WebKit-https.git
11 years ago[Qt] Implement textZoomIn() and textZoomOut() in DRT's EventSender, add results
abecsi@webkit.org [Fri, 19 Feb 2010 17:13:38 +0000 (17:13 +0000)]
[Qt] Implement textZoomIn() and textZoomOut() in DRT's EventSender, add results
for passing new tests and unskip the passing ones from the Skipped list.
https://bugs.webkit.org/show_bug.cgi?id=35159

Reviewed by Kenneth Rohde Christiansen.

WebKitTools:

* DumpRenderTree/qt/EventSenderQt.cpp:
(EventSender::textZoomIn):
(EventSender::textZoomOut):
* DumpRenderTree/qt/EventSenderQt.h:

LayoutTests:

* platform/qt/Skipped: unskip passing tests
* platform/qt/svg/custom/text-zoom-expected.txt: Added.
* platform/qt/svg/zoom: Added.
* platform/qt/svg/zoom/page: Added.
* platform/qt/svg/zoom/page/absolute-sized-document-no-scrollbars-expected.txt: Added.
* platform/qt/svg/zoom/page/absolute-sized-document-scrollbars-expected.txt: Added.
* platform/qt/svg/zoom/page/relative-sized-document-scrollbars-expected.txt: Added.
* platform/qt/svg/zoom/page/zoom-foreignObject-expected.txt: Added.
* platform/qt/svg/zoom/page/zoom-hixie-mixed-008-expected.txt: Added.
* platform/qt/svg/zoom/page/zoom-hixie-mixed-009-expected.txt: Added.
* platform/qt/svg/zoom/page/zoom-hixie-rendering-model-004-expected.txt: Added.
* platform/qt/svg/zoom/page/zoom-svg-float-border-padding-expected.txt: Added.
* platform/qt/svg/zoom/text: Added.
* platform/qt/svg/zoom/text/absolute-sized-document-no-scrollbars-expected.txt: Added.
* platform/qt/svg/zoom/text/absolute-sized-document-scrollbars-expected.txt: Added.
* platform/qt/svg/zoom/text/relative-sized-document-scrollbars-expected.txt: Added.
* platform/qt/svg/zoom/text/zoom-foreignObject-expected.txt: Added.
* platform/qt/svg/zoom/text/zoom-hixie-mixed-008-expected.txt: Added.
* platform/qt/svg/zoom/text/zoom-hixie-mixed-009-expected.txt: Added.
* platform/qt/svg/zoom/text/zoom-hixie-rendering-model-004-expected.txt: Added.
* platform/qt/svg/zoom/text/zoom-svg-float-border-padding-expected.txt: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@55019 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-02-19 Marcus Bulach <bulach@chromium.org>
eric@webkit.org [Fri, 19 Feb 2010 16:57:38 +0000 (16:57 +0000)]
2010-02-19  Marcus Bulach  <bulach@chromium.org>

        Reviewed by Jeremy Orlow.

        Fixes GeolocationServiceBridgeImpl::stopUpdating (it can be called multiple times).

        * src/GeolocationServiceBridgeChromium.cpp:
        (WebKit::GeolocationServiceBridgeImpl::stopUpdating):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@55018 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-02-19 Philippe Normand <pnormand@igalia.com>
philn@webkit.org [Fri, 19 Feb 2010 16:50:16 +0000 (16:50 +0000)]
2010-02-19  Philippe Normand  <pnormand@igalia.com>

        Rubber-stamped by Gustavo Noronha Silva.

        [GTK] fast/forms/restore-selection-after-layout.html fails on 32bits Debug build
        https://bugs.webkit.org/show_bug.cgi?id=35158

        * platform/gtk/Skipped: Skip failing test

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@55017 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-02-19 Alexander Pavlov <apavlov@chromium.org>
eric@webkit.org [Fri, 19 Feb 2010 15:16:10 +0000 (15:16 +0000)]
2010-02-19  Alexander Pavlov  <apavlov@chromium.org>

        Reviewed by Pavel Feldman.

        WebInspector: Elements panel: Correctly show empty elements' ending tags
        for XML and HTML documents.
        https://bugs.webkit.org/show_bug.cgi?id=26315

        * inspector/console-dirxml-expected.txt:
        * inspector/console-format-collections-expected.txt:
        * inspector/console-format-expected.txt:
        * inspector/elements-panel-limited-children-expected.txt:
        * inspector/elements-panel-structure-expected.txt:
        Re-baselined due to closing tags and ellipses between opening and closing tags

        * inspector/elements-panel-xhtml-structure-expected.txt: Added.
        * inspector/elements-panel-xhtml-structure.xhtml: Added.
2010-02-19  Alexander Pavlov  <apavlov@chromium.org>

        Reviewed by Pavel Feldman.

        WebInspector: Elements panel: Correctly show empty elements' ending tags
        for XML and HTML documents.
        https://bugs.webkit.org/show_bug.cgi?id=26315

        Test: inspector/elements-panel-xhtml-structure.xhtml

        * inspector/InspectorController.cpp:
        (WebCore::InspectorController::populateScriptObjects):
        * inspector/front-end/ElementsTreeOutline.js:
        (WebInspector.ElementsTreeOutline.prototype.set rootDOMNode):
        (WebInspector.ElementsTreeOutline.prototype.get isXMLMimeType):
        (WebInspector.ElementsTreeOutline.prototype._contextMenuEventFired):
        (WebInspector.ElementsTreeElement.prototype.onexpand):
        (WebInspector.ElementsTreeElement.prototype.oncollapse):
        (WebInspector.ElementsTreeElement.prototype.updateTitle.callback):
        (WebInspector.ElementsTreeElement.prototype.updateTitle):
        (WebInspector.ElementsTreeElement.prototype._nodeTitleInfo):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@55016 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago Unreviewed Changelog typo fix.
abecsi@webkit.org [Fri, 19 Feb 2010 14:52:01 +0000 (14:52 +0000)]
Unreviewed Changelog typo fix.

    r5993 should be r54993.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@55015 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago Rubber-stamped by Kenneth Rohde Christiansen.
abecsi@webkit.org [Fri, 19 Feb 2010 14:45:59 +0000 (14:45 +0000)]
Rubber-stamped by Kenneth Rohde Christiansen.

    [Qt] plugins/set-status.html introduced in r5993 fails skipping it until fix.
    https://bugs.webkit.org/show_bug.cgi?id=35144

    * platform/qt/Skipped:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@55014 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-02-19 Laszlo Gombos <laszlo.1.gombos@nokia.com>
eric@webkit.org [Fri, 19 Feb 2010 14:34:39 +0000 (14:34 +0000)]
2010-02-19  Laszlo Gombos  <laszlo.1.gombos@nokia.com>

        Reviewed by Eric Seidel.

        [Qt] Clean-up the handling of HTML5 persistent data for LayoutTests
        https://bugs.webkit.org/show_bug.cgi?id=35004

        DumpRenderTreeQt.cpp calls QWebSettings::enablePersistentStorage which sets up
        and enables all HTML5 persistent data. All the other calls for setting the
        persistent path or enabling the persistent feature are redundant.

        * DumpRenderTree/qt/DumpRenderTreeQt.cpp:
        (WebCore::WebPage::WebPage): No need to enable AppCache again.
        * DumpRenderTree/qt/main.cpp:
        (main): No need to setup and enable HTML5 LocalStorage again.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@55013 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-02-19 Diego Gonzalez <diego.gonzalez@openbossa.org>
eric@webkit.org [Fri, 19 Feb 2010 13:13:19 +0000 (13:13 +0000)]
2010-02-19  Diego Gonzalez  <diego.gonzalez@openbossa.org>

        Reviewed by Kenneth Rohde Christiansen.

        Send the context menu event from contextClick() of EventSender
        and do not show context menu in DRT.

        LayoutTests:
            fast/events/contextmenu-scrolled-page-with-frame.html

        [Qt] DRT: Send context menu event from EventSender
        https://bugs.webkit.org/show_bug.cgi?id=35131

        * platform/qt/Skipped:
2010-02-19  Diego Gonzalez  <diego.gonzalez@openbossa.org>

        Reviewed by Kenneth Rohde Christiansen.

        Send the context menu event from contextClick() of EventSender
        and do not show context menu in DRT.

        LayoutTests:
            fast/events/contextmenu-scrolled-page-with-frame.html

        [Qt] DRT: Send context menu event from EventSender
        https://bugs.webkit.org/show_bug.cgi?id=35131

        * DumpRenderTree/qt/DumpRenderTreeQt.cpp:
        (WebCore::DumpRenderTree::DumpRenderTree):
        * DumpRenderTree/qt/EventSenderQt.cpp:
        (EventSender::contextClick):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@55012 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-02-19 Pavel Feldman <pfeldman@chromium.org>
pfeldman@chromium.org [Fri, 19 Feb 2010 13:01:10 +0000 (13:01 +0000)]
2010-02-19  Pavel Feldman  <pfeldman@chromium.org>

        Not reviewed. Chromium build fix: reverting r54997 and r54998.
        This is a two-sided change that should be coordinated.

        * bindings/v8/custom/V8LocationCustom.cpp:
        (WebCore::V8Location::protocolAccessorSetter):
        * platform/KURLGoogle.cpp:
        (WebCore::KURL::setProtocol):
        (WebCore::KURL::isHierarchical):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@55011 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago[Qt] editing tests fails after r54980
ossy@webkit.org [Fri, 19 Feb 2010 12:41:53 +0000 (12:41 +0000)]
[Qt] editing tests fails after r54980
https://bugs.webkit.org/show_bug.cgi?id=35145

Reviewed by Kenneth Rohde Christiansen.

After r54980 two tests fail, skip them until fix.
Another two tests pass, they have same results as Win port.

* platform/qt/Skipped:
 - editing/selection/extend-selection.html skipped.
 - editing/selection/extend-after-mouse-selection.html skipped.
* platform/qt/editing/selection/click-in-margins-inside-editable-div-expected.txt: Copied from LayoutTests/platform/win/editing/selection/click-in-margins-inside-editable-div-expected.txt.
* platform/qt/editing/selection/click-in-padding-with-multiple-line-boxes-expected.txt: Copied from LayoutTests/platform/win/editing/selection/click-in-padding-with-multiple-line-boxes-expected.txt.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@55010 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-02-19 Pavel Feldman <pfeldman@chromium.org>
pfeldman@chromium.org [Fri, 19 Feb 2010 12:40:14 +0000 (12:40 +0000)]
2010-02-19  Pavel Feldman  <pfeldman@chromium.org>

        Not reviewed. Chromium tests fix: update devtools code to match new
        web inspector object proxy format.

        * src/js/DebuggerAgent.js:
        (devtools.DebuggerAgent.prototype.formatCallFrame_):
        (devtools.DebuggerAgent.formatObjectProxy_):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@55009 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-02-19 Philippe Normand <pnormand@igalia.com>
philn@webkit.org [Fri, 19 Feb 2010 12:06:23 +0000 (12:06 +0000)]
2010-02-19  Philippe Normand  <pnormand@igalia.com>

        Rubber-stamped by Xan Lopez.

        [GTK] plugins/set-status.html fails
        https://bugs.webkit.org/show_bug.cgi?id=35144

        * platform/gtk/Skipped: Skip failing test.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@55008 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-02-19 Philippe Normand <pnormand@igalia.com>
philn@webkit.org [Fri, 19 Feb 2010 09:18:53 +0000 (09:18 +0000)]
2010-02-19  Philippe Normand  <pnormand@igalia.com>

        Rubber-stamped by Xan Lopez.

        Adding a GTK layout test result for a new test
        added in r54980.

        * platform/gtk/editing/selection/extend-after-mouse-selection-expected.txt: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@55007 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-02-19 Philippe Normand <pnormand@igalia.com>
philn@webkit.org [Fri, 19 Feb 2010 09:18:38 +0000 (09:18 +0000)]
2010-02-19  Philippe Normand  <pnormand@igalia.com>

        Rubber-stamped by Xan Lopez.

        Adding a GTK layout test result for a new test
        added in r54982.

        * platform/gtk/fast/repaint/selection-gap-overflow-scroll-2-expected.txt: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@55006 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-02-17 Philippe Normand <pnormand@igalia.com>
philn@webkit.org [Fri, 19 Feb 2010 08:30:14 +0000 (08:30 +0000)]
2010-02-17  Philippe Normand  <pnormand@igalia.com>

        Reviewed by Eric Seidel.

        [GTK] RTP/RTSP streams playback support
        https://bugs.webkit.org/show_bug.cgi?id=33662

        Added live pipelines support in updateStates().

        * manual-tests/video-rtsp.html: Added.
        * platform/graphics/gtk/MediaPlayerPrivateGStreamer.cpp:
        (WebCore::MediaPlayerPrivate::seek):
        (WebCore::MediaPlayerPrivate::updateStates):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@55005 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoRubber-stamped by Simon Hausmann.
ossy@webkit.org [Fri, 19 Feb 2010 07:32:44 +0000 (07:32 +0000)]
Rubber-stamped by Simon Hausmann.

[Qt] Update platform specific expected files after r54991.

* platform/qt/svg/css/circle-in-mask-with-shadow-expected.txt: updated.
* platform/qt/svg/css/mask-with-shadow-expected.txt: updated.
* platform/qt/svg/custom/empty-mask-expected.txt: updated.
* platform/qt/svg/custom/mask-with-default-value-expected.txt: updated.
* platform/qt/svg/custom/transformedMaskFails-expected.txt: updated.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@55003 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-02-18 Oliver Hunt <oliver@apple.com>
oliver@apple.com [Fri, 19 Feb 2010 06:23:25 +0000 (06:23 +0000)]
2010-02-18  Oliver Hunt  <oliver@apple.com>

        Reviewed by Gavin Barraclough.

