start an accelerated transition or animation, we stop looking at
        overlap because we can no longer guarantee correct front-to-back ordering while the
        accelerated animation is running.

        (WebCore::RenderLayerCompositor::needsToBeComposited):
        Method renames.

        * rendering/RenderView.cpp:
        (WebCore::RenderView::setMaximalOutlineSize):
        Add comment indicating that this could be optimized.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@45598 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago Updated BugsSite with local changes on server
ddkilzer@apple.com [Tue, 7 Jul 2009 16:52:32 +0000 (16:52 +0000)]
    Updated BugsSite with local changes on server

        The new files were created by checksetup.pl during the upgrade
        to Bugzilla 3.2.3.  The rest are self-explanatory.

        * data: Added "bugzilla-update.xml" to svn:ignore.
        * data/attachments: Added "groups.*" to svn:ignore.
        * data/params: Updated for a few changed parameters.

        * lib/.htaccess: Added.
        * skins/contrib/Dusk/IE-fixes.css: Added.
        * skins/contrib/Dusk/admin.css: Added.
        * skins/contrib/Dusk/create_attachment.css: Added.
        * skins/contrib/Dusk/dependency-tree.css: Added.
        * skins/contrib/Dusk/duplicates.css: Added.
        * skins/contrib/Dusk/editusers.css: Added.
        * skins/contrib/Dusk/help.css: Added.
        * skins/contrib/Dusk/index.css: Added.
        * skins/contrib/Dusk/panel.css: Added.
        * skins/contrib/Dusk/params.css: Added.
        * skins/contrib/Dusk/release-notes.css: Added.
        * skins/contrib/Dusk/show_bug.css: Added.
        * skins/contrib/Dusk/show_multiple.css: Added.
        * skins/contrib/Dusk/summarize-time.css: Added.
        * skins/contrib/Dusk/voting.css: Added.
        * skins/contrib/Dusk/yui: Added.
        * skins/contrib/Dusk/yui/calendar.css: Added.
        * skins/custom/IE-fixes.css: Added.
        * skins/custom/admin.css: Added.
        * skins/custom/buglist.css: Added.
        * skins/custom/create_attachment.css: Added.
        * skins/custom/dependency-tree.css: Added.
        * skins/custom/duplicates.css: Added.
        * skins/custom/editusers.css: Added.
        * skins/custom/help.css: Added.
        * skins/custom/panel.css: Added.
        * skins/custom/params.css: Added.
        * skins/custom/release-notes.css: Added.
        * skins/custom/show_bug.css: Added.
        * skins/custom/show_multiple.css: Added.
        * skins/custom/summarize-time.css: Added.
        * skins/custom/voting.css: Added.
        * skins/custom/yui: Added.
        * skins/custom/yui/calendar.css: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@45597 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-07-07 Anton Muhin <antonm@chromium.org>
darin@chromium.org [Tue, 7 Jul 2009 16:16:57 +0000 (16:16 +0000)]
2009-07-07  Anton Muhin  <antonm@chromium.org>

        Reviewed by Darin Fisher.

        Speed up creation of V8 wrappers for DOM nodes.

        https://bugs.webkit.org/show_bug.cgi?id=26882

        This patch doesn't require new tests as it a set of refactorings
        to speed up wrapper creation.

        * bindings/v8/V8Proxy.cpp:
        * bindings/v8/V8Proxy.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@45596 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-07-07 Jan Michael Alonzo <jmalonzo@webkit.org>
jmalonzo@webkit.org [Tue, 7 Jul 2009 15:51:24 +0000 (15:51 +0000)]
2009-07-07  Jan Michael Alonzo  <jmalonzo@webkit.org>

        Reviewed by Gustavo Noronha.

        [GTK] textarea height property works only if other property are defined
        https://bugs.webkit.org/show_bug.cgi?id=18984

        Let WebCore handle textarea's metrics instead of readjusting it RenderthemeGtk.

        Test: fast/forms/textarea-metrics.html

        * platform/gtk/RenderThemeGtk.cpp:
        * platform/gtk/RenderThemeGtk.h:
        * LayoutTests/platform/gtk/Skipped:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@45595 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-07-07 Ben Murdoch <benm@google.com>
jmalonzo@webkit.org [Tue, 7 Jul 2009 14:29:15 +0000 (14:29 +0000)]
2009-07-07  Ben Murdoch  <benm@google.com>

        Reviewed by Antti Koivisto.

        HTML5 Database becomes locked if a transaction is in progress when the page is refreshed.
        https://bugs.webkit.org/show_bug.cgi?id=25711

        Tests fix for https://bugs.webkit.org/show_bug.cgi?id=25711, where
        web storage databases could become locked until the browser is
        shut down if the page is refreshed whilst a transaction is in
        progress.

        * storage/database-lock-after-reload-expected.txt: Added.
        * storage/database-lock-after-reload.html: Added.
        * storage/resources/database-lock-after-reload-2.html: Added.

        Fix for https://bugs.webkit.org/show_bug.cgi?id=25711 where web
        storage databases could become locked until the browser is shut
        down if the page is refreshed whilst a transaction is in progress.

        Test: storage/database-lock-after-reload.html

        * storage/Database.cpp:
        (WebCore::Database::Database):
        (WebCore::Database::close): add code to inform the database thread we've closed the database.
        (WebCore::Database::performOpenAndVerify): add code to inform the database thread we've opened a database.
        * storage/Database.h:
        (WebCore::Database::opened): return true iff the underlying sqlite database has been opened but not closed.
        * storage/DatabaseThread.cpp:
        (WebCore::DatabaseThread::databaseThread): Before the database thread terminates, close any databases that ran transactions in this thread.
        (WebCore::DatabaseThread::recordDatabaseOpen): Records a database that executed a transaction in this thread.
        (WebCore::DatabaseThread::recordDatabaseClosed): Removes a database from the set of open databases.
        * storage/DatabaseThread.h:
        (WebCore::DatabaseThread::getThreadID): return the thread id for the database thread.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@45594 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoWebKit/qt:
hausmann@webkit.org [Tue, 7 Jul 2009 14:04:02 +0000 (14:04 +0000)]
WebKit/qt:

2009-07-07  Simon Hausmann  <hausmann@webkit.org>

        Reviewed by Holger Freyther.

        Add Qt DRT hook for clearing the frame name.

        * Api/qwebframe.cpp:
        (qt_drt_clearFrameName):

WebKitTools:

2009-07-07  Simon Hausmann  <hausmann@webkit.org>

        Reviewed by Holger Freyther.

        Clear the main frame's name between loading pages, like in r36652. This
        reduces the side-effects between test cases.

        * DumpRenderTree/qt/DumpRenderTree.cpp:
        (WebCore::DumpRenderTree::open):

LayoutTests:

2009-07-07  Simon Hausmann  <hausmann@webkit.org>

        Reviewed by Holger Freyther.

        Removed subframe-self-close.html from the Qt skip list.

        * platform/qt/Skipped:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@45593 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-07-07 Jiahua Huang <jhuangjiahua@gmail.com>
jmalonzo@webkit.org [Tue, 7 Jul 2009 12:52:23 +0000 (12:52 +0000)]
2009-07-07  Jiahua Huang  <jhuangjiahua@gmail.com>

        Reviewed by Jan Alonzo.

        [Gtk] Paste of rich text from firefox results garbled markup
        https://bugs.webkit.org/show_bug.cgi?id=26791

        Fix problem with UTF-16 clipboard pasted.

        * manual-tests/gtk/paste-richtext-from-firefox.html: Added.
        * platform/gtk/PasteboardGtk.cpp:
        (WebCore::Pasteboard::documentFragment):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@45592 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-07-07 Yury Semikhatsky <yurys@chromium.org>
pfeldman@chromium.org [Tue, 7 Jul 2009 12:35:04 +0000 (12:35 +0000)]
2009-07-07  Yury Semikhatsky  <yurys@chromium.org>

        Reviewed by Timothy Hatcher.

        Change 'Continue debug' shortcut from F5 to F8 for consistency with Firebug.

        https://bugs.webkit.org/show_bug.cgi?id=23849

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

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@45591 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-07-07 Simon Hausmann <hausmann@webkit.org>
hausmann@webkit.org [Tue, 7 Jul 2009 10:48:17 +0000 (10:48 +0000)]
2009-07-07  Simon Hausmann  <hausmann@webkit.org>

        Reviewed by Holger Freyther.

        Added all current failing Qt DRT tests to the skip list, for a fresh start.

        * platform/qt/Skipped:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@45590 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-07-06 Simon Hausmann <hausmann@webkit.org>
hausmann@webkit.org [Tue, 7 Jul 2009 10:27:20 +0000 (10:27 +0000)]
2009-07-06  Simon Hausmann  <hausmann@webkit.org>

        Reviewed by Holger Freyther.

        Fix various layout tests with Qt's DRT.

        Remember to set the focus on the web page before
        loading.

        * DumpRenderTree/qt/DumpRenderTree.cpp:
        (WebCore::DumpRenderTree::open):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@45589 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-07-07 Simon Hausmann <hausmann@webkit.org>
hausmann@webkit.org [Tue, 7 Jul 2009 10:25:11 +0000 (10:25 +0000)]
2009-07-07  Simon Hausmann  <hausmann@webkit.org>

        Reviewed by Tor Arne Vestbø and Holger Freyther.

        Fix timeout timer handling in Qt DRT.

        Use QBasicTimer instead of startTimer/killTimer, to fix the
        problem that starting the timeout timer did not stop a
        previously started timer, causing multiple emissions of timeout.
        The timerEvent() implementation now also checks the timer id, to
        protect against double timeouts.

        * DumpRenderTree/qt/jsobjects.cpp:
        (LayoutTestController::LayoutTestController):
        (LayoutTestController::reset):
        (LayoutTestController::waitUntilDone):
        (LayoutTestController::notifyDone):
        (LayoutTestController::timerEvent):
        * DumpRenderTree/qt/jsobjects.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@45588 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-07-06 Albert J. Wong <ajwong@chromium.org>
eric@webkit.org [Tue, 7 Jul 2009 03:40:07 +0000 (03:40 +0000)]
2009-07-06  Albert J. Wong  <ajwong@chromium.org>

        Reviewed by Maciej Stachowiak.

