WebKit-https.git
11 years agoWebKit/mac: WebKit part of making the appearance of the full-screen video HUD match
mitz@apple.com [Mon, 2 Nov 2009 17:03:14 +0000 (17:03 +0000)]
WebKit/mac: WebKit part of making the appearance of the full-screen video HUD match
QuickTime Player X’s HUD.

Reviewed by John Sullivan.

* WebView/WebVideoFullscreenHUDWindowController.h: Removed unnecessary
#import statements, cleaned up style, and changed _timeline,
_volumeSlider and _playButton to have more specific types.
* WebView/WebVideoFullscreenHUDWindowController.mm: Updated #import
statements.
(webkit_CGFloor): Added this helper function.
(-[WebVideoFullscreenHUDWindowController init]): Cleaned up style.
(createTimeTextField): Changed to use the bold system font.
(-[WebVideoFullscreenHUDWindowController windowDidLoad]): Changed the
subviews’ metrics and the text fields’ text alignment.
(-[WebVideoFullscreenHUDWindowController updateTime]): Avoid conversion
from double to float.
(stringToTimeTextAttributed): Removed this useless function that
returned an NSAttributedString masquerading as an NSString.
(-[WebVideoFullscreenHUDWindowController remainingTimeText]): Removed
call to stringToTimeTextAttributed().
(-[WebVideoFullscreenHUDWindowController elapsedTimeText]): Ditto.

WebKitLibraries: WebKitSystemInterface part of making the appearance of the full-screen video HUD match
QuickTime Player X’s HUD.

Reviewed by John Sullivan.

* libWebKitSystemInterfaceLeopard.a:
* libWebKitSystemInterfaceSnowLeopard.a:
* libWebKitSystemInterfaceTiger.a:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@50410 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoSupport ARIA "tab" roles
cfleizach@apple.com [Mon, 2 Nov 2009 16:57:47 +0000 (16:57 +0000)]
Support ARIA "tab" roles
https://bugs.webkit.org/show_bug.cgi?id=30842

Reviewed by Beth Dakin.

WebCore:

Implement support for ARIA "tab", "tabpanel" and "tablist".
As a consequence, we also needed to implement aria-selected
and aria-controls.

Tests: accessibility/aria-controls-with-tabs.html
       accessibility/aria-tab-roles.html

* accessibility/AXObjectCache.cpp:
* accessibility/AccessibilityObject.h:
* accessibility/AccessibilityRenderObject.cpp:
* accessibility/AccessibilityRenderObject.h:
* accessibility/mac/AccessibilityObjectWrapper.mm:
* html/HTMLAttributeNames.in:

WebKit:

Add a localizable string for tab panel.

* English.lproj/Localizable.strings:
* StringsNotToBeLocalized.txt:

WebKit/mac:

* WebCoreSupport/WebViewFactory.mm:
(-[WebViewFactory AXARIAContentGroupText:]):

WebKitTools:

* DumpRenderTree/AccessibilityUIElement.cpp:
* DumpRenderTree/AccessibilityUIElement.h:
* DumpRenderTree/gtk/AccessibilityUIElementGtk.cpp:
* DumpRenderTree/mac/AccessibilityUIElementMac.mm:
* DumpRenderTree/win/AccessibilityUIElementWin.cpp:

LayoutTests:

* accessibility/aria-controls-with-tabs-expected.txt: Added.
* accessibility/aria-controls-with-tabs.html: Added.
* accessibility/aria-tab-roles.html: Added.
* platform/gtk/Skipped:
* platform/mac/accessibility/aria-tab-roles-expected.txt: Added.
* platform/win/Skipped:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@50409 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoThis is the WebKit-side change needed to fix canvas.getImageData() for
senorblanco@chromium.org [Mon, 2 Nov 2009 15:27:16 +0000 (15:27 +0000)]
This is the WebKit-side change needed to fix canvas.getImageData() for
Chromium.  The unpremultiply code in Skia assumes that unpremultiplied
values should be rounded, while CG does not.  In addition, the fixed
point inversion used by Skia introduces slight inaccuracies that make
us fail this test.  This change brings Chromium in line with
the CG path.
https://bugs.webkit.org/show_bug.cgi?id=30825

Reviewed by Dmitry Titov.

Covered by LayoutTests/fast/canvas/canvas-getImageData.html

* platform/graphics/skia/ImageBufferSkia.cpp:
(WebCore::getImageData):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@50408 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-11-01 Kelly Norton <knorton@google.com>
pfeldman@chromium.org [Mon, 2 Nov 2009 09:37:27 +0000 (09:37 +0000)]
2009-11-01  Kelly Norton  <knorton@google.com>

        Reviewed by Timothy Hatcher.

        Adds window event dispatches to InspectorTimelineAgent.
        https://bugs.webkit.org/show_bug.cgi?id=31002

        * English.lproj/localizedStrings.js:
        * dom/Node.cpp: Updated call site to willDispatchEvent and didDispatchEvent.
        (WebCore::Node::dispatchGenericEvent):
        * inspector/InspectorTimelineAgent.cpp:
        (WebCore::InspectorTimelineAgent::willDispatchEvent): Renamed.
        (WebCore::InspectorTimelineAgent::didDispatchEvent): Renamed.
        * inspector/InspectorTimelineAgent.h:
        (WebCore::):
        * inspector/TimelineRecordFactory.cpp:
        (WebCore::TimelineRecordFactory::createEventDispatchRecord): Renamed.
        * inspector/TimelineRecordFactory.h:
        * inspector/front-end/TimelineAgent.js:
        * inspector/front-end/TimelinePanel.js:
        (WebInspector.TimelinePanel.prototype._formatRecord):
        (WebInspector.TimelinePanel.prototype._getRecordDetails):
        * page/DOMWindow.cpp:
        (WebCore::DOMWindow::dispatchEvent):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@50407 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-11-01 Pavel Feldman <pfeldman@chromium.org>
pfeldman@chromium.org [Mon, 2 Nov 2009 09:25:29 +0000 (09:25 +0000)]
2009-11-01  Pavel Feldman  <pfeldman@chromium.org>

        Reviewed by Timothy Hatcher.

        Web Inspector: Convert script tag event into a more generic
        script eval event in timeline.

        https://bugs.webkit.org/show_bug.cgi?id=30999

        * bindings/js/ScriptController.cpp:
        (WebCore::ScriptController::evaluateInWorld):
        * bindings/v8/V8Proxy.cpp:
        (WebCore::V8Proxy::evaluate):
        * html/HTMLTokenizer.cpp:
        (WebCore::HTMLTokenizer::scriptExecution):
        * inspector/InspectorTimelineAgent.cpp:
        (WebCore::InspectorTimelineAgent::willEvaluateScript):
        (WebCore::InspectorTimelineAgent::didEvaluateScript):
        * inspector/InspectorTimelineAgent.h:
        (WebCore::):
        * inspector/TimelineRecordFactory.cpp:
        (WebCore::TimelineRecordFactory::createEvaluateScriptTimelineRecord):
        * inspector/TimelineRecordFactory.h:
        * inspector/front-end/TimelineAgent.js:
        * inspector/front-end/TimelinePanel.js:
        (WebInspector.TimelinePanel.prototype._formatRecord):
        (WebInspector.TimelinePanel.prototype._getRecordDetails):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@50406 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoRubber-stamped by Mark Rowe.
bweinstein@apple.com [Mon, 2 Nov 2009 06:21:09 +0000 (06:21 +0000)]
Rubber-stamped by Mark Rowe.

Fix for loop to use an size_t instead of unsigned and some spacing
style fixes.

* dom/Node.cpp:
(WebCore::eventHasListeners):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@50405 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoRe-fix the Mac build. Sorry :-(.
bweinstein@apple.com [Mon, 2 Nov 2009 06:03:38 +0000 (06:03 +0000)]
Re-fix the Mac build. Sorry :-(.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@50404 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoChange ssize_t to unsigned to try and fix Windows build.
bweinstein@apple.com [Mon, 2 Nov 2009 05:57:39 +0000 (05:57 +0000)]
Change ssize_t to unsigned to try and fix Windows build.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@50403 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoMade the space bar toggle playing state in full-screen video when
mitz@apple.com [Mon, 2 Nov 2009 03:53:23 +0000 (03:53 +0000)]
Made the space bar toggle playing state in full-screen video when
modifier keys are down. Made it do so without highlighting the
Play/Pause button.

Reviewed by Mark Rowe.

* WebView/WebVideoFullscreenHUDWindowController.mm:
(-[WebVideoFullscreenHUDWindowController keyDown:]):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@50402 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-11-01 Eric Seidel <eric@webkit.org>
eric@webkit.org [Mon, 2 Nov 2009 03:30:54 +0000 (03:30 +0000)]
2009-11-01  Eric Seidel  <eric@webkit.org>

        Reviewed by David Levin.

        buildbots should use --exit-after-N-failures
        https://bugs.webkit.org/show_bug.cgi?id=30809

        Make the bots exit after 20 failures to prevent never-ending
        test runs where every test spends a minute crashing.

        * BuildSlaveSupport/build.webkit.org-config/master.cfg:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@50401 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-11-01 Kelly Norton <knorton@google.com>
eric@webkit.org [Mon, 2 Nov 2009 03:23:16 +0000 (03:23 +0000)]
2009-11-01  Kelly Norton  <knorton@google.com>

        Reviewed by Timothy Hatcher.

        Does not send DOM dispatches to the InspectorTimelineAgent if there
        are no event listeners.
        https://bugs.webkit.org/show_bug.cgi?id=30995

        * dom/Node.cpp:
        (WebCore::eventHasListeners):
        (WebCore::Node::dispatchGenericEvent):
        * xml/XMLHttpRequest.cpp:
        (WebCore::XMLHttpRequest::callReadyStateChangeListener):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@50400 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoMade the full-screen video HUD respond to the up and down arrow keys by
mitz@apple.com [Mon, 2 Nov 2009 03:14:51 +0000 (03:14 +0000)]
Made the full-screen video HUD respond to the up and down arrow keys by
increasing and decreasing the volume by 1/10 of the range or, when
combined with the Option key, all the way up or down.

Reviewed by Mark Rowe.

Made the volume buttons in the full-screen video HUD match the behavior
of their equivalents in the QuickTime Player HUD by turning the volume
all the way up or down.