        Improve interpreter getter performance
        https://bugs.webkit.org/show_bug.cgi?id=35138

        Improve the performance of getter dispatch by making it possible
        for the interpreter to cache the GetterSetter object lookup.

        To do this we simply need to make PropertySlot aware of getters
        as a potentially cacheable property, and record the base and this
        objects for a getter access.  This allows us to use more-or-less
        identical code to that used by the normal get_by_id caching, with
        the dispatch being the only actual difference.

        I'm holding off of implementing this in the JIT until I do some
        cleanup to try and making coding in the JIT not be as horrible
        as it is currently.

        * bytecode/CodeBlock.cpp:
        (JSC::CodeBlock::dump):
        (JSC::CodeBlock::derefStructures):
        (JSC::CodeBlock::refStructures):
        * bytecode/Opcode.h:
        * interpreter/Interpreter.cpp:
        (JSC::Interpreter::resolveGlobal):
        (JSC::Interpreter::tryCacheGetByID):
        (JSC::Interpreter::privateExecute):
        * jit/JIT.cpp:
        (JSC::JIT::privateCompileMainPass):
        * jit/JITStubs.cpp:
        (JSC::JITThunks::tryCacheGetByID):
        (JSC::DEFINE_STUB_FUNCTION):
        * runtime/JSObject.cpp:
        (JSC::JSObject::fillGetterPropertySlot):
        * runtime/PropertySlot.cpp:
        (JSC::PropertySlot::functionGetter):
        * runtime/PropertySlot.h:
        (JSC::PropertySlot::isGetter):
        (JSC::PropertySlot::isCacheable):
        (JSC::PropertySlot::isCacheableValue):
        (JSC::PropertySlot::setValueSlot):
        (JSC::PropertySlot::setGetterSlot):
        (JSC::PropertySlot::setCacheableGetterSlot):
        (JSC::PropertySlot::clearOffset):
        (JSC::PropertySlot::thisValue):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@55002 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-02-18 Chris Jerdonek <cjerdonek@webkit.org>
eric@webkit.org [Fri, 19 Feb 2010 05:17:04 +0000 (05:17 +0000)]
2010-02-18  Chris Jerdonek  <cjerdonek@webkit.org>

        Reviewed by Shinichiro Hamaji.

        Moved parsing-related code to a separate file. Also increased
        the unit test coverage in some affected areas.

        https://bugs.webkit.org/show_bug.cgi?id=34675

        This revision contains no new functionality.

        * Scripts/check-webkit-style:
          - Adjusted to call check_webkit_style_parser().

        * Scripts/webkitpy/style/checker.py:
          - Added check_webkit_style_parser() to pass checker.py
            configuration settings to optparser.py.
          - Moved _create_usage() and the CommandOptionValues,
            DefaultCommandOptionValues, ArgumentPrinter, and
            ArgumentParser classes to optparser.py.

        * Scripts/webkitpy/style/checker_unittest.py:
          - Moved the ProcessorOptionsTest, ArgumentPrinterTest, and
            ArgumentParserTest classes to optparser.py.
          - Added the CheckWebKitStyleFunctionTest class to check
            the check_webkit_style_configuration() and
            check_webkit_style_parser() code paths.

        * Scripts/webkitpy/style/optparser.py: Added.
          - From checker.py, added _create_usage() and the
            CommandOptionValues, DefaultCommandOptionValues,
            ArgumentPrinter, and ArgumentParser classes.
          - In the ArgumentParser constructor--
            - Added all_categories as a required parameter.
            - Removed the default value from the default_options parameter.

        * Scripts/webkitpy/style/optparser_unittest.py: Added.
          - From checker_unittest.py, added the ProcessorOptionsTest,
            ArgumentPrinterTest, and ArgumentParserTest classes.
          - Added the CreateUsageTest class to test _create_usage().

        * Scripts/webkitpy/style/unittests.py:
          - Added optparser_unittest import.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@55001 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-02-18 Yaar Schnitman <yaar@chromium.org>
eric@webkit.org [Fri, 19 Feb 2010 05:00:52 +0000 (05:00 +0000)]
2010-02-18  Yaar Schnitman  <yaar@chromium.org>

        Reviewed by Darin Adler.

        Normalize custom ctors for Image, Option, Audio
        https://bugs.webkit.org/show_bug.cgi?id=34782

        * fast/js/custom-constructors-expected.txt: Added.
        * fast/js/custom-constructors.html: Added.
        * fast/js/script-tests/custom-constructors.js: Added.
2010-02-18  Yaar Schnitman  <yaar@chromium.org>

        Reviewed by Darin Adler.

        Normalize custom ctors for Image, Option, Audio
        https://bugs.webkit.org/show_bug.cgi?id=34782

        Test: fast/js/custom-constructors.html

        * bindings/js/JSAudioConstructor.cpp:
        (WebCore::JSAudioConstructor::JSAudioConstructor):
        (WebCore::constructAudio):
        * bindings/js/JSImageConstructor.cpp:
        (WebCore::constructImage):
        * bindings/js/JSOptionConstructor.cpp:
        (WebCore::constructHTMLOptionElement):
        * bindings/v8/custom/V8HTMLAudioElementConstructor.cpp:
        (WebCore::v8HTMLAudioElementConstructorCallback):
        * bindings/v8/custom/V8HTMLImageElementConstructor.cpp:
        (WebCore::v8HTMLImageElementConstructorCallback):
        * bindings/v8/custom/V8HTMLOptionElementConstructor.cpp:
        (WebCore::v8HTMLOptionElementConstructorCallback):
        * html/HTMLAudioElement.cpp:
        (WebCore::HTMLAudioElement::HTMLAudioElement):
        (WebCore::HTMLAudioElement::createForJSConstructor):
        * html/HTMLAudioElement.h:
        * html/HTMLImageElement.cpp:
        (WebCore::HTMLImageElement::createForJSConstructor):
        (WebCore::HTMLImageElement::mapToEntry):
        (WebCore::HTMLImageElement::createRenderer):
        (WebCore::HTMLImageElement::attach):
        (WebCore::HTMLImageElement::width):
        (WebCore::HTMLImageElement::height):
        (WebCore::HTMLImageElement::naturalHeight):
        * html/HTMLImageElement.h:
        * html/HTMLOptionElement.cpp:
        (WebCore::HTMLOptionElement::HTMLOptionElement):
        (WebCore::HTMLOptionElement::createForJSConstructor):
        (WebCore::HTMLOptionElement::ownerSelectElement):
        (WebCore::HTMLOptionElement::nonRendererRenderStyle):
        (WebCore::HTMLOptionElement::disabled):
        (WebCore::HTMLOptionElement::insertedIntoTree):
        * html/HTMLOptionElement.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@54999 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-02-12 Kent Tamura <tkent@chromium.org>
tkent@chromium.org [Fri, 19 Feb 2010 04:29:53 +0000 (04:29 +0000)]
2010-02-12  Kent Tamura  <tkent@chromium.org>

        Unreviewed. Fix Chromium build.

        * DEPS: Bump Google URL revision to 123.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@54998 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-02-12 Brett Wilson <brettw@chromium.org>
brettw@chromium.org [Fri, 19 Feb 2010 02:06:14 +0000 (02:06 +0000)]
2010-02-12  Brett Wilson  <brettw@chromium.org>

        Reviewed by Adam Barth.

        Update the Google-URL version of KURL and the V8 bindings to the new
        behavior of KURL.IsStandard.

        https://bugs.webkit.org/show_bug.cgi?id=34859

        This is covered by fast/dom/Window/invalid-protocol.html

        * bindings/v8/custom/V8LocationCustom.cpp:
        (WebCore::V8Location::protocolAccessorSetter):
        * platform/KURLGoogle.cpp:
        (WebCore::KURL::setProtocol):
        (WebCore::KURL::isHierarchical):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@54997 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago Reviewed by Dan Bernstein.
ap@apple.com [Fri, 19 Feb 2010 01:21:37 +0000 (01:21 +0000)]
    Reviewed by Dan Bernstein.

        https://bugs.webkit.org/show_bug.cgi?id=35134
        <rdar://problem/7246280> Crash when a plugin calls NPN_SetStatus(0)

        * DumpRenderTree/TestNetscapePlugIn.subproj/PluginObject.cpp:
        (testSetStatus):
        (pluginInvoke):
        Added a setStatus() method.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@54996 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-02-18 Simon Fraser <simon.fraser@apple.com>
simon.fraser@apple.com [Fri, 19 Feb 2010 00:47:41 +0000 (00:47 +0000)]
2010-02-18  Simon Fraser  <simon.fraser@apple.com>

        No Review.

        Remove a couple of extraneous spaces that got added to the project file
        by hand-ending.

        * WebCore.xcodeproj/project.pbxproj:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@54995 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-02-18 Simon Fraser <simon.fraser@apple.com>
simon.fraser@apple.com [Fri, 19 Feb 2010 00:44:46 +0000 (00:44 +0000)]
2010-02-18  Simon Fraser  <simon.fraser@apple.com>

        Reviewed by Dan Bernstein.

        <rdar://problem/7535894> Page contents missing from snapshot on Newsweek.com article

        Add logic to determine when painting via the software rendering path will give an equivalent
        result to the accelerated compositing presentation. This tests for the presence of 3D transforms
        via the existing RenderLayerCompositor::has3DContent() method.

        * WebCore.base.exp: Export FrameView's isSoftwareRenderable(), paintBehavior() and setPaintBehavior().
        * page/FrameView.h:
        * page/FrameView.cpp:
        (WebCore::FrameView::isSoftwareRenderable): New method.
        (WebCore::FrameView::paintBehavior): Make this non-inline so that we can reliably export it.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@54994 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago Reviewed by Dan Bernstein.
ap@apple.com [Fri, 19 Feb 2010 00:37:08 +0000 (00:37 +0000)]
    Reviewed by Dan Bernstein.

        https://bugs.webkit.org/show_bug.cgi?id=35134
        <rdar://problem/7246280> Crash when a plugin calls NPN_SetStatus(0)

        Test: plugins/set-status.html

        * Plugins/Hosted/NetscapePluginInstanceProxy.mm:
        (WebKit::NetscapePluginInstanceProxy::status):
        (WebKit::NetscapePluginInstanceProxy::loadURL):
        Added null checks for CFStringCreateWithCString arguments.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@54993 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago<rdar://problem/7658811> Multiple style recalcs due to getComputedStyle() on “display...
mitz@apple.com [Fri, 19 Feb 2010 00:19:48 +0000 (00:19 +0000)]
<rdar://problem/7658811> Multiple style recalcs due to getComputedStyle() on “display: none;” element
when there are pending style sheets

Reviewed by John Sullivan.

WebCore:

Test: fast/css/getComputedStyle/pending-stylesheet.html

When querying a property of a computed style declaration for a non-rendered element,
CSSStyleSelector::styleForElement() was called, and if there were pending style sheet, it
would behave as if the lack of renderer is due to FOUC suppression, and set a flag on
the document causing it to recalculate style. On the next computed style property access,
style would be recalculated for the document, but then the flag would get set again if the
element did not have a renderer.

* dom/Document.cpp:
(WebCore::Document::styleForElementIgnoringPendingStylesheets): Added. Temporarily sets
m_ignorePendingStylesheets around the call to CSSStyleSelector::styleForElement().
* dom/Document.h:
* dom/Element.cpp:
(WebCore::Element::computedStyle): Use Document::styleForElementIgnoringPendingStylesheets().

LayoutTests:

* fast/css/getComputedStyle/pending-stylesheet-expected.txt: Added.
* fast/css/getComputedStyle/pending-stylesheet.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@54992 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-02-18 Dirk Schulze <krit@webkit.org>
krit@webkit.org [Thu, 18 Feb 2010 23:12:30 +0000 (23:12 +0000)]
2010-02-18  Dirk Schulze  <krit@webkit.org>

        Reviewed by Nikolas Zimmermann.

        Move SVGResources to Renderers, starting with Masker
        https://bugs.webkit.org/show_bug.cgi?id=35020

        We have rendering specific code in WebCore/svg/graphics. The goal is to move
        this code into suitable Renderers. This helps us to clean up the code and makes
        maintenance easier. It also makes it possible to remove rendering specific code
        from SVG*Elements into this renderers. So the Renderer contains everything that
        is needed to use the resource.
        RenderSVGResource will be the base class for all new resource render classes like
        RenderSVGResourceMasker, RenderSVGResourceClipper and the other resources.

        This patch starts moving SVGResourceMasker to RenderSVGResourceMasker.
        Another benefit is the much more useful result in DRT on using masker.