        Update RenderThemeChromiumMac for wkDrawMediaUIPart and wkDrawMediaSlider API
        change.

        Fix compile RenderThemeChromiumMac.mm due to API changes from r45572
        https://bugs.webkit.org/show_bug.cgi?id=27018

        This is mainly an API change update to fix the chromium compile.  It
        also has a small fix of a typo in RenderThemeMac.mm.

        Compilation against the head of chromium trunk passed.  Since this is
        a compile fix, that should be sufficient.

        * rendering/RenderThemeChromiumMac.mm:
        (WebCore::):
        (WebCore::RenderThemeChromiumMac::paintMediaFullscreenButton):
        (WebCore::RenderThemeChromiumMac::paintMediaMuteButton):
        (WebCore::RenderThemeChromiumMac::paintMediaPlayButton):
        (WebCore::RenderThemeChromiumMac::paintMediaSeekBackButton):
        (WebCore::RenderThemeChromiumMac::paintMediaSeekForwardButton):
        (WebCore::RenderThemeChromiumMac::paintMediaSliderTrack):
        (WebCore::RenderThemeChromiumMac::paintMediaSliderThumb):
        * rendering/RenderThemeMac.mm:
        (WebCore::):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@45585 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago Bug 27006: Build fix when MAC_JAVA_BRIDGE is disabled
ddkilzer@apple.com [Tue, 7 Jul 2009 02:49:08 +0000 (02:49 +0000)]
    Bug 27006: Build fix when MAC_JAVA_BRIDGE is disabled

        <https://bugs.webkit.org/show_bug.cgi?id=27006>

        Reviewed by Darin Adler.

        * WebCoreSupport/WebFrameLoaderClient.mm:
        (WebFrameLoaderClient::createJavaAppletWidget): Wrap contents of
        the method in #if ENABLE(MAC_JAVA_BRIDGE)/#endif and return 0
        when this feature is disabled.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@45584 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago Bug 27005: Build fixes when NETSCAPE_PLUGIN_API is disabled
ddkilzer@apple.com [Tue, 7 Jul 2009 02:44:31 +0000 (02:44 +0000)]
    Bug 27005: Build fixes when NETSCAPE_PLUGIN_API is disabled

        <https://bugs.webkit.org/show_bug.cgi?id=27005>

        Reviewed by Geoff Garen.

        * Plugins/WebPluginController.mm:
        (-[WebPluginController destroyPlugin:]): Wrap call to
        ScriptController::cleanupScriptObjectsForPlugin() in
        #if ENABLE(NETSCAPE_PLUGIN_API)/#endif macro.
        (-[WebPluginController destroyAllPlugins]): Ditto.
        * Plugins/WebPluginDatabase.mm:
        (-[WebPluginDatabase removePluginInstanceViewsFor:]): Wrap
        WebBaseNetscapePluginView class checks in
        #if ENABLE(NETSCAPE_PLUGIN_API)/#endif macro.
        (-[WebPluginDatabase destroyAllPluginInstanceViews]): Ditto.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@45583 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago BUILD FIX: Use ENABLE(NETSCAPE_PLUGIN_API) instead of USE(PLUGIN_HOST_PROCESS)
ddkilzer@apple.com [Tue, 7 Jul 2009 02:39:38 +0000 (02:39 +0000)]
    BUILD FIX: Use ENABLE(NETSCAPE_PLUGIN_API) instead of USE(PLUGIN_HOST_PROCESS)

        In r45579, #if/#endif macros for USE(PLUGIN_HOST_PROCESS) were
        used, but ENABLE(NETSCAPE_PLUGIN_API) should have been used
        instead.

        * Plugins/WebNetscapeContainerCheckContextInfo.h:
        * Plugins/WebNetscapeContainerCheckContextInfo.mm:
        * Plugins/WebNetscapeContainerCheckPrivate.mm:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@45582 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago Bug 27004: Build fix for ENABLE(PLUGIN_PROXY_FOR_VIDEO) after r42618
ddkilzer@apple.com [Tue, 7 Jul 2009 02:18:45 +0000 (02:18 +0000)]
    Bug 27004: Build fix for ENABLE(PLUGIN_PROXY_FOR_VIDEO) after r42618

        <https://bugs.webkit.org/show_bug.cgi?id=27004>

        Reviewed by Geoff Garen.

        * Plugins/WebPluginController.mm:
        (mediaProxyClient): Use core() method to convert from
        DOMElement to WebCore::Element.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@45581 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-07-06 Eric Seidel <eric@webkit.org>
eric@webkit.org [Tue, 7 Jul 2009 02:18:41 +0000 (02:18 +0000)]
2009-07-06  Eric Seidel  <eric@webkit.org>

        Reviewed by Maciej Stachowiak.

        bugzilla-tool fails to close bugs with bugzilla 3.0
        https://bugs.webkit.org/show_bug.cgi?id=27008

        Update bugzilla.py to match changes in the bug page "changeform"
        Update scm.py to add bug_ids to attachments to make error reporting nicer.

        * Scripts/modules/bugzilla.py:
        * Scripts/modules/scm.py:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@45580 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago Bug 27003: Build fix when USE(PLUGIN_HOST_PROCESS) is disabled
ddkilzer@apple.com [Tue, 7 Jul 2009 02:14:33 +0000 (02:14 +0000)]
    Bug 27003: Build fix when USE(PLUGIN_HOST_PROCESS) is disabled

        <https://bugs.webkit.org/show_bug.cgi?id=27003>

        Reviewed by Geoff Garen.

        * Plugins/WebNetscapeContainerCheckContextInfo.h: Added #if
        USE(PLUGIN_HOST_PROCESS)/#endif guards.
        * Plugins/WebNetscapeContainerCheckContextInfo.mm: Ditto.
        * Plugins/WebNetscapeContainerCheckPrivate.mm: Ditto.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@45579 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago Bug 27002: Build fix when DASHBOARD_SUPPORT is disabled with -Wunused-parameter
ddkilzer@apple.com [Tue, 7 Jul 2009 01:01:14 +0000 (01:01 +0000)]
    Bug 27002: Build fix when DASHBOARD_SUPPORT is disabled with -Wunused-parameter

        <https://bugs.webkit.org/show_bug.cgi?id=27002>

        Reviewed by Geoff Garen.

        * html/CanvasRenderingContext2D.cpp:
        (WebCore::CanvasRenderingContext2D::prepareGradientForDashboard):
        Mark gradient as an unused parameter when DASHBOARD_SUPPORT is
        disabled.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@45578 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago Bug 27001: Fix improper use of PassRefPtr<Node> to RefPtr<Node>
ddkilzer@apple.com [Tue, 7 Jul 2009 00:59:02 +0000 (00:59 +0000)]
    Bug 27001: Fix improper use of PassRefPtr<Node> to RefPtr<Node>

        <https://bugs.webkit.org/show_bug.cgi?id=27001>

        Reviewed by Geoff Garen.

        PassRefPtr<> should only be used for arguments to functions that
        take ownership of the object, or as return values from functions
        that relinquish ownership of the object.

        * editing/Editor.cpp:
        (WebCore::Editor::increaseSelectionListLevelOrdered): Changed
        stack-allocated PassRefPtr<Node> to RefPtr<Node> and call
        release() on returned object.
        (WebCore::Editor::increaseSelectionListLevelUnordered): Ditto.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@45577 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-07-06 Eric Carlson <eric.carlson@apple.com>
eric.carlson@apple.com [Tue, 7 Jul 2009 00:44:23 +0000 (00:44 +0000)]
2009-07-06  Eric Carlson  <eric.carlson@apple.com>

        Reviewed by Simon Fraser.

        <rdar://problem/7035474> Make new media controller UI default on SnowLeopard

        * rendering/RenderThemeMac.mm:
        (WebCore::mediaControllerTheme):
            New media controller UI is enabled by default on SnowLeopard.

2009-07-06  Eric Carlson  <eric.carlson@apple.com>

        Reviewed by Simon Fraser.

        <rdar://problem/7035474> Make new media controller UI default on SnowLeopard

        Disable media tests that depend on controller size/position on SnowLeopard until
        <rdar://problem/6502936> Update run-webkit-tests to deal with new media controller look
        has been fixed

        * platform/mac-snowleopard/Skipped:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@45576 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-07-06 Nate Chapin <japhet@chromium.org>
japhet@chromium.org [Mon, 6 Jul 2009 23:18:36 +0000 (23:18 +0000)]
2009-07-06  Nate Chapin  <japhet@chromium.org>

        Unreviewed, build fix.

        Roll out r45559 to fix Chromium canary.

        https://bugs.webkit.org/show_bug.cgi?id=26857

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@45574 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-07-06 Eric Carlson <eric.carlson@apple.com>
eric.carlson@apple.com [Mon, 6 Jul 2009 22:56:34 +0000 (22:56 +0000)]
2009-07-06  Eric Carlson  <eric.carlson@apple.com>

        Reviewed by Darin Adler.

        <rdar://problem/7008093> Media controller can’t be used to scrub when movie is
        narrow — track is too narrow

        Do not show media controller time display elements when the a movie is too narrow.

        * platform/mac/WebCoreSystemInterface.h:
        * platform/mac/WebCoreSystemInterface.mm:
            The 'state' parameter to wkDrawMediaUIPart is now an unsigned bitfield.