Made the volume slider update immediately when the keyboard or volume
buttons are used to change the volume, rather than at the nearest 1/4
second interval.

Made the elapsed and remaining time displays update immediately when the
play head is dragged across the timeline, rather than at 1/4 second
intervals.

* WebView/WebVideoFullscreenHUDWindowController.mm:
(-[WebVideoFullscreenHUDWindowController keyDown:]): Handle the up and
down arrow keys.
(-[WebVideoFullscreenHUDWindowController windowDidLoad]): Changed the
actions of the volume up and volume down buttons.
(-[WebVideoFullscreenHUDWindowController setCurrentTime:]): Call
-updateTime.
(-[WebVideoFullscreenHUDWindowController setVolumeToZero:]): Added this
action for the volume down button.
(-[WebVideoFullscreenHUDWindowController setVolumeToMaximum:]): Added
this action for the volume up button.
(-[WebVideoFullscreenHUDWindowController decrementVolume]): No longer
and action method.
(-[WebVideoFullscreenHUDWindowController incrementVolume]): Ditto.
(-[WebVideoFullscreenHUDWindowController setVolume:]): Call
-updateVolume.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@50399 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoNo review (build fix).
rolandsteiner@chromium.org [Mon, 2 Nov 2009 02:50:58 +0000 (02:50 +0000)]
No review (build fix).

Add missing files for Ruby implementation to WebCore.vcproj
(Fix build break after 50397)
https://bugs.webkit.org/show_bug.cgi?id=31001

No new tests. (No functional change)

* WebCore.vcproj/WebCore.vcproj:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@50398 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoBug 28420 - Implement HTML5 <ruby> rendering
rolandsteiner@chromium.org [Mon, 2 Nov 2009 01:52:25 +0000 (01:52 +0000)]
Bug 28420 -  Implement HTML5 <ruby> rendering
(https://bugs.webkit.org/show_bug.cgi?id=28420)

Reviewed by Dave Hyatt.

First rudimentary implementation of HTML5 ruby rendering support.

WebCore:

Following the HTML 5 spec, the box object model for a <ruby> element allows several runs of ruby
bases with their respective ruby texts looks as follows:

1 RenderRuby object, corresponding to the whole <ruby> HTML element
     1+ RenderRubyRun (anonymous)
         0 or 1 RenderRubyText - shuffled to the front in order to re-use existing block layouting
             0-n inline object(s)
         0 or 1 RenderRubyBase - contains the inline objects that make up the ruby base
             1-n inline object(s)

Note: <rp> elements are defined as having 'display:none' and thus normally are not assigned a renderer.

New layout tests will be committed in a follow-up patch under fast/ruby.

Makefiles, etc.
* GNUmakefile.am:
* WebCore.gypi:
* WebCore.pro:
* WebCore.xcodeproj/project.pbxproj:

CSS
* css/html.css: Added <ruby> and <rt>

Existing render files:
* rendering/RenderBlock.cpp: make moveChild a member function moveChildTo
(WebCore::RenderBlock::moveChildTo):
(WebCore::RenderBlock::makeChildrenNonInline):
(WebCore::RenderBlock::removeChild):
* rendering/RenderBlock.h:
* rendering/RenderObject.cpp:
(WebCore::RenderObject::createObject): instantiate ruby renderers based on element name
* rendering/RenderObject.h: add query methods for ruby renderers
(WebCore::RenderObject::isRuby):
(WebCore::RenderObject::isRubyBase):
(WebCore::RenderObject::isRubyRun):
(WebCore::RenderObject::isRubyText):

New ruby renderers:
* rendering/RenderRuby.cpp: Added.
(WebCore::lastRubyRun):
(WebCore::findRubyRunParent):
(WebCore::RenderRubyAsInline::RenderRubyAsInline):
(WebCore::RenderRubyAsInline::~RenderRubyAsInline):
(WebCore::RenderRubyAsInline::isChildAllowed):
(WebCore::RenderRubyAsInline::addChild):
(WebCore::RenderRubyAsInline::removeChild):
(WebCore::RenderRubyAsBlock::RenderRubyAsBlock):
(WebCore::RenderRubyAsBlock::~RenderRubyAsBlock):
(WebCore::RenderRubyAsBlock::isChildAllowed):
(WebCore::RenderRubyAsBlock::addChild):
(WebCore::RenderRubyAsBlock::removeChild):
* rendering/RenderRuby.h: Added.
(WebCore::RenderRubyAsInline::renderName):
(WebCore::RenderRubyAsInline::isRuby):
(WebCore::RenderRubyAsBlock::renderName):
(WebCore::RenderRubyAsBlock::isRuby):
* rendering/RenderRubyBase.cpp: Added.
(WebCore::RenderRubyBase::RenderRubyBase):
(WebCore::RenderRubyBase::~RenderRubyBase):
(WebCore::RenderRubyBase::isChildAllowed):
(WebCore::RenderRubyBase::splitToLeft):
(WebCore::RenderRubyBase::mergeWithRight):
* rendering/RenderRubyBase.h: Added.
(WebCore::RenderRubyBase::renderName):
(WebCore::RenderRubyBase::isRubyBase):
* rendering/RenderRubyRun.cpp: Added.
(WebCore::RenderRubyRun::RenderRubyRun):
(WebCore::RenderRubyRun::~RenderRubyRun):
(WebCore::RenderRubyRun::hasRubyText):
(WebCore::RenderRubyRun::hasRubyBase):
(WebCore::RenderRubyRun::isEmpty):
(WebCore::RenderRubyRun::rubyText):
(WebCore::RenderRubyRun::rubyBase):
(WebCore::RenderRubyRun::rubyBaseSafe):
(WebCore::RenderRubyRun::firstLineBlock):
(WebCore::RenderRubyRun::updateFirstLetter):
(WebCore::RenderRubyRun::isChildAllowed):
(WebCore::RenderRubyRun::addChild):
(WebCore::RenderRubyRun::removeChild):
(WebCore::RenderRubyRun::createRubyBase):
(WebCore::RenderRubyRun::staticCreateRubyRun):
* rendering/RenderRubyRun.h: Added.
(WebCore::RenderRubyRun::renderName):
(WebCore::RenderRubyRun::isRubyRun):
* rendering/RenderRubyText.cpp: Added.
(WebCore::RenderRubyText::RenderRubyText):
(WebCore::RenderRubyText::~RenderRubyText):
(WebCore::RenderRubyText::isChildAllowed):
* rendering/RenderRubyText.h: Added.
(WebCore::RenderRubyText::renderName):
(WebCore::RenderRubyText::isRubyText):

LayoutTests:

Further layout tests will be committed in a follow-up patch under fast/ruby.

* fast/ruby/parse-rp-expected.txt: result changed due to <rt> being rendered as block within <ruby>

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@50397 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago Reviewed by Dan Bernstein.
ap@apple.com [Mon, 2 Nov 2009 01:00:11 +0000 (01:00 +0000)]
    Reviewed by Dan Bernstein.

        https://bugs.webkit.org/show_bug.cgi?id=30992
        Node.isDefaultNamespace doesn't convert empty strings to null

        Test: fast/dom/Node/default-namespace-empty-argument.html

        * dom/Node.cpp: (WebCore::Node::isDefaultNamespace): Per DOM 3 Core, treat empty input
        as null.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@50396 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-11-01 Yael Aharon <yael.aharon@nokia.com>
eric@webkit.org [Sun, 1 Nov 2009 21:38:52 +0000 (21:38 +0000)]
2009-11-01  Yael Aharon  <yael.aharon@nokia.com>

        Reviewed by Darin Adler.

        Don't add '/' to the URL path if the it does not include '/' after the protocol component
        https://bugs.webkit.org/show_bug.cgi?id=30971

        Updating the expected result to reflect the change.

        * fast/loader/url-parse-1-expected.txt:
2009-11-01  Yael Aharon  <yael.aharon@nokia.com>

        Reviewed by Darin Adler.

        Don't add '/' to the URL path if the it does not include '/' after the protocol component
        https://bugs.webkit.org/show_bug.cgi?id=30971

        Match IE8 behaviour, that does not add '/' if there is none after the protocol component.

        * platform/KURL.cpp:
        (WebCore::KURL::parse):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@50395 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoFix layering violations in GraphicsContext3D
oliver@apple.com [Sun, 1 Nov 2009 21:30:16 +0000 (21:30 +0000)]
Fix layering violations in GraphicsContext3D
https://bugs.webkit.org/show_bug.cgi?id=30986

Reviewed by Darin Adler.

Remove uses of HTMLImageElement and HTMLCanvasElement

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@50394 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-11-01 Joanmarie Diggs <joanmarie.diggs@gmail.com>
eric@webkit.org [Sun, 1 Nov 2009 21:22:30 +0000 (21:22 +0000)]
2009-11-01  Joanmarie Diggs  <joanmarie.diggs@gmail.com>

        Reviewed by Jan Alonzo.

        https://bugs.webkit.org/show_bug.cgi?id=30964
        [Gtk] Implemment AtkDocument

        Implements what has been implemented in AT-SPI.

        * accessibility/gtk/AccessibilityObjectWrapperAtk.cpp:
        (GetAtkInterfaceTypeFromWAIType):
        (getInterfaceMaskFromObject):
        (atk_document_interface_init):
        (webkit_accessible_document_get_document_attribute_value):
        (webkit_accessible_document_get_document_attributes):
        (webkit_accessible_document_get_locale):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@50393 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-11-01 Laszlo Gombos <laszlo.1.gombos@nokia.com>
eric@webkit.org [Sun, 1 Nov 2009 20:49:04 +0000 (20:49 +0000)]
2009-11-01  Laszlo Gombos  <laszlo.1.gombos@nokia.com>

        Reviewed by Eric Seidel.

        Turn on warnings for QtWebKit for gcc
        https://bugs.webkit.org/show_bug.cgi?id=30958

        * WebKit.pri: Turn on warnings for the GCC compiler
2009-11-01  Laszlo Gombos  <laszlo.1.gombos@nokia.com>

        Reviewed by Eric Seidel.

        Turn on warnings for QtWebKit for gcc
        https://bugs.webkit.org/show_bug.cgi?id=30958

        No new tests as there is no functional change.

        * platform/image-decoders/qt/RGBA32BufferQt.cpp:
        (WebCore::RGBA32Buffer::RGBA32Buffer): Reorder
        initialization list to fix compiler warnings.

2009-11-01  Laszlo Gombos  <laszlo.1.gombos@nokia.com>

        Reviewed by Eric Seidel.

        Turn on warnings for QtWebKit for gcc
        https://bugs.webkit.org/show_bug.cgi?id=30958

        * Api/qwebpage.cpp:
        (QWebPagePrivate::QWebPagePrivate): Reorder initialization list
        to fix compiler warnings.
        * WebCoreSupport/FrameLoaderClientQt.cpp:
        (WebCore::FrameLoaderClientQt::FrameLoaderClientQt): Ditto.
2009-11-01  Laszlo Gombos  <laszlo.1.gombos@nokia.com>

        Reviewed by Eric Seidel.

        Turn on warnings for QtWebKit for gcc
        https://bugs.webkit.org/show_bug.cgi?id=30958

        * DumpRenderTree/qt/main.cpp:
        (crashHandler): Mark function NO_RETURN

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@50392 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-11-01 Keishi Hattori <casey.hattori@gmail.com>
eric@webkit.org [Sun, 1 Nov 2009 20:41:15 +0000 (20:41 +0000)]
2009-11-01  Keishi Hattori  <casey.hattori@gmail.com>

        Reviewed by Pavel Feldman.

        [Regression] monitorEvent doesn't work

        * inspector/front-end/InjectedScript.js:
        (InjectedScript._ensureCommandLineAPIInstalled):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@50391 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-11-01 Keishi Hattori <casey.hattori@gmail.com>
eric@webkit.org [Sun, 1 Nov 2009 20:33:37 +0000 (20:33 +0000)]
2009-11-01  Keishi Hattori  <casey.hattori@gmail.com>

        Reviewed by Pavel Feldman.

        Fix Web Inspector: Bug with Message Bubble in Syntax Highlighter
        https://bugs.webkit.org/show_bug.cgi?id=30990

        * inspector/front-end/SourceFrame.js:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@50390 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago Reviewed by John Sullivan.
ap@apple.com [Sun, 1 Nov 2009 20:04:11 +0000 (20:04 +0000)]
    Reviewed by John Sullivan.

        https://bugs.webkit.org/show_bug.cgi?id=30982
        createHTMLDocument doesn't escape ampersand and less-than in title

        Test: fast/dom/DOMImplementation/createHTMLDocument-title.html

        * dom/DOMImplementation.cpp: (WebCore::DOMImplementation::createHTMLDocument):
        Set document title after creating the document, avoiding parser intricacies.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@50389 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-11-01 Hironori Bono <hbono@chromium.org>
eric@webkit.org [Sun, 1 Nov 2009 16:18:38 +0000 (16:18 +0000)]
2009-11-01  Hironori Bono  <hbono@chromium.org>

        Reviewed by Darin Adler.

        Bug 30902: Need a layout test for Bug 28284
        This change just adds a simple test that calls TextInputController.firstRectForCharacterRange()
        before and after inserting a Thai character and compare their results.
        https://bugs.webkit.org/show_bug.cgi?id=30902

        * fast/text/international/thai-cursor-position-expected.txt: Added.
        * fast/text/international/thai-cursor-position.html: Added.
        * platform/gtk/Skipped: Skip this test because TextInputController is not implemented.
        * platform/qt/Skipped: ditto.
        * platform/win/Skipped: ditto.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@50388 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-11-01 Keishi Hattori <casey.hattori@gmail.com>
eric@webkit.org [Sun, 1 Nov 2009 16:11:04 +0000 (16:11 +0000)]
2009-11-01  Keishi Hattori  <casey.hattori@gmail.com>

        Reviewed by Timothy Hatcher.

        Web Inspector: Double clicking on a breakpoints should not select text
        https://bugs.webkit.org/show_bug.cgi?id=30950

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

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@50387 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-11-01 Yuta Kitamura <yutak@chromium.org>
eric@webkit.org [Sun, 1 Nov 2009 16:03:27 +0000 (16:03 +0000)]
2009-11-01  Yuta Kitamura  <yutak@chromium.org>

        Reviewed by Darin Adler.

        Fix assertion falure in RenderObjectChildList::updateBeforeAfterContent().

        [Crash (debug)] Combination of list-item and :after causes assertion failure
        https://bugs.webkit.org/show_bug.cgi?id=30944

        * fast/css/list-item-pseudo-nocrash-expected.txt: Added.
        * fast/css/list-item-pseudo-nocrash.html: Added.
2009-11-01  Yuta Kitamura  <yutak@chromium.org>

        Reviewed by Darin Adler.

        Fix assertion falure in RenderObjectChildList::updateBeforeAfterContent().

        [Crash (debug)] Combination of list-item and :after causes assertion failure
        https://bugs.webkit.org/show_bug.cgi?id=30944

        Test: fast/css/list-item-pseudo-nocrash.html

        * rendering/RenderObjectChildList.cpp:
        (WebCore::RenderObjectChildList::updateBeforeAfterContent):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@50386 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-11-01 Philippe Normand <pnormand@igalia.com>
eric@webkit.org [Sun, 1 Nov 2009 15:54:59 +0000 (15:54 +0000)]
2009-11-01  Philippe Normand  <pnormand@igalia.com>

        Reviewed by Eric Seidel.

        https://bugs.webkit.org/show_bug.cgi?id=30586
        [GTK] Failing test media/video-src-empty.html

        * platform/gtk/Skipped: Unskip fixed test.
2009-11-01  Philippe Normand  <pnormand@igalia.com>

        Reviewed by Eric Seidel.

        https://bugs.webkit.org/show_bug.cgi?id=30586
        [GTK] Failing test media/video-src-empty.html

        Correctly set network/ready state depending on GStreamer errors
        received on the bus.

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

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@50385 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-11-01 Jessie Berlin <jberlin@webkit.org>
jberlin@webkit.org [Sun, 1 Nov 2009 15:25:58 +0000 (15:25 +0000)]
2009-11-01  Jessie Berlin  <jberlin@webkit.org>

        Adding myself to the committers list.

        * Scripts/modules/committers.py:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@50384 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoBuild fix
oliver@apple.com [Sat, 31 Oct 2009 23:22:33 +0000 (23:22 +0000)]
Build fix

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@50383 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoRemove obsolete null checks from CanvasRenderingContext3DMac
oliver@apple.com [Sat, 31 Oct 2009 23:10:55 +0000 (23:10 +0000)]
Remove obsolete null checks from CanvasRenderingContext3DMac
https://bugs.webkit.org/show_bug.cgi?id=30983

Reviewed by Darin Adler

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@50382 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoWebGL allows objects to be used with the wrong context
oliver@apple.com [Sat, 31 Oct 2009 22:19:47 +0000 (22:19 +0000)]
WebGL allows objects to be used with the wrong context
https://bugs.webkit.org/show_bug.cgi?id=30981

Reviewed by Jon Honeycutt.

Simply add null checks and a few context guards to ensure we don't
deref null or attempt to use an object from a different context.

Tests: fast/canvas/webgl/incorrect-context-object-behaviour.html
       fast/canvas/webgl/null-object-behaviour.html

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@50381 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoFixed failing layout test: restore a special case I accidentally deleted.
ggaren@apple.com [Sat, 31 Oct 2009 02:11:21 +0000 (02:11 +0000)]
Fixed failing layout test: restore a special case I accidentally deleted.

Patch by Geoffrey Garen <ggaren@apple.com> on 2009-10-30
Reviewed by Oliver Hunt.

* runtime/DatePrototype.cpp:
(JSC::setNewValueFromDateArgs): In the case of applying a change to a date
that is NaN, reset the date to 0 *and* then apply the change; don't just
reset the date to 0.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@50377 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoFix an issue that Adam noticed in DRT.
jhoneycutt@apple.com [Sat, 31 Oct 2009 00:51:27 +0000 (00:51 +0000)]
Fix an issue that Adam noticed in DRT.

Reviewed by Darin Adler.

* DumpRenderTree/win/AccessibilityUIElementWin.cpp:
(AccessibilityUIElement::role):
Get the length of the role text, and create a buffer dynamically.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@50376 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoAllow custom memory allocation control for WebCore's MediaQueryResult
zoltan@webkit.org [Fri, 30 Oct 2009 23:18:55 +0000 (23:18 +0000)]
Allow custom memory allocation control for WebCore's MediaQueryResult
https://bugs.webkit.org/show_bug.cgi?id=30857

Reviewed by Darin Adler.

Inherits MediaQueryResult class from Noncopyable because it is
instantiated by 'new' in WebCore/css/CSSStyleSelector.cpp:5984 and
it is no need to be copyable.

* css/CSSStyleSelector.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@50369 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoAllow custom memory allocation control for WebCore's MediaQueryEvaluator
zoltan@webkit.org [Fri, 30 Oct 2009 23:16:19 +0000 (23:16 +0000)]
Allow custom memory allocation control for WebCore's MediaQueryEvaluator
https://bugs.webkit.org/show_bug.cgi?id=30854

Reviewed by Darin Adler.

Inherits MediaQueryEvaluator class from Noncopyable because it is
instantiated by 'new' in WebCore/css/CSSStyleSelector.cpp:401 and
it is no need to be copyable.

* css/MediaQueryEvaluator.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@50368 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoAllow custom memory allocation control for WebCore's MediaQueryEvaluator
zoltan@webkit.org [Fri, 30 Oct 2009 23:10:56 +0000 (23:10 +0000)]
Allow custom memory allocation control for WebCore's MediaQueryEvaluator
https://bugs.webkit.org/show_bug.cgi?id=30854

Reviewed by Darin Adler.

Inherits MediaQueryEvaluator class from Noncopyable because it is
instantiated by 'new' in WebCore/css/CSSStyleSelector.cpp:401 and
it is no need to be copyable.

* css/MediaQueryEvaluator.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@50367 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoAllow custom memory allocation control for WebCore's CSSRuleSet
zoltan@webkit.org [Fri, 30 Oct 2009 23:08:50 +0000 (23:08 +0000)]
Allow custom memory allocation control for WebCore's CSSRuleSet
https://bugs.webkit.org/show_bug.cgi?id=30852

Reviewed by Darin Adler.

Inherits CSSRuleSet class from Noncopyable because it is
instantiated by 'new' in WebCore/css/CSSStyleSelector.cpp:455 and
it is no need to be copyable.

* css/CSSStyleSelector.cpp:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@50366 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoAllow custom memory allocation control for WebCore's CSSRuleData
zoltan@webkit.org [Fri, 30 Oct 2009 23:06:18 +0000 (23:06 +0000)]
Allow custom memory allocation control for WebCore's CSSRuleData
https://bugs.webkit.org/show_bug.cgi?id=30851

Reviewed by Darin Adler.

Inherits CSSRuleData class from Noncopyable because it is
instantiated by 'new' in WebCore/css/CSSStyleSelector.h:320 and
it is no need to be copyable.

* css/CSSStyleSelector.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@50365 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoAllow custom memory allocation control for WebCore's CSSRuleDataList
zoltan@webkit.org [Fri, 30 Oct 2009 23:02:49 +0000 (23:02 +0000)]
Allow custom memory allocation control for WebCore's CSSRuleDataList
https://bugs.webkit.org/show_bug.cgi?id=30850

Reviewed by Darin Adler.

Inherits CSSRuleDataList class from Noncopyable because it has been
instantiated by 'new' in WebCore/css/CSSStyleSelector.cpp:2715 and
it is no need to be copyable.

* css/CSSStyleSelector.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@50364 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoAllow custom memory allocation control for WebCore's CSSNamespace
zoltan@webkit.org [Fri, 30 Oct 2009 22:59:07 +0000 (22:59 +0000)]
Allow custom memory allocation control for WebCore's CSSNamespace
https://bugs.webkit.org/show_bug.cgi?id=30849

Reviewed by Darin Adler.

Inherits CSSNamespace struct from Noncopyable because it is
instantiated by 'new' in WebCore/css/CSSStyleSheet.cpp:141 and
it is no need to be copyable.

* css/CSSNamespace.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@50363 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-10-30 Eric Seidel <eric@webkit.org>
eric@webkit.org [Fri, 30 Oct 2009 22:41:04 +0000 (22:41 +0000)]
2009-10-30  Eric Seidel  <eric@webkit.org>

        No review, rolling out r50105.
        http://trac.webkit.org/changeset/50105

        This commit was causing:
        https://bugs.webkit.org/show_bug.cgi?id=30869
        We'll re-implement the feature a different way.

        * Scripts/bugzilla-tool:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@50362 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoWindows build fix: update for object-to-pointer change.
ggaren@apple.com [Fri, 30 Oct 2009 22:35:18 +0000 (22:35 +0000)]
Windows build fix: update for object-to-pointer change.

Patch by Geoffrey Garen <ggaren@apple.com> on 2009-10-30
* runtime/DatePrototype.cpp:
(JSC::formatLocaleDate):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@50361 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoRefactor DatabaseTask in preparation for removing threadsafe refcounting from it.
dimich@chromium.org [Fri, 30 Oct 2009 22:26:32 +0000 (22:26 +0000)]
Refactor DatabaseTask in preparation for removing threadsafe refcounting from it.
Move the synchronizer object out of the DatabaseTask so there is no need to keep
the pointer to Databasetask around after passing it to MessageQueue.
Also pass the references to return parameters to the task so it can update them.
https://bugs.webkit.org/show_bug.cgi?id=30941

Reviewed by David Levin.

No new tests, since this is just moving the code around, no change in functionality.

* storage/Database.cpp:
(WebCore::Database::Database):
(WebCore::Database::openAndVerifyVersion): Use new DatabaseTaskSynchronizer to wait for task completion.
(WebCore::Database::markAsDeletedAndClose): Ditto.
(WebCore::Database::tableNames): Ditto.
(WebCore::Database::stop): Use the boolean flag rather then 'killed' flag built into MessageQueue.
(WebCore::Database::scheduleTransaction): Transaction queue is a Deque now, change the way to fetch the transaction.
* storage/Database.h: Change the SQLTransaction queue to be a Deque rather then a MessageQueue.
* storage/DatabaseTask.cpp:
(WebCore::DatabaseTaskSynchronizer::DatabaseTaskSynchronizer):
(WebCore::DatabaseTaskSynchronizer::waitForTaskCompletion):
(WebCore::DatabaseTaskSynchronizer::taskCompleted):
(WebCore::DatabaseTask::DatabaseTask): Ctor takes DatabaseTaskSynchronizer which can be 0.
(WebCore::DatabaseTask::performTask): Signal completion. m_synchronizer should still be around since main thread is waiting on it.
(WebCore::DatabaseOpenTask::DatabaseOpenTask): Pass synchronizer and return parameters via constructor.
(WebCore::DatabaseCloseTask::DatabaseCloseTask): Ditto.
(WebCore::DatabaseTransactionTask::DatabaseTransactionTask): Ditto.
(WebCore::DatabaseTableNamesTask::DatabaseTableNamesTask): Ditto.
* storage/DatabaseTask.h:
(WebCore::DatabaseOpenTask::create):
(WebCore::DatabaseCloseTask::create):
(WebCore::DatabaseTransactionTask::create):
(WebCore::DatabaseTableNamesTask::create):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@50360 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agohttps://bugs.webkit.org/show_bug.cgi?id=30942
ggaren@apple.com [Fri, 30 Oct 2009 22:23:31 +0000 (22:23 +0000)]
https://bugs.webkit.org/show_bug.cgi?id=30942
Use pointers instead of copies to pass GregorianDateTime objects around.

Patch by Geoffrey Garen <ggaren@apple.com> on 2009-10-29
Reviewed by Darin Adler.

SunSpider reports a shocking 4.5% speedup on date-format-xparb, and 1.3%
speedup on date-format-tofte.

* runtime/DateInstance.cpp:
(JSC::DateInstance::gregorianDateTime):
* runtime/DateInstance.h:
* runtime/DatePrototype.cpp:
(JSC::formatLocaleDate):
(JSC::dateProtoFuncToString):
(JSC::dateProtoFuncToUTCString):
(JSC::dateProtoFuncToISOString):
(JSC::dateProtoFuncToDateString):
(JSC::dateProtoFuncToTimeString):
(JSC::dateProtoFuncGetFullYear):
(JSC::dateProtoFuncGetUTCFullYear):
(JSC::dateProtoFuncToGMTString):
(JSC::dateProtoFuncGetMonth):
(JSC::dateProtoFuncGetUTCMonth):
(JSC::dateProtoFuncGetDate):
(JSC::dateProtoFuncGetUTCDate):
(JSC::dateProtoFuncGetDay):
(JSC::dateProtoFuncGetUTCDay):
(JSC::dateProtoFuncGetHours):
(JSC::dateProtoFuncGetUTCHours):
(JSC::dateProtoFuncGetMinutes):
(JSC::dateProtoFuncGetUTCMinutes):
(JSC::dateProtoFuncGetSeconds):
(JSC::dateProtoFuncGetUTCSeconds):
(JSC::dateProtoFuncGetTimezoneOffset):
(JSC::setNewValueFromTimeArgs):
(JSC::setNewValueFromDateArgs):
(JSC::dateProtoFuncSetYear):
(JSC::dateProtoFuncGetYear): Renamed getGregorianDateTime to gregorianDateTime,
since it no longer has an out parameter. Uses 0 to indicate invalid dates.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@50359 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoREGRESSION: In Mail, Undo does not restore some characters I have deleted at the...
adele@apple.com [Fri, 30 Oct 2009 22:09:07 +0000 (22:09 +0000)]
REGRESSION: In Mail, Undo does not restore some characters I have deleted at the end of a line
https://bugs.webkit.org/show_bug.cgi?id=30955
<rdar://problem/7067033>

Patch by Enrica Casucci <enrica@apple.com> on 2009-10-30
Reviewed by Darin Adler.

WebCore:

When the command is deleteWordBackward or deleteWordForward
we should not add to the open typing command, but
create a new one.

Test: editing/undo/undo-deleteWord.html

* editing/TypingCommand.cpp:
(WebCore::TypingCommand::deleteKeyPressed): Always start a new command if the granularity is
not CharacterGranularity.
(WebCore::TypingCommand::forwardDeleteKeyPressed): Always start a new command if the granularity is
not CharacterGranularity.

LayoutTests:

* editing/undo/undo-deleteWord-expected.txt: Added.
* editing/undo/undo-deleteWord.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@50358 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-10-30 Eric Carlson <eric.carlson@apple.com>
eric.carlson@apple.com [Fri, 30 Oct 2009 22:03:37 +0000 (22:03 +0000)]
2009-10-30  Eric Carlson  <eric.carlson@apple.com>

        Reviewed by Darin Adler.

        Make MediaPlayer constructor private
        https://bugs.webkit.org/show_bug.cgi?id=30965

        * html/HTMLMediaElement.cpp:
        (WebCore::HTMLMediaElement::loadResource):
        (WebCore::HTMLMediaElement::finishParsingChildren): Use MediaPlayer::create.

        * platform/graphics/MediaPlayer.h:
        (WebCore::MediaPlayer::create): New.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@50357 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoTest for MSAA: Accessibility of headings is not correct
jhoneycutt@apple.com [Fri, 30 Oct 2009 21:38:27 +0000 (21:38 +0000)]
Test for MSAA: Accessibility of headings is not correct

https://bugs.webkit.org/show_bug.cgi?id=30937

Reviewed by Adam Roben.

WebKitTools:

* DumpRenderTree/win/AccessibilityUIElementWin.cpp:
(AccessibilityUIElement::role):
Allow the role returned to be a BSTR.
(AccessibilityUIElement::description):
Fix a copy/paste error.

LayoutTests:

* platform/win/accessibility/heading-elements-expected.txt: Added.
* platform/win/accessibility/heading-elements.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@50356 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoTest for MSAA: Accessibility of links is wrong
jhoneycutt@apple.com [Fri, 30 Oct 2009 21:38:15 +0000 (21:38 +0000)]
Test for MSAA: Accessibility of links is wrong

https://bugs.webkit.org/show_bug.cgi?id=30928

Reviewed by Darin Adler.

WebKitTools:

* DumpRenderTree/AccessibilityUIElement.cpp:
(getAccessibilityValueCallback):
Return the accessibility value.
(AccessibilityUIElement::getJSClass):
Added "accessibilityValue" value.

* DumpRenderTree/AccessibilityUIElement.h:

* DumpRenderTree/gtk/AccessibilityUIElementGtk.cpp:
(AccessibilityUIElement::accessibilityValue):
Stubbed.

* DumpRenderTree/mac/AccessibilityUIElementMac.mm:
(AccessibilityUIElement::accessibilityValue):
Stubbed.

* DumpRenderTree/win/AccessibilityUIElementWin.cpp:
(AccessibilityUIElement::accessibilityValue):
Get the object's value, and return it as a JS string.

LayoutTests:

* platform/win/accessibility/linked-elements-expected.txt: Added.
* platform/win/accessibility/linked-elements.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@50355 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoMSAA: Accessibility of headings is not correct
jhoneycutt@apple.com [Fri, 30 Oct 2009 21:38:02 +0000 (21:38 +0000)]
MSAA: Accessibility of headings is not correct

https://bugs.webkit.org/show_bug.cgi?id=30937

Reviewed by Alice Liu.

WebCore:

* accessibility/AccessibilityObject.h:
(WebCore::AccessibilityObject::stringRoleForMSAA):
(WebCore::AccessibilityObject::descriptionForMSAA):

* accessibility/AccessibilityRenderObject.cpp:
(WebCore::shouldReturnTagNameAsRoleForMSAA):
If the element's tag name is one of h1, h2, h3, h4, h5, h6, return
true.
(WebCore::AccessibilityRenderObject::stringRoleForMSAA):
If the element should return its tag name as the role, return the tag
name.
(WebCore::AccessibilityRenderObject::positionalDescriptionForMSAA):
If the object is a heading, return the string "L" followed by the
heading level.
(WebCore::AccessibilityRenderObject::descriptionForMSAA):
If the object has a positional description, return it. Otherwise, get
the accessibility description, and prefix it with "Description" so that
MSAA clients know that it's not a positional description.

* accessibility/AccessibilityRenderObject.h:

WebKit/win:

* AccessibleBase.cpp:
(AccessibleBase::get_accDescription):
Call the object's descriptionForMSAA(). Moved the comment to the
WebCore file.
(AccessibleBase::get_accRole):
If the object has a string role, return that. Otherwise, return the
integer role.

* AccessibleBase.h:
Removed description(), as this was moved to WebCore.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@50354 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoMSAA: Accessibility of links is wrong
jhoneycutt@apple.com [Fri, 30 Oct 2009 21:37:44 +0000 (21:37 +0000)]
MSAA: Accessibility of links is wrong

https://bugs.webkit.org/show_bug.cgi?id=30928

Reviewed by Darin Adler.

WebCore:

* accessibility/AccessibilityImageMapLink.cpp:
(WebCore::AccessibilityImageMapLink::stringValueForMSAA):
Return the URL.
(WebCore::AccessibilityImageMapLink::nameForMSAA):
Return the alt text.

* accessibility/AccessibilityImageMapLink.h:
(WebCore::AccessibilityImageMapLink::isLinked):
Return true.

* accessibility/AccessibilityObject.h:
(WebCore::AccessibilityObject::isLinked):
(WebCore::AccessibilityObject::stringValueForMSAA):
(WebCore::AccessibilityObject::nameForMSAA):

* accessibility/AccessibilityRenderObject.cpp:
(WebCore::isLinkable):
Return true if the element is considered "linkable" with respect to
accessibility.
(WebCore::AccessibilityRenderObject::stringValueForMSAA):
If the element is linkable, check whether it has a parent anchor
element. If so, return the anchor element's href.
(WebCore::AccessibilityRenderObject::isLinked):
Return true if the element is linkable and if it's parent anchor tag's
href is non-empty.
(WebCore::AccessibilityRenderObject::nameForMSAA):
For text nodes, return the text.

* accessibility/AccessibilityRenderObject.h:

WebKit/win:

* AccessibleBase.cpp:
(AccessibleBase::get_accState):
Check whether the element is linked, instead of whether the node is an
anchor, so the child "linkable" elements of anchor elements will also
be "linked".
(AccessibleBase::name):
Return the name for MSAA.
(AccessibleBase::value):
Return the string value for MSAA.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@50353 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoTouch WebKitPrefix to force rebuild of WebKit.
bweinstein@apple.com [Fri, 30 Oct 2009 21:20:48 +0000 (21:20 +0000)]
Touch WebKitPrefix to force rebuild of WebKit.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@50352 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoNotify the chrome when the focused node has changed.
levin@chromium.org [Fri, 30 Oct 2009 20:59:46 +0000 (20:59 +0000)]
Notify the chrome when the focused node has changed.
https://bugs.webkit.org/show_bug.cgi?id=30832

Patch by Evan Stade <estade@chromium.org> on 2009-10-30
Reviewed by David Levin.

WebCore:

This is similar to AX code that is already in place, except that this also informs the
chrome when there stops being a focused node. This is needed for a browser to show the
anchor for links that have keyboard focus.

* dom/Document.cpp:
(WebCore::Document::setFocusedNode):
* loader/EmptyClients.h:
(WebCore::EmptyChromeClient::focusedNodeChanged):
* page/Chrome.cpp:
(WebCore::Chrome::focusedNodeChanged):
* page/Chrome.h:
* page/ChromeClient.h:

WebKit/gtk:

Added stub implementation for new ChromeClient function.

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

WebKit/haiku:

Added stub implementation for new ChromeClient function.

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

WebKit/mac:

Added stub implementation for new ChromeClient function.

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

WebKit/qt:

Added stub implementation for new ChromeClient function.

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

WebKit/win:

Added stub implementation for new ChromeClient function.

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

WebKit/wx:

Added stub implementation for new ChromeClient function.

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

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@50351 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-10-30 Ben Murdoch <benm@google.com>
eric@webkit.org [Fri, 30 Oct 2009 20:20:32 +0000 (20:20 +0000)]
2009-10-30  Ben Murdoch  <benm@google.com>

        Reviewed by David Kilzer.

        openDatabase() with empty version sets db version up incorrectly
        https://bugs.webkit.org/show_bug.cgi?id=28417

        * storage/open-database-set-empty-version-expected.txt: Added.
        * storage/open-database-set-empty-version.html: Added.
2009-10-30  Ben Murdoch  <benm@google.com>

        Reviewed by David Kilzer.

        openDatabase() with empty version sets db version up incorrectly
        https://bugs.webkit.org/show_bug.cgi?id=28417

        Test: storage/open-database-set-empty-version.html

        * storage/Database.cpp:
        (WebCore::Database::performOpenAndVerify): Raise an exception if the current database version does not match the expected version when the current version is the empty string.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@50350 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoFix typo in command name used by wx build system.
kevino@webkit.org [Fri, 30 Oct 2009 19:52:22 +0000 (19:52 +0000)]
Fix typo in command name used by wx build system.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@50349 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-10-30 John Gregg <johnnyg@google.com>
eric@webkit.org [Fri, 30 Oct 2009 19:37:12 +0000 (19:37 +0000)]
2009-10-30  John Gregg  <johnnyg@google.com>

        Reviewed by David Levin.

        Need to turn off notifications properly at runtime
        https://bugs.webkit.org/show_bug.cgi?id=30409

        Moving the notificationsEnabled bit from NotificationCenter
        to the new V8 RuntimeEnabledFeatures object.

        Just moving a bit around, so no new tests.

        * bindings/v8/RuntimeEnabledFeatures.cpp:
        * bindings/v8/RuntimeEnabledFeatures.h:
        (WebCore::RuntimeEnabledFeatures::setNotificationsEnabled):
        (WebCore::RuntimeEnabledFeatures::notificationsEnabled):
        * bindings/v8/custom/V8DOMWindowCustom.cpp:
        (WebCore::ACCESSOR_RUNTIME_ENABLER):
        * bindings/v8/custom/V8WorkerContextCustom.cpp:
        (WebCore::ACCESSOR_RUNTIME_ENABLER):
        * notifications/NotificationCenter.cpp:
        * notifications/NotificationCenter.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@50348 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago[V8] More cleanup after r49949: remove ListenerGuard.
dimich@chromium.org [Fri, 30 Oct 2009 18:31:58 +0000 (18:31 +0000)]
[V8] More cleanup after r49949: remove ListenerGuard.
ListenerGuard is no longer needed since EventListeners do not depend on frame or v8 context.
https://bugs.webkit.org/show_bug.cgi?id=30943

Reviewed by Dimitri Glazkov.

Covered by fast/events/add-event-without-document.html which will now pass in Chromium.

* bindings/v8/V8AbstractEventListener.cpp:
(WebCore::V8AbstractEventListener::V8AbstractEventListener):
(WebCore::V8AbstractEventListener::handleEvent):
* bindings/v8/V8AbstractEventListener.h:
* bindings/v8/V8DOMWrapper.cpp:
(WebCore::V8DOMWrapper::getEventListener):
* bindings/v8/V8EventListenerList.h:
(WebCore::V8EventListenerList::findOrCreateWrapper):
* bindings/v8/V8LazyEventListener.cpp:
(WebCore::V8LazyEventListener::V8LazyEventListener):
* bindings/v8/V8Proxy.cpp:
(WebCore::V8Proxy::V8Proxy):
(WebCore::V8Proxy::disconnectFrame):
(WebCore::V8Proxy::clearForNavigation):
* bindings/v8/V8Proxy.h:
* bindings/v8/V8WorkerContextEventListener.cpp:
(WebCore::V8WorkerContextEventListener::V8WorkerContextEventListener):
(WebCore::V8WorkerContextEventListener::handleEvent):
(WebCore::V8WorkerContextEventListener::reportError):
* bindings/v8/V8WorkerContextEventListener.h:
(WebCore::V8WorkerContextEventListener::create):
* bindings/v8/WorkerContextExecutionProxy.cpp:
(WebCore::WorkerContextExecutionProxy::WorkerContextExecutionProxy):
(WebCore::WorkerContextExecutionProxy::dispose):
(WebCore::WorkerContextExecutionProxy::findOrCreateEventListener):
* bindings/v8/WorkerContextExecutionProxy.h:
* bindings/v8/custom/V8CustomEventListener.cpp:
(WebCore::V8EventListener::V8EventListener):
* bindings/v8/custom/V8CustomEventListener.h:
(WebCore::V8EventListener::create):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@50347 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-10-30 Adam Barth <abarth@webkit.org>
eric@webkit.org [Fri, 30 Oct 2009 18:19:33 +0000 (18:19 +0000)]
2009-10-30  Adam Barth  <abarth@webkit.org>

        Reviewed by Darin Adler.

        Patch v1 is a dumb default name for patches
        https://bugs.webkit.org/show_bug.cgi?id=30952

        Let's use "Patch" instead.

        * Scripts/bugzilla-tool:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@50346 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-10-30 Shinichiro Hamaji <hamaji@chromium.org>
hamaji@chromium.org [Fri, 30 Oct 2009 17:57:17 +0000 (17:57 +0000)]
2009-10-30  Shinichiro Hamaji  <hamaji@chromium.org>

        Reviewed by Darin Adler.

        counterValueForElementById should return space-separated string for multiple counters
        https://bugs.webkit.org/show_bug.cgi?id=30939

        Test: fast/css/counters/counterValueForElementById.html

        * rendering/RenderTreeAsText.cpp:
        (WebCore::writeCounterValuesFromChildren):
        (WebCore::counterValueForElement):
2009-10-30  Shinichiro Hamaji  <hamaji@chromium.org>

        Reviewed by Darin Adler.

        counterValueForElementById should return space-separated string for multiple counters
        https://bugs.webkit.org/show_bug.cgi?id=30939

        * fast/css/counters/counterValueForElementById-expected.txt: Added.
        * fast/css/counters/counterValueForElementById.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@50345 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-10-30 Jocelyn Turcotte <jocelyn.turcotte@nokia.com>
eric@webkit.org [Fri, 30 Oct 2009 17:50:45 +0000 (17:50 +0000)]
2009-10-30  Jocelyn Turcotte  <jocelyn.turcotte@nokia.com>

        Reviewed by Tor Arne Vestbø.

        [Qt] Remove the QWebInspector::windowTitleChanged signal,
        QEvent::WindowTitleChange can be used to achieve the same.
        https://bugs.webkit.org/show_bug.cgi?id=30927

        * Api/qwebinspector.cpp:
        * Api/qwebinspector.h:
        * WebCoreSupport/InspectorClientQt.cpp:
        (WebCore::InspectorClientQt::updateWindowTitle):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@50344 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-10-30 Sebastian Dröge <sebastian.droege@collabora.co.uk>
eric@webkit.org [Fri, 30 Oct 2009 17:43:29 +0000 (17:43 +0000)]
2009-10-30  Sebastian Dröge  <sebastian.droege@collabora.co.uk>

        Reviewed by Gustavo Noronha.

        * platform/graphics/gtk/MediaPlayerPrivateGStreamer.cpp:
        (WebCore::MediaPlayerPrivate::paint):
        Add some comments to explain what is happening here.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@50343 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-10-30 Alexander Pavlov <apavlov@chromium.org>
eric@webkit.org [Fri, 30 Oct 2009 17:35:12 +0000 (17:35 +0000)]
2009-10-30  Alexander Pavlov  <apavlov@chromium.org>

        Reviewed by Timothy Hatcher.

        Fix Web Inspector crash on the errors/warnings counter click

        RenderObject::createVisiblePosition(const Position& position)
        understands "null Positions", so we can construct such a Position manually.

        https://bugs.webkit.org/show_bug.cgi?id=30499

        * rendering/RenderBox.cpp:
        (WebCore::RenderBox::positionForPoint):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@50342 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-10-30 Alexander Pavlov <apavlov@chromium.org>
eric@webkit.org [Fri, 30 Oct 2009 17:27:52 +0000 (17:27 +0000)]
2009-10-30  Alexander Pavlov  <apavlov@chromium.org>

        Reviewed by Timothy Hatcher.

        Fix Web Inspector crash on the errors/warnings counter click

        RenderObject::createVisiblePosition(const Position& position)
        understands "null Positions", so we can construct such a Position manually.

        https://bugs.webkit.org/show_bug.cgi?id=30499

        * rendering/RenderBox.cpp:
        (WebCore::RenderBox::positionForPoint):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@50341 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-10-30 Shinichiro Hamaji <hamaji@chromium.org>
hamaji@chromium.org [Fri, 30 Oct 2009 17:26:23 +0000 (17:26 +0000)]
2009-10-30  Shinichiro Hamaji  <hamaji@chromium.org>

        Reviewed by Mark Rowe.

        fast/css/counters/t1204* are not asynchronous anymore
        https://bugs.webkit.org/show_bug.cgi?id=30940

        * fast/css/counters/t1204-increment-00-c-o.html:
        * fast/css/counters/t1204-increment-01-c-o.html:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@50340 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-10-30 Joanmarie Diggs <joanmarie.diggs@gmail.com>
eric@webkit.org [Fri, 30 Oct 2009 17:19:40 +0000 (17:19 +0000)]
2009-10-30  Joanmarie Diggs  <joanmarie.diggs@gmail.com>

        Reviewed by Xan Lopez.

        https://bugs.webkit.org/show_bug.cgi?id=25534
        [GTK] Objects of ROLE_TABLE should implement the accessible table interface

        Second part of the implementation of AtkTable.

        * accessibility/gtk/AccessibilityObjectWrapperAtk.cpp:
        (cellAtIndex):
        (webkit_accessible_table_get_column_at_index):
        (webkit_accessible_table_get_row_at_index):
        (webkit_accessible_table_get_caption):
        (atk_table_interface_init):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@50339 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoTouch WebCorePrefix to try and force a clean build.
bweinstein@apple.com [Fri, 30 Oct 2009 16:40:30 +0000 (16:40 +0000)]
Touch WebCorePrefix to try and force a clean build.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@50338 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-10-30 Pavel Feldman <pfeldman@chromium.org>
pfeldman@chromium.org [Fri, 30 Oct 2009 16:18:00 +0000 (16:18 +0000)]
2009-10-30  Pavel Feldman  <pfeldman@chromium.org>

        Reviewed by Timothy Hatcher.

        Web Inspector: Add TimelinePanel into the panels enum.

        https://bugs.webkit.org/show_bug.cgi?id=30915

        * inspector/InspectorFrontend.cpp:
        (WebCore::InspectorFrontend::showPanel):
        * inspector/front-end/inspector.js:
        (WebInspector.showTimelinePanel):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@50337 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoUnreviewed trivial buildfix.
ossy@webkit.org [Fri, 30 Oct 2009 15:48:50 +0000 (15:48 +0000)]
Unreviewed trivial buildfix.

Patch by Andras Becsi <becsi.andras@stud.u-szeged.hu> on 2009-10-30
[Qt] Buildfix for r50333.

* DumpRenderTree/qt/DumpRenderTree.pro:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@50336 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoUnreviewed layout test fixes.
kenneth@webkit.org [Fri, 30 Oct 2009 15:00:55 +0000 (15:00 +0000)]
Unreviewed layout test fixes.

Patch by Kenneth Rohde Christiansen <kenneth@webkit.org> on 2009-10-30
The existing excepted results were wrong, due to the visited links not
being reset between tests.

This was fixed in r50329, so updating the results.

* platform/qt/fast/block/margin-collapse/block-inside-inline/010-expected.txt:
* platform/qt/fast/block/margin-collapse/block-inside-inline/011-expected.txt:
* platform/qt/fast/block/margin-collapse/block-inside-inline/012-expected.txt:
* platform/qt/fast/block/margin-collapse/block-inside-inline/015-expected.txt:
* platform/qt/fast/block/margin-collapse/block-inside-inline/016-expected.txt:
* platform/qt/fast/block/margin-collapse/block-inside-inline/020-expected.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@50335 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago[Qt] Add expected files for new svg tests we pass.
ossy@webkit.org [Fri, 30 Oct 2009 14:13:48 +0000 (14:13 +0000)]
[Qt] Add expected files for new svg tests we pass.
Compared to the Mac expected png files.

Patch by Andras Becsi <becsi.andras@stud.u-szeged.hu> on 2009-10-30
Reviewed by Holger Hans Peter Freyther.

* platform/qt/svg/css/arrow-with-shadow-expected.txt: Added.
* platform/qt/svg/css/circle-in-mask-with-shadow-expected.txt: Added.
* platform/qt/svg/css/clippath-with-shadow-expected.txt: Added.
* platform/qt/svg/css/mask-with-shadow-expected.txt: Added.
* platform/qt/svg/css/path-with-shadow-expected.txt: Added.
* platform/qt/svg/css/stars-with-shadow-expected.txt: Added.
* platform/qt/svg/filters/shadow-on-filter-expected.txt: Added.
* platform/qt/svg/filters/shadow-on-rect-with-filter-expected.txt: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@50334 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago[Qt] Remove qt/WorkQueue.cpp|h in favor of platform independent WorkQueue
tonikitoo@webkit.org [Fri, 30 Oct 2009 14:02:18 +0000 (14:02 +0000)]
[Qt] Remove qt/WorkQueue.cpp|h in favor of platform independent WorkQueue
https://bugs.webkit.org/show_bug.cgi?id=30953

Patch by Antonio Gomes <tonikitoo@webkit.org> on 2009-10-30
Reviewed by Holger Freyther.

DumpRenderTree/WorkQueue and DumpRenderTree/qt/WorkQueue share mostly the
same implementation. Some Q_ASSERTs differ from ASSERTs basically. Patch
makes qt DRT to share this implementation (as gtk and mac ports do).

* DumpRenderTree/qt/DumpRenderTree.pro:
* DumpRenderTree/qt/WorkQueue.cpp: Removed.
* DumpRenderTree/qt/WorkQueue.h: Removed.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@50333 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoUnreviewed potential buildbot fix.
kenneth@webkit.org [Fri, 30 Oct 2009 13:51:56 +0000 (13:51 +0000)]
Unreviewed potential buildbot fix.

Patch by Kenneth Rohde Christiansen <kenneth@webkit.org> on 2009-10-30
Second try: Reset page history before running each test.

Apparently the QWebHistory::clear() keeps the current page
in history which is not what we want, so we not additionally
sets the history capacity to 0 (forces removing everything)
and then sets it back to its original value.

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

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@50332 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agochangelog date
yurys@chromium.org [Fri, 30 Oct 2009 13:38:42 +0000 (13:38 +0000)]
changelog date

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@50331 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoIf the owner widget of the page has a palette set, we
kenneth@webkit.org [Fri, 30 Oct 2009 13:29:02 +0000 (13:29 +0000)]
If the owner widget of the page has a palette set, we
should use that one. This was only working when the
owner was a QWebView. This patch fixes that.

Patch by Kenneth Rohde Christiansen <kenneth@webkit.org> on 2009-10-30
Reviewed by Holger Hans Peter Freyther.

* platform/qt/RenderThemeQt.cpp:
(WebCore::RenderThemeQt::applyTheme):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@50330 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoUnreviewed potential buildbot fix.
kenneth@webkit.org [Fri, 30 Oct 2009 13:07:34 +0000 (13:07 +0000)]
Unreviewed potential buildbot fix.

Patch by Kenneth Rohde Christiansen <kenneth@webkit.org> on 2009-10-30
Reset page history before running each test.

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

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@50329 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-10-30 Pavel Feldman <pfeldman@chromium.org>
pfeldman@chromium.org [Fri, 30 Oct 2009 10:50:24 +0000 (10:50 +0000)]
2009-10-30  Pavel Feldman  <pfeldman@chromium.org>

        Not reviewed: Wire CookieJarChromium to the cookies
        backend. This is a final step of a 3-steps raw cookies
        access implementation in Chromium.

        * platform/network/chromium/CookieJarChromium.cpp:
        (WebCore::getRawCookies):
        (WebCore::deleteCookie):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@50328 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-10-30 Yury Semikhatsky <yurys@chromium.org>
yurys@chromium.org [Fri, 30 Oct 2009 10:32:19 +0000 (10:32 +0000)]
2009-10-30  Yury Semikhatsky  <yurys@chromium.org>

        Reviewed by Pavel Feldman.

        Fix Chromium crash in console.log in "deeply recursive" function

        Check that result of 'frameSourceName' is not null handle
        before casting it to String.

        Allow V8Proxy::sourceName/sourceLineNumber() to report
        that they have failed due to JavaScript stack overflow.

        https://bugs.webkit.org/show_bug.cgi?id=30904

        * bindings/scripts/CodeGeneratorV8.pm:
        * bindings/v8/ScriptCallStack.cpp:
        (WebCore::ScriptCallStack::create):
        (WebCore::ScriptCallStack::ScriptCallStack):
        * bindings/v8/ScriptCallStack.h:
        * bindings/v8/V8Proxy.cpp:
        (WebCore::V8Proxy::sourceLineNumber):
        (WebCore::V8Proxy::sourceName):
        * bindings/v8/V8Proxy.h:
        * bindings/v8/custom/V8WorkerContextCustom.cpp:
        (WebCore::CALLBACK_FUNC_DECL):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@50327 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoAllow custom memory allocation control for WebCore's MediaQueryExp
zoltan@webkit.org [Fri, 30 Oct 2009 09:27:28 +0000 (09:27 +0000)]
Allow custom memory allocation control for WebCore's MediaQueryExp
https://bugs.webkit.org/show_bug.cgi?id=30855

Reviewed by Darin Adler.

Inherits MediaQueryExp class from FastAllocBase because it is
instantiated by 'new' in WebCore/css/CSSParser.cpp:4874.

* css/MediaQueryExp.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@50326 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-10-30 Adam Barth <abarth@webkit.org>
abarth@webkit.org [Fri, 30 Oct 2009 08:55:56 +0000 (08:55 +0000)]
2009-10-30  Adam Barth  <abarth@webkit.org>

        Reviewed by Mark Rowe.

        Teach git to ignore some files
        https://bugs.webkit.org/show_bug.cgi?id=30951

        Ignore WebKitBuild because we never want to version that directory.
        Also, ignore the xcode project files so git clean doesn't blow away
        your project settings.  Finally, ignore the compiled python files in
        WebKitTools/Script modules because they clutter up git status.

        * .gitignore: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@50325 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoAllow custom memory allocation control for JavaScriptCore's ListHashSet
zoltan@webkit.org [Fri, 30 Oct 2009 08:53:27 +0000 (08:53 +0000)]
Allow custom memory allocation control for JavaScriptCore's ListHashSet
https://bugs.webkit.org/show_bug.cgi?id=30853

Reviewed by Darin Adler.

Inherits ListHashSet class from FastAllocBase because it is
instantiated by 'new' in WebCore/rendering/RenderBlock.cpp:1813.

* wtf/ListHashSet.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@50324 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoRegression: crash enumerating properties of an object with getters or setters
oliver@apple.com [Fri, 30 Oct 2009 08:13:38 +0000 (08:13 +0000)]
Regression: crash enumerating properties of an object with getters or setters
https://bugs.webkit.org/show_bug.cgi?id=30948

Reviewed by Gavin Barraclough

Add a guard to prevent us trying to cache property enumeration on
objects with getters or setters.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@50323 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoCommit patch 1 of ruby implementation, reviewed by Eric Seidel. (see issue 28420)
rolandsteiner@chromium.org [Fri, 30 Oct 2009 07:55:17 +0000 (07:55 +0000)]
Commit patch 1 of ruby implementation, reviewed by Eric Seidel. (see issue 28420)

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@50322 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoRemoved test for an impossible condition (a glyph in a right-to-left run not having
mitz@apple.com [Fri, 30 Oct 2009 07:24:09 +0000 (07:24 +0000)]
Removed test for an impossible condition (a glyph in a right-to-left run not having
the RTL flag 0x800)

Reviewed by Adele Peterson.

* platform/graphics/mac/ComplexTextControllerATSUI.cpp:
(WebCore::ComplexTextController::ComplexTextRun::overrideLayoutOperation):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@50321 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoREGRESSION (r50218-r50262): E*TRADE accounts page is missing content
oliver@apple.com [Fri, 30 Oct 2009 07:13:38 +0000 (07:13 +0000)]
REGRESSION (r50218-r50262): E*TRADE accounts page is missing content
https://bugs.webkit.org/show_bug.cgi?id=30947
<rdar://problem/7348833>

Reviewed by Maciej Stachowiak

The logic for flagging that a structure has non-enumerable properties
was in addPropertyWithoutTransition, rather than in the core Structure::put
method.  Despite this I was unable to produce a testcase that caused
the failure that etrade was experiencing, but the new assertion in
getEnumerablePropertyNames triggers on numerous layout tests without
the fix, so in effecti all for..in enumeration in any test ends up
doing the required consistency check.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@50320 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-10-29 Adam Barth <abarth@webkit.org>
abarth@webkit.org [Fri, 30 Oct 2009 06:51:51 +0000 (06:51 +0000)]
2009-10-29  Adam Barth  <abarth@webkit.org>

        Reviewed by Dimitri Glazkov.

        [V8] Out-of-memory crash in isolated worlds
        https://bugs.webkit.org/show_bug.cgi?id=30906

        We need to handle the fact that creating a V8:Context might fail.  I
        don't know how to test this change because creating a context usually
        only fails when V8 decides it's using too much memory.

        * bindings/v8/V8IsolatedWorld.cpp:
        (WebCore::V8IsolatedWorld::V8IsolatedWorld):
        * bindings/v8/V8Proxy.cpp:
        (WebCore::V8Proxy::evaluateInIsolatedWorld):
        (WebCore::V8Proxy::evaluateInNewContext):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@50319 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoFix “Undefined subroutine” errors in svn-*apply by moving the removeEOL subroutine
mitz@apple.com [Fri, 30 Oct 2009 06:12:02 +0000 (06:12 +0000)]
Fix “Undefined subroutine” errors in svn-*apply by moving the removeEOL subroutine
from the two scripts that define it but don’t use it to the script that uses it but doesn’t
define it.

Reviewed by Mark Rowe.

* Scripts/VCSUtils.pm:
* Scripts/svn-apply:
* Scripts/svn-unapply:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@50318 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoValidate the stopSpeaking: selector so that is not always enabled when a WebView...
mrowe@apple.com [Fri, 30 Oct 2009 05:57:29 +0000 (05:57 +0000)]
Validate the stopSpeaking: selector so that is not always enabled when a WebView is first responder.

Reviewed by Oliver Hunt.

* WebView/WebHTMLView.mm:
(-[WebHTMLView validateUserInterfaceItemWithoutDelegate:]):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@50317 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago<http://webkit.org/b/30938> REGRESSION(r50233): Windows nightlies crash on launch...
mrowe@apple.com [Fri, 30 Oct 2009 04:07:27 +0000 (04:07 +0000)]
<webkit.org/b/30938> REGRESSION(r50233): Windows nightlies crash on launch due to changes
to IWebFramePrivate vtable ordering

Rubber-stamped by Dan Bernstein.

* Interfaces/IWebFramePrivate.idl: Move the newly-added method to the end of the interface so that
it doesn't affect the ordering of the vtable.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@50316 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-10-29 Erik Arvidsson <arv@chromium.org>
eric@webkit.org [Fri, 30 Oct 2009 02:39:01 +0000 (02:39 +0000)]
2009-10-29  Erik Arvidsson  <arv@chromium.org>

        Reviewed by Eric Seidel.

        Make links mouse focusable only on GTK and QT.

        Links are now always mouse focusable on GTK and QT. On other platforms
        the link needs a tabIndex or it needs to be contentEditable.

        https://bugs.webkit.org/show_bug.cgi?id=26856

        * fast/events/click-focus-anchor-expected.txt:
        * fast/events/click-focus-anchor.html:
        * platform/gtk/fast/events/click-focus-anchor-expected.txt: Added.
        * platform/mac/fast/events/click-focus-anchor-expected.txt: Removed.
        * platform/qt/fast/events/click-focus-anchor-expected.txt: Added.
2009-10-29  Erik Arvidsson  <arv@chromium.org>

        Reviewed by Eric Seidel.

        Make links mouse focusable only on GTK and QT.

        Links are now always mouse focusable on GTK and QT. On other platforms
        the link needs a tabIndex or it needs to be contentEditable.

        https://bugs.webkit.org/show_bug.cgi?id=26856

        * html/HTMLAnchorElement.cpp:
        (WebCore::HTMLAnchorElement::isMouseFocusable):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@50315 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-10-29 Jonathan Dixon <joth@chromium.org>
eric@webkit.org [Fri, 30 Oct 2009 02:30:42 +0000 (02:30 +0000)]
2009-10-29  Jonathan Dixon  <joth@chromium.org>

        Reviewed by Eric Seidel.

        Add expected files for GTK and QT builds, for test text-input-webkit-radius.html
        Expected fils collected from buildbot:
        http://build.webkit.org/results/Qt%20Linux%20Release/r50286%20(3257)/fast/css/
        http://build.webkit.org/results/GTK%20Linux%20Release/r50286%20(5308)/fast/css/

        * platform/gtk/fast/css/text-input-with-webkit-border-radius-expected.txt: Added.
        * platform/qt/fast/css/text-input-with-webkit-border-radius-expected.txt: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@50314 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoUpdated Tiger-specific result now that Tiger matches current behavior. A Tiger-specific
mitz@apple.com [Fri, 30 Oct 2009 02:00:54 +0000 (02:00 +0000)]
Updated Tiger-specific result now that Tiger matches current behavior. A Tiger-specific
result is still needed because of the difference in Thai fonts.

* platform/mac-tiger/fast/text/atsui-multiple-renderers-expected.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@50313 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoWebCore: Complete the fix for hit-testing and selection highlighting in ligatures...
mitz@apple.com [Fri, 30 Oct 2009 00:49:42 +0000 (00:49 +0000)]
WebCore: Complete the fix for hit-testing and selection highlighting in ligatures for the ATSUI
code path.

Reviewed by Sam Weinig.

* platform/graphics/mac/ComplexTextController.h: Added m_ltr member to ComplexTextRun.
* platform/graphics/mac/ComplexTextControllerATSUI.cpp:
(WebCore::ComplexTextController::ComplexTextRun::overrideLayoutOperation): Skip over deleted
glyphs, but update indexes and advances correctly.
(WebCore::ComplexTextController::ComplexTextRun::ComplexTextRun): Initialize m_ltr.

LayoutTests: Updated a test to account for font metric differences between Mac OS X versions. Removed
Leopard-specific results for another test now that Leopard matches current behavior.

Reviewed by Sam Weinig.

* platform/mac-leopard/fast/text/atsui-multiple-renderers-expected.checksum: Removed.
* platform/mac-leopard/fast/text/atsui-multiple-renderers-expected.png: Removed.
* platform/mac-leopard/fast/text/atsui-multiple-renderers-expected.txt: Removed.
* platform/mac/fast/text/ligature-subdivision-expected.txt:
* platform/mac/fast/text/script-tests/ligature-subdivision.js:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@50312 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoUpdating Radar numbers
adele@apple.com [Fri, 30 Oct 2009 00:26:30 +0000 (00:26 +0000)]
Updating Radar numbers

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@50311 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoREGRESSION(3.2.3 - 4.0.2): Message composing: when I undo a color change to text...
adele@apple.com [Thu, 29 Oct 2009 23:37:17 +0000 (23:37 +0000)]
REGRESSION(3.2.3 - 4.0.2): Message composing: when I undo a color change to text in Mail, undo/redo behaves strangely
<rdar://problem/7067033>
https://bugs.webkit.org/show_bug.cgi?id=30892

WebCore:

Patch by Enrica Casucci <enrica@apple.com> on 2009-10-29
Reviewed by Darin Adler.

This problem shows in any scenario where it is necessary to split a text
node to apply a style. SplitElementCommand and WrapContentsInDummySpanCommand both
have member variables initialized in the constructor to keep reference to elements
they need to operate upon. These reference are not updated when reapplying the command.
For this reason it is necessary to guarantee that unapply doesn not delete the references
and that these commands implement doReapply to correctly reuse the existing
elements.

Test: editing/undo/redo-style.html

* editing/SplitElementCommand.cpp:
(WebCore::SplitElementCommand::executeApply): Added.
(WebCore::SplitElementCommand::doApply): Modified to call executeApply.
(WebCore::SplitElementCommand::doUnapply): Doesn't release m_element1.
(WebCore::SplitElementCommand::doReapply): Added.
* editing/SplitElementCommand.h: Added doReapply and executeApply.
* editing/WrapContentsInDummySpanCommand.cpp:
(WebCore::WrapContentsInDummySpanCommand::executeApply): Added.
(WebCore::WrapContentsInDummySpanCommand::doApply): Modified to call executeApply.
(WebCore::WrapContentsInDummySpanCommand::doUnapply): Doesn't release m_dummySpan.
(WebCore::WrapContentsInDummySpanCommand::doReapply): Added.
* editing/WrapContentsInDummySpanCommand.h: Added doReapply and executeApply.

LayoutTests:

Patch by Enrica Casucci <enrica@apple.com> on 2009-10-29
* editing/undo/redo-style-expected.txt: Added.
* editing/undo/redo-style.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@50310 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-10-29 Jeremy Orlow <jorlow@chromium.org>
jorlow@chromium.org [Thu, 29 Oct 2009 23:22:33 +0000 (23:22 +0000)]
2009-10-29  Jeremy Orlow  <jorlow@chromium.org>

        Reviewed by Darin Fisher.

        DOM Storage's condition variable needs to handle spurious wakeups
        https://bugs.webkit.org/show_bug.cgi?id=30920

        Add a boolean to keep track of whether it's been terminated.  Clean
        up the locking code a tiny bit to make it easier to read.  There's
        no way to reproduce this reliably in a LayoutTest.

        * storage/LocalStorageThread.cpp:
        (WebCore::LocalStorageThread::LocalStorageThread):
        (WebCore::LocalStorageThread::terminate):
        (WebCore::LocalStorageThread::performTerminate):
        * storage/LocalStorageThread.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@50309 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agohttps://bugs.webkit.org/show_bug.cgi?id=30932
andersca@apple.com [Thu, 29 Oct 2009 23:22:26 +0000 (23:22 +0000)]
https://bugs.webkit.org/show_bug.cgi?id=30932
<rdar://problem/7350269>

Reviewed by Sam Weinig.

REGRESSION: Crash when turning on Private Browsing on site with flash.

Null check setValue.

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

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@50308 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoWebKitTools: Implement the Qt version of DRT dumpBackForwardList().
kenneth@webkit.org [Thu, 29 Oct 2009 22:49:35 +0000 (22:49 +0000)]
WebKitTools: Implement the Qt version of DRT dumpBackForwardList().

Patch by Kenneth Rohde Christiansen <kenneth@webkit.org> on 2009-10-29
Reviewed by Oliver Hunt.

* DumpRenderTree/qt/DumpRenderTree.cpp:
(WebCore::dumpHistoryItem):
(WebCore::DumpRenderTree::dumpBackForwardList):

LayoutTests: Removed test that passes after implementing the Qt DRT version
of dumpBackForwardList().

Patch by Kenneth Rohde Christiansen <kenneth@webkit.org> on 2009-10-29
Reviewed by Oliver Hunt.

* platform/qt/Skipped:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@50307 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-10-29 Dimitri Glazkov <dglazkov@chromium.org>
dglazkov@chromium.org [Thu, 29 Oct 2009 22:43:55 +0000 (22:43 +0000)]
2009-10-29  Dimitri Glazkov  <dglazkov@chromium.org>

        Unreviewed, build fix.

        [Chromium] Include ComplextTextController into Chromium Mac project.

        * WebCore.gyp/WebCore.gyp: Added include rule for ComplextText* files.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@50306 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-10-29 Eric Carlson <eric.carlson@apple.com>
eric.carlson@apple.com [Thu, 29 Oct 2009 22:32:20 +0000 (22:32 +0000)]
2009-10-29  Eric Carlson  <eric.carlson@apple.com>

        Reviewed by Kevin Decker.

        Refactor some duplicate plug-in clean up code into shared functions.

        * Plugins/WebPluginController.mm:
        (-[WebPluginController stopOnePlugin:]): New, stop a plug-in.
        (-[WebPluginController destroyOnePlugin:]): New, destroy plug-in.
        (-[WebPluginController stopAllPlugins]): Call stopOnePlugin.
        (-[WebPluginController destroyPlugin:]): Call stopOnePlugin and destroyOnePlugin.
        (-[WebPluginController destroyAllPlugins]): Call destroyOnePlugin.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@50305 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2009-10-29 Shinichiro Hamaji <hamaji@chromium.org>
hamaji@chromium.org [Thu, 29 Oct 2009 22:32:04 +0000 (22:32 +0000)]
2009-10-29  Shinichiro Hamaji  <hamaji@chromium.org>

        Reviewed by Darin Adler.

        fast/css/counters/invalidate-cached-counter-node can be dumpAsText
        https://bugs.webkit.org/show_bug.cgi?id=30931

        * fast/css/counters/invalidate-cached-counter-node-expected.txt: Added.
        * fast/css/counters/invalidate-cached-counter-node.html:
        * platform/mac/fast/css/counters/invalidate-cached-counter-node-expected.checksum: Removed.
        * platform/mac/fast/css/counters/invalidate-cached-counter-node-expected.png: Removed.
        * platform/mac/fast/css/counters/invalidate-cached-counter-node-expected.txt: Removed.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@50304 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoFix tabbing through element attributes in the Web Insector.
timothy@apple.com [Thu, 29 Oct 2009 21:59:33 +0000 (21:59 +0000)]
Fix tabbing through element attributes in the Web Insector.

https://bugs.webkit.org/show_bug.cgi?id=30429

Reviewed by Pavel Feldman.

* inspector/front-end/ElementsTreeOutline.js:
(WebInspector.ElementsTreeElement.prototype._attributeEditingCommitted): Don't call _updateTitle,
it is called for us when removeAttribute succeeds in the back-end.
(WebInspector.ElementsTreeElement.prototype._textNodeEditingCommitted): Ditto except for nodeValue.
(WebInspector.ElementsTreeElement.prototype._editingCancelled): Don't call _updateTitle, editing code reverts.
(WebInspector.ElementsTreeElement.prototype._updateTitle): Return early if we are editing.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@50303 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago[Qt] Remove some tests that now pass.
ossy@webkit.org [Thu, 29 Oct 2009 21:54:00 +0000 (21:54 +0000)]
[Qt] Remove some tests that now pass.

Patch by Csaba Osztrogonác <ossy@webkit.org> on 2009-10-29
* platform/qt/Skipped:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@50302 268f45cc-cd09-0410-ab3c-d52691b4dbfc