        * Android.mk:
        * GNUmakefile.am:
        * WebCore.gypi:
        * WebCore.pro:
        * WebCore.vcproj/WebCore.vcproj:
        * WebCore.xcodeproj/project.pbxproj:
        * rendering/RenderObject.cpp:
        (WebCore::RenderObject::toRenderSVGResource): Conversion to RenderSVGResource base class.
        * rendering/RenderObject.h:
        (WebCore::RenderObject::isSVGResource): Check if renderer is a resource.
        * rendering/RenderPath.cpp:
        * rendering/RenderSVGImage.cpp:
        (WebCore::RenderSVGImage::destroy): Renderer gets destroyed, unregister it from it's resources.
        * rendering/RenderSVGImage.h: Some code clean up according to the webkit style.
        (WebCore::RenderSVGImage::toSVGRenderBase):
        (WebCore::RenderSVGImage::renderName):
        (WebCore::RenderSVGImage::isSVGImage):
        (WebCore::RenderSVGImage::localToParentTransform):
        (WebCore::RenderSVGImage::strokeBoundingBox):
        (WebCore::RenderSVGImage::requiresLayer):
        (WebCore::RenderSVGImage::localTransform):
        * rendering/RenderSVGInlineText.h:
        (WebCore::RenderSVGInlineText::objectBoundingBox): Needed for SVGRenderTreeAsText
        * rendering/RenderSVGModelObject.cpp:
        (WebCore::RenderSVGModelObject::destroy): Renderer gets destroyed, unregister it from it's resources.
        * rendering/RenderSVGModelObject.h:
        * rendering/RenderSVGResource.h: Added.
        (WebCore::): Base class for all Resource renderers like masker, clipper and others.
        (WebCore::RenderSVGResource::RenderSVGResource):
        (WebCore::RenderSVGResource::cast):
        (WebCore::RenderSVGResource::toRenderSVGResource):
        (WebCore::RenderSVGResource::isSVGResource):
        (WebCore::RenderSVGResource::drawsContents):
        (WebCore::getRenderSVGResourceById):
        * rendering/RenderSVGResourceMasker.cpp: Added.
        (WebCore::RenderSVGResourceMasker::RenderSVGResourceMasker):
        (WebCore::RenderSVGResourceMasker::~RenderSVGResourceMasker):
        (WebCore::RenderSVGResourceMasker::invalidateClients): Status of masker changed, remove all clients.
        (WebCore::RenderSVGResourceMasker::invalidateClient): Status of an object changed, remove pending client.
        (WebCore::RenderSVGResourceMasker::applyResource): Apply masker to object.
        (WebCore::RenderSVGResourceMasker::resourceBoundingBox): boundingBox of the resource, depending on the object.
        (WebCore::RenderSVGResourceMasker::createMaskImage): Creates the mask image, the context gets clipped with.
        * rendering/RenderSVGResourceMasker.h: Added.
        (WebCore::MaskerData::MaskerData):
        (WebCore::RenderSVGResourceMasker::renderName):
        (WebCore::RenderSVGResourceMasker::maskUnits): Unit of mask for DRT.
        (WebCore::RenderSVGResourceMasker::maskContentUnits): Unit of childs from mask for DRT.
        (WebCore::RenderSVGResourceMasker::resourceType):
        * rendering/RenderSVGRoot.cpp:
        (WebCore::RenderSVGRoot::destroy): Renderer gets destroyed, unregister it from it's resources.
        * rendering/RenderSVGRoot.h:
        * rendering/RenderSVGText.cpp:
        (WebCore::RenderSVGText::destroy): dito.
        * rendering/RenderSVGText.h:
        * rendering/RenderTreeAsText.cpp:
        (WebCore::write):
        * rendering/SVGRenderSupport.cpp:
        (WebCore::SVGRenderBase::prepareToRenderSVGContent):
        (WebCore::SVGRenderBase::maskerBoundingBoxForRenderer):
        (WebCore::SVGRenderBase::deregisterFromResources): Unregister object from all it's resources after status changed.
        * rendering/SVGRenderSupport.h:
        (WebCore::SVGRenderBase::toSVGRenderBase):
        (WebCore::SVGRenderBase::strokeBoundingBox):
        (WebCore::SVGRenderBase::markerBoundingBox):
        * rendering/SVGRenderTreeAsText.cpp: Update TreeAsText to dump maskers correctly.
        (WebCore::operator<<):
        (WebCore::writeSVGResource):
        (WebCore::writeSVGContainer):
        (WebCore::writeSVGText):
        (WebCore::writeSVGInlineText):
        (WebCore::writeSVGImage):
        (WebCore::write):
        (WebCore::writeResourcesToObject):
        * rendering/SVGRenderTreeAsText.h:
        * svg/SVGMaskElement.cpp: Update Masker to use the new renderer.
        (WebCore::SVGMaskElement::svgAttributeChanged):
        (WebCore::SVGMaskElement::childrenChanged):
        (WebCore::SVGMaskElement::maskBoundingBox):
        (WebCore::SVGMaskElement::createRenderer):
        * svg/SVGMaskElement.h:
        * svg/SVGStyledElement.cpp: We need to tell the renderer to unregister object, after the status changed.
        (WebCore::SVGStyledElement::invalidateResources):
        (WebCore::SVGStyledElement::invalidateResourcesInAncestorChain):
        * svg/SVGUnitTypes.h: Conversion of integer to SVGUnitType.
        (WebCore::toUnitType):
        * svg/graphics/SVGResource.h:
        (WebCore::):
        (WebCore::SVGResource::isMarker):
        * svg/graphics/SVGResourceMasker.cpp: Removed.
        * svg/graphics/SVGResourceMasker.h: Removed.

2010-02-18  Dirk Schulze  <krit@webkit.org>

        Reviewed by Nikolas Zimmermann.

        Move SVGResources to Renderers, starting with Masker
        https://bugs.webkit.org/show_bug.cgi?id=35020

        With the move from SVGResourceMasker to RenderSVGResourceMasker, I changed
        the way DRT dumps resources. DRT is now able to dump multi-access
        of resources.
        Also the size of a Mask resource (depending of the object) and the access of
        the resource (togeter with the resource name) is pointed out as an insertion
        under the object.

        * platform/mac/svg/W3C-SVG-1.1/masking-intro-01-f-expected.txt:
        * platform/mac/svg/W3C-SVG-1.1/masking-mask-01-b-expected.txt:
        * platform/mac/svg/batik/masking/maskRegions-expected.txt:
        * platform/mac/svg/css/circle-in-mask-with-shadow-expected.txt:
        * platform/mac/svg/css/mask-with-shadow-expected.txt:
        * platform/mac/svg/custom/empty-mask-expected.txt:
        * platform/mac/svg/custom/grayscale-gradient-mask-expected.txt:
        * platform/mac/svg/custom/js-late-mask-and-object-creation-expected.txt:
        * platform/mac/svg/custom/js-late-mask-creation-expected.txt:
        * platform/mac/svg/custom/mask-changes-expected.txt:
        * platform/mac/svg/custom/mask-child-changes-expected.txt:
        * platform/mac/svg/custom/mask-excessive-malloc-expected.txt:
        * platform/mac/svg/custom/mask-inside-defs-expected.txt:
        * platform/mac/svg/custom/mask-invalidation-expected.txt:
        * platform/mac/svg/custom/mask-on-multiple-objects-expected.txt:
        * platform/mac/svg/custom/mask-with-all-units-expected.txt:
        * platform/mac/svg/custom/mask-with-default-value-expected.txt:
        * platform/mac/svg/custom/resource-invalidate-on-target-update-expected.txt:
        * platform/mac/svg/custom/visibility-override-mask-expected.txt:
        * platform/mac/svg/filters/filter-clip-expected.txt:
        * svg/custom/transformedMaskFails-expected.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@54991 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoAllow creating/dropping virtual tables when the module is FTS2.
dumi@chromium.org [Thu, 18 Feb 2010 22:51:43 +0000 (22:51 +0000)]
Allow creating/dropping virtual tables when the module is FTS2.

Reviewed by Dimitri Glazkov.

https://bugs.webkit.org/show_bug.cgi?id=34867

* storage/DatabaseAuthorizer.cpp:
(WebCore::DatabaseAuthorizer::createVTable):
(WebCore::DatabaseAuthorizer::dropVTable):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@54990 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoNot reviewed, Chromium build fix.
pkasting@chromium.org [Thu, 18 Feb 2010 22:40:50 +0000 (22:40 +0000)]
Not reviewed, Chromium build fix.

r54963 had a typo in the WebCore.gypi change.
https://bugs.webkit.org/show_bug.cgi?id=35003

* WebCore.gypi:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@54989 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoFixed a portion of:
ggaren@apple.com [Thu, 18 Feb 2010 22:15:34 +0000 (22:15 +0000)]
Fixed a portion of:
<rdar://problem/7165917> | https://bugs.webkit.org/show_bug.cgi?id=28676
Safari 4 does not release memory back to the operating system fast enough (28676)

Reviewed by Oliver Hunt.

This patch fixes a surprisingly common edge case in which the page heap
would have only one free span, but that span would be larger than the
minimum free size, so we would decide not to free it, even though it
could be as large as 100MB or more!

SunSpider reports no change on Mac or Windows.

* wtf/FastMalloc.cpp:
(WTF::TCMalloc_PageHeap::scavenge): Call shouldContinueScavenging() instead
of doing the math ourselves. Don't keep a local value for pagesDecommitted
because that lets free_committed_pages_ be wrong temporarily. Instead,
update free_committed_pages_ as we go. ASSERT that we aren't releasing
a span that has already been released, because we think this is impossible.
Finally, don't be afraid to release all free memory in the page heap when
scavenging. We only scavenge after 5 seconds of the application's working
set not growing, and we keep both thread caches and a central cache on
top of the page heap, so the extra free pages in the page heap were just
overkill.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@54988 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-02-18 Vangelis Kokkevis <vangelis@chromium.org>
eric@webkit.org [Thu, 18 Feb 2010 22:08:21 +0000 (22:08 +0000)]
2010-02-18  Vangelis Kokkevis  <vangelis@chromium.org>

        Reviewed by Simon Fraser.

        Changing forward declaration of TimingFunction in GraphicsLayer.h from
        class to struct to match its actual definition in TimingFunction.h

        https://bugs.webkit.org/show_bug.cgi?id=35069

        * platform/graphics/GraphicsLayer.h:
           Change forward declaration from: class TimingFunction to:
           struct TimingFunction

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@54987 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-02-18 Noam Rosenthal <noam.rosenthal@nokia.com>
eric@webkit.org [Thu, 18 Feb 2010 21:50:59 +0000 (21:50 +0000)]
2010-02-18  Noam Rosenthal  <noam.rosenthal@nokia.com>

        Reviewed by Kenneth Rohde Christiansen.

        [Qt] Minor improvement to hybrid QPixmap
        https://bugs.webkit.org/show_bug.cgi?id=34507
        Instead of having toHTMLImageElement which creates a new element,
        assignToHTMLImageElement would set an existing HTML <img/> element to
        contain the native QPixmap/QImge.
        Also made some style fixes.

        Additions to existing tests: see WebKit/qt/tests

        * bridge/qt/qt_pixmapruntime.cpp:
        (JSC::Bindings::QtPixmapWidthField::valueFromInstance): style
        (JSC::Bindings::QtPixmapHeightField::valueFromInstance): style
        (JSC::Bindings::QtPixmapAssignToElementMethod::name): assignTo
        (JSC::Bindings::QtPixmapAssignToElementMethod::invoke): new function
        (JSC::Bindings::QtPixmapAssignToElementMethod::numParameters): 1
        (JSC::Bindings::QtPixmapToDataUrlMethod::invoke): style
        (JSC::Bindings::QtPixmapToStringMethod::invoke): style
        (JSC::Bindings::QtPixmapInstance::invokeMethod): style
        (JSC::Bindings::QtPixmapClass::methodsNamed): new func, removed old
        (JSC::Bindings::QtPixmapInstance::getPropertyNames): ditto
        (JSC::Bindings::QtPixmapInstance::defaultValue): style
        (JSC::Bindings::QtPixmapInstance::valueOf): style
        (JSC::Bindings::QtPixmapInstance::toPixmap): style
        (JSC::Bindings::QtPixmapInstance::variantFromObject): style
2010-02-18  Noam Rosenthal  <noam.rosenthal@nokia.com>

        Reviewed by Kenneth Rohde Christiansen.

        [Qt] Minor improvement to hybrid QPixmap
        https://bugs.webkit.org/show_bug.cgi?id=34507

        * tests/hybridPixmap/test.html: use assignToHTMLImageElement on an
        existing element instead of toHTMLImageElement which creates a new one

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@54986 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-02-18 Gustavo Noronha Silva <gns@gnome.org>
kov@webkit.org [Thu, 18 Feb 2010 21:15:14 +0000 (21:15 +0000)]
2010-02-18  Gustavo Noronha Silva  <gns@gnome.org>

        [Qt][GTK] fast/xmlhttprequest/xmlhttprequest-nonexistent-file.html fails (DRT sideeffect)
        https://bugs.webkit.org/show_bug.cgi?id=35086

        Skip both new tests that fail. It may be a DRT problem, and the
        file path that is reported in the error message also seems to
        require tweaking.

        * platform/gtk/Skipped:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@54985 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoNot reviewed, bustage fix.
pkasting@chromium.org [Thu, 18 Feb 2010 21:07:39 +0000 (21:07 +0000)]
Not reviewed, bustage fix.

An extraneous line in r54839 broke GIF animation.

* platform/image-decoders/gif/GIFImageReader.cpp:
(GIFImageReader::read):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@54984 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-02-18 Gustavo Noronha Silva <gns@gnome.org>
kov@webkit.org [Thu, 18 Feb 2010 21:03:19 +0000 (21:03 +0000)]
2010-02-18  Gustavo Noronha Silva  <gns@gnome.org>

        [GTK] Fails fast/multicol/client-rects.html
        https://bugs.webkit.org/show_bug.cgi?id=35127

        Skip test that fails in GTK+ - trying to get bot green again.

        * platform/gtk/Skipped:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@54983 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago<rdar://problem/7650652> REGRESSION: Selection painting issue in bug review textbox
mitz@apple.com [Thu, 18 Feb 2010 20:55:49 +0000 (20:55 +0000)]
<rdar://problem/7650652> REGRESSION: Selection painting issue in bug review textbox
https://bugs.webkit.org/show_bug.cgi?id=34946

Reviewed by Simon Fraser.

WebCore:

Test: fast/repaint/selection-gap-overflow-scroll-2.html

* rendering/RenderBlock.cpp:
(WebCore::RenderBlock::paintSelection): localToContainerQuad() adjusts for
overflow scroll, but RenderLayer::addBlockSelectionGapsBounds() takes
non-scrolled coordinates, so account for that.

LayoutTests:

* fast/repaint/selection-gap-overflow-scroll-2.html: Copied from LayoutTests/fast/repaint/selection-gap-overflow-scroll.html.
* platform/mac/fast/repaint/selection-gap-overflow-scroll-2-expected.checksum: Added.
* platform/mac/fast/repaint/selection-gap-overflow-scroll-2-expected.png: Added.
* platform/mac/fast/repaint/selection-gap-overflow-scroll-2-expected.txt: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@54982 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoWebCore: Change the V8 and JSC SQLStatementErrorCallback to interpret
dumi@chromium.org [Thu, 18 Feb 2010 20:52:07 +0000 (20:52 +0000)]
WebCore: Change the V8 and JSC SQLStatementErrorCallback to interpret
'undefined' return values as 'true', as required by the spec.

Reviewed by Darin Adler.

https://bugs.webkit.org/show_bug.cgi?id=35048

* bindings/js/JSCustomSQLStatementErrorCallback.cpp:
(WebCore::JSCustomSQLStatementErrorCallback::handleEvent):
* bindings/v8/custom/V8CustomSQLStatementErrorCallback.cpp:
(WebCore::V8CustomSQLStatementErrorCallback::handleEvent):
* bindings/v8/custom/V8CustomVoidCallback.cpp:
(WebCore::invokeCallbackHelper):
(WebCore::invokeCallback):
(WebCore::invokeCallbackTreatUndefinedAsTrue):
* bindings/v8/custom/V8CustomVoidCallback.h:

LayoutTests: 'undefined' return values from statement error callbacks are not
treated as 'true'. Fix the tests that did not use this
assumption.

Reviewed by NOBODY Darin Adler.

https://bugs.webkit.org/show_bug.cgi?id=35048

* storage/database-lock-after-reload.html:
* storage/private-browsing-readonly.html:
* storage/statement-error-callback.html:
* storage/statement-error-callback-expected.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@54981 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-02-17 Ojan Vafai <ojan@chromium.org>
ojan@chromium.org [Thu, 18 Feb 2010 19:38:14 +0000 (19:38 +0000)]
2010-02-17  Ojan Vafai  <ojan@chromium.org>

        Reviewed by Adam Barth.

        keyboard selection sometimes moves the wrong end of the selection for Win/Linux
        https://bugs.webkit.org/show_bug.cgi?id=35066

        The changes to js-test-selection-shared.js are just to make the selection editing tests
        easier to debug in the future.

        The mac results for extend-after-mouse-selection.html contain a "FAIL" line. This is
        a case where WebKit doesn't match TextEdit.

        * editing/editing.js:
        (execMoveSelectionForwardByLineBoundaryCommand):
        (moveSelectionForwardByLineBoundaryCommand):
        (execExtendSelectionForwardByLineBoundaryCommand):
        (extendSelectionForwardByLineBoundaryCommand):
        (execExtendSelectionBackwardByLineBoundaryCommand):
        (extendSelectionBackwardByLineBoundaryCommand):
        (execMoveSelectionBackwardByLineBoundaryCommand):
        (moveSelectionBackwardByLineBoundaryCommand):
        * editing/selection/extend-after-mouse-selection.html: Added.
        * editing/selection/move-begin-end-expected.txt:
        * editing/selection/move-begin-end.html:
        * editing/selection/resources/js-test-selection-shared.js:
        (objectAsString):
        (assertSelectionAt):
        * platform/mac/editing/selection/click-in-margins-inside-editable-div-expected.txt:
        * platform/mac/editing/selection/click-in-padding-with-multiple-line-boxes-expected.txt:
        * platform/mac/editing/selection/extend-after-mouse-selection-expected.txt: Added.
        * platform/win/editing/selection/click-in-margins-inside-editable-div-expected.txt:
        * platform/win/editing/selection/click-in-padding-with-multiple-line-boxes-expected.txt:
        * platform/win/editing/selection/extend-after-mouse-selection-expected.txt: Added.
        * platform/win/editing/selection/extend-selection-expected.txt: Added.
2010-02-17  Ojan Vafai  <ojan@chromium.org>

        Reviewed by Adam Barth.

        keyboard selection sometimes moves the wrong end of the selection for Win/Linux
        https://bugs.webkit.org/show_bug.cgi?id=35066

        On Windows/Linux keyboard based selections should always move the same
        end of the seleciton. On Mac, lineboundary and documentboundary changes
        move different ends of the selection depending on which direction your
        extending.

        Test: editing/selection/extend-after-mouse-selection.html

        * editing/SelectionController.cpp:
        (WebCore::SelectionController::positionForPlatform):
        (WebCore::SelectionController::startForPlatform):
        (WebCore::SelectionController::endForPlatform):
        (WebCore::SelectionController::modifyExtendingRight):
        (WebCore::SelectionController::modifyExtendingForward):
        (WebCore::SelectionController::modifyMovingForward):
        (WebCore::SelectionController::modifyExtendingBackward):
        (WebCore::SelectionController::modifyMovingBackward):
        * editing/SelectionController.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@54980 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoAdd "with" to the list of keywords to syntax highlight.
timothy@apple.com [Thu, 18 Feb 2010 19:36:27 +0000 (19:36 +0000)]
Add "with" to the list of keywords to syntax highlight.

http://webkit.org/b/35123

Reviewed by Pavel Feldman.

* inspector/front-end/SourceJavaScriptTokenizer.js:
(WebInspector.SourceJavaScriptTokenizer): Add "width" to _keywords.
* inspector/front-end/SourceJavaScriptTokenizer.re2js:
(WebInspector.SourceJavaScriptTokenizer): Ditto.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@54979 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoRework PNG-in-ICO decoding to copy the decoded framebuffer into the ICO
pkasting@chromium.org [Thu, 18 Feb 2010 19:35:16 +0000 (19:35 +0000)]
Rework PNG-in-ICO decoding to copy the decoded framebuffer into the ICO
decoder, making the logic less crazy and more like the other decoders.
https://bugs.webkit.org/show_bug.cgi?id=28751

Reviewed by Adam Barth.

* platform/image-decoders/ImageDecoder.cpp:
(WebCore::RGBA32Buffer::operator=):
* platform/image-decoders/ImageDecoder.h:
(WebCore::RGBA32Buffer::RGBA32Buffer):
* platform/image-decoders/ico/ICOImageDecoder.cpp:
(WebCore::ICOImageDecoder::frameBufferAtIndex):
(WebCore::ICOImageDecoder::decodeAtIndex):
* platform/image-decoders/ico/ICOImageDecoder.h:
* platform/image-decoders/qt/RGBA32BufferQt.cpp:
(WebCore::RGBA32Buffer::operator=):
* platform/image-decoders/skia/ImageDecoderSkia.cpp:
(WebCore::RGBA32Buffer::operator=):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@54978 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-02-18 Nate Chapin <japhet@chromium.org>
japhet@chromium.org [Thu, 18 Feb 2010 19:08:52 +0000 (19:08 +0000)]
2010-02-18  Nate Chapin  <japhet@chromium.org>

        Unreviewed, build fix.

        Provide good expected results for fast/dom/Window/window-event-override-no-crash.html.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@54977 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoWindows build fix.
sfalken@apple.com [Thu, 18 Feb 2010 19:05:45 +0000 (19:05 +0000)]
Windows build fix.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@54976 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-02-18 Gustavo Noronha Silva <gns@gnome.org>
kov@webkit.org [Thu, 18 Feb 2010 18:59:01 +0000 (18:59 +0000)]
2010-02-18  Gustavo Noronha Silva  <gns@gnome.org>

        [GTK][Qt] Fails glyph reordering test fast/text/glyph-reordering.html
        https://bugs.webkit.org/show_bug.cgi?id=34918

        Unskipping that now passes, after r54912.

        * platform/gtk/Skipped:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@54975 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-02-18 Pavel Feldman <pfeldman@chromium.org>
pfeldman@chromium.org [Thu, 18 Feb 2010 18:53:56 +0000 (18:53 +0000)]
2010-02-18  Pavel Feldman  <pfeldman@chromium.org>

        Reviewed by Timothy Hatcher.

        Web Inspector: multiple popovers on screen at the same time.

        https://bugs.webkit.org/show_bug.cgi?id=35105

        * inspector/front-end/Popover.js:
        (WebInspector.Popover.prototype.show):
        (WebInspector.Popover.prototype.hide):
        * inspector/front-end/SourceFrame.js:
        (WebInspector.SourceFrame.prototype._showPopup):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@54974 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-02-18 Pavel Feldman <pfeldman@chromium.org>
pfeldman@chromium.org [Thu, 18 Feb 2010 18:45:31 +0000 (18:45 +0000)]
2010-02-18  Pavel Feldman  <pfeldman@chromium.org>

        Reviewed by Timothy Hatcher.

        Web Inspector: Expand Object.__proto__ properly.

        https://bugs.webkit.org/show_bug.cgi?id=35113

        * inspector/front-end/EventListenersSidebarPane.js:
        * inspector/front-end/InjectedScript.js:
        (injectedScriptConstructor):
        * inspector/front-end/ObjectProxy.js:
        (WebInspector.ObjectProxy):
        * inspector/front-end/PropertiesSidebarPane.js:
        (WebInspector.PropertiesSidebarPane.prototype.update.callback):
        * inspector/front-end/inspector.js:
        (WebInspector.log.logMessage):
        (WebInspector.log):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@54973 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-02-18 Nate Chapin <japhet@chromium.org>
japhet@chromium.org [Thu, 18 Feb 2010 18:40:49 +0000 (18:40 +0000)]
2010-02-18  Nate Chapin  <japhet@chromium.org>

        Reviewed by Adam Barth.

        [V8] Merge the DOMWindow and WorkerContext object wrapping code paths,
        and use a faster method of disambiguating between the types of contexts.

        https://bugs.webkit.org/show_bug.cgi?id=35009

        * bindings/scripts/CodeGeneratorV8.pm: Remove logic determining whether we need to
            handle the WorkerContext case.
        * bindings/v8/V8DOMWrapper.cpp:
        (WebCore::globalObjectPrototypeIsDOMWindow):
        (WebCore::V8DOMWrapper::instantiateV8Object): Merge instantiateV8Object paths.
        * bindings/v8/V8DOMWrapper.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@54972 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-02-18 Xan Lopez <xlopez@igalia.com>
xan@webkit.org [Thu, 18 Feb 2010 18:25:53 +0000 (18:25 +0000)]
2010-02-18  Xan Lopez  <xlopez@igalia.com>

        Reviewed by Gustavo Noronha.

        Remove some duplication between PluginView and Widget methods, and
        move the setSelfVisible calls to the parent class.

        * platform/gtk/WidgetGtk.cpp:
        (WebCore::Widget::show):
        (WebCore::Widget::hide):
        * plugins/gtk/PluginViewGtk.cpp:
        (WebCore::PluginView::setFocus):
        (WebCore::PluginView::show):
        (WebCore::PluginView::hide):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@54971 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-02-18 Simon Fraser <simon.fraser@apple.com>
simon.fraser@apple.com [Thu, 18 Feb 2010 17:51:20 +0000 (17:51 +0000)]
2010-02-18  Simon Fraser  <simon.fraser@apple.com>

        Reviewed by Darin Adler.

        <rdar://problem/7655195> Switch Leopard back to using CGShading to avoid CGGradient leaks

        Define USE_CG_SHADING on for Tiger and Leopard, and use it to toggle the methods
        used for Core Graphics gradient drawing.

        * platform/graphics/Gradient.h:
        * platform/graphics/cg/GradientCG.cpp:
        (WebCore::Gradient::platformDestroy):
        (WebCore::Gradient::paint):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@54970 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-02-18 Pavel Feldman <pfeldman@chromium.org>
pfeldman@chromium.org [Thu, 18 Feb 2010 17:36:40 +0000 (17:36 +0000)]
2010-02-18  Pavel Feldman  <pfeldman@chromium.org>

        Reviewed by Timothy Hatcher.

        Web Inspector: on-hover evaluation renders nodes and arrays as strings.

        https://bugs.webkit.org/show_bug.cgi?id=35103

        * inspector/front-end/SourceFrame.js:
        (WebInspector.SourceFrame.prototype._showPopup.showObjectPopup):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@54969 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-02-18 Yury Semikhatsky <yurys@chromium.org>
yurys@chromium.org [Thu, 18 Feb 2010 17:32:06 +0000 (17:32 +0000)]
2010-02-18  Yury Semikhatsky  <yurys@chromium.org>

        Reviewed by Pavel Feldman.

        Don't truncate long lines when evaluating expressions on a breakpoint.

        https://bugs.webkit.org/show_bug.cgi?id=35102

        * src/js/DebuggerAgent.js:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@54968 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-02-18 A S Alam <amanpreet.alam@gmail.com>
kov@webkit.org [Thu, 18 Feb 2010 17:31:33 +0000 (17:31 +0000)]
2010-02-18  A S Alam  <amanpreet.alam@gmail.com>

        Punjabi translation.

        * pa.po: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@54967 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoParticularly constructed WebFrames can try to access a null HistoryItem
beidson@apple.com [Thu, 18 Feb 2010 17:13:22 +0000 (17:13 +0000)]
Particularly constructed WebFrames can try to access a null HistoryItem
<rdar://problem/7638892> and https://bugs.webkit.org/show_bug.cgi?id=35063

Reviewed by Sam Weinig.

WebCore:

Test: fast/loader/api-test-new-window-data-load-base-url.html

* loader/HistoryController.cpp:
(WebCore::HistoryController::updateBackForwardListForFragmentScroll): We have a known case where a fragment scroll
  might take place with a null m_currentItem. updateBackForwardListClippedAtTarget() will either move m_currentItem
  to m_previousItem then create a new m_currentItem...  or it will do nothing. So we either have both an m_currentItem
  and m_previousItem, or we have neither.  In the case where we have no m_previousItem, return early.

WebKitTools:

Add the specific ability to test this API pattern.

For now only on Mac, perhaps on the main Windows port later, probably not relevant for other ports.

* DumpRenderTree/LayoutTestController.cpp:
(apiTestNewWindowDataLoadBaseURLCallback):
(LayoutTestController::staticFunctions):
* DumpRenderTree/LayoutTestController.h:

* DumpRenderTree/mac/LayoutTestControllerMac.mm:
(-[APITestDelegate initWithCompletionCondition:]):
(-[APITestDelegate webView:didFailLoadWithError:forFrame:]):
(-[APITestDelegate webView:didFailProvisionalLoadWithError:forFrame:]):
(-[APITestDelegate webView:didFinishLoadForFrame:]):
(LayoutTestController::apiTestNewWindowDataLoadBaseURL): Create a WebView, do a loadData: in its
  mainFrame, and synchronously wait for main load completion.

* DumpRenderTree/win/LayoutTestControllerWin.cpp:
(LayoutTestController::apiTestNewWindowDataLoadBaseURL):
* DumpRenderTree/wx/LayoutTestControllerWx.cpp:
(LayoutTestController::apiTestNewWindowDataLoadBaseURL):
* DumpRenderTree/gtk/LayoutTestControllerGtk.cpp:
(LayoutTestController::apiTestNewWindowDataLoadBaseURL):

LayoutTests:

* fast/loader/api-test-new-window-data-load-base-url-expected.txt: Added.
* fast/loader/api-test-new-window-data-load-base-url.html: Added.
* platform/gtk/Skipped:
* platform/qt/Skipped:
* platform/win/Skipped:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@54966 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-02-18 Nate Chapin <japhet@chromium.org>
japhet@chromium.org [Thu, 18 Feb 2010 17:08:39 +0000 (17:08 +0000)]
2010-02-18  Nate Chapin  <japhet@chromium.org>

        Reviewed by Eric Seidel.

        Actually submitting test for https://bugs.webkit.org/show_bug.cgi?id=34899 this time.

        * fast/dom/Window/window-event-override-no-crash-expected.txt: Added.
        * fast/dom/Window/window-event-override-no-crash.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@54965 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-02-18 Nate Chapin <japhet@chromium.org>
japhet@chromium.org [Thu, 18 Feb 2010 17:04:37 +0000 (17:04 +0000)]
2010-02-18  Nate Chapin  <japhet@chromium.org>

        Reviewed by Eric Seidel.

        [V8] Correctly handle the case where the event field on the
        global object is a v8::Object, but not a DOM wrapper.

        https://bugs.webkit.org/show_bug.cgi?id=34899

        Test: fast/dom/Window/window-event-override-no-crash.html

        * bindings/v8/ScriptController.cpp:
        (WebCore::ScriptController::processingUserGesture):
        * bindings/v8/V8DOMWrapper.cpp:
        (WebCore::V8DOMWrapper::isValidDOMObject):
        (WebCore::V8DOMWrapper::isWrapperOfType):
        * bindings/v8/V8DOMWrapper.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@54964 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-02-18 Pavel Feldman <pfeldman@chromium.org>
pfeldman@chromium.org [Thu, 18 Feb 2010 16:56:50 +0000 (16:56 +0000)]
2010-02-18  Pavel Feldman  <pfeldman@chromium.org>

        Not reviewed, Qt build fix.

        * inspector/front-end/WebKit.qrc:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@54963 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-02-18 Pavel Feldman <pfeldman@chromium.org>
pfeldman@chromium.org [Thu, 18 Feb 2010 16:12:53 +0000 (16:12 +0000)]
2010-02-18  Pavel Feldman  <pfeldman@chromium.org>

        Reviewed by Timothy Hatcher.

        Web Inspector: Implement evaluate-on-hover for scripts panel.

        https://bugs.webkit.org/show_bug.cgi?id=35003

        * WebCore.gypi:
        * WebCore.vcproj/WebCore.vcproj:
        * inspector/front-end/Images/gearButtonGlyph.png: Added.
        * inspector/front-end/Images/popoverArrows.png: Added.
        * inspector/front-end/Images/popoverBackground.png: Added.
        * inspector/front-end/Images/thumbActiveHoriz.png: Added.
        * inspector/front-end/Images/thumbActiveVert.png: Added.
        * inspector/front-end/Images/thumbHoriz.png: Added.
        * inspector/front-end/Images/thumbHoverHoriz.png: Added.
        * inspector/front-end/Images/thumbHoverVert.png: Added.
        * inspector/front-end/Images/thumbVert.png: Added.
        * inspector/front-end/Images/trackHoriz.png: Added.
        * inspector/front-end/Images/trackVert.png: Added.
        * inspector/front-end/Popup.js:
        (WebInspector.Popup):
        (WebInspector.Popup.prototype.show):
        (WebInspector.Popup.prototype.hide):
        (WebInspector.Popup.prototype._positionElement):
        * inspector/front-end/SourceFrame.js:
        (WebInspector.SourceFrame.prototype._createViewerIfNeeded):
        (WebInspector.SourceFrame.prototype._scroll):
        (WebInspector.SourceFrame.prototype._mouseDown):
        (WebInspector.SourceFrame.prototype._mouseUp):
        (WebInspector.SourceFrame.prototype._mouseMove):
        (WebInspector.SourceFrame.prototype._mouseOut):
        (WebInspector.SourceFrame.prototype._resetHoverTimer):
        (WebInspector.SourceFrame.prototype._hidePopup):
        (WebInspector.SourceFrame.prototype._mouseHover):
        (WebInspector.SourceFrame.prototype._showPopup.showTextPopup):
        (WebInspector.SourceFrame.prototype._showPopup.showObjectPopup):
        (WebInspector.SourceFrame.prototype._showPopup.evaluateCallback):
        (WebInspector.SourceFrame.prototype._showPopup):
        (WebInspector.HoverPropertiesSection):
        (WebInspector.HoverPropertiesSection.prototype.update):
        * inspector/front-end/TextEditorHighlighter.js:
        (WebInspector.TextEditorHighlighter):
        (WebInspector.TextEditorHighlighter.prototype._lex):
        * inspector/front-end/TextViewer.js:
        (WebInspector.TextViewer.prototype._paintLine):
        * inspector/front-end/WebKit.qrc:
        * inspector/front-end/inspector.css:
        * inspector/front-end/inspector.html:
        * inspector/front-end/inspectorSyntaxHighlight.css:
        * inspector/front-end/popover.css: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@54962 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago Rubber-stamped by Kenneth Rohde Christiansen.
abecsi@webkit.org [Thu, 18 Feb 2010 16:03:31 +0000 (16:03 +0000)]
Rubber-stamped by Kenneth Rohde Christiansen.

    [Qt] Skip the following tests because of missing eventSender.textZoomIn():

        svg/zoom/text/absolute-sized-document-no-scrollbars.svg
        svg/zoom/text/absolute-sized-document-scrollbars.svg
        svg/zoom/text/relative-sized-document-scrollbars.svg
        svg/zoom/text/zoom-foreignObject.svg
        svg/zoom/text/zoom-hixie-mixed-008.xml
        svg/zoom/text/zoom-hixie-mixed-009.xml
        svg/zoom/text/zoom-hixie-rendering-model-004.xhtml
        svg/zoom/text/zoom-svg-float-border-padding.xml

    * platform/qt/Skipped:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@54961 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoTeach resolve-ChangeLogs to act as a merge-driver for Git
vestbo@webkit.org [Thu, 18 Feb 2010 15:51:35 +0000 (15:51 +0000)]
Teach resolve-ChangeLogs to act as a merge-driver for Git

Reviewed by Eric Seidel.

https://bugs.webkit.org/show_bug.cgi?id=28721

To enable automatic merging of ChangeLog files, use the following command:

git config merge.changelog.driver "resolve-ChangeLogs --merge-driver %O %A %B"

The driver always works in "downstream" merge mode, meaning
it will always put the user's changelog entries on top.

* Scripts/resolve-ChangeLogs:
* .gitattributes: Add instruction to use merge-driver for ChangeLogs

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@54960 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-02-18 Diego Escalante Urrelo <descalante@igalia.com>
eric@webkit.org [Thu, 18 Feb 2010 14:27:17 +0000 (14:27 +0000)]
2010-02-18  Diego Escalante Urrelo  <descalante@igalia.com>

        Reviewed by Gustavo Noronha Silva.

        [GTK] WebKitDownload documentation is incomplete
        https://bugs.webkit.org/show_bug.cgi?id=35018

        Concretely the error signal was undocumented. Also two relevant enums
        were not included.

        * docs/webkitgtk-sections.txt:
        * docs/webkitgtk.types:
        * webkit/webkitdownload.cpp:
        (webkit_download_class_init):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@54959 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-02-18 Ben Murdoch <benm@google.com>
benm@google.com [Thu, 18 Feb 2010 14:22:53 +0000 (14:22 +0000)]
2010-02-18  Ben Murdoch  <benm@google.com>

        Reviewed by Jeremy Orlow.

        [v8] Complete upstreaming of V8 Touch Event bindings
        https://bugs.webkit.org/show_bug.cgi?id=35094

        No new tests required.

        * bindings/v8/V8Index.cpp: Add generated touch event headers.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@54958 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-02-18 Steve Block <steveblock@google.com>
steveblock@google.com [Thu, 18 Feb 2010 14:12:37 +0000 (14:12 +0000)]
2010-02-18  Steve Block  <steveblock@google.com>

        Reviewed by Jeremy Orlow.

        Updates Android V8 build to use DerivedSourcesAllInOne.cpp
        https://bugs.webkit.org/show_bug.cgi?id=35083

        No new tests, build fix only.

        * Android.derived.v8bindings.mk:
        * Android.v8bindings.mk:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@54957 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-12-05 Gustavo Noronha Silva <gns@gnome.org>
kov@webkit.org [Thu, 18 Feb 2010 13:32:19 +0000 (13:32 +0000)]
2009-12-05  Gustavo Noronha Silva  <gns@gnome.org>

        Reviewed by Eric Seidel.

        Enable running of GTK+ API tests.

        * BuildSlaveSupport/build.webkit.org-config/master.cfg:
        * Scripts/run-gtk-tests: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@54956 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-02-18 Gustavo Noronha Silva <gustavo.noronha@collabora.co.uk>
kov@webkit.org [Thu, 18 Feb 2010 13:27:22 +0000 (13:27 +0000)]
2010-02-18  Gustavo Noronha Silva  <gustavo.noronha@collabora.co.uk>

        Reviewed by Xan Lopez.

        fast/frames/iframe-reparenting.html crashing on GTK Debug bots
        https://bugs.webkit.org/show_bug.cgi?id=35081

        Check that the client is alive after every call to it, since any
        of them could cause the load to be cancelled, and the client to go
        away.

        This is much better than protecting a specific subclass of
        ResourceHandleClient (ResourceLoader), which makes us fail when
        any other client is used.

        Test: fast/frames/iframe-reparenting.html

        * platform/network/soup/ResourceHandleSoup.cpp:
        (WebCore::parseDataUrl):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@54955 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago[Qt] fast/xmlhttprequest/xmlhttprequest-nonexistent-file.html fails (DRT sideeffect)
ossy@webkit.org [Thu, 18 Feb 2010 12:00:28 +0000 (12:00 +0000)]
[Qt] fast/xmlhttprequest/xmlhttprequest-nonexistent-file.html fails (DRT sideeffect)
https://bugs.webkit.org/show_bug.cgi?id=35086

fast/xmlhttprequest/xmlhttprequest-no-file-access.html cause
this fail indirectly, but both of them pass in separated DRT.

* platform/qt/Skipped: fast/xmlhttprequest/xmlhttprequest-no-file-access.html
skipped until fix to make buildbot happy.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@54954 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-02-18 Philippe Normand <pnormand@igalia.com>
philn@webkit.org [Thu, 18 Feb 2010 11:46:57 +0000 (11:46 +0000)]
2010-02-18  Philippe Normand  <pnormand@igalia.com>

        Not reviewed, adding a GTK layout test result for a new test
        added in r54926.

        * platform/gtk/fast/css/first-letter-first-line-hover-expected.txt: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@54953 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-02-18 Philippe Normand <pnormand@igalia.com>
philn@webkit.org [Thu, 18 Feb 2010 11:37:29 +0000 (11:37 +0000)]
2010-02-18  Philippe Normand  <pnormand@igalia.com>

        Rubber-stamped by Xan Lopez.

        plugins/iframe-shims.html crashing on GTK Debug bots
        https://bugs.webkit.org/show_bug.cgi?id=35085

        * platform/gtk/Skipped: Skip crashing tests.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@54952 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-02-18 Philippe Normand <pnormand@igalia.com>
philn@webkit.org [Thu, 18 Feb 2010 10:29:35 +0000 (10:29 +0000)]
2010-02-18  Philippe Normand  <pnormand@igalia.com>

        Rubber-stamped by Xan Lopez.

        fast/frames/iframe-reparenting.html crashing on GTK Debug bots
        https://bugs.webkit.org/show_bug.cgi?id=35081

        * platform/gtk/Skipped: Skip crashing test.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@54951 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago[v8] [Android] V8 bindings for touch events are missing.
benm@google.com [Thu, 18 Feb 2010 10:21:25 +0000 (10:21 +0000)]
[v8] [Android] V8 bindings for touch events are missing.
https://bugs.webkit.org/show_bug.cgi?id=33795

Reviewed by Nate Chapin.

No new tests as this just enables touch events in V8. Existing touch tests suffice.

* WebCore.gypi: Add Touch idl files.
* bindings/scripts/CodeGeneratorV8.pm: Add TouchList to typeCanFailConversion map.
* bindings/v8/DOMObjectsInclude.h: Add touch headers.
* bindings/v8/DerivedSourcesAllInOne.cpp: Add touch generated bindings.
* bindings/v8/V8Index.h: Add touch DOM object types.
* bindings/v8/custom/V8EventCustom.cpp: Add conversion of event to touch event.
* Android.derived.jscbindings.mk: Add the touch derived sources to the makefile.
* Android.derived.v8bindings.mk: Add the touch derived sources to the makefile.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@54950 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-02-18 Philippe Normand <pnormand@igalia.com>
philn@webkit.org [Thu, 18 Feb 2010 10:07:28 +0000 (10:07 +0000)]
2010-02-18  Philippe Normand  <pnormand@igalia.com>

        Rubber-stamped by Xan Lopez.

        fast/css/font-face-opentype.html crashing on GTK bots
        https://bugs.webkit.org/show_bug.cgi?id=35080

        * platform/gtk/Skipped: Skip crashing test.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@54949 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-02-18 Jay Campan <jcampan@google.com>
eric@webkit.org [Thu, 18 Feb 2010 09:27:03 +0000 (09:27 +0000)]
2010-02-18  Jay Campan  <jcampan@google.com>

        Reviewed by Adam Barth.

        Exposing the title in WebDocument.

        https://bugs.webkit.org/show_bug.cgi?id=35058

        * public/WebDocument.h:
        * src/WebDocument.cpp:
        (WebKit::WebDocument::title):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@54948 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-02-18 Shinichiro Hamaji <hamaji@chromium.org>
hamaji@chromium.org [Thu, 18 Feb 2010 09:17:37 +0000 (09:17 +0000)]
2010-02-18  Shinichiro Hamaji  <hamaji@chromium.org>

        Reviewed by Adam Barth.

        debug-safari doesn't pass --debug option to gdb-safari on MacOSX
        https://bugs.webkit.org/show_bug.cgi?id=34411

        * Scripts/webkitdirs.pm:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@54947 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-02-18 William Chan <willchan@chromium.org>
eric@webkit.org [Thu, 18 Feb 2010 09:10:54 +0000 (09:10 +0000)]
2010-02-18  William Chan  <willchan@chromium.org>

        Reviewed by Adam Barth.

        https://bugs.webkit.org/show_bug.cgi?id=35071
        Disable loader limiting of requests per host for the chromium port.

        No tests because we're only changing a constant.

        * platform/network/chromium/ResourceRequest.cpp:
        (WebCore::initializeMaximumHTTPConnectionCountPerHost):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@54946 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-02-18 Eric Seidel <eric@webkit.org>
eric@webkit.org [Thu, 18 Feb 2010 08:54:47 +0000 (08:54 +0000)]
2010-02-18  Eric Seidel  <eric@webkit.org>

        Reviewed by Adam Barth.

        run-chromium-webkit-tests --platform=mac-leopard needs to run build-dumprendertree
        https://bugs.webkit.org/show_bug.cgi?id=35053

        * Scripts/webkitpy/layout_tests/port/base.py: Add script_path() function for finding scripts.
        * Scripts/webkitpy/layout_tests/port/chromium.py: Remove unused argument.
        * Scripts/webkitpy/layout_tests/port/mac.py:
         - Make sure that calling webkit-build-directory works even if Scripts/ is not in the user's path.
         - Call build-dumprendertree (and make sure it succeeds) before running the tests.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@54945 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-02-18 Chris Evans <cevans@chromium.org>
eric@webkit.org [Thu, 18 Feb 2010 08:38:32 +0000 (08:38 +0000)]
2010-02-18  Chris Evans  <cevans@chromium.org>

        Reviewed by Adam Barth.

        Add the promised test for the unique file:// origins support.
        See https://bugs.webkit.org/show_bug.cgi?id=34778 for more details.

        * fast/xmlhttprequest/xmlhttprequest-no-file-access.html: Added.
        * fast/xmlhttprequest/resources/xmlhttprequest-no-file-access-real.html: Added.
        * fast/xmlhttprequest/xmlhttprequest-no-file-access-expected.txt: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@54944 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoWebKit on Windows needs a mechanism to listen for WM_SETTINGCHANGED messages
sfalken@apple.com [Thu, 18 Feb 2010 08:36:39 +0000 (08:36 +0000)]
WebKit on Windows needs a mechanism to listen for WM_SETTINGCHANGED messages
https://bugs.webkit.org/show_bug.cgi?id=35076

Reviewed by Dan Bernstein.

WebKit/win:

* WebKit.vcproj/WebKit.def: Added WebKitSystemParameterChanged.
* WebKit.vcproj/WebKit_debug.def: Added WebKitSystemParameterChanged.
* WebKitGraphics.cpp:
(WebKitSystemParameterChanged): Call through to wkSystemFontSmoothingChanged for font smoothing changes.
* WebKitGraphics.h: Added WebKitSystemParameterChanged.

WebKitLibraries:

* win/include/WebKitSystemInterface/WebKitSystemInterface.h:
* win/lib/WebKitSystemInterface.lib: Updated to add wkSystemFontSmoothingChanged.
* win/lib/WebKitSystemInterface_debug.lib: Updated to add wkSystemFontSmoothingChanged.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@54943 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-02-16 Chris Jerdonek <cjerdonek@webkit.org>
hamaji@chromium.org [Thu, 18 Feb 2010 08:17:39 +0000 (08:17 +0000)]
2010-02-16  Chris Jerdonek  <cjerdonek@webkit.org>

        Reviewed by Shinichiro Hamaji.

        Refactored check-webkit-style's ProcessorOptions class into two
        classes. This revision contains no new functionality.

        https://bugs.webkit.org/show_bug.cgi?id=34674

        Divided the ProcessorOptions class into a CommandOptionValues
        class (the result of parsing the command-line options) and
        a StyleCheckerConfiguration class (which configures the main
        StyleChecker).

        * Scripts/check-webkit-style:
          - Updated main() to convert the parsed command option values
            to a StyleCheckConfiguration instance prior to constructing
            a StyleChecker.

        * Scripts/webkitpy/style/checker.py:
          - Added check_webkit_style_configuration() to convert a
            CommandOptionValues instance into a StyleCheckerConfiguration
            instance.
          - Renamed the ProcessorOptions class to CommandOptionValues.
          - In the CommandOptionValues class--
            - Replaced the filter_configuration attribute with the
              simpler filter_rules attribute.
            - Removed the max_reports_per_error attribute.
            - Moved the is_reportable() method to the new
              StyleCheckerConfiguration class.
          - Removed the base_filter_rules attribute from the
            DefaultCommandOptionValues class.
          - In the ArgumentParser class--
            - Added base_filter_rules to the constructor.
            - Changed the parse() method to return a CommandOptionValues
              instance instead of a ProcessorOptions instance.
          - Created a StyleCheckerConfiguration class.
            - Added the data attributes max_reports_per_category,
              stderr_write, and verbosity.
            - Added is_reportable() (moved from the ProcessorOptions
              class) and write_style_error() (moved from the
              DefaultStyleErrorHandler class).
          - In the StyleChecker class--
            - Replaced the ProcessorOptions options attribute with the
              StyleCheckerConfiguration _configuration attribute.
            - Removed the _stderr_write attribute.

        * Scripts/webkitpy/style/checker_unittest.py:
          - Updated the existing unit test classes as necessary.
          - Added a StyleCheckerConfigurationTest class.

        * Scripts/webkitpy/style/error_handlers.py:
          - Updated the DefaultStyleErrorHandler class to accept a
            StyleCheckerConfiguration instance instead of a ProcessorOptions
            instance and an stderr_write method.

        * Scripts/webkitpy/style/error_handlers_unittest.py:
          - Updated the unit test classes as necessary.

        * Scripts/webkitpy/style/filter.py:
          - Addressed the FIXME in the FilterConfiguration class to change
            the user_rules attribute to _user_rules (since it is now
            accessed only internally).

        * Scripts/webkitpy/style/filter_unittest.py:
          - Updated to reflect the change from user_rules to _user_rules.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@54942 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-02-18 Tony Chang <tony@chromium.org>
tony@chromium.org [Thu, 18 Feb 2010 08:17:27 +0000 (08:17 +0000)]
2010-02-18  Tony Chang  <tony@chromium.org>

        Not reviewed, adding a QT layout test result for a new test
        added in r54932.

        * platform/qt/editing/pasteboard/page-zoom-expected.txt: Copied from LayoutTests/platform/gtk/editing/pasteboard/page-zoom-expected.txt.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@54941 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-02-18 Xan Lopez <xlopez@igalia.com>
xan@webkit.org [Thu, 18 Feb 2010 08:06:03 +0000 (08:06 +0000)]
2010-02-18  Xan Lopez  <xlopez@igalia.com>

        Reviewed by Eric Seidel.

        [Linux] Webkit incompatible with Java plugins
        https://bugs.webkit.org/show_bug.cgi?id=24912

        The NP Version supported by WebKit is at the moment hardcoded in
        PluginPackage.cpp (to 24), but not all backends actually implement
        the methods needed to claim this. Introduce a new method to be
        implemented by each backend where the maximum supported version
        can be specified, and set the GTK+ port NPVersion to 19. This
        fixes an instantaneous crasher in the Sun Java NPAPI plugin.

        * plugins/PluginPackage.cpp:
        (WebCore::PluginPackage::initializeBrowserFuncs):
        * plugins/PluginPackage.h:
        * plugins/gtk/PluginPackageGtk.cpp:
        (WebCore::PluginPackage::NPVersion):
        * plugins/mac/PluginPackageMac.cpp:
        (WebCore::PluginPackage::NPVersion):
        * plugins/qt/PluginPackageQt.cpp:
        (WebCore::PluginPackage::NPVersion):
        * plugins/symbian/PluginPackageSymbian.cpp:
        (WebCore::PluginPackage::NPVersion):
        * plugins/win/PluginPackageWin.cpp:
        (WebCore::PluginPackage::NPVersion):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@54940 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago[Qt] editing/pasteboard/page-zoom.html fails
ossy@webkit.org [Thu, 18 Feb 2010 08:04:55 +0000 (08:04 +0000)]
[Qt] editing/pasteboard/page-zoom.html fails
https://bugs.webkit.org/show_bug.cgi?id=35077

* platform/qt/Skipped: editing/pasteboard/page-zoom.html skipped until fix.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@54939 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoWhen a live iframe element is moved between pages, it still depends on the old page.
dimich@chromium.org [Thu, 18 Feb 2010 07:55:06 +0000 (07:55 +0000)]
When a live iframe element is moved between pages, it still depends on the old page.
https://bugs.webkit.org/show_bug.cgi?id=34382

Reviewed by David Levin, Darin Fisher, Simon Hausmann.

WebCore:

Test: fast/frames/iframe-reparenting-new-page.html

* html/HTMLFrameElementBase.cpp:
(WebCore::HTMLFrameElementBase::setName):
Move the code setting the frame name into a separate function.

(WebCore::HTMLFrameElementBase::setNameAndOpenURL):
(WebCore::HTMLFrameElementBase::updateOnReparenting):
Called on the frame that was just re-parented and inserted into another Document.
Simply invoke Frame::transferChildFrameToNewDocument(...);

(WebCore::HTMLFrameElementBase::insertedIntoDocument):
* html/HTMLFrameElementBase.h:
* html/HTMLFrameOwnerElement.h:
(WebCore::HTMLFrameOwnerElement::setName):
Make this a virtual function, to be able to reach it via Frame::m_ownerElement.

* loader/EmptyClients.h:
(WebCore::EmptyFrameLoaderClient::adoptFrame):
* loader/FrameLoaderClient.h:
Add a new method, didTransferChildFrameToNewDocument().
It compliments createFrame() in that a frame which was re-parented
in DOM and potentially changes Page, should notify the WebKit
implementation about potential ownership change.
Many embedders assume that Page owns all the Frames, or at least
all Frames are destroyed before 'their' Page is destroyed. Obviously, if Frame
can be transferred to another Page, the embedders should be notified.

* page/Frame.cpp:
(WebCore::Frame::transferChildFrameToNewDocument):
Added, makes actual adjustments for Frame - resets the Page,
updates the frame tree and frame name, calls to FrameLoaderClient
to update external objects and recurses into children.
Can only be used on child frames.

* page/Frame.h:

WebKit/chromium:

* src/FrameLoaderClientImpl.cpp:
(WebKit::FrameLoaderClientImpl::detachedFromParent3):
(WebKit::FrameLoaderClientImpl::didTransferChildFrameToNewDocument):
Since Chromium's WebFrameClient is destroyed once a window is closed,
if Frame moves between Pages, the client of corresponding WebFrame
should be replaced as well.

* src/FrameLoaderClientImpl.h:
* src/WebFrameImpl.h:
(WebKit::WebFrameImpl::setClient):

WebKit/gtk:

* WebCoreSupport/FrameLoaderClientGtk.cpp:
(WebKit::FrameLoaderClient::didTransferChildFrameToNewDocument):
Added empty implementation of a new virtual method.

* WebCoreSupport/FrameLoaderClientGtk.h:

WebKit/haiku:

* WebCoreSupport/FrameLoaderClientHaiku.cpp:
(WebCore::FrameLoaderClientHaiku::didTransferChildFrameToNewDocument):
Added empty implementation of a new virtual method.

* WebCoreSupport/FrameLoaderClientHaiku.h:

WebKit/mac:

* WebCoreSupport/WebFrameLoaderClient.h:
* WebCoreSupport/WebFrameLoaderClient.mm:
Added empty implementation of a new virtual method.

(WebFrameLoaderClient::didTransferChildFrameToNewDocument):

WebKit/qt:

* Api/qwebframe_p.h:
(QWebFramePrivate::setPage): Added.
* WebCoreSupport/FrameLoaderClientQt.cpp:
(WebCore::FrameLoaderClientQt::didTransferChildFrameToNewDocument):
The QWebFrame caches a QWebPage which should be replaced when Frame is re-parented.
Also, the QWebFrame is a child (in QT terms) of QWebPage - so update that relationship as well.
Emit a signal that QWebFrame moved to a different QWebPage.

* WebCoreSupport/FrameLoaderClientQt.h:

WebKit/win:

* WebCoreSupport/WebFrameLoaderClient.cpp:
(WebFrameLoaderClient::didTransferChildFrameToNewDocument):
Added empty implementation of a new virtual method.

* WebCoreSupport/WebFrameLoaderClient.h:

WebKit/wx:

* WebKitSupport/FrameLoaderClientWx.cpp:
(WebCore::FrameLoaderClientWx::didTransferChildFrameToNewDocument):
Added empty implementation of a new virtual method.

* WebKitSupport/FrameLoaderClientWx.h:

LayoutTests:

* fast/frames/iframe-reparenting-new-page-expected.txt: Added.
* fast/frames/iframe-reparenting-new-page.html: Added.
* fast/frames/resources/iframe-reparenting-new-page-1.html: Added.
* fast/frames/resources/iframe-reparenting-new-page-2.html: Added.
* fast/frames/resources/iframe-reparenting-new-page-iframe.html: Added.
* fast/frames/script-tests/TEMPLATE.html: Copied from LayoutTests/fast/canvas/script-tests/TEMPLATE.html.
* fast/frames/script-tests/iframe-reparenting-new-page.js: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@54938 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago [Qt] DRT: Get total number of pages to be printed
ossy@webkit.org [Thu, 18 Feb 2010 07:49:15 +0000 (07:49 +0000)]
 [Qt] DRT: Get total number of pages to be printed
 https://bugs.webkit.org/show_bug.cgi?id=34955

 Fix landed in r54902, so we can unskip the test.

* platform/qt/Skipped: printing/numberOfPages.html unskipped.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@54937 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago[Qt] Fails glyph reordering test fast/text/glyph-reordering.html
ossy@webkit.org [Thu, 18 Feb 2010 07:41:32 +0000 (07:41 +0000)]
[Qt] Fails glyph reordering test fast/text/glyph-reordering.html
https://bugs.webkit.org/show_bug.cgi?id=34918

Fix landed in r54912, so we can unskip the test.

* platform/qt/Skipped: fast/text/glyph-reordering.html unskipped.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@54936 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-02-17 Tony Chang <tony@chromium.org>
tony@chromium.org [Thu, 18 Feb 2010 07:30:19 +0000 (07:30 +0000)]
2010-02-17  Tony Chang  <tony@chromium.org>

        Not reviewed, adding a GTK+ layout test result for a new test
        added in r54932.

        https://bugs.webkit.org/show_bug.cgi?id=34914

        * platform/gtk/editing/pasteboard/page-zoom-expected.txt: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@54935 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoRefactored check-webkit-style's ProcessorOptions class into two
cjerdonek@webkit.org [Thu, 18 Feb 2010 07:29:12 +0000 (07:29 +0000)]
Refactored check-webkit-style's ProcessorOptions class into two
classes. This revision contains no new functionality.

Reviewed by Shinichiro Hamaji.

https://bugs.webkit.org/show_bug.cgi?id=34674

Divided the ProcessorOptions class into a CommandOptionValues
class (the result of parsing the command-line options) and
a StyleCheckerConfiguration class (which configures the main
StyleChecker).

* Scripts/check-webkit-style:
  - Updated main() to convert the parsed command option values
    to a StyleCheckConfiguration instance prior to constructing
    a StyleChecker.

* Scripts/webkitpy/style/checker.py:
  - Added check_webkit_style_configuration() to convert a
    CommandOptionValues instance into a StyleCheckerConfiguration
    instance.
  - Renamed the ProcessorOptions class to CommandOptionValues.
  - In the CommandOptionValues class--
    - Replaced the filter_configuration attribute with the
      simpler filter_rules attribute.
    - Removed the max_reports_per_error attribute.
    - Moved the is_reportable() method to the new
      StyleCheckerConfiguration class.
  - Removed the base_filter_rules attribute from the
    DefaultCommandOptionValues class.
  - In the ArgumentParser class--
    - Added base_filter_rules to the constructor.
    - Changed the parse() method to return a CommandOptionValues
      instance instead of a ProcessorOptions instance.
  - Created a StyleCheckerConfiguration class.
    - Added the data attributes max_reports_per_category,
      stderr_write, and verbosity.
    - Added is_reportable() (moved from the ProcessorOptions
      class) and write_style_error() (moved from the
      DefaultStyleErrorHandler class).
  - In the StyleChecker class--
    - Replaced the ProcessorOptions options attribute with the
      StyleCheckerConfiguration _configuration attribute.
    - Removed the _stderr_write attribute.

* Scripts/webkitpy/style/checker_unittest.py:
  - Updated the existing unit test classes as necessary.
  - Added a StyleCheckerConfigurationTest class.

* Scripts/webkitpy/style/error_handlers.py:
  - Updated the DefaultStyleErrorHandler class to accept a
    StyleCheckerConfiguration instance instead of a ProcessorOptions
    instance and an stderr_write method.

* Scripts/webkitpy/style/error_handlers_unittest.py:
  - Updated the unit test classes as necessary.

* Scripts/webkitpy/style/filter.py:
  - Addressed the FIXME in the FilterConfiguration class to change
    the user_rules attribute to _user_rules (since it is now
    accessed only internally).

* Scripts/webkitpy/style/filter_unittest.py:
  - Updated to reflect the change from user_rules to _user_rules.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@54934 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-02-17 Tony Chang <tony@chromium.org>
tony@chromium.org [Thu, 18 Feb 2010 07:12:13 +0000 (07:12 +0000)]
2010-02-17  Tony Chang  <tony@chromium.org>

        Reviewed by Eric Seidel.

        https://bugs.webkit.org/show_bug.cgi?id=26937
        No longer allow span styles to surround block elements like
        divs when pasting (see paste-text-012 and 5065605 which had divs in
        spans).  This also causes a few cases of empty spans to be removed
        (see 19089 and 5245519).

        * editing/execCommand/19089-expected.txt:
        * editing/pasteboard/5245519-expected.txt:
        * editing/pasteboard/5245519.html:
        * platform/mac/editing/pasteboard/5065605-expected.txt:
        * platform/mac/editing/pasteboard/paste-text-011-expected.txt:
        * platform/mac/editing/pasteboard/paste-text-012-expected.txt:
2010-02-17  Tony Chang  <tony@chromium.org>

        Reviewed by Eric Seidel.

        Copying and pasting into a contenteditable area can create <div>s surrounded by <span>s
        https://bugs.webkit.org/show_bug.cgi?id=26937

        This happens because of a span added when we copy that is used to
        preserve styles.  To avoid this, when we paste, make sure to apply
        the styles to the span's children and then remove the style span.

        This change is covered by existing layout tests.

        * editing/ReplaceSelectionCommand.cpp:
        (WebCore::ReplaceSelectionCommand::handleStyleSpans):
        (WebCore::ReplaceSelectionCommand::copyStyleToChildren):
        * editing/ReplaceSelectionCommand.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@54933 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-02-17 Tony Chang <tony@chromium.org>
eric@webkit.org [Thu, 18 Feb 2010 06:44:48 +0000 (06:44 +0000)]
2010-02-17  Tony Chang  <tony@chromium.org>

        Reviewed by Eric Seidel.

        https://bugs.webkit.org/show_bug.cgi?id=34914
        When pasting absolute font sizes into a content editable area with
        page zoom, adjust the font sizes to be the same after page zoom is
        applied.

        * editing/pasteboard/page-zoom-expected.txt: Added.
        * editing/pasteboard/page-zoom.html: Added.
2010-02-17  Tony Chang  <tony@chromium.org>

        Reviewed by Eric Seidel.

        https://bugs.webkit.org/show_bug.cgi?id=34914
        When pasting absolute font sizes into a content editable area with
        page zoom, adjust the font sizes to be the same after page zoom is
        applied.

        Test: editing/pasteboard/page-zoom.html

        * editing/ReplaceSelectionCommand.cpp:
        (WebCore::ReplaceSelectionCommand::negateStyleRulesThatAffectAppearance):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@54932 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-02-17 Tony Chang <tony@chromium.org>
eric@webkit.org [Thu, 18 Feb 2010 06:28:30 +0000 (06:28 +0000)]
2010-02-17  Tony Chang  <tony@chromium.org>

        Reviewed by Eric Seidel.

        https://bugs.webkit.org/show_bug.cgi?id=34737
        Test to verify that pasting styled list items works
        the same as pasting unstyle list items.

        * editing/pasteboard/paste-list-003-expected.txt: Added.
        * editing/pasteboard/paste-list-003.html: Added.
2010-02-17  Tony Chang  <tony@chromium.org>

        Reviewed by Eric Seidel.

        https://bugs.webkit.org/show_bug.cgi?id=34737
        Copying styled list items then pasting into a list
        should work the same as copying unstyle list items:
        it shouldn't indent an extra level, but styles should
        be copied.

        Small cleanups to insertAsListItems to make variable names
        more descriptive.

        Test: editing/pasteboard/paste-list-003.html

        * editing/ReplaceSelectionCommand.cpp:
        (WebCore::ReplaceSelectionCommand::doApply):
        (WebCore::ReplaceSelectionCommand::insertAsListItems):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@54931 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-02-17 Kwang Yul Seo <skyul@company100.net>
eric@webkit.org [Thu, 18 Feb 2010 06:12:07 +0000 (06:12 +0000)]
2010-02-17  Kwang Yul Seo  <skyul@company100.net>

        Reviewed by Eric Seidel.

        [BREWMP] Port PlatformKeyboardEvent
        https://bugs.webkit.org/show_bug.cgi?id=34794

        Retrieve the type, key code, text and modifiers from BREW's keyboard event.

        * platform/PlatformKeyboardEvent.h:
        * platform/brew/PlatformKeyboardEventBrew.cpp: Added.
        (WebCore::keyIdentifierForBrewKeyCode):
        (WebCore::windowsKeyCodeForKeyEvent):
        (WebCore::singleCharacterString):
        (WebCore::PlatformKeyboardEvent::PlatformKeyboardEvent):
        (WebCore::PlatformKeyboardEvent::disambiguateKeyDownEvent):
        (WebCore::PlatformKeyboardEvent::currentCapsLockState):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@54930 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-02-17 Hayato Ito <hayato@chromium.org>
eric@webkit.org [Thu, 18 Feb 2010 05:55:17 +0000 (05:55 +0000)]
2010-02-17  Hayato Ito  <hayato@chromium.org>

        Reviewed by Eric Seidel.

        Support CSS page-break-inside with a value of 'avoid'.

        https://bugs.webkit.org/show_bug.cgi?id=34080

        * printing/page-break-inside-expected.txt: Added.
        * printing/page-break-inside.html: Added.
        * printing/script-tests/page-break-inside.js: Added.
        (createParagraph):
        (pageNumber):
2010-02-17  Hayato Ito  <hayato@chromium.org>

        Reviewed by Eric Seidel.

        Support CSS page-break-inside with a value of 'avoid'.

        https://bugs.webkit.org/show_bug.cgi?id=34080

        Test: printing/page-break-inside.html

        * rendering/RenderBlock.cpp:
        (WebCore::RenderBlock::paintChildren):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@54929 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-02-17 Evan Martin <evan@chromium.org>
eric@webkit.org [Thu, 18 Feb 2010 05:38:26 +0000 (05:38 +0000)]
2010-02-17  Evan Martin  <evan@chromium.org>

        Reviewed by Eric Seidel.

        [chromium] expose cache-bypass flag in reload() API
        https://bugs.webkit.org/show_bug.cgi?id=34969

        This flag is used to force a reload to ignore caches.
        (Aka "shift-reload".)

        * public/WebFrame.h:
        * src/WebFrameImpl.cpp:
        (WebKit::WebFrameImpl::reload):
        * src/WebFrameImpl.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@54928 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-02-17 Fumitoshi Ukai <ukai@chromium.org>
ukai@chromium.org [Thu, 18 Feb 2010 05:32:05 +0000 (05:32 +0000)]
2010-02-17  Fumitoshi Ukai  <ukai@chromium.org>

        Reviewed by Alexey Proskuryakov.

        WebSocket bufferedAmount should not be 0 when send after close.
        https://bugs.webkit.org/show_bug.cgi?id=34633

        * websocket/tests/bufferedAmount-after-close-expected.txt: Added.
        * websocket/tests/bufferedAmount-after-close.html: Added.
        * websocket/tests/script-tests/bufferedAmount-after-close.js: Added.
2010-02-17  Fumitoshi Ukai  <ukai@chromium.org>

        Reviewed by Alexey Proskuryakov.

        WebSocket bufferedAmount should not be 0 when send after close.
        https://bugs.webkit.org/show_bug.cgi?id=34633

        Fixed bug in webkit r54694.  check m_client after it calls user
        callback, because user callback may call close(), which would
        call onclose event handler.

        Test: websocket/tests/bufferedAmount-after-close.html

        * websockets/ThreadableWebSocketChannelClientWrapper.h:
        (WebCore::ThreadableWebSocketChannelClientWrapper::didClose):
        * websockets/WebSocket.cpp:
        (WebCore::WebSocket::WebSocket):
        (WebCore::WebSocket::send):
        (WebCore::WebSocket::close):
        (WebCore::WebSocket::bufferedAmount):
        (WebCore::WebSocket::didConnect):
        (WebCore::WebSocket::didClose):
        * websockets/WebSocket.h:
        * websockets/WebSocketChannel.cpp:
        (WebCore::WebSocketChannel::WebSocketChannel):
        (WebCore::WebSocketChannel::send):
        (WebCore::WebSocketChannel::bufferedAmount):
        (WebCore::WebSocketChannel::didClose):
        (WebCore::WebSocketChannel::didReceiveData):
        * websockets/WebSocketChannel.h:
        * websockets/WebSocketChannelClient.h:
        (WebCore::WebSocketChannelClient::didClose):
        * websockets/WorkerThreadableWebSocketChannel.cpp:
        (WebCore::workerContextDidClose):
        (WebCore::WorkerThreadableWebSocketChannel::Peer::didClose):
        * websockets/WorkerThreadableWebSocketChannel.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@54927 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-02-17 Yuzo Fujishima <yuzo@google.com>
eric@webkit.org [Thu, 18 Feb 2010 05:14:59 +0000 (05:14 +0000)]
2010-02-17  Yuzo Fujishima  <yuzo@google.com>

        Reviewed by Eric Seidel.

        In diffing render styles, consider all pseudo style changes.
        Without this patch, only :before and :after are considered. This is the cause of the following bug.
        https://bugs.webkit.org/show_bug.cgi?id=32187

        * fast/css/first-letter-first-line-hover.html: Added.
        * platform/mac/fast/css/first-letter-first-line-hover-expected.checksum: Added.
        * platform/mac/fast/css/first-letter-first-line-hover-expected.png: Added.
        * platform/mac/fast/css/first-letter-first-line-hover-expected.txt: Added.
2010-02-17  Yuzo Fujishima  <yuzo@google.com>

        Reviewed by Eric Seidel.

        In diffing render styles, consider all pseudo style changes.
        Without this patch, only :before and :after are considered. This is the cause of the following bug.
        https://bugs.webkit.org/show_bug.cgi?id=32187

        Test: fast/css/first-letter-first-line-hover.html

        * dom/Node.cpp:
        (WebCore::Node::diff):
        * rendering/style/RenderStyle.cpp:
        (WebCore::RenderStyle::hasAnyPublicPseudoStyles):
        * rendering/style/RenderStyle.h:
        * rendering/style/RenderStyleConstants.h:
        (WebCore::):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@54926 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agohttps://bugs.webkit.org/show_bug.cgi?id=35070
barraclough@apple.com [Thu, 18 Feb 2010 05:07:41 +0000 (05:07 +0000)]
https://bugs.webkit.org/show_bug.cgi?id=35070
Addition of 2 strings of length 2^31 may result in a string of length 0.

Reviewed by Oliver Hunt.

Check for overflow when creating a new JSString as a result of an addition
or concatenation, throw an out of memory exception.

* runtime/JSString.h:
(JSC::):
* runtime/Operations.h:
(JSC::jsString):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@54925 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-02-17 Kwang Yul Seo <skyul@company100.net>
eric@webkit.org [Thu, 18 Feb 2010 04:55:12 +0000 (04:55 +0000)]
2010-02-17  Kwang Yul Seo  <skyul@company100.net>

        Reviewed by Eric Seidel.

        [BREWMP] Add WebCore::prefetchDNS
        https://bugs.webkit.org/show_bug.cgi?id=34873

        DNS prefetching is not implemented because the maximum number
        of UDP sockets is quite small in most BREW devices.

        * platform/network/brew/DNSBrew.cpp: Added.
        (WebCore::prefetchDNS):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@54924 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-02-17 Kent Tamura <tkent@chromium.org>
tkent@chromium.org [Thu, 18 Feb 2010 04:36:36 +0000 (04:36 +0000)]
2010-02-17  Kent Tamura  <tkent@chromium.org>

        Reviewed by Eric Seidel.

        Introduces new Icon loading interface in order to support
        asynchronous loading.
        https://bugs.webkit.org/show_bug.cgi?id=32054

        It's hard for Chromium port to load an icon inside
        Icon::createIconForFiles() because of sanbox and multi-process
        architecture. So this change adds a method to request an icon to
        Chrome class, and makes FileChooser receives an Icon instance
        asynchronously.  Synchronous loading also works with the new interface.

        Because all ports don't have implementations of Chrome::iconForFiles()
        yet, FileChooser tries to load an Icon synchronously with
        Icon::createIconForFiles(), then tries to load an Icon asynchronously
        with Chrome::iconForFiles() if Icon::createIconForFiles() returns 0.

        The existing Icon::createIconForFiles() implementations should be
        moved to Chrome::iconForFiles(). We're going to remove
        Icon::createIconForFiles().

        * loader/EmptyClients.h:
        (WebCore::EmptyChromeClient::iconForFiles): Add an empty implementation.
        * page/Chrome.cpp:
        (WebCore::Chrome::iconForFiles): Delegate to ChromeClient::iconForFiles().
        * page/Chrome.h:
        * page/ChromeClient.h:
        (WebCore::ChromeClient::iconForFiles): Add a declaration as a pure virtual method.
        * platform/FileChooser.cpp:
        (WebCore::FileChooser::FileChooser): Use loadIcon().
        (WebCore::FileChooser::chooseFiles): ditto.
        (WebCore::FileChooser::loadIcon): Added.
        (WebCore::FileChooser::iconLoaded): Added.
        * platform/FileChooser.h: Add two methods to FileChooserClient; repaint() and iconForFiles().
        * platform/graphics/Icon.h: Add a FIXME comment.
        * platform/graphics/gtk/IconGtk.cpp: ditto.
        * platform/graphics/mac/IconMac.mm: ditto.
        * platform/graphics/qt/IconQt.cpp: ditto.
        * platform/graphics/win/IconWin.cpp: ditto.
        * rendering/RenderFileUploadControl.cpp:
        (WebCore::RenderFileUploadControl::iconForFiles): Delegate to Chrome::iconForFiles().
        (WebCore::RenderFileUploadControl::click): Use chrome().
        (WebCore::RenderFileUploadControl::chrome):
        * rendering/RenderFileUploadControl.h:
        (WebCore::RenderFileUploadControl::repaint):

2010-02-17  Kent Tamura  <tkent@chromium.org>

        Reviewed by Eric Seidel.

        Introduces new Icon loading interface in order to support
        asynchronous loading.
        https://bugs.webkit.org/show_bug.cgi?id=32054

        Add an empty implementation of ChromeClient::iconForFiles().

        * src/ChromeClientImpl.cpp:
        (WebKit::ChromeClientImpl::iconForFiles):
        * src/ChromeClientImpl.h:

2010-02-17  Kent Tamura  <tkent@chromium.org>

        Reviewed by Eric Seidel.

        Introduces new Icon loading interface in order to support
        asynchronous loading.
        https://bugs.webkit.org/show_bug.cgi?id=32054

        Add an empty implementation of ChromeClient::iconForFiles().

        * WebCoreSupport/ChromeClientGtk.cpp:
        (WebKit::ChromeClient::iconForFiles):
        * WebCoreSupport/ChromeClientGtk.h:

2010-02-17  Kent Tamura  <tkent@chromium.org>

        Reviewed by Eric Seidel.

        Introduces new Icon loading interface in order to support
        asynchronous loading.
        https://bugs.webkit.org/show_bug.cgi?id=32054

        Add an empty implementation of ChromeClient::iconForFiles().

        * WebCoreSupport/ChromeClientHaiku.cpp:
        (WebCore::ChromeClientHaiku::iconForFiles):
        * WebCoreSupport/ChromeClientHaiku.h:

2010-02-17  Kent Tamura  <tkent@chromium.org>

        Reviewed by Eric Seidel.

        Introduces new Icon loading interface in order to support
        asynchronous loading.
        https://bugs.webkit.org/show_bug.cgi?id=32054

        Add an empty implementation of ChromeClient::iconForFiles().

        * WebCoreSupport/WebChromeClient.h:
        * WebCoreSupport/WebChromeClient.mm:
        (WebChromeClient::iconForFiles):

2010-02-17  Kent Tamura  <tkent@chromium.org>

        Reviewed by Eric Seidel.

        Introduces new Icon loading interface in order to support
        asynchronous loading.
        https://bugs.webkit.org/show_bug.cgi?id=32054

        Add an empty implementation of ChromeClient::iconForFiles().

        * WebCoreSupport/ChromeClientQt.cpp:
        (WebCore::ChromeClientQt::iconForFiles):
        * WebCoreSupport/ChromeClientQt.h:

2010-02-17  Kent Tamura  <tkent@chromium.org>

        Reviewed by Eric Seidel.

        Introduces new Icon loading interface in order to support
        asynchronous loading.
        https://bugs.webkit.org/show_bug.cgi?id=32054

        Add an empty implementation of ChromeClient::iconForFiles().

        * WebCoreSupport/WebChromeClient.cpp:
        (WebChromeClient::iconForFiles):
        * WebCoreSupport/WebChromeClient.h:

2010-02-17  Kent Tamura  <tkent@chromium.org>

        Reviewed by Eric Seidel.

        Introduces new Icon loading interface in order to support
        asynchronous loading.
        https://bugs.webkit.org/show_bug.cgi?id=32054

        Add an empty implementation of ChromeClient::iconForFiles().

        * WebKitSupport/ChromeClientWx.cpp:
        (WebCore::ChromeClientWx::iconForFiles):
        * WebKitSupport/ChromeClientWx.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@54923 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-02-17 Shinichiro Hamaji <hamaji@chromium.org>
hamaji@chromium.org [Thu, 18 Feb 2010 04:19:03 +0000 (04:19 +0000)]
2010-02-17  Shinichiro Hamaji  <hamaji@chromium.org>

        Unreviewed. Touch WebKit.idl to fix the build.

        [Win] Implement test functions for printing
        https://bugs.webkit.org/show_bug.cgi?id=34570

        * Interfaces/WebKit.idl:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@54922 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-02-17 Shinichiro Hamaji <hamaji@chromium.org>
hamaji@chromium.org [Thu, 18 Feb 2010 02:42:00 +0000 (02:42 +0000)]
2010-02-17  Shinichiro Hamaji  <hamaji@chromium.org>

        Reviewed by Eric Seidel.

        [Win] Implement layoutTestController.pageNumberForElementById
        https://bugs.webkit.org/show_bug.cgi?id=34570

        * platform/win/Skipped:
2010-02-17  Shinichiro Hamaji  <hamaji@chromium.org>

        Reviewed by Eric Seidel.

        [Win] Implement test functions for printing
        https://bugs.webkit.org/show_bug.cgi?id=34570

        * Interfaces/IWebFramePrivate.idl:
        * WebFrame.cpp:
        (WebFrame::pageNumberForElementById):
        (WebFrame::numberOfPages):
        * WebFrame.h:
2010-02-17  Shinichiro Hamaji  <hamaji@chromium.org>

        Reviewed by Eric Seidel.

        [Win] Implement test functions for printing
        https://bugs.webkit.org/show_bug.cgi?id=34570

        * DumpRenderTree/win/LayoutTestControllerWin.cpp:
        (LayoutTestController::pageNumberForElementById):
        (LayoutTestController::numberOfPages):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@54921 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-02-17 Fumitoshi Ukai <ukai@chromium.org>
ukai@chromium.org [Thu, 18 Feb 2010 02:19:37 +0000 (02:19 +0000)]
2010-02-17  Fumitoshi Ukai  <ukai@chromium.org>

        Reviewed by Eric Seidel.

        websocket/tests/cookies/httponly-cookie.pl fails every time under run-chromium-webkit-tests
        https://bugs.webkit.org/show_bug.cgi?id=35001

        * Scripts/webkitpy/layout_tests/port/websocket_server.py: add -x flag to run pywebsocket.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@54920 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoBug 35065: Delay between page loading and animated GIF playing
mrowe@apple.com [Thu, 18 Feb 2010 02:19:30 +0000 (02:19 +0000)]
Bug 35065: Delay between page loading and animated GIF playing
<http://webkit.org/b/35065> / <rdar://problem/7109548>

Reviewed by Darin Adler.

BitmapImage::startAnimation was adding the current frame duration to the desired start time
of the frame for every time it was called.  If the function then bailed out due to not having
sufficient data to render the frame, this would lead to the desired start time of the frame
being pushed out multiple times.  On an animated GIF that took mulitple seconds to load this
could happen many times for a single frame, resulting in the start time of the second frame
of the animation being pushed out by as much as five seconds.

* platform/graphics/BitmapImage.cpp:
(WebCore::BitmapImage::startAnimation): Change the order of the code slightly so that the
desired start time is only updated after determining that we have sufficient data to handle
the next frame.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@54919 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-02-17 Shinichiro Hamaji <hamaji@chromium.org>
hamaji@chromium.org [Thu, 18 Feb 2010 02:13:03 +0000 (02:13 +0000)]
2010-02-17  Shinichiro Hamaji  <hamaji@chromium.org>

        Reviewed by Eric Seidel.

        check-webkit-style: Misses variables that contain underscores.
        https://bugs.webkit.org/show_bug.cgi?id=33724

        - Check identifiers whose types are unsigned.
        - Check bitfields properly.

        * Scripts/webkitpy/style/processors/cpp.py:
        * Scripts/webkitpy/style/processors/cpp_unittest.py:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@54918 268f45cc-cd09-0410-ab3c-d52691b4dbfc