        * rendering/MediaControlElements.cpp:
        (WebCore::MediaControlTimeDisplayElement::MediaControlTimeDisplayElement):
            Give current time and time remaining controls a common base class.
        (WebCore::MediaControlTimeDisplayElement::setVisible):
            New method, hide and show the element.
        * rendering/MediaControlElements.h:

        * rendering/RenderMedia.cpp:
        (WebCore::RenderMedia::RenderMedia):
            Initialize m_previousVisible.
        (WebCore::RenderMedia::layout):
            Show/hide the time display elements as the movie width changes.
        (WebCore::RenderMedia::createCurrentTimeDisplay):
        (WebCore::RenderMedia::createTimeRemainingDisplay):
            Base class is now MediaControlTimeDisplayElement, not MediaControlElement.
        (WebCore::RenderMedia::shouldShowTimeDisplayControls):
            New, decide if time display elements should be visible or not.
        * rendering/RenderMedia.h:

        * rendering/RenderThemeMac.mm:
        (WebCore::):
        (WebCore::RenderThemeMac::paintMediaFullscreenButton):
        (WebCore::RenderThemeMac::paintMediaMuteButton):
        (WebCore::RenderThemeMac::paintMediaPlayButton):
        (WebCore::RenderThemeMac::paintMediaSeekBackButton):
        (WebCore::RenderThemeMac::paintMediaSeekForwardButton):
        (WebCore::RenderThemeMac::paintMediaSliderTrack):
        (WebCore::RenderThemeMac::paintMediaSliderThumb):
        (WebCore::RenderThemeMac::paintMediaRewindButton):
        (WebCore::RenderThemeMac::paintMediaReturnToRealtimeButton):
        (WebCore::RenderThemeMac::paintMediaControlsBackground):
        (WebCore::RenderThemeMac::paintMediaCurrentTime):
        (WebCore::RenderThemeMac::paintMediaTimeRemaining):
            The 'state' parameter to wkDrawMediaUIPart is now an unsigned bitfield.

2009-07-06  Eric Carlson  <eric.carlson@apple.com>

        Update WebKitSystemInterface for <rdar://problem/7008093>.

        * WebKitSystemInterface.h:
        * libWebKitSystemInterfaceLeopard.a:
        * libWebKitSystemInterfaceSnowLeopard.a:
        * libWebKitSystemInterfaceTiger.a:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@45572 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago Bug 27000: Minor clean up to runtime_root.{cpp|h}
ddkilzer@apple.com [Mon, 6 Jul 2009 22:46:53 +0000 (22:46 +0000)]
    Bug 27000: Minor clean up to runtime_root.{cpp|h}

        <https://bugs.webkit.org/show_bug.cgi?id=27000>

        Reviewed by Geoff Garen.

        * bridge/runtime_root.cpp: Added blank line between license and
        the first #include statement.
        * bridge/runtime_root.h: Added comment to #endif.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@45571 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-07-06 Roland Steiner <rolandsteiner@google.com>
eric@webkit.org [Mon, 6 Jul 2009 22:42:38 +0000 (22:42 +0000)]
2009-07-06  Roland Steiner  <rolandsteiner@google.com>

        Reviewed by Maciej Stachowiak.

        generalize the special height treatment for SVG (to be re-used for ruby):
            renamed InlineBox::m_isSVG to m_hasVirtualHeight
            renamed InlineBox::isSVG() to hasVirtualHeight()
            renamed InlineBox::setIsSVG() to setHasVirtualHeight()

        * rendering/InlineBox.cpp:
        (WebCore::InlineBox::height):
        * rendering/InlineBox.h:
        (WebCore::InlineBox::InlineBox):
        (WebCore::InlineBox::isText):
        (WebCore::InlineBox::setIsText):
        (WebCore::InlineBox::isSVGRootInlineBox):
        (WebCore::InlineBox::hasVirtualHeight):
        (WebCore::InlineBox::setHasVirtualHeight):
        (WebCore::InlineBox::virtualHeight):
        * rendering/RenderSVGInline.cpp:
        (WebCore::RenderSVGInline::createFlowBox):
        * rendering/RenderSVGInlineText.cpp:
        (WebCore::RenderSVGInlineText::createTextBox):
        * rendering/RenderSVGText.cpp:
        (WebCore::RenderSVGText::createRootBox):
        * rendering/SVGInlineFlowBox.h:
        (WebCore::SVGInlineFlowBox::virtualHeight):
        * rendering/SVGInlineTextBox.h:
        (WebCore::SVGInlineTextBox::virtualHeight):
        * rendering/SVGRootInlineBox.h:
        (WebCore::SVGRootInlineBox::virtualHeight):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@45570 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-07-06 Alice Liu <alice.liu@apple.com>
alice.liu@apple.com [Mon, 6 Jul 2009 22:11:40 +0000 (22:11 +0000)]
2009-07-06  Alice Liu  <alice.liu@apple.com>

        Reviewed by Darin Adler.

        REGRESSION(r45285): focus rings are black on windows safari
        https://bugs.webkit.org/show_bug.cgi?id=26821
        <rdar://problem/7018252>

        Add a mechanism for setting a custom focus ring color, and relocate the
        determination of focus ring color to base class RenderTheme.

        No new tests.  Already-existing tests run in pixel mode would have caught this regression

        * css/CSSStyleSelector.cpp:
        (WebCore::CSSStyleSelector::getColorFromPrimitiveValue): Move the responsibility
        of determining focus color to base class RenderTheme.
        * rendering/RenderTheme.cpp:
        (WebCore::customFocusRingColor): Added static local.
        (WebCore::RenderTheme::setCustomFocusRingColor): Added
        (WebCore::RenderTheme::focusRingColor): Use custom color over platform color.
        * rendering/RenderTheme.h:
        (WebCore::RenderTheme::platformFocusRingColor): default implementation, black color
        * rendering/RenderThemeSafari.cpp:
        (WebCore::RenderTheme::themeForPage): Circumstances that lead to returning the RenderThemeWin
        are the same under which we want to use the focus ring color from SafariTheme.
        (WebCore::RenderThemeSafari::platformFocusRingColor): Renamed from focusRingColor
        * rendering/RenderThemeSafari.h:

        Renaming focusRingColor to platformFocusRingColor in these files:
        * rendering/RenderThemeChromiumMac.h:
        * rendering/RenderThemeChromiumMac.mm:
        (WebCore::RenderThemeChromiumMac::platformFocusRingColor):
        * rendering/RenderThemeChromiumSkia.cpp:
        (WebCore::RenderThemeChromiumSkia::platformFocusRingColor):
        * rendering/RenderThemeChromiumSkia.h:
        * rendering/RenderThemeMac.h:
        * rendering/RenderThemeMac.mm:
        (WebCore::RenderThemeMac::platformFocusRingColor):

        Cleanup leftover from removal of WebCore::Color in r45285 in these files:
        * WebCore.order:
        * platform/graphics/Color.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@45569 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-07-06 Shinichiro Hamaji <hamaji@chromium.org>
eric@webkit.org [Mon, 6 Jul 2009 21:53:50 +0000 (21:53 +0000)]
2009-07-06  Shinichiro Hamaji  <hamaji@chromium.org>

        Reviewed by David Hyatt.

        Bug 15135: REGRESSION (r19843-r19850): Changing a flexbox's
        contents makes its container scroll to the top
        https://bugs.webkit.org/show_bug.cgi?id=15135

        Delay updateing scroll bar of descendants of flexbox until their
        positions are determined.  In this way we can prevent descendants
        of flexible boxes from changing positions of their scrollbars
        using tentative positions.

        * fast/flexbox/repaint-scrollbar-expected.txt: Added.
        * fast/flexbox/repaint-scrollbar.html: Added.
        * fast/flexbox/resources/repaint-scrollbar.js: Added.

2009-07-06  Shinichiro Hamaji  <hamaji@chromium.org>

        Reviewed by David Hyatt.

        Bug 15135: REGRESSION (r19843-r19850): Changing a flexbox's
        contents makes its container scroll to the top
        https://bugs.webkit.org/show_bug.cgi?id=15135

        Delay updateing scroll bar of descendants of flexbox until their
        positions are determined.  In this way we can prevent descendants
        of flexible boxes from changing positions of their scrollbars
        using tentative positions.

        Test: fast/flexbox/repaint-scrollbar.html

        * rendering/RenderBlock.cpp:
        (WebCore::RenderBlock::startDelayUpdateScrollInfo):
        (WebCore::RenderBlock::finishDelayUpdateScrollInfo):
        (WebCore::RenderBlock::updateScrollInfoAfterLayout):
        (WebCore::RenderBlock::layoutBlock):
        (WebCore::RenderBlock::layoutOnlyPositionedObjects):
        * rendering/RenderBlock.h:
        * rendering/RenderFlexibleBox.cpp:
        (WebCore::RenderFlexibleBox::layoutHorizontalBox):
        (WebCore::RenderFlexibleBox::layoutVerticalBox):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@45568 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-07-06 Hironori Bono <hbono@chromium.org>
eric@webkit.org [Mon, 6 Jul 2009 21:33:52 +0000 (21:33 +0000)]
2009-07-06  Hironori Bono  <hbono@chromium.org>

        Reviewed by Maciej Stachowiak.

        Make unconfirmed IME text affect textarea's value.
        This matches input and contentEditable elements as well
        IE and Firefox.

        This fixes https://bugs.webkit.org/show_bug.cgi?id=25061.

        Input elements would go down this code path because it
        would always get a null compositionNode from frame->editor().
        Special casing compositionNodes is wrong because we explicitly
        want unconfirmed IME input in the textarea's value (assuming we
        want to match IE and Firefox here).

        This change is originally created by Ojan Vafai <ojan@chromium.org> and
        I just changed its manual tests with an automated test on his behalf.

        * platform/mac/editing/input/text-control-ime-input-expected.txt: Added.
        * platform/mac/editing/input/text-control-ime-input.html: Added.

2009-07-06  Hironori Bono  <hbono@chromium.org>

        Reviewed by Maciej Stachowiak.

        Make unconfirmed IME text affect textarea's value.
        This matches input and contentEditable elements as well
        IE and Firefox.

        This fixes https://bugs.webkit.org/show_bug.cgi?id=25061.

        Input elements would go down this code path because it
        would always get a null compositionNode from frame->editor().
        Special casing compositionNodes is wrong because we explicitly
        want unconfirmed IME input in the textarea's value (assuming we
        want to match IE and Firefox here).

        This change is originally created by Ojan Vafai <ojan@chromium.org> and
        I just changed its manual tests with an automated test on his behalf.

        Test: platform/mac/editing/input/text-control-ime-input.html

        * rendering/RenderTextControl.cpp:
        (WebCore::RenderTextControl::text):
        (WebCore::RenderTextControl::textWithHardLineBreaks):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@45567 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago Reviewed by Darin Adler and Mark Rowe.
adachan@apple.com [Mon, 6 Jul 2009 21:13:59 +0000 (21:13 +0000)]
    Reviewed by Darin Adler and Mark Rowe.

        Decommitted spans are added to the list of normal spans rather than
        the returned spans in TCMalloc_PageHeap::Delete().
        https://bugs.webkit.org/show_bug.cgi?id=26998

        In TCMalloc_PageHeap::Delete(), the deleted span can be decommitted in
        the process of merging with neighboring spans that are also decommitted.
        The merged span needs to be placed in the list of returned spans (spans
        whose memory has been returned to the system).  Right now it's always added
        to the list of the normal spans which can theoretically cause thrashing.

        * wtf/FastMalloc.cpp:
        (WTF::TCMalloc_PageHeap::Delete):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@45566 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoWebCore:
andersca@apple.com [Mon, 6 Jul 2009 20:59:09 +0000 (20:59 +0000)]
WebCore:

2009-07-06  Anders Carlsson  <andersca@apple.com>

        Reviewed by Adele Peterson.

        Add the ability for wheel events to latch to a node.

        * WebCore.base.exp:
        * page/EventHandler.cpp:
        (WebCore::EventHandler::EventHandler):
        (WebCore::EventHandler::clear):
        (WebCore::EventHandler::handleWheelEvent):
        * page/EventHandler.h:
        * page/mac/EventHandlerMac.mm:
        (WebCore::EventHandler::wheelEvent):
        * platform/mac/WebCoreSystemInterface.h:
        * platform/mac/WebCoreSystemInterface.mm:

WebKit/mac:

2009-07-06  Anders Carlsson  <andersca@apple.com>

        Reviewed by Adele Peterson.

        Initialize wkIsLatchingWheelEvent.

        * WebCoreSupport/WebSystemInterface.m:
        (InitWebCoreSystemInterface):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@45565 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-07-06 Nate Chapin <japhet@chromium.org>
japhet@chromium.org [Mon, 6 Jul 2009 20:55:28 +0000 (20:55 +0000)]
2009-07-06  Nate Chapin  <japhet@chromium.org>

        Reviewed by Dimitri Glazkov.

        Fix a bunch of layout test crahses in Chromium caused by a bad usage of DEFINE_STATIC_LOCAL.

        https://bugs.webkit.org/show_bug.cgi?id=26997

        * bindings/v8/V8Binding.cpp:
        (WebCore::v8ValueToWebCoreString): Use a regular static declaration instead of DEFINE_STATIC_LOCAL.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@45564 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-07-06 Anders Carlsson <andersca@apple.com>
andersca@apple.com [Mon, 6 Jul 2009 20:48:17 +0000 (20:48 +0000)]
2009-07-06  Anders Carlsson  <andersca@apple.com>

        Update WebKitSystemInterface.

        * WebKitSystemInterface.h:
        * libWebKitSystemInterfaceLeopard.a:
        * libWebKitSystemInterfaceSnowLeopard.a:
        * libWebKitSystemInterfaceTiger.a:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@45563 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-07-06 Pavel Feldman <pfeldman@chromium.org>
eric@webkit.org [Mon, 6 Jul 2009 20:47:32 +0000 (20:47 +0000)]
2009-07-06  Pavel Feldman  <pfeldman@chromium.org>

        Reviewed by Timothy Hatcher.

        WebInspector: suggest global properties based on async evaluation.

        https://bugs.webkit.org/show_bug.cgi?id=26976

        Before this change, empty string was evaluated to the global object
        (or scope chain object) synchronously. This is now fixed and global
        object is evaluated using the same control flow.

        * inspector/front-end/Console.js:
        (WebInspector.Console.prototype.completions):
        (WebInspector.Console.prototype._evalInInspectedWindow):
        * inspector/front-end/ScriptsPanel.js:
        (WebInspector.ScriptsPanel.prototype._variablesInScope):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@45562 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoWebCore:
ojan@chromium.org [Mon, 6 Jul 2009 20:37:16 +0000 (20:37 +0000)]
WebCore:

2009-07-06  Ojan Vafai  <ojan@chromium.org>

        Reviewed by Eric Seidel.

        Fix crash when indenting in an empty table cell.
        https://bugs.webkit.org/show_bug.cgi?id=26872

        The crash is that we would call splitTreeToNode where the node
        and the nodeToSplitTo were the same node.

        Test: editing/execCommand/indent-empty-table-cell.html

        * editing/CompositeEditCommand.cpp:
        (WebCore::CompositeEditCommand::splitTreeToNode):
        Added an assert in that node and nodeToSplitTo
        are different nodes to make this assumption explicit.

        * editing/IndentOutdentCommand.cpp:
        (WebCore::IndentOutdentCommand::isAtUnsplittableElement):
        (WebCore::IndentOutdentCommand::indentRegion):
        * editing/IndentOutdentCommand.h:

LayoutTests:

2009-07-06  Ojan Vafai  <ojan@chromium.org>

        Reviewed by Eric Seidel.

        Fix crash when indenting in an empty table cell.
        https://bugs.webkit.org/show_bug.cgi?id=26872

        The crash is that we would call splitTreeToNode where the node
        and the nodeToSplitTo were the same node.

        * editing/execCommand/indent-empty-table-cell-expected.txt: Added.
        * editing/execCommand/indent-empty-table-cell.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@45561 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoadded pseudocode form decomposition method
cmarrin@apple.com [Mon, 6 Jul 2009 18:10:12 +0000 (18:10 +0000)]
added pseudocode form decomposition method

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@45560 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-07-06 Nate Chapin <japhet@chromium.org>
japhet@chromium.org [Mon, 6 Jul 2009 17:38:30 +0000 (17:38 +0000)]
2009-07-06  Nate Chapin  <japhet@chromium.org>

        Reviewed by David Levin.

        Upstream V8Binding and update some function names.

        https://bugs.webkit.org/show_bug.cgi?id=26857

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@45559 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoWebCore
kov@webkit.org [Mon, 6 Jul 2009 14:37:55 +0000 (14:37 +0000)]
WebCore

2009-07-06  Gustavo Noronha Silva  <gustavo.noronha@collabora.co.uk>

        Reviewed by Jan Alonzo.

        Use soup's content sniffing
        https://bugs.webkit.org/show_bug.cgi?id=26982

        Drop our hackish content sniffing code, and use the new libsoup
        feature to do that job for us.

        Testing this is in a cross-platform way is not obvious or possible
        while we are using platform-specific code for sniffing.

        * platform/network/ResourceHandleInternal.h:
        (WebCore::ResourceHandleInternal::ResourceHandleInternal):
        * platform/network/soup/ResourceHandleSoup.cpp:
        (WebCore::gotHeadersCallback):
        (WebCore::contentSniffedCallback):
        (WebCore::gotChunkCallback):
        (WebCore::ResourceHandle::startHttp):

WebKit/gtk

2009-07-06  Gustavo Noronha Silva  <gustavo.noronha@collabora.co.uk>

        Reviewed by Jan Alonzo.

        Use soup's content sniffing
        https://bugs.webkit.org/show_bug.cgi?id=26982

        Add the new Content Sniffing feature provided by libsoup.

        * webkit/webkitprivate.cpp:
        (webkit_init):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@45558 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-07-05 Antonio Gomes <antonio.gomes@openbossa.org>
darin@apple.com [Mon, 6 Jul 2009 05:22:00 +0000 (05:22 +0000)]
2009-07-05  Antonio Gomes  <antonio.gomes@openbossa.org>

        Reviewed by Darin Adler.

        REGRESSION (r40499): fast/dom/cssTarget-crash.html fails
        https://bugs.webkit.org/show_bug.cgi?id=20342

        Re-added code removed by commit r40499.
        Without this, both Qt and Mac were crashing while running the test.

        Note that this does not entirely fix the bug. It fixes the WebCore
        crash, but the test no longer seems to work due to loader changes.
        So this patch does not reenable the test. The test probably has to
        be rewritten.

        * dom/ContainerNode.cpp:
        (WebCore::ContainerNode::removedFromDocument): Re-added code to
        set the CSS target of the document to 0.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@45557 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-07-05 Chris Marrin <cmarrin@apple.com>
simon.fraser@apple.com [Mon, 6 Jul 2009 01:46:50 +0000 (01:46 +0000)]
2009-07-05  Chris Marrin  <cmarrin@apple.com>

        Reviewed by Simon Fraser.

        https://bugs.webkit.org/show_bug.cgi?id=26943

        When one transition finishes slightly before another the longer
        one will fire a second time. This is because the second
        ImplicitAnmation object is culled too early, before its final
        RenderStyle is in place. This is done by cleanupFinishedAnimations()
        so I got rid of that method completely and now cleanup each
        transition or animation at the point where I am setting the final
        style, or when I detect that the transition or animation has been
        terminated early (which happens when you remove it from the style).

        Test: transitions/extra-transition.html

        * page/animation/AnimationController.cpp:
        (WebCore::AnimationControllerPrivate::updateStyleIfNeededDispatcherFired):
        * page/animation/CompositeAnimation.cpp:
        (WebCore::CompositeAnimation::updateTransitions):
        (WebCore::CompositeAnimation::updateKeyframeAnimations):
        (WebCore::CompositeAnimation::animate):
        * page/animation/CompositeAnimation.h:

        * page/animation/AnimationBase.cpp:
        (WebCore::AnimationBase::getTimeToNextEvent):
        Avoid a divide by zero if m_animation->duration() is zero, which can happen
        if the duration is changed to zero while the animation is running.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@45556 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-07-05 Simon Fraser <simon.fraser@apple.com>
simon.fraser@apple.com [Sun, 5 Jul 2009 19:39:56 +0000 (19:39 +0000)]
2009-07-05  Simon Fraser  <simon.fraser@apple.com>

        Revert the previous commit because it broke the
        animations/transition-and-animation-1.html
        testcase.

        * page/animation/AnimationBase.cpp:
        (WebCore::AnimationBase::getTimeToNextEvent):
        * page/animation/AnimationController.cpp:
        (WebCore::AnimationControllerPrivate::updateStyleIfNeededDispatcherFired):
        * page/animation/CompositeAnimation.cpp:
        (WebCore::CompositeAnimation::updateTransitions):
        (WebCore::CompositeAnimation::updateKeyframeAnimations):
        (WebCore::CompositeAnimation::animate):
        (WebCore::CompositeAnimation::cleanupFinishedAnimations):
        * page/animation/CompositeAnimation.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@45555 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-07-05 Chris Marrin <cmarrin@apple.com>
simon.fraser@apple.com [Sun, 5 Jul 2009 19:18:51 +0000 (19:18 +0000)]
2009-07-05  Chris Marrin  <cmarrin@apple.com>

        Reviewed by Simon Fraser.

        https://bugs.webkit.org/show_bug.cgi?id=26943

        When one transition finishes slightly before another the longer
        one will fire a second time. This is because the second
        ImplicitAnmation object is culled too early, before its final
        RenderStyle is in place. This is done by cleanupFinishedAnimations()
        so I got rid of that method completely and now cleanup each
        transition or animation at the point where I am setting the final
        style, or when I detect that the transition or animation has been
        terminated early (which happens when you remove it from the style).

        Test: transitions/extra-transition.html

        * page/animation/AnimationController.cpp:
        (WebCore::AnimationControllerPrivate::updateStyleIfNeededDispatcherFired):
        * page/animation/CompositeAnimation.cpp:
        (WebCore::CompositeAnimation::updateTransitions):
        (WebCore::CompositeAnimation::updateKeyframeAnimations):
        (WebCore::CompositeAnimation::animate):
        * page/animation/CompositeAnimation.h:

        * page/animation/AnimationBase.cpp:
        (WebCore::AnimationBase::getTimeToNextEvent):
        Avoid a divide by zero if m_animation->duration() is zero, which can happen
        if the duration is changed to zero while the animation is running.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@45554 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoJavaScriptCore:
hausmann@webkit.org [Sun, 5 Jul 2009 15:45:31 +0000 (15:45 +0000)]
JavaScriptCore:

2009-07-05  Lars Knoll  <lars.knoll@nokia.com>

        Reviewed by Maciej Stachowiak.

        https://bugs.webkit.org/show_bug.cgi?id=26843

        Fix run-time crashes in JavaScriptCore with the Metrowerks compiler on Symbian.

        The Metrowerks compiler on the Symbian platform moves the globally
        defined Hashtables into read-only memory, despite one of the members
        being mutable. This causes crashes at run-time due to write access to
        read-only memory.

        Avoid the use of const with this compiler by introducing the
        JSC_CONST_HASHTABLE macro.

        Based on idea by Norbert Leser.

        * runtime/Lookup.h: Define JSC_CONST_HASHTABLE as const for !WINSCW.
        * create_hash_table: Use JSC_CONST_HASHTABLE for hashtables.
        * runtime/JSGlobalData.cpp: Import various global hashtables via the macro.

WebCore:

2009-07-05  Lars Knoll  <lars.knoll@nokia.com>

        Reviewed by Maciej Stachowiak.

        https://bugs.webkit.org/show_bug.cgi?id=26843

        Fix run-time crashes in JavaScriptCore with the Metrowerks compiler on Symbian.

        The Metrowerks compiler on the Symbian platform moves the globally
        defined Hashtables into read-only memory, despite one of the members
        being mutable. This causes crashes at run-time due to write access to
        read-only memory.

        Avoid the use of const with this compiler by introducing the
        JSC_CONST_HASHTABLE macro.

        Based on idea by Norbert Leser.

        * bindings/scripts/CodeGeneratorJS.pm: Use JSC_CONST_HASHTABLE for hash tables
        define in the bindings.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@45553 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-07-05 Holger Hans Peter Freyther <zecke@selfish.org>
zecke@webkit.org [Sun, 5 Jul 2009 14:53:28 +0000 (14:53 +0000)]
2009-07-05  Holger Hans Peter Freyther  <zecke@selfish.org>

        Reviewed by Xan Lopez.

        Always initialize WebCore.

        It is possible that WebKitWebSettings and other classes get
        constructed before the WebKitWebView. In this case WebCore is
        not yet properly initialized. Add webkit_init to every class
        that can be constructed by API users.

        * webkit/webkitdownload.cpp:
        (webkit_download_class_init):
        * webkit/webkitnetworkrequest.cpp:
        (webkit_network_request_class_init):
        * webkit/webkitwebbackforwardlist.cpp:
        (webkit_web_back_forward_list_class_init):
        * webkit/webkitwebhistoryitem.cpp:
        (webkit_web_history_item_class_init):
        * webkit/webkitwebsettings.cpp:
        (webkit_web_settings_class_init):
        * webkit/webkitwebwindowfeatures.cpp:
        (webkit_web_window_features_class_init):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@45552 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-07-05 Simon Hausmann <hausmann@webkit.org>
hausmann@webkit.org [Sun, 5 Jul 2009 14:33:33 +0000 (14:33 +0000)]
2009-07-05  Simon Hausmann  <hausmann@webkit.org>

        Reviewed by Holger Freyther.

        Fix two qdoc warnings.

        Added missing \property for QWebFrame::hasFocus and added \a
        tag for pos of QWebPage::frameAt.

        * Api/qwebframe.cpp:
        * Api/qwebpage.cpp:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@45551 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago Reviewed by Maciej Stachowiak.
rwlbuis@webkit.org [Sun, 5 Jul 2009 13:06:48 +0000 (13:06 +0000)]
    Reviewed by Maciej Stachowiak.

        Bug 26897 - Dynamic SVG images do not display correctly
        https://bugs.webkit.org/show_bug.cgi?id=26897

        Use repaint() to fix a rendering problem with a SVG image embedded in xhtml.

        Test: svg/custom/createImageElement2.xhtml

        * rendering/RenderSVGImage.cpp:
        (WebCore::RenderSVGImage::imageChanged):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@45550 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-07-05 Xan Lopez <xlopez@igalia.com>
xan@webkit.org [Sun, 5 Jul 2009 10:43:37 +0000 (10:43 +0000)]
2009-07-05  Xan Lopez  <xlopez@igalia.com>

        Reviewed by Jan Alonzo.

        https://bugs.webkit.org/show_bug.cgi?id=26960
        [Gtk] caret offset not updated when selecting text

        Report the caret offset from the end of the selection so it works
        correctly for multi-char selections too (ie, anything that is not
        the zero width caret).

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

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@45549 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoFix Qt link error by adding two new localized strings
zecke@webkit.org [Sun, 5 Jul 2009 07:28:24 +0000 (07:28 +0000)]
Fix Qt link error by adding two new localized strings

In r45474 two new strings got added to LocalizedStrings.h add
the definition of them to Qt.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@45548 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-07-04 Holger Hans Peter Freyther <zecke@selfish.org>
zecke@webkit.org [Sun, 5 Jul 2009 07:27:53 +0000 (07:27 +0000)]
2009-07-04  Holger Hans Peter Freyther  <zecke@selfish.org>

        Reviewed by Simon Hausmann.

        Use the recently introduced FocusController::setFocused

        Use the recently introduced FocusController::setFocused
        in the Qt platform. The SelectionController will be updated
        from within the FocusController now.

        * Api/qwebpage.cpp:
        (QWebPagePrivate::focusInEvent):
        (QWebPagePrivate::focusOutEvent):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@45547 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago - debug build fix
mitz@apple.com [Sat, 4 Jul 2009 17:53:20 +0000 (17:53 +0000)]
    - debug build fix

        * runtime/RegExpConstructor.cpp:
        (JSC::RegExpConstructor::getLastParen):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@45546 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-07-03 Yong Li <yong.li@torchmobile.com>
mjs@apple.com [Sat, 4 Jul 2009 14:21:30 +0000 (14:21 +0000)]
2009-07-03  Yong Li  <yong.li@torchmobile.com>

        Reviewed by Maciej Stachowiak (and revised slightly)

        RegExp::match to be optimized
        https://bugs.webkit.org/show_bug.cgi?id=26957

        Allow regexp matching to use Vectors with inline capacity instead of
        allocating a new ovector buffer every time.

        ~5% speedup on SunSpider string-unpack-code test, 0.3% on SunSpider overall.

        * runtime/RegExp.cpp:
        (JSC::RegExp::match):
        * runtime/RegExp.h:
        * runtime/RegExpConstructor.cpp:
        (JSC::RegExpConstructorPrivate::RegExpConstructorPrivate):
        (JSC::RegExpConstructorPrivate::lastOvector):
        (JSC::RegExpConstructorPrivate::tempOvector):
        (JSC::RegExpConstructorPrivate::changeLastOvector):
        (JSC::RegExpConstructor::performMatch):
        (JSC::RegExpMatchesArray::RegExpMatchesArray):
        (JSC::RegExpMatchesArray::fillArrayInstance):
        (JSC::RegExpConstructor::getBackref):
        (JSC::RegExpConstructor::getLastParen):
        (JSC::RegExpConstructor::getLeftContext):
        (JSC::RegExpConstructor::getRightContext):
        * runtime/StringPrototype.cpp:
        (JSC::stringProtoFuncSplit):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@45545 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-07-03 Maciej Stachowiak <mjs@apple.com>
mjs@apple.com [Sat, 4 Jul 2009 14:20:51 +0000 (14:20 +0000)]
2009-07-03  Maciej Stachowiak  <mjs@apple.com>

        Reviewed by Eric Seidel.

        bugzilla-tool forgot how to obsolete patches
        https://bugs.webkit.org/show_bug.cgi?id=26951

        * Scripts/modules/bugzilla.py:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@45544 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoAdded Radar link I left out by accident the first time.
darin@apple.com [Sat, 4 Jul 2009 04:50:34 +0000 (04:50 +0000)]
Added Radar link I left out by accident the first time.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@45543 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-07-03 Darin Adler <darin@apple.com>
darin@apple.com [Sat, 4 Jul 2009 04:48:13 +0000 (04:48 +0000)]
2009-07-03  Darin Adler  <darin@apple.com>

        Reviewed by Adele Peterson.

        REGRESSION (r44670-r44680): Typing is suprisingly slow in password field on reddit.com
        https://bugs.webkit.org/show_bug.cgi?id=26959

        The code to handle iteration boundaries was malfunctioning when the boundary was
        at the edge of a shadow tree. This happens all the time with <input> elements.

        It's not immediately obvious how to make a regression test for this since the
        symptom was a performance problem, not incorrect behavior. I'll add a test if I
        figure out a way to make one.

        * editing/TextIterator.cpp:
        (WebCore::parentCrossingShadowBoundaries): Renamed from parentOrShadowParent.
        (WebCore::depthCrossingShadowBoundaries): Updated for name change.
        (WebCore::nextInPreOrderCrossingShadowBoundaries): Added.
        (WebCore::previousInPostOrderCrossingShadowBoundaries): Added.
        (WebCore::setUpFullyClippedStack): Updated for name change.
        (WebCore::TextIterator::TextIterator): Updated for name change. Use
        nextInPreOrderCrossingShadowBoundaries instead of Range::pastLastNode to fix the bug.
        (WebCore::TextIterator::advance): Updated for name change.
        (WebCore::SimplifiedBackwardsTextIterator::SimplifiedBackwardsTextIterator):
        Tweaked formatting.
        (WebCore::SimplifiedBackwardsTextIterator::advance): Ditto. Changed code that
        initializes m_pastStartNode to use previousInPostOrderCrossingShadowBoundaries.
        (WebCore::SimplifiedBackwardsTextIterator::handleNonTextNode): Tweaked formatting.
        (WebCore::SimplifiedBackwardsTextIterator::exitNode): Ditto.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@45542 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago Reviewed by Simon Fraser.
mitz@apple.com [Sat, 4 Jul 2009 03:56:30 +0000 (03:56 +0000)]
    Reviewed by Simon Fraser.

        - fix <rdar://problem/6964278> REGRESSION (r42118): Scrolling redraw
          problem in FileMaker Pro

        * WebView/WebHTMLView.mm:
        (-[WebHTMLView _recursiveDisplayRectIfNeededIgnoringOpacity:isVisibleRect:rectIsVisibleRectForView:topView:]):
        Perform layout if needed, even on Mac OS X versions that have
        -viewWillDraw. This prevents attempts to draw without layout in case
        -viewWillDraw was not called due to NSView issues or the client
        did something during the display operation that re-invalidated the
        layout.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@45541 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago Bug 26958: Change edit link to review link in request messages
ddkilzer@apple.com [Sat, 4 Jul 2009 03:27:33 +0000 (03:27 +0000)]
    Bug 26958: Change edit link to review link in request messages

        <https://bugs.webkit.org/show_bug.cgi?id=26958>

        Reviewed by Dan Bernstein.

        * template/en/custom/request/email.txt.tmpl: Copied from BugsSite/template/en/default/request/email.txt.tmpl.
        Changed link from "action=edit" to "action=review".

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@45540 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoNot reviewed. Forgot to land ontimer-inline-event.html results within my last WML...
zimmermann@webkit.org [Sat, 4 Jul 2009 01:09:16 +0000 (01:09 +0000)]
Not reviewed. Forgot to land ontimer-inline-event.html results within my last WML commit.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@45539 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago Bug 26950: Make the summary and alias fields support click-to-edit
ddkilzer@apple.com [Sat, 4 Jul 2009 00:19:47 +0000 (00:19 +0000)]
    Bug 26950: Make the summary and alias fields support click-to-edit
        <https://bugs.webkit.org/show_bug.cgi?id=26950>

        Reviewed by Maciej Stachowiak.

        Original patch by Maciej Stachowiak.

        * js/field.js:
        (hideEditableField): Updated to add click event listeners to the
        alias and short description elements to make them easier to
        edit.  Renamed field_id parameter to field2_id (short
        description id) and added a field1_id parameter (alias id).
        (showEditableField): If a third parameter is passed in the
        ContainerInputArray parameter, use it to find the element to
        focus.  Otherwise, fall back to the original behavior of
        focusing the first input field.
        (hideAliasAndSummary): Changed to pass the id for the alias
        element to hideEditableField().

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@45538 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-07-03 Simon Fraser <simon.fraser@apple.com>
simon.fraser@apple.com [Fri, 3 Jul 2009 20:54:24 +0000 (20:54 +0000)]
2009-07-03  Simon Fraser  <simon.fraser@apple.com>

        Reviewed by Sam Weinig.

        Background audio stops playing when JS GC runs
        https://bugs.webkit.org/show_bug.cgi?id=26956

        Ensure we keep the Audio object around while it is playing, so that it
        doesn't get collected.

        Not testable because there's no way to know whether the audio keeps
        playing without holding a reference to it.

        * bindings/js/JSDOMBinding.cpp:
        (WebCore::isObservableThroughDOM):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@45537 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-07-03 Adam Barth <abarth@webkit.org>
abarth@webkit.org [Fri, 3 Jul 2009 19:34:21 +0000 (19:34 +0000)]
2009-07-03  Adam Barth  <abarth@webkit.org>

        Reviewed by Sam Weinig.

        Adam and Collin's affiliation is incorrect
        https://bugs.webkit.org/show_bug.cgi?id=26954

        This change corrects their affiliation.

        * security/security-group-members.html:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@45536 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-07-02 Xan Lopez <xlopez@igalia.com>
xan@webkit.org [Fri, 3 Jul 2009 14:08:34 +0000 (14:08 +0000)]
2009-07-02  Xan Lopez  <xlopez@igalia.com>

        Reviewed by Jan Alonzo.

        https://bugs.webkit.org/show_bug.cgi?id=26814
        [Gtk] Caret-moved events are not issued for the correct offset
        when text is selected forward

        Report the caret offset from the end of the selection, otherwise
        we'll report the same offset when moving the caret while
        modififying the selection (eg, with Shift + Right).

        * editing/gtk/SelectionControllerGtk.cpp:
        (WebCore::SelectionController::notifyAccessibilityForSelectionChange):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@45535 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-07-02 Xan Lopez <xlopez@igalia.com>
xan@webkit.org [Fri, 3 Jul 2009 14:08:22 +0000 (14:08 +0000)]
2009-07-02  Xan Lopez  <xlopez@igalia.com>

        Reviewed by Jan Alonzo.

        https://bugs.webkit.org/show_bug.cgi?id=26815
        [Gtk] text-selection-changed events are not issued for the correct
        object when the selection spans multiple objects

        Get the focused node from the end of the selection, not the start,
        so we can detect when we cross object boundaries.

        * editing/gtk/SelectionControllerGtk.cpp:
        (WebCore::SelectionController::notifyAccessibilityForSelectionChange):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@45534 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-07-03 Dmitry Titov <dimich@chromium.org>
dimich@chromium.org [Fri, 3 Jul 2009 10:47:41 +0000 (10:47 +0000)]
2009-07-03  Dmitry Titov  <dimich@chromium.org>

        Reviewed by Darin Fisher.

        Increase delay in worker-terminate layout test to make it more reliable.
        https://bugs.webkit.org/show_bug.cgi?id=26917

        This test used time delay to see of terminated worker does not return
        messages (terminated). In some test conditions, the current delay (500ms)
        is too short and test fails. Move setting the delay to the end of initialization
        and increase it 2x to make test more reliable.

        * fast/workers/worker-terminate.html:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@45533 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-07-03 Jan Michael Alonzo <jmalonzo@webkit.org>
jmalonzo@webkit.org [Fri, 3 Jul 2009 10:24:18 +0000 (10:24 +0000)]
2009-07-03  Jan Michael Alonzo  <jmalonzo@webkit.org>

        Skip tests that require expected results that Gtk doesn't have.

        * platform/gtk/Skipped:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@45532 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-07-03 Jan Michael Alonzo <jmalonzo@webkit.org>
jmalonzo@webkit.org [Fri, 3 Jul 2009 09:57:39 +0000 (09:57 +0000)]
2009-07-03  Jan Michael Alonzo  <jmalonzo@webkit.org>

        Reviewed by Xan Lopez and Gustavo Noronha.

        Set user-agent from application
        https://bugs.webkit.org/show_bug.cgi?id=17375

        Add 'user-agent' setter/getter API to WebSettings.

        * WebCoreSupport/FrameLoaderClientGtk.cpp:
        (WebKit::FrameLoaderClient::FrameLoaderClient):
        (WebKit::FrameLoaderClient::userAgent):
        * WebCoreSupport/FrameLoaderClientGtk.h:
        * webkit/webkitprivate.h:
        * webkit/webkitversion.h.in:
        * webkit/webkitwebsettings.cpp:
        (webkit_get_user_agent):
        (webkit_web_settings_class_init):
        (webkit_web_settings_finalize):
        (webkit_web_settings_set_property):
        (webkit_web_settings_get_property):
        (webkit_web_settings_copy):
        (webkit_web_settings_get_user_agent):
        * webkit/webkitwebsettings.h:
        * webkit/webkitwebview.cpp:
        (webkit_web_view_dispose):

        Define UA version macros to be used by the UA string.
        Add new WebSettings unit test for the User-Agent string API.

        * GNUmakefile.am:
        * configure.ac:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@45531 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoEdit WebCore/ChangeLog last entry to fix a spelling typo.
pdherbemont@apple.com [Fri, 3 Jul 2009 05:21:11 +0000 (05:21 +0000)]
Edit WebCore/ChangeLog last entry to fix a spelling typo.

Pointed out by Mark Rowe.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@45530 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-07-02 Pierre d'Herbemont <pdherbemont@apple.com>
pdherbemont@apple.com [Fri, 3 Jul 2009 05:12:44 +0000 (05:12 +0000)]
2009-07-02  Pierre d'Herbemont  <pdherbemont@apple.com>

        Reviewed by Simon Fraser.

        https://bugs.webkit.org/show_bug.cgi?id=26947

        Fix the media controls hit tests. They may loose click events.

        Test cases already covered in the video-controls-zoomed tests.

        Don't trust wkHitTestMediaUIPart for the tests. We are accurate enough.

        * rendering/RenderThemeMac.h: Remove hitTestMediaControlPart.
        * rendering/RenderThemeMac.mm: Remove hitTestMediaControlPart.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@45529 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoIgnore data/template cache directory and update data/params.
ddkilzer@apple.com [Fri, 3 Jul 2009 03:34:57 +0000 (03:34 +0000)]
Ignore data/template cache directory and update data/params.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@45528 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-07-02 Brent Fulgham <bfulgham@webkit.org>
bfulgham@webkit.org [Fri, 3 Jul 2009 03:15:32 +0000 (03:15 +0000)]
2009-07-02  Brent Fulgham  <bfulgham@webkit.org>

        Unreviewed build fix (Cairo this time) for DumpRenderTree.
        Put Cairo code back in, but use original include order
        for the CG stuff to avoid compile issues.

        * DumpRenderTree/win/PixelDumpSupportWin.cpp:
        (createBitmapContextFromWebView):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@45527 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago Config file for contrib/recode.pl when upgrading bugs.webkit.org
ddkilzer@apple.com [Fri, 3 Jul 2009 02:14:46 +0000 (02:14 +0000)]
    Config file for contrib/recode.pl when upgrading bugs.webkit.org

        Current as of midday on Wednesday, July 1, 2009.

        * contrib/recode-overrides.txt: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@45526 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago Workaround for WebKit Bug 9630 when running contrib/recode.pl
ddkilzer@apple.com [Fri, 3 Jul 2009 02:14:45 +0000 (02:14 +0000)]
    Workaround for WebKit Bug 9630 when running contrib/recode.pl

        A number of WebKit nightly builds included Bug 9630 which caused
        non-breaking space characters (0xA0) to be submitted within the
        content of textarea elements.

        * contrib/recode.pl: To work around these 0xA0 characters, try
        an encoding of cp1252 and use it if it works.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@45525 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago Don't print "Failed to guess" warning if an override is available
ddkilzer@apple.com [Fri, 3 Jul 2009 02:14:44 +0000 (02:14 +0000)]
    Don't print "Failed to guess" warning if an override is available

        * contrib/recode.pl: Check that %overrides does NOT contain a
        key matching the current digest before printing out the warning
        about a failed charset guess.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@45524 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago Added --[no-]truncate switch to contrib/recode.pl
ddkilzer@apple.com [Fri, 3 Jul 2009 02:14:42 +0000 (02:14 +0000)]
    Added --[no-]truncate switch to contrib/recode.pl

        * contrib/recode.pl: Added --[no-]truncate switch to make
        debugging of failed charset guessing easier.  Often times the
        illegal character was truncated in the output.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@45523 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago Changed Perl scripts to use #!/usr/bin/env perl
ddkilzer@apple.com [Fri, 3 Jul 2009 02:14:39 +0000 (02:14 +0000)]
    Changed Perl scripts to use #!/usr/bin/env perl

        perl -pi -e 's@#\!/usr/bin/perl@#\!/usr/bin/env perl@' *.cgi *.pl */*.pl

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@45522 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago Replace Bugzilla favicon with webkit.org favicon
ddkilzer@apple.com [Fri, 3 Jul 2009 02:14:33 +0000 (02:14 +0000)]
    Replace Bugzilla favicon with webkit.org favicon

        * favicon.ico: Removed.
        * images/favicon.ico: Replaced.  Copied from favicon.ico.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@45521 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago Merged BugsSite to Bugzilla-3.2.3
ddkilzer@apple.com [Fri, 3 Jul 2009 02:14:25 +0000 (02:14 +0000)]
    Merged BugsSite to Bugzilla-3.2.3

        Updated to the latest-and-greatest stable version.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@45520 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago Merged BugsSite to Bugzilla-3.0.3
ddkilzer@apple.com [Fri, 3 Jul 2009 02:13:41 +0000 (02:13 +0000)]
    Merged BugsSite to Bugzilla-3.0.3

        Nothing to see here.  Move along.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@45519 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-07-02 Pierre d'Herbemont <pdherbemont@apple.com>
pdherbemont@apple.com [Fri, 3 Jul 2009 01:38:29 +0000 (01:38 +0000)]
2009-07-02  Pierre d'Herbemont  <pdherbemont@apple.com>

        Reviewed by Simon Fraser.

        https://bugs.webkit.org/show_bug.cgi?id=26944

        Make sure we support full page zoom in video controls.

        This is tracked by a the video-controls-zoom test case.

        * css/mediaControlsQT.css: Make sure we don't have any inherited margin.
        * rendering/MediaControlElements.cpp:
        (WebCore::MediaControlElement::updateStyle): Propagate the style to the innertext.
        * rendering/RenderThemeMac.mm: Adjust the painting rect.
        (WebCore::getUnzoomedRectAndAdjustCurrentContext):
        (WebCore::RenderThemeMac::paintMediaSliderTrack):
        (WebCore::RenderThemeMac::paintMediaCurrentTime):
        (WebCore::RenderThemeMac::paintMediaTimeRemaining):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@45518 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-07-02 Roland Steiner <rolandsteiner@google.com>
eric@webkit.org [Fri, 3 Jul 2009 01:29:07 +0000 (01:29 +0000)]
2009-07-02  Roland Steiner  <rolandsteiner@google.com>

        Reviewed by Eric Seidel.

        RenderBlock and RenderInline have confusingly named object creation methods:
        RenderBlock::createRootBox/createRootInlineBox
        RenderInline::createFlowBox/createInlineFlowBox
        where the 2nd method in both cases just calls the first and then appends the created object.
        I therefore renamed those methods to something IMHO more informative:

        createRootBox()         -> createRootInlineBox()
        createRootInlineBox()   -> createAndAppendRootInlineBox();
        createFlowBox()         -> createInlineFlowBox();
        createInlineFlowBox()   -> createAndAppendInlineFlowBox();

        https://bugs.webkit.org/show_bug.cgi?id=26828

        * rendering/RenderBlock.cpp:
        (WebCore::RenderBlock::createRootInlineBox):
        (WebCore::RenderBlock::createAndAppendRootInlineBox):
        * rendering/RenderBlock.h:
        * rendering/RenderInline.cpp:
        (WebCore::RenderInline::createInlineFlowBox):
        (WebCore::RenderInline::createAndAppendInlineFlowBox):
        * rendering/RenderInline.h:
        * rendering/RenderSVGInline.cpp:
        (WebCore::RenderSVGInline::createInlineFlowBox):
        * rendering/RenderSVGInline.h:
        * rendering/RenderSVGText.cpp:
        (WebCore::RenderSVGText::createRootInlineBox):
        * rendering/RenderSVGText.h:
        * rendering/bidi.cpp:
        (WebCore::createInlineBoxForRenderer):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@45517 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-07-02 Pavel Feldman <pfeldman@chromium.org>
eric@webkit.org [Fri, 3 Jul 2009 01:23:48 +0000 (01:23 +0000)]
2009-07-02  Pavel Feldman  <pfeldman@chromium.org>

        Reviewed by Timothy Hatcher.

        WebInspector: Fix typo in the resource panel enabler caption.

        https://bugs.webkit.org/show_bug.cgi?id=26753

        * English.lproj/localizedStrings.js:
        * inspector/front-end/ResourcesPanel.js:
        (WebInspector.ResourcesPanel):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@45516 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-07-02 Erik Arvidsson <arv@chromium.org>
eric@webkit.org [Fri, 3 Jul 2009 01:06:27 +0000 (01:06 +0000)]
2009-07-02  Erik Arvidsson  <arv@chromium.org>

        Reviewed by Adam Roben.

        [Win] HTML5 Drag and drop, dragend is not fired when pressing Esc
        https://bugs.webkit.org/show_bug.cgi?id=26699

        * manual-tests/drag-escape.html: Added.
        * page/EventHandler.cpp:
        (WebCore::EventHandler::dragSourceEndedAt):

2009-07-02  Erik Arvidsson  <arv@chromium.org>

        Reviewed by Adam Roben.

        [Win] HTML5 Drag and drop, dragend is not fired when pressing Esc
        https://bugs.webkit.org/show_bug.cgi?id=26699

        * WebDropSource.cpp:
        (WebDropSource::QueryContinueDrag):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@45515 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-07-02 Eric Seidel <eric@webkit.org>
eric@webkit.org [Fri, 3 Jul 2009 00:55:13 +0000 (00:55 +0000)]
2009-07-02  Eric Seidel  <eric@webkit.org>

        Correct blatant typo (missing []), no review.

        * Scripts/modules/scm.py:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@45514 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-07-02 Pierre d'Herbemont <pdherbemont@apple.com>
pdherbemont@apple.com [Fri, 3 Jul 2009 00:54:38 +0000 (00:54 +0000)]
2009-07-02  Pierre d'Herbemont  <pdherbemont@apple.com>

        Added the test result of r45511 for platform/mac.

        * platform/mac/media/controls-strict-expected.txt: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@45513 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-07-02 Brent Fulgham <bfulgham@webkit.org>
bfulgham@webkit.org [Fri, 3 Jul 2009 00:37:04 +0000 (00:37 +0000)]
2009-07-02  Brent Fulgham  <bfulgham@webkit.org>

        Windows build fix, no review.

        Reverting PixelDumpSupportWin.cpp change to avoid strange
        error on the build-bot.

        * DumpRenderTree/win/PixelDumpSupportWin.cpp:
        (createBitmapContextFromWebView):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@45512 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-07-02 Pierre d'Herbemont <pdherbemont@apple.com>
pdherbemont@apple.com [Fri, 3 Jul 2009 00:18:44 +0000 (00:18 +0000)]
2009-07-02  Pierre d'Herbemont  <pdherbemont@apple.com>

        Fix the Gtk build after r45474. The localized strings should
        have been added there.

        * platform/gtk/LocalizedStringsGtk.cpp:
        (WebCore::mediaElementLoadingStateText):
        (WebCore::mediaElementLiveBroadcastStateText):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@45511 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoWebCore:
pdherbemont@apple.com [Fri, 3 Jul 2009 00:07:24 +0000 (00:07 +0000)]
WebCore:

2009-07-02  Pierre d'Herbemont  <pdherbemont@apple.com>

        Reviewed by Simon Fraser.

        https://bugs.webkit.org/show_bug.cgi?id=26939

        Media controller is rendered badly at http://www.mozilla.com/en-US/firefox/video/firefox-3.5.html

        We fix two things:
        - We use px instead of em, because px is used everywhere else
        and because 0.09em hit the font size limit.
        - We use -webkit-box instead of inline-block because in strict mode
        inline-block has a different behavior.

        Test: media/controls-strict.html

        * css/mediaControlsQT.css:

LayoutTests:

2009-07-02  Pierre d'Herbemont  <pdherbemont@apple.com>

        Reviewed by Simon Fraser.

        https://bugs.webkit.org/show_bug.cgi?id=26939

        Media controller is rendered badly at http://www.mozilla.com/en-US/firefox/video/firefox-3.5.html

        Test that the controller renders correctly in strict mode as well.

        * media/controls-strict.html: Added.
        * platform/mac/media/controls-strict-mode-expected.txt: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@45510 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-07-02 Brent Fulgham <bfulgham@webkit.org>
bfulgham@webkit.org [Fri, 3 Jul 2009 00:03:24 +0000 (00:03 +0000)]
2009-07-02  Brent Fulgham  <bfulgham@webkit.org>

        Build fix, no review.

        * DumpRenderTree/config.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@45509 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago<rdar://problem/6939593> REGRESSION (r37793): Hitting back button on Google often...
oliver@apple.com [Thu, 2 Jul 2009 23:55:08 +0000 (23:55 +0000)]
<rdar://problem/6939593> REGRESSION (r37793): Hitting back button on Google often gives non-interactive Google page (26670)
<https://bugs.webkit.org/show_bug.cgi?id=26670>

Reviewed by Anders Carlsson

Make windows loader client match mac behaviour more closely.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@45508 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-07-02 Eric Seidel <eric@webkit.org>
eric@webkit.org [Thu, 2 Jul 2009 23:53:16 +0000 (23:53 +0000)]
2009-07-02  Eric Seidel  <eric@webkit.org>

        Reviewed by David Levin.

        bugzilla-tool needs to handle login failure
        https://bugs.webkit.org/show_bug.cgi?id=26913

        We now exit(1) on login failure.

        * Scripts/modules/bugzilla.py:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@45507 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-07-02 David Hyatt <hyatt@apple.com>
hyatt@apple.com [Thu, 2 Jul 2009 23:53:01 +0000 (23:53 +0000)]
2009-07-02  David Hyatt  <hyatt@apple.com>

        Reviewed by Dan Bernstein.

        TextIterator should use hasOverflowClip when checking for overflow instead of looking at the style.
        https://bugs.webkit.org/show_bug.cgi?id=26942

        * editing/TextIterator.cpp:
        (WebCore::fullyClipsContents):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@45506 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoWebKit/win:
bfulgham@webkit.org [Thu, 2 Jul 2009 23:49:52 +0000 (23:49 +0000)]
WebKit/win:

2009-06-30  Brent Fulgham  <bfulgham@gmail.com>

        Reviewed by Adam Roben.

        Build DumpRenderTree under Cairo.
        https://bugs.webkit.org/show_bug.cgi?id=26457

        * WebKit.vcproj/WebKit.sln: Add Cairo targets to the
          DumpRenderTree build, and use the in the Debug_Cairo
          and Release_Cairo targets.

WebKitTools:

2009-06-30  Brent Fulgham  <bfulgham@webkit.org>

        Reviewed by Adam Roben.

        Add build support for a Windows Cairo version of
        DumpRenderTree.  Share as much logic between the
        CG and Cairo builds as possible.
        https://bugs.webkit.org/show_bug.cgi?id=26457

        * DumpRenderTree/PixelDumpSupport.cpp: Added.
        (dumpWebViewAsPixelsAndCompareWithExpected): Moved common logic
          from CG-specific file.
        (printPNG): Moved common logic from CG-specific file.
        * DumpRenderTree/PixelDumpSupport.h: Add declaration for new
          common printPNG function.
        * DumpRenderTree/cairo: Added.
        * DumpRenderTree/cairo/PixelDumpSupportCairo.cpp: Added.
          Implement cairo-specific logic for dealing with Cairo surfaces.
        (writeFunction):
        (printPNG): Cairo-specific PNG handling logic.
        (computeMD5HashStringForBitmapContext): New Cairo routine using
          the generic BitmapContext signature.
        (dumpBitmap): New function containing Cairo-specific portions of
          the dumping routine.
        * DumpRenderTree/cairo/PixelDumpSupportCairo.h: Added.
          Provide Cairo version of the BitmapContext structure so that
          dumping routines can work on an abstract type.
        (BitmapContext::createByAdoptingBitmapAndContext):
        (BitmapContext::~BitmapContext):
        (BitmapContext::cairoContext):
        (BitmapContext::BitmapContext):
        * DumpRenderTree/cg/PixelDumpSupportCG.cpp:
        (printPNG): CG-specific PNG handling logic.
        (computeMD5HashStringForBitmapContext): Revised to use generic
          BitmapContext signature.
        (dumpBitmap): New function containing CG-specific portions of
          the dumping routine.
        * DumpRenderTree/cg/PixelDumpSupportCG.h: Add new signatures.
        * DumpRenderTree/win/DumpRenderTree.cpp: Disable CFNetwork logic
          when building the non-CFNetwork version.
        (main):
        * DumpRenderTree/win/DumpRenderTree.vcproj:
        * DumpRenderTree/win/PixelDumpSupportWin.cpp:
        (createBitmapContextFromWebView): Provide parallel Cairo implementation
          of CG bitmap/context setup.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@45505 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-07-02 Sam Weinig <sam@webkit.org>
weinig@apple.com [Thu, 2 Jul 2009 23:45:32 +0000 (23:45 +0000)]
2009-07-02  Sam Weinig  <sam@webkit.org>

        Reviewed by Dave Hyatt.

        Fix for <rdar://problem/5230700>
        Remove local .xhtml file workaround

        * platform/network/mac/ResourceResponseMac.mm:
        (WebCore::ResourceResponse::platformLazyInit): Work around is no longer necessary.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@45504 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoVersioning.
mrowe@apple.com [Thu, 2 Jul 2009 23:36:13 +0000 (23:36 +0000)]
Versioning.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@45503 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoWebCore:
andersca@apple.com [Thu, 2 Jul 2009 23:34:01 +0000 (23:34 +0000)]
WebCore:

2009-07-02  Anders Carlsson  <andersca@apple.com>

        Reviewed by Dan Bernstein.

        Sort, add functions used by WebKit.

        * WebCore.base.exp:

WebKit/mac:

2009-07-02  Anders Carlsson  <andersca@apple.com>

        Reviewed by Dan Bernstein.

        <rdar://problem/6978783>
        Software-rendered plug-in does not update correctly when inside a hardware layer

        Replace calls to setNeedsDisplay: and setNeedsDisplayInRect: with a call to the new method
        invalidatePluginContentRect:. This new method will ask WebCore to do the repainting, taking
        transforms into account.

        * Plugins/Hosted/NetscapePluginInstanceProxy.mm:
        (WebKit::NetscapePluginInstanceProxy::invalidateRect):
        * Plugins/Hosted/WebHostedNetscapePluginView.mm:
        (-[WebHostedNetscapePluginView pluginHostDied]):
        * Plugins/WebBaseNetscapePluginView.h:
        * Plugins/WebBaseNetscapePluginView.mm:
        (-[WebBaseNetscapePluginView windowBecameKey:]):
        (-[WebBaseNetscapePluginView windowResignedKey:]):
        (-[WebBaseNetscapePluginView preferencesHaveChanged:]):
        (-[WebBaseNetscapePluginView invalidatePluginContentRect:]):
        * Plugins/WebNetscapePluginView.mm:
        (-[WebNetscapePluginView invalidateRect:]):
        (-[WebNetscapePluginView invalidateRegion:]):
        (-[WebNetscapePluginView forceRedraw]):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@45502 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-07-02 David Hyatt <hyatt@apple.com>
hyatt@apple.com [Thu, 2 Jul 2009 23:31:39 +0000 (23:31 +0000)]
2009-07-02  David Hyatt  <hyatt@apple.com>

        Reviewed by Mark Rowe.

        Land new results from my change to make replaced elements no longer be overflow:hidden by
        default.  This affected plaintext dumping, which is clearly a bug in and of itself.  For now
        I am just updating the results and will file a followup bug about the fact that text dumping actually
        does something different with overflow on replaced elements (when it clearly should not).

        * fast/block/float/crash-replaced-display-block-expected.txt:
        * fast/dom/Range/acid3-surround-contents-expected.txt:
        * fast/dom/object-embed-plugin-scripting-expected.txt:
        * fast/parser/badentity-expected.txt:
        * http/tests/security/local-video-poster-from-remote-expected.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@45500 268f45cc-cd09-0410-ab3c-d52691b4dbfc