Timeline



Aug 2, 2009:

11:33 PM Changeset in webkit [46712] by xan@webkit.org
  • 2 edits in trunk/JavaScriptCore

2009-08-02 Xan Lopez <xlopez@igalia.com>

Fix the GTK+ build.

  • wtf/Platform.h:
9:12 PM Changeset in webkit [46711] by abarth@webkit.org
  • 2 edits in trunk/WebCore

2009-08-02 Adam Barth <abarth@webkit.org>

Reviewed by Darin Adler.

SVGElementInstance::scriptExecutionContext should call document() instead of scriptExecutionContext()

https://bugs.webkit.org/show_bug.cgi?id=27932

There's no reason to pay for a virtual function call in this case. We

can just call document() directly because we know that element is a

Node*.

  • svg/SVGElementInstance.cpp:

(WebCore::SVGElementInstance::scriptExecutionContext):

8:49 PM Changeset in webkit [46710] by abarth@webkit.org
  • 4 edits in trunk/WebCore

2009-08-02 Adam Barth <abarth@webkit.org>

Reviewed by Darin Adler.

JSNodeCustom should call document() instead of scriptExecutionContext()
https://bugs.webkit.org/show_bug.cgi?id=27931

We want to call document() because it's in inline function, instead of
scriptExecutionContext() because it's a virtual function. If we have a
Node*, then we know that the virtual function will always return the
document anyway.

There are more instance of this bug, but I'd like to start with a small
patch and work up.

  • bindings/js/JSDOMGlobalObject.cpp: (WebCore::toJSDOMGlobalObject):
  • bindings/js/JSDOMGlobalObject.h:
  • bindings/js/JSNodeCustom.cpp: (WebCore::JSNode::addEventListener): (WebCore::JSNode::removeEventListener):
4:10 PM Changeset in webkit [46709] by ggaren@apple.com
  • 2 edits in trunk/JavaScriptCore

2009-08-02 Geoffrey Garen <ggaren@apple.com>

Reviewed by Sam Weinig.

Disabled JSVALUE32_64 on Qt builds, since all layout tests mysteriously
crash with it enabled.

  • wtf/Platform.h:
4:08 PM Changeset in webkit [46708] by weinig@apple.com
  • 2 edits in trunk/LayoutTests

2009-08-02 Sam Weinig <sam@webkit.org>

Land updated expected result.

  • fast/js/array-filter-expected.txt:
3:48 PM Changeset in webkit [46707] by weinig@apple.com
  • 2 edits in trunk/LayoutTests

2009-08-02 Sam Weinig <sam@webkit.org>

Reviewed by Oliver Hunt.

Speed up fast/js/array-filter.html by 10x in debug builds by not using an
unnecessarily large sparse array.

  • fast/js/resources/array-filter.js:
3:10 PM Changeset in webkit [46706] by ggaren@apple.com
  • 2 edits in trunk/JavaScriptCore

2009-08-02 Geoffrey Garen <ggaren@apple.com>

Qt build fix.

Added JSAPIValueWrapper.cpp to the build.

3:00 PM Changeset in webkit [46705] by ggaren@apple.com
  • 3 edits in trunk/JavaScriptCore

2009-08-02 Geoffrey Garen <ggaren@apple.com>

Windows build fix.

Exported symbols for JSAPIValueWrapper.

2:56 PM Changeset in webkit [46704] by ggaren@apple.com
  • 2 edits in trunk/JavaScriptCore

2009-08-02 Geoffrey Garen <ggaren@apple.com>

GTK build fix.

  • jit/JITStubs.cpp: #include <stdarg.h>, for a definition of va_start.
2:52 PM Changeset in webkit [46703] by ggaren@apple.com
  • 2 edits in trunk/JavaScriptCore

2009-08-02 Geoffrey Garen <ggaren@apple.com>

Qt build fix.


  • runtime/Collector.cpp: #include <limits.h>, for a definition of ULONG_MAX.
2:45 PM Changeset in webkit [46702] by ggaren@apple.com
  • 5 edits in trunk/JavaScriptCore

2009-08-02 Geoffrey Garen <ggaren@apple.com>

Windows build fix: Nixed JSImmediate::prototype, JSImmediate::toObject,
and JSImmediate::toThisObject, and removed their exported symbols.

  • JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def:
  • JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore_debug.def:
  • runtime/JSImmediate.cpp:
  • runtime/JSImmediate.h:
2:33 PM Changeset in webkit [46701] by ggaren@apple.com
  • 2 edits in trunk/JavaScriptCore

2009-08-02 Geoffrey Garen <ggaren@apple.com>

Reviewed by Mark Rowe.

Enabled JSVALUE32_64 by default on all platforms other than x86_64 (which uses JSVALUE64).

  • wtf/Platform.h:
7:26 AM Changeset in webkit [46700] by ddkilzer@apple.com
  • 2 edits in trunk/WebKitTools

2009-08-02 David D. Kilzer <ddkilzer@webkit.org>

<http://webkit.org/b/27930> bugzilla-tool hates Tor Arne Vestbø

Reviewed by Tor Arne Vestbø.

  • Scripts/bugzilla-tool: (set_reviewer_in_changelog): Made sure reviewer is properly encoded when calling replace().
7:09 AM Changeset in webkit [46699] by vestbo@webkit.org
  • 7 edits in trunk/WebCore

2009-07-27 Tor Arne Vestbø <tor.arne.vestbo@nokia.com>

Reviewed by Anders Carlsson.

Share implementation of PluginView::userAgent() and userAgentStatic()

https://bugs.webkit.org/show_bug.cgi?id=27706

Differences to the original implementation are as follows:

  • The Win port used to return 0 for userAgentStatic, but now returns the quirk mode MozillaUserAgent as the other ports.
  • plugins/PluginView.cpp: (WebCore::PluginView::userAgent): (WebCore::PluginView::userAgentStatic):
  • plugins/PluginViewNone.cpp:
  • plugins/gtk/PluginViewGtk.cpp:
  • plugins/mac/PluginViewMac.cpp:
  • plugins/qt/PluginViewQt.cpp:
  • plugins/win/PluginViewWin.cpp:
7:00 AM Changeset in webkit [46698] by vestbo@webkit.org
  • 2 edits in trunk/WebCore

2009-07-27 Tor Arne Vestbø <tor.arne.vestbo@nokia.com>

Reviewed by Anders Carlsson.

Unregister plugin from PluginMainThreadScheduler if NPP_New fails

https://bugs.webkit.org/show_bug.cgi?id=27706

Since the plugin has not been started yet (m_isStarted is not set),
calling stop() will not have any affect, so we manually unregister
in start() before returning.

  • plugins/PluginView.cpp: (WebCore::PluginView::start):
6:47 AM Changeset in webkit [46697] by vestbo@webkit.org
  • 7 edits in trunk/WebCore

2009-07-27 Tor Arne Vestbø <tor.arne.vestbo@nokia.com>

Reviewed by Anders Calsson.

Join the various versions of PluginView::stop() into one shared implementation

https://bugs.webkit.org/show_bug.cgi?id=27706

The platform-dependent bits are now ifdef'ed in the shared implementation,
using the XP_ defines (XP_WIN, XP_MACOSX, XP_UNIX) from npapi.h

Differences to the original implementation are as follows:

  • The Qt port unregistered the plugin before calling NP_SetWindow with a 0 window handle. Now it's done after (similar to Win)
  • The GTK and Qt ports unregistered the plugin before clearing the m_npWindow.ws_info struct. Now it's done after.
  • The Win port constructed the DropAllLocks after unsubclassing the window. Now it's done before.
  • The Mac, Qt, and GTK ports did not pass a NPSavedData struct to NP_Destroy (like the Win port). Now all ports do.
  • The Win port did not call PluginView::setCurrentPluginView() before calling to NP-functions. Now it does.
  • plugins/PluginView.cpp: (WebCore::PluginView::start): (WebCore::PluginView::~PluginView): (WebCore::PluginView::stop):
  • plugins/PluginViewNone.cpp:
  • plugins/gtk/PluginViewGtk.cpp:
  • plugins/mac/PluginViewMac.cpp: (WebCore::PluginView::getValueStatic):
  • plugins/qt/PluginViewQt.cpp:
  • plugins/win/PluginViewWin.cpp: (WebCore::PluginView::handlePostReadFile):
1:54 AM Changeset in webkit [46696] by abarth@webkit.org
  • 1 edit
    1 add in trunk/WebKitTools

2009-08-02 Kevin Ollivier <kevino@theolliviers.com>

Reviewed by David Levin.

Script for building the wxBrowser sample app for wx.
https://bugs.webkit.org/show_bug.cgi?id=27619

  • wx/browser/wscript: Added.
1:54 AM Changeset in webkit [46695] by abarth@webkit.org
  • 1 edit
    1 add in trunk/WebKit/wx

2009-08-02 Kevin Ollivier <kevino@theolliviers.com>

Reviewed by Jan Alonzo.

Script for building the WebKit library for wx.
https://bugs.webkit.org/show_bug.cgi?id=27619

  • wscript: Added.
1:54 AM Changeset in webkit [46694] by abarth@webkit.org
  • 1 edit
    1 add in trunk/JavaScriptCore

2009-08-02 Kevin Ollivier <kevino@theolliviers.com>

Reviewed by Jan Alonzo.

Script for building the JavaScriptCore library for wx.
https://bugs.webkit.org/show_bug.cgi?id=27619

  • wscript: Added.
1:54 AM Changeset in webkit [46693] by abarth@webkit.org
  • 1 edit
    5 adds in trunk/WebKitTools

2009-08-02 Kevin Ollivier <kevino@theolliviers.com>

Reviewed by Jan Alonzo.

Scripts needed for the waf build.
https://bugs.webkit.org/show_bug.cgi?id=27619

  • wx/build: Added.
  • wx/build/build_utils.py: Added.
  • wx/build/settings.py: Added.
  • wx/build/waf_extensions.py: Added.
  • wx/build/wxpresets.py: Added.
1:05 AM Changeset in webkit [46692] by abarth@webkit.org
  • 2 edits in trunk/JavaScriptCore

2009-08-02 Yong Li <yong.li@torchmobile.com>

Reviewed by George Staikos.

DateMath depends on strftime and localtime, which need to be imported manually on WinCE
https://bugs.webkit.org/show_bug.cgi?id=26558

  • wtf/DateMath.cpp:
12:58 AM Changeset in webkit [46691] by abarth@webkit.org
  • 2 edits in trunk/WebKit/qt

2009-07-30 Laszlo Gombos <Laszlo Gombos>

Reviewed by Tor Arne Vestbø.

[Qt] Initialize HTML5 Application Cache for QtWebKit to a sensible default
https://bugs.webkit.org/show_bug.cgi?id=27866

  • Api/qwebpage.cpp: (initializeApplicationCachePathIfNecessary): (QWebPagePrivate::QWebPagePrivate):
12:41 AM Changeset in webkit [46690] by abarth@webkit.org
  • 5 edits in trunk/WebCore

2009-07-30 Joseph Pecoraro <joepeck02@gmail.com>

Reviewed by Timothy Hatcher.

Inspector: let me *edit* css styles in the web inspector.
https://bugs.webkit.org/show_bug.cgi?id=27124

  • English.lproj/localizedStrings.js:
  • inspector/front-end/ElementsPanel.js: (WebInspector.ElementsPanel): added property stylesheet pointing to 1 stylesheet added to the page if needed
  • inspector/front-end/StylesSidebarPane.js: (WebInspector.StylesSidebarPane.prototype.update): handle blank sections (WebInspector.StylesSidebarPane.prototype.addBlankSection): blank section in the correct position (WebInspector.StylesSidebarPane.prototype.appropriateSelectorForNode): helper to get a nice selector for the selectd node (WebInspector.StylePropertiesSection): (WebInspector.StylePropertiesSection.prototype.expand): (WebInspector.StylePropertiesSection.prototype.isPropertyInherited): (WebInspector.StylePropertiesSection.prototype.isPropertyOverloaded): (WebInspector.StylePropertiesSection.prototype.addNewBlankProperty): (WebInspector.StylePropertiesSection.prototype._dblclickEmptySpace): easily create new properties (WebInspector.StylePropertiesSection.prototype._dblclickSelector): (WebInspector.StylePropertiesSection.prototype.startEditingSelector): allow for editing selectors (WebInspector.StylePropertiesSection.prototype.editingSelectorCommitted): (WebInspector.StylePropertiesSection.prototype.editingSelectorCancelled): (WebInspector.StylePropertiesSection.prototype._doesSelectorAffectSelectedNode): helper to check if a selector applies to the selected node (WebInspector.BlankStylePropertiesSection.prototype._dblclick): (WebInspector.BlankStylePropertiesSection.prototype.startEditing): (WebInspector.BlankStylePropertiesSection.prototype.editingCancelled): (WebInspector.BlankStylePropertiesSection.prototype.editingCommitted): (WebInspector.BlankStylePropertiesSection.prototype.makeNormal): morph into a StylePropertiesSection (WebInspector.StylePropertyTreeElement.prototype): (WebInspector.StylePropertyTreeElement.prototype.):
  • inspector/front-end/inspector.css:
12:33 AM Changeset in webkit [46689] by abarth@webkit.org
  • 2 edits in trunk/WebCore

2009-07-31 Anton Muhin <antonm@chromium.org>

Reviewed by Adam Barth.

[v8] check if proxy is present before invoking a handler
https://bugs.webkit.org/show_bug.cgi?id=27883

Do not invoke handler function if proxy is null pointer (that would lead to access violation
anyway)

  • bindings/v8/custom/V8CustomEventListener.cpp: (WebCore::V8EventListener::callListenerFunction):

Aug 1, 2009:

11:05 PM Changeset in webkit [46688] by mitz@apple.com
  • 3 edits in trunk/WebKitTools

Reviewed by Darin Adler.

Make pixel test results more consistent across Mac OS X versions

  • DumpRenderTree/cg/ImageDiffCG.cpp: (createDifferenceImage): Avoid color correction when rendering the

images into graphics contexts, so that the actual color values are
compared, regardless of how the images are tagged.

  • DumpRenderTree/mac/DumpRenderTree.mm: (createWebViewAndOffscreenWindow): Set the window’s color space to the

main screen’s color space, even though the window is off-screen.

10:18 PM Changeset in webkit [46687] by abarth@webkit.org
  • 18 edits
    4 adds in trunk/WebCore

2009-08-01 Pavel Feldman <pfeldman@chromium.org>

Reviewed by Timothy Hatcher.

Web Inspector: Reimplement Elements Panel so that its
interaction with DOM is serialized.

This is a first cut wuth read support for DOM elements
tree as well as some limited editing. No properties, no
styles, no search capabilities are implemented yet.
Set Preferences.useDOMAgent to true in order to try it
out, otherwise disabled by default.

https://bugs.webkit.org/show_bug.cgi?id=27771

  • GNUmakefile.am:
  • WebCore.gypi:
  • WebCore.pro:
  • WebCore.vcproj/WebCore.vcproj:
  • WebCore.xcodeproj/project.pbxproj:
  • WebCoreSources.bkl:
  • inspector/InspectorBackend.cpp: (WebCore::InspectorBackend::getChildNodes): (WebCore::InspectorBackend::setAttribute): (WebCore::InspectorBackend::removeAttribute): (WebCore::InspectorBackend::setTextNodeValue):
  • inspector/InspectorBackend.h:
  • inspector/InspectorBackend.idl:
  • inspector/InspectorController.cpp: (WebCore::InspectorController::windowScriptObjectAvailable): (WebCore::InspectorController::setFrontendProxyObject): (WebCore::InspectorController::close): (WebCore::InspectorController::populateScriptObjects): (WebCore::InspectorController::didCommitLoad):
  • inspector/InspectorController.h: (WebCore::InspectorController::domAgent):
  • inspector/InspectorDOMAgent.cpp: Added.
  • inspector/InspectorDOMAgent.h: Added.
  • inspector/InspectorFrontend.cpp:
  • inspector/InspectorFrontend.h:
  • inspector/front-end/Callback.js: Added.
  • inspector/front-end/DOMAgent.js: Added.
  • inspector/front-end/ElementsPanel.js:
  • inspector/front-end/WebKit.qrc:
  • inspector/front-end/inspector.html:
  • inspector/front-end/inspector.js: (WebInspector.loaded):
7:31 PM Changeset in webkit [46686] by mrowe@apple.com
  • 2 edits in trunk/WebKitTools

Try again to make the Qt build bot happy.

  • BuildSlaveSupport/test-result-archive: Create a placeholder file inside the newly-created

directory to prevent 'zip' from failing due to the directory containing no files.

7:18 PM Changeset in webkit [46685] by mrowe@apple.com
  • 2 edits in trunk/WebKitTools

Try and make the Qt build bot happy.

  • BuildSlaveSupport/test-result-archive: Handle the case where run-webkit-tests generates

no output files at all and thus does not create the layout-test-results directory.

6:59 PM Changeset in webkit [46684] by mrowe@apple.com
  • 2 edits in trunk/WebKitTools

Move the include of DumpRenderTreeSupport to after DumpRenderTree has been built.

  • Scripts/run-webkit-tests:
6:41 PM Changeset in webkit [46683] by ddkilzer@apple.com
  • 2 edits in trunk/JavaScriptCore

wtf/Threading.h: added include of Platform.h

Reviewed by Mark Rowe.

  • wtf/Threading.h: Added #include "Platform.h" since this header uses PLATFORM() and other macros.
6:40 PM Changeset in webkit [46682] by mrowe@apple.com
  • 3 edits
    3 adds in trunk/WebKitTools

Bug 27923: run-webkit-tests should not invoke "ps" repeatedly on Mac OS X
Fix <https://bugs.webkit.org/show_bug.cgi?id=27923>.

Reviewed by Sam Weinig and David D. Kilzer.

run-webkit-tests invokes "ps" after each test on Mac OS X to determine whether DumpRenderTree is
in the process of crashing. We can do the same test more efficiently with a simple call to sysctl.
I couldn't find any built-in way of doing this directly from perl, so we do it from a simple C module
instead. This speeds up run-webkit-tests by around 8%.

  • DumpRenderTree/DumpRenderTree.xcodeproj/project.pbxproj:
  • DumpRenderTree/mac/PerlSupport/DumpRenderTreeSupport.c: Added.

(processIsCrashing):

  • DumpRenderTree/mac/PerlSupport/DumpRenderTreeSupportTiger.pm: Added. Tiger does not have the SWIG tool

that is used to generate the Perl binding to the C code, so we fall back to the old implementation on Tiger.

  • DumpRenderTree/mac/PerlSupport/Makefile: Added.
  • Scripts/run-webkit-tests:
6:04 PM Changeset in webkit [46681] by rniwa@webkit.org
  • 6 edits in trunk/WebCore

2009-08-01 Ryosuke Niwa <rniwa@webkit.org>

Reviewed by Adele Peterson.

isContentEditable function in htmlediting.* must be removed
https://bugs.webkit.org/show_bug.cgi?id=27870

This patch removes isContentEditable function in htmlediting.cpp and
replaces all usage by calling Node::isContentEditable or its derivatives.

  • editing/AppendNodeCommand.cpp: ditto (WebCore::AppendNodeCommand::AppendNodeCommand): ditto
  • editing/IndentOutdentCommand.cpp: ditto (WebCore::IndentOutdentCommand::outdentParagraph): ditto
  • editing/InsertNodeBeforeCommand.cpp: ditto (WebCore::InsertNodeBeforeCommand::InsertNodeBeforeCommand): ditto
  • editing/htmlediting.cpp: ditto (WebCore::enclosingNodeWithTag): ditto (WebCore::enclosingNodeOfType): ditto (WebCore::canMergeLists): ditto
  • editing/htmlediting.h: Updated prototype
5:24 PM Changeset in webkit [46680] by mrowe@apple.com
  • 2 edits in trunk/WebKitTools

Display the reminder once rather than once per ChangeLog.

Reviewed by David D. Kilzer.

  • Scripts/prepare-ChangeLog:
4:00 PM Changeset in webkit [46679] by mrowe@apple.com
  • 2 edits in trunk/JavaScriptCore

Roll out r46668 as it was misinformed. ScopeChain is only used with placement new.

Rubber-stamped by Oliver Hunt.

  • runtime/ScopeChain.h:
3:52 PM UpdatingBugzilla edited by ddkilzer@apple.com
(diff)
3:49 PM UpdatingBugzilla edited by ddkilzer@apple.com
(diff)
3:48 PM UpdatingBugzilla edited by ddkilzer@apple.com
(diff)
3:48 PM UpdatingBugzilla edited by ddkilzer@apple.com
(diff)
3:22 PM UpdatingBugzilla edited by ddkilzer@apple.com
(diff)
3:08 PM UpdatingBugzilla edited by ddkilzer@apple.com
(diff)
3:01 PM UpdatingBugzilla edited by ddkilzer@apple.com
(diff)
2:57 PM UpdatingBugzilla edited by ddkilzer@apple.com
(diff)
2:57 PM UpdatingBugzilla edited by ddkilzer@apple.com
(diff)
2:55 PM UpdatingBugzilla edited by ddkilzer@apple.com
(diff)
1:13 PM Changeset in webkit [46678] by mrowe@apple.com
  • 3 edits
    2 deletes in trunk/WebKitTools

Roll out r46676. Tiger is why we can't have nice things :-(

1:09 PM Changeset in webkit [46677] by bweinstein@apple.com
  • 2 edits in trunk/LayoutTests

2009-08-01 Brian Weinstein <bweinstein@apple.com>

Rubber-stamped by Mark Rowe.

Remove a non-existent test from the Windows Skipped List.

  • platform/win/Skipped:
1:05 PM Changeset in webkit [46676] by mrowe@apple.com
  • 3 edits
    3 adds in trunk/WebKitTools

Bug 27923: run-webkit-tests should not invoke "ps" repeatedly on Mac OS X
Fix <https://bugs.webkit.org/show_bug.cgi?id=27923>.

Reviewed by Sam Weinig.

run-webkit-tests invokes "ps" after each test on Mac OS X to determine whether DumpRenderTree is
in the process of crashing. We can do the same test more efficiently with a simple call to sysctl.
I couldn't find any built-in way of doing this directly from perl, so we do it from a simple C module
instead. This speeds up run-webkit-tests by around 8%.

  • DumpRenderTree/DumpRenderTree.xcodeproj/project.pbxproj:
  • DumpRenderTree/mac/PerlSupport/DumpRenderTreeSupport.c: Added.

(processIsCrashing):

  • DumpRenderTree/mac/PerlSupport/Makefile: Added.
  • Scripts/run-webkit-tests:
11:35 AM Changeset in webkit [46675] by abarth@webkit.org
  • 2 edits in trunk/JavaScriptCore

2009-08-01 Zoltan Horvath <hzoltan@inf.u-szeged.hu>

Allow custom memory allocation control for JavaScriptCore's HashMap class
http://bugs.webkit.org/show_bug.cgi?id=27871

Inherits HashMap class from FastAllocBase because it has been
instantiated by 'new' in JavaScriptCore/API/JSClassRef.cpp:148.

  • wtf/RefPtrHashMap.h: (WTF::):
11:17 AM Changeset in webkit [46674] by beidson@apple.com
  • 2 edits
    1 add in trunk/LayoutTests

2009-08-01 Brady Eidson <beidson@apple.com>

For my new test I landed last night, check in Tiger-specific results, and skip on Windows.

  • platform/mac-tiger/http/tests/misc/favicon-loads-with-images-disabled-expected.txt: Added.
  • platform/win/Skipped: Add to the skipped list because the failure needs exploration.
10:58 AM Changeset in webkit [46673] by abarth@webkit.org
  • 2 edits in trunk/LayoutTests

2009-08-01 John Abd-El-Malek <jam@chromium.org>

Reviewed by Eric Seidel.

Update test to work with multi-process message ports.

https://bugs.webkit.org/show_bug.cgi?id=27819

  • fast/events/message-channel-gc.html:
7:29 AM Changeset in webkit [46672] by ddkilzer@apple.com
  • 2 edits in trunk/WebKitTools

resolve-ChangeLogs: add error checking to git ls-files command on close()

Reviewed by Eric Seidel.

Item 2 of <https://bugs.webkit.org/show_bug.cgi?id=18599#c0>.

  • Scripts/resolve-ChangeLogs: Added error checking to close() after running git ls-files. Added error checking to all system() calls by checking for a non-zero WEXITSTATUS($?).
Changed "
die;" expressions to "or die $!;".
7:29 AM Changeset in webkit [46671] by ddkilzer@apple.com
  • 2 edits in trunk/WebKitTools

<http://webkit.org/b/18599> resolve-ChangeLogs doesn't work with relative paths

Reviewed by Eric Seidel.

  • Scripts/resolve-ChangeLogs: Used chdirReturningRelativePath() and determineVCSRoot() to change directories to the root of the project before running the command and to provide a path for removeChangeLogArguments() to make sure any ChangeLog arguments on the command line are still found. (canonicalRelativePath): Added. Returns a canonical path (e.g., stripping 'dir/../' from the path) relative to the current directory. (removeChangeLogArguments): Added argument which contains a relative path that must be prepended to any ChangeLog arguments. Used canonicalRelativePath() and File::Spec->catfile() to construct a normalized, relative path to each file.
7:29 AM Changeset in webkit [46670] by ddkilzer@apple.com
  • 3 edits in trunk/WebKitTools

Implement VCSUtils::chdirReturningRelativePath()

Reviewed by Eric Seidel.

Step 2 to fix:
<http://webkit.org/b/18599> resolve-ChangeLogs doesn't work with relative paths

  • Scripts/VCSUtils.pm: (VCSUtils::chdirReturningRelativePath): Moved here from chdirAndGetDifference() in svn-create-patch.
  • Scripts/svn-create-patch: Switched to use chdirReturningRelativePath() instead of chdirAndGetDifference(). (chdirAndGetDifference): Removed.
7:23 AM Changeset in webkit [46669] by ddkilzer@apple.com
  • 3 edits in trunk/WebKitTools

Implement VCSUtils::determineVCSRoot()

Reviewed by Eric Seidel.

Step 1 to fix:
<http://webkit.org/b/18599> resolve-ChangeLogs doesn't work with relative paths

  • Scripts/VCSUtils.pm: Removed reference to webkitdirs module. (VCSUtils::EXPORT): Added &determineVCSRoot. Realphabetized. (VCSUtils::determineGitRoot): Added. Based on code in commit-log-editor. (VCSUtils::determineVCSRoot): Implemented using determineGitRoot() and determineSVNRoot().
  • Scripts/commit-log-editor: Replaced use of topLevelSourceDirectory() with determineVCSRoot(). Resorted use statements. (topLevelSourceDirectory): Removed.
1:19 AM Changeset in webkit [46668] by abarth@webkit.org
  • 2 edits in trunk/JavaScriptCore

2009-08-01 Zoltan Horvath <hzoltan@inf.u-szeged.hu>

Allow custom memory allocation control for JavaScriptCore's ScopeChain class
https://bugs.webkit.org/show_bug.cgi?id=27834

Inherits ScopeChain class from FastAllocBase because it has been
instantiated by 'new' in JavaScriptCore/runtime/JSFunction.h:109.

  • runtime/ScopeChain.h:
12:51 AM Changeset in webkit [46667] by abarth@webkit.org
  • 2 edits in trunk/JavaScriptCore

2009-08-01 Zoltan Horvath <hzoltan@inf.u-szeged.hu>

Reviewed by Darin Adler.

Allow custom memory allocation control for JavaScriptCore's RegExpConstructorPrivate struct
https://bugs.webkit.org/show_bug.cgi?id=27833

Inherits RegExpConstructorPrivate class from FastAllocBase because it has been
instantiated by 'new' in JavaScriptCore/runtime/RegExpConstructor.cpp:152.

  • runtime/RegExpConstructor.cpp:
12:37 AM Changeset in webkit [46666] by abarth@webkit.org
  • 2 edits in trunk/WebCore

2009-08-01 John Abd-El-Malek <jam@chromium.org>

Reviewed by Adam Barth.

Fix a message port handle always getting marked as reachable once it gets entangled.

https://bugs.webkit.org/show_bug.cgi?id=27824

No tests since this only arises with the Chromium multi-process message port implementation
which isn't in this repostiory.

  • bindings/v8/V8GCController.cpp: (WebCore::GCEpilogueVisitor::visitDOMWrapper):
12:23 AM Changeset in webkit [46665] by abarth@webkit.org
  • 1 edit
    2 adds in trunk/LayoutTests

2009-08-01 Evan Martin <evan@chromium.org>

Reviewed by Adam Barth.

Add a regression test for a crash when a page has smart quotes in a
mime type.

  • fast/parser/smart-quotes-in-tag-expected.txt: Added.
  • fast/parser/smart-quotes-in-tag.html: Added.

Jul 31, 2009:

11:56 PM Changeset in webkit [46664] by abarth@webkit.org
  • 3 edits in trunk/WebKitTools

2009-07-31 Daniel Bates <dbates@intudata.com>

Reviewed by Eric Seidel.

https://bugs.webkit.org/show_bug.cgi?id=27757

Resolves Perl redefinition warnings about functions tmpfile and tmpnam which are
defined in both File::Temp and POSIX.

This stops the warnings by preventing the "import" method of File::Temp from being called,
which would otherwise add these functions into the primary namespace. Note, "use POSIX"
will import these functions into the primary namespace anyway. We only need one instance
of these functions in our namespace (or we get redefinition warnings), so lets use the
one's in the POSIX package.

  • Scripts/update-webkit-auxiliary-libs: Changed "use File::Temp" to "use File::Temp ()".
  • Scripts/update-webkit-support-libs: ditto
11:43 PM Changeset in webkit [46663] by abarth@webkit.org
  • 2 edits in trunk/WebCore

2009-07-31 Yong Li <yong.li@torchmobile.com>

Reviewed by Adam Barth.

Fix Geolocation permission problem
https://bugs.webkit.org/show_bug.cgi?id=26993

  • page/Geolocation.cpp: (WebCore::Geolocation::requestPermission):
11:29 PM Changeset in webkit [46662] by abarth@webkit.org
  • 2 edits in trunk/WebKitTools

2009-07-31 Laszlo Gombos <Laszlo Gombos>

Reviewed by Eric Seidel.

Add a top level build option to control datagrid support
https://bugs.webkit.org/show_bug.cgi?id=27915

  • Scripts/build-webkit:
11:09 PM Changeset in webkit [46661] by abarth@webkit.org
  • 2 edits in trunk/WebCore

2009-07-31 Norbert Leser <norbert.leser@nokia.com>

Reviewed by Eric Seidel.

Added project properties as needed by symbian platform

  • WebCore.pro:
10:26 PM Changeset in webkit [46660] by abarth@webkit.org
  • 3 edits in trunk/JavaScriptCore

2009-07-31 Yong Li <yong.li@torchmobile.com>

Reviewed by George Staikos.

Resurrect the old GetTickCount implementation of currentTime, controlled by WTF_USE_QUERY_PERFORMANCE_COUNTER
currentSystemTime taken from older WebKit; currentTime written by Yong Li <yong.li@torchmobile.com>; cleanup by Joe Mason <joe.mason@torchmobile.com>
https://bugs.webkit.org/show_bug.cgi?id=27848

  • wtf/CurrentTime.cpp: (WTF::currentSystemTime): get current time with GetCurrentFT (WTF::currentTime): track msec elapsed since first currentSystemTime call using GetTickCount
  • wtf/Platform.h:
9:07 PM Changeset in webkit [46659] by xan@webkit.org
  • 2 edits in trunk/WebCore

2009-07-31 Xan Lopez <xlopez@igalia.com>

Reviewed by Holger Freyther.

Fix compiler warning.

Initialize member variables in the correct order.

  • plugins/PluginView.cpp: (WebCore::PluginView::PluginView):
9:07 PM Changeset in webkit [46658] by treat@webkit.org
  • 2 edits in trunk/LayoutTests

2009-07-31 Adam Treat <adam.treat@torchmobile.com>

Reviewed by Holger Hans Peter Freyther.

The reason these four tests fail is a mystery to me right, but placing
them in the Skipped file and the Qt port is... for the first time ever,
finally green!

https://bugs.webkit.org/show_bug.cgi?id=27914

  • platform/qt/Skipped:
8:43 PM Changeset in webkit [46657] by treat@webkit.org
  • 2 edits in trunk/LayoutTests

2009-07-31 Adam Treat <adam.treat@torchmobile.com>

Reviewed by Eric Seidel.

Printing related test needs to go in Skipped for now so we can
get Qt buildbot green for first time.

https://bugs.webkit.org/show_bug.cgi?id=27913

  • platform/qt/Skipped:
8:43 PM Changeset in webkit [46656] by treat@webkit.org
  • 2 edits in trunk/LayoutTests

2009-07-31 Adam Treat <adam.treat@torchmobile.com>

Reviewed by Eric Seidel.

Sometimes they pass, sometimes they fail. In attempt to get Qt buildbot
completely green for first time I'm placing these two in the Skipped file.

https://bugs.webkit.org/show_bug.cgi?id=27912

  • platform/qt/Skipped:
8:43 PM Changeset in webkit [46655] by treat@webkit.org
  • 2 edits in trunk/LayoutTests

2009-07-31 Adam Treat <adam.treat@torchmobile.com>

Reviewed by George Staikos and Eric Seidel.

Place some audio related tests into the Skipped list in
attempt to get Qt buildbot finally green.

https://bugs.webkit.org/show_bug.cgi?id=27910

  • platform/qt/Skipped:
7:58 PM Changeset in webkit [46654] by jhoneycutt@apple.com
  • 2 edits in trunk/WebCore

2009-07-31 Jon Honeycutt <jhoneycutt@apple.com>

Windows build fix!

Unreviewed.

  • page/win/PageWin.cpp:
7:38 PM Changeset in webkit [46653] by jhoneycutt@apple.com
  • 3 edits in trunk/WebCore

2009-07-31 Jon Honeycutt <jhoneycutt@apple.com>

Mac build fix.

Unreviewed.

  • page/Page.cpp: Move setCanStartPlugins() from here...
  • page/win/PageWin.cpp: (WebCore::Page::setCanStartPlugins): ... to here.
7:33 PM Changeset in webkit [46652] by levin@chromium.org
  • 3 edits in trunk/WebKitTools

2009-07-31 David Levin <levin@chromium.org>

Reviewed by Adam Treat.

check-webkit-style --git-commit has bugs if you select a commit in the past
https://bugs.webkit.org/show_bug.cgi?id=27908

  • Scripts/check-webkit-style: (main): Made the git-commit command only select the first commit if a range is given (and print a warning).
  • Scripts/modules/scm.py: (SCM::create_patch_from_local_commit): (SCM::create_patch_since_local_commit): (SCM::commit_locally_with_message): (SCM::discard_local_commits): Added an error messages in case these aren't overriden.

(Git::create_patch_since_local_commit):
Added this command to allow doing a diff of everything after a given commit.

6:40 PM Changeset in webkit [46651] by jmalonzo@webkit.org
  • 2 edits in trunk/WebKitTools

2009-07-31 Jan Michael Alonzo <jmalonzo@webkit.org>

Speculative Qt build fix. Not reviewed.

Add LayoutTestController::disableImageLoading() in jsobjects.h.

  • DumpRenderTree/qt/jsobjects.h:
6:34 PM Changeset in webkit [46650] by jmalonzo@webkit.org
  • 2 edits in trunk/LayoutTests

2009-07-31 Jan Michael Alonzo <jmalonzo@webkit.org>

Skip http/tests/misc/favicon-loads-with-images-disabled.html as
this requires resource load callbacks in Gtk DRT. I've raised bug
#27905 to track this.

  • platform/gtk/Skipped:
6:13 PM Changeset in webkit [46649] by jhoneycutt@apple.com
  • 16 edits in trunk

WebCore:

2009-07-29 Jon Honeycutt <jhoneycutt@apple.com>

<rdar://problem/5698113> Safari shouldn't auto-activate plug-ins in
background tabs (make Win consistent with Mac)

https://bugs.webkit.org/show_bug.cgi?id=27855

Reviewed by Anders Carlsson.

  • loader/FrameLoaderClient.h: (WebCore::FrameLoaderClient::dispatchDidFailToStartPlugin): Declare a new function to dispatch failures to start plug-ins.
  • page/Page.cpp: (WebCore::Page::Page): Initialize new member variable. (WebCore::Page::addUnstartedPlugin): Add the PluginView to the set of unstarted plug-ins. (WebCore::Page::removeUnstartedPlugin): Remove the PluginView from the set of unstarted plug-ins. (WebCore::Page::setCanStartPlugins): If we can now start plug-ins, iterate the set of unstarted plug-ins, starting them. If a plug-in fails to start, dispatch a failed-to-start- plug-in error. Clear the list of unstarted plug-ins.
  • page/Page.h: Added m_canStartPlugins to track whether we can currently start plug-ins. Added m_unstartedPlugins to keep track of plug-ins in the page that are waiting to be started. (WebCore::Page::canStartPlugins):
  • plugins/PluginView.cpp: (WebCore::PluginView::start): m_mimeType is now a WebCore::String, so we call .utf8(). If NPP_New() fails, set m_status to reflect this; this used to be set by init(), but start() will not be called by init() if we are delaying start(). If NPP_New() succeeds, set m_status to reflect this, and call platformStart(). (WebCore::PluginView::startOrAddToUnstartedList): If we cannot start plug-ins, add this view to the Page's list of unstarted plug-ins and return true. Otherwise, call start() and return the result. (WebCore::PluginView::removeFromUnstartedListIfNecessary): If we've been started or we don't have a Page, return early. Remove ourselves from the Page's list of unstarted plug-ins. (WebCore::PluginView::bindingInstance): If we haven't been started yet, return 0. This matches the Mac behavior. (WebCore::PluginView::setParameters): If we find the "plug-ins page" parameter, record it; it may be used if we fail to start the plug-in. (WebCore::PluginView::PluginView): Initialize new members m_haveUpdatedPluginWidget. Initialize m_mimeType with mimeType directly; m_mimeType is now a WebCore::String.
  • plugins/PluginView.h: Made start() public, so Page can call it when we start unstarted plug-ins. Changed m_mimeType from CString to String and added m_pluginsPage; these are needed by dispatchDidFailToStartPlugin(). (WebCore::PluginView::pluginsPage): Added for dispatchDidFailToStartPlugin(). (WebCore::PluginView::mimeType): Ditto. (WebCore::PluginView::url): Ditto.
  • plugins/gtk/PluginViewGtk.cpp: (WebCore::PluginView::platformStart): Stubbed.
  • plugins/mac/PluginViewMac.cpp: (WebCore::PluginView::platformStart): Stubbed.
  • plugins/qt/PluginViewQt.cpp: (WebCore::PluginView::platformStart): Stubbed.
  • plugins/win/PluginViewWin.cpp: (WebCore::PluginView::updatePluginWidget): Use m_haveUpdatedPluginWidget to ensure that we will call MoveWindow() and SetWindowRgn() to position the plug-in and set its clip rect if an earlier call to updatePluginWidget() was made before setPlatformPluginWidget() was called to set the plug-in's HWND. (WebCore::PluginView::~PluginView): Call removeFromUnstartedListIfNecessary() to remove the plug-in from the unstarted list if the plug-in is destroyed before it is started. (WebCore::PluginView::init): Call startOrAddtoUnstartedList to start the plug-in or, if plug-ins cannot be started, to add it to the list of plug-ins waiting to be started. Set m_status to indicate success; this is also set by start(), but that may be delayed, and we don't want the loader to dispatch a failed-to-start-plug-in error when we return from init(). (WebCore::PluginView::platformStart): Added; code moved from init(). This is now called by start() to perform any platform-specific tasks that need to take place after the plug-in has started. Added a call to updatePluginWidget() to ensure we position the plug-in and set its clip rect after the plug-in has started. Removed the setting of m_status; start() does this before calling this function (and we assert this at the head of the function).

WebKit/win:

2009-07-29 Jon Honeycutt <jhoneycutt@apple.com>

Part of <rdar://problem/5698113> Safari shouldn't auto-activate
plug-ins in background tabs (make Win consistent with Mac)

https://bugs.webkit.org/show_bug.cgi?id=27855

Reviewed by Adam Roben and Anders Carlsson.

  • Interfaces/IWebViewPrivate.idl: Add setCanStartPlugins().
  • WebCoreSupport/WebFrameLoaderClient.cpp: (WebFrameLoaderClient::dispatchDidFailToStartPlugin): Added; code moved from createPlugin(). Modified to obtain the plug-in's page URL, the plug-ins "plug-ins page" parameter value, and the plug-in's MIME type by calling getter functions on the passed PluginView. (WebFrameLoaderClient::createPlugin): If the plug-in status is not "success," call dispatchDidFailToStartPlugin().
  • WebCoreSupport/WebFrameLoaderClient.h: Declare an override of dispatchDidFailToStartPlugin().
  • WebView.cpp: (WebViewWndProc): Add a missing ) to a comment. (WebView::setCanStartPlugins): Call the Page's setCanStartPlugins() function.
  • WebView.h:
5:50 PM Changeset in webkit [46648] by adachan@apple.com
  • 3 edits in trunk/JavaScriptCore

Fixes the Windows release-PGO build.

Reviewed by Jon Honeycutt.

  • JavaScriptCore.vcproj/WTF/WTF.vcproj: Suppresses the warning about unreachable code that we get by adding "return 0" to WTF::TCMalloc_PageHeap::runScavengerThread().
  • wtf/FastMalloc.cpp: (WTF::TCMalloc_PageHeap::runScavengerThread): Fixes the error about the method not returning a value in the release-PGO build.
5:40 PM Changeset in webkit [46647] by Darin Adler
  • 42 edits in trunk

WebCore:

2009-07-30 Darin Adler <Darin Adler>

Reviewed by David Levin.

Use checked casts for render tree
https://bugs.webkit.org/show_bug.cgi?id=23522

Next step: Add casts for all the RenderTable classes and use them everywhere.

Also added a few uses in places that were still using static_cast.
Also made some virtual member functions private.
Also changed RenderTable::m_tableLayout to use OwnPtr.

  • accessibility/AccessibilityTable.cpp: (WebCore::AccessibilityTable::isTableExposableThroughAccessibility): (WebCore::AccessibilityTable::addChildren): (WebCore::AccessibilityTable::cellForColumnAndRow):
  • accessibility/AccessibilityTableCell.cpp: (WebCore::AccessibilityTableCell::parentTable): (WebCore::AccessibilityTableCell::rowIndexRange): (WebCore::AccessibilityTableCell::columnIndexRange): (WebCore::AccessibilityTableCell::titleUIElement):
  • accessibility/AccessibilityTableColumn.cpp: (WebCore::AccessibilityTableColumn::headerObject):
  • accessibility/AccessibilityTableRow.cpp: (WebCore::AccessibilityTableRow::parentTable):
  • editing/DeleteSelectionCommand.cpp: (WebCore::DeleteSelectionCommand::removeNode):
  • editing/TextIterator.cpp: (WebCore::shouldEmitTabBeforeNode): (WebCore::shouldEmitNewlinesBeforeAndAfterNode):
  • html/HTMLTableCellElement.cpp: (WebCore::HTMLTableCellElement::parseMappedAttribute):
  • html/HTMLTableColElement.cpp: (WebCore::HTMLTableColElement::parseMappedAttribute):
  • page/Frame.cpp: (WebCore::Frame::searchForLabelsAboveCell):
  • page/mac/FrameMac.mm: (WebCore::Frame::searchForNSLabelsAboveCell):
  • rendering/AutoTableLayout.cpp: (WebCore::AutoTableLayout::recalcColumn): (WebCore::AutoTableLayout::fullRecalc): (WebCore::shouldScaleColumns):
  • rendering/FixedTableLayout.cpp: (WebCore::FixedTableLayout::calcWidthArray):
  • rendering/RenderBlock.cpp: (WebCore::RenderBlock::calcPrefWidths):
  • rendering/RenderBox.cpp: (WebCore::RenderBox::calcPercentageHeight):
  • rendering/RenderObject.cpp: (WebCore::RenderObject::addChild): (WebCore::RenderObject::containingBlock):
  • rendering/RenderTableCell.cpp: (WebCore::RenderTableCell::collapsedBottomBorder):
  • rendering/RenderTableCol.cpp: (WebCore::RenderTableCol::table):
  • rendering/RenderTableRow.cpp: (WebCore::RenderTableRow::addChild): (WebCore::RenderTableRow::layout): (WebCore::RenderTableRow::paint):
  • rendering/RenderTableSection.cpp: (WebCore::RenderTableSection::addChild): (WebCore::RenderTableSection::layoutRows): (WebCore::RenderTableSection::lowestPosition): (WebCore::RenderTableSection::rightmostPosition): (WebCore::RenderTableSection::leftmostPosition): (WebCore::RenderTableSection::paintObject): (WebCore::RenderTableSection::recalcCells):
  • rendering/RenderTreeAsText.cpp: (WebCore::operator<<): (WebCore::writeTextRun): Use checked casts.
  • rendering/RenderTable.cpp: (WebCore::RenderTable::styleDidChange): Updated to use OwnPtr. (WebCore::RenderTable::addChild): Use checked cast. (WebCore::RenderTable::layout): Ditto. (WebCore::RenderTable::setCellWidths): Ditto. (WebCore::RenderTable::paintObject): Ditto. (WebCore::RenderTable::splitColumn): Ditto. (WebCore::RenderTable::appendColumn): Ditto. (WebCore::RenderTable::colElement): Ditto. (WebCore::RenderTable::recalcSections): Ditto. (WebCore::RenderTable::outerBorderBottom): Ditto. (WebCore::RenderTable::outerBorderLeft): Ditto. (WebCore::RenderTable::outerBorderRight): Ditto. (WebCore::RenderTable::sectionAbove): Ditto. (WebCore::RenderTable::sectionBelow): Ditto.
  • rendering/RenderTable.h: Added checked cast. Made virtual functions private. Changed m_tableLayout to be a OwnPtr.
  • rendering/RenderTableCell.h: Added checked cast.
  • rendering/RenderTableCol.h: Ditto. Made virtual functions private.
  • rendering/RenderTableRow.h: Ditto.
  • rendering/RenderTableSection.h: Ditto.

WebKit/mac:

2009-07-30 Darin Adler <Darin Adler>

Reviewed by David Levin.

Use checked casts for render tree
https://bugs.webkit.org/show_bug.cgi?id=23522

  • Misc/WebNSAttributedStringExtras.mm: (fileWrapperForElement):
  • Misc/WebNSPasteboardExtras.mm: (imageFromElement): (-[NSPasteboard _web_declareAndWriteDragImageForElement:URL:title:archive:source:]):
  • WebView/WebFrame.mm: (-[WebFrame _computePageRectsWithPrintWidthScaleFactor:printHeight:]): (-[WebFrame _accessibilityTree]):
  • WebView/WebRenderNode.mm: (copyRenderNode): Use checked casts.

WebKit/win:

2009-07-30 Darin Adler <Darin Adler>

Reviewed by David Levin.

Use checked casts for render tree
https://bugs.webkit.org/show_bug.cgi?id=23522

  • DOMHTMLClasses.cpp: (DOMHTMLInputElement::isUserEdited): (DOMHTMLTextAreaElement::isUserEdited): Use checked casts.
4:27 PM UpdatingBugzilla edited by ddkilzer@apple.com
(diff)
4:27 PM UpdatingBugzilla edited by ddkilzer@apple.com
(diff)
4:22 PM Changeset in webkit [46646] by beidson@apple.com
  • 12 edits
    2 adds in trunk

WebCore:

2009-07-31 Brady Eidson <beidson@apple.com>

Reviewed by John Sullivan.

<rdar://problem/6973106> and https://bugs.webkit.org/show_bug.cgi?id=27896
Favicons are still loaded when automatic image loading is disabled.

People who want to avoid loading images generally want to avoid loading all images.

Test: http/tests/misc/favicon-loads-with-images-disabled.html

  • loader/FrameLoader.cpp: (WebCore::FrameLoader::startIconLoader): After committing the URL mapping, don't actually perform the load if images shouldn't be loading.

WebKitTools:

2009-07-31 Brady Eidson <beidson@apple.com>

Reviewed by John Sullivan.

<rdar://problem/6973106> and https://bugs.webkit.org/show_bug.cgi?id=27896
Favicons are still loaded when automatic image loading is disabled.

  • DumpRenderTree/LayoutTestController.cpp: (disableImageLoadingCallback): (LayoutTestController::staticFunctions):
  • DumpRenderTree/LayoutTestController.h:
  • DumpRenderTree/mac/DumpRenderTree.mm: (resetWebViewToConsistentStateBeforeTesting):
  • DumpRenderTree/mac/LayoutTestControllerMac.mm: (LayoutTestController::disableImageLoading):
  • DumpRenderTree/win/DumpRenderTree.cpp: (resetWebViewToConsistentStateBeforeTesting):
  • DumpRenderTree/win/LayoutTestControllerWin.cpp: (LayoutTestController::disableImageLoading):
  • DumpRenderTree/gtk/LayoutTestControllerGtk.cpp: (LayoutTestController::disableImageLoading): Stubbed for now.
  • DumpRenderTree/qt/jsobjects.cpp: (LayoutTestController::disableImageLoading): Ditto.

LayoutTests:

2009-07-31 Brady Eidson <beidson@apple.com>

Reviewed by John Sullivan.

<rdar://problem/6973106> and https://bugs.webkit.org/show_bug.cgi?id=27896
Favicons are still loaded when automatic image loading is disabled.

  • http/tests/misc/favicon-loads-with-images-disabled-expected.txt: Added.
  • http/tests/misc/favicon-loads-with-images-disabled.html: Added.
4:12 PM UpdatingBugzilla edited by ddkilzer@apple.com
(diff)
4:03 PM Changeset in webkit [46645] by Simon Fraser
  • 24 edits in trunk

2009-07-31 Simon Fraser <Simon Fraser>

Reviewed by Anders Carlsson.

Accelerated animations stutter on pages with lots of animations and 3d transforms
https://bugs.webkit.org/show_bug.cgi?id=27884

This patch changes the strategy for synchronizing painting view the view,
and compositing layer updates. Previously the strategy was to disable screen
updates between the time we updated the layer tree, and painted the view. That
left screen updates disabled for too long (hundreds of milliseconds) in some
cases, causing animation stutter.

The new strategy is to batch up changes to the CA layer tree, and commit them
all at once just before painting happens (referred to as a "sync" in the code).
GraphicsLayerCA now keeps a bitmask of changed properties, and then migrates
the values stored in GraphicsLayer into the CA layer tree at commit time.

Compositing layers are then synced in FrameView::paintContents(). However, not
all style/layout changes will result in painting; to deal with style changes that
touch only compositing properties, we set up a runloop observer that takes care
of comitting layer changes when no painting happens.

  • WebCore.base.exp: Export FrameView::syncCompositingStateRecursive()
  • loader/EmptyClients.h: scheduleViewUpdate() renamed to syncCompositingStateRecursive()
  • page/ChromeClient.h: scheduleViewUpdate() renamed to syncCompositingStateRecursive()
  • page/FrameView.h:
  • page/FrameView.cpp: (WebCore::FrameView::syncCompositingStateRecursive): syncCompositingState() on the view and all subviews. Like layoutIfNeededRecursive(). If layout is pending, does not sync and returns false, since we only want to sync when layout is done.

(WebCore::FrameView::paintContents): syncCompositingState() before painting.

  • page/animation/KeyframeAnimation.cpp: (WebCore::KeyframeAnimation::endAnimation): Call animationPaused() to notify the graphics layers about animation pausing.
  • platform/graphics/FloatPoint3D.h: (WebCore::operator==): (WebCore::operator!=): Add missing comparison operators.
  • platform/graphics/GraphicsLayer.cpp: (WebCore::GraphicsLayer::setOpacity): (WebCore::GraphicsLayer::setBackgroundColor): Simple setters no longer care about animation info.

(WebCore::GraphicsLayer::paintGraphicsLayerContents): Null-check client.

  • platform/graphics/GraphicsLayer.h: (WebCore::AnimationValue:): (WebCore::TransformAnimationValue:): (WebCore::KeyframeValueList:): (WebCore::KeyframeValueList::insert): Cleaned up versions of FloatValue and TransformValue, used to store information about keyframes values.

(WebCore::GraphicsLayer::contentsRect):
(WebCore::GraphicsLayer::setContentsRect):
ContentsRect is now a simple setter.

(WebCore::GraphicsLayer::addAnimation):
(WebCore::GraphicsLayer::removeAnimationsForProperty):
(WebCore::GraphicsLayer::removeAnimationsForKeyframes):
(WebCore::GraphicsLayer::pauseAnimation):
Simplified animation api.

(WebCore::GraphicsLayer::setGeometryOrientation):
(WebCore::GraphicsLayer::geometryOrientation):
setGeometryOrientation is now just a normal member variable.

(WebCore::GraphicsLayer::contentsOrientation): add a getter.
(WebCore::GraphicsLayer::syncCompositingState): Entry into the layer sync code.

  • platform/graphics/GraphicsLayerClient.h: scheduleViewUpdate() renamed to syncCompositingStateRecursive)
  • platform/graphics/mac/GraphicsLayerCA.h:
  • platform/graphics/mac/GraphicsLayerCA.mm: Lots of cleanup and refactoring. Main points:
  • Layer changes are all batched, and only committed to the CA layer on syncCompositingState().
  • Bitmask is used to store which properties have changed. More efficient than before.
  • Simpler animation interface; simple setters are no longer confounded with animations.
  • Refactored code that creates CA animations, and stores which animations are running.
  • platform/graphics/transforms/TransformOperations.h: (WebCore::TransformOperations::size): (WebCore::TransformOperations::at): Useful accessors for size and indexed access.
  • rendering/RenderLayerBacking.h:
  • rendering/RenderLayerBacking.cpp: Renamed 'contentsLayer' to 'foregroundLayer' to avoid confusion with GraphicsLayer's contentsLayer. Adapt to GraphicsLayer's simpler animation API. Pass animation pausing through to the graphics layers. contentsBox() is no longer a callback via GraphicsLayerClient.
  • rendering/RenderLayerCompositor.h:
  • rendering/RenderLayerCompositor.cpp: (WebCore::RenderLayerCompositor::setCompositingLayersNeedRebuild): (WebCore::RenderLayerCompositor::scheduleSync): (WebCore::RenderLayerCompositor::rebuildCompositingLayerTree): scheduleViewUpdate() is no longer required. Instead, we plumb through "compositingLayerSync" notifications, which travel up to WebKit and set up a runloop observer.
3:59 PM UpdatingBugzilla created by ddkilzer@apple.com
Initial contents
3:58 PM WikiStart edited by ddkilzer@apple.com
(diff)
3:46 PM Changeset in webkit [46644] by eric@webkit.org
  • 3 edits in trunk/WebKitTools

2009-07-30 Eric Seidel <eric@webkit.org>

Reviewed by David Kilzer.

bugzilla-tool post-commits trunk..head errors out
https://bugs.webkit.org/show_bug.cgi?id=27847

Rename commit_ids_from_range_arguments to commit_ids_from_commitish_arguments
and make it handle the new arguments format proposed in the bug.

  • Scripts/bugzilla-tool:
  • Scripts/modules/scm.py:
3:29 PM Changeset in webkit [46643] by eric@webkit.org
  • 2 edits in trunk/JavaScriptCore

2009-07-31 Zoltan Horvath <hzoltan@inf.u-szeged.hu>

Change malloc to fastMalloc and free to fastFree in Yarr's RegexInterpreter.cpp
https://bugs.webkit.org/show_bug.cgi?id=27874

Use fastMalloc and fastFree instead of malloc and free in RegexInterpreter.cpp's methods.

  • yarr/RegexInterpreter.cpp: (JSC::Yarr::Interpreter::allocDisjunctionContext): (JSC::Yarr::Interpreter::freeDisjunctionContext): (JSC::Yarr::Interpreter::allocParenthesesDisjunctionContext): (JSC::Yarr::Interpreter::freeParenthesesDisjunctionContext):
2:57 PM Changeset in webkit [46642] by bolsinga@apple.com
  • 2 edits in trunk/WebCore

2009-07-31 Greg Bolsinga <bolsinga@apple.com>

Reviewed by Eric Seidel.

Geolocation clean up when no longer updating
https://bugs.webkit.org/show_bug.cgi?id=27888

When Geolocation::disconnectFrame() is called, need to call
Document::setUsingGeolocation(false) to mirror the true call
when it is set up.

When handling an error, and there are no more listeners, call
GeolocationService::stopUpdating().

  • page/Geolocation.cpp: (WebCore::Geolocation::disconnectFrame): (WebCore::Geolocation::handleError):
2:14 PM Changeset in webkit [46641] by eric@webkit.org
  • 2 edits in trunk/WebKitTools

2009-07-30 Eric Seidel <eric@webkit.org>

Reviewed by David Levin.

clean up build-webkit to use less copy/paste code and remove unneeded SVG warning
https://bugs.webkit.org/show_bug.cgi?id=27857

  • Scripts/build-webkit:
2:02 PM Changeset in webkit [46640] by ddkilzer@apple.com
  • 2 edits in trunk/WebKitTools

<http://webkit.org/b/27894> bugzilla-tool: Bugzilla.clear_attachment_review_flag() fails with ClientForm-0.2.10 and python-2.5

Reviewed by Eric Seidel.

  • Scripts/modules/bugzilla.py: (Bugzilla.clear_attachment_review_flag): Specify which 'comment' form field to use when setting its value.
1:29 PM Changeset in webkit [46639] by jorlow@chromium.org
  • 1 edit
    6 adds in trunk/LayoutTests

2009-07-31 Jeremy Orlow <jorlow@chromium.org>

Reviewed by Adam Barth.

Add new LocalStorage layout tests.
https://bugs.webkit.org/show_bug.cgi?id=27778

These test behavior that I don't think was very well verified earlier.

  • storage/domstorage/localstorage/complex-keys-expected.txt: Added.
  • storage/domstorage/localstorage/complex-keys.html: Added.

A bunch of corner cases related to keys.

  • storage/domstorage/localstorage/complex-values-expected.txt: Added.
  • storage/domstorage/localstorage/complex-values.html: Added.

A bunch of corner cases related to values.

  • storage/domstorage/localstorage/remove-item-expected.txt: Added.
  • storage/domstorage/localstorage/remove-item.html: Added.

Test basic removeItem functionality.

1:21 PM Changeset in webkit [46638] by andersca@apple.com
  • 3 edits in trunk/WebKit/win

2009-07-31 Anders Carlsson <andersca@apple.com>

Reviewed by Dan Bernstein.

-Fix <rdar://problem/7104097>

Make EmbeddedWidget::create return a PassRefPtr so it will have the correct ref count.


  • WebCoreSupport/EmbeddedWidget.cpp: (EmbeddedWidget::create):
  • WebCoreSupport/EmbeddedWidget.h:
1:12 PM Changeset in webkit [46637] by ddkilzer@apple.com
  • 2 edits in trunk/WebKitTools

2009-07-31 Eric Seidel <eric@webkit.org>

Reviewed by David Kilzer.

bugzilla-tool post-commits obsoletes its own work :(
https://bugs.webkit.org/show_bug.cgi?id=27849

Simple fix. Use .add() instead of .update()

  • Scripts/bugzilla-tool:
11:40 AM Changeset in webkit [46636] by jorlow@chromium.org
  • 2 edits in trunk/WebCore

2009-07-30 Jeremy Orlow <jorlow@chromium.org>

Reviewed by David Levin.

Guard needs 's between #endif and ENABLE(DATAGRID)
https://bugs.webkit.org/show_bug.cgi?id=27862

Compiling this file causes a build break without this change.

  • bindings/v8/custom/V8DataGridColumnListCustom.cpp:
11:16 AM Changeset in webkit [46635] by treat@webkit.org
  • 2 edits in trunk/LayoutTests

2009-07-31 Mike Fenton <mike.fenton@torchmobile.com>

Reviewed by Adam Treat.

Re-Enable tests from Skipped that are passing correctly.

https://bugs.webkit.org/show_bug.cgi?id=27890

  • platform/qt/Skipped:
10:07 AM Changeset in webkit [46634] by kenneth@webkit.org
  • 1 edit
    45 adds in trunk/LayoutTests

2009-07-31 Mike Fenton <mike.fenton@torchmobile.com>

Reviewed by Adam Treat.

Add LayoutTest results from BuildBot output for tests that currently lack results under Qt.

https://bugs.webkit.org/show_bug.cgi?id=27882

  • platform/qt/fast/backgrounds/size/backgroundSize20-expected.txt: Copied from LayoutTests/platform/mac/fast/backgrounds/size/backgroundSize20-expected.txt.
  • platform/qt/fast/backgrounds/size/backgroundSize21-expected.txt: Copied from LayoutTests/platform/mac/fast/backgrounds/size/backgroundSize21-expected.txt.
  • platform/qt/fast/backgrounds/size/backgroundSize22-expected.txt: Copied from LayoutTests/platform/mac/fast/backgrounds/size/backgroundSize22-expected.txt.
  • platform/qt/fast/borders/border-radius-constraints-expected.txt: Added.
  • platform/qt/fast/box-shadow/inset-expected.txt: Added.
  • platform/qt/fast/box-shadow/spread-expected.txt: Copied from LayoutTests/platform/mac/fast/box-shadow/spread-expected.txt.
  • platform/qt/fast/canvas/image-pattern-rotate-expected.txt: Added.
  • platform/qt/fast/css/nested-floating-relative-position-percentages-expected.txt: Added.
  • platform/qt/fast/css/pseudo-cache-stale-expected.txt: Added.
  • platform/qt/fast/css/rem-dynamic-scaling-expected.txt: Added.
  • platform/qt/fast/css/rem-units-on-root-expected.txt: Added.
  • platform/qt/fast/dynamic/window-resize-scrollbars-test-expected.txt: Added.
  • platform/qt/fast/dynamic/window-scrollbars-test-expected.txt: Added.
  • platform/qt/fast/forms/isindex-placeholder-expected.txt: Added.
  • platform/qt/fast/forms/search-vertical-alignment-expected.txt: Added.
  • platform/qt/fast/frames/content-opacity-1-expected.txt: Copied from LayoutTests/platform/mac/fast/frames/content-opacity-1-expected.txt.
  • platform/qt/fast/frames/content-opacity-2-expected.txt: Copied from LayoutTests/platform/mac/fast/frames/content-opacity-2-expected.txt.
  • platform/qt/fast/frames/frame-navigation-expected.txt: Added.
  • platform/qt/fast/frames/frame-scrolling-attribute-expected.txt: Added.
  • platform/qt/fast/frames/iframe-scrolling-attribute-expected.txt: Added.
  • platform/qt/fast/gradients/crash-on-zero-radius-expected.txt: Added.
  • platform/qt/fast/lists/ol-start-parsing-expected.txt: Added.
  • platform/qt/fast/overflow/overflow-with-local-background-attachment-expected.txt: Added.
  • platform/qt/fast/repaint/float-new-in-block-expected.txt: Copied from LayoutTests/platform/mac/fast/repaint/float-new-in-block-expected.txt.
  • platform/qt/fast/text/international/khmer-selection-expected.txt: Added.
  • platform/qt/fast/text/shadow-translucent-fill-expected.txt: Added.
  • platform/qt/fast/text/text-letter-spacing-expected.txt: Added.
  • platform/qt/plugins/netscape-dom-access-expected.txt: Added.
  • platform/qt/svg/batik/filters/feTile-expected.txt: Added.
  • platform/qt/svg/custom/createImageElement-expected.txt: Copied from LayoutTests/platform/mac/svg/custom/createImageElement-expected.txt.
  • platform/qt/svg/custom/createImageElement2-expected.txt: Added.
  • platform/qt/svg/custom/js-update-image-and-display-expected.txt: Copied from LayoutTests/platform/mac/svg/custom/js-update-image-and-display-expected.txt.
  • platform/qt/svg/custom/js-update-image-and-display2-expected.txt: Copied from LayoutTests/platform/mac/svg/custom/js-update-image-and-display2-expected.txt.
  • platform/qt/svg/custom/js-update-image-and-display3-expected.txt: Copied from LayoutTests/platform/mac/svg/custom/js-update-image-and-display3-expected.txt.
  • platform/qt/svg/custom/outer-svg-unknown-feature-expected.txt: Copied from LayoutTests/platform/mac/svg/custom/outer-svg-unknown-feature-expected.txt.
  • platform/qt/svg/filters/feMerge-expected.txt: Added.
  • platform/qt/svg/filters/feOffset-expected.txt: Added.
  • platform/qt/svg/filters/feTile-expected.txt: Added.
  • platform/qt/svg/filters/subRegion-one-effect-expected.txt: Added.
  • platform/qt/svg/filters/subRegion-two-effects-expected.txt: Added.
10:04 AM Changeset in webkit [46633] by bolsinga@apple.com
  • 3 edits in trunk/WebCore

2009-07-30 Greg Bolsinga <bolsinga@apple.com>

Reviewed by George Staikos.

Consolidate GeoLocation code to send positions and errors
https://bugs.webkit.org/show_bug.cgi?id=27863

Create helper methods that will send positions and errors to either
one shots or watchers.

  • page/Geolocation.cpp: (WebCore::Geolocation::sendError): (WebCore::Geolocation::sendErrorToOneShots): (WebCore::Geolocation::sendErrorToWatchers): (WebCore::Geolocation::sendPosition): (WebCore::Geolocation::sendPositionToOneShots): (WebCore::Geolocation::sendPositionToWatchers):
  • page/Geolocation.h:
9:52 AM Changeset in webkit [46632] by kenneth@webkit.org
  • 2 edits in trunk/WebKit/qt

2009-07-31 Jakob Petsovits <jakob.petsovits@torchmobile.com>

Reviewed by Adam Treat.

Build QtWebKit without benchmark tests for Qt < 4.5.
https://bugs.webkit.org/show_bug.cgi?id=27886

The QBENCHMARK macro is only available from 4.5 on,
so builds for Qt < 4.5 shouldn't try to build those.

  • tests/tests.pro:
7:56 AM Changeset in webkit [46631] by vestbo@webkit.org
  • 2 edits in trunk/WebKitTools

2009-07-31 Tor Arne Vestbø <tor.arne.vestbo@nokia.com>

Reviewed by Anders Carlsson.

Improve platform-detection in run-webkit-tests

This prevents miss-detection of mac-* over Qt/Mac.

  • Scripts/run-webkit-tests:
7:37 AM Changeset in webkit [46630] by kenneth@webkit.org
  • 2 edits in trunk/WebKit/qt

2009-07-31 Kenneth Rohde Christiansen <kenneth@webkit.org>

Reviewed by Adam Treat.

Fix warning by migrating use of QString.sprintf(...) to
QString.arg().arg()...

  • WebCoreSupport/EditorClientQt.cpp: (dumpRange):
6:51 AM Changeset in webkit [46629] by treat@webkit.org
  • 3 edits in trunk/LayoutTests

2009-07-31 Adam Treat <adam.treat@torchmobile.com>

Reviewed by Tor Arne Vestbo.

Adding -webkit-animation-play-state back in following r46547.

  • platform/qt/fast/css/getComputedStyle/computed-style-expected.txt:
  • platform/qt/fast/css/getComputedStyle/computed-style-without-renderer-expected.txt:
6:29 AM Changeset in webkit [46628] by kenneth@webkit.org
  • 3 edits in trunk/WebKit/qt

2009-07-31 Kenneth Rohde Christiansen <kenneth@webkit.org>

Reviewed by Tor Arne Vestbø.

Expose the Offline Web Application Cache in the Qt API.

  • Api/qwebsettings.cpp: (QWebSettings::QWebSettings): (QWebSettings::setOfflineWebApplicationCachePath): (QWebSettings::offlineWebApplicationCachePath): (QWebSettings::setOfflineWebApplicationCacheQuota): (QWebSettings::offlineWebApplicationCacheQuota):
  • Api/qwebsettings.h:
5:43 AM Changeset in webkit [46627] by xan@webkit.org
  • 2 edits in trunk/WebCore

2009-07-31 Xan Lopez <xlopez@igalia.com>

Roll out previous change as it might be causing some weirdness in
the bots.

  • platform/image-decoders/bmp/BMPImageDecoder.cpp: (WebCore::BMPImageDecoder::processFileHeader):
5:06 AM Changeset in webkit [46626] by xan@webkit.org
  • 2 edits in trunk/WebCore

2009-07-31 Xan Lopez <xlopez@igalia.com>

Reviewed by Mark Rowe.

Fix compiler warning.
https://bugs.webkit.org/show_bug.cgi?id=27851

GCC does not like multi-character character constants, so use the
explicit numerical value of 'BM' in the enum.

  • platform/image-decoders/bmp/BMPImageDecoder.cpp: (WebCore::BMPImageDecoder::processFileHeader):
12:21 AM Changeset in webkit [46625] by mrowe@apple.com
  • 2 edits in trunk/LayoutTests

Fix the test case that was landed in r46624.

  • http/tests/misc/meta-refresh-stray-single-quote.html:

Jul 30, 2009:

11:54 PM Changeset in webkit [46624] by beidson@apple.com
  • 3 edits
    3 adds in trunk

WebCore:

2009-07-30 Brady Eidson <beidson@apple.com>

Reviewed by Mark Rowe, but Dan Bernstein also reviewed and asked thoughtful questions.

<rdar://problem/7106968> and https://bugs.webkit.org/show_bug.cgi?id=27868
http://www.ruthhuntcandy.com/ goes into infinite refresh in WebKit, works in others.


Test: http/tests/misc/meta-refresh-stray-single-quote.html

  • platform/network/HTTPParsers.cpp: (WebCore::parseHTTPRefresh): Allow for a stray quote character at the start of the URL string.

LayoutTests:

2009-07-30 Brady Eidson <beidson@apple.com>

Reviewed by Mark Rowe, but Dan Bernstein also reviewed and asked thoughtful questions.

<rdar://problem/7106968> and https://bugs.webkit.org/show_bug.cgi?id=27868
http://www.ruthhuntcandy.com/ goes into infinite refresh in WebKit, works in others.

  • http/tests/misc/meta-refresh-stray-single-quote-expected.txt: Added.
  • http/tests/misc/meta-refresh-stray-single-quote.html: Added.
  • http/tests/misc/resources/success-notify-done.html: Added.
11:37 PM Changeset in webkit [46623] by xan@webkit.org
  • 2 edits in trunk/JavaScriptCore

2009-07-30 Xan Lopez <xlopez@igalia.com>

Reviewed by Jan Alonzo.

Fix compiler warning.

GCC does not like C++-style comments in preprocessor directives.

  • wtf/Platform.h:
11:15 PM Changeset in webkit [46622] by jmalonzo@webkit.org
  • 4 edits in trunk/WebKit/gtk

2009-07-30 Martin Robinson <mrobinson@appcelerator.com>

Reviewed by Xan Lopez, Gustavo Noronha and Jan Alonzo.

[GTK] new-window-policy-decision-requested provides no information about the target frame
https://bugs.webkit.org/show_bug.cgi?id=27792

Add frameName property to the WebKitWebNavigation object.

  • WebCoreSupport/FrameLoaderClientGtk.cpp: (WebKit::getNavigationAction): (WebKit::FrameLoaderClient::dispatchDecidePolicyForNewWindowAction): (WebKit::FrameLoaderClient::dispatchDecidePolicyForNavigationAction):
  • webkit/webkitwebnavigationaction.cpp: (webkit_web_navigation_action_get_property): (webkit_web_navigation_action_set_property): (webkit_web_navigation_action_class_init): (webkit_web_navigation_action_get_target_frame): (webkit_web_navigation_action_set_target_frame):
  • webkit/webkitwebnavigationaction.h:
10:38 PM Changeset in webkit [46621] by barraclough@apple.com
  • 5 edits in trunk/JavaScriptCore

2009-07-30 John McCall <rjmccall@apple.com>

Reviewed by Gavin Barraclough.

Optimize the X86_64 trampolines: avoid the need for filler arguments
and move the stub-args area closer to the stack pointer.

  • jit/JIT.h: adjust patch offsets because of slight code-size change
  • jit/JITCode.h: (JSC::JITCode::execute): don't pass filler args
  • jit/JITStubs.cpp: (ctiTrampoline): (X86_64): push args onto stack, use args directly (ctiVMThrowTrampoline): (X86_64): adjust %rsp by correct displacement (ctiOpThrowNotCaught): (X86_64): adjust %rsp by correct displacement
  • jit/JITStubs.h: (JITStackFrame): (X86_64): move args area earlier (ctiTrampoline): remove filler args from prototype
7:57 PM Changeset in webkit [46620] by barraclough@apple.com
  • 13 edits in trunk/JavaScriptCore

2009-07-30 Gavin Barraclough <barraclough@apple.com>

Reviewed by NOBODY (build fix).

Temporarily revert r46618 since this is b0rking on Linux.

7:29 PM Changeset in webkit [46619] by mrowe@apple.com
  • 2 edits in trunk/WebKitTools

Teach buildbot to detect test cases that are missing results.

Reviewed by Cameron Zwarich.

  • BuildSlaveSupport/build.webkit.org-config/master.cfg:
7:20 PM Changeset in webkit [46618] by barraclough@apple.com
  • 13 edits in trunk/JavaScriptCore

2009-07-23 Gavin Barraclough <barraclough@apple.com>

Reviewed by Oliver Hunt.

Make get_by_id/put_by_id/method_check/call defer optimization using a data flag rather than a code modification.
( https://bugs.webkit.org/show_bug.cgi?id=27635 )

This improves performance of ENABLE(ASSEMBLER_WX_EXCLUSIVE) builds by 2-2.5%, reducing the overhead to about 2.5%.
(No performance impact with ASSEMBLER_WX_EXCLUSIVE disabled).

  • bytecode/CodeBlock.cpp: (JSC::printStructureStubInfo):
    • Make StructureStubInfo store the type as an integer, rather than an OpcodeID.
  • bytecode/CodeBlock.h: (JSC::): (JSC::CallLinkInfo::seenOnce): (JSC::CallLinkInfo::setSeen): (JSC::MethodCallLinkInfo::seenOnce): (JSC::MethodCallLinkInfo::setSeen):
    • Change a pointer in CallLinkInfo/MethodCallLinkInfo to use a PtrAndFlags, use a flag to track when an op has been executed once.
  • bytecode/StructureStubInfo.cpp: (JSC::StructureStubInfo::deref):
    • Make StructureStubInfo store the type as an integer, rather than an OpcodeID.
  • bytecode/StructureStubInfo.h: (JSC::StructureStubInfo::StructureStubInfo): (JSC::StructureStubInfo::initGetByIdSelf): (JSC::StructureStubInfo::initGetByIdProto): (JSC::StructureStubInfo::initGetByIdChain): (JSC::StructureStubInfo::initGetByIdSelfList): (JSC::StructureStubInfo::initGetByIdProtoList): (JSC::StructureStubInfo::initPutByIdTransition): (JSC::StructureStubInfo::initPutByIdReplace): (JSC::StructureStubInfo::seenOnce): (JSC::StructureStubInfo::setSeen):
    • Make StructureStubInfo store the type as an integer, rather than an OpcodeID, add a flag to track when an op has been executed once.
  • bytecompiler/BytecodeGenerator.cpp: (JSC::BytecodeGenerator::emitGetById): (JSC::BytecodeGenerator::emitPutById):
    • Make StructureStubInfo store the type as an integer, rather than an OpcodeID.
  • jit/JIT.cpp: (JSC::JIT::privateCompileCTIMachineTrampolines): (JSC::JIT::unlinkCall):
    • Remove the "don't lazy link" stage of calls.
  • jit/JIT.h: (JSC::JIT::compileCTIMachineTrampolines):
    • Remove the "don't lazy link" stage of calls.
  • jit/JITCall.cpp: (JSC::JIT::compileOpCallSlowCase):
    • Remove the "don't lazy link" stage of calls.
  • jit/JITStubs.cpp: (JSC::JITThunks::JITThunks): (JSC::JITThunks::tryCachePutByID): (JSC::JITThunks::tryCacheGetByID): (JSC::JITStubs::DEFINE_STUB_FUNCTION): (JSC::JITStubs::getPolymorphicAccessStructureListSlot):
    • Remove the "don't lazy link" stage of calls, and the "_second" stage of get_by_id/put_by_id/method_check.
  • jit/JITStubs.h: (JSC::JITThunks::ctiStringLengthTrampoline): (JSC::JITStubs::):
    • Remove the "don't lazy link" stage of calls, and the "_second" stage of get_by_id/put_by_id/method_check.
  • wtf/PtrAndFlags.h: (WTF::PtrAndFlags::PtrAndFlags): (WTF::PtrAndFlags::operator!): (WTF::PtrAndFlags::operator->):
    • Add ! and -> operators, add constuctor with pointer argument.
6:28 PM Changeset in webkit [46617] by mrowe@apple.com
  • 6 edits in trunk/WebCore

Bug 27828: Title attribute is not respected on option elements
Fix <https://bugs.webkit.org/show_bug.cgi?id=27828> for Mac.

Reviewed by Adele Peterson and Jon Honeycutt.

No tests added as it is not clear how to test a tool tip from DumpRenderTree.

  • platform/PopupMenuClient.h: Add a method for retrieving the tool tip of an item.
  • platform/mac/PopupMenuMac.mm:

(WebCore::PopupMenu::populate): Set the tool tip of the menu item to that of the represented item.

  • rendering/RenderMenuList.cpp:

(WebCore::RenderMenuList::itemToolTip): Expose the title attribute of the element as the tool tip.

  • rendering/RenderMenuList.h:
  • rendering/RenderTextControlSingleLine.h:

(WebCore::RenderTextControlSingleLine::itemToolTip): Return an empty string, indicating no tool tip, for
the popup menu in RenderTextControlSingleLine.

5:52 PM Changeset in webkit [46616] by mrowe@apple.com
  • 2 edits in trunk/WebKitTools

Things work better when you create the correct directories.

  • Scripts/run-webkit-tests:
5:22 PM Changeset in webkit [46615] by mrowe@apple.com
  • 2 edits in trunk/WebKitTools

Include some information about the error when we die.

  • Scripts/run-webkit-tests:
4:51 PM Changeset in webkit [46614] by treat@webkit.org
  • 1 edit
    2 adds in trunk/LayoutTests

2009-07-30 Adam Treat <adam.treat@torchmobile.com>

Add expected results file for new test.

  • platform/gtk/fast/forms/button-style-color-expected.txt: Added.
4:41 PM Changeset in webkit [46613] by mrowe@apple.com
  • 2 edits in trunk/WebKitTools

Ensure that the path exists before writing results to it.
Hopefully this will make the Windows build bot happy.

  • Scripts/run-webkit-tests:
4:32 PM Changeset in webkit [46612] by treat@webkit.org
  • 1 edit
    1 add in trunk/LayoutTests

2009-07-30 Adam Treat <adam.treat@torchmobile.com>

Add expected results file for new test.

  • platform/mac/fast/forms/button-style-color-expected.txt: Added.
4:21 PM Changeset in webkit [46611] by darin@chromium.org
  • 2 edits in trunk/WebCore

2009-07-30 Xiaomei Ji <xji@chromium.org>

Reviewed by Dan Bernstein.

Remove ChromeClientChromium::setToolTip().
https://bugs.webkit.org/show_bug.cgi?id=27861

This patch is just to remove a temporarily introduced overloaded empty
virtual function. No test is needed.

  • page/chromium/ChromeClientChromium.h:
4:11 PM Changeset in webkit [46610] by ggaren@apple.com
  • 3 edits in trunk/JavaScriptCore

2009-07-30 Geoffrey Garen <ggaren@apple.com>

Reviewed by Gavin Barraclough.

Fixed failing tests seen on Windows buildbot.

  • jit/JITStubs.cpp: (JSC::DEFINE_STUB_FUNCTION):
  • jit/JITStubs.h: (JSC::): Use "int" instead of "bool" to guarantee a 32-bit result, regardless of compiler. gcc on mac uses 32-bit values for bool, but gcc on linux and MSVC on Windows use 8-bit values.
4:08 PM Changeset in webkit [46609] by darin@chromium.org
  • 25 edits
    2 adds in trunk

2009-07-30 Michael Nordman <Michael Nordman>

Reviewed by Darin Fisher.

https://bugs.webkit.org/show_bug.cgi?id=27821

ApplicationCacheHost refactoring.


1) Better encapsulate the interfaces between webcore common code
and the appcache system within a new class ApplicationCacheHost.

2) Use that interface throughout the loader system, replacing inline appcache logic.

3) Implement the interface in terms of webcore's appcache system.

4) Add the new files to various makefiles.

5) Implement protocolHostAndPortAreEqual() in KURLGoogle.cpp

No new features, no new tests. The existing layout tests all pass.

  • GNUmakefile.am:
  • WebCore.base.exp:
  • WebCore.gypi:
  • WebCore.pro:
  • WebCore.vcproj/WebCore.vcproj:
  • WebCore.xcodeproj/project.pbxproj:
  • WebCoreSources.bkl:
  • html/HTMLHtmlElement.cpp: (WebCore::HTMLHtmlElement::insertedIntoDocument):
  • loader/DocumentLoader.cpp: (WebCore::DocumentLoader::DocumentLoader): (WebCore::DocumentLoader::~DocumentLoader): (WebCore::DocumentLoader::mainReceivedError): (WebCore::DocumentLoader::detachFromFrame): (WebCore::DocumentLoader::setPrimaryLoadComplete):
  • loader/DocumentLoader.h: (WebCore::DocumentLoader::applicationCacheHost):
  • loader/FrameLoader.cpp: (WebCore::FrameLoader::canCachePageContainingThisFrame): (WebCore::FrameLoader::logCanCacheFrameDecision): (WebCore::FrameLoader::loadResourceSynchronously):
  • loader/MainResourceLoader.cpp: (WebCore::MainResourceLoader::didReceiveResponse): (WebCore::MainResourceLoader::didReceiveData): (WebCore::MainResourceLoader::didFinishLoading): (WebCore::MainResourceLoader::didFail): (WebCore::MainResourceLoader::load):
  • loader/MainResourceLoader.h:
  • loader/ResourceLoader.cpp: (WebCore::ResourceLoader::load): (WebCore::ResourceLoader::willSendRequest): (WebCore::ResourceLoader::didReceiveResponse): (WebCore::ResourceLoader::didFail):
  • loader/ResourceLoader.h:
  • loader/appcache/ApplicationCacheGroup.cpp: (WebCore::ApplicationCacheGroup::selectCache): (WebCore::ApplicationCacheGroup::selectCacheWithoutManifestURL): (WebCore::ApplicationCacheGroup::finishedLoadingMainResource): (WebCore::ApplicationCacheGroup::failedLoadingMainResource): (WebCore::ApplicationCacheGroup::disassociateDocumentLoader): (WebCore::ApplicationCacheGroup::update): (WebCore::ApplicationCacheGroup::didFinishLoadingManifest): (WebCore::ApplicationCacheGroup::manifestNotFound): (WebCore::ApplicationCacheGroup::checkIfLoadIsComplete): (WebCore::ApplicationCacheGroup::startLoadingEntry): (WebCore::ApplicationCacheGroup::associateDocumentLoaderWithCache): (WebCore::CallCacheListenerTask::create): (WebCore::CallCacheListenerTask::performTask): (WebCore::CallCacheListenerTask::CallCacheListenerTask): (WebCore::ApplicationCacheGroup::postListenerTask):
  • loader/appcache/ApplicationCacheGroup.h:
  • loader/appcache/ApplicationCacheHost.cpp: Added.
  • loader/appcache/ApplicationCacheHost.h: Added.
  • loader/appcache/ApplicationCacheStorage.cpp: (WebCore::ApplicationCacheStorage::transferApplicationCache):
  • loader/appcache/ApplicationCacheStorage.h:
  • loader/appcache/DOMApplicationCache.cpp: (WebCore::DOMApplicationCache::DOMApplicationCache): (WebCore::DOMApplicationCache::disconnectFrame): (WebCore::DOMApplicationCache::applicationCacheHost): (WebCore::DOMApplicationCache::status): (WebCore::DOMApplicationCache::update): (WebCore::DOMApplicationCache::swapCache): (WebCore::DOMApplicationCache::addEventListener): (WebCore::DOMApplicationCache::removeEventListener): (WebCore::DOMApplicationCache::dispatchEvent): (WebCore::DOMApplicationCache::callListener): (WebCore::DOMApplicationCache::toEventName): (WebCore::DOMApplicationCache::toEventType):
  • loader/appcache/DOMApplicationCache.h: (WebCore::DOMApplicationCache::): (WebCore::DOMApplicationCache::setAttributeEventListener): (WebCore::DOMApplicationCache::getAttributeEventListener): (WebCore::DOMApplicationCache::clearAttributeEventListener): (WebCore::DOMApplicationCache::callEventListener): (WebCore::DOMApplicationCache::setOnchecking): (WebCore::DOMApplicationCache::onchecking): (WebCore::DOMApplicationCache::setOnerror): (WebCore::DOMApplicationCache::onerror): (WebCore::DOMApplicationCache::setOnnoupdate): (WebCore::DOMApplicationCache::onnoupdate): (WebCore::DOMApplicationCache::setOndownloading): (WebCore::DOMApplicationCache::ondownloading): (WebCore::DOMApplicationCache::setOnprogress): (WebCore::DOMApplicationCache::onprogress): (WebCore::DOMApplicationCache::setOnupdateready): (WebCore::DOMApplicationCache::onupdateready): (WebCore::DOMApplicationCache::setOncached): (WebCore::DOMApplicationCache::oncached): (WebCore::DOMApplicationCache::setOnobsolete): (WebCore::DOMApplicationCache::onobsolete): (WebCore::DOMApplicationCache::~DOMApplicationCache):
  • platform/KURLGoogle.cpp: (WebCore::protocolHostAndPortAreEqual):
4:05 PM Changeset in webkit [46608] by treat@webkit.org
  • 4 edits in trunk/LayoutTests

2009-07-30 Adam Treat <adam.treat@torchmobile.com>

Reviewed by Mark Rowe.

Revert previous commit to test out new functionality of buildbots and
run-webkit-tests made possible by r46607.

  • platform/gtk/Skipped:
  • platform/mac/Skipped:
  • platform/win/Skipped:
3:54 PM Changeset in webkit [46607] by mrowe@apple.com
  • 2 edits in trunk/WebKitTools

Teach run-webkit-tests to always store the actual results in the results directory for new tests.

Reviewed by Adam Treat and Jon Honeycutt.

This allows us to give a working link to the results from the result summary page, even if we're
not generating new results in to the tree.

  • Scripts/run-webkit-tests:
3:44 PM Changeset in webkit [46606] by Simon Fraser
  • 2 edits in trunk/WebCore

2009-07-30 Simon Fraser <Simon Fraser>

Minor change to earlier commit suggested by Darin Adler.
Use the variable rather than dereferencing 'it' again.

  • css/CSSStyleSelector.cpp: (WebCore::CSSStyleSelector::keyframeStylesForAnimation):
3:28 PM Changeset in webkit [46605] by treat@webkit.org
  • 4 edits in trunk/LayoutTests

2009-07-30 Adam Treat <adam.treat@torchmobile.com>

Reviewed by George Staikos.

Add new test with platform specific results to the Skipped files of ports
that haven't generated them yet. This is to make the buildbot green again.

  • platform/gtk/Skipped:
  • platform/mac/Skipped:
  • platform/win/Skipped:
2:50 PM Changeset in webkit [46604] by Chris Fleizach
  • 3 edits in trunk/WebCore

Bug 27854 - crash at WebCore::AXObjectCache::notificationPostTimerFired
https://bugs.webkit.org/show_bug.cgi?id=27854

2:47 PM Changeset in webkit [46603] by Simon Fraser
  • 3 edits in trunk/WebCore

2009-07-30 Simon Fraser <Simon Fraser>

Reviewed by Dan Bernstein.

Animation with a timing function property in a keyframe eats CPU
https://bugs.webkit.org/show_bug.cgi?id=27856
<rdar://problem/7104476> Animation demo uses lots of CPU

Don't include animation-timing-function in the list of properties to
animate in a keyframe animation, because this property is not animated;
instead, it describes the timing function to apply to this keyframe.

This prevents the animation code from thinking that there's a property
that it has to software-animate, and thus firing the animation timer frequently.

Not testable because there is no visible impact.

  • css/CSSStyleSelector.cpp: (WebCore::CSSStyleSelector::keyframeStylesForAnimation): Don't add CSSPropertyWebkitAnimationTimingFunction to the list of properties to animate.
  • page/animation/KeyframeAnimation.cpp (WebCore::KeyframeAnimation::getKeyframeAnimationInterval): Add a comment
2:37 PM Changeset in webkit [46602] by mrowe@apple.com
  • 2 edits in trunk/WebKitTools

Add a Qt Linux build bot.

  • BuildSlaveSupport/build.webkit.org-config/config.json:
2:18 PM Changeset in webkit [46601] by ggaren@apple.com
  • 3 edits in trunk/JavaScriptCore

2009-07-30 Geoffrey Garen <ggaren@apple.com>

Windows build fix: added missing symbols on Windows.

2:16 PM Changeset in webkit [46600] by treat@webkit.org
  • 5 edits
    2 adds in trunk

WebCore:
2009-07-30 Mike Fenton <mike.fenton@torchmobile.com>

Reviewed by Adam Treat.

Apply colour style to buttons that use Theme settings in Qt.

https://bugs.webkit.org/show_bug.cgi?id=27814

  • platform/qt/RenderThemeQt.cpp: (WebCore::RenderThemeQt::adjustButtonStyle):

LayoutTests:
2009-07-30 Mike Fenton <mike.fenton@torchmobile.com>

Reviewed by Adam Treat.

Add Button Colour style test and results for Qt based upon fix for bug.
Unskip tables/mozilla/bugs/bug92647-2.html which passes with bug fix.
Update expected results for qt for test fast/dom/HTMLTableColElement/resize-table-using-col-width-expected
which also passes with the the bug fix.

https://bugs.webkit.org/show_bug.cgi?id=27814

  • fast/forms/button-style-color.html: Added.
  • platform/qt/Skipped:
  • platform/qt/fast/dom/HTMLTableColElement/resize-table-using-col-width-expected.txt:
  • platform/qt/fast/forms/button-style-color-expected.txt: Added.
2:05 PM Changeset in webkit [46599] by ggaren@apple.com
  • 3 edits in trunk/JavaScriptCore

2009-07-30 Geoffrey Garen <ggaren@apple.com>

Windows build fix: removed stale symbols on Windows.

1:57 PM Changeset in webkit [46598] by ggaren@apple.com
  • 63 edits
    3 adds in trunk

Merged nitro-extreme branch into trunk.

1:57 PM Changeset in webkit [46597] by treat@webkit.org
  • 2 edits in trunk/WebKitTools

2009-07-30 Jakob Petsovits <jakob.petsovits@torchmobile.com>

Reviewed by Adam Roben.

Add --minimal option to webkit-build.
https://bugs.webkit.org/show_bug.cgi?id=27852

This option disables all optional build features unless
they are explicitly enabled.

  • Scripts/build-webkit:
12:31 PM Changeset in webkit [46596] by Dimitri Glazkov
  • 3 edits in trunk/WebCore

2009-07-30 Dean McNamee <deanm@chromium.org>

Reviewed by Dimitri Glazkov.

Don't try to hash member function pointers, instead use a precomputed value based on the field.
https://bugs.webkit.org/show_bug.cgi?id=27843

  • bindings/scripts/CodeGeneratorV8.pm:
  • bindings/v8/V8SVGPODTypeWrapper.h: (WebCore::PODTypeWrapperCacheInfo::PODTypeWrapperCacheInfo): (WebCore::PODTypeWrapperCacheInfo::operator==): (WebCore::PODTypeWrapperCacheInfoHash::hash): (WebCore::V8SVGDynamicPODTypeWrapperCache::lookupOrCreateWrapper):
12:30 PM Changeset in webkit [46595] by xan@webkit.org
  • 2 edits in trunk/WebCore

2009-07-30 Xan Lopez <xlopez@igalia.com>

Reviewed by Simon Fraser.

Fix compiler warning.

  • dom/SelectElement.cpp: (WebCore::SelectElement::menuListDefaultEventHandler):
12:20 PM Changeset in webkit [46594] by mpcomplete@chromium.org
  • 7 edits in trunk/WebCore

2009-07-29 Matt Perry <mpcomplete@chromium.org>

Reviewed by Adam Barth.

Add a way to register V8 extensions for Isolated Worlds only.
https://bugs.webkit.org/show_bug.cgi?id=27785

  • bindings/v8/ScriptController.cpp: (WebCore::ScriptController::evaluateInNewWorld): (WebCore::ScriptController::evaluateInNewContext):
  • bindings/v8/ScriptController.h:
  • bindings/v8/V8IsolatedWorld.cpp: (WebCore::V8IsolatedWorld::evaluate):
  • bindings/v8/V8IsolatedWorld.h:
  • bindings/v8/V8Proxy.cpp: (WebCore::V8Proxy::evaluateInNewWorld): (WebCore::V8Proxy::evaluateInNewContext): (WebCore::V8Proxy::createNewContext): (WebCore::V8Proxy::initContextIfNeeded): (WebCore::V8Proxy::registerExtensionWithV8): (WebCore::V8Proxy::registerExtension):
  • bindings/v8/V8Proxy.h:
12:07 PM Changeset in webkit [46593] by treat@webkit.org
  • 2 edits in trunk/WebCore

2009-07-30 Mike Fenton <mike.fenton@torchmobile.com>

Reviewed by Adam Treat.

Add previously defined out support to PopupMenuQt for marking entries as
disabled and for selecting the desired item.

https://bugs.webkit.org/show_bug.cgi?id=27772

  • platform/qt/PopupMenuQt.cpp: (WebCore::PopupMenu::populate):
11:44 AM Changeset in webkit [46592] by Darin Adler
  • 34 edits in trunk/WebCore

2009-07-30 Darin Adler <Darin Adler>

Reviewed by David Levin.

Use checked casts for render tree
https://bugs.webkit.org/show_bug.cgi?id=23522

Next step: Add new toRenderWidget cast and use it everywhere.

Use checked casts in all the places that were using static_cast
but there is a checked cast available.

  • accessibility/AccessibilityRenderObject.cpp: (WebCore::AccessibilityRenderObject::stringValue): (WebCore::AccessibilityRenderObject::widget): (WebCore::AccessibilityRenderObject::widgetForAttachmentView): (WebCore::AccessibilityRenderObject::visiblePositionForPoint):
  • accessibility/gtk/AccessibilityObjectWrapperAtk.cpp: (getPangoLayoutForAtk):
  • dom/Document.cpp: (WebCore::widgetForNode): (WebCore::Document::setFocusedNode):
  • html/HTMLEmbedElement.cpp: (WebCore::findWidgetRenderer):
  • html/HTMLMediaElement.cpp: (WebCore::HTMLMediaElement::defaultEventHandler):
  • html/HTMLObjectElement.cpp: (WebCore::HTMLObjectElement::renderWidgetForJSBindings):
  • html/HTMLPlugInElement.cpp: (WebCore::HTMLPlugInElement::defaultEventHandler):
  • loader/FrameLoader.cpp: (WebCore::FrameLoader::loadSubframe):
  • loader/PluginDocument.cpp: (WebCore::PluginTokenizer::writeRawData):
  • page/EventHandler.cpp: (WebCore::EventHandler::hitTestResultAtPoint): (WebCore::subframeForTargetNode): (WebCore::EventHandler::handleWheelEvent):
  • page/Frame.cpp: (WebCore::isFrameElement):
  • page/PrintContext.cpp: (WebCore::PrintContext::computePageRects):
  • page/android/EventHandlerAndroid.cpp: (WebCore::EventHandler::passWidgetMouseDownEventToWidget):
  • page/chromium/EventHandlerChromium.cpp: (WebCore::EventHandler::passWidgetMouseDownEventToWidget):
  • page/chromium/FrameChromium.cpp: (WebCore::computePageRectsForFrame):
  • page/gtk/EventHandlerGtk.cpp: (WebCore::EventHandler::passWidgetMouseDownEventToWidget):
  • page/haiku/EventHandlerHaiku.cpp: (WebCore::EventHandler::passWidgetMouseDownEventToWidget):
  • page/mac/EventHandlerMac.mm: (WebCore::EventHandler::passWidgetMouseDownEventToWidget): (WebCore::EventHandler::passSubframeEventToSubframe):
  • page/win/FrameWin.cpp: (WebCore::computePageRectsForFrame):
  • page/wx/EventHandlerWx.cpp: (WebCore::EventHandler::passWidgetMouseDownEventToWidget):
  • platform/chromium/ClipboardChromium.cpp: (WebCore::getCachedImage):
  • platform/chromium/PasteboardChromium.cpp: (WebCore::Pasteboard::writeImage):
  • platform/gtk/PasteboardGtk.cpp: (WebCore::Pasteboard::writeImage):
  • platform/mac/PasteboardMac.mm: (WebCore::Pasteboard::writeImage):
  • platform/qt/ClipboardQt.cpp: (WebCore::getCachedImage):
  • platform/qt/PasteboardQt.cpp: (WebCore::Pasteboard::writeImage):
  • platform/win/ClipboardWin.cpp: (WebCore::getCachedImage):
  • platform/win/PasteboardWin.cpp: (WebCore::Pasteboard::writeImage):
  • rendering/InlineBox.h: (WebCore::InlineBox::boxModelObject):
  • rendering/RenderInline.cpp: (WebCore::RenderInline::splitInlines): (WebCore::RenderInline::addChildToContinuation):
  • rendering/RenderLayerBacking.cpp: (WebCore::RenderLayerBacking::updateImageContents):
  • rendering/TextControlInnerElements.cpp: (WebCore::RenderTextControlInnerBlock::positionForPoint): Use checked cast instead of static_cast.


  • rendering/RenderWidget.h: Added toRenderWidget.
11:25 AM Changeset in webkit [46591] by staikos@webkit.org
  • 6 edits in trunk/WebCore

2009-07-30 Yong Li <yong.li@torchmobile.com>

Reviewed by George Staikos.

WINCE PORT: some files modified to build for WINCE
https://bugs.webkit.org/show_bug.cgi?id=27816

11:08 AM Changeset in webkit [46590] by staikos@webkit.org
  • 13 edits in trunk/WebCore

2009-07-30 Yong Li <yong.li@torchmobile.com>

Reviewed by George Staikos.

WINCE PORT: modified graphics files
https://bugs.webkit.org/show_bug.cgi?id=27779

10:52 AM Changeset in webkit [46589] by kenneth@webkit.org
  • 2 edits in trunk/WebCore

2009-07-30 Kenneth Rohde Christiansen <kenneth@webkit.org>

Reviewed by Ariya Hidayat.

Improve efficiency by rewriting code doing three hash table
lookups, which can be replaced by just one as pointed out
by Darin Adler.

Though being slightly less clear, this should be considerable
faster.

  • plugins/PluginDatabase.cpp: (WebCore::PluginDatabase::remove):
10:50 AM Changeset in webkit [46588] by kov@webkit.org
  • 3 edits in trunk/WebCore

2009-07-30 Antonio Gomes <antonio.gomes@openbossa.org>

Reviewed by Gustavo Noronha.

[Gtk] Code cleanup in MediaPlayerPrivateGStreamer.cpp|h and VideoSinkGStreamer.cpp|h (Part 2)
https://bugs.webkit.org/show_bug.cgi?id=27651

Removed unneeded includes (gdk.h and gtk.h) and added glib.h and cairo.h instead.

  • platform/graphics/gtk/MediaPlayerPrivateGStreamer.cpp:
  • platform/graphics/gtk/MediaPlayerPrivateGStreamer.h:
10:38 AM Changeset in webkit [46587] by kov@webkit.org
  • 5 edits in trunk/WebCore

2009-07-30 Antonio Gomes <antonio.gomes@openbossa.org>

Reviewed by Gustavo Noronha.

[Gtk] Code cleanup in MediaPlayerPrivateGStreamer.cpp|h and VideoSinkGStreamer.cpp|h (Part 1)
https://bugs.webkit.org/show_bug.cgi?id=27651

Fixed many code style issues pointed by WebKitTools/Scripts/modules/cpplint.py
No functionality change at all.

  • platform/graphics/gtk/MediaPlayerPrivateGStreamer.cpp:
  • platform/graphics/gtk/MediaPlayerPrivateGStreamer.h:
  • platform/graphics/gtk/VideoSinkGStreamer.cpp:
  • platform/graphics/gtk/VideoSinkGStreamer.h:
9:43 AM Changeset in webkit [46586] by kenneth@webkit.org
  • 3 edits in trunk/WebCore

Allow to explicitly choose a preferred plugin for a mimetype.

https://bugs.webkit.org/show_bug.cgi?id=27651

When the preferred plugin is set for a specific MIME type, it will be
always picked up, regardless of its version, quirks etc.

Client applications may want to use that API to resolve mimetype
ambiguity in a custom way, rather than in the default way that is
currently used in WebKit.

9:32 AM Changeset in webkit [46585] by kenneth@webkit.org
  • 2 edits in trunk/WebKitTools

[Qt] Fix build with GCC 4.4.

9:31 AM Changeset in webkit [46584] by kenneth@webkit.org
  • 2 edits in trunk/WebCore

[Qt] Fix build with Qt 4.4 after r46535.

9:31 AM Changeset in webkit [46583] by kenneth@webkit.org
  • 2 edits in trunk/WebKit/qt

2009-07-30 Kenneth Rohde Christiansen <kenneth@webkit.org>

Reviewed by Adam Treat.

Add missing Q_OBJECT to class inheriting from QObject.

  • tests/qwebframe/tst_qwebframe.cpp:
9:31 AM Changeset in webkit [46582] by kenneth@webkit.org
  • 2 edits in trunk/WebKit/qt

2009-07-30 Kenneth Rohde Christiansen <kenneth@webkit.org>

Reviewed by Adam Treat.

Fix indentation.

  • tests/qwebframe/tst_qwebframe.cpp:
9:05 AM Changeset in webkit [46581] by eric.carlson@apple.com
  • 3 edits
    2 adds in trunk

2009-07-30 Eric Carlson <eric.carlson@apple.com>

Reviewed by Simon Fraser.

Video elements fires another "load" event when attached to DOM
https://bugs.webkit.org/show_bug.cgi?id=27623

Test: media/media-load-event.html

  • html/HTMLMediaElement.cpp: (WebCore::HTMLMediaElement::insertedIntoDocument):

Only schedule load when element's network state is NETWORK_EMPTY.

2009-07-30 Eric Carlson <eric.carlson@apple.com>

Reviewed by Simon Fraser.

Video elements fires another "load" event when attached to DOM
https://bugs.webkit.org/show_bug.cgi?id=27623

  • media/media-load-event-expected.txt: Added.
  • media/media-load-event.html: Added.
7:41 AM SpecSupport edited by tony@gonk.net
Added entries for XSLT 2.0 and XPath 2.0. Although each spec's v1.0 … (diff)
7:07 AM Changeset in webkit [46580] by levin@chromium.org
  • 3 edits in trunk/WebCore

2009-07-30 Anton Muhin <antonm@chromium.org>

Reviewed by David Levin.

Cache v8 strings when converting from WebCore::String to v8 string.
https://bugs.webkit.org/show_bug.cgi?id=27655

  • bindings/v8/V8Binding.cpp: (WebCore::v8String): now just immediately calls v8ExternalString (WebCore::enableStringImplCache): enables caching of conversions from WebCore::StringImpl to v8::String (WebCore::makeExternalString): utilty function to create external v8::String out of WebCore::String (WebCore::getStringCache): static function to access string cache (WebCore::cachedStringCallback): callback for weak handles of v8::Strings stored in the cache (WebCore::v8ExternalString): if caching enabled, checks if there is already v8::String for the given WebCore::StringImpl. If present, returns it, otherwise creates a new v8 external string.
  • bindings/v8/V8Binding.h:
6:11 AM Changeset in webkit [46579] by ariya@webkit.org
  • 4 edits in trunk/WebKitTools

2009-07-29 Ariya Hidayat <ariya.hidayat@nokia.com>

Reviewed by Simon Hausmann.

https://bugs.webkit.org/show_bug.cgi?id=27813

[Qt] Support pixel tests in DumpRenderTree for the Qt port.

  • DumpRenderTree/qt/DumpRenderTree.cpp: (WebCore::DumpRenderTree::DumpRenderTree): (WebCore::DumpRenderTree::open): (WebCore::DumpRenderTree::setDumpPixels): (WebCore::DumpRenderTree::dump):
  • DumpRenderTree/qt/DumpRenderTree.h:
  • DumpRenderTree/qt/main.cpp: (main):
6:01 AM Changeset in webkit [46578] by Simon Hausmann
  • 2 edits in trunk/WebKit/qt

2009-07-30 Volker Hilsheimer <volker.hilsheimer@nokia.com>

Reviewed by Simon Hausmann.

Fix documentation links to examples.

  • Api/qwebview.cpp:
5:59 AM Changeset in webkit [46577] by ariya@webkit.org
  • 3 edits
    2 adds in trunk

ChangeLog:

2009-07-29 Ariya Hidayat <ariya.hidayat@nokia.com>

Reviewed by Simon Hausmann.

Adding ImageDiff project file.

https://bugs.webkit.org/show_bug.cgi?id=27813

  • WebKit.pro: Added ImageDiff.pro.

WebKitTools/ChangeLog:

2009-07-29 Ariya Hidayat <ariya.hidayat@nokia.com>

Reviewed by Simon Hausmann.

https://bugs.webkit.org/show_bug.cgi?id=27813

[Qt] Implement ImageDiff.

  • DumpRenderTree/qt/ImageDiff.cpp: Added.
  • DumpRenderTree/qt/ImageDiff.pro: Added.
5:48 AM Changeset in webkit [46576] by Simon Hausmann
  • 2 edits in trunk/WebKit/qt

2009-07-30 Balazs Kelemen <kelemen.balazs.3@stud.u-szeged.hu>

Reviewed by Simon Hausmann.

Allocate MainWindow on heap in robotized QtLauncher as well.

  • QtLauncher/main.cpp: (main):
4:24 AM Changeset in webkit [46575] by vestbo@webkit.org
  • 2 edits in trunk/WebKitTools

2009-07-30 Tor Arne Vestbø <tor.arne.vestbo@nokia.com>

Reviewed by Eric Seidel.

build-webkit: Reorder logic for determining the baseProductDir

https://bugs.webkit.org/show_bug.cgi?id=27699

We now do port-spesific probing for the product dir first, and then
fall back to either the generic WEBKITOUTPUTDIR (which now also works
on Mac), or WebKitBuild.

Then we add the git branch if the git branch build option is enabled.

Finally we massage the product dir based on the port. For example the
Mac port will now always set the SYMROOT and OBJROOT configuration for
XCode.

Note that isAppleWinWebkit() and isCygwin() is not mutually exclusive,
hence the normal if and not elsif in the last two blocks.

  • Scripts/webkitdirs.pm:
12:52 AM Changeset in webkit [46574] by xan@webkit.org
  • 3 edits in trunk/WebCore

2009-07-30 Xan Lopez <xlopez@igalia.com>

Reviewed by Maciej Stachowiak.

https://bugs.webkit.org/show_bug.cgi?id=25535
[GTK] object:state-changed:checked events missing for radio buttons and checkboxes

Implement state-changed:checked for radio buttons and checkboxes.

  • accessibility/gtk/AXObjectCacheAtk.cpp: (WebCore::AXObjectCache::postPlatformNotification):
  • html/HTMLInputElement.cpp: (WebCore::HTMLInputElement::setChecked):
12:13 AM Changeset in webkit [46573] by xan@webkit.org
  • 2 edits in trunk/WebCore

2009-07-30 Xan Lopez <xlopez@igalia.com>

Forgot the 'break'.

  • platform/ContextMenu.cpp: (WebCore::ContextMenu::checkOrEnableIfNeeded):
12:06 AM Changeset in webkit [46572] by xan@webkit.org
  • 2 edits in trunk/WebCore

2009-07-30 Xan Lopez <xlopez@igalia.com>

Try to fix Mac build.

  • platform/ContextMenu.cpp: (WebCore::ContextMenu::checkOrEnableIfNeeded):
12:00 AM Changeset in webkit [46571] by xan@webkit.org
  • 2 edits in trunk/WebCore

2009-07-29 Xan Lopez <xlopez@igalia.com>

Reviewed by Adam Barth.

Add missing enumeration value to switch.

  • platform/ContextMenu.cpp: (WebCore::ContextMenu::checkOrEnableIfNeeded):

Jul 29, 2009:

6:10 PM Changeset in webkit [46570] by jianli@chromium.org
  • 13 edits in trunk

WebCore:

2009-07-29 Jian Li <jianli@chromium.org>

Reviewed by Darin Adler.

Workers need to throw an exception when presented with invalid URLs.
https://bugs.webkit.org/show_bug.cgi?id=27770

Tests covered by worker-constructor.html and worker-redirect.html.

  • bindings/js/JSWorkerConstructor.cpp: (WebCore::constructWorker):
  • bindings/v8/custom/V8WorkerCustom.cpp: (WebCore::CALLBACK_FUNC_DECL):
  • workers/Worker.cpp: (WebCore::Worker::Worker):
  • workers/Worker.h: (WebCore::Worker::create):
  • workers/WorkerContext.cpp: (WebCore::WorkerContext::importScripts):
  • workers/WorkerScriptLoader.cpp: (WebCore::WorkerScriptLoader::loadSynchronously): (WebCore::WorkerScriptLoader::loadAsynchronously): (WebCore::WorkerScriptLoader::createResourceRequest):
  • workers/WorkerScriptLoader.h:

LayoutTests:

2009-07-29 Jian Li <jianli@chromium.org>

Reviewed by Darin Adler.

Workers need to throw an exception when presented with invalid URLs.
https://bugs.webkit.org/show_bug.cgi?id=27770

Update worker-constructor.html and worker-redirect.html per the behavior change.

  • fast/workers/worker-constructor-expected.txt:
  • fast/workers/worker-constructor.html:
  • http/tests/workers/worker-redirect-expected.txt:
  • http/tests/workers/worker-redirect.html:
5:33 PM Changeset in webkit [46569] by beidson@apple.com
  • 2 edits in trunk/WebCore

2009-07-29 Brady Eidson <beidson@apple.com>

Reviewed by Darin Adler.

(REGRESSION: r46039) Should restore previous connections-per-host limit for non-http(s) hosts
https://bugs.webkit.org/show_bug.cgi?id=27822 and <rdar://problem/7091659>

  • loader/loader.cpp: (WebCore::Loader::Host::servePendingRequests): For non-http(s) hosts, restore the previous behavior of only limiting connections while parsing and still trying to figure out stylesheet urls.
5:17 PM Changeset in webkit [46568] by staikos@webkit.org
  • 1 edit
    2 adds in trunk/WebCore

2009-07-29 Yong Li <yong.li@torchmobile.com>

Reviewed by George Staikos.


WINCE PORT: Add WebCore/page/wince/FrameWince.cpp
https://bugs.webkit.org/show_bug.cgi?id=27729

5:13 PM Changeset in webkit [46567] by rniwa@webkit.org
  • 5 edits
    2 adds
    8 deletes in trunk

WebCore:

2009-07-29 Ryosuke Niwa <rniwa@webkit.org>

Reviewed by Justin Garcia.

REGRESSION(r46370-46426): /editing/style/remove-underline-from-stylesheet.html fails
https://bugs.webkit.org/show_bug.cgi?id=27809

The patch primarily rebaselines the tests but also fixes the bug in currentlyHasStyle.
To determine that a particular text decoration is present, currentlyHasStyle should refer to
-webkit-text-decorations-in-effect to take care of styles set by u, s, strike tags and ancestors' CSS.
We also need to update layout to accommodate the changes made within ApplyStyleCommand.

  • editing/ApplyStyleCommand.cpp: ditto (WebCore::StyleChange::currentlyHasStyle): ditto

LayoutTests:

2009-07-29 Ryosuke Niwa <rniwa@webkit.org>

Reviewed by Justin Garcia.

REGRESSION(r46370-46426): /editing/style/remove-underline-from-stylesheet.html fails
https://bugs.webkit.org/show_bug.cgi?id=27809

Since new behavior is consistent with Firefox, we rebaseline the test cases.
The patch converts remove-underline-from-stylesheet and remove-underline-in-bold to dumpAsText tests
with markup printed explicitly because pixel tests did not detect the disappearance of underline.

  • editing/style/remove-underline-from-stylesheet-expected.txt: Added.
  • editing/style/remove-underline-from-stylesheet.html:
  • editing/style/remove-underline-in-bold-expected.txt: Added.
  • editing/style/remove-underline-in-bold.html:
  • platform/mac/editing/style/remove-underline-from-stylesheet-expected.checksum: Removed.
  • platform/mac/editing/style/remove-underline-from-stylesheet-expected.png: Removed.
  • platform/mac/editing/style/remove-underline-from-stylesheet-expected.txt: Removed.
  • platform/mac/editing/style/remove-underline-in-bold-expected.checksum: Removed.
  • platform/mac/editing/style/remove-underline-in-bold-expected.png: Removed.
  • platform/mac/editing/style/remove-underline-in-bold-expected.txt: Removed.
  • platform/qt/editing/style/remove-underline-from-stylesheet-expected.txt: Removed.
  • platform/qt/editing/style/remove-underline-in-bold-expected.txt: Removed.
5:08 PM Changeset in webkit [46566] by staikos@webkit.org
  • 4 edits in trunk/WebCore

2009-07-29 Yong Li <yong.li@torchmobile.com>

Reviewed by George Staikos.

WINCE PORT: changes to platform/text files
https://bugs.webkit.org/show_bug.cgi?id=27715

5:08 PM Changeset in webkit [46565] by ddkilzer@apple.com
  • 4 edits in trunk/WebKit

<http://webkit.org/b/27788> Don't export WebPluginController.h as a private header

Reviewed by Mark Rowe.

WebKit:

WebPluginController.h includes WebPluginContainerCheck.h, which
is not a private header. Since WebPluginController.h doesn't
appear to be used anywhere, remove its private header status.

  • WebKit.xcodeproj/project.pbxproj: Remove private header attribute from WebPluginController.h.

WebKit/mac:

  • Plugins/WebPluginController.h: Changed #import of WebPluginContainerCheck.h to use a framework-style include in case other platforms wish to export WebPluginController.h as a private header.
4:48 PM Changeset in webkit [46564] by staikos@webkit.org
  • 2 edits in trunk/WebCore

2009-07-29 Yong Li <yong.li@torchmobile.com>

Reviewed by Adam Roben.

FIX: HDC leaks in PopupMenuWin.cpp
https://bugs.webkit.org/show_bug.cgi?id=27817

4:48 PM Changeset in webkit [46563] by ddkilzer@apple.com
  • 3 edits in trunk/WebKitTools

<http://webkit.org/b/27082> bugzilla-tool: Add --no-close switch to land-patches

Reviewed by David Levin.

  • Scripts/bugzilla-tool: (LandPatchesFromBugs.init): Added --no-close switch. (LandPatchesFromBugs.land_patches): Don't close the bug if the --no-close switch was used. Always clear the review+ flag on every landed patch using the commit_text message when cleared. This prevents patches from showing up in the commit queue if reopened and provides consistency with all landed patches.
  • Scripts/modules/bugzilla.py: (Bugzilla.clear_attachment_review_flag): Added.
4:42 PM Changeset in webkit [46562] by Darin Adler
  • 2 edits in trunk/LayoutTests

2009-07-29 Darin Adler <Darin Adler>

  • platform/gtk/Skipped: Added ol-start-parsing.html since it has no result yet.
4:29 PM Changeset in webkit [46561] by Darin Adler
  • 3 edits
    4 adds in trunk

2009-07-29 Darin Adler <Darin Adler>

Reviewed by Dan Bernstein.

Illegal values for <ol start> cause list numbering start at 0 (should start at 1)
https://bugs.webkit.org/show_bug.cgi?id=27810

Test: fast/lists/ol-start-parsing.html

  • html/HTMLOListElement.cpp: (WebCore::HTMLOListElement::parseMappedAttribute): Check the result of toInt and use the value 1 if it failed to parse. Before we were getting toInt's default behavior, which is to return 0.

2009-07-29 Darin Adler <Darin Adler>

Reviewed by Dan Bernstein.

Illegal values for <ol start> cause list numbering start at 0 (should start at 1)
https://bugs.webkit.org/show_bug.cgi?id=27810

Couldn't make a text-only test because list numbers do not show up in
text dumps. This matches the other fast/lists tests.

  • fast/lists/ol-start-parsing.html: Added.
  • platform/mac/fast/lists/ol-start-parsing-expected.checksum: Added.
  • platform/mac/fast/lists/ol-start-parsing-expected.png: Added.
  • platform/mac/fast/lists/ol-start-parsing-expected.txt: Added.
3:49 PM Changeset in webkit [46560] by bweinstein@apple.com
  • 2 edits in trunk/WebKit/win

Was doing the build fix wrong, NotImplemented.h is in WebCore.

3:32 PM Changeset in webkit [46559] by bweinstein@apple.com
  • 2 edits in trunk/WebKit/win

Windows build fix.

3:12 PM Changeset in webkit [46558] by kmccullough@apple.com
  • 21 edits
    1 add in trunk

WebCore:

2009-07-29 Kevin McCullough <kmccullough@apple.com>

Reviewed by Darin Adler.

Added foundation work to allow a testing infrastructure for the Web
Inspector.

  • inspector/InspectorClient.h:
  • inspector/InspectorController.cpp: (WebCore::InspectorController::scriptObjectReady): Tell the testing harness that the window object is ready.
  • loader/EmptyClients.h: Empty class support for SVG. (WebCore::EmptyInspectorClient::inspectorWindowObjectCleared):

WebKit/gtk:

2009-07-29 Kevin McCullough <kmccullough@apple.com>

Reviewed by Darin Adler.

Added foundation work to allow a testing infrastructure for the Web
Inspector.


  • WebCoreSupport/InspectorClientGtk.cpp: (WebKit::InspectorClient::inspectorWindowObjectCleared):
  • WebCoreSupport/InspectorClientGtk.h:

WebKit/mac:

2009-07-29 Kevin McCullough <kmccullough@apple.com>

Reviewed by Darin Adler.

Added foundation work to allow a testing infrastructure for the Web
Inspector.

  • WebCoreSupport/WebInspectorClient.h:
  • WebCoreSupport/WebInspectorClient.mm: (WebInspectorClient::inspectorWindowObjectCleared): Send the delegate callback.
  • WebView/WebDelegateImplementationCaching.h:
  • WebView/WebView.mm: (-[WebView _cacheFrameLoadDelegateImplementations]):
  • WebView/WebViewPrivate.h: The delegate SPI.

WebKit/qt:

2009-07-29 Kevin McCullough <kmccullough@apple.com>

Reviewed by Darin Adler.

Added foundation work to allow a testing infrastructure for the Web
Inspector.

  • WebCoreSupport/InspectorClientQt.cpp: (WebCore::InspectorClientQt::inspectorWindowObjectCleared):
  • WebCoreSupport/InspectorClientQt.h:

WebKit/win:

2009-07-29 Kevin McCullough <kmccullough@apple.com>

Reviewed by Darin Adler.

Added foundation work to allow a testing infrastructure for the Web
Inspector.

  • WebCoreSupport/WebInspectorClient.cpp: (WebInspectorClient::inspectorWindowObjectCleared): Empty definition to let Windows compile but we should implement this at some point so that the tests can be run on Windows too.
  • WebCoreSupport/WebInspectorClient.h:
2:45 PM Changeset in webkit [46557] by Chris Fleizach
  • 2 edits in trunk/WebCore

Fix Windows build breakage.

2:44 PM Changeset in webkit [46556] by bweinstein@apple.com
  • 3 edits in trunk/LayoutTests

2009-07-29 Brian Weinstein <bweinstein@apple.com>

Rubber-stamped by Adam Roben.

Update getComputedStyle expected results on Windows to match updated Mac results.

  • platform/win/fast/css/getComputedStyle/computed-style-expected.txt:
  • platform/win/fast/css/getComputedStyle/computed-style-without-renderer-expected.txt:
2:35 PM Changeset in webkit [46555] by Chris Fleizach
  • 6 edits in trunk/WebCore

Bug 27807 - AX: move re-usable code into more common areas for other platforms
https://bugs.webkit.org/show_bug.cgi?id=27807

1:53 PM Changeset in webkit [46554] by ddkilzer@apple.com
  • 2 edits in trunk/WebKitTools

<http://webkit.org/b/27815> bugzilla-tool: add -m/--description to post-commits command

Reviewed by Eric Seidel.

  • Scripts/bugzilla-tool: (PostDiffAsPatchToBug.execute): Changed default patch name from "patch" to "Patch v1". (PostCommitsAsPatchesToBug.init): Added -m/--description switch. (PostCommitsAsPatchesToBug.execute): Use description provided by -m/--description switch first, else fall back to the commit message.
1:35 PM Changeset in webkit [46553] by mitz@apple.com
  • 9 edits in trunk/WebCore

Reviewed by Simon Fraser.

Inset box shadows are incorrectly accounted for in visual overflow
computations
https://bugs.webkit.org/show_bug.cgi?id=27811

  • rendering/InlineFlowBox.cpp: (WebCore::InlineFlowBox::placeBoxesHorizontally): Use

getBoxShadowHorizontalExtent().

(WebCore::InlineFlowBox::placeBoxesVertically): Use

getBoxShadowVerticalExtent(). Removed duplicate code to get the
text-shadow overflow.

(WebCore::InlineFlowBox::paint): Use getBoxShadowHorizontalExtent().

  • rendering/RenderBlock.cpp: (WebCore::RenderBlock::overflowHeight): Use

getBoxShadowVerticalExtent().

(WebCore::RenderBlock::overflowWidth): Use

getBoxShadowHorizontalExtent().

(WebCore::RenderBlock::overflowLeft): Ditto.
(WebCore::RenderBlock::overflowTop): Use getBoxShadowVerticalExtent().
(WebCore::RenderBlock::overflowRect): Use getBoxShadowExtent().
(WebCore::RenderBlock::layoutBlock): Ditto.

  • rendering/RenderFlexibleBox.cpp: (WebCore::RenderFlexibleBox::layoutBlock): Ditto.
  • rendering/RenderLayer.cpp: (WebCore::RenderLayer::calculateRects): Exclude inset shadows.
  • rendering/RenderObject.cpp: (WebCore::RenderObject::repaintAfterLayoutIfNeeded): Use

getBoxShadowHorizontalExtent() and getBoxShadowVerticalExtent().

(WebCore::RenderObject::adjustRectForOutlineAndShadow): Exclude inset

shadows.

  • rendering/RenderReplaced.cpp: (WebCore::RenderReplaced::adjustOverflowForBoxShadowAndReflect): Ditto.
  • rendering/style/RenderStyle.cpp: (WebCore::RenderStyle::setTextShadow): Also assert that text-shadows are

not inset.

(WebCore::RenderStyle::getBoxShadowExtent): Added. Excludes inset

shadows.

(WebCore::RenderStyle::getBoxShadowHorizontalExtent): Ditto.
(WebCore::RenderStyle::getBoxShadowVerticalExtent): Ditto.

  • rendering/style/RenderStyle.h:
1:26 PM Changeset in webkit [46552] by beidson@apple.com
  • 2 edits in trunk/LayoutTests

2009-07-29 Brady Eidson <beidson@apple.com>

Rubber-stamped by Mark Rowe

  • platform/gtk/Skipped: Disable the new test on GTK, as they don't dump resource load callbacks in their DumpRenderTree.
1:13 PM Changeset in webkit [46551] by beidson@apple.com
  • 1 edit
    2 adds in trunk/LayoutTests

2009-07-29 Brady Eidson <beidson@apple.com>

Rubber-stamped by Jon Honeycutt

Check in Tiger-specific results for the new test, as it is known they send willCacheResponse
callbacks in a different order.

  • platform/mac-tiger/http/tests/loading: Added.
  • platform/mac-tiger/http/tests/loading/redirect-methods-expected.txt: Added.
1:07 PM Changeset in webkit [46550] by beidson@apple.com
  • 2 edits in trunk/LayoutTests

2009-07-29 Brady Eidson <beidson@apple.com>

Adding my new test to the Windows skipped list until I can figure out what's going on.

  • platform/win/Skipped:
12:48 PM Changeset in webkit [46549] by Simon Fraser
  • 4 edits
    2 adds in trunk

2009-07-29 Simon Fraser <Simon Fraser>

Reviewed by Dan Bernstein.

Fix crash when an element with display: table-row is composited (e.g. via a 3d-transform)
https://bugs.webkit.org/show_bug.cgi?id=27796

Avoid repainting when a layer becomes composited if the renderer is not parented
yet, because it makes no sense to do so.

If the table row is a repaint container, default to the RenderBox implementation
of clippedOverflowRectForRepaint(), because we cannot hand off the repaint rect
computation to something that is above the repaint container.

Test: fast/table/table-row-compositing-repaint-crash.html

  • rendering/RenderLayerCompositor.cpp: (WebCore::RenderLayerCompositor::repaintOnCompositingChange):
  • rendering/RenderTableRow.cpp: (WebCore::RenderTableRow::clippedOverflowRectForRepaint):
11:22 AM Changeset in webkit [46548] by beidson@apple.com
  • 4 edits
    4 adds in trunk

WebCore:

2009-07-29 Brady Eidson <beidson@apple.com>

Reviewed by Sam Weinig.

https://bugs.webkit.org/show_bug.cgi?id=27791
307 redirects of POSTs should use POST, not GET

Test: http/tests/loading/redirect-methods.html

  • platform/network/cf/ResourceHandleCFNet.cpp: (WebCore::willSendRequest):
  • platform/network/mac/ResourceHandleMac.mm: (-[WebCoreResourceHandleAsDelegate connection:willSendRequest:redirectResponse:]):

LayoutTests:

2009-07-29 Brady Eidson <beidson@apple.com>

Reviewed by Sam Weinig.

https://bugs.webkit.org/show_bug.cgi?id=27791
307 redirects of POSTs should use POST, not GET

  • http/tests/loading/redirect-methods-expected.txt: Added.
  • http/tests/loading/redirect-methods.html: Added.
  • http/tests/loading/resources/redirect-methods-form.html: Added.
  • http/tests/loading/resources/redirect-methods-result.php: Added.
11:03 AM Changeset in webkit [46547] by cmarrin@apple.com
  • 14 edits
    2 adds in trunk

Adding -webkit-animation-play-state back in
https://bugs.webkit.org/show_bug.cgi?id=26867

We've decided to keep -webkit-animation-play-state. So this
just adds back in the code from https://bugs.webkit.org/show_bug.cgi?id=22907.

10:02 AM Changeset in webkit [46546] by kenneth@webkit.org
  • 2 edits in trunk/WebCore

2009-07-29 Jakub Wieczorek <faw217@gmail.com>

Reviewed by Simon Hausmann.

Expose the default plugin directories and the current directory set of
the plugin database as public API.

https://bugs.webkit.org/show_bug.cgi?id=27651

  • plugins/PluginDatabase.h: (WebCore::PluginDatabase::pluginDirectories):
9:30 AM Changeset in webkit [46545] by levin@chromium.org
  • 2 edits in trunk/WebCore

2009-07-29 Alpha Lam <hclam@chromium.org>

Reviewed by David Levin.

[chromium] Font size for current time display in media controls panel
is affected by body font size.
https://bugs.webkit.org/show_bug.cgi?id=27799

Fixing the problem of rendering by explicitly setting the font size
for the time displays.

No new tests since this is covered by existing media tests.

  • css/mediaControlsChromium.css:
9:30 AM Changeset in webkit [46544] by levin@chromium.org
  • 2 edits in trunk/WebCore

2009-07-29 Mike Fenton <mike.fenton@torchmobile.com>

Reviewed by David Levin.

Update WebCore/page/ContextMenuController.cpp to conform to WebKit
Style Guidelines as identified by cpp_style.py.
https://bugs.webkit.org/show_bug.cgi?id=27613

  • page/ContextMenuController.cpp: (WebCore::openNewWindow): (WebCore::ContextMenuController::contextMenuItemSelected):
9:17 AM Changeset in webkit [46543] by Dimitri Glazkov
  • 2 edits in trunk/WebCore

2009-07-29 Dean McNamee <deanm@chromium.org>

Reviewed by Dimitri Glazkov.

Removed unused wrapCPointer/extractCPointer from the v8 bindings.
https://bugs.webkit.org/show_bug.cgi?id=27805

  • bindings/v8/V8DOMWrapper.h:
8:13 AM Changeset in webkit [46542] by kenneth@webkit.org
  • 2 edits in trunk/WebKitTools

2009-07-29 Kenneth Rohde Christiansen <kenneth@webkit.org>

Reviewed by Dave Levin.

Make check-webkit-style support files as arguments.
https://bugs.webkit.org/show_bug.cgi?id=27549

  • Scripts/check-webkit-style: Support files as argument. Change documentation and use basename on the binary name when used in the documentation. Also do not die when printing something containing non-ASCII characters.
7:54 AM Changeset in webkit [46541] by kenneth@webkit.org
  • 2 edits in trunk/WebCore

2009-07-29 Jakub Wieczorek <faw217@gmail.com>

Reviewed by Simon Hausmann.

Expose the PluginDatabase::pluginForMIMEType() function as public API.
https://bugs.webkit.org/show_bug.cgi?id=27651

It can be used to determine the appropriate plugin for a mime type,
without guessing the mimetype from the extension.

  • plugins/PluginDatabase.h:
7:36 AM Changeset in webkit [46540] by kenneth@webkit.org
  • 4 edits in trunk/WebCore

2009-07-29 Jakub Wieczorek <faw217@gmail.com>

Reviewed by Adam Treat.

Allow to enable/disable particular plugin packages.
https://bugs.webkit.org/show_bug.cgi?id=27651

Disabled plugins will not be picked up when looking for a plugin
supporting the requested mimetypes.

Client applications may want to use that API to disable specific
plugins.

  • plugins/PluginDatabase.cpp: (WebCore::PluginDatabase::pluginForMIMEType): (WebCore::PluginDatabase::MIMETypeForExtension):
  • plugins/PluginPackage.cpp: (WebCore::PluginPackage::PluginPackage): (WebCore::PluginPackage::setEnabled):
  • plugins/PluginPackage.h: (WebCore::PluginPackage::isEnabled):
6:18 AM Changeset in webkit [46539] by Nikolas Zimmermann
  • 12 edits in trunk

2009-07-29 Nikolas Zimmermann <nikolas.zimmermann@torchmobile.com>

Reviewed by Adam Treat.

[WML] Running WML tests in random order multiple times exposes subtle bugs
https://bugs.webkit.org/show_bug.cgi?id=27801

Some changes to fix random order WML tests, simplilfy WMLTestCase.js and reset testDocument
properly in enter-first-card-with-events.js. fast/wml/err-multi-access.wml still creates a layout
test difference on consecutive runs, though that's related to bug 27721, which has to be fixed first.

  • wml/resources/WMLTestCase.js:
  • wml/resources/enter-first-card-with-events.js: (setupTestDocument): (prepareTest): (executeTest):

2009-07-29 Nikolas Zimmermann <nikolas.zimmermann@torchmobile.com>

Reviewed by Adam Treat.

[WML] Running WML tests in random order multiple times exposes subtle bugs
https://bugs.webkit.org/show_bug.cgi?id=27801

Remove superflous assertions regarding the parent node. Under certain circumstances
these can even fire (related to garbage collection while destructing). Fixes random order
WML tests (run-webkit-tests fast/wml wml http/tests/wml fast/wml ... --random)

The wml/enter-first-card-with-events.html test relied on a bug in our implementation of
WMLPageState::reset() - the history stack should still contain the current card afterwards.
Fix that bug by preserving the first item in BackForwardList::clearWMLPageHistory().

  • history/BackForwardList.cpp: Preserve first item in history stack, as demanded by the spec. (WebCore::BackForwardList::clearWMLPageHistory):
  • wml/WMLDoElement.cpp: (WebCore::WMLDoElement::insertedIntoDocument): (WebCore::WMLDoElement::removedFromDocument):
  • wml/WMLNoopElement.cpp: (WebCore::WMLNoopElement::insertedIntoDocument):
  • wml/WMLOnEventElement.cpp: (WebCore::eventHandlingParent):
  • wml/WMLPostfieldElement.cpp: (WebCore::WMLPostfieldElement::insertedIntoDocument): (WebCore::WMLPostfieldElement::removedFromDocument):
  • wml/WMLSetvarElement.cpp: (WebCore::WMLSetvarElement::insertedIntoDocument): (WebCore::WMLSetvarElement::removedFromDocument):
  • wml/WMLTaskElement.cpp: (WebCore::WMLTaskElement::insertedIntoDocument): (WebCore::WMLTaskElement::removedFromDocument):
  • wml/WMLTimerElement.cpp: (WebCore::WMLTimerElement::insertedIntoDocument): (WebCore::WMLTimerElement::removedFromDocument):
4:35 AM Changeset in webkit [46538] by ddkilzer@apple.com
  • 3 edits in trunk/WebKitTools

<http://webkit.org/b/27119> bugzilla-tool: Add create-bug command

Reviewed by David Levin.

Implement "create-bug" command for bugzilla-tool.

  • Scripts/bugzilla-tool: Added CreateBug class. (CreateBug.init): Added. (CreateBug.create_bug_from_commit): Added. (CreateBug.create_bug_from_patch): Added. (CreateBug.prompt_for_bug_title_and_comment): Added. (CreateBug.execute): Added. (BugzillaTool.init): Added create-bug command.
  • Scripts/modules/bugzilla.py: (Bugzilla.prompt_for_component): Added. (Bugzilla.check_create_bug_response_returning_bug_id_on_success): Added. (Bugzilla.create_bug_with_patch): Added.
4:18 AM Changeset in webkit [46537] by jmalonzo@webkit.org
  • 12 edits
    1 add in trunk

2009-07-29 Jan Michael Alonzo <jmalonzo@webkit.org>

Reviewed by Eric Seidel and Xan Lopez.

[Gtk] Enable http/tests/appcache tests
https://bugs.webkit.org/show_bug.cgi?id=27674

  • GNUmakefile.am: Add webkit/webkitapplicationcache.cpp

Enable passing http/tests/appcache tests.

  • platform/gtk/Skipped:

Add 'enable-offline-web-application-cache' WebSetting to enable
or disable the appcache. Also, Add appcache SPI for DRT to use.

  • webkit/webkitapplicationcache.cpp: Added. (webkit_application_cache_set_maximum_size):
  • webkit/webkitprivate.cpp: (webkit_init):
  • webkit/webkitprivate.h:
  • webkit/webkitwebsettings.cpp: (webkit_web_settings_class_init): (webkit_web_settings_set_property): (webkit_web_settings_get_property): (webkit_web_settings_copy):
  • webkit/webkitwebview.cpp: (webkit_web_view_update_settings): (webkit_web_view_settings_notify):

Add support for testing http/tests/appcache tests.

  • DumpRenderTree/gtk/DumpRenderTree.cpp: (resetWebViewToConsistentStateBeforeTesting):
  • DumpRenderTree/gtk/LayoutTestControllerGtk.cpp: (LayoutTestController::setAppCacheMaximumSize):
3:50 AM Changeset in webkit [46536] by Simon Hausmann
  • 3 edits in trunk/LayoutTests

2009-07-29 Simon Hausmann <simon.hausmann@nokia.com>

Update Qt-specific results for computed-style tests, like in r46512.

  • platform/qt/fast/css/getComputedStyle/computed-style-expected.txt:
  • platform/qt/fast/css/getComputedStyle/computed-style-without-renderer-expected.txt:
3:33 AM Changeset in webkit [46535] by Simon Hausmann
  • 8 edits in trunk/WebCore

2009-07-29 Yongjun Zhang <yongjun.zhang@nokia.com>

Reviewed by Simon Hausmann.

https://bugs.webkit.org/show_bug.cgi?id=26848
[Qt] ResourceHandle::willLoadFromCache needs to be implemented QtWebKit.

  • loader/FrameLoader.cpp: (WebCore::FrameLoader::loadItem):
  • platform/network/ResourceHandle.h:
  • platform/network/cf/ResourceHandleCFNet.cpp: (WebCore::ResourceHandle::willLoadFromCache):
  • platform/network/curl/ResourceHandleCurl.cpp: (WebCore::ResourceHandle::willLoadFromCache):
  • platform/network/mac/ResourceHandleMac.mm: (WebCore::ResourceHandle::willLoadFromCache):
  • platform/network/qt/ResourceHandleQt.cpp: (WebCore::ResourceHandle::willLoadFromCache):
  • platform/network/soup/ResourceHandleSoup.cpp: (WebCore::):
2:53 AM Changeset in webkit [46534] by levin@chromium.org
  • 2 edits in trunk/WebCore

2009-07-29 Alpha Lam <hclam@chromium.org>

Reviewed by David Levin.

Media control panel for <video> in MediaDocument is mis-placed
https://bugs.webkit.org/show_bug.cgi?id=27798

Fixing a rendering problem: When <video> is displayed in MediaDocument,
the media control panel overlaps with the video by 16 pixels.

No new tests as this is covered by existing media tests.

  • css/mediaControlsChromium.css:
2:48 AM Changeset in webkit [46533] by vestbo@webkit.org
  • 1 edit in trunk/WebKitTools/Scripts/webkitdirs.pm

Revert r46532 (build-webkit: Reorder logic for determining the baseProductDir)

Broke the Windows build.

2:37 AM Changeset in webkit [46532] by vestbo@webkit.org
  • 2 edits in trunk/WebKitTools

2009-07-29 Tor Arne Vestbø <tor.arne.vestbo@nokia.com>

Reviewed by Eric Seidel.

build-webkit: Reorder logic for determining the baseProductDir

https://bugs.webkit.org/show_bug.cgi?id=27699

We now do port-spesific probing for the product dir first, and then
fall back to either the generic WEBKITOUTPUTDIR (which now also works
on Mac), or WebKitBuild.

Then we add the git branch if the git branch build option is enabled.

Finally we massage the product dir based on the port. For example the
Mac port will now always set the SYMROOT and OBJROOT configuration for
XCode.

  • Scripts/webkitdirs.pm:
1:36 AM Changeset in webkit [46531] by abarth@webkit.org
  • 2 edits in trunk/WebCore

2009-07-29 Adam Barth <abarth@webkit.org>

Unreviewed build fix for Chromium. Those last two patches weren't
independent despite the clean merge.

  • bindings/v8/V8AbstractEventListener.cpp: (WebCore::V8AbstractEventListener::V8AbstractEventListener):
1:34 AM Changeset in webkit [46530] by Simon Hausmann
  • 2 edits in trunk/WebKit/qt

2009-07-29 Zoltan Herczeg <zherczeg@inf.u-szeged.hu>

Reviewed by Simon Hausmann.

[Qt] In debug mode, this fix frees cached resources
and other unreferenced objects (by calling the
garbage collector) to remove the LEAK messages.
https://bugs.webkit.org/show_bug.cgi?id=27767

  • Api/qwebframe.cpp: (qt_drt_cache_clear):
  • QtLauncher/main.cpp: (main):
1:11 AM Changeset in webkit [46529] by abarth@webkit.org
  • 9 edits
    3 adds in trunk

2009-07-29 Avi Drissman <avi@chromium.org>

Reviewed by Darin Adler.

ImageSourceCG makes bad data refs (race condition causes blank images)
https://bugs.webkit.org/show_bug.cgi?id=27777

Make ImageSourceCG guarantee that the lifetime of the SharedBuffer that
backs the CFDataRef will be long enough.

No new tests, as this fixes a bug with a race condition that is
difficult to trigger.

  • platform/graphics/cg/ImageSourceCG.cpp: (WebCore::ImageSource::setData):

2009-07-29 Kent Tamura <tkent@chromium.org>

Reviewed by Nikolas Zimmermann.

Fix a bug that HTMLOptionElement::value() returns an incorrect
value in a case that the element has a label attribute and no
value attribute.
https://bugs.webkit.org/show_bug.cgi?id=27760

  • fast/forms/option-value-and-label-expected.txt: Added.
  • fast/forms/option-value-and-label.html: Added.
1:04 AM Changeset in webkit [46528] by levin@chromium.org
  • 15 edits in trunk/JavaScriptCore

2009-07-29 Laszlo Gombos <Laszlo Gombos>

Reviewed by George Staikos.

Resolve class/struct mixup in forward declarations
https://bugs.webkit.org/show_bug.cgi?id=27708

  • API/JSClassRef.h:
  • bytecode/SamplingTool.h:
  • interpreter/Interpreter.h:
  • jit/JIT.h:
  • profiler/ProfileGenerator.h:
  • profiler/Profiler.h:
  • runtime/ClassInfo.h:
  • runtime/ExceptionHelpers.h:
  • runtime/JSByteArray.h:
  • runtime/JSCell.h:
  • runtime/JSFunction.h:
  • runtime/JSGlobalData.h:
  • runtime/JSObject.h:
  • runtime/JSString.h:
1:04 AM Changeset in webkit [46527] by levin@chromium.org
  • 2 edits in trunk/WebCore

2009-07-29 Avi Drissman <avi@chromium.org>

Reviewed by Darin Adler.

ImageSourceCG makes bad data refs (race condition causes blank images)
https://bugs.webkit.org/show_bug.cgi?id=27777

Make ImageSourceCG guarantee that the lifetime of the SharedBuffer that
backs the CFDataRef will be long enough.

No new tests, as this fixes a bug with a race condition that is
difficult to trigger.

  • platform/graphics/cg/ImageSourceCG.cpp: (WebCore::ImageSource::setData):
1:04 AM Changeset in webkit [46526] by levin@chromium.org
  • 3 edits in trunk/WebCore

2009-07-29 Keishi Hattori <casey.hattori@gmail.com>

Reviewed by Timothy Hatcher.

Inspector: Console should show completions for the command line APIs
https://bugs.webkit.org/show_bug.cgi?id=27696

  • inspector/front-end/Console.js: (WebInspector.Console.prototype._reportCompletions): Adds properties from _inspectorCommandLineAPI to completions when available.
  • inspector/front-end/utilities.js: (Object.properties): Added. (Object.sortedProperties):
1:03 AM Changeset in webkit [46525] by levin@chromium.org
  • 2 edits in trunk/WebCore

2009-07-29 Shinichiro Hamaji <hamaji@chromium.org>

Reviewed by Eric Seidel.

Refactor the first step of layout in RenderFlexibleBox.cpp
https://bugs.webkit.org/show_bug.cgi?id=27704

No new tests as this change is just a refactoring.

  • rendering/RenderFlexibleBox.cpp: (WebCore::gatherFlexChildrenInfo): (WebCore::RenderFlexibleBox::layoutHorizontalBox): (WebCore::RenderFlexibleBox::layoutVerticalBox):
1:03 AM Changeset in webkit [46524] by levin@chromium.org
  • 6 edits
    2 adds in trunk

2009-07-29 Kent Tamura <tkent@chromium.org>

Reviewed by Nikolas Zimmermann.

Fix a bug that HTMLOptionElement::value() returns an incorrect
value in a case that the element has a label attribute and no
value attribute.
https://bugs.webkit.org/show_bug.cgi?id=27760

WebCore:

Test: fast/forms/option-value-and-label.html

  • dom/OptionElement.cpp: (WebCore::OptionElement::collectOptionLabelOrText): (WebCore::OptionElement::collectOptionInnerText): (WebCore::OptionElement::normalizeText): (WebCore::OptionElement::collectOptionTextRespectingGroupLabel): (WebCore::OptionElement::collectOptionValue):
  • dom/OptionElement.h:
  • html/HTMLOptionElement.cpp: (WebCore::HTMLOptionElement::text):
  • wml/WMLOptionElement.cpp: (WebCore::WMLOptionElement::text):

LayoutTests:

  • fast/forms/option-value-and-label-expected.txt: Added.
  • fast/forms/option-value-and-label.html: Added.
12:59 AM Changeset in webkit [46523] by abarth@webkit.org
  • 15 edits
    2 adds in trunk

2009-07-29 Adam Barth <abarth@webkit.org>

Reviewed by Dimitri Glazkov.

[V8] Teach V8Proxy::context about isolated worlds
https://bugs.webkit.org/show_bug.cgi?id=27701

Change V8Proxy::context(Frame*) to understand isolated worlds. Audit
all callers of this method to make sure they want isolated worlds. In
cases where we really want the main world, I've changed the call to
V8Proxy::mainWorldContext(Frame*).


The main visible change is to the document.open method when called with
more than two arguments. This design seems more likely to lead to
future correct code.

Test: http/tests/security/isolatedWorld/document-open.html

  • bindings/v8/ScriptController.cpp: (WebCore::ScriptController::processingUserGesture): (WebCore::ScriptController::evaluate): (WebCore::ScriptController::bindToWindowObject): (WebCore::ScriptController::collectGarbage): (WebCore::createScriptObject): (WebCore::ScriptController::createScriptObjectForPluginElement):
  • bindings/v8/ScriptObjectQuarantine.cpp: (WebCore::getQuarantinedScriptObject):
  • bindings/v8/V8AbstractEventListener.cpp: (WebCore::V8AbstractEventListener::handleEvent):
  • bindings/v8/V8DOMWrapper.cpp: (WebCore::V8DOMWrapper::getConstructor): (WebCore::V8DOMWrapper::setHiddenWindowReference): (WebCore::V8DOMWrapper::convertNodeToV8Object): (WebCore::V8DOMWrapper::convertWindowToV8Object):
  • bindings/v8/V8Helpers.cpp: (WebCore::toV8Context):
  • bindings/v8/V8LazyEventListener.cpp: (WebCore::V8LazyEventListener::getListenerFunction): (WebCore::V8LazyEventListener::getWrappedListenerFunction):
  • bindings/v8/V8Proxy.cpp: (WebCore::V8Proxy::context): (WebCore::V8Proxy::mainWorldContext): (WebCore::V8Proxy::bindJsObjectToWindow):
  • bindings/v8/V8Proxy.h:

2009-07-29 Adam Barth <abarth@webkit.org>

Reviewed by Dimitri Glazkov.

[V8] Teach V8Proxy::context about isolated worlds
https://bugs.webkit.org/show_bug.cgi?id=27701

Test that calling document.open with more than two arguments calls the
window.open method from the correct world.

  • http/tests/security/isolatedWorld/document-open-expected.txt: Added.
  • http/tests/security/isolatedWorld/document-open.html: Added.
12:25 AM Changeset in webkit [46522] by Simon Hausmann
  • 2 edits in trunk/WebCore

2009-07-29 Balazs Kelemen <kelemen.balazs.3@stud.u-szeged.hu>

Reviewed by Simon Hausmann.

Fix the Qt build, add missing file.

  • WebCore.pro:
12:13 AM Changeset in webkit [46521] by jhoneycutt@apple.com
  • 2 edits in trunk/WebCore

2009-07-28 Jon Honeycutt <jhoneycutt@apple.com>

Speculative fix for <rdar://problem/7005077> WER: Crash in
WebCore::PluginStream::destroyStream+279 (1310510882)

Reviewed by Oliver Hunt.

  • plugins/PluginStream.cpp: (WebCore::PluginStream::destroyStream): Move the "protector" RefPtr out of the block that dispatches notifications and into the function level; if NPN_DestroyStream were called from NPP_NewStream as the comment warns, we would be deleted at the end of the block.
Note: See TracTimeline for information about the timeline